Download Editor Assistant Editor Editorial Assistant Editorial Consultant

Editor
Thomas N Walsh
Assistant Editor
Arnold DK Hill
Editorial Assistant
Helen Moore
Editorial Consultant
John Daly
Statistical Consultant
Alan Kelly
Editorial Advisers
E Quigley
D Bouchier-Hayes
S Tierney
H O’Connor
G O’Sullivan
J Lucey
A Clarke
L Viani
M O’Doherty
J Fenton
P Murray
Executive of the Academy
President
D Bouchier-Hayes
General Secretary
ADK Hill
Immediate Past President
BL Sheppard
Members
D Bouchier-Hayes
A Keenan
F O’Kelly
JD Gaffney
FE Murray
F Howell
This journal is indexed by Current Contents, Embase and is
Annual Subscription:
included in the abstracting and indexing of the Bio Sciences
Ireland and EU Countries
I156
Information Service of Biological Abstracts. It is available in
Non-EU
I192
microfilm from University Microfilms Ltd.
Single Copy
I 42
All communications for the Editor
should be addressed to:
2nd Floor, International House,
20-22 Lower Hatch St., Dublin 2
Tel: 00353 -1- 6623706
Fax: 00353-1- 6611684 e-mail: [email protected]
website: www.rami.ie www.iformix.com
Eireann Healthcare Publications,
25-26 Windsor Place, Dublin 2.
Tel: 01 475 3300,
Fax: 01 475 3311
Web: www.eireannpublications.ie
E-mail: [email protected]
Irish Journal of Medical Science • Volume 173 • Number 2
In House Editor: Susan Power
In House Sub Editor: Tim Ilsley
Design: Danny Lane
Operations Director: Caoimhe Tierney
Editorial Director: Maura Henderson
65
Irish Journal of Medical Science
April, May, June 2004; 173 (2): 65-116
SHORT REPORTS
72
Ethical approval for national studies in Ireland: an illustration of current challenges
M Smith, F Doyle, HM McGee, D De La Harpe
75
Fifty years of carotid surgery – hail and farewell?
PA Grace
ORIGINAL PAPERS
78
Patients’ experiences of dialysis services: are national health strategy targets being met?
K Rundle, O Keegan, HM McGee
82
Impaired renal allograft, but not patient survival, in patients with antibodies to hepatitis C virus
L Giblin, MR Clarkson, PJ Conlon, JJ Walshe, P O’Kelly, D Hickey, D Little, M Keoghan, J Donohoe
85
Risk factors for pulmonary embolism in an Irish patient cohort
S Timmons, R Liston, H Kelly
89
The cost of managing diabetic foot ulceration in an Irish hospital
D Smith, MJ Cullen, JJ Nolan
93
Emergency department post-coital contraception in Northern Ireland
S Mawhinney, O Dornan, R Ashe
96
Detection of mycobacterial DNA from sputum of patients with cystic fibrosis
M Devine, JE Moore, J Xu, BC Millar, K Dunbar, T Stanley, PG Murphy, AOB Redmond, JS Elborn
99
Documentation of do-not-resuscitate orders in an Irish hospital
J McNamee, ST O’Keeffe
102
"Is this a dagger I see before me?" — an audit of stabbings and gunshot wounds in Limerick
J Shabbir, CO McDonnell, JB O’Sullivan, K Cahill, A Moore, R Raminlagan, G Quinn, PA Grace
105
Neonatal transportation: the effects of a national neonatal transportation programme
D Mullane, H Byrne, TA Clarke, W Gorman, E Griffin, K Ramesh, T Rohinath
66
Irish Journal of Medical Science • Volume 173 • Number 2
Irish Journal of Medical Science
April, May, June 2004; 173 (2): 65-116
BOOK REVIEWS
111
Alzheimer Disease: Neuropsychology and Pharmacology
B McCabe
111
A Colour Handbook of Renal Medicine
PJ Conlon
112
Rapid Differential Diagnosis
P Naughton
CORRESPONDENCE
114
Isolated Crohn’s disease of the appendix
I Alam, N Nolan, J Geoghegan
GUIDELINES
116
68
General instructions to authors
Irish Journal of Medical Science • Volume 173 • Number 2
General instructions to author
General instructions to authors
The Irish Journal of Medical Science will consider for
publication original research of relevance to clinical
medicine. All material is assumed to be submitted
exclusively to the journal. Submitted material will be
sent for peer review and for statistical assessment and
may be styled and edited where appropriate. The evaluation of
the editorial committee is final.
Manuscripts, with disk, should be posted to:
The Editor
Irish Journal of Medical Science
International House,
20-22 Lower Hatch Street,
Dublin 2.
Email: [email protected]
Layout of paper
Manuscripts: Submit three copies, with disk, Word for
Windows format, size 12 font, paginated and double-spaced. See
First Issue, year 2000 for current format.
Scientific articles are set out under the headings: Abstract,
Introduction, Methods, Results, Discussion and References.
The title page contains the title, authors, institution and name,
contact number and email address of the corresponding author.
The Abstract of 150 words or less should be structured as
Background, Aims, Methods, Results and Conclusions.
Abbreviations should be kept to a minimum. Measurements
should be given in SI units. Blood pressure may be expressed in
mmHg. Drugs should be given their approved name.
Statistical method used should be detailed in the Methods
section and any not in common use should be referenced.
Ethics Committee Approval by the relevant authority is
needed for investigations on human subjects and animal studies
must be in accordance with the appropriate laws.
Tables and illustrations
Tables should not duplicate text information. Illustrations
should be professionally produced and may be photographic
prints or computer generated. Top should be indicated on the
back. Staining techniques should be stated for histological
material. An internal scale marker or the final print
magnification of the photograph should be included on the
photomicrograph. Patients shown in photographs should have
their identity concealed or give written consent for
publication. Provide legends on a separate sheet. Get written
consent from the authors and copyright holder to publish
material published elsewhere.
70
References
Number references consecutively in the Vancouver style (e.g.
style1). In the references, give the names and initials of all
authors but if more than six authors only the first three followed
by ‘et al’. Book titles should be followed by publisher, place of
publication, year and pages.
1. Allen JM, Keenan AK, McHale NG et al. Lymphatic
functions and cardiovascular disease. N Engl J Med 1999;
123: 10-4.
2. Davis GK, Mertt 0. Transition metals in nutrition. In: Trace
Elements in Nutrition (4th edition). 0 Mertt (ed).
Academic Press, San Diego 1998: 123-32.
Proofs and reprints
Proof corrections should be kept to a minimum and should
conform to the conventions shown in Whitaker's Almanack. A
charge of Ä70 per page (600 words) will be levied on authors of
full papers. Reprints are available; a scale of charges is included
when a proof is sent.
Letter of release
A covering letter must be included, signed by the senior authors
stating that:
“I have read this paper and the data and interpretation concurs
with my views.”
Checklist for authors
We have checked that the submitted article follows the
appropriate format:
• Three hard copies plus disk; use Times New Roman, size 12
font; normal typeface.
• Paginated and double line spacing; two spaces after full stops;
references superscript outside punctuation. Please follow
reference style as detailed above.
• A structured abstract as detailed above.
• Case reports must have a structured abstract.
• Three copies of graphics.
• A release letter by the senior author.
Submissions will not be processed if they are not presented in
the correct format for the journal.
Irish Journal of Medical Science • Volume 173 • Number 2
General instructions to authors
The Irish Journal of Medical Science will consider for publication original research of relevance to clinical
medicine. All material is assumed to be submitted exclusively to the journal. Submitted material will be sent for
peer review and for statistical assessment and may be styled and edited where appropriate. The evaluation of the
editorial committee is final.
Manuscripts, with disc, should be posted to:
The Editor
Irish Journal of Medical Science
International House,
20-22 Lower Hatch Street,
Dublin 2.
Email: [email protected]
Layout of paper
Manuscripts Submit three copies, with disk, Word for Windows format, size 12 font, paginated and doublespaced. See First Issue, year 2000 for current format.
Scientific articles are set out under the headings: Abstract, Introduction, Methods, Results, Discussion and
References.
The title page contains the title, authors, institution and name, contact number and email address of the
corresponding author.
The Abstract of 150 words or less should be structured as Background, Aims, Methods, Results and
Conclusions.
Abbreviations should be kept to a minimum. Measurements should be given in SI units. Blood pressure may
be expressed in mmHg. Drugs should be given their approved name.
Statistical method used should be detailed in the Methods section and any not in common use should be
referenced.
Ethics Committee Approval by the relevant authority is needed for investigations on human subjects and
animal studies must be in accordance with the appropriate laws.
Tables and illustrations
Tables should not duplicate text information. Illustrations should be professionally produced and may be
photographic prints or computer generated. Top should be indicated on the back. Staining techniques should
be stated for histological material. An internal scale marker or the final print magnification of the photograph
should be included on the photomicrograph. Patients shown in photographs should have their identity
concealed or give written consent for publication. Provide legends on a separate sheet. Get written consent from
the authors and copyright holder to publish material published elsewhere.
References
Number references consecutively in the Vancouver style (e.g. style1). In the references, give the names and
initials of all authors but if more than six authors only the first three followed by ‘et al’. Book titles should be
followed by publisher, place of publication, year and pages.
1. Allen JM, Keenan AK, McHale NG et al. Lymphatic functions and cardiovascular disease. N Engl J Med
1999; 123: 10-4.
1. Davis GK, Mertt 0. Transition metals in nutrition. In: Trace Elements in Nutrition (4th edition). 0 Mertt
(ed). Academic Press, San Diego 1998: 123-32.
Proofs and reprints
Proof corrections should be kept to a minimum and should conform to the conventions shown in Whitaker's
Almanack. A charge of 70 per page (600 wor ds) will be levied on authors of full papers. Reprints are available;
a scale of charges is included when a proof is sent.
Letter of release
A covering letter must be included, signed by the senior authors stating that:
"I have read this paper and the data and interpretation concurs with my views."
Checklist for authors
We have checked that the submitted article follows the appropriate format:
∑ Three hard copies plus disk; use Times New Roman, size 12 font; normal typeface.
∑ Paginated and double line spacing; two spaces after full stops; references superscript outside punctuation.
Please follow reference style as detailed above.
∑ A structured abstract as detailed above.
∑ Case reports must have a structured abstract.
∑ Three copies of graphics.
∑ A release letter by the senior author.
Submissions will not be processed if they are not presented in the correct format for the journal.
Ethical approval for national studies in Ireland: an illustration of current challenges
Ethical approval for national studies in Ireland:
an illustration of current challenges
M Smith1, F Doyle1,2, HM McGee1, D De La Harpe2
Health Services Research Centre, Department of Psychology1 and Department of Epidemiology and Public Health
Medicine2, Royal College of Surgeons in Ireland, Dublin, Ireland
Abstract
Background Ethical approval of research projects is, appropriately, an essential prerequisite in health settings.
Aims This paper outlines difficulties encountered with procedures for gaining ethical approval for two multicentre
surveys in Ireland.
Methods The experiences of two national surveys were documented.
Results Delays in processing ethics applications led to substantial delays in both surveys. Research ethics
committees (RECs) assessed applications in an idiosyncratic manner.
Conclusion In Ireland, there is currently no accepted mechanism for single location ethical approval for multicentre
studies. Instead, they require separate approval from all participating centres. The challenges of this system of
application to multiple committees are outlined in this paper, and possible solutions presented.
Introduction
Evidence-based practice is informed by research. Health services
re s e a rch provides policymakers and service providers with
i n f o rmation to plan for healthcare needs and for service
evaluation. Health research involves finding the evidence for
best practice and may be distinguished from audit, which
addresses whether best practice is being followed. Ethical review
of health services research is entirely appropriate and serves the
interest of the general public, funding agencies and the broad
research community. The process of requiring ethical approval
has expanded very rapidly from coverage of clinical trials only, to
coverage of all human and animal research.
At present, research ethics committees (RECs) may
distinguish between clinical trials and other human or animal
research (such as health services research and audit). In practice,
however, most of the RECs preside over both clinical trials and
other research submitted for approval, including audit. Audit,
while not strictly research, is often undertaken as a research-type
activity and will use staff time and resources. Institutions differ
in the extent to which they require ethical evaluation of audittype projects. In some institutions, there may be a wish to have
documented such information-gathering work that is conducted
on an irregular basis.
RECs have only recently been established in many centres and
there appears to be significant variation in their modes of
operation. For example, some RECs require the investigators to
attend REC meetings. Of concern is the fact that conflicting
decisions can be taken by different committees considering the
same or similar research protocols. The time taken to obtain
ethical approval for multicentre studies is also a concern. This
problem is not unique to Ireland.1 The EU is working to
establish formal procedures and time limits for ethics submission
and review.2 This paper provides examples from our recent
experience to illustrate areas of concern, to foster discussion of
the challenges among researchers in Ireland and to consider
some possible solutions.
Two studies undertaken recently by the Health Services
Research Centre (HSRC) at the Royal College of Surgeons in
72
Ireland (RCSI) involved data collection in national samples of
hospitals. One study was a national evaluation of the
management of acute coronary syndromes in all 39 hospitals
providing cardiac services. The other was a staff survey of
attitudes to organ donation. It was conducted in all relevant 37
public hospitals nationally. There is currently no single centre or
group who can provide ethical approval that is acceptable across
hospitals nationally. Both studies applied to and were given
ethical approval by the RCSI REC. Researchers then contacted
all potentially participating hospitals and asked about their ethics
review procedures. They noted that the study had already
received overall ethical approval from the RCSI REC. Details
relating to processing the project at the various centres
throughout Ireland are presented in Table 1.
A number of idiosyncratic requests arose in the process of
seeking ethical approval. In one hospital, the staff survey
protocol was sent to both the REC and a patient-focused
hospital liaison committee. While the REC provided approval,
the patient-focused committee requested a number of changes
to the (staff-only) interview schedule. In another hospital, REC
approval was granted but the researcher was subsequently
(informally) notified that approval was also needed from an
internal research access committee which aimed to evaluate the
level of staff input required to facilitate studies. This committee
was not mentioned in correspondence by the hospital’s chief
executive or the REC. Its existence only became evident when
plans to implement the study at ward level were initiated.
Approval from a risk management group was required following
REC approval at another centre. In another centre, one local
sponsoring consultant physician was required to make a
designated research application before the researcher could seek
REC approval. Two hospital executive committees and two
senior consultants considered the question of acceptability of the
study in two hospitals where RECs also operated. Finally, one
consultant continued to insist that the local hospital REC
provide evaluation (rather than the Health Board REC), despite
confirmation from the relevant CEO’s office that the hospital in
question did not have a local REC.
Irish Journal of Medical Science • Volume 173 • Number 2
M Smith et al
Table 1. Profile of the research ethics committee requirements to conduct national studies: two recent Irish projects
Study feature
Study 1
Study 2
Barriers to organ donation in Ireland:
ICU staff survey
Hospital management of acute coronary
syndromes in Ireland
Study format
Descriptive survey of staff attitudes
Patient chart data and patient self-report data
Data-gathering instrument
Anonymous postal questionnaire
distributed to relevant staff by human
resources departments of hospitals
Patient chart data and patient
completion of self-report questionnaire.
Patient’s name and permission for re-contact
12 months later were also requested
Funding body
Health Research Board
Department of Health & Children
Time frame for data collection
2001-2003
2003
Number of hospitals included
37
39
Initial contact
Letter to hospital chief executive officer
(CEO) introducing study and asking
about requirements for local research
ethics committee (REC) approval
Letter to CEO. Subsequent telephone
contact asking about REC
considerations. Decision taken to submit
protocol to each committee individually
as lack of clarity and delays regarding same
Subsequent contact
Referral to appropriate REC
Correspondence to RECs
Outcome
Hospital required application to individual 13
hospital REC on their standard form
(2 of these required re-submission)
13 (1 committee covered 2 hospitals,
(1 committee covered 2 hospitals,
another covered 8 hospitals)
Health board REC (n=8 health board
authorities)
2
(2 boards which covered 3 relevant
hospitals)
4
(4 boards which covered 15 relevant
hospitals)
Informal approval/approval by hospital
CEO’s office
21
3
*Submissions to other committees or
individuals required following local or
regional REC approval**
6
2
Researcher attendance at REC required
4
3 (plus 2 additional sites where local
consultant/PI was required to attend)
*Details given in text of paper.
**These are in addition to overall research approval by researcher institutions REC.
Most centres had their own application form. While there
w e re similarities across centres, the process of completing
multiple application forms was a time-consuming one. In many
cases, it was difficult to find out who to contact for details of the
hospital’s REC procedures and forms, when the next REC
meeting was scheduled and whether the researcher needed to
Irish Journal of Medical Science • Volume 173 • Number 2
attend. Other sources of delay in processing applications
including postponement of scheduled REC meetings.
One application took fully seven months to process because of
various administrative difficulties. Obviously this is a significant
strain on a project in terms of resources and logistics. An
estimate of researcher time required to process ethics
73
Ethical approval for national studies in Ireland: an illustration of current challenges
applications alone in a multicentre project is surprising — our
estimate is a two-month full-time equivalent period but with the
workload extending across approximately a five-month
timeframe: between time taken to complete different forms;
contacting committees for individual arrangements; travelling to
committees to attend meetings; resubmissions; re-contact for
other required procedures/committees etc. Ultimately all ethics
committee applications for the two projects were approved.
The time that may be required to obtain ethical approval is an
important concern for researchers and research funders. Failure
to predict and ‘cost’ for the time required for completion of the
REC approval process can result in overruns and even
abandonment of otherwise sound projects in relevant centres.
The time required is likely to be either underestimated by
researchers and/or considered excessive if requested in grant
applications by reviewers.
That acquiring ethical approval should be so fraught with
difficulties is not inevitable but rather suggests the system for
REC approval itself requires review. Some observations from our
experiences in conducting the studies presented above are
illustrative of how the system for REC approval has failed to
keep pace with the increasing demands made upon it.
Firstly, the administrative tasks associated with REC appear to
be poorly supported. In our experience, only one REC had
dedicated secretarial/clerical support. Most operated with the
REC chairperson (usually a senior hospital consultant) supported
by a member of hospital staff (frequently a professional employed
in the hospital laboratory setting), with clerical services supplied
on an ad hoc basis by personnel from various work areas within the
hospital. Tracking the pro g ress of applications is difficult. In many
cases, personnel from clinical and laboratory services within
hospitals appeared to have responsibility for progressing
applications and supporting RECs. It is assumed that this is in
addition to their routine tasks, which given the workload in most
hospital departments today, allows little scope for added duties.
Secondly, application forms for ethical review are generally
formulated in the ‘clinical trial’ model. Many RECs were set up
specifically under the Clinical Trials Act (1987). This format is ill
suited to a wider programme of research activity. In our
experience, completing the documentation and meeting the
stated requirements based on the ‘clinical trial model’ presented
a number of difficulties:
• Much of the information requested is inappropriate in the
context of health services research.
• T h e re is often a re q u i rement that the principal investigator (PI)
is a hospital consultant. This may be inappropriate in the context
of service re s e a rch and indeed has the potential to introduce bias
w h e re a project is effectively reliant on ‘sponsorship’ by a service
which is being assessed or investigated. The re q u i rement for a
named PI from a hospital in one of our studies resulted in all
c o rrespondence relating to the project being routed thro u g h
this local consultant (nominated and agreeing to act as PI to
facilitate pro g ress). This caused considerable confusion, delay to
the project and inconvenience to the consultant.
Finally, RECs may take on a role, or be re q u i red to act in a
manner, that goes beyond consideration of ethical concerns. One
a rea in which this is evident is comment on study methodology. It
is, of course, the case that poor methodology makes for bad
science. Quite appropriately there are ethical concerns in approving
a methodologically flawed study. In our experience however, some
questions raised by RECs showed poor comprehension of
methodological issues, or a willingness to propose alternative or
‘pre f e rred’ methodological strategies. For instance, one REC
initially re q u i red (and later retracted following correspondence)
74
alterations to the methodology in a manner that would have
c o m p romised its scientific integrity. In conducting multicentre
studies with single institutional RECs, it is particularly important
that the undertakings of the RECs should be restricted to the
review of ethical considerations.
A number of recommendations can be proposed from the
observations above. Firstly, the composition of an REC needs to be
t a i l o red to the topics/disciplines being assessed. Individuals with
expertise/experience in the re s e a rch methods to be used in the
submitted research are needed. Since the RECs under
consideration are hospital-based committees, there has typically
been a preponderance of clinical medical personnel on such groups.
Epidemiological and social science members are increasingly
needed to assess the wider range of projects submitted to RECs.
Secondly, consideration should be given to the development of
national guidelines for the operation of RECs. A single national
REC is an attractive option but likely to be unmanageable in terms
of the time requirement for staff or the legal responsibilities they
would face. However, this option may be re q u i red if the recent EU
directive2 is implemented fully. Directive 2001/20/EC may restrict
the volume of research being conducted. It is also likely that several
types of re s e a rch (e.g. re s e a rch on incapacitated participants) will
become increasingly difficult to conduct, as proxy consent of a legal
re p resentative will be required.3
A more manageable option may be a system of mutual
acceptance of approval across centres for multicentre studies
(this would be facilitated if centres were committed to national
guidelines). Another system would be for centres to focus on
specific types of ethical application, e.g. animal re s e a rc h .
Approval is now required from single RECs for projects as
diverse as hospital-based clinical trials, community health studies
and animal research. These options need to be discussed in open
fora if progress is to be made. The most acceptable organisers of
such discussions are likely to be the research funding agencies,
and we understand some developments are already underway.
This whole issue is all the more important given changing
requirements for data protection from the EU and from the
Data Protection (Amendment) Act 2003.
In essence, we are currently at a juncture where the
considerable efforts of many individuals to provide ethical review
of research projects need some co ordination and rationalisation
if the system is to remain manageable for those providing, using
and funding it.
References
1.
2.
3.
External Advisory Group (EAG) of the Cell Factory Key Action. From
Medical Biotechnology to Clinical Practice - Report of a Workshop
organised under the aegis of the Cell Factory Key Action. Quality of
Life Programme, European Commission, Research Directorate
General, Brussels, 2000.
D i rective 2001/20/EC of the European Parliament and of the
Council of 4 April 2001 on the approximation of the laws, regulations
and administrative provisions of the Member States relating to the
implementation of good clinical practice in the conduct of clinical trials
on medicinal products for human use. O fficial Journal of the European
Communities 2001; L121: 34-44 http://europa.eu.int/eurlex/en/archive/2001/l_12120010501en.html (accessed 13 January,
2004).
Singer EA, Mullner M. Implications of the EU directive on clinical
trials for emergency medicine. BMJ 2002; 324: 1169-70.
Correspondence to: Mr F Doyle, Health Services Research Centre,
Department of Psychology, Royal College of Surgeons in Ireland,
Mercer Street Lower, Dublin 2. Tel: (01) 4022718
Email: [email protected]
Irish Journal of Medical Science • Volume 173 • Number 2
Fifty years of carotid surgery — hail and farewell?
Fifty years of carotid surgery — hail and
farewell?
PA Grace
Department of Surgery, Midwestern Regional Hospital and the University of Limerick, Ireland
Abstract
Background The operation of carotid endarterectomy (CEA) has had an eventful 50 years.
Aim This report reviews the literature comparing the outcome of CEA with that of medical therapy, angioplasty and stenting.
Methods A review of the published randomised trials.
Results and conclusion After 50 years, CEA is still the gold standard in the management of severe symptomatic
carotid artery disease. It is doubtful, however, that in another 50 years it will continue to retain that pre-eminence.
Introduction
It is now 50 years since Eascott, Pickering and Rob from St
Mary’s Hospital, London, published their landmark report that
led to the development of carotid artery surgery worldwide.1 In
fact, Carrea, Molins and Murphy, a group of Argentinean neurosurgeons, performed the first carotid operation in 1951 in
Buenos Aires. They anastomosed the distal internal carotid
artery to the proximal external carotid artery in a 41-year-old
man who had presented with a hat-trick of symptoms: hemiparesis, aphasia and loss of vision.2 The Argentineans did not
report their case until 1955. Eastcott, Pickering and Rob in their
operation anastomosed the distal internal carotid artery to the
proximal common carotid artery to good effect.1 It was actually
Michael de Bakey in Houston who performed the first carotid
endarterectomy (CEA) operation as we know today in August
1953.3
The operation of CEA has had an eventful 50 years; it was
used sparingly at first, but between 1970 and the mid-1980s,
there was great enthusiasm for its performance, especially in the
US.4 Careful scrutiny revealed that this enthusiasm was possibly
misplaced.5 In 1988, an audit of a random sample of 1,302
Medicare patients showed that only one-third of procedures
were being done for what could be deemed at the time as appropriate reasons. Arteries were operated on because they were
occluded, had minimal stenosis or the patients were very high
risk. Sometimes even the wrong artery was operated on.5 These
concerns led to a reduction in the number of operations being
performed and became a catalyst for the most famous trials in
vascular surgery, the North American Symptomatic Carotid
Endarterectomy Trial (NASCET)6 and the European Carotid
Surgery Trial (ECST).7
NASCET and ECST trials
The aim of the NASCET and ECST trials was to compare surgical and medical therapy in the management of symptomatic
carotid disease. Medical therapy in the ESCT study was any dose
of aspirin while in the NASCET study patients received
1,300mg of aspirin. Both studies used angiography to assess the
degree of stenosis but different methods of calculation were
used to measure the degree of stenosis. Thus, a 60% NASCET
stenosis is equivalent to an 80% ESCT stenosis. The outcome of
the two studies is well known. Reduced to their simplest form,
the results showed that for patients with severe symptomatic
carotid disease (≥70% ipsilateral internal carotid artery stenosis)
overall surgical therapy confers a significant advantage over medIrish Journal of Medical Science • Volume 173 • Number 2
ical therapy. At two years in the NASCET study, there was a 9%
ipsilateral stroke rate in the surgery group compared to a 26%
ipsilateral stroke rate in the medical group with an absolute risk
reduction of 15% at two years. A recent analysis of the ESCT
data using the NASCET criteria showed that at five years there
was an absolute risk reduction of 21% in favour of surgery for
>70% symptomatic stenosis.8 There was no evidence of benefit
for CEA in patients with mild disease or near occlusion.
One interesting analysis of the ESCT results shows how a
given surgeon can see how his/her operative risk influences the
long-term benefit of CEA. Thus for a unit with a 2% operative
risk rate, nine CEAs are required to prevent one ipsilateral
stroke. If the operative risk increases to 10% then that surgeon
will have to perform 32 CEAs to prevent one stroke.9 For many
doctors, the results of the two studies signalled the end of the
debate regarding the management of severe symptomatic carotid
disease. Surgery was simply better.
However, a number of controversies remained after the
NASCET and ESCT studies were published. Could the trial
results be achieved in everyday practice? How should asymptomatic stenoses be managed? What are the implications of modern ‘best medical’ therapy? What is the role for carotid angioplasty and stenting?
Can the results of trials be achieved in practice? The answer is
probably yes but if people have bad results and do not report
them, then it is difficult to know. That is why audit, governance
and registries should be encouraged in modern surgical practice.
However, in a recent meta-analysis of almost 40,000 carotid
endarterectomies reported between 1980 and 2000 it would
appear that most surgeons do achieve results within the
American Heart Association Stroke Committee guidelines. A
very important observation to emerge from this analysis is that
the indication for surgery has a significant impact on the risk.
Thus, ocular transient ischaemic attacks have the same risk as
asymptomatic disease but urgent CEA for patients with ongoing
symptoms carries a very high risk.10 When people are reporting
the results of new therapies for carotid disease it is not enough
to divide patients into symptomatic and asymptomatic but a
breakdown of the results in respect of indications should also be
given.
Asymptomatic lesions
Perhaps nothing has been more controversial in the treatment of
carotid disease than the management of asymptomatic lesions.
In 1993, a small Veterans Administration study on 444 men
75
PA Grace
with >50% stenosis showed that there was a reduction in all neurological events with surgery (8%) compared to medical therapy
(20.6%), but there was no difference in overall stroke and death
rates for the two groups.11 In 1995, the Asymptomatic Carotid
Atherosclerosis Study (ACAS) of 1,662 patients reported that
CEA was better than medical therapy for asymptomatic lesions
>60%. The aggregate risk for ipsilateral stroke and any peri-operative stroke or death was estimated to be 5.1% for surgical
patients and 11.0% for patients treated medically.12
Very similar results were obtained from the Asymptomatic
Carotid Surgery Trial (ACST) of 3,200 patients (6.4% vs 11.7%)
recently presented in Dublin. The risk of stroke from an asymptomatic lesion of ≥70% seems to be about 2% per annum. It
would appear therefore that many asymptomatic lesions might
warrant surgical therapy.
Advances in medical therapy
Advances in medical therapy for cardiovascular disease are having a significant impact on management and one wonders what
the results might be if the NASCET and ESCT trials were now
repeated comparing surgery to current maximum medical therapy. There is good evidence to suggest that cessation of smoking, exercise, control of blood pressure, administration of antiplatelet agents and statin therapy have a significant effect on the
progression of cardiovascular disease.
Indeed, in a recent paper in the British Medical Journal, Wald
and Law advocated the use of a ‘Polypill’ not only for patients
with cardiovascular disease, but for everyone over the age of 55
years.13 Their proposed Polypill would contain a statin (e.g. atorvastatin [10mg] or simvastatin [40mg], three blood pressure
lowering drugs (e.g. a thiazide, a β-blocker and an angiotensinconverting enzyme inhibitor), each at half standard dose, folic
acid (0.8mg); and aspirin (75mg). They estimate that such a
combination of drugs would reduce ischaemic heart disease
events by 88% and stroke by 80%.13
Role of carotid angioplasty
What is the role of carotid angioplasty with or without stenting
in the management of carotid disease? Undoubtedly carotid
angioplasty with or without stenting can be performed with
good results. However, the ability to perform a procedure is not
necessarily an indication for it. So, what is the evidence?
Carotid angioplasty and stenting (CAS) has been performed
with success since the mid 1990s. Early observational studies
showed that the technique was feasible with reasonable
results.14,15 However, a number of randomised trials comparing
CAS with CEA failed to show an advantage for CAS15-19 and
indeed two trials were stopped because of poor results in the
CAS arm.17,18 In 2000, the authors of a meta-analysis of 33 single centre studies comparing CEA with angioplasty±stenting for
symptomatic carotid artery disease concluded. “At present,
endovascular treatment is not recommended for the majority of
patients with symptomatic carotid artery disease.”20 It looked as
though surgery for carotid disease would have a bright future.
A different picture seems to be emerging, however, with the
use of protection devices to be placed distally in the artery during angioplasty and stent placement. These devices are designed
to prevent what is perceived to be the most common and severe
complication of CAS, embolisation of debris to the brain. Two
studies using protection devices have been reported. The SAPPHIRE trial randomised 307 high risk patients (age >80 years,
congestive cardiac failure, severe chronic obstructive pulmonary
disease, previous CEA with restenosis, previous radiation therapy, radical neck surgery, abnormal lesion location) to CEA or
CAS with a distal protection device. Peri-operative stroke and
76
death rates were 7.3% for CEA and 4.4% for CAS. Thirty day
death, any stroke or myocardial infarction for CEA was 12.6%
versus 5.8% for CAS.21 The second trial was an observational
study of 437 high risk patients. All patients were treated with
CAS using the Guidant Acculink system and all used a protection device. The stroke/death rate was 6.6% and the
stroke/death/MI rate was 7.7%.22
These two studies showed that CAS with the use of a protection device gives reasonably good results in a high risk group of
patients. Whether the technique can be applied to all patients
who need interventional therapy for carotid artery disease is still
unknown and is the subject of four current randomised, controlled trials. They are: Carotid Revascularization
Endarterectomy vs Stent Trial (CREST) (USA), International
Carotid Stenting Study (ICSS or CAVATAS 2) (UK),
Endarterectomy Versus Angioplasty in patients with Severe
Symptomatic carotid Stenosis (EAV-3S) (France) and Stent-protected Percutaneous Angioplasty of the Carotid vs
Endarterectomy (SPACE) (Germany/Austria).23
These trials will collectively randomise over 7,000 patients to
either CEA or CAS and the results should be available by
2005/6. Now is the time for patience, the trials should be supported and results awaited.
Conclusion
CEA is a well-proven, safe operation with good evidence for who
should and should not be operated on, and who should perform
the operation. After 50 years, CEA is still the gold standard in
the management of severe symptomatic carotid artery disease. It
is doubtful, however, that in another 50 years it will continue to
retain that pre-eminence.
References
1.
Eastcott HHG, Pickering GW, Robb C. Reconstruction of intern a l
c a rotid art e ry in a patient with intermittent attacks of hemiplegia. The
Lancet 1954; ii: 994-6.
2. C a rrea R, Molins M, Murphy M. Surgical treatment of spontaneous
thrombosis of the internal carotid art e ry in the neck: carotid-carotideal anastomosis. Acta Neurol Latinoamer 1955; 1: 71-8.
3. Debakey ME. Successful carotid endart e rectomy for cerebrovascular
insufficiency. JAMA 233: 1083-5.
4. Dyken ML, Pokras R. The perf o rmance of endart e rectomy for disease
of the extracranial arteries of the head. Stroke 1984; 15: 948-50.
5. Winslow CM, Solomon DH, Chassin MR, Kosecoff J, Merrick NJ,
B rook RH. The appropriateness of carotid endarterectomy. N Eng J
Med 1998; 318: 721-7.
6. North American Symptomatic Carotid Endarterectomy Trial
Collaboration. Beneficial effect of carotid endart e rectomy in symptomatic patients with high grade stenosis. N Eng J Med 1991; 325: 44553.
7. E u ropean Carotid Surgery Trialists Collaborative Group. MRC
E u ropean Carotid Surg e ry Trial: interim results for symptomatic
patients with severe (70-90%) or mild (0-29%) stenosis. The Lancet
1991; 337: 1235-43.
8. Rothwell PM, Gutnikov SA, Warlow CP. Reanalysis of the final results
of the European Carotid Surgery Trial. Stroke 2003; 34: 514-23.
9. Naylor AR, Rothwell PM, Bell PR. Overview of the principal results
and secondary analyses from the European and North American randomised trials of endart e rectomy for symptomatic carotid stenosis. Eur
J Vasc Endovasc Surg 2003; 26: 115-29.
10. Bond R, Rerkasem K, Rothwell PM. Systematic review of the risks of
c a rotid endart e rectomy in relation to the clinical indication for and timing of surg e ry. Stroke. 2003; 34: 2290-301.
11. Hobson R. W. Weiss D. Fields WS et al. Efficacy of carotid endart e re ctomy for asymptomatic carotid stenosis. The Veterans Affairs
Irish Journal of Medical Science • Volume 173 • Number 2
Fifty years of carotid surgery — hail and farewell?
Cooperative Study Group. N Engl J Med 1993; 328: 221-7.
12. Executive Committee for the Asymptomatic Carotid Atherosclerosis
(ACAS) Study: Endart e rectomy for asymptomatic carotid art e ry stenosis. JAMA 1995; 273: 1421-8.
13. Wald NJ, Law MR. A strategy to reduce cardiovascular disease by more
than 80%. Br Med J 2003; 326: 1419-20.
14. Yadav JS, Roubin GS, King P et al. Angioplasty and stenting for
restenosis after carotid endart e rectomy: initial experience. S t roke 1996;
27: 2075-9.
15. T h e ron JG, Payelle GG, Coshun O, Huet HF, Guimareaus L. Carotid
artery stenosis: treatment with protected balloon angioplasty and stent
placement. Radiology 1996: 201: 627-36.
16. Endovascular versus surgical treatment in patients with carotid stenosis
in the Carotid and Ve rtebral Artery Transluminal Angioplasty Study
(CAVATAS): a randomized trial. The Lancet 2001; 357: 1729-37.
17. Naylor AR, Bolia A, Abbott RJ et al. Randomised syudy of carotid
angioplasty and stenting versus carotid endarterectomy: a stopped trial.
J Vasc Surg 1998; 28: 326-34.
18. Alberts MJ for the Publications Committee of the Wallstent Trial.
Irish Journal of Medical Science • Volume 173 • Number 2
19.
20.
21.
22.
Results of a multicenter prospective randomized trial of carotid artery
stenting vs carotid endarectomy (abstr). S t ro k e 2001; 32: 325.
B rooks WH, McClure RR, Jones MR et al. Carotid angioplasty and
stenting versus carotid endart e rectomy: randomized trial in a community hospital. J Am Coll Cardiol 2001; 38: 1598.
Golledge J, Mitchell A, Greenhalgh RM, Davies AH. Systematic comparision of the early outcome of angioplasty and endarterectomy for
symptomatic carotid artery disease. S t ro k e 2000; 31: 1439-43.
Yadav J. Presented at the American Heart Association Scientific
Sessions, November 2002. www.medscape.com/viewart icle/445125
ARCHeR: ACCULINK for Revascularization of Carotids in HighRisk Patients. Preliminary 30-Day Results. Presented at the American
College of Cardiology 52nd Annual,Scientific Session, March 2003.
www.medscape.com/viewarticle/451826 www.strokecenter.org/trials/
Correspondence to: Professor Pierce A Grace, Department of
Surgery, Midwestern Regional Hospital, Dooradoyle, Limerick.
77
Patients’ experiences of dialysis services: are national health strategy targets being met?
Patients’ experiences of dialysis services: are
national health strategy targets being met?
K Rundle, O Keegan, HM McGee
Health Services Research Centre, Department of Psychology, Royal College of Surgeons in Ireland, Dublin, Ireland
Abstract
Background The National Health Strategy envisages a health system incorporating patient views; and providing
accessible, consultant-led dialysis services with patient choice of dialysis modality, in all regions.
Aims To describe patients’ experiences of renal services against National Health Strategy objectives.
Methods Telephone interviews with 192 dialysis patients from three hospitals in the Eastern region.
Results One-quarter of participants (16% of haemodialysis [HD] and 46% of peritoneal dialysis patients) lived outside the
Eastern region, and travelled there because dialysis was not available locally. Two-thirds (65%) had a choice of dialysis
modality. High satisfaction with interpersonal care was observed (83–98% satisfaction). Dissatisfaction with physical
environment included parking (39–56%), waiting areas (62–69%), HD unit space (74%). Regarding support services,
dietary services were satisfactory (92–95%), with lower satisfaction ratings for social and financial support services (62%).
Conclusions Structural and management issues must be addressed to advance a quality agenda for renal care in Ireland.
Introduction
Recent health service policies have championed a consumeroriented approach to healthcare. The latest Irish National
Health Strategy envisages a new health system “that encourages
you to have your say, listens to you, and ensures that your views
are taken into account”.1 Thus, health service evaluation has
begun to consider patients’ views of their care. This is often
assessed using measurements of patient satisfaction. Patient
satisfaction studies can provide useful information on the quality
of services and how best to improve them.
Patients with end-stage renal disease (ESRD) face many
challenges in undertaking the life changes necessary to
accommodate a dialysis routine. Haemodialysis (HD) requires
approximately three weekly visits to hospital for three-hour
sessions, while continuous ambulatory peritoneal dialysis
(CAPD) and continuous cyclical peritoneal dialysis (CCPD)
require completion of daily self-management tasks at home.
Since dialysis has an ongoing impact on the life of the patient,
their views of care are particularly important.
The present study evaluates current renal services in the Eastern
region of Ireland in the context of plans for improvements to renal
services. The National Health Strategy1 proposed specific
objectives for quality renal services: to provide dialysis services
close to patients’ homes; to provide nephrology services with
specialist oversight; and to make available various modalities so
patients can chose to have dialysis in their own home (if that is
their pre f e rence and in line with medical suitability).1
This study specifically measured levels of patient satisfaction.
Patient satisfaction may be particularly relevant in dialysis
patients since studies have shown that satisfied patients are more
likely to adhere to medical regimes.2-5 It is particularly important
that renal dialysis patients adhere to medical regimes, for
example, dietary restrictions, since non-adherence decreases
their long-term survival rates. Non-adherence to dietary
restrictions has been found to be common in Irish renal
patients.6
Patient satisfaction studies of dialysis services have been
completed in many countries, evaluating dialysis services across
78
various dimensions, including the physical environment,
interpersonal aspects of care, and psychosocial support.4,7,8 The
present study measures levels of patient satisfaction with these
dimensions of care among Irish dialysis patients.
The study objectives were to investigate current renal services
in the context of objectives stated in the National Health
Strategy, and to determine levels of patient satisfaction with
various dimensions of the care provided in three Dublin
hospitals. Thus information collected can inform service
planning activities and provide a baseline from which future
studies can monitor changes in patient satisfaction following
service changes.
Methods
Sample
Renal dialysis units in three (of four) Irish hospitals providing
such services in the Eastern Regional Health Authority (ERHA)
participated. The units treat approximately 370 HD, CAPD and
CCPD patients. The research protocol received ethical approval
from the three hospitals.
Measures
A previously validated outpatient satisfaction questionnaire was
adapted to the Irish renal dialysis setting.9 It examined multiple
dimensions of patient satisfaction using five-point rating scales
and incorporated open-ended questions to collect qualitative
experiences. Questionnaire adaptation was completed with staff
feedback from all three centres.
Procedure
All patients at the three hospitals received a letter from their
consultant nephrologist inviting them to participate. They
returned consent forms to a research team independent of the
hospitals indicating if they wished to take part. A total of 192
patients took part (response rate 59%). This response rate is
satisfactory for a group of patients with a very serious and timedemanding medical condition. The study questionnaire was
administered by telephone interview.
Irish Journal of Medical Science • Volume 173 • Number 2
K Rundle et al
Results
Demographic and dialysis profile of participants
Of the 192 patients who participated, 116 (60%) were men.
Median age was 58 years (range 17–88 years), with over onet h i rd (37%) aged 65 years or older. One-third (32%) of those
below the State retirement age (66 years) were unable to work
due to their renal condition. Patients were a median 24
months on continuous dialysis (i.e. length of time since
starting dialysis or since most recent transplant failure,
whichever was the most recent) with a range of 2 weeks to 14
years. HD participants received a median three sessions of
dialysis per week (range 2–4), of a median 3.5 hours per
session (range 2.5–4 hours). Results are presented in relation
to the National Health Strategy objectives.
National Health Strategy objectives
Choice:
“to widen the availability of alternative dialysis treatment
programmes to allow patients to manage their dialysis care at
home” (if that is their preference and is in line with medical
suitability)
Most patients (69%) were currently availing of HD, with 15%
using CAPD and 16% using CCPD. Significant variations in
levels of dialysis modality existed across hospitals, with one unit
maintaining 49% of its patients on HD, 32% on CAPD and 19%
on CCPD. Another unit provided HD to 74% of participants,
with 8% on CAPD and 18% on CCPD. The third unit offered
only HD services.
Participants were asked if they knew why they were treated
with their current type of dialysis (HD, CAPD or CCPD). Over
three-quarters (76%) reported they knew why they were
assigned to their particular type of dialysis. There were variations
across hospitals, with only 37% of the participants from the
hospital that provides only HD knowing why they were on their
current type of dialysis. Similarly, while two-thirds (65%) felt
they had a choice of type of dialysis, fewer (21%) of participants
from the hospital providing only HD felt they had a choice of
which dialysis type to use. These results must be interpreted
cautiously since the hospital provided no alternative to HD
internally, but referred potential peritoneal dialysis patients to
another hospital for suitability evaluation. However, the results
should not be discounted if the Government commitment is to
provide choice of treatment to all patients (within the limits of
medical suitability).
Qualitative analysis of participant comments also highlighted
differences across hospitals. Most respondents who felt they had
not had a choice of dialysis type explained further that other
dialysis types were not possible due to medical reasons (e.g.
previous surgery, infection problems), emergency start of dialysis
or due to disability restrictions. However, a number of
respondents from the hospital that provided only HD said they
knew nothing, or very little, about peritoneal dialysis.
Accessibility:
“to ensure that regional dialysis centres are adequately resourced
to give patients access to services close to their own home”
HD patients lived closer (median 5.5 miles; range 0.5–50
miles) to their treating hospital than CAPD/CCPD patients
(median 40 miles; range 1–200 miles). Thus, those who lived
farthest from the hospital were patients who travelled to the
hospital less frequently. Nonetheless, 9% of HD participants
lived over 50 miles away from a service they needed to use two
to four times weekly, and 41% of peritoneal dialysis participants
lived over 50 miles away from their treating hospital (with
appointments approximately every two months).
Irish Journal of Medical Science • Volume 173 • Number 2
Consultant-led service:
“to ensure the availability of consultant-led services in all
regions”
Sixteen per cent of HD patients and 46% of peritoneal dialysis
patients lived outside the health board region studied. Most
reported that they travelled to the Eastern region for dialysis
services because there were no dialysis services locally.
Patient satisfaction
Dimensions of patient satisfaction fell into three main categories
of health services care: interpersonal care, physical environment
and management, and psychosocial care and support.
Interpersonal care
High levels of satisfaction with the quality of care received from
medical and nursing staff were observed for both the HD unit
and renal outpatient clinic (see Figure 1). Care provided in the
HD unit was highly rated. Satisfaction was high for consultants
(92%), other doctors (92%) and renal nurses (95%). Similar
ratings were noted for care provided in the outpatient clinic by
consultants (96%), other doctors (95%) and renal nurses (96%).
Physical environment and management
Parking convenience for outpatient clinics received low
satisfaction ratings (39% satisfaction) with little variation across
hospitals (see Figure 2). Analysis of participant comments
Figure 1. Percentage of patients satisfied with aspects of
interpersonal care.
highlighted the main issues to be expensive parking and a lack of
disabled or dialysis patient parking close to the hospital.
However, ratings of parking for HD units varied across hospitals
(33%, 75% and 100%).
Waiting areas received varying ratings. Tw o - t h i rds were
satisfied with renal outpatient clinic waiting areas (HD clinics,
69%; peritoneal dialysis clinics, 62%). However, when analysed
by individual hospital, areas of particular dissatisfaction were
highlighted, with one hospital’s peritoneal dialysis clinic waiting
area receiving 35% satisfaction and the other hospital rated as
82% on this aspect. Waiting time for HD sessions was rated
highly overall (86%). However, a breakdown by hospital showed
a significant difference in satisfaction across hospitals (χ2=26.76;
p<0.001). One hospital received lower ratings (42% rated
waiting time as poor or fair). Some patients there reported
waiting 1–1.5 hours to begin a HD session.
While most patients (84%) found HD units comfortable, onequarter (26%) were not satisfied with space available at the HD
79
Patients’ experiences of dialysis services: are national health strategy targets being met?
information; these to be distributed when patients begin dialysis.
Satisfied participants reported receiving information regarding
entitlements and support when first starting dialysis, and
receiving support from a hospital social worker.
Discussion
Regarding choice of dialysis modality, most participants used
HD. While one hospital offered only this option (this hospital
now offers CAPD and CCPD in addition to HD), half of
Figure 2. Percentage of patients satisfied with aspects of
physical environment and management.
units (unit seen as too small and overcrowded), with one
hospital receiving lower ratings (45% not satisfied). Qualitative
analysis of participant suggestions for improvements to HD
units highlighted two main issues across hospitals. The first
c o n c e rned improvements to the physical environment, in
particular, having larger units, more comfortable beds and
improved air-conditioning. The second issue concerned the lack
of dialysis machines, and participants suggested increasing the
numbers of dialysis machines would reduce HD waiting times,
increase patient choice of dialysis time and increase the flexibility
of the units.
Satisfaction with waiting times for outpatient clinics varied,
with waiting times for HD clinics receiving lower ratings,
ranging from 43% satisfaction to 71% satisfaction across
hospitals. Alternatively, peritoneal dialysis outpatient clinics
received high ratings (92%). Qualitative analysis of participant
comments suggest that this was due to smaller patient numbers
attending these clinics.
Other areas in which hospital ratings differed significantly
were respect shown for patient privacy in HD units (61%, 84%
and 89% satisfaction; χ 2=9.91; p<0.05) and the way time is
occupied during HD (64%, 90% and 95% satisfaction; χ2=17.01;
p<0.005).
Psychosocial care and support
Relevant renal support services include dietary advice, financial
and social support services, and emergency medical support.
About half of participants had seen a dietician in the last two
months (median 61 days; range 0–4 years), with large
differences across hospitals (one hospital reporting a median 116
days since last dietician consultation). Most aspects of
psychosocial care were rated highly (see Figure 3); 92% were
satisfied with the availability of a dietician, 90% satisfied with the
availability of a hospital social worker and 85% satisfied with the
medical emergency support procedure.
Notably lower satisfaction ratings (62% satisfied) were given
for the availability of information and advice regarding social and
financial support, with some variation across hospitals (range
46%–73% satisfaction). This was further investigated through
qualitative analysis of participant supporting comments. Many
participants highlighted the need for more information.
Suggestions included increasing the availability of information
leaflets, of information packs containing financial entitlement
information and application forms, and of social support contact
80
Figure 3. Percentage of patients satisfied with aspects of
psychosocial care and support.
participants from another hospital used HD, compared with
three-quarters of participants from the third. Reasons for this
may reflect, for example, differences in patient profiles across
hospitals, hospital re s o u rces and expertise, or professional
pre f e rences. Comparisons of mortality rates for HD and
peritoneal dialysis have yielded conflicting results, with neither
modality consistently being shown to produce greater survival
rates.10-12 Thus, it is important that choices about peritoneal
dialysis or HD are not constrained by readily amenable factors.
One-third of the group felt they did not have a choice of
dialysis modality. While this was often for medical reasons, some
participant comments indicated a lack of awareness of alternative
dialysis types. A major barrier to patient selection of peritoneal
dialysis has been shown to be lack of explanation of the
alternative dialysis modalities.13 Patients forced to use CAPD
have also been shown to have lower levels of quality of life scores
than those who voluntarily used CAPD.14 Thus,
recommendations for service improvements include increasing
the availability of information and patient understanding of all
dialysis modality alternatives open to them, and ensuring that
dialysis modality is chosen by the patient (to the extent that this
is medically possible).
Regarding service accessibility, one-quarter of participants
travelled from other health board areas to avail of dialysis
services in the Eastern region. The main reason for travelling to
the Eastern region was lack of local dialysis services. While
almost half of all peritoneal dialysis participants lived outside the
region, their hospital visits would generally be less frequent
(appointments approximately every two months). Similarly,
while only 16% of HD participants lived outside the Eastern
region and 8% travelled more than 50 miles, the question of
accessibility of dialysis services must be addressed since these
patients must travel to hospital for HD three times per week.
Previous research has shown an increase in patient satisfaction
following a reduction in dialysis travel times (due to opening of
Irish Journal of Medical Science • Volume 173 • Number 2
K Rundle et al
sub-regional renal unit), thus illustrating the importance of
locating dialysis services close to patients’ homes.15
Many patients clearly travelled long distances to avail of
dialysis treatments in the Eastern region of Ireland. Reasons for
this could not be ascertained from this study. They may reflect
patients’ hospital or consultant preferences, patients’ previous
contact with the hospital, etc. However, the high proportion of
patients travelling into the Eastern region from other areas to
avail of dialysis services must be addressed since one-quarter of
all patients lived outside the region and since most reported that
they travelled to the Eastern region because there were no
dialysis services locally.
The most consistently satisfactory dimensions of services
related to interpersonal care. Extremely high levels of satisfaction
were recorded for patient encounters with consultants, other
doctors and renal nurses.
Less satisfactory aspects related to physical environment and
waiting times. When analysed by individual hospital, interesting
variations were noted, highlighting that it is possible to
successfully address this ongoing hospital service issue. HD
patients spend three long visits weekly in hospital. Their
feedback highlighted the importance of comfortable
surroundings. Simple changes, such as the removal of televisions
from one unit, had very negative effects on the quality of their
care environment. Time to wait for a dialysis machine to become
available was also criticised. Participant suggestions for
improvements included increasing the number of dialysis
machines.
Regarding outpatient clinics, there were lower satisfaction
ratings for the waiting area and waiting times for outpatient
appointments. Again wide variations in satisfaction across
hospitals demonstrate that it is possible to address this ongoing
hospital issue. Similarly, parking convenience for both HD
sessions and outpatient clinic appointments created problems.
This was both the expense of parking and the lack of disabled or
dialysis-specific parking spaces close to hospitals. Again, one
hospital received notably higher ratings, thus illustrating that it
is possible to successfully address this problematic area of
hospital service. This aspect was particularly noted by HD
patients, who attend hospital an average three times per week.
Practical resources such as dietary, social support and
financial entitlement information are a necessary and
important aspect of renal services. Some hospitals had better
patient satisfaction ratings, in particular for information
provision, than others. The area of psychosocial support
clearly re q u i res further investigation and perh a p s
consideration of the extent to which the hospital should take
responsibility for this area. With such close and regular
contact with dialysis patients, hospital staff may be best placed
to identify those patients most in need of psychosocial
s u p p o rt, or indeed to develop a system whereby they
themselves can deliver this to a certain extent (for example, by
distributing financial information packs or social support
contact information) to all patients in their early stages of
kidney failure diagnosis, or alternatively by referral to the
appropriate support services.
This evaluation of renal services in the ERHA from the patient
perspective illustrates the very high levels of satisfaction with the
interpersonal aspects of renal care, while highlighting some
deficits in the structural and management aspects of the delivery
of care. While the high quality of interpersonal care is laudable
and needs acknowledgement, structural and management issues
need to be addressed by clinical renal staff in association with
health administration staff if further progress is to be made on a
quality agenda for renal patient care in Ireland.
Irish Journal of Medical Science • Volume 173 • Number 2
Acknowledgements
This study was funded by the ERHA. Thanks to renal patients
and hospital staff at three ERHA hospitals, Ms Evelyn Jameson
(ERHA), Ms Denise O’Shea (study interviewer) and Mr Ronan
Conroy (RCSI) for their assistance with the study.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
D e p a rtment of Health and Children. Quality and Fairness. A Health
System for You. Government Publications, Dublin, 2001.
Hall JA, Roter DL, Katz NR. Meta-analysis of correlates of provider
behaviour in medical encounters. Med Care 1988; 26 (7): 657–75.
H a rris LE, Luft FC, Rudy DW, Tierney WM. Correlates of health care
satisfaction in inner-city patients with hypertension and chronic renal
insufficiency. Soc Sci Med 1995; 41 (12): 1639–45.
Kovac JA, Patel SS, Peterson RA, Kimmel PL. Patient satisfaction with
care and behavioural compliance in end-stage renal disease patients
treated with hemodialysis. Am J Kidney Dis 2002; 39: 1236–44.
Milas NC, Nowalk MP, Akpele L et al. Factors associated with
a d h e rence to the dietary protein intervention in the Modification of
Diet in Renal Disease Study. J Am Diet Assoc 1995; 95 (11):
1295–300.
McGee HM, Rushe H, Sheil K, Keogh B. Association of psychosocial
factors and dietary adherence in haemodialysis patients. Br J Health
Psychol 1998; 3: 97–109.
Alexander GC, Sehgal AR. Dialysis patient ratings of the quality of
medical care. Am J Kidney Dis 1998; 32 (2): 284–9.
Wuerth DB, Finkelstein SH, Kliger AS, Finkelstein FO. Patient
assessment of quality of care in a chronic peritoneal dialysis facility. Am
J Kidney Dis 2000; 35 (4): 638–43.
Carey RG, Seibert JH. A patient survey system to measure quality
i m p rovement: questionnaire reliability and validity. Med Care 1993; 31
(9): 834–45.
Fenton SS, Schaubel DE, Desmeules M et al. Hemodialysis versus
peritoneal dialysis: a comparison of adjusted mortality rates. Am J
Kidney Dis 1997; 30 (3): 334–42.
H a rris SA, Lamping DL, Brown EA, Constantinovici N. Clinical
outcomes and quality of life in elderly patients on peritoneal dialysis
versus hemodialysis. Perit Dial Int 2002; 22 (4): 463–70.
Maiorca R, Vonesh EF, Cavalli P et al. A multicenter, selection-adjusted
comparison of patient and technique survivals on CAPD and
hemodialysis. Perit Dial Int 1991; 11 (2): 118–27.
Mclaughlin K, Manns B, Mortis G, Hons R, Taub K. Why patients
with ESRD do not select self-care dialysis as a treatment option. Am J
Kidney Dis 2003; 41(2): 380–5.
Szabo E, Moody H, Hamilton T et al. Choice of treatment improves
quality of life. A study on patients undergoing dialysis. Arch Intern
Med 1997; 157 (12): 1352–6.
Willis CE, Watson JD, Casson K et al. Locations for renal services –
patient satisfaction surveys. Ulster Med J 1998; 67 (2): 110–4.
Correspondence to: K Rundle, Health Services Research Centre,
Department of Psychology, Royal College of Surgeons in Ireland,
Mercer Street Lower, Dublin 2, Ireland. Email:[email protected]
81
Impaired renal allograft, but not patient survival, in patients with antibodies to hepatitis C virus
Impaired renal allograft, but not patient survival,
in patients with antibodies to hepatitis C virus
L Giblin1, MR Clarkson1, PJ Conlon1, JJ Walshe1, P O’Kelly1, D Hickey2, D Little2, M Keoghan3, J Donohoe1
Departments of Nephrology1, Renal Transplantation2 and Immunology3, Beaumont Hospital, Dublin, Ireland
Abstract
Background The impact of hepatitis C virus (HCV) infection in renal transplant patients is controversial and there are
no data on the outcome of renal transplantation in this sub-group of Irish patients.
Aim To examine the outcome of renal transplantation in patients with hepatitis C.
Methods We examined the outcome of first grafts from renal transplant patients with hepatitis C antibody positive and
compared them to a control group. During this period, 24 HCV positive patients received 33 grafts. All were treated
with standard immunosuppression.
Results Graft survival rate was less in the HCV positive cases (p=0.0087). Graft survival at 1 year was 75% in the
HCV positive group versus 85% in the HCV negative group, 40% versus 62% at 5 years and 14% compared with 40%
at 10 years. Patient survival was similar in both groups (p=0.78). Patient survival at 1 year was 96% versus 94%, 87%
versus 80% at 5 years and 70% in both groups at 10 years.
Conclusion In the Irish renal transplant population, the presence of hepatitis C antibodies, before or after
transplantation is associated with worse long-term graft, but not patient survival.
Introduction
The long-term impact of hepatitis C virus (HCV) infection in renal
transplant recipients remains controversial. HCV infection is a
relatively common disease and approximately 3% of the world’s
population is chronically infected.1 HCV accounts for
a p p roximately 20% of cases of acute and 70% of chronic hepatitis.2
The infection persists in about 80% of cases and is the major cause
of cirrhosis and hepatocellular carcinoma.3 HCV is relatively
common in dialysis patients worldwide although there is
considerable variation between geographical locations.2 The
p revalence in Northern European dialysis units (Ireland, UK,
Scandinavia) is reported to be <5% while prevalence in the USA is
between 10 and 20%. The Mediterranean countries such as Spain,
Italy, France and Greece have a higher prevalence in their dialysis
units, reported to be >20%.4 The incidence in Ireland is low by
international standards. In 1993 we reported an incidence of 1.7%
in our chronic haemodialysis population (n=266) and an incidence
of 1.1% in a matched control group of our renal transplant
recipients (n=272).5 The prevalence in our chronic haemodialysis
population in December 2002 was 2.6% (using a third generation
assay). The long-term effect of HCV infection on renal transplant
recipients and on the allograft remains uncertain. We there f o re
perf o rmed a re t rospective review of the Irish renal transplant
population and compared those patients who had antibodies to
HCV to those who were negative for these antibodies. The longterm patient and graft survival rates were compared for each group.
Methods
We analysed our renal transplant database to detect all kidney
transplant recipients who had been diagnosed with antibodies
positive for HCV. Data were attained from the Irish Renal
Transplant Registry. All renal transplant patients who received a
renal transplant at our centre between 1986 and 2001 were
included in the analysis. Prospective testing for HCV was first
introduced in this country in 1991 and was routinely perf o rmed in
all renal transplant donors and recipients from 1992. The initial
82
s c reening involves detection of antibodies to HCV using an ELISA
test. If a patient is shown to be antibody positive the presence of
HCV is then confirmed using PCR techniques. All patients
commencing dialysis therapy at our centre are now routinely
s c reened for antibodies to HCV and also for hepatitis B (HBV).
Patients who are being considered for renal transplantation and all
renal transplant donors are also routinely screened for HCV, HBV,
cytomegalovirus and Epstein-Barr Vi rus.
We reviewed the average age of each of these HCV antibody
positive patients who reached end-stage renal disease (ESRD),
the length of time spent on dialysis prior to receiving a kidney
transplant, the mode of dialysis therapy used and the number of
blood transfusions administered to each patient. We also
observed the age at which the patient received his or her first
renal allograft and the total number of transplants each patient
received. These data were directly compared to a 2:1 time
matched control group of 48 patients taken from our general
renal transplant population (which was reflective of our overall
transplant group). This matched group consisted of the patients
who received the kidney transplant at our institution immediately
before and immediately after the HCV antibody positive patient.
Kaplan-Meier survivor functions were used to estimate
percentage graft loss and patient mortality in the study. Log Rank
tests were used to compare survival rates between the two
groups. Differences in patient characteristics were assessed with
Wilcoxon’s rank sum and Fisher’s exact tests. Results were
deemed significant for p values <0.05. All of the statistical analysis
was conducted using Stata (Version 8, Texas).
Results
There were a total of 1,686 kidney transplant recipients at our
centre during the study period 1986–2001. During this time
period a total of 24 patients received a renal transplant in our unit
who were diagnosed as HCV antibody positive. The diagnosis of
HCV was made from 1991 onwards in all of these patients. It was
not known for how long prior to making the diagnosis that the
Irish Journal of Medical Science • Volume 173 • Number 2
L Giblin at al
Table 1. Patient characteristic table
Number of patients
Male:female % Ratio
Haemodialysis: peritoneal dialysis
Median age at ESRF
Median age at 1st transplant
Median time on renal replacement therapy
Percentage requiring re-transplantation
HCV+
HCV-controls
p value
24
58.3: 41.7%
3.6:1
37 years
39 years
28 months
37.5%
48
58.3: 41.7%
2.5:1
39 years
40 years
17 months
16%
1.000
0.578
0.437
0.761
<0.001
0.079
to 1991 when the test became available here in Ireland. The
other 12 patients received their first graft subsequent to the
availability of this screening technique. Of these, seven patients
received their first graft while they were known to be HCV
antibody positive.
The first transplant was a living related donor graft (LRD) in
one of the 24 patients while the remaining 23 transplants were
cadaveric grafts. The patient who received a graft from a LRD
was the only patient who was transplanted pre-emptively. The
remaining 23 patients were established on renal replacement
therapy in the form of haemodialysis in 18 patients and peritoneal
dialysis in five patients.
Graft survival
Figure 1. Graft survival in HCV positive and control kidney
recipients.
We compared the HCV antibody positive transplant patients to
our era matched control group who were antibody negative. The
first transplant was prior to 1991 in 12 of these HCV positive
patients and after/during 1991 in the remaining 12 recipients.
The graft survival rate was significantly less in the HCV positive
cases: 75.0% in the HCV positive group versus 85.4% in the HCV
negative group at 1 year, 57.4% versus 70.9% at 3 years, 39.7%
versus 61.7% at 5 years and 13.7% compared with 39.5% at 10
years. Using Log Rank tests, the difference in graft outcome for
first grafts was highly significant with an overall p value of
0.0087. Figure 1 demonstrates the difference in outcome
between HCV positive patients and HCV negative controls.
To date there are only 5/24 of the first transplants still
functioning at 10, 73, 95, 104 and 221 months post
transplantation. In one case the patient died with a functioning
graft in situ, which had been placed 8.5 years prior to his death.
The median time to failure of HCV positive patients was 3.41
years compared to 8.67 years in the control group.
Patient survival
Figure 2. Patient survival for HCV positive and control
kidney recipients.
patient was actually antibody positive, as we were unable to
determine the exact time of seroconversion. We presume that in
the majority, if not all of these cases the patient contracted the
HCV infection while they were on dialysis therapy.
Table 1 shows the patient details for HCV positive and contro l
patients. Demographic characteristics were similar for the two
groups with age at time of transplantation slightly lower for the
HCV positive patients. The time on dialysis therapy while awaiting
first transplant was significantly longer for the HCV positive patients
compared to the control group. They were also more likely to have
been on haemodialysis therapy rather than peritoneal dialysis
therapy and were also more likely to re q u i re re-transplantation.
Of the HCV positive patients, 12 received their first graft prior
Irish Journal of Medical Science • Volume 173 • Number 2
We also compared the transplant patient survival rates for those
known to be HCV antibody positive to the control group. The
patient survival rate was similar in both groups with an overall p
value of 0.78. Patient survival was 95.8% in the HCV positive gro u p
versus 93.8% in the HCV negative group at 1 year, 87.1% versus
86% at 3 years, 87% versus 80% at 5 years and 70% in both gro u p s
at 10 years. There have been eight deaths in our known HCV
antibody positive transplant patients during the period studied, with
a mean age of death of 50.3 years (range 19–65). Death occurred
while on dialysis in four of these patients (two of which were on
peritoneal dialysis and two of which were on haemodialysis). The
other four patients died with a functioning graft in situ.
Discussion
HCV infection is more common in renal patients than in the
general population.1,3 Currently in our unit, 2.6% of our
haemodialysis patients and 4.7% of our peritoneal dialysis patients,
a re chronically infected. Our data show that there is significantly
83
Impaired renal allograft, but not patient survival, in patients with antibodies to hepatitis C virus
Table 2. Graft survival rates of HCV+ versus HCV- kidneys at 5 and 10 years post transplant compared to other published studies
Author/year
HCV+ at 5 years
HCV– at 5 years
HCV+ at 10 years
HCV– at 10 years
Giblin et al, 2004
Mateos et al, 1999
Mathurin et al, 1999
39.7%
68.3%
61.7%
84.7%
13.7%
51.0%
49.5%
39.5%
66.5%
69%
Table 3. Patient survival rates of HCV+ versus HCV- patients at 5 and 10 years post transplant compared to other published studies
Author/year
HCV+ at 5 years
HCV– at 5 years
HCV+ at 10 years
HCV– at 10 years
Giblin et al, 2004
Mateos et al9, 1999
Mathurin et al11, 1999
Nagano et al, 1998
80%
87%
87%
96%
70%
71.9%
65.5%
83.7%
70%
90%
85.3%
88.9%
i m p a i red graft survival in patients known to have antibodies to
HCV compared to the general renal transplant population.
However, contrary to other reports, our data do not show
i m p a i red patient survival in HCV antibody positive patients.
There are several reasons why the end-stage renal failure
(ESRF) patient is at increased risk of contracting HCV infection;
the requirement for blood transfusion for renal-related anaemia
or peri-operatively at the time of transplantation, recurrent use of
needles/extracorporeal circuits on haemodialysis and the risk of
infection from the donor organ at time of transplant. Factors that
influence the risk of infection include; the length of time spent on
dialysis therapy, the mode of renal replacement therapy employed
and the number of blood transfusions required. The most
important factor is the type of renal replacement therapy used,
with patients on haemodialysis estimated to have a risk 2-3 times
that of their counterparts on peritoneal dialysis.
Recent developments in the management of renal failure have
significantly decreased the risk of infection. Since the
introduction of recombinant human erythropoeitin in the mid
1980s, the requirement for blood transfusions in renal patients
has markedly decreased. Many renal patients now reach target
haemoglobin without ever requiring blood transfusion. As the
main route of transmission of HCV is parenteral, blood
transfusions were the major source of transmission of HCV
infection for many years. Since the introduction of a screening
test for all blood donors the risk of transfusion related
contraction of HCV has dramatically diminished. It is now
estimated that the risk is less than 1 in 100,000 blood units.6 As
it is now standard practice to routinely screen all potential donors
for viral infections, including HCV, the risk of donor-related
HCV infection has also virtually disappeared. Furthermore the
risk of nosocomial transfer of the virus has decreased in the
haemodialysis setting due to routine screening of all patients
commencing renal replacement therapy. Isolation procedures for
infected patients and the non re-use of dialysis needles and filters
in this country also minimise the risk of transfer of the infection.
Some studies re p o rt increased incidence of death due to liver
failure and/or sepsis.7,8 Mateus et al showed the main cause of death
was cardiovascular with no apparent increase in mortality due to
infections or chronic liver disease.9 The main cause of loss of the renal
transplant was due to chronic allograft nephropathy or death with a
functioning graft. Contrary to these findings Periera et al reported
a higher rate of graft loss in patients who were HCV positive at time
of transplantation (versus those who were HCV negative), an
increased risk of death (particularly from uncontrolled sepsis) and a
higher incidence of post-transplant liver disease.10 In our series death
was predominantly due to cardiovascular complications and the
presence of significant liver disease was infrequent.
84
Whether HCV infection after renal transplantation adversely
affects graft and patient survival remains controversial. Our study
clearly demonstrated impaired kidney allograft survival, but
demonstrated no effect on the survival of the recipients
themselves. Several authors have reported conflicting results
from their own centres. Mathurin et al performed a case control
study, matching 216 HCV-positive patients with 216 HCVnegative controls.11 They demonstrated that 10-year patient and
graft survival rates were significantly lower in the HCV positive
patients compared to the HCV negatively matched controls (see
Tables 2 and 3). Our figures show a 10-year patient survival rate
of about 70% in both patient groups.
In the Irish renal transplant population the presence of
hepatitis C antibodies, before or after transplantation, is
associated with a worse long-term graft, but not patient survival.
References
1.
Naoumov NV. Hepatitis C virus infection in Eastern Europe. J Hepatol
1999; 31(Suppl 1): 65–70.
2. P e reira BJG. Hepatitis C in organ transplantation: Its significance and
influence on transplantation policies. Curr Opin Nephol Hyperten 1993;
2: 912–922.
3. Thomas DL. Hepatitis C epidemiology. Curr Top Microbiol Immunol
2000; 242: 25–41.
4. Morales JM, Campistol JM. Transplantation in the Patient with
Hepatitis C. J Am Soc Neph 2000; 11: 1343–53.
5. Conlon PJ, Walshe JJ, Symth EG, McNamara ES, Donohue J, Carmody
M. Lower prevalence of anti-hepatitis C antibody in dialysis and renal
transplant patients in Ireland. Ir J Med Sci 1993; 162(4): 145–7.
6. New Eng J Med Vol 340:6 Feb1999
7. Rodrigues A, Morgado T, Henriques AC, Sarmento AM, Pereira M,
Guimaracs S. Outcome of renal graft recipients with hepatitis C viru s
infection. Transpl Int 1996; 9 Suppl 1: S23–31.
8. Periera BJG, Wright TL, Schmid CH, Levey AS. The impact of
p retransplantation Hepatitis C virus infection on the outcome of renal
transplantation. Transplantation 1995; 60 (8): 799–805.
9. Gentil MA, Rocha JL, Rodrguez-Algarra G et al. Impaired kidney
transplant survival in patients with antibodies to hepatitis C virus.
N e p h rol Dial Transplant 1999; 14(10): 2455–6.
10. Pereira BJ, Levey AS. Hepatitis C virus in dialysis and re n a l
transplantation. Kidney Int 1997; 51: 981–99.
11. Mathurin P, Mouquet C, Poynard T et al. Impact of hepatitis B and C
v i rus on kidney transplantation outcome. Hepatology 1999; 29(1):
257–63.
Correspondence to: Dr Peter Conlon, consultant nephrologist,
Beaumont Hospital, Beaumont Road, Dublin 9, Ireland.
Email [email protected]
Irish Journal of Medical Science • Volume 173 • Number 2
Risk factors for pulmonary embolism in an Irish patient cohort
Risk factors for pulmonary embolism in an Irish
patient cohort
S Timmons, R Liston, H Kelly
South Munster Geriatric Training Scheme, Tralee General Hospital, Co Kerry, Ireland
Abstract
Introduction The prevention of pulmonary embolism (PE) is an important component of medical care.
Aim To examine the risk factors for venous thromboembolism in an Irish patient cohort with acute PE, and identify
cases that may have been preventable.
Methods Retrospective review of 60 consecutive cases of computed tomography (CT)-confirmed acute PE.
Results The primary thromboembolic risk factors were elective surgery (27%), medical illness (20%), primary
immobility (13%) and isolated distal lower limb fracture (7%). A significant proportion (43%) had been hospitalised
within the six weeks prior to PE onset. Some patients had undergone ‘low risk’ procedures, without prophylaxis, but
had other significant thromboembolic risk factors that indicated a requirement for prophylaxis.
Conclusions Hospital- and ward-based thromboprophylaxis guidelines, based on certain categories of patient or
procedure, need to be routinely supplemented by an individual risk factor assessment for each patient, to determine
those at particularly high risk for venous thromboembolism.
Introduction
Pulmonary embolism (PE) is a serious, potentially fatal disorder
and is the most common preventable cause of hospital death.1
Unfortunately, PE is often the first presentation of venous
t h romboembolism (VTE) and many deaths occur before
investigations or treatment can be instigated.2 Therefore, despite
advances in the diagnosis and treatment of deep venous thrombosis
(DVT) and PE, prevention remains of paramount importance.
Up to 25% of patients with proven PE have no identifiable risk
factors for VTE3 and others have occult risk factors that are only
identified in retrospect. However, many cases occur in patients
with overt pre-existing risk factors and it has been well reported
that hospitalised patients at risk of VTE are under-prescribed
thromboprophylaxis.4-6
In an effort to highlight the importance of thromboembolism
risk factor detection, leading to the initiation of appropriate
thromboprophylaxis, this study reports the risk factors found in
a cohort of Irish patients with proven acute PE. Cases where, in
retrospect, thromboprophylaxis may have been indicated, are
identified and discussed.
Methods
The results of all spiral CT examinations performed in Tralee
General Hospital (TGH) over a three-year period, July
2000–June 2002, were reviewed. A consultant radiologist
reviewed the films of possible or definite PE cases, to confirm the
diagnosis. Sixty consecutive cases of acute PE were identified and
the case notes reviewed. The risk factors for VTE and
thromboprophylactic measures (compression hosiery or low
molecular weight heparin [LMWH]) employed for each case
were recorded. Of note, seven patients had been hospitalised
elsewhere during the relevant thromboembolic risk period, with
details of the thromboprophylaxis received, if any, available for
only two of these patients. All statistical analysis was performed
using the non-parametric Fisher’s exact test.
Results
The 60 cases of acute PE ranged in age from 24 to 88 years.
Irish Journal of Medical Science • Volume 173 • Number 2
Forty-eight per cent were female, mean age 64 years (±16 years),
while males were non-significantly younger, 55 (±17) years
(p<0.06). Twenty-three per cent were in-patients at the time of
PE, while a further 20% had been in-patients within the
preceding 6 weeks. In 15% of cases, no risk factor for VTE was
ultimately identified. Thirteen per cent had two or more risk
factors, while 7% had three or more (see Figure 1). The main risk
factors are summarised in Table 1.
Figure 1. Number of thromboembolism risk factors in each
case of pulmonary embolism.
Table 1. Primary risk factors for venous thromboembolism
in cases of acute pulmonary embolism
Risk factor
Percentage of cases
Surgery
Current medical illness
Immobility*
28%
20%
13%
Lower limb fracture
Previous thromboembolism
Others
(None)
12%
7%
5%
(15%)
*Excludes immobility secondary to another risk factor.
85
S Timmons et al
Immobility
Fifty-three per cent had a preceding period of significant
immobility (more than 48 hours of restricted mobility, plaster
immobilisation of a limb, or a long-distance flight). However,
most immobility occurred in the context of another risk factor,
such as a fracture or surgery, with immobility being the primary
risk factor for PE in only 13% of cases. This included three
people post long-haul flights, one obese long-distance lorry
d r i v e r, three patients with chronic immobility (remote
hemiparesis, spinal stenosis and Parkinson's disease) and one
psychiatric in-patient with profound depression.
Lower limb or pelvis fracture
Twelve per cent of the patients with acute PE (n=7) had
sustained a fracture, 4–50 days prior to symptom onset. In four
cases, the distal lower limb was fractured and managed
c o n s e rvatively with plaster immobilisation, without
thromboprophylaxis. Apart from obesity in one patient, these
cases had no other risk factors for VTE. Two patients had
f r a c t u red pelvic bones, both approximately one month prior to
PE. One patient had not received prophylaxis. A young
woman, one-month postpartum, underwent surg i c a l
stabilisation of a tibial fracture, without thromboprophylaxis.
Orthopaedic surgery
Twelve per cent of patients (n=7) had undergone elective total hip
replacement, two to ten weeks prior to PE. Five of these had no
other risk factors for VTE, while two men were obese with remote
histories of spontaneous DVT. All cases with available postoperative
notes had received low-dose thro m b o p rophylaxis (see Table 2).
Pelvic surgery
Ten per cent of patients (n=6) had undergone pelvic surgery in
the preceding 2 weeks, including transurethral resection of
prostate (TURP) or biopsy, and vaginal hysterectomy. Three
cases had no other risk factors, while two had malignancy
(prostate carcinoma; prostate and laryngeal carcinoma). One
patient underwent uterine coil insertion for menorrhagia and
developed symptoms suggestive of PE within a few hours. The
three patients who underwent transurethral procedures did not
receive thromboprophylaxis.
Thoraco-abdominal surgery
Seven per cent of patients (n=4) had undergone surgery in the
p receding month, namely coro n a ry artery bypass grafting,
l a p a rotomy for intestinal obstruction, cholecystectomy and
pharyngeal pouch repair. One patient was significantly immobile
perioperatively due to Parkinson’s disease but did not receive
thromboprophylaxis.
Medical conditions
Twenty per cent of patients (n=12) had a medical condition as
the primary risk factor for VTE. One patient had been
hospitalised with severe pneumonia one month previously, with
co-existing chronic lung disease and cardiac failure, and had not
received thromboprophylaxis. Three patients, none hospitalised,
had congestive cardiac failure, with two of these having no other
risk factors and one treated for a recent DVT. Another patient
was six weeks post myocardial infarction.
Three patients had hemiparetic cerebral ischaemic events and
one patient had a large intracerebral haemorrhage. Thre e
patients, ambulant and not hospitalised, had carcinoma. Two
had localised breast carcinoma (it is not known if they were
receiving anti-oestrogen therapy, which may increase the risk of
VTE).7 The third patient had an unspecified adenocarcinoma.
Personal or family history of venous
thromboembolism
Five patients had a recent proximal DVT. Three had commenced
anticoagulation 3–25 days previously, with one patient at
therapeutic international normalised ratio (INR), one
subtherapeutic and the third still receiving LMWH at the time
of PE. One case had received three months warfarin treatment
for a spontaneous DVT, discontinued one month prior to PE,
while another patient with dilated cardiomyopathy and a DVT
five months previously was poorly compliant with warfarin
therapy (INR 1.6).
Two further patients had seemingly isolated distal DVT and
did not receive anticoagulation (one had intracere b r a l
haemorrhage). Two patients had DVT diagnosed only after PE
symptom onset. Four patients had a remote history of VTE,
with this being the sole risk factor in three cases. Two young
patients had a family history of VTE but both had other risk
factors also.
Other risk factors
Three patients were taking oral contraceptive medication or
h o rmone replacement therapy. None had other risk factors for
VTE. Five patients were obese, with four having other VTE risk
factors also. Similarly, three patients with varicose veins had other
risk factors. A thrombophilia screen (to detect protein S, protein C
and antithrombin 3 deficiencies, fibrinogen levels, lupus
anticoagulant or anticardiolipin antibodies and activated protein C
resistance) was planned in only ten patients. Results were available
for seven patients, and were normal. This is likely to underre p resent the frequency of clotting abnormalities in this cohort.
Gender and age differences in risk factors
Male patients had more often undergone orthopaedic surgery,
while females had a higher prevalence of primary immobility.
Younger patients (<65 years) more commonly had a lower limb
fracture, while older patients more often had a medical illness
(young:old=2:10, p<0.01). Age and gender did not influence
the total number of risk factors.
Table 2. Details of thromboprophylaxis received by patients with fracture, stroke and postoperative patients
VTE risk factor
Notes available
LMWH
Dose and duration of LMWH
Compression stockings
Lower limb fracture
Orthopaedic surgery
6/7 cases
3/7 cases
0
3
0
3
Pelvic surgery
Thoraco-abdominal surgery
5/6 cases
3/4 cases
0
1
Acute hemiplegia
3/4 cases
1
Not applicable
Enoxaparin 20mg OD;
10–12 days duration
Not applicable
Enoxaparin 20mg OD;
Until PE onset
Enoxaparin 20 mg OD;
Until PE onset
86
0
0
0
Irish Journal of Medical Science • Volume 173 • Number 2
Risk factors for pulmonary embolism in an Irish patient cohort
Discussion
As ventilation/perfusion scanning is not performed at TGH,
spiral CT is the primary means of diagnosing PE. This ensures
that the cases described in this study are typical and unselected
(i.e. they are not a select group of patients who underwent
second-line PE investigations). However, any study of PE,
whether prospective or retrospective, is naturally biased towards
including patients with well-known risk factors, as these patients
are more likely to be suspected of having PE in the first instance.
Similarly, the risk factor profile in any hospital reflects the type of
patients treated there. However, TGH is a typical general
hospital and this patient cohort is likely to be representative of
many Irish hospitals.
Lower limb plaster immobilisation may confer some risk for
VTE, compounded by the inflammatory response to the
fracture. At the time of this study, there was limited literature to
support thromboprophylaxis in-patients with isolated lower limb
fracture or injury that required immobilisation. In one
prospective series of 165 conservatively treated tibial fractures,
2% of patients developed PE.8 Of 102 patients with surgically
repaired distal lower limb fracture, 28% had demonstrated DVT
by venography and 4% had clinical evidence of PE, with one
confirmed radiologically.9 Two small studies of LMWH
p rophylaxis in conservatively treated lower limb injuries
(fractures and soft tissue injuries), demonstrated significant
reductions in DVT incidence.10,11
More recently, a study of 440 patients demonstrated a relative
risk reduction for DVT of approximately 50% with reviparin
(LMWH) prophylaxis for immobilised isolated leg injuries
(from 19% to 9%).12 Two patients developed symptomatic PE in
the placebo group, and none in the treatment group. The most
recent ACCP guidelines do not contain specific
recommendations for minor trauma or lower limb injuries.13
The Scottish Intercollegiate Guidelines Network (SIGN)
recommends LMWH for ‘major’ lower limb fractures (other
than hip fracture), or aspirin 150mg/day if LMWH is
contraindicated.14 Pending updated recommendations, it seems
prudent to assess patients with distal lower limb fractures for
other VTE risk factors, and give LMWH if these are present.
The threshold for thromboprophylaxis should be lower for
older or obese patients and all patients should be mobilised as
soon as possible.
A small study of low-dose heparin during TURP found that it
did not increase bleeding15 but transurethral procedures are
considered low risk for VTE and thromboprophylaxis is not
usually prescribed. Both cases of PE post TURP had localised
prostate carcinoma. One also had documented laryngeal
carcinoma, further increasing the risk for VTE and probably
indicating that thromboprophylaxis was required.16
Similarly, the patient who underwent uterine coil insertion did
not receive thromboprophylaxis. This is considered a low-risk
procedure as pelvic veins are not directly involved and the
duration of anaesthesia is short. However, this patient had a
personal and family history of VTE, significantly increasing the
risk of VTE for even a minor procedure. (The rapidity of PE
onset after the procedure suggests that the procedure may have
dislodged an existing pelvic thrombus, in which case
thromboprophylaxis may not have altered the outcome.)
Of the five patients with cancer and PE, two had undergone
pelvic surgery as described above, while three were ambulant at
home. For ambulant cancer patients without other risk factors or
indwelling central lines, thromboprophylaxis is under evaluation
but is not currently recommended for routine use.13,14
One of the three patients with hemiplegic ischaemic stroke
received low dose LMWH prophylaxis, while one had
Irish Journal of Medical Science • Volume 173 • Number 2
appropriate omission of thromboprophylaxis (extreme disability
from a previous stroke, with little chance of survival). Published
guidelines vary widely in their recommendations for suitable
thromboprophylaxis for ischaemic stroke patients, with the
ACCP strongly recommending LMWH unless there is a specific
contraindication,13 while the Royal College of Physicians
recommends avoiding LMWH.17 The SIGN suggests LMWH be
considered only in stroke patients at particularly high risk of
VTE (e.g. a history of a previous event).14 Compression hosiery
is an accepted alternative to LMWH in cerebral haemorrhage or
massive cerebral infarction,13,17 but the two eligible patients did
not receive these. A further patient, hospitalised with
pneumonia, with chronic interstitial lung fibrosis and cardiac
failure, had not received any thromboprophylaxis. The published
guidelines universally recommend LMWH in such a patient.13,14,18
A final case to note was a psychiatric in-patient with profound
depression, involving psychomotor retardation and consequent
immobility, with no other risk factors for VTE. Such a patient
would not usually receive thromboprophylaxis. However,
profound immobility is a risk factor for VTE. It seems prudent
to routinely determine VTE risk factors in such a patient, and if
present, to consider compression hosiery.
This study highlights the common risk factors for PE in an
Irish patient cohort. It is important to be aware of such risk
factors, as in their presence, even an atypical presentation for PE
may still warrant full investigation. In particular, older patients
may present atypically with PE,19 making the presence of VTE
risk factors an important indicator of the correct diagnosis.
This study demonstrates the importance of assessing the
requirement for thromboembolism prophylaxis in all
hospitalised adults. At the time of hospitalisation of the patients
in this study, there were ward-based guidelines in place in TGH
for thromboprophylaxis in stroke, orthopaedic and surgical
patients, but clearly, these were not always followed. More
importantly, this study demonstrates that ‘blanket’ policies for
thromboprohylaxis for certain patient categories and types of
procedure must always be supplemented by an individual risk
factor determination, to allow modification of the standard
prophylaxis regime for those at higher risk.
References
1.
2.
3.
4.
5.
6.
7.
8.
Anderson FA, Wheeler HB, Goldberg RJ et al. A population-based
perspective of the hospital incidence and case-fatality rates of deep vein
thrombosis and pulmonary embolism. Arch Intern Med 1991; 151:
933.
Dalen J, Alpert J. Natural history of pulmonary embolism. P ro g
Cardiovasc 1975; 17: 259–70.
Green R, Meyer T, Dunn M et al. Pulmonary embolism in younger
adults. Chest 1992; 101: 1507–11.
Gillies T, Ruckley C, Nixon S. Still missing the boat with fatal
pulmonary embolism. Br J Surg 1996; 83: 1394–5.
Stratton M, Anderson F, Bussey H et al. Prevention of venous
thromboembolism: adherence to the 1995 American College of Chest
Physicians consensus guidelines for surgical patients. Arch Int Med
2000; 160: 334–40.
Arnold D, Kahn S, Shrier I. Missed opportunities for prevention of
venous thromboembolism: an evaluation of the use of
thro m b o p rophylaxis guidelines. Chest 2001; 120: 1964–71.
Fisher B, Costantino J, Redmond C et al. A randomised clinical trial
evaluating tamoxifen in the treatment of patients with node-negative
breast cancer who have oestrogen-receptor-positive tumours. N Engl J
Med 1989; 320: 479–84.
Kyro A, Tunturi T, Soukka A. Conservative treatment of tibial
fractures. Results in a series of 163 patients. Ann Chir Gynaecology
1991; 80: 294–300.
87
S Timmons et al
9.
10.
11.
12.
13.
14.
88
Abelseth G, Buckley R, Pineo G, Hull R, Rose M. Incidence of deepvein thrombosis in patients with fractures of the lower extremity distal
to the hip. J Orthop Trauma 1996; 10: 230–5.
Kujath P, Spannagel U, Habscheid W. Incidence and prophylaxis of
deep venous thrombosis in outpatients with injury of the lower limb.
Haemostasis 1993; 23 (suppl 1): 20–6.
Kock H-J, Schmit-Neuerberg KP, Hanke J et al. Thromboprophylaxis
with low-molecular-weight heparin in outpatients with plaster-cast
immobilization of the leg. The Lancet 1995; 346: 459–61.
Lassen M, Borris L, Nakov R. Use of low-molecular-weight heparin
reviparin to prevent deep-vein thrombosis after leg injury requiring
immobilisation. N Eng J Med 2002; 347: 726–30.
Geerts W, Heit J, Clagett G et al. Prevention of venous
thromboembolism (6th ACCP conference). Chest 2001; 119:
S132–75.
Scottish Intercollegiate Guidelines Network. Prophylaxis of venous
thromboembolism. Publication no. 62, Edinburgh, Oct 2002.
http://www.sign.ac.uk/guidelines/fulltext/62/index.html (Updated
Jan 03, accessed Mar 04)
15. Bejjani B, Chen D, Nolan N, Edson M. Minidose heparin in
transurethral prostatectomy. Urology 1983; 22: 251–4.
16. Kakkar A, Williamson R. Prevention of venous thromboembolism in
cancer patients. Semin Thromb Hemost 1999; 25: 239–43.
17. Royal College of Physicians. National Guidelines on stroke.
http://www.rcpondon.ac.uk/pubs/books/stroke/ceeu_stroke_clinic
al08.htm#83. Updated 16 Aug 2002.
(Accessed August 03).
18. Nicolaides A, Breddin H, Fareed J et al. Prevention of venous
thromboembolism: international consensus statement. Int Angiol
2001; 20: 1–37.
19. Timmons S, Kingston M, Hussein M, Kelly H, Liston R. Pulmonary
embolism: diff e rences in presentation between older and younger
patients. Age and Ageing 2003; 32: 601–5.
Correspondence to: Suzanne Timmons, c/o Department of Geriatric
Medicine, Cork University Hospital, Wilton, Cork. Email:
[email protected]
Irish Journal of Medical Science • Volume 173 • Number 2
The cost of managing diabetic foot ulceration in an Irish hospital
The cost of managing diabetic foot ulceration
in an Irish hospital
D Smith, MJ Cullen, JJ Nolan
Department of Endocrinology, St James’s Hospital, Dublin, Ireland
Abstract
Background Little is known about the economic impact of diabetic foot ulceration in the Irish healthcare setting.
Aim Audit of diabetic foot ulcer admissions in St James’s Hospital between April 2001 and March 2002.
Methods Hospital charts were reviewed and costs were calculated on the length of patients’ hospital stay and the cost
of individual investigations performed.
Results Thirty patients were admitted with diabetic foot ulceration as the primary complaint. Amputation was
performed in eight patients, two patients with a non-healing ulcer died. The average duration of each hospital
admission was 20.3±30.7 days. Net in-hospital expenditure was €704,689, an average of €23,489.63 per hospital
admission.
Conclusions The management of diabetic foot ulceration has a significant economic impact on the Irish healthcare
budget. Treatment should therefore be focused on primary prevention through specialised foot clinics and a
multidisciplinary team approach to reduce this economic burden.
Introduction
Diabetic foot ulceration, past or present, affects 7.4% (type 1 and
type 2 combined) of people with diabetes.1 The lifetime risk of
developing a foot ulcer for any diabetic patient is up to 15%.2
The morbidity associated with diabetic foot ulcers is
considerable. People with diabetes are 15–40 times more likely
to undergo a lower extremity amputation than patients with
non-diabetic foot ulceration.3 A prospective study of 314
consecutive patients admitted with diabetic foot ulceration to a
university hospital reported an amputation rate of 25% and over
40 patients died with unhealed ulcers.4
The economic impact of diabetic foot disease is enorm o u s .
Diabetic foot problems are responsible for 47% of all diabetesrelated hospital admissions.5 Reiber published a compre h e n s i v e
summary of the direct costs of diabetic foot disorders in the USA.6
She reported that the treatment of foot ulceration in patients with
type 2 diabetes accounted in 1986 for $150 million. A Swedish
study estimated that the treatment of diabetic gangrene accounted
for 25% of the institutional costs of diabetes care in 1978 (87.9
million out of a total of 351.6 million Swedish kronor).7 Little is
known about the economic impact of managing diabetic foot
ulceration in the Irish healthcare setting. We therefore performed
an audit of diabetic foot ulcer admissions in St James’s hospital
between the 1 April 2001 and the 31 March 2002 to look at
morbidity and mortality associated with foot ulceration, length of
hospital stay and hospital expenditure during the admission.
Methods
Patients were identified from the Hospital In-patient Enquiry
(HIPE) database, diabetes day centre and podiatry records. The
ulcers were classified as either neuropathic (defined as absent
vibration sensation measured by the technique of a tuning fork
applied at the malleoli in association with an ankle-brachial
pressure index >0.8) or ischaemic (vibration sensation intact
with an ankle brachial pressure index of <0.8) or neuroischaemic (absent vibration sensation with an ankle brachial
pressure index of <0.8) or other causes.
Irish Journal of Medical Science • Volume 173 • Number 2
Details of glycaemic control, the treatment for diabetes, coexistent history of hypertension (defined as three consecutive
readings >140/90mmHg), dyslipidaemia (defined as a fasting
total cholesterol >5.0mmol/l), microalbuminuria (defined as an
albumin concentration >30mg in a 24-hour urine collection)
and current smoking history were recorded.
Individual patient charts were reviewed in detail recording the
manner of presentation of the foot ulcer, presence of active
infection and the organism involved, the investigations and
therapies directly applied to the management of the diabetic foot
ulcer. Patients were followed to determine the outcome
measures of healing or non-healing of the ulcer, length of
hospital stay and the estimated cost in euro per patient. Costs for
the length of hospital stay were calculated from the 2002 end of
year hospital budget for St James’s Hospital. Radiological and
operating theatre costs were obtained from the financial
manager of the relevant department. Costs did not include
laboratory tests performed during the hospital admission or the
cost of antibiotic therapy or outpatient care and follow-up of the
diabetic foot ulcer.
Results
There were 30 diabetic foot ulcer admissions over the one-year
period to St James’s Hospital. Within these 30 admissions, three
patients had more than one hospital admission with recurrent
infection of a non-healing foot ulcer. Four patients had type 1
diabetes, the remainder had type 2 diabetes. Three patients were
diagnosed with type 2 diabetes at time of presentation with their
foot ulcer. Mean age of patients was 68.6±12.8 (mean±SD) years
with a male to female ratio of 5:1 and a mean duration of
diabetes of 10.1±12.5 years. Fourteen of the admissions to
hospital were from the diabetes day centre, seven via the vascular
service, six from the diabetes and podiatry outpatient
department and three directly through the hospital’s accident
and emergency department. The characteristics of the group
according to glucose control and diabetes complications are
outlined in Table 1.
89
D Smith et al
Table 1. Patients’ characteristics according to their diabetes treatment
Treatment
HbA1C (mean±SD)
PVD (%)
Diet only (n=3)
Sulphonlyurea only
(n=8)
Metformin only (n= 7)
Metformin+sulphonlyurea
(n=3)
Insulin+oral
hypoglycaemics (n=3)
Insulin only (n=6)
8.3±0.4
8.2±1.4
100
100
8.4±2.8
9.3±1.9
Neuropathy (%)
Smoker (%) BP (%)
ALB (%)
Lipid (%)
67
12
67
25
67
88
33
50
33
75
86
67
57
67
0
67
86
100
0
67
29
67
8.3±1.4
100
33
67
67
33
100
8.8±1.4
67
83
50
67
17
33
HbA1C=glycosylated haemoglobin; PVD=peripheral vascular disease with an ankle brachial index pressure <0.8; BP=blood
pressure >140/90mmHg on three consecutive readings; ALB = >30mg of albumin on a 24 hour urine collection; Lipid=total
cholesterol >5.0mmol/l.
Table 2. Characteristics of the foot ulcer
Type of ulcer
Organism identified
Ischaemic (n=15) Staphylococcal in 3
(MRSA, MSSA x 2)
Streptococcal in 2
GM- Bacilli in 2
Neuropathic (n=2) Staphylococcal in (1),
Streptococcal (2),
GM– Bacillus (1).
Mixed (n=11)
Staphylococcal in (6),
(MRSAx1),
Streptococcal (3),
GM– Bacilli (3),
anaerobe (1).
Others (n=2)
Staphylococcal in (2),
Streptococcal (2),
GM– Bacilli (2),
anaerobe (1).
Choice of antibiotic
Length of antibiotic Rx (days:mean±SD)
Benzylpenicillin/flucloxacillin (7).
Ciprofloxacin/clindamycin (5).
Co-amoxiclavulinic acid (1).
23.0±14.2
Benzylpenicillin/flucloxacillin (1).
Ciprofloxacin/clindamycin in (1).
23.5±6.4
Benzylpenicillin/flucloxacillin (5).
Ciprofloxacin/clindamycin in (2).
Clindamycin/Levofloxacin in (1).
Vancomycin/rifampicin in (1).
Benzylpenicillin/flucloxacillin (1).
Co-amoxiclavulinic acid and
metronidazole (1).
35.6±15.5
15.5±7.8
MRSA=methicillin-resistant Staphylococcus aureus; MSSA=methicillin-sensitive Staphylococcus aureus.
Of the 30 hospital admissions, 15 had an ischaemic ulcer, two
admissions were due to an infected neuropathic ulcer whilst 11
were due to mixed neuro-ischaemic ulcers, one was a venous ulcer
and one was an ulcerated infected ingrown toenail (see Table 2).
A swab of the ulcer was perf o rmed in only 50% of cases. The ulcer
swabs were cultured on three diff e rent media, blood agar in
carbon dioxide, MacConkey agar and neomycin blood agar. The
majority of swabs (73%) revealed polymicrobial infection, the most
commonly identified organisms were Staphylococcus,
S t reptococcus and Gram-negative bacilli. Twenty-six of the
patients received antibiotic treatment. The most commonly
prescribed antibiotic combination was flucloxacillin and
benzylpenicillin or ciprofloxacin and clindamycin. Antibiotic
treatment was changed in 11 patients on the basis of repeat swabs
taken from the ulcer.
One patient with a neuropathic foot ulcer was referred to
the orthopaedic service for contact casting to relieve pressure
on the affected foot. However in this patient previous skin
grafting for a burn injury on the affected foot prohibited the
use of a contact cast.
The results of laboratory and radiological investigations are
outlined in Table 3. An elevated white cell count was present in
90
only 20% of patients. Osteomyelitis complicated foot ulceration in
7% of cases and no patient presented with an acute Charc o t ’s foot.
Nearly all of the foot ulcer patients had a formal vascular
assessment on admission to hospital (see Table 4). The majority
of patients (97%) had an ankle brachial pressure index
performed. Patients were referred to the vascular surgical service
for an opinion if they had an ankle brachial index of <0.8 and/or
clinical evidence of arterial insufficiency. At the
recommendations of the vascular service, 17 patients had a
femoral angiogram with a femoral angioplasty attempted in 10
patients. One patient had a stent inserted into the femoral artery.
Angioplasty with or without stent insertion was successful in
only 45% of cases. The remaining cases underwent femoralpopliteal bypass surgery or amputation. Amputation was
performed in eight patients (bilateral amputation performed in
one patient); five of the eight had an above knee amputation, the
remaining three had an amputation restricted to the affected
foot. Amputation was performed after the patients had been in
hospital for an average of 19.0±8.7 days. Full ulcer healing
occurred in 43% of foot ulcers, the length of time for this to
occur was 76.5±74.2 days. Unfortunately at time of completion
of the audit two patients with a non healing ulcer had died. One
Irish Journal of Medical Science • Volume 173 • Number 2
The cost of managing diabetic foot ulceration in an Irish hospital
Table 3. Investigations
Type of ulcer
WCC (3.5–11.0x109/l)
Ischaemic (n=15)
Neuropathic (n=2)
Mixed (n=11)
Others (n=2)
9.1±1.8
8.3±2.3
8.4±3.9
5.9±0.5
ESR (0–10mm/hr)
C-RP (0-4mg/l)
Foot X-ray (%) Bone scan (%) MRI foot (%)
71.9±36.3
70.0±14.1
97.9±18.9
52.5±33.2
37.3±30.5
48.2±61.9
99.8±61.2
18.9±0
53
100
64
100
20
100
27
50
27
50
18
0
WCC=white cell count; ESR=erythrocyte sedimentation rate; C-RP=C-reactive protein. Normal reference ranges are given
in brackets. Data is presented as mean±SD or percentage (%).
Table 4. Vascular investigations and treatment
Type of ulcer
ABI* (mean±SD)
Angiogram (%)
Angioplasty (%) Stent (%)
Bypass (%)
Amputation (%)
Ischaemic (n=15)
Neuropathic (n=2)
Mixed (n=11)
Others (n=2)
0.34±0.29
0.92±0.21
0.34±0.25
0.81±0.38
73
0
55
0
33
0
37
0
13
0
18
0
47
0
27
0
7
0
0
0
*ABI=ankle brachial pressure index of the affected leg.
Table 5. Outcome
Type of ulcer
Healed (%) Non-healed (%) Died (%) Length of hospital stay (days: mean±SD) Estimated cost* (euro)
Ischaemic (n=15)
Neuropathic (n=2)
Mixed (n=11)
Others (n=2)
27
50
27
100
67
50
64
0
7
0
9
0
24.8±18.6
24.5±5.0
42.2±21.6
18.0±8.5
296,902
34,990
347,413
25,384
*Estimated cost does not include laboratory tests or antibiotic treatment.
patient died in hospital from a myocardial infarction, the cause
of death in the second patient is not known. The average
duration of each hospital admission was 20.3±30.7 days (see
Table 5); one patient spent 220 days in hospital over three
separate admissions with problems directly related to his diabetic
foot ulceration. Net in hospital expenditure was Ä704,689, an
average of Ä23,489.63 per hospital admission.
Discussion
This audit estimates the economic impact of diabetic foot
ulceration in the Irish hospital setting. Over a one-year period
between April 2001 and March 2002 there were 30 admissions
to St James’s Hospital with diabetic foot ulceration as the
primary problem. This year was not exceptional with a similar
number of foot-related admissions the previous year. We
calculated the net cost of managing diabetic foot disease in one
year in these patients to be over Ä700,000.
This calculation is an underestimate of the impact of diabetic
foot disease on the hospital budget. We did not include the cost
of routine blood testing or of antibiotic treatment during the
hospital stay. Since diabetic foot ulcer patients are often very
sick, are either pre or post surgery and have numerous coexisting
medical conditions then it is reasonable to assume that blood
tests were taken on a daily or alternate day basis. Other studies
have shown the cost of antibiotic treatment to be between 5%
and 11% of total direct hospital costs in the management of
diabetic foot disease.8,9 Therefore at least an extra Ä100,000
could be added to the net hospital cost when blood testing and
antibiotic therapy are included. While this audit focused on the
economic burden of diabetic foot ulceration in the hospital
setting, it is important to remember the community healthcare
Irish Journal of Medical Science • Volume 173 • Number 2
costs and the psychological impact of foot ulceration on our
patients with diabetes, in particular those patients who
unfortunately underwent an amputation. Patients with non
healing ulcers were reviewed regularly by public health nurses,
community and hospital podiatry services and attended the
hospital outpatient department up to six times a year following
discharge. Unfortunately there are no published data on the
overall cost of managing a foot ulcer on an outpatient basis in
Ireland but in Europe outpatient dressings and nursing time
contribute most to the cost of care for foot ulcer patients.9,10
The audit also emphasises the significant morbidity and
mortality associated with diabetic foot ulceration. In general
diabetic patients with foot ulceration have poor glycaemic
control and a high incidence of coexistent vascular risk
factors.11,12 They are unable to mount a systemic white cell
response despite the presence of active infection,13 the ulcers are
infected with polymicrobes and require prolonged courses of
combination antibiotic therapy but unfortunately despite
extensive use of resources the amputation rate remains high,
close to 27% in our audit.14
Therefore the treatment of diabetic foot ulceration should
focus on primary and secondary prevention. The presence of one
foot ulcer is strongly predictive of new ulceration.15 Similarly
primary prevention is crucial, education, regular surveillance, a
specialised diabetes foot clinic, identification of the high risk foot
with appropriate targeted care should in theory reduce the
incidence of diabetic foot ulceration, amputation rates and be
cost effective.16 An intensive diabetes foot programme involving
patient foot education, regular podiatr y and appropriate
footwear, has recently been shown to significantly reduce the
hospital admission rate for diabetes foot ulcers, the number of
91
D Smith et al
investigations performed, the length of hospital stay, and overall
saved an American hospital approximately US$5,000 per person
when compared to a standard foot programme.17
In conclusion, foot ulceration and amputation are known and
feared by almost every person with diabetes and have a huge
economic impact on our healthcare services. Yet these are
potentially the most preventable of all diabetic complications by
the simplest techniques of education and care. Successful
management of diabetic foot ulceration requires a
multidisciplinary approach within the hospital with specialised
foot care teams and close collaboration between primary care
and the hospital service.
References
1.
2.
3.
4.
5.
6.
7.
92
Walters DP, Gatling W, Mullee MA, Hill RD. The distribution and
severity of diabetic foot disease: a community based study with
comparison to a non-diabetic group. Diabetic Medicine 1992; 9:
354–8.
Reiber GE, Lipsky BA, Gibbons GW. The burden of diabetic foot
ulcers. Am J Surg 1998; 176: 5S–10S.
J e ffcoate WJ, Harding KG. Diabetic foot ulcers. The Lancet 2003; 361:
1545–51.
Apelqvist J, Agardh CD. The association between clinical risk factors
and outcome of diabetic foot ulcers. Diabetes Res Clin Prac 1992; 18:
43–45.
Lithner FG. The diabetic foot: epidemiology and economic impact.
IDF Bulletin 1992; 38: 7–9.
Reiber GE. Diabetic foot care. Financial implications and practice
guidelines. Diabetes Care 1992; 15 Suppl (1): 29–31.
Jonsson B. Diabetes: the cost of illness and the cost of control. An
estimate for Sweden 1978. Acta Med Scand 1983; 671: 19–27.
VanAcker K, Oleen-Buckley M, DeDecker L et al. Cost and resource
utilization for prevention and treatment of foot lesions in a diabetic foot
clinic in Belgium. Diabetes Res Clin Pract 2000; 50: 87–95.
9. Tennvall GR, Apelqvist J, Eneroth M. Costs of deep foot infections in
patients with diabetes mellitus. Pharmacoeconomics 2000; 18: 225–38.
10. H a rding K, Cutting K, Price P. The cost-effectiveness of wound
management protocols of care. Brit J Nurs 2000; 19: S6–S24.
11. Reiber GE, Vileikyte L, Boyko EJ et al. Causal pathways for incident
lower-extremity ulcers in patients with diabetes from two settings.
Diabetes Care 1999; 22: 157–62.
12. Macfarlane RM, Jeffcoate WJ. Factors contributing to the presentation
of diabetic foot ulcers. Diabetic Medicine 1997; 14: 867–70.
13. Delamaire M, Maugendre D, Moreno M et al. Impaired leucocyte
function in diabetic patients. Diabetic Medicine 1997; 14: 29–34.
14. Adler EI, Boyko EJ, Ahroni JH et al. Lower-extremity amputation in
diabetes: the independent effects of peripheral vascular disease, sensory
n e u ropathy and foot ulcers. Diabetes Care 1999; 22: 1029–35.
15. Apelqvist J, Larsson J, Agardh C-D. Long term prognosis for diabetic
patients with foot ulcers. J Intern Med 1993; 233: 485–91.
16. Ragnarsson T, Tennvall G, Apelqvist J. Prevention of diabetes-related
foot ulcers and amputations: a cost utility analysis based on Markov
model simulations. Diabetologia 2001; 44: 2077–81.
17. Horswell RL, Birke JA, Patout CA Jr. A staged management diabetes
foot program versus standard care: a 1-year cost and utilization
comparison in a state public hospital system. Arch Phys Med Rehab
2003; 84: 1743–46.
8.
Correspondence to: Professor John Nolan, Department of
Endocrinology, St James’s Hospital, James’s Street, Dublin 8.
Email:[email protected]
Irish Journal of Medical Science • Volume 173 • Number 2
Emergency department post-coital contraception in Northern Ireland
Emergency department post-coital
contraception in Northern Ireland
S Mawhinney, O Dornan, R Ashe
Antrim Area Hospital, Northern Ireland
Abstract
Background The granting of a licence to Levonelle as an emergency hormonal contraceptive in the Republic of
Ireland may require accident and emergency (A&E) departments to formally provide such a service. This article
outlines the experiences of a Northern Ireland A&E unit.
Aims To examine the pattern of attendance of patients requesting emergency contraception at an A&E department
and to assess if adequate standards of care are achieved.
Method Retrospective case note review of 100 patients attending the A&E department requesting emergency
contraception.
Results Sixty-one per cent of requests for emergency contraception were outside normal pharmacy opening hours.
Seventy-seven per cent of these patients were less than 26 years old. Most (63%) attended within 24 hours of
unprotected sexual intercourse. Forty-three per cent of the patients studied had used no contraception prior to this
request. Recording of menstrual details and sexual behaviour as part of the consultation was variable.
Conclusions A&E departments receive requests for emergency hormonal contraception particularly from younger
women (<25 years). A&E staff must have appropriate training and support to manage these consultations effectively.
Introduction
Method
Unintended pregnancy remains common throughout the British
Isles. Emergency contraception provides women with a safe
means of preventing pregnancy following unprotected sexual
i n t e rcourse or potential contraceptive failure. 1,2 After the
introduction of levonorgestrel 0.75mg, recent UK governmentdriven legislative changes have allowed this emergency hormonal
contraceptive to be sold in pharmacies in Northern Ireland
without a prescription.
The Irish Medicines Board3 in the Republic of Ireland issued a
licence for Levonelle as an emergency contraceptive on 2 May
2003, specifying that the drug be available as a prescription-only
medicine prescribed by a registered medical practitioner. The
expectation is that wider access to such contraception may help
to reduce the high rate of unintended pregnancies among all age
groups.4
Although patients may attend their GP for emergency
contraception, other service providers have a role to play to
ensure that this is easily accessible to all.5 The A&E department
is perceived as a source of healthcare and advice 24 hours per day.
The A&E department in this paper did not choose this role, but
responded to patient demand. Younger patients in particular find
A&E departments convenient, accessible and feel more assured
of confidentiality.6 A&E departments in the UK report that while
96% of responding units received requests for emergency
contraceptives, only 57% provided treatment.7 Management of
women who present to A&E departments for emergency
contraception is in disarray.8 It is important that the patients who
present for emergency contraceptive advice and treatment are
provided with appropriate information, care and follow-up.
This study examines the pattern of attendance for emergency
contraception at an area hospital A&E department in Northern
Ireland and the standard of care provided by A&E staff.
The setting for the study was an A&E department in an area
hospital close to Belfast which serves a semi-urban population of
around 300,000. The department is staffed by A&E senior house
officers, staff grades, consultants and nurse practitioners, each of
whom regularly dispense emergency hormonal contraception in
response to patient demand. The nurse practitioners work under
the guidance of agreed protocols for each of the conditions that
they treat including the provision of emergency contraception.
While the medical staff participate in an extensive postgraduate
education programme, this has not included specific training on
emergency hormonal contraception as the latter has not been
regarded as part of the core A&E service.
The study took place between 1 January 2000 and 4
September 2000. At this time, the Yuzpe method of hormonal
emergency contraception was being used within the department.9
This consists of 100mcg of ethinyloestradiol and 500mcg of
levonorgestrel repeated twice, 12 hours apart, after the first dose
given within 72 hours of unprotected intercourse. Levonelle
(0.75mg levonorgestrel) has replaced the Yuzpe method of
emergency hormonal contraception. It is more effective,
although efficacy decreases markedly with time (95% of
pregnancies prevented if given within the first 24 hours, falling to
58% by 72 hours) and is better tolerated by patients with less
nausea and vomiting. However, the standards of care for
provision of emergency hormonal contraception remain the same
and practitioners need to be aware of these.
Clinical records for 100 patients who attended requesting
emergency contraception were analysed retrospectively against
standards drawn from the April 2000 guidance note on
emergency contraception produced by the Faculty of Family
Planning and Reproductive Healthcare.10 The following
information was also recorded for each case: the patient’s age, the
day and time of the request and whether a nurse practitioner or
Irish Journal of Medical Science • Volume 173 • Number 2
93
S Mawhinney et al
Table 1. Standards for hormonal emergency contraception
consultation
1.
Date and character of LMP and usual cycle length
(to calculate most likely date of ovulation).
2. Time elapsed since unprotected sexual intercourse.
3. Other episodes of unprotected sexual intercourse
this cycle.
4. Reason for request (contraceptive failure or no
contraceptive used).
5. Previous use of emergency contraception.
6. Past medical history checked to outline possible
contraindications.
7. Drug history checked.
8. Potential side effects discussed and documented.
9. Future contraceptive advice given.
10. Follow up advice and appointment suggested with
patient’s family doctor or local family planning clinic
(details of clinic times supplied).
peak time for patients to attend A&E was between 12.00 and
17.59 hours when 57% attended. Twenty-five per cent attended
between 18.00 and 23.59 hours, 5% from 00.00 to 05.59 hours
and 13% from 06.00 to 11.59 hours. Sixty-one per cent of all
requests were outside local pharmacy opening hours.
Sixty-one per cent of patients requesting emergency
contraception were treated by nurse practitioners and 39% by
doctors. Ninety-four per cent were given emergency
contraception at the A&E department and 3% were referred on
for treatment and/or advice to either the Brook Advisory Clinic
or an intrauterine contraceptive device (IUCD)-fitting Family
Planning Clinic. In 3% of cases, emergency contraception was
not thought necessary and these patients were counselled
appropriately.
Standards of care
Date and character of last menstrual period and usual cycle
length
The date of the last menstrual period (LMP) was recorded in
81% of consultations. The character of the LMP was only
recorded in 52% of consultations and the usual cycle length in
8% of patients.
Time elapsed since unprotected sexual intercourse
Sixty-three per cent of patients requested emergency
contraception within 24 hours of unprotected sexual
intercourse, 24% within 25–48 hours and 10% within 49–72
hours. One patient attended after 72 hours and was referred on
for possible IUCD insertion. In two cases, the time elapsed since
u n p rotected intercourse was not recorded but emergency
contraception was given.
Reason for request
Figure 1. Age range of patients requesting emergency
contraception (n=100).
Fifty-five per cent of patients gave contraceptive failure as the
reason for needing emergency contraception; mostly condom
accidents while 43% documented no contraceptive use. In 2% of
cases, the reason was not recorded.
Other episodes of unprotected sexual intercourse this cycle
In only 13% of cases was there a record of whether the patient
had been specifically asked about any previous episodes of
unprotected sexual intercourse this cycle.
Previous use of emergency contraception
Previous emergency contraceptive use was only asked about in
46% of cases and of these, 65% said that they had used it before.
Past medical history
Figure 2. Day of attendance at A&E (n=100).
Past medical history and potential contraindications were
checked in 87% of cases. Drug history was checked in 78% of
cases, and possible side-effects of treatment were discussed and
documented in 85% of cases.
a doctor treated the patient.
Ten standards were chosen as essential components of a
consultation resulting in provision of hormonal emergency
contraception (see Table 1).
Future contraceptive advice and follow-up arrangements
Results
Discussion
Pattern of attendance
Emergency contraception can be obtained from a variety of
services in Northern Ireland: family planning clinics, GPs, A&E
departments, genitourinary medicine departments, gynaecology
clinics and pharmacies. There are differences in specific
knowledge relating to emergency contraception and its
provision between these specialities.11
Seventy-seven per cent of patients who requested emergency
contraception were aged 25 years or younger, of which 32% were
teenagers. The range of ages is shown in Figure 1. Seventy-two
per cent of requests were received on Saturday, Sunday or
Monday with Sunday being the busiest day (see Figure 2). The
94
Fifty-eight per cent of patients were given advice regarding
future contraception and follow-up arrangements were
documented for 66% of cases.
Irish Journal of Medical Science • Volume 173 • Number 2
Emergency department post-coital contraception in Northern Ireland
Most patients seemed to be aware of the higher efficacy of
post-coital contraception given within the first 24 hours and
were motivated to attend early. They see this as an emergency
situation and feel justified in attending A&E, particularly at
times when other outlets are not available. Widespread Sunday
closure is still the norm in Northern Ireland and only a very
small number of pharmacies, and no family planning clinics, are
open on Sunday afternoons. Even though many clinicians
responsible for A&E services may not consider them
appropriate, it seems inevitable that their departments will
continue to receive requests for emergency contraception. A&E
clinicians can make an important contribution to the efforts
aimed at reducing unintended pregnancy in younger women by
providing a responsive, effective service for these patients.
Predominantly younger patients (<26 years) used the A&E
department to obtain emergency contraception and the majority
of requests were made outside local pharmacy opening hours.
The cost of over-the-counter emergency hormonal
contraception is Ä17.30 with prescription in Ireland and
stg£19.99 without prescription in the UK. This may act as a
deterrent, particularly for this younger age group. All A&E
treatment in Northern Ireland is free to the patient. This may
mean that the pattern of attendance described will not be
replicated in the Republic of Ireland. Although A&E is an
expensive source of emergency contraception it is less costly to
the State than an unwanted pregnancy.
Both the doctors and nurse practitioners working in the A&E
department provided effective access to hormonal emergency
contraception with 97% of patients receiving immediate care and
3% referred appropriately elsewhere. Nonetheless, a significant
proportion of patients did not receive the standard of care
recommended in recent guidance, particularly in respect of the
assessment of risk of existing pregnancy, and of prevention of
unintended pregnancy in the future.
Forty-three per cent of the patients in this study had used no
contraception prior to this request for emergency contraception.
Further education and advice must be given to such patients, not
only to reduce the risk of unintended pregnancy, but also to
minimise the potential risks and sequelae of sexually transmitted
infection and other sexual health issues. Advice about
contraception, sexual health and follow-up were documented in
just over half of patients. The provision of emergency
contraception is a complex issue that demands an understanding
of the menstrual cycle, the potential for conception and
knowledge of the pharmacology of the treatment.
It follows that the delivery of this service requires a
confidential setting with well-informed staff who have sufficient
time to address all these areas. Allowing over-the-counter sales in
pharmacies widens availability of emergency contraception, but
obtaining a sexual history to assess the risk of sexually
transmitted infection is not appropriate in this setting. In
addition this may not appeal to the under 25s who are at most
risk of unwanted pregnancy but have the least money. We have
shown that although the overall number of patients requesting
this service has decreased, the actual number of teenagers
increased since Levonelle became available without
prescription.12
A&E clinicians can expect that requests for emergency
Irish Journal of Medical Science • Volume 173 • Number 2
contraception will continue. Many feel that this is not an
appropriate use of A&E resources, but younger patients
continue to choose this option, particularly at weekends, when
demand is high and there is a gap in family planning services. It
is desirable that patients should receive effective treatment at
A&E as part of the effort to reduce the high rate of unintended
pregnancy amongst young women.
It may be more effective to resource such departments with
emergency access to nurses and midwives who have training in
all contraceptive methods to ensure that proper follow-up
arrangements can be made. For those departments responding
to this challenge, there is a clear need for all staff involved to be
appropriately trained and have access to accurate, regularly
updated protocols in order to ensure a high standard of care.
As a result of this study, the A&E department studied is to
produce new guidelines for all staff in the department, based on
current standards for best practice. Such standards should
become the minimum accepted code of practice, wherever
patients seek emergency contraception.
References
1.
Ho PC, Kwan MSW. A prospective randomized comparison of
levonorg e s t rel with the Yuzpe regimen in post-coital contraception.
Hum Repro d 1993; 8: 389–92.
2. WHO Task Force on Postovulatory Methods of Fertility Regulation.
Randomised controlled trial of levonorg e s t rel versus the Yuzpe regimen
of combined oral contraceptives for emergency contraception. The
Lancet 1998; 352: 428–33.
3. IMB confirms licence for Levonelle http://www.imb.ie/news
(20.5.2003)
4. D e p a rtment of Health. The National Strategy for Sexual Health and
HIV. London: Department of Health, 2001. www.doh.gov.uk/
nshs/index.htm
5. Harrison-Woolrych M, Duncan A, Howe J, Smith C. Improving access
to emergency contraception. Br Med J 2001; 322: 186–7.
6. Heard-Dimyan J. Issue of emergency hormonal contraception through
a casualty department in a community hospital. Br J Family Planning
1999; 25: 105–9.
7. Gbolade BA, Elstein M, Yates D. UK accident & emergency
d e p a rtments and emergency contraception: what do they think and do?
J Accid Emerg Med 1999; 16: 35–8.
8. Nathan B, Evans G, McKeever J. Practice in prescribing emergency
contraceptives in A and E departments varies. Br Med J 1998; 316:
149.
9. Yuzpe AA, Lancee WJ. Ethinylestradiol and dl-norgestrel as a postcoital
contraceptive. F e rtil Steril 1977; 28 (9): 932-6.
10. Emergency Contraception: Recommendations for clinical practice.
Faculty of Family Planning and Reproductive Health Care. Guidance,
April 2000.
11. Beckman LJ, Harvey SM, Sherman CA et al. Changes in providers’
views and practices about emergency contraception with education.
Obstet Gynecol 2001; 97: 942–6.
12. Mawhinney S, Dornan O. Requests for Emergency Contraception at
an Accident and Emergency Department – Assessing the impact of a
change in legislation. Ulster Med J (In Press).
Correspondence to: Dr Sandra Mawhinney, Antrim Area Hospital,
Antrim, Northern Ireland. Email: [email protected]
95
Detection of mycobacterial DNA from sputum of patients with cystic fibrosis
Detection of mycobacterial DNA from sputum
of patients with cystic fibrosis
M Devine1, JE Moore1, J Xu1, BC Millar1, K Dunbar1, T Stanley1, PG Murphy1, AOB Redmond2, JS Elborn3
Northern Ireland Public Health Laboratory1, Department of Bacteriology and Northern Ireland Regional Adult Cystic
Fibrosis Centre3, Belfast City Hospital, Northern Ireland Regional Paediatric Cystic Fibrosis Centre2, The Royal
Belfast Hospital for Sick Children, Northern Ireland
Abstract
Background Patients with cystic fibrosis (CF) are at high risk from atypical mycobacterial infections. There have been
few attempts to delineate the intensity of mycobacterial infection in CF patients in Ireland.
Aims To examine the incidence of mycobacterial DNA in an archived collection of genomic DNA extracted from the
sputa of CF patients within the Northern Ireland population.
Methods One hundred and eighty-two CF patients (66 adults and 116 children) were examined for the presence of
mycobacterial DNA in their sputum by a genus specific PCR assay based on 16S rRNA, followed by direct automated
sequencing of the PCR amplicons.
Results One of 116 (0.9%) children and 2 of 66 adults were positive. Sequence identity revealed Mycobacterium
xenopi in the paediatric patient and M. xenopi and M. chelonei in the two adult patients. False-positive results occurred
in 11 patients (four adults), mainly due to Corynebacterium spp.
Conclusions There was a low prevalence of Mycobacterium spp in the CF patient population. All PCR positive results
should be confirmed by direct automated sequencing and an alternative specific assay employed. Enhanced molecular
screening will contribute in understanding their role as opportunistic pathogens in patients with worsening lung function.
Introduction
Patients with cystic fibrosis (CF) suffer a high degree of
morbidity due to chronic infections of the lower respiratory
tract.1 The majority of these infections are due to Staphylococcus
aureus, as well as Pseudomonas aeruginosa, Burkholderia cepacia
and Haemophilus influenzae. Atypical mycobacteria may infect
patients with underlying and chronic lung disease, including
bronchiectasis, pneumoconiosis or healed tuberculosis.2 Often, it
is difficult to differentiate CF patients with active mycobacterial
infection from those with common bacterial infections.
Mycobacterial infections in CF patients are relatively uncommon
and sputum specimens for specific mycobacterial analysis are not
routinely requested by respiratory physicians. In addition,
routine laboratory surveillance of CF sputa for respiratory
pathogens does not normally include mycobacterial culture, due
to safety considerations and cost. Hence, it is difficult to estimate
the prevalence of Mycobacterium spp in CF sputa.
The aim of this study was to examine retrospectively the
incidence of mycobacterial DNA in an archived collection of
genomic DNA extracted from the sputa of CF patients.
Patients and methods
Fresh sputum was obtained from 182 CF patients (66 adults and
116 children) following physiotherapy in an inpatient hospital
setting at the Regional Adult and Paediatric CF centres in
Northern Ireland. Sputa were transported from the ward to the
laboratory at ambient temperature and were analysed within
four hours following expectoration.
All DNA isolation procedures of bacterial genomic DNA from
sputa were carried out in a Class II Biological Safety Cabinet in
a room physically separate from that used to set up reaction
mixes and also from the ‘post-PCR’ room in order to minimise
96
the production of false-positive results. In all specimens tested
and where applicable, molecular grade water was employed
(Biowhittaker Inc., Maryland, USA, LAL Grade Cat No. W50100) to reduce contamination.
Sputa (1g) were initially mixed with Sputasol (Oxoid Ltd.,
Dorset, UK) in a ratio of 1:1 [w/w] and incubated at 37oC for
30 minutes. Following this, 1ml was removed and centrifuged
(13,000xg; 15 minutes). Bacterial genomic DNA from the pellet
was prepared employing the Roche High Purity PCR DNA
extraction kit (Roche Diagnostics Ltd., UK), in accordance with
the manufacturer’s instructions. All reaction mixes were set up in
a PCR hood in a room separate from that used to extract DNA
and the amplification and post-PCR room in order to minimise
contamination.
Reaction mixes (50µl) were set up as follows: 10mM TrisHCl, pH 8.3, 50mM KCl, 2.5mM MgCl2, 200µM (each) dATP,
dCTP, dGTP and dTTP; 1.25U of Taq DNA polymerase
(Amplitaq; Perkin Elmer Cetus Inc., CA, USA). PCR primers
(0.2µM each), MYCGEN-F, 5'-AGA GTT TGA TCC TGG
CTC AG-3' and MYCGEN -R, 5'-TGC ACA CAG GCC ACA
AGG GA-3'), were employed in a PCR assay with the following
cycling conditions: 96oC for 3 minutes, followed by 40 cycles of
96oC for 1 minute, 65oC for 1 minute and 72oC for 1 minute,
following by a final extension step of 72oC for 10 minutes, which
yielded an 1,027bp amplicon, as previously described.3
During each run, molecular grade water was included
randomly as negative controls and appropriate DNA template
isolated from a heat-killed, wild-type isolate of M. tuberculosis
was included as a positive control. In addition, internal DNA
extraction and amplification controls were employed, whereby
the beta-globin gene was amplified from human sputum, as
previously described.4 Following amplification, aliquots (15µl)
Irish Journal of Medical Science • Volume 173 • Number 2
M Devine et al
were removed from each reaction mixture and examined by
electrophoresis (80V, 45min) in gels composed of 2% (w/v)
agarose (Gibco, UK) in TAE buffer (40mM Tris, 20mM acetic
acid, 1mM EDTA, pH 8.3), stained with ethidium bromide
(5µg/100ml). Gels were visualised under UV illumination using
a gel image analysis system (UVP Products, UK) and all images
archived as digital graphic files (*bmp).
Extracted sputa giving a PCR amplicon of the expected size
(1,027bp) were further characterised by direct automated
sequencing, employing the primer, BSR1 primer, 5'-GGA TTA
GAT ACC CTG GTA GTC -3', as shown (see Figure 1).
Amplicons chosen for sequencing were purified using a
QIAquick PCR purification kit (Qiagen Ltd., UK) eluted in
Tris-HCl (10mM, pH 8.5) prior to sequencing, particularly to
remove dNTPS, polymerases, salts and primers. BSR1 was used
for sequencing in the forward direction with the ABI PRISM™
Dye Terminator Cycle Sequencing Reaction with AmpliTaq
DNA Polymerase®, FS (PE Biosystems, Foster City, CA, USA)
(96oC 1minute, followed by 25 cycles of 96oC for 10s, 50oC for
5s, 60oC for 4 minutes, followed by a 4oC hold).
The products were ethanol-precipitated and analysed on an
ABI 377 Automatic Sequencer (PE Biosystems, Foster City, CA,
USA). The resulting sequences obtained were compared with
those stored in the Genbank Data system using BLAST
alignment software (www.blast.genome.ad.jp/) and sequence
homology identity determined in accordance with the criteria, as
previously described by Goldenberger et al.5
Figure 1. Map of positions of diagnostic and sequencing
primers employed in this assay.
Results
Initially, a PCR positive result was recorded for 10/116 (8.6%)
children and 4/66 (6.1%) adults, however on subsequent
sequencing of the PCR amplicon, this was reduced to 1/116
(0.9%) children and 2/66 (3.0%) adults, due to the presence of
false-positives. Sequence identity revealed M. xenopi in the
paediatric patient and M. xenopi and M. chelonei in the two adult
patients. Further details of these patients, their CF mutation and
the other concurrent flora in their sputa, are shown in Table 1.
The false-positive amplicons from the remaining 11 patients
(four adults and seven children) were subsequently identified as
mainly Corynebacterium spp, including C. flavescens (n=5), C.
xerosis (n=4), as well as Rhodococcus coprophilus (n=1) and
Streptococcus ambifaciens (n=1).
Discussion
To date, there have been relatively few studies using a molecular
screening technique to detect mycobacteria in sputum from CF
patients. Employment of a molecular technique to detect
Mycobacterium spp. directly from the sputum of CF patients is
important, especially for patients undergoing lung
transplantation, as many of the clinically significant mycobacteria
i.e. M. tuberculosis, M. intracellulare, M. avium, are slow
g rowing organisms, which can compromise the ability of
diagnostic laboratories to rapidly detect these organisms
employing conventional culture techniques. Consequently
several PCR techniques have been developed for the direct
detection of mycobacteria from sputum.
In this study, a PCR assay based on 16S rRNA detection was
employed, targeting the genus Mycobacterium and hence all
species within this genus. Analysis of the primer sequences,
MYCGEN-F and MYCGEN-R, showed the forward primer to
be universal, i.e. relatively well-conserved within the eubacteria,
and hence has the ability to anneal to most bacteria. However,
specificity was conferred by the reverse primer, MYCGEN-R,
which was specific for the mycobacteria. On application of this
primer set to DNA extracted from patients’ sputa, mycobacteria
were successfully amplified in addition to several other nonmycobacterial species. In addition to the possibility of nonspecific amplification of other genera including mainly
Corynebacterium spp., Streptococcus ambifaciens and Rhodococcus
coprophilus, as demonstrated from this study, additional in silico
analysis demonstrated that in addition to the amplification of the
mycobacteria, Pseudonocardia spp may also be amplified.
Therefore all PCR positive specimens should be confirmed by
d i rect automated sequencing to reduce the rate of falsepositivity.
Our data are in general agreement with other data on the
prevalence of mycobacteria in CF patients, in that mycobacteria
are not commonly associated with relevant clinical disease in CF.
For recent seminal reviews on non-tuberculous mycobacteria in
CF, see Griffiths,6 as well as Ebert and Olivier.7 In a major recent
study from the US, Olivier et al8 described the prevalence of
mycobacteria at 21 CF centres and quoted a value of 13% of
patients culturing nontuberculous mycobacteria from their
sputum, with M. avium complex (72%) and M. abscessus (16%)
representing the most common nontuberculous mycobacteria.
Mycobacterial culture-positive patients were more frequently
older and had a higher frequency of S. aureus and a lower
frequency of P. aeruginosa.
A substudy that followed 60 non-tuberculous mycobacteriapositive patients for 15 months and compared them with an
uninfected control group identified no difference in the rate of
decline of FEV1.9 In a previous study, Oliver et al10 demonstrated
Table 1. Details of patients with CF identified as PCR-positive for non-tuberculous mycobacteria
Patient no.
Age (years)
Sex
CF genotype
Other organisms chronically present in sputum
1
12
M
ND/ND
Pseudomonas aeruginosa
2
33
F
DF508/ND
Staphylococcus aureus Aspergillus fumigatus
3
24
M
Y917C/ND
Burkholderia cenocepacia (genomovar IIIA)
ND, no mutation identified.
Irish Journal of Medical Science • Volume 173 • Number 2
97
Detection of mycobacterial DNA from sputum of patients with cystic fibrosis
the presence of non-tuberculous mycobacteria in 6/37 (16.2%)
CF patients, which were mainly M. chelonae and M. aviumintracellulare complex. In Germany, Bange et al11 showed that
5/214 (2.3%) of CF patients were positive for M. abscessus.
Cullen et al2 have suggested that atypical mycobacterial disease
in CF patients may be subclinically active for a long period of
time and that it may contribute to a progressive decline in lung
function.
Furthermore, increased employment of ibuprofen in CF
may predispose the patient with underlying atypical
mycobacterial to proliferation of infection, due to a
dampening of the immune system through inhibition of the
migration, adherence, swelling and aggregation of
neutrophils, as well as leading to a reduction in lysosomal
enzymes. However, Olivier9 stated that abnormalities on high
resolution computerised tomography (HRCT) scanning,
however, were predictive of progression. Thus, current
recommendations suggest that adult patients be screened on a
regular basis with both acid-fast smear and appropriately
processed sputum or BAL fluid culture.
Furthermore, various clinical indices may suggest infection
rather than mere colonisation and these indices include multiple
positive cultures, a single positive culture associated with a
pulmonary exacerbation which is not responsive to conventional
antibiotic therapy or HRCT imaging demonstrating peripheral
pulmonary nodules, and/or a mucosal biopsy demonstrating
granulomatous disease.9,12
Overall, this molecular method has the main advantage of
rapid detection of mycobacterial DNA from sputum specimens
from CF patients. Therefore, we propose that such a method
may be a useful tool in screening studies for mycobacteria in
atypical patient populations, such as CF. As this study was
retrospective and as culture was not performed, the viability of
the mycobacteria detected is uncertain. Therefore, if such an
approach is to be used in an initial screen, we propose that all
PCR positive/sequence positive sputa are subsequently
cultured, where clinical significance is obtained when at least
two or preferably greater sputa are positive.
Conclusion
There was a low prevalence of Mycobacterium spp in the CF
patient population examined. Those patients who were PCR
positive for these organisms were not symptomatic of
mycobacterial infection and thus these organisms were believed
to represent environmental mycobacteria of little clinical
significance. Such an assay may be of benefit however, when the
microbiological status of CF patients undergoing lung
transplantation, is being assessed. Care should be used when
employing this molecular assay for the detection of
mycobacteria, due to the relatively large number of false-positive
results obtained.
98
Therefore we recommend that all PCR positive results are
confirmed by direct automated sequencing and an alternative
specific assay employed, where laboratories do not have access to
sequencing facilities. Enhanced molecular screening for these
organisms in CF patients will contribute to understanding their
role as opportunistic pathogens in patients with worsening lung
function.
Acknowledgements
MD was supported financially by the European Social Fund. KD
was supported financially by the Irish Cystic Fibrosis Association.
References
1.
Hutchison ML, Govan JR. Pathogenicity of microbes associated with
cystic fibrosis. Microbes Infect 1999; 1: 1005-14.
2. Cullen AR, Cannon CL, Mark EJ et al. Mycobacterium abscessus
infection in cystic fibrosis. Colonisation or infection? Am J Respir Crit
Care Med 2000; 161: 641-45.
3. Kulski JK, Khinsoe C, Pryce T et al. Use of a multiplex PCR to detect
and identify Mycobacterium avium and M. intracellulare in blood
c u l t u re fluids of AIDS patients. J Clin Microbiol 1995; 33: 668-74.
4. Millar B, Moore J, Mallon P et al. Molecular diagnosis of infective
endocarditis — a new Duke’s criterion. Scand J Infect Dis 2001; 33 (9):
673-80.
5. Goldenberger D, Kunzli A, Vogt P et al. Molecular diagnosis of
bacterial endocarditis by broad-range PCR amplification and direct
sequencing. J Clin Microbiol 1997; 35: 2733-9.
6. Griffith DE. Emergence of nontuberculous mycobacteria as pathogens
in cystic fibrosis. Am J Respir Crit Care Med 2003; 167: 810-2.
7. E b e rtDL, Olivier KN. Nontuberculous mycobacteria in the setting of
cystic fibrosis. Clin Chest Med 2002; 23: 655-63.
8. Olivier KN, Weber DJ, Wallace RJ Jr et al. Nontuberc u l o u s
mycobacteria. I: multicenter prevalence study in cystic fibrosis. Am J
Respir Crit Care Med 2003; 167: 828-34.
9. Olivier KN, Weber DJ, Lee JH et al. Nontuberculous mycobacteria. II:
nested-cohort study of impact on cystic fibrosis lung disease. Am J
Respir Crit Care Med 2003; 167: 835-840.
10. Oliver A, Maiz L, Canton R et al. Nontuberculous mycobacteria in
patients with cystic fibrosis. Clin Infect Dis 2001; 32: 1298-303.
11. Bange FC, Brown BA, Smaczny C et al. Lack of transmission of
Mycobacterium abscessus among patients with cystic fibrosis attending
a single clinic. Clin Infect Dis 2001; 32: 1648-50.
12. Gibson RL, Burns JL, Ramsey BW. Pathophysiology and management
of pulmonary infections in cystic fibrosis. Am J Respir Crit Care Med
2003; 168: 918-51.
Correspondence to: Dr JE Moore, Northern Ireland Public Health
Laboratory, Department of Bacteriology, Belfast City Hospital, Belfast
BT9 7AD, Northern Ireland.
Email: [email protected]
Irish Journal of Medical Science • Volume 173 • Number 2
Documentation of do-not-resuscitate orders in an Irish hospital
Documentation of do-not-resuscitate orders in
an Irish hospital
J McNamee1, ST O’Keeffe2
Departments of Nursing1 and Geriatric Medicine2, Galway Regional Hospitals, Ireland
Abstract
Background Some studies have suggested that do-not-resuscitate (DNR) decisions are often documented poorly in
European countries.
Aim To examine the use and documentation of DNR orders in a large Irish teaching hospital.
Methods Resuscitation status of all inpatients on a single day was determined using interviews with nursing staff and
examination of the nursing and medical case notes.
Results Seventeen (3.5%) of 485 patients were identified as not for resuscitation. There was written confirmation of
the DNR order in the nursing notes for 14 (82%) and in the medical notes for 15 (88%) patients; in two cases, it was
reported that doctors were reluctant to write down the agreed decision. Documentation of DNR orders was by
consultant (7), registrar (7) and intern (1). Discussion with patient (2), family (10) or both (1) was recorded in 14 cases.
Conclusion The majority of DNR orders were clearly documented by senior doctors and had been discussed with the
patient or with the relatives. A number of problems were identified that might be avoided by development of guidelines
regarding use and documentation of DNR orders.
Introduction
When cardiopulmonary resuscitation (CPR) was originally
described in 1961, it was intended that it should be restricted to
"the victim of acute insult".1 However, in acute hospitals, use of
CPR has come to be standard care for virtually any patient in
whom cardiac or respiratory function ceases and no prior decision
not to resuscitate has been made.
The outcome of CPR depends on how candidates for
resuscitation are selected. In general, the pro p o rtion of patients
receiving CPR who survive to discharge is about 15–20%.2,3 Survival
rates are even lower in elderly people, although this probably reflects
the effects of co-morbid illness more than that of age itself.4 CPR is
not a harmless intervention. For example, rib or sternal fracture s
occur in 44–75% of cases.5 As Saunders has noted: "If the expected
outcome is death, a pro c e d u re less dignified and peaceful could
hardly be devised."6 These considerations have prompted extensive
debate regarding the appropriate use of CPR.
Hospital CPR policies have long been commonplace in North
America but are less widespread in European countries. In the
US, hospital policies usually require CPR unless an explicit
written do-not-resuscitate (DNR) order has been agreed by the
patient or surrogate decision maker.7 A less formal, more
paternalistic approach has been usual in many European
countries, with the consultant as the main decision maker.8 Some
studies have suggested that resuscitation decisions are less likely
to be clearly and openly documented in European countries.9,10
This has given rise to public concern, for example, in Britain
where hospitals have been obliged to produce CPR guidelines in
recent years in response to requests from the Chief Medical
Officer and the Health Service Commissioner.11,12
In this study, we examined the use and documentation of
resuscitation decisions in a large Irish teaching hospital. As in
most hospitals in the Republic of Ireland, Galway Regional
Hospitals had no official policy regarding CPR at the time of the
study, although, in practice, CPR was (and is) performed in the
absence of a DNR order.
Irish Journal of Medical Science • Volume 173 • Number 2
Methods
We conducted a study of the resuscitation status of all inpatients
on the general medical and surgical wards of Galway Regional
Hospitals (comprising University College Hospital Galway and
Merlin Park Regional Hospital) on a single day in July 2002. We
did not include the intensive care or coronary care units, the
paediatric, obstetric or psychiatric wards.
Semi-structured interviews were conducted with nurses on the
wards, including the senior nurse on duty and primary nurses of
individual patients, regarding the resuscitation and clinical status
of patients under their care. Details recorded included the
number of patients with pre-specified conditions generally
associated with a poor prognosis following CPR. 3,4 Nurses were
also asked to identify patients they felt to be at high risk of
arresting or dying within the following week.
Nursing and medical case notes of patients verbally identified
as being not for resuscitation were examined. We determined
how many such patients had written ‘not for resuscitation’
orders in the medical and nursing notes and any discrepancies
between the two. We also examined the wording of DNR
orders in the medical notes and the seniority of the doctor
writing the order.
Results
On the day of the study there were 485 eligible patients in
Galway Regional Hospitals, of whom 219 (45.2%) were aged 70
years or more. Details of these patients are shown in Table 1.
Overall, 60 (12.4%) patients had one or more of the conditions
strongly associated with a poor outcome following CPR
(metastatic cancer, acute stroke, hypotension due to infection or
heart failure and acute renal failure). Another 65 (13.4%) patients
had chronic disability or cognitive impairment or both of
sufficient severity to interfere with their ability to live
independently. Nurses identified 48 patients that they considered
at high risk for cardiac arrest within the following week.
99
J McNamee et al
Table 1. Details of patients
Total
485
No >70 years
Metastatic cancer
Active cancer treatment
Acute stroke
Bed-chair bound (chronic)
Dementia
Intravenous antibiotics
Hypotensive shock
Acute renal failure
Recent MI
219
31
16
17
53
20
82
6
6
13
Seventeen (3.5%) patients were identified by nursing staff as
not for resuscitation. Sixteen of these patients had one or more
of the following conditions: metastatic cancer (8); cancer without
metastases (2); recent stroke (5); and severe dementia (4). One
of these 16 patients and the remaining 17th patient had a DNR
order written at the request of a patient. Fourteen (82.4%) of the
17 patients with DNR orders were aged more than 70 years
(median age 79, range 49–92 years), compared with 205 (43.8%)
of the 468 patients without DNR orders (p<0.01). Three
patients were identified in whom resuscitation had been discussed
with the patient or family and a decision to proceed with
resuscitation had been made.
T h e re was written confirmation of the DNR order in the nursing
notes for 14 (82.3%) of the 17 patients verbally identified as not for
resuscitation. In the other three cases, it was re p o rted that the
DNR order was transmitted verbally at nursing re p o rt but was not
re c o rded in the nursing notes because these notes were left by the
patients’ bedsides. In 12 cases, the DNR ord e r, usually written as
‘not for resuscitation’, ‘NFR’ or a close variant, was re c o rd e d
p rominently on the front of the nursing notes. In one case, where
the DNR was for review depending on change in the patient’s
condition, the order was re - re c o rded each day in the continuation
notes; in a further case, only the original note re c o rding that the
doctors had made a DNR order could be found.
Medical documentation regarding end-of-life issues was found
for 15 (88.2%) of the 17 cases identified by the nurses as not for
resuscitation. No such documentation was recorded for two
patients; in both cases it was reported that the decision not to
resuscitate had been agreed with the family but that doctors were
reluctant to commit this to paper. The words used were ‘not for
resuscitation’ or a close variant for 13 patients and ‘palliative care
only’ and ‘not for aggressive measures’ for the other two patients.
Documentation of DNR orders was by consultant (7), registrar
(7) and intern (1). The order documented by an intern was in
response to the request of a patient. Discussion with patient (3),
family (10) or both (1) was recorded in 14 cases; in the
remaining case, it was noted that the patient had severe dementia
and that there was no close family. In nine cases, it was explicitly
recorded that all other care apart from resuscitation should be
provided; in three cases, it was also reported that antibiotics
should not be prescribed. Three of the DNR orders were for
review in the event of a change in the patients’ condition.
It was not possible to determine if discussions with relatives
occurred in circumstances where the patient should have been
consulted. Discussion with the patient is preferable whenever
possible, although this may prove very difficult during acute
severe illness or when there is cognitive impairment.13 Some
DNR orders were not recorded in writing because notes were
available at the patient’s bedside. A benign explanation for this
practice may be that such information should not be accessible to
visitors; it is also possible that there was an intention that the
patient should not be aware of the decision. Similar findings have
been reported from other countries, with the use of cryptic
signals to indicate resuscitation status.9,10 Such behaviour could
lead to the belief that professionals have something to hide or to
feel guilty about when considering withholding resuscitation. A
similar interpretation could be made when doctors, despite
having discussed DNR orders, are unwilling to write down the
decision. Also, this practice is clearly unfair to nurses who may be
left with the responsibility for deciding about resuscitation in the
event of a cardiac arrest.
Finally, the number of patients with a DNR order seems
surprisingly low given the number of patients identified as close
to death or with conditions that might reasonably lead to
consideration of a DNR order. This finding has been reported in
other studies, including one conducted in the same hospitals over
a decade ago.14,15 This conclusion is necessarily tentative since we
are not aware of the individual factors that might have influenced
clinical decision making and because prediction of the likelihood
of success following CPR is inexact. Nevertheless, our results
suggest reluctance among doctors to make resuscitation decisions
in a timely manner. This may result in patients undergoing
unnecessary and, as the results of a previous survey of elderly
Irish patients suggest, often unwanted resuscitation.16
Resuscitation decisions are often difficult, as they involve
complex clinical and ethical issues. The making and the
documentation of such decisions have the potential to cause
disagreement and confusion amongst doctors, nurses and
patients. Many problems could be prevented by better
communication among relevant health professionals and with
patients and those close to the patient. Guidelines regarding
appropriate use and documentation of DNR orders have recently
been developed by a multidisciplinary group in Galway Regional
Hospitals, and it is hoped that these will address some of the
difficulties noted in this survey.
References
1.
2.
3.
4.
5.
Discussion
The results of this survey are generally reassuring, although a
number of problems were identified. The majority of DNR
orders were clearly documented and had been discussed with the
patient or with the relatives. Furthermore, most decisions were
documented by doctors of registrar or consultant status rather
than delegated to more junior doctors.
100
6.
7.
8.
Kouwenhoven WB, Jude JR, Knickerbocker CO. Closed chest massage.
JAMA 1960; 173: 1064–7.
Tunstall-Pedoe H, Bailey L, Chamberlain DA, Marsden AK, Wa rd NIE,
Zideman DA. Survey of 3,765 cardiopulmonaryresuscitations in British
hospitals (the BRESUS study): methods and overall results. Br Med J
1992; 304: 1347–51.
Dautzenberg PLJ, Broekman TCJ, Hooyer C, Schonwetter RS,
Duursma SA. Review: Patient-related predictors of cardiopulmonary
resuscitation of hospitalized patients. Age Ageing 1993; 22: 464–75.
O'Keeffe S, Redahan C, Keane P, Daly K. Age and other determinants
of survival after in-hospital cardiopulmonary resuscitation. Q J Med
1991; 81: 1005–10.
Saunders J. Who’s for CPR? J Royal Coll Physicians London 1992; 26:
254–7.
Rabl W, Baubin M, Broinger G, Scheithauer R. Serious complications
f rom active compression-decompression cardiopulmonary resuscitation.
Int J Legal Med 1996; 109: 284–9.
Tomlinson T, Brody H. Ethics and communication in do-notresuscitate orders. N Engl J Med 1988; 318: 43–6.
E d g ren E. The ethics of resuscitation: differences between Europe and
Irish Journal of Medical Science • Volume 173 • Number 2
Documentation of do-not-resuscitate orders in an Irish hospital
9.
10.
11.
12.
13.
the USA — Europe should not accept American guidelines without
debate. Resuscitation 1992; 23: 85–90.
Asplund K, Britton M. Do-not-resuscitate orders in Swedish medical
w a rds. J Intern Med 1990; 228: 139–45.
Dautzenberg PL, Duursma SA, Bezemer PD, Van Engen C,
Schonwetter RS, Hooyer C. Resuscitation decisions on a Dutch
geriatric ward. Q J Med 1993; 86: 535–42.
The Health Service Commission. Communications surrounding a decision
not to resuscitate a patient. London: W.258/89-90, HMSO, 1991.
Calman K. Health Service Commissioner — Annual re p o rt for 19901991. Resuscitation policy PL/CMO (91) 22. London: DoH, 1992.
O’Keeffe ST. Development and implementation of resuscitation
Irish Journal of Medical Science • Volume 173 • Number 2
guidelines: a personal experience. Age Ageing 2001; 30: 19–25.
14. Stewart K, Abel K, Rai GS. Resuscitation decisions in a general hospital.
Br Med J 1990; 300: 672–7.
15. O’Keeffe ST, Redahan C, Keane P, Daly K. Do-not-resuscitate orders in
an Irish teaching hospital. Ir Med J 1993; 106: 87–8.
16. O’Keeffe S, Noel J, Lavan J. Cardiopulmonary resuscitation preferences
in elderly hospital patients. Eur J Med 1993; 1: 33–6.
Correspondence to: Dr S O’Keeffe, Unit 4, Merlin Park Regional
Hospital, Galway. Tel. (091) 775 561; fax (091) 770 515; Email
[email protected]
101
"Is this a dagger I see before me?" — an audit of stabbings and gunshot wounds in Limerick
"Is this a dagger I see before me?" — an audit
of stabbings and gunshot wounds in Limerick
J Shabbir1, CO McDonnell1, JB O’Sullivan1, K Cahill1, A Moore1, R Raminlagan1, G Quinn2, PA Grace1
Departments of Surgery1 and Emergency Medicine2, Mid-Western Regional Hospital and National Institute of Health
Sciences, University of Limerick, Ireland
Abstract
Background According to a recent study in Cardiff, the incidence of stab wounds is 14 per 100,000 population per
annum. No such figures are available for Ireland.
Aim To evaluate the incidence, type of injury, medical consequences and outcome of patients with stab or gunshot
wounds presenting to the Mid-Western Regional Hospital, Limerick, over a 12 month period.
Method A retrospective case study of all stab and gunshot wounds presenting over a 12 month period.
Results Out of 62,000 new presentations to the Accident and Emergency (A&E) department, 101 (0.16%) were
stabbings, giving an incidence of 33 per 100,000 population. Twenty-six patients required surgical intervention.
There were three deaths. There were 10 gunshot wounds, of which 40% required surgical intervention, with
no mortalities.
Conclusion The incidence of stab wounds presenting to our institution is high. Although constituting a small
percentage of presentations to the A&E department they result in considerable morbidity and surgical activity.
Introduction
T h e re is a current perception that the incidence of violent
assaults using weapons, in the form of stabbings and gunshot
wounds, has increased in Ireland in recent years. There is,
however, a paucity of reports in the worldwide medical
l i t e r a t u re pertaining to this topic which makes objective
examination of the issue difficult. No audit of stab and gunshot
wounds has ever been published for any institution in the
Republic of Ireland previously. The aim of this study was to
analyse the incidence, age and sex distribution, type of injuries,
and morbidity and mortality, caused by stab and gunshot
wounds presenting to the Mid-We s t e rn Regional Hospital,
Limerick, over a 12-month period.
Methods
This was a retrospective case study of all the patients with stab or
gunshot wounds presenting to the Mid-We s t e rn Regional
Figure 1. Stabbings: instruments used
102
Hospital, Limerick, over a 12-month period from 1 January to
31 December 2001 inclusive.
For the purposes of the study, we defined a stab wound as a
non-accidental injury, caused by a foreign object, resulting in
penetration of the skin layers and underlying muscles. The term
‘gunshot wound’ was self-explanatory. All patients presenting
to the hospital who fulfilled this definition during the study
period had their medical re c o rds examined to determine the
patient’s details, time of presentation, weapon used and
subsequent intervention and outcome.
Results
Stab wounds
From 1 January to 31 December 2001, there were 62,000 new
presentations to the A&E department at the Mid-Western
Regional Hospital. Of these, 101 (0.16%) were stabbings.
Allowing a population attendance ratio of 1:5, this equates to an
incidence of 33 per 100,000 population.
The median age was 26 years (range 16–50) and 88% of
victims were male. The majority of alleged assaults (92/101)
occurred outside the home, the remainder (n=9) were associated
with domestic violence. Wives were responsible for 6 (67%) of
the domestic violence stabbings. Knives were the most common
weapon used (n=72), followed by broken glass bottles (n=15),
screw drivers (n=5), scissors (n=4) and in five cases the weapon
used was not identified (see Figure 1).
The majority of cases (n=69) presented at weekends (see
Figure 2) and between the hours of 10.00pm and 8.00am
(n=70). Sixty-five patients had a history of alcohol intake.
Injuries to limbs (n=39) and to the head and neck (n=36) were
more frequent than chest (n=16), abdominal (n=15) or back
injuries (n=5) (see Table 1).
Thirty-eight patients (38%) were admitted. Fifty-six patients
(56%) had their wounds explored in the A&E department
under local anaesthesia and had minor injuries requiring
suturing. Twenty-six of those admitted (68%) re q u i re d
Irish Journal of Medical Science • Volume 173 • Number 2
J Shabbir et al
Table 1. Stabbings: site of injury
Site
No.
Limbs
Head and neck
Chest
Abdomen
Back
39
36
16
15
5
ophthalmic surgeons and one patient required a laparotomy with
a small bowel resection and formation of an ileostomy. There
were no fatalities.
Discussion
Figure 2. Day of presentation
s u rgical intervention. This included chest drain insertion
(n=9), laparoscopy (n=7), exploration of abdominal wounds
under general anaesthesia (n=4), thoracotomy (n=2), repair of
major vascular injury (n=2) and peripheral wound explorations
(n=2). Three patients were transferred to a plastic surgery unit,
two with a nerve injury and one with a complex facial wound.
Three patients died as a result of their wounds. One patient
who died during resuscitation efforts in the A&E department
suffered a single stab wound to the left hemithorax which was
found on post-mortem to have penetrated the left ventricle. A
second patient who died in the A&E department was stabbed in
the neck, the carotid artery being transected. The third fatality
was a young man who suffered an inferior vena cava laceration
f rom an abdominal stab wound. He developed a massive
pulmonary embolism whilst undergoing a laparotomy and
despite a thoracotomy and attempted embolectomy being
performed immediately, he succumbed from his injuries.
Gunshot wounds
There were just 10 gunshot wounds presenting to the hospital
during the study period. All victims were male and had a mean
age of 19.2 years (range 12–26 years). Of these, four (40%)
required surgical intervention. Two patients had shotgun
wounds to the limbs, which needed debridement, one patient
suffered a penetrating eye injury which was debrided by the
The incidence of stabbings and gunshot wounds has not pre v i o u s l y
been reported in Ireland. Of 62,000 new presentations to the
A&E department in a 12-month period, 101 (0.16%) were due to
stab wounds. Assuming a population attendance ratio of 1:5, this
gives Limerick an incidence of 33 stab wounds per 100,000 of the
population. A similar study in Card i ff re p o rted an incidence of just
14 per 100,000 of the population per annum.1 Karlsson
demonstrated that just 10% of patients seeking medical attention
after physical abuse in Stockholm had penetrating injuries.2
Another paper from the same institute found that 40% of patients
who sought medical attention for stab wounds were admitted to
the hospital.3 This is similar to our own admission rate of 38%.
Webb has previously stated that stab wounds are more
common in males,4 a finding corroborated by this series (88%).
As in other reports,5 the majority of patients were in the young
and middle-aged group, median age 26 (range 16–50) years.
Similar to the Cardiff series,1 knives were the most common
weapons used, being employed in 72% of assaults. Stab wounds
are commonly perceived as serious injuries,6 however the majority
of patients in our study had minor, non-life-threatening injuries,
similar to those from Cardiff and Bulawayo.1,7
Of the 101 patients stabbed, seven (7%) re q u i red a laparotomy
(see Table 2), two of which (29%) were negative. As many as onethird of patients with penetrating abdominal stab wounds will
have serious visceral injury in the absence of physical signs.8
Salaman and colleagues reported visceral injury in 42% of
penetrating abdominal wounds. The negative laparotomy rate can
be reduced by the use of diagnostic peritoneal lavage,9
computerised tomography scanning and laparoscopy, which can
be perf o rmed through the explored debrided wound.10 Diagnostic
l a p a roscopy reduces the need for laparotomy in 55% of cases.11
Two patients underwent thoracotomy. One had bleeding
intercostal vessels causing a large haemothorax and the second
had developed a massive pulmonary embolus during
laparotomy for an inferior vena cava injury. Open embolectomy
and cardiac massage was attempted unsuccessfully. Two patients
had major vascular injuries (one brachial and one radial artery)
which were repaired with full recovery. The 3% mortality rate
reported here is compatible with what few studies are available
in the literature.1,5
Gunshot wounds were far less common in our study, with just
10 injuries presenting during the study period, of which 40%
required surgical intervention. The uncommon nature of
gunshot injuries in this part of the world has been previously
reported.12 A far higher incidence of gunshot wounds has been
reported in the US which may be a reflection of the differing gun
control laws which exist there.13,14
Table 2. Findings in seven patients requiring laparotomy following abdominal stab wounds
Injury
Intervention
Outcome
Mesenteric tear
Colonic perforation
Stomach perforation
Jejunal perforation
IVC and colonic perforation
Repair
Colostomy
Repair
Repair
Repair of IVC and right hemicolectomy, thoracotomy and
pulmonary embolectomy
Nil
Nil
Recovery
Recovery
Recovery
Recovery
Death
Stab abdomen
Stab abdomen
Irish Journal of Medical Science • Volume 173 • Number 2
Recovery
Recovery
103
"Is this a dagger I see before me?" — an audit of stabbings and gunshot wounds in Limerick
Conclusion
The incidence of stab wounds presenting to our institution is
high (33 per 100,000 population). Most occur at the weekend
and are associated with alcohol intake. Although stab wounds
constitute only a small percentage (0.16%) of all presentations to
the A&E department, they result in a considerable amount of
morbidity and surgical activity. The incidence of gunshot
wounds, at present, is low.
7.
8.
10.
11.
12.
References
1.
2.
3.
4.
5.
6.
104
Fligelstone LJ, Johnson RC, Wheeler MH, Salaman RJ. An audit of stab
wounds in Card i ff. J R Coll Surg Ed 1995; 40: 167–70.
Karlsson T. Sharp force homicide in the Stockholm area. 1983-92.
Forensic Sci Int 1998; 94: 129–39.
B o s t rom L, Jersenius U, Riddez L, Boijsen M. More stab wound attacks
despite the knife law. Injury panorama in 399 patients with stab wounds.
Swedish Medical Journal 1994; 91:380–1.
Webb E, Wyatt JP, Henry J, Busuttil A. A comparison of fatal with
nonfatal knife injuries in Edinburgh. F o rensic Sci Int 1999; 99: 179–87.
B o s t rom L, Heinius G, Nilsson BO. Trends in the incidence and severity
of stab wounds in Sweden 1987-94. Eur J Surg 2000; 166: 765–70.
Walton C, Blaisdell F, Jordan R, Bodal B. The injury potential and
lethality of stab wounds. J Trauma 1989; 29: 99–101.
13.
14.
15.
Muguti GI, Zishiri C, Dube M. Stab wounds in Bulawayo, Zimbabwe:a
four year audit. Cent Afr J Med 1995; 41: 380–5.
Cope A, Stebbings W. ABC of trauma, Abdominal trauma. Br Med J
1990; 301: 172–6.
Muckhart D, McDonald M. Evaluation of diagnostic peritoneal lavage
in suspected penetrating abdominal stab wounds using a dipstick
technique Br J Surg 1991; 78: 696–8.
Hay D. A simple device for open laparoscopy. Br J Surg 1992; 79: 155.
Fabian TC, Stewart RM, Pritchard FE, Minard G, Kudsk KA. A
p rospective analysis of diagnostic laproscopy in trauma. Ann Surg 1993;
217: 557–65.
Magee TR, Collin J, Hands LJ, Gray DW, Roake J. A ten year audit of
surgery for vascular trauma in a British teaching hospital. Eur J Vasc
Endovasc Surg 1996; 12 (4): 424–7.
Cook PJ, Lawrence BA, Ludwig J, Miller TR. The medical costs of
gunshot injuries in the United States. JAMA 1999; 282 (5): 447–54.
Bowley DM, Khavandi A, Boff a rd KD et al. The malignant epidemicchanging patterns of trauma. S Afr Med J 2002; 92 (10): 798–802.
Correspondence to: Professor PA Grace, Professor of Surgical
Sciences, University of Limerick, consultant vascular surgeon, MidWestern Regional Hospital, Dooradoyle, Limerick. Tel. (35361) 482
121; fax (35361) 482 212. Email: [email protected]
Irish Journal of Medical Science • Volume 173 • Number 2
Neonatal transportation: the effects of a national neonatal transportation programme
Neonatal transportation: the effects of a
national neonatal transportation programme
D Mullane, H Byrne, TA Clarke, W Gorman, E Griffin, K Ramesh, T Rohinath
National Neonatal Transport Programme, Coombe Women’s Hospital, National Maternity Hospital and Rotunda
Hospital, Dublin, Ireland
Abstract
Background The transport of critically ill newborns by specialised transport teams has been shown to be associated
with a significant improvement in their clinical condition on arrival at the receiving hospital.
Aim To determine if the National Neonatal Transport Programme introduced in 2001 improved clinical condition of
newborns at the end of transfer.
Methods A retrospective study of all 176 patients transported by the National Neonatal Transport Programme between
March 2001 and March 2002.
Results Before transfer, 17% of patients were hypothermic, 2% hypoglycaemic and 11% acidotic as were 7%, 3% and
5% respectively at the end of transfer. A review of 172 neonatal transports between 1987 and 1989 revealed that 21%
of patients were hypothermic, 13% hypoglycaemic and 20% acidotic at the end of transfer.
Conclusions The National Neonatal Transport Programme has resulted in improved clinical condition of newborns at
the end of transfer when compared to their condition before transfer and compared to outcomes prior to the
introduction of the programme.
Introduction
The transport of critically ill newborns by specialised transport
teams has been shown to be associated with a significant
improvement in their clinical condition on arrival at the receiving hospital resulting in a decrease in morbidity and mortality.1-4
The need for a specialised transport team in Ireland has been
highlighted in previous studies.5,6
The National Neonatal Transport Programme was set up to
transport critically ill newborns from hospitals all over the country, when needed, mainly to one of the five referral centres in
Dublin for neonatal care. Traditionally, the hospital where the
baby was born or receiving treatment transported their own
patients. The aim of the service was to bring a level of care similar to that of a neonatal tertiary centre to the point of retrieval
and maintain this during transport. The Department of Health
and Children approved the programme in 1998 and the first
patient was transported in March 2001.
The transport team consists of a neonatal registrar and a
neonatal intensive care nurse from the Coombe Women’s
Hospital, National Maternity Hospital and Rotunda Hospital
(all Dublin). Each hospital is on call for one week in three. A
purpose-built neonatal transport ambulance and driver are provided by the Eastern Region Health Authority. Team members
have to be experienced and have undertaken the STABLE programme.7 The service is available from 9am to 5pm seven days a
week and transports infants up to six weeks of age. The transport
team aims to achieve a response time of 45 minutes from the
time the transport has been requested to the time they leave the
base hospital.
A review of neonatal transports to the three main Dublin
maternity hospitals between 1987 and 1989 (n=172) revealed
that 21% of infants were hypothermic, 13% hypoglycaemic and
20% acidotic at the end of the transfer.5 These parameters are a
guide to the level of care during the transfer. These specific paraIrish Journal of Medical Science • Volume 173 • Number 2
meters were recorded for infants transferred by the National
Neonatal Transport Programme for comparison.
The purpose of this study was to review the first year of operation of the National Neonatal Transport Programme and also
to compare its effect with the previous study.5
Method
A review was performed of all transports undertaken by the
national neonatal transport team in its first year of operation,
March 2001 to March 2002. The hospital notes of all patients
transported in the study period were reviewed by one of the
authors. The nursing flow sheet and physician medical data form
completed for each transport were also reviewed. Measurements
of infant’s temperature, blood sugar and blood pH before and
after transfer were recorded and compared with those of infants
transported in the late 1980s.
Results
One hundred and seventy-six patients were transported during
the study period. One hundred and seventy-four were transported by road ambulance and two by helicopter. Twins were
transported on two occasions. Gestational age ranged from 24
to 42 weeks (mean 32 weeks). Fifty-four patients (30%) were less
than 27 weeks’ gestation. Birth weight ranged from 560 to
4,320g (mean 1,840g). The age at time of transport ranged
from two hours to 129 days (see Table 1). Two patients had a
corrected gestational age of more than six weeks at time of transport, one ten-weeks old with heart failure and one seven-weeks
old with bronchomalacia.
The most common reasons for transfer were for paediatric surgical problems (23%), congenital heart disease and patent ductus arteriosus ligation (20%), prematurity (16%) and transports
back to referring hospital (27%) (see Table 2). Most patients
(39%) were transported to Our Lady’s Hospital, Crumlin,
105
D Mullane et al
Table 1. Patients transported
Parameter
Range
Mean
Median
n (%)
Gestational age (weeks)
24-41
32
32
<25
Birth weight (kg)
Age at time of transport
0.56–4.25
1.84
2 hr–129 days
18 days
Dublin (see Table 3). Of the 176 patients transported, 136
(83%) were receiving intermittent positive pressure ventilation,
15 (8.7%) were on a dopamine infusion and 12 (6.9%) were on
a prostaglandin infusion. The average mobilisation time (time
from telephone acceptance of transport until ambulance departure from base hospital) during the study period where there
was not a specific cause for delay (n=144) was 34 minutes (range
10-90 minutes). In 20 cases, the team were on another transport
when the transport was requested. In 10 cases, there was a problem with the ambulance or equipment. In two cases, there was
a problem with bed availability at the receiving hospital. The
total time of transports outside of Dublin ranged from three
hours to 14 hours 45 minutes (mean 7.45 hours). Although the
hours of service are 9am to 5pm, 63% of transports were completed after 5pm.
The temperature, blood glucose and blood gas pH of the
patients when the transport team arrived at the referring hospital and at the end of the transfer are outlined in Table 4. It can
Table 2. Principal problem for which transport was requested
Prematurity
29
Congenital heart disease
22
PDA ligation
13
NEC
12
GI obstruction
9
Encephalopathy/seizures
6
Persistent foetal circulation
6
Sepsis
5
Exomphalos/gastroschisis
4
Diaphragmatic hernia
4
Others
18
Return transports
48
106
1.4
3 days
33 (19%)
26–27
21 (12%)
28–29
23 (13%)
30–37
40 (23%)
>37
59 (33%)
<0.75
30 (17%)
0.75–0.99
33 (19%)
1–1.49
28 (16%)
1.5–2.49
25 (14%)
2.5–3.49
38 (22%)
≥ 3.5
22 (12%)
<24 hrs
54 (31%)
2–6 days
40 (23%)
1–6 weeks
59 (33%)
>6 weeks
23 (13%)
be seen that there is a significant decrease in the number of
patients with a low temperature (<36°C) (p<0.01) and pH
(<7.25) (p=0.049). There was a slight increase in the number of
those with a low blood glucose (<2.5mmol/l) at the end of the
transfer which was not statistically significant. Of the 12 patients
who were hypothermic (mean 35.7°C) at the end of the transfer, 11 were premature.
Infant’s temperature, blood glucose and blood pH at the end
of transfers in the late 80s5 were compared with those of infants
transported by the National Neonatal Transport Programme
(see Table 4). The number of hypothermic infants has reduced
from 12% to 7%, the number of hypoglycaemic infants has
reduced from 13% to 3% and the number of acidotic infants has
reduced from 20% to 5%.
Of the 176 patients transported, 26 (15%) died before discharge from hospital. No patient died during transport. The
causes of death are listed in Table 6.
Discussion
The establishment of the National Neonatal Transport Service
has been a long-awaited important development for paediatrics
Table 3. Accepting hospital (hospital to which baby transported)
Hospital
Number
%
Our Lady’s, Crumlin
68
39
National Maternity
31
18
Temple Street
20
11
Rotunda
18
10
Coombe Women’s
17
9.5
Our Lady of Lourdes, Drogheda
8
4.5
Erinville
7
4
Other transports (all return)
7
4
Irish Journal of Medical Science • Volume 173 • Number 2
Neonatal transportation: the effects of a national neonatal transportation programme
Table 4. Clinical parameters
Before transfer 2001-2
After transfer 2001-2
After transfer 1987-9*
No.
176
176
172
Temperature <36°C
30 (17%)
12 (7%)
37 (12%)
Blood glucose <2.5mmol/l
4 (2%)
6 (3%)
22 (13%)
pH <7.25
19 (11%)
9 (5%)
34 (20%)
*Coulter Smith et al. Transportation of newborn infants. Ir Med J 1990; 83: 152-3.
Table 5. Parameters of patients hypothermic (T<36°C) at
end of transport (n=12)
Gestation
Range
Median
Mean
24-40 weeks
28 weeks
29 weeks
Birth weight
Range
Median
Mean
620-3,300g
1.2kg
1.4kg
Diagnosis
Premature (11)
Congenital heart disease (1)
Outcome
One died (hypoplastic left heart)
Table 6. Patients who died before discharge from hospital
hospitals and the Childre n ’s University Hospital, Temple Stre e t ,
Dublin, in the late 1980s was 14%.5 The corresponding mortality
after transports to these four hospitals in the present study was
11% (n=10). Five died as a result of necrotising enterocolitis and
two as a result of prematurity (four of these patients’ gestational
age was 25 weeks or less).
There are some significant deficiencies with the National
Neonatal Transport Service. At present, it is possible to request
a transport between the hours of 9am and 5pm only. It is hoped
to extend to a 24-hour service and the cost implications of this
are currently being assessed. The service is unable at present to
provide nitric oxide11 although it is planned to be available in the
near future.
Before the National Neonatal Transport Programme was
established each unit was responsible for the transport of its own
infants. One of the major risks of neonatal transport in the past,
which has been well documented in North America, is exposure
to inadequately trained personnel.12 There have been many
advances in neonatal care but we believe that the establishment
of the National Neonatal Transport Programme by providing an
experienced well-trained team for the transfer of critically ill
infants is a significant step in improving care available to critically ill infants in Ireland.
Diagnosis
Number
Hypoplastic left heart
6
Necrotising enterocolitis
5
Prematurity
3
Acknowledgements
Diaphragmatic hernia
2
Encephalopathy
2
Others
8
The dedicated work of all members of the transport team
including nursing staff, ambulance personnel, biomedical engineers and medical staff is acknowledged. The Health Boards,
Department of Health and Children, Faculty of Paediatrics
Royal College of Physicians of Ireland and the Dublin Maternity
Hospitals are thanked for their support. We thank Ms Claire
Collins for statistical advice.
in Ireland. Success in the centralisation of resources and expertise in the care of critically ill newborns is dependent on a reliable and effective transport service.8
The reduced number of patients who were hypothermic,
hypoglycaemic or acidotic at the end of transfer compared to
figures prior to the establishment of the National Neonatal
Transport Programme are encouraging. Moreover, the mean
birth weight in the study from the late 1980s was 2.3kg and
mean gestational age 34.2 weeks, compared to 1.84kg and 32
weeks respectively in the present study.
The maintenance of a normal core temperature is one of the
essentials of newborn care.9 The observation that 7% of patients
were hypothermic at the end of their transport is a cause for concern, although this is an improvement on the 17% who were
hypothermic prior to transport.5 Maintaining normal temperature in premature newborns is difficult when frequent access to
the patient may be required.10 The improvement in the observations recorded before and after transport reflect the care of the
transport team, given the often very unstable condition of these
patients.
The overall mortality in the present study was 15%. The overall
mortality after transports to the three main Dublin matern i t y
Irish Journal of Medical Science • Volume 173 • Number 2
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
Roy RN, Kitchen WH. NETS: a new system for neonatal transport.
Med J Aust 1977; 2: 855-8.
Chance GW, Matthew JD, Gash J, Williams G, Cunningham K.
Neonatal transport: a controlled study of skilled assistance. J Pediatr
1978; 93: 662-6.
Blake AM, McIntosh N, Reynolds EO, Andrew DS. Tr a n s p o rtof newb o rn infants for intensive care. Br Med J 1975; 4 (5987): 13-7.
Leslie AJ, Stephenson TJ. Audit of neonatal intensive care transport –
closing the loop. Acta Paediatr 1997; 86: 1253-6.
Coulter Smith S, Clarke TA, Matthews TG et al. Transportation of
n e w b o rn infants. Ir Med J 1990; 83: 152-3.
Hussein T, Clarke TA, Al Girim H, Matthews TG. Questionnaire on
practices and priorities in neonatal care in Ireland. Ir Med J 2001; 94:
74-8.
Karlsen KA. Transporting newborns the STABLE way. 2001.
Rashid A, Bhuta T, Berry A. A regionalised transport service, the way
ahead? Arch Dis Child 1999; 80: 488-92.
Buetow KC, Klein SW. Effects of maintenance of ‘normal’ skin
107
D Mullane et al
temperature on survival of infants of low birth weight. Pediatrics 1964;
34: 163-70.
10. Bowman ED, Roy RN. Control of temperature during neonatal transp o rt: an old problem with new difficulties. J Paediatr Child Health
1998; 33: 398-401.
11. Dhillon J, Kronick JB, Singh NC, Johnson CC. A portable nitric oxide
scavenging system designed for use on neonatal transport. Crit Care
Med 1994; 24: 1068-71.
108
12. Butterfield LJ. Historical perspectives of neonatal transport. Pediatr
Clin North Am 1993; 40: 221-39.
Correspondence to: Dr David Mullane, paediatric SpR, Cork
University Hospital, Wilton, Cork
Email: [email protected]
Irish Journal of Medical Science • Volume 173 • Number 2
Book reviews
Alzheimer Disease:
Neuropsychology and
Pharmacology
G Emilien, C Durlach, KL Minaker, B Winblad, S
Gauthier, J-M Maloteaux. Published by Birkhauser,
ISBN 3-7643-2426-0
Alzheimer Disease: Neuropsychology and Pharmacology addresses
some very important aspects of Alzheimer disease (AD), a
condition that forms a large component of a consultant’s
workload in the psychiatry of old age area. This book is a very
useful and detailed introduction to this topic.
The contents section is well laid out and is very detailed which
makes it very easy to use. This is important as throughout the
text there are very few diagrams or tables and without the
contents table, finding information could be quite daunting. A
detailed glossary of abbreviations after the contents also
reassures the reader.
The information contained appears to be up to date, is well
written and very easy to read considering the absence of
diagrams and tables to break up the text. The authors
introduce new ideas and any controversies or discussion
around those ideas in a balanced way. I refer, as an example of
this, to ‘The concept of mild cognitive impairment’ in the
introduction chapter. I felt the absence of explanatory
diagrams, photographs (in particular the absence of neuroimaging scans) and tables was unhelpful, as they would have
complemented the text very well.
I enjoyed the final chapter on ‘Discussion and Conclusion’,
which again summarised succinctly the current position and
referred to future possibilities. The extensive reference listing at
the end of the book was impressive. It was laid out well and must
be considered very useful for anyone undertaking research in
this field.
Overall the book is a useful introduction for anyone planning
to remain in this field. It is also useful for those who do not
necessarily work in this field, but require some updating of their
knowledge. Despite its lack of explanatory diagrams etc, it is an
easy-to-read text with a detailed look at the neuropsychological
and pharmacological aspects of AD.
A Colour Handbook of Renal
Medicine
J Pattison, D Goldsmith, B Hartley, FC Fervenza, JP Grande
Manson Publishing, Stg£29.95, pp240; ISBN 1-84076035-4
A Colour Handbook of Renal Medicine is designed for senior
hospital doctors (SHOs) and registrars beginning a career in
nephrology. The major feature of the textbook is the 411 very
fine colour photographs of clinical cases with renal and pathological material and microscope pictures.
The text is well written and produces particularly fine clinical
details relating to the clinical manifestations of renal disease.
There are many tables throughout the text clearly summarising
many aspects of renal disease.
Considering what a large part renal transplantation plays in
the treatment of patients with kidney disease, there is a surprisingly short section (23 pages) on this topic. The text is not
designed to be an exhaustive review of the topic but it is certainly the best set of colour pictures relating to patients with kidney disease I have seen in any textbook. I highly recommend it.
PJ Conlon
B McCabe
Irish Journal of Medical Science • Volume 173 • Number 2
111
Book reviews
Rapid Differential
Diagnosis
AH Sam, Ed. HLC Beynon, Blackwell Publishing
(online bookstore www.medirect.com), stg £12.95
This book contains a collection of differential diagnoses
for 350 symptoms, signs and laboratory tests. It is ideal
for undergraduates and postgraduates particularly for
last minute revision before an exam or a busy,
demanding ward round.
Presented in a compact and concise style, it is simply
structured in bullet format with the differentials listed
in alphabetical order, making the information easily and
rapidly accessible. The precise and abrupt nature of this
book, focusing on relevant points should assist
memorisation and retention of the information. In
addition, the clear, abbreviated format may also assist to
provide explanations for the wide range of tests, signs
and symptoms encountered in the clinical domain.
Despite this abbreviated format, it is surprisingly
comprehensive, covering an array of subjects from abdominal
pain to wrist drop.
In summary, this is a practical, handy book primarily suited
for undergraduates and postgraduates providing essential,
relevant information and is a worthwhile addition to a
s t u d e n t ’s personal library.
P Naughton
112
Irish Journal of Medical Science • Volume 173 • Number 2
Correspondence
Isolated Crohn’s disease of the appendix
Dear Editor,
Crohn’s disease may involve any part of the
gastrointestinal tract including the appendix. Isolated
Crohn’s disease of the appendix is rare.
A 16 year old female presented with a short history of
lower abdominal pain. Her signs and symptoms were
suggestive of acute appendicitis. At laparoscopy the
small and large intestine were normal. The appendix
was enlarged, acutely inflamed and thick walled
involving the base. An open appendicectomy was
performed along with portion of caecum. Her recovery
was uneventful. The final histology showed transmural
inflammation, lymphoid aggregates, micro-abscesses and poorly
formed non-caseous granuloma. Sections of the ileum and
caecum were normal. A swab from the wound was negative for
Yersinia and serological titres were normal. The ZN and PAS
stains were negative. The final diagnosis was Crohn’s disease.
The appendix is involved in 25% of cases of Crohn’s disease.
Isolated Crohn’s disease of the appendix is rare. The first case
was reported in 1953 by Meyerding and Betram1 156 cases have
since been reported.2 The clinical features are those of acute
appendicitis. There is no single diagnostic investigation but the
clinical, haematological, radiological and laparoscopic features
raise suspicion and histological findings confirm the final
diagnosis. Yersinia infection is the most common differential
diagnosis.3
The treatment is appendicectomy. Other foci of disease should
be sought. Complications are few but include enterocutaneous
fistula, abscess, wound infection or recurrence of disease
elsewhere in the gastrointestinal tract. Appendiceal Crohn’s
disease is a relatively benign form and prognosis is good but
regular follow up for four to five years is mandatory.
I Alam1, N Nolan2, J Geoghegan1
Department of Surg e ry1, Pathology 2, St Columcille’s
Hospital, Loughlinstown, Co Dublin, Ireland
References
1.
2.
3.
114
Meyerding EV, Bertram HF. Nonspecific granulomatous inflammation
(Crohn’s disease) of appendix: a case re p o rt. Surgery 1953; 34: 891-4.
Roth T, Zimmer G, Tschanz P. Crohn’s disease of the appendix.
Annales de Chiru rgie 2000; 125(7): 665-7.
EL-Maraghi NRH, Mair NS. The histology of enteric infection with
Yersinia pseudotuberculosis. Am J Clin Pathology 1979; 2: 69-77.
Irish Journal of Medical Science • Volume 173 • Number 2
General instructions to author
General instructions to authors
The Irish Journal of Medical Science will consider for
publication original research of relevance to clinical
medicine. All material is assumed to be submitted
exclusively to the journal. Submitted material will be
sent for peer review and for statistical assessment and
may be styled and edited where appropriate. The evaluation of
the editorial committee is final.
Manuscripts, with disk, should be posted to:
The Editor
Irish Journal of Medical Science
International House,
20-22 Lower Hatch Street,
Dublin 2.
Email: [email protected]
Layout of paper
Manuscripts: Submit three copies, with disk, Word for
Windows format, size 12 font, paginated and double-spaced. See
First Issue, year 2000 for current format.
Scientific articles are set out under the headings: Abstract,
Introduction, Methods, Results, Discussion and References.
The title page contains the title, authors, institution and name,
contact number and email address of the corresponding author.
The Abstract of 150 words or less should be structured as
Background, Aims, Methods, Results and Conclusions.
Abbreviations should be kept to a minimum. Measurements
should be given in SI units. Blood pressure may be expressed in
mmHg. Drugs should be given their approved name.
Statistical method used should be detailed in the Methods
section and any not in common use should be referenced.
Ethics Committee Approval by the relevant authority is
needed for investigations on human subjects and animal studies
must be in accordance with the appropriate laws.
Tables and illustrations
Tables should not duplicate text information. Illustrations
should be professionally produced and may be photographic
prints or computer generated. Top should be indicated on the
back. Staining techniques should be stated for histological
material. An internal scale marker or the final print
magnification of the photograph should be included on the
photomicrograph. Patients shown in photographs should have
their identity concealed or give written consent for
publication. Provide legends on a separate sheet. Get written
consent from the authors and copyright holder to publish
material published elsewhere.
116
References
Number references consecutively in the Vancouver style (e.g.
style1). In the references, give the names and initials of all
authors but if more than six authors only the first three followed
by ‘et al’. Book titles should be followed by publisher, place of
publication, year and pages.
1. Allen JM, Keenan AK, McHale NG et al. Lymphatic
functions and cardiovascular disease. N Engl J Med 1999;
123: 10-4.
2. Davis GK, Mertt 0. Transition metals in nutrition. In: Trace
Elements in Nutrition (4th edition). 0 Mertt (ed).
Academic Press, San Diego 1998: 123-32.
Proofs and reprints
Proof corrections should be kept to a minimum and should
conform to the conventions shown in Whitaker's Almanack. A
charge of Ä70 per page (600 words) will be levied on authors of
full papers. Reprints are available; a scale of charges is included
when a proof is sent.
Letter of release
A covering letter must be included, signed by the senior authors
stating that:
“I have read this paper and the data and interpretation concurs
with my views.”
Checklist for authors
We have checked that the submitted article follows the
appropriate format:
• Three hard copies plus disk; use Times New Roman, size 12
font; normal typeface.
• Paginated and double line spacing; two spaces after full stops;
references superscript outside punctuation. Please follow
reference style as detailed above.
• A structured abstract as detailed above.
• Case reports must have a structured abstract.
• Three copies of graphics.
• A release letter by the senior author.
Submissions will not be processed if they are not presented in
the correct format for the journal.
Irish Journal of Medical Science • Volume 173 • Number 2