Download Otitis media in children: detection of effusion and influence on hearing

OTITIS MEDIA IN CHILDREN:
detection of effusion and influence on hearing
PETRI
KO IV UNEN
Department of Otorhinolaryngology
O UL U 1 9 9 9
PETRI KOIVUNEN
OTITIS MEDIA IN CHILDREN:
detection of effusion and influence on hearing
Academic Dissertation to be presented with the assent
of the Faculty of Medicine, University of Oulu, for public
discussion in the Auditorium 10 of the University
Hospital of Oulu, on May 14th, 1999, at 12 noon.
O U LU N Y LI O P IS T O , O U LU 1 99 9
Copyright © 1999
Oulu University Library, 1999
Manuscript received 16.4.1999
Accepted 19.4.1999
Communicated by
Professor Olli Ruuskanen
Docent Markku Sipilä
ISBN 951-42-5231-4
(URL: http://herkules.oulu.fi/isbn9514252314/)
ALSO AVAILABLE IN PRINTED FORMAT
ISBN 951-42-5230-6
ISSN 0355-3221
(URL: http://herkules.oulu.fi/issn03553221/)
OULU UNIVERSITY LIBRARY
OULU 1999
To my Parents
Koivunen Petri, Otitis media in children: detection of effusion and influence on hearing
Departments of Otorhinolaryngology and Paediatrics, University of Oulu, P.O. Box 5000,
FIN 90401, Oulu, Finland
1999
Oulu, Finland
(Manuscript received 16 April 1999)
Abstract
This study was undertaken to improve the diagnosis of otitis media and to investigate possible hearing loss caused by middle ear effusion (MEE) in small children.
Temporal development of AOM was assessed from 250 episodes in 184 children. Sixty-three per
cent of cases of AOM occurred during the first week after the onset of URI, peaking on days 2 to 5.
The onset of AOM in children with a history of recurrent episodes of AOM did not differ from that
in those who had experienced only a few episodes of AOM. No individual tendency was noticed
among children suffering more than one AOM episode during follow-up.
Studies on symptomatology and the temporal development of acute otitis media (AOM) during
upper respiratory tract infection (URI) were based on three-month follow-up of 857 children.
Symptoms of URI only were compared with symptoms of URI complicated by AOM in the same
child in 138 children. The most important symptom associated with AOM was earache, with a relative risk of 21.3. Sore throat, night restlessness and fever at days 3-6 were also significantly associated with AOM, with relative risks of 3.2, 2.6 and 1.8, respectively. In 44 children under two years
of age, earache, conjunctival symptoms and cloudy rhinitis were significantly associated with
AOM.
The accuracy of minitympanometry in detecting MEE was evaluated in 162 children. The finding was compared with the amount of effusion found in myringotomy. Minitympanometry proved
to be an accurate method to detect MEE in young children, the sensitivity and specificity values
being 79 % and 93 % in cooperative children but it had no value in non-cooperative children. Minitympanometric examination could be performed successfully with good cooperation in 87 % of a
total of 206 children in paediatric outpatient clinic.
Impaired mobility of the tympanic membrane (TM) was the best sign of MEE in pneumatic
otoscopy of 76 children, with sensitivity and specificity values of 75 % and 90 %, respectively.
The influence of nitrous oxide (N2O) on MEE was tested by weighting the effusion found in
myringotomy during general anaesthesia with and without N 2O in 39 and 37 children, respectively.
The mean weight of the effusion in the oxygen-air group did not differ from the weight in the N2O
group, and thus peroperative findings in myringotomy are reliable.
To assess the influence of the quantity and quality of MEE on hearing in small children, transient evoked otoacoustic emission (TEOAE) was performed under general anaesthesia before
myringotomy in 185 ears of 102 children. Reduced TEOAEs indicating hearing loss were found in
83 % of the ears with mucoid effusion and in 56 % of the ears with non-mucoid effusion, the difference being statistically significant (p<0.01). A significant negative correlation between the reproducibility of TEOAE responses and the amount of effusion was found (Spearman rank correlation
coefficient r=-0.589, p<0.001). Findings in minitympanometry correlated with the responses of
TEOAE.
Although parents are able to predict AOM quite reliably, various symptoms and the duration of
URI seems to be of little value in helping the diagnosis of AOM. Detection of effusion in OM may
be improven by minitympanometry in cooperative children. Any kind of effusion may cause hearing loss in small children, which must be considered when treating OM.
Keywords: Otitis media, symptoms, temporal development, hearing loss
Acknowledgements
This study was carried out in the Department of Otolaryngology, Oulu University Hospital, in co-operation with the Department of Paediatrics during the years 1995-1998.
I want to express my sincere gratitude to Professor Juhani Nuutinen, MD, Head of the
Department of Otolaryngology, for his encouragement during this research and also for his
support in my clinical work. His constructive criticism as a supervisor, which is based on a
wide experience of a scientific and clinical career, has deepened my understanding about
scientific work in otolaryngology. I am also very much obliged to the former Head of the
Department of Otolaryngology, Professor Kalevi Jokinen, MD, who facilitated my first
steps in the field of scientific work.
My best thanks and respect are due to my supervisor and teacher, Docent Jukka Luotonen, MD. His support and help at all times during the study have been invaluable. As an
excellent organiser, he made it possible to carry out this research without interfering too
much with clinical work. I have always received excellent help for whatever scientific or
clinical problem I have presented to him.
I am deeply grateful for having had the privilege to do my research under the supervision
of Professor Matti Uhari, MD. His intelligence, enthusiasm and endless ideas during these
years have made a lasting impression on me. In addition to doing scientific work, I have
tried to learn some of his ways of independent thinking in research.
I wish to thank the official referees of this thesis, Professor Olli Ruuskanen, MD, and
Docent Markku Sipilä, MD, for their careful and constructive criticism.
I am very grateful to Docent Olli-Pekka Alho, MD, who has been a great support for me
during this study and also during my residency. He has also given me a lot of practical
advice during the research. My warm thanks go to Kyösti Laitakari, MD, whose unique
knowledge in the field of audiology made it possible to perform the series V.
My sincere thanks are due to Tero Kontiokari, MD, and Marjo Niemelä, MD, who have
served as an example for me in research work. The collaboration with them has not only
been productive, but also pleasant. I also wish to thank Tytti Pokka, BSc, for the skilful statistical work in the series III and IV. I want to thank Aarno Partanen, MD, for his valuable
help in the series IV.
Being an otolaryngologist is not only research work. I am most grateful to Heikki Löppönen, MD, Kalevi Hyrynkangas, MD, and Tapio Pirilä, MD, for sharing some of their wide
knowledge and experience in the field of otolaryngologic surgery.
Apart from being of continuous assistance in my research, the staff of the Department of
Otolaryngology have offered me a very pleasant atmosphere to work in, for which I am very
grateful. Especially, I want to thank Ms. Irja Jouhten for her technical help with the series I,
II and V, and Ms. Raili Puhakka for her practical help with the thesis.
Since the beginning of my medical studies, I have been proud of being a member of
’Humerus club’, which has offered me a possibility also to follow the development of other
specialities in medicine. With these twelve athletics, we have shared the successes and failures in medicine and in life during numerous sport competition events in 14 years.
I want to thank surgeon Jukka Salminen, MD, whose very special attitude to work and
life has provided me much support and pleasure during these years of research and also
served as an example for me. We have shared the same problems when doing our research,
and he has shown how to take a right attitude towards them.
The friendship with Timo Koivisto, MD, and Juha Välimäki, MD, has made my work
with this thesis much easier. Their unrealistic optimism regarding my survey has been very
encouraging.
I most sincerely thank to my parents, Sirpa and Seppo, for their encouragement during
my life. They deserve the dedication of this thesis.
Without my wife Riitta, these years would have been much harder. I have always
received endless encouragement despite the shortage of time for being together. Even
though we have spent most of our little spare time with research work, our love and friendship have still deepened. At the final stage of this thesis, our daughter Sini has reminded me
of what the really important things in life are.
This research was financially supported by the Korvatautien Tutkimussäätiö Foundation
and the Orion-Farmos company.
Oulu, April 1999
Petri Koivunen
Abbreviations
AOM
CI
ET
H. influenzae
M. catarrhalis
MEE
N 2O
OAE
OM
OME
PCR
RR
SA
SOM
S. pneumoniae
TEOAE
TM
TPP
TR
URI
acute otitis media
confidence interval
Eustachian tube
Haemophilus influenzae
Moraxella catarrhalis
middle ear effusion
nitrous oxide
otoacoustic emission
otitis media
otitis media with effusion
polymerase chain reaction
risk ratio
static admittance
secretory otitis media
Streptococcus pneumoniae
transient evoked otoacoustic emission
tympanic membrane
tympanic peak pressure
tympanometric gradient
upper respiratory tract infection
List of original papers
The thesis is based on the following reports, which are referred to in the text by Roman
numerals (I-V), and on certain previously unpublished data.
I
Koivunen P, Kontiokari T, Niemelä M, Pokka T & Uhari M (1999) Time to development of acute otitis media during an upper respiratory tract infection in children. Pediatr Infect Dis J 18:303-305.
II Kontiokari T, Koivunen P, Niemelä M, Pokka T & Uhari M (1998) Symptoms of acute
otitis media. Pediatr Infect Dis J 17: 676-679.
III Koivunen P, Alho O-P, Uhari M, Niemelä M & Luotonen J (1997) Minitympanometry
in detecting middle ear fluid. J Pediatr 131: 419-422.
IV Koivunen P, Alho O-P, Uhari M, Partanen A & Luotonen J (1996) General anesthesia
with and without nitrous oxide (N2O) and the weight of middle ear effusion in children
undergoing adenoidectomy and tympanostomy. Laryngoscope 106: 724-726.
V Koivunen P, Uhari M, Laitakari K, Alho O-P & Luotonen J. Hearing loss in otitis
media with effusion in infants and children. Submitted for publication.
Contents
Abstract
Acknowledgements
Abbreviations
List of original papers
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. Review of the literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1. Acute otitis media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.2. Epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.3. Pathogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.4. Microbial aetiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.5. Risk factors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.6. Symptoms and temporal development . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.7. Treatment of acute otitis media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2. Diagnosis of otitis media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.1. Pneumatic otoscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.2. Tympanometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3. Consequences of otitis media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.1. Infectious complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.2. Duration of effusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.3. Influence on hearing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.3.1. Measured by conventional methods . . . . . . . . . . . . . . . . . . . . .
2.3.3.2. Measured by otoacoustic emission . . . . . . . . . . . . . . . . . . . . . .
2.3.4. Late sequelae of otitis media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. Purpose of the study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4. Patients and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1. Temporal development of acute otitis media (I) . . . . . . . . . . . . . . . . . . . . . . . .
4.2. Symptoms of acute otitis media (II) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3. Pneumatic otoscopy in the detection of middle ear effusion (previously
unpublished data) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.4. Minitympanometry in the diagnosis of otitis media (III) . . . . . . . . . . . . . . . . . .
15
16
16
16
16
17
19
20
22
23
24
25
25
28
28
28
29
29
30
31
34
35
35
36
36
37
5.
6.
7.
8.
4.5. The effect of nitrous oxide on middle ear effusion (IV). . . . . . . . . . . . . . . . . . . 37
4.6. Hearing loss in otitis media measured by otoacoustic emission (V) . . . . . . . . . 38
4.7. Statistical analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.7.1. Series I and II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.7.2. Series III, IV and previously previously unpublished data . . . . . . . . . . 39
4.7.3. Series V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.8. Ethical aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5.1. Temporal development of acute otitis media during upper respiratory
tract infection (I) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5.2. Symptoms associated with acute otitis media (II) . . . . . . . . . . . . . . . . . . . . . . . 41
5.3. The accuracy of parents prediction of their child’s possible AOM
and the predictive value of symptoms (II)
42
5.4. Accuracy of pneumatic otoscopy and the value of various signs of tympanic
membrane in detecting middle ear effusion (previously unpublished data) . . . . 44
5.5. Success rate and the time required to perform the minitympanometric
examination (III) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
5.6. Accuracy of minitympanometry in the detection of middle ear effusion in
cooperative and non-cooperative children (III) . . . . . . . . . . . . . . . . . . . . . . . . . 44
5.7. The influence of nitrous oxide on middle ear effusion (IV). . . . . . . . . . . . . . . . 46
5.8. Effect of middle ear effusion on hearing when measured
by otoacoustic emission (V) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
5.9. Correlation of responses in otoacoustic emission measurement
with findings in minitympanometry (V). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
6.1. Symptoms and temporal development of acute otitis media . . . . . . . . . . . . . . . 50
6.2. Accuracy of pneumatic otoscopy and tympanic membrane signs
in the detection of middle ear effusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
6.3. Minitympanometry in the detection of middle ear effusion . . . . . . . . . . . . . . . . 52
6.4. The influence of nitrous oxide on middle ear effusion. . . . . . . . . . . . . . . . . . . . 53
6.5. Hearing loss attributable to middle ear effusion. . . . . . . . . . . . . . . . . . . . . . . . . 54
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
1. Introduction
Acute otitis media (AOM) is one of the most common diseases in children. Almost every
child experiences at least one episode by the age of three and the mean time spent with
middle ear effusion (MEE) during the first two years of life is one to two months a year.
About half a million episodes appear in Finland every year leading to a cost of 700 million
FIM. Because of the growing incidence of antibiotic-resistant bacteria, possibilities to
decrease the use of antimicrobials in the treatment of AOM has been considered.
Accurate diagnosis of OM is a cornerstone of treatment, but it is not always easy to perform, especially in small children. Minitympanometry might improve the accuracy of diagnosis in clinical studies and in primary care, but the usefulness of the device is unclear.
Parental suspicion of possible AOM during upper respiratory tract infection (URI) is based
on the assumption that some symptoms of their child are related to AOM. However, the
ability of the parents to predict AOM in their child has not been evaluated. Knowledge of
the temporal development of AOM and symptoms of the child with AOM would help to
instruct parents when to seek medical advice.
There is growing evidence of adverse long-term consequences of otitis media (OM),
which have been related to fluctuating conductive hearing loss caused by MEE. Children
who experience several OM episodes under three years of age are at risk of long-term
adverse consequences of speech and language development. Unfortunately, small children
are likely to suffer from recurrent episodes of AOM, and prolonged effusion and treatment
failures after AOM appear especially in children under two years of age. The influence of
the type and amount of MEE on hearing in small children and infants has not been determined.
2. Review of the literature
2.1. Acute otitis media
2.1.1. Terminology
Otitis media is a general term covering various conditions of an inflammatory process in
the middle ear (1). The presence of MEE, of any kind, can be considered as a sign of middle ear inflammation. It can proceed from an acute to a chronic form with reversible or
irreversible pathology. AOM is defined as rapid onset of signs and symptoms of middle
ear inflammation, usually with signs of URI, together with clinical evidence of MEE. Synonyms such as acute suppurative otitis media and purulent otitis media are commonly
used. Acute onset of purulent discharge through the perforation of tympanic membrane
(TM) or ventilation tube is also considered as AOM. The presence of MEE after an episode of AOM without continuing acute infectious symptoms can be described as residual
effusion. MEE that persists without acute signs and symptoms refers to otitis media with
effusion (OME), which some authors consider as chronic when it persists for more than 12
to 16 weeks (1-4). Various synonyms for OME, such as glue ear, serous otitis media and
mucoid otitis media, refer to the type of MEE. OME can also develop without a preceding
episode of AOM. Chronic suppurative otitis media refers to a chronic discharge from the
middle ear through the ventilation tube, or perforation of the TM.
2.1.2. Epidemiology
AOM is closely related to viral respiratory infections and it is one of the most common
diseases in childhood. Approximately every fifth upper respiratory tract infection is complicated by AOM (5,6). The incidence of AOM is highest among children under two years
of age and it gradually diminishes with age, the peak incidence being between the ages of
6 to 12 months (5,7-9). In a prospective cohort study of 498 children in the area of greater
Boston, the annual incidences of at least one episode of AOM in the first, second and third
year were 62 %, 59 % and 40 %, respectively (9). In a Finnish cohort study of 2512 chil-
17
dren, the cumulative incidence of children who experienced at least one episode of AOM
was 42 % during the first year of life and 71 % by the age of two (8). An even higher frequency of middle ear disease has been reported in a prospective study from the USA,
where 79 % and 91 % of the 2253 children experienced at least one episode of MEE
before the ages of one and two, respectively (10). Small differences in these studies may
be partly due to variable numbers of drop outs, differences in diagnostic criteria or variation in microbiological aetiology. The incidence rates of recurrent AOM in various studies
are summarized in table 1.
Table 1. Incidence of recurrent AOM episodes in different studies.
Author
Criteria
n
%
Sipilä et al. 1987 (7)
≥3 AOM before age of 18 mo
1642
30
Teele et al. 1989 (9)
≥6 AOM before age of 3 y
498
16
Harsten et al. 1989 (11)
≥6 AOM in 12 mo period before age of 36 mo
113
12
Ingvarsson et al. 1990 (12)
≥4 AOM before age of 4 y
2978
18
Duncan et al. 1993 (13)
≥3 AOM in 6 mo or 4 AOM before age of 12 mo
1220
17
Alho 1997 (14)
≥3 AOM in 6 mo period in children aged under 24 mo
2411
15
AOM, acute otitis media; y,years; mo,months; n, number of children in cohort
The incidence of AOM appers to have increased in the past few decades. Schappert
found about 2.5 times more clinic visits with a principal diagnosis of OM in the United
States during 1990 than during 1975 (15). The incidence of recurrent OM among pre-school
children has also increased from 18.7 % in 1981 to 26 % in 1988 in the United States (16).
The same tendency has been found in Finland, where the total number AOM episodes was
estimated to be 200 000 in 1982 and 500 000 in 1997 (17,18).
2.1.3. Pathogenesis
The pathogenesis of OM is multifactorial and includes host factors, anatomical factors,
environmental factors, infection and inflammation. Abnormal function of the Eustachian
tube (ET), first suggested more than 100 years ago, is probably the most important factor
in the development of AOM. OM can be attributable to loss of protective function of the
ET, impairment of pressure regulation and/or impairment of clearance of the tube and
middle ear (19). As a consequence of an inflammatory process which is usually caused by
viral respiratory infection, the patient develops congestion of the mucosa in the upper respiratory tract, which results in obstruction of the ET (20). This leads to negative middle
ear pressure, and if it continues, it is followed by transportation of potential pathogens
from the nasopharynx to the middle ear cavity. Because of obstruction of the ET, the clearance of secretion is impaired and it accumulates in the middle ear resulting in OM (20).
Infants and children with shorter, relatively wider and more flexible ET, are at a greater
risk of insufflate nasopharyngeal secretions to the middle ear.
18
A relationship between abnormal function of the ET and AOM has been documented in
several studies. Buchman et al. inoculated influenzae A viruses into the noses of 27 adults
and subsequently 59 % of subjects developed negative middle ear pressure and 25 % developed AOM (21). Similar results have been obtained by using rhinovirus (22). Negative middle ear pressure occurring during the first two days after an onset of URI has also been
detected by Sanyal et al. in a prospective study of 28 pre-school children (23). Moody et al.
followed 20 children in winter time and reported that children who developed MEE during a
common cold had negative middle ear pressure more often and the pressure returned to normal more slowly than in controls without MEE (24). Ciliary ultrastructural changes in
human nasal mucosal epithelium have been detected during the common cold (25). Similar
changes most probably occur in the ET as well. Nuutinen et al. reported mucociliary transport to be totally absent in chronic OM and in untreated secretory OM. The mucociliary
transport function returned to normal when the ear was clinically healed (26). Functional
abnormality of the ET associated with OM was demonstrated by Stenström et al., who used
a pressure chamber to test ET function. They found significantly poorer function in otitis
prone children than in controls (27). Active muscular opening function of the ET has also
been demonstrated to deteriorate during URI (28).
Several inflammatory mediators, such as leukotriens and cytokines are released in viral
respiratory infection (29). Jung et al. investigated arachidonic acid metabolites in MEE and
concluded that they are important mediators in the pathogenesis of OM (30). In addition to
the inflammatory changes causing interference in ET function, viral infection damages
mucosal cells and leads to increased bacterial adherence to pharyngeal cells, which has been
reported at least with adenovirus, influenzae A virus and rhinovirus (23,31,32,33).
Nasopharyngeal normal flora plays an important role in the development of OM. Bacteria rapidly colonize the nasopharynx of infants soon after birth and colonization remains relatively consistent among healthy children. In a Swedish study 62 % of children carried S.
pneumoniae, M. catarrhalis or H. influenzae at the age of 10 months, but even higher rates
have been reported (34,35). Each child has at least one pathogen at the age of two, but the
number of pathogens gradually decreases with age and only 40 % of children have pathogens in their nasopharynx at the age of 11 to 14 years (36). Children with early nasopharyngeal colonization by S. pneumoniae, M. catarrhalis and H. influenzae have been shown to
be at greater risk of developing OM (37). A correlation between nasopharyngeal flora and
OM has been shown in a study by Faden et al., who followed 70 children with OM and 40
healthy controls from birth to three years of age (38). They found a significant increase in
carriage rates of pathogenic bacteria during episodes of AOM. At the same time the carriage
rate of non-pathogens, in particular Streptococcus viridans, declined. Similar results have
later been obtained by Bernstein et al. (39). Pneumococcal infections are usually attributable to a newly acquired strain and are seldom associated with prolonged carriage. Acquisition of a new type peaks in winter months and it is usually associated with viral URI (35).
Even though the bacterial strain found in MEE in children with AOM can usually also be
detected in the nasopharynx, the specificity of nasopharyngeal culture is too low to guide
antimicrobical treatment in AOM (38,40).
19
2.1.4. Microbial aetiology
The association between virus infection and AOM was first described by Berglund et al.,
who recovered respiratory syncytial virus from middle ear aspirates of two children with
AOM (41). The temporal association between a viral URI and AOM has been well documented in a 14-year prospective study (5). The correlation between respiratory virus infections and AOM has been clearly demonstrated in Finnish studies where respiratory
syncytial virus and rhinovirus has been particularly associated with AOM (42-44). Direct
evidence of viral infection in the middle ear obtained by cultures or by detection of viral
antigens in MEE have been found in 15-24 % of patients with AOM (44-48). In these
studies several types of viruses have been found in MEE. Respiratory syncytial virus was
the principal virus causing AOM in the study of 456 children (49). It was found in 74 % of
MEE samples in 65 children infected by this virus, whereas parainfluenza viruses were
detected in 52 % and influenza viruses in 42 % of middle ear samples when these viruses
were the causative organisms of URI (49). Using the reverse transcriptase polymerase
chain reaction (PCR), viral RNA of rhinovirus, respiratory syncytial virus or coronavirus
has been found in 48 % of all MEE samples and also in 57 % of bacteria-negative samples
(50). The most important viruses associated with AOM in this study were rhinovirus and
respiratory syncytial virus.
Concurrent viral infection in the middle ear may worsen the outcome of antibiotic treatment of bacterial AOM. In a study by Chonmaitree, the proportion of patients who did not
respond to antibiotic treatment of AOM was 50 % in patients with both viruses and bacteria
found in MEE compared with 13 % in the group with bacteria alone (51). This is supported
by some other studies (44,45). The worse outcome may partly depend on the type of virus
present in MEE. At least rhinovirus has been associated with a poor bacteriological
response to treatment (52). Bacteria and viruses together have an additive effect on histamine concentration in MEE and the concentration is higher in patients with persistent otitis
than in controls with a good response to treatment. Thus, increased inflammation of the
middle ear may be associated with poorer outcome of antimicrobial treatment of AOM (53).
However, there are also studies where no differences in the clinical course after antibiotic
treatment have been found between patients with and without coexisting viral infection verified in MEE (50).
Although there is growing evidence of a viral aetiology in AOM, with or without concomitant bacterial infection, it is still considered mostly to be a bacterial disease. About one
third of conventionally cultured effusions are sterile, but when also detecting S. pneumoniae capsular polysaccaride antigens, evidence of bacteria is found in one third of the culture-negative samples (54). Furthermore, using meticulous bacteriological techniques, evidence of pathogenic bacteria has been found in 90 % of patients with AOM (55). When
using the PCR technique, Post et al. found 77.3 % of 97 specimens from children with
OME to be positive for at least one of the three commonest bacterial pathogens tested (56).
Another study showed a 1.5 fold increase in the proportion of positive samples of S. pneumoniae in MEE when using PCR compared with conventional bacterial culture in 135 children with AOM (57). Pathogens have also been found in 84 % of MEE samples from children with OME when using multiplex PCR, which may implicate inflammation in the middle ear caused by bacterial remnants (58). Because PCR can also detect non-viable pathogens, findings using the PCR technique may be clinically unimportant. The reliability of a
20
PCR-based assay system in detecting causative pathogens in AOM has been demonstrated
in an animal model. Non-viable bacteria did not persist in the MEE and were not detectable
by PCR, whereas viable bacteria, even when treated with antimicrobical agents, remained in
MEE for weeks and were detected by PCR (59).
The most common bacteria responsible for AOM have been S. pneumoniae in 27-52 %,
H. influenzae in 16-52 % and M. catarrhalis in 2-15 % of episodes. The proportion of these
bacteria has been relative constant in recent years (1,60). The most notable changes in data
collected from 1980-1989 in 2807 children in the United States were an increase in the
prevalence of S. pneumoniae and a rise in the beta-lactamase-producing strains of H. influenzae and M. catarrhalis (60). Other pathogens include Staphylococcus aureus, Streptococcus pyogenes, Pseudomonas aeruginosa, Klebsiella, Escherichia coli and as a new pathogen Alloicoccus otitidis, all at a proportion of a few per cent (1,61,62).
According to recent studies it can be estimated that viable bacteria or traces of bacterial
pathogens can be found in about 70-90 % of AOM episodes. Evidence of viral infection in
the middle ear, with or without concomitant bacterial infection, can be found in about 20-40
% of cases. Bacteria-negative AOM episodes may be caused by respiratory viral pathogens,
for example by respiratory syncytial virus and rhinovirus.
2.1.5. Risk factors
Even though risk factors often have confounding interrelations to each other several contributing factors to AOM have been found. Young age is the most important risk factor for
AOM (8,9). Young age at the first episode is also inversely associated with persistent
MEE and recurrent episodes of AOM (9,11,63,64). Our knowledge of anatomical and
functional factors concerning the eustachian tube and nasopharyngeal colonization in
infants support these findings (1,38).
Among the strongest risk factors of AOM is day-care in nurseries (65-71). The adverse
effect of day-care has been suggested to be most evident during the first two years of life
(71,72). The relative risk almost doubles among children attending day-care in nurseries
compared with care at home (73). The number of children in a day-care setting is also an
important determinant for early AOM, and the risk has reported to be greater in large daycare centres compared with family day-care (68,74). Tainio et al. found daily contact with
five or more children to be related to AOM with a relative risk of 2.1 (67). Family day-care
was significantly associated with AOM in a meta-analysis (73). Wald et al. reported an
increased rate of ventilation tube placement as a result of OM among children in day-care in
nurseries compared with children in home care: 21 % and 3 % respectively (72). In addition
the re-intubation rate is about three-fold among children in day-care compared with those at
home (75). The same tendency has been found in Finland, where the number of adenotomies and tympanostomies increased significantly after local authorities were obliged to
offer day-care nursery to all children under three years of age (76).
The most probable explanation for the increased risk of AOM episodes in day-care is
increased susceptibility to respiratory infections transmitted by other children. This is supported by Collet et al. who found children admitted to day-care to have twice the risk of
having a common cold in comparison with children who remained at home (71). The signif-
21
icance of transmission of respiratory pathogens in a nursery was enhanced by the result of
an intervention study, where the authors were able to reduce the number of respiratory
infections and AOM to one third by improving hygienic practice (77). Day-care attendance
is also associated with the development of OME (78).
The number of siblings seems to be associated with the occurrence of AOM although
this has not been found in all studies (9,66,68,69,74,79). In a meta-analysis, AOM was significantly related to one or more siblings with a risk ratio of 1.92 (73). Sibling’s day-care
outside the home increases contacts and the transmission of respiratory pathogens and was
the most important determinant for early AOM in one prospective study (80). A sibling’s
history of recurrent episodes of AOM may reflect a similar genetic predisposition, i.e. ET
function, or a larger amount of transmission of respiratory pathogens, or both, so it is not
surprising that a positive family history of AOM is one of the most important risk factors for
AOM with a risk ratio of 2.63 (73).
Most studies concerning the association between breast feeding and OM have found a
protective effect of prolonged breast feeding against recurrent episodes of AOM and the
time spent with MEE (69,66,79). Teele et al. prospectively followed 638 children and found
breast feeding for less than three months to be associated with an increased number of AOM
episodes in the first year of life (9). Saarinen followed 256 infants for three years and
reported an inverse association between duration of breast feeding and incidence of AOM
and the protective effect of at least six months of breast feeding lasted up to three years of
age (81). The protective effect in other two studies have lasted less than one year (65,68). In
a 1220-infant cohort study from the USA, children breast-fed for 4 months or more experienced half the number of AOM episodes compared with those not breast-fed during the 12
months of follow-up. The number of recurrent AOM episodes also declined (13). Similar
results have been obtained in Sweden, where none of 24 exclusive breast fed children experienced any AOM episodes during 12 months follow-up (82).
Studies of breast feeding have several confounding factors such as the interrelationship
with other risk factors and the adverse effect of cow’s milk or formula rather than the protective effect of human milk. Different feeding positions may also explain the findings. A
supine feeding position has been associated with earlier onset of chronic OME (65). The
benefical effect of human milk was partly demonstrated in a study by Paradise and Elster,
where open cleft palate children were fed breast milk and formula by using the same kind of
artificial feeder (83). A significant protective effect of breast milk was found indicating that
the quality of milk is more significant than the mode of feeding.
Exposure to environmental tobacco smoke increases the risk of lower respiratory illness
in the first years of life (84-86). Similarly it increases the time spent with MEE. (65,66,8790). Possible mechanisms behind the adverse effect are goblet cell hyperplasia, mucus
hypersecretion in the respiratory tract and decreased mucociliary transport (87,91). Maternal
smoking seems to be a greater risk as regards recurrent AOM than paternal smoking
(85,92). In a meta-analysis from the USA, passive smoking caused by parents increased
morbidity in children, with a risk ratio of 1.19 (86). The authors also calculated that 340 000
– 2 million AOM episodes and 5200 to 165 000 tympanostomies are attributable to environmental smoke.
Some well documented risk factors include race, some congenital disorders and immunological disorders (1,93-95). The relative risk estimate as regards use of a pacifier among
day-care children has been found to be 1.43 (96). An association of recurrent AOM with
22
nasopharyngeal dimensions in children has also been reported (97). In a questionnaire study
covering 14 000 infants, a prone sleeping position was found to be a risk factor for earrelated symptoms (98). Several other risk factors, such as low socioeconomic status, atopy,
allergy and low birth weight, have been presented as risk factors for AOM, but there is still
some discrepancy concerning the association of these factors with middle ear diseases (1).
Fig. 1. Pooled risk ratios from 18 studies analyzed in a meta-analysis of the risk factors for acute
otitis media (AOM). AOM was classified as yed or no. (Figure 1 from Uhari et al. Clin Infect
Dis 1996;22:1079-1083)
2.1.6. Symptoms and temporal development
Considering how common AOM is, surprisingly few analyses have been reported concerning the significance of those symptoms which predict the occurrence of AOM and its
temporal development. The symptoms of concomitant viral upper respiratory tract infection confound the symptoms of AOM and therefore many non-specific symptoms, such as
night restlessness, poor appetite, vomiting, diarrhoea, loss of appetite and abdominal pain
have been claimed to be associated with AOM (1,4,99). Rhinitis, cough and earache were
the symptoms which increased the risk of AOM compared with age-matched controls in a
study of 197 hospitalized children with AOM (100). All three variables correctly predicted
AOM in 63 % of cases. Usually only ear-related symptoms have been constantly related to
AOM and they exist in 54-75 % of cases. The association of other symptoms suggested to
23
correlate with AOM, such as fever, rhinitis and restless sleeping, is still controversial
(101-104).
The most common reason to seek medical advice for children during URI is the assumption of AOM. Prolongation of certain symptoms of URI and parents’ suspicion of AOM are
important reasons which lead to over-diagnosis of AOM (105). However, there are only a
few studies specifically addressing the issue. This aspect is difficult to evaluate and the surveys are therefore easily biased. Detection bias occurs if the children have not been examined during every episode of URI. Diagnosis of AOM may be inaccurate if there are several
investigators in the study or if no objective method is used. Controlling possible interindividual variation cannot be ensured when comparing symptoms in children with and without
AOM.
Heikkinen and co-workers evaluated the temporal development of AOM and found that
54 % of episodes were diagnosed during the first 4 days and 75 % during the first week after
the onset of URI (106). Children experienced three or more AOM episodes had similar pattern in developing AOM than children without previous episodes in one study (107). The
mean duration of preceding symptoms before the diagnosis of AOM has been reported to be
3.9-5.9 days (107-108). Specific knowledge of the temporal development of AOM would
help in advising parents when they should contact a physician during an URI in order to
diagnose and treat possible AOM. In previous literature, individual tendencies as regards
the temporal development of AOM have not been evaluated.
2.1.7. Treatment of acute otitis media
Since penicillin and sulphonamides were introduced for treatment of AOM, the nature of
management of OM has changed (109). Complications of suppurative OM, most frequently acute mastoiditis, and the incidence of operations because of chronic OM have
declined dramatically, which is partly due to improved welfare and hygiene in Western
countries (109,110). The aetiology of AOM has also altered; about thirty years ago betahaemolytic streptococci were the most important pathogen (109). Despite these facts, an
increased proportion of antibiotic-resistant bacteria and obscurity in placebo-controlled
studies concerning the efficacy of treatment have aroused debate about treatment policies
in AOM (111-114).
The goal in the management of AOM is to relieve the symptoms, accelerate the resolution of MEE in order to elimate hearing loss caused by it and to prevent sequelae. Because
bacteria play a major role in the pathogenesis of OM, antibiotics have been indicated in the
treatment. Several studies have been conducted in order to clarify whether antimicrobials
are beneficial in the management of AOM. A recent meta-analysis of six placebo-controlled studies indicated that the early use of antibiotics provides only modest benefit in
AOM, mainly pain relief on days 2-7 (115). Antibiotics had no effect on resolution of MEE
at one month, but there was a statistically non-significant tendency towards benefit at three
months as measured by tympanometry. The authors concluded that one child out of seventeen benefitted from antibiotic treatment (115). In contrary, in a meta-analysis of 5400 children in 33 randomized trials, an improvement of 13.7 % (95 % confidence interval (CI), 8.2
% to 19.2 %) was found as regards antibiotic therapy in the resolution of MEE within 7 to
24
14 days after the therapy started (116). Spontaneous resolution of effusion was 81 % and the
authors calculated that one out of seven children benefits from active treatment. In a study
from Pittsburgh, where myringotomy was performed to ensure the diagnosis, antibiotic
treatment brought 46.9 % effusion-free ears compared with 62.5 % in the placebo-treated
group two weeks after initialization of treatment in 1049 episodes (117). In a survey of the
microbiological efficacy of antibacterial drugs in AOM, clinical success was achieved in 93
% of cases where the bacterial pathogen was eradicated compared with 62 % of cases where
eradication failed (118). Patients younger than 24 months are at the highest risk of treatment failure and also of persistent MEE after AOM (64,117,119,120).
Antibiotics used in the treatment of AOM should be effective against the three most
common middle ear pathogens, S. pneumoniae, H. influenzae and M. catarrhalis. Despite
the growing proportion of Beta-lactamase production by these pathogens, penicillin-V and
amoxycillin are still recommended as first line antibiotics together with sulphonamide-trimethoprim by many authors and also in Finland (121-124). The conventional duration of
antimicrobical course for AOM has been 7 to 10 days. In a meta-analysis of the optimal
duration of an antibiotic course, a five days regimen was found to be as effective as seven
days or even longer (125). However, Cohen et al. conducted a randomized, double-blind
survey among 194 children with AOM, whose mean age was 13.3 months. A five-day regimen of amoxycillin/clavulanate was not equivalent to a 10-day regimen and especially
among children cared for outside their homes, a 10-day course was more effective than a
five-day regimen when assessed 12 to 14 days after the outset of treatment (126).
Studies of the value of myringotomy in the treatment of OM have been controversial, but
according to the result of recent studies it has not improved the outcome of AOM in uncomplicated cases (112,117,127-129). However, in the present era of growing resistance of bacteria to antimicrobial agents, myringotomy should be performed probably more often in
order to detect the causative organism of AOM at least in treatment failures. For the same
reason, a more aggressive strategy as regards surgical prevention has been recommended
instead of the widespread use of prophylactic antibiotic treatment (130,131). Adjuvant therapies, such as use of decongestants and antihistamine have not been shown to be of any benefit in the treatment of AOM (132,133).
2.2. Diagnosis of otitis media
Differentiation between viral and bacterial AOM could help in targeting antibiotic treatment, but this seems to be impossible without a sample from the middle ear effusion
(108). Tejani et al. demonstrated that higher levels of serum C-reactive protein are associated with bacterial AOM, but the sensitivity is so poor that it has no clinical use (134).
Serum interleukin-6 levels have been reported to be significantly higher in bacterial AOM,
but it was almost entirely attributable to pneumococcal infection. Overall sensitivity and
specificity values were 61 % and 27 %, respectively (135).
Remarkable uncertainty in the diagnosis of AOM was found in a questionnaire study
among physicians who had treated a total of 3660 children with AOM (99). In the age
groups of 0-12 months and 13 to 30 months, the doctors were uncertain of the diagnosis in
42 % and 34 % of the cases, respectively (99). Furthermore, in a review article of pertinent
25
literature concerning studies about OM, clinical diagnostic criteria of AOM were described
in only 26 of 43 studies and 18 different sets of criteria were used in these 26 surveys (136).
In the same study a questionnaire was sent to 165 paediatricians, which resulted in 147 different sets of diagnostic criteria (136). Thus, there seems to be tremendous diversity rather
than agreement in the diagnosis of OM, which also makes it difficult to compare the numerous studies in the field and hampers the discussion about the efficacy of treatment of AOM
and the discrepancy concerning late consequences of OM.
2.2.1. Pneumatic otoscopy
The diagnosis of OM requires detection of MEE (4), which is not easy to verify by pneumatic otoscope, even by an experienced examiner. The appearance of the TM is only of
limited diagnostic value. A red, cloudy or yellow TM was found in only 46 %, 52 % and
24 %, respectively, of 528 ears of 363 children with AOM (108). In the same study, fluid
level was detected in 33 % of ears with AOM. However, red TM is often used as one of
the most important signs in otoscopy indicating AOM (99,136) Karma et al. evaluated different otoscopic findings in 11 804 ear-related visits to an otolaryngologist and a paediatrician. The result of myringotomy was used as a reference method and it was performed
whenever MEE was suspected. A red TM was found in 17.5-26.8 % of AOM episodes and
an abnormal position was detected in 60.7-67.7 % of cases. In ears with MEE without
acute symptoms these figures were much lower. A cloudy TM was present in 67.1-92.9 %
of all cases when MEE was found. Only impaired motility was acceptably specific in diagnosing effusion. It was found in 93.7-98.3 % of ears when effusion was present, but also in
9.6-28.9 % of cases when effusion was not found in myringotomy (137).
When pneumatic otoscopy is performed and the findings are compared with those in
myringotomy, the sensitivity and specificity have been 81-94 % and 58-89 %, respectively
(138-141). However, the applicability of these studies to primary care is somewhat
restricted because the examinations were performed by experienced otoscopists and also the
devices would be different in primary care. It can be assumed that these figures are beyond
the overall accuracy of diagnosis of MEE when pneumatic otoscopy is used. Otoscopic
examination without testing the mobility of the TM has only limited value in the diagnosis
of OM (137,141).
2.2.2. Tympanometry
Tympanometry is a measurement of acoustic immittance of the ear canal and middle ear.
Acoustic immittance is a general term used to refer either to acoustic impedance or acoustic admittance measurements. At present most of the devices are based on admittance
measurement. Acoustic admittance is the ease with which acoustic energy is transferred
from one system to another, which is the opposite of acoustic impedance. If the air in the
ear canal is easily set into a vibration, the admittance is high. If the air is difficult to set
into a vibration, the admittance of the system is low. Middle ear pathologies that increase
the stiffness of the TM, such as negative middle ear pressure and MEE, have a great effect
26
on the transmission of low frequency signals. Tympanometry is the measurement of
acoustic admittance as a function of ear canal pressure and the resulting graph is a tympanogram. Positive or negative pressure introduced to the sealed ear canal decreases the
admittance of the air in the ear canal by stiffening the TM. The effect of air pressure on
acoustic admittance measured in the ear canal is systematically altered by the middle ear
disease (142).
Four tympanometric variables have been used to detect middle ear disease. Static admittance (SA), i.e. the ease with which acoustic energy flows into the middle ear, is an estimate
of acoustic admittance. The term compliance is also used as a synonym for SA. Equivalent
ear canal volume is a measure of the ear canal volume in front of the measurement probe.
Tympanometric peak pressure (TPP) is the position of the tympanometric peak, i.e. height
of SA, on the pressure axis. Tympanometric gradient and width describe the shape of the
tympanogram. Tympanometric width is calculated as an ear canal pressure range corresponding to a 50 % reduction in SA and it is reported in daPa. Tympanometric gradient is
defined as the change in admittance between the peak value and the mean admittance value
(142,143).
The classification mostly used in clinical practice is based on studies by Margolis et al.
and Jerger (143,144). In children the tympanogram is classified as abnormal when SA is
< 0.2 mmho (B curve) and/or if the TPP is < -139 or > +11 daPa. An A curve (SA ≥ 0.2,
TPP > -139 daPa) is interpreted as normal. When the pressure is < -139 daPa and the SA
≥ 0.2 mmho it is determined as C curve indicating negative middle ear pressure. It can further be divided to C1 and C2 curves, and the cut-off point is -300 daPa. Normal tympanometric width ranges from 50 to 150 daPa (143,144). Depending on instrumentation, some
authors have used slight modifications of these classifications.
Fig. 2. Three tympanograms; B curve, C1 curve and A curve.
The sensitivity of tympanometry in the detection of middle ear fluid against the gold
standard of the presence or absence of MEE, has varied between 58-93 % and the specificity between 71-93 % (139,140,145-147). Orchik et al. classified the amount of MEE found
in myringotomy in four categories; none, minimal, moderate and impaction. They reported
that in 89 % of ears with a B tympanogram at least a moderate amount of effusion was
found, but 33 % of ears with a type A tympanogram also showed a notable amount of effu-
27
sion in myringotomy (148). The predictive value of a C curve was uninformative. Nozza et
al. developed diagnostic criteria for a device they used and reported sensitivity of 83 % and
specificity of 87 % at best, when combining SA and the tympanometric gradient (149).
When combining tympanometry and pneumatic otoscopy, sensitivity increases to 90-98 %
and specificity to 80-93 % (138,139,141).
Tympanometric results have been shown to correlate with conductive hearing loss
caused by MEE in young children in some studies. A type B tympanogram suggests a pure
tone air conduction threshold greater than 25 dB with a sensitivity of 91-93 % and a specificity of 71-76 % in 3-10-year-old children (150,151). An even better result of tympanometric screening has been reported by Haapaniemi, who stated that TPP>150 daPa is even better than 15 dB hearing screening in the detection of MEE in 7-14-year-old children (152).
However, the correlation of tympanometric findings to hearing level in children under three
years of age has not been determined.
Most of the previous studies concerning the reliability of tympanometry have been carried out with clinic tympanometers, which are inpractical in out-patient use. Minitympanometry is portable tympanometry that has been developed for use in primary care. It has
five parts: pressure transducer, pump, loudspeaker, microphone and microcomputer. A 226
Hz probe tone is introduced into a sealed ear canal. A microphone records the sound pressure in the ear canal. The sound level is maintained at a constant 85 dB SPL (Sound Pressure Level) throughout the test. When the amount of sound energy absorbed by the middle
ear increases, the voltage to the loudspeaker increases, thus maintaining the constant SPL.
The voltage required to maintain the probe tone at 85 dB SPL is proportional to the acoustic admittance of the ear. Air pressure in the ear canal is changed throughout the test by the
pump mechanism, and the acoustic admittance is displayed as a function of the ear canal
volume (142).
There is still some disagreement about the accuracy of minitympanometry in detecting
MEE, and in particular the specificity has been relatively poor. When clinic tympanometry
serves as a reference instrument, the sensitivity and specificity have been 96 % and 81 %,
respectively (153). When minitympanometry is compared with the gold standard, its sensitivity and specificity values have varied between 90-94 % and 48–63 % in pre-school children, respectively (138,153). In a Finnish out-patient clinic study, false negative findings
were found with minitympanometry in 3 % and false positive findings in 7 % of children,
when compared with pneumatic otoscopy (154). Interobserver validity of minitympanometry has been shown to be high, up to 95 % (155).
Previous studies have been mainly carried out with older children, which decreases the
applicability to primary care. Although cooperation of the children at the time of tympanometric examination probably affects the accuracy of tympanometry, it has not been mentioned in all studies. In addition, differences in quality and quantity of MEE may affect the
accuracy of tympanometry. Children at the age of highest risk of AOM have been examined in only a few studies.
In most of the previous studies concerning the accuracy of tympanometry, myringotomies have been performed under general anaesthesia and one explanation offered for the
relatively poor specificity values has been the displacement of MEE into the ET by anaesthesia gases (138,146,147,156-158). It is well documented that anaesthesia gases, especially nitrous oxide (N2O), increase middle ear pressure (159-161). Gates and Cooper also
found rising middle ear pressure, but they concluded that it was due to assisted mask venti-
28
lation (162). However, Levy et al. in a well-planned study demonstrated direct gas diffusion from blood into the middle ear. They tracheotomized 4 guinea pigs and sealed the proximal part of the trachea (163). Freon-22 breathed through the distal part of the trachea was
found in each middle ear appearing there through diffusion from the blood. Similarly, N2O
should diffuse from the blood into the middle ear cavity, and the middle ear mucosa was
suggested to play an active role in a gas composite exchanging (163). This is also supported
by the result of another study, where significant changes of N2O concentration across the
middle ear mucosa were found during general anaesthesia (164). Disagreement still exists
about the effect of increased middle ear pressure on MEE.
2.3. Consequences of otitis media
2.3.1. Infectious complications
Severe infectious complications as sequelae of AOM, such as labyrinthitis, petrositis,
intracranial infections and abscesses are uncommon nowadays in Western countries (1).
Since the pre-antibiotic era, the incidence of acute mastoiditis, which is the most common
complication of AOM, has declined dramatically (165). Previously streptococci were the
typical bacteria detected in acute mastoiditis but in a recent analysis of 57 cases, S. pneumoniae was the most common causative pathogen (165,166). A conservative approach
using intravenous antibiotics and a drainage procedure seems to be equally effective as
surgical therapy in the management of acute mastoiditis (166). Complications such as subperiostal abscess, meningitis and facial palsy developed in 23 % of patients with acute
mastoiditis in the 124 patients with acute mastoiditis (167). The high proportion of
patients who present with complications of mastoiditis may reflect the masking effect of
antibiotics or a delay in diagnosis. In developing countries mastoiditis still occurs in 0.19
% to 0.74 % of schoolchildren (168). The incidence of facial palsy as a complication of
AOM also diminished after the introduction of antibiotics, from 5/1000 to 5/100 000
(169). In an analysis of 23 patients complete remission was seen in all but one case (169).
van Buchem reported that withholding initial antibiotics in AOM treatment in the Netherlands has not been shown to increase the incidence of complications of AOM (112). However, a conservative approach and withholding of antibiotics in the treatment of AOM has
increased the occurrence of acute mastoiditis in Germany (170).
2.3.2. Duration of effusion
Because a tendency for spontaneous recovery and the appearance of asymptomatic episodes in MEE, exact figures for the time spent with effusion after AOM is difficult to evaluate. In ten placebo-controlled studies, the spontaneous cure rate was 14 % to 88 % (119).
When using tympanometry in screening for the presence of middle ear fluid, pathological
findings indicating effusion have been detected in about 7-20 % of 2 to 3-year-old Danish
children, depending on the season of the year (171,172). A tendency of spontaneous reso-
29
lution has also been demonstrated in screening studies (171-173). After an episode of
AOM, MEE persists for more than four weeks in about 40 % of children (129,174). In
children aged under and over two years of age, effusion was still found in 57 % and 46 %
of placebo-treated children six weeks after the onset of AOM in 527 episodes, respectively
(117). In a study of 498 children from greater Boston, the authors estimated that the mean
time between diagnosis and resolution of effusion is 23 days in actively treated children
(9).
Children less than 24 months of age have a 3.8 times higher risk of persistence of effusion after AOM than older children, which may be a consequence of a similar pathophysiology as that predisposing them to AOM (64). A previous episode of AOM is also a strong
risk factor, odds ratio 11.9, for chronic MEE (175). The risk disappears after a period of
three months without OM. In a survey of 95 infants followed up to three years of age, the
mean duration of bilateral MEE at each episode in groups with low and high AOM incidence was 6 and 10 weeks, respectively (176). Thus duration of effusion is associated with
recurrent AOM.
The average time spent with MEE in all children during the first three years of life,
which is the most critical age regarding language development, is considerable. In the
cohort from greater Boston, the total time spent with MEE during the first, second and third
year of life was estimated to be 29, 27 and 18 days, respectively (177). Friel-Patti and
Finitzo reported an even higher duration in children aged 6-18 months. The mean time spent
with MEE was 62 days per year (178), which was about the same as in a study by Roland et
al (179). Recently, Paradise et al. reported the mean cumulative proportion of days with
MEE to be 19 % in the first year and 15.6 % in the second year in a cohort of 2253 infants.
About 45 % of urban children suffered persistent MEE for three months in the first year of
life (10).
2.3.3. Influence on hearing
2.3.3.1. Measured by conventional methods
Adverse long-term developmental effects of OM have been considered to be a consequence of hearing loss caused by MEE (178). Because of the lack of cooperation in small
children, only few investigators have evaluated hearing impairment caused by AOM in
children under three years of age. Olmstead et al. found hearing loss more than 15 dB in
67 % of children aged 2.5 to 12 years in an initial audiogram following the diagnosis of
AOM (180). In a study comparing different treatment modalities in AOM, 31 % of the
patients showed an air-bone gap of more than 20 dB in 2- to 12-year-old children one
month after diagnosis (112).
Hearing loss in OME has been evaluated in several studies. In children under 15 years of
age, the mean hearing level of speech frequencies was 27.6 dB, and in 38 % of the patients
hearing loss was more than 30 dB (181). In a study by Fria et al., the authors used behavioural observation audiometer, incorporating conditioned responses to an animated toy in
222 infants aged 7-22 months, and pure tone audiometer in 540 children aged 2 to 11 years
to examine hearing levels in OME (182). MEE was associated with an average hearing level
30
of approximately 27 dB at 500, 1000 and 4000 Hz. 30 % of the older children had a pure
tone average of 30 dB or more, the range varying from 0 to 50 dB. In the group of infants
with bilateral OME, the mean speech awareness threshold was 24 -26 dB (182). Cohen and
Sade found similar hearing thresholds in cases of secretory OM in 222 children (183). Hearing loss caused by chronic MEE has also been evaluated by the auditory brain stem
response. The mean hearing threshold was 32 dB in 328 ears in children between the ages of
6 to 18 months (179). Hearing loss seems to depend both on the quantity and the quality of
effusion (184,185).
2.3.3.2. Measured by otoacoustic emission
To obtain objective and ear-specific data in young children concerning possible hearing
loss caused by MEE, the only applicable methods are measurements of auditory brain
stem responses and otoacoustic emissions (OAEs). Even though auditory brain stem
responses have been claimed to be useful in evaluating hearing in infants with MEE, the
long time required for the examination has limited its use (186). Otoacoustic emissions are
sounds measured in the external ear canal that are a reflection of an active process occurring in the outer hair cells in a healthy inner ear (187). They consist of the backward leakage of the energy of the basilar membrane travelling wave. This leakage is caused by
spatial irregularity within the rows of outer hair cells. The phenomenon can be measured
in almost all closed ear canals with normal middle ear and cochlear function (188).
Because the cochlea is frequency-specific in its response, frequency-specific information
can also be obtained by measuring OAE (188). There are two classes of OAE; spontaneous and evoked. Spontaneous emissions can be measured in the external ear canal without
external sound stimulation. They are present in about 60 % of persons with normal hearing. Evoked OAEs are elicited by external sound stimulation. They can further be divided
into stimulus-frequency, distortion-product and transient evoked otoacoustic emissions
(TEOAEs). According to Probst and Harris, TEOAEs are preferable for screening purposes, whereas distortion-product otoacoustic emissions may be more valuable for monitoring cochlear changes clinically (189). TEOAEs are elicited by brief acoustic stimuli
such as clicks or tonebursts and they can be detected in essentially all normally hearing
ears in sealed ear canals (190). Because MEE causes stiffening of the TM and ossicular
chain, it reduces transmission of acoustic stimulation level and also the backward transmission of OAE, the latter being particularly affected (188). Changes in middle ear pressure also alter the responses of TEOAE (191-193).
When comparing the results of TEOAE measurement with audiometric findings, it has
been stated that when the overall hearing is better than 20 dB, TEOAEs are present in more
than 90 % of ears. When hearing loss is more than 20 dB, TEOAE responses decrease
sharply, being absent when the hearing level is 30-40 dB or poorer (194). Other authors
have also concluded that when the sensorineural hearing level exceeds 20-30 dB, TEOAEs
are usually not measurable (188,189,195,196).
In previous studies the effect of MEE on responses in TEOAE measurement has been
variable. Owens et al. recorded TEOAEs in children with chronic MEE diagnosed by B
tympanogram and otoscopy and whose average hearing loss was 25 dB. None of the 20 ears
31
with effusion demonstrated normal TEOAEs (197). In contrast, Amedee found measurable
TEOAEs in 50 % of 60 ears with MEE in 30 children aged 23 - 88 months (198). None of
the patients with mucoid effusion showed responses in this study. The author concluded that
the presence or absence of emissions was primarily a function of the type of middle ear fluid
(198). When comparing TEOAE with Play audiometry at the 20 dB level, TEOAE has been
demonstrated to be sensitive method in detecting conductive impairment (199). This is in
agreement with Hall et al., who stated that TEOAE can be recorded in middle ear disease
only when hearing thresholds are 30 dB or better (195). Nozza et al. found measurement of
TEOAE to be a reliable method in screening for hearing impairment and middle ear disorders (200). Thus, TEOAE measurement seems to be an accurate method in detecting conductive hearing loss induced by MEE in small children and infants.
2.3.4. Late sequelae of otitis media
Development of the phonological system is dependent on the child’s repeated experience
with spoken language. Phonetic prototypes are necessary for language perception preceding speech comprehension (201). The latter half of the infant’s first year is a critical period
in learning language and during this period infants begin to acquire a vocabulary (202).
The fundamental task of language acquisition and segmentation of words from fluent
speech can be accomplished by 8-month-old infants, based on statistical relationships
between neighbouring sounds (203). In addition 8-month-old infants are beginning to
engage in long-term storage of words that occur frequently in speech, which is an important prerequisite for learning language (202). The critical period for phonology has been
found to be in the second half of the first year of life, for syntax it is up to the end of 4th
year of life and for semantics up to the end of the 15th –16th year of life (204). Since a
fluctuating hearing ability increases the variation of perceived sound patterns even more
than a constant hearing loss and thus impairs the recognition of words in infancy, a fluctuating hearing loss during the sensitive period may be even more hazardous to language
development than a stable, mild hearing impairment (205).
In 1969 Holm and Kunze reported that a fluctuating hearing loss accompanying chronic
OM before the age of two causes delay in speech and language development at the age of 5
to 9, measured by various tests (206). One of the first prospective studies in 1988 revealed a
significant association between the time spent with MEE and language development (207).
Teele et al. followed 207 children from birth to the age of seven, at which age they assessed
intelligence, academic achievement, speech and language (177). Confounding variables
were controlled by multivariate analysis and they found that the time spent with MEE during the first three years of life was inversely associated with ability in speech, language and
school performance in reading and mathematics. No association was found between OM
suffered during years 4-7 and performance at the age of seven (177).
In a study of 394 children, retrospective data were collected from parents by questionnaire to obtain a history of OM (208). The validity of retrospective data from a random sample was checked by comparison with medical records, and the information given by the parents was found to be accurate. Multiple regression analysis adjusted for confounding variables showed that children with more than four otitis episodes before the age of three years
32
performed significantly less well in the reading comprehension test at the age of nine, and
the adverse effect was stronger among girls than in boys. Reading comprehension test result
correlated with the teacher’s grading of the student’s reading (208). When teachers evaluated the performance of 1708 children at school at the age of seven and the result was compared with otitis history, recurrent OM episodes before the age of three were associated with
lower performance in various mathematical skills (RR 1.2 –1.4, 95 % CI 1.0 to 1.7) and
classroom concentration among girls (RR 1.4, 95%, CI 1.1 to 1.4) (209). The boys with
recurrent otitis episodes performed poorly in reading (RR 1.3, 95 % CI 1.0-1.6) and oral
performance (RR 1.2, 95% CI 1.1 to 1.4) (209).
Most of the studies evaluating the association between hearing and developmental disorders have been based on epidemiological evidence and not on serial hearing measurements,
which would more reliably show the relation between fluctuating hearing loss and late
sequelae of OM. This restriction has been overcome in a survey by Friel-Patti et al., where
auditory brain stem responses were recorded in order to evaluate hearing levels in 14 otitis
prone and 14 non-otitis prone children at the ages of 6, 12 and 18 months (205). Language
development was assessed at the age of 24 months. In the otitis-prone group 71.5 % of children showed language delay, with 42.9 % delayed greater than 6 months and it was related
to conductive hearing loss. In the control group 21.4 % of children suffered from language
delay (205). Similar risk factors for OM and for delayed developmental disorders may have
confounded the results in some studies, even though multivariate regression analysis were
performed. The fact that there are only minor differences in risk factors for OM among
healthy siblings has been utilized in a study by Sak and Ruben, who compared 18 children
with a history of recurrent MEE with their effusion-free siblings. They found significant
associations between verbal ability, auditory decoding and spelling skills, and recurrent
MEE (210). Otitis-prone children are particularly considered to produce consonants less
accurately than their counterparts with little or no history of OM (211). There are also
reports which indicate that OM is associated with behavioural problems (212,213).
On the other hand, there are several studies which have failed to demonstrate any relationship between early OM and cognitive, language and psychosocial development. In a
prospective study of 88 children, a modest association between time spent with MEE and
low measures of language and cognitive skills was found at the age of two years (214). The
authors explained the result by the finding that children with more OME lived in less caring
environments. However, the total time with MEE in this study was 82.5 % between 6 and
14 months, which could have caused bias in the control group. Some investigators have
included only socially disadvantaged children (215,216), or evaluated AOM occurring in
children older than three years (217), or in children with repaired palatal clefts (218). All of
these factors may bring about the result of the study towards non-significancy. Because of
the time-consuming tests required to evaluate cognitive abilities, language and reading,
study population tend to be small, which can also make the result non-significant (219). The
heterogeneous tests and differences in language in various countries also hamper the comparability of studies. Other confounding factors may be difficulties in the diagnosis of MEE,
interaction between risk factors and limitations in research designs (220).
There is also some evidence that OM in childhood is associated with permanent high frequency hearing loss (221,222), which seems to be mainly of the conductive type (223). Neither recurrent AOM or its treatment seem to cause permanent sensorineural hearing loss in
speech frequencies (224).
33
Despite the lack of agreement about the late developmental consequences of childhood
recurrent OM, in most of the studies at least some tendency towards an adverse effect of
OM has been detected. Thus, a symposium of developmental implications of early life OM
concluded that a child with OME should at least be provided with an optimal listening and
language-learning environment (225).
3. Purpose of the study
The purpose of the present study was:
1. To determine the temporal development of AOM.
2. To evaluate how the symptoms of the child and parental suspicion of AOM predict
the existence of AOM.
3. To assess which signs are accurate in diagnosing middle ear effusion by pneumatic
otoscopy.
4. To find out if diagnostic accuracy in OM can be improved with minitympanometry
and to assess whether there is an association between the tympanogram and the
amount of MEE.
5. To evaluate if N2O in general anaesthesia has an effect on middle ear pressure and
the amount of MEE.
6. To evaluate the hearing loss in OM measured by TEOAE.
The overall aim of the study was to improve accuracy in the diagnosis of MEE and to
evaluate hearing loss resulting from it.
4. Patients and methods
4.1. Temporal development of acute otitis media (I)
A total of 857 previously healthy day-care children aged 0.6-6.9 years (mean 3.7 years)
were enrolled from 34 day-care nurseries. The parents were asked to register all the symptoms, especially fever, cough, rhinitis (clear, cloudy, obstructive), earache, sore throat,
vomiting, diarrhoea, night restlessness, irritability, poor appetite and conjunctival symptoms of their child during the study period. When a child had any of the following acute
symptoms: rhinitis, cough, sore throat or conjunctivitis, the child was considered to have
URI. The symptoms were registered on symptom sheets that were collected monthly. During visits to the clinic, the parents were also asked whether they suspected that their child
had AOM or not.
At the beginning of the study, all the children were screened for MEE by hand-held
(Micro Tymp® Welch Allyn) minitympanometry and, if the finding was abnormal, by pneumatic otoscopy. Children were included in the study only after complete resolution of MEE.
When a child had any symptoms suggestive of URI, the parents were asked to bring him/her
to the outpatient department within 3 days of the beginning of the episode, and if AOM was
not diagnosed, once a week until the symptoms disappeared. Whenever the child had earache or a sore throat or if the parents suspected AOM, they were asked to bring their child to
the clinic the same day. During the visit, a trained nurse performed a minitympanometric
examination. Tympanometry was repeated at least three times before an abnormal tympanogram was accepted. The tympanogram was classified as abnormal if SA was < 0.2 mmho
(B curve) or if TPP was < -139 (C curve) or > +11 daPa (positive curve) together with SA
over 0.2 mmho. An A curve (SA ≥ 0.2 mmho, TPP -139 - +11 daPa) was interpreted as normal. Pneumatic otoscopy was always performed when the finding in tympanometry was
interpreted as abnormal, or the examination was not considered reliable, or if the child had a
sore throat or earache (or if the parents suspected it), or if there was discharge from the ear.
The otoscopic criteria for the diagnosis of AOM were discharge from the ear, an air-fluid
level behind the TM or a cloudy or red TM together with impaired mobility accompanied
by the symptoms of URI. At least three asymptomatic days between any given URI and
AOM episodes were required before they were registered as separate events. The disappearance of possible previous middle ear effusion had also to be verified before registering a
36
new episode of AOM. AOM was treated with amoxycillin (40 mg/kg) for 7 days whenever
there was no contraindication for this medication.
Children who had had at least one episode of respiratory tract infection with AOM during a three-month follow-up period were included in the analysis of temporal development
of AOM. The time lag preceding the diagnosis of AOM after the beginning of URI was registered, and the onset of earache was calculated from the symptom diary. A possible individual tendency to develop AOM with a consistent pattern was analyzed among the children
who had had more than one episode of AOM. The hypothesis that children with a history of
recurrent episodes of AOM have a different pattern in developing AOM compared with
children who have had only few attacks was tested by comparing the temporal development
of AOM between these two groups. The children who had undergone operative treatment
because of recurrent AOM or who had had more than five episodes of AOM before entering
the study were classified as having recurrent episodes of AOM. The possible effects of gender and age on the developmental pattern of AOM was also analyzed.
Because the children were not examined daily during their URI, we carried out two further analyses to assess the possible delay in the diagnosis of AOM. The developmental pattern of AOM was compared between the children whose diagnosis of AOM was based on
discharge from a ventilatory tube or from a spontaneous perforation, and all the other children with a diagnosis of AOM. We also compared the time lag between developing AOM
after the onset of URI in children with and without earache.
4.2. Symptoms of acute otitis media (II)
The population in series II was the same as in series I. Only the first attack of AOM during
each URI episode was included in the analyses. Children who had persistent symptoms of
URI for more than 30 days were excluded from the analyses. Only those children who had
had respiratory infections both with and without AOM during the follow-up period were
included in series II. The symptoms during an episode of AOM were compared with the
symptoms during an episode without AOM. Calculations involving parental suspicion of
AOM were carried out for the whole study population, with a total of 345 episodes of
AOM.
4.3. Pneumatic otoscopy in the detection of middle ear effusion
(previously unpublished data)
A total of 76 children referred for adenoidectomy and/or tympanostomy tube placement in
the Department of Otolaryngology of Oulu University Hospital were included in the study.
Thirty-nine (51 %) of the patients were boys and the ages ranged from 7 months to 9 years
(mean 2.5 years). The previous otological history was obtained by parental interviews. All
the children were examined preoperatively by using pneumatic otoscopy by the same otolaryngologist and the appearance, position and mobility of the TM was registered. Myrin-
37
gotomy was performed on each ear and the presence of effusion was registered. Signs and
motility of the TM were registered and compared with the absence or presence of MEE.
4.4. Minitympanometry in the diagnosis of otitis media (III)
Two different sets of patients were enrolled to the series III. The first set consisted of children brought to the Out-patient Emergency Department of Paediatrics, University Hospital of Oulu, with any sign or symptom of infection. A trained nurse performed
minitympanometric examination (MicroTymp ® Welch-Allyn) in a total of 206 children,
130 boys and 76 girls, whose mean age was 4.7 years, ranging from one month to 16
years. The success rate and the time needed to perform minitympanometric examination
on each ear were recorded.
The second set of patients consisted of 162 consecutive children (68 girls and 94 boys)
referred for adenoidectomy and/or tympanostomy tube placement in the Department of Otolaryngology, University Hospital of Oulu. The ages of the children varied from 7 months to
8 years (median 34 months). The patients had earlier experienced an average of 8.5 episodes
of OM (range 1 to 25). Informed consent was obtained from all the parents. Ten ears out of
324 were excluded because of perforation or tympanostomy tubes. Minitympanometric
examination was performed on each ear, and it was repeated if a type B tympanogram was
obtained. The child was classified as non-cooperative if he or she was crying or resisted the
tympanometric examination. The tympanometric findings were compared with the results
of myringotomy performed on each ear under general anaesthesia. The quality of the MEE
was classified by the operating surgeon as serous or mucoid.
The tympanograms were classified as type A when the SA was ≥ 0.2 mmho and the TPP
-200 - -100 daPa, type B when the SA was < 0.2 mmho, and type C when the TPP was
≤ 200 daPa and the SA ≥ 0.2 mmho. Measures of the curves were interpreted separately by
the authors, and in cases of disagreement, a consensus was reached through a group conference.
4.5. The effect of nitrous oxide on middle ear effusion (IV)
The study population in series IV was the same as in the study on pneumatic otoscopy
(previously unpublished data, 4.3). Tympanometry (MicroTymp® Welch-Allyn) was performed on each ear before the operation by the same otolaryngologist. Four ears out of
152 were excluded because of a perforation or a tympanostomy tube. The examination
was repeated during an anaesthesia before myringotomy. Myringotomy was performed on
each ear by another surgeon, who was unaware of the result of the preoperative tympanometry. The presence of possible effusion was registered and the effusion was aspirated.
The suction tips were weighed before and after the aspiration of MEE and the weight of
the effusion was calculated. The quality of the MEE was visually classified as serous or
mucoid by the operating surgeon. The tympanograms were classified as type A, when
compliance was ≥ 0.3 mmho and the TPP > -100 daPa, type B, when SA was < 0.3 mmho,
and type C when TPP was < -100 daPa and SA 0.3 mmho. The curves were interpreted
38
separately by the authors, and in cases of disagreement, a consensus was reached in a
group conference.
Anaesthesia was induced with fentanyl (1-2 µg/kg) followed by thiopenthal sodium.
Sixty-four of the patients were relaxed with succinylcholine and endotracheally intubated
when adenoidectomy was performed. Tympanostomy only was performed on 12 children
and anaesthesia was maintained by mask ventilation under spontaneous breathing. The first
39 children were ventilated with a mixture of 30 % oxygen and 70 % N2O with 0.5 -1.0 %
isoflurane (N2O group). The following 37 children were ventilated with a mixture of oxygen
and air at an oxygen fraction of 0.3 with 0.5-2.0 % isoflurane (oxygen-air group).
The N2O and oxygen-air groups were similar with respect to age. The number of previous AOM episodes was also similar in the N2O (mean 10 episodes) and the oxygen-air
groups (mean 8 episodes). Preoperative tympanogram findings were also similar in the two
groups.
4.6. Hearing loss in otitis media measured by otoacoustic emission (V)
All the children referred to the Department of Otorhinolaryngology of the University of
Oulu for adenoidectomy and/or ventilation tube placement because of chronic MEE or
recurrent otitis media episodes between January 1 and April 30, 1998, were regarded as
candidates for the study. After informed consent had been obtained from the parents, a
minitympanometric examination with a hand-held device (MicroTymp® Welch Allyn)
was performed by a trained nurse on each ear after the child had been given premedication
for the operation. If the first tympanogram was classified as B curve, the examination was
repeated at least three times before the finding was classified as reliable.
The minitympanometric findings were classified as in series II. B and C2 curves were
interpreted as abnormal and C and A curves as normal tympanograms. Only the children
who were cooperative at the time of the minitympanometric examination were included in
the study (102 children).
TEOAEs were measured using the Quickscreen mode of a computer-based ILO 88 Analyzer (Otodynamics Ltd., Hatfield, England) under general anaesthesia before myringotomy.
Before the measurement, the external ear canal was cleaned. The stimuli consisted of a standard set of non-linear clicks with an intensity of 95 dB peak equivalent sound pressure.
Alternate responses were stored and averaged in two separate buffers, A and B. The correlation between the two averages determined the reproducibility of TEOAE, which was calculated by the device and expressed as a percentage.
TEOAEs were considered to be present when the amplitude of the response was at least
3 dB above the noise level and at least halfway across each of the test frequency bands of 1
to 2, 2 to 3 and 3 to 4 kHz. When only one or two of the test frequency bands were present,
TEOAEs were classified as being partially present. The TEOAE measurement was considered technically unreliable when the stimulus level in the closed ear canal was less than 71
dB. The higher the reproducibility, the more identical waves A and B appear, which indicates the presence of TEOAE.
39
4.7. Statistical analysis
4.7.1. Series I and II
Statistical analyses were performed by using SPSS version 7.0. McNemar´s test was used
to test differences between symptoms during episodes with and without AOM. Relative
risk (RR) and 95 % CI were calculated for each symptom. The sensitivity and specificity
of parental suspicion concerning their child’s possible AOM were calculated. The ability
to predict AOM by means of various symptoms was analyzed using a logistic regression
model where AOM was the dependent variable and symptoms were covariants.
The duration of symptoms of URI in different groups with and without AOM was compared by using the Mann-Whitney U-test.
4.7.2. Series III, IV and previously previously unpublished data
Differences between the N2O and the oxygen-air groups were tested with a two-tailed ttest for continuous variables, and a binomial test was used for the non-parametric variables. 95 % confidence intervals were calculated accordingly. The sensitivity and specificity of the results of pneumatic otoscopy and the finding of a B curve in minitympanometry
were assessed by using the presence or absence of middle ear effusion in myringotomy as
a gold standard. The Spearman rank correlation coefficient was calculated between the SA
of minitympanometry and the net weight of MEE in the cooperative group. Differences
between the mean weights of effusion in ears with different types of tympanograms were
first tested in one-way analysis of variance, and if the difference was significant, the analysis was continued with a t-test. The sensitivity and specificity values of different pneumatic otoscopic signs predicting acute otitis media were also calculated.
Sensitivity:
Specificity:
Probability that the diagnostic test result will be positive when the
disease is present
=true positive results
total patients with disease
Probability that the diagnostic test result will be negative when the
disease is not present =true negative results
total patients without disease
4.7.3. Series V
Differences between the mean weight of effusion in the presence and absence of TEOAE
were calculated by one-way analysis of variance. The Chi-square test was used to analyze
the difference between mucoid and non-mucoid effusion associated with responses in
TEOAE measurement. The Spearman rank correlation coefficient was calculated between
the reproducibility of TEOAE and the weight of MEE.
40
4.8. Ethical aspects
All the studies were approved by the ethics committee of Oulu University Hospital.
Informed consent was obtained from the parents. Myringotomies performed in this
research were of normal diagnostic practice in children undergoing adenotomy or tympanostomy. Measuring TEOAEs in children during general anaesthesia is a non-invasive
examination.
5. Results
5.1. Temporal development of acute otitis media during upper
respiratory tract infection (I)
A total of 250 episodes of AOM were diagnosed in 184 children (mean age 3.0 years) during three months of follow-up. Sixty-three per cent of cases of AOM complicating URI
occurred during the first week, peaking on days 2 to 5. Eighty-nine percent of all episodes
occurred by the end of the second week after the onset of URI (Figure 3).
The onset of AOM in children with a history of recurrent episodes of AOM did not differ
from those who had experienced only a few episodes of AOM. Nor did gender or age affect
the onset of AOM. Sixty-one children had more than one occurrence of AOM in two distinct URI episodes. The temporal development of AOM during the first event did not correlate with the time lag during the second event, which was interpreted as lack of an individual tendency in the temporal development of AOM.
The 127 children with earache showed a similar time in developing AOM as children
without ear-related symptoms, and also fifteen children whose diagnosis was based on discharge through ventilation tube had a similar time lag in diagnosis of AOM as other children who were diagnosed by pneumatic otoscopy. The mean duration of symptoms of URI
in the subgroups with and without AOM was similar in this study, where AOM was treated
effectively.
5.2. Symptoms associated with acute otitis media (II)
A total of 138 children experienced at least one episode of URI with or without AOM. The
most important symptom associated with AOM was earache with a relative risk of 21.3.
However, 41 % of children with AOM did not have ear-related symptoms and 15 % had
earache but AOM was not diagnosed. Sore throat, night restlessness and fever were also
significantly associated with AOM, with relative risks of 3.2, 2.6 and 1.8, respectively.
Eight per cent of the children with AOM suffered fever more often at days 3-6 than chil-
42
dren with URI alone (Table 2). Non-specific symptoms such as cough, poor appetite and
gastrointestinal symptoms were not related to AOM.
In the total of 44 children younger than two years, only earache suspected by parents
(RR= 8.5), conjunctival symptoms (p=0.016) and cloudy rhinitis (RR=3.2) correlated to
AOM. In the 94 children of two years or older the value of earache was emphasized, the risk
ratio being 47. Sore throat was also significantly associated with AOM, with risk ratio of 3.3.
Other symptoms with positive association were night restlessness and irritability (Table 2).
Cases 35
120 %
30
100 %
25
80
20
60
15
40
10
20
5
0
0
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
Days after the onset of URI
Fig. 3. Occurrence of acute otitis media after the onset of upper respiratory tract infection
(URI) in 250 episodes. The bar represents the number of cases on each day, and the line represents the cumulative force of morbidity.
5.3. The accuracy of parents prediction of their child’s possible AOM
and the predictive value of symptoms (II)
The parents correctly predicted their child’s AOM in 71 % of cases. The sensitivity and
specificity of their suspicion were 71 % and 80 %. The corresponding figures for children
younger and older than two years were 75/75 % and 69/82 %, respectively. The best combination of symptoms, assessed by logistic regression analysis, was earache together with
night restlessness which predicted AOM correctly in 71 % of cases.
19 (14)
33 (33)
58 (42)
20 (15)
Sore throat
Night restlessness
Fever
Conjunctival symptom
12 (9)
1.0 (0.3-2.5)
1.0
0.4
0.3
0.07
0.3
0.2
0.1
0.06
0.03
0.02
0.03
<0.01
P
6 (14)
33 (75)
13 (30)
22 (50)
23 (52)
37 (84)
28 (64)
9 (20)
26 (59)
17 (39)
4 (9)
22 (50)
AOM
(%)
3.0 (0.2-158)
8.5 (2.0-76)
RR
(CI)
** (**)
8 (18)
0.7 (0.1-2.8)
30 (68) 1.4 (0.5-4.4)
14 (32) 1.1 (0.4-3.7)
22 (50 1.0 (0.4-2.5)
14 (32) 2.3 (0.9-6.6)
24 (55) 3.2 (1.2-9.7)
27 (61) 1.1 (0.4-3.1)
2 (5)
19 (43) 1.8 (0.7-4.69
15 (34) 1.3 (0.4-4.7)
2 (5)
7 (16)
URI
(%)
Children < 2 Years (n = 44)
* McNemar´s test ** Not possible to calculate since there were no URI children with this symptom
symptom
11 (8)
102 (74) 95 (69) 1.3 (0.7-2.5)
Gastrointestinal (diarrhoea,
vomiting or pain)
Cough
23 (17) 1.5 (0.7-3.4)
40 (30) 1.7 (1.0-3.2)
30 (22)
48(35) 1.4 (0.8-2.4)
57 (42) 1.5 (0.9-2.5)
70 (51) 1.5 (0.9-2.7)
2.4 (1.0-6.9)
39 (28) 1.8 (1.1-3.2)
10(7)
54 (39)
obstructive
3.2 (1.1-11)
20 (14) 2.6 (1.1-6.9)
8 (6)
Poor appetite
69 (50)
57 (42)
cloudy
RR
(CI)
21 (15) 21.3 (7.0-106)
URI
(%)
All (n = 138)
Irritability
70 (51)
clear
Rhinitis
82 (59)
AOM
(%)
Earache
Symptom
0.8
0.6
1.0
1.0
0.09
0.02
1.0
0.02
0.2
0.8
0.6
<0.02
P
URI
(%)
RR
(CI)
P
6.5(1.5-59)
3.3(1.0-14)
1.4(0.5-4.4)
2.6(0.9-9.3)
5 (5)
4 (4)
1.3(0.3-6.3
69 (73) 65(69) 1.3(0.6-2.8)
17 (18) 9 (10)
32 (34) 18(19) 2.8 (1.2-7.2)
34 (37) 34(37) 1.0(0.5-2.1)
32 (34) 33(35) 1.0(0.5-1.9)
54 (58) 43(46) 1.8(0.9-3.7)
11 (12) 8 (9)
32 (34) 20(22) 1.9(0.9-3.9)
16 (17) 5 (5)
15 (16) 6 (6)
1.0
0.6
0.1
0.02
1.0
1.0
0.1
0.6
0.08
<0.01
0.05
60 (64) 14(15) 47.0(8.0-1819) <0.01
AOM
(%)
Children ≥ 2 Years
(n = 94)
Table 2. Symptoms in 138 children who had upper respiratory infection (URI) both with and without acute otitis media (AOM). Relative
risks (RRs), 95 % confidence intervals (CIs) and P -values*.
43
44
5.4. Accuracy of pneumatic otoscopy and the value of various signs of
tympanic membrane in detecting middle ear effusion (previously
unpublished data)
Most of the TM signs in pneumatic otoscopy had limited value in the detection of MEE.
Only impaired mobility of the TM had a moderate sensitivity value (Table 3). When the
different signs were combined into an overall subjective judgement, however, an acceptable sensitivity (81 %) and specificity (86 %) were obtained (Table 3).
Table 3. Sensitivity and specificity of various tympanic membrane (TM) signs detected by
pneumatic otoscopy in identifying middle ear effusion in children.
Signs in pneumatic otoscopy
Red or opaque TM
Sensitivity (%)
Specificity (%)
64
36
Bulging or retracted TM
56
93
Impaired mobility of the TM
75
90
Air-fluid level
12
100
Overall judgement
81
86
5.5. Success rate and the time required to perform the
minitympanometric examination (III)
Minitympanometric examination was performed successfully in 179 children (86.9 %)
and failed in at least one ear in 27 children of the outpatient children. The patients who did
not cooperate were significantly younger than those who allowed the nurse to examine
their ears (mean 1.5 years vs. 5.8 years, p<0.0001). The mean time needed for tympanometry was 2.1 minutes (range 0.5 to 10 minutes).
5.6. Accuracy of minitympanometry in the detection of middle ear
effusion in cooperative and non-cooperative children (III)
The tympanometric examination could be performed with good cooperation on 183 ears
of 162 children, whereas in 131 examinations the child was not cooperative at the time of
the examination among children referred for adenoidectomy and/or tympanostomy. Ten
ears were excluded because of perforation or tympanostomy tubes. The mean age of the
children was 40 months (median 35 months) in the cooperative group and 23 months
(median 19 months) in the non-cooperative group.
The sensitivity and specificity values of the B curve in detecting MEE by tympanometry
were 79 % and 93 % in the cooperative group. In the non-cooperative group sensitivity was
71 %, but specificity only 38 % (Table 4). Sensitivity and specificity of minitympanometry
in the children under 24 months were 78 % and 91 %, and in children older than 24 months
45
they were 79 % and 94 %, respectively. The proportion of unsuccessful examinations was
much smaller in the cooperative group (Table 4). The net weight of the fluid varied from 5
to 695 mg with a mean of 59 mg. Most of the false A curves (A curve in ear with effusion
found in myringotomy) were associated with only insubstantial amounts of effusion (Figure
4). SA had a significant negative correlation with the weight of the middle ear fluid in the
cooperative group (Spearman rank correlation r=-0.66, P<0.001).
MEE was present in myringotomy in 30 % of the ears. Mucoid effusion was more common than serous (59 % versus 33 %). The quality of effusion did not affect the findings in
minitympanometry.
Fig. 4. The weight of the middle ear fluid of the cooperative patients classified according to the
findings in minitympanometry.
46
Table 4. Preoperative tympanometric findings compared with the presence and the mean
weight (mg) of middle ear fluid obtained in myringotomy in 314 ears of 162 children
classified by the child’s cooperation at the time of the tympanometric examination. Ten
ears were excluded because of perforation or tympanostomy tubes.
Finding in myringotomy
Cooperative*
Non-cooperative**
A
B
C
A
B
C
106
9
13
16
29
2
7
42
3
3
10
1
79 (49)
213 (117)
180 (181)
239 (54)
193 (135)
20
Fluid absent
n
Fluid present
n
Weight (SD)
The sensitivity and specificity in the cooperative group were 79 % and 93 %, respectively. In the non cooperative
group, sensitivity was 71 % and specificity 38 %. When calculating these figures, the A and C curves were
summed. * Unsuccessful or unclear: 3. ** Unsuccessful or unclear: 70
5.7. The influence of nitrous oxide on middle ear effusion (IV)
The number of ears where effusion was found was similar in the N2O and the oxygen-air
groups. The mean weight of effusion in the oxygen-air group did not exceed the weight in
the N2O group. Instead a non-significant tendency towards more effusion in the nitrous
oxide group was noticed. The mean weights of effusion in the N 2O and oxygen-air groups
were 66.4 mg and 49.0 mg, respectively (difference is 17 mg, 95 % CI –21.1-56.0,
p=0.383). (Figure 5)
The increment of middle ear pressure was indirectly demonstrated by a change in tympanograms from A curve to B curve in the N2O group. Eighteen of the 20 type A preoperative
tympanograms changed to type B tympanograms during N2O anaesthesia. In the oxygen-air
group, only 4 of the 16 type A tympanograms changed to type B peroperatively. The result
was similar in the spontaneous mask ventilation group. (Table 5)
47
Fig. 5. Weight of middle ear effusion in 148 ears according to the use of nitrous oxide (N2O).
Nitrous oxide was used in group A and it was not used in group B. Intratracheal intubation l,
mask ventilation s.
Table 5. Pre- and peroperative tympanometric findings in 76 children (148 ears) under
anaesthesia with and without nitrous oxide (N2O).
Preoperative
Anaesthesia
With N 2O
Mask ventilation
Intratracheal intubation
Total
Without N2O
Mask ventilation
Intratracheal intubation
Total
*=unsuccessful
Peroperative
Number of
ears with
effusion (%)
A
B
C
U*
A
B
C
U*
3
17
20
7
29
36
0
16
4
2
0
18
2
59
2
9
0
68
0
7
0
1
24 (35)
8
10 (83)
24 (35)
34 (44)
2
16
18
4
26
30
0
6
6
3
13
16
2
27
29
6
28
34
1
4
5
0
2
2
6 (67)
25 (41)
31 (44)
48
5.8. Effect of middle ear effusion on hearing when measured by
otoacoustic emission (V)
TEOAE could be measured technically reliably in 185 ears of 102 children. Among the 67
ears with effusion, no responses were obtained in 31 (45 %) ears and in 19 (28 %) ears
responses were measured in all the three frequency bands. In 19 (28 %) ears with effusion,
reponses were recorded in one or two frequency bands. Among 116 ears with no effusion,
in only 3 (3 %) ears were no responses obtained (Table 6). Reduced TEOAEs were found
in 83 % of the ears with mucoid effusion and in 56 % of the ears with non-mucoid effusion, the difference being statistically significant (p<0.01) (Table 6). Responses were significantly dependent on the amount of MEE. There was a significant difference between
the mean weight of effusion in the ears with no responses and the ears with responses in
all three frequency bands (p<0.001). Significant negative correlation between the reproducibility of TEOAE responses and the amount of effusion was also found (Spearman
rank correlation coefficient, r=-0.589, p<0.001) (Table 6).
5.9. Correlation of responses in otoacoustic emission measurement
with findings in minitympanometry (V)
The minitympanometric examination could be performed reliably in 136 ears of 102 children, and the finding could be compared with those of TEOAE measurement in 124 ears
in 98 children. Among the 89 ears with normal tympanograms, 65 (73 %) showed
responses in all three frequency bands, while no responses were obtained in 6 (7 %) ears.
Among the 35 abnormal tympanograms, reduced responses in the TEOAE measurement
were present in 25 (71 %) ears. There was a significant association (p<0.01) between the
weight of MEE found in myringotomy and the type of tympanogram (Table 7).
Table 6. Mean weight (mg) of effusion compared with responses in transient evoked
otoacoustic emission (TEOAE) measurement in 185 ears where TEOAE could be measured
reliably in the study population of 102 children operated upon because of otitis media.
No
effusion
TEOAE responses
No responses
Response in one frequency band
Responses in two frequency bands
Responses in three
frequency bands
Overall
Mucoid**
Non-mucoid**
3
n
24
mean
191
SD
89
range
35-360
n
7
mean
127
SD
86
3
5
104
101
30-280
1
190
13
6
25
15
10-50
7
67
48
97
7
75
120
10-340
12
98
116
42
138
109
10-360
27
101
range
10-290
Total*
mean
161
SD
101
79
97
10-150
24
38
89
10-340
15
52
80
10-340
46
85
*Significant (p<0.001) difference between groups, except between the groups of responses in two and three
frequency bands. **Significant (p<0.01) difference in reduced TEOAE between mucoid and non-mucoid
effusion.
49
Table 7. Mean weight (mg) of middle ear effusion classified by responses in transient
evoked otoacoustic emission (TEOAE) measurement compared with normal (A curve and
C1 curve) and abnormal (B curve and C2 curve) minitympanometric findings in 124 ears
where a minitympanometric examination and a TEOAE measurement could be performed
reliably in a study population of 102 children operated upon because of otitis media.
Minitympanometric findings
Normal*
Abnormal*
TEOAE responses
n
mean
SD
range
n
mean
SD
range
No responses
6
105
88
0-200
16
180
106
0-320
Response in one frequency band
3
17
29
0-50
4
158
97
80-280
Responses in two frequency bands
15
16
26
0-90
5
36
64
0-150
Responses in three frequency bands
65
4
16
0-110
10
24
48
0-150
Overall
89
13
37
0-200
35
112
112
0-320
*Significant (p<0.01) difference between the weights of middle ear fluid
6. Discussion
6.1. Symptoms and temporal development of acute otitis media
Our study design concerning symptoms of AOM had some advantages compared with
previous studies. The diagnosis of AOM was confirmed by both minitympanometry and
pneumatic otoscopy, and in most cases by the same physician. To control possible bias of
parents seeking medical help and variations in children’s reactions to different pathogens
during URI, we examined the children during each episode of URI and compared the
symptoms of URI with and without AOM in the same child. Symptoms of concomitant
URI always confuse assessment of the specific symptoms of AOM, which may explain the
variation of results in previous studies concerning the issue.
The most important symptom related to AOM in our study was earache, with a relative
risk of 21.3, which is in agreement with the result of other studies (100-104). The result is
not surprising because of an ongoing inflammation and infectious process in the middle ear
during AOM. Assessment of possible earache in a small infant may be difficult and therefore parents may also regard restless sleeping, crying, irritability and rubbing of the ear as
earache, which probably partly explains the variation of frequency figures in previous studies.
Our finding that sore throat and fever are significantly correlated to AOM, which is not
supported by all studies (102), is in agreement with our understanding of the pathophysiology of AOM. Inflammation of the nasopharynx caused by viral infection may well result in
a feeling of sore throat in children. The fever profile, with a greater proportion of febrile
AOM children on days 2-9 is suggestive of the development of a bacterial complication of
the viral disease, which agrees with the present opinion of AOM being a bacterial disease
(1). The finding is also in accordance with the results of a clinical trial on AOM, where antimicrobials were found to significantly reduce fever compared with placebo (129). The association of conjunctival symptoms and cloudy rhinitis with AOM in children less than 2
years olds in this study, could be indicative of bacterial infection at all these sites, which is
supported by the result of some previous studies where usually non-typeable H. influenzae
has been found in conjunctival, nasal and middle ear cultures (226,227). We found no asso-
51
ciation between AOM and gastrointestinal symptoms, cough and poor appetite, as suggested
earlier (1,99).
Even though earache was the most important symptom related to AOM, as many as
41 % of the children with AOM did not have any ear-related symptoms. This is also in
agreement with previous studies where the sensitivity of earache has been as low as 60 %
(100-102). Fifteen per cent of children had earache but were found not to have AOM. Thus,
AOM cannot be reliably differentiated from URI on the basis of symptoms, while even the
best combinations of symptoms can predict AOM correctly in only about 71 % of cases in
children over and under two years.
Despite wide variation in the symptoms, the parents could predict if their child had AOM
with relatively good accuracy, the sensitivity and specificity being 71 % and 80 %, respectively. These figures were surprisingly similar in both age groups despite the differences in
the combinations of symptoms predicting AOM in these age groups. Thus parental suspicion of AOM cannot be ignored, whatever the symptoms are. The finding is also epidemiologically important, suggesting that data showing an increasing incidence of AOM based on
the number of clinic visits are reliable and do not merely reflect changes in the habits of parents in seeking medical advice (15).
Variability of time in the development of AOM during URI in children and even variation in the same child is easily understood, when considering the multifactorial aetiology of
the pathophysiology of AOM. Patency of the ET, severity of inflammation, function of the
immunological system and virulence of causative pathogens are major aspects of the
progress of AOM (1). The children in our study developed AOM during URI somewhat
later than in a study of Heikkinen et al., who found 54 % of AOM episodes to appear in the
first four days and 75 % in the first week after the onset of URI (106). In both studies the
peak incidence of AOM was during the second to fifth days, which is in accordance with the
finding that negative middle ear pressure manifests within the first few days after the start of
URI in pre-school children (23,25). In a study by Sanyal et al. 97 % of the children developed negative middle ear pressure during the first four days after the onset of URI (23).
However, they also found great variation in the development of effusion, being up to 28
days.
The value of duration of URI in the diagnosis of AOM is limited. In our study, the cumulative incidence of AOM was 41 % in the first four days but as many as 40 % of AOM episodes developed later than one week after the onset of URI. If the parents had visited a physician only during the first two days of URI, more than 80 % of the AOM episodes would
have been missed. Furthermore, AOM develops irregularly even in the same child, and the
temporal development of AOM in children with multiple episodes is not different from
those with only a few episodes.
Prolonged symptoms and parental expectation are two reasons bringing about overtreatment of AOM (105). Even though there are some symptoms significantly associated with
AOM, the duration of URI is of no help in making correct diagnosis. According to the criterias, AOM is defined as detection of effusion in middle ear together with symptoms of URI
(4). Differentiation of AOM and OME can be impossible in some cases with recurrent URI
with residual AOM effusion. There are no definitive symptoms or limit in the temporal
development of URI which would suggest one or the other. Appearance of TM is also of
limited value in distinguishing AOM from OME. However, performing tympanostomy in
either case would not be overtreatment.
52
6.2. Accuracy of pneumatic otoscopy and tympanic membrane signs in
the detection of middle ear effusion
Impaired motility of the TM had acceptable sensitivity and specificity values in detecting
MEE, which agrees with results published by Karma et al. (137). The appearance of the
TM had no value in our series, which is supported by the results of some studies
(108,228), but not all (99,139). Karma et al. found cloudiness to be a good sign in predicting MEE, but they performed myringotomy only when they suspected AOM, which
impairs the accuracy of diagnosis (137). Even the position of the TM appears to be of limited value in the diagnosis of MEE, because of its poor sensitivity. Deciding on a diagnosis
of OM should always be based on detection of middle ear fluid, which is only achieved by
assessing the mobility of the TM when using otoscopy. This would simplify the diagnostic
uncertainty and the wide variation in diagnostic criteria demonstrated previously (99,136).
Our study population comprised of children who did not have ongoing URI, and thus
AOM was not suspected. However, according to the definition of AOM, i.e. effusion
appearing in the middle ear during URI, the cornerstone of diagnosis is detection of MEE.
We were able to confirm that accurate diagnosis of OM by otoscopy is impossible without
assessment of the mobility of the TM. Thus, performing otoscopy instead of pneumatic
otoscopy is of very limited value in the diagnosis of OM, a fact which is supported by the
results of earlier studies (145)
The sensitivity (81 %) and specificity (86 %) values achieved in this study are acceptably
high considering that the children had suffered from several episodes of AOM and most of
the children were under three years old, which impairs cooperation. In previous studies in
which pneumatic otoscopy has been use in the diagnosis of AOM, the sensitivity and specificity have been 81-94 % and 74-89 %, respectively (138, 140,141). These values have been
achieved by experienced otoscopists, which means that accurate diagnosis of OM is not
easy to achieve. Declining frequency in carrying out myringotomy has also worsened the
feedback of physicians’ diagnoses of OM.
According to a review article on OM, failure to distinguish AOM from OME is a common reason for the overuse of antibiotics (229). The author claimed that the signs of acute
infection, i.e. red, yellow, grey or thickened TM, can be detected and are distinctive of
AOM. A discolored TM of normal thickness and position together with decreased mobility
would suggest OME (229). However, these signs have been considered to be unacceptably
inaccurate method in previous studies on the diagnosis of AOM (108,137), and in our study
also the appearance of the TM as regards detection of effusion was worthless. Thus, distinguishing AOM from OME on the base of the TM signs may be difficult.
6.3. Minitympanometry in the detection of middle ear effusion
We found minitympanometry to be a method well comparable to pneumatic otoscopy performed by an experienced otoscopist in detecting middle ear fluid when performed in
cooperative children, which is in accordance with the results of previous surveys
(138,140,141). An objective method could increase accuracy in the diagnosis of OM in
primary care, where the diagnostic criteria of AOM show great variation (136). In the rare
53
cases of an incorrect A curve against the gold standard, the amount of fluid is insubstantial
with no effect on hearing (V). The significant negative correlation between SA and the
weight of MEE agrees with our result in series V, where the amount of fluid correlated to
hearing loss measured by OAE (V). The clinical utility of SA has not been widely
accepted, probably due to a lack of norms and standardized measurement procedures, as
suggested in some previous reports (142,143). The amount of fluid in ears with type C1
tympanograms appeared to be roughly the same as in ears with type A tympanograms. Our
opinion is that, in clinical practice, A and C1 tympanograms can be grouped together and
interpreted as normal findings in minitympanometry. Similar suggestions have been made
earlier, even though there is no general agreement on the significance of a type C tympanogram (140,153).
The great weight variation of MEE found in our study (5 mg-695 mg) partly accounts for
the discrepancy in previous studies concerning the sensitivity and specificity of tympanometry, because tympanometry appears to display a normal A curve in cases of small amounts
of MEE (139,140,145-147,150). Disregard of the child’s cooperation or crying and differences in interpretation may also explain the variation in specificity values of tympanometry
in previous studies (145).
According to our results, minitympanometry may improve diagnostic accuracy in cases
of AOM. Although minitympanometry has no value in the detection of MEE in non-cooperative children, the proportion of cooperative children among group of unselected patients in
a paediatric outpatient clinic in our series was sufficiently high (87 %) to justify the use of
the device in primary care. This is supported by the results of two previous studies in which
clinical tympanometry and pneumatic otoscopy have been served as a reference methods
(154,155). Performing minitympanometry in the age group of under two years, which is
also the critical period as regards the late sequelae of OM, is less often successful than in
older children.
Previously the validity of minitympanometry and pneumatic otoscopy have been compared in only one study, and the authors found the latter to be slightly more accurate (138).
The sensitivity and specificity values achieved in minitympanometry and pneumatic otoscopy in this study were calculated from different populations, and non-cooperative children
were also included to the analysis of pneumatic otoscopy. Thus, comparability of the figures
is somewhat limited. However, both of these methods are sufficiently accurate in the diagnosis of AOM.
6.4. The influence of nitrous oxide on middle ear effusion
The amount of MEE was not influenced by N2O used during general anaesthesia. The
result was similar among the children in the mask ventilation group and in the tracheal
intubation group, which indirectly supports a previous finding that gas diffusion into the
middle ear occurs through the blood circulation (163). A negative effect of N 2O on MEE
is supported by some authors, even though they did not measure the amount of effusion
and the conclusions were based only on tympanometric findings (159,230). We were able
to confirm indirectly a finding, that N 2O anaesthesia raises the pressure in the middle ear
(159,161,163,164). The previously reported discrepancy concerning sensitivity and espe-
54
cially specificity values of tympanometry in detecting MEE seems not to be attributable to
displacement of MEE to the ET by way of increased middle ear pressure caused by N2O
(138,146,156-158). Thus, there must be other reasons for the diversity of the results, such
as cooperation of the children, variation in the amount of effusion, interpretation of tympanograms and variations in study populations. The scale used in minitympanometry is not
wide enough to measure the actual pressure during N2O anaesthesia, which has been
reported to vary from 230 to 420 mmH 2O (160). Drake-Lee and Casey showed that widerange tympanometry predicted the findings at myringotomy even during N2O anaesthesia
(231). However, minitympanometry is of no value in detecting effusion peroperatively
when N2O is used.
The suspicion, that increased middle ear pressure forces some middle ear fluid to be
expelled through the ET is groundless. Thus, the operative surgeon can rely on the finding
at myringotomy when deciding whether to insert a ventilation tube or not. Our result
enables comparing preoperative findings of the detection of MEE by various diagnostic
tools with the result of myringotomy performed under general anaesthesia.
6.5. Hearing loss attributable to middle ear effusion
One of the limitations in studies concerning the association between OM and developmental impairment later in life has been lack of evidence of hearing loss during OM (23). We
were able to confirm by objective measurement that OM of any kind in small children and
infants causes significant hearing loss, about 30 dB, in speech frequencies in 45 % of
cases, detected by absence of TEOAEs. In addition, reduced TEOAEs were obtained in 28
% of ears with effusion. The result is well in agreement with some previous studies where
the average hearing loss has been about 30 dB in small children (179,182). Unlike these
studies, we performed myringotomy immediately after the hearing measurement in order
to confirm the diagnosis and to evaluate the quality and quantity of effusion.
Children with severe hearing loss receive inconsistent auditory signals, which results in
inaccurate encoding of the information into the database from which language develops.
Thus rehabilitation of children with profound hearing loss is indicated, at least when hearing
loss exceeds 30 dB in speech frequencies. Language acquisition and segmentation of words
are accomplished in infancy by reference to the statistical relationship between neighbouring speech sounds (203). It is logical to assume, that fluctuating hearing loss caused by OM
episodes can interfere with this statistical relationship and thus impair the learning of the
language. Moreover, as early as 8 months, which is also the time period of the highest incidence of AOM episodes, children are beginning to engage in long-term storage of words
(202).
The adverse effect of OM on later development has been related to OM episodes experienced under three years of age (177,208,209). Children under three years are susceptible
to OM and they have a much higher risk of persistence of effusion after AOM (5,7,8,64).
Furthermore, previous studies have shown that the time spent with MEE can be 20 % during the first year and 17 % during the second year and about 45 % of urban children suffer
consistent MEE for more than three months of the first year of their life (10). Combining
these facts with the results in our study, which demonstrate that OM with effusion usually
55
causes considerable hearing loss in small children, the diagnosis and treatment of OM cannot be ignored.
The magnitude of conductive hearing loss caused by MEE is dependent on both the quality and quantity of effusion, the latter being more important. This has not been previously
documented in children. In experimental works only the amount of effusion has affected the
hearing threshold (184,185). The result agrees with our knowledge of the importance of the
mobility of the TM in conducting sound waves into the inner ear. Previously, the effect of
MEE on the responses in TEOAE measurement has been examined in only few studies,
where the authors have found TEOAEs to be absent in 50 - 92 % of ears containing effusion
(198). In one study the result of measurement of TEOAE in ears with MEE has been suspected to be dependent of the type of MEE, even though the amount of effusion was not
measured (198). The great variation in the amount of effusion, 5 to 695 mg in series IV, may
explain the variation of results in TEOAE measurement in previous series. This is supported
by our finding of significant negative correlation between the amount of MEE and the
reproducibility of TEOAE measurement.
The effect of MEE on hearing during AOM has not been previously studied in small
children. We consider that non-mucoid effusion, i.e. serous effusion, better reflects the situation in AOM. It was found to impair the responses in TEOAE measurement in 52 % of
cases.Thus, repeated episodes of AOM may cause a fluctuating hearing loss of more than
20-30 dB, which agrees with the result of a study by van Buchem et al. where even one
month after AOM, an air/bone gap of more than 20 dB was detected in older children (112).
The value of minitympanometry in assisting detection of MEE was confirmed in the
series V. We found a good correlation between the type of tympanogram and the hearing
level estimated by TEOAE. There was also a significant difference in the amount of MEE in
relation to the different types of tympanogram. Our finding, that conductive hearing loss
caused by MEE is related to minitympanometric findings is supported by most of the previous studies conducted with older children (150-152). Only Fria et al. failed to support the
relationship, which is somewhat puzzling (182). Ignoring the cooperation of the children
may be an explanation for the discrepancy.
Our result showing that small amounts of MEE did not affect the TEOAE responses,
facilitates the follow-up of OME and prevents possible overtreatment. Even experienced
otoscopists do not achieve better sensitivity than about 90 %, which is also the case with
tympanometry. However, the hearing impairment is usually less than 20-30 dB in these
cases. Thus, if small amounts of effusion are missed in pneumatic otoscopy or in minitympanometric examination, there is probably no significant hearing loss. Similarly when there
is uncertainty about the possible effusion in a follow-up visit regarding OM, only re-examination is needed.
Acute otitis media is a very common problem in childhood; about every fifth child suffers from recurrent AOM, and the incidence appears to be increasing (5,15). To prevent
long-term adverse effects of OM and also to avoid overtreatment with multiple antibiotic
courses, diagnostic accuracy must be emphasized among physicians dealing with OM.
Selecting appropriate patients for interventions in the prevention of OM is also very important, but it is not easy to carry out as shown in a study by Alho et al. (232). Future studies
may give insight regarding the ongoing discussion concerning the relationship between
early-life OM and developmental impairment in children. However, there is convincing evidence of the effects of all kinds of MEE on hearing and it is reasonable to conclude that
56
recurrent fluctuating hearing loss during the sensitive period of language development cannot be an optimal condition for any child. It can interfere with language acquisition and
learning competence, which may be permanent in some children. This should guide physicians towards active diagnosis, treatment and prevention of OM and a child’s hearing and
developmental status should always be taken into account in decisions regarding middle ear
diseases.
7. Conclusions
1. Even though most AOM episodes develop during the first five days, a considerable
proportion of cases appear later than one week after the onset of URI. The variation in
temporal development of AOM is further emphasized by the finding that there was no
individual tendency in time lag of AOM.
2. Ear-related symptoms are most strongly associated with AOM, and sore throat and
fever to a much lesser extent. It appears, that various symptoms and the duration of
URI are of little value in helping the diagnosis of AOM. However, parental suspicion of
AOM in their child should not be ignored.
3. Carrying out otoscopy without assessing the mobility of the TM leads to an unacceptably low accuracy in diagnosis of OM. Impaired mobility of the TM is the most accurate mark in the prediction of MEE, but other TM signs are inaccurate.
4. Minitympanometry is a method well comparable to pneumatic otoscopy performed by
an experienced otoscopist in detecting middle ear fluid when carried out with a cooperative child. An association between amount of MEE and finding in minitympanometry
was found.
5. Nitous oxide increases middle ear pressure in general anaesthesia. However, it has no
influence on the amount of MEE. Validity tests of diagnostic devices used in the detection of MEE can be performed during general anaesthesia with N2O.
6. Both the quantity and quality of MEE have an influence on hearing thresholds measured by TEOAE. Thus, repeated episodes of OM may cause fluctuating hearing loss of
more than 20-30 dB, which can be consider to be a risk factor for development requiring normal hearing.
8. References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Bluestone CD & Klein JO (1995) Otitis media in infants and children. 2nd ed. W.B. Saunders
Company, Philadelphia.
Klein JO, Tos M, Hussl B, Naunton RF, Ohyama M & van Cauwenberge PB (1989) Recent
advances in otitis media. Definition and classification. Ann Otol Rhinol Laryngol (Suppl 139)
98: 10.
Berman S. (1997) Classification and criteria of otitis media. Clin Microbiol Infect (Suppl) 3:
1-4.
Karma P, Palva T, Kouvalainen K, Kärjä J, Mäkelä PH, Prinssi V-P, Ruuskanen O & Launiala
K (1987) Finnish approach to the treatment of acute otitis media. Report of the Finnish consensus conference. Ann Otol Rhinol Laryngol (Suppl 129) 96: 1-19.
Henderson FW, Collier AM, Sanyal MA, Watkins JM, Fairclough DL, Clyde WA Jr & Denny
FW (1982) A longitudinal study of respiratory viruses and bacteria in the etiology of acute otitis media with effusion. N Engl J Med 306: 1377-1383.
Heikkinen T, Ruuskanen O, Ziegler T, Waris M & Puhakka H (1995) Short term use of amoxicillin-clavulanate during upper respiratory tract infection for prevention of acute otitis media.
J Pediatr 126: 313-316.
Sipilä M, Pukander J & Karma P (1987) Incidence of acute otitis media up to the age of 1 1/2
years in urban infants. Acta Otolaryngol (Stockh) 104: 138-145.
Alho O-P, Koivu M, Sorri M & Rantakallio P (1991) The occurrence of acute otitis media in
infants - A life table analysis. Int J Pediatr Otorhinolaryngol 21: 7-14.
Teele DW, Klein JO, Rosner B & the Greater Boston Otitis Media Study Group (1989) Epidemiology of otitis media during the first seven years of life in children in greater Boston: A
prospective, cohort study. J Infect Dis 160: 83-94.
Paradise JL, Rockette HE, Colborn DK, Bernard BS, Smith CG, Kurs-Lasky M & Janosky JE
(1997) Otitis media in 2253 Pittsburgh-area infants: Prevalence and risk factors during the
first two years of life. Pediatrics 99: 318-333.
Harsten G, Prellner K, Heldrup J, Kalm O & Kornfält R (1989) Recurrent acute otitis media. A
prospective study of children during the first three years of life. Acta Otolaryngol (Stockh)
107: 111-119.
Ingvarsson L, Lundgren K & Stenstrom C (1990) Occurrence of acute otitis media in children:
cohort studies in urban population. In: Bluestone CD, Casselbrant ML, Scheetz MD, eds.
Workshop on epidemiology of otitis media. Ann Otol Rhinol laryngol 99(suppl 149): 17-18.
Duncan B, Ey J, Holberg CJ, Wright AL, Martinez FD & Taussig LM (1993) Exclusive
breast-feeding for at least 4 months protects against otitis media. Pediatrics 91: 867-872.
59
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
Alho O-P (1997) How common is recurrent acute otitis media? Acta Otolaryngol (Stockh)
529: 8-10.
Schappert SM (1992). Office visits for otitis media: United states, 1975-1990. From Vital and
Health Statistics of the Centers for Disease Control/National Center for Health Statistics 214:
1-9.
Lanphear BP, Byrd RS, Auinger P & Hall CB (1997) Increasing prevanlence of recurrent otitis media among children in the United States. Pediatrics 99: el.
Joki-Erkkilä VP, Laippala P, Pukander J (1998) Increase in paediatric acute otitis media diagnosed by primary care in two Finnish municipalities 1994-5 versus 1978-9. Epidemiol Infect
12: 529-534.
Niemelä M, Uhari M, Möttönen M & Pokka T Costs arising from acute otitis media. Acta
Paediatr. In press.
Bluestone CD (1996) Pathogenesis of otitis media: role of eustachian tube. Pediatr Infect Dis
J 15: 281-91.
Bluestone CD & Cantekin EI & Beery QC (1977) Effect of inflammation on the ventilatory
function of the eustachian tube. Laryngoscope 87: 493-507.
Buchman CA, Doyle WJ, Skoner DP, Post JP, Alper CM, Seroky JT, Anderson K, Preston RA,
Hayden FG, Fireman P & Ehrlich G (1995) Influenza A virus – induced acute otitis media. J
Pediatr Infect Dis 172: 1348-1351.
Buchman CA, Doyle WJ, Skoner D, Fireman P & Gwaltney JM (1994) Otologic manifestations of experimental rhinovirus infection. Laryngoscope 104: 1295-1299.
Sanyal MA, Henderson FW, Stempel EC, Collier AM & Denny FW (1980) Effect of upper
respiratory tract infection on eustachian tube ventilatory function in the preschool child. J
Pediatr 97: 11-15.
Moody SA, Alper CM & Doyle WJ (1998) Daily tympanometry in children during cold season: association of otitis media with upper respiratory tract infection. Int J Pediatr Orhinolaryngol 45: 143-150.
Rautiainen M, Nuutinen J, Kiukaanniemi H & Collan Y (1992) Ultrastructural changes in
human nasal cilia caused by the common cold and recovery of ciliated epithelium. Ann Otol
Rhinol Laryngol 101: 982-987.
Nuutinen J, Kärjä J & Karjalainen P (1983) Measurement of mucociliary function of the eustachian tube. Arch Otolaryngol 109: 669-672.
Stenström C, Bylander-Groth A & Ingvarsson L (1991) Eustachian tube function in otitisprone and healthy children. Int J Pediatr Otorhinolaryngol 21: 127-138.
Bylander A (1984) Upper respiratory tract infection and eustachian tube function in children.
Acta Otolaryngol (Stockh) 97: 343-349.
Ruuskanen O & Ogra PL (1993) Respiratory syncytial virus. Curr Probl Pediatr 23: 50-79.
Jung TTK (1988) Prostaglandins, leukotrienes, and other arachidonic acid metabolites in the
pathogenesis of otitis media. Laryngoscope 98: 980-993.
Fainstein V, Musher DM & Cate TR (1980) Bacterial adherence to pharyngeal cells during
viral infection. J Infect Dis 141: 172-176.
Selinger DS, Reed WP & McLaren LC (1981) Model of studying bacterial adherence to epithelial cells infected with viruses. Infect Immun 32: 941-944.
Håkansson A, Kidd A, Wadell G, Sabharwal H & Svanborg C (1994) Adenovirus infection
enhances in vitro adherence of Streptococcus pneumoniae. Infect Immun 62: 2707-2714.
Aniansson G, Alm B, Andersson B, Larsson P, Nylén O, Peterson H, Rignér P, Svanborg M &
Svanborg C (1992) Nasopharyngeal colonization during the first year of life. J Infect Dis
165(suppl 1): S38-S42.
60
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
Gray BM, Converse III GM & Dillon HC Jr (1980) Epidemiologic studies of Streptococcus
pneumoniae in infants: Acquisition, carriage, and infection during the first 24 months of life. J
Infect Dis 142: 923-933.
Stenfors L-E & Räisänen S (1990) Occurrence of middle ear pathogens in the nasopharynx of
young indiduals. Acta Otolaryngol 109:142-148.
Faden H, Duffy L, Wasielewski R, Wolf J, Krystofik D, Tung Y & Tonawanda/Williamsville
Paediatrics (1997) Relationship between nasopharyngeal colonization and the development of
otitis media in children. J Infect Dis 175: 1440-1445.
Faden H, Stanievich J, Brodsky L, Bernstein J & Ogra PL (1990) Changes in nasopharyngeal
flora during otitis media of childhood. Pediatr Infect Dis J 9: 623-626.
Bernstein JM, Faden HF, Dryja DM & Wactawski-Wende J (1993) Micro-etiology of the
nasopharyngeal bacterial flora in otitis-prone and non-otitis-prone children. Acta Otolaryngol
(Stockh) 113: 88-92.
Luotonen J (1982) Streptococcus pneumoniae and Haemophilus influenzae in nasal cultures
during acute otitis media. Acta Otolaryngol 93: 295-299.
Berglund B, Salmivalli A & Toivanen P (1966) Isolation of respiratory syncytial virus from
middle ear exudates of infants. Acta Otolaryngol (Stockh) 61: 475-87.
Ruuskanen O, Arola M, Putto-Laurila A, Mertsola J, Meurman O, Viljanen MK & Halonen P
(1989) Acute otitis media and respiratory virus infections. Pediatr Infect Dis J 8: 94-99.
Uhari M, Hietala J & Tuokko H (1995) Risk of acute otitis media in relation to viral etiology
of infections in children. Clin Infect Dis 20: 521-524.
Arola M, Ziegler T & Ruuskanen O (1990) Respiratory virus infection as a cause of prolonged
symptoms in acute otitis media. J Pediatr 116: 697-701.
Chonmaitree T, Owen MJ, Patel JA, Hedgpeth D, Horlick D & Howie VM (1992) Effect of
viral respiratory tract infection on outcome of acute otitis media. J Pediatr 120: 856-862.
Chonmaitree T, Howie VM & Truant AL (1986) Presence of respiratory viruses in middle ear
fluids and nasal wash specimens from children with acute otitis media. Pediatrics 77: 698-702.
Klein BS, Dollete FR & Yolken RH (1982) The role of respiratory syncytial virus and other
viral pathogens in acute otitis media. J Pediatr 101: 16-20.
Sarkkinen H, Ruuskanen O, Meurman O, Puhakka H, Virolainen E & Eskola J (1985) Identification of respiratory virus antigens in middle ear fluids of children with acute otitis media. J
Infect Dis 151: 444-448.
Heikkinen T, Thint M & Chonmaitree T (1999) Prevalence of various respiratory viruses in
the middle ear during acute otitis media. N Engl J Med 340: 260-4.
Pitkäranta A, Virolainen A, Jero J, Arruda E & Hayden FG (1998) Detection of rhinovirus,
respiratory syncytial virus, and coronavirus infections in acute otitis media by reverse transcriptase polymerase chain reaction. Pediatrics 102: 291-295.
Chonmaitree T, Owen MJ & Howie VM (1990) Respiratory viruses interfere with bacteriologic response to antibiotic in children with acute otitis media. J Infect Dis 162: 546-549.
Sung BS, Chonmaitree T, Broemeling LD, Owen MJ, Patel JA, Hedgpeth DC & Howie VM
(1993) Association of rhinovirus infection with poor bacteriologic outcome of bacterial-viral
otitis media. Clin Infect Dis 17: 38-42
Chonmaitree T, Patel JA, Lett-Brown MA, Uchida T, Garofalo R, Owen MJ & Howie VM
(1994) Virus and bacteria enhance histamine production in middle ear fluids of children with
acute otitis media. J Infect Dis 169: 1265-1270.
Luotonen J, Herva E, Karma P, Timonen M, Leinonen M & Mäkelä PH (1981) The bacteriology of acute otitis media in children with special reference to Steptococcus pneumoniae as
studied by bacteriological and antigen detection methods. Scand J Infect Dis 13: 177-183.
Del Beccaro MA, Mendelman PM, Inglis AF, Richardson MA, Duncan NO, Clausen CR &
Stull TL (1992) Bacteriology of acute otitis media: A new perspective. J Pediatr 120: 81-84.
61
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
Post JC, Preston RA, Aul JJ, Larkins-Pettigrew M, Rydquist-White J, Anderson KW, Wadowsky RM, Reagan DR, Walker ES, Kingsley LA, Magit AE & Ehrlich GD (1995). Molecular analysis of bacterial pathogens in otitis media with effusion. JAMA 273: 1598-1604.
Virolainen A, Salo P, Jero J, Karma P, Eskola J & Leinonen M (1994) Comparison of PCR
assay with bacterial culture for detecting Streptococcus pneumoniae in middle ear fluid of
children with acute otitis media. J Clin Microbiol 32: 2667-2670.
Hendolin PH, Markkanen A, Ylikoski J & Wahlfors JJ (1997) Use of multiplex PCR for
simultaneous detection of four bacterial species in middle ear effusions. J Clin Microbiol 35:
2854-2858.
Post JC, Aul JJ, White GJ, Wadowsky RM, Zavoral T, Tabari R, Kerber B, Doyle WJ & Ehrlich GD (1996) PCR-based detection of bacterial DNA after antimicrobial treatment is indicative of persistent, viable bacteria in the chinchilla model of otitis media. Am J Otolaryngol 17:
106-111.
Bluestone CD, Stephenson JS & Martin LM (1992) Ten-year review of otitis media pathogens. Pediatr Infect Dis J (Suppl) 11: 7-11.
Arguedas A, Loaiza C, Perez A, Vargas F, Herrera M, Rodriguez G, Gutierrez A & Mohs E
(1998) Microbiology of acute otitis media in Costa Rica children. Pediatr Infect Dis J 17: 680689.
Bosley GS, Whitney AM, Pruckler JM, Moss CW, Daneshvar M, Sih T & Talkington DF
(1995) Characterization of ear isolates of Alloicoccus otitidis from patients with recurrent otitis media. J Clin Microbiol 33: 2876-2880.
Kværner KJ, Nafstad P, Hagen JA, Mair IWS & Jaakkola JJK (1997) Recurrent acute otitis
media: The significance of age at onset. Acta Otolaryngol (Stockh) 117: 578-584.
Shurin PA, Pelton SI, Donner A & Klein JO (1979) Persistence of middle-ear effusion after
acute otitis media in children. N Engl J Med 300: 1121-1123.
Owen MJ, Baldwin CD, Swank PR, Pannu AK, Johnson DL & Howie VM (1993) Relation of
infant feeding practices, cigarette smoke exposure, and group child care to the onset and duration of otitis media with effusion in the first two years of life. J Pediatr 123: 702-711.
Pukander J, Luotonen J, Timonen M & Karma P (1985) Risk factors affecting the occurrence
of acute otitis media among 2-3-year-old urban children. Acta Otolaryngol (Stockh) 100: 260265.
Tainio V-M, Savilahti E, Salmenperä L, Arjomaa P, Siimes MA & Perheentupa J (1988) Risk
factors for infantile recurrent otitis media: Atopy but not type of feeding. Pediatr Res 23: 509512.
Sipilä M, Karma P, Pukander J, Timonen M & Kataja M (1988) The Bayesian approach to the
evaluation of risk factors in acute and recurrent acute otitis media. Acta Otolaryngol (Stockh)
106: 94-101.
Alho O-P, Koivu M, Sorri M & Rantakallio P (1990) Risk factors for recurrent acute otitis
media and respiratory infection in infancy. Int J Pediatr Otorhinolaryngol 19: 151-161.
Ståhlberg M-R, Ruuskanen O, Virolainen E (1986) Risk factors for recurrent otitis media.
Pediatr Infect Dis 5: 30-32.
Collet JP, Ducruet T, Floret D, Cogan-Collet J, Honneger D & Boissel J (1991) Daycare attendance and a risk of first infectious disease. Eur J Pediatr 150: 214-216.
Wald ER, Guerra N & Byers C (1991) Frequency and severity of infections in day-care:
Three-years follow-up. J Pediatr 118: 509-514.
Uhari M, Mäntysaari K & Niemelä M (1996) A meta-analytic review of the risk factors for
acute otitis media. Clin Infect Dis 22: 1079-1083.
Alho OP, Kilkku O, Oja H, Koivu M & Sorri M (1993) Control of the temporal aspect when
considering risk factors for acute otitis media. Arch Otolaryngol Head Neck Surg 119: 444449.
62
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
Postma DS, Poole MD, Wu S & Tober R (1997) The impact of day-care on ventilation tube
insertion. Pediatr Otorhinolaryngol 41: 253-262.
Niemelä M, Uhari M, Luotonen M, Luotonen J, Manninen M-P & Puhakka H (1998) Changes
in day-care attendance rates and in the occurrence of adenoidectomies and tympanostomies.
Acta Paediatr 87: 1003-1004.
Uhari M & Möttönen M An open randomized controlled trial of infection prevention in child
day-care centers. Submitted.
Daly K, Giebink GS, Le CT, Lindgren B, Batalden PB, Anderson RS & Russ JN (1988)
Determining risk for chronic otitis media with effusion. Pediatr Infect Dis J 7: 471-475.
Kero P & Piekkala P (1987) Factors affecting the occurrence of acute otitis media during the
first year of life. Acta Paediatr Scand 76: 618-623.
Kværner KJ, Nafstad P, Hagen JA, Mair IWS & Jaakkola JJK (1996) Early acute otitis media
and siblings’ attendance at nursery. Arch Dis Child 75: 338-341.
Saarinen UM (1982) Prolonged breast feeding as prophylaxis for for recurrent otitis media.
Acta Pediatr Scand 71: 567-571.
Aniansson G, Alm B, Andersson B, Håkansson A, Larsson P, Nylén O, Peterson H, Rignér P,
Svanborg M, Sabharwal H & Svanborg C (1994) A prospective cohort study on breast-feeding
and otitis media in Swedish infants. Pediatr Infect Dis J 13: 183-188.
Paradise JL & Elster BA (1984) Breast milk protects against otitis media with effusion. Pediatr Res 18: 283A.
Tager IB, Weiss ST, Muñoz A, Rosner B & Speizer FE (1983) Longitudinal study of the
effects of maternal smoking on pulmonary function in children. N Engl J Med 309: 699-703.
Pedreira FA, Guandolo VL, Feroli EJ, Mella GW & Weiss IP (1985) Involuntary smoking and
incidence of respiratory illness during the first year of life. Pediatrics 75: 594-597.
DiFranza JR & Lew RA (1996) Morbidity and mortality in children associated with the use of
tobacco products by other people. Pediatrics 97: 560-568.
Etzel RA, Pattishall EN, Haley NJ, Fletcher RH & Henderson FW (1992) Passive smoking
and middle ear effusion among children in day-care. Pediatrics 90: 228-232.
Maw AR, Parker AJ, Lance GN & Dilkes MG (1992) The effect of parental smoking on outcome after treatment for glue ear in children. Clin Otolaryngol 17: 411-414.
Kraemer MJ, Richardson MA, Weiss NS, Furukawa CT, Shapiro GG, Pierson WE & Bierman
CW (1983) Risk factors for persistent middle-ear effusions. JAMA 249: 1022-1025.
Stenström R, Bernard PAM & Ben-Simhon H (1993) Exposure to environmental tobacco
smoke as a risk factor for recurrent acute otitis media in children under the age of five years.
Int J Pediatr Otorhinolaryngol 27: 127-136.
Agius AM, Wake M, Pahor AL & Smallman LA (1995) Smoking and middle ear ciliary beat
frequency in otitis media with effusion. Acta Otolaryngol (Stockh) 115: 44-49.
Ey JL, Holberg CJ, Aldous MB, Wright AL, Martinez FD, Taussig LM & Group health medical associates (1995) Passive smoke exposure and otitis media in the first year of life. Pediatrics 95: 670-677.
Kværner KJ, Tambs K, Harris JR & Magnus P (1997) Distribution and heritability of recurrent
ear infections. Ann Otol Rhinol Laryngol 106: 624-632.
Spivey GH & Hirschhorn N (1977) A migrant study of adopted Apache children. Johns Hopkins Med J 140: 43-46.
Prellner K, Kalm O, Hartsen G Heldrup J & Oxelius V-A (1989) Pneumococcal serum antibody concentrations during the first three years of life: A study of otitis-prone and non-otitisprone children. Int J Pediatr Otorhinolaryngol 17: 267-269.
Niemelä M, Uhari M & Hannuksela A (1994) Pacifiers and dental structure as risk factors for
otitis media. Int J Pediatr Otorhinolaryngol 29: 121-127.
63
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
Niemelä M, Uhari M, lautala P & Huggare J (1994) Association of recurrent acute otitis
media with nasopharynx dimensions in children. J Laryngol Otol 108: 299-302.
Gannon MM, Haggard MP, Golding J & Fleming P. (1995) Sleeping position – a new environmental risk factor for otitis media? In: Lim DJ ed. Abstracts of the Sixth international symposium on Recent Advances in otitis media, Ft Lauderdale, Fla : 24.
Froom J, Culpepper L, Grob P, Bartelds A, Bowers P, Bridges-Webb C, Grava-Gubins I,
Green L, Lion J, Somaini B, Stroobant A, West R & Yodfat Y (1990) Diagnosis and antibiotic
treatment of acute otitis media: report from International Primary Care Network. Br Med J
300: 582- 586.
Uhari M, Niemelä M & Hietala J (1995) Prediction of acute otitis media with symptoms and
signs. Acta Paediatr 84: 90-92.
Niemelä M, Uhari M, Jounio-Ervasti K, Luotonen J, Alho O-P & Vierimaa E (1994) Lack of
specific symptomalogy in children with acute otitis media. Pediatr Infect Dis J 13: 765-768.
Heikkinen T & Ruuskanen O (1995) Signs and symptoms predicting acute otitis media. Arch
Pediatr Adolesc Med 149: 26-29.
Pukander J (1983) Clinical features of acute otitis media among children. Acta Otolaryngol
95: 117-122.
Hayden GF & Schwartz RH (1985) Characteristics of earache among children with acute otitis media. Am J Dis Child 139: 721-723.
Berman S (1995) Current concepts: otitis media in children. N Engl J Med 332: 1560-1565.
Heikkinen T & Ruuskanen O (1994) Temporal development of acute otitis media during
upper respiratory tract infection. Pediatr Infect Dis J 13: 659-661.
Heikkinen T (1994) Development and prevention of acute otitis media in children. Ann Univ
Turku D 158: 50-51.
Arola M, Ruuskanen O, Ziegler T, Mertsola J, Näntö-Salonen K, Putto-Laurila A, Viljanen
MK & Halonen P (1990) Clinical role of respiratory virus infection in acute otitis media. Pediatrics 86: 848-855.
Rudberg RD (1954) Acute otitis media: comparative therapeutic results of sulphonamide and
penicillin administered in various forms. Acta Otolaryngol (Suppl) 113: 9-79.
Alho OP, Jokinen K, Laitakari K & Palokangas J (1997) Chronic suppurative otitis media and
cholesteatoma. Vanishing diseases among Western populations? Clin Otolaryngol 22: 358361.
Froom J, Culpepper L, Jacobs M, De Melker RA, Green LA, van Buchem L, Grob P &
Heeren T (1997) Antimicrobials for acute otitis media? A review from the International Primary Care Network. Br Med J 315: 98-102.
van Buchem FL, Dunk JHM & van´t Hof MA (1981) Therapy of acute otitis media: myringotomy, antibiotics, or neither? A double-blind study in children. Lancet 2: 883-887.
van Buchem FL, Peeters MF & van´t Hof MA (1985) Acute otitis media: a new treatment
strategy. Br Med J 290: 1033-1037.
Appelman CL, Claessen JQ, Touw-Otten FW, Hordijk GJ & DeMelker RA (1991) Co-amoxiclav in recurrent acute otitis media: placebo controlled study. Br Med J 303: 1450-1452.
Del Mar C, Glasziou P & Hayem M (1997) Are antibiotics indicated as initial treatment for
children with acute otitis media? A meta-analysis. Br Med J 314: 1526-1529.
Rosenfeld RM, Vertrees JE, Carr J, Cipolle RJ, Uden DL, Giebink GS & Canafax DM (1994)
Clinical efficacy of antimicrobial drugs for acute otitis media: meta-analysis of 5400 children
from thirty-three randomized trials. J Pediatr 124: 355-367.
Kaleida PH, Casselbrant ML, Rockette HE, Paradise JL, Bluestone CD, Blatter MM, Resinger
KS, Wald ER & Supance JS (1991) Amoxicillin or myringotomy or both for acute otitis
media: result of a randomized clinical trial. Pediatrics 87: 466-474.
64
118. Marchant CD, Carlin SA, Johnson CE & Shurin PA (1992) Measuring the comparative efficacy of antibacterial agents for acute otitis media: the “Pollyanna phenomenon” J Pediatr 120:
72-77.
119. Pichichero ME (1997) Assessing the treatment alternatives for acute otitis media. Pediatr
Infect Dis J 13: S27-S34.
120. Harsten G, Prellner K, Heldrup J, Kalm O & Kornfält R (1989) Treatment failure in acute otitis media. A clinical study of children during their first three years of life. Acta Otolaryngol
(Stockh) 108: 253-258.
121. Berman S (1995) Management of acute and chronic otitis media in pediatric practice. Curr
Opinion Pediatr 7: 513-522.
122. Bluestone CD (1989) Modern managament of otitis media. Pediatr Clin North Am 36: 13711385.
123. Huovinen P & Vaara M (1995) Avohoidon mikrobilääkkeet. Kapseli 24: 27-28.
124. Canafax DM & Giebink GS (1994) Antimicrobial treatment of acute otitis media. Ann Otol
Rhinol Laryngol 103: 11-14.
125. Kozyrskyj AL, Hildes-Ripstein GE, Longstaffe SEA, Wincott JL, Sitar DS, Klassen TP &
Moffatt MEK (1998) Treatment of acute otitis media with a shortened course of antibiotics.
JAMA 279: 1736-1742.
126. Cohen R, Levy C, Boucherat M, Langue J & de La Rocque F (1998) A multicenter, randomized, double-bind trial of 5 versus 10 days antibiotic therapy for acute otitis media in young
children. J Pediatr 133: 634-639.
127. Puhakka H, Virolainen E, Aantaa E, Tuohimaa P, Eskola J & Ruuskanen O (1979) Myringotomy in the treatment of acute otitis media in children. Acta Otolaryngol 88: 122-126.
128. Qvarnberg Y (1981) Acute otitis media. A prospective clinical study of myringotomy and
antimicrobial treatment. Acta Otolarynogol 375 (suppl)
129. Burke P, Bain J, Robinson D & Dunleavey J (1991) Acute red ear in children: controlled trial
of non-antibiotic treatment in general practice. Br Med J 303: 558-562.
130. Klein JO (1998) Clinical implications of antibiotic resistance for management of acute otitis
media. Pediatr Infect Dis J 17: 1084-1089.
131. Bluestone CD (1998) Role of surgery for otitis media in the era of resistant bacteria. J Pediatr
Infect Dis J 17: 1090-1098.
132. Schnore SK, Sangster JF, Gerace TM & Bass MJ (1986) Are antihistamine-decongestants of
value in the treatment of acute otitis media in children? J Fam Pract 22: 39-43.
133. Turner RB & Darden PM (1996) Effect of topical adrenergic decongestants on middle ear
pressure in infants with common colds. Pediatr Infect Dis J 15: 621-624.
134. Tejani NR, Chonmaitree T, Rassin DK, Howie VM, Owen MJ & Goldman AS (1995) Use of
C-reactive protein in differentation between acute bacterial and viral otitis media. Pediatrics
95: 664-669.
135. Heikkinen T, Ghaffar F, Okorodudu AO & Chonmaitree T (1998) Serum interleukin-6 in bacterial and nonbacterial acute otitis media. Pediatrics 102: 296-299.
136. Hayden GF (1981) Acute suppurative otitis media in children. Diversity of clinical diagnostic
criteria. Clin Pediatr 2: 99-104.
137. Karma PH, Penttilä MA, Sipilä MM & Kataja MJ (1989) Otoscopic diagnosis of middle ear
effusion in acute and non-acute otitis media. I. The value of different otoscopic findings. Int J
Pediatr Otorhinolaryngol 17: 37-49.
138. Vaughan-Jones R & Mills RP (1992) The Welch Allyn Audioscope and Microtymp: their
accuracy and that of pneumatic otoscopy, tympanometry and pure tone audiometry as predictors of otitis media with fluid. J Laryngol Otol 106: 600-602.
139. Finitzo T, Friel-Patti S, Chinn K & Brown O (1992) Tympanometry and otoscopy prior to
myringotomy: issues in diagnosis of otitis media. Int J Pediatr Otorhinolaryngol 24: 100-110.
65
140. Toner JG & Mains B (1990) Pneumatic otoscopy and tympanometry in the detection of middle ear effusion. Clin Otolaryngol 15: 121-123.
141. Cantekin EI, Bluestone CD, Fria TJ, Stool SE, Beery QC & Sabo DL (1980) Identification of
otitis media with effusion in children. Ann Otol Rhinol Laryngol (suppl) 89: 190-195.
142. Shanks JS & Shelton C (1991) Basic principles and clinical applications of tympanometry.
Otolaryngol Clin North Am 24: 299-328.
143. Margolis RH & Heller JW (1987) Screening tympanometry: Criteria for medical referral.
Audiology 26: 197-208.
144. Jerger J (1970) Clinical experience with impedance audiometry. Arch Otolaryngol 92: 311324.
145. Sassen ML, van Aarem A & Grote JJ (1994) Validity of tympanometry in the diagnosis of
middle ear fluid. Clin Otolaryngol 19: 185-189.
146. Szucs E, Diependale R & Clement PAR (1995) The accuracy of tympanometry assessed by its
sensitivity and specificity. Acta Oto-rhino-laryngol Belg 49: 287-292.
147. Watters GWR, Jones JE & Freeland AP (1997) The predicitive value of tympanometry in the
diagnosis of middle ear effusion. Clin Otolaryngol 22: 343-345.
148. Orchik DJ, Dunn JW & McNutt L (1978) Tympanometry as a predictor of middle ear fluid.
Arch Otolaryngol 104: 4-6.
149. Nozza RJ, Bluestone CD, Kardatzke D & Bachman R (1992) Towards the validation of aural
acoustic immittance measures for diagnosis of middle ear effusion in children. Ear Hear 13:
442-447.
150. Dempster JH & Mackenzie KM (1991) Tympanometry in detection of hearing impairments
associated with otitis media with fluid. Clin Otolaryngol 16: 157-159.
151. Kazanas SG & Maw R (1994) Tympanometry, stapedius reflex and hearing impairment in
children with otitis media with effusion. Acta Otolaryngol (Stockh) 114: 410-414.
152. Haapaniemi JJ (1997) Pure-tone audiometric and impedance measurements in school- aged
children in Finland. Eur Arch Otorhinolaryngol 254: 269-273.
153. van Balen FAM & de Melker RA (1994) Validation of a portable tympanometer for use in primary care. Int J Pediatr Otorhinolaryngol 29: 219-25.
154. Puhakka HJ (1991) Pieni tympanometri parantaa välikorvantulehduksen diagnostiikkaa
avohoidossa. Suom Lääkäril 29: 2708-2711.
155. De Melker RA (1992) Diagnostic value of microtympanometry in primary care. Br Med J
304: 96-98.
156. Haughton PM (1977) Validity of tympanometry for middle ear effusions. Arch Otolaryngol
103: 505-513.
157. Gersdorff MCH (1992) Diagnostic value of tympanometry in otitis media with effusion. Acta
Oto-rhino-laryngol Belg 46: 361-368.
158. Johnson LP, Parkin JL, Stevens MH, Otto WC & McCandless GA (1980) Action of general
anesthesia on middle ear effusions. Arch Otolaryngol 106: 100-102.
159. Tom LWC, Tsao F, Marsh RR, Kessler A & Konkle DF (1994) Effect of anesthetic gas on
middle ear effusion. Laryngoscope 104: 832-836.
160. Davis I, Moore JRM & Lahiri SK (1979) Nitrous oxide and the middle ear. Anesthesia 34:
147-151.
161. Thomsen KA, Terkildsen K & Arnfred I (1965) Middle ear pressure variations during anesthesia. Arch Otolaryngol 82: 609-611.
162. Gates GA & Cooper JC (1980) Effect of anesthetic gases on middle ear pressure in the presence of effusion. Ann Otol Rhinol Laryngol (suppl) 89: 62-64.
163. Levy D, Herman M, Luntz M & Sade J (1995) Direct demonstration of gas diffusion into the
middle ear. Acta Otolaryngol 115: 276-278.
66
164. Ostfeld E, Crispin M, Blonder J & Szeinberg A (1980) Middle ear gas composition during
nitrous oxide-oxygen ventilation. Ann Otol Rhinol Laryngol 89: 165-167.
165. House HP (1946) Acute otitis media. a comparative study of the results obtained in therapy
before and after the introduction of the sulfonamide compounds. Arch Otolaryngol Head
Neck Surg 43: 371-378.
166. Harley EH, Sdralis T & Berkowitz RG (1997) Acute mastoiditis in children: A 12-year retrospective study. Otolaryngol Head Neck Surg 116: 26-30.
167. Gliklich RE, Eavey RE, Ianuzzi RA & Camacho AE (1996) A contemporary analysis of acute
mastoiditis. Arch Otolaryngol Head Neck Surg 122: 135-139.
168. Berman S (1995) Otitis media in developing countries. Pediatrics 96: 126-131.
169. Ellefsen B & Bonding P (1996) Facial palsy in acute otitis media. Clin Otolaryngol 21: 39339
170. Hoppe JE, Köster S, Bootz F & Niethammer D (1994) Acute mastoiditis – relevant once
again. Infection 22: 178-182.
171. Tos M & Poulsen G (1980) Screening tympanometry in infants and two-year-old children.
Ann Otol Rhinol Laryngol 89: 217-222.
172. Fiellau-Nikolajsen M (1983) Epidemiology of secretory otitis media, a descriptive cohort
study. Ann Otol Rhinol Laryngol 92: 172-177.
173. Zielhuis GA, Rach GH & van den Broek P (1989) Screening for otitis media with effusion in
preschool children. Lancet 11: 311-313.
174. Teele DW, Klein JO & Rosner BA (1980) Epidemiology of otitis media in children. Ann Otol
Rhinol Laryngol (Suppl) 89: 5-6.
175. Alho O-P, Oja H, Koivu M & Sorri M (1995) Risk factors for chronic otitis media with effusion in infancy. Each acute otitis media episode induces a high but transient risk. Arch Otolaryngol Head Neck Surg 121: 839-843.
176. Hogan SC, Stratford KJ & Moore DR (1997) Duration and recurrence of otitis media with
effusion in children from birth to 3 years: prospective study using monthly otoscopy and tympanometry. Br Med J 314: 350-355.
177. Teele DW, Klein JO, Chase C, Menyuk P, Rosner BA & the Greater Greater Boston Otitis
Media Study Group (1990) Otitis media in infancy and intellectual ability, school achievement, speech, and language at age 7 years. J Infect Dis 162: 685-694.
178. Friel-Patti S & Finitzo T (1990) Language learning in a prospective study of otitis media with
effusion in the first two years of life. J Speech Hear Res 33: 188-194.
179. Roland PS, Finitzo T, Friel-patti S, Brown KC, Stephen KT, Brown O & Coleman M (1989)
Otitis media: incidence, duration and hearing status. Arch Otolaryngol Head Neck Surg 115:
1049-1053.
180. Olmstead RW, Alvarez MC, Moroney JD & Eversden M (1964) The pattern of hearing following acute otitis media. J Pediatr 65: 252-255.
181. Kokko E (1974) Chronic secretory otitis media in children. Acta Otolaryngol (Suppl) 327: 144.
182. Fria TJ, Cantekin EI & Eichler JA (1985) Hearing acuity of children with otitis media with
effusion. Arch Otolaryngol 111: 10-16.
183. Cohen D, Sade J (1972) Hearing in secretory otitis media. Can J Otolaryngol 1: 27-29.
184. Wiederhold ML, Zajtchuk JT, Vap JG & Paggi RE (1980) Hearing loss in relation to physical
properties of middle ear effusions. Ann Otol Rhinol Laryngol (Suppl) 89: 185-189.
185. Marsh RR, Baranak CC & Potsic WP (1985) Hearing loss and visco-elasticity of middle ear
fluid. Int J Pediatr Otorhinolaryngol 9: 115-120.
186. Hunter LL, Margolis RH & Giebink GS (1994) Identification of hearing loss in children with
otitis media. Ann Otol Rhinol Laryngol 103: 59-61.
67
187. Kemp DT (1978) Stimulated acoustic emissions from within the human auditory system. J
Acoust Soc Am 64: 1386-1391.
188. Robinette MS & Glattke (1997) Otoacoustic emissions:clinical applications. Thieme, New
York
189. Probst R & Harris FP (1993) Transiently evoked and distortion-product otoacoustic emissions. Arch Otolaryngol Head Neck Surg 119: 858-860.
190. Kemp DT, Ryan S & Bray P (1990) A guide to effective use of otoacoustic emissions. Ear
Hear 11: 93-105.
191. Marshall L, Heller LM & Westhusin LJ (1997) Effect of negative middle-ear pressure on transient evoked otoacoustic emissions. Ear Hear 18: 218-226.
192. Trine MB, Hirsch JE & Margolis RH (1993) The effect of middle ear pressure on transient
evoked otoacoustic emissions. Ear Hear 14: 401-407.
193. Plinkert PK, Bootz F & Voßieck T (1994) Influence of static middle ear pressure on transiently evoked otoacoustic emissiond and distortion products. Eur Arch Otorhinolaryngol
251: 95-99.
194. Prieve BA, Gorga MP, Schmidt GA, Neely S, Peters J Schultes L & Jesteadt W (1993) Analysis of transient-evoked otoacoustic emissions in normal-hearing and hearing-impaired ears. J
Acoust Soc Am 93: 3308-3319.
195. Hall JW, Baer JE, Chase PE & Schwaber MK (1994) Clinical application of otoacoustic emissions: What do we know about factors influencing measurement and analysis. Otolaryngol
Head Neck Surg 110: 22-38.
196. White KR, Vohr BR, Maxon AB, Behrens TR McPherson MG & Mauk GW (1994) screening
all newborns for hearing loss using transient evoked otoacoustic emissions 29: 203-217.
197. Owens JJ, McCoy MJ, Londsbury-Martin BL & Martin GK (1993) Otoacoustic emissions in
children with normal ears, middle ear dysfunction, and ventilating tubes. Am J Otol 14: 34-39.
198. Amedee RG (1995) The effects of chronic otitis media with effusion on the measurement of
transiently evoked otoacoustic emissions. Laryngoscope 105: 589-595.
199. Beppu R, Hattori T & Yanagita N (1997) Comparison of TEOAE with play audiometry for
screening hearing problems in children. Auris Nasus Larynx 24: 367-371.
200. Nozza RJ, Sabo DL & Mandel EM (1997) A role for otoacoustic emissions in screening for
hearing impairment and middle ear disorders in school-age children. Ear Hear 18: 227-239.
201. Kuhl PK, Williams KA, Lacerda F, Stevens KN & Lindblom B (1992) Linguistic experience
alters phonetic perception in infants by 6 months of age. Science 255: 606-608.
202. Jusczyk PW & Hohne EA (1997) Infants' memory for spoken words. Science 277: 1984-1986.
203. Saffran JR, Aslin RN & Newport EL (1996) Statistical learning by 8-month-old infants. Science 274: 1926-1928.
204. Ruben RJ (1997) A time framee of critical/sensitive periods of language development. Acta
Otolaryngol (Stockh) 117: 202-215.
205. Friel-Patti S, Finitzo-Hieber T, Conti G & Brown KC (1982) Language delay in infants associated with middle ear disease and mild, fluctuating hearing impairment. Pediatr Infect Dis 1:
104-109.
206. Holm VA & Kunze LH (1969) Effect of chronic otitis media on language and speech development. Pediatrics 43: 833-839.
207. Rach GH, Zielhuis GA & van den Broek P (1988) The influence of chronic persistent otitis
media with effusion on language development of 2- to 4-year-olds. Int J Pediatr Otorhinolaryngol 15: 253-261.
208. Luotonen M, Uhari M, Aitola L, Lukkaroinen A-M, Luotonen J, Uhari M & Korkeamäki R-L
(1996) Recurrent otitis media during infancy and linguistic skills at the age of nine years.
Pediatr Infect Dis J 15: 854-858.
68
209. Luotonen M, Uhari M, Aitola L, Lukkaroinen A-M, Luotonen J & Uhari M (1998) A nation
wide, population-based survey of otitis media and school achievement. Int J Pediatr Otorhinolaryngol 43: 41-51.
210. Sak RJ & Ruben RJ (1981) Recurrent middle ear effusion in childhood: Implications of temporary auditory deprivation for language and learning. Ann Otol 90: 546-551.
211. Abraham SS, Wallace IF & Gravel JS (1996) Early otitis media and phonological development at age 2 years. Laryngoscope 106: 727-732.
212. Haggard MP, Birkin JA, Browning GG, Gatehouse S & Lewis S (1994) Behavioral problems
in otitis media. Pediatr Infect Dis J 13: S43-S50.
213. Chase C (1992) Hearing loss and development: A neuropsychologic perspective. In: Eavey
RD & Klein JO (eds) Hearing loss in childhood: A primer report of the 102nd Ross conference on Paediatric Research: 88-93. Columbus, Ohio.
214. Roberts JE, Burchinal MR, Zeisel SA, Neebe EC, Hooper SR, Roush J, Bryant D, Mundy M,
& Henderson FW (1998) Otitis media, the caregiving environment, and language and cognitive outcomes at 2 years. Pediatrics 102: 346-354.
215. Roberts JE, Sanyal MA, Burchinal MR, Collier AM, Ramey CT & Henderson FW (1986) Otitis media in early childhood and its relationship to later verbal and academic performance.
Pediatrics 78: 423-430.
216. Roberts JE, Burchinal MR, Collier AM, Ramey CT, Koch MA & Henderson FW (1989) Otitis
media in early childhood and cognitive, academic, and classroom performance of the schoolaged child. Pediatrics 83: 477-485.
217. Lous J & Fiellau-Nikolajsen M (1984) A 5-year prospective case-control study of the influence of early otitis media with effusion on reading achievement. Int J Pediatr Otorhinolaryngol 8: 19-30.
218. Hubbard TW, Paradise JL, McWilliams BJ, Elster BA & Taylor FH (1985) Consequences of
unremitting middle-ear disease in early life. Otologic, audiologic, and developmental findings
in children with cleft palate. N Engl J Med 312: 1529-1534.
219. Harsten G, Nettelbladt U, Schalen L, Kalm O & Prellner K (1993) Language development in
children with recurrent acute otitis media during the first three years of life. Follow-up study
from birth to seven years of age. J Laryngol Otol 107: 407-412.
220. Roberts JE, Burchinal MR, Davis BP, Collier AM & Henderson FW (1991) Otitis media in
early childhood and later language. J Speech Hear Res 34: 1158-1168.
221. Hunter LL, Margolis RH, Rykken JR, Le CT, Daly KA & Giebink GS (1996) High frequency
hearing loss associated with otitis media. Ear Hearing 17: 1-11.
222. Sorri M, Mäki-Torkko E & Alho O-P (1995) Otitis media and long term follow-up of hearing.
Acta Otolaryngol (Stockh) 115: 193-195.
223. Löppönen H, Sorri M, Pekkala R & Penna J (1992) Secretory otitis media and high-frequency
hearing loss. Acta Otolaryngol (Stockh) 493: 99-107.
224. Rahko T, Karma P & Sipilä M (1989) Sensorineural hearing loss and acute otitis media in
children. Acta Otolaryngol (Stockh) 108: 107-112.
225. Paradise JL, Haggard MP, Lous J, Roberts JE & Schilder AGM (1995) Developmental implications of early life otitis media. Int J Pediatr Otorhinolaryngol 32 (suppl): S37-S44.
226. Bodor FF (1982) Conjunctivitis-otitis syndrome. Pediatrics 69: 695-698.
227. Bodor FF, Marchant CD & Shurin PA (1985) Bacterial etiology of conjunctivitis-otitis syndrome. Pediatrics 76: 26-28.
228. Ingvarsson L (1982) Acute otalgia in children – findings and diagnosis. Acta Paediatr Scand
71: 705-710.
229. Faden H, Duffy L & Boeve M (1998) Otitis media: back to basics. Pediatr Infect Dis J 17:
1105-1113.
69
230. Kennedy TL & Gore LB (1982) Middle ear effusions and the nitrous oxide myth. Laryngoscope 92: 169-172.
231. Drake-Lee AB & Casey WF (1983) Anaesthesia and tympanometry. Int J Pediatr Otorhinolaryngol 6: 171-178.
232. Alho O-P, Koivu M, Sorri M, Oja H & Kilkku O (1994) Which children are being operated on
for recurrent acute otitis media? Arch Otolaryngol Head Neck Surg 120: 807-811.