Download Case study: Investigation of an outbreak of

Case study
An outbreak with respiratory
symptoms
Toronto, Canada
This case study was developed by a working group led by WHO EPR. It is based on the
investigation of an outbreak of acute respiratory symptoms in 2003 in Toronto, Canada.
The data have been modified for the purposes of the case study.
Respiratory Outbreak, Toronto Version May 2007
1
Using this case study in the classroom:
We recommend that this case study be used in conjunction with the “Laboratory Skills for
Epidemiologists” module, developed by the WHO. However, it can be delivered on its own,
provided key lectures are presented first.
Recommended prerequisite lectures (cf. training matrix):
Lecture 6: Principles of immunology
Lecture 7: Antibody and antigen detection
Lecture 10: Viral culture
Lecture 3: Taking appropriate and adequate samples safely
Lecture 4:Transport, disinfection and biosafety
Lecture 13: Sequencing and phylogeny (if Part 10 is to be done)
Time required for this case study:
3hours for each session
This case study does not come with a facilitator’s guide. The answers to all the questions
for each section are provided as an introduction to the following section.
To run this case study in the classroom, we propose that it be distributed one page at a
time. Participants should take turns reading it aloud, paragraph by paragraph. Reading all
paragraphs aloud and in turns has two advantages: first, everyone can quickly participate
and get beyond the inhibition of having her/his voice heard in a large room; second, time is
given to the whole class to understand the issue and think about the answers. The
participants reading the question may try to answer it if s/he can; otherwise, it can be
discussed as a group. The next participant reads the next question and so on until the end
of the page. After the next part/page is distributed, the next participant continues and so
on until the case study is over. Once the epilogue is read, re-visit the objectives – this
reinforces the learning and provides an opportunity to clarify any remaining issues.
Learning objectives
1. Understand the importance of laboratory data for surveillance of respiratory pathogens
and outbreak detection;
2. Understand the limitations of laboratory data in surveillance and outbreak detection;
3. Devise an optimal sampling strategy for the type and the number of samples to take
during an outbreak of respiratory illness;
4. Follow ethical guidelines and consider requirements for informed consent in patient
sampling;
5. Use personal protective measures to prevent infections when taking samples and
preparing them for transport;
6. Transport samples according to national and/or international regulations, as
applicable;
7. Identify the laboratory techniques that may be used to test specimens collected during
a respiratory outbreak;
8. Understand limitations of results of PCR and serology testing;
9. Interpret phylogenetic trees to understand the spread of disease; origin and
relatedness of the pathogen;
10. Interpret laboratory data in context of epidemiological data in order to make
recommendations for public health action
Respiratory Outbreak, Toronto Version May 2007
2
Prologue
On February 14th, 2003, the World Health Organization (WHO) reported 305 cases and
two deaths of an unknown respiratory disease in Guangdong Province, China.
On February 21st a doctor from that province arrived in Hong Kong and stayed at the
Metropole Hotel, unaware that he had been exposed to this unknown, respiratory
pathogen and was now infected. His illness marked the beginning of a global outbreak.
The story continues in Toronto, Canada in late February….
Part 1. A case of acute respiratory illness in Toronto
Between February 18th and 21st, 2003, a 78-year old woman was on vacation in Hong
Kong with her husband – they stayed at the Metropole Hotel. Upon her return home to
Toronto, Canada, on February 23rd she developed a fever, myalgia, sore throat and a
cough. She was cared for by her family at her home. Over the next 10 days, her condition
deteriorated and she died at home on March 5th.
On February 27th, this woman’s 41-year old son (Case A) became ill with fever and
respiratory symptoms. On March 7th, 2003, he presented to the emergency room (ER) of
Hospital A and was admitted to the Intensive Care Unit (ICU) where he died on March
13th.
Between March 3rd and 14th, four more family members developed similar symptoms. The
doctor who provided care to this family on March 6th developed respiratory symptoms on
March 10th.
By March 19th, several nurses who worked in the ER and ICU of Hospital A reported ill
with fever and one or more of these symptoms: cough, malaise, myalgia, and headache.
They were told to isolate themselves at home and to wear masks. The local public health
department was notified of this hospital cluster.
As a Field Epidemiologist placed with the local public health department, you are called in
to assist.
Question 1a
Is this an outbreak?
Question 1b
What source(s) of routine surveillance data could help you confirm the existence of an
outbreak in this investigation?
Question 1c
List the types of pathogens (bacterial, viral etc.) that could be considered in the differential
diagnosis in an outbreak of respiratory illness.
Question 1d
Who should be on the outbreak team?
Respiratory Outbreak, Toronto Version May 2007
3
Discussion for Part 1
Because we expect respiratory infections to spread within a family, the situation was not
initially defined as an outbreak. In Toronto, it was initially thought that the index case and
her son (Case A) had tuberculosis (TB). The family members were being investigated as
potential TB case contacts. However, the two deaths and the rapid onset of respiratory
symptoms among hospital staff raised suspicion that there was something going on
unrelated to TB.
The categories of pathogens included in the differential diagnosis for a respiratory infection
include: viral, bacterial, fungal and parasitic infections (though unlikely as it would not
present like an outbreak). A list of causes of respiratory illness is provided in Appendix 1.
As in any outbreak investigation, you should be part of a multidisciplinary team, including
epidemiologists, laboratory specialists and clinicians, if possible with expertise in different
content areas.
During respiratory disease outbreaks, laboratory surveillance data related to influenza and
other reportable respiratory disease are useful - in reviewing these, epidemiologists can
verify baseline (expected) case counts, and may be able to identify unusual
patterns/activity – leading or helping confirm the existence of an outbreak. Systematic
laboratory testing to identify the pathogen is an important first step in any outbreak. In this
scenario, while the causative agent is unknown, supplementary information from the lab
can help rule out the differential diagnoses.
Part 2. Collecting laboratory specimens
On March 12th, 2003, the WHO issued an alert about a new illness called Severe Acute
Respiratory Syndrome (SARS); global surveillance activities were implemented. Based on
the epidemiological and laboratory information available, the Toronto Public Health (TPH)
department and clinicians identify the Toronto situation as a SARS outbreak.
Recognizing the need to coordinate laboratory-based surveillance with the epidemiologic
investigation, TPH contacts the microbiologist at the provincial laboratory on March 20th to
ensure parties keep one another updated.
The outbreak investigation team includes hospital and public health representatives. They
decide to look for other cases of SARS and adopt the outbreak case definition set by
WHO:
Suspect case: Person of any age with a history of high fever (>38°C) AND cough
or breathing difficulty AND one or more of the following exposures during the 10
days prior to onset of symptoms: close contact with a probable case of SARS,
history of travel to affected areas (e.g., Hong Kong).
Probable case: A suspect case with radiographic evidence of infiltrates consistent
with pneumonia or respiratory distress syndrome on chest x-ray.
As part of your hypothesis generation, you determine that you should interview one of the
ill hospital staff members at home. You would collect the epidemiologic information and a
Respiratory Outbreak, Toronto Version May 2007
4
nurse will collect specimens for laboratory testing. The nurse seeks your advice in
determining what kind of specimens she needs to collect and what she needs to bring
along with her.
Question 2a
List the different samples that could be taken during outbreaks of respiratory illness
caused by an unknown pathogen.
Question 2b
In this investigation, what kind of samples should be collected? What materials will you
need?
Question 2c
1) From whom should you obtain samples?
2) In this situation, would you consider testing i)individuals not meeting the outbreak case
definition or ii)healthy contacts of ill individuals? If so, why?
3) How many individuals should you sample? Why?
4) How many samples would you take from each person?
Question 2d
Will you need to use transport media?
If so, what kind of transport media should be used for viral specimens?
Respiratory Outbreak, Toronto Version May 2007
5
Discussion for Part 2
Appendix 2 shows the range of specimens that could be taken during respiratory
outbreaks of unknown etiology, and the materials required to obtain them. You consult
with the microbiologist at the laboratory about types of samples to collect.
You determine that the nurse should obtain a respiratory sample (a nasopharyngeal (NP)
swab). Nasopharyngeal aspirates (NPA) are more invasive, are harder to collect and can
produce more contamination – so while they were initially considered, NP swabs were
favoured. In addition to the NP swab, a whole blood sample is obtained to look for
circulating antigens.
Before heading off to the field, you ask the microbiologist about specimen collection and
shipment protocols. She recommends that you transport the NP swab in viral transport
media (VTM) to improve virus survival, but notes that such media are not required if cold
chain is ensured and if the laboratory can process the specimens promptly. Transport
media are also used to maintain viability of bacterial samples – but viability is also
dependant on the pathogen.
To transport the NP swab and whole blood sample you ask the nurse to bring:
 a styrofoam container/cooler with ice packs to maintain cold chain for collected
samples;
 sterile rayon swabs with plastic shafts for the NP swab;
and
 sterile EDTA specimen tubes for collection of whole blood (lavender top).
Sampling protocol
As the causative pathogen is unknown, the laboratory advises taking samples from all ill
individuals. Since multiple tests may be required to exclude all differential organisms, with
the potential identification of a new pathogen and the subsequent need to send additional
samples to multiple (reference) laboratories, the laboratory requests that, if possible,
additional specimens be taken.
The number of samples needed is generally based on:
i)
sensitivity of the test
ii)
the importance of avoiding misclassification
iii)
whether this is a known or unknown pathogen
Serial sampling on consecutive days can increase the probability of identifying an
unknown agent and assists in determining the natural course of the unknown disease.
While laboratory testing is usually done to test for a causative agent based on a
predetermined hypothesis, in this outbreak, testing of contacts who did not meet the case
definition is considered because of the potentially high infectious nature and the
seriousness of illness. This is especially important early in an outbreak with an unknown
pathogen, as the case definition may need to be refined (i.e., too specific).
Other reasons for testing of contacts not meeting the outbreak case definition include:
 Describing the natural history of infection
 Determining whether asymptomatic infection is possible
 Estimating the secondary attack rate
Respiratory Outbreak, Toronto Version May 2007
6
Part 3. Ethical considerations
As described above, specimens were taken from ill and healthy persons and personal
information was collected from both groups. This process can be both uncomfortable and
invasive.
Question 3a
What are the ethical considerations in testing these contacts?
Question 3b
How do you ensure confidentiality in a high profile outbreak?
Respiratory Outbreak, Toronto Version May 2007
7
Discussion for Part 3
Before collecting epidemiologic information or biological specimens, it is important that
people understand why and how they were identified and selected, why their personal
information and specimens are required, and how the information generated will be used;
data and specimens should only be collected once informed consent is obtained.
Normally, if specimens are collected for purely academic research, approval from an
ethics committee and written consent from participant’s and assurances of confidentiality
are absolutely necessary; this is usually time-consuming and can add significant delay to
such studies. Outbreak investigations are usually exempt from ethics committee
clearance, and can usually proceed under public health legislation. Nonetheless, good
practice (respecting confidentiality), provision of informed consent (written if possible) and
self-imposed ethical standards must be maintained during outbreaks.
Because this is an outbreak of a new communicable disease, you and the laboratory
agree that you need as much information as possible (e.g., to detect subclinical cases)
and therefore healthy contacts of cases or mildly symptomatic individuals are asked to
provide samples. As lab results became available, some case contacts test positive for
the causative agent for SARS. While some of these contacts will eventually meet the WHO
case definition for SARS, others remain mildly symptomatic and do not meet the case
definition.
While specimens collected for diagnostic purposes should be labelled with the patient’s
name, you point out to the team that it is important to not generate or release any data that
could potentially identify individuals. The team agrees to avoid publishing patient names or
personal identifiers in reports.
Part 4. Taking the sample
The nurse is preparing to collect the samples from the case.
Question 4A
What personal precautions should be taken when interacting with possible cases infected
with an unknown infectious respiratory pathogen?
Respiratory Outbreak, Toronto Version May 2007
8
Discussion for Part 4
Since the mode of SARS transmission is still unknown, the field team visiting the suspect
cases must wear personal protective equipment (PPE) while conducting interviews and
when obtaining specimens,
It is likely that the causative agent in the outbreak is a respiratory pathogen. Respiratory
pathogens can be spread by respiratory droplets, through direct or indirect contact with the
patient, or by the airborne route, particularly if the individual coughs or sneezes during an
interview or when collecting biologic specimens.
Precautions against all possible modes of transmission are taken (see Table 2); before
entering the homes, gloves, gown, eye protection and masks are worn. To be effective,
you must ensure that the equipment you are donning fit you properly.
Table 2: Infection control precautions to interrupt transmission of pathogens
Precautions
Contact
precautions
Droplet
precautions
Airborne
precautions
Use
For patients known or suspected to have
serious illnesses easily transmitted by direct
patient contact or by contact with items in the
patient's environment
Barrier to stop infections spread by large (>5
microns), moist droplets produced by people
when they cough, sneeze or speak
For patients known or suspected to have
serious illnesses transmitted by airborne droplet
nuclei
Requirements
Gloves, gown
Contact precautions + wellfitting mask + eye protection
Contact and droplet
precautions + N95 mask,
isolation room (in hospital)
Part 5. Laboratory selection
The nurse is preparing and packaging the NP swabs for shipment to the laboratory. You
consult with your colleagues as to the appropriate laboratory to send these specimens.
Your colleagues offer you several choices (the hospital lab, the regional lab…etc). You
vaguely recall a discussion with a microbiologist relating to laboratory designation
according to pathogen biosafety levels.
Question 5a
What biosafety level do you think is appropriate for SARS?
Question 5b
What documentation should accompany your team’s samples to ensure timely
processing?
Respiratory Outbreak, Toronto Version May 2007
9
Discussion for Part 5
Laboratories designate biosafety precautions for specimens based on the severity of
disease and the potential for a cure. In this outbreak, it is important to identify a laboratory
capable and designated to handle highly infectious agents and with experience in different
methodology. Assistance from international and reference laboratories should also be
considered. Appendix 4 references the different laboratory biosafety levels and the work
they can carry out.
In this situation with an unknown pathogen, BioSafety Level (BSL)-3 handling should be
carried out, since the samples may contain a live virus. This means samples will be
handled under a safety hood and/or with PPE plus a full-face shield.
Before collecting the sample, you should consult with the laboratory point of contact, to
ensure timely transport and processing of the samples.
The following minimal information needs to accompany the sample:
1. Type of sample
2. Place
3. Symptoms and dates of onset
4. Date of specimen collection
5. Patient Name/unique identifier/outbreak code
Note: Steps should be taken to ensure the link between the patient and specimen is
maintained. When there are multiple samples taken from each patient or when several
laboratories are involved in testing and use different identifiers, it may be hard to re-match
results to patients. As a minimum, use date of birth and sex consistently for each sample
taken.
The samples which your team collects are to be sent to the National Microbiology
Laboratory (NML) in Winnipeg, Manitoba (a two hour flight away), which includes BSL-3
and 4 facilities. In other countries experiencing SARS, the WHO provides a list of
appropriate laboratories where samples could be sent.
--------END OF Session 1-------
Respiratory Outbreak, Toronto Version May 2007
10
Part 6. Transportation of respiratory specimens
Samples from the ill and their contacts have been taken and the appropriate laboratory
has been identified and contacted.
On March 21st, the samples are ready to be sent to the national laboratory. The WHO has
also requested parallel specimens for testing.
Question 6a
What kind of packaging is necessary to prepare these samples for transport? Consider
both national and international shipment requirements.
Question 6b
Would you request a cold chain for the transport of these nasopharyngeal specimens?
Why?
Respiratory Outbreak, Toronto Version May 2007
11
Discussion for Part 6
Shipping potentially infectious samples is a highly regulated issue that should only be
performed by trained individuals. If regulations are ignored, it can not only lead to heavy
fines (and even imprisonment), but the sample may not be accepted by the transporting
body, the sample may become destroyed, and in a worst case scenario, the sample could
infect people who come into contact with the package. National regulations for the
transport of infectious substances exist in most countries. In Canada, the Transportation
of Dangerous Goods Act provides procedures for shipping of specimens.
Generally, substances that contain potentially infectious agents should be transported in a
“triple package”:
● Primary receptacle*: a leak proof specimen container. In case of breakage, the
receptacle is packaged with sufficient absorbent material to absorb its entire
contents..
● Secondary packaging*: a leak proof secondary container which encloses and
protects the primary receptacle(s). Several cushioned primary receptacles may be
placed in one secondary package, but additional absorbent material should be
added to the package -- enough to absorb all fluids contained, in case of breakage.
● Outer packaging*: the secondary package(s) are placed in outer shipping
packaging with suitable cushioning material. Outer packaging protects the contents
from outside influences, such as physical damage, while in transit. The smallest
overall external dimension shall be 10x10 cm.
*Please refer to current International Airline and Transport Agency (IATA) shipping guidelines
for more details about definitions, packaging requirements, markings and labels, accompanying
documentation, and refrigerants.
Example of triple package:
Respiratory Outbreak, Toronto Version May 2007
12
Part 7. Laboratory testing
The outbreak continues over the next few weeks…
On March 24th, 2003, facilitated by international epidemiologic and laboratory
collaboration, a new coronavirus is identified from a SARS case. Between April 8th–10th,
researchers find positive antibody titers to SARS CoV in a high percentage of SARS
patients and no titer in control patients. Finally, on April 16th, the WHO announces that this
coronavirus, never before seen in humans, is the causative agent of SARS, based on
study results that found that monkeys infected with the pathogen developed similar
symptoms to those observed in human SARS cases. The speed with which the causative
agent of SARS was identified can be attributed in part through this excellent, and
unprecedented, laboratory-epidemiology collaboration between 13 laboratories in 10
countries.
After reviewing the descriptive epidemiology for this outbreak to date, your team finds that
contact and droplet precautions implemented in Hospital A reduced the nosocomial spread
of SARS. As a result, you decide that the Toronto data support the leading hypothesis
that SARS spreads by respiratory droplets.
On March 25, the samples obtained by the nurse that accompanied you reach the national
laboratory and are tested for respiratory pathogens and SARS CoV. The laboratory stores
parts of the samples under appropriate conditions for further testing.
With no standard test for SARS established yet, what are the potential tests that could be
considered?
Question 7a
What types of tests could be used to look for antibodies to the SARS pathogen?
What types of tests could be used to look for the SARS pathogen (coronavirus) itself?
Respiratory Outbreak, Toronto Version May 2007
13
Discussion for Part 7
Table 3: Available tests for detection of SARS-CoV.
Test Category
Direct Tests
Virus isolation (culture)
Purpose of Test
To provide evidence that a person
has been infected with the virus.
In the course of the identification of
the SARS-CoV, it is important to
prove that it is also the causative
agent.
Comments/Interpretation
A negative test does not disprove this
hypothesis; a negative result could be
due to timing of specimen collection,
no live virus in the sample and/or a
poor handling of the culture or using
the wrong cells to grow the virus in.
Requires BSL-3 laboratory.
Molecular methods (i.e.,
PCR)
To provide evidence that genetic
material of the virus is present in
the sample.
Note that PCR, particularly RT-PCR
is prone to contamination.
Indirect Tests
Enzyme-linked
immunosorbent
assay (ELISA)
Immunofluorescence
assay (IFA)
To detect a mixture of IgM and IgG
antibodies in the serum of SARS
patients.
To detect IgM or IgG (or both) in the
serum of SARS patients.
This technique uses SARS CoV-infected
cells fixed on a microscope slide.
Immunofluorescent-labeled secondary
antibodies against human IgG and/or IgM
detect patient antibodies binding to viral
antigens using an immunofluorescence
microscope
Neutralization
test (NT)
To assess (via titration) the ability of
patient sera to neutralize the
infectivity of SARS-CoV on cell
culture
A negative test does not rule out virus
in the sample nor that the patient had
the infection. It can be due to low
sensitivity of the PCR or timing of the
sample collection.
A positive test must be interpreted as
such only after ascertaining that
laboratory cross-contamination did
not occur.
IgG usually remains detectable after
resolution of the illness. Reliably
yields positive results at, or after day
28 following SARS onset
Separate IgG/IgM tests were not
available at the beginning
Expect positive results at, or after day
28 following SARS onset. Results
may be quantified by using serial
titrations of patient sera.
Best correlate of immunity. However,
must be done in institutions with BSL3 facilities and is expensive and timeconsuming.
Direct Tests: demonstration of the presence of the infectious agent/pathogen
Indirect Tests: demonstration of the presence of antibodies to a particular agent
Antibody tests
A positive IgG antibody test result indicates previous infection with SARS-CoV. The
microbiologist informs you that seroconversion from negative to positive or a four-fold rise
in the antibody titer from acute to convalescent serum indicates a recent infection. A
negative antibody test taken more than 28 days after symptom onset is likely to indicate no
infection with SARS-CoV.
Respiratory Outbreak, Toronto Version May 2007
14
You and the laboratory discuss that there seems to be no background seroprevalence
against SARS-CoV in control populations screened so far. Antibody testing allows the
indirect diagnosis of SARS-CoV infection but has the advantage of being independent of
the sample type, in contrast to other virus detection methods. Disadvantages include a
lack of usefulness in the early stages of SARS (an IgG, not IgM test) and some SARS
cases have negative antibody tests. A positive antibody test indicates previous infection
with SARS CoV (helpful to confirm the diagnosis) but a negative test in a patient who
meets the case definition of SARS does not necessarily disprove the diagnosis.
Molecular tests
To design a PCR probe, parts of the genomic sequence of the pathogen have to be known
and must be specific for the virus. The latest tests for SARS used two different regions of
the gene (the replicase and nuclease gene) as targets.
There are different methods and names for PCR tests:
1) RT-PCR: Since PCR has to be done with a reverse transcriptase on a RNA-virus
2) Nested and unnested PCRs: Nested PCR means that first a larger section of the
gene is replicated and then the specific area is targeted. Nested PCRs are usually
more sensitive, but they bear a higher risk for contamination.
3) Real-time PCR: Where each cycle of the PCR is analysed and the amount of the
virus in the sample can be estimated.
Part 8. Confirmation of SARS Coronavirus infection
Now that confirmatory tests have been developed to detect SARS-CoV infection, you can
include laboratory results in your case definition.
Question 8A
What could be a possible laboratory-confirmed case definition for SARS?
Respiratory Outbreak, Toronto Version May 2007
15
Discussion for Part 8
WHO recommends the following case definition for laboratory-confirmed SARS cases:
A person with clinically-compatible illness and one of:
 Detection of antibody to SARS-CoV in specimens obtained during acute illness or >28
days after illness onset,
OR
 Detection of SARS-CoV RNA by RT-PCR, confirmed by a second PCR assay by using
a second aliquot of the specimen and a different set of PCR primers,
OR
 Isolation of SARS CoV
By May 2003, 257 probable and suspect SARS cases associated with the Toronto area
outbreak had been reported. Of these, approximately 63% were health-care acquired
infections, 32% were household or close contacts of cases and 5% were travel-related
(imported cases). Twenty-seven SARS-related deaths were also reported.
Part 9 Interpretation of laboratory results
Available serology and PCR results from those tested for SARS (including both
symptomatic and asymptomatic contacts of cases) are now available. A summary is
presented in Table 4.
Table 4:
Specimen
PCR positive
Seropositive
EDTA blood
Nasopharyngeal
swab
Throat swab
Stool
Sputum
PCR negative
Seropositive
Seronegative
4
21
Seronegativ
e
19
29
74
48
296
320
11
8
6
9
0
2
32
7
0
278
32
9
Using convalescent serology as the “gold standard”, the team decides to look at the
sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV)
of the PCR tests as a diagnostic tool.
Respiratory Outbreak, Toronto Version May 2007
16
Question 9a
For the PCR test, calculate and describe the
a) sensitivity
b) specificity
c) positive predictive value
d) negative predictive value
Question 9b
Interpret these values. Which of these do you think is most relevant for public health
decision-making?
Question 9c
How do you interpret positive PCR test results for SARS CoV in individuals who do not
meet the outbreak case definition?
Question 9d
If there were a case of SARS reported in a setting where there were no known cases
(ruling out travel-related importations), how would you interpret a positive PCR test result
for SARS CoV?
Question 9e
How do you interpret negative PCR results for SARS CoV in individuals who meet the
case definition?
Question 9f
Human metapneumovirus was also identified in some specimens from SARS patients.
Discuss possible interpretations of this result.
Respiratory Outbreak, Toronto Version May 2007
17
Answer to Part 9
Answer 9a
Test
(PCR)
Gold standard (Serology)
+
a (true positive)
b (false positive)
c (false negative)
d (true negative)
+
-
Sensitivity=a/(a+c)
Specificity=d/(b+d)
PPV=a/(a+b)
NPV=d/(c+d)
Table 5: Accuracy calculations for PCR tests for SARS-CoV
Specimen
EDTA
blood
NP swab
Throat
swab
Stool
Sputum
Sensitivity
(%)
5
Specificity
(%)
94
PPV (%)
NPV (%)
17
80
30
53
92
97
42
55
87
90
53
100
100
82
100
75
82
100
Sensitivity of a test asks “How many true cases were detected using the test?”. For SARS
we want a test that is highly sensitive because of the implications of missing a true case
(i.e., transmission of illness). Using these data, the test sensitivity was low for the
specimens most frequently taken (blood, NP and throat swab) when using the PCR test
results used at that time.
Note that the PPV of the PCR depends on the type of PCR (the protocol and primer used)
--so the PPV calculated is specific for the PCR used here.
Specificity of a test asks “How many true non-cases will test negative?”. Based on your
data, you found that specificity of the PCR tests were high.
Answer 9b
Public health relevance
In order to interpret the data correctly, and especially during outbreaks, clinicians and
epidemiologists need to understand the level of confidence that a positive diagnostic test
is a true case and not a false positive. This is the PPV of the test. The NPV is the
probability that with a negative test, the person is not infected.
The PPV (rate of false positives detected) should be low– the implications of detecting
false positives would in this outbreak would lead to unintentional or inappropriate use of
resources, unnecessary quarantine of contacts (i.e., work days lost) as well as the
psychological effects of being stigmatized for having SARS.
From your calculations, you find that the PPV is low for blood, throat and NP swabs. If a
PCR test is positive, there is a 45-83% chance that it is a false positive. The NPV is high
Respiratory Outbreak, Toronto Version May 2007
18
(i.e., under the assumption that proper sampling and techniques were used, a negative
PCR test indicates that the individual was likely not infected with SARS CoV).
Answer 9c
Positive PCR in those that did not meet case definition
Among individuals that did not meet the case definition for SARS, some have positive
PCR test results for SARS CoV (using NP and throat swabs). Based on the low PPV of
these PCR tests, the microbiologist warns that these may be false positive results.
However, since SARS is a new disease and the full spectrum of illness is not known, it is
possible that the case definition is too specific and excludes those at the mild end of the
clinical spectrum. That is, some of these people who did not meet the case definition may
be still infected with SARS CoV.
Answer 9d
The PPV of a test is dependent on the sensitivity and specificity of the test as well as the
prevalence of the disease in the population. When the prevalence of disease is low, the
PPV will also be low, even using a test with high sensitivity and specificity. Therefore, if
there were a case of SARS reported in a setting where there were no known cases, one
should consider the possibility that this may be a false positive result. It is important to use
a testing strategy which starts to test patients with a higher likelihood of being a case first,
together with stepwise laboratory testing (i.e., a more sensitive test first with a second test
for confirmation).
Answer 9e
The negative predictive value of the PCR tests was generally high indicating that a
negative test likely means that the individual was not infected with SARS CoV. However, it
is possible that an infected individual could test negative on PCR due to poor timing of
specimen collection or improper handling of the specimen.
Answer 9r
Human metapneumovirus
The discovery of metapneumovirus in some samples could mean several things. The
specimen could have been contaminated during sampling or in the laboratory, at some
point in the diagnostic process. Another possibility is that there has been a misdiagnosis
and the SARS patient in fact has metapneumoviral infection and is not a SARS case at all.
However, as specimens from multiple SARS cases are positive for both organisms, it is
most likely that it is a co-infection. The potential role of human metapneumovirus in SARS
is not completely understood.
Antibodies to SARS-associated CoV were detected in 96% of the Toronto SARS cases’
specimens providing further strength to the body of global evidence that this outbreak was
caused by a new coronavirus. Laboratory test results for SARS-CoV must always be
interpreted in conjunction with clinical and epidemiological history.
Respiratory Outbreak, Toronto Version May 2007
19
Part 10 Sequence data (Optional)
RT-PCR sequencing analysis
The very first whole virus genomic sequence on the SARS CoV was done to confirm the
novelty of the virus and to develop molecular-based tests, such as the PCR.
Sequence analysis of different SARS strains were conducted to see if strains found in
different geographic areas of the world were genetically related. This was done for three
reasons:
1. To determine the stability of the genome
2. To determine whether variability in the genome produces strains with different
pathogenicity
3. To determine if strains were related and if so, whether this could facilitate the
understanding or identification of transmission chains
The phylogenetic tree of the sequencing results is presented in the figure below.
Recall: To interpret the above, compare the base pair differences of each strain to the
sequence results of the index case (shown in the far left column).
Question 10a
How would you interpret this and the following phylogenetic tree given what you know of
the origin of the Toronto cases and the common source at the Metropole hotel?
What factors can influence variations of the genome?
Respiratory Outbreak, Toronto Version May 2007
20
Figure
The letters before the numbers refer
to the place where the strains were
isolated:
SIN= Singapore
BJ=Beijing
GZ= Guangzhou
CUHK= Chinese University of
Hong Kong
HKU= Hong Kong University.
TOR= Toronto
Linked to Hotel M
YiJun Ruan, Chia Lin Wei, Ling Ai Ee, et al, Comparative full-length genome sequence analysis of 14 SARS coronovirus
isolates and common mutations associated with putative origins of infection, Lancet, May 2003 Vol 361, 1779-1785
Respiratory Outbreak, Toronto Version May 2007
21
Discussion for Part 10
Interpreting phylogenetic trees
A phylogenetic tree is a graphical way to depict the evolutionary relationships of a
group of organisms, in this case, viral strains. The distance between the branches
of the tree is directly proportional to the genetic distance between the strains. If
two species have a small distance between them, then they have a recent
common ancestor; if they are far apart, then their common ancestor is in the
remote past.
The figure above shows that the strains from Singapore are more closely related to
each other than the ones from earlier infections in China. It also shows that the
Toronto strain is more closely related to the Hong Kong strain, which fits in well
given the known transmission history. Similarly, in the figure showing the
character sequences, there are fewer differences (shown in red) in base pairs of
strains linked to Hotel M compared to those that are not linked to Hotel M.
Factors influencing variations in the genome:



Viruses may mutate within the host, or after passing from person to person. These
mutations can happen spontaneously or through adaptations to environmental
conditions.
Viral mutations can also occur during laboratory propagation for isolation, especially
when passing through several life cycles on the cell/lines.
Human error can occur in transcribing genomes to databanks, especially if many
different investigators are involved in the complete sequencing.
Respiratory Outbreak, Toronto Version May 2007
22
Part 11 Drawing conclusions
Looking at the broader global context, surveillance data indicated that by the end of the
outbreak, there were 8,098 cases and 774 deaths worldwide, with an overall case fatality
ratio of 9.6% (although this varied from country to country). Mortality was highest in those
aged 24 years and older. It is reported that up to 80% of infections were acquired in
hospitals.
Back to Toronto, where you now have all currently available laboratory data.
Question 11a
Integrating the laboratory and epidemiological data, provide a brief summary of what you
know about the outbreak in Toronto.
Question 11b
What else would you like to know about this new pathogen? What type of epidemiological
studies would you suggest to address these questions?
Respiratory Outbreak, Toronto Version May 2007
23
Discussion for Part 11
Outbreak Summary
An outbreak of severe acute respiratory illness originated in Guangdong province, China in
November 2002. As a result of international travel, the illness was introduced in Toronto in
February 2003l. The WHO subsequently declared an outbreak of a new illness, Severe
Acute Respiratory Syndrome (SARS) on March 12th, 2003.
Epidemiology – person, place, time
By May 2003, when the first phase of the outbreak was declared over, 257 SARS cases
had been reported in Toronto, as a result of unrecognized spread to hospital staff, patients
and visitors to Hospital A and to contacts of these cases. Approximately 63% of the
infections were health-care acquired. The case fatality rate was 10.5%.
Laboratory
Almost all cases with available laboratory data had positive serology results for SARS
CoV. Based on strong epidemiology-laboratory collaboration worldwide, transmission
from Asia to Canada was confirmed through sequencing and comparison of the strains
isolated in Asia and Toronto. PCR sequencing analysis confirmed that the outbreak was
caused by a newly recognized coronavirus.
As SARS was caused by an unknown pathogen, there was added urgency to develop
reliable, accurate laboratory tests in light of the severity of illness and global panic. This
necessitated close collaboration of epidemiologists and laboratory specialists.
Additional information
Although the outbreak in Toronto was controlled and initially declared over in early May,
2003, a new cluster of cases was reported in a Toronto hospital on May 22nd. On July 5th,
2003, WHO reported that the chain of transmission of SARS was broken.
Further investigations that could be recommended include:
 A case-control study to look at risk factors for infection (e.g., use of personal
protective measures, exposure through aerosol-generating procedures etc.) to
better understand disease transmission.
 Seroprevalence studies to look at extent of transmission and explore the spectrum
of illness (including the possibility of asymptomatic cases)
Respiratory Outbreak, Toronto Version May 2007
24
Epilogue – A global picture
Worldwide, 5,865 cases of SARS were reported as of May 1st, 2003.
On May 23rd, 2003, coronaviruses linked to SARS were found by researchers in China in
some animals: civets, racoon-dogs, and Chinese ferret badgers. No new cases appeared
once the last patients recovered. However, SARS CoV is still available in many
laboratories worldwide. Laboratory-acquired infections caused by inappropriate handling
continued to be sources of three laboratory-acquired cases of SARS since the end of the
outbreak. It is important that adequate containment practices and procedures are followed
when working with infectious agents to minimize these risks.
The rapid spread of SARS to 29 different countries in a short period of time demonstrates
how a pandemic might behave. However, during SARS, epidemiologists and laboratories
worked closely in conjunction with international health organizations and governmental
agencies to coordinate outbreak control efforts. As a result of active case finding,
epidemiological and laboratory information and international travel restrictions, a feared
uncontrolled worldwide epidemic never materialized.
Expertise was pooled and
information was shared to discover this new coronavirus. The importance of this
multisectoral collaboration remains among the top lessons learned.
Respiratory Outbreak, Toronto Version May 2007
25
References:
1. SARS Reference 10/2003, editors Kamps and Hoffman, 3rd Edition, Flying Publisher,
2003. (www.sarsreference.com)
2. WHO. Severe acute respiratory syndrome (SARS): Status of the outbreak and lessons
for the immediate future. Geneva, 20 May 2003.
http://www.who.int/csr/media/sars_wha.pdf
3. Reference: WHO. Case Definitions for Surveillance of Severe Acute Respiratory
Syndrome (SARS). Accessed 22nd February 2005. Last updated 1st May 2003.
http://www.who.int/csr/sars/casedefinition/en/
4. Medical News Today. Consequences of SARS Revealed.
5. Reference: http://www.medicalnewstoday.com/medicalnews.php?newsid=15935
6. WHO. Severe Acute Respiratory Syndrome (SARS)-multi-country outbreak - Update
49.7 May 2003. http://www.who.int/csr/don/2003_05_07a/en/
7. WHO. Summary of probable SARS cases with onset of illness from 1 November 2002
to 31 July 2003 http://www.who.int/csr/sars/country/table2004_04_21/en/
8. Reference: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5212a1.htm
9. Reference: http://www.cioms.ch/frame_guidelines_nov_2002.htm
10. WHO biosafety guidelines for handling SARS clinical specimens and materials derived
from laboratory investigations of SARS
11. Guidance on regulations for the Transport of Infectious Substances 2007-2008
http://www.who.int/csr/resources/publications/biosafety/WHO_CDS_EPR_2007_2/en/i
ndex.html
Respiratory Outbreak, Toronto Version May 2007
26
Appendix 1: Differential diagnosis for acute respiratory infections
UPPER RESPIRATORY TRACT INFECTIONS
Acute pharyngitis
Viral:
Influenza A & B
Rhinovirus
Rhinitis
Laryngitis
Adenovirus
Coronavirus
Parainfluenza virus
Epstein-Barr Virus
Coxsackie A
Bacterial:
Streptococcus pyogenes (Group A beta haemolytic streptococcus), Corynebacterium diphtheria
Rhinovirus
Coronavirus
Adenovirus
Respiratory syncytial virus (RSV)
Parainfluenza virus
Primarily viral: Influenza, rhinovirus, adenovirus
Primarily viral: Parainfluenza virus, influenza, RSV, adenovirus, rhinovirus
Bacterial: Mycoplasma pneumoniae
ACUTE LOWER RESPIRATORY TRACT INFECTIONS
Acute bronchitis
Viral: Parainfluenza virus, influenza, RSV, adenovirus, measles
Laryngotracheobronchitis
Influenza
PNEUMONIA
Community-acquired
pneumonia
Nosocomial pneumonia
Atypical
Other
Bacterial: Bordetella pertussis, Haemophilus influenzae, Mycoplasma pneumoniae, Chlamydia
pneumoniae
Influenza virus
<30 years old: Influenza virus
Bacterial: Pneumococcus spp., Mycoplasma pneumoniae, Chlamydia pneumoniae
30-60 years old: RSV
Bacterial: Pneumococcus spp., Mycoplasma pneumoniae, Chlamydia pneumoniae, Haemophilus
influenzae
>60 years old: Respiratory viruses
Bacterial: Pneumococcus spp., Haemophilus influenzae, aerobic gram negative rods, Staphylococcus
aureus
Enterobacter spp., Klebsiella spp., Acinetobacter spp., Pseudomonas spp., S. aureus, Anaerobic bacteria
Mycoplasma pneumoniae, Respiratory viruses including SARS coronavirus, Influenza, Chlamydia
pneumoniae
Opportunistic pneumonias (i.e., fungal) wouldn't present as an outbreak
Respiratory Outbreak, Toronto Version May 2007
27
Note: Also consider 1) non-infectious possibilities including new onset asthma (e.g., due to chemical exposure), thiamine deficiency in infants; 2)
chemical/bioterrorism agents or 3) the possibility of a new pathogen. The list above is only a short summary of the most probable causes; it is not meant to be an
exhaustive list.
Respiratory Outbreak, Toronto Version May 2007
28
Appendix 2: Specimen Collection and handling during outbreaks of respiratory illness
Upper
respiratory
Nasopharyngeal
wash/aspirate
(prefer
nasopharyngeal
aspirate for viral
specimens)
Timing of
specimen
collection
Organisms
Test
Transport
medium/
container
type
Instructions
for Transport
Time to test
results
Other comments
Virus: usually
during first 4
days of illness
Viral:
- Influenza A, B
- Parainfluenza 1,2,3
- Adenovirus
- RSV
- Other respiratory
viruses
Viral:
- Culture
- PCR
- Immunofluorescence
- Direct antigen testing:
EIA,IFA,
immunochromatogr
aphy (Rapid tests
for influenza, etc)
Sterile
leakproof
container
Virus:
Can be kept at
room temperature 4
hours.
Virus:
- Culture:
days/weeks
- PCR: hours
- Rapid tests:
minutes
For bacterial detection,
proceed with the culture
as soon as possible
Bacteria:
preferably in the
acute phase of
illness (prior to
initiation of
antibiotic
therapy)
Bacterial:
Oropharyngeal:
C. diphtheria
S. pyogenes
Afterwards, ship in
transport media
and on ice/cold
packs (+4°C) within
a few hours of
collection
Bacterial:
Culture
Bacteria:
- Culture: days
- PCR: hours
If international, ship
on dry ice
Nasopharyngeal
swabs:
B. pertussis
Nasopharyngeal
or
oropharyngeal
swabs*
As above
Viral:
Sterile vials
with viral
transport
media
Bacterial:
Tellurite in
tube
(except B.
pertussis:
Reagan-Lowe
media in tube)
Respiratory Outbreak, Toronto Version May 2007
While throat swabs are
easy to collect for
bacterial culture, as
they are also
contaminated with
normal flora, do not
keep at room temp for
>1hr. Process as soon
as possible. If not
possible, then keep at
+4°C.
29
Lower
respiratory
Sputum
Timing of
specimen
collection
Organisms
Test
Transport
medium/
container
type
Instructions
for Transport
Time to test
results
Any time in
course of illness
M. tuberculosis
S. pneumoniae
H. influenzae
M. pneumoniae
L. pneumophila
C. pneumoniae
S. aureus
M. catarrhalis
K. pneumoniae
P. aeruginosa
A. baumannii
Fungi:
C. albicans
Aspergillus spp.
Culture
PCR
Sterile
leakproof
container
Bacteria: Keep at
room temperature
within 1 hour of
collection:
Afterwards, Ship on
ice packs (+4°C)
Culture: days
(except M.
tuberculosis:
weeks)
M. tuberculosis
S. pneumoniae
H. influenzae
M. pneumoniae
L. pneumophila
C. pneumoniae
S. aureus
Fungi
Viral:
Culture
PCR
Bacteria:
preferably in the
acute phase of
illness (prior to
initiation of
antibiotic
therapy)
BAL
Tracheal
aspirate
Pleural fluid tap
When clinically
appropriate:
intubated, severe
lung disease, or
pleural effusion.
Isolates for viral
testing should be
collected within 4
days after illness
onset as virus
shedding
decreases
rapidly after this.
Other comments
PCR: hours
Fungi:
Room temperature
Bacterial :
Standard cultures
Sterile
leakproof
container
For testing
in-patients;
Consider these
specimen types if
sputum not available
S. pneumoniae
S. aureus
H. influenzae
Anaerobic bacteria
Bacteria:
preferably in the
acute phase of
illness
Respiratory Outbreak, Toronto Version May 2007
30
Blood
Serum
Whole Blood
Whole Blood
Timing of
specimen
collection
Organisms
Test
Transport
medium/
container type
Instructions
for Transport
Processing
Time
Other
Acute: as soon
as possible (up
to 7 days)
Convalescent:
after 3-4 weeks
Bacterial, Fungal,
Parasitic
Specific antibody
detection (IgM, IgA,
rising IgG titers)
Sterile red top
tube
Whole blood can
be kept at room
temperature within
2 hours after
collection.
Hours or
days
Do not freeze whole blood.
Legionella
infection:
obtain
convalescent
serum sample
at least 6
weeks after
illness onset.
At fever peak
preferably.
Repeat sample
(collect 2
bottles 3 times
or 3 bottles 2
times, at fever
peaks or
regular
intervals (few
hours apart)
Any time in
illness
Viral:
Influenza
Parainfluenza
Adenovirus
RSV
Avian flu
SARS coronavirus
Bacteria
Bacteria, virus,
parasite
Respiratory Outbreak, Toronto Version May 2007
Centrifugate as soon as
possible then freeze the
serum only.
Afterwards, ship
blood or preferably
serum on cold
packs (domestic)
or on dry ice
(international)
Standard
hemoculture
PCR
Standard blood
culture bottle)
EDTA (lavender
top
Preferably, transport serum
rather than whole blood.
Room temperature
within 2 hours if
around 25-35°C. If
room temperature
is higher and
lower, transport in
an incubator
(37°C)
Days
ship cold packs
(domestic or
international)
Hours or few
days
Sample in aseptic
conditions.
Do not cool or freeze the
bottles
Note: Blood culture is
positive for the causative
organism in up to 20% cases
of pneumonia
Do not centrifugate and do
not freeze EDTA whole
blood
31
Timing of
specimen
collection
Organisms
Other
Tissue – lung,
upper airway
Tissue, major
organs
Test
Transport
medium/
container type
Instructions
for Transport
Electron microscopy
Immunofluorescence
Molecular biology
Sterile container
with viral
transport
medium or
saline.
Culture or PCR:
Sterile container
with physiologic
water
Histopathology:
Formalin fixed
or paraffin
embedded.
Sterile jar
Fresh frozen
tissue: Ship at
–70°C
For patients who are
deceased
Ship at room temp.
For patients who are
deceased
Ship on ice
Isolation of adenovirus in the
stool is more informative if
simultaneously isolated from
the respiratory tract
Clean catch,
sterile leakproof container
Ship on cold packs
(domestic)
Ship on dry ice
(international)
Culture
PCR
Histopathology
Stool
Any time:
Best sample at
around day 14
Urine
Acute phase of
illness
Adenovirus
Culture
PCR
Antigen detection for
S. pneumoniae and
Legionella
pneumophila
Processing
Time
Other
*Swab: use only sterile Dacron or rayon swab with plastic shaft (wooden sticks should not be used as it inhibits viral growth)
**Stability of Specimen – At ambient temperature: 2 hours; Refrigerated: 3 days; Frozen: Unacceptable
In general do not store specimen for bacterial culture for more than 24 hours. Viruses, however, usually remain stable for 2-3 days at 40C
Respiratory Outbreak, Toronto Version May 2007
32
Appendix 3: Taking a nasopharyngeal (NP) swab
You will need the following:
Mask, goggles, gown, gloves
nasopharyngeal swab
viral transport medium (or Regan Lowe media for B. pertussis)
1. Put on all protective equipment.
2. The specimen should be collected with a nasopharyngeal swab. Only sterile
Dacron or rayon swabs with plastic shafts should be used. Wooden sticks
should not be used as it inhibits viral growth). Swabs intended for bacterial
culture should not be used.
3. Incline the person’s head and
insert the NP swab through
the nostrils until resistance is
met by virtue of contact with
the nasopharynx. (*Note:
sometimes you may hit nasal
turbinates first, and need to
change angle a bit to get past
them). Quickly rotate the
swab 3-5 times against the
nasopharynx as long as
tolerated by the patient. Note
that coughing or patient
resistance is adequate
incentive to remove the swab.
4. Withdraw the swab and insert into the transport medium (viral transport
media, or Regan Lowe media for B. pertussis) by inserting the swab at least
½ inch below the surface of the medium. Bend or cut off the wire to fit the
transport medium tube and reattach the cap securely.
5. Label the transport tube with the name and specimen source as per
requirements.
6. Complete the specimen submission form with all required information. Failure
to submit or complete this form could result in a delay in processing the
specimen.
Respiratory Outbreak, Toronto Version May 2007
33
Appendix 4: Biosafety levels and Risk groups (From the WHO Laboratory
Biosafety Manual, 3rd edition, 2005)
BIOSAFETY LEVEL
Isolation of laboratory
Room sealable for
Decontamination
Ventilation :
— inward airflow
— controlled ventilating system
— HEPA-filtered air exhaust
Double-door entry
Airlock
Airlock with shower
Anteroom
Anteroom with shower
Effluent treatment
Autoclave:
— on site
— in laboratory room
— double-ended
Biological safety cabinets
Personnel safety monitoring capability
RISK BIOSAFETY
GROUP LEVEL
1
1= Basic
LABORATORY
TYPE
Basic teaching,
research
Primary health
services;
diagnostic
services, research
Special diagnostic
services, research
2
2 = Basic
3
3=
Containment
4
4 = Maximum Dangerous
containment pathogen units
1
2
3
4
No
No
No
No
Yes
Yes
Yes
Yes
No
No
No
No
No
No
No
No
No
Desirable
Desirable
No
No
No
No
No
No
No
Yes
Yes
Yes/No
Yes
No
No
Yes
Yes/No
Yes/No
Yes
Yes
Yes
Yes
Yes
Yes
_
No
Yes
No
No
No
No
No
Desirable
No
No
Desirable
No
Yes
Desirable
Desirable
Yes
Desirable
Yes
Yes
Yes
Yes
Yes
LABORATORY SAFETY EQUIPMENT
PRACTICES
GMT
None; open bench work
GMT plus
protective
clothing,
biohazard sign
As Level 2 plus
special
clothing,
controlled
access,
directional
airflow
As Level 3 plus
airlock entry,
shower exit,
special waste
Open bench plus BSC for potential
aerosols
BSC and/or other primary devices
for all activities
Class III BSC, or positive pressure
suits in conjunction with Class II
BSCs double ended autoclave
(through the wall), filtered air
BSC= Biological Safety Cabinet
GMT= Good Microbiological Techniques
Respiratory Outbreak, Toronto Version May 2007
34