 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ELECTRONIC ARTICLE |
* Division of Infectious Diseases, Department of Paediatrics, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada
Department of Paediatric Laboratory Medicine, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada
Department of Diagnostic Imaging, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Methods. A prospective investigational study protocol was established for the management of children with a presumptive diagnosis of suspect or probable SARS. All were ultimately classified as having probable SARS, suspect SARS, or another cause on the basis of their epidemiologic exposure, clinical and radiologic features, and results of microbiologic investigations.
Results. Twenty-five children were included; 10 were classified as probable SARS and 5 were classified as suspect SARS, and in 10 another cause was identified. The exposure consisted of direct contact with at least 1 adult probable SARS case in 11 children, travel from a World Health Organization-designated affected area in Asia in 9 children, and presence in a Toronto area hospital in which secondary SARS spread had occurred in 5 children. The predominant clinical manifestations of probable cases were fever, cough, and rhinorrhea. With the exception of 1 teenager, none of the children developed respiratory distress or an oxygen requirement, and all made full recoveries. Mild focal alveolar infiltrates were the predominant chest radiograph abnormality. Lymphopenia; neutropenia; thrombocytopenia; and elevated alanine aminotransferase, aspartate aminotransferase, and creatine kinase were present in some cases. Nasopharyngeal swab specimens were negative for the SARS-associated coronavirus by an in-house reverse transcriptase-polymerase chain reaction in all 25 children.
Conclusions. Our results indicate that SARS is a relatively mild and nonspecific respiratory illness in previously healthy young children. The presence of fever in conjunction with a SARS exposure history should prompt one to consider SARS as a possible diagnosis in children irrespective of the presence or absence of respiratory symptoms. Reverse-transcriptase polymerase chain reaction analysis of nasopharyngeal specimens seems to be of little utility for the diagnosis of SARS during the early symptomatic phase of this illness in young children.
Key Words: severe acute respiratory syndrome • reverse-transcriptase polymerase chain reaction • probable SARS • suspect SARS
Abbreviations: SARS, severe acute respiratory syndrome • WHO, World Health Organization • RT-PCR, reverse-transcriptase polymerase chain reaction • ALT, alanine aminotransferase • AST, aspartate aminotransferase • CPK, creatine kinase
Severe acute respiratory syndrome (SARS) was first described in an American businessman who was hospitalized in Hanoi, Vietnam, in late February 2003 by Dr Carlo Urbani, a World Health Organization (WHO) communicable diseases expert (sadly, Dr Urbani himself died of SARS on March 29, 2003).1 The WHO issued a global health alert describing this syndrome on March 12, 2003.2 By June 30, 2003, a total of 8447 probable cases of SARS were reported to the WHO from >30 countries.3 The majority of cases occurred in China (n = 5327), Hong Kong (n = 1755), Taiwan (n = 678), Canada (n = 252), and Singapore (n = 206). An overall mortality rate of approximately 9.5% is suggested by WHO surveillance data.
A novel coronavirus has been implicated in several clinical studies as the likely cause of SARS.4–8 In these studies, the SARS-associated coronavirus was isolated in Vero-cell culture and detected by reverse-transcriptase polymerase chain reaction (RT-PCR) in respiratory samples taken from patients with suspected SARS. Furthermore, the illness has now been reproduced in cynomolgus macaques inoculated with Vero cell-cultured SARS-associated coronavirus, and the virus subsequently isolated from these animals was found to be identical to that inoculated, as shown by negative-contrast electron microscopy and RT-PCR.9 Thus, the modified Koch’s postulates set forth by Rivers9,10 for viral diseases seem to have been satisfied. The genome of the SARS-associated coronavirus has now been fully sequenced, and phylogenetic analysis indicates that it is not closely related to the 3 previously known coronavirus groups.11,12
The clinical, laboratory, and radiologic features of SARS have been described principally in adults.4,5,8,13,14 There is a paucity of pediatric data; a preliminary report out of Hong Kong describing 10 cases of pediatric SARS suggested that younger children may have milder disease than teenagers and adults.15 The purpose of the present report is to describe the clinical, laboratory, and epidemiologic features of children who were admitted to the Hospital for Sick Children, Toronto, with a presumptive diagnosis of suspect or probable SARS.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
SARS-Related Definitions
Suspect SARS was defined by the presence of fever (>38°C orally) and a history of potential exposure to SARS during the 10 days preceding the onset of symptoms. A child was considered to have probable SARS when he or she fulfilled the criteria for suspect SARS and had chest radiograph findings suggestive of lower respiratory tract disease (mild peribronchial thickening was not considered evidence of lower respiratory tract disease). The WHO surveillance case definition requirement for cough or difficulty in breathing was removed because of early clinical experience indicating that SARS is a relatively mild illness in children and concern that some SARS cases would be missed if the WHO criteria were followed strictly.16 Potential exposure to SARS was defined by documented close contact with a person fulfilling the WHO definition of suspect or probable SARS or history of travel from a SARS-affected area in Asia or having been in a hospital in the greater Toronto area in which secondary SARS transmission had occurred.
During the course of hospitalization, children were reclassified into 3 groups (probable SARS, suspect SARS, and other cause) in a manner similar to that recently recommended by the WHO.16 A child was considered to have probable SARS when he or she fulfilled the aforementioned criteria for suspect SARS, had radiographic changes suggestive of lower respiratory tract disease or evidence of a severe progressive respiratory illness (including decreased oxygen saturation) suggestive of atypical pneumonia or acute respiratory distress syndrome, and no other potential causative agent was identified. In the absence of chest radiograph changes suggestive of lower respiratory tract disease or severe progressive respiratory illness, a child was classified as suspect SARS when no other potential causative agent was identified. The other cause category included all children for whom a firm non-SARS causative diagnosis was established.
Study Design and Protocol
A prospective investigational study protocol was established for the management of children who were admitted with a diagnosis of suspect or probable SARS by the Pediatric SARS Investigation Team. The study was approved by the Research Ethics Board of the Hospital for Sick Children. All children who fulfilled criteria for suspect or probable SARS were admitted and placed in single negative-pressure isolation rooms. Strict airborne and droplet infection control precautions were implemented for all cases; this included the use of an N-95 mask, face shield or goggles, cap, gown, and gloves for any patient contact. Direct contact with patients was restricted to a small number of SARS management team members. Parents of children hospitalized with suspect or probable SARS were not permitted to stay with or visit their children and were quarantined at home by public health authorities. No visitors were permitted. A record of all health care workers and support staff who came in contact with these patients was maintained in the patient’s chart.
A standard set of investigations, including a portable chest radiograph and laboratory tests, were performed on all patients at the time of admission. Routine laboratory investigations included a complete blood count and differential, venous blood gases, serum electrolytes, creatinine, urea, bilirubin, 
-glutamyl transferase, alanine aminotransferase (ALT), aspartate aminotransferase (AST), and creatine kinase (CPK). Follow-up complete blood counts were performed every 2 to 4 days on patients who received intravenous ribavirin. Follow-up chest radiographs and blood work, other than the complete blood counts in those on ribavirin, and any investigations not listed in the protocol were performed at the discretion of the managing physician.
Microbiologic investigations, performed within 24 hours of admission, included a nasopharyngeal swab for direct antigen detection and culture of common respiratory viruses (respiratory syncytial virus, influenza A and B, parainfluenza 1–3, and adenovirus), a nasopharyngeal swab for SARS-associated coronavirus RT-PCR, a throat swab for bacterial culture, a throat swab for Mycoplasma pneumoniae PCR, a stool sample for electron microscopy, and 2 blood cultures. Direct antigen detection for common respiratory viruses was performed by means of direct immunofluorescence microscopy, and culture for these same viruses was performed using AGMK, MDCK, and Hep-2 cell lines. M pneumoniae PCR was performed using the P1–1 and P1–3 primers for the P1 adhesin gene as previously described.17 Blood and throat swab bacterial cultures and stool for electron microscopy were performed using standard microbiologic methods.
RT-PCR for the SARS-associated coronavirus was performed using a modified version of a previously published assay.4 Primers were directed at a highly conserved region of the POL-1B gene of the coronaviruses. Reverse transcription was done using the primer 5'-GCATAGGCAGTAGTTGCATC-3', and PCR was done using the primers 5'-TGATGGGTTGGGACTATCCTAAATGTGA-3' and 5'-GTAGTTGCATCACCGGAAGTTGTGCCACC-3'. The expected amplicon size was 220 bp. The reverse transcription master mix was prepared by mixing 4 µL of reverse transcription buffer, 0.5 µL of RNasin (Promega, Mississauga, Ontario), 2 µL of dithiothreitol (100 mM), 1 µL dNTP mix (10 mM each), 1.5 µL of the coronavirus primer, and 1 µL of Superscript II (Life Technologies, Burlington, Ontario). The reverse transcription reaction was conducted by heating a 10-µL RNA aliquot to 65°C for 2 minutes, cooling it on ice, adding 10 µL of the reverse transcription reaction mix, and incubating at 42°C for 60 minutes. The PCR reaction was performed in a total volume of 50 µL containing 5 µL of reaction buffer, 1.25 µL of dNTP mix (10 mM each), 2 µL of each PCR primer (10 µm stock), 1 µL of Advantage 2 polymerase mix (Clontech, Mississauga, Ontario), 28.75 µL of double-distilled H2O, and 10 µL of reverse transcription reaction mixture, all of which was overlaid with mineral oil. The PCR was done using a Robocycler 40 thermal cycler (Stratagene, La Jolla, CA). Thirty-seven cycles, each consisting of denaturation at 99°C for 35 seconds, annealing at 50°C for 90 seconds, and elongation at 68°C for 90 seconds, were performed. Reactions were analyzed by agarose gel electrophoresis.
The analytical sensitivity of the assay was measured using a serial dilution of RNA extracted from a lung biopsy sample of a patient with SARS. The viral load of the lung tissue had been quantified with the Real Art HPA-coronavirus RT-PCR (Artus, San Francisco, CA), and the analytical sensitivity of our RT-PCR assay was estimated at approximately 10 genome copies. The identity of the amplicon was verified by cleavage with the restriction enzyme Alu I and/or by sequencing the amplicon.
Clinical samples were run in batches. Each batch included a positive control consisting of an aliquot of serially diluted RNA from a lung tissue infected with the SARS-associated coronavirus (see above), which was reextracted and subjected to RT-PCR, 1 negative control consisting of double-distilled water used as a template for RNA extraction and RT-PCR, and 1 negative control consisting of double-distilled water used as a template for RT-PCR. Strict adherence to previously described precautions against PCR contamination was maintained.18
Ribavirin therapy was considered for all children who were admitted with a presumptive diagnosis of suspect or probable SARS, particularly in the presence of chest radiograph abnormalities and/or a strong history of close contact with 1 or more probable SARS cases. Dosing of intravenous ribavirin was based on the dosing regimen recommended for several viral hemorrhagic fever syndromes.19–21 The recommended dosing schedule included a loading dose of 33 mg/kg, followed by 16 mg/kg administered every 6 hours for a total of 4 days (16 doses), and then 8 mg/kg administered every 8 hours for a total of 3 days (9 doses). Standard antimicrobial therapy for community-acquired pneumonia or other focal infections was administered at the discretion of the treating physician.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
The epidemiologic exposure as well as clinical and laboratory features of children with probable and suspect SARS are illustrated in Tables 2 and 3. In general, the clinical manifestations of children in whom a non-SARS cause was identified did not differ substantially from children with probable and suspect SARS. Fever (38.0°C or higher) was the initial clinical manifestation in 7 probable (70%), 5 suspect (100%), and 9 other cause (90%) cases. Respiratory symptoms occurred in 8 probable (80%), 3 suspect (60%), and 7 other cause cases (70%). The onset of respiratory symptoms preceded fever in 3 probable cases (30%) and coincided with the onset of fever in 2 probable (20%) and 6 other cause cases (60%). Headache, lethargy, vomiting, and diarrhea were observed in a minority of cases. Irritability and myalgia were not identified in any of the children in this cohort.
|
|
|
|
Antimicrobial Therapy
Ribavirin was administered intravenously to 10 children (1 also received aerosolized ribavirin). In all but 2, the decision to initiate ribavirin was based on the finding of chest radiograph infiltrates and a preliminary diagnosis of probable SARS. The 2 children who received ribavirin despite normal chest radiographs had heavy household exposure to >1 probable SARS case (patient 3, Table 2, and patient 2, Table 3). Of the 10 children initiated on ribavirin therapy, 8 were ultimately classified as probable SARS, 1 as suspect SARS, and 1 as other cause. The duration of ribavirin therapy ranged from 2 to 10 days (median 7 days). Hemolytic anemia occurred in 1 probable case (patient 1, Table 2), but this was of mild degree and did not necessitate discontinuation of the drug. No other ribavirin-associated adverse effects were noted. Corticosteroids were not administered to any of the children in this cohort. A full course of antibiotic therapy (clarithromycin) for community-acquired pneumonia was given to 6 probable cases and 1 other cause case.
Course in Hospital and Outcome
The clinical course of most children was mild and brief in duration. Only 1 child (a teenager) with probable SARS developed respiratory distress and required oxygen supplementation (patient 1, Table 2). None of the children required ventilatory support or intensive care admission. The duration of fever ranged from 1 to 6 days (median: 1 day) for children with probable SARS, 1 to 3 days (median: 1 day) for children with suspect SARS, and 1 to 4 days (median: 2 days) for children with other causes. The duration of cough ranged from 0 to 10 days (median: 2 days) for children with probable SARS, 0 to 3 days (median: 1 day) for children with suspect SARS, and 0 to 4 days (median: 1 day) for children with other causes. The median duration of hospitalization for the entire cohort was 4 days (range: 1–25 days). For both probable and suspect cases, the duration of hospitalization ranged from 2 to 25 days (median: 8 days and 3 days, respectively). Two siblings, 1 with probable SARS and 1 with suspect SARS, were hospitalized for 25 days principally as a result of their parents’ prolonged hospitalization. The older of these 2 children (patient 3, Table 2) had a second febrile episode that lasted 3 days beginning 11 days into his admission; he had no other symptoms during this episode, but a patchy right lower lobe infiltrate was demonstrated.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Two children in our cohort, classified as probable SARS because of household exposure, fever, and the presence of chest radiograph infiltrates, lacked respiratory symptoms throughout the course of their illness. Cough occurred in only 6 of 10 probable SARS cases. The chest radiograph changes of 1 of these children, who had prolonged household exposure to SARS (both parents with probable SARS), became apparent only during a second febrile episode that developed 11 days into hospitalization (patient 3, Table 2). His younger sister (patient 2, Table 3) had fever, mild respiratory symptoms, and a normal chest radiograph. These observations suggest that the WHO surveillance case definition may not be sufficiently sensitive for young children. Specifically, the requirement for respiratory symptoms (cough or difficulty breathing) to qualify as a suspect case and the requirement for progressive respiratory disease or radiographic abnormalities to qualify as a probable case may not be appropriate in the pediatric setting. Our results suggest that, in children, the presence of fever in conjunction with an appropriate exposure history should lead one to consider the possibility of SARS. If SARS were to become an endemic disease, it would be necessary to treat any child with unexplained fever as a potential case of SARS until proved otherwise.
The relatively mild nature of SARS in young children raises intriguing questions regarding the pathophysiology and spread of this disease. In particular, the role of the host immune response to the offending agent in the disease process requires investigation. The demonstration of bilateral peripheral air-space ground-glass consolidation reminiscent of bronchiolitis obliterans organizing pneumonia (on computed tomography scanning) and the similarity of histologic features (pulmonary edema with hyaline membrane formation) of SARS to those of early adult respiratory distress syndrome are consistent with a disease that has an immune-mediated component.14 It was, in part, these observations that led physicians in Hong Kong to use corticosteroids in addition to antimicrobial therapy in the management of SARS.14
Preliminary data suggest that children pose a lower risk of transmitting SARS than do adults.15 In a report out of Hong Kong, there was no evidence of transmission by 8 children who had probable SARS and whose symptoms began while they were attending school. In our institution, no health care workers contracted SARS, and, as far as we know, there has been no documented transmission of SARS to health care workers from children. Whether this is attributable to a lower viral burden or to other factors remains to be determined.
Early in the epidemic, intravenous ribavirin was advocated in the treatment of adults and children with SARS in part because of its broad-spectrum in vitro activity against RNA viruses19,22–24 and in part because of its efficacy in treatment of some viral hemorrhagic fever syndromes.20,21 However, preliminary in vitro susceptibility testing undertaken by Health Canada’s National Microbiology Laboratory and the US Army Medical Research Institute of Infectious Diseases suggests that ribavirin has no demonstrable activity against the SARS-associated coronavirus at concentrations effective for Lassa fever virus and other hemorrhagic fever viruses.25 In addition, there is no convincing evidence that ribavirin is effective in reducing intensive care admissions, need for ventilatory support, or mortality among adult SARS patients.13 At present, ribavirin is not recommended for the routine management of SARS in Canada.25 The potential role of aerosolized ribavirin remains undefined.
The RT-PCR used in this cohort is an improved version of the coronavirus RT-PCR previously developed in our laboratory.4 The primer sequences have been slightly modified, taking into account the recently published SARS-associated coronavirus genomic sequence.11 The sensitivity of the new version of the test has been improved by 3 orders of magnitude. The original version was able to detect the presence of the SARS-associated coronavirus in bronchoalveolar lavage and/or lung biopsy specimens of some severely affected adult SARS patients.4,26 In a Hong Kong cohort, the presence of SARS-associated coronavirus RNA was detected in nasopharyngeal specimens of 50% (22 of 44) of adults with probable SARS using a different RT-PCR assay.8 In the only pediatric study published to date, the SARS-associated coronavirus was detected in nasopharyngeal specimens of 4 children, 3 of whom were teenagers.15 The failure to detect coronavirus by RT-PCR in nasopharyngeal swab specimens in our cohort of predominantly young children may be related to poor sensitivity of the assay in the setting of mild disease (low viral load). Alternatively, it is conceivable that the SARS-associated coronavirus is only transiently present in the upper airway. The possibility that none of the children in our cohort was infected with the SARS-associated coronavirus cannot be excluded with certainty in the absence of serologic data, but this possibility seems extremely unlikely in view of the strong epidemiologic link of many of these cases to SARS. An intriguing observation is the recent finding that the SARS-associated coronavirus can be detected by RT-PCR in the blood, stool, and urine of some affected patients; the ultimate role that this will play in routine diagnosis remains to be seen.
The main limitation of the present study was the reliance on clinical and epidemiologic criteria to classify children as to their likelihood of having SARS. It is possible that some of the children in our cohort who were classified as probable SARS did not in fact have this disease, whereas others, classified as suspect SARS, may have been true SARS cases. It is also possible that some of the children in the other cause category had co-infections with the SARS-associated coronavirus. Despite the limitations of our classification strategy, we believe that its utility is supported by the fact that the majority of those with the strongest epidemiologic links to SARS were ultimately classified as probable SARS cases, whereas most of those with weak epidemiologic links ended up in the other cause category. Notwithstanding this observation, it will be important to revisit our classification strategy once reliable serologic assays become available. In this regard, we have been collecting and storing acute and convalescent blood samples for future analysis.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Reprint requests to (S.R.) Division of Infectious Diseases, Hospital for Sick Children, 555 University Ave, Toronto, Ontario, M5G 1X8, Canada. E-mail: sread{at}sickkids.ca
Other members of the Hospital for Sick Children SARS Investigation Team include Alex Ahmet, Abdulla Al Junaibi, Karen Breen-Reid, Carol Goldman, Ra Han, Megan Harrison, Debra Louch, Patricia Malloy, Robyn Salter-Goldie, Laurie Streitenberger, Margaret Wilson, and Rick Wray.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  V. C. C. Cheng, S. K. P. Lau, P. C. Y. Woo, and K. Y. Yuen Severe Acute Respiratory Syndrome Coronavirus as an Agent of Emerging and Reemerging Infection Clin. Microbiol. Rev., October 1, 2007; 20(4): 660 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. F. Koller, D. B. Nicholas, R. S. Goldie, R. Gearing, and E. K. Selkirk When Family-Centered Care Is Challenged by Infectious Disease: Pediatric Health Care Delivery During the SARS Outbreaks Qual Health Res, January 1, 2006; 16(1): 47 - 60. [Abstract] [PDF]
|
|
|
|
 |
|
 A M Li and P C Ng Severe acute respiratory syndrome (SARS) in neonates and children Arch. Dis. Child. Fetal Neonatal Ed., November 1, 2005; 90(6): F461 - F465. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F W T Cheng, P C Ng, W K Chiu, W C W Chu, A M Li, K L Lo, E K L Hon, E A S Nelson, T F Leung, W H Ng, et al. A case-control study of SARS versus community acquired pneumonia Arch. Dis. Child., July 1, 2005; 90(7): 747 - 749. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L.-Y. Chang, F.-Y. Huang, Y.-C. Wu, I.-J. Su, N.-C. Chiu, K.-T. Chen, H.-S. Wu, T.-H. Lin, S.-F. Peng, C.-L. Kao, et al. Childhood Severe Acute Respiratory Syndrome in Taiwan and How to Differentiate It From Childhood Influenza Infection Arch Pediatr Adolesc Med, November 1, 2004; 158(11): 1037 - 1042. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. D Parashar and L. J Anderson Severe acute respiratory syndrome: review and lessons of the 2003 outbreak Int. J. Epidemiol., August 1, 2004; 33(4): 628 - 634. [Full Text] [PDF]
|
|
|
|
 |
|
 D S C Hui, M C H Chan, A K Wu, and P C Ng Severe acute respiratory syndrome (SARS): epidemiology and clinical features Postgrad. Med. J., July 1, 2004; 80(945): 373 - 381. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Other articles noted: 14 Nov 2003 to 30 Jan 2004 Evid. Based Nurs., April 1, 2004; 7(2): e2 - e2. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Wallace Studies indicate disease was mild in children: SARS baffles researchers AAP News, October 1, 2003; 23(4): 145 - 152. [Full Text] [PDF]
|
|
Read all P3Rs
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2003 American Academy of Pediatrics. All rights reserved. |
||
|
||
