Background. Adenovirus (ADV) infections were difficult to diagnose in the past, and many infections were unrecognized. Direct fluorescent assay (DFA) for the rapid diagnosis of ADV infection, as part of a viral respiratory panel, became available at Primary Children’s Medical Center (Salt Lake City, UT) in December 2000.
Objective. To describe children with ADV infection diagnosed by DFA and viral culture and document the impact of rapid ADV testing on patient care.
Methods. DFA testing for respiratory viruses including ADV was performed on nasal wash specimens with parallel viral culture. Chart review was performed for all ADV-positive patients identified from microbiology records between December 2000 and May 2002.
Results. Of 1901 patients positive for respiratory viruses, 143 (7.5%) were ADV-positive by DFA or culture. The mean age of ADV-positive children was 23 months; 90% were ≤60 months old. Eighty percent were previously healthy, and 56% required admission with a mean length of stay of 3.4 days. The most common diagnoses included fever (31%), bronchiolitis (24%), and pneumonia (14%). Other conditions included suspected Kawasaki disease (KD) and hepatitis. Forty-six percent of ADV-positive children were given antibiotics at presentation, but only 2 (1.4%) had documented bacterial infection (one had Escherichia coli urinary tract infection and one had Moraxella catarrhalis bacteremia). Thirty-six percent of children had a change in management based on positive ADV DFA. In children with suspected KD (n = 5), 100% had positive ADV DFA, and immune globulin was withheld in 4. One immunocompetent patient with fulminant liver failure received cidofovir treatment after a positive ADV DFA and recovered before liver transplant.
Conclusions. ADV is a common infection in young children and often results in admission and unnecessary antibiotic therapy. Identifying ADV as the cause of illness can favorably impact care and in some instances may be life-saving. DFA testing for ADV should be considered for infants and children requiring admission for fever, respiratory illness, suspected KD, and hepatitis.
Comments
Adenoviral Infections in Children with Suspected Kawasaki Disease
In Reply.---
We appreciate the letter submitted by Drs. Rowley and Shulman regarding our article on adenoviral infections in children (1) and the opportunity to discuss some of the important issues related to adenovirus (ADV) infection and suspected Kawasaki disease (KD).
Drs. Rowley and Shulman point out that asymptomatic shedding of adenoviruses may occur following primary infection, especially with adenovirus type 1 or 2 acquired in the first year of life. (2, 3) We are aware of this data. We would point out, however, that at this time there are 49 distinct types of ADV recognized in humans. Infection in susceptible hosts may occur at any time and community outbreaks of ADV infection occur. (4-6) In the 1977 study mentioned by Drs Rowley and Shulman, JP Fox and colleagues noted that in susceptible subjects with respiratory shedding of ADV, 65% were clinically ill and the majority of illness was febrile. (3) In the Fox studies, detection of ADV was done by viral culture. (3) This method is known to be more sensitive than direct fluorescent assay (DFA) used in our study. (7) It is possible that respiratory shedding detected by DFA may be more likely to be related to acute illness with high viral loads than shedding detected by culture.
Drs. Rowley and Shulman express concern that we did not confirm ADV infections in study subjects with serology. We would remind them that this was a retrospective study and that we did not have access to acute and convalescent serum samples. Even if serologic testing had been possible, interpretation of results may be difficult. The 49 known ADV types are immunologically distinct and are categorized into 4 subgroups. However, heterotypic antibody responses may occur for adenoviruses within the same subgroup and caution is needed in interpreting type specific antibody results. (8) The sensitivity of complement fixation tests, the most widely available, is low and more sensitive tests are not standardized for commercial use. (8) Finally in the acute setting, serology is unlikely to be useful in helping to establish whether a febrile episode is related to ADV or not, since 2 specimens obtained 2-4 weeks apart are needed to confirm diagnosis.
Regarding platelet counts in the patients presented, the authors are well aware that many inflammatory conditions, including infectious diseases may elevate platelet counts. In our study, 35% of children who were diagnosed with ADV infection and in whom there was no suspicion of KD had platelet counts over 400,000 mm3 and 18% had platelet counts over 500,00 mm3.
The most important concern expressed by Drs. Rowley and Shulman is the possibility of children with KD being misdiagnosed as having adenoviral infection if asymptomatic shedding of ADV was occurring. We share this concern and take it seriously. We performed a review of the literature prior to publication of our study to try and find any evidence of ADV infection and KD occurring simultaneously and could find only 1 report where the possibility of both entities was considered to be possible. (9) We believe more study is needed in this area.
In the US, where measles immunization is nearly universal, the most common viral mimic of KD will likely be ADV infection. The use of DFA testing, which can yield results quickly (4 hours in our institution), compared to culture (9 days), or serologic testing (> 4weeks) can provide information that can be used to inform the differential diagnosis of the patient. (1, 10)
We have used DFA testing for the last 4 years to perform viral surveillance in our community.(11) This has been the gold- standard in pediatrics for recognizing respiratory syncytial virus (RSV) season. By using a DFA that tests for multiple common respiratory viruses, we have been able to track not only RSV, but also influenza, parainfluenza, and adenovirus in our community. Four out of five of the patients we presented with suspected KD and a positive DFA for ADV were diagnosed between January and April of 2002, during a peak in ADV activity in our community. (1)
DFA testing should be used in context with the patient’s clinical presentation, contacts, viral activity in the community, etc. to help the physician decide whether KD is the most likely diagnosis. Obviously, if KD cannot be excluded, treatment should be initiated. In a separate article, Dr. Rowley notes that the overdiagnosis of KD seems less harmful than under-diagnosis. (12) This may be correct, but as we pointed out in our paper, the treatment of KD is not without risk (13-16) and the overdiagnosis may be associated with long term consequences such as lower perceived health, parental anxiety and the vulnerable child syndrome. (17) As physicians, we must do all that we can to assure the diagnosis of KD is correct. The DFA test for ADV provides a timely piece of information that can be added to the decision-making process.
Finally, we would like to comment on the area of KD research. We believe that it is critical that prospective research studies be performed that can address the issues of ADV as a mimic of KD. As we stated in our article, “ the prospective study of ADV infection in children with suspected KD could help clarify important issues such as how often ADV and KD occur together, whether asymptomatic shedding of ADV occurs in children with KD, and which serotypes of ADV mimic KD.” A review of recent large studies of children with diagnosed KD, including studies by Drs. Rowley and Shulman, all include the specific clinical criteria used to diagnose KD, but none mentions standard testing of children to exclude those with possible bacterial or viral infections. (18-22) A large study performed in 1991 that examined mimics of KD, noted that in nearly 50% of patients with suspected KD, but with another diagnosis established, fulfilled all standard diagnostic criteria for KD. (23) Our understanding of KD, especially in children without coronary artery involvement may be jeopardized if children with ADV infection are inadvertently included in the study populations. We stand by our conclusion that all children with suspected KD in the US be tested for ADV infection.
Sincerely,
Carrie L. Byington, MD Associate Professor of Pediatrics and Infectious Diseases University of Utah
Christian Rocholl, MD Fellow in Pediatric Emergency Medicine Brown University
References
1. Rocholl C, Gerber K, Daly J, Pavia AT, Byington CL. Adenoviral infections in children: the impact of rapid diagnosis. Pediatrics. Jan 2004;113(1 Pt 1):e51-56.
2. Fox JP, Brandt CD, Wassermann FE, et al. The virus watch program: a continuing surveillance of viral infections in metropolitan New York families. VI. Observations of adenovirus infections: virus excretion patterns, antibody response, efficiency of surveillance, patterns of infections, and relation to illness. Am J Epidemiol. Jan 1969;89(1):25-50.
3. Fox JP, Hall CE, Cooney MK. The Seattle Virus Watch. VII. Observations of adenovirus infections. Am J Epidemiol. Apr 1977;105(4):362-386.
4. Hong JY, Lee HJ, Piedra PA, et al. Lower respiratory tract infections due to adenovirus in hospitalized Korean children: epidemiology, clinical features, and prognosis. Clin Infect Dis. 2001;32(10):1423-1429.
5. Two fatal cases of adenovirus-related illness in previously healthy young adults--Illinois, 2000. MMWR Morb Mortal Wkly Rep. 2001;50(26):553-555.
6. Civilian outbreak of adenovirus acute respiratory disease-- South Dakota, 1997. MMWR Morb Mortal Wkly Rep. 1998;47(27):567-570.
7. Cherry JD. Adenoviruses. In: Feigin R, Cherry J, Demmler G, Kaplan S, eds. Textbook of Pediatric Infectious Diseases. Vol 2. Philadelphia: Saunders; 2004:1843-1863.
8. Demmler GJ. Adenoviruses. In: Long S, Pickering L, Prober C, eds. Principles and Practice of Pediatric Infectious Diseases. New York: Churchill Livingstone; 2003:1076-1080.
9. Embil JA, McFarlane ES, Murphy DM, Krause VW, Stewart HB. Adenovirus type 2 isolated from a patient with fatal Kawasaki disease. Can Med Assoc J. 1985;132(12):1400.
10. Barone SR, Pontrelli LR, Krilov LR. The differentiation of classic Kawasaki disease, atypical Kawasaki disease, and acute adenoviral infection: use of clinical features and a rapid direct fluorescent antigen test. Arch Pediatr Adolesc Med. 2000;154(5):453-456.
11. Daly J, Gerber K, Pavia AT. Viral Surveillance at Primary Childrens Medical Center. Available at: http:// www.ped.med.utah.edu/. Access General Information and Follow links to Infectious Diseases.
12. Rowley AH. Incomplete (atypical) Kawasaki disease. Pediatr Infect Dis J. Jun 2002;21(6):563-565.
13. Bresee JS, Mast EE, Coleman PJ, et al. Hepatitis C virus infection associated with administration of intravenous immune globulin. A cohort study. JAMA. Nov 20 1996;276(19):1563-1567.
14. Berger A, Scharrer I, Doerr HW, Hess G, Weber B. Infection with hepatitis G virus in immunoglobulin recipients. Lancet. Jan 18 1997;349(9046):207.
15. Lefrere JJ, Ravera N, Corbi C, Mariotti M, Loiseau P. Infection with hepatitis G virus in immunoglobulin recipients. Lancet. Jan 18 1997;349(9046):206.
16. Matsubara T, Mason W, Kashani IA, Kligerman M, Burns JC. Gastrointestinal hemorrhage complicating aspirin therapy in acute Kawasaki disease. J Pediatr. May 1996;128(5 Pt 1):701-703.
17. Baker AL, Gauvreau K, Newburger JW, Sundel RP, Fulton DR, Jenkins KJ. Physical and psychosocial health in children who have had Kawasaki disease. Pediatrics. Mar 2003;111(3):579- 583.
18. Burns JC, Capparelli EV, Brown JA, Newburger JW, Glode MP. Intravenous gamma-globulin treatment and retreatment in Kawasaki disease. US/Canadian Kawasaki Syndrome Study Group. Pediatr Infect Dis J. Dec 1998;17(12):1144-1148.
19. Dengler LD, Capparelli EV, Bastian JF, et al. Cerebrospinal fluid profile in patients with acute Kawasaki disease. Pediatr Infect Dis J. Jun 1998;17(6):478-481.
20. Quasney MW, Bronstein DE, Cantor RM, et al. Increased frequency of alleles associated with elevated tumor necrosis factor-alpha levels in children with Kawasaki disease. Pediatr Res. May 2001;49(5):686-690.
21. Leung DY, Meissner HC, Shulman ST, et al. Prevalence of superantigen-secreting bacteria in patients with Kawasaki disease. J Pediatr. Jun 2002;140(6):742-746.
22. Shingadia D, O'Gorman M, Rowley AH, Shulman ST. Surface and cytoplasmic immunoglobulin expression in circulating B- lymphocytes in acute Kawasaki disease. Pediatr Res. Oct 2001;50(4):538-543.
23. Burns JC, Mason WH, Glode MP, et al. Clinical and epidemiologic characteristics of patients referred for evaluation of possible Kawasaki disease. United States Multicenter Kawasaki Disease Study Group. J Pediatr. May 1991;118(5):680-686.
The Diagnosis of Kawasaki Disease is not Excluded by a Positive Respiratory DFA for Adenovirus
January 7, 2004
Jerold F. Lucey Editor, Pediatrics
To the Editor,
With regard to “Adenoviral Infections in Children: The Impact of Rapid Diagnosis” by Rocholl et al (1), published in the electronic pages of the January 2004 issue of Pediatrics, several important aspects of adenovirus biology were not considered by the authors. The failure to recognize and address these facts seriously flaws their study, particularly with respect to the recommendations that the authors propose for the evaluation of children with suspected Kawasaki Disease.
Epidemiologic and clinical aspects of childhood adenovirus infections were elegantly studied as a part of the comprehensive “Virus Watch” program funded by the National Institute of Health, National Institute of Allergy and Infectious Diseases, in the 1960s and 1970s and carried out by John P. Fox and colleagues (1,2). In these studies, a young child or infant in a family served as the index case, and biweekly respiratory and fecal specimens were obtained from the index case and interval history data were collected. Additional visits were made when any family member became ill, at which time the index patient and all available family members were cultured. Serology was performed on admission to the study, and on a routine schedule thereafter, as well as with an acute illness. Tens of thousands of respiratory cultures were performed in this study (2,3).
Several important aspects of adenovirus biology in children were elucidated by these studies: 1) primary adenovirus infection is almost universal in children aged 0-1 years, with 97-100% of children in this age group experiencing primary infection with adenovirus type 1 or 2; 2) adenoviruses comprised the largest number of isolates of all the viruses cultured in the Virus Watch studies (including rhinovirus, herpesviruses, enteroviruses, RSV, influenza, parainfluenza, and others); and 3) most positive adenovirus cultures occur in children without serologic evidence of primary infection or reinfection, because prolonged, intermittent adenovirus shedding may go on for years following primary infection. An important implication of these studies, as stated by Fox et al, is that for fecal or respiratory cultures, “without serologic support, recovery of an adenovirus from a patient cannot be regarded as firm evidence of a newly acquired infection”(2).
In the study by Rocholl et al, serologic studies were not performed, and every positive test for adenovirus by DFA was apparently considered indicative of an acute infection causing the child’s clinical symptoms. Of most concern to us, the authors used a positive respiratory DFA for adenovirus to exclude a diagnosis of Kawasaki Disease in children with compatible clinical and laboratory features of the illness. We believe that this misinterpretation of adenovirus biology could have potentially devastating consequences for a child with Kawasaki Disease.
It is known that Kawasaki Disease can mimic primary adenovirus infection, just as it can mimic a variety of other infectious and inflammatory diseases (4, 5). In a patient with exudative pharyngitis and exudative conjunctivitis, adenovirus infection is the more likely diagnosis. However, in a child with findings suggesting Kawasaki disease, a positive respiratory DFA or shell vial assay for adenovirus does not exclude the diagnosis.
We also would like to remind Rocholl et al that a normal platelet count during the first week of fever is characteristic of Kawasaki Disease; it is well-recognized that platelet counts do not rise until the second to third week of illness (4,5). In fact, a platelet count of 468,000/mm3 seen on day 8 of illness for patient 3 in Table 3 of the paper by Rocholl et al would be typical of the beginning rise of the platelet count in a child with Kawasaki Disease. The patients in Table 3 who had compatible clinical findings of Kawasaki Disease but were not treated could be at high risk of developing coronary artery sequelae.
Sincerely,
Anne H. Rowley, M.D. Professor of Pediatrics and of Microbiology/Immunology Feinberg School of Medicine, Northwestern University Pediatrics W140, Ward 12-204 303 E Chicago Avenue Chicago, IL 60611 a-rowley@northwestern.edu
Stanford T. Shulman, M.D. Professor of Pediatrics Feinberg School of Medicine, Northwestern University Chief, Division of Infectious Diseases Children's Memorial Hospital 2300 Children's Plaza, Box # 20 Chicago, IL 60614 sshulman@northwestern.edu
References:
1) Rocholl C, Gerber K, Daly J, Pavia AT, Byington CL. Adenoviral infections in children: the impact of rapid diagnosis. Pediatrics 2004; 113:e51. 2) Fox JP, Brandt CD, Wassermann FE, Hall CE, Spigland I, Kogon A, Elveback LR. The Virus Watch Program: A Continuing Surveillance of Viral Infections in Metropolitan New York Families. VI. Observations of Adenovirus Infections: Virus Excretion Patterns, Antibody Response, Efficiency of Surveillance, Patterns of Infection, and Relation to Illness. American Journal of Epidemiology 1969: 89:25-50. 3) Fox JP, Hall CE, Cooney MK. The Seattle Virus Watch. VII. Observations of Adenovirus Infections. American Journal of Epidemiology 1977;105: 362- 386. 4) Rowley AH, Shulman ST. Kawasaki Disease. In: Nelson Textbook of Pediatrics, 17th edition. Behrman RE, Kliegman RM, Jenson HB (eds). WB Saunders Company, 2004, pp. 823-826. 5) Shulman ST. Kawasaki Disease. In: Textbook of Pediatric Infectious Diseases, 5th edition. Feigin RD, Cherry J, Demmler G, Kaplan S (eds). WB Saunders Company, 2003, pp. 1055-1074.