Human adenovirus infection in Kawasaki disease: a confounding bystander?

BACKGROUND Human adenovirus (HAdV) infection mimics Kawasaki disease (KD) but can also be detected in KD patients. Evidence suggests that HAdV-C species can persist in pediatric adenoids and/or tonsils. We sought to determine (1) the frequency of HAdV detection by real-time polymerase chain reaction in KD patients, (2) the differences in HAdV semiquantitative nasopharyngeal viral loads between KD patients with detectable HAdV vs those with HAdV disease, and (3) whether nasopharyngeal HAdV-C shedding is occurring in KD. METHODS From August 2009 through April 2011, HAdV-positive patients were identified in 1 of the following groups: group I, complete or incomplete KD as defined by the American Heart Association (AHA); group II, treated for incomplete KD but not fulfilling AHA criteria; and group III, otherwise healthy children with some KD-like features ultimately diagnosed with HAdV disease. RESULTS Among 77 KD patients diagnosed, 8.8% (5/57) of group I and 25% (5/20) of group II KD patients had HAdV detected. Viral loads were significantly lower in group I (n = 5) vs group III (n = 26; P = .034). Of the 13 specimens available for HAdV typing, 7 of 7 group III and 1 of 3 group II specimens were determined to be HAdV-B using viral culture. The remaining 5 KD samples were unable to be cultured and molecular typing showed either HAdV-C (n = 3) or were nontypeable (n = 2). CONCLUSIONS In KD, molecular-based HAdV detection is not uncommon, may represent persistence of HAdV-C, and should be interpreted with caution. Together, quantitative polymerase chain reaction and HAdV typing may aid in distinguishing HAdV disease mimicking KD from KD with concomitant HAdV detection.