Hand, foot and mouth disease in Singapore: a comparison of fatal and non-fatal cases C-Y Chong 1 , K-P Chan 2 , VA Shah 1 , WYM Ng 1 , G Lau 3 , TE-S Teo 3 , S-H Lai 3 and A-E Ling 2 Department of Paediatric Medicine 1 , KK Women’s and Children’s Hospital; Department of Pathology 2 , Singapore General Hospital; Centre for Forensic Medicine 3 , Health Sciences Authority, Singapore Chong CY, Chan KP, Shah V A, Ng WYM, Lau G, Teo TES, Lai SH, Ling AE. Hand, foot and mouth disease in Singapore: a comparison of fatal and non-fatal cases. Acta Pædiatr 2003; 92: 1163–1169. Stockholm. ISSN 0803-5253 Aim: An epidemic of hand, foot and mouth disease (HFMD) occurred in Singapore between September and November 2000. During the epidemic, there were four HFMD-related deaths and after the epidemic, another three HFMD-related deaths. This study sought to determine the risk factors predictive of death from HFMD disease. Methods: The risk factors for fatal HFMD were determined by comparing clinical and laboratory findings between fatal cases (n = 7) and non-fatal controls (n = 131) admitted between September 2000 and April 2001. Enterovirus 71 positive fatal cases (n = 4) and non-fatal controls (n = 63) were also compared. Results: In total, 138 HFMD cases with a mean age of 32 mo were studied. The majority of fatal cases died from interstitial pneumonitis, of whom three also had brainstem encephalitis. Of the 131 non-fatal cases, 3 had concomitant infections (respiratory syncytial virus bronchiolitis, right-sided pneumonia, Haemo- philus influenzae type b meningitis), 2 had aseptic meningitis, and 1 each had transient drowsiness, intravenous immunoglobulin-related complications and transverse myelitis. By multivariate logistic regression analysis, atypical physical findings (p = 0.0006), raised total white cell count (p = 0.0128), vomiting (p = 0.0116) and absence of mouth ulcers (p = 0.043) were predictive of a fatal course. Although previous epidemics have described neurogenic pulmonary oedema as the main cause of death, the fatal cases in this study died mainly from interstitial pneumonitis alone or with myocarditis or encephalitis. Conclusion: Although HFMD is generally a benign disease, risk factors such as vomiting, absence of mouth ulcers, atypical presentation and raised total white cell count should alert the physician of a fatal course of illness. Key words: Encephalitis, enterovirus, hand, foot and mouth disease, interstitial pneumonitis CY Chong, Department of Paediatric Medicine, KK Women’s and Children’s Hospital, 100 Bukit Timah Road, Singapore 229899, Republic of Singapore (Tel. 65 3941 964, fax. 65 3941 973, e-mail. cychong@kkh.com.sg) Epidemics of hand, foot and mouth disease (HFMD) have been associated with deaths, as previously seen in the Taiwan and Malaysian outbreaks during which 78 and 34 deaths were recorded, respectively (1, 2). From September to November 2000, Singapore experienced an HFMD epidemic during which 3526 HFMD cases were notified, 652 cases were seen at KK Women’s and Children’s Hospital Emergency and 129 (20%) were admitted (3). During the epidemic, four HFMD-related deaths occurred, whereas after the epidemic, another three deaths occurred between Jan- uary and February 2001. This study sought to discover the predictive factors for death related to HFMD in comparison with non-fatal HFMD admissions. Methods and definitions Fatal and non-fatal HFMD cases were enrolled between September 2000 and April 2001. Classical HFMD was defined as a case of vesicular–papular rash over the hands, soles and/or buttocks with mouth ulcers. Atypical HFMD was defined as a maculopap- ular rash over the hands, soles and/or buttocks with or without mouth ulcers. Herpangina was defined as oral ulcers without any skin rash. A fatal case was defined as death as a result of HFMD disease and unrelated to other causes. Fatalities due to enterovirus 71 (EV71) but not associated with HFMD disease were excluded as they have been previously described (4, 5). A non- fatal control was defined as HFMD that did not result in death. Data comprising clinical characteristics on admission and the day of clinical deterioration, viro- logical results and postmortem findings were collected. As the numbers of herpangina cases were low, they were combined with atypical HFMD for statistical analysis. Virological methods have been described previously (4). An enterovirus was classified as  2003 Taylor & Francis. ISSN 0803-5253 Acta Pñdiatr 92: 1163±1169. 2003 DOI 10.1080/08035250310005242