recently reported on their findings in a Spanish cohort of 101 pediatric patients hospitalized for HMPV-related respiratory tract infections. Although this article is highly confirmative to their previous study 2 and our matched pairs analysis, 3 it contains some interesting observations and conclusions we would like to comment on. As in our study, the authors have not found a significant difference in the clinical severity between HMPV and RSV-infections, although they did not used a matched pairs approach 3 but a random sample of 95 hospitalized infants-aged <2 and documented as RSV-infected. In addition, they observed a high rate (29.7%) of coinfections with at least one other viral respiratory pathogen. In earlier publications, and also in the reviewer's comments to our own publication 3 such high rates of coinfections were called 'strikingly high.' Coinfections may be detected more frequently in hospitalized patients simply as a result of a more sophisticated diagnostic approach. 5,6 In our study 3 39 patients (44.3 %) with a coinfection were observed, none of them displayed a more severe clinical course than mono-infected patients. Moreover, other studies described striking coinfection rates of up to 70% of patients, although in smaller cohorts (e.g., Ref. [4]). Most interestingly, the coinfected cohort of patients observed by Garcia-Garcia et al. also did not show a more severe clinical course. This is in contrast to earlier observations (reviewed by Kahn [5] and Schildgen et al. [6]), but in high agreement with our study. 3 This controversially discussed observation consequently has to be investigated more in detail, probably by the use of an appropriate animal model system.
ACKNOWLEDGMENTS