The article by Jennings and colleagues1 described interesting findings regarding the common nature of mixed viral/bacterial aetiology in patients with community acquired pneumonia (CAP) and the association between mixed rhinovirus/pneumococcal infection and severe disease. We have also examined the role of respiratory picornaviruses as causative agents of CAP in adults and their contribution to disease severity. As part of a larger prospective clinical study2 of the aetiology of CAP, the occurrence of rhinoviruses and enteroviruses was analysed in 231 patients. Detailed information on the study design has been reported previously.2 In addition, throat swab specimens were examined for the presence of rhinoviruses and enteroviruses using previously described reverse transcriptase (RT)-PCR assays.3

The characteristics of the patients and microbiological findings are described in table 1. Viruses were detected in 46 (20%) patients, of whom 19 (41%) were positive for respiratory picornaviruses by RT-PCR. Among the 12 patients with enteroviruses, additional aetiological agents were identified in seven (58%), including three (25%) Streptococcus pneumoniae. Among the seven patients with rhinoviruses, a concomitant S pneumoniae infection was detected in four (57%).

It has been shown in an experimental model that adherence of S pneumoniae to human tracheal epithelial cells is increased in the presence of rhinovirus.4 The results of Jennings and colleagues1 prove this association in vivo by showing that 39% of their patients with rhinovirus identified from a nasopharyngeal sample had concurrent S pneumoniae infection. Consistently, as many as 57% of our patients with rhinovirus also had S pneumoniae infection. Rhinovirus was associated with severe disease (Pneumonia Severity Index IV-V) in 29% of cases, the percentage being somewhat lower than the 39% of severe rhinovirus associated infections reported by Jennings and colleagues.1 One of our patients with mixed rhinovirus/pneumococcal infection died.

To date, only limited data exist on the role of enteroviruses in lower respiratory tract infections in non-immunocompromised adults. To our knowledge, only one previous study included the enterovirus PCR test in the diagnostic array of CAP.5 Moreover, only one (0.5%) of the 198 patients in that study had enterovirus infection. Here, enterovirus was the second most common viral agent after influenza A virus, being detected in 5% of our patients. This percentage is similar to that observed in association with lower respiratory tract infection in children6 in whom enteroviruses are among the most important viruses causing this disease. Collectively, our findings corroborate those of Jennings and colleagues1 and support their conclusion that the importance of both viral pneumonia and mixed viral/bacterial pneumonia may be greater than previously realised.