The ORChID study is an ongoing dynamic birth cohort study in Brisbane (Queensland, Australia) that has been described (13). The study was approved by the Human Research Ethics Committees of the Children’s Health Queensland Hospital and Health Service, the Royal Brisbane and Women’s Hospital, and The University of Queensland. In brief, anterior nasal swab specimens were collected at birth and weekly until the child’s second birthday (Technical Appendix).

The present study reports on the first 56 children to complete 18 months (censored for swab specimens collected after 530 study days) of the ORChID study. A total of 3,851 nasal swab specimens (mean 69 swab specimens/child, range 41–77 swab specimens/child) during 29,678 person-days of observation (mean 530 person-days of observation/participant, range 384−560 person-days of observation/participant). These samples were tested for WUPyV, KIPyV, MCPyV, and MWPyV by using reported real-time PCRs (12,13). Samples were screened for 13 common respiratory viruses according to the ORChID study protocol (Technical Appendix).

A sole detection episode was defined as ≥1 consecutive swab specimens in which an individual PyV was the only virus detected, and no other viruses were reported 7 days before or after detection of the PyV. Detection of a different PyV or the same PyV after 30 days and ≥2 intervening negative samples was considered a new infection episode. During the period of the sole detection episode, clinical symptoms were broadly categorized as upper respiratory, lower respiratory, nonspecific, and gastrointestinal (Technical Appendix). Nonspecific symptoms were only separately categorized when unaccompanied by upper or lower respiratory symptoms. Gastrointestinal symptoms were recorded in the presence or absence of respiratory symptoms.

All 4 novel PyV viruses were detected in respiratory samples from children ≤18 months of age; MWPyV was the predominant virus (157 positive detections) (Table 1). A total of 23% (13/56 for MCPyV) and 56% (31/56 for MWPyV) of children had ≥1 positive result for 1 of the PyVs. WUPyV, KIPyV, and MWPyV were initially detected when the child was ≈1 year of age, and each virus was detected for a mean of 2 consecutive weeks after primary detection. MCPyV was detected less frequently; primary detections occurred in children ≈7 months of age, which was earlier than detection of WUPyV, KIPyV, and MWPyV (p<0.001, by Mann-Whitney 2-tailed test with pairwise comparisons of ages). MCPyV was not seen in any consecutive swab specimen collections.

WUPyV (50%, 18/36) and KIPyV (49%, 31/63) were commonly detected with other respiratory viruses. MWPyV (33%, 52/157) and MCPyV (21%, 3/13) had lower semiquantitative viral loads (p<0.001, by Mann-Whitney 2-tailed test with pairwise comparisons of cycle thresholds), and were less frequently detected with other respiratory viruses (Table 1; Technical Appendix).

Symptoms were reported during sole detection episodes for WUPyV, KIPyV, and MWPyV, but not for MCPyV (Table 2). During symptomatic episodes, numerous overlapping respiratory virus detections were commonly observed, which made sole detection episodes of PyV too rare to be considered in a formal statistical analysis. However, sole detection episodes corresponded to parental reporting of clinical symptoms for 57% (4/7) of WUPyV and 36% of KIPyV (5/14) and MWPyV (13/36) infection episodes (Table 2). Most symptoms reported during the sole detection episodes were upper respiratory (Table 2).

Primary acquisition of WUPyV, KIPyV, and MWPyV occurred most commonly when children were ≈12 months of age, which is consistent with previous serologic data showing PyV primary infection in children at an early age (3). MCPyV was also detected within this cohort, but age, frequency of detection, and lower levels of viral shedding contrasted with the findings for WUPyV, KIPyV, and MWPyV. This finding is supported by other retrospective studies that showed that MCPyV is more frequently detected in the respiratory tract of adults (7). Previously reported high co-detection rates of WUPyV and KIPyV with other respiratory viruses (>70%) have hampered efforts to associate these viruses with clinical symptoms (3,14). However, we observed a lower rate of PyV co-detection, which enables us to examine the association of PyV sole detection episodes with symptoms. This examination showed that most symptoms were upper respiratory, although gastrointestinal symptoms were also reported during WUPyV, KIPyV, and MWPyV sole detection episodes.

A higher viral load was observed for WUPyV and KIPyV than for MWPyV and MCPyV. Although use of cycle thresholds as a semiquantitative marker has some limitations, thresholds suggest that WUPyV and KIPyV actively replicate in the respiratory tract and may account for the higher rate of reported symptoms with sole detection of WUPyV. A previous study in the Netherlands also reported a higher symptom association of WUPyV than KIPyV, but MWPyV was not investigated in that study (15). MWPyV was first detected in fecal specimens from healthy children, but we found that MWPyV was the most prevalent PyV detected in respiratory specimens and that it was associated with upper respiratory infection symptoms in >33% of sole detection episodes. We found that 1/36 participants had gastrointestinal symptoms during an episode of MWPyV sole detection, which is an observation that warrants further investigation.

MCPyV is shed from healthy skin, and MWPyV has been detected in anal warts. Thus, a limitation of this study is potential cutaneous contamination of swab specimens by the parent or child during sample collection. Although the complex nature of virus acquisition and overlapping intervals of virus detection confound the association of PyV sole detection episodes with particular symptoms, our data show that WUPyV, KIPyV, and MWPyV, but not MCPyV, are frequently detected within the respiratory tract of healthy children <18 months of age and are associated with mild upper respiratory symptoms.

We thank Anne Cook, Francis Maguire, and Minda Sarna for providing assistance during collection of clinical data.

This study was supported by the National Health and Medical Research Council, Australia (grant 615700), a Queensland Children’s Hospital Research Institute PhD scholarship (grant 5001), and the Children’s Health Foundation (Brisbane, Australia) (program grant 50006).