Human Rhinovirus RNA Detection in the Lower Respiratory Tract of Hematopoietic Cell Transplant Recipients: Association with Mortality

Background: Human rhinovirus (HRV) is the most common respiratory viral infection following hematopoietic cell transplantation (HCT); the day 100 cumulative incidence is 22%. HRV can progress to lower respiratory tract (LRT) infection. The clinical significance of HRV RNA in LRT secretions is unclear. We analyzed the outcome and risk factors for mortality in HCT recipients with HRV RNA detection in the LRT, comparing survival of patients with LRT infection due to HRV and other respiratory viruses.

Methods: A retrospective cohort of 87 HCT recipients transplanted between 6/1993–4/2012 at FHCRC had HRV RNA detected in LRT by bronchoalveolar lavage (BAL) or biopsy using culture (N=13) and/or real time RT-PCR (N=80); all patients had LRT symptoms and/or pulmonary infiltrates. Factors associated with overall mortality were analyzed using Cox proportional hazards models. Comparison cohorts with parainfluenza virus (PIV, N=55), RSV (N=111), or influenza A (FLU, N=41) LRT infection were included in the analyses. Samples from a separate control cohort of 21 asymptomatic HCT patients without radiographic abnormalities undergoing research BAL between day 35–50 were also studied.

Results: Median age of the 87 patients was 48.5 years (range 1–71); median time to HRV LRT detection after HCT was 81 days (range 0–3596). Viral/fungal/bacterial copathogens were present in 48% of patients. Probability of 180-day survival after HRV detection was 59%. Multivariable analysis showed that oxygen use at diagnosis (HR 2.6, p=0.04), steroid use ≥ 1mg/kg prior to diagnosis (HR 5.5, p=0.002), and bone marrow as stem cell source (HR 2.3, p=0.05) were associated with poor outcome. The presence of copathogens did not significantly affect mortality (HR 1.1, p=0.71) (Figure). Survival at day 180 in patients with HRV LRT infection and LRT infection caused by other respiratory viruses was similar (Figure); after adjusting for stem cell source and baseline steroid and oxygen use, no statistically significant difference in mortality between HRV and any of the other viruses was found (adjusted HR versus HRV without copathogens: HRV with copathogens, HR 0.96; PIV, HR 1.05; RSV, HR 0.93; FLU, HR 1.47). HRV was detected in only one asymptomatic control patient (5%).

Conclusions: HCT recipients who underwent BAL or biopsy for lung disease with HRV RNA detection had a high mortality even in the absence of copathogens. Outcome was not statistically significantly different from patients with RSV, influenza or PIV LRT disease. These results suggest that HRV is a pathogen that is associated with severe pulmonary disease, and is associated with mortality rates comparable to well-recognized viral pathogens. Additional studies are ongoing to define the role of viral load and HRV genotypes on outcome.