Epidemiology and Outcomes of Respiratory Virus Infections in Adult Acute Leukemia and Hematopoietic Stem Cell Transplant Patients

BACKGROUND

Respiratory virus infections (RVI) are significant cause of morbidity and mortality in adult patients with hematopoietic stem cell transplant (HSCT) recipients, causing delay in administering chemotherapy and HSCT, increased hospital stay and in occasionally resulting in mortality. In 2008, the FDA approved the use of multiplex PCR assay to detect respiratory viruses. Our institution has been using the multiplex PCR technique for approximately four years. With this technology the time to diagnosis of a RVI became shorter and detection of multiple viruses at the same time became feasible.

RVIs have been studied more exclusively in the HSCT population where lymphopenia has been identified for poor outcomes. These infections have been less well studied in the acute leukemia population. This report, however, encompasses both patient populations. The natural progression of RVIs apart from influenza is poorly understood. The viruses isolated from the upper respiratory tract, development of lower respiratory tract infection (LRI), as well as overall survival of the patients identified with RVIs are reported here.

STUDY DESIGN

This is a retrospective cohort study of adult patients (>=18yrs) with acute leukemia and/or after HSCT diagnosed with laboratory documented RVI between May 1^st 2010 and March 1^st2014. A list of positive RVIs from multiplex PCR was retrieved from the hospital laboratory based on patients with specific ICD-9 codes.

RESULTS

Between May 2010 to March 2014, there were 208 episodes of RVI in 122 patients. Out of these, 77 patients were post-HSCT and 45 patients had acute leukemia without undergoing HSCT. Out of these episodes, 85 were caused by rhinovirus, 37 by coronavirus, 18 by RSV, 35 by parainfluenza, 15 by influenza, 12 by metapneumovirus, 3 by adenovirus, 3 by poly-viral infection. 189 episodes were initially diagnosed as an upper respiratory tract infection (URI). There were 47 episodes of LRI; of these 32 (16.9%) had a prior URI with rhinovirus (n=13), coronavirus (n=5), parainfluenza (n=4), RSV (n=4), influenza (n=3) and poly-viral infections (n=2). 15 (7.2%) episodes were LRI at initial presentation (4 metapneumovirus, 4 parainfluenza, 2 RSV, 3 rhinovirus, 1 coronavirus, 1 adenovirus). Out of the 122 patients there were 7 deaths (5.7%).

CONCLUSION

The availability of multiplex PCR has allowed for the identification of respiratory viruses responsible for URI in this patient population. Most of these infections have limited clinical consequences. However, in our study 25% of the episodes either presented with or progressed to LRI. The exact role of respiratory virus in causing LRI is difficult to determine in a retrospective study.