Parainfluenza virus (PIV) is an upper respiratory tract infection (URTI) than can progress to a more serious and often fatal lower respiratory tract disease (LRTD) in immunocompromised individuals. Hematopoietic cell transplantation (HCT) patients undergo immunosuppressive radiation or chemotherapy treatment prior to transplant and frequently acquire PIV following the procedure. However, the risk factors that predict progression from URTI to LRTD in HCT patients are undefined, complicating identification and prevention of LRTD.
In a recent Biology of Blood and Marrow Transplantation publication, Dr. Sachiko Seo, an affiliate investigator with Dr. Michael Boeckh’s group at Fred Hutch, conducted a retrospective cohort study to evaluate the progression rate of PIV LRTD in HCT patients, identify the risk factors that predict progression to LRTD, and determine the effect of antiviral drugs on prevention of LRTD. Additionally, the investigators sought to identify risk factors associated with LRTD alone, as PIV LRTD sometimes insidiously occurs without an initial manifestation of URTI.
Hospital records were procured for 540 patients who developed PIV after receiving HCT at Fred Hutch clinics between 1990 and 2011. The authors found that 424 patients presented with an initial URTI while 81 were diagnosed with LRTD within 40 days of HCT. The authors defined LRTD cases as possible, probable, and proven based on documented symptoms and diagnostic results. Using cumulative incidence, or the number of LRTD cases as a percentage of total patients at risk, the probability of progression to LRTD was 12%, while the probability was higher when patients who had probable but unconfirmed LTRD were included. The medical records revealed that infection with PIV subtype 3, blood co-infection with other pathogens such as Staphylococcus aureus or Pseudomonas aeruginosa, low monocyte counts, and use of immunosuppressive steroids before infection were all associated with a high progression rate to probable or proven LRTD. Cumulative incidence curves identified that the greatest risk factors were monocyte count and steroid use, the latter being prescribed in and acting as a surrogate marker for incidences of graft-versus-host disease. When the researchers analyzed LRTD progression rate relative to the number of risk factors patients presented at the detection of URTI, they found that risk of progression increased with number of risk factors, with all patients who progressed to LRTD having at least one risk factor. These results suggest that risk of progression to LTRD in HCT patients may be higher than previously expected and that definable risk factors can be used to predict which patients progress.
Next, Dr. Seo and colleagues investigated the effectiveness of antiviral treatment in preventing LRTD progression. Ribavirin is an antiviral drug that has historically been suggested to inhibit PIV by blocking viral RNA synthesis and mRNA capping. To examine the effects of ribavirin treatment on LRTD progression, the researchers identified 18 patients who received ribavirin therapy after diagnosis with URTI or possible LRTD, 3 of whom progressed to probable or proven LRTD. Cox regression modeling-a proportional hazards statistical method that estimates the association between a variable and an outcome-of risk factors for progression to probable or proven LRTD demonstrated that ribavirin use was not associated with progression. However, adjusting for each of the four risk factors revealed that ribavirin also does not have a preventative effect on progression to LRTD, suggesting that current PIV treatment methods are inadequate.
Finally, to investigate the risk factors for presentation of LRTD without prior URTI symptoms or diagnosis, Dr. Boeckh’s team identified 64 patients who presented with LRTD but not URTI, 19 of whom had been confirmed negative for URTI. Using multivariate logistic regression, risk factors for restrictive lung function, absence of documented URTI symptoms, requirement of oxygen at diagnosis, and co-infection with pathogens such as cytomegalovirus or Aspergillus were all associated with LRTD without URTI. These results suggest that although these patients seem to spontaneously develop a lower respiratory disease without the traditional harbinger of an initial upper respiratory infection, diagnostic factors can be evaluated in order to better predict which HCT patients may develop LRTD even when canonical warning symptoms are absent.
This study stratifies specific risk factors which could aid in prevention and diagnosis of LRTD progression. These findings also provide novel insight into PIV LRTD treatment that may inform clinic practices in immunocompromised and transplant patients. Importantly, this research suggests that ribavirin, which has been thought to prevent PIV disease progression, is not effective in this infection and that steroids used to combat immune system-mediated rejection of transplants should be decreased when possible until more effective antivirals are available to control PIV. Furthermore, because half of LRTD patients lacked URTI manifestations, rigorous virologic testing should be performed whenever respiratory symptoms present in HCT patients. This approach may be helpful to diagnose and treat LRTD in patients who may not have previously been considered at risk for PIV progression. These results identify which transplant recipients are at risk for for LRTD and will help prevent progression to this deadly condition as well as highlight a need for improved antivirals and treatments for transplant-associated complications such as graft-versus-host disease.
Seo S, Xie H, Leisenring WM, Kuypers JM, Sahoo FT, Goyal S, Kimball LE, Campbell AP, Jerome KR, Englund JA, Boeckh M. 2018. Risk Factors for Parainfluenza Virus Lower Respiratory Tract Disease after Hematopoietic Cell Transplantation. Biology of Blood and Marrow Transplantation. pii: S1083-8791(18)30492-0. doi: 10.1016/j.bbmt.2018.08.021
This work was funded by the National Institutes of Health and the Joel Meyers Endowment Scholarship.
Fred Hutch/UW Cancer Consortium authors Angela P. Campbell, Keith R. Jerome, Michael Boeckh (UW/FH), and Janet A. Englund (FH) contributed to this work.