Surgical Mask Usage Reduces the Incidence of Parainfluenza Virus 3 in Recipients of Stem Cell Transplantation

Conference Paper

Abstract Abstract 462 Respiratory viral infections (RVI) are frequent complications of hematopoietic stem cell transplantation (HSCT). Parainfluenza virus 3 (PIV3) in particular affects patients after HSCT, spreads nosocomially, and is associated with increased morbidity and mortality. Given the absence of proven effective treatments, prevention is key. However, surveillance and isolation of infected patients have been insufficient, likely foiled by asymptomatic viral shedding. Requiring all individuals with direct patient contact to wear a surgical mask is a simple and inexpensive intervention that may reduce droplet exposure by asymptomatic shedders and limit nosocomial spread. Prompted by recurrent clusters of PIV3 affecting our adult HSCT program, we performed an institutional review board approved prospective observational study to assess the impact of surgical mask usage. Between January 10, 2010-January 9, 2012 (mask period), all individuals with direct patient contact, including caregivers, visitors, and medical staff, were required to wear a surgical mask when within 3 feet of a patient, regardless of suspicion of RVI. This continued until the patient returned to their primary residence following completion of HSCT. The primary endpoint was the incidence of RVI compared to the control period of December 1, 2003-November 30, 2009 (pre-mask period). A brief washout period (December 1, 2009-January 9, 2010) allowed for training and implementation. Standard infection prevention precautions including strict hand washing were in place during both the pre-mask and mask periods. RVI (influenza A and B, PIV 1, 2, and 3, adenovirus, and respiratory syncytial virus (RSV)) were documented by positive culture, polymerase chain reaction, or direct fluorescence antibody testing from respiratory specimens. Metapneumovirus was excluded as our lab only started testing for it February 2011. Secondary endpoints include clusters of RVI, defined as 3 or more infections in a 30-day period, and death from RVI. The study was designed to detect a 40% reduction in the incidence of RVI with 90% power. During the six-year pre-mask period, 920 patients received HSCTs. During the two-year mask period, 454 patients received HSCTs. In the pre-mask and mask groups, mean age was 50 and 54 (t-test p<0.0001) (ranges 19 to 79 and 19 to 81). With the exception of non-Hodgkin lymphoma (21% and 15%) and plasma cell dyscrasia (32% and 46%) (chi-square p=0.004), there were no differences in disease representation among the two cohorts. Allogeneic HSCT constituted 41% and 33% of transplants (two-sided Fisher's Exact Test p=0.004), of which 52% and 59% were myeloablative (p=0.18); 19% and 21% used cord blood as a donor source (p=0.63); 16% and 8% used a haploidentical donor (p=0.02); and 43% and 35% used alemtuzumab (p=0.12). The incidence of any RVI was 10.3% in the pre-mask period and 3.3% in the mask period (one-sided Fisher's Exact Test p<0.0001) (Table 1). Among patients who had an allogeneic HSCT, the incidence was 16.9% and 6.7% (p=0.002); among patients who had an autologous HSCT, the incidence was 5.7% and 1.6% (p=0.005). Surgical mask usage had the most pronounced affect on reducing the incidence of PIV3 (8.2% to 2.0%, p<0.0001), which was the most common virus (74% and 60% of infections), followed by RSV (16% and 13%). There were more clusters in the pre-mask period (1.7/year vs. 0.5/year). In a blinded audit by three physicians (ADS, JAMS, MEH), viral infection contributed to patient death in 1.2% of patients in the pre-mask period and 0.2% in the mask period (p=0.12). Patients with RVI required more peri-transplant care (median 76 days vs. 21 days, Wilcoxon Rank Sum p<0.0001). These data suggest that requiring all individuals with direct patient contact to wear a surgical mask can reduce the incidence of RVI, particularly PIV3, during the most vulnerable period following autologous or allogeneic stem cell transplantation. Table 1. Incidence of Viral Infections Virus % (n) Pre-Mask (n = 920) Post-Mask (n = 454) p-value Any 10.3% (95) 3.3% (15) <0.0001 Influenza A 0.5% (5) 0.4% (2) 1.0* Influenza B 0.1% (1) 0% (0) 1.0* Parainfluenza 1 0.1% (1) 0% (0) 1.0* Parainfluenza 2 0.3% (3) 0% (0) 1.0* Parainfluenza 3 8.2% (75) 2.0% (9) <0.0001* Adenovirus 0% (0) 0.4% (2) 1.0* Respiratory Syncytial Virus 1.7% (16) 0.4% (2) 0.24* * One-sided Fisher's Exact Test adjusted for multiple comparisons using the Bonferroni approach. Disclosures: No relevant conflicts of interest to declare.

Full Text

Duke Authors

Cited Authors

  • Sung, AD; Sung, JAM; Corbet, K; Broadwater, G; Hars, V; Zanter, A; Chao, NJ; Horwitz, ME

Published Date

  • November 16, 2012

Published In

Volume / Issue

  • 120 / 21

Start / End Page

  • 462 - 462

Published By

Electronic International Standard Serial Number (EISSN)

  • 1528-0020

International Standard Serial Number (ISSN)

  • 0006-4971

Digital Object Identifier (DOI)

  • 10.1182/blood.v120.21.462.462