Parainfluenza Virus Infections In Children with Cancer

Parainfluenza virus (PIV) infections are relatively common and cause significant mortality in adults undergoing hematopoietic stem cell transplantation (HSCT). Children are more prone than adults to develop PIV infections. However, data regarding the epidemiology of PIV infections in children with cancer is limited. This study sought to determine the epidemiology of PIV infections in children with leukemia/lymphoma (LL) and those undergoing HSCT, risk factors for progression to lower respiratory tract infection (LRTI) and impact of PIV infections on mortality after HSCT.

A total of 1381 children with LL diagnosed between 2000–2009 and 1349 children who underwent HSCT between 1995–2009 were studied. Medical record review included patient demographics, disease characteristics, results of virologic analysis, characterization of the infection as nosocomial or acquired and with upper or lower respiratory tract involvement, duration of symptoms and viral shedding, presence of co-pathogens, absolute neutrophil count (ANC) and absolute lymphocyte count (ALC) at the time of infection, use of corticosteroids prior to infection, antiviral therapy and mortality. Statistical analysis was performed using Fischer's exact and Kruskal-Wallis test.

PIV infections occurred in 83 (6%) of patients with LL and in 46 (3.4%) HSCT patients. For patients with LL there were more infections in 2005–2009 compared to 2000–2004 (p<0.0001). Patients with acute lymphoblastic leukemia had more infections compared to those with acute myeloid leukemia or lymphoma (p<0.0001). Of the 83 LL patients, 76 (92%) had community-acquired (CA) infections and 17 (20%) patients had LRTI. For HSCT patients, allogeneic transplant recipients were more prone to develop PIV infections (p<0.0001). Of the 46 HSCT patients, 33 (72%) had CA infections and 18 (39%) patients had LRTI. A significantly greater number of HSCT patients had LRTI (p=0.04). Six (13%) HSCT patients died of PIV infection while all LL patients survived the infection (p=0.002). Of the 129 patients, culture and direct fluorescent antibody detection identified PIV infection in 126 (98%) patients. For patients with LL, LRTI was associated with age < 2 years (p=0.005), ANC <500 cells/μL (p=0.002) and ALC <100 cells/μL (p=0.008). For HSCT patients LRTI was associated with nosocomial infection (p=0.017), infection in the first 100 days after transplant (p=0.006), non-engraftment (p=0.014), use of steroids in the 2 weeks preceding the infection (p=0.035), fever (p=0.004), ANC <500 cells/μL (p<0.0001) and ALC <100 cells/μL (p<0.0001). Mortality in HSCT patients was associated with African-American race (p=0.013), LRTI (p=0.002), use of steroids (p<0.0001), mechanical ventilation (p<0.0001) and ALC <100 cells/μL (p=0.01). Prolonged duration of viral shedding was associated with ALC <500 cells/μL (p=0.045). Seventy-nine (61%) patients had PIV-3 infection. PIV-3 infection was not associated with increased LRTI or mortality. Pulmonary co-pathogens were isolated from 13 (10%) patients. Presence of co-pathogens was not associated with increased LRTI or mortality. Anti-viral therapy with aerosolized ribavirin with or without immunoglobulin therapy did not appear to alter mortality.

An increasing incidence of PIV infections was observed in patients with leukemia and lymphoma at our institution. These infections were predominantly community-acquired. LRTI was associated with younger age, severe neutropenia and lymphopenia. HSCT patients had increased LRTI and mortality following PIV infection. LRTI was associated with nosocomial acquisition, infection before day + 100, use of steroids, severe neutropenia and lymphopenia. Mortality in the HSCT population was associated with African-American race, LRTI, use of steroids, mechanical ventilation and severe lymphopenia. Currently available therapy appears to be inadequate in reducing mortality from PIV pneumonia. This represents the largest cohort of children with cancer evaluated retrospectively to study the epidemiology of PIV infections. In comparison with studies on the adult population, the association of younger age, race, nosocomial infection and severe neutropenia with LRTI and mortality are novel observations.

No relevant conflicts of interest to declare.