Effective Containment of Human Respiratory Syncytial Virus Infections in a Bone Marrow Transplantation Program Resulting from the Combined Implementation of Molecular Epidemiology, Infection Control, and Early Treatment Strategies.

Abstract

Although the precise incidence, epidemiology, and impact of human respiratory syncytial virus (HRSV) infection may differ greatly from center to center, there is little debate about the potential for associated morbidity and mortality. We report our approach to infection control and clinical management to reduce the clinical impact of HRSV in our BMT population. An outbreak on our BMT unit was identified in the Winter of 2000–2001. RT-PCR and nucleotide (nt) sequencing of a 270 nt variable region of the HRSV G gene of 6 cases and 12 unrelated community controls was performed. Of the 6 cases, 4 were identified as group A (genotype GA2) and 2 were identified as group B (genotype GB3). Sequences of the GA2 hospital cluster cases were identical as were the 2 hospital case GB3 isolates and the hospital case sequences differed from their respective community controls. These data suggested that there were 2 separate introductions of virus from the community followed by nosocomial spread within the outpatient facility and we were able to direct our infection control efforts and treatment strategies accordingly. An educational campaign was implemented for staff and patients emphasizing the importance of early identification and isolation of patients with HRSV. All patients with either upper (URTI) or lower respiratory tract infection (LRTI) symptoms were isolated and a nasopharyngeal swab for HRSV direct detection and CXR were performed. Post-transplantation, patients with an URTI and a positive swab for HRSV received aerosolized ribavirin (1-beta-D-ribofuranosyl-1,2,4-triazole-3-carboxamide, 6g per day) and 1.5g per kg bodyweight per day of an intravenous immunoglobulin preparation (IVIg) for 3 days. Patients with a positive NP swab and an abnormal CXR or a requirement for supplemental oxygen also received IVIg and ribavirin; however, the treatment was continued until patients no longer required oxygen supplementation. Overall, 3 patients had received autologous and 19 had received allogeneic grafts (14 nonmyeloablative, 5 myeloablative). Steroid administration was not a risk factor for disease. Since the sentinel outbreak, there was a significant decrease in the case attack rates per 10,000 patient days (0.30, 0.27, 0.15, and <0.10, per season, respectively) despite no annual decrease in the number of total cases at our institution during the same time period. Similarly, the median time to diagnosis decreased (6.0, 5.0, and 2.5 days, respectively). In the sentinel outbreak, 4 of the 9 HRSV patients, either presented or progressed to LRTI without intervention. Since the institution of our current strategies, only 3 of the 15 HRSV patients presented with LRTI; no patients progressed from upper to lower tract disease. Our findings demonstrate: (1) HRSV sequence analysis is a useful epidemiologic tool in identifying breeches in infection control measures and (2) the highly cost-effective strategy of educational materials for staff and patients was associated with a reduction in the time to diagnosis/treatment and a lower incidence of LRTI. Moreover, treatment of URTI with a truncated course of ribavirin paired with IVIg was associated with a reduction in the overall incidence of infection and appeared to prevent the progression of URTI to LRTI.