1583. The Utility of the Immunodeficiency Scoring Index (ISI) to Predict Outcomes of Coronavirus (HCoV) Infections in Hematopietic Cell Transplant (HCT) Recipients

BACKGROUND: Respiratory viral infections in HCT recipients are associated with high morbidity and mortality, especially after progression from upper respiratory tract infection (URI) to lower respiratory tract infections (LRI). Data on risk factors (RF) for LRI and mortality is lacking for HCoV infections after HCT. We aimed to validate our ISI in HCoV infections. METHODS: All adult HCT recipients with HCoV infection from 2015 to 2017 were evaluated. An ISI based on RF was used to classify patients as low (0–2), moderate (3–6), or high (7 or higher) risk for progression to LRI or death. We defined LRI as HCoV detected in nasal wash and/or bronchoalveolar lavage and new lung infiltrates on diagnostic imaging. Clinical parameters were collected and ISI were calculated for comparison. RESULTS: A total of 144 adult HCT recipients with 166 episodes of HCoV infections were analyzed. The most common HCoV serotype for LRI and URI was 229E (42.4%) and OC43 (37.6%), respectively, and most patients were infected between November and March each year (Figures 1 and 2). When compared with URI, patients with LRI were more likely in the pre-engraftment period, had multiple respiratory viruses infections, had nosocomially acquired HCoV, required hospitalization, ICU transfer, and mechanical ventilation (all, P < 0.05). Overall mortality rate was 4% at Day 30 from diagnosis and all patients who died had LRI with an 18% mortality. Among those who died, 33% had nosocomial infection, 67% were co-infected with another respiratory virus and 67% required mechanical ventilation. Using an ISI cut off of <4, the negative predictive value (NPV) for progression to LRI was 86% with a specificity of 76%. CONCLUSION: HCT recipients with HCoV LRI were more likely to have a fatal outcome. The NPV of the ISI for progression to LRI was high and could be used as a prognostic tool for future studies and for therapeutic clinical trials. DISCLOSURES: All authors: No reported disclosures.