Impact of Immunocompromising Conditions on Severity of Presentations and Outcomes in Hospitalized Coronavirus Disease 2019 (COVID-19) Patients

Abstract

Introduction An initial hypothesis regarding outcomes of COVID-19 infection linked worse outcomes to a dysregulated hyperinflammatory response. As a result, immunosuppressive medications have been proposed for treatment of severe cases of COVID-19. We sought to evaluate the impact of immune compromise in patients admitted for COVID-19-related pneumonia. Methods We constructed a retrospective observational study including patients admitted with COVID-19 pneumonia at Barnes Jewish/Christian (BJC) Hospitals between March 15 to May 13. Washington University School of Medicine IRB waived the need for informed consent. Inclusion criteria were duration of admission of more than 24 hours and positive nasopharyngeal RT-PCR for SARS-CoV-2. Data collection and follow-up were completed on August 27. Collected data included demographics, comorbidities (Elixhauser comorbidity score, nursing home residence, cardiovascular, renal, and pulmonary conditions, diabetes, obesity, substance abuse) and markers of severity of presentation (presence of shock, need for mechanical ventilation). Immunocompromising conditions were grouped in: hematological malignancy or bone marrow transplantation, solid organ transplantation, solid cancer on chemotherapy, TNF-α inhibitor use, and chronic glucocorticoid use. Primary outcome was all-cause mortality, and secondary outcomes were need for ICU stay, length of ICU stay, need for mechanical ventilation (MV), and MV-free days. ICU stay was defined as beginning when more than 6 L of oxygen were needed and ending with discharge from the ICU. Results 627 patients met the inclusion criteria and 80 (14.6%) were immunocompromised at admission. Immunocompromised patients were more likely to be non- African American and with lower BMI. Immunocompromised patients were as likely to develop shock (21.3% vs 28.7%, p=0.164), require ICU admission (33.8% vs 38.8%, p=0.389), mechanical ventilation (22.5% vs 28.5%, p=0.275), and die when compared to non-immunocompromised patients (20% vs 26.1%, p=0.238). Age (OR: 1.08;95% CI:1.06-1.10, p < 0.001), admission from nursing homes (OR: 2.1;95% CI: 1.3-3.3, p=0.002), non-white race (OR: 1.5;95% CI: 1.1-2, p=0.022) and need for > 6 L of oxygen (OR: 4.7;95% CI: 2.4- 9.1, p < 0.001) and mechanical ventilation (OR: 2.3;95% CI: 1.2-4.5, p=0.02) were significant predictors for mortality in multivariable logistic regression analyses. Immunocompromised status did not impact admission to the ICU and all-cause mortality. Conclusion Immunocompromised status does not seem to impact mortality and need for ICU admission for COVID-19 patients in our multi-center cohort. Future larger studies and analyses including treatment data will further characterize the trajectory of immunocompromised patients admitted for COVID-19 related pneumonia.