BackgroundHigh mortality has been described in coronavirus disease 2019 (COVID-19) with cytokine release syndrome (CRS). Tocilizumab (TCZ), an interleukin-6 (IL-6) receptor antagonist may be associated with improved outcomes in such patients; however, the subgroups of patients who benefit the most need to be identified.ObjectiveTo analyze the efficacy and optimal timing of administration of TCZ in moderate to severe COVID-19 with features of CRS, where the response to steroids was poor.MethodsThis is a retrospective study of 125 patients admitted between May 5 to July 31, 2020, in a tertiary care hospital in western India, with moderate to severe COVID-19 who were treated with TCZ along with steroids. The primary outcomes were the need for mechanical ventilation (MV) or death, and secondary outcomes were a decrease in oxygen requirement and inflammatory markers; the incidence of secondary infections, and renal or hepatic dysfunction. Kaplan Meier survival analysis and log rank test were used for evaluating primary outcomes. Secondary outcomes were analyzed using the Wilcoxon Signed-Rank test.ResultsAmong 1081 patients admitted during the study period, 125 were administered TCZ (median age, 56 [95% CI 54 - 60] years; 100 [80%] male). The commonest symptoms were fever (96%), cough (64%), and dyspnea (48.8%). 78.4% patients had comorbidities (hypertension 51.2%, diabetes 43.2%, obesity 25.6% and chronic cardiac disease 13.6%). Of 117 patients who were treated with TCZ before requiring MV, 18.8% progressed to MV. Overall, 25% of the patients needed MV support. 65.3% of patients were discharged by day 14 after TCZ administration. Mortality was nil, 16.2%, 50%, and 62.5% in patients who received TCZ on room air, low flow oxygen, high flow nasal cannula (HFNC) and bilevel positive airway pressure (BiPAP), and MV respectively; overall 24.8% of patients died. Survival analysis showed no difference in outcome with respect to age and gender, while progression to MV showed a statistically significant reduction for the event death (90.9% of patients who progressed to MV died as compared to 6.3% who did not; log rank test with p < 0.0001). No adverse events were noticed.ConclusionMortality was least in patients of COVID-19 with CRS who received TCZ while on low flow oxygen. When administered in the early hypoxemic phase, TCZ is associated with reduced mortality and decreased need for mechanical ventilation.

No abstract

e21526 Background: Palliative care improves the quality of life in cancer patients; however, there is no literature on specific factors that predict its use in patients with malignant melanoma. We explored the prevalence trends, and predictors of palliative care utilization among hospitalized patients with malignant melanoma. Methods: Retrospective analyses were conducted using the National Inpatient Sample (NIS) data collected between 2016 and 2020. Descriptive analyses and multivariable regression models were used to investigate the prevalence trends, and sociodemographic and hospital-level factors associated with palliative care utilization in hospitalized malignant melanoma patients. Results: Of the 9,760 hospitalizations with a diagnosis of malignant melanoma over the study duration, 14% utilized palliative care during their hospital stay. Overall, 9.2% of malignant melanoma patients used palliative care and were discharged alive. There was a stable trend of palliative care use over the 5-year period (14%). Compared to patients on Medicare, those on Medicaid were twice (adjusted odds ratio (AOR): 2.12; 95% confidence interval (CI): 1.26–3.64) more likely to utilize palliative care. Relative to other regions in the US, patients hospitalized in the West were 40% less likely to receive the service of the palliative team (AOR: 0.61; 95% CI: 0.38-0.96). Those admitted to teaching hospitals (AOR: 0.55; 95% CI: 0.40-0.77) had lower odds of having palliative care consultations when compared to non-teaching hospitals. Patients admitted to urban hospitals had 50% (AOR: 0.51; 95% CI: 0.27–0.98) lesser odds of getting palliative care when compared to their counterparts in rural hospitals. Individuals who were either discharged to a facility/with home health (AOR: 8.81; 95% CI: 5.88–13.23) or died during hospitalization (AOR: 128.42; 95% CI: 72.10–228.77) had higher odds of utilizing palliative care when compared to those with a routine discharge. Conclusions: The prevalence of palliative care utilization was low, and factors associated with utilization in our population were identified. Our findings emphasize the necessity to improve awareness among medical oncologists and primary inpatient teams on the importance of involving the palliative care service early in the management of hospitalized patients with malignant melanoma.

e16327 Background: COVID-19 infections have been known to cause worse outcomes in patients with underlying co-morbid conditions. Currently, there is limited data regarding re-admissions in such patients after COVID-19 infection. So, we pursued to assess the rates, predictors, and causes of re-admission after COVID-19 infection in patients who had underlying gastrointestinal malignancy. Methods: The National Readmission Database for 2020 was analyzed to identify patients with COVID-19 hospitalizations who also had a co-diagnosis of GI malignancy. Data for re-admission was gathered, at 30 and 90 days after initial hospitalization. Multivariate logistic and linear regression analysis was used accordingly to adjust for possible confounders. Results: For 30 days: A total of 2,726 COVID-19 hospitalizations had a co-diagnosis of GI malignancy. The mean LOS (length of stay) for index admission was 8.7 days. The mean TOTCHG (total hospital charge) for index admission was $85,747. Within 30 days from discharge, 462 (21%) were re-admitted. The in-hospital mortality in re-admissions (18.4%) was quite similar to that for index hospitalization (18.5%). Positive predictors of re-admission include shorter length of stay during initial hospitalization (<3 days), anemia, renal insufficiency/chronic kidney disease, and atherosclerosis of the aorta. Females had a lower chance of re-admission compared to males. Among the top causes of re-admission were ongoing COVID-19 infection (27%), sepsis (16.3%), acute kidney injury (2.4%), and metabolic encephalopathy (1.2%). For 90 days: A total of 1,782 COVID-19 hospitalizations had a co-diagnosis of GI malignancy. The mean LOS (length of stay) for index admission was 9.2 days. The mean TOTCHG (total hospital charge) for index admission was $90,908. Within 90 days from discharge, 411 (29%) were re-admitted. The in-hospital mortality in readmissions (14%) was lower than that for index hospitalization (20.3%). Predictors of re-admissions include secured insurance, anemia, renal insufficiency/chronic kidney disease, and cerebrovascular accidents. Top causes of re-admission were ongoing COVID-19 infection (18.5%), aspiration pneumonitis (2.9%), pneumonia (2.3%), and acute kidney injury (1.4%). Conclusions: Infections were the leading cause of re-admission, which is an important factor to consider when managing patients with GI malignancies due to sub-optimal immunity. Among these patients, those with concomitant anemia and chronic kidney disease were more prone to being re-admitted. Patients with GI malignancies also tended to have a higher length of stay, leading to higher charges, more exposure to nosocomial infections, and delays in cancer treatment leading to worse cancer outcomes. Pre-mature discharges, without medical optimization, can also potentially predispose to higher rates of re-admissions.