e18816 Background: Acute Kidney Injury (AKI) in patients with COVID-19 infection is associated with poor clinical outcomes. We examined outcomes (hemodialysis, mechanical ventilation, ICU admission and death) in cancer patients with normal estimated glomerular filtration rate (eGFR) treated in a tertiary referral center with COVID-19 infection, who developed AKI within 30 days of diagnosis. Methods: All patient data — demographics, labs, comorbidities and outcomes — were aggregated and analyzed in the Syntropy platform, Palantir Foundry (“Foundry”), as part of the Data-Driven Determinants of COVID-19 Oncology Discovery Effort (D3CODE) protocol at MD Anderson. The cohort was defined by the following: (1) positive COVID-19 test; (2) baseline eGFR >60 ml/min/1.73m2most temporally proximal lab results within 30 days prior to the patient’s infection. AKI was defined by an absolute change of creatinine ≥0.3 within 30 days after the positive COVID-19 test. Kaplan-Meier analysis was used for survival estimates at specific time periods and multivariate Cox Proportional cause-specific Hazard model regression to determine hazard ratios with 95% confidence intervals for major outcomes. Results: 635 patients with Covid-19 infection had a baseline eGFR >60 ml/min/1.73m2. Of these patients, 124 (19.5%) developed AKI. Patients with AKI were older, mean age of 61+/-13.2 vs 56.9+/- 14.3 years (p=0.002) and more Hypertensive (69.4% vs 56.4%, p=0.011). AKI patients were more likely to have pneumonia (63.7% vs 37%, p<0.001), cardiac arrhythmias (39.5% vs 20.7%, p<0.001) and myocardial infarction (15.3% vs 8.8%, p=0.046). These patients had more hematologic malignancies (35.1% vs 19%, p=0.005), with no difference between non metastatic vs metastatic disease (p=0.284). There was no significant difference in other comorbidities including smoking, diabetes, hypothyroidism and liver disease. AKI patients were more likely to require dialysis (2.4% vs 0.2%, p=0.025), mechanical ventilation (16.1% vs 1.8%, p<0.001), ICU admission (43.5% vs 11.5%, p<0.001) within 30 days, and had a higher mortality at 90 days of admission (20.2% vs 3.7%, p<0.001). Multivariate Cox Proportional cause-specific Hazard model regression analysis identified history of Diabetes Mellitus (HR 10.8, CI 2.42 - 48.4, p=0.001) as an independent risk factor associated with worse outcomes. Mortality was higher in patients with COVID-19 infection that developed AKI compared with those who did not developed AKI (survival estimate 150 days vs 240 days, p=0.0076). Conclusions: In cancer patients treated at a tertiary cancer center with COVID-19 infection and no history of CKD, the presence of AKI is associated with worse outcomes including higher 90 day mortality, ICU stay and mechanical ventilation. Older age and hypertension are major risk factors, where being diabetic was associated with worse clinical outcomes.
Epidemiology and outcomes of acute kidney injury in hospitalized cancer patients in China