The determination of causality of drug induced liver injury in patients with COVID-19 clinical syndrome

Abstract

Background Drug induced liver injury (DILI) is a rising morbidity amongst patients with COVID-19 clinical syndrome. The updated RUCAM causality assessment scale is validated for use in the general population, but its utility for causality determination in cohorts of patients with COVID-19 and DILI remains uncertain. Methods This retrospective study was comprised of COVID-19 patients presenting with suspected DILI to the emergency department of Weill Cornell medicine-affiliated Hamad General Hospital, Doha, Qatar. All cases that met the inclusion criteria were comparatively adjudicated by two independent rating pairs (2 clinical pharmacist and 2 physicians) utilizing the updated RUCAM scale to assess the likelihood of DILI. Results A total of 72 patients (mean age 48.96 (SD ± 10.21) years) were examined for the determination of DILI causality. The majority had probability likelihood of “possible” or “probable” by the updated RUCAM scale. Azithromycin was the most commonly reported drug as a cause of DILI. The median R-ratio was 4.74 which correspond to a mixed liver injury phenotype. The overall Krippendorf's kappa was 0.52; with an intraclass correlation coefficient (ICC) of 0.79 (IQR 0.72-0.85). The proportion of exact pairwise agreement and disagreement between the rating pairs were 64.4%, kappa 0.269 (ICC 0.28 [0.18, 0.40]) and kappa 0.45 (ICC 0.43 [0.29-0.57]), respectively. Conclusion In a cohort of patients with COVID-19 clinical syndrome, we found the updated RUCAM scale to be useful in establishing “possible” or “probable” DILI likelihood as evident by the respective kappa values; this results if validated by larger sample sized studies will extend the clinical application of this universal tool for adjudication of DILI.