Correlation of Biochemical Marker Ratios with Disease Severity in Patients with Coronavirus Disease 2019 at Admission: An Observational Study

Aim: Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2, has infected over 760 million people and resulted in nearly 7 million deaths worldwide, presenting an unprecedented public health challenge. This study assessed the prognostic value of biochemical marker ratios derived from routine laboratory tests for predicting in-hospital mortality among patients with COVID-19 at admission. Materials and Methods: A retrospective observational analysis was conducted on 134 reverse transcription-polymerase chain reaction–confirmed COVID-19 patients admitted to a tertiary care center. Baseline biochemical data were available for 133 cases; one survivor was excluded due to incomplete records. Patients were classified as survivors ( n = 94) or nonsurvivors ( n = 40). Seven ratios were calculated: blood urea nitrogen-to-albumin (BAR), neutrophil-to-albumin (NAR), C-reactive-protein-to-albumin (CAR), aspartate aminotransferase-to-lymphocyte (ALRI), aspartate aminotransferase-to-platelet (APRI), platelet-to-albumin (PAR), and albumin-to-globulin (AGR). Comparative and receiver-operating characteristic-curve analyses were performed using IBM SPSS Statistics v25 (IBM Corp., Armonk, NY, USA). Results: BAR, NAR, CAR, and ALRI were significantly elevated in nonsurvivors, whereas AGR was markedly reduced ( P < 0.05). APRI and PAR showed no significant difference. CAR demonstrated the best discriminative power (area under the curve = 0.917; 95% confidence interval 0.87–0.96), followed by NAR (0.831), ALRI (0.801), and BAR (0.770). Conclusion: CAR and NAR displayed the strongest prognostic performance and may serve as inexpensive, readily available early markers for mortality risk in hospitalized COVID-19 patients. Validation through large-scale multicenter prospective studies is recommended.