Urine, serum markers may help predict COVID-19 outcomes

Urinary and serum biomarkers may help the early detection of acute kidney injury (AKI) in patients with COVID-19, signalling renal involvement and the potential need for greater medical attention, a recent study has found.

Researchers conducted a retrospective observational study on 199 adult patients with COVID-19. Urinalysis had been conducted using a biochemical and flow cytometer sediment analyser, while appropriate assays were performed on serum samples.

Ultimately, 12.1 percent of patients needed intensive care. Levels of urinary biomarkers, such as glucose (p=0.005), protein (p=0.001), and ketones (p=0.024) were significantly altered in these patients. The same was true for serum molecules such as sodium (p<0.001) and albumin (p<0.001).

Similar patterns were observed for AKI, which developed in 15 patients over COVID-19’s clinical course. Urinary pH (p=0.017) as well as the presence of blood (p=0.001), protein (p=0.03), and nitrite (p=0.032) were all indicative of the risk of AKI, as were serum levels of urate (p=0.003), chloride (p=0.017), and creatinine (p=0.013).

Receiver operating characteristic (ROC) curve analysis found that the combination between blood in the urine and chronic kidney disease had the best predictive capacity for AKI, with an area under the curve (AUC) of 0.676 (95 percent confidence interval [CI], 0.512–0.840).

Notably, ROC analysis also showed that combinations of biomarkers could help predict COVID-19 death. In particular, blood in urine and serum lactate dehydrogenase seemed to be a powerful combination, yielding an AUC of 0.923 (95 percent CI, 0.866–0.979).