Long-term effects of COVID-19 on kidney function

We read with interest the long-term follow-up data from hospitalised patients with COVID-19 study by Chaolin Huang and colleagues.¹ The investigators report that 107 (13%) of 822 study participants with an estimated glomerular filtration rate (eGFR; calculated with the Chronic Kidney Disease Epidemiology Collaboration equation²) of 90 mL/min per 1·73 m² or more and no acute kidney injury during the acute phase had an eGFR of less than 90 mL/min per 1·73 m² at follow-up. Huang and colleagues interpret this observation as persistent renal dysfunction. A persistent and potentially progressive reduction in eGFR in the absence of acute kidney injury at the time of acute infection would indeed have important implications for COVID-19 follow-up surveillance. However, we want to point out that an alternative explanation is possible. eGFR is calculated on the basis of serum creatinine values, which undergo small fluctuations over time as a result of shifts in hydration and other factors.³ Such fluctuations will stochastically place some individuals with normal GFR in the eGFR group of 90 mL/min per 1·73 m² or more during acute disease and in the eGFR group of less than 90 mL/min per 1·73 m² at follow-up, which is not necessarily a sign of worsening kidney function. Huang and colleagues show an opposite seeming improvement of kidney function with an eGFR of 90 mL/min per 1·73 m² or more in 142 (29·7%) of 478 patients at follow-up with an eGFR of less than 90 mL/min per 1·73 m² and no evidence of acute kidney injury during the acute disease. We encourage the investigators to show eGFR trajectories between acute phase and follow-up independent from cutoffs to substantiate the robustness of their findings.