Outcomes of COVID-19 in patients with CKD: A Scottish observational study

ABSTRACT

Background: Many studies have reported a poor prognosis from COVID-19 infection among patients with reduced eGFR. These findings may be partly explained by the high burden of comorbidities affecting patients with CKD but the impact of CKD itself is unclear. The aim of this study was to evaluate and quantify the effect of CKD on COVID-19-related deaths, hospitalisations and cardiovascular (CV) events in a Scottish population, using a propensity-score-based method to eliminate confounding factors. Methods: All adult patients in Tayside and Fife, Scotland, UK with a COVID-19 positive PCR test between 03/ 2020 and 02/2021 were included in the cohort and further divided into two groups (CKD and non-CKD). Patients were included in the CKD group if their most proximal eGFR was <60 mL/min/1.73 m2 and they had another eGFR measurement<60 at least 90 days before. A Covariate Balancing Propensity Score (CBPS) algorithm was implemented to account for systematic differences in their baseline characteristics. Doubly robust Cox and logistic regression enabled to estimate the effect of CKD on COVID-19 outcomes. Results: The cohort included 4556 patients (858 with CKD and 3698 without CKD). Patients with CKD were older (84 versus 59 years old, p < 0.001) and had a higher burden of comorbidities. After implementation of the CBPS algorithm, the characteristics of the two groups were similar. Patients with CKD experienced worse outcomes than those without CKD (mortality 37% versus 12%, hospitalisation 38% versus 23%, CV events 12% versus 3%, p < 0.001). Although reduced after CBPS adjustment, those differences remained significant for COVID-19-related deaths and CV events but not for the risk of hospitalisation or the length of stay. The cause specific hazard ratio for COVID-19-related deaths was 3.67 (95% CI 3.17 to 4.24 - logrank test p < 0.001) before CBPS and 1.28 (95% CI 1.07 to 1.52 - p < 0.001) after CBPS adjustment (Figure 1). For all outcomes, the risk magnitude progressively increased as lower eGFR values were reached. Indeed, after CBPS adjustment, having CKD G3A was not associated with a higher risk of COVID-19-related outcomes. However, CKD G4/5 or being on dialysis was associated with an approximately twice higher risk of death and CV event, compared with having a normal renal function (Figure 2 for risk of death). Conclusion: Patients with CKD experience worse outcomes following a COVID-19 infection, compared to those with a preserved renal function. A large part of this increase in risk for poor outcomes is related to the greater burden of comorbidities affecting patients withCKD, but our analyses suggest thatCKDis also an independent risk factor. The effect of CKD on COVID-19 severity was particularly marked in more advanced CKD stages. Patients with CKD are at increased vulnerability to COVID-19 and should be prioritised for preventive measures such as vaccination. (Table Presented).