Serum albumin, clotting activation and COVID-19 severity: a systematic review and meta-regression analysis of 4579 patients

BACKGROUND: Preliminary data showed that serum albumin (SA), an acute phase protein with anticoagulant property, is inversely associated with thrombotic complications in respiratory syndrome Coronavirus 2 (SARS-CoV-2) infection associated disease (COVID-19). We performed a meta-analysis to corroborate this finding on a large sample population. METHODS: We performed a systematic review and meta-regression analysis of clinical studies reporting data on SA according to the severity of COVID-19 disease. COVID-19 severity was defined as: 1) admission to the Intensive Care Unit; 2) acute respiratory distress syndrome; or 3) in-hospital death. RESULT S: We included 16 studies with 4,579 patients with SARS-CoV-2 infection. Mean age was 51.5 years, and 2146 (44.2%) of patients were women. Overall, 1199 (31.3%) of patients had severe COVID-19 (range 5.4% to 72.6%). Mean SA level was 37.15 g/L. The pooled analysis showed a mean difference of SA: -4.06 g/L (95% CI -4.98/-3.15) in severe COVID-19 compared to non-severe ones. This difference ranged from -7.10 g/L to 1.09 g/L. At meta regression analysis, the difference in SA levels between severe and non-severe COVID-19 patients was more evident in studies with high DDimer (P<0.001) and procalcitonin (P=0.07), suggesting a more SA reduction in patients with thrombotic/septic disease. CONCLUSIONS: SA is significantly reduced in severe COVID-19 and associated with elevated D-Dimer. Albumin supplementation may be tested as adjunctive therapeutic strategy to reduce the thrombotic risk.