Pericarditis in uremic patients: serum albumin and size of pericardial effusion predict drainage necessity

Abstract

BACKGROUND: Pericardial effusion in uremic patients (UPE) was first described by R. Bright in 1836. It is generally agreed that patients require emergency pericardial drainage when tamponade signs are present, but in patients with no tamponade the optimal timing for drainage remains unclear. METHODS: To define patients who will require pericardial drainage, we retrospectively studied risk factors for pericardial drainage in patients admitted with pericardial effusion and chronic renal failure. RESULTS: Between 2000 and 2012, 44 UPE patients were identified using the database of our institution: 43% were under hemodialysis, 7% under peritoneal dialysis, 11% transplanted, 39% had chronic kidney disease (CKD) stage 4 or 5. Cause of UPE was uremic pericarditis in 45.5%, dialysis pericarditis in 45.5%, and other in 9%. On initial echocardiography, UPE was estimated small (<300 ml) in 38%, moderate (300-500 ml) in 32%, and large (>500 ml) in 30%. Tamponade signs were observed in 16% of patients. During follow-up, 100 % of large effusions required drainage (70% immediate, 30% delayed). For moderate and small UPE, the initial size on echocardiography was not discriminating. Serum albumin level was highly predictive of the risk of drainage: when albuminemia was ≤31 g/l, 35% patients were drained vs. only 7% when albuminemia was >31 g/l. CONCLUSION: In this first study reporting UPE drainage risk factors, all large UPE required drainage even when extra-renal epuration intensification or medical treatment were tried. This suggests that large UPE should be drained without delay. For small and moderate UPE, size of effusion on echocardiography does not predict drainage requirement but serum albumin level does.