Racial Factors Influence Appropriate Albumin Infusion After Undergoing Large Volume Paracentesis: Retrospective Study From Multiple Community-Based Hospitals

Introduction: Cirrhotic patients undergoing inpatient large volume paracentesis (LVP) are at increased risk of paracentesis induced circulatory dysfunction (PICD) and mortality. Current AASLD guidelines recommend an albumin infusion of 6-8 g per liter of fluid removed when undergoing LVP. Adherence to these recommendations in a community-based setting is unclear. We aimed to assess adherence to albumin infusion after LVP and associated factors in receiving appropriate albumin infusion. Methods: This was a retrospective study. We included all patients who received an inpatient LVP (defined as ≥ 5 liters removed) at 4 community-based hospitals between 01/01/2013 and 12/31/2017. We queried the billing database for paracentesis using CPT code 49082 and 49083 and ICD-9 code 54.91. We identified patients who met the following inclusion criteria: (1) age ≥ 18; (2) ≥ 5L ascites removed; (3) documented cirrhosis prior to LVP; (4) inpatient LVP. Exclusion criteria included: (1) ESRD; (2) GI bleeding requiring RBC transfusion; (3) on vasopressor support; (4) spontaneous bacterial peritonitis; and (5) allergy to albumin. We assessed the overall frequency of adherence, used Wilcoxon rank sum tests, repeated measure GEEs for comparisons, and a multivariable GEE analysis to identify factors associated with adherence. Results: There were 328 LVPs performed that met our criteria. Patients were predominately white (79%) and male (65%). Most cirrhosis was caused by alcohol (46%), followed by Hep C (16%), NASH (15%) and other (22%). The median MELD was 19 (IQR = 14-23). Only 91/328 LVPs (28%) received the appropriate albumin infusion. In addition, 166/328 (55%) patients received any dose of albumin after LVP. African-Americans were much less likely to receive appropriate albumin infusion compared to Caucasians (13% vs 34%; p = 0.008). Similarly, in a multivariable GEE repeated measures analysis, African-American race was the only factor that influenced appropriate albumin infusion after LVP (Z = -2.68; p = 0.007). Conclusion: Cirrhotic patients undergoing LVP are at increased risk of PICD and mortality. Appropriate albumin infusion after LVP is recommended to help minimize this. Without systematic interventions, hospitals are not only at risk for poor adherence to albumin infusion guidelines but are inadvertently widening racial gaps. Future studies are needed to assess interventions that increase adherence of albumin infusion after LVPs, for all patients in the community-setting.