Background: Therapeutic paracentesis is a common procedure performed to relieve symptoms of ascites most commonly from cirrhosis, heart failure or malignancy. Previous studies have shown that acute kidney injury (AKI) is common after paracentesis and portends a poor prognosis, although these studies have excluded patients with pre-existing chronic kidney disease (CKD). Experiences in our hospital ked us to question whether patients with CKD receiving a paracentesis may be more prone to developing AKI. To prevent AKI after paracentesis, intravenous albumin is administered when more than 5 liters are removed, according to guidelines from the American Association for the Study of Liver Diseases (AASLD). A recent Cochrane review found that evidence supporting this practice is of poor quality. Less still is known about whether patients with CKD benefit from albumin administered to prevent AKI after paracentesis. Due to the cost of albumin, understanding its efficacy for preventing AKI is critical to providing high-value care to hospitalized patients with ascites receiving paracentesis. As such, we designed a study to determine whether history of CKD predisposes to AKI after paracentesis and whether albumin administration reduces risk of developing AKI after paracentesis in patients with CKD.
Methods: We reviewed over 1100 inpatient bedside paracenteses performed at Strong Memorial Hospital from June 2016 through June 2020. Patients on dialysis, requiring vasopressors or without follow-up serum creatinine within 7 days were excluded. AKI was defined as a rise in serum creatinine of 0.3 mg/dL or 1.5 times the pre-procedure creatinine within 7 days post-procedure. Multivariate logistic regression was used to determine whether covariates of age, gender, race/ethnicity, etiology of ascites, volume of ascites removed and use of antihypertensives and diuretics 24 hours peri-procedure affected the incidence of AKI.
Results: After exclusions, 872 paracenteses performed on 473 unique patients were analyzed. AKI occurred after 289 procedures in 193 unique patients. In multivariate logistic regression analysis, any degree of CKD greater than CKD2 was associated with AKI (OR 1.64, 95% CI 1.22 – 2.2). Each progressive stage of CKD was associated with an increasing fraction of patients who developed of AKI: CKD1 97/361 (26.9%), CKD2 79/239 (33.0%), CKD3a 44/119 (36.9%), CKD3b 46/104 (44.2%), CKD 4 21/45 (46.7%), CKD5 2/4 (50%). Albumin failed to prevent AKI in patients with CKD (OR 1.05, 95% CI 0.77-1.56).
Conclusions: Our data have practice-changing implications suggesting that caution be taken in performing therapeutic paracentesis in patients with CKD who may be at higher risk of AKI. This risk was not reduced by administering albumin. Since albumin is an expensive resource, we join with others in recommending larger prospective, randomized studies to ascertain whether albumin should be given after paracentesis to prevent AKI and advise that patients with CKD be included in future studies.