Clinical outcomes after bedside and interventional radiology paracentesis procedures

Background: Increasingly, paracentesis procedures are performed in interventional radiology (IR) rather than at the bedside. No guidelines exist to aid decision-making about the best location, and patient outcomes are unknown. Our aims were to develop a prediction model for which location (bedside vs IR) clinicians select for inpatient paracentesis procedures, and to compare clinical outcomes.

Methods: We performed an observational medical records review of all paracentesis procedures performed on the hepatology service of an 894-bed urban tertiary care hospital from July 2008 through December 2011. We developed a prediction model to determine factors for IR referral. Clinical outcomes including blood product transfusions, intensive care unit (ICU) transfer, hospital length of stay, inpatient mortality, 30-day readmission, and emergency department visit within 30 days of discharge were compared between patients who had bedside versus IR procedures.

Results: Five hundred two patients who underwent a paracentesis were included in the analysis. Being female, higher body mass index, lower volume of ascites removed, and attending physician of record predicted the probability of IR referral. IR referrals were associated with 1.86 additional hospital days (P=.003). Platelet and fresh frozen plasma transfusions were more common in patients who underwent IR procedures (odds ratio [OR] 4.56; 95% confidence interval [CI], 2.13-9.78 and OR 4.07; 95% CI, 2.03-8.18, respectively). Subsequent ICU transfers also were more common among patients who had IR procedures (OR 2.21; 95% CI, 1.13-4.31). All other clinical outcomes were similar between groups.

Conclusions: The decision to perform a paracentesis procedure at the bedside or in IR is largely discretionary. Paracentesis procedures performed at the bedside result in equal or better patient outcomes. Clinicians should receive the training needed to perform paracentesis procedures safely at the bedside. Large prospective studies are needed to confirm the findings of this study and inform national practice patterns.