Format

Send to

Choose Destination
Proc (Bayl Univ Med Cent). 2017 Jul;30(3):262-264.

Imaging and clinical predictors of spontaneous bacterial peritonitis diagnosed by ultrasound-guided paracentesis.

Author information

1
New York University School of Medicine, New York, New York (Sideris, Patel); South Florida Vascular Associates, Coconut Creek, Florida (Charles); New York University Hepatology Associates, New York University Langone Medical Center, New York, New York (Park, Feldman); and Interventional Radiology Service, Memorial Sloan Kettering Cancer Center, New York, New York (Deipolyi).

Abstract

Spontaneous bacterial peritonitis (SBP) is a potentially life-threatening complication of ascites diagnosed by paracentesis. We determined predictors of SBP to facilitate patient selection. The 301 paracenteses performed in 119 patients (51 women, 68 men) from July to November 2015 were retrospectively reviewed. Presentation, lab data, depth of the deepest ascites pocket on ultrasound, total volume of ascites removed, absolute neutrophil count, and complications were studied. Of 301 paracenteses, 16 (5%) diagnosed SBP. On univariate analysis, SBP was associated negatively with history of cirrhosis and positively with history of cancer, abdominal pain, greater depth of the fluid pocket, prior SBP, and leukocytosis. Multivariate analysis using these variables to predict SBP was significant (P < 0.0001); only depth of the largest fluid pocket (P = 0.008) and complaint of abdominal pain (P = 0.006) were independent predictors. Receiver-operator curve analysis showed that a 5-cm cutoff of pocket depth yielded 100% sensitivity and 32% specificity. Two (0.1%) hemorrhagic complications occurred, one causing death and one necessitating laparotomy. In conclusion, deeper ascites pockets and abdominal pain are independent predictors of SBP. When the largest ascites pocket is <5 cm, the probability of SBP is nearly negligible. Given the potential for hemorrhagic complications, findings may help triage patients for paracentesis.

PMID:
28670052
PMCID:
PMC5468008

Supplemental Content

Full text links

Icon for PubMed Central Icon for NYU School of Medicine
Loading ...
Support Center