This article is more than 5 years old. Information may no longer be current.

New model can identify in-hospital mortality of patients with SBP

Researchers from Barcelona developed a model that predicted the in-hospital mortality of patients with spontaneous bacterial peritonitis at a sensitivity of 0.85 and specificity of 0.75, according to recent findings published in Alimentary Pharmacology & Therapeutics. The model includes Child-Pugh score, serum urea, blood leukocyte count and mean arterial pressure.

“The main finding in the present study was the development and validation of a predictive model of mortality for high-risk [spontaneous bacterial peritonitis (SBP)] patients treated with antibiotics and albumin,” Maria Poca, MD, at the Universitat Autónoma de Barcelona, and colleagues wrote. “We focused on high-risk SBP patients because low-risk SBP patients have a low mortality rate, independently of being treated with albumin or not.”

Despite albumin treatment, in-hospital mortality of high-risk patients with SBP remains between 20% and 30%, the researchers wrote. To better identify which patients would benefit from albumin, Poca and colleagues analyzed all patients with cirrhosis treated for high-risk SBP at their hospital between 2001 and 2011. The researchers defined “high-risk” SBP as a serum urea of at least 11 mmol per liter and/or a serum bilirubin of at least 68 µmol per liter. The researchers included 118 episodes of SBP treated with antibiotics and albumin. Afterward, they developed and validated a model of in-hospital mortality in a separate group of 161 patients.

Their model included Child-Pugh score, serum urea, blood leukocyte count and mean arterial pressure. The model showed a discrimination accuracy of 0.85 (95% CI, 0.777-0.922). When using a cut-off point of 0.245, the model had a sensitivity of 0.85 and specificity of 0.75. In-hospital mortality was 28% before using the model, 57.1% in patients when using the model above the cut-off and 7.2% in patients below the cut-off (P < .001). In the validation cohort, the in-hospital mortality was 24.8% before using the model, 39% when using the model above the cut-off and 11.9% below the cut-off (P < .001).

These findings may help identify patients who can benefit from additional therapeutic strategies, the researchers concluded.

“A main contribution of our study is that we focused only on high-risk SBP patients, as low-risk patients were not included in the analysis,” the researchers wrote. “We consider this approach more suitable if we aim to identify the subgroup of patients with a poor prognosis that could benefit from other therapeutic measures, in addition to antibiotics and albumin.” – by Will Offit

Disclosure: The researchers report no relevant financial disclosures.