Background: Spontaneous bacterial peritonitis (SBP) is a commonly encountered complication of end stage liver disease that carries a high morbidity and mortality. Risk factors for SBP are defined by the American Association for the Study of Liver Diseases (AASLD) as a prior history of SBP, active gastrointestinal bleeding, and low ascites total protein (< 1.5 g/dL). For patients with low ascites total protein, the AASLD recommends initiating antibiotic prophylaxis in those with an ascitic fluid protein of <1.5 g/dL and at least one of the following: creatinine ≥1.2, BUN ≥25, serum sodium ≤130, Child score ≥8, or total bilirubin of ≥3. Any of these risk factors benefit greatly from prophylactic antibiotics; in some studies reductions as high as 80% in SBP incidence and 40% in mortality rates have been demonstrated. Nonetheless, primary SBP prophylaxis rates for low protein ascites are lower than expected. This study evaluated the impact of an electronic medical record (EMR) dot phrase with a clinical reminder on rates of SBP antibiotic prophylaxis among eligible patients.
Methods: We conducted a retrospective chart review of all patients with cirrhosis and a documented ascitic total protein of <1.5 g/dL hospitalized at an 800-bed urban health system. Patients were excluded if they died or were discharged to hospice during the admission, were already on SBP antibiotic prophylaxis prior to admission, or were prescribed antibiotics for a reason other than low protein ascites. Rates of discharge with appropriate antibiotic prophylaxis among eligible patients was the primary outcome. Baseline rates were calculated in the pre-intervention period (June 2017 – June 2018), and again following the introduction of the EMR dot phrase/clinical reminder (July 2018 – July 2019). Use of the dot phrase was optional, so usage rates were computed as a secondary outcome.
Results: In the pre-intervention period, 72 patients not previously on SBP prophylaxis were admitted with cirrhosis and low-protein ascites. Among patients eligible for antibiotic prophylaxis, 56% were not prescribed appropriate therapy on discharge. Inpatient hepatology consultation was not associated with higher rates of SBP antibiotic prophylaxis. In the post-intervention period, an additional 65 patients met inclusion criteria. Antibiotic prescription rates improved significantly following introduction of the EMR dot phrase (68.9% vs. 44.9%, p <0.05). Rates of dot phrase use were 75.3% of the total sample. Providers who used the dot phrase were more likely to prescribe antibiotics to eligible patients, compared to those who did not use the phrase (77.6% vs. 14.3%, p<0.001).
Conclusions: In this single-center study, a clinical reminder added to the EMR improved rates of primary SBP antibiotic prophylaxis eligible patients. The intervention described is low cost and easily implementable in hospital centers with an existing EMR. Efforts to increase use of the phrase would be important to optimize reduction in the high burden of morbidity and mortality in patients with cirrhosis.