Early diagnosis and appropriate treatment of infections in cirrhosis are crucial because of their high morbidity and mortality. Multidrug-resistant (MDR) infections are on the increase in health care settings. Health-care-associated (HCA) infections are still frequently treated as community-acquired with a detrimental effect on survival. We aimed to prospectively evaluate in a randomized trial the effectiveness of a broad spectrum antibiotic treatment in patients with cirrhosis with HCA infections. Consecutive patients with cirrhosis hospitalized with HCA infections were enrolled. After culture sampling, patients were promptly randomized to receive a standard or a broad spectrum antibiotic treatment (NCT01820026). The primary endpoint was in-hospital mortality. Efficacy, side effects, and the length of hospitalization were considered. Treatment failure was followed by a change in antibiotic therapy. Ninety-six patients were randomized and 94 were included. The two groups were similar for demographic, clinical, and microbiological characteristics. The prevalence of MDR pathogens was 40% in the standard versus 46% in the broad spectrum group. In-hospital mortality showed a substantial reduction in the broad spectrum versus standard group (6% vs. 25%; P = 0.01). In a post-hoc analysis, reduction of mortality was more evident in patients with sepsis. The broad spectrum showed a lower rate of treatment failure than the standard therapy (18% vs. 51%; P = 0.001). Length of hospitalization was shorter in the broad spectrum (12.3 ± 7 days) versus standard group (18 ± 15 days; P = 0.03). Five patients in each group developed a second infection during hospitalization with a similar prevalence of MDR (50% broad spectrum vs. 60% standard).
Conclusions: A broad spectrum antibiotic therapy as empirical treatment in HCA infections improves survival in cirrhosis. This treatment was significantly effective, safe, and cost saving.
© 2015 by the American Association for the Study of Liver Diseases.