Risk of infection shortly before and after a diagnosis of liver cirrhosis: a nationwide, population-based study in Sweden.

Abstract

Liver cirrhosis is associated with an increased risk of infections, but it is unknown when infection rates start to increase. This study aims to investigate the risk of infection before, and after, cirrhosis diagnosis. All patients diagnosed with cirrhosis in Sweden between 1997 and 2020 were retrospectively identified from National Patient Register, from which incident infections were also ascertained. Conditional Poisson regression was used to compare the incidence rate [IR] of infections during the diagnostic period (±3 months from cirrhosis diagnosis) to a pre-diagnostic period (the same six-months-span four years prior) among the same patients. 21,187 (50.4%) patients with decompensated, and 20,880 (49.6%) with compensated cirrhosis were identified. Alcohol-related liver disease (44.8%) and viral hepatitis (18.1%) were the most common etiologies. A total of 824 infections were observed in the pre-diagnostic, and 9892 in the diagnostic period. Compared to the pre-diagnostic period, the IR ratio during the diagnostic period was 15.7-fold higher (95%CI=14.7-16.9) for all patients, 17.2-fold higher (95%CI=15.6-19.0) in decompensated cirrhosis and 14.2-fold higher (95%CI=14.7-16.9) in compensated cirrhosis. Patients with ascites had the greatest increase in infection risk. Sepsis and peritonitis showed the highest increase in risk during the diagnostic period. Infection risk increases several years prior to cirrhosis and peaks around the time of cirrhosis diagnosis, although estimates may reflect detection bias. The progressive increase in infection risk four years proceeding cirrhosis suggests that infections may serve as clinical indicators of underlying cirrhosis, and highlights the importance of infection prevention, especially in decompensated cirrhosis.