Bundled strategies against infection after liver transplantation: Lessons from multidrug-resistant Pseudomonas aeruginosa

Abstract

Infection is a life-threatening complication after liver transplantation (LT). A recent outbreak of multidrug-resistant Pseudomonas aeruginosa triggered changes in our infection control measures. This study investigated the usefulness of our bundled interventions against postoperative infection after LT. This before-and-after analysis enrolled 130 patients who underwent living donor or deceased donor LT between January 2011 and October 2014. We initiated 3 measures after January 2013: (1) we required LT candidates to be able to walk independently; (2) we increased the hand hygiene compliance rate and contact precautions; and (3) we introduced procalcitonin (PCT) measurement for a more precise determination of empirical antimicrobial treatment. We compared factors affecting the emergence of drug-resistant microorganisms, such as the duration of antimicrobial and carbapenem therapy and hospital stay, and outcomes such as bacteremia and death from infection between before (n = 77) and after (n = 53) the LT suspension period. The utility of PCT measurement was also evaluated. Patients' backgrounds were not significantly different before and after the protocol revision. Incidence of bacteremia (44% versus 25%; P = 0.02), detection rate of multiple bacteria (18% versus 4%; P = 0.01), and deaths from infections (12% versus 2%; P = 0.04) significantly decreased after the protocol revision. Duration of antibiotic (42.3 versus 25.1 days; P = 0.002) and carbapenem administration (15.1 versus 5.2 days; P < 0.001) and the length of postoperative hospital stay (85.4 versus 63.5 days; P = 0.048) also decreased after the protocol revision. PCT mean values were significantly higher in the bacteremia group (10.10 ng/mL), compared with the uneventful group (0.65 ng/mL; P = 0.002) and rejection group (2.30 ng/mL; P = 0.02). One-year overall survival after LT significantly increased in the latter period (71% versus 94%; P = 0.001). In conclusion, the bundled interventions were useful in preventing infections and lengthening overall survival after LT.