Ventilatory threshold predicts episodes of sepsis in end-stage liver disease patients awaiting liver transplantation

Matthew Wallen, Adrian Hall, Tina L. Skinner, Aidan J. Woodward, Jeff Coombes, Graeme A. Macdonald

Research output: Contribution to journalMeeting Abstractpeer-review


Background and Aims: Patients with end-stage liver disease frequently have reduced cardiorespiratory fitness and are at increased risk of sepsis. The aim of this study was to determine if ventilatory threshold measured during cardiopulmonary exercise testing is associated with the incidence of septic episodes in patients with non-cholestatic end-stage liver disease awaiting liver transplantation.
Methods: Patients completed a cardiopulmonary exercise test on a cycle ergometer prior to being listed for liver transplantation. Successful completion of the test was defined as the ability to reach ventilatory (anaerobic) threshold, calculated using the V-slope method. Sepsis was defined as episodes of clinical sepsis requiring hospital admission. These were diagnosed in accordance with accepted international criteria, and were recorded from the time of listing until death, removal from the waiting list or liver transplantation. Multivariate Poisson regression analysis was used to examine the impact of age, MELD score, hepatocellular carcinoma (HCC) diagnosis and ventilatory threshold on the incidence rate ratios for septic episodes. Waitlist duration was used as the offset variable in the model.
Results: Sixty-four patients with non-cholestatic end-stage liver disease (age 56.1, interquartile range 51.9–59.5; male 87.5%) achieved a mean ventilatory threshold of 11.7 ± 2.6 mL/kg/min. Their median MELD score was 15.7 ± 4.7 and 26 (41.5%) had HCC. There were 15 episodes of sepsis (5 positive blood cultures, 2 urinary tract infection, 3 spontaneous bacterial peritonitis, 2 Campylobacter enteritis, and 1 each cellulitis, lower respiratory tract infection, and appendicitis).
Ventilatory threshold was not significantly different based on number of septic episodes (no episodes 12.1 ± 2.2, one episode 10.6 ± 3.9, and two episodes 8.7 ± 1.6 mL/kg/min; p = 0.06). Poisson regression showed that pre- transplant ventilatory threshold and MELD score were significant predictors of sepsis (incidence rate ratios: 0.792, 95%CI 0.661–0.948; p = 0.011; and 1.155, 95%CI 1.018– 1.309, p = 0.025, respectively), independent of age (p = 0.95). When HCC diagnosis was included in the model, only ventilatory threshold retained significance (incidence rate ratio: 0.801, 95%CI 0.673–0.953, p = 0.012).
Conclusions: Cardiorespiratory fitness is an independent predictor of risk of sepsis in advanced liver disease. This has implications for the potential of exercise training in patients with decompensated cirrhosis.
Original languageEnglish
Article numberTHU-358
Pages (from-to)s275-s276
Number of pages2
JournalJournal of Hepatology
Issue number2 Suppl.
Publication statusPublished - 2016
Externally publishedYes


  • liver disease
  • sepsis
  • cardiorespiratory fitness


Dive into the research topics of 'Ventilatory threshold predicts episodes of sepsis in end-stage liver disease patients awaiting liver transplantation'. Together they form a unique fingerprint.

Cite this