P45 Determinants and impact of impaired aerobic capacity on survival in patients with cirrhosis: results from a 2600 patient cohort

<h3>Background</h3> Aerobic capacity quantified using maximal oxygen uptake (VO2 max) at cardio-pulmonary exercise testing (CPET) is often impaired in cirrhotic chronic liver disease (CLD) and impacts survival. Important as a potentially modifiable risk factor, its clinical associations have limited characterisation. In a very large patient cohort undergoing CPET as part of liver transplant (LT) assessment we explored these and its impact on long term survival with and without LT. <h3>Methods</h3> We studied adult patients with CLD undergoing assessment for first LT as single high-volume centre over the period 2010–2022. Incremental CPET used bike ergometer and time-to-event analysis Kaplan-Meier and Cox-regression. <h3>Findings</h3> The study cohort comprised 2624 subjects (64% male) of median age 56 (IQR 47–62) years. MELD score was 11 (7–15) and Child-Pugh 8 (7–10); cause of CLD was alcohol in 33%, Viral in 18% and MASLD in 12%. Median follow-up was 857 (242–2313) days; 1860 (71%) were waitlisted for LT of whom 1493 (80%) underwent LT 104 (34–236) days later. There were 226 (12%) deaths on the waitlist. At CPET peak heart rate was 138 (121–156) bpm and workload 91 (69–118) W. VO2 max was 16.5 ml/kg/min (13–20.2) with independent association with age, sex, body mass index, diabetes, MELD and presence of ascites (all p&lt;0.001) and was 63% (49–78) predicted. VO2 max varied between etiologies, lowest in MASLD: 13.9 (11.2–17.1) vs. Others 16.8 (13.4–20.6), (p&lt;0.001). Patients waitlisted for LT had higher VO2 max those not waitlisted 17 (13.6–20.8) vs. 15 (11.6–18.8), and as% predicted 64 (50–79) vs. 59 (45–74) (both p&lt;0.001). Those who died on the waitlist had lower VO2 predicted than survivors; 57% (46–71) vs. 64% (51–80) (p&lt;0.001). VO2 max was independently associated with waitlist death when adjusted for age and MELD score (p&lt;0.001). VO2 max was independently associated with non-LT survival, with incremental worsening with lower VO2 (figure 1A, p&lt;0.001 log-rank), with no association with post-LT survival (figure 1B, p=0.436). In non-LT patients VO2 predicted of &lt;50% had adjusted hazard ratio for death of 1.75 (95% CI 1.35–2.302) compared to those with &gt;80%, p&lt;0.001. <h3>Conclusions</h3> Maximal oxygen uptake is often impaired in CLD and linked with specific clinical features, and has close, independent association with non-LT but not post-LT survival. Further exploration of interventions to improve this modifiable risk factor for mortality.