Exercise-induced myocardial dysfunction detected by cardiopulmonary exercise testing is associated with increased risk of mortality in major oncological colorectal surgery

Jason Mann, Murray Williams, Jonathan Wilson, David Yates, Alexander Harrison, Patrick Doherty, Simon Davies

Research output: Contribution to journalArticlepeer-review


BACKGROUND: Cardiopulmonary exercise testing (CPET) identifies high-risk patients before major surgery. In addition to using oxygen uptake and ventilatory efficiency to assess functional capacity, CPET can be used to identify underlying myocardial dysfunction through the assessment of the oxygen uptake to heart rate response (oxygen pulse response). We examined the relationship of oxygen pulse response, in combination with other CPET variables and known cardiac risk factors, with mortality after colorectal cancer surgery.

METHODS: This work focused on a retrospective cohort study of patients who had CPET and underwent colorectal cancer surgery. The primary outcome was a composite of in-hospital and 30-day mortality. Ventilatory inefficiency (Ve/Vco2>34) and exercise-induced myocardial dysfunction (abnormal oxygen pulse response) were investigated for an association with mortality using bivariable analysis and multivariable Cox regression.

RESULTS: A total of 1214 patients who underwent colorectal cancer surgery were included, and the primary outcome occurred in 26 patients (2.1%). Multivariable Cox regression showed abnormal oxygen pulse response was independently associated with the primary outcome (odds ratio [OR]=2.75; 95% confidence interval [CI], 1.17-6.47). Bivariable analysis showed that Ve/Vco2 >34 was associated with the primary outcome (OR=3.43; 95% CI, 1.47-8.01). Combining Ve/Vco2 >34 and abnormal oxygen pulse response conferred an increased risk for the primary outcome (OR=4.47; 95% CI, 1.62-12.34), compared with Ve/Vco2 >34 and normal oxygen pulse response.

CONCLUSION: Ventilatory inefficiency and an abnormal oxygen pulse response were independently associated with short- (30-day) and long-term (2-yr) mortality. Oxygen pulse response may provide additional information when considering perioperative risk stratification.

Original languageEnglish
Pages (from-to)473-479
Number of pages7
JournalBritish journal of anaesthesia
Issue number4
Early online date18 Feb 2020
Publication statusE-pub ahead of print - 18 Feb 2020

Bibliographical note

Copyright © 2020 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.

Cite this