Cardiopulmonary Exercise Testing for Preoperative Risk Assessment before Pancreaticoduodenectomy for Cancer

Abstract

Objective. Pancreaticoduodenectomy is the standard of care for tumors confined to the head of pancreas and can be undertaken with low operative mortality. The procedure has a high morbidity, particularly in older patient populations with preexisting comorbidities. This study evaluated the role of cardiopulmonary exercise testing to predict postoperative morbidity and outcome in high-risk patients undergoing pancreaticoduodenectomy. Methods. In a prospective cohort of consecutive patients undergoing pancreaticoduodenectomy, those aged over 65 years (or younger with comorbidity) were categorized as high risk and underwent preoperative assessment by cardiopulmonary exercise testing (CPET) according to a predefined protocol. Data were collected on functional status, postoperative complications, and survival. Results. A total of 143 patients underwent preoperative assessment, 50 of whom were deemed to be at low risk for surgery per study protocol. Of 93 high-risk patients, 64 proceeded to surgery after preoperative CPET. Neither anaerobic threshold (AT) nor maximal oxygen consumption ( V˙ o 2 MAX) predicted patient mortality or morbidity. However, ventilatory equivalent of carbon dioxide ( V˙ e/ V˙ co 2) at AT was a predictive marker of postoperative mortality, with an area under the curve (AUC) of 0.84 (95 % confidence interval [CI] 0.63–1.00, p = 0.020); a threshold of 41 was 75 % sensitive and 95 % specific (positive predictive value 50 %, negative predictive value 98 %). Above this threshold, raised V˙ e/ V˙ co 2 predicted poor long-term survival (hazard ratio 2.05, 95 % CI 1.09–3.86, p = 0.026). Conclusions. CPET is a useful adjunctive test for predicting postoperative outcome in patients being assessed for pancreaticoduodenectomy. Raised CPET-derived V˙ e/ V˙ co 2 predicts early postoperative death and poor long-term survival.