Submaximal cardiopulmonary exercise testing predicts complications and hospital length of stay in patients undergoing major elective surgery

• Published in: Annals of surgery Vol. 251; no. 3; p. 535
• Main Authors: Snowden, Chris P, Prentis, James M, Anderson, Helen L, Roberts, Digby R, Randles, Derek, Renton, Morag, Manas, Derek M
• Format: Journal Article
• Language: English
• Published: United States 01.03.2010
• Subjects: Aged; Elective Surgical Procedures - adverse effects; Exercise Test; Female; Heart Function Tests; Humans; Length of Stay - statistics & numerical data; Male; Postoperative Complications - epidemiology; Predictive Value of Tests; Prospective Studies; Respiratory Function Tests
• ISSN: 1528-1140
• DOI: 10.1097/sla.0b013e3181cf811d

To investigate the null hypothesis that an objective, noninvasive technique of measuring cardiorespiratory reserve, does not improve the preoperative assessment of patient risk of postoperative complications, when compared with a standard questionnaire-based assessment of functional capacity. Postoperative complications may be increased in patients with reduced cardiorespiratory function. Activity questionnaires are subjective, whereas cardiopulmonary exercise testing (CPET) provides an objective definition of cardiorespiratory reserve. The use of preoperative CPET to predict postoperative complications is not fully defined. CPET and an algorithm-based activity assessment (Veterans Activity Questionnaire Index [VASI]) were performed on consecutive patients (n = 171) with low subjective functional capacity (metabolic equivalent score [METS] < 7), being assessed for major surgery. A morbidity survey determined postoperative day 7 complications. Logistic regression defined independent predictors of complication group. Receiver-operating curve (ROC) analysis defined the predictive value of CPET to outcome. P < 0.05 value demonstrated significance. Objective cardiorespiratory reserve did not differ between operated (n = 116) and nonoperated patients (n = 55). Median complication rate on postoperative day 7 was 1. Patients with >1 complication had an increase in hospital LOS compared to the group with < or =1 complication (26 vs. 10 days; P < 0.001). Anaerobic threshold (AT) was higher in the group with < or =1 complication (11.9 vs. 9.1 mL/kg/min; P = 0.001) and demonstrated high accuracy (AUC = 0.85), sensitivity (88%), and specificity (79%), at an optimum AT of 10.1 mL/kg/min (defined by the furthest left point on the ROC curve). AT, VASI, and surgical reintervention were independent predictors of complication group. Preoperative AT significantly improved outcome prediction when compared with the use of VASI alone. An objective measure of cardiorespiratory reserve was an independent predictor of a major surgical group with increased postoperative complications and hospital LOS. AT measurement significantly improved outcome prediction compared with an algorithm-based activity assessment.