Comparison of Duke Activity Status Index with cardiopulmonary exercise testing in cancer patients

The Duke Activity Status Index (DASI), a patient-administered questionnaire, is used to quantify functional capacity in patients undergoing cancer surgery. This retrospective cohort study assessed whether the DASI was accurate in predicting peak oxygen consumption (pVO(2)) that was objectively measured using cardiopulmonary exercise testing (CPET) in 43 consecutive patients scheduled for elective major cancer surgery at a tertiary cancer centre. The primary outcome measured the limits of agreement between DASI-predicted pVO(2) and actual measured pVO(2). The study population was elderly (median 63 years, interquartile range 18), 58% were male, with the majority having intraabdominal cancer surgery. Although the DASI scores were statistically related to the measured pVO(2) (N = 43, adjusted R (2) = 0.20, p = 0.002), both the bias (8 ml kg(- 1) min(- 1)) and 95% limits of agreement (19.5 to - 3.4 ml kg(- 1) min(- 1)) between the predicted and measured pVO(2) were large. Using some of the individual components, recalibrating the intercept and regression coefficient of the total DASI score did not substantially improve its ability to predict the measured pVO(2). In summary, both the limits of agreement and bias between the measured and DASI-predicted pVO(2) were substantial. The DASI-predicted pVO(2) based on patient's assessment of their functional status could not be considered a reliable surrogate of measured pVO(2) during CPET for the population of patients pending major cancer surgery and cannot, therefore, be used as a triage tool for referral to CPET centres for objective risk assessment.