Preoperative pulmonary evaluation of the thoracic surgical patient

A test designed to separate those undergoing thoracic surgery without complications and those with complications must be both highly specific and sensitive. Clearly, the difference between patients at opposite ends of the population curves is easy to identify. Spirometry can be helpful for screening, although it is not a very discriminating test. If patients fall in the overlap region between the populations, however, it is impossible to discern the risks with any certainty using low-yield tests. A test with higher sensitivity, specificity, and predictive values is necessary to ascertain such marginal differences. With this kind of analysis at hand, preoperative testing can be divided into three predictive value groups. Calculating the predictive value of each preoperative test can provide a comparative measure of usefulness of discriminative power (Table 1). In this way, spirometry, blood gas analysis, and stair climbing tolerance are shown to be poor predictors of outcome. An intermediate predictive value can be achieved using diffusion capacity, exercise-induced decreases in O2 saturation, and exercise PVR. High predictive value can be accomplished with combination indexes (PPP, possibly PRQ), measurement of VO2 at 40 watts of exercise, or VO2max. Logic dictates a step-wise preoperative evaluation using prediction value analysis (Fig.4). A flow decision chart for the preoperative evaluation of patients for pulmonary resection begins with exercise oximetry, spirometry, and blood gas analysis as general screening tests to separate those patients at minimal or no risks for complications from those patients that require further evaluation. Functional indexes (PPP, PRQ) or exercise testing can aid further in the selection of those patients in whom a nonsurgical option should be considered. Flow decision chart for the preoperative evaluation of patients for pulmonary resection should continue to evolve as new information about outcome studies is gathered. Examination of outcome data will provide us with reduction of the size of the nonoperable population, so that we can deny only those patients who truly pose a prohibitive risk.