Predictive ability of preoperative indices for major pulmonary surgery.

In the last years a number of significant improvements have been achieved in risk stratification for lung cancer patients who undergo lung resection. Nevertheless, future improvements should be based in prospective cooperative studies including a large number of comparable cases. First limitations of available published evidence come from the fact that FEV1 and ppo-FEV1 are pivotal in patient classification. Besides, ppoFEV1 is not a reliable predictor of complications in patients with COPD, since their functional behavior is different compared with non-COPD patients after lung resection and, although current calculation methods may be accurate in estimating the residual definitive FEVI at 3 to 6 months after surgery, they tend to overestimate the actual FEV1 in the first postoperative days, when most of the complications occur. The assumption that FEV1 and DLCO are highly correlated is the reason why ppoDLCO estimation is not recommended in all patients, limiting the predictive ability of this test. Finally, the predictive ability of exercise testing has not been proven since it is not systematically recommended in all patients before lung resection.