Abstract: |
PURPOSE: To develop a new quantitative predictor of post-thoracotomy pulmonary complications using routine lung function tests. METHODS: Data of consecutive patients who underwent lung resection only for malignancy and had preoperative pulmonary function tests; FEV1 and single breath DCO, resting arterial blood gas analysis; P(A-a)O2 and quantitative lung perfusion scans performed in our laboratory were included in this study. Predicted postoperative values (ppo) FEV1%, ppo DCO% and P(A-a)2O were calculated for all patients. Pulmonary complication was defined as pneumonia, atelectasis or respiratory insufficiency requiring therapy. Statistical analysis was performed using Stepwise Logistic Regression (p≤.01) and Relative Risk (odds ratio). RESULTS: Sixty patients were studied, pneumonectomies (n=30) and lobectomies (n=30). The mean age was 65±11 years. Nine patients had pulmonary complications including 2 deaths. The following formula was PRQ=(ppo FEV1%)x(ppo DCO%)2/P(A-a)O2 A PRQ < 3200 was associated with a sensitivity of 100% and specificity of 95% for pulmonary complications. PRQ was associated with a positive predictive value of 75% and a negative predictive value of 100% for pulmonary complications. A PRQ < 3200 was associated with an estimated relative risk of pulmonary complications of 54:1 (95% CL3-972). CONCLUSIONS: The construct PRQ has an excellent association with postoperative pulmonary complications. This novel index not only incorporates predicted postoperative estimates of FEV1% and DCO% but adds P(A-a)O2 and weights the relative contributions of each of the three variables. Pulmonary Respiratory Quotient may be the most significant predictor of pulmonary morbidity in patients undergoing lung resection. CLINICAL IMPLICATIONS: This new index warrants serious prospective evaluation as a predictor of pulmonary morbidity and mortality. |