Routine preoperative pulmonary function tests appear to have only limited utility in predicting outcomes in patients undergoing cardiothoracic surgery when the Society of Thoracic Surgeons risk score is available, according to the results of a retrospective study.
Dr. Alexander Ivanov of New York Methodist Hospital, Brooklyn, and his colleagues conducted a database analysis of 1,685 patients undergoing index cardiac surgery at New York Methodist Hospital between April 2004 and January 2014. They used the STS risk model version 2.73 to estimate postoperative risk of respiratory failure (defined as the need for mechanical ventilation greater than or equal to 72 hours, or reintubation), prolonged postoperative length of stay (defined as greater than 14 days), and 30-day all cause mortality in these patients, according to their report in The Journal of Thoracic and Cardiovascular Surgery (2016;151:1183-9).
They plotted the receiver operating characteristics curve for the STS score for each of these adverse events and compared the resulting area under the curve (AUC) with the AUC after adding pulmonary function testing parameters and COPD classifications.
A total of 1,412 patients had a calculated STS score, of which 751 underwent pulmonary function testing (53%). In general, patients who had pulmonary function testing were older and had higher rates of comorbidities and more complex cardiothoracic surgery compared with their counterparts, according to Dr. Ivanov. These patients also had significantly elevated STS risk for prolonged ventilation (12.4% vs. 10.3%), prolonged postoperative length of stay (8.9% vs. 7.2%), and 30-day mortality (2.7% vs. 2.2%).
The decision to perform pulmonary testing was left to the treating physician. Of those patients tested, 652 had bedside spirometry and 99 had formal laboratory testing. Forced expiratory volume in 1 second (FEV1) and forced volume vital capacity (FVC) values were determined by taking the best of three trials. COPD was diagnosed in cases of an FEV1/FVC ratio of less than 70%.
Among these patients, 4.5% developed postoperative respiratory failure, and there was no statistically significant difference in the respiratory failure rate between patients with and without pulmonary function testing. In addition, there was no significant difference in 30-day mortality between these patients (1.9% vs. 2.1%). However, a total of 6.9% had a prolonged postoperative length of stay, with a significantly higher rate in the patients with pulmonary function testing than without (8.8% vs. 4.7%).
Dr. Ivanov and his colleagues found that the AUC of the STS score was 0.65 (95% confidence interval [CI], 0.55-0.74)for respiratory failure, 0.67 (95% CI, 0.6-0.74) for prolonged postoperative length of stay, and 0.74 (95% CI, 0.6-0.87) for 30-day mortality. Even though the STS score based upon clinical definitions of lung disease afforded only modest discriminatory ability for the three studied outcomes, they found that there was no significant added benefit to the predictive ability of these STS scores obtained by incorporating any of the pulmonary function testing parameters or COPD classifications studied.
“A possible physiological explanation for these findings may be that the examined pulmonary function testing variables do not depend solely on pulmonary parameters such as airway diameter, degree of obstruction, or lung elasticity, but rather on a patient’s effort and muscle “strength,” characteristics that are already well captured and accounted for in the current STS model,” the researchers stated.
“The STS score calculated with clinical information on lung disease status offers modest discriminatory ability for respiratory failure, prolonged postoperative length of stay, and 30-day mortality after CT surgery, which cannot be improved by adding PFT parameters or PFT-derived COPD categorization,” they wrote. “Therefore, routine preoperative PFTs may have only limited clinical utility in patients undergoing CT surgery when the STS score is readily available. Further prospective studies will be helpful in confirming these conclusions,” Dr. Ivanov and his colleagues noted.
The authors reported that they had nothing to disclose.