Race-Specific vs Race-Neutral Pulmonary Function Predicted Values in Operable Lung Cancer Patients.

Abstract

Rationale: Percent-predicted forced expiratory volume in 1 second (FEV1pp) is used for surgical risk assessment in lung cancer patients. FEV1pp is adjusted for race despite concerns regarding inaccurate estimations of lung health.

Objectives: To compare prediction of race-specific versus race-neutral FEV1pp for pulmonary complications following lung cancer resection.

Methods: Patients who underwent lung resection in the Society of Thoracic Surgeons General Thoracic Surgery Database were identified (2002-2008). We used Global Lung Initiative equations to derive race-specific and race-neutral FEV1pp and compared their performance to predict pulmonary complications. Percent predicted postoperative FEV1 (ppoFEV1) was calculated with patients categorized into low (>60%), intermediate (30-60%), and high (<30%) risk groups.

Measurements and main results: Of 24,276 patients, most were White (n=21,130; 87.0%) or Black (n=1,912; 7.9%). Race-specific equations reduced the mean FEV1pp by 5.3% for White patients and increased it 6.2% for Black patients compared to race-neutral. Multivariate models using race-neutral FEV1pp performed similarly to race-specific models in predicting pulmonary complications (sublobar resection: C-statistic 0.72; lobectomy: C-statistic 0.65; and bilobectomy/pneumonectomy: C-statistic 0.67; for both models) with similar adjusted odds ratios of FEV1pp for both equation types. In 5,422 patients with calculable ppoFEV1, 617 (11.4%) were recategorized into a higher (n=65) or lower (n=552) risk group when using race-neutral equations. Of those moving into a lower risk group, 98.0% were White. All patients reclassified into higher risk groups were Black.

Conclusions: Race-neutral FEV1pp performed equally well as race-specific equations in predicting pulmonary complications and disentangled the effect of respiratory function from race on outcomes.