Outcomes of Patients Undergoing Surgery for NSCLC after “Race De-correction” for Pulmonary Function

A. Medlock¹, M. Diagut², L. Godoy², L. Brown², H. Li², D. Cooke^3
1University of California, Davis School of Medicine, Sacramento, California ²University of California, Davis, Sacramento, California ³University of California, Davis Medical Center, Sacramento, California

Purpose: Spirometry pulmonary function tests (PFTs) in the US undergo race correction for patients identified as Black (increased by 10-15%) or Asian (increased by 4-6%.) We determined if a) Black and Asian (B&A) patients have similar outcomes to their counterparts and b) if race de-correction resulted in fewer B&A surgical candidates.

Methods: This is a single institution, retrospective cohort analysis from 2007 to 2020, comparing B&A patients undergoing lobectomy for primary lung cancer to all other patients (“non-B&A”). Predicted postoperative forced expiratory volume in 1 second (ppoFEV1) was calculated based on total lung segments removed. PFTs were “de-corrected” (reduced) by 12% for Black patients and 5% for Asian patients. We determined the number of B&A patients who, after de-correction, had ppoFEV1 < 40% or < 30%, the most prominent guideline thresholds for surgical candidacy. Propensity score matching with 1:1 matching ratio for age, sex, COPD, smoking status, FEV1, and ppoFEV1 was used to generate the final analysis dataset. Descriptive statistics (e.g., mean, median, N, percentage) were utilized for the data analysis. Pulmonary complications followed the definitions of the Society of Thoracic Surgeons (STS) General Thoracic Database. Complications were compared amongst propensity matched B&A groups to non-B&A patient groups.

Results: A total of 592 patients were studied; 34 B&A (6%) and 558 Other (94%). There was no 30-day mortality amongst B&A patients compared to 8 (1.4%) amongst non-B&A patients. 30-day readmission rates were identical for both cohorts (8.8%). Seven (20%) B&A patients experienced pulmonary complications, compared to 123 (22%) non-B&A patients. Three (8.8%) B&A patients were discharged on home O2 compared to 31 (5.5%) non-B&A patients. Four percent (n=23) and 0.5% (n=3) of non-B&A patients underwent surgery with a ppoFEV1 < 40% and < 30% respectively, compared to zero B&A patients for both thresholds. After de-correction, 3 (8%) B&A patient’s undergoing surgery had ppoFEV1 < 40% and zero B&A patients had a ppoFEV1 < 30%. After propensity matching, both standard corrected and de-corrected B&A patient cohorts compared to matched non-B&A patients showed no significant differences in incidences of return to OR, 30-day readmission, 30-day mortality, need for home O2, and pulmonary complications.

Conclusion: De-correction in B&A patients in our institutional cohort would have led to 8% of B&A patients to fall below the ppoFEV1 < 40% threshold for surgical candidacy, though no difference in mortality. Also, there were no significant differences in pulmonary complications when comparing propensity matched cohorts. In moving towards race neutral PFT equations, holistic assessment beyond spirometry is necessary to maintain equal access to curative surgery.

Identify the source of the funding for this research project: Research supported by the Division of General Thoracic Surgery at the University of California, Davis.