Evaluating the Impact of Race-Neutral Interpretation of Preoperative Pulmonary Function

Authors: Katrina Steiling

Pulmonary function tests (PFTs) are essential in the preoperative evaluation of patients being considered for surgical lung resection of localized non–small-cell lung cancer (NSCLC) (1, 2). Surgical resection is the treatment of choice for localized NSCLC because it provides the best chance of curing the disease. However, poor lung function is a strong predictor of postoperative mortality (3). Thus, preoperative PFTs are necessary to identify the patients who are most likely to benefit from surgical resection of their cancers and minimize morbidity and mortality in others.

Historically, values obtained from PFTs have been interpreted and compared versus expected values calculated from reference equations that account for the self-reported social constructs of race and ethnicity as well as biologic variables that influence lung function such as age, sex, and height (4, 5). Drawbacks to this approach include a lack of specific race-based reference equations for several racial and ethnic groups, limitations in reference equations for diffusion capacity of the lung for carbon monoxide (DlCO) that account for only persons who are White and non-White, and obscuration of myriad social and biologic factors that influence lung function but also associate with race (6). Based on these considerations and several studies that used a single reference equation (7, 8), the American Thoracic Society updated its PFT result interpretation guidelines to recommend the use of race-neutral reference equations (6).

However, the impact of race-neutral PFT result interpretation on clinical practice and patient outcomes is not yet clear. Rigorous studies to elucidate the implications of these updated guidelines are critical, as illustrated in the example of a hypothetical patient of a given age, sex, and height being evaluated for surgical management of localized NSCLC. If this hypothetical patient were to be White, a raw forced expiratory volume in 1 second (FEV1) value in liters could be interpreted as an abnormal percentage of the predicted normal value in a reference population but normal if the patient were Black. The use of race-neutral reference equations would increase the FEV1% predicted normal value if the patient were White and decrease FEV1% predicted if the patient were Black. It has been unclear whether and how such changes in PFT result interpretation might affect clinical decision-making and patient outcomes.

In this issue of AnnalsATS, Sheshadri and colleagues (pp. 38–46) make an important step forward in assessing the impact of the updated PFT result interpretation guidelines (9). In this paper, the authors examine the association of postoperative pulmonary complications following surgical lung resection for NSCLC and preoperative PFT results interpreted using race-neutral versus race-specific Global Lung Initiative reference equations.

The design of the study has several strengths. First, the single-center study included a large number of consecutive patients undergoing lobectomy. Second, important covariates were considered as potential confounders of postoperative complications following lung resection (10), including pulmonary comorbidities such as cigarette smoking, chronic obstructive pulmonary disease, and obstructive sleep apnea, as well as detailed measurements of cardiac comorbidities. Third, the authors used a causal model with directed acyclic graphs to specify the relationship of pulmonary function with postoperative pulmonary complications. This approach helps to ensure adequate control for potential confounding and improves the transparency of the analysis by making the assumptions explicit.

The central finding of Sheshadri and colleagues is that race-neutral interpretation of preoperative PFT results is comparable to race-specific interpretation in the context of postoperative pulmonary complications. This finding argues against the potential for excess harm to non-Hispanic White persons via the false reassurance of adequate lung function based on race-neutral reference equations. Given that there is no Global Lung Initiative reference equation for Hispanic ethnicity (5), another interesting finding was that classifying Hispanic persons as non-White in race-specific FEV1% predicted calculations diminished the association of race with postoperative pulmonary complications, whereas the association was significant when Hispanic persons were classified as White. Similarly, the association of Black race with postoperative pulmonary complications was altered when using race-specific models. Overall, these findings suggest that the association of race with complications following lobectomy hinges on the subjective definition of race and that the inclusion of race in PFT result interpretation may cloud the association of important environmental and biologic variables that influence lung function (11) and postoperative complications.

This study also has important limitations that should rightfully moderate the pace at which the findings are incorporated into clinical practice. First, this study does not examine whether race-neutral or race-specific interpretation of preoperative PFT results alters the selection of patients for curative resection of localized NSCLC. Examining this question is key to ensure that race-neutral interpretation of preoperative PFT results does not inappropriately exclude Black persons from curative lung resection and unintentionally exacerbate the known disparities in surgical management of early-stage NSCLC (12–14). Second, the cohort of patients included in this study had near-normal lung function and generally good performance status, as would be expected in a group of patients undergoing lobectomy. However, results derived from this type of retrospective cohort may be affected by selection bias. This type of bias in studies of postoperative complications is particularly germane because patients with borderline lung function at the threshold of meeting criteria for surgical lung resection will ultimately have the highest risk of postoperative complications (15). Finally, the study did not evaluate whether race-neutral PFT result interpretation performed better than race-specific approaches in predicting postoperative complications. Future studies that include greater proportions of non-White participants from multiple centers could evaluate this question, as well as associations with longer-term surgical outcomes.

Although this study is an important step forward in understanding the clinical implications of race-neutral PFT result interpretation, several knowledge gaps remain. For example, further studies are needed to better understand precisely which biologic variables drive differences in lung function and how these variables influence postoperative pulmonary complications following lobectomy. Additionally, although the authors examined continuous associations between FEV1% predicted and postoperative complications following lobectomy, clinical decision-making and patient selection for curative resection of NSCLC are driven by absolute thresholds of FEV1, FEV1% predicted, and DlCO (1, 2). Thus, additional studies are needed to evaluate whether the current lung function thresholds for surgical resection are appropriate when considering race-neutral interpretation of PFT results and how best to approach DlCO thresholds given that race-neutral reference equations are not available (5).

It is important to balance the important findings of this study with its limitations when considering the clinical implications of the results. The demonstration that race-neutral and race-specific PFT result interpretation strategies yield similar associations with postoperative pulmonary complications following lobectomy for localized NSCLC supports recent guidelines from the American Thoracic Society on PFT result interpretation that encourage the use of race-neutral reference equations. However, premature incorporation of this study’s findings into clinical practice without sufficient replication and prospective analysis may lead to harm by reducing access to curative surgery for patients with localized NSCLC who previously would have been deemed eligible. Nonetheless, this study highlights the importance of delineating the precise social and biologic factors that influence lung function and the necessity of reexamining surgical risk in light of these factors.