Blood eosinophil counts and postoperative outcomes in early-stage lung cancer: a retrospective cohort study

BACKGROUND: Lung cancer resection is curative but associated with postoperative morbidity and mortality. This study evaluated whether elevated blood eosinophil count (BEC) was associated with postoperative outcomes in early-stage lung cancer. METHODS: This was a retrospective cohort study of consecutive adult patients undergoing lung resection for stage I and II non-small cell lung cancer in a large tertiary referral center from September 2017 to June 2021. Data were drawn from the institution's Data Warehouse. The primary outcome was 90-day healthcare utilization defined as emergency department visit or hospital readmission. Secondary outcomes were postoperative complications, index hospitalization length of stay, and 1-year survival. Preoperative 90-day BEC was categorized by a threshold of 200 cells/µL. Covariates were age, sex, smoking status, Charlson Comorbidity Index, chronic obstructive pulmonary disease (COPD), asthma, tumor size, nodal status, surgical approach, and blood results (white blood cells, hemoglobin, and creatinine). The main analyses were validated by a second international cohort. Log-Poisson with robust variance estimation and Cox proportional hazards regression models were fit for primary and secondary outcomes. Analyses were replicated for BEC thresholds of 150 and 300 cells/µL. RESULTS: Among 715 patients undergoing lung resection (median age = 69 years, 42% male, 29% with COPD), 146 patients (20%) had high preoperative BEC ≥ 200 cells/µL. BEC ≥ 200 cells/µL was associated with a higher rate of 90-day healthcare utilization:19% vs. 14% for BEC < 200 cells/µL. This association remained after adjustment (Risk Ratio [RR], 1.52; 95% Confidence Interval [CI], 1.02-2.25) and the validation cohort (RR, 2.23; 95% CI, 1.06-4.69). BEC as a continuous measure was also associated with the primary outcome in both cohorts: RR, 2.15 (95% CI, 1.49-3.12) and RR, 1.42 (95% CI, 1.10-1.94), respectively. BEC ≥ 200 cells/µL was associated with higher probability of death 1 year post-surgery (adjusted Hazard Ratio, 2.41; 95% CI, 1.08-5.35). There was no difference in the risk of postoperative pulmonary complications between high and low BEC (RR, 0.86; 95% CI, 0.58-1.27). CONCLUSIONS: Elevated preoperative BEC was associated with higher risk of postoperative healthcare utilization and lower 1-year survival after lung cancer resection among patients with or without respiratory disease.