Association Between Sarcoidosis and Lung Cancer: A Retrospective Analysis

Background: Sarcoidosis has been variably associated with malignancy, but its specific relationship with lung cancer remains contested. We evaluated this association using a nationally representative U.S. inpatient database.

Methods: Retrospective cross-sectional analysis of the 2018 National Inpatient Sample (7,105,498 discharges; 35,527,481 weighted). Sarcoidosis (D86.0, D86.2, D86.9) and lung cancer (C34.x) were identified across all 40 ICD-10-CM diagnosis fields. Survey-weighted logistic regression accounted for NIS stratification, clustering, and discharge weights. Adjusted models controlled for age, gender, race/ethnicity, income, admission type, and Elixhauser comorbidities. Design effects were reported. Sensitivity analyses varied the sarcoidosis code definition.

Results: Of 14,961 sarcoidosis and 82,121 lung cancer discharges, 131 (0.87%) carried both diagnoses. Sarcoidosis patients were older (61.1 vs. 49.9 years), more often female (62.2% vs. 56.1%), and disproportionately Black (42.7% vs. 14.6%). Sarcoidosis was associated with lower odds of lung cancer (adjusted OR 0.582, 95% CI 0.479–0.707, p < 0.001; design effect 6.09). Among lung cancer discharges, sarcoidosis was not significantly associated with in-hospital mortality (adjusted OR 0.693, 95% CI 0.323–1.489, p = 0.35). Sensitivity analyses using narrow and broad sarcoidosis definitions yielded consistent results.

Conclusions: Sarcoidosis was associated with significantly lower odds of concurrent lung cancer, consistent with large cohort studies reporting protective-direction findings. This inverse association likely reflects unmeasured confounding by smoking, which is inversely related to sarcoidosis yet is the dominant lung cancer risk factor. Cross-sectional administrative data cannot establish temporality, distinguish sarcoidosis from sarcoid-like reactions, or account for Berkson’s bias. Longitudinal studies with smoking ascertainment are needed to clarify this relationship.

Keywords: Sarcoidosis, lung cancer, National Inpatient Sample, administrative database, retrospective, complex survey design, in-hospital mortality