(268) Is Statin Use Associated with An Increased Risk of Adrenal Insufficiency? A Nested Matched Case-Control Study

Publication Number: 599

Background: The adrenal gland uses the cholesterol of low-density lipoproteins as a substrate to produce steroid hormones including mineralocorticoids, glucocorticoids and sexual hormones. By inhibiting cholesterol synthesis, statins could reduce the synthesis of steroid hormones, thereby causing adrenal insufficiency (AI).

Objectives: We aimed to study the potential statin-related risk of AI in a large nationwide database.

Methods: We conducted a nested case-time control study using a cohort of patients ≥18y and affiliated to the General Scheme of the French Health Insurance database in 2010 (SNDS), without AI history. Cases were patients with a first event of AI between 01/01/2015 and 31/12/2017. AI was defined as the presence of both a hospital stay with a diagnosis of AI (ICD-10 codes E271 to E274), and a first dispensing of hydrocortisone ±30 days before/after hospital admission. Each case was matched to up to 10 controls on age (± 2 years), sex, and prior treatment with corticosteroids in the 3 years before the index date. Exposure to statin was measured over the 5 years preceding the index date using different definitions: history of use, cumulative use and intensity of last statin used. A 6-month censoring lag-time for exposure was applied before the index date to avoid a potential detection bias. Association was estimated using a conditional logistic regression adjusted on confounders including a disease risk score. Analyses were stratified on age, sex and corticosteroid history of use. Sensitivity analyses explored the impact of lag-time length for exposure by repeating the main analysis with a 3-month and without lag-time.

Results: 4,492 cases of AI were compared to 44,798 controls (median age 66y, 58% women), of which 39% vs. 33% were exposed to statins, respectively. No association between statin use and AI was found when adjusting the model for confounders, either for new-users of statin (adjusted OR 1.12; 95% CI 0.81-1.55), current users (0.97; 0.89-1.06) or past users (0.98; 0.87-1.09). A slight, but non significant, increase in AI risk was observed for new users of statins (1.12; 0.81-1.55), and a risk reduction for patients whose last statin used was of low intensity (0.84; 0.73-0.98). In stratified analyses, there was no difference in AI risk, neither according to sex nor according to corticosteroid history of use. A non significant higher risk was found in statin users aged 51 to 65 (1.12, 0.97-1.29). Sensitivity analyses on lag-time provided consistent results.

Conclusions: We found no relevant association between statin exposure and AI risk. The risk, if any, seems to be very limited and does not compromise the benefit of statin treatment.