Abstract
OBJECTIVE: To assess the patterns of angiotensin converting enzyme inhibitors and angiotensin receptor blockers (ACE-I/ARB) discontinuation in the setting of chronic kidney disease (CKD) progression in real-world clinical practice. PATIENTS AND METHODS: We identified incident ACE-I/ARB users with a baseline estimated glomerular filtration rate (eGFR) ≥15 mL/min/1.73 m(2) and without end-stage renal disease in the Geisinger Health System between January 1, 2004, and December 31, 2015. We investigated the associations of CKD stage, hospitalizations with and without acute kidney injury (AKI), serum potassium, bicarbonate level, thiazide, and loop diuretic use with ACE-I/ARB discontinuation. RESULTS: Among the 53,912 ACE-I/ARB users, the mean age was 59.9 years, and 50.6% were female. More than half of users discontinued ACE-I/ARB within 5 years of therapy initiation. The risk of ACE-I/ARB discontinuation increased with more advanced CKD stage. For example, patients who initiated ACE-I/ARB with CKD stage G4 (eGFR: 15-29 mL/min/1.73 m(2)) were 2.09-fold (95% CI, 1.87-2.34) more likely to discontinue therapy than those with eGFR ≥ 90 mL/min/1.73 m(2). Potassium level greater than 5.3 mEq/L, systolic blood pressure ≤ 90 mm Hg, bicarbonate level < 22 mmol/L, and intervening hospitalization-particularly AKI-related-were also strong risk factors for ACE-I/ARB discontinuation. Thiazide diuretic use was associated with lower risk, whereas loop diuretic use was associated with higher risk of discontinuation. CONCLUSION: In a real-world cohort, discontinuation of ACE-I/ARB was common, particularly in patients with lower eGFR. Hyperkalemia, hypotension, low bicarbonate level, and hospitalization (AKI-related, in particular) were associated with a higher risk of ACE-I/ARB discontinuation. Additional studies are needed to evaluate the risk-benefit balance of discontinuing ACE-I/ARB in the setting of CKD progression.