(P-613) The Risk of Severe Acute Kidney Injury with Use of Anti-hypertensives in the Elderly: A Self-controlled Case Series Analysis

Background: Age and anti-hypertensive (AH) use can predispose elderly patients to acute kidney injury (AKI).

Objectives: To assess the risk of severe AKI associated with the use of AH in three cohorts of elderly patients with complex health needs.

Methods: Patients older than 65 years on the 1 January 2010 (start date) in the clinical practice research datalink (CPRD) GOLD primary care dataset were included, with linked hospital episode statistics (HES) for the identification of severe AKI. Three cohorts were defined: 1) Hospital (≥ 1 unplanned hospitalisation in the last year); 2) Frailty (electronic frailty index ≥ 3); 3) Polypharmacy (≥ 10 medications in the last year). Patients were followed up until the earliest of: patient leaving practice, practice last collection date, end of HES linkage coverage, death, or date of data extraction. Patients were included in the self-controlled case series (SCCS) if they were AH-naive and had AKI during follow-up. Incidence rate ratios (IRR) and 95% confidence intervals (CI) were calculated for the risk (exposed to AH) compared to the control (non-exposed) period. Our main analysis was adjusted for age, with sensitivity analyses: 1) including a 30-day pre-risk period prior to initiation of AH; 2) assessing the risk for the first AKI only; and 3) stratifying patients by survival status during follow-up.

Results: A total of 274/25,210 participants suffered severe AKI in the hospital cohort during median follow-up of 4.2 years, 238/18,211 in the frailty cohort (median follow-up 3.7 years), and 162/14,955 in the polypharmacy cohort (median follow-up 4.2 years). The number of severe AKI events were 302 (hospital), 254 (frailty), and 173 (polypharmacy) respectively. The main SCCS model showed IRR 2.00 (95% CI: 1.29-3.08) for the hospital cohort, IRR 2.66 (95% CI: 1.59-4.47) for the frailty cohort, and IRR 2.91 (95% CI: 1.63-5.19) for the polypharmacy cohort. The sensitivity analyses were consistent with our main results. Risks were attenuated when only patients who survived all follow-up were included: IRR 1.02 (95% CI: 0.54-1.91) (hospital), 1.84 (95% CI: 0.79-4.30) (frail), 2.02 (95% CI: 0.86-4.73) (polypharmacy) although numbers of events were low.

Conclusions: Our study found a 2 to 3-fold excess risk of severe AKI with AH use in elderly patients. Future studies should investigate the risk of specific AH drug classes associated with severe AKI in this population.