1196 — Blood Pressure Control Lowers Long-Term Mortality in Veterans Following Acute Kidney Injury: Outcomes and Impacts of Social Determinants of Health
Lead/Presenter: Benjamin Griffin,
COIN - Iowa City
All Authors: Griffin BR (Center for Access and Delivery Research and Evaluation, Iowa City VAIowa City VA), Vaughan-Sarrazin M (Center for Access and Delivery Research and Evaluation, Iowa City VA) Yamada M (Center for Access and Delivery Research and Evaluation, Iowa City VA) Swee M (Center for Access and Delivery Research and Evaluation, Iowa City VA) Perencevich E (Center for Access and Delivery Research and Evaluation, Iowa City VA) Reisinger HS (Center for Access and Delivery Research and Evaluation, Iowa City VA) Jalal DI (Center for Access and Delivery Research and Evaluation, Iowa City VA)
Acute kidney injury (AKI) complicates 20-25% of hospital admissions, and is associated with significant increases in long-term mortality. Systolic blood pressure (SBP) < 130 mmHg and use of Angiotensin-converting enzyme inhibitors (ACEI) or angiotensin receptor blockers (ARB) medications may be associated with decreased mortality post-AKI, but further data are needed. Furthermore, whether SBP control and ACEI/ARB use are impacted by social determinants of health (SDOH) is unknown. We hypothesized that SBP control and ACEI/ARB use would be associated with reduced mortality following AKI, and that non-white patients, those with lower socioeconomic status, and those in more rural settings will be less likely to achieve adequate blood pressure control or use ACEI/ARB medications.
We included all adult VA patients from 2013 to 2018 with AKI during a hospitalization and > = 1 blood pressure within 30 days of discharge. ACEI/ARB use was defined as a prescription within 90 days of hospital discharge. Baseline SBP was defined within 30 days of discharge and followed up to two years after discharge, and categorized as controlled ( < 130 mmHG) or uncontrolled (>130 mmHG) over time. The primary outcome was time to mortality within two years. We used Cox Proportional Hazards regression to adjust for demographics, comorbidities, and illness severity, while controlling for the impact of BP control over time. The SDOH evaluated included race/ethnicity, socioeconomic status, and rurality. Socioeconomic status was defined based on the Social Deprivation Index (SDI), and Rurality was defined based on the 4-category Rural-Urban Commuting Area (RUCA) codes. To evaluate the impact of each SDOH, we conducted univariate comparisons between blood pressure control and ACEI/ARB use and each SDOH, and included interaction terms for each SDOH and BP control in multivariable models.
A total of 97,376 patients met inclusion criteria, of which 25,600 (26%) died within 2 years of discharge. There were 52,352 (54%) Veterans with baseline SBP < 130 mmHg with a 2-year mortality of 27% compared to 24% in SBO >130 mmHg at baseline. Within 90 days, 41,147 (42%) were treated with ACEI/ARB, and these patients had 20.6% mortality within 2 years, compared to 30.5% in those not treated with ACEI/ARB Veterans (unadjusted OR 0.59 (95% CI 0.57-0.61)). Cumulative time with controlled blood pressure was clinically significant in the final Cox proportional hazard model (aHR 0.74 per year of controlled BP). There were no clinically significant differences in SDOH in Veterans with and without SBP < 130, nor in Veterans with and without ACEI/ARB use, and interaction terms in the Cox proportional hazard model were non-significant. Stratified analyses did not show significant differences among SDOH categories.
In a post-AKI cohort, average SBP < 130 and use of ACEI/ARB were significantly associated with decreased long-term mortality. The impact of BP control and ACEI/ARB use was consistent, regardless of race, rurality, or social deprivation.
AKI increases Veteran risk for long-term mortality. Adequate blood pressure control and use of ACEI/ARB medications are associated with reduced 2-year mortality, and further implementation research should focus on ways to safely increase utilization of these interventions.