Arterial Stiffness Is Independently Associated with Acute Kidney Injury in SPRINT

AKI is defined by an abrupt decrease in kidney function and is associated with significant morbidity and mortality, including kidney disease progression and higher cardiovascular disease risk (¹). Greater arterial stiffness is also associated with higher risk for kidney function decline and cardiovascular disease (²,³). The gold standard for measuring arterial stiffness is carotid-femoral pulse wave velocity (PWV). Whether greater carotid-femoral PWV predicts incident AKI is unknown.

We examined the association of carotid-femoral PWV and incident AKI among individuals at high cardiovascular risk who participated in the Systolic Blood Pressure Intervention Trial (SPRINT). SPRINT was a randomized trial comparing intensive systolic BP treatment (<120 mm Hg) with standard treatment (<140 mm Hg) on cardiovascular outcomes in 9361 nondiabetic hypertensive participants (⁴). A subset of 613 individuals without atrial fibrillation recruited from 11 of the 102 clinical sites participated in a carotid-femoral PWV ancillary study and were included in this analysis (⁵). Participants recruited to the ancillary study were slightly older and more likely to be White; they had lower prevalence of cardiovascular disease but higher prevalence of CKD compared with SPRINT participants not included. Detailed methods are described elsewhere (⁴,⁵). In SPRINT, AKI was on the basis of the modified Kidney Disease Improving Global Outcomes criteria incorporating only serum creatinine concentration and included if the AKI event was recorded as an adjudicated serious adverse event (SAE). We used multivariable-adjusted Cox proportional hazards models to examine the association between baseline carotid-femoral PWV and incident AKI. Carotid-femoral PWV was examined as a categorical variable, and participants with carotid-femoral PWV greater than or equal to median (10.4 m/s) were compared with those with carotid-femoral PWV less than median. Carotid-femoral PWV was also examined as a continuous variable: risk of AKI per 1-m/s higher. All statistical analyses were performed with SAS software version 9.4 (SAS Institute, Cary, NC). P values of 0.05 were considered statistically significant.

Baseline characteristics were similar between the two groups (carotid-femoral PWV greater than or equal to median and less than median). The mean ± SD age was 72±9 years, 40% were women, 69% were White, and mean eGFR was 67±20 ml/min per 1.73 m². Those with baseline carotid-femoral PWV greater than or equal to median were older and had higher systolic BP than those with carotid-femoral PWV less than median. Over a median of 453 (interquartile range, 289–724) days, there was a total of 20 adjudicated SAE AKI events (18 in the group with carotid-femoral PWV greater than or equal to median and two in the group with carotid-femoral PWV less than median). Carotid-femoral PWV was strongly associated with AKI among individuals at high risk of cardiovascular disease who participated in SPRINT (Table 1). This strong association remained when carotid-femoral PWV was analyzed as a continuous variable after adjusting for demographics, including age, and clinical risk factors, including systolic BP, preexisting cardiovascular disease, CKD, and heart rate (Table 1).

Table 1. - Association of arterial stiffness as measured by carotid-femoral pulse wave velocity with AKI

Models	Hazard Ratio (95% Confidence Interval)
	Per 1-m/s Higher Carotid-Femoral Pulse Wave Velocity	Carotid-Femoral Pulse Wave Velocity ≥ Median a versus Carotid-Femoral Pulse Wave Velocity < Median a
Unadjusted	1.22 (1.07 to 1.39)	9.28 (2.15 to 40.00)
Model 1: Age, sex, race, randomization assignment	1.26 (1.10 to 1.45)	10.67 (2.45 to 46.55)
Model 2: Model 1 + smoking category, history of cardiovascular disease, no. of antihypertensives	1.27 (1.10 to 1.45)	11.33 (2.59 to 49.66)
Model 3: Model 2 + eGFR, UACR, systolic BP, HR	1.32 (1.13 to 1.54)	13.28 (2.91 to 60.58)

UACR, urine albumin-creatinine ratio; HR, heart rate.

^aCarotid-femoral pulse wave velocity median =10.4 m/s.

Arterial stiffness is associated with decline of kidney function in CKD (²). This may be due to the important role arteries play in mitigating the cyclical pressure changes generated during the cardiac cycle. With increasing arterial stiffness, arterial elasticity and compliance are reduced, leading to transmission of high-pressured forces to end organ microvasculature. These vascular beds are not designed to handle high pulsatility, resulting in damage. The kidneys may be particularly affected given the high volume of flow and the low resistance in the renal afferent arterioles. This pathophysiology may also explain our findings that greater arterial stiffness is strongly associated with incident AKI independent of eGFR and urine albumin-creatinine ratio. Additionally, the observed independent association between arterial stiffness and AKI could contribute to the higher cardiovascular risk observed after AKI. However, further investigation into the physiologic mechanisms underlying these associations is required.

Strengths of our study include use of a large multicenter trial, which allowed us to account for many important confounders. Additionally, AKI was a prespecified and adjudicated SAE. Our study was limited to SPRINT participants, who were older, hypertensive, nondiabetic adults at high risk for cardiovascular disease, thus limiting the generalizability. The confidence intervals in some analyses were wide, suggesting that the small number of AKI events may have limited power.

In conclusion, greater large elastic artery stiffness is independently associated with incident AKI in nondiabetic older adults at high risk for cardiovascular events. Our results support consideration of including arterial stiffness into risk prediction models for AKI and as a target for prevention or therapeutic treatment of AKI. However, further inquiry in larger and more diverse populations, including in healthy individuals and those with diabetes, is needed to validate these hypothesis-generating findings.

Disclosures

M.B. Chonchol reports consultancy agreements with Amgen, Corvidia, Otsuka, Reata, Tricidia, and Vifor; receiving research funding from Corvidia, the National Institutes of Health, OTSUKA, Reata, and Sanofi; receiving honoraria from Amgen, Corvidia, Reata, Tricidia, and Vifor; and serving as a Deputy Editor of CJASN. A.J. Jovanovich reports receiving research funding as a site investigator for AstraZeneca and received the study drug free of charge from Shire. A.J. Jovanovich reports serving as an American Heart Association Council for Kidney in Cardiovascular Disease Leadership Committee member and the Chair of the Early Investigators Committee, on the editorial board of CJASN, and as a section editor for Clinical Nephrology. K.L. Nowak reports receiving research funding from Corvidia Therapeutics, Otsuka Pharmaceutical Development and Commercialization (data analysis), and Verdure Sciences. K.L. Nowak also interacts with the PKD Foundation. M.A. Supiano reports serving on the American Geriatrics Society Board of Directors and Journal of the American Geriatrics Society editorial board. All remaining authors have nothing to disclose.

Funding

SPRINT was supported by A-HL contracts HHSN268200900040C, HHSN268200900046C, HSN268200900047C, HHSN268200900048C, and HHSN268200900049C and National Institutes of Health (NIH) interagency agreement 13-002-001, including the National Heart, Lung, and Blood Institute, the National Institute on Aging, the National Institute of Diabetes and Digestive and Kidney Diseases, and the National Institute of Neurological Disorders and Stroke. Several study sites were supported by Clinical and Translational Science Awards funded by the National Center for Advancing Translational Sciences of the NIH: Boston University grant UL1RR025771; Case Western Reserve University grant UL1TR000439; George Washington University grant UL1TR000075; Ohio State University grant UL1RR025755; Stanford University grant UL1TR000093; Tulane University grant P30GM103337 (COBRE Award NIGMS); Tufts University grants UL1RR025752, UL1TR000073, and UL1TR001064; University of California, Davis grant UL1TR000002; University of Florida grant UL1TR000064; University of Illinois grant UL1TR000050; University of Michigan grant UL1TR000433; University of Pennsylvania grants UL1RR024134 and UL1TR000003; University of Pittsburgh grant UL1TR000005; University of Texas Southwestern grant 9U54TR000017-06; University of Utah grant UL1TR000105-05; and Vanderbilt University grant UL1TR000445. The trial was also supported in part with respect to resources and the use of facilities by the Department of Veterans Affairs. The CFPWV ancillary study grant R01HL107241 (to M.A. Supiano) was supported by the National Heart, Lung, and Blood Institute. A.J. Jovanovich is supported by CFPWV Veterans Affairs Merit Award I01CX001985.