Abstract
Rationale and Objective:
In the Systolic Blood Pressure Intervention Trial, the possible relationships between acute kidney injury (AKI) and risk of major cardiovascular events and death are not known.
Study Design:
Post hoc analysis of a multicenter, randomized, controlled, open-label clinical trial.
Setting and Participants:
Hypertensive adults without diabetes who were ≥50 years of age with prior cardiovascular disease, chronic kidney disease (CKD), 10-year Framingham risk score >15%, or age >75 years were assigned to a systolic blood pressure target of <120 mm Hg (intensive) or <140 mm Hg (standard).
Predictor:
AKI episodes.
Outcomes:
The primary outcome was a composite of myocardial infarction, acute coronary syndrome, stroke, decompensated heart failure, or cardiovascular death. The secondary outcome was death from any cause.
Analytical Approach:
AKI was defined using the Kidney Disease: Improving Global Outcomes modified criteria based solely upon serum creatinine. AKI episodes were identified by serious adverse events or emergency room visits. Cox proportional hazards models assessed the risk for the primary and secondary outcomes by AKI status.
Results:
Participants were 68 ± 9 years of age, 36% women (3,332/9,361), and 30% Black race (2,802/9,361), and 17% (1,562/9,361) with cardiovascular disease. Systolic blood pressure was 140 ± 16 mm Hg at study entry. AKI occurred in 4.4% (204/4,678) and 2.6% (120/4,683) in the intensive and standard treatment groups respectively (p < 0.001). Those who experienced AKI had higher risk of cardiovascular events (hazard ratio [HR] 1.52, 95% CI 1.05–2.20, p = 0.026) and death from any cause (HR 2.33, 95% CI 1.56–3.48, p < 0.001) controlling for age, sex, race, baseline systolic blood pressure, body mass index, number of antihypertensive medications, cardiovascular disease and CKD status, hypotensive episodes, and treatment assignment.
Limitations:
The study was not prospectively designed to determine relationships between AKI, cardiovascular events, and death.
Conclusions:
Among older adults with hypertension at high cardiovascular risk, intensive treatment of blood pressure independently increased risk of AKI, which substantially raised risks of major cardiovascular events and death.
Keywords: Hypertension, Blood pressure, Anti-hypertensive treatment, Kidney disease, Survival
Introduction
Hypertension is a major public health problem greatly contributing to disability and deaths. Blood pressure is a modifiable risk factor for both cardiovascular disease and chronic kidney disease (CKD) [1, 4, 10, 13]. According to previously reported observational data, for each 20 mm Hg increment in systolic blood pressure or 10 mm Hg increment in diastolic blood pressure above 120 mm Hg, the risk of cardiovascular disease approximately doubles [12]. Antihypertensive therapy reduces risks of heart failure, stroke, myocardial infarction, and CKD progression [1, 6, 7]. While the importance of hypertension control is clear, the optimal blood pressure target has been unclear and potential benefits versus risks of different targets are not well understood.
The Systolic Blood Pressure Intervention Trial (SPRINT) was designed to determine the effect of intensive treatment to a systolic blood pressure <120 mm Hg compared to standard treatment to systolic blood pressure <140 mm Hg [6]. SPRINT was stopped nearly 2 years early due to benefits of 25 and 27% relative risk reductions for major cardiovascular events and deaths from any cause, respectively. Furthermore, there were no differences between the treatment groups in prespecified renal outcomes, including progression to end-stage renal disease. However, kidney-associated adverse events, particularly acute kidney injury (AKI), occurred more often with intensive treatment. Rates of AKI by investigator report were 4.1% in the intensively-treated group versus 2.5% in the standard treatment group over the 3.26-year median duration of follow-up. In a subsequent analysis, rates of adjudicated AKI were similarly higher in the intensive treatment group compared to the standard treatment group [17].
Relationships between AKI and risk of major cardiovascular events and death have not been determined in SPRINT. Therefore, the objective of this study was to delineate these relationships.
Methods
Design and Participants
SPRINT was a randomized, controlled, open-label clinical trial conducted at 102 sites in the United States. Participants were enrolled from November 2010 to March 2013 [6]. Main inclusion criteria included: ≥50 years of age, systolic blood pressure ≥130 mm Hg and <180 mm Hg, high cardiovascular risk defined by clinical or subclinical cardiovascular disease other than stroke, CKD with an estimated glomerular filtration rate (eGFR) 20–60 mL/min/1.73 m2, 10-year Framingham risk score >15%, or age >75 years. Clinical cardiovascular disease was defined as history of one or more of MI, ACS, coronary revascularization, carotid revascularization, PAD with revascularization, >50% stenosis of coronary/carotid/lower extremity artery; or AAA ≥5 mm. Subclinical cardiovascular disease was defined as a history of one or more of coronary artery calcium score ≥400, ABI ≤0.90, or LVH. Main exclusion criteria included prior stroke, urine protein excretion >1 g/day, diabetes mellitus, symptomatic heart failure, cardiac ejection fraction <35%, and active alcohol or substance use disorder. For the present analysis, SPRINT participants who experienced AKI after the primary outcome were excluded from analyses (n = 45). SPRINT conformed to the Declaration of Helsinki and was approved by institutional review boards at each of the study sites. All participants provided written informed consent to join the study.
Participants were assigned to a systolic blood-pressure target of either <120 mm Hg (intensive treatment) or <140 mm Hg (standard treatment). Randomization was stratified according to clinical site. Participants and study personnel were aware of study group assignments. Investigators adjudicating outcomes were unaware of study group assignment [6].
Measurements
Clinical, laboratory, and demographic data were collected at baseline. In addition, clinical and laboratory data were obtained every 3 months after the baseline assessment. To obtain self-reported cardiovascular event outcomes, structured interviews were performed every 3 months [6]. Electrocardiograms were also obtained to assess for cardiovascular events. AKI was assessed at all hospitalizations and defined using Kidney Disease: Improving Global Outcomes modified criteria incorporating solely serum creatinine. Urine output was not consistently recorded in SPRINT. AKI episodes were identified as serious adverse events and emergency department visits. The present dataset, provided for the SPRINT data analysis challenge sponsored by the New England Journal of Medicine, only included investigator-reported, unadjudicated AKI. Hypotensive episodes were also only investigator reported.
Outcomes
The primary composite outcome included myocardial infarction, acute coronary syndrome not resulting in myocardial infarction, stroke, acute decompensated heart failure, or death from cardiovascular causes [6]. The secondary outcomes were the individual components of the primary composite outcome, death from any cause, and the composite of the primary outcome or death from any cause.
Statistical Analysis
Data were accessed via the Biological Specimen and Data Repository Information Coordinating Center for the SPRINT data analysis challenge. Baseline characteristics of the study participants were reported as mean ± SD for normally distributed variables, median and interquartile range for non-normally distributed variables, and frequencies and percentages for categorical variables. Parametric and nonparametric analysis of variance models were tested for group comparisons at baseline for normally and non-normally distributed continuous data respectively. Categorical variables were assessed with the use of chi-square tests.
Effects of intensive versus standard treatment for blood pressure on AKI were analyzed first. The next analyses examined relationships of AKI with the primary SPRINT outcome and death from any cause. For both sets of analyses, Kaplan-Meier survival curves were constructed for the primary and secondary outcomes with log-rank tests to assess differences in survival. Unadjusted and adjusted Cox proportional hazards models were fitted and with inclusion of baseline clinical risk factors as co-variates (age, sex, race, systolic blood pressure, body mass index (BMI), number of antihypertensive medications, clinical cardiovascular disease and CKD status). For associations of AKI with the primary outcome and death from any cause, models were adjusted for study treatment assignment. The adjusted models were further adjusted for hypotensive episodes. Internal validation was conducted by split-file analysis. Mediation analyses were also utilized to evaluate the proportion of the total effect on the primary outcome and death from any cause associated with AKI. Accelerated failure time models with Weibull distributions for time to the outcomes were used and logistic regression was utilized for AKI as the mediator [21]. Models used for mediation analysis were also estimated and included baseline clinical risk factors as co-variates (age, sex, race, systolic blood pressure, BMI, number of antihypertensive medications, clinical cardiovascular disease and CKD status). An alpha level of <0.05 was used as the threshold for statistical significance. R statistical computing software (version 3.1.2) was used for all analyses.
Results
Clinical Characteristics
Participants were 68 ± 9 years of age, 36% women (3,320/9,316), and 30% Black race (2,786/9,316) with a systolic blood pressure of 140 ± 16 mm Hg at study entry. Baseline clinical characteristics of participants in the intensive and standard treatment groups were well-matched overall [6]. Clinical cardiovascular disease and CKD were present in 16.7% (1,548/9,361) and 28% (2,615/9,361), respectively, at baseline. AKI occurred in 3.9% (180/4,654) and 2.1% (99/4,662) in the intensive and standard treatment groups respectively (p < 0.001). AKI represented 7.9% of all serious adverse events reported (274/3,485). Stratification of baseline characteristics by presence or absence of an AKI episode showed that participants who experienced AKI were older, had higher rates of clinical cardiovascular disease, and CKD, lower eGFR, and were more often taking antihypertensive medications at baseline (Table 1).
Table 1.
Baseline characteristics of study participants
| Characteristic | Intensive treatment |
Standard treatment |
||||
|---|---|---|---|---|---|---|
| AKI (n = 180) | number AKI (n = 4,474) | p value | AKI (n = 99) | number AKI (n = 4,563) | p value | |
| Criterion for increased cardiovascular risk, n (%) | ||||||
| Age >75 years | 61 (33.9) | 1,099 (24.6) | 0.006 | 41 (41.4) | 1,130 (24.8) | <0.001 |
| CKD | 102 (56.7) | 1,213 (27.1) | <0.001 | 35 (35.4) | 1,236 (27.1) | <0.001 |
| Cardiovascular disease | ||||||
| Clinical | 45 (25.0) | 726 (16.2) | 0.002 | 30 (30.3) | 747 (16.4) | <0.001 |
| Subclinical | 11 (6.1) | 234 (5.2) | 0.73 | 5 (5.1) | 241 (5.3) | 0.99 |
| Women, n (%) | 48 (26.7) | 1,630 (56.5) | 0.009 | 25 (25.3) | 1,617 (35.4) | 0.046 |
| Age, years | 70.5±10.6 | 67.8±9.3 | <0.001 | 70.9±10.2 | 67.9±9.4 | 0.003 |
| Statin use, n (%)† | 82 (45.6) | 1,885 (42.4) | 0.45 | 50 (50.5) | 2,015 (44.6) | 0.28 |
| Race/ethnicity, n (%) | ||||||
| Hispanic | 9 (5.0) | 492 (11.0) | 0.002 | 6 (6.1) | 474 (10.4) | 0.29 |
| Non-hispanic black | 72 (40.0) | 1,299 (29.0) | 34 (34.3) | 1,381 (30.3) | ||
| Non-hispanic white | 98 (54.4) | 2,586 (57.8) | 59 (59.6) | 2,631 (57.7) | ||
| Other | 1 (0.6) | 97 (2.2) | 0 (0.0) | 77 (1.7) | ||
| BMI, kg/m2‡ | 29.8±5.9 | 29.9±5.8 | 0.72 | 29.7±6.4 | 29.8±5.7 | 0.83 |
| Baseline blood pressure, mm Hg | ||||||
| Systolic | 141.0 (16.1) | 139.6 (15.7) | 0.19 | 141.7 (16.2) | 139.6 (15.4) | 0.24 |
| Diastolic | 76.0 (12.4) | 78.3 (11.8) | 0.015 | 76.7 (13.4) | 78.0 (12.0) | 0.29 |
| Estimated GFR, mL/min/1.73 m2 | 58.7±23.8 | 72.4±20.3 | <0.001 | 55.9±20.3 | 72.4±20.4 | <0.001 |
| Smoking status, n (%) | ||||||
| Never smoked | 70 (38.9) | 1,974 (44.1) | 0.37 | 25 (25.2) | 2,039 (44.7) | <0.001 |
| Former smoked | 83 (46.1) | 1,884 (42.1) | 58 (58.6) | 1,928 (44.3) | ||
| Current smoker | 27 (15) | 604 (13.5) | 16 (16.2) | 582 (12.8) | ||
| Missing data | 0 (0.0) | 12 (0.3) | 0 (0.0) | 14 (0.3) | ||
| Number of antihypertensive medications, n (%) | ||||||
| 0 | 11 (6.1) | 419 (9.4) | 0.023 | 5 (2.5) | 445 (9.6) | <0.001 |
| 1 | 45 (25.0) | 1,319 (29.5) | 18 (18.2) | 1,367 (30.0) | ||
| 2 | 59 (32.8) | 1,596 (35.7) | 33 (33.3) | 1,590 (34.8) | ||
| 3 | 50 (27.8) | 897 (19.9) | 35 (35.4) | 916 (20.1) | ||
| 4+ | 15 (8.3) | 243 (5.4) | 8 (8.1) | 245 (5.4) | ||
| Number of antihypertensive medications | 2.1±1.1 | 1.8±1.0 | 0.002 | 2.23±1.0 | 1.8±1.0 | <0.001 |
Plus-minus values are means ± SD.
Due to some missing data, the total sample size available for the intensive and standard treatment arms were 4,621 and 4,619 for the statin data respectively.
Due to some missing data, the total sample size available for the intensive and standard treatment arms were 4,622 and 4,617 for the BMI data respectively.
Clinical cardiovascular disease was defined as history of one or more of MI, ACS, coronary revascularization, carotid revascularization, PAD with revascularization, >50% stenosis of coronary/carotid/lower extremity artery; or AAA ≥5 mm.
Subclinical cardiovascular disease was defined as history of one or more of coronary artery calcium score ≥400, ABI ≤0.90, or LVH.
AKI, acute kidney injury; CKD, chronic kidney disease; GFR, glomerular filtration rate; BMI, Body mass index.
Risks of AKI and Kidney Outcomes by Treatment Groups
Participants receiving intensive treatment had increased risk of AKI compared to standard treatment in the unadjusted model (hazard ratio [HR] 1.83; 95% CI 1.43–2.33, p < 0.001; Table 2). Hypotensive episodes occurred in 3.3% (154/4,654) and 2.0% (91/4,662) of participants in the intensive and standard treatment groups respectively. Unadjusted risk of hypotensive episodes was increased in the intensive treatment group compared to that of the standard treatment group (HR 1.7; 95% CI 1.32–2.21, p < 0.001). Risk of AKI was not attenuated with control for covariates consisting of clinical risk factors (age, sex, race, systolic blood pressure, BMI, number of antihypertensive medications, and clinical cardiovascular disease and CKD status), treatment assignment, and hypotensive episodes (HR 1.80; 95% CI 1.40–2.33, p < 0.001).
Table 2.
Associations of intensive blood pressure treatment with AKI
| Variable | HR | Lower 95% CI | Higher 95% CI | p value |
|---|---|---|---|---|
| Model 1 | ||||
| Intensive therapy | 1.83 | 1.43 | 2.33 | <0.001 |
| Model 2 | ||||
| Intensive therapy | 1.90 | 1.49 | 2.44 | <0.001 |
| Age, years | 1.02 | 1.01 | 1.04 | 0.007 |
| Gender, women | 0.52 | 0.39 | 0.70 | <0.001 |
| Race/ethnicity | ||||
| Hispanic | 0.67 | 0.37 | 1.24 | 0.21 |
| Other | 0.13 | 0.02 | 0.96 | 0.46 |
| Non-hispanic white | 0.50 | 0.37 | 0.68 | <0.001 |
| BMI, kg/m2 | 1.00 | 0.98 | 1.02 | 0.85 |
| Systolic blood pressure (10 mm Hg) | 1.14 | 1.06 | 1.23 | <0.001 |
| CKD status | 3.32 | 2.54 | 4.32 | <0.001 |
| Cardiovascular disease at baseline | 1.54 | 1.16 | 2.05 | 0.003 |
| Number of antihypertensive medications | 1.15 | 1.02 | 1.3 | 0.025 |
| Model 3 | ||||
| Intensive therapy | 1.80 | 1.40 | 2.33 | <0.001 |
| Age, years | 1.02 | 1.00 | 1.03 | 0.022 |
| Gender, women | 0.53 | 0.38 | 0.72 | <0.001 |
| Race/ethnicity | ||||
| Hispanic | 0.79 | 0.43 | 1.47 | 0.46 |
| Other | 0.16 | 0.02 | 1.16 | 0.07 |
| Non-hispanic white | 0.46 | 0.34 | 0.63 | <0.001 |
| BMI, kg/m2 | 1.01 | 0.98 | 1.03 | 0.63 |
| Systolic blood pressure (10 mm Hg) | 1.13 | 1.05 | 1.23 | 0.001 |
| CKD status | 3.10 | 2.36 | 4.06 | <0.001 |
| Cardiovascular disease at baseline | 1.47 | 1.1 | 1.96 | 0.008 |
| Number of antihypertensive medications | 1.13 | 0.99 | 1.28 | 0.054 |
| Hypotensive event | 13.04 | 9.56 | 17.77 | <0.001 |
CKD, chronic kidney disease; BMI, body mass index; AKI, acute kidney injury; HR, hazard ratio.
Decline in kidney function as measured by eGFR was greater in the intensive arm compared to the standard treatment arm. Of the 6,663 participants with available data, a >30% reduction in eGFR occurred in 4.0% (127/3,323) and 1.1% (37/3,340) participants in the intensive and standard treatment groups respectively. Participants receiving intensive anti-hypertensive treatment had increased risk of a >30% reduction in eGFR compared to standard treatment in the unadjusted model (HR 3.59; 95% CI 2.48–5.20, p < 0.001). This risk persisted in the model fully adjusted for clinical risk factors (age, sex, race, systolic blood pressure, BMI, number of antihypertensive medications, and clinical cardiovascular disease and CKD status; HR 3.69; 95% CI 2.54–5.36, p < 0.001). When hypotensive episodes were added to the adjusted model, the higher risk of intensive therapy for experiencing a 30% decline in eGFR remained higher (HR 3.70; 95% CI 2.54–5.37, p < 0.001).
Risks of the Primary Outcome and Death Associated with AKI
Study participants who experienced AKI had higher risk of the primary cardiovascular outcome (myocardial infarction, acute coronary syndrome, stroke, acute decompensated heart failure, or death from cardiovascular causes) and death from any cause (Fig. 1a, b). The primary cardiovascular outcome occurred in 7.3% (39/531) of participants who experienced AKI versus 2.7% (240/8,545) of those without AKI (p < 0.001). Death from any cause occurred in 9.8% (34/347) of participants with AKI versus 2.7% (245/8,969) in those without AKI (p < 0.001). Participants who experienced AKI had a significantly higher risk of the primary cardiovascular outcome (HR 2.38; 95% CI 1.7–3.32, p < 0.001) and death from any cause (HR 3.43; 95% CI 2.38–4.94, p < 0.001). In fully adjusted models (age, sex, race, systolic blood pressure, BMI, number of antihypertensive medications, and clinical cardiovascular disease and CKD status), including baseline clinical cardiovascular disease and CKD status, treatment assignment, and hypotensive episodes, AKI significantly increased risks for the primary outcome (HR 1.52; 95% CI 1.05–2.22, p = 0.026) and for death from any cause (HR 2.33; 95% CI 1.56–3.48, p < 0. 001), although the magnitude of risk was attenuated (Table 3). Risk estimates were similar when the models were adjusted for eGFR as a continuous variable rather than CKD status (online suppl. Table 1, see www.karger.com/doi/10.1159/000499574). There was no significant interaction between AKI events and treatment assignment on the primary outcome (HR 0.87; 95% CI 0.68–1.12, p = 0.28) or death from any cause (HR 1.09; 95% CI 0.55–2.15, p = 0.80). When death from any cause was examined as a competing risk for the primary outcome, AKI remained significantly associated with an increased risk of the primary outcome (HR 2.50; 95% CI 1.82–3.45, p < 0.001)
Fig. 1.
Survival without the primary outcome (a) or death (b) in participants with and without AKI. Kaplan-Meier curves for the composite primary outcome (myocardial infarction, acute coronary syndrome, stroke, heart failure, or death from cardiovascular causes) and for death from any cause. AKI, acute kidney injury.
Table 3.
Associations of AKI with the primary outcome and death from any cause
| Variable | Primary outcome |
Death from any cause |
||||||
|---|---|---|---|---|---|---|---|---|
| HR | lower 95% CI | higher 95% CI | p value | HR | lower 95% CI | higher 95% CI | p value | |
| Model 1 | ||||||||
| AKI | 2.38 | 1.70 | 3.32 | <0.0001 | 3.43 | 2.38 | 4.94 | <0.0001 |
| Model 2 | ||||||||
| AKI | 1.73 | 1.23 | 2.43 | 0.002 | 2.31 | 1.58 | 3.39 | <0.0001 |
| Age, years | 1.04 | 1.03 | 1.05 | <0.0001 | 1.06 | 1.05 | 1.08 | <0.0001 |
| Gender, women | 0.77 | 0.62 | 0.95 | 0.013 | 0.61 | 0.47 | 0.79 | 0.0002 |
| Race/ethnicity | ||||||||
| Hispanic | 0.86 | 0.52 | 1.43 | 0.56 | 0.87 | 0.5 | 1.51 | 0.62 |
| Other | 1.00 | 0.48 | 2.1 | 0.98 | 0.51 | 0.16 | 1.65 | 0.26 |
| Non-hispanic white | 1.11 | 0.87 | 1.43 | 0.39 | 0.84 | 0.63 | 1.14 | 0.27 |
| BMI, kg/m2 | 1.01 | 0.99 | 1.03 | 0.27 | 0.99 | 0.97 | 1.01 | 0.24 |
| Systolic blood pressure (10 mm Hg) | 1.00 | 1.00 | 1.01 | 0.09 | 1.01 | 0.99 | 1.01 | 0.10 |
| CKD status | 1.24 | 1.03 | 1.51 | 0.026 | 1.31 | 1.04 | 1.67 | 0.024 |
| Cardiovascular disease at baseline | 2.19 | 1.80 | 2.66 | <0.0001 | 1.62 | 1.26 | 2.08 | 0.0002 |
| Number of antihypertensive medications | 1.13 | 1.03 | 1.23 | 0.01 | 1.04 | 0.93 | 1.17 | |
| Model 3 | ||||||||
| AKI | 1.52 | 1.05 | 2.2 | 0.026 | 2.33 | 1.56 | 3.48 | <0.0001 |
| Age, years | 1.04 | 1.03 | 1.05 | <0.0001 | 1.06 | 1.05 | 1.08 | <0.0001 |
| Gender, women | 0.78 | 0.63 | 0.96 | 0.019 | 0.61 | 0.47 | 0.79 | 0.0002 |
| Race/ethnicity | ||||||||
| Hispanic | 0.87 | 0.52 | 1.44 | 0.59 | 0.87 | 0.5 | 1.51 | 0.62 |
| Other | 1.02 | 0.49 | 2.12 | 0.96 | 0.51 | 0.16 | 1.65 | 0.26 |
| Non-hispanic white | 1.1 | 0.86 | 1.42 | 1.42 | 0.84 | 0.63 | 1.14 | 0.27 |
| BMI, kg/m2 | 1.01 | 0.99 | 1.03 | 0.23 | 0.98 | 0.97 | 1.01 | 0.24 |
| Systolic blood pressure (10 mm Hg) | 1.01 | 1.00 | 1.01 | 0.08 | 1.01 | 0.99 | 1.01 | 0.10 |
| CKD status | 1.24 | 1.03 | 1.51 | 0.026 | 1.32 | 1.04 | 1.67 | 0.024 |
| Cardiovascular disease at baseline | 2.18 | 1.79 | 2.65 | <0.0001 | 1.62 | 1.26 | 2.08 | 0.0002 |
| Number of antihypertensive medications | 1.13 | 1.03 | 1.23 | 0.01 | 1.04 | 0.93 | 1.16 | 0.46 |
| Hypotensive episode | 1.61 | 1.03 | 2.53 | 0.037 | 0.96 | 0.52 | 1.78 | 0.89 |
Goodness of fit measures for primary outcome.
Model 1: AIC = 4,763; likelihood ratio test = 32, p < 0.0001.
Model 2: AIC = 4,587; likelihood ratio test = 211, p < 0.0001.
Model 3: AIC = 4,585; likelihood ratio test = 215, p < 0.0001.
Goodness of fit measures for all cause mortality.
Model 1: AIC = 3,035; likelihood ratio test = 42, p < 0.0001.
Model 2: AIC = 2,888; likelihood ratio test = 185 p < 0.0001.
Model 3: AIC = 2,890; likelihood ratio test = 185, p < 0.0001.
AKI, acute kidney injury; BMI, body mass index; CKD, chronic kidney disease.
Internal validation using split-file analyses demonstrated high congruency between the derivation (HR 1.77, 95% CI 1.10–2.85, p = 0.019) and validation (HR 2.93, 95% CI 1.95–4.41, p < 0.001) datasets for the effect of AKI on the primary cardiovascular outcome in fully adjusted models (age, sex, race, systolic blood pressure, BMI, number of antihypertensive medications, and clinical cardiovascular disease and CKD status). Internal validation using split-file analyses also demonstrated high congruency between the derivation (HR 3.46, 95% CI 2.24–5.35, p < 0.001) and validation (HR 2.71, 95% CI 1.66–4.42, p < 0.001) datasets for the effect of AKI on the death from any cause in the fully adjusted model. Mediation analysis demonstrated the proportion of the primary outcome mediated by AKI was 86% in the intensive treatment group and 62% in the standard treatment group (p < 0.001; Fig. 2a). The proportion of the death from any cause outcome mediated by AKI was 89% in the intensive treatment group and 69% in the standard treatment group (p < 0.001; Fig. 2b).
Fig. 2.
Mediation analyses of the association between intensive versus standard blood pressure treatment and the primary outcome and death from any cause with AKI as the mediator. Pathway (a) represents the pathway mediated by AKI and pathway (b) represents the pathway not mediated by AKI for the composite primary outcome (myocardial infarction, acute coronary syndrome, stroke, heart failure, or death from cardiovascular causes; panel a) and for death from any cause (panel b). AKI, acute kidney injury.
Discussion
Intensive blood pressure treatment increased the risk of AKI in older adults with hypertension in SPRINT. The present analyses indicate that AKI risk was independent of other clinical risk factors including CKD, cardiovascular disease, episodes of hypotension, and treatment assignment. Notably, SPRINT participants with AKI had a 50% higher risk of major cardiovascular events and a twofold higher risk of death, which was also independent of these covariates.
A target systolic blood pressure of <120 vs. <140 mm Hg produced more adverse events for kidney disease, including AKI and 30% eGFR decline, and did not prevent loss of kidney function over time [6]. Higher risk of AKI due to intensive blood pressure treatment was similarly elevated when analyzed by adjudicated events [7]. The present study builds upon prior findings by demonstrating that increased risk of AKI with intensive treatment was independent of hypotension and other clinical risk factors. Furthermore, there was an increased risk of a 30% decline in eGFR among those in the intensive arm. However, it is possible that a lower blood pressure target may have a different effect on outcomes over time in patients similar to SPRINT participants, who had a low prevalence of proteinuria. Additionally, in patients without CKD, a dose-response relationship was observed in which a greater reduction in mean arterial pressure associated with greater eGFR decline, and the balance between cardiovascular benefits and kidney risks were less favorable with larger blood pressure reductions [14]. A systematic review of 9 clinical trials conducted in participants with non-diabetic CKD demonstrated that compared to standard treatment, intensive blood pressure treatment did not reduce rates of overall eGFR decline, reaching 50% eGFR decline, or serum creatinine doubling [20]. In the Action to Control Cardiovascular Risk in Diabetes trial of patients with type 2 diabetes randomized to a target systolic blood pressure of <120 versus <140 mm Hg, eGFR was lower and risk of reaching eGFR <30 mL/min/1.73 m2 was higher without benefit on cardiovascular outcomes. Taken together, this compilation of data indicates that, for kidney disease, intensive blood pressure treatment may not provide benefit and increases the risk of adverse events.
In the present SPRINT analyses, episodes of AKI increased risks of major cardiovascular events and death from any cause. The mediation analysis showed that the proportional influence of intensive blood pressure treatment on both the primary and death outcomes was 85–90% for AKI, indicating that the cardiovascular and survival benefits of intensive blood pressure treatment could be abrogated by AKI. The increased risk of cardiovascular events related to AKI stands out as an important observation, even though the majority of AKI events were considered mild and most resolved in SPRINT [17]. Similar observations have been reported in several other studies and meta-analyses. A recent study reported a strong association between AKI and subsequent cardiovascular events with an 86% increased risk of cardiovascular death [15]. Another showed that compared to no AKI, stage 1 AKI following coronary angiography increased the risk of death twofold, while stage 2 or 3 AKI increased the risk of death approximately fourfold [9]. Several other studies have demonstrated that patients who experience AKI after major surgical procedures have higher risks of myocardial infarction, heart failure, stroke, and death from any cause [2, 3, 7, 8]. A meta-analysis of 47 studies, including 242,388 individuals demonstrated that AKI increased risk of mortality four-fold and stroke 2-fold [16]. Since AKI may increase volume retention, these episodes could exacerbate heart failure and myocardial ischemia [18]. AKI may also activate the sympathetic nervous system via ischemia in the kidney as a potential mechanism to increase risks of cardiovascular events and death [5, 11, 19].
The present analyses have several limitations. First, SPRINT was not prospectively designed to determine the effect of AKI on risks of cardiovascular events and death from any cause. As such, this analysis must be interpreted as exploratory and hypothesis generating. Second, access to adjudicated AKI event data was not available for these analyses. The modified Kidney Disease: Improving Global Outcomes criteria were used, which incorporates serum creatinine because urine output was not consistently measured. However, it is possible that some AKI events were not captured and AKI events were reported as singular events, which would underestimate the effect of AKI on the primary outcome. Hypotension was investigator reported, and as such the effect of hypotension may be underestimated in the present analysis. In the present analyses, intensive blood pressure lowering led to a higher risk of a 30% reduction in eGFR. However, it is possible that a blood pressure lower target could lead to benefits on kidney outcomes over time despite few SPRINT participants with proteinuria. These analyses focused on the relationship of AKI to cardiovascular events and death from any cause. Although the absolute rate of investigator-reported AKI was small, among these patients heightened vigilance for adverse outcomes, including cardiovascular disease, is warranted. Finally, we cannot exclude the possibility of reverse causality such that cardiovascular disease may contribute to occurrence of AKI.
In conclusion, among older adults with hypertension at high cardiovascular risk, targeting an intensive treatment goal of a systolic blood pressure <120 mm Hg independently increased risk of AKI, which meaningfully raised risks of major cardiovascular events and death. The present study highlights the need to balance the trade-off of reducing cardiovascular risk through intensive blood pressure treatment with increased risk of AKI and subsequent cardiovascular events.
Supplementary Material
Acknowledgments
SPRINT was sponsored by the National Heart, Lung, and Blood Institute along with co-sponsorship by the National Institute of Diabetes and Digestive and Kidney Diseases, the National Institute of Neurological Diseases and Stroke, and the National Institute on Aging. We would like to thank Dr. Allison Lambert, Celestina Barbosa-Leiker, and Cami Jones for their insightful comments on the manuscript and data analyses.
Funding Source
K.R.T. is supported by the following National Institutes of Health grants: National Center for Advancing Translational Sciences UL1 TR00043; National Institute of Diabetes, Digestive, and Kidney Diseases 1U2CDK114886-01, 5UM1DK100846-02, 1UC4DK101108-01, 1 U54 DK08 3912, U01 DK085689, R34 DK094016.
Footnotes
Financial Disclosure
K.R.T. has received consulting fees regarding therapies for diabetic kidney disease from Eli Lilly and Company, Boehringer Ingelheim, Astra Zeneca, and Gilead Sciences.
References
- 1.Bakris GL, Williams M, Dworkin L, Elliott WJ, Epstein M, Toto R, et al. ; National Kidney Foundation Hypertension and Diabetes Executive Committees Working Group. Preserving renal function in adults with hypertension and diabetes: a consensus approach. Am J Kidney Dis. 2000. September;36(3):646–61. [DOI] [PubMed] [Google Scholar]
- 2.Bihorac A, Yavas S, Subbiah S, Hobson CE, Schold JD, Gabrielli A, et al. Long-term risk of mortality and acute kidney injury during hospitalization after major surgery. Ann Surg. 2009. May;249(5):851–8. [DOI] [PubMed] [Google Scholar]
- 3.Chew ST, Ng RR, Liu W, Chow KY, Ti LK. Acute kidney injury increases the risk of end-stage renal disease after cardiac surgery in an Asian population: a prospective cohort study. BMC Nephrol. 2017. February;18(1):60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Egan BM, Zhao Y, Axon RN. US trends in prevalence, awareness, treatment, and control of hypertension, 1988-2008. JAMA. 2010. May;303(20):2043–50. [DOI] [PubMed] [Google Scholar]
- 5.Fujii T, Kurata H, Takaoka M, Muraoka T, Fujisawa Y, Shokoji T, et al. The role of renal sympathetic nervous system in the pathogenesis of ischemic acute renal failure. Eur J Pharmacol. 2003. November;481(2-3):241–8. [DOI] [PubMed] [Google Scholar]
- 6.Wright JT Jr, Williamson JD, Whelton PK, Snyder JK, Sink KM, Rocco MV, et al. ; SPRINT Research Group. A Randomized Trial of Intensive versus Standard Blood-Pressure Control. N Engl J Med. 2015. November; 373(22):2103–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hansen MK, Gammelager H, Jacobsen CJ, Hjortdal VE, Layton JB, Rasmussen BS, et al. Acute Kidney Injury and Long-term Risk of Cardiovascular Events After Cardiac Surgery: A Population-Based Cohort Study. J Cardiothorac Vasc Anesth. 2015;29(3):617–25. [DOI] [PubMed] [Google Scholar]
- 8.Hobson CE, Yavas S, Segal MS, Schold JD, Tribble CG, Layon AJ, et al. Acute kidney injury is associated with increased long-term mortality after cardiothoracic surgery. Circulation. 2009. May; 119(18):2444–53. [DOI] [PubMed] [Google Scholar]
- 9.James MT, Ghali WA, Knudtson ML, Ravani P, Tonelli M, Faris P, et al. ; Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease (APPROACH) Investigators. Associations between acute kidney injury and cardiovascular and renal outcomes after coronary angiography. Circulation. 2011. February; 123(4):409–16. [DOI] [PubMed] [Google Scholar]
- 10.James PA, Oparil S, Carter BL, Cushman WC, Dennison-Himmelfarb C, Handler J, et al. 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA. 2014. February;311(5):507–20. [DOI] [PubMed] [Google Scholar]
- 11.Kelleher SP, Robinette JB, Conger JD. Sympathetic nervous system in the loss of autoregulation in acute renal failure. Am J Physiol. 1984. April;246(4 Pt 2):F379–86. [DOI] [PubMed] [Google Scholar]
- 12.Lewington S, Clarke R, Qizilbash N, Peto R, Collins R; Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet. 2002. December; 360(9349):1903–13. [DOI] [PubMed] [Google Scholar]
- 13.Lim SS, Vos T, Flaxman AD, Danaei G, Shibuya K, Adair-Rohani H, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012. December;380(9859): 2224–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Magriço R, Bigotte Vieira M, Viegas Dias C, Leitão L, Neves JS. BP Reduction, Kidney Function Decline, and Cardiovascular Events in Patients without CKD. Clin J Am Soc Nephrol. 2018. January;13(1):73–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Odutayo A, Wong CX, Farkouh M, Altman DG, Hopewell S, Emdin CA, et al. AKI and Long-Term Risk for Cardiovascular Events and Mortality. J Am Soc Nephrol. 2017. January; 28(1):377–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Pickering JW, James MT, Palmer SC. Acute kidney injury and prognosis after cardiopulmonary bypass: a meta-analysis of cohort studies. Am J Kidney Dis. 2015. February;65(2): 283–93. [DOI] [PubMed] [Google Scholar]
- 17.Rocco MV, Sink KM, Lovato LC, Wolfgram DF, Wiegmann TB, Wall BM, et al. Effects of Intensive Blood Pressure Treatment on Acute Kidney Injury Events in the Systolic Blood Pressure Intervention Trial (SPRINT). Am J Kidney Dis. 2018. March;71(3):352–361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ronco C, Cruz DN, Ronco F. Cardiorenal syndromes. Curr Opin Crit Care. 2009. October; 15(5):384–91. [DOI] [PubMed] [Google Scholar]
- 19.Salman IM, Ameer OZ, Sattar MA, Abdullah NA, Yam MF, Najim HS, et al. Role of the renal sympathetic nervous system in mediating renal ischaemic injury-induced reductions in renal haemodynamic and excretory functions. Pathology. 2010. April;42(3):259–66. [DOI] [PubMed] [Google Scholar]
- 20.Tsai WC, Wu HY, Peng YS, Yang JY, Chen HY, Chiu YL, et al. Association of Intensive Blood Pressure Control and Kidney Disease Progression in Nondiabetic Patients With Chronic Kidney Disease: A Systematic Re-view and Meta-analysis. JAMA Intern Med. 2017. June;177(6):792–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Xie Y, Bowe B, Li T, Xian H, Yan Y, Al-Aly Z. Long-term kidney outcomes among users of proton pump inhibitors without intervening acute kidney injury. Kidney Int. 2017. June; 91(6):1482–94. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.