Patient-Oriented, Translational Research: Research Article

Nephrology
American Journal of

Am J Nephrol 2019;49:297–306 Received: December 6, 2018

Accepted: February 26, 2019
DOI: 10.1159/000499188 Published online: March 27, 2019

Modifying Effect of Statins on Fatal Outcomes in

Chronic Kidney Disease Patients in the Systolic Blood
Pressure Intervention Trial: A Post Hoc Analysis
Manuel Rivera a Leonardo Tamariz b Maritza Suarez c Gabriel Contreras a
       

a Katz
Family and Division of Nephrology and Hypertension, Department of Medicine, University of Miami Miller
School of Medicine, Miami, FL, USA; b Division of Population Health and Computational Medicine, Department of
 

Medicine, University of Miami Miller School of Medicine, Miami, FL, USA; c Division of General Internal Medicine,
 

Department of Medicine, University of Miami Miller of School of Medicine, Miami, FL, USA

Keywords
Hypertension · Chronic kidney disease · Mortality · Statin
Abstract
Background: Management of chronic kidney disease (CKD)
patients includes efforts directed toward modifying tradi-
tional cardiovascular risk factors. Such efforts include opti-
mal management of hypertension together with the initia-
tion of statin therapy. Methods: In this observational study,
we determine the modifying effect of statins on the relation-
ship of systolic blood pressure (SBP) goal with mortality and
other outcomes in patients with CKD participating in a clini-
cal trial. At baseline, 2,646 CKD patients (estimated glomeru-
lar filtration rate <60 mL/min/1.73 m2) were randomized to
an intensive SBP goal < 120 mm Hg or standard SBP goal
<140 mm Hg. One thousand two hundred and seventy-three
were not on statin, 1,354 were on a statin, and in 19 the use
of statin was unknown. The 2 primary outcomes were all-
cause mortality and cardiovascular disease (CVD) mortality.
Results: The relationships of SBP goal with all-cause mortal-
ity (interaction p = 0.009) and cardiovascular (CV) mortality
(interaction p = 0.021) were modified by the use of statin af-
ter adjusting for age, gender, race, CVD history, smoking, as-
pirin use, and blood pressure at baseline. In the statin group,
targeting SBP to < 120 mm Hg compared to SBP < 140 mm
Hg significantly reduced the risk of all-cause mortality (ad-
justed hazard ratio [aHR] 0.44 [0.28–0.71]; event rates 1.16 vs.
2.5 per 100 patient-years) and CV mortality (aHR 0.29 [0.12–
0.74]; event rates 0.28 vs. 0.92 per 100 patient-years) after a
median follow-up of 3.26 years. In the non-statin group, the
risk of all-cause mortality (aHR 1.07 [0.69–1.66]; event rates
2.01 vs. 1.94 per 100 patient-years) and CV mortality (aHR
1.42 [0.56–3.59]; event rates 0.52 vs. 0.41 per 100 patient-
years) were not significantly different in both SBP goal arms.
Conclusion: The combination of statin therapy and intensive
SBP management leads to improved survival in hypertensive
patients with CKD. © 2019 S. Karger AG, Basel
Introduction
The association of hypertension with cardiovascular
disease (CVD) in chronic kidney disease (CKD) patients
has been well documented in observational studies [1–
3]. The systolic blood pressure (SBP) Intervention trial
(SPRINT) demonstrated that intensive SBP control was

© 2019 S. Karger AG, Basel Gabriel Contreras, MD, MPH

Department of Medicine, University of Miami
Jackson Memorial Hospital
E-Mail karger@karger.com
Miami, FL 33136 (USA)
www.karger.com/ajn E-Mail gcontrer @ med.miami.edu
associated with significantly lower risk of major cardio-
vascular (CV) events including acute myocardial infarc-
tion (MI), acute coronary syndrome (ACS), stroke, acute
decompensated heart failure (HF), CV mortality and all-
cause mortality [4]. These effects were independent of
renal impairment at baseline and therefore favored in-
tensive SBP targets for patients with underlying CKD
[5]. Consequently, one of the cornerstones in the man-
agement of patients with CKD focuses on optimal blood
pressure (BP) control in conjunction with rigorous mod-
ification of all of the other coexistent risk factors for
CVD [6].
Parallel to intensive SBP management, the addition of
statin therapy has been demonstrated to be equally im-
portant for the reduction of major CV events and is there-
fore regarded as a standard practice in the management
of patients with CKD not on dialysis [7, 8]. As shown in
the Study of Heart and Renal Protection, the use of statin
and ezetimibe led to less major cardiac events in patients
with CKD, this beneficial effect has been particularly
found to be more pronounced for patients not on dialysis
[7]. This being the case, our study seeks to determine the
modifying effect of statins on the relationship of SBP goal
with mortality in participants of SPRINT with underlying
CKD.
Methods
Population
The SPRINT is a randomized, controlled, open-label trial con-
ducted at 102 clinical sites in the United States [4]. The study de-
sign and main results have been reported in prior publications [4,
5]. The trial was stopped early for benefit owing to improved out-
comes with intensive SBP management. Median time of follow-
up  for SPRINT participants was 3.3 years. In this investigation,
we conducted an observational post hoc analysis focusing on the
2,646 patients with CKD enrolled in this study. CKD was de-
fined as the baseline estimated glomerular filtration rate (eGFR)
<60 mL/min/1.73 m2, calculated with the use of the 4 variable
Modification of Diet in Renal Disease equation [9]. Serum creati-
nine values were measured in the central laboratories using an en-
zymatic procedure (Roche, Indianapolis, IN, USA).
Following stratification of intervention groups by statin use at
baseline, study patients were classified into 4 groups: (a) intensive
SBP arm, on statin therapy, (b) standard SBP arm, on statin ther-
apy, (c) intensive SBP arm, not on statin therapy, and (d) standard
SBP arm, not on statin therapy.
Blood Pressure Interventions
Patients enrolled in the SPRINT were randomized to a SBP goal
of < 120 mm Hg (intensive arm) or < 140 mm Hg (the standard
arm). Patients and study personnel were aware of the intervention
group assignments. After randomization, patients’ baseline anti-
298 Am J Nephrol 2019;49:297–306
DOI: 10.1159/000499188
hypertensive regimens were adjusted on the basis of the interven-
tion group assignment. Patients were seen monthly for the first
3 months and every 3 months thereafter. After 3 months in the
intensive arm, if the SBP goal had not been reached, medications
were adjusted every month until achieving a SBP <120 mm Hg. For
patients in the standard arm, medications were adjusted until a
SBP between 135 and 139 mm Hg was achieved. Dose adjustment
was based on a mean of 3 BP measurements at an office visit while
the patient was seated and after 5 min of quiet rest; the measure-
ments were made with the use of an automated measurement sys-
tem (Model 907, Omron Healthcare). All major classes of antihy-
pertensive agents were included in the formulary and were pro-
vided at no cost to the patients. The protocol encouraged, but did
not mandate, the use of drug classes with the strongest evidence
for reduction in outcomes.
Outcomes
For the purposes of this investigation, the primary outcomes
were all-cause mortality and CV mortality. Secondary outcomes
included CHF, MI, stroke, ACS, composite of first occurrence of
CHF, MI, stroke, ACS or CV mortality and composite of first oc-
currence of CHF, MI, stroke, ACS, and all-cause mortality.
Statistical Analyses
All participants with CKD at baseline, except 19 for whom the
use of statin was unknown, were included in this study. Baseline
characteristics were summarized as frequencies and proportions
for categorical variables and as medians and interquartile ranges
(IQRs) for continuous variables. Comparisons of baseline categor-
ical variables between the intensive arm and the standard arm were
performed using χ2 tests. Comparisons of baseline continuous
variables between the intensive arm and the standard arm were
performed using Student t tests or Wilcoxon rank sum tests as ap-
propriate. Similarly, categorical and continuous variables were
compared after stratifying by use of statin at baseline.
Follow-up SBP was compared between SBP goal arms after
stratification based on statin use at baseline. This comparison was
conducted using a multilevel mixed linear model with an unstruc-
tured covariance matrix controlling for within subject correlation.
The 19 participants with missing data for statin use at baseline
were not accounted for in this longitudinal SBP comparison.
Adjusted multivariate proportional hazards regression models
were primarily used to investigate the role of statin use at baseline
as a potential effect-modifier for the relationships between SBP
goal with risk of all-cause mortality, CV mortality, and other sec-
ondary outcomes. This model was adjusted for the following base-
line predictors: age, sex, race, aspirin use, history of CVD, current
smoking, SBP, and DBP.
In subsequent subgroups analyses, separate Cox regression
models stratified by statin use at baseline were used to estimate
hazard ratios (HRs) and their respective 95% CIs for primary and
secondary outcomes. HR estimated the impact of targeting an in-
tensive SBP goal compared to standard SBP goal (reference group).
Event rates of outcomes were calculated as number of events per
100 patient-years (pt-ys). Similarly, serious adverse events (SAEs)
and clinical safety alerts (CSAs) were compared between the inten-
sive SBP goal and the standard SBP goal arms.
Finally, 2 sensitivity analyses were conducted: First, due to
the  imbalance in baseline characteristics between the 1,354 pa-
tients on statins compared to the 1,273 patients not on statin ther-
Rivera/Tamariz/Suarez/Contreras
apy at baseline, a cohort of 846 pairs of patients matched for
the  baseline characteristics presented in online supplementary
­Table S1 (for all online suppl. material, see www.karger.com/
doi/10.1159/000499188) was assembled by employing the predict-
ed probability of belonging to the statin group within the 2 differ-
ent SBP goal strata. Initially, we used multivariate logistic regres-
sion in which statin use at baseline was used as the dependent
­variable, and baseline characteristics presented in online supple-
mentary Table S1 as covariates (urinary albumin-to-creatinine and
body mass index were the only 2 variables not included for this
multivariate regression due to the considerable amount of missing
data). After stratifying by SBP goal, a greedy matching algorithm
using Mahalanobis distance method and a caliper radius of 0.5 *
sigma was used. This led to the matching of 418 (63%) patients not
on statins with 418 (64%) patients on statin therapy within the in-
tensive SBP goal group; and to the matching of 428 (70%) patients
not on statins with 428 (61%) patients on statin therapy within the
standard SBP goal group [10, 11].
Furthermore, an additional sensitivity analysis was run in the
subgroup of patients with an elevated 10-year risk of atheroscle-
rotic CVD (ASCVD), defined as 7.5% or above.
Cox models were used to investigate the role of statin use at
baseline as a potential effect-modifier on the relationships between
SBP goal and risk of all-cause mortality as well as CV mortality for
these sensitivity analyses.
Stata/IC 14.2 (Stata-Corp., College Station, TX, USA) and
NCSS 11 Statistical Software (NCSS, LLC, Kaysville, UT, USA)
were employed to conduct the statistical analysis.
Results
Baseline Characteristics
Of the 9,361 SPRINT participants, 2,646 had CKD at
baseline: 1,756 (66%) CKD stage IIIa, 735 (28%) CKD
stage IIIb, and 154 (6%) CKD stage IV. Fifty-one per-
cent (n = 1,354) were on a statin, and 48% (n = 1,273)
were not on statin. Data regarding use of statin at base-
line were missing for 1% (n = 19) of study participants
(S3). In contrast to the non-statin group, subjects on
the statin group were predominantly non-Hispanic
whites (72 vs. 62%). Forty-eight percent of patients on
the statin group compared to 40% in the non-statin
group were 75 years of age or older at the time of ran-
domization. Women made up a smaller proportion in
the statin group (36 vs. 44%). Moreover, the statin
group had a much higher prevalence of subclinical and
clinical CVD (34 vs. 14%), had lower SBP (137 vs. 139
mm Hg) and fasting total cholesterol levels (167 vs. 195
mg/dL), and were also more likely to be taking aspirin
(69 vs. 42%) at baseline. Both groups had similar eGFR
(50 mL/min/1.73 m2; Table 1). In the propensity score
matched cohort, statin and non-statin group differenc-
es were balanced on all baseline characteristics in the
Modifying Effect of Statins on Fatal
Outcomes in CKD Patients in the SPRINT
matched pairs (online suppl. Table S1). Following strat-
ification of the SPRINT CKD patients by statin use,
within subgroup comparisons revealed no significant
differences in most baseline demographic, clinical or
laboratory variables by BP intervention strategy (Table
1 and online suppl. Table S2).
Follow-Up Blood Pressure
The intensive arm and standard arm achieved a rapid
separation in the BP values by 3 months after randomiza-
tion (Fig. 1). At 1-year, the median BP for the non-statin
groups was 137 mm Hg (IQR 129–145) over 76 mm Hg
(IQR 67–83) in the standard arm and 119 mm Hg (IQR
113–129) over 68 mm Hg (IQR 60–75) in the intensive
arm; the median BP for the statin groups was 136 mm Hg
(IQR 128–143) over 73 mm Hg (IQR 65–81) in the stan-
dard arm and 121 mm Hg (IQR 114–133) over 66 mm Hg
(IQR 58–72) in the intensive arm. Throughout the dura-
tion of the study, median BP for the non-statin groups was
136 mm Hg (IQR 127–144) over 74 mm Hg (IQR 65–82)
in the standard arm and 120 mm Hg (IQR 114–132) over
68 mm Hg (IQR 60–75) in the intensive arm; median BP
for the statin groups was 135 mm Hg (IQR 126–143) over
71 mm Hg (IQR 64–79) in the standard arm and 121 mm
Hg (IQR 114–133) over 65 mm Hg (IQR 58–72) in the
intensive arm. The mean number of antihypertensive
agents prescribed at the last study visit for the non-statin
group was 1.9 ± 1.2 in the standard arm compared with
2.9 ± 1.2 in the intensive arm; mean number of antihyper-
tensive agents for the statin subgroup was 2.1 ± 1.2 in the
standard arm and 2.9 ± 1.2 in the intensive arm.
Outcomes
A total of 163 patients died during the study period: 82
of 1,354 patients (6.06%) in the statin group and 81 of
1,273 patients (6.36%) in the non-statin group. In the ad-
justed Cox regression model relating all-cause mortality
jointly to the SBP goal strategy and statin use at baseline,
the risk of all-cause mortality associated to the SBP
goal  strategy differed significantly between the patients
with and without statin use at baseline (adjusted p for in-
teraction = 0.009). In the statin group, targeting a SBP to
<120 mm Hg compared to SBP <140 mm Hg significant-
ly reduced the risk of all-cause mortality (adjusted HR
[aHR] 0.44, [0.28–0.71]; event rates 1.16 vs. 2.5 per 100
patient-years). In the non-statin group, the risk of all-
cause mortality was not significantly different in both
SBP goal arms (aHR 1.07 [0.69–1.66]; event rates 2.01 vs.
1.94 per 100 patient-years; Table 2; Fig. 2; online suppl.
Fig. S1b).
Am J Nephrol 2019;49:297–306 299
DOI: 10.1159/000499188
Table 1. Baseline subgroups characteristics of SPRINT patients with CKD

Non-statin (n = 1,273) Statin (n = 1,354)

standard intensive standard intensive

Variables, n 609 664 697 657

Age, years, median (IQR) 72 (63–78) 72 (63–78) 74 (67–79) 74 (67–79)
Gender, female, n (%) 257 (42.2) 308 (46.4) 259 (37.2) 227 (34.5)
Race, n (%)
Non-hispanic black 164 (26.9) 201 (30.3) 146 (20.9) 124 (18.9)
Hispanic 52 (8.5) 43 (6.5) 44 (6.3) 50 (7.6)
Non-hispanic white 382 (62.7) 410 (61.7) 503 (72.2) 467 (71.1)
Other 11 (1.8) 10 (1.5) 4 (0.6) 16 (2.4)
African Americans, n (%) 168 (27.6) 201 (30.3) 146 (20.9) 127 (19.3)
Baseline BP, mm Hg, median (IQR)
Systolic 139 (130–150) 139 (129–150) 137 (128–147) 137 (129–148)
Diastolic 77 (68–84) 77 (67–85) 73 (65–81) 74 (66–82)
Serum creatinine, mg/dL, median (IQR) 1.34 (1.2–1.57) 1.34 (1.17–1.59) 1.33 (1.21–1.58) 1.34 (1.21–1.58)
Estimated GFR, mL/min/1.73 m2, median (IQR) 50.2 (42.6–55.8) 50.1 (41.7–55.4) 49.5 (41.2–55.6) 50.0 (42.1–55.4)
Urinary albumin to creatinine, mg/g, median (IQR)* 13.3 (6.0–43.4) 12.9 (6.2–37.1) 14 (6.7–43.5) 12.8 (7.0–46.7)
Fasting total cholesterol, mg/dL, median (IQR)¶ 193 (171–222) 197 (174–224) 167 (148–192) 167 (146–193)
Fasting HDL cholesterol, mg/dL, median (IQR)¶ 50 (42–60) 51 (43–62) 50 (42–59) 50 (42–60)
Fasting total triglycerides, mg/dL, median (IQR)¶ 113 (83–161) 107 (76–153) 113 (81–154) 111 (82–149)
Fasting plasma glucose, mg/dL, median (IQR)¶ 96 (90–103) 95 (88–103) 98 (92–105) 98 (91–106)
Aspirin use, n (%)§ 259 (42.5) 281 (42.3) 464 (66.9) 465 (70.8)
Smoking status, n (%)
Never smoked 308 (50.6) 330 (49.7) 289 (41.5) 270 (41.1)
Former smoker 239 (39.2) 275 (41.4) 356 (51.1) 339 (51.6)
Current smoker 62 (10.2) 59 (8.9) 51 (7.3) 48 (7.3)
Missing data 0 (0) 0 (0) 1 (0.1) 0 (0)
Cardiovascular disease, n (%) 83 (13.6) 90 (13.5) 236 (33.9) 230 (35.0)
Framingham 10-year cardiovascular
disease risk score, median (IQR)# 19.8 (12.3–30.6) 18.8 (12.3–28.2) 18.5 (12.1–26.7) 18.6 (12.6–27.0)
BMI, kg/m2, median (IQR)+ 28.7 (25.3–32.7) 28.6 (25.5–32.8) 28.4 (25.5–31.7) 28.5 (25.7–32.4)
No antihypertensive, n (%) 41 (6.7) 39 (5.9) 20 (2.9) 21 (3.2)

* 50 missing values in the non-statin group; 41 missing values in the statin group.
¶
 1 missing value in the statin group.
§ 3 missing values in the statin group.
# 3 missing values in the non statin group; 5 missing values in the statin group.
+ 8 missing values in the non statin group; 10 missing values in the statin group.

BMI, body mass index; BP, blood pressure; IQR, interquartile ranges; SPRINT, systolic blood pressure intervention trial; CKD,
chronic kidney disease.

Forty-six patients died from CV events: 27 (1.99%) in
the statin group and 19 (1.49%) in the non-statin group.
In the Cox regression model relating CV mortality jointly
to the SBP goal strategy and statin use at baseline, the risk
of CV mortality associated to the SBP goal strategy dif-
fered significantly between the patients with and without
statin use at baseline (adjusted p for interaction = 0.021).
In the statin group, targeting a SBP to <120 mm Hg com-
pared to SBP <140 mm Hg significantly reduced the risk
of CV mortality (aHR 0.29 [0.12–0.74]; event rates 0.28 vs.
0.92 per 100 pt-ys). In the non-statin group, the risk of CV
mortality was not significantly different in both SBP goal
arms (aHR 1.42 [0.56–3.59]; event rates 0.52 vs. 0.41 per
100 patient-years; Table 2; Fig. 2; online suppl. Fig. S1a).
The modifying effects of statins on the relationship of
SBP goal with other outcomes were not significant (on-
line suppl. Table S4).
Sensitivity Analyses
A modifying effect of statins on the relationship of
SBP goal and all-cause mortality was also appreciated
with the propensity score matched analysis (p for inter-

300 Am J Nephrol 2019;49:297–306 Rivera/Tamariz/Suarez/Contreras

DOI: 10.1159/000499188
 140
140

135
135

SBP, mm Hg
SBP, mm Hg

130 130

125 125

120 120

115
115
0 6 12 18 24 30 36 42 48 54 0 6 12 18 24 30 36 42 48 54
a Follow-up, month b Follow-up, month

Number with data Number with data

Standard 609 556 528 501 491 473 347 220 110 23 Standard 697 649 618 607 587 540 417 257 120 23
Intensive 664 619 591 562 557 522 399 256 137 36 Intensive 657 620 594 575 573 527 403 257 130 22
Mean number of meds Mean number of meds
Standard 2.1 1.9 1.9 1.9 1.9 1.9 2.0 2.0 2.0 2.2 Standard 2.2 2.0 2.0 2.1 2.1 2.0 2.0 2.1 2.1 1.9
Intensive 2.4 2.8 2.9 2.9 2.9 2.9 2.9 2.8 2.8 3.1 Intensive 2.5 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.1 3.0

Fig. 1. Separation achieved in SBP levels in the non-statin group (a) and statin group (b). The broken line depicts
the standard group; the solid line depicts the intensive group. SBP, systolic blood pressure.

action = 0.026). In the statin group, targeting a SBP to
< 120 mm Hg compared to SBP < 140 mm Hg was as-
sociated with a reduced risk of all-cause mortality (HR
0.39, [0.20–0.73]). In contrast, the risk of all-cause mor-
tality was not significantly different in both SBP goal
arms (HR 0.96 [0.58–1.59]) in the group not on statin
therapy. With regard to the CV mortality outcome,
while taking into account the constraints associated
with the small number of events recorded, the results
of this sensitivity analysis also suggested a trend for an
interaction effect between statin use and SBP goal on
CV mortality (p for interaction = 0.064; online suppl.
Table S5).
An additional sensitivity analysis was conducted in
the group of patients with a 10-year ASCVD risk ≥7.5%
and did not show that statins modified the effect of the
SBP goal for the all-cause mortality outcome (p for inter-
action = 0.103) or CV mortality (p for interaction =
0.475); however, subgroup analyses showed that target-
ing a SBP <120 mm Hg compared to <140 mm Hg sig-
nificantly reduced the risk of all-cause mortality (aHR
0.55 [0.40–0.77]) and CV mortality (aHR 0.51 [0.28–
0.92]) only in the group that was on statins at base-
line. Targeting a SBP to <120 mm Hg compared to SBP
<140 mm Hg was associated with a trend toward reduc-
ing the risk of all-cause mortality (aHR 0.82 [0.60–1.11])
and CV mortality (aHR 0.71 [0.37–1.36]) for the sub-
group of patients not on statins.
SAE, CSA, and Expected Events
Overall, no statistically significant differences of total
SAE were reported between the intensive SBP arm and stan-
dard SBP arm in patients not on statin or on statin at base-
line. In the non-statin group, no significant differences of
CSA and expected events were reported between the inten-
sive SBP arm and standard SBP arm. On the other hand, in
the statin group, patients randomized to the intensive SBP
arm were at higher risk of developing acute kidney injury
(AKI) and hypokalemia. The risk of hypotension, syncope,
bradycardia, injurious fall, hyponatremia, hypernatremia,
hyperkalemia, and orthostatic hypotension did not differ
between treatment arms in the statin group (Table 3).
Numbers Needed to Treat for Benefit and Harm
Estimated number needed to treat to prevent a death
from any cause and death from CV causes at 4.5 years of
follow-up were 16 (95% CI 10–35) and 33 (95% CI 20–95)
in the statin group. The numbers needed to harm for AKI
and hypokalemia were 17 (95% CI 10–90) and 12 (95% CI
6–134), respectively, in this same group [12].

Modifying Effect of Statins on Fatal Am J Nephrol 2019;49:297–306 301

Outcomes in CKD Patients in the SPRINT DOI: 10.1159/000499188
 Discussion

HR, hazard ratio; SBP, systolic blood pressure; MI, myocardial infarction; ACS, acute coronary syndrome; CVA, stroke; aHF, acute decompensated heart failure; CVD death, death from cardiovascular causes.
interaction*
Adjusted

0.009

0.021
The prevalence of hypertension is as high as 86% in
p for

CKD patients not on dialysis [3, 13], and it is strongly as-

0.44 (0.28–0.71)

0.29 (0.12–0.74)
sociated with mortality [14]. Moreover, CKD patients
have an increased risk for adverse CVD outcomes [15,
16]. Consequently, one of the essential components in the
aHR*

management of CKD patients focuses in the early and

 

* Adjusted for 10 covariates including: age, gender, race or ethnic group, history of subclinical/clinical CVD, aspirin use at baseline, smoking status, SBP and DBP at baseline.
optimal modification of CVD risk factors: alluding to in-
0.46 (0.29–0.74)

0.30 (0.12–0.74)

tensive SBP control as well as to initiating statin therapy

[6]. Recommendations issued by the Kidney Disease Im-
proving Global Outcomes organization advocate for ini-
HR

tiating statin therapy for all adults aged 50 years or above

 

patient-
intensive (n = 657)

with CKD not on dialysis [17]. In our post hoc analysis,

year, n

1.16

0.28
100

we demonstrated that statin use modifies the risk rela-

tionship between SBP goal and both all-cause mortality
patients

25 (3.8)

6 (0.9)

and CV mortality in the SPRINT CKD participants.

The publication by Cheung et al. [5] demonstrated
that for SPRINT participants with CKD at baseline
standard (n = 697)

patient-
years, n
Statin (n = 1,354)

(SPRINT CKD), a SBP target of <120 mm Hg compared

2.5

0.92
100

to SBP <140 mm Hg led to a 28% reduction in the risk

patients

57 (8.2)

21 (3.0)

of all-cause mortality (event rates 1.61 vs. 2.21 per 100

pt-ys). Our results extend the finding of the SPRINT
CKD study: as we showed that for patients with CKD on
1.07 (0.69–1.66)

1.42 (0.56–3.59)

statin therapy, targeting a SBP to < 120 mm Hg com-

pared to SBP <140 mm Hg further reduced the risk of
Table 2. All-cause mortality and cardiovascular mortality outcomes by subgroup

all-cause mortality to 56% (event rates 1.16 vs. 2.5 per

aHR*

100 patient-years) and the risk of CV mortality to 71%

 

(event rates 0.28 vs. 0.92 per 100 patient-years). No sig-

1.02 (0.66–1.59)

1.25 (0.50–3.10)

nificant differences in risk of all-cause mortality (event

rates 2.01 vs. 1.94 per 100 patient-years) or CV mortal-
ity (event rates 0.52 vs. 0.41 per 100 patient-years) were
HR
 

appreciated between SBP intervention arms in the non-

patient-
years, n
intensive (n = 664)

statin group. Supporting our findings, experimental

2.01

0.52
100

studies have described improved effects from BP-lower-

ing therapy in hypertensive patients managed concur-
patients

43 (6.5)

11 (1.7)

rently with a statin [18–20]. Ge et al. [19] showed that

the combination of statin and antihypertensive therapy
compared to antihypertensive therapy alone allowed
patient-
No statin (n = 1,273)

years, n

better SBP control (123 ± 12 vs. 137 ± 11 mm Hg, p <

standard (n = 609)

1.94

0.41
100

0.05), lower high-sensitive C-reactive protein levels, and

improved cardiac remodeling (left ventricular mass in-
patients

38 (6.2)

8 (1.3)

dex: 103 ± 32 vs. 111 ± 38 g/m2, p < 0.05) [19]. Kamberi

et al. [21] described that atherosclerotic plaque regres-
sion resulting from statin use led to better responses to
cardiovascular causes

BP-lowering therapy in hypertensive patients. Ferrier et

All-Cause Mortality

al. [20] additionally showed that the use of statin in pa-

tients with isolated systolic hypertension significantly
Death from
Outcomes

improved systemic arterial compliance (0.43 ± 0.05 vs.

0.36 ± 0.03 mL/mm Hg, p = 0.03) measured noninva-

302 Am J Nephrol 2019;49:297–306 Rivera/Tamariz/Suarez/Contreras

DOI: 10.1159/000499188
 0.05 0.05

0.04 0.04
Cumulative hazards

Cumulative hazards
0.03 0.03

p = 0.634 p = 0.006
0.02 0.02

0.01 0.01

0 0
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Follow-up, years Follow-up, years
Number at risk
Standard 609 584 553 378 119 697 675 657 458 138
a Intensive 664 632 612 416 132 657 634 620 424 140

0.12 0.12

0.10 0.10

Cumulative hazards
Cumulative hazards

0.08 0.08

0.06 p = 0.897 0.06 p = 0.001

0.04 0.04

0.02 0.02

0 0
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Follow-up, years Follow-up, years
Number at risk
Standard 609 584 557 378 119 697 682 659 460 138
b Intensive 664 637 614 416 134 657 635 624 425 140

Fig. 2. Cumulative hazards plot for prespecified outcomes of in the statin group. Black interrupted curves represented intensive
SPRINT trial: left-sided curves depict patients with CKD in the non- arm, and solid black curves represented standard arm. a Cardiovas-
statin group, whereas right-sided curves depict patients with CKD cular mortality. b All-cause mortality. p values by log-rank test.

sively by carotid applanation tonometry and Doppler
velocimetry of the ascending aorta. This synergistic ef-
fect stemming from the combination of antihyperten-
sive therapy and statins appears to transcend cholesterol
reduction uniquely and suggests a more intricate form
of statin-mediated vascular pleiotropic effect [21, 22].
As previously mentioned, lipid-lowering therapy is an
important cornerstone in the management of all patients
with CKD not on dialysis. Individuals with underlying
CKD are at a much higher risk for developing hard car-
diovascular outcomes when compared to individuals
with normal renal function. Consequently, the most re-
cent cholesterol guideline identifies CKD as a risk en-
hancer for ASCVD [23]. Evidence from randomized clin-
ical trials and meta-analysis has substantiated the effect of
statins in reducing major CVD events, all-cause mortal-
ity, and CV mortality [7, 8, 24]. The Study of Heart and
Renal Protection trial demonstrated that the use of simv-
astatin plus ezetimibe achieved a significant 17% reduc-
tion in the risk of CVD events [7]. Furthermore, in the
Treating to New Targets trial, atorvastatin 80 mg/day
compared to 10 mg/day achieved a significant 32% reduc-
tion in the risk of major CVD events in the subgroup of
patients with CKD [24].

Modifying Effect of Statins on Fatal Am J Nephrol 2019;49:297–306 303

Outcomes in CKD Patients in the SPRINT DOI: 10.1159/000499188
Table 3. Serious adverse events and monitored expected events

Events No statin (n = 1,273) Statin (n = 1,354)

standard intensive HR standard intensive HR
(n = 609) (n = 664) (n = 697) (n = 657)

Total SAEs 285 (46.8) 290 (43.7) 0.90 (0.76–1.06) 353 (50.6) 332 (50.5) 1.01 (0.87–1.18)
Conditions of interest (ER visits or SAEs)
Hypotension 20 (3.3) 23 (3.5) 1.06 (0.58–1.92) 25 (3.6) 36 (5.5) 1.52 (0.91–2.54)
Syncope 16 (2.6) 23 (3.5) 1.31 (0.69–2.48) 28 (4.0) 36 (5.5) 1.35 (0.82–2.21)
Bradycardia 21 (3.4) 15 (2.3) 0.65 (0.33–1.26) 21 (3.0) 24 (3.6) 1.20 (0.67–2.16)
Electrolyte abnormalities 28 (4.6) 35 (5.3) 1.15 (0.70–1.88) 25 (3.6) 36 (5.5) 1.54 (0.92–2.56)
Injurious fall 59 (9.7) 60 (9.0) 0.91 (0.63–1.31) 80 (11.5) 72 (11.0) 0.95 (0.69–1.31)
AKI 40 (6.6) 60 (9.0) 1.40 (0.94–2.08) 40 (5.7) 57 (8.7) 1.52 (1.02–2.28)
Expected events
Serum sodium <130 mEq/L 23 (3.8) 25 (3.8) 0.96 (0.54–1.69) 12 (1.7) 21 (3.2) 1.82 (0.89–3.71)
Serum sodium >150 mEq/L 0 (0) 1 (0.1) – 0 (0) 2 (0.3) –
Serum potassium <3.0 mEq/L 12 (2.0) 15 (2.3) 1.09 (0.51–2.33) 2 (0.3) 14 (2.1) 6.97 (1.58–30.70)
Serum potassium >5.5 mEq/L 39 (6.4) 49 (7.4) 1.10 (0.72–1.68) 37 (5.3) 51 (7.8) 1.44 (0.94–2.21)
Orthostatic hypotension
Without dizziness 127 (20.8) 141 (21.2) 0.96 (0.76–1.22) 172 (24.7) 155 (23.6) 0.94 (0.75–1.16)
With dizziness 7 (1.1) 15 (2.3) 1.78 (0.73–4.38) 15 (2.1) 9 (1.4) 0.63 (0.28–1.45)

SAE, serious adverse events; AKI, acute kidney injury; HR, hazard ratio.

Interestingly, in our propensity score-matched cohort,
the use of statins alone was associated with a nonstatisti-
cally significant reduced risk of all-cause death (HR 0.76
[0.52–1.11]) and a nonstatistically significant increased
risk of CV mortality (1.11 [0.53–2.34]). It is important to
mention that this analysis was underpowered and not de-
void of bias given that statin therapy was not a random-
ized intervention in SPRINT.
In spite of these findings, only 51% of the CKD pa-
tients enrolled in the SPRINT were using statins at base-
line. Considering that the risk of cardiovascular mortality
increases with decreasing eGFR and that the 10-year risk
for coronary death and nonfatal MI for CKD patients
over the age of 50 has been noted to be consistently great-
er than 10% [25], this low proportion calls into question
the quality of care that patients with underlying CKD
were receiving in the primary care and specialty clinics
that referred patient to the SPRINT. More importantly,
this also reveals the need to improve dissemination efforts
focusing on guideline implementation at the primary care
level.
Overall, few SAE were reported in the subgroup of
patients not on statins as in the subgroup of patients on
statins. With the exception of a higher risk of AKI and
hypokalemia in patients on statins and on the intensive
SBP arm, differences in the risk of SAE when comparing
intensive to standard SBP goals after stratifying by statin
use were not statistically significant. Importantly, given
the small number of SAE, we cannot presume that statis-
tical nonsignificance implies equal between-subgroup
safety profiles. Patients on statin therapy assigned to the
intensive SBP arm were at a 52% higher risk of develop-
ing AKI. Similarly, Rocco et al. [26] recently reported
that for SPRINT participants with CKD at baseline, tar-
geting an intensive SBP goal strategy compared to stan-
dard SBP goal strategy increased the risk of AKI to 64%.
Seventy-eight percent of those events consisted of AKI
stages 1 and 2 and had an overall 95.2% partial and com-
plete resolution. Importantly, the development AKI was
associated with all-cause mortality in this population. It
is thus recommended that renal function tests be closely
monitored in CKD patients when intensifying BP con-
trol. Characteristics such older age, nonwhite race, lower
baseline eGFR, and history of CVD warrant particular
attention as these have been found to confer a higher risk
for AKI.
This investigation has limitations worth mentioning.
First, this was a post hoc analysis of the SPRINT study.
Since statin use was not a randomized intervention, cau-
sality cannot be inferred, and confirmation of these find-
ings would require a dedicated experimental design. Im-
portantly, however, our primary results were reproduced
with a propensity score-matched sensitivity analysis. This
suggests that the interaction of statins with SBP interven-

304 Am J Nephrol 2019;49:297–306 Rivera/Tamariz/Suarez/Contreras

DOI: 10.1159/000499188
tion on mortality is essentially related to statin use rather
than to disproportionately distributed comorbidities and
nonbalanced mortality risks between statin subgroups.
Second, this study only assessed statin use at baseline.
Third, the SPRINT excluded patients with diabetes mel-
litus at the time of enrollment. Diabetes mellitus is an-
other major CVD risk factor and one of the most impor-
tant causes of CKD. Fourth, the small number of fatal CV
events accrued in the statin use groups does not allow
proper estimate of risk with high degree of certainty.
However, a sufficient number of all-cause fatal events
were accrued in the study that allow proper estimate of
the all-cause mortality risk.
Strengths of our study include the large CKD popula-
tion enrolled in the SPRINT, generating enough power
for the evaluation of our primary outcomes: all-cause
mortality and CV mortality. Second, the SBP interven-
tions in the SPRINT were very effective as a sustained
separation of both treatment arms was achieved early.
Third, SPRINT study clinics prospectively and system-
atically ascertained outcomes. Fourth, a blinded adjudi-
cation CVD outcome committee classified all outcomes
using pre-specified case definitions.
In summary, statin use at baseline significantly modi-
fies the risk relationship between SBP goal and both
­all-cause mortality and CV mortality for SPRINT CKD
participants. In the statin group, targeting a SBP to
­ The authors have no conflicts of interest to disclose.
<120 mm Hg compared to SBP <140 mm Hg significant-
ly reduced the incidence of all-cause mortality and CV
mortality. Medical optimization of patients with CKD
and hypertension by means of intensive SBP control and
statin therapy could also lead to major public health ben-
efits as suggested by the small number needed to treat to
prevent 1 all-cause and CV mortality event. Recent re-
ports suggest that there are about 3,032,280 individuals
with CKD not on dialysis over the age of 50 living in the
United States [27]. Consequently, the implementation of
intensive SBP control combined with statin therapy in
this patient population could potentially translate into
189,517 fewer deaths from any cause and 91,887 fewer
deaths from cardiovascular causes while also potentially
resulting in 178,370 as many events of AKI and 252,690
as many events of hypokalemia.
Acknowledgments
The views expressed in this paper are those of the authors and
do not represent the official position of the National Institutes of
Health, the Department of Veterans Affairs, the U.S. Government,
or the SPRINT Research Group.
This Manuscript was prepared using SPRINT_POP Research
Materials obtained from the NHLBI Biologic Specimen and Data
Repository Information Coordinating Center.
Ethics Statement
The authors have no ethical conflicts to disclose.
Disclosure Statement
Author Contributions
G.C. and M.R.: contributed substantially to the study design,
data analysis, interpretation, and writing of the manuscript. L.T.
and M.S.: contributed to the revision and made important intel-
lectual contributions to this manuscript. All authors ensure for the
accuracy and integrity of this work.

References
  1 Anderson AH, Yang W, Townsend RR, Pan
Q, Chertow GM, Kusek JW, et al.; Chronic
Renal Insufficiency Cohort Study Investiga-
tors. Time-updated systolic blood pressure
and the progression of chronic kidney dis-
ease: a cohort study. Ann Intern Med. 2015
Feb;162(4):258–65.  
  2 Go AS, Chertow GM, Fan D, McCulloch CE,
Hsu CY. Chronic kidney disease and the risks
of death, cardiovascular events, and hospital-
ization. N Engl J Med. 2004 Sep; 351(13):
1296–305.
  3 Muntner P, Anderson A, Charleston J, Chen  
Z, Ford V, Makos G, et al.; Chronic Renal In-
sufficiency Cohort (CRIC) Study Investiga-
tors. Hypertension awareness, treatment, and
control in adults with CKD: results from the
Chronic Renal Insufficiency Cohort (CRIC)  
Study. Am J Kidney Dis. 2010 Mar;55(3):441–
51.
4 Wright JT Jr, Williamson JD, Whelton PK,  
Snyder JK, Sink KM, Rocco MV, et al.;
SPRINT Research Group. A Randomized Tri-
al of Intensive versus Standard Blood-Pres-
sure Control. N Engl J Med. 2015 Nov;
373(22):2103–16.
5 Cheung AK, Rahman M, Reboussin DM, Cra-
ven TE, Greene T, Kimmel PL, et al.; SPRINT
Research Group. Effects of Intensive BP Con-
trol in CKD. J Am Soc Nephrol. 2017 Sep;
28(9):2812–23.
6 National Guideline C. KDIGO clinical prac-
tice guideline for the management of blood
pressure in chronic kidney disease. 2012.
7 Baigent C, Landray MJ, Reith C, Emberson J,
Wheeler DC, Tomson C, et al.; SHARP Inves-
tigators. The effects of lowering LDL choles-
terol with simvastatin plus ezetimibe in pa-
tients with chronic kidney disease (Study of
Heart and Renal Protection): a randomised
placebo-controlled trial. Lancet. 2011 Jun;
377(9784):2181–92.

Modifying Effect of Statins on Fatal Am J Nephrol 2019;49:297–306 305

Outcomes in CKD Patients in the SPRINT DOI: 10.1159/000499188
  8 Palmer SC, Navaneethan SD, Craig JC, John-
son DW, Perkovic V, Hegbrant J, et al. HMG
CoA reductase inhibitors (statins) for people
with chronic kidney disease not requiring di-
alysis. Cochrane Database of Systematic Re-
views. John Wiley & Sons, Ltd; 2014.
 9 Levey AS, Bosch JP, Lewis JB, Greene T,
Rogers N, Roth D; Modification of Diet in
Renal Disease Study Group. A more accu-
rate method to estimate glomerular filtra-
tion rate from serum creatinine: a new pre-
diction equation. Ann Intern Med. 1999
Mar; 130(6): 461–70.
10 Rosenbaum PR, Rubin DB. The central role of
the propensity score in observational studies
for causal effects. Biometrika. 1983; 70(1): 41–
55.
11 Fitzmaurice G. Confounding: regression ad-
justment. Nutrition. 2006 May;22(5):581–3.
12 Bowry SK, Schoder V, Apel C. An inadvertent
but explicable error in calculating number
needed to treat for reporting survival data. J
Am Soc Nephrol. 2014 May;25(5):875–6.
13 Lash JP, Go AS, Appel LJ, He J, Ojo A, Rah-
man M, et al.; Chronic Renal Insufficiency
Cohort (CRIC) Study Group. Chronic Renal
Insufficiency Cohort (CRIC) Study: baseline
characteristics and associations with kidney
function. Clin J Am Soc Nephrol. 2009 Aug;
4(8):1302–11.
14 Lewington S, Clarke R, Qizilbash N, Peto R,
Collins R; Prospective Studies Collaboration.
Age-specific relevance of usual blood pres-
sure to vascular mortality: a meta-analysis of
individual data for one million adults in 61
prospective studies. Lancet. 2002 Dec;
360(9349):1903–13.
15 Navaneethan SD, Schold JD, Jolly SE, Arri-
gain S, Blum MF, Winkelmayer WC, et al.
Blood pressure parameters are associated
with all-cause and cause-specific mortality in
chronic kidney disease. Kidney Int. 2017 Nov;
92(5):1272–81.
16 Whelton PK, Carey RM, Aronow WS, Casey
DE Jr, Collins KJ, Dennison Himmelfarb C, et
al. 2017 ACC/AHA/AAPA/ABC/ACPM/
AGS/APhA/ASH/ASPC/NMA/PCNA
Guideline for the Prevention, Detection,
Evaluation, and Management of High Blood
Pressure in Adults: A Report of the American
College of Cardiology/American Heart Asso-
ciation Task Force on Clinical Practice Guide-
lines. Hypertension. 2018 Jun; 71(6): 1269–
324.
17 Wanner C, Tonelli M; Kidney Disease: Im-
proving Global Outcomes Lipid Guideline
Development Work Group Members. KDI-
GO Clinical Practice Guideline for Lipid
Management in CKD: summary of recom-
mendation statements and clinical approach
to the patient. Kidney Int. 2014 Jun; 85(6):
1303–9.
18 Strazzullo P, Kerry SM, Barbato A, Versiero
M, D’Elia L, Cappuccio FP. Do statins reduce
blood pressure?: a meta-analysis of random-
ized, controlled trials. Hypertension. 2007
Apr;49(4):792–8.
19 Ge CJ, Lu SZ, Chen YD, Wu XF, Hu SJ, Ji Y.
Synergistic effect of amlodipine and atorvas-
tatin on blood pressure, left ventricular re-
modeling, and C-reactive protein in hyper-
tensive patients with primary hypercholester-
olemia. Heart Vessels. 2008 Mar;23(2):91–5.
20 Ferrier KE, Muhlmann MH, Baguet JP, Cam-
eron JD, Jennings GL, Dart AM, et al. Inten-
sive cholesterol reduction lowers blood pres-
sure and large artery stiffness in isolated sys-
tolic hypertension. J Am Coll Cardiol. 2002
Mar;39(6):1020–5.
21 Kamberi LS, Bedri Bakalli A, Muhamet Bu-
dima N, Rashit Gorani D, Karabulut AM, Ta-
lat Pallaska K. Pleiotropic and Lipid-lowering
Effects of Statins in Hypertension. Mater So-
ciomed. 2012;24(2):84–6.
22 Drapala A, Sikora M, Ufnal M. Statins, the
renin-angiotensin-aldosterone system and
hypertension - a tale of another beneficial ef-
fect of statins. J Renin Angiotensin Aldoste-
rone Syst. 2014 Sep;15(3):250–8.
23 Grundy SM, Stone NJ, Bailey AL, Beam C,
Birtcher KK, Blumenthal RS, et al. 2018 AHA/
ACC/AACVPR/AAPA/ABC/ACPM/ADA/
AGS/APhA/ASPC/NLA/PCNA Guideline on
the Management of Blood Cholesterol. Circu-
lation. 2018 Nov:CIR0000000000000625.
24 Shepherd J, Kastelein JJ, Bittner V, Deed-
wania P, Breazna A, Dobson S, et al.; TNT
(Treating to New Targets) Investigators. In-
tensive lipid lowering with atorvastatin in pa-
tients with coronary heart disease and chron-
ic kidney disease: the TNT (Treating to New
Targets) study. J Am Coll Cardiol. 2008 Apr;
51(15):1448–54.
25 Tonelli M, Wanner C; Kidney Disease: Im-
proving Global Outcomes Lipid Guideline
Development Work Group Members. Lipid
management in chronic kidney disease: syn-
opsis of the Kidney Disease: Improving Glob-
al Outcomes 2013 clinical practice guideline.
Ann Intern Med. 2014 Feb;160(3):182.
26 Rocco MV, Sink KM, Lovato LC, Wolfgram
DF, Wiegmann TB, Wall BM, et al.; SPRINT
Research Group. Effects of Intensive Blood
Pressure Treatment on Acute Kidney Injury
Events in the Systolic Blood Pressure Inter-
vention Trial (SPRINT). Am J Kidney Dis.
2018 Mar;71(3):352–61.
27 Saran R, Robinson B, Abbott KC, Agodoa LY,
Bhave N, Bragg-Gresham J, et al. US Renal
Data System 2017 Annual Data Report: Epi-
demiology of Kidney Disease in the United
States. Am J Kidney Dis. 2018 Mar; 71(3S1
Suppl 1):A7–8.

306 Am J Nephrol 2019;49:297–306 Rivera/Tamariz/Suarez/Contreras

DOI: 10.1159/000499188