Blood Pressure Targets in Chronic Kidney Disease: An Update on the Evidence

Dominique Guerrot; Jelmer K. Humalda


Curr Opin Nephrol Hypertens. 2020;29(3):327-332. 

In This Article

Evidence for Blood Pressure Targets to Reduce Cardiovascular Disease and Mortality in Patients With Chronic Kidney Disease

Hypertension and kidney disease are closely linked. Regardless of the type of nephropathy, the pathophysiology of hypertension in CKD is multifactorial and can involve excessive fluid retention, progressive arterial stiffness, activation of the sympathetic and renal-angiotensin-aldosterone nervous systems, and disruption of circadian rhythms. On average 86% of patients with CKD have hypertension, compared to 29% of the general population,[3,5,6] and the more the GFR decreases, the more the prevalence of hypertension increases.[2] Multiple studies have demonstrated that hypertension is a major cause of cardiovascular mortality and morbidity in CKD.[7–9] The management of hypertension is particularly challenging in patients with CKD and frequently requires a combination of several drug classes and nonpharmacological interventions.[8]

In spite of the assumed benefit of BP control in CKD, interventional studies evaluating different BP targets are scarce in this specific population. In a study of 400 000 patients treated for hypertension, compared to the group reaching a target SBP of 130 to 140 mmHg, patients between 120 and 130 mmHg experienced more dialysis and death.[10] The J-curve described in the general population has been demonstrated in CKD as well. A large observational study conducted in 651 000 patients with CKD (mean eGFR 50.5 mL/min/1.73m2) showed such an association between mortality and systolic and diastolic BP, with excess mortality for DBP less than 70 mm Hg, regardless of the value of SBP (except for SBP > 180 mmHg). All-cause mortality was highest in the CKD group with SBP more than 160 mmHg and was lowest in the group of patients with BP of 130–149/70–89 mmHg.[11] Similarly, a post-hoc analysis of the IDNT study performed in 1590 patients with diabetic nephropathy demonstrated an increased risk of cardiovascular mortality, all-cause mortality and heart failure in association with SBP less than 120 mmHg and DBP less than 85 mmHg under treatment.[12]

Findings From the SPRINT Trial

The main RCTs including patients with CKD are summarized in Figure 1. To date, the RCT designed to compare BP targets which included the largest number patients with CKD is the SPRINT trial.[13] In SPRINT, 9361 patients without diabetes with an increased cardiovascular risk were assigned to intensive BP control or standard treatment, aiming at SBP less than 120 and less than 140 mmHg, respectively. The trial's data safety monitoring board halted the trial after 3.26 years of the intended 5 years follow-up because of clear benefit in the intensive-treatment arm, that reached a SBP of 121.4 mmHg achieved with on average one additional antihypertensive drug. The intensive-treatment arm demonstrated a 25% lower risk of the composite outcome myocardial infarction, acute coronary syndrome, stroke, acute decompensate heart failure, or cardiovascular death (hazard ratio [HR] = 0.75; confidence interval [CI], 0.64–0.89, P < 0.001). This was mainly driven by heart failure (HR = 0.62; CI, 0.45–0.84) and cardiovascular death. The HR for death of all causes was 0.73 (CI, 0.60–0.90, P = 0.003). A post-hoc analysis by Obi et al. reported that intensive-treatment in participants with eGFR less than 45 ml/min/1.73 m2 did not improve the composite outcome (HR = 0.92; CI 0.62–1.38). Actually, in SPRINT 28.2% of the participants had a baseline MDRD eGFR between 20 and 60 ml/min/1.73 m2. Cheung et al.[14,15] found that in this subgroup of patients with CKD, intensive-treatment lowered the risk of death by 28% (HR = 0.72; CI, 0.53–0.99). Heart failure (41 vs. 52 events) and cardiovascular death (18 vs. 30 events) had the largest difference in incidence between treatment groups, nevertheless the composite outcome did not reach significance (HR = 0.81; CI 0.63–1.05).[15] Kjeldsen et al.[16] expressed concern that increase use of diuretics and thus inadvertent treatment of heart failure in the intensive-treatment group may have driven these results. They also pointed that SPRINT was a methodological exception among hypertension RCTs, because BP was evaluated by automated BP measurement which yields significantly lower values compared with manual office BP measurement.[17] This is a major point, which partly explains different interpretations and hence different guidelines regarding BP targets, both in the general and the CKD populations following the publication of SPRINT.[18] Moreover, SPRINT excluded patients with diabetes (and thus diabetic nephropathy) and proteinuria more than 1 g/day, polycystic kidney disease and glomerulonephritis (about to be) treated with immunosuppressive therapy, limiting generalization to a nephrology outpatient clinic. Of particular importance, whether the benefits observed in SPRINT apply to patients with stage four to five CKD or albuminuria more than 1 g/g is unclear.

Figure 1.

Cardiovascular and renal outcomes of the major randomized controlled trials (RCTs) comparing blood pressure (BP) targets and including patients with chronic kidney disease (CKD).

Findings From Other Blood Pressure Trials in Chronic Kidney Disease

Strolling away from SPRINT, a recent analysis of the AASK and MDRD trials by Ku et al. described that strict BP control was mainly associated with lower risk of death in the eGFR below 30 ml/min/1.73 m2 subgroup.[19] The 19-year follow-up of the MDRD study shows a long-term benefit on the risk of death from any cause, whether the patients are on dialysis or not.[20] BP targets were defined as mean arterial pressure (MAP) less than 92 mmHg in AASK. In addition, targets varied in MDRD by age, translating to targets of less than 125/75 mmHg in patients less than 61 years old or less than 135/80 mmHg in patients more than 61 years in intensive-treatment, versus 140/90 or 160/90 in normal control.

Notwithstanding the aforementioned uncertainty what BP should be targeted to limit cardiovascular disease in CKD, recent publications offer valuable information on how to achieve these targets, that's is how to interpret creatinine rise or orthostatic hypotension during treatment, how to deliver optimal treatment without increasing drugs by timing or dietary interventions. A reduction in GFR may occur acutely after intensification or initiation of antihypertensive therapy. In the combined AASK/MDRD dataset, Ku et al. found that a reduction in eGFR of 5–20% in the first 4 months of intensive-treatment was actually associated with less death (HR = 0.77; CI, 0.62–0.96) in adjusted analysis compared to the usual treatment group with less than 5% reduction. This mirrors findings in other CKD trials were acute drops after treatment intensification are associated with favorable outcome in the long term, for example ACEi or SGLT2i.[21,22] A separate analysis of the AASK found that pursuing the intensive-treatment BP goal did not result in more orthostatic hypotension compared to standard treatment,[23] in line with earlier findings in patients with diabetes in ACCORD[24] and SPRINT.[13] Orthostatic hypotension itself was associated with cardiovascular disease and all-cause mortality, and more prevalent when β blockade instead of ACEi was used.

Findings From Nonpharmacological Intervention Studies

A nonpharmaceutical intervention, changing timing instead of increasing pill count, may offer benefits as suggested by the Hygia Chronotherapy Trial. Indeed, Hermida et al. reported that in 19 084 hypertensive patients (diagnosed with ambulatory BP monitoring, ABPM) randomized to take their BP lowering medications at bedtime or after awakening, bedtime-treatment resulted a 45% lower risk (HR = 0.55; CI, 0.50–0.61] of reaching the composite cardiovascular endpoint during 6.3 years of follow-up.[25] This endpoint consisted of myocardial infarction, coronary revascularization, heart failure, stroke, and cardiovascular death. 29.4% of the participants had CKD, they demonstrated similar risk reduction of bedtime-treatment compared to participants without CKD. ABPM demonstrated lower night time SBP (114.7 vs. 118 mmHg), however without an adverse effect on day time SBP (129.5 vs. 129.2), resulting in lower average ABPM SBP. Replication of the findings of this study in other cohorts would further strengthen these interesting results. Sodium restriction contributes to BP and proteinuria control in patients with CKD, as reviewed earlier.[26] Adherence is challenging. We recently published results of the SUBLIME trial, which assessed effects of a web-based self-management intervention (including e-coaching and group meetings; Humalda et al.,[27]). In the SUBLIME intervention group, sodium intake decreased from 188 to 148 mmol/day, and a decrease of SBP from 140 to 132 mmHg in 3 months. After 6 months of maintenance phase, SBP remained 132 mmHg, but because of a concomitant modest decrease in sodium intake and BP in the control group (possibly Hawthorne-effect, awareness of participation in a trial influences behaviour), the between group difference was formally not significant.