#NephJC Chat
Tuesday Sept 10th, 2024 at 9pm Eastern (AEST = Sept 11th, 11am)
Wednesday Sept11th, 2024, at 9pm Indian Standard Time and 3:30 pm GMT (AEST = Sept 12th, 2am)
Lancet. 2024 Jul 20;404(10449):245-255. doi: 10.1016/S0140-6736(24)01028-6. Epub 2024 Jun 27.
Lowering systolic blood pressure to less than 120 mm Hg versus less than 140 mm Hg in patients with high cardiovascular risk with and without diabetes or previous stroke: an open-label, blinded-outcome, randomized trial
PMID: 38945140
Introduction
What is the “ideal” blood pressure that minimizes cardiovascular and end-organ damage? Despite years of research, this remains a contentious issue in hypertension management.
Studies like ACCORD (Cushman et al, NEJM 2010) and SPRINT (Wright et al, NEJM 2015) have tried to prove the benefits of intensive blood pressure treatment. In the ACCORD trial, which focused only on patients with diabetes, the intensive treatment did not reduce overall cardiovascular events but lowered stroke risk with a higher incidence of adverse effects, including a non-significant increase in mortality. A subsequent post hoc study (Beddhu et al JAHA 2018) reported that this increased mortality was only seen in the intensive glycemic arm, and there was a mortality benefit with lowering BP in the standard glycemic arm, but by then the damage was done. In contrast, the SPRINT trial which targeted patients at high cardiovascular risk without diabetes, reported broader cardiovascular and survival benefits. Both fell short of providing generalizable results because of limitations of their study populations and, some might mistakenly add, the methods used to measure blood pressure. Among these, the SPRINT trial stands out as the only trial that offered compelling evidence that targeting systolic blood pressure <120 mm Hg was more effective in reducing the risk of major cardiovascular events (Wright et al, NEJM 2015| NephJC Summary). This, in turn, led to an influence on clinical practice guidelines, especially those from AHA (but not so much on the ESC guidelines). Also see: NephJC summary of AHA guidelines and KDIGO BP guidelines.
Table comparing 2017 AHA/ACC guidelines (Whelton et al, Hypertension 2017) with 2024 ESC guidelines (McEvoy et al, EHJ 2024)
Why do we need another BP target trial? Clearly as shown with the interpretation, not everyone agrees with the generalizability of the SPRINT/ACCORD results, particularly in patients with diabetes. The meticulous BP measurement in these trials was labeled as an unusual ‘unattended’ method by mischievous and misinformed experts rather than just a meticulous way of ensuring a proper resting BP (see Johnson et al, Hypertension 2018). The withdrawal of BP medications to ensure adequate BP separation between arms, and the (ethically appropriate) early stoppage of SPRINT was used by EBM hardliners to wish away the demonstrated mortality benefit. Possibly there is an element of less-is-more therapeutic nihilism thrown in. Nevertheless, it is not unreasonable to ask for replication in this age of replication crisis.
Against this backdrop, the Effects of Intensive Systolic Blood Pressure Lowering Treatment in Reducing Risk of Vascular Events (ESPRIT) study was designed (Liu et al Am Heart Jour 2023) to compare outcomes in patients with intensive versus standard blood pressure treatment to assess effects on major vascular events in a high-risk population, regardless of history of diabetes or stroke across communities in China.
The Study
Participants
ESPRIT was an open-label, blinded-outcome, randomized controlled trial conducted at 116 sites (103 hospitals and 13 community medical centers) in China. Participants who were at least 50 years old and with two consecutive SBP measurements of 130–180 mmHg were considered to be eligible if they had high cardiovascular risk factors including previous ischemic heart disease or stroke, revascularisation, peripheral artery disease, abdominal aortic aneurysm ≥5 cm and at least two of the following: Age >60 years for men or >65 years for women, diabetes, dyslipidemia and current smoker.
ESPRIT inclusion and exclusion criteria, table from Liu et al. Am Heart J, 2023.
Known secondary causes of hypertension, one minute standing SBP < 110 mmHg, eGFR < 45 ml/min/1.73m², proteinuria ≥2+, left ventricular ejection fraction <35%, and any organ transplant were the major exclusion criteria. (Sigh, renalism strikes again).
The eligible participants were allocated to either intensive treatment (SBP target <120 mmHg) or standard treatment (SBP target <140 mmHg) in a 1:1 ratio. A minimized randomization program was used where imbalance scores were calculated with each new participant for each arm based on stratification, and the participant was assigned to the treatment arm, which yielded the lowest imbalance.
BP Measurement
At each study visit, BP was measured by a trained physician or nurse (observed manner) using an electronic BP monitor (Omron HBP-1100; Omron Corp). Each participant was measured 3 times with an interval of 1 minute according to the standard procedure. Before the measurement, participants were required to rest in a seated position for at least 5 minutes in a quiet room. The mean value of the 3 measurements was used as the BP level. As mentioned below, during the pandemic, some BP measurements were home BP, not office BP.
Interventions
After randomization, participants in both groups were followed up at months one, two, and three and then every three months for three years. The participants’ antihypertensive medications were adjusted based on standard office blood pressure measurement and the study-group assignment with the guidance of unified treatment algorithms, which can be consulted in the supplementary material (Supp figures S1 and 2) which includes down titration of BP meds similar to SPRINT in control arm. The use of medication was according to recommendations of hypertension guidelines, and five classes of evidence-based antihypertensive medications were provided (ie, ACEIs, ARBs, CCBs, thiazide-type diuretics, and beta-blockers). When participants could not come to the study clinic in person, they were followed by telephone. This was particularly important because the study coincided with the COVID pandemic in 2020. There is no mention of standardization of blood pressure measurement during these telephone follow-ups, and titration of medication was interrupted, causing a 3-month delay in achieving BP targets in the intensive treatment group.
Outcomes
Primary Outcome = Major vascular events: a composite of myocardial infarction, coronary or non-coronary revascularisation, hospitalization, or emergency room visit for heart failure, stroke, or death from cardiovascular causes.
Secondary Outcomes = Included components of the primary composite outcome, death from any cause, a composite of the primary outcome and death from any cause, and composite kidney outcome (ie, end-stage renal disease, a sustained decline in eGFR to <10 ml/min/1.73m², death from renal causes, or a sustained decline ≥40% in eGFR from baseline).
Statistical Analysis and Sample Size Calculation
The hypothesis was that a target SBP of <120 mmHg would reduce the rate of major CV events compared to the strategy targeted at an SBP of <140 mmHg. Presuming an annual event rate of 3.4% and a 2% loss of follow-up per year, they estimated that at least 10,300 participants would provide 90% power at two-sided p 0.05 to detect a 20% difference in the primary outcome risk between groups. All analyses were based on the intention to treat principle with Kaplan Meir estimate for the time to the primary outcome. Cox proportional hazard regression was used for treatment comparison, estimating hazard ratios (HRs) and 95% confidence intervals (CIs). Sensitivity analyses were performed. The Fine Gray model was used to compare risk analyses for non-cardiovascular death. Post-hoc sensitivity analysis accounted for undetermined deaths and CV outcomes.
Supplementary Table 2. Sample size calculation, from Liu et al. Lancet, 2024.
Regarding secondary outcomes, CIs were not adjusted for multiplicity, hence, secondary outcome results should not be over-interpreted as hypothesis tests, because of the increased risk of type I errors.
Funding
The study was funded by The Ministry of Science and Technology of China and Fuwai Hospital. They had no role in designing the study, data collection, analysis, and interpretation, or writing of the report.
Results
A total of 16,329 participants were assessed for eligibility from September 17, 2019, to July 13, 2020. Of these, 11,255 enrolled and were randomly assigned, and 127 discontinued the intervention in the intensive group compared to 35 in the standard treatment group. Three participants in each of the intensive and standard treatment groups were lost to follow-up. The median duration of follow-up was 3.4 years.
Figure 1. Trial profile, from Liu et al. Lancet, 2024.
Randomized participants had a mean age of 64.6 years, and around 24.5% were above 70. Women represented 41% of the randomized population. Co-morbid conditions were common and included 39% of participants with a history of diabetes, 29% with coronary heart disease, and 27% with stroke. The mean eGFR was about 83 ml/min per 1.73 m² and only 6% of the participants had eGFR between 45 to 60 ml/min/1.73m² at baseline. The average number of baseline BP medications was 1.7 for both groups.
Table 1. Baseline characteristics, from Liu et al. Lancet, 2024.
The mean baseline SBP overall was about 147 mmHg. After the first 3 months of up-titration and down-titration by applying a treatment algorithm similar to the SPRINT trial, throughout the follow-up, mean SBP in the intensive treatment group was 119.1±11.1 mmHg and 134.8±10.5 mmHg in the standard treatment group. The intensive treatment group achieved a stable target SBP of <120 mmHg after 9 months of follow-up, mainly because the COVID-19 pandemic interrupted the uptitration of blood pressure medications. The standard treatment group achieved the target after only 2 months. More antihypertensive medications were used in the intensive treatment group (2.7±1.0) than the standard treatment group (2±0.9). Among the five classes of evidence-based antihypertensive medications, the use of diuretics at the final follow-up was reported to be higher in the intensive treatment group (42.5%) than in the standard treatment group (15.4%).
Figure 2. Systolic blood pressures in the two treatment arms, from Liu et al. Lancet, 2024.
Primary Outcomes
The primary outcome occurred in 9.7% of participants in the intensive treatment group compared to 11.1% in the standard treatment group (HR 0.88; 95% CI 0.78-0.99). The risk difference was evident after 1 year in the exploratory analysis and the HR of 0.78 (95% CI 0.67-0.90) for the period longer than 1 year. To prevent a primary outcome event 75 patients need to be treated with intensive treatment for 3 years (see the NephJC stats overview for why such use of NNT is problematic).
Figure 3. Primary outcomes, from Liu et al. Lancet, 2024.
Having in mind that the study was not powered for individual components of the primary outcome, death from cardiovascular effects was lower in the intensive treatment group (1.1%) compared to the standard treatment group (1.7%), with a HR of 0.61 (95% CI 0.44 - 0.84). The risks of death from any cause (HR 0.79, 0.64 - 0.97) and composite of primary outcome or death from any cause (HR 0.89, 0.80 - 0.99) were lower in the treatment group. The other individual components were not statistically significant, but most of the point estimates (except coronary revascularizations) aligned on the beneficial side.
Figure 4. Primary and secondary outcomes, from Liu et al. Lancet, 2024.
Secondary Outcomes
The composite kidney outcome occurred in 169 (3%) participants in the intensive treatment group and 102 (1.8%) from the standard treatment group (HR 1.70, 95% CI 1.33-2.17), primarily driven by a sustained decline in eGFR more or equal to 40% from baseline. In the intensive treatment group, 3 participants had sustained decline in eGFR to a value less than 10 ml/min/1.73 m², and one developed end-stage kidney disease. In a post-hoc sensitivity analysis excluding participants whose eGFR either improved to 60 or higher, or declined only at the last follow-up, the relative risk of the intensive group increased compared to the main analysis, though the absolute risk was lower.
Table S14. Kidney outcome, from Liu et al. Lancet, 2024.
Subgroup analyses
Prespecified subgroups for the primary outcome included diabetes status, duration of diabetes, history of stroke, CAD, and others. Two subgroups, including previous cardiovascular disease and eGFR of less than 60 ml/min per 1.73 m², were also analyzed. Analysis of both prespecified and non-prespecified subgroups did not find heterogeneity in the effect of the intensive treatment on the primary outcome (Supp figure S4).
Safety Outcomes
Both the intensive treatment group and the standard treatment group showed a similar incidence of serious adverse events, 42.1% and 42.2% respectively (HR 1.01, 95% CI 0.95-1.07). The incidence of syncope was higher among participants in the intensive treatment group (0.4%) than in the standard treatment group (0.1%), as expected. Hypotension caused syncope in five participants in the intensive treatment group and two in the standard treatment group. There was no statistically significant difference between the groups regarding hypotension, electrolyte abnormality, injurious fall, or acute kidney injury.
Table 2. Adverse events, from Liu et al. Lancet, 2024.
Discussion
The ESPRIT trial in a cohort of older Chinese individuals with preserved kidney function, but including diabetes, achieved a nice separation of BP and demonstrated a cardiovascular and mortality benefit with intensive BP lowering, thus providing external validation of the SPRINT results. It also provides a stronger extension of intensive BP lowering in Diabetes than what the post hoc ACCORD analysis did.
Key Strengths
The ESPRIT trial showed a significant decrease in major vascular events in the intensive BP treatment group compared to the standard BP group in patients with hypertension at high cardiovascular risk, regardless of history of diabetes or stroke, with few serious adverse events compared to previous trials. The study's main strengths include a large sample size from 116 sites, including hospitals and primary health care institutions, high adherence to intervention; few participants lost to follow-up, and a large number of clinical outcomes.
Limitations
This trial included only patients of Chinese origin. An important consideration arises regarding the generalizability of the trial results to diverse ethnic populations, but is quickly irrelevant once we consider this in totality with SPRINT. One should not generalize these results to those populations that were excluded, in particular those with standing BP <110, and those with a limited life expectancy (which was similar for SPRINT, which also excluded those who were wheelchair-bound and institutionalized). If the participants did not tolerate systolic BP <120 mmHg in the intensive arm, they maintained the lowest tolerable systolic BP limit. The trial had a short follow-up duration, affected by the COVID-19 pandemic. Patients with eGFR below 45 ml/min per 1.73 m² and left ventricular ejection fraction less than < 35% were excluded, both subgroups of interest to nephrologists. However, hypertension is less commonly an issue in HFrEF, though the evidence lacuna for BP targets in advanced CKD (particularly GFR <20, which was the SPRINT exclusion) is sorely felt.
ESPRIT in Perspective: Comparing with Previous Hypertension Trials
Comparisons between ESPRIT, SPRINT, and ACCORD trials are inevitable because the trials examined identical systolic blood pressure targets (<120 mmHg vs <140 mmHg) in patients with high risk for cardiovascular events. While ACCORD exclusively enrolled diabetic participants, SPRINT excluded those with diabetes and stroke, and ESPRIT included individuals with a history of both diabetes and stroke. STEP (Zhang et al, NEJM 2021) is another trial of interest as it was done in a Chinese hypertensive elderly population with a slightly higher blood pressure goal (intensive BP arm: 110-130 mmHg vs standard BP arm: 130-150 mmHg). Let us look again at the inclusion and exclusion criteria for these trials:
Intensive versus standard BP control across RCTs
ESPRIT was the largest trial, with a sample size of 11,255 patients. The mean age varied in these trials, with SPRINT having an older cohort mean age. 28% of participants in SPRINT were 75 years or older, while 24.3% of participants in ESPRIT were 70 years or older, and STEP had 24.1% of patients older than 70.
Intensive versus standard BP control across RCTs
SPRINT used unattended office BP measurements, which led to arguments that the measurements presumably led to lower blood pressure values than those observed in other trials. ESPRIT used attended standard office BP measurement to negate this argument. Nevertheless, during the COVID-19 pandemic, we are assuming that in follow-up visits done by telephone the BP measurements were done at home, which was an alteration of the established protocol. The potential impact of COVID-19 on the trial was evaluated by dividing participants into two subgroups: those who were randomized before April 2020, and those who were randomized after. They compared baseline characteristics and the process of BP-lowering titration. It took longer for those affected by COVID-19 to reach the BP target in the first three months, but the proportions of reaching BP targets at 6 months were similar in both groups. No other analysis was made.
ESPRIT trial succeded in attaining mean systolic blood pressure below 120 mmHg (119.1 mmHg) in the intensive treatment group, unlike the SPRINT trial where the mean systolic blood pressure was 121.5 mmHg in the intensive treatment group. There was a good separation between the mean systolic blood pressure in the two arms (119.1 mmHg vs 134.8 mmHg) after 3 months. They applied a treatment algorithm similar to the SPRINT trial and down titrated medications in the standard treatment arm if the systolic BP was lower than 130 mmHg at a single visit or 135 mm Hg at 2 consecutive visits.
Intensive versus standard BP control across RCTs
SPRINT showed a 25 % risk reduction in the primary outcome and a 43 % risk reduction in cardiovascular death. ESPRIT relative risk reduction is smaller than SPRINT (HR 0.88, 95% CI 0.78-0.99). Major cardiovascular events excluding revascularisation showed a more pronounced risk reduction in ESPRIT (HR 0.84, 95% CI 0.74-0.96). Proportional risk reduction appears to be larger with longer follow-up, as seen in SPRINT and ACCORD trials. Interestingly the ESPRIT trial showed a 22 % risk reduction in primary outcome at one year with a 15 mm Hg difference in the systolic BP.
The ESPRIT trial (38.7% of participants had diabetes, 4359 participants) showed a reduction in major vascular events in the intensive BP arm. Subgroup analysis did not show any change in primary event incidence with history or duration of diabetes. The ACCORD trial showed no benefit in targeting intensive blood pressure control (systolic BP <120 mmHg vs <140 mmHg) in reducing the composite CV outcome. Beddhu et al, compared the effects of intensive SBP lowering within each of the glycemia arms in ACCORD BP. Intensive SBP lowering decreased the composite CVD endpoint similarly in SPRINT (HR 0.75, 95% CI 0.64–0.89) and in the ACCORD BP standard glycemia arm (HR: 0.77, 95% CI 0.63–0.95) - but not in the intensive glycemic arm (Beddhu et al, JAHA 2018). An individual participant level data meta-analysis showed that participants with diabetes had weaker relative risk reduction than those without, but similar absolute risk reduction (Nazarzadeh et al, Lancet Diabetes and Endocrinology 2022). ESPRIT results align with this meta-analysis findings and reinforce the role of intensive BP lowering even in patients with diabetes.
Tailoring Blood Pressure Targets in CKD
Effective management of high blood pressure is a priority to optimize cardiovascular health (not renal health) in patients with chronic kidney disease. Unlike SPRINT, which included patients with eGFR down to 20 ml/min/1.73m², ESPRIT excluded patients with eGFR <45 ml/min/1.73m² and the mean baseline GFR was 83. Hence this trial does not provide useful information in managing BP in advanced CKD.
The ESPRIT trial demonstrated an increased risk of sustained (40%) renal functional decline. The interpretation of kidney outcomes in the ESPRIT trial is limited, as kidney outcome events were very small overall and were predominantly driven by a sustained decline in eGFR more or equal to 40% from baseline. This necessitates a cautious interpretation of kidney outcomes in this trial. Major nephrology trials (ie, MDRD, AASK and REIN 2) have not reported a renal benefit with intensive BP lowering in nondiabetic CKD. In a post hoc analysis of SPRINT, no increase in urine tubular injury biomarkers in the intensive versus standard arm were noted (Malhotra et al, AJKD 2019). Thus the initial eGFR decline observed with intensive BP lowering likely represents reversible hemodynamic changes rather than intrinsic kidney damage. This may be applicable to the ESPRIT trial as well.
Translating Results Into Clinical Practice
As with any trial, generalizability remains key. Measuring BP properly (attended or unattended!) has to precede any efforts for BP management. One should be careful in extrapolating these results beyond the populations in whom these trials were conducted(e.g. limited life expectancy and standing SBP < 110) - though in the totality of evidence, diabetes and the elderly populations do also deserve intensive BP lowering.
Conclusion
Expanding on the SPRINT trial, ESPRIT strengthens the case for intensive BP control in most hypertensive individuals, including those with a history of diabetes and stroke. Within one year of intensive systolic blood pressure control, there were notable benefits to vascular health and mortality.
Summary by
Oscar R. Durón
Nephrologist, Asturias, Spain
Shahin Mohammed
Nephrologist, Kerala, India
NSMC interns Class of 2024
Reviewed by Brian Rifkin, Cristina Popa, Husam Alzayer,
Jade Teakell and Swapnil Hiremath
Header Image created by AI, based on prompts by Evan Zeitler