Intensive Blood-Pressure Control in Patients with Type 2 Diabetes (BPROAD)

Authors, Journal, Affiliations, Type, DOI

• Authors: Y. Bi, M. Li, Y. Liu, T. Li, J. Lu, P. Duan, and the BPROAD Research Group; senior authors G. Ning, J. He, Y. Xu, W. Wang
• Journal: New England Journal of Medicine, 2025;392:1155–67
• Affiliations: Ruijin Hospital, Shanghai Jiao Tong University School of Medicine; National Clinical Research Center for Metabolic Diseases (Shanghai); 145 clinical sites across China; J. He: UT Southwestern Medical Center, Dallas
• Type: Multicenter, parallel-design, open-label randomized controlled trial with blinded outcome assessment; 145 sites in 7 geographic regions of China
• DOI: https://doi.org/10.1056/NEJMoa2412006

Overview

The BPROAD (Blood Pressure Control Target in Diabetes) trial enrolled 12,821 patients aged ≥50 with type 2 diabetes and elevated systolic BP at increased cardiovascular risk across 145 Chinese sites. Patients were randomized to intensive treatment (SBP <120 mmHg) vs standard treatment (SBP <140 mmHg) for up to 5 years. At median 4.2-year follow-up, the intensive arm achieved mean SBP 121.6 vs 133.2 mmHg. The primary composite (nonfatal stroke/MI, HF treatment/hospitalization, CV death) occurred in 393 vs 492 patients (HR 0.79; 95% CI 0.69–0.90; P<0.001), a 21% relative risk reduction. Serious adverse events were equivalent, though symptomatic hypotension and hyperkalemia were more frequent with intensive treatment.

Keywords

Type 2 diabetes, systolic blood pressure, intensive treatment, standard treatment, cardiovascular events, blood pressure target, BPROAD, China, hypertension, stroke

Key Takeaways

Background and Context

• Hypertension is the most common comorbidity in T2DM and the most modifiable CVD risk factor in this population
• ACCORD trial (2010): Compared SBP <120 vs <140 mmHg in T2DM; primary outcome neutral (HR 0.88; 95% CI 0.73–1.06; P=0.20). Critically underpowered — actual event rate 2.09%/yr vs assumed 4%/yr. Further confounded by factorial glucose intervention: intensive BP lowering only reduced CVD risk in the standard (not intensive) glycemic control arm (P=0.08 for interaction)
• SPRINT trial (non-diabetic patients): same SBP targets; major CVD events HR 0.73 (95% CI 0.63–0.86) — significant benefit
• ESPRIT trial (Lancet 2024): mixed T2DM/no-diabetes; HR 0.88 (0.78–0.99) for major vascular events overall; T2DM subgroup alone HR 0.91 (NS)
• Most current guidelines recommend SBP <130 mmHg in T2DM; evidence for <120 mmHg was lacking

Methods

• Population: ≥50 years, T2DM, elevated SBP (130–180 mmHg on antihypertensives or ≥140 mmHg off), plus ≥1 of: prior clinical CVD ≥3 months pre-enrollment, subclinical CVD within 3 years, ≥2 CVD risk factors, or CKD (eGFR 30–<60)
• Randomization: Block randomization stratified by site via web-based central system
• Targets: Intensive SBP <120 mmHg vs standard SBP <140 mmHg; treatment algorithm mirrored SPRINT
• Follow-up schedule: Monthly first 3 months, then every 3 months if target achieved; home BP monitoring during COVID-19
• Primary outcome: Composite of first nonfatal stroke + nonfatal MI + treatment/hospitalization for HF + CV death
• Secondary outcomes: Fatal/nonfatal stroke; fatal/nonfatal MI; HF treatment/hospitalization; CV death; all-cause death; expanded composite; CKD progression/development; incident albuminuria
• Analysis: ITT principle; Cox proportional-hazards; multiple imputation for missing outcomes (50 datasets); competing-risk analysis (Fine and Gray); prespecified subgroups: age, sex, prior CVD, prior CKD, SBP, HbA1c, diabetes duration, hypertension duration
• Power: 90% power at 2.0% event rate/yr; enrolled 12,821 (target 12,702)

Blood Pressure Achieved

• Baseline SBP ~140.0 mmHg in both groups
• At 1 year: mean SBP 121.6 mmHg (median 118.3) intensive vs 133.2 mmHg (median 135.0) standard
• Sustained between-group separation throughout follow-up
• ~60% of intensive-arm patients met SBP <120 mmHg after year 1
• More antihypertensive drugs used in intensive arm; HbA1c (mean 7.6% both groups at 48 months), lipids, BMI similar throughout

Clinical Outcomes

• Primary composite: HR 0.79 (95% CI 0.69–0.90; P<0.001) — 393 vs 492 events; 1.65 vs 2.09 per 100 person-years; Kaplan-Meier separation apparent after 1 year
• Fatal/nonfatal stroke: HR 0.79 (95% CI 0.67–0.92); 284 vs 356 events (1.19 vs 1.50/100 py) — primary driver of benefit
• Fatal/nonfatal MI: Similar in both groups
• HF treatment/hospitalization: Similar in both groups
• CV death: Similar in both groups
• All-cause mortality: HR 0.95 (95% CI 0.77–1.17) — NS; 169 vs 179 events
• Incident albuminuria: HR 0.87 (95% CI 0.77–0.97) — 554 vs 648 events (11.29 vs 13.84/100 py)
• CKD progression or development: Similar in both groups
• Consistent benefit across all prespecified subgroups (age, sex, prior CVD, prior CKD, SBP, HbA1c, DM duration, HTN duration)
• Sensitivity analyses (censored missing data; non-missing-at-random assumptions; competing-risk) all concordant

Safety

• Serious adverse events: 36.5% vs 36.3% (HR 1.00; 95% CI 0.94–1.06; P=0.96) — no difference
• Symptomatic hypotension: 0.1% (8/6,414) vs <0.1% (1/6,407); P=0.05
• Hyperkalemia (K >5.5 mmol/L): 2.8% (177/6,230) vs 2.0% (125/6,220); P=0.003
• Other conditions of interest and adverse event rates similar between groups

Why ACCORD Failed and BPROAD Succeeded

• ACCORD event rate was 2.09%/yr (actual) vs 4%/yr (assumed) — reduced power to detect true difference; BPROAD assumed 2.0%/yr and achieved 2.09%/yr exactly
• ACCORD factorial design: subgroup showed intensive BP lowering only benefited standard glycemic control arm; BPROAD used standard glycemic control per guidelines (mean HbA1c 7.6% both groups, matching ACCORD standard-glycemia subgroup)
• BPROAD 2.7× larger than ACCORD BP subcomponent; full statistical power achieved

Limitations of the Document

• Open-label design: patients and physicians unblinded (outcome assessors blinded)
• Only ~60% of intensive-arm patients achieved SBP <120 mmHg — true effect of full target attainment may be larger
• Home/telephone BP monitoring during COVID-19 pandemic (percentage of telephone visits balanced between groups)
• Diastolic BP also differed markedly between groups — independent effect of SBP vs DBP reduction cannot be fully isolated
• Single-population limitation: conducted entirely in Chinese patients; generalizability to other ethnic populations uncertain
• CKD follow-up (4.2 years) may be insufficient to detect long-term nephroprotective benefit beyond albuminuria reduction
• Hyperkalemia monitoring essential when multiple antihypertensives (especially RAASi) are combined

Key Concepts Mentioned

• concepts/Blood-Pressure-Target-T2DM — central concept of this trial; intensive vs standard SBP targets in T2DM
• concepts/ASCVD-Risk-Assessment — CVD risk definition used for patient eligibility

Key Entities Mentioned

• entities/Hypertension — central entity; BPROAD updates the diabetes comorbidity section with first RCT evidence for SBP <120 mmHg
• entities/Type-2-Diabetes — primary study population

Wiki Pages Updated

• wiki/sources/bp-dm-bproad-nejm-2025.md — created
• wiki/concepts/Blood-Pressure-Target-T2DM.md — created
• wiki/entities/Type-2-Diabetes.md — created (stub with cardiovascular risk section)
• wiki/entities/Hypertension.md — diabetes comorbidity and contradictions sections updated
• wiki/sourceindex.md — BPROAD entry added at top
• wiki/wikiindex.md — Blood-Pressure-Target-T2DM concept entry added at top