Series: Landmark Trials in Endocrinology & Metabolism | Study #21
Category: Type 2 Diabetes · Blood Pressure | Design: Multicentre, randomised controlled trial (within UKPDS) | n: 1,148 | Follow-up: 8.4 years (median)
📋 Summary
Authors: UK Prospective Diabetes Study Group
Journal: BMJ 1998;317:703–713 | PMID: 9732337
UKPDS 38 was the blood pressure arm of the UKPDS, which enrolled 1,148 hypertensive patients with type 2 diabetes (mean entry blood pressure 160/94 mmHg; mean age 56 years) from 20 hospital-based clinics across the United Kingdom. Participants were randomised to tight blood pressure control aiming at below 150/85 mmHg, using captopril or atenolol as the primary agent, or to less tight control aiming at below 180/105 mmHg without use of ACE inhibitors or beta-blockers, with a median follow-up of 8.4 years. Mean blood pressure during follow-up was 144/82 mmHg in the tight control group versus 154/87 mmHg in the less-tight group. Tight blood pressure control reduced diabetes-related endpoints by 24% (8 to 38%; p=0.0046), deaths related to diabetes by 32% (6 to 51%; p=0.019), strokes by 44% (11 to 65%; p=0.013), and microvascular endpoints by 37% (11 to 56%; p=0.0092), with the microvascular benefit driven predominantly by a reduced rate of retinal photocoagulation. After nine years of follow-up, tight control also reduced retinopathy progression by 34% (11 to 50%; p=0.0004) and visual acuity deterioration by 47% (7 to 70%; p=0.004). There was a non-significant trend towards reduced all-cause mortality. The benefits of tight blood pressure control were broadly comparable between the captopril and atenolol arms, suggesting that the benefit was attributable to blood pressure reduction rather than to the specific drug class.
📊 Key Findings
| Endpoint (tight vs less-tight BP) | Effect Size |
|---|---|
| Any diabetes-related endpoint | 24% reduction (8–38%) · p=0.0046 |
| Diabetes-related death | 32% reduction (6–51%) · p=0.019 |
| Stroke | 44% reduction (11–65%) · p=0.013 |
| Microvascular endpoints | 37% reduction (11–56%) · p=0.0092 |
| Retinopathy progression (9-year) | 34% reduction (11–50%) · p=0.0004 |
| Visual acuity deterioration (9-year) | 47% reduction (7–70%) · p=0.004 |
| Mean BP achieved | 144/82 vs 154/87 mmHg · ~10/5 mmHg difference |
💬 Expert Commentary
UKPDS 38 established blood pressure control as a therapeutic target of at least equal importance to glycaemic control in type 2 diabetes. The direct comparison of blood pressure versus glycaemic outcomes from the UKPDS programme is instructive: a 10/5 mmHg blood pressure reduction in UKPDS 38 produced reductions in diabetes-related endpoints of 24% and a 44% stroke reduction, while the 0.9% HbA1c reduction in UKPDS 33 produced a 12% reduction in diabetes-related endpoints without a statistically significant macrovascular benefit. On a per-unit-of-intervention basis, blood pressure reduction appeared to produce more immediate macrovascular benefit than glycaemic improvement in this population, a conclusion that supported the prioritisation of blood pressure and cardiovascular risk factor management alongside glucose control in type 2 diabetes guidelines. This comparison should be interpreted cautiously given the different endpoints and follow-up periods, but it shaped clinical practice substantially.
The finding that captopril and atenolol produced broadly equivalent benefits was important because it deflected concern about whether the specific antihypertensive class mattered and reinforced the message that blood pressure level, rather than drug mechanism, is the primary determinant of benefit. Subsequent trials have refined this conclusion, with RAS blockade now preferred in diabetic nephropathy for its specific antiproteinuric effects beyond blood pressure lowering, but the UKPDS 38 primary finding that tight blood pressure control per se is protective remains the foundational principle. The absolute magnitude of the stroke reduction (44%) is one of the largest clinical benefits documented in any diabetes trial and speaks to the high cerebrovascular risk in hypertensive type 2 diabetes that can be substantially modified by relatively modest blood pressure reductions in the 10 mmHg range.
Limitations: The blood pressure target of 150/85 mmHg in the tight control arm would now be considered moderately aggressive but not intensive by current guideline standards. The captopril-versus-atenolol comparison was not a primary prespecified analysis. All-cause mortality was not significantly reduced. The study was funded by the UK MRC, BDA, and industry partners.
🔑 BOTTOM LINE
UKPDS 38 demonstrated that tight blood pressure control targeting below 150/85 mmHg in hypertensive type 2 diabetes reduces stroke by 44%, all diabetes-related endpoints by 24%, and microvascular complications by 37% over 8.4 years, establishing blood pressure control as a therapeutic target of at least equal macrovascular importance to glycaemic control in type 2 diabetes.
⭐ Clinical Impact Rating: ●●●●● Practice-defining
