PROGRESS

Published

PROGRESS Collaborative Group. "Randomised trial of a perindopril-based blood-pressure-lowering regimen among 6,105 individuals with previous stroke or transient ischaemic attack". Lancet. 2001. 358(9287):1033–1041.
PubMed • Full text

Clinical Question

Among adults with stroke or TIA in the prior 5 years, is use of perindopril±indapamide associated with lower stroke risk when compared to placebo?

Bottom Line

Among adults with stroke or TIA in the prior 5 years, perindopril±indapamide is associated with lower stroke risk when compared to placebo. In a subgroup analysis, it seems that the benefit was seen among those on perindopril+indapamide (not perindopril alone), a group with greater BP lowering.

Major Points

Stroke is a leading cause of death and disability worldwide. Hypertension is the most common modifiable stroke risk factor. While it was generally accepted that BP lowering with antihypertensives led to lower stroke risk among those without a prior event, the role of BP lowering for secondary prevention of stroke was unclear in the 1990s.

Published in 2001, the Protection Against Recurrent Stroke Study (PROGRESS) randomized 6105 persons with prior stroke or TIA to blood pressure lowering with a fixed dose of 1 or 2 medications ("active therapy") or placebo. The active therapy in this study was perindopril±indapamide. About half of those randomized also had indapamide, which was used among those without a clear indication for diuretics and also not a clear reason not to use diuretics. Most had stage 1 or 2 hypertension. Unlike 2015 SPRINT's design, this was not a treat-to-BP-target study as all participants received the same medication(s) that was (were) not titrated. Notably, BP lowering was about 2x larger among those with both perindopril+indapamide than perindopril alone, when compared to placebo. With 3.9 years of follow-up, fatal or non-fatal stroke events were less common in the active treatment arm than the placebo arm (10% vs. 14%; risk reduction 28%; P<0.0001). Major vascular events (non-fatal stroke, non-fatal MI, vascular mortality) and hospital admissions were also less common. There was no mortality difference.

There were two important findings in the subgroup analysis. First, the benefit from active therapy was not limited to those with very elevated BPs at baseline (threshold 160/90 mm Hg), as stroke reduction was similarly lowered among those with with very elevated ("hypertensive" in this manuscript) or not very elevated BP at baseline ("not hypertensive"). Second, there was no observed stroke reduction with perindopril alone while there was a 43% risk reduction among those on perindopril+indapamide. While it is possible that there might be some sort of class effect with the use of indapamide, findings from SPRINT showed that intensive BP control (ie greater BP lowering) is associated with lower CVD events, so the benefit in stroke reduction in the perindopril+indapamide arm might from to the overall BP lowering itself. Of note, a separate PROGRESS manuscript reported less cognitive decline with active therapy than with placebo, which is also consistent with SPRINT's findings.

Ultimately, PROGRESS helped solidify the importance of BP lowering among adults with prior stroke and elevated BP.

Guidelines

AHA/ASA Stroke secondary prevention (2021, adapted)

• For those with hypertension and stroke or TIA, ACE-inhibitor/ARB, or thiazide diuretics is useful to lower BP and stroke risk (COR 1, LOE A)
  • An office BP goal of 130/80 mm Hg is recommended for most patients (COR 1, LOE B-R)
  • Can individualize drug regimens (COR 1, LOE B-NR)

Design

• Multicenter, double-blind, randomized controlled trial
• N=6105
  • Placebo (n=3054)
    • Single placebo (n=1280)
    • Double placebo (n=1774)
  • Anti-hypertensive treatment (n=3051)
    • perindopril only (n=1281)
    • combination of perindopril and indapamide (n=1770)
• Setting:172 collaborating centers from ten countries in Asia, Australasia, and Europe
• Enrollment: 1995-2001
• Mean follow-up: 3.9 years
• Analysis: Intention-to-treat
• Primary outcome: Repeat stroke

Population

Inclusion Criteria

• In the prior 5 years a stroke or TIA event, defined by:
  • Stroke - Acute neurological disturbance for >24h attributed to ICH or ischemia
  • TIA - Acute, focal neurological or monocular disturbance for <24h attributed to embolic or thrombotic arterial event
• Clinically stable for ≥2 weeks since qualifying event

Note: There were no BP threshold criteria, though there was a recommendation for uncontrolled hypertension to be treated with non-ACE inhibitor BP agents prior to trial enrollment.

Exclusion Criteria

• Other indication for treatment with ACE inhibitor (eg HF) or contraindication to ACE inhibitor

Baseline Characteristics

From the active treatment arm.

• Demographics: Age 64 years, 30% women, 39% asian
• CVA history: Ischemic stroke 71%, hemorrhagic stroke 11%, unknown stroke type 4%, TIA or amaurosis fugax 22%
  • Time since stroke: 8 months (IQR 2-21)
• Medical problems: Smoker 20%, diabetes 13%, CHD 16%
• BP (SD): systolic 147 (19), diastolic 86 (11),
  • Hypertension (SBP ≥160 or dbp ≥90, regardless of BP med use): 48%
  • On BP medications: 50%

Interventions

• All participants had 4 week prerandomization run-in, receiving open-label perindopril 2mg PO qday x2 weeks then 4mg PO qday x2 weeks. Those with high adherence and those who tolerated the agent were randomzied to a group:
  • Active treatment - Perindopril 4 mg po qday ± indapamide 2.5 mg po qday (2 mg in Japan)
    • Addition of indapamide occurred among patients without clear indication or contraindication for a diuretic
  • Placebo - including indapamide placebo if they would be treated with this agent in the active treatment arm.
• Determination of whether or not indapamide would be used occurred prior to randomization, so participants were stratified by indapamide use (in addition to other stratification criteria outlined in the Methods section on page 1034).
• Other medical care left to the treating physician's discretion.

Outcomes

Presented as active treatment vs. placebo. The risk differences reported were derived from hazard ratios from Cox proportional hazards models.

Primary Outcomes

Fatal or nonfatal stroke

Components of the primary outcome were not themselves primary outcomes, but are presented here for simplicity

307 (10%) vs. 420 (14%) (risk reduction 28%; 95% CI 17 to 38%; P<0.0001)

Fatal or disabling (a secondary outcome): 123 vs. 181 (risk reduction 33%; 95% CI 15 to 46%)

Not fatal or disabling: 201 vs. 262 (risk reduction 24%; 95% CI 9 to 37%)

Ischemic: 246 vs. 319 (risk reduction 24%; 95% CI 10 to 35%)

Hemorrhagic: 37 vs. 74 (risk reduction 50%; 95% CI 26 to 67%)

Unknown stroke type: 42 vs. 51 (risk reduction 18%; 95% CI -24 to 45%)

Secondary Outcomes

See fatal or disabling stroke outcome above.

Major vascular event

Non-fatal stroke, non-fatal MI, vascular mortality.

458 (15%) vs. 604 (20%) (risk reduction 26%; 95% CI 16 to 34%)

All-cause mortality

306 vs. 319 (risk reduction 4%; 95% CI -12 to 18%)

Vascular mortality: 181 vs. 198 (risk reduction 9%; 95% CI -12 to 25%)

Stroke mortality: 42 vs. 50 (risk reduction 16%; 95% CI -17 to 44%)

MI mortality: 58 vs. 62 (risk reduction 7%; 95% CI -34 to 39%)

Other vascular mortality: 81 vs. 86 (risk reduction 6%; 95% CI -28 to 30%)

Cancer mortality: 64 vs. 65 (risk reduction 2%; -39 to 30%)

Other non-vascular mortality: 61 vs. 56 (risk reduction -9%; 95% CI -57 to 24%)

Proportion with any hospital admission

41% vs. 44% (risk reduction 9%; 95% CI 1 to 15%)

Cognitive outcomes

Published in a separate manuscript.

Dementia: 6.3% vs. 7.1% (risk reduction 12%; 95% CI -8 to 28%; P=0.20)

Cognitive decline: 9.1% vs. 11.0% (risk reduction 19%; 95% CI 4 to 32%; P=0.01)

Additional Outcomes

Change in BP

9.0/4.0 mm Hg (SE 0.3/0.2) lower in the active treatment arm

Perindopril+indapamide: 12.3/5.0 mm Hg (SE 0.5/0.3) lower in active treatment arm

Perindopril only: 4.9/2.8 mm Hg (SE 0.6/0.3) lower in active treatment arm

Subgroup Analysis

Stroke

Perindopril+indapamide: 150/1770 vs. 255/1774 (risk reduction 43%; 95% CI 30 to 54%)

Perindopril only: 157/1281 vs. 165/1280 (risk reduction 5%; 95% CI -19 to 23%)

Hypertensive: 163/1464 vs. 235/1452 (risk reduction 32%; 95% CI 17 to 44%)

Not hypertensive: 144/1587 vs. 185/1602 (risk reduction 27%; 95% CI 8 to 42%)

Note: "Hypertensive" was defined as SBP ≥160 or DBP ≥90 mm Hg at baseline, regardless of concurrent BP medication use.

Major vascular event

Perindopril+indapamide: 231/1770 vs. 367/1774 (risk reduction 40%; 95% CI 29 to 49%)

Perindopril only: 227/1281 vs. 237/1280 (risk reduction 4%; 95% CI -15 to 20%)

Hypertensive: 240/1464 vs. 331/1452 (risk reduction 29%; 95% CI 16 to 40%)

Not hypertensive: 218/1587 vs. 273/1602 (risk reduction 24%; 95% CI 9 to 37%)

Adverse Events

Stopped study medication

23% vs. 21% (P=0.02)

For participants discretion: 7.6% vs. 8.2%

For cough: 2.2% vs. 0.4%

For hypotension: 2.1% vs. 0.9%

Angioedema

3 vs. 0

Criticisms

• Not a treat-to-BP-target trial, like SPRINT.
• The blood pressure cut off delineating hypertension at intake was 160/90 mmHg, and did not consider use of other BP medications.
• 14% of patients (1016/7121) withdrew prior to randomization in open-label period due to side effects or other reason affecting the applicability of this study. The study population was highly selected, and may not be representative of the general population
• Mostly men 70%
• There was no indapamide alone arm in this trial, which make the results difficult to interpret, especially when the PATS trial previously shown indapamide reduces recurrent stroke risk by 29% when used as secondary prevention with only a 5/2 mm Hg BP reduction.

Funding

• Servier, which at the time had a patent on the brand version of perindopril and perindopril+indapamide agents.
• The Health Research Council of New Zealand, and the National Health and Medical Research Council of Australia.