Contemporary achievement of guideline recommendations for blood pressure management after stroke in the CONVINCE trial

Clinical Question (PICO)

Among clinically stable adults aged 40 years or older enrolled after a recent non-severe ischemic stroke or high-risk TIA, how often was mean follow-up blood pressure below the thresholds assessed in the CONVINCE cohort, and which baseline characteristics were associated with mean systolic blood pressure at or above 140 mmHg? This was an observational analysis of measured follow-up BP—not a randomized comparison of BP targets or antihypertensive intensification. The descriptive contrast was participants meeting versus not meeting the paper's <140/90 and <130/80 mmHg targets; age, prior hypertension, and lacunar stroke were examined as associated factors.

Bottom Line

Among 2,920 participants with at least one post-baseline BP reading, 36% had mean follow-up systolic BP ≥140 mmHg, while 70% had mean systolic BP ≥130 mmHg and therefore did not meet the <130 mmHg systolic component of the 2022 ESO target used by the authors. This documents a substantial target-attainment gap in a selected trial cohort; it does not establish that intensifying treatment to a particular target improves outcomes or is appropriate for every participant.

Design

Trial type: Longitudinal secondary analysis of the CONVINCE randomized trial cohort; this analysis did not randomize BP targets or treatment intensification.
N: CONVINCE randomized 3,154 patients; 3,144 consented patients were included in the reported cohort, and 2,920 had at least one post-baseline BP reading for this analysis.
Randomization: The parent trial assigned colchicine 0.5 mg daily versus usual care. Allocation ratio and blinding are not specified in this analysis; BP target attainment was not randomized.
Setting: 144 hospital sites in 13 European countries and Canada.
Enrollment / measurement period: Clinic BP follow-up visits from 2017 through 2024; this analysis does not give the parent trial's enrollment dates.
Follow-up: Median 33.6 months.
Analysis: Individual mean follow-up BP used all available readings, without imputation. Forward multivariable logistic regression assessed factors associated with mean SBP ≥140 mmHg; variables with univariate P<.05 were considered for the multivariable model. Sensitivity analyses examined readings after the 2022 ESO update and participants with at least 3 BP readings (n=2,383).
Primary outcome: Achievement of mean follow-up BP <140/90 mmHg (the target applied through most follow-up); the authors also assessed the 2022 ESO target of <130/80 mmHg. The regression endpoint was mean SBP ≥140 mmHg.

Population

Inclusion Criteria

Clinically stable patients aged ≥40 years after recent non-severe ischemic stroke due to large-artery atherosclerosis, lacunar disease, or cryptogenic embolism, or after high-risk TIA.
TIA eligibility included transient focal motor or speech symptoms with ABCD2 ≥4, ≥50% large-artery stenosis, or symptom-consistent diffusion-weighted MRI hyperintensity.
The qualifying event occurred 72 hours to 28 days before parent-trial randomization.

Exclusion Criteria

Atrial fibrillation, another cardioembolic source, or another defined cause such as dissection.
Moderate-to-severe renal or liver impairment or another contraindication to colchicine.
Severely disabling stroke was not represented in this cohort.

Baseline Characteristics (reported overall)

65% had a history of hypertension; 70.2% were taking at least one antihypertensive medication at baseline.
Antiplatelet use was 97.6% and statin use 94.1% at randomization.
The cohort was predominantly White European; among participants tabulated by mean SBP group, 95.8–96.5% were White and about 29–30% were women.

Interventions

Parent-trial colchicine arm: Colchicine 0.5 mg daily.
Parent-trial comparator: Usual care.
For this BP analysis: No BP-lowering intervention was assigned. Clinic BP was measured after a 5-minute rest at six-monthly trial visits; antihypertensive dose and adherence during follow-up were not objectively recorded.

Outcomes

Primary Outcome (mean follow-up BP target attainment):

Of 2,920 participants with follow-up BP, mean SBP was 136.6 mmHg (SD 12.8) and mean DBP was 79.3 mmHg (SD 7.8).
36% had mean SBP ≥140 mmHg; 9% had mean DBP ≥90 mmHg. Against the <130/80 mmHg target, 70% had mean SBP ≥130 mmHg and 47% had mean DBP ≥80 mmHg. The authors report that 70% did not meet the current ESO target.
Across individual follow-up visits, 50% of participants had BP exceeding 140/90 mmHg, and more than 75% exceeded 130/80 mmHg. These visit-level proportions are distinct from the participant-level averages.
In the prespecified/primary multivariable model for mean SBP ≥140 mmHg, older age was associated with higher odds per year (OR 1.026, 95% CI 1.018–1.034), as were history of hypertension (OR 1.47, 95% CI 1.24–1.74) and lacunar stroke subtype (OR 1.26, 95% CI 1.06–1.49). Diabetes had OR 1.20 (95% CI 1.00–1.45; P=.055) in this model.
The paper did not report a randomized effect of BP lowering on recurrent stroke, disability, or mortality. The post-2022 comparison found no observed difference in mean SBP between corresponding pre- and post-guideline visit periods; it was a sensitivity analysis, not a randomized policy comparison.

Secondary Outcomes:

At baseline, 52% met the pre-2022 <140/90 mmHg threshold and 25% met <130/80 mmHg.
There were 11,422 follow-up BP readings: 7,001 (61.3%) before the 2022 guideline update and 4,421 (38.7%) afterward. A sensitivity analysis restricted to participants with ≥3 readings (n=2,383) found similar mean BP and target-attainment rates.
In the sensitivity model replacing hypertension history with baseline antihypertensive count, each additional baseline medication was associated with SBP ≥140 mmHg (OR 1.107, 95% CI 1.03–1.19); this is an association, not evidence that the medications caused poor control.

Adverse Events / Safety:

This analysis did not test BP-target intensification and does not quantify harms from pursuing <130/80 mmHg. It provides no treatment-effect or safety estimate for an antihypertensive strategy.

Figures

Blood pressure control at follow-up visits below 140/90 mmHg — *Figure 1.* BP control at trial follow-up visits (<140/90 mmHg). Abbreviation: BP, blood pressure.

Blood pressure control at follow-up visits below 130/80 mmHg — *Figure 2.* BP control at trial follow-up visits (<130/80 mmHg).

Mean systolic blood pressure before and after the ESO guideline update — *Figure 3.* Mean systolic BP at trial visits before and after updated ESO guidelines. Abbreviations: BP, blood pressure; ESO, European Stroke Organisation.

Association between baseline antihypertensive medication count and mean systolic blood pressure — *Figure 4.* Association of number of baseline antihypertensive medications with mean SBP. Abbreviation: SBP, systolic blood pressure.

Criticisms

Selected cohort: This is a trial cohort of clinically stable, non-severely disabled participants; atrial-fibrillation/cardioembolic stroke and other groups were excluded. The population was predominantly White and European, limiting generalizability.
Target attainment is not treatment efficacy: The analysis does not randomize BP goals or titration and reports no causal effect of reaching <130/80 on recurrent stroke, disability, or survival. The measured gap alone cannot establish that every participant should receive more medication.
Measurement and treatment data are incomplete: BP was measured at scheduled clinic visits, and medication adherence and dose during follow-up were not objectively recorded. The mean can conceal visit-to-visit variation; visit-level threshold exceedance was higher than participant-average exceedance.
Post-2022 comparison is weak for causal interpretation: Only 38.7% of readings were after the 2022 ESO update. No change in corresponding-visit mean SBP does not prove the guideline had no effect; calendar time, changing participants, COVID-19, and implementation all complicate interpretation.
Modeling limits: Forward selection using univariate P<.05 can make associations sensitive to model specification; the diabetes estimate in the primary multivariable model was borderline (P=.055). The odds ratios do not show which intervention would improve control.

Funding

Supported by the Health Research Board Ireland, Deutsche Forschungsgemeinschaft (DFG, German Research Foundation; grant 397530000), and FWO—Research Foundation Flanders, Belgium. The funding statement does not describe the funders' role. Trial registration: ClinicalTrials.gov NCT02898610.

The paper

Authors. Synnott P, Walsh C, Weimar C, et al.
Title. Contemporary achievement of guideline recommendations for blood pressure management after stroke in the CONVINCE trial.
Journal. European Stroke Journal.
Year. 2026; volume 11, article aakag051.
DOI. 10.1093/esj/aakag051
PMCID. PMC13344104

Deep Dive — click to expand

What this is

[F] This 2026 secondary analysis uses repeated clinic BP measurements from 2,920 participants in CONVINCE, a randomized colchicine trial after recent ischemic stroke/high-risk TIA; it is not a trial of BP targets. The headline is persistent target non-attainment: 36% had mean SBP ≥140 mmHg and 70% had mean SBP ≥130 mmHg.

1. Shadow Audit

[F] The headline averages conceal a larger visit-level problem: 50% exceeded 140/90 at individual visits and more than 75% exceeded 130/80, versus 36% and 70% above the corresponding participant-mean SBP thresholds.
[F] The study recorded 11,422 follow-up BP readings, but the authors lacked objective follow-up antihypertensive dose and adherence data. [M] It therefore cannot distinguish therapeutic inertia from adherence, measurement, access, or individualized-treatment explanations.
[I] Calling this simply a “guideline failure” risks making an implementation diagnosis the study did not measure; it measured BP and associated characteristics, not the mechanism behind each elevated value.

2. Inversion Engine

[F] Mean follow-up SBP was 136.6 mmHg; 70% had mean SBP ≥130 mmHg. The result is target-relative, not an observed randomized treatment effect.
[I] The conclusion that a broad “new approach” is needed would weaken if systematic out-of-office measurement, individualized targets, or missing medication data showed that many above-threshold clinic averages did not represent safely modifiable BP. [M] The paper does not quantify that fraction or compare intervention strategies.
[H] A practical discriminating test would re-measure a sample of patients with validated home BP monitoring, review medication-taking and contraindications, and compare classification with clinic averages before launching broad titration. A large reclassification would favor a measurement/individualization response; persistent elevation among eligible adherent patients would support a treatment-escalation pathway.

3. Second-Order Catalyst

[F] Older age, prior hypertension, and lacunar stroke subtype were associated with higher odds of mean SBP ≥140 mmHg; the reported ORs were 1.026 per year, 1.47, and 1.26 respectively.
[I] These factors can help prioritize follow-up audits, but they do not identify who benefits most from medication intensification. If clinics act on them, the downstream need is an explicit owner for measurement review, adherence assessment, and safe treatment adjustment—not just a stricter discharge target.
[H] A low-cost pilot could compare completion of documented BP review and clinician follow-up in these higher-risk groups before and after assigning a named stroke/primary-care handoff; the paper supplies no estimate of the pilot's clinical effect.

4. Asymmetric Leverage

[F] Six-monthly scheduled readings already produced a sizeable measured gap, and the paper notes possible roles for organized secondary-prevention services, home monitoring, adherence, and therapeutic inertia.
[I] The leverage point is likely the step between a high reading and a completed treatment review. A simple process measure—among patients above target, how many receive a documented recheck, medication/adherence review, and follow-up plan?—could expose a remediable bottleneck at lower cost than launching a new drug trial.
[M] The study did not test that workflow or show whether home monitoring alone is sufficient; any implementation should track both BP and patient-important harms/outcomes.

5. Paradigm Destroyer

[F] The analysis was not randomized to BP goals; it did not measure treatment-intensification effects, recurrent stroke benefit, or harms of pursuing <130/80 mmHg.
[I] The strongest skeptical reading is that the paper quantifies a selected cohort's distance from guideline thresholds, not the number of patients who were undertreated. Its mean-based classification, missing longitudinal medication details, selected non-severe population, and predominantly White European sample limit causal and external-validity claims.
[M] A policy claim that everyone above the threshold should be intensified would require evidence on individualized eligibility, adherence, achieved out-of-office BP, adverse effects, and net clinical outcomes—none is estimated here.

MVP — Minimum Viable Proof

[F] In this selected cohort, 70% of 2,920 participants with follow-up BP had mean SBP ≥130 mmHg, despite repeated follow-up and high baseline antiplatelet/statin use. That is enough to justify auditing how elevated BP is rechecked and acted upon; it is not proof that automatic intensification improves outcomes.

Best Combination

Use Shadow Audit → Asymmetric Leverage: first separate measured target non-attainment from its unmeasured causes, then test the handoff from an elevated reading to confirmation, adherence review, and an individualized treatment plan. This avoids turning an observational gap into an unsupported treatment directive.

Overvalue Warning

Do not overvalue the 70% figure as a causal estimate or universal medication threshold. A trial-cohort average against a guideline target is not an individualized benefit–risk assessment, and the paper did not measure either the cause of non-attainment or the effect of correcting it.

Educational use notice. The Time Is Brain Review is dedicated to medical education. Figures from published articles are embedded for teaching purposes, with attribution to the original publication. We do not claim copyright over any embedded figure; all rights remain with the original publisher and authors under the article's published license. If you are a rights holder and would like a figure removed, please contact the editorial team. See our editorial notice for the full policy.