Secondary Prevention & Rehabilitation

1. Why Secondary Prevention is the Highest-Yield Intervention

The first stroke advertises a brain whose vasculature is already deranged. Without treatment, the recurrence risk is steep: ~10% in the first year, ~5% per year thereafter, with a particularly dense early window (TIA confers ~5% stroke risk in 48 h, ~10% in 90 days). Recurrent strokes are also typically worse than the index event — more disability, higher mortality, more dementia.

Unlike acute reperfusion (which buys minutes back from a single event), secondary prevention compounds: a 30% relative risk reduction maintained over 10 years prevents far more disability-adjusted life years than any single thrombolysis. The modifiable factors identified in INTERSTROKE account for ~90% of population-attributable risk — nearly all of it is preventable.

Stroke recurrence is, in principle, the single most preventable neurological catastrophe in adult medicine. The clinical task is simply to actually do it— uptake of evidence-based prevention remains poor in most healthcare systems.

2. Mechanism-Based Approach

Effective prevention follows the TOAST classification (Adams 1993): different mechanisms demand different strategies, and the wrong drug for the wrong mechanism wastes risk reduction or harms the patient.

Diagnostic work-up therefore aims to nail the mechanism: vessel imaging (carotid duplex, CTA, MRA), cardiac evaluation (echocardiogram, telemetry, often insertable cardiac monitor in ESUS where AF detection rate >30% by 3 years), and serology when indicated (vasculitis, thrombophilia in young stroke).

3. Antiplatelet Therapy

For non-cardioembolic ischaemic stroke and TIA, antiplatelet therapy is the backbone. The major options and their evidence:

Routine triple therapy or aspirin+ticagrelor for chronic prevention is not supported. For symptomatic intracranial atherosclerosis, the SAMMPRIS trial favoured aggressive medical therapy (DAPT 90 d + statin + BP) over stenting.

4. Anticoagulation in Atrial Fibrillation

Atrial fibrillation accounts for the largest fraction of preventable recurrent strokes. AF roughly quintuples stroke risk; AF-related strokes are larger (left atrial appendage thrombi) and twice as deadly as non-AF strokes. Anticoagulation cuts this risk by ~64% — among the largest risk reductions available in any branch of medicine.

Direct Oral Anticoagulants (DOACs)

DOACs have replaced warfarin as first-line for non-valvular AF in modern guidelines: non-inferior efficacy, ~50% less intracranial haemorrhage, no INR monitoring, predictable kinetics. The four pivotal trials each enrolled ~14,000–21,000 patients vs warfarin:

Apixaban is often preferred for the elderly and those with renal impairment (renal clearance ~25%); dabigatran is preferred only if reversal with idarucizumab is desired and renal function is good. Andexanet alfa reverses Xa inhibitors. Warfarin remains first-line for mechanical heart valves and moderate-severe mitral stenosis — DOACs failed in RE-ALIGN.

Timing of initiation after stroke

The classical “1-3-6-12 day” rule (TIA day 1, small stroke day 3, moderate day 6, large day 12) was always pragmatic rather than evidence-based. The ELAN (NEJM 2023) and OPTIMAS (Lancet 2024) trials showedearly DOAC initiation (within ~4 days) is at least as safe and reduces recurrence vs delayed start; bleeding does not rise. Practice is shifting earlier.

Left atrial appendage occlusion

For patients with absolute or relative anticoagulation contraindications, percutaneous LAA closure (Watchman device, PROTECT-AF, PREVAIL) prevents stroke at rates comparable to warfarin, eliminating long-term bleeding risk. Reserved for selected patients due to procedural complications and the residual stump risk.

5. Carotid Revascularisation

For symptomatic extracranial internal carotid artery stenosis, revascularisation on top of medical therapy reduces recurrent stroke. The classic trials defined the degree-of-stenosis threshold:

Trans-carotid artery revascularisation (TCAR) with proximal flow reversal is an emerging hybrid approach with peri-procedural stroke rates approaching CEA.

6. Statin & Lipid Therapy

The SPARCL trial (NEJM 2006, n=4731) gave atorvastatin 80 mg to patients with recent stroke or TIA and no known coronary disease: 5-year stroke recurrence 11.2% → 13.1% (HR 0.84, ARR 2.2%, NNT ~46). Mechanism is plaque stabilisation, anti-inflammatory effect, endothelial improvement — not solely LDL lowering, though LDL reduction tracks benefit.

Caveat in haemorrhagic stroke. SPARCL hinted at a small absolute increase in ICH on intensive statin (especially in patients with a prior haemorrhagic event). Net benefit still favours statin in those with concurrent ischaemic risk; case-by-case assessment after lobar ICH or CAA.

7. Hypertension Control

Hypertension is the largest single modifiable risk factor for both ischaemic and haemorrhagic stroke (~48% population-attributable risk in INTERSTROKE). Lowering BP after stroke is among the highest-yield interventions available.

Modern target: <130/80 mmHg in most stroke survivors (AHA/ASA 2021), individualised for tolerability and orthostasis. Class effect dominates — ACEi/ARB, calcium-channel blockers, and thiazides all reduce events; diuretic + ACEi combination has the strongest stroke-specific evidence (PROGRESS).

8. Lifestyle & Behavioural

Behavioural interventions are unglamorous, hard to deliver, and often the largest residual lever after pharmacology has been optimised.

9. Patent Foramen Ovale Closure

A patent foramen ovale persists in ~25% of adults; in cryptogenic stroke (especially age <60), paradoxical embolism through PFO is a plausible mechanism. Three concurrent 2017 trials transformed practice:

Pooled NNT ~33 over 5 years to prevent one stroke. Atrial fibrillation occurs more commonly post-closure (~3–5%), usually transient. Patient selection uses the ROPE score (Risk of Paradoxical Embolism; Kent 2013) — younger patients without classical risk factors and large/shunting PFOs have the highest probability that the PFO is causal.

Recommended for: cryptogenic stroke patients aged 18–60 with high-risk PFO features (atrial septal aneurysm, large shunt) and otherwise unexplained mechanism. Not recommended in older patients with vascular risk factors where the alternative mechanism (small-vessel, occult AF) is more likely.

10. Neuroplasticity & Recovery Biology

Recovery after stroke is not the dead returning to life — it is the surviving brain reorganising to do work the infarcted tissue can no longer do. The biology proceeds on several timescales.

The proportional recovery rule (Prabhakaran et al. 2008) observed that ~70% of upper-limb impairment is recovered in patients without severe motor cortex/CST damage — a reproducible regularity that constrains expectations and identifies non-fitters who need different strategies.

11. Rehabilitation

Acute reperfusion saves brain at minute scale; rehabilitation determines what the patient can do with what is left. The single strongest predictor of recovery, after initial severity, is therapy intensity — hours per day of task-specific, dose-driven, goal-oriented practice.

Motor rehabilitation

Aphasia & communication

Intensive language therapy (≥5 h/wk over weeks) improves naming, comprehension, and functional communication beyond spontaneous recovery; constraint- induced aphasia therapy (CIAT) extends the CIMT principle to language. Melodic intonation therapy exploits right-hemisphere singing pathways for non-fluent aphasia. Computerised therapy and tele-rehabilitation extend dose access.

Hemispatial neglect

Prism adaptation — right-shifting prism goggles during reaching tasks induce a leftward after-effect that reduces left neglect for hours to weeks. Visual scanning training, limb activation, and (experimentally) caloric/galvanic vestibular stimulation provide modest gains. Neglect is a strong negative prognostic feature for functional independence.

Post-stroke depression

Affects ~33% of survivors; under-recognised; worsens motor recovery, cognition, and mortality. Screen with PHQ-9. SSRIs (sertraline, citalopram) are first-line; psycho- therapy effective. The fluoxetine-and-recovery story is instructive: FLAME (Lancet Neurol 2011) suggested fluoxetine 20 mg accelerated motor recovery beyond mood effect — but the much larger FOCUS (Lancet 2019, n=3127), EFFECTS(n=1500), and AFFINITY (n=1280) trials were neutral for functional recovery. Fluoxetine is not recommended routinely for motor recovery; treat depression on its own merits.

Stroke-unit care & intensity

Cochrane meta-analysis: organised stroke-unit care reduces death-or-dependency by ~21% compared with general-ward care — the largest organisational effect in stroke medicine, with no marginal cost beyond co-locating expertise. Multidisciplinary teams (PT, OT, SLT, neuropsychology, social work, nursing) co-ordinate goals and dose.

Across modalities, a single principle dominates: dose drives outcome. Most rehabilitation programmes worldwide deliver far less practice than animal models suggest is necessary; the frontier is intensity scaling via robotics, tele-rehabilitation, gamification, and home-based programmes.

12. Long-Term Outcomes & Disability

The lived outcome of stroke is captured in standardised measures that anchor trials and clinical conversation:

mRS 0–2 is the conventional definition of functional independence and the primary endpoint of most acute-stroke trials. The Barthel Index (0–100) measures activities of daily living more granularly. NIHSS (acute), FMA (Fugl-Meyer, motor), and the Stroke Impact Scale (patient-reported) round out the toolkit.

Return to work. ~40–60% of working-age survivors return, often part-time or with accommodations; predictors include younger age, mild stroke (NIHSS <5), preserved cognition, white-collar work, supportive employer, and access to vocational rehabilitation. Failure to return correlates with depression and reduced quality of life independent of physical recovery.

Community reintegration. Beyond mRS lies participation: re-engaging in roles, relationships, and meaning. Social isolation, caregiver burden, and post-stroke fatigue (~50% prevalence, often invisible) shape the human outcome. Survivor-led peer support, community rehabilitation, and long-term follow-up clinics address what hospitals leave undone.

Post-stroke cognitive impairment and vascular dementia — same vasculature, same risk factors, same prevention — bind stroke care to dementia prevention. Aggressive vascular risk-factor control after stroke probably reduces incident dementia, though dedicated trials remain limited.