Skip to main content

Foundational Trials

British Aneurysm Nimodipine Trial

Image: Shutterstock / ChaNaWiT

Publication

  • Title: Effect of oral nimodipine on cerebral infarction and outcome after subarachnoid haemorrhage: British aneurysm nimodipine trial
  • Year: 1989
  • Journal published in: BMJ
  • Citation: Pickard JD, Murray GD, Illingworth R, Shaw MDM, Teasdale GM, Foy PM, et al. Effect of oral nimodipine on cerebral infarction and outcome after subarachnoid haemorrhage: British aneurysm nimodipine trial. BMJ. 1989 Mar 11;298(6674):636-42.

Context & Rationale

  • Background
    • Delayed neurological deterioration after aneurysmal subarachnoid haemorrhage was (and remains) a major driver of death and severe disability, commonly associated with cerebral ischaemia and infarction.
    • Nimodipine (a dihydropyridine calcium antagonist) was hypothesised to reduce ischaemic events after subarachnoid haemorrhage (via vascular and/or neuroprotective mechanisms), but definitive multicentre evidence with clinically meaningful endpoints was limited at the time.
    • Oral administration was clinically attractive (simplicity; avoidance of high-dose intravenous hypotension), but required confirmation of benefit and safety in routine neurosurgical practice.
  • Research Question/Hypothesis
    • Does early oral nimodipine, compared with placebo, reduce cerebral infarction and improve 3‑month neurological outcome after subarachnoid haemorrhage?
  • Why This Matters
    • A pharmacological strategy that meaningfully reduces infarction and severe disability after subarachnoid haemorrhage would represent a major advance in neurocritical care and neurosurgical outcomes.
    • Demonstration of benefit (or lack of benefit) would shape standard-of-care globally, given the burden of aneurysmal subarachnoid haemorrhage–related morbidity.

Design & Methods

  • Research Question: In patients with subarachnoid haemorrhage admitted within 96 hours, does oral nimodipine reduce cerebral infarction and improve 3‑month outcome compared with placebo?
  • Study Type: Multicentre, randomised, double‑blind, placebo‑controlled trial in four UK neurosurgical centres (London, Southampton, Liverpool, Glasgow).
  • Population:
    • Inclusion: Patients admitted within 96 hours of onset of symptoms and signs of subarachnoid haemorrhage; diagnosis confirmed by lumbar puncture and/or computed tomography.
    • Consent: Patient consent or (if unconscious) next‑of‑kin consent.
    • Key exclusions: Pregnancy; major renal/hepatic/pulmonary disease; cardiac decompensation; myocardial infarction within previous 6 months; age <18 years; coma in the week preceding the most recent subarachnoid haemorrhage; refusal of consent.
  • Intervention:
    • Oral nimodipine 60 mg every 4 hours (two 30 mg tablets), for 21 days in survivors.
    • Administration by mouth or via nasogastric tube if unable to swallow.
    • Dose modification for hypotension was permitted by protocol, but was not usually necessary.
  • Comparison:
    • Matching placebo tablets on the same schedule (every 4 hours for 21 days in survivors), with identical appearance and packaging.
    • Other aspects of clinical management (including surgery and supportive care) were provided according to local practice.
  • Blinding: Double‑blind (patients, clinicians, and outcome assessors); identical tablets; four code breaks reported.
  • Statistics: A minimum of 540 patients were required to detect a 50% reduction in incidence of cerebral infarction of 15% with 80% power at the 5% significance level; primary analyses were intention‑to‑treat (no post‑entry exclusions reported other than one withdrawal); multivariable (stepwise) logistic regression evaluated treatment effects adjusting for key prognostic variables.
  • Follow-Up Period: Outcomes assessed at least 3 months after trial entry (Glasgow Outcome Scale), including postal/telephone follow‑up if patients moved away.

Key Results

This trial was not stopped early. Recruitment continued until the planned sample size threshold was exceeded (554 patients randomised).

Outcome Oral nimodipine Placebo Effect p value / 95% CI Notes
Cerebral infarct (all) 61/278 (22%) 92/276 (33%) Relative reduction 34% 95% CI 13 to 50; P=0.003 Primary endpoint (radiological/pathological confirmation when appropriate).
Poor outcome (death, vegetative state, or severe disability at ≥3 months) 55/278 (20%) 91/276 (33%) Relative reduction 40% 95% CI 20 to 55; P=0.0001 Glasgow Outcome Scale dichotomised as poor vs not poor.
Rebleed 25/278 (9%) 38/276 (14%) Relative reduction 35% 95% CI -5 to 59; P=0.077 Difference did not meet conventional statistical significance.
Death at ≥3 months (Glasgow Outcome Scale) 43/278 60/276 Relative reduction 29% 95% CI -1 to 50; P≈0.06 Reported “observed significance level” 6% for mortality.
Severe disability at ≥3 months (Glasgow Outcome Scale) 11/278 29/276 Not reported Not reported Reduction in severe disability numerically prominent in categorical outcome table.
Adverse reactions attributed to trial drug 14/278 10/276 Not reported Not reported 27 reactions in 24 patients (17 reactions in nimodipine group; 10 in placebo group).
  • Nimodipine was associated with fewer cerebral infarcts (22% vs 33%) and fewer poor outcomes (20% vs 33%), with reported relative reductions of 34% and 40%, respectively.
  • The categorical Glasgow Outcome Scale distribution suggested benefit driven largely by fewer deaths and fewer patients left severely disabled.
  • Adjusted analyses (logistic regression) retained statistically significant treatment effects for cerebral infarction and poor outcome after accounting for key prognostic factors.

Internal Validity

  • Randomisation and allocation: Centre‑specific randomisation lists balanced in blocks of four; sequentially numbered, identical tablets allocated to eligible patients; practical allocation concealment appears reasonable, although small fixed blocks can create predictability risk if sequences are compromised.
  • Dropout/exclusions: One reported withdrawal after 4 days of placebo; otherwise no post‑entry exclusions reported and follow‑up processes included postal/telephone follow‑up when patients moved away.
  • Performance/detection bias: Double‑blind design reduces bias; three‑month outcome assessed by a doctor not responsible for early management (or by postal/telephone enquiry), supporting independence of outcome assessment.
  • Protocol adherence: Treatment was discontinued early in 130 patients (70 nimodipine; 60 placebo), commonly around surgery (26), or because of intercurrent complications (including hypotension (6), cardiac complications (7), hepatic/biliary abnormalities (4), gastrointestinal symptoms (4)); this discontinuation would tend to dilute measured treatment effects.
  • Baseline characteristics: Broadly similar between groups; for example, clinical grade II: 168/278 vs 159/276; grade III: 76/278 vs 72/276; grade IV: 19/278 vs 25/276; cisternal blood >2 mm on initial CT: 177/278 vs 168/276; time from ictus to operation: 10.8 (range 2–60) vs 11.3 days (range 2–116).
  • Heterogeneity: Multicentre design improves robustness but introduces practice variation (e.g., timing and approach to aneurysm surgery); stratification by centre and multivariable adjustment partially address this.
  • Timing: Treatment started within 96 hours of subarachnoid haemorrhage; early ischaemic events before enrolment would not be preventable by the intervention, potentially biasing toward underestimating benefit.
  • Dose: Fixed oral dose (60 mg every 4 hours) for 21 days aligns with later “standard” dosing paradigms; protocol permitted dose adjustment for hypotension but this was “not usually necessary”.
  • Separation of the variable of interest: Pharmacodynamic signal present (after starting nimodipine, systolic blood pressure fell by mean 7.1 mmHg (SD 18.5), diastolic by 3.6 mmHg (SD 11.0); pulse rose from mean 76 (SD 12) to 85 (SD 15) beats/min), while clinical separation was reduced by early discontinuation (70 vs 60).
  • Outcome assessment: Deterioration was operationalised (new focal sign or fall >1 point on Glasgow Coma Scale for >6 hours) and investigated clinically and by CT/lumbar puncture/necropsy; all deterioration episodes and all patients without “good recovery” were reviewed by a committee.
  • Statistical rigour: Pre‑trial sample size justification provided and achieved (554 randomised vs minimum 540); intention‑to‑treat framework with adjusted analyses; key endpoints reported with confidence intervals and p values.

Conclusion on Internal Validity: Overall, internal validity appears moderate to strong, supported by randomisation, double‑blinding, and structured outcome ascertainment; the main threats are treatment discontinuation (130 patients) and background care heterogeneity typical of pragmatic multicentre trials in that era.

External Validity

  • Population representativeness: Adults presenting to UK neurosurgical centres within 96 hours; broad clinical grade spectrum (I–V) but with predominance of grade II–III; exclusions removed patients with major organ failure and recent myocardial infarction, limiting applicability to multi‑morbid cohorts.
  • Applicability to current practice: Aneurysm securing strategies and neurocritical care have evolved substantially since 1989 (endovascular coiling, earlier intervention, modern ICU protocols), but delayed cerebral ischaemia remains common, and the tested nimodipine regimen is still clinically feasible in high‑resource settings.
  • System-level transferability: Findings likely translate best to centres able to deliver frequent enteral dosing and manage hypotension (including vasopressor support where needed); generalisability to low‑resource settings may be constrained by monitoring and nursing intensity.

Conclusion on External Validity: External validity is good for typical adult aneurysmal subarachnoid haemorrhage care in well‑resourced systems, but less certain in very elderly, severely co‑morbid, or haemodynamically fragile populations, and in settings where intensive monitoring cannot support frequent dosing.

Strengths & Limitations

  • Strengths:
    • Large, multicentre, randomised, double‑blind, placebo‑controlled design with clinically meaningful endpoints.
    • Primary outcome incorporated objective infarction ascertainment (CT/necropsy when appropriate).
    • Three‑month functional outcome assessed independently of early clinical team, supporting credible outcome measurement.
    • Recruitment met/exceeded the prespecified sample size target (554 vs minimum 540).
  • Limitations:
    • Substantial early discontinuation of study drug (130 patients), reducing treatment separation and complicating “dose–response” inference.
    • Background management (including timing of aneurysm surgery and supportive strategies) was not protocolised and reflects 1980s practice, potentially limiting comparability to contemporary pathways.
    • Block randomisation with small fixed block size may carry a theoretical risk of allocation predictability if concealment is breached.
    • The mechanism of benefit cannot be confidently inferred from trial endpoints (infarction reduction does not specify whether benefit is via vasospasm modification, neuroprotection, or other pathways).

Interpretation & Why It Matters

  • Clinical implication
    Oral nimodipine (60 mg every 4 hours for 21 days) reduced both cerebral infarction (22% vs 33%) and poor functional outcome (20% vs 33%) after subarachnoid haemorrhage, supporting routine early administration in eligible patients.
  • Why it became practice-changing
    Benefit was demonstrated on hard, patient‑centred outcomes (death/severe disability) rather than solely physiological surrogates, providing a rare example of a neuroprotective/vasoactive agent improving meaningful recovery after aneurysmal subarachnoid haemorrhage.
  • Modern practice alignment
    Contemporary aneurysmal subarachnoid haemorrhage guidelines continue to recommend enteral nimodipine as standard therapy following aneurysm rupture23.

Controversies & Subsequent Evidence

  • Mechanism of benefit remains contested: Subsequent synthesis indicates that calcium antagonists (notably nimodipine) improve clinical outcome after aneurysmal subarachnoid haemorrhage, but this improvement does not map cleanly onto reductions in angiographic vasospasm, implying potential neuroprotective and microvascular effects beyond large‑artery spasm1.
  • Route of administration and haemodynamic trade-offs: Enteral dosing is guideline‑preferred, in part because intravenous administration can be associated with hypotension; more recent network meta-analytic comparisons of enteral versus intravenous nimodipine highlight uncertainty about superiority of alternative routes and reinforce pragmatic concerns about haemodynamic tolerance and treatment continuity56.
  • Care-era generalisability: The British trial occurred in an era of later aneurysm surgery and different ICU standards; despite this, major modern guidelines still interpret the nimodipine signal as sufficiently robust to recommend it routinely, suggesting perceived biological generalisability across eras of aneurysm‑securing strategy234.
  • Rebleeding signal interpretation: Although rebleeding was numerically lower (9% vs 14%) in this trial, it did not meet conventional significance, and later aggregated evidence has not positioned nimodipine as an anti‑rebleeding strategy; its role is centred on ischaemic injury and outcomes rather than aneurysm instability1.

Summary

  • Multicentre UK trial (n=554) testing oral nimodipine 60 mg every 4 hours for 21 days versus placebo, started within 96 hours of subarachnoid haemorrhage.
  • Reduced cerebral infarction: 22% (61/278) vs 33% (92/276); relative reduction 34%; 95% CI 13 to 50; P=0.003.
  • Improved functional outcome: poor outcome 20% (55/278) vs 33% (91/276); relative reduction 40%; 95% CI 20 to 55; P=0.0001.
  • Mortality was numerically lower (43/278 vs 60/276) but with borderline statistical support (reported significance level 6%).
  • High early discontinuation (130 total; 70 nimodipine, 60 placebo) likely diluted effects; despite this, clinically meaningful benefit persisted.

Further Reading

Other Trials

Systematic Review & Meta Analysis

Observational Studies

Guidelines

Notes

  • Where author lists or bibliographic fields were not retrievable from available sources, they are marked “Not reported” rather than inferred.

Overall Takeaway

The British Aneurysm Nimodipine Trial is a landmark because it demonstrated, in a large double‑blind multicentre design, that oral nimodipine meaningfully reduces cerebral infarction and improves functional outcome after subarachnoid haemorrhage. Despite care‑era differences, modern guidelines continue to embed nimodipine as standard therapy because few other pharmacological interventions have shown comparable patient‑centred benefit.

Overall Summary

  • Oral nimodipine started within 96 hours and continued (intended) for 21 days reduced cerebral infarction and poor outcome at ≥3 months in this foundational UK RCT.

Bibliography