OPTIMISTmain
(2025)Objective
To assess whether a low-intensity monitoring protocol is non-inferior to standard high-intensity monitoring for functional outcome in patients with mild-to-moderate acute ischemic stroke (NIHSS <10) receiving intravenous thrombolysis
Study Summary
• No increase in symptomatic ICH (0.2% vs 0.4%) or serious adverse events (11.1% vs 11.3%)
• Intervention reduced monitoring assessments from 37 to 17 over 24 hours
Intervention
Low-intensity monitoring: vital signs and neurological assessment q15min for 2h, q2h for 8h, q4h until 24h (17 total assessments) vs Standard: q15min for 2h, q30min for 6h, q1h until 24h (37 total assessments)
Inclusion Criteria
Adults ≥18 years with acute ischemic stroke receiving IV thrombolysis, clinically stable with NIHSS <10 within 2 hours of thrombolysis initiation
Study Design
Arms: Low-intensity monitoring (17 assessments/24h) vs Standard monitoring (37 assessments/24h)
Patients per Arm: Low-intensity: 2789, Standard: 2133
Outcome
• Death at 90 days: 2.2% vs 1.9%
• Symptomatic ICH: 0.2% vs 0.4%
Bottom Line
OPTIMISTmain provides weak evidence that low-intensity monitoring (17 assessments over 24h) is non-inferior to standard monitoring (37 assessments) in stable post-thrombolysis patients with NIHSS <10. The upper bound of the 95% CI (1.15) just touched the non-inferiority margin, with no increase in symptomatic ICH or serious adverse events. Hospitals may consider incorporating this approach according to local circumstances.
Major Points
- International stepped-wedge cluster-randomized non-inferiority trial across 114 hospitals in 8 countries
- 4922 participants: 2789 low-intensity monitoring, 2133 standard monitoring
- Low-intensity protocol reduced assessments from 37 to 17 over 24 hours post-thrombolysis
- Primary outcome (mRS 2-6 at 90 days): 31.7% vs 30.9%, RR 1.03 (95% CI 0.92-1.15), p_non-inferiority=0.057
- Upper CI boundary just touches the prespecified non-inferiority margin of 1.15, providing 'weak evidence' of non-inferiority
- Symptomatic ICH very low in both groups: 0.2% vs 0.4% (RR 0.57)
- No significant differences in serious adverse events (11.1% vs 11.3%)
- ICU utilization reduced with low-intensity monitoring, particularly in USA (47.1% vs 77.1%)
- Process evaluation showed intervention was acceptable, feasible, and well-received by nurses and clinicians
Study Design
- Study Type
- International, pragmatic, multicentre, stepped-wedge, cluster-randomised, controlled, blinded-endpoint, non-inferiority trial
- Randomization
- Yes
- Blinding
- Blinded outcome assessment; open-label intervention (outcome assessors masked to treatment allocation)
- Sample Size
- 4922
- Follow-up
- 90 days
- Centers
- 114
- Countries
- Australia, Chile, China, Malaysia, Mexico, UK, USA, Viet Nam
Primary Outcome
Definition: Unfavourable functional outcome defined by mRS score 2-6 at 90 days
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 606/1963 (30.9%) | 809/2552 (31.7%) | - (0.92-1.15) | 0.057 (non-inferiority) |
Limitations & Criticisms
- Upper bound of 95% CI just touches the non-inferiority margin (1.15), providing only 'weak evidence' of non-inferiority
- Open-label design may introduce observer bias despite blinded outcome assessment
- Non-inferiority margin derived from thrombolysis trials in all stroke patients, not specifically mild stroke (NIHSS <10) who have better prognosis
- Stepped-wedge design with guardian consent may have compromised data quality
- COVID-19 pandemic disrupted services and created preference for low-intensity monitoring, potentially biasing results
- Imbalance in group sizes (2789 vs 2133) due to accelerated adoption of low-intensity monitoring
- Secular confounding from sequential implementation cannot be excluded
- Modest non-significant increase in mortality with low-intensity monitoring (2.2% vs 1.9%)
- Approximately half of participants had NIHSS ≤5, where thrombolysis benefit over DAPT is unclear
Citation
Lancet 2025; 405: 1909–22