RIGHT-2 Trial: Ultra-Acute Prehospital Glyceryl Trinitrate in Presumed Stroke
In patients with presumed stroke and elevated blood pressure, does transdermal glyceryl trinitrate started by paramedics in the ambulance within 4 hours of symptom onset improve 90-day functional outcome compared with sham?
RIGHT-2 Investigators (Lancet 2019) · doi:10.1016/S0140-6736(19)30194-1 · 1149 patients
Population
Included
- Adults (18 years or older) with presumed acute stroke
- FAST score 2 or 3 on paramedic assessment
- Systolic blood pressure 120 mm Hg or higher
- Able to start treatment within 4 hours of symptom onset
- Randomized and treated by trained paramedic in the ambulance
- Adult patients (18 years of age or older)
- Presumed stroke after an emergency call to a participating ambulance service
- Presentation within 4 hours of onset of symptoms
- Able to be taken to a participating hospital
- Face-Arm-Speech-Time (FAST) score of 2 or 3 (thus ensuring the presence of motor weakness)
- Systolic blood pressure of 120 mm Hg or higher
Excluded
- Residence in a nursing home
- Reduced consciousness (GCS below 8)
- Hypoglycemia (capillary glucose below 2.5 mmol/L)
- Witnessed seizure at onset
- Recent nitrate use or PDE-5 inhibitor use
- Patients from a nursing home
- Reduced consciousness (Glasgow Coma Scale score below 8 of 15)
- Hypoglycaemia (capillary glucose concentration below 2.5 mmol/L)
- A witnessed seizure at onset
Source: RIGHT-2 Investigators, Lancet 2019;393:1009–1020· Retrieved 2026-06-09
Primary Outcome · mRS Shift at 90 Days
Confirmed stroke or TIA (cohort 1, N=852); full ITT (cohort 2, N=1149)
In patients with confirmed stroke or transient ischemic attack (cohort 1, N=852), the distribution of mRS scores at 90 days did not differ significantly between groups. Median mRS was 3 (IQR 2 to 5) in both arms. The adjusted common odds ratio for a poor outcome was 1.25 (95% CI 0.97 to 1.60, p=0.083), a trend favoring sham that did not reach statistical significance. In the full intention-to-treat population (cohort 2, N=1149), the common odds ratio was 1.04 (95% CI 0.84 to 1.29, p=0.69).
Distribution chart not shown. See source Figure 2 for per-segment breakdown.
Safety Signals
Symptomatic hypotension occurred in 21 of 568 (4%) GTN patients versus 9 of 581 (2%) sham patients (adjusted OR 2.49, 95% CI 1.11 to 5.57). In the ICH subgroup, GTN was associated with larger hematoma (adjusted OR 1.95, 95% CI 1.07 to 3.58) and more mass effect on imaging (adjusted OR 2.42, 95% CI 1.26 to 4.68). Serious adverse events were 188 (33%) with GTN versus 170 (29%) with sham (p=0.16, not significant). The trial was not stopped for harm; these findings are considered hypothesis-generating, particularly for the ICH subgroup.
Study Arms
- Agent
- Glyceryl trinitrate (nitroglycerin) transdermal patch (Transiderm-Nitro 5; Novartis)
- Dose
- 5 mg once daily
- Route
- Transdermal patch (placed on the shoulder, back, or chest; site changed daily)
- Frequency
- One patch once daily
- Duration
- Up to 4 days (first patch in the ambulance, further patches in hospital)
- Co-interventions
- Standard prehospital and in-hospital stroke care, including thrombolysis and endovascular therapy as indicated after the final diagnosis was made.
The first patch was applied by the paramedic in the ambulance immediately after randomisation; further patches were given in hospital for up to 3 more days. The design was sham-controlled, participant-masked, and blinded-endpoint. Patches were unlabelled and a gauze dressing was taped over the patch to maintain masking. Adherence was low: only about 36% of cohort 2 received all 4 days of treatment. Source: RIGHT-2 Investigators, Lancet 2019 p.1011 (Procedures).
- Agent
- Sham dressing (DuoDERM hydrocolloid dressing; Convatec)
- Route
- Transdermal (applied as a similar-appearing dressing)
- Frequency
- One dressing once daily
- Duration
- Up to 4 days
- Co-interventions
- Standard prehospital and in-hospital stroke care, identical to the GTN group.
A similar-appearing sham dressing was applied to maintain masking; patients were effectively masked because patches and dressings were unlabelled and an over-taped gauze dressing concealed the treatment. Paramedics were unmasked to treatment; participants and outcome assessors were masked. Source: RIGHT-2 Investigators, Lancet 2019 p.1011 (Randomisation and masking).
Trial Design
Type
- Paramedic-delivered, randomized, sham-controlled phase 3 trial
- Prehospital GTN vs sham dressing
- Blinded endpoint assessment
Timeline
United Kingdom; October 2015 to May 2018
N
1149
Enrollment
Randomized October 2015 to May 2018 across the United Kingdom. Paramedic-delivered, randomized, sham-controlled phase 3 trial with blinded endpoint assessment. 1149 patients enrolled across multiple UK ambulance services.
Bedside Pearl
In ultra-acute presumed stroke, transdermal GTN shows no functional benefit and a safety signal in intracerebral hemorrhage patients treated before imaging. The trial's main contribution is feasibility: UK paramedics can randomize and treat stroke patients in the ambulance. For blood pressure management in suspected stroke before imaging, RIGHT-2 argues against routine prehospital intervention.
See also