Highlights
– In a three-arm, randomised phase 3 trial (PreMeFen), inhalational methoxyflurane produced a mean pain reduction of 3.31 NRS points at 10 minutes and was non-inferior to both intranasal fentanyl and intravenous morphine in the prehospital ambulance setting.
– Methoxyflurane compared with intranasal fentanyl showed a greater mean reduction in pain at 10 minutes (adjusted difference -1.33, 95% CI -2.01 to -0.64), meeting non-inferiority and suggesting superiority at this early time point.
– Intranasal fentanyl failed to demonstrate non-inferiority to intravenous morphine at 10 minutes; adverse event rates were similar across groups.
Background
Adequate early analgesia for acute painful conditions is a core element of emergency medical care. In the prehospital environment, delays to intravenous (IV) access, variability in provider skills, patient distress, and operational constraints commonly contribute to undertreatment of pain. Non-intravenous options that can be administered rapidly and safely by paramedics or self-administered by patients have therefore attracted interest as bridging strategies until definitive treatment is available.
Methoxyflurane (delivered via a portable inhaler; Penthrox is a commonly known preparation) is an inhalational analgesic historically used at high doses as an anaesthetic in the 1960s–70s; at the low analgesic doses used in modern emergency settings it has a rapid onset, short duration of action, and an established pharmacologic profile. Intranasal (IN) fentanyl delivers a potent opioid without IV access, and IV morphine remains a widely used benchmark for moderate-to-severe acute pain. Head-to-head trials comparing these approaches specifically in the prehospital setting have been limited.
Study design
PreMeFen (ClinicalTrials.gov NCT05137184) is a randomised, open-label, non-inferiority, three-arm, phase 3 trial conducted in the Innlandet Hospital Trust ground ambulance service area in Norway. Adults with traumatic or medical acute pain scoring ≥4 on the 0–10 Numeric Rating Scale (NRS) were randomised 1:1:1 to receive:
- 3 mL inhalational methoxyflurane (single administration via inhaler, titrated as per protocol),
- intranasal fentanyl (age-dependent dose: 50 µg for older patients, 100 µg for younger adults), or
- intravenous morphine (age-dependent dose: 0.05 mg/kg for older patients, 0.1 mg/kg for younger adults).
The primary endpoint was the change in NRS pain score from baseline to 10 minutes after treatment start, analysed per protocol. The prespecified non-inferiority margin was 1.3 NRS points. Secondary outcomes included later pain measurements, rescue analgesia use, and safety/adverse events. The trial enrolled 338 participants (methoxyflurane n=112, fentanyl n=115, morphine n=111), with 281 included in the per-protocol analysis.
Key findings
Primary outcome
Baseline NRS was 7.6 (SD 1.8). At 10 minutes, mean NRS changes (absolute reduction from baseline) were:
- Methoxyflurane: -3.31 (SD 2.67)
- Intranasal fentanyl: -1.98 (SD 2.28)
- Intravenous morphine: -2.74 (SD 2.12)
Adjusted comparisons (methoxyflurane minus comparator) showed:
- Methoxyflurane vs intranasal fentanyl: -1.33 (95% CI -2.01 to -0.64)
- Methoxyflurane vs intravenous morphine: -0.36 (95% CI -1.03 to 0.31)
- Intranasal fentanyl vs intravenous morphine: 0.91 (95% CI 0.27 to 1.55)
Interpretation of these estimates with the pre-specified non-inferiority margin of 1.3 NRS points:
- Methoxyflurane was non-inferior to intranasal fentanyl; the CI is wholly below +1.3 and in fact lies below zero, indicating that methoxyflurane provided greater mean reduction at 10 minutes (the CI excludes no difference and suggests superiority at this early time point).
- Methoxyflurane was non-inferior to IV morphine; the upper bound of the CI (0.31) is below the margin of 1.3, but the CI crosses zero so superiority is not demonstrated.
- Intranasal fentanyl failed to demonstrate non-inferiority to IV morphine; the upper CI bound (1.55) exceeds the non-inferiority margin.
Clinical significance
The trial chose an NRS non-inferiority margin of 1.3, consistent with commonly reported minimal clinically important differences (MCID) for acute pain. All three study arms achieved numerically meaningful pain reductions by 10 minutes (methoxyflurane and morphine reductions exceed 2 points), whereas intranasal fentanyl reduced pain by a mean of ~2 points, bordering on the threshold of robust clinical effect. The rapid, larger mean reduction with methoxyflurane—particularly vs intranasal fentanyl—supports its role as an effective early analgesic in the field.
Safety
Adverse events occurred at similar rates across groups: 24% (26/109) in the morphine arm, 24% (27/112) in the fentanyl arm, and 22% (24/111) in the methoxyflurane arm. Two serious adverse events—respiratory depression (grade 2) and loss of consciousness (grade 3)—occurred in the same patient assigned to methoxyflurane. There were no treatment-related deaths.
The overall safety profile showed no major differences between groups in this sample, but the occurrence of serious events in one methoxyflurane-treated patient highlights the need for standard monitoring after administration, especially if used with other sedating agents or in patients with comorbidities.
Expert commentary and interpretation
PreMeFen addresses a pragmatic clinical question: which readily deliverable analgesic provides the most rapid and reliable pain relief in the ambulance? The findings are practice‑relevant for prehospital clinicians and services designing analgesia protocols.
Strengths of the trial include randomised allocation, a clearly prespecified non-inferiority margin anchored to a clinically meaningful NRS value, and direct comparison of three contemporaneously used strategies in routine prehospital care. The per-protocol primary analysis is consistent with non-inferiority trial conventions, although sensitivity analyses using intention-to-treat would be informative for robustness.
Limitations include the open-label design, which is difficult to avoid given the different routes and devices used; potential performance and expectation effects could influence subjective pain scores. The primary endpoint at 10 minutes emphasises very early analgesia—which is clinically valuable in acute trauma or severe pain—but does not capture durability of effect beyond the early phase. Also, dosing strategies were age-stratified rather than weight-titrated for intranasal fentanyl, which may have affected comparative potency for some patients. Finally, this single-region Norwegian ambulance system may differ from other systems in staffing, training, and patient populations, affecting generalisability.
Mechanistically, inhaled methoxyflurane rapidly reaches central nervous system targets via pulmonary uptake, producing analgesia without the need for injection. Intranasal fentanyl provides rapid systemic opioid absorption but can be affected by nasal mucosa conditions and dosing constraints. IV morphine remains a robust comparator but requires IV access and monitoring. The PreMeFen data suggest that when rapid, simple, non-IV analgesia is required, methoxyflurane is a viable and at times superior early option.
Implications for practice
For prehospital services, the trial supports offering inhalational methoxyflurane as an early, non-intravenous analgesic option that can be initiated rapidly, even before IV access is obtained. Methoxyflurane can function as a bridge to longer‑acting parenteral opioids or regional analgesia when indicated. Intranasal fentanyl remains useful, particularly when opioid titration or intravenous access will not be feasible, but at the doses used in PreMeFen it did not match IV morphine at 10 minutes and was less effective than methoxyflurane.
Implementation should be accompanied by training on patient selection, dosing, monitoring for respiratory depression or sedation, and clear protocols for escalation or conversion to longer-acting analgesics. Particular caution is warranted in patients with polypharmacy or significant cardiorespiratory disease.
Research and policy gaps
Further work could address:
- Longer-term analgesic efficacy and functional outcomes (beyond the 10-minute window), including need for rescue analgesia and total opioid consumption.
- Effectiveness in diverse EMS systems, different patient populations (including pediatric groups), and specific pain etiologies (e.g., isolated extremity fracture vs chest pain).
- Direct comparisons with alternative non-invasive options and with weight‑based intranasal dosing strategies.
- Health-economics analyses considering time to analgesia, resource use (IV cannulation), and patient-centered outcomes.
Conclusion
The PreMeFen randomised trial demonstrates that inhalational methoxyflurane provides rapid and clinically meaningful analgesia in the prehospital setting and is non-inferior to intranasal fentanyl and intravenous morphine at 10 minutes. Methoxyflurane appears particularly effective compared with intranasal fentanyl at this early time point and offers a practical non‑intravenous bridging option. Prehospital services should consider methoxyflurane as part of multimodal analgesia protocols while ensuring appropriate monitoring and escalation pathways.
Funding and trial registration
Funding: Norwegian Air Ambulance Foundation and Innlandet Hospital Trust.
Trial registration: ClinicalTrials.gov NCT05137184.
References
Simensen R, Fjose LO, Thorsen K, Olsen IC, Rehn M, Hagemo J, Smalberget L, Heyerdahl F. Comparison of inhalational methoxyflurane, intranasal fentanyl, and intravenous morphine for treatment of prehospital acute pain in Norway (PreMeFen): a randomised, non-inferiority, three-arm, phase 3 trial. Lancet. 2025 Nov 20:S0140-6736(25)01575-2. doi: 10.1016/S0140-6736(25)01575-2. Epub ahead of print. PMID: 41275876.
Thumbnail Prompt
Realistic photo-style illustration of an ambulance interior: a paramedic assisting an adult patient who is using a small handheld inhaler (methoxyflurane/Penthrox-style); another paramedic prepares an intranasal spray and an IV line is visible but not yet connected. The scene conveys rapid, compassionate prehospital care; neutral clinical colors; diverse adult patient (middle-aged), shallow depth of field, high detail.