This observational cohort study examined patients who were treated with OTFC from July 2020 to May 2022. Reporting of the study conforms with the STROBE statement for the reporting of observational cohort studies. The local ethics committee evaluated and approved the study and waived written informed consent (BASEC Nr. 2020–00,096 EKOS 21/006).

All patients with severe pain, defined as a score of ≥ 4 on a numeric rating scale (NRS), with 0 defined as absence of pain and 10 being the maximum imaginable pain, who did not have i.v. access were included. If there was evidence of altered consciousness, alcohol consumption or noticeable abnormal vital signs, administration of OTFC was deemed to be contraindicated. Head trauma (GCS < 15) was an absolute contraindication for OTFC. Further OTFC contraindications were known allergy to the drug or its additives, patient refusal, or age < 14 years. Major adverse events were defined as apnea and severe bradypnea requiring ventilation as an expression of overdosing.

All EMS providers, including ski patrol rescuers, EMS paramedics, and prehospital emergency physicians, received theoretical and practical instruction about OTFC. Specifically, they were instructed about the mechanisms of action, indications and contraindications, as well as potential side effects. The study group consisted of trauma patients predominantly participating in recreational sport in the Ski and Bike resort Lenzerheide and St. Moritz (Switzerland), Ski World Cup (FIS Ski Worldcup St. Moritz, Lenzerheide, Switzerland), Mountainbike World cup (UCI World Cup Lenzerheide Downhill and Cross Country, Switzerland), Bike Park and MTB Stage Races (Swiss Epic, Bike Giro Engadin, Switzerland).

The medication used is OTFC 600/800 mcg (Actiq, Teva Pharma AG, Basel, Switzerland). The single dose was 600 mcg oral fentanyl for the first 65 patients. This amount was later changed to 800 mcg.

The original dose of 600 mcg was chosen according to the buccal absorption equivalent of about 20% of intravenous fentanyl. When employing this route of medication, the onset is nearly as fast as that of intravenous fentanyl, as there is no hepatic first pass effect.

Because of limited availability, the dose was changed to 800 mcg lozenges. There is hardly any uptake of the drug via the intestines [13]. Each EMS member participating in the study was additionally carrying a dose of 1.8 mg naloxone (Nyxoid, Mundipharma Medical Company, Basel, Switzerland) for nasal application as an antidote. Additional medication after extrication was only administered if it was indicated and a physician was on scene, which was predominantly the case in HEMS deployment.

Data collection was performed by the medical provider using an app-based online protocol with predefined endpoints, prospectively recording the data in Table 1 (except retrospective HEMS data). The patient’s and mission’s characteristics were examined, including vital signs (heart rate, oxygen saturation, AVPU scale, GCS and NRS) on initial presentation to the prehospital team and upon admission to the emergency department (mechanism of injury, type of transport, side effects and need for additional drugs). The overall safety profile for OTFC was determined by the incidence of adverse events, defined as interrupted use, nausea, drowsiness, or respiratory depression. No further personal data were recorded in order to ensure de-identification of patients. Also, no personal data of the first responders on scene were collected, so no conclusions about the rescuers could be drawn during later analysis. Additionally, the HEMS and EMS charts were analysed retrospectively in relation to further analgesia requirements or side effects during transport to the hospital (Table 1).

Distribution of absolute and relative reduction depending on pain level. Footnote: Graph A shows the pain level before and after administering OTFC. Graph B includes the absolute reduction depending on pain level before OTFC. Moreover, the effect of higher efficiency with higher NRS when administering fentanyl can be seen in both. Graph C includes the relative reduction in percentage. Here, we were able to show equal pain reduction depending on the initial pain level, with an average just below 40%

Patient characteristics are summarized and presented in tables. Continuous variables were summarized by mean ± standard deviation if normally distributed, or by median and the interquartile range if skewed. To increase the quality of data analysis, each set of data was tested for normal distribution (D’Agostino and Pearson omnibus normality test) and for homogeneity of variance (Levene’s test) before statistical analyses. To find differences between paired groups (e.g., NRS before and after administration of fentanyl), a paired t-test was used in case of normally distributed data, and a Wilcoxon matched-pairs signed rank test was used in case of not normally distributed data. To find differences between two unpaired groups, Student’s t-test was used in case of normally distributed data (with Welch’s correction in case of unequal variance), and a Mann–Whitney test was used in case data were not normally distributed. To find differences between multiple unpaired groups (e.g., reduction in NRS after application of fentanyl in different age groups), a one-way ANOVA and subsequent Tukey–Kramer multiple comparisons post-hoc test with a single pooled variance was used in case of homoscedastic data with normally distributed residuals. In case of heteroscedastic data with normally distributed residuals, a Brown-Forsythe and Welch ANOVA with subsequent Dunnett’s T3 multiple comparisons post-hoc test, and with individual variances computed for each comparison (n > 50 / group) was used. In case of not normally distributed residuals, a Kruskal–Wallis test with a subsequent Dunn’s multiple comparisons test was applied. To show the connection between pre-therapeutics NRS and effectiveness of therapy, simple linear regression analysis was used. Furthermore, a multiple linear regression with NRS-reduction as dependent and gender, age group and initial NRS as main effects was calculated, whereas male gender as well as the age group < 20 years old was used as reference. Statistical analyses were performed using IBM SPSS Statistics (Version 24.0.01, IBM SPSS, Chicago, IL, USA) and GraphPad Prism (version 9.4.0 for Windows, GraphPad Software, San Diego, CA, USA). Significance was accepted at P < 0.05 (two-sided for t-tests and exact for Mann–Whitney tests).