Prophylaxis of opioid-induced nausea in helicopter emergency medical service: a prospective quasi-experimental study Authors: Jürgen Knapp, Robin Strohm, Markus Huber, Roland Albrecht, Urs Pietsch Published: 2026 ABSTRACT AbstractObjectiveIn prehospital emergency medicine, opioids are usually used to treat acute traumatic pain. A common side effect of opioids is nausea. To prevent this, the prophylactic administration of 5-HT3 antagonists such as ondansetron has become established in clinical practice, even though there is currently no evidence to support its use. Aromatherapy is also used as an alternative non-pharmacological method for the treatment of nausea. We therefore investigated whether the prophylactic administration of ondansetron or prophylactic aromatherapy after opioid administration is beneficial to the patient.MethodsPatients over the age of 16 who received opioid therapy due to acute traumatic pain by the crews of a total of six HEMS bases between February 2024 and April 2025 were prospectively included. Nausea prophylaxis after opioid-based pain therapy was performed according to a fixed rotating 7-day schedule: no nausea prophylaxis, medication-based nausea prophylaxis with ondansetron 4 mg intravenously, and aromatherapy-based nausea prophylaxis by inhalation of isopropanol. A nausea score was evaluated at the time of opioid administration and upon handover at the destination hospital, using a numeric rating scale (NRS) between 0 (no nausea) and 10 (maximum nausea). Primary endpoint was the change in NRS between the time before opioid administration and the time of patient handover at the emergency department. Secondary endpoints were incidence of moderate and severe nausea (defined as an NRS > 3), the incidence of vomiting during prehospital patient care and need of rescue medication (ondansetron for therapeutic use).ResultsA total of 205 patients were included in the analysis. Eighty-six in the “no prophylaxis” group, 64 in the ondansetron group, and 55 in the aromatherapy group. We found a median change in NRS of 0 (interquartile range 0–0) points across all groups. An increase in NRS was observed in 15% of patients in the no prophylaxis group, 14% in the ondansetron group, and 15% in the aromatherapy group. An NRS > 3 was observed in 14% of patients in the no prophylaxis group and in 11% of patients in both the ondansetron and aromatherapy groups. Vomiting occurred in 4% of patients in the no prophylaxis group, 2% in the ondansetron group, and in none of the aromatherapy group. Therapeutic administration of ondansetron was necessary in 4 patients in the “no prophylaxis” group and in one patient in the aromatherapy group. None of the patients experienced any improvement in nausea as a result, three of these patients from the “no prophylaxis” group vomited after ondansetron administration.ConclusionNeither the prophylactic administration of ondansetron nor aromatherapy seem to be beneficial for the prevention of opioid-induced nausea in unselected patients in the setting of prehospital emergency medicine. Larger randomized controlled trials are needed to clarify whether certain subgroups may benefit.Trial registrationNot applicable.Keywords: Prehospital emergency medicine, HEMS, Nausea prophylaxis, Pain management, Opioids Background The use of opioids is widespread in prehospital emergency medicine to provide adequate analgesia [1–3]. Previous studies have shown an incidence of opioid-induced nausea in emergency department patients between 2 and 20%.4 Nausea not only significantly impairs patient comfort, but can also lead to tachycardia and blood pressure fluctuations or vomiting, with the risk of aspiration or delays of rescue process and thus a clinical safety hazard to the patient. The prophylactic administration of antiemetics (especially 5-HT3 antagonists as ondansetron) in the context of pain therapy with opioids has become widely established in prehospital emergency medicine. However, this indication is an off-label use with no evidence of its effectiveness in this setting and was simply transferred from recommendations for in-hospital anesthesia for prophylaxis of postoperative nausea and vomiting (PONV) after general anesthesia. On the other hand, there is evidence from randomized controlled trials and a meta-analysis that aromatherapy through inhalation of isopropyl alcohol is superior in relief of nausea especially in the postoperative and emergency department setting compared to 5-HT3 antagonists [4–6]. This method is extremely cost-effective and easily available, as the patient simply has to sniff a disinfectant swab, and has no known side effects. So, some colleagues have adopted this method in their daily practice, even though data on its effectiveness for prophylactic use is still lacking. We therefore conducted a prospective quasi-experimental study to determine whether any of these prophylactic measures are effective in preventing opioid-induced nausea in the setting of helicopter emergency medicine service (HEMS). Methods Study design and participants The study was conducted at three HEMS bases of Swiss Air Rescue Rega between February 2024 and April 2025 and between June 2024 and April 2025 at three additional bases. Adult patients over the age of 16 who received opioid therapy due to acute traumatic pain were included. Fentanyl is available as an opioid within our air ambulance service. At the discretion of the treating team, ketamine may be administered as an adjunct to pain management. Exclusion criteria were life-threatening conditions, head injuries with neurological impairment (defined as a loss of consciousness occurring within the context of the accident, retrograde or anterograde amnesia, or a Glasgow Coma Scale score of < 15), allergies to the prophylactic medications, and the patient’s refusal to use disinfectant swabs for prophylactic aromatherapy. A nausea score between 0 and 10 was evaluated immediately before opioid administration and upon handover at the destination hospital, with 0 corresponding to no nausea and 10 to maximum nausea, subjectively assessed using a numeric rating scale (NRS). Procedures The following three procedures were applied after opioid-based pain therapy (using fentanyl, dosage as determined by the medical team) according to a fixed rotating 7-day schedule: no nausea prophylaxis in week one, medication-based nausea prophylaxis with ondansetron 4 mg intravenously immediately after the first dose of opioid (independent of the NRS prior to opioid administration) in week two, and aromatherapy-based nausea prophylaxis by inhalation of isopropanol also immediately after the first dose of opioid and independent of the NRS prior to opioid administration in week three. The patients in the aromatherapy group sniffed a disinfectant swab (SOFTA Swabs Reinigungstupfer, B. Braun, Melsungen, Germany), containing 80% isopropanol. They were instructed to take 3–5 deep breaths at a distance of 1–2 cm from the nose, repeated as needed after 10 min. In order to avoid a placebo effect arising from the aromatherapy instructions compared with the ondansetron group, patients in both groups were informed, in accordance with the study protocol, that the inhalation or intravenous administration of the medication was intended to prevent nausea. Therapeutic administration of ondansetron was permitted at any time in the aromatherapy as well as in the “no prophylaxis” group if the patient complained of increasing or severe nausea after receiving opioids, according to the assessment of the treating team. All interventions, severity of nausea (NRS), vomiting, and need of rescue medication were documented prospectively in accordance with the study protocol and reviewed monthly by an external evaluator. All data and patient details were extracted from the corresponding mission logs in the Rega database in a fully de-identified, anonymized, and non-traceable format. Outcomes The primary endpoint was the change in NRS between the time before opioid administration and the time of patient handover at the emergency department. Secondary endpoints were incidence of clinically relevant nausea at the time of patient handover, the incidence of vomiting during prehospital patient care and need of rescue medication (ondansetron for therapeutic use). A cut-off of > 3 on the NRS was used to define clinically relevant nausea. In pain research, NRS values above 3–4 are commonly considered to reflect clinically relevant symptoms requiring treatment [7]. Given the absence of validated cut-offs for nausea, this threshold was applied pragmatically to identify patients with clinically meaningful symptoms of nausea. Ethics The study was approved by the EKOS Ethics Committee (EKOS23/220). Statistical analysis The sample size was informed by the minimal clinically important difference (MCID) of 2.0 points on the numeric rating scale (NRS) as suggested in previous studies on nausea in the emergency department [8]. Assuming a standard deviation of 3.0 based on a previous randomized controlled trial on the therapeutic use of ondansetron and aromatherapy in an emergency department setting,6 a three-group comparison (one-way ANOVA, α = 0.05, power = 80%) would require approximately 90 participants (30 per group). To ensure adequate power under potential deviations from the assumed parameters (e.g., higher variability), the sample size was increased to 40 participants per group. In terms of descriptive statistics, categorical variables were summarized with counts and frequencies and numerical variables with medians and interquartile ranges (IQR). Unadjusted group comparisons were based on standard statistical tests: chi-square tests or Fisher’s exacts test for categorical variables and Kruskal-Wallis tests for numerical variables. The dichotomous outcome discriminating between patients with an increase in their nausea score (NRShospital arrival - NRSbefore opioids >0) and those whose nausea score remained unchanged or improved (NRShospital arrival - NRSbefore opioids ≤0) was examined by means of logistic regression model with treatment, age, sex and ketamine administration as covariates. Given the exploratory nature of the secondary outcomes, no P-value adjustment for multiple comparisons was performed. All computations were performed with R version 4.4.2. Results A total of 205 participants were included in the analysis (“no prophylaxis” group n = 86, ondansetron group n = 64, aromatherapy group n = 55, Fig. 1). The groups did not differ with respect to age, sex, or baseline nausea scores (Table 1). Median dose of fentanyl was 100 µg (IQR 100–200 µg). In 29 patients, ketamine was administered in addition to fentanyl (median dose 20 mg, IQR 15–25 mg). Flow diagram of patient inclusion and allocation to the groups Table 1Demographics, pain medication, numeric rating scale (NRS) of nausea, primary outcome and secondary outcomesAllAromatherapyNo prophylaxisOndansetron P n = 205n = 55n = 85n = 65 Demographics Age (years)51 [33;61]47 [34;59]54 [38;62]49 [30;61]0.404Sex (female)98 (47.8%)29 (52.7%)36 (42.4%)33 (50.8%)0.412Mono-trauma (Yes)192 (93.7%)53 (96.4%)81 (95.3%)58 (89.2%)0.255 Pain medication Fentanyl dosage (µg) 100 [100;200]100 [100;200]100 [100;200]100 [100;200]Ketamine (Yes):29 (14.1%)12 (21.8%)10 (11.8%)7 (10.8%)0.159 NRS NRS before opioids NRS (original scale)0.0 [0.0;0.0]0.0 [0.0;0.0]0.0 [0.0;0.0]0.0 [0.0;0.0]0.502 NRS > 038 (18.5%)12 (21.8%)13 (15.3%)13 (20.0%)0.584 NRS hospital arrival NRS (original scale)0.0 [0.0;1.0]0.0 [0.0;1.5]0.0 [0.0;0.0]0.0 [0.0;2.0]0.851 NRS > 057 (27.8%)17 (30.9%)21 (24.7%)19 (29.2%)0.692 NRS > 325 (12.2%)6 (10.9%)12 (14.1%)7 (10.8%)0.778 Primary outcome Change in NRS NRS (original scale)0.0 [0.0;0.0]0.0 [0.0;0.0]0.0 [0.0;0.0]0.0 [0.0;0.0]0.217 NRS (categorical scale)0.969 Δ (NRShospital arrival - NRSbefore opioids) ≤ 0175 (85.4%)47 (85.5%)72 (84.7%)56 (86.2%) Δ (NRShospital arrival - NRSbefore opioids) > 030 (14.6%)8 (14.5%)13 (15.3%)9 (13.8%) Secondary outcomes Rescue (Yes):5 (2.4%)1 (1.8%)4 (4.7%)n.a.0.179 Vomitus (Yes):4 (2.0%)0 (0.0%)3 (3.5%)1 (1.5%)0.467NRS: numeric rating scale Demographics, pain medication, numeric rating scale (NRS) of nausea, primary outcome and secondary outcomes NRS: numeric rating scale There was no difference between the groups in terms of a worsening of nausea scores or unchanged or improved nausea scores before opioid administration compared to the time of hospital admission (Table 1; Fig. 2). Incidence of moderate and severe nausea (NRS > 3) on hospital admission was 12% (14% in female patients, 10% in male patients). An increase of NRS after opioid administration was observed in 20% of female patients and 10% of male patients. Changes in numeric rating scale of nausea between the timepoint immediately before opioid administration compared to the time of hospital admission in the group treated with ondansetron prophylactically, no nausea prophylaxis and “aromatherapy” for nausea prophylaxis. No differences between the groups Therapeutic administration of ondansetron (in the sense of a “rescue medication”) was necessary in 4 patients in the “no prophylaxis” group and in one patient in the aromatherapy group. None of the patients experienced any improvement in nausea afterwards, three of these patients from the “no prophylaxis” group vomited after “rescue” ondansetron administration. Logistic regression analysis is shown in Table 2. Neither type of prophylactic treatment, nor sex or administration of ketamine appear to have any influence on outcome. Only with regard to age do younger patients appear to have a higher odds ratio for a deterioration in nausea scores. Table 2Logistic regression analysis of unchanged or improved nausea score (NRShospital arrival - NRSbefore opiods ≤0) with treatment, age, sex and ketamine administration as covariatesCharacteristicUnadjustedAdjustedOR95% CI P OR95% CI P Treatment Aroma———— No prophylaxis0.940.35, 2.41> 0.90.980.35, 2.61> 0.9 Ondansetron1.060.37, 2.98> 0.91.050.36, 3.05> 0.9Age (Years)0.970.95, 1.00.023Sex (Male vs. Female)1.980.89, 4.590.10Ketamine (Yes vs. No)1.080.37, 3.960.9CI = Confidence Interval, OR = Odds Ratio Logistic regression analysis of unchanged or improved nausea score (NRShospital arrival - NRSbefore opiods ≤0) with treatment, age, sex and ketamine administration as covariates CI = Confidence Interval, OR = Odds Ratio Discussion To our knowledge, this is the first study to evaluate the effectiveness of different prophylactic strategies for opioid-induced nausea in the setting of prehospital emergency medicine. Although, oral ondansetron offers a convenient and non-invasive alternative with good bioavailability in stable patients, intravenous ondansetron was selected in our study to ensure rapid onset of action and reliable bioavailability, particularly in patients treated with opioids, where oral absorption may be impaired or delayed. Furthermore, as an intravenous line had already been established in all patients due to the administration of opioids, the intravenous administration of ondansetron was the obvious choice. On the other hand, the inhalation of isopropyl alcohol offers reliable bioavailability even in the presence of nausea or during opioid therapy, and does not require any invasive procedures, making it highly suitable for the prehospital setting. This was one of the reasons behind our study. Patients were allocated to treatment groups according to a rotational scheme. However, due to practical and logistical factors in the prehospital setting, including variability in the feasibility and implementation of aromatherapy, which required additional explanation and patients’ cooperation. This reflects the pragmatic nature of our study design. However, this may have affected inclusion rates in this group. As a result, unequal group sizes were observed despite the predefined allocation scheme. Overall incidence of nausea and was high in our group of patients receiving prehospital opioid-based pain therapy for mainly mono-trauma. One reason for the higher incidence than in the emergency room could be the additional effect of motion sickness caused by the typically backward, reclining transport in the helicopter, without the possibility of looking outside, which typically reduces motion sickness. The relatively low intensity of nausea observed in this study may in part be explained by the timing and conditions of symptom assessment. Due to the limited ability to reliably assess symptoms during helicopter transport, NRS values were obtained upon hospital arrival. Consequently, peak nausea intensity during transport may not have been captured. Depending on transport duration, symptoms may have either already decreased or not yet reached their maximum at the time of assessment. Our findings demonstrated no significant benefit from either pharmacologic prophylaxis or aromatherapy. In concordance to our results, previous studies also found that prophylactic intravenous antiemetics do not make a significant difference in decreasing opioid-induced nausea and vomiting in the setting of emergency departments [9, 10]. Also, the therapeutic use of ondansetron for opioid-induced nausea and vomiting was found to be ineffective compared to placebo [11]. This corresponds with our findings that none of the five patients who received ondansetron for therapeutic purposes after complaining about increasing or severe nausea experienced a reduction in nausea, and three of these patients even had to vomit later on. Rescue antiemetic therapy differed between groups, as therapeutic administration of ondansetron was permitted in the no prophylaxis and aromatherapy groups but not repeated in the ondansetron group. This approach was chosen to avoid dose escalation of the same agent; however, it may have influenced symptom progression and should be considered when interpreting the results. Alternative antiemetics (such as metoclopramide) were not available in the prehospital setting. No data is currently available on the prophylactic use of aromatherapy to prevent opioid-induced nausea. According to our findings, aromatherapy does not appear to be beneficial in this regard. Existing data on the therapeutic use of aromatherapy for PONV is inconsistent. In ambulatory surgery patients one study found no difference in the rate of PONV with aromatherapy, whereas another study showed a significant reduction [12, 13]. The results for female patients after gynecological surgery (proofing a benefit from aromatherapy) are similarly contradictory when compared to female patients after general surgery (where no benefit could be demonstrated) [14, 15]. Although, according to the regression analysis, gender does not appear to be a significant influencing factor, given the magnitude of a possible effect between males and females – despite its large uncertainty – warrants further investigation. Numerous studies have shown that women are more likely to experience nausea and vomiting when given opioids [16–19]. Therefore, further studies should investigate whether female patients in particular may benefit from nausea prophylaxis in the context of opioid administration. According to existing data, also the subgroup of patients with known motion sickness and known history nausea after opioid use, may benefit from administration of prophylactic antiemetics when being treated with parenteral opioids [10]. Our findings regarding age also correspond with in-hospital experience regarding PONV, which young adults suffer from more frequently than older people [17–19]. Limitations The study has several limitations. First, we could not rule out the influence of certain confounding factors, such as the susceptibility to motion sickness, the type of injury, the duration of transport, and the transport position (lying down facing backwards or sitting facing forwards). Second, we also did not document how reliably the patients actually used the aromatherapy and inhaled deeply with the disinfectant swab directly in front of the nose. On the other hand, this reflects the practicality of the prophylactic aromatherapy in a real prehospital setting. Third, due to the rotating weekly schedule transportation conditions, particularly adverse weather such as turbulence or strong wind, as well as staffing differences and case mix variation may have influenced the occurrence and perception of nausea. Fourth, the open-label design without placebo control may have introduced expectation and observer bias, particularly given the subjective nature of nausea assessment using the NRS. However, as patients receiving both pharmacological prophylaxis and aromatherapy were informed about the intended antiemetic effects of these interventions, no advantage was observed compared to the group without prophylaxis. Nevertheless, the absence of a placebo comparator limits the ability to distinguish specific treatment effects from non-specific or contextual influences. These factors should be considered when interpreting the observed effects, especially for non-pharmacological interventions. Conclusion Based on our evaluation, neither the prophylactic administration of ondansetron nor aromatherapy were beneficial for the prevention of opioid-induced nausea in unselected patients in the setting of prehospital emergency medicine. Routine nausea prophylaxis with parenteral administration of opioids therefore does not appear to be indicated. Larger randomized controlled trials are needed to clarify whether certain subgroups may benefit. Acknowledgements We would like to thank all HEMS crews who supported us in conducting this study. Author contributions JK, RS, and UP conceived the study. MH provided statistical advice on study design, analyzed the data and prepared figure 2 and the tables . JK and RS drafted the manuscript. RA provided organisational support and contributed to the writing of the manscript. All authors reviewed the manuscript. Funding None. Data availability The datasets generated and analysed in this study are available upon reasonable request and can be obtained from the corresponding author. Declarations Ethics approval and consent to participate The study was approved by the EKOS Ethics Committee (EKOS23/220). Consent to participate was waived due to a retrospective, anonymized evaluation of the data. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets generated and analysed in this study are available upon reasonable request and can be obtained from the corresponding author.