Differences in training among prehospital emergency physicians in Germany

Surgeons reported performing 89 (88.9) endotracheal intubations with direct laryngoscopy during their training, while internists performed 76.9 (64.9) intubations using direct laryngoscopy. A study by Konrad et al. [5] found a success rate of 90% after a mean of 57 intubations. However, after an initial steep increase in successful intubations, upon reaching an 80% success rate, the curve flattens considerably and larger numbers of intubations are required for modest further increases. Similar studies likewise found a number around 50 endotracheal intubations to correspond to a 90% success rate [6‐9]. Bernhard et al. [10] found similar results and reported that the success rate stabilizes around 95% only after more than 150 attempts in an operating room environment. On the basis of these studies, the German Society for Anesthesiology and Intensive Care has issued a guideline for out-of-hospital airway management [14] recommending a benchmark of 100 endotracheal intubations for learning the procedure. Non-anesthesiologists did not meet the recommended number of procedures and the high standard deviations indicate that there are significant differences between those applying for board certification. Likewise, a minimum of 50 procedures is recommended for learning endotracheal intubation using video laryngoscopy [14] which non-anesthesiologists also did not achieve.

Despite the fact that the laryngeal tube is still the most available “alternative” airway, all participants of this study had little experience with the laryngeal tube. This may be attributable to generally low utilization of this device in hospitals when compared with prehospital care. Non-anesthesiologists used a laryngeal tube more frequently out-of-hospital than anesthesiologists. We hypothesize that this is due to most anesthesiologists having little experience using this device and therefore choosing more familiar measures for securing the airway. The higher prehospital experience of non-anesthesiologists with this device might therefore be an expression of lacking expertise in endotracheal intubation.

Mohr et al. reported an 86% success rate for correct placement of a laryngeal mask airway after an experience of 40 performed procedures [15]. Current guidelines [14] recommend a minimum of 45 performed procedures for learning the skill of securing the airway using supraglottic devices. Surgeons did meet this requirement using laryngeal-mask airways, while internists did not meet this requirement regardless of the device used. It should be emphasized that supraglottic devices should only be used in the prehospital field when caregivers have previously undergone proper in-hospital training, as these devices are prone to malpositioning in inexperienced hands.

Cricothyroidotomy was rarely performed among all participants—though anesthesiologists and surgeons reported higher than expected numbers, which may signal that this question was misunderstood. To the best of our knowledge no data describing procedural volume required for skill mastery is available. Some reference procedural volumes can only be derived from mannequin studies, although the transferability of these results is questionable—especially for airway management skills [16]. Wong et al. [17] reported that after five attempts 96% of participants were able to successfully perform cricothyroidotomy and success rates plateaued after five attempts. Based on this study and the rare opportunities to learn this skill in clinical practice, a minimum of five cricothyroidotomies on a skill simulator could be recommended. However, simulator training was not the subject of this study.

Komatsu et al. [8] reported a success rate of 80% after a mean performance of 27 bag-valve-mask ventilations. However, this appears to be a relatively small number and previous studies have shown that bag-valve-mask ventilation can be associated with a high rate of serious complications [18]. Current guidelines therefore recommend a minimum of 100 performed bag-valve-mask ventilations to adequately learn the skill [14]. Although on average participants met the requirements, the high standard deviation indicates that some participants did not.

While anesthesiologists had significantly more experience in mechanical ventilation, the experience for non-invasive ventilation was similar between anesthesiologists and internists. Surgeons reported only 8.8 (18.5) performed non-invasive ventilations in the prehospital field and 7.3 (12.4) in-hospital. To the best of our knowledge, there are no recommendations available regarding a minimum number to have proficiency with these procedures. However, we assumed that the reported numbers indicate a lack of expertise in this skill for the surgical participants of this study.

Prior work demonstrated that the success rate for intravenous access is 80% after 56 ± 43 performed procedures [9]. Establishing intravenous access can be uniquely challenging in the prehospital environment, necessitating greater experience and expertise. The participants of this study reported much higher numbers than those previously published to obtain proficiency.

Interestingly, non-anesthesiologists had more experience with placement of an intraosseous access—both in- and out-of-hospital. Intraosseous access is often used when intravenous access is unsuccessful; therefore, this finding may reflect a higher likelihood of success with intravenous access among anesthesiologists relative to non-anesthesiologists. There is no recommended minimum procedural number for learning how to establish intraosseous access, but studies have demonstrated that the procedure is easy to learn and training models can be utilized to learn successful needle placement [19].

Needle thoracostomy was rarely performed on a patient during training by all participants. In the prehospital setting, tension pneumothorax is a rare event in countries like Germany where gunshot or stab wounds are not common [20, 21]. Furthermore, in the hospital environment, it is more common to perform a tube thoracostomy, which is a more definitive treatment. Therefore, training has to utilize cadaver and other training models to learn this technique. To the best of our knowledge, there is no recommended number of procedures for mastering tube thoracostomy. Ball et al. [22] showed a complication rate of 28% for this procedure with the level of training not being a risk factor for complications. Davis et al. [23] defined a number of 10 thoracostomies to distinguish between “novices” and “experts”. Surgeons were the most experienced providers with 25.4 (31.3) performed procedures, while anesthesiologists performed 11.4 (17.4) tube thoracostomies and internists performed 7.2 (10.8) procedures. However, the high standard deviation indicates significant variability among individual participants.

Electrical cardioversion, external pacing and defibrillation were all more frequently performed by internists. There are no recommendations concerning training in these procedures in international guidelines on resuscitation [24]. However, surgeons in particular reported very little experience with these procedures.

Tourniquets were rarely used during training whether in- or out-of-hospital. This is consistent with a previous study we conducted regarding the experience of paramedic personnel [25]. Although the use of tourniquets is easy to learn, due to how rarely they are used in real-world settings, it may be necessary to provide additional practical training for prehospital emergency medicine trainees. Several studies have shown that tourniquets are often applied with too little pressure, and may occasionally cause increased bleeding due to venous stasis [26].

Pelvic binders were used more often than tourniquets, but overall were still used relatively infrequently. Much like the application of a tourniquet, the procedure is generally easy to learn, but may require focused education due to its relative rarity.

Point-of-care ultrasound is increasingly used in prehospital medicine as ultrasound machines have become smaller and more portable. Prehospital emergency physicians therefore have to be familiar with ultrasound protocols like the Focused Assessment with Sonography in Trauma (FAST) exam. Ma et al. reported an accuracy of 86.1% after 12 months of training or a total of 35 FAST exams [27]. Gracias et al. reported that the learning curve flatten out between 30 and 100 FAST examinations [28]. Chisholm et al. reported an increased competency in focused transthoracic echocardiography after a 10 h course and more than 45 performed ultrasounds. The German Society for Ultrasound in Medicine (DEGUM) has issued a certificate in Emergency Ultrasound [30]. As part of this certificate, candidates must provide proof of a minimum of 25 supervised FAST exams and 80 supervised focused echocardiography exams. This provides some reference for interpreting the results of our study. Based on this reference, and given the findings of the current study, it has to be assumed that anesthesiologists as well as non-anesthesiologists are able to perform a sufficient FAST exam. However, anesthesiologists did not meet the recommended minimum of 80 supervised focused echocardiography exams.

The formal educational requirements for the subspecialty of prehospital emergency medicine do not incorporate a minimum number of performed procedures or require evidence of the number of performed procedures. Therefore, our data reflect participants’ estimates of the number of procedures they performed during training. Given the survey format, we cannot exclude the possibility of selection or non-response bias. However, with a response rate of 50.67%, the study would in general be considered representative. As the survey was conducted after passing the board exam, respondent anticipation of negative consequences is unlikely to have impacted response rate. As there are no formal requirements concerning procedural volume and participants of the study had just successfully passed the board exam, a self-disclosure bias is unlikely to be a source of relevant systematic error. Given a mean clinical experience of the respondents of 5.9 years, it can be assumed that there is some degree of recall-bias that has to be taken into account when interpreting the results of this study. However, especially for rarely performed procedures, it is likely that the participants were able to recall the number of performed procedures very accurately. In higher numbers—meaning that one cannot recall but only roughly estimate the number of performed procedures—the estimation will likely be less accurate. Independent of the specific procedure, learning curves usually flatten out with a higher number of performed procedures. Meaning that differences between higher procedural numbers usually should not represent marked differences in competency, and this limitation is therefore less relevant for answering our research question. Furthermore, the reported estimates are generally plausible and in accordance with previous study results [13, 31].

The consistently high standard deviation likely indicates large interindividual differences in the number of performed procedures. Given these limitations, these results should be interpreted with some caution.