Introduction

Background

Since the first description of Focused Assessment with Sonography for Trauma (FAST), ultrasound examination by non-radiologists have considerably expanded to include, for example, deep venous thrombosis assessment, echocardiography and long bone fracture diagnosis [1, 2]. Interest in goal-directed sonography has also spread to primary care in remote areas [3–5], even when performed by non-medical staff and combined with telemedicine [6, 7].

Importance

The development of focused sonography in many specialities, leads to an increased diversity of practitioners and a broad new range of applications [2, 8]. This emphasizes the need for developing evaluation, standardization and quality control of training [8]. Additionally, the development of International Emergency Medicine and educational collaborations plead for a global consensus on ultrasound curriculum and physicians’ accreditation [8, 9]. In a Boston (MA, USA) and Tuscany (Italy) cooperative study, teaching language proficiency and intercultural communication were identified as key issues to be overcome for attaining training achievement [10]. Language and cultural barriers between teachers and learners, between the on-site ultrasound practitioner and referral radiologist, may impact on training efficacy, practitioner proficiency and interpretation of results.

Goal of this investigation

As part of a Sino-French educational collaboration, we set up a focused sonography introductory course. Firstly, we sought to determine if the training was effective and, secondly, if it was affected by simultaneous translation and cultural differences between French teachers and Chinese learners. The primary end point was the overall success rate of ultrasound image acquisition according to quality and time assessed by independent observers.

Materials and methods

Study design

This study had a single-setting, prospective, observational cohort design. It was approved by our institutional research ethics review board (Anzhen Hospital, Beijing, China).

Setting

The study was completed at the Sino-French Emergency Training Center (Beijing, China). Managed by both Chinese and French emergency medicine experts, the center provides training sessions for Chinese emergency physicians. Lectures and hands-on courses are given by French specialists, therefore requiring translation.

We organized a 2-day training session on goal-directed emergency sonography. The content and organization of this kind of training have been previously validated [11]. The curriculum is in accordance with the guidelines for implementation of an introductory emergency ultrasound course for emergency physicians [2]. It consisted of five lectures based on physics and ultrasound device handling, FAST, and cardiac, vascular and bone sonography (Table 1). Each lecture was followed by hands-on training on two live models during which participants were separated into two groups (maximum of five students per instructor). The ratio between didactic and hands-on sessions was 1:2. Total duration of the training was approximately 16 h.

Table 1 Training program
Full size table

During the hands-on session, participants were trained to perform the newly taught procedure, but they also rehearsed every procedure they had already learned since the beginning of the seminar. In addition, students reviewed all ultrasound procedures once on each model, at the beginning and the end of every day. For example, before the final evaluation, students had practiced on at least 12 FAST examinations (Table 1).

Selection of participants

Instructors (MG and FL) were two experienced French emergency physicians, fellowship trained and board certified in emergency ultrasonography. They had already performed more than 12 emergency ultrasonography training sessions and had published on its scope [11]. Learners were Chinese physicians practicing in Beijing. We required all participants to have at least 2 years of emergency medicine practice as an attending physician. None of them was a registered diagnostic medical sonographer, and none had hospital credential to perform emergency ultrasonography. Simultaneous translation (Chinese–French) was mainly performed by a professional Chinese interpreter with a master’s degree, but no medical background. Specific vocabulary translation was made by an attending physician who had completed a 1-year emergency medicine residency in France. Teaching aids were in English. Presentations projected for lectures were in Chinese. Live models were two healthy Chinese men. Every participant involved in this study gave their written informed consent.

Methods of measurement

The final evaluation was scheduled to be done at the end of the training. Students were asked to perform an ultrasound examination on live models resting in a supine position. More precisely, emergency physicians had to obtain all the 11 following standard views (Fig. 1):

Fig. 1
figure 1

Eleven ultrasound views/procedure diagrams at the learners’ disposal

Full size image
  1. 1.

    FAST: Morrison’s pouch, perisplenic, pelvic and subxiphoid view.

  2. 2.

    Cardiac: long and short axis, parasternal view; four and five chambers, apical views.

  3. 3.

    Vascular: identification of the internal jugular vein and location of a potential ideal puncture site [12]; identification of the popliteal vein and compression test as indicated for evaluation of deep vein thrombosis [13].

  4. 4.

    Bone: transverse and longitudinal view of the radius; performance of a 10-cm cortical examination [14].

The examination was timed from the first probe contact with the model’s skin. Students searched for the standard view and performed, when indicated, ultrasound test (vascular and bone). Students froze the image when they felt that they had achieved the objective. When the time was complete, the investigators assessed the appropriateness and quality of the image. If the view and/or procedure were considered incorrect, the timing was restarted and students resumed the examination. After 60 s, if the test was considered as failed, investigators helped students to find the right view or to properly accomplish the ultrasound procedure. Each student performed the whole examination once on both live models.

Physicians used a LogiQe ultrasonographer (General Electrics, Milwaukee, MI) with appropriate probes: 2–5 MHz probe for abdominal (3C-RS), 3 MHz probe for cardiac (3S-RS) and 4–10 MHz probe for vascular and bone views (8L-RS).

Primary data analysis

We mainly used descriptive statistics including frequencies, means, medians and appropriate estimation of certainty. The aim of this study was to estimate the efficiency of an ultrasonography introductory course. The overall success rate of image and procedure achievement was the primary outcome. Most articles on the evaluation of ultrasound courses report pre/post tests and image acquisition outcome without time limit [15–19] We define and use a composite overall success rate, which include both quality and time criteria. This combined outcome allowed us to take into account not only the image interpretability, but also emergency medicine time constraints. We arbitrarily set up a cutoff point for overall success rate at 90% to assert that the training was efficient.

We tested the hypothesis of a difference in the success rate between the two live models (chi square test) and compared the time for image capture (Mann–Whitney test). Analysis was performed using Statistical software for Windows 7.0 (StatSoft Inc., Tulsa, USA).

Results

Characteristics of study subjects

Two sessions of 2 days each were organized. Eighteen Chinese emergency physicians attended the training (eight and ten participants for each). One attendee was on-call and did not attend the entire course. None of the applicants withdrew because of language or cultural issues. A total of 17 students, from 10 Beijing academic hospitals, participated in the study. Both out-of (n = 3) and in-hospital (n = 14) emergency practitioners were involved. The 17 students completed the whole study protocol.

The two live models were Chinese Han, 21 and 25 years old, with a body mass index equal to 21 and 22 kg/m2, respectively.

Main results

A total of 374 standard images were obtained and analyzed. Appropriate images and procedures were acquired in less than 60 s in 97.3% (364 out of 374). Investigators interrupted students in ten cases related to six different views, most of them during cardiac ultrasonography (Table 2). Six students were responsible for these cases: one physician failed three times, two failed two times, and three failed one time. These results corresponded to a 24% (8/34) complete sequence failure (1 or more of the 11 images/procedures exceeded 60 s).

Table 2 Number of failed examinations and median time for image acquisition
Full size table

Median cumulate time to obtain the 11 images and procedures was 182 s (141–238) (Fig. 2). Time to get the specific views of each examination category was 35 s (21–60) for the four abdominal views, 69 s (53–117) for the four cardiac views, 34 s (21–52) for the two vascular views with corresponding ultrasound procedures and 26 s (16–39) for the long bone procedure. Median time to capture a single image was 13 s (5–24). The time to obtain each view is detailed in Table 2. The difference in failure rate was not significant between the two models (6/187 vs. 4/187). There was no statistically significant difference between the two live models regarding the time to obtain a single image, perform a specific category of examination or complete the whole examination protocol (p > 0.2 for every comparison).

Fig. 2
figure 2

Time needed to acquire ultrasound images according to the examination category

Full size image

Limitations

The main limitation of this study is the lack of clinical proof of its efficiency. Even if the overall success rate was excellent, examinations were made on healthy live models and no pathological image acquisition was performed. However, we have previously published an increase in diagnosis performance after the implementation of a similar ultrasound training and on completion of 25 examinations [11]. The learning curve for focused sonography is known to flatten only after 20–70 procedures, so that we can reasonably expect that physicians’ performance will increase after a clinical training period [2]. Besides, this high success rate in image capture could be highly relevant to the context of telemedicine and remote area health care, in which practitioners perform only the examination, the interpretation being the specialist’s prerogative.

A randomized study comparing ultrasound tutorials with and without simultaneous translation would have increased the power of the study. The final goal was not to demonstrate the specific role of translation. The goal was to demonstrate that a bilingual ultrasound course was feasible. Many factors contributed to the success of the training, such as knowledge and understanding of anatomy, translation of slides and direct discussion with few Chinese physicians speaking English.

Course registration was on a voluntary basis and all the participants were from academic hospitals. Thus, we cannot exclude a selection bias and our results may have been different with less skilled people. However, this is unlikely, as performance disparity between various experienced practitioners was not detected after a previous study on an ultrasound introductory course [15].

The objectives were basic ones. For example, participants had to capture the four main cardiac views, but no Doppler or time-motion acquisition was required. Compression test during vascular examination was performed only on a single site, while the diagnosis of deep vein thrombosis relies on at least four compression sites [13]. The results of this study must be understood as the validation of an introductory course of focused sonography and are reliable only for basic applications (FAST, visual assessment of pericardial effusion and ejection fraction, long bone fracture…).

Discussion

Our ultrasound introductory course led to excellent skill acquisition despite translation and cross-cultural constraints, reaching a success rate superior to 97% in obtaining images and validating procedure. To our knowledge, it is the first time that such training efficiency has been evaluated with a composite primary outcome including quality and time limitation. Numerous researches have analyzed training effectiveness through pre/post course quiz, static image interpretation and/or images achievement [15–19]. In our protocol, we integrated a time limit to be in line with time-sensitive diseases typically encountered in emergency medicine and in remote area primary care. Despite this additional judgment criterion, learners failed to achieve the image/procedure in only 3% and full-sequence acquisition had a 76% success rate. In daily life, examination accomplishment would even be enhanced as a single image acquisition failure does not necessarily lead to the failure of the overall examination, and the time to obtain difficult views is balanced by a rapid capture of other images. Moreover, in focused sonography like FAST, not all the views are mandatory to reach a diagnosis, one positive being sufficient to conclude and others only increasing the sensibility of the examination [20].

Besides the overall success rate, the total time to perform a sonography examination may also be pertinent to evaluate a physician’s skill. Some authors reported a mean time of 4.9 min to complete a FAST scan, while others described a median time of 9 min 53 s to complete an all-coming focused ultrasound examination [20, 21]. In our study, learners obtained a single image in less than 30 s in 97% of the cases and, excluding time for probe change, achieved the whole ultrasound protocol in less than 4 min. Even if image goals were simple ones (but cornerstones of FAST examination), these results greatly satisfied common practice requirements. Finally, it is worth noting that more complex examinations, such as cardiac sonography, had a good success rate and were not time-consuming, supporting the legitimacy of basic echocardiography teaching in ultrasound introductory courses [2]. Using practical repetition as a learning process may have a key role in this good training performance. In laparoscopic and thoracostomy training, repetition has been shown to effectively increase success rate and reduce the length of the procedure [22, 23].

Our secondary objective was to evaluate the effect of language and cultural differences on training effectiveness. Translation accuracy, unfamiliar abbreviations and subtleties of language are all known to possibly affect training success [10]. Divergences on cultural norms and teaching styles may also alter the learning process [10, 24]. In Asia, the shift from a teacher-centered to a subject-oriented teaching has been involved in education programs’ improvement [25, 26]. Despite all these potential barriers, no student withdrew from our training and, as stated above, the overall success rate of image acquisition was excellent. One key point was certainly the association of a professional interpreter, who cared about general translation, and a French speaking Chinese physician, who was in charge of translation of complex medical terms. Besides, the training characteristics themselves, with practical and visual learning, avoid many language and cultural pitfalls and make ultrasound courses a good candidate for international emergency medicine programs. Especially, focused ultrasound training lends itself to a repetition learning process and thus eludes many communication difficulties. However, the real challenge will remain in the implementation, in daily life, of ultrasound practice by non-specialists. In China, like in other parts of the world, traditional sonographers are against the spread of focused ultrasound, regardless of international guidelines and multiple proofs of clinical and economic efficiency [2, 8].

In conclusion, after a 2-day introductory course, nontraditional imagers as emergency physicians are able to capture basic ultrasound images in a time consistent with daily practice. Subject to professional translation and highly practical learning, language and cultural barriers do not seem to alter such training efficiency. Furthermore, a time limit appears to be an objective and reliable criterion to help in the assessment of training efficacy, especially in a bilingual and cross-cultural context. Further studies are needed to define the importance and the extent of time limit, but we think that every training efficiency and learning curve studies on focused sonography should integrate a time criterion.