In this open randomized cross-over clinical trial among 2nd year medical students, the addition of short educational units of arthroscopy to the macroscopic dissection course increased the anatomical knowledge gain compared to standard anatomy training. The addition of equivalent educational units of musculoskeletal ultrasound (MSUS) to the standard anatomy training using peer-assisted learning (PAL) did not improve anatomical skills. Despite a statistically significant difference in performance between the study groups regarding the MC exam, the absolute differences were moderate. However, after additional arthroscopy teaching only 1% of students failed the MC exam, in contrast to 10% in the MSUS- or CON-group, respectively.
The benefit of the ASK module was limited to the shoulder area. The multidimensional augmentation of the standard anatomy lecture with arthroscopy and ultrasound was considered beneficial by the students regarding the acquisition of anatomical knowledge of the musculoskeletal system. When asked to compare, students preferred arthroscopy due to a better anatomical orientation and a higher gain in anatomical knowledge as well as in spatial imagination.
The benefit of virtual multidimensionality in anatomical education has been demonstrated before [
1]. For this reason our study investigates, which particular kind of cognitive 2D-3D-transformation, ultrasound or arthroscopy, enhances anatomical knowledge during the curricular dissection course. The musculoskeletal ultrasound (MSUS) has already been incorporated successfully into the curricular anatomical education [
18]. Teaching these skills at an early level may improve medical expertise in diagnostic methods and may improve the quality of patient care [
2]. Ultrasound is a rapidly available and cost-saving instrument, and is a perfect teaching tool for medical students [
7]. Contrary to the results of other trials [
2-
6] students in this study did not benefit from the additional ultrasound courses of the musculoskeletal system with regard to anatomical skills. A possible explanation might be the fact that we used student-teachers as lecturers. Minimally instructed student-teachers can be employed to teach anatomically skilled students, whereas medical experts should be used in earlier stages of anatomical education [
11]. Tolsgaard
et al. and Shiozawa
et al. postulated that the specific training of student-teachers was an absolute necessity for success in the environment of peer-assisted learning [
19,
20]. Therefore, we included student-teachers who received an extensive training in education theory (implementation of a teach-the-teacher course) in our study. Possibly, the results of Tolsgaard and Shiozawa cannot be transferred to MSUS. Because of the strong operator-dependence, an adequate training is highly important to guarantee a firm and competent MSUS application [
21].
However, we showed in our study, that arthroscopy tutorials using simulators held by medical experts moderately increased the anatomical knowledge gain in comparison to solely macroscopic dissection. Arthroscopy models can be very useful, regarding the anatomical education of students without any previous experiences [
14]. To the best of our knowledge, our trial is the first randomized study evaluating the direct anatomical knowledge benefit of arthroscopic skill training in comparison to a control group without any intervention. Until now it has only been shown that students’ motivation towards anatomical education could be increased by the use of surgical simulators [
13]. It has been demonstrated that not every student has the psychomotor precondition to learn complex arthroscopy skills [
12].
Our results show that a multimedia presentation and consecutive “hands on” training supported the anatomical learning efficiency. Translating a two dimensional view into three-dimensional spatial orientation requires a bimanual psychomotor activation, estimation of depth and coordination of the visual and tactile sense. These complex coordinative skills may act as important memory anchors. Other approaches to virtual three-dimensional visualization have also been able to show potential additional benefit for the anatomic education [
22,
23].
One recent study on virtual dissection software as opposed to a cadaver-based course reported that students perceived that the virtual approach is highly valuable in their learning of anatomy [
24].
Students in our cohort reported that structures were easier identifiable using arthroscopy and that arthroscopy training was associated with a subjectively higher learning effect in comparison to ultrasound. Especially spatial imagination was described as clearly improved.
The impact of the different teaching methods (medical experts
vs. student-teachers) can not be estimated entirely. Even though peer-assisted learning is regarded as a good teaching concept, medical experts of the arthroscopy module scored higher than the student-teachers of the MSUS module in terms of level of competence. One aspect might be the steep learning curve and the complexity of MSUS [
17,
21]. According to the students’ opinion, arthroscopy and MSUS have the potency to raise interest in surgery. This is in line with an earlier report by our study group that showed that students develop an interest in orthopaedic trauma at a very early level but somehow lose it over time due to negative experience during their clinical training [
25]. In times of shortage of young academics, this is a very important factor that is amenable to intervention.
The question, why students particularly profit from an arthroscopy augmented anatomical education for the shoulder as opposed to the knee joint, seems difficult to answer. On the one hand the level of musculoskeletal education varies according to anatomical regions, on the other hand the anatomical complexity appears to be region depending as well [
26]. Day and Yeh showed that students’ confidence in their own examination techniques are clearly reduced at the shoulder while their confidence is above average regarding the knee joint [
26]. Since students are apparently very familiar with examination techniques of the knee, an additional arthroscopic intervention may not lead to an additional knowledge gain. In contrast, the anatomy of the shoulder is more difficult to understand and ultrasound examinations of the shoulder have been attributed one of the steepest learning curves [
27].
Limitations
Our study is subject to a number of limitations. First: Despite significance there were only slightly differences between the study groups in terms of the 15-point MCQ and its sub-analysis, and one must be wary of drawing major definitive conclusions from these test results. A more comprehensive knowledge-based shoulder and knee anatomy test may have demonstrated differences more clearly.
This was a single-centre study. Results may differ in different organizational or didactical settings. Furthermore, we did not assess the level of any anatomical knowledge or skills concerning ultrasound and arthroscopy techniques acquired prior to the intervention. However, according to the curriculum, students had not received any anatomical training on the musculoskeletal system prior to the study and students denied any such qualifications in the questionnaire. The anatomical outcome measures only refer to the shoulder and knee joint and an extrapolation of our results to other anatomic regions is not valid.
We could not control for autonomous self-study and students’ motivation which might have influenced the final test results. We do not see this as a threat to internal validity since selection bias was controlled for by including a large number of participants and using methods of complete random sampling. The study guideline allowed students to miss two classes during the entire 3-week dissection course. Frequency and timing of absence had no significant influence on the final result.
Although the course evaluation did not show any differences between teachers, a certain amount of variability in the level of competence of the student-teachers can not be ruled out.