Teaching and evaluating measurement of the ankle–brachial index

Peripheral artery disease (PAD) is one of the most frequent cardiovascular conditions and paradoxically one of the most neglected ones. Although the reasons for the underdiagnosis of PAD are multifactorial, it cannot be related to difficulties in screening and detecting this condition.

Actually, measurement of the ankle–brachial index (ABI) is harmless and highly accessible at a limited cost, with satisfying sensitivity and specificity to detect PAD. ¹ The technical aspects for correct use seem far more accessible to technicians, nurses, and physicians as compared to many other cardiovascular diagnostic tools. ¹

For all these reasons, several international guidelines concur to advocate for the use of the ABI as the first diagnostic tool to identify PAD.^1–3 A document in 2012 provided standards for the measurement, calculation, and interpretation of the ABI in order to limit discrepancies observed in the past. ¹ The document emphasized the importance of training to improve the reproducibility and reliability of results and recommended the measurement and interpretation of the ABI as part of the curriculum for medical and nursing students, based on didactic and experiential learning under the supervision of qualified trainers. However, studies providing insights on the best way to teach and learn measurement of the ABI were sparse, and no specific recommendations have been proposed.

Indeed, a recent literature review highlighted the paucity of teaching experiences, mostly performed on a low number of trainees, and overall rated as low quality. ⁴

In this regard, the report of Donnou and colleagues ⁵ in this issue of Vascular Medicine can be considered as the first attempt to scientifically assess an ABI teaching experience, through a randomized trial with a satisfying number of trainees (n=30). The trial was designed to assess the experiential component of the teaching after didactic learning for all students. None of the students who benefitted from didactic-only teaching were considered to be proficient, but half (11/20) of those who also had experiential teaching over 3 hours were qualified as proficient.

While the authors should be commended for their study, several aspects need further investigation to clarify and advocate the best way for teaching the ABI measurement.

Ideally, more details about the material used for the didactic learning component should have been shared (e.g. through a video online supplement). Indeed, the reproducibility of the study if performed elsewhere is questionable.

The experiential group of students was split into two subgroups: one training on healthy individuals (i.e. the students measuring each other’s ABIs) and the other on actual patients in the vascular lab. The rational for this distinction is unclear, especially given that the results of both subgroups were ultimately provided together. By mixing healthy individuals and patients, the students would theoretically be confronted with a higher number of training sessions and a larger spectrum of ABIs. Another possibility would be to split the two subgroups according to different durations of training, in order to see whether the proficiency rates would be improved by a higher number of training sessions. All 20 students in the experiential group were assessed after seven pairs of ABI measurement, a number which appears low to reach proficiency based on common experience. The time and number of ABI measurements needed to consider most trainees as proficient needs further study.

Along with the minimal period of learning, another unresolved issue is how to optimally assess the trainees’ proficiency. In the study by Donnou and colleagues, students were assessed by measuring the ABI in only two cases, and their proficiency was judged according to a process checklist as well as whether the ABI obtained by each trainee was within the ±0.15 error margin of the trainer result. This error margin is suggested by the American Heart Association statement on ABI and is based on inter-observer variability obtained by experienced individuals. ¹ However, this threshold was obtained on average, through several pairs of measurements. Therefore, the comparison of trainee and trainer ABI measurements needs several pairs of measurements (reasonably >10 pairs) to assess the quality of results obtained by the trainee. In this aspect, it also seems preferable to have several trainers as the reference to avoid systematic bias. Such an evaluation would not compare the average of results obtained by trainees versus trainers but rather the number of cases with concordant results (e.g. within the ± 0.15 margin).

A larger multicenter study is hence recommended to address the standardization of teaching and learning how to measure the resting ABI for all nursing and medical students. Refresher training would also be suitable to mitigate loss of competence over time. ⁶ Furthermore, a second level would also be necessary for teaching how to measure post-exercise ABI and the toe–brachial index for vascular specialists in training.

In conclusion, Donnou and colleagues made a first step in the evaluation of teaching methods for ABI measurement. Further studies are needed in this field to better implement efficacious training for ABI measurement, with a reliable and standardized assessment of trainees.