Internal Medicine Point-of-Care Ultrasound Curriculum: Consensus Recommendations from the Canadian Internal Medicine Ultrasound (CIMUS) Group

The use of point-of-care ultrasound (POCUS) has increased significantly over the last decade. This is likely the result of accumulating evidence demonstrating that effective POCUS skills can be acquired with minimal training1 ^– 3 and that POCUS may improve diagnostic performance when used with the traditional physical examination,4 ^– 6 especially in situations where patient characteristics limit the accuracy of the physical examination.7 ^, 8

POCUS has significant utility in assessing patients seen by internists. For the assessment of dyspneic patients, POCUS has demonstrated higher accuracy than the traditional workup.6 ^, 9 For example, the use of POCUS for assessment of lung B lines in heart failure patients on discharge can predict readmission rates at 6 months.10 Second, POCUS guidance of some bedside procedures reduces errors and complications.11 ^– 13 Lastly, POCUS use in internal medicine may result in reduced expenditures.14

The performance of POCUS is highly operator-dependent, and appropriate competency-based training is necessary prior to its use.15 ^, 16 Despite the purported clinical benefits of POCUS, however, there is currently no clear agreement as to what an internal medicine POCUS curriculum should be.17 ^, 18 To address this gap,19 this document outlines a set of consensus-based recommendations for a Canadian internal medicine POCUS curriculum, accounting for the existing limitations in available trained faculty, administrative structures, and resources within a Canadian context.

The Canadian Internal Medicine Ultrasound (CIMUS) group comprises leadership representatives from a number of internal medicine residency programs across Canada. Support for this group’s work was obtained from the Canadian Society of Internal Medicine (CSIM) Council and Education Committee in October 2015. In June 2016, program directors from each of 17 Canadian internal medicine residency training programs (postgraduate years [PGY] 1–3) and 16 general internal medicine residency training programs (PGY 4–5), as well as 17 internal medicine division chiefs across Canada, were invited to identify leaders in their respective programs and/or divisions as having specialized POCUS skills, educational expertise, and/or leadership roles within their institution for advancing POCUS use and education within internal medicine.20 ^, 21 Each identified lead was then invited to participate in a 4-h consensus meeting, using a modified nominal group technique (NGT),22 held during the CSIM Annual Meeting in Montréal, QC, on October 29, 2016. Those unable to attend the meeting in person participated via teleconference. The Royal College of Physicians and Surgeons of Canada (RCPSC) is Canada’s national accreditation body for residency programs. As such, one representative each from the RCPSC specialty committees in internal medicine and general internal medicine also participated in this meeting.

At the meeting, preliminary learner needs assessment data from five Canadian internal medicine training programs were presented. Participants then discussed and agreed upon four overarching principles upon which curricular items would be selected:

To this end, we aimed to achieve the minimum number of topics that we felt could feasibly be introduced, given existing limitations in equipment resources, trainee time, and expert faculty time. The process of voting (described below) was then discussed with the expert group. We determined a priori to conduct no more than three rounds of voting.22

The meeting was facilitated by two POCUS experts (IM, SA), both of whom have completed a 1-year dedicated POCUS fellowship. At the start of the meeting, a list of candidate POCUS applications (25 applications and 10 ultrasound-guided procedures) was presented based on commonly accepted POCUS applications23 ^– 26 and Canadian internal medicine procedural competency training requirements.27 ^, 28 Paper copies of key articles were also provided at the meeting.23 ^– 28 We did not conduct a round-robin discussion for item generation, given the existence of commonly accepted applications. Our participant group size (N = 39) was substantially larger than group sizes typically used in NGT studies (N = 5–12).22 To optimize participant engagement in the discussion of each of the 35 curricular items, we divided participants into five subgroups rather than having one large group discussion.

Following the small group discussions, a preliminary large group discussion was held on individual curriculum applications, led by the same facilitators (IM, SA). Participants then voted anonymously on each item as to whether it should be included in or excluded from a core internal medicine POCUS curriculum (postgraduate years [PGY] 1–3) or expanded general internal medicine curriculum (PGY 4–5). All participants voted using an anonymous paper-based approach (or via e-mail for the teleconference participants). We defined consensus as agreement by at least 80% of the members. This 80% threshold is in keeping with guideline recommendations.29

All applications not reaching consensus were put forward for consideration by voting in round 2. Only quantitative results (percentage agreement) for applications that did not reach consensus were fed back to the panel. For each of these applications, if more than 50% of participants indicated interest in readdressing it, the application was voted upon again in round 2. The second round was conducted in an open, unblinded fashion (i.e. not anonymous) for convenience reasons due to time limitations (a maximum of 4 h was allotted for the meeting). Items with 80% or greater agreement were considered to have reached consensus. A final round was then conducted using an online survey in a blinded fashion approximately 2 weeks after the meeting in order to minimize the potential impact of dominating members of the group on the unblinded second-round vote.22 The same experts were invited to participate in all rounds.

A total of 47 individuals were identified by 14 of the 17 (82%) Canadian academic institutions as meeting POCUS education leadership criteria. Of these, 39 (83%) individuals participated in the meeting: 31 in person and eight via teleconferencing. Baseline demographics of the 39 individuals are described in Table 1.

We recommend that four applications (inferior vena cava, lung B lines, pleural effusion, and abdominal free fluid) and three procedures (central venous catheterization, thoracentesis, and paracentesis) be included in the core Internal Medicine PGY 1–3 curriculum. For the expanded PGY 4–5 curriculum, we recommend that in addition to the core applications and procedures listed above, seven applications (internal jugular vein, lung consolidation, pneumothorax, knee effusion, gross left ventricular systolic function, pericardial effusion, and right ventricular strain), and four procedures (knee arthrocentesis, arterial line insertion, arterial blood gas sampling, and peripheral venous catheterization) be included.

A number of contextual features and limitations should be highlighted in the interpretation and application of our results. First, our group aimed to achieve the minimum number of topics that we felt could feasibly be introduced, given the existing limitations in resources, trainee time, and expert faculty within the Canadian internal medicine programs.30 These guidelines are not intended to dissuade programs from teaching additional applications. Second, these recommendations are expected to change over time. As programs gain comfort and expertise, and as additional evidence on POCUS becomes available, we anticipate that our current recommendations will need to be modified. With time, we anticipate that some of these applications will be taught in the undergraduate medical curriculum31 and may need only to be reviewed in the postgraduate curriculum. Third, our recommendations were determined solely by expert opinion-based consensus. We did not grade the strength of our recommendations or conduct a systematic review of all applications. However, collectively, we feel that our group has the necessary clinical and educational expertise and awareness of our current training limitations to make the above recommendations. Fourth, because of the large number of items considered and the number of experts in our group, we chose to ask the experts to indicate binary responses (should include vs. should not include) rather than ranking or rating items on Likert scales. Future studies could consider these alternative rating options. Fifth, because of the anonymous nature of the process, we were not able to identify which experts did or did not participate in the voting for each round, only that we had response rates of 90% in round 1, 87% in round 2, and 97% in round 3. Future studies should consider tracking the identities of each expert. Sixth, our report does not cover curriculum design or implementation issues.

As a pan-Canadian internal medicine expert-based group, the Canadian Internal Medicine Ultrasound (CIMUS) group has reached consensus on the POCUS applications for internal medicine postgraduate curriculum. We recommend that four POCUS applications and three procedures be included in the core PGY 1–3 curriculum, and 11 POCUS applications and seven procedures be included in the expanded PGY 4–5 curriculum.