Original paper
DOI: 10.11152/mu.2013.2066.183.ruzAbstract
Aim: The objective of this study was to evaluate whether musculoskeletal (MS) ultrasound (US) can be useful in helping medical students to detect joint inflammation through physical examination. Material and methods: The study was per- formed by two groups of four 6th year medical students. None had received any previous training in the clinical examination of joints or the use of ultrasound. Students were put through a 5-session training programme on the clinical detection of either knee [group 1] or metacarpophalangeal (MCP) [group 2] inflammation. After an initial training session on physical examina- tion of normal and inflamed joints, the students examined 170 joints from 41 patients attending the hospital outpatient clinic in 4 separate sessions. The same joints were assessed for synovitis with US with the ensuing data compared to that of the students and analyzed for concordance with Cohen’s unweighted kappa. Results: In total 60 knees [group 1] and 110 MCP [group 2]
were evaluated. The agreement between the presence of arthritis detected by the students in the four sessions and the pres- ence of synovitis detected by US improved from the session I to sessions III with a marked improvement in the last session.
Conclusions: MSUS may be an effective technique for helping students to acquire the ability to detect joint inflammation.
Keywords: medical students, musculoskeletal ultrasound, education, training, arthritis
Musculoskeletal ultrasound: an effective tool to help medical students improve joint inflammation detection?
Aníbal Ruiz-Curiel
1, María Dolores Díaz-Barreda
1, Marta González-Rodríguez
1, Teresa Martín-Delgado
1, Adriana Medrano-Antoñazas
1, Raquel Muñoz-González
1, Silvia Perales-Torres
1, Javier Sánchez-Montero
1, Iustina Janta
2, Juan Carlos Nieto
2, Luis Collado-Yurrita
3*, Francisco Javier López-Longo
2*, Esperanza Naredo
2*1Faculty of Medicine, Universidad Complutense, 2Department of Rheumatology, Hospital General Universitario Gre- gorio Marañón, 3Department of Medicine, Faculty of Medicine, Complutense University, Madrid, Spain
*Senior authors
Received 14.04.2016 Accepted 20.05.2016 Med Ultrason
2016, Vol. 18, No 3, 294-298 Corresponding author: Iustina Janta
Department of Rheumatology
Hospital General Universitario Gregorio Marañón Doctor Esquerdo, 46, 28007 Madrid, Spain E-mail: [email protected]
Introduction
For the sixth year students of medicine there is no current training on detecting inflamed joints through physical examination. The scope of this paper was to examine whether musculoskeletal ultrasound (MSUS) might be useful as a diagnostic tool to help students ac- quire competence and skills in this area.
This exercise was conducted in conjunction with ex- perienced rheumatologists, for whom joint inflammation
or arthritis is the most characteristic clinical manifesta- tion of many rheumatic diseases. For these doctors the ability to identify inflammation in symptomatic joints is paramount. This is achieved through external observation and palpation, with pathologic joints determined through abnormalities to anatomical structures such as soft tis- sue under or between the bones, liquid matter, tender- ness, swelling or warmth. However, even for experts, the clinical detection of joint inflammation through physical examination is not always easy, evidenced by relatively low peer consensus or interobserver reliability, estimated at 0.29 and 0.98 [1-3].
The diagnosis process can be improved, however, through MSUS which is a diagnostic imaging technique based on the use of ultrasound to view internal structures.
Employed in many countries, including Spain, the tech- nique is more sensitive and accurate than the physical
examination in detecting inflammation in limb joints [4,5]. Furthermore, MSUS has proven to be a useful tool in helping medical undergraduates broaden their under- standing of anatomy [6-8]. Its use has extended to areas other than rheumatology to enable physicians another perspective on internal structures. Additionally, MSUS can also be used by the patient to self-assess joint inflam- mation [9].
To the best our knowledge, there are no training methods or studies reported which evaluate MSUS as a teaching tool to aid medical students in the clinical detec- tion of joint inflammation [3].
We hypothesized that the complementary use of MSUS as a diagnostic tool could help medical students to distinguish normal joints from those presenting arthri- tis. Therefore, the aim of this study was to test whether MSUS is effective as a teaching tool.
Material and methods Study design
This study was conducted by the Department of Rheumatology, University Hospital Gregorio Marañón, Madrid, Spain, from September to December 2015. It was approved by the Ethics Committee of Clinical Inves- tigation and conducted in accordance with the Declara- tion of Helsinki. All patients who participated signed the informed consent.
Two groups of 6th year medical students of the Com- plutense University, Madrid (Spain) conducted the study.
Each group comprised 4 students – 1 male and 3 female in both cases – none of whom had any previous training in joint or ultrasound examination.
The study consisted of five sessions as follows:
a) Training session: a 2-hour training session was given on the clinical detection of inflammation in ei- ther the knee [group 1] or metacarpophalangeal (MCP) [group 2] joints.
Firstly, the students performed knee [group 1] and MCP [group 2] physical examinations on each other to enable them to appraise healthy joints.
Students were then trained by consultant rheumatolo- gists (IJ, JCN) in how to detect inflammation in target joints through physical examination. This was demon-
strated with rheumatic patients attending the outpatient clinic on that particular day presenting varying degrees of clinical inflammation.
b) Sessions I to IV: the first session was conducted one week after the training session with the second con- ducted one week following the first. A 10-week break ensued to prevent any possible influence of external fac- tors – such as exams - in the results. The third and fourth training sessions were then held, with a week’s interlude, as illustrated in figure 1.
Patients were selected by the consultant rheumatolo- gists and split into 2 groups according to whether their pathology/ inflammation was that of a knee-related rheu- matic disease [group 1] or rheumatoid arthritis (RA) [group 2]. Patient selection coincided with medical ap- pointments that day and data gathered included age, sex, medication, and disease duration.
Patients in their respective groups underwent exami- nation by all 4 students, one by one. Students introduced themselves to the patients and started the palpation of ei- ther the knee [group 1] or MCP [group 2] joints, record- ing on a data collection sheet whether there was inflam- mation or not. It is important to note that there was no communication between the students during the process so as to avoid any potential results bias.
Immediately after the clinical examination performed by the 4 students, a third, senior consultant rheumatolo- gist (EN) (different from the training session rheumatolo- gists), with 20 years of experience in MSUS and blinded to students’ results, performed an US examination of the knee [group 1] and MCP [group 2] joints of patients to detect the presence of B-mode synovitis, i.e. joint in- flammation. This US examination was carried out in the presence of the students and the above ultrasonographer explained the anatomic structures and told the students the findings and the location of the synovitis if present.
The US examination consisted of a systematic longi- tudinal and transverse multiplanar examination of knee [group 1] and MCP joints [group 2] using a real-time scanner (Logiq E9, GE Medical Systems Ultrasound and Primary Care Diagnostics, Wauwatosa, WI, USA) equipped with a multi-frequency linear matrix array transducer (6-15 MHz). Machine settings were optimized before the study and standardized for the whole study.
Fig 1. Training process
For the knee, the following recesses were examined: su- prapatellar, parapatellar medial and lateral. For the MCP joints, the dorsal recess was assessed.
Synovitis was defined under Outcome Measures in Rheumatology (OMERACT) definitions as the presence of either effusion or synovial hypertrophy [10]. Synovitis was graduated using a semi-quantitative grade scale, as follows: grade 0 – absence, grade 1 – mild, grade 2 – moderate, grade 3 – marked [11].
Students were then given the opportunity to clinically re-explore the joints so as to hone their clinical skills, whilst using feedback from ultrasound results as a reference.
Statistical analysis
Concordance between the clinical detection of arthri- tis by the students and the presence of ultrasound joint in- flammation was analysed by Cohen’s unweighted kappa.
The US score of synovitis was dichotomized as follows:
Grades 0 and 1, absence of arthritis; grades 2 and 3, pres- ence of arthritis. The interobserver agreement between the students was evaluated by the Fleiss kappa test. The kappa values were considered poor if < 0-0.2, modest if
> 0.20-0.40, moderate if > 0.40-0.60, good if > 0.60-0.80 and excellent if > 0.80 [12]. Any p value < 0.05 was con- sidered significant.
Results
Patient characteristics
Group 1.Thirty patients were included with a mean (SD, range) age of 58.5 (18.0, 25-89) years, 10 (33.3%) being male, with a mean (SD, range) disease duration of 9.9 (6.9, 1-25) years and the following diagnosis: 16 with rheumatoid arthritis (RA) (53.3%), 5 (16.7%) with osteoar- thritis, 2 (6.7%) with connective tissue disease, 6 (20.0%) with spondyloarthropathy, and 1 (3.3 %) with gout.
Eleven patients (36.7%) received synthetic disease- modifying anti-rheumatic drugs (DMARDs), 6 (20.0%) biological DMARDs, 3 (10.0%) synthetic and biologi-
cal DMARDs, 2 (6.7%) anti-hyperuricemic drugs, and 8 (26.7%) did not receive any treatment.
Out of the 60 knees studied, 7 were explored in ses- sion I; 7 in session II; 8 in session III; and 8 in session IV.
Group 2. Eleven patients with RA were included with a mean (SD, range) age of 70.8 (8.3, 54-81) years and with a mean (SD, range) disease duration of 157 (125.2, 13-408) months.
One (9.1%) patient did not receive any DMARDs, 4 (36.4%) received synthetic DMARDs, 3 (27.3%) received biological DMARDs and 3 (27.3%) received synthetic and biological DMARDs. Three (27.3%) were receiv- ing non-steroidal anti-inflammatory drugs (NSAIDs), 2 (18.2%) corticosteroids and another 2 (18.2%) were un- dergoing NSAIDs plus corticosteroid therapy.
Out of the 11 patients studied, 2 were explored in ses- sion I, 3 in session II, 3 in session III and 3 in session IV.
Clinical and US findings
Group 1. The students explored 14 knees during ses- sion I, 14 knees in session II, 16 knees in session III, and 16 knees in session IV. All knees (a total of 60) were examined by US.
Group 2. The students explored 20 MCP joints during session I, 30 MCP joints session II, 30 MCP joints ses- sion III, and 30 MCP joints session IV. All MCP joints (a total of 110) were examined by US.
The prevalence of arthritis detected by the students by clinical examination is shown in table I.
Agreement between students and US findings Group 1. Table II displays the agreement between the presence of student-detected knee arthritis by clini- cal examination in the four sessions and the presence of US-detected synovitis. The kappa values increased in the session III and especially in the session IV. There was a significant improvement in the agreement between ses- sion I and session III (p=0.017), session I and session IV (p=0.005), session II and session IV (p=0.001), and session III and session IV (p=0.01).
Table I. Prevalence of student-detected arthritis.
Std 1 GROUP 1 GROUP 2
Std 2 Std 3 Std 4 Std 1 Std 2 Std 3 Std 4
Session Arthritis n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%)
I Yes 8 (57.1) 7(50) 5 (35.7) 7 (50) 4 (20) 3 (15) 4 (20) 4 (20)
No 6 (42.9) 7 (50) 9 (64.3) 7 (50) 16 (80) 17 (85) 16 (80) 16 (80)
II Yes 6 (42.9) 6 (42.9) 6 (42.9) 7 (50) 15 (50) 15 (50) 22 (73.3) 18 (60)
No 8 (57.1) 8 (57.1) 8 (57.1) 7 (50) 15 (50) 15 (50) 8 (26.7) 12 (40)
III Yes 6 (37.5) 7 (43.8) 6 (37.5) 6 (37.5) 9 (30) 15 (50) 19 (63.3) 19 (63.3)
No 10 (62.5) 9 (56.3) 10 (62.5) 10 (62.5) 21 (70) 15 (50) 11 (36.7) 11 (36.7)
IV Yes 3 (18.8) 2 (12.5) 5 (31.3) 2 (12.5) 9 (30) 11 (36.7) 14 (46.7) 12 (40)
No 13 (81.3) 14 (87.5) 11 (68.8) 14 (87.5) 21 (70) 19 (63.3) 16 (53.3) 18 (60) Std – student, n – number of examined joints
Group 2. Table III displays the agreement between the presence of student-detected MCP arthritis and the presence of US-detected synovitis over the 4 sessions.
There seems to be a slight increase in the kappa values in session III. However, three out of 4 students produced clearly better results in session IV. There was a significant improvement in the agreement between session I and ses- sion III (p=0.024), session I and session IV (p=<0.001), session II and session IV (p=0.002) and session III and session IV (p=0.013).
Agreement between students
Group 1. The agreement between students improved from session I to session III, being good in the latter.
However, less agreement was observed in the last ses- sion (Table IV).
Group 2. The agreement between students improved from session I to session IV (Table V).
Table II. Agreement between the presence of arthritis detected by the students (group 1) and the presence of synovitis detected by US.
Session I Session II Session III Session IV
Student KAPPA (CI 95%) p KAPPA (CI 95%) p KAPPA (CI 95%) p KAPPA (CI 95%) p 1 -0.12 (-0.62 - 0.29) 0.64 0.44 (-0.01- 0.81) 0.086 0.20 (-0.29- 0.60) 0.424 0.59 (0.08- 1.00) 0.018 2 -0.29 (-0.78- 0.12) 0.28 -0.12 (-0.62- 0.29) 0.64 0.36 (-0.11- 0.73) 0.152 0.29 (-0.30- 0.89) 0.226 3 -0.04 (-0.56- 0.38) 0.872 0.16 (-0.33- 0.56) 0.533 0.20 (-0.29- 0.60) 0.424 0.67 (0.27- 1.00) 0.004 4 0.14 (-0.35- 0.55) 0.577 0.29 (-0.21- 0.69) 0.28 0.47 (0.02- 0.83) 0.062 0.77 (0.33- 1.00) 0.002
Table III. Agreement between the presence of arthritis detected by the students (group 2) and the presence of synovitis detected by US.
Session I Session II Session III Session IV
Student KAPPA (CI 95%) p KAPPA (CI 95%) p KAPPA (CI 95%) p KAPPA (CI 95%) p 1 0.04 (-0.33 – 0.42) 0.822 -0.07 (-0.42 – 0.28) 0.713 0.25 (-0.10 – 0.61) 0.16 0.44 (0.12 – 0.75) 0.013 2 0.14 (-0.20 – 0.48) 0.413 -0.20 (-0.55 – 0.57) 0.269 0.20 (-0.14 – 0.54) 0.256 0.59 (0.29 – 0.88) 0.001 3 -0.17 (-0.52 – 0.18) 0.369 0.06 (-0.23 – 0.35) 0.697 0.25 (-0.04 – 0.54) 0.11 -0.01 (-0.37 – 0.35) 0.961 4 0.26 (-0.12 – 0.63) 0.178 0.29 (-0.04 – 0.61) 0.098 0.38 (0.10 – 0.65) 0.017 0.66 (0.39 – 0.93) <0.001
Table IV. Agreement between students [group 1].
SESSION KAPPA (CI 95%) p
I 0.55 ( 0.33-0.76) <0.0005
II 0.54 ( 0.33-0.76) <0.0005
III 0.63 (0.43-0.83) <0.0005
IV 0.25 (0.05-0.45) 0.008
Discussions
US has been used in various studies to improve medi- cal students’ clinical examination skills [13-16]. The ma- jority of the studies focused on students’ skills to detect liver margins [13,14], inferior limit of the lung or thyroid size [15], or to localize femoral artery and vein [16].
The results of our study showed that medical students significantly improved their skills in the detection of synovitis in target joints using US as an educational tool for feedback. Overall, interobserver agreement improved over the duration of the training course proving no worse than that reported from expert physicians in the field [1- 3]. The main weakness of the study is the lack of group control. This means that we cannot be sure that clini- cal examination improvement was only due to the US feedback; results may have been similar even without US. Additionally, US joint inflammation scores were di- chotomised. The argument for considering grade 0 and 1 of US synovitis as absence of clinical inflammation is that both can be observed in normal joints [17]. Another weakness of this study is that we did not assess the stu- dent’s opinion regarding potential self-confidence im- provement or educational value of the training program.
On the other hand, one of the study strengths was that the students were unaware of their colleagues’ evaluation and the ultrasonographer was unaware of the students’
evaluation. Additionally, students reached levels of inter- observer reliability when assessing joints similar to that of their more experienced, veteran colleagues, experts in the musculoskeletical field [1-3], although a larger group of patients would be required to substantiate this further.
Table V. Agreement between students [group 2].
SESSION KAPPA (CI 95%) p
I 0.15 (0.07 – 0.33) 0.048
II 0.25 (0.10 – 0.39) <0.001
III 0.31 (0.16 – 0.46) <0.001
IV 0.39 (0.24 – 0.54) <0.001
vitis in the knee: a comparison with arthroscopy and clini- cal examination. Arthritis Rheum 2004; 50: 387-394.
6. Swamy M, Searle RF. Anatomy teaching with portable ul- trasound to medical students. BMC Med Educ 2012; 12: 99.
7. Stringer MD, Duncan LJ, Samalia L. Using real-time ul- trasound to teach living anatomy: an alternative model for large classes. N Z Med J 2012; 125: 37-45.
8. Jurjus RA, Dimorier K, Brown K, et al. Can anatomists teach living anatomy using ultrasound as a teaching tool?
Anat Sci Educ 2014; 7: 340-349.
9. Cheung PP, Lahiri M, Teng GG, et al. A randomized controlled trial for improving patient self-assessment of synovitis in rheumatoid arthritis with education by ultra- sonography: the RAEUS Study. Rheumatology 2015; 54:
1161-1169.
10. Wakefield RJ, Balint PV, Szkudlarek M, et al; OMERACT 7 Special Interest Group. Musculoskeletal ultrasound in- cluding definitions for ultrasonographic pathology. J Rheu- matol 2005; 32: 2485-2487.
11. Naredo E, Rodríguez M, Campos C, et al. Validity, repro- ducibility and responsiveness of a 12-joint simplified power Doppler ultrasonographic assessment of joint inflammation in rheumatoid arthritis. Arthritis Rheum 2008; 59: 512-522.
12. Landis JR, Koch GG. The measurement of observer agree- ment for categorical data. Biometrics 1977; 33: 159-174.
13. Barloon TJ, Brown BP, Abu-Yousef MM, et al. Teaching physical examinationof the adult liver with use of real-time sonography. Acad Radiol 1998; 5: 101–103.
14. Butter J, Grant TH, Egan M, et al. Does ultrasound train- ing boost year medical student competence and confidence when learning abdominal examination? Med Educ 2007;
41: 843–848.
15. Fodor D, Badea R, Poanta L, Dumitrascu DL, Buzoianu AD, Mircea PA. The use of ultrasonography in learning clinical examination: a pilot study involving third year medical students. Med Ultrason 2012; 14: 177–181.
16. Ahn JS, French AJ, Thiessen ME, et al. Using ultrasound to enhance medical students’ femoral vascular physical ex- amination skills. J Ultrasound Med 2015; 34: 1771-1776.
17. Witt M, Mueller F, Nigg A, et al. Relevance of grade 1 gray- scale ultrasound findings in wrists and small joints to the assessment of subclinical synovitis in rheumatoid arthritis.
Arthritis Rheum 2013; 65: 1694-1701.
MSUS is a user-friendly, non-invasive, objective and readily-available technique which can be used to im- prove students’ clinical skills in the detection of synovitic joints. The technique offers immediate feedback mean- ing the patient can be re-examined on the spot if neces- sary. The training was overseen by a sonographer with over 20 years’ experience and, last but not least, the cost of this training was affordable.
Although we did not formally assess the student’s feedback, at the end of the study they declared to have greatly appreciated the educational value of this innova- tive learning method.
In conclusion, MSUS may be an effective technique for helping students to acquire the ability to detect joint inflammation. Additionally, we all found the practical training extremely useful and would encourage those who prepare the medical curricula to consider its inclu- sion.
Conflict of interest: none
References
1. Felson DT, Zhang Y. An update on the epidemiology of knee and hip osteoarthritis with a view to prevention. Ar- thritis Rheum 1998; 41: 1343-1355.
2. Carmona L, Ballina J, Gabriel R, Laffon A; EPISER Study Group. The burden of musculoskeletal diseases in the gen- eral population of Spain: results from a national survey.
Ann Rheum Dis 2001; 60: 1040-1045.
3. Cheung PP, Gossec L, Mak A, March L. Reliabili- ty of joint count assessment in rheumatoid arthritis: a sys- tematic literature review. Semin Arthritis Rheum 2014; 43:
721-729.
4. Kane D, Balint PV, Sturrock RD. Ultrasonography is supe- rior to clinical examination in the detection and localization of knee joint effusion in rheumatoid arthritis. J Rheumatol 2003; 30: 966-971.
5. Karim Z, Wakefield RJ, Quinn M, et al. Validation and re- productibility of ultrasonography in the detection of syno-
