Drs. Gutiérrez and Di Geso contributed equally to this work.
Ultrasound Learning Curve in Gout: A Disease-Oriented Training Program
Version of Record online: 26 JUL 2013
Copyright © 2013 by the American College of Rheumatology
Arthritis Care & Research
Volume 65, Issue 8, pages 1265–1274, August 2013
How to Cite
Gutiérrez, M., Di Geso, L., Rovisco, J., Di Carlo, M., Ariani, A., Filippucci, E. and Grassi, W. (2013), Ultrasound Learning Curve in Gout: A Disease-Oriented Training Program. Arthritis Care Res, 65: 1265–1274. doi: 10.1002/acr.22009
- Issue online: 26 JUL 2013
- Version of Record online: 26 JUL 2013
- Accepted manuscript online: 18 MAR 2013 02:43PM EST
- Manuscript Accepted: 26 FEB 2013
- Manuscript Received: 13 SEP 2012
To describe the learning curve of rheumatologists with limited experience using ultrasound (US) attending an intensive disease-oriented training program focusing on the skills required to obtain and interpret US signs of monosodium urate (MSU) crystal deposits in joint and periarticular tissues.
Three investigators participated in a 7-day training program involving 12 men with gout. The agreement between the expert and beginners was calculated in 4 sessions involving 8 patients with gout. The US assessment was performed at the second and third metacarpophalangeal joints, knee, tibiotalar and first metatarsophalangeal joints, second and third finger flexors, quadriceps and patellar posterior tibialis, peroneus longus and brevis, and Achilles tendons. The presence or absence of synovial fluid/synovial hypertrophy, double contour sign, intra- or periarticular and intratendinous tophi, bursitis, bone erosions, and tendon tears was recorded.
A total of 416 anatomic sites were studied. Kappa values and overall agreement percentages of qualitative assessments of US gout findings at the end of the exercise both showed moderate to excellent agreement, while in the first session they showed poor/fair agreement. At the end of the training session, sensitivity, specificity, and capability of the beginners were also improved.
After 1 week of the disease-oriented training program, rheumatologists with limited experience in US were satisfactorily able to detect and interpret the main US signs indicative of MSU crystal deposits at different tissues in patients with gout.
Over the past 10 years, ultrasound (US) has elicited considerable interest among rheumatologists because of its usefulness in both clinical practice and research activity ([1-6]). While a solid and still growing body of evidence supports its use in daily rheumatologic practice, operator dependency and the long learning curve represent the main limitations. To date, different teaching approaches have been proposed and different training experiences have been described ([5, 7-16]), but the best method for allowing rheumatologists to obtain and maintain competency in musculoskeletal US has yet to be defined.
Several studies have investigated the potential of US for the diagnosis and treatment monitoring of patients with gout ([17-29]). The usefulness of US has also been underscored in the most recent European League Against Rheumatism (EULAR) evidence-based recommendations for gout ([30, 31]).
The aim of the present study was to describe the learning curve of rheumatologists with limited experience in musculoskeletal US who attended an intensive disease-oriented training program focused on the skills required to obtain and interpret the main US signs indicative of monosodium urate (MSU) crystal deposits in joint and periarticular tissues.
Box 1. Significance & Innovations
- Competence in knowledge and interpretation of ultrasound (US) findings is becoming part of the core curriculum for the rheumatologist in Europe.
- Rheumatologists with limited experience in US can achieve good to excellent interobserver agreement rates in the assessment of gout.
- A disease-oriented training program could be useful in improving US assessment in different tissues of patients with gout.
MATERIALS AND METHODS
Overall, 12 patients with a definitive diagnosis of gout according to international criteria () were involved in different phases of the training program. Patients ages <18 years with a history of severe trauma or surgery at the anatomic sites to be scanned and with another active concomitant chronic arthritis (such as rheumatoid arthritis [RA] or spondyloarthritis) were not included in the study. All patients were seen at the outpatient or inpatient clinics of the Rheumatology Department of the Università Politecnica delle Marche (Ancona, Italy).
Three investigators with different levels of experience in performing musculoskeletal US participated in the training program. The expert sonographer (MG) was a rheumatologist with >7 years of US experience and a faculty member of the EULAR/Outcome Measures in Rheumatology (OMERACT) US group, whose assessments were used as the gold standard to evaluate the findings obtained by the remaining 2 investigators. Beginner sonographer 1 (MDC) was a fellow in rheumatology with 3 months of US experience who routinely performed a mean of 5 US examinations per week. Beginner sonographer 2 (JR) was a fellow in rheumatology with a very basic knowledge of musculoskeletal US and had 2 months of previous experience as a sonographer (performing ∼3 US examinations per week) mainly focusing on RA and having no direct US experience in the field of crystal-related arthropathies.
In a preliminary step, 2 independent investigators (denominated neutral investigators) who were experts in musculoskeletal US (EF and LDG) and who did not participate in the exercise, selected 12 gout patients with a wide range of US signs indicative of MSU deposits in periarticular soft tissues and at the joint level. The selected patients were coupled into groups in order to include 1 easy patient (i.e., with widespread and evident MSU crystals depositions) and 1 difficult patient (i.e., presenting only small MSU deposits) so that each group would be scanned during each step. Additionally, the researchers selected a core set of 100 US images from their digital database showing representative examples of MSU crystal depositions at different tissue levels and utilized these images during the training program.
The training program lasted for 7 days (at least 4 hours per day). Table 1 shows the aims and activities for each day during the training program. On day 1, the first session was employed to assess the beginners' skills. Two beginners were provided with a core set of 10 scientific articles describing both the standard scanning technique and the key US findings in patients with gout ([17-20, 22, 23, 27]). After a careful reading (lasting ∼4 hours), these articles were discussed with the expert for 1 additional hour.
|1a||Two beginners were provided with a core set of scientific articles describing both the standard scanning technique and the key US findings in patients with gout||To scan and provide the beginners with the basic US theory tools knowledge To obtain data on baseline scanning ability|
|Careful reading of the scientific articles and discussion on their content with the expert|
|After a preliminary agreement on how to set up the US system, first the expert and then the beginners independently carried out the first US examinations of 2 patients with gout|
|The expert took notes about the beginners' errors related to technical aspects of the beginners' US examinations|
|2a||The expert provided to the beginners the basic knowledge necessary for identifying the main US signs of MSU crystal deposits by means of a slide presentation||To assess the early improvement of the beginners' ability to identify US abnormalities in patients with gout|
|Beginners had access to the core set of US images of gout acquired in the first step for discussion with the expert in order to strengthen knowledge of US findings of gout|
|A second US examination was performed independently by the expert and beginners in another 2 patients with gout|
|3||Expert performed US examinations in another 2 patients in front of the beginners and provided hands-on training and didactic instructions||To improve potential errors in technical acquisition of US images in patients with gout To acquire more confidence with the normal US aspects of the different tissues in normal subjects|
|Additional US examinations in 2 healthy volunteers to clarify the normal aspects of the different tissues|
|Discussion of both the quality of the US images and the errors detected by the expert on the images acquired by the beginners during the first 2 days of the program|
|4||Beginners' US examinations of another 2 patients with gout under the direct supervision of the expert||To reevaluate the technical capability of acquisition and to improve the method of interpretation of US findings|
|5a||Independent US examinations by the beginners in 2 patients with gout||To test the ability of beginners in both acquisition and interpretation of US findings without the supervision or assistance of the expert|
|6||Selection of 30 images from those acquired in the previous days The expert showed these images to the beginners and described the US findings in detail||To answer the beginners' questions in order to improve agreement and resolve their doubts|
|7a||Independent US examinations of 2 patients with gout by all investigators||To assess agreement between beginners and the expert at the end of the training program|
Successively, after a preliminary agreement on how to set the US system, first the expert and then the beginners independently carried out the first US examinations of 2 patients with gout. The sonographers were asked to indicate the presence or absence (qualitative assessment) of US signs indicative of joint and periarticular inflammation and/or MSU crystal deposits. These signs included synovial fluid and/or synovial hypertrophy; hyperechoic enhancement of the superficial margin of hyaline cartilage (double contour sign); intraarticular or periarticular and intratendinous tophi; bursitis; bone erosions; and tendon tears at the hand, knee, ankle, and foot level. Power Doppler sonography (PDS) was used to assess the blood flow changes at any anatomic site.
All of the images obtained during the first day were stored. While the beginners performed their US examinations, the expert took notes about their mistakes related to technical aspects (for instance, setting of the US system, the way to hold the probe, or the patient's position). The aims of day 1 were to obtain data on baseline scanning ability and to provide the beginners with the most important articles on US and gout.
On day 2, the second session was used to assess the beginners' skills. Utilizing a slide presentation, the expert provided the beginners with the basic knowledge necessary to identify the main US signs of MSU crystal deposits. Moreover, the beginners had access to the core set of US gout images selected in the first step by the neutral investigators to strengthen the knowledge of US findings of gout. After this presentation, US examinations were performed independently by the 3 investigators in the 2 additional patients with gout. The aim of day 2 was to detect early improvement of the beginners' ability to identify US abnormalities in patients with gout.
On day 3, the expert carried out US examinations of another 2 patients in front of the beginners, providing hands-on training and didactic instructions. Additionally, US examinations were also performed in 2 healthy volunteers in order to clarify the normal aspects of the different tissues included in the study. Finally, the quality of the US images acquired and the errors of beginners detected by the expert investigator during the first 2 days of the program were discussed. The aims of day 3 were to improve the potential errors in the technical acquisition of the US images in patients with gout and to acquire greater confidence in terms of the normal US aspects of the different tissues in normal subjects.
On day 4, the beginners performed direct US examinations on 2 different patients with gout under the direct supervision of the expert, who provided his suggestions on both technical aspects and the method used to interpret the US findings. The aims of day 4 were to improve technical acquisition and to address the beginners' questions on the interpretation of US findings.
On day 5, the third session was employed to assess the beginners' skills. The beginners independently performed US examinations on 2 new patients with gout. The aim of this session was to test the beginners' ability to both acquire and interpret US findings without the supervision of or assistance by the expert investigator.
On day 6, 30 images from among those acquired in the previous days were selected by the neutral investigators. The expert showed these images to the beginners and described the US findings in detail. The aim of this day was to respond to the beginners' questions in order to improve agreement and resolve their doubts.
On day 7, the fourth session was used to assess the beginners' skills. This session consisted of final, independent US examinations of 2 patients with gout by all of the investigators. The objective of this day was to assess the agreement between the beginners and the expert at the end of the training program.
The study was conducted according to the Declaration of Helsinki and local regulations, and signed informed consent was obtained from all patients.
US examinations were performed using a MyLab Twice US system (Esaote) equipped with 8–16 MHz and 4–13 MHz broadband linear transducers and a Logiq 9 US system (General Electric Medical Systems) equipped with an 8–15 MHz multifrequency linear transducer.
The following anatomic areas were bilaterally scanned: the hand, second and third metacarpophalangeal (MCP) joints (synovial tissue, hyaline cartilage, and bone of the metacarpal head), and finger flexor tendons of second and third fingers (synovial tissue and tendons); the knee, knee joint (synovial tissue and femoral hyaline cartilage), quadriceps and patellar tendons, and entheses; the ankle, tibiotalar joint (synovial tissue and talar hyaline cartilage), posterior tibialis, peroneus longus and brevis tendons, and Achilles tendons (synovial tissue, tendons, entheses, and retrocalcaneal bursa); and the foot and first metatarsophalangeal (MTP) joint (synovial tissue, hyaline cartilage, and bone of the metatarsal head).
All US examinations were performed using a multiplanar technique adopting the indications suggested by the EULAR guidelines for musculoskeletal US in rheumatology (). Briefly, MCP joints were examined while the patient was seated with hands lying in the prone position on the examination table. Both longitudinal and transverse scans were performed, moving the transducer slightly from the radial to ulnar and from proximal to distal sides on the dorsal aspect to enable maximum coverage of the anatomic surface area. Moreover, a plentiful amount of gel was used in order to avoid compression of the tissues under examination. A full flexion position was adopted for the assessment of the hyaline cartilage.
The patient's knee was examined in a neutral position. Additional scans, performed to widely assess the cartilage surface, included suprapatellar and medial parapatellar views, which were carried out with the knee in maximal flexion. The direction of the US beam was adjusted to be perpendicular to the cartilage surface. Dynamic examination during both compression with the probe and flexion to extension of the knee was performed to identify the superficial margin of the hyaline cartilage.
US examination of the first MTP and tibiotalar joints, peroneus longus and brevis, as well as posterior tibialis tendons was performed with the patient in a supine position and the knee in slight flexion (∼30°). The supine position with extended lower extremities was adopted for the quadriceps and patellar tendons and the suprapatellar pouch, while the Achilles tendons were examined with the patient lying prone and their feet hanging over the edge of the examination table in flexion (90°).
Each anatomic area was scanned in gray-scale mode to detect morphostructural changes and subsequently with the PDS technique to detect abnormal blood flow. Blood flow was examined with a pulse repetition frequency of 750 Hz and a Doppler frequency between 6.6 and 9.1 MHz.
Interpretation of US findings.
US definitions described by the OMERACT Special Interest Group () were adopted for the current study. Synovitis was defined as the presence of either synovial fluid and/or synovial hypertrophy. Synovial fluid was defined as abnormal hypoechoic or anechoic intraarticular material that is displaceable and compressible, but that does not exhibit a Doppler signal. Synovial hypertrophy was defined as abnormal hypoechoic intraarticular tissue that is nondisplaceable and poorly compressible and that may exhibit a Doppler signal. Hyperechoic aggregates of variable shape and echogenicity within the joint cavity were considered intraarticular tophi ().
Bone erosion was defined as intraarticular discontinuity of the bone cortex that is visible on 2 perpendicular planes. Additionally, hyperechoic enhancement of the superficial margin of hyaline cartilage was regarded as a sign of MSU crystal deposition (double contour sign) ([17, 19-22]). Inhomogeneous tendon thickening with loss of typical fibrillar echotexture due to the presence of hyperechoic bands defined the presence of intratendinous tophi. Focal and complete losses of tendon substance visualized on 2 perpendicular scanning planes were considered a partial and total tear, respectively ().
All statistical analyses were performed using MedCalc software, version 10.0 for Windows XP (Microsoft). Descriptive results including the capability of the beginners at the beginning and end of the exercise were expressed as the mean ± SD. For the interobserver agreement, the beginners' results were compared with those of the expert as the gold standard utilizing unweighted kappa for qualitative assessment (i.e., presence/absence). A kappa value of 0–0.20 was considered poor, 0.21–0.40 fair, 0.41–0.60 moderate, 0.61–0.80 good, and 0.81–1.00 excellent. One-sample t-test was used to compare agreement between the groups. A statistical significance value was set at a P value less than 0.05. Sensitivity, specificity, and negative and positive predictive values for any investigator were calculated using the results of the last evaluation.
A total of 12 patients with gout (all men) were included. Eight patients were scanned directly by all of the investigators during the 4 sessions of the US evaluation, whereas 4 patients were scanned during hands-on and practical sessions by the expert sonographer. The mean ± SD age and disease duration were 60.4 ± 6.4 years and 7.4 ± 2.6 years, respectively. In total, 416 anatomic sites (10 joints, 4 finger flexor tendons, 2 quadriceps tendons, 2 patellar tendons, 2 posterior tibialis tendons, 4 peroneous tendons, and 2 Achilles tendons for each patient) were studied.
Table 2 shows the kappa values and overall agreement percentages between all of the investigators for any single abnormality at the different times of the training program. Figures 1 and 2 show the main US findings obtained by the beginners and the learning curve graphic of all of the participants by kappa coefficients, respectively. Briefly, both the kappa values and overall agreement percentages of the qualitative assessments of US gout findings showed moderate to excellent agreement at the end of the exercise, while there was mostly poor/fair agreement obtained in the first session (Table 2).
|US findings||Session I (day 1)||Session II (day 2)||Session III (day 5)||Session IV (day 7)||Pa|
|Periarticular tophaceous deposits|
The comparison between the expert (gold standard) and beginners' examinations at the fourth and final session, including the kappa values, sensitivity, specificity, and negative and positive predictive value, is shown in Table 3. The beginners' most frequent questions or doubts formulated during the exercise are listed in Table 4. The correlation among beginners in the assessment of US findings of gout is shown in Figure 3.
|Pathologic finding||Expert (gold standard)||Totala||κ||Sensitivity, %||Specificity, %||NPV, %||PPV, %|
|Periarticular tophaceous deposits|
|How can we set the machine to better visualize gout-related ultrasound findings?|
|How can we distinguish the subtle double contour sign?|
|How can we differentiate the double contour sign from normal chondrosynovial interface?|
|What are the differences between enthesophytes and intratendinous hyperechoic bands?|
|How can we discriminate intraarticular tophi and synovial hypertrophy?|
Taking into account the expert results of all 4 sessions, the distribution of pathologic findings was as follows: 127 synovitis (23 with intraarticular PDS signal), 56 tenosynovitis, 47 intratendon tophi, 35 double contour signs, 29 bone erosions, 10 bursitis, and 2 tendon ruptures. The mean ± SD time employed by the beginners in the US assessment was 32.5 ± 6.7 minutes (range 16–38 minutes) at the beginning of the exercise, whereas the mean ± SD time was 18.4 ± 4.2 minutes (range 15–22 minutes) at the end of the exercise.
There is increasing interest in understanding the potential role of US in the diagnosis and followup of patients with chronic arthritis. However, the impact of US findings strictly depends on both the experience and the expertise of the sonographer and on the quality of the US equipment. The use of US is currently supported and justified by a growing number of studies ([17-29]). Competence in the knowledge and interpretation of US findings is becoming part of the core curriculum for rheumatologists in Europe () and US was mentioned in EULAR evidence-based recommendations for gout as a potential tool for diagnosis and monitoring of the disease ([30, 31]). Currently, to our knowledge, there are no proposed studies aimed at developing and testing US training programs in gout for rheumatologists, and international consensus on what the best US educational program should be remains lacking. Taking into account the growing number of rheumatologists who are incorporating US into their clinical assessment of gout as a useful imaging method for both diagnostic and research purposes, forming standardized disease-oriented US programs may become an issue to address.
To the best of our knowledge, this is the first study describing the learning curve of rheumatologists with limited experience in US attending a disease-oriented training program focused on the assessment of the main US signs indicative of MSU crystal depositions in different tissues in patients with gout. Previous studies examining the reliability of US in the assessment of MSU deposition in patients with gout have been performed ([19, 22, 36]). In these studies, excellent to nearly perfect interobserver agreement was achieved, but these studies were performed by well-experienced investigators in the field of US. Our study demonstrates that researchers with limited experience in US can achieve good to excellent interobserver agreement rates for the detection of the main pathologic US findings of gout in different tissues after a 1-week program.
On the basis of the results of the present study, the following considerations can be formulated. First, at the end of the exercise, both beginners improved their performance of all US findings due to MSU deposits. In some cases, the kappa value range changed from poor to excellent. Additionally, the time beginners spent performing US examinations was significantly reduced at the final session. Second, the US abnormality with the lowest level of agreement among beginners was the double contour sign. In fact, in the first session, agreement was very poor. The difficulty in interpreting this specific US finding of gout was also listed among the most common questions that the beginners formulated during the exercise (Table 4). The discrepancy between the gold standard and the beginners' assessment in this case was caused by overestimation of pathologic US findings by the beginners, probably due to the fact that the participants were not blinded to the patient's diagnosis. Conversely, detection of US pathologic findings not strictly related to gout (i.e., bone erosions and bursitis) exhibited a higher agreement level at baseline and a steep learning curve. This aspect could also be explained by the fact that detection of bone erosions and bursitis could be frequently found in other diseases, so their detection is an essential requirement for sonographers with basic knowledge. In our study, both beginners had a basic knowledge of US, which facilitated the detection of these elementary US lesions.
The data obtained in the present study may be interpreted with the following limitations. First, patients may have US features with different levels of difficulty to be detected; while large MSU deposits may be easy to find, crystal aggregates that are smaller in size could require skilled and experienced sonographers to reveal these. However, the wide inter- and intraindividual heterogeneity in the size and distribution of the MSU deposits characteristic of patients with gout renders patients identical for US findings and MSU deposits impossible to find. Possible bias was minimized by a careful selection of the group of subjects by the 2 neutral investigators prior to initiating the second step of the study. Second, the high rates of agreement may have been determined because of the fact that only dichotomous data were included. Third, the participants assessed only patients with gout; the lack of a control group, especially disease controls such as those with other crystalline arthropathies or osteoarthritis, may have led to the overestimation of agreement.
In conclusion, the present study suggests that, after a 1-week disease-oriented training program, rheumatologists with limited experience in US were satisfactorily able to detect and interpret the main US signs indicative of MSU crystal deposits in different tissues in patients with gout.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Gutiérrez had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Gutiérrez, Di Geso, Rovisco, Di Carlo, Ariani, Filippucci, Grassi.
Acquisition of data. Gutiérrez, Di Geso, Rovisco, Di Carlo, Ariani, Filippucci, Grassi.
Analysis and interpretation of data. Gutiérrez, Di Geso, Rovisco, Di Carlo, Ariani, Filippucci, Grassi.
- 13Pan-American League of Associations for Rheumatology (PANLAR) Ultrasound Study Group.Pan-American League of Associations for Rheumatology (PANLAR) recommendations and guidelines for musculoskeletal ultrasound training in the Americas for rheumatologists.J Clin Rheumatol2010;16:113–8., , , , , , et al, for the
- 14Practice, training, and assessment among experts performing musculoskeletal ultrasonography: toward the development of an international consensus of educational standards for ultrasonography for rheumatologists.Arthritis Rheum2004;51:1018–22., , , , , .
- 26Crystal-associated synovitis: ultrasonographic feature and clinical correlation.Ortop Traumatol Rehabil2008;10:99–110., , .
- 30EULAR Standing Committee for International Clinical Studies Including Therapeutics.EULAR evidence based recommendations for gout. Part I: diagnosis. Report of a Task Force of the Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT).Ann Rheum Dis2006;65:1301–11., , , , , , et al, for the
- 31EULAR Standing Committee for International Clinical Studies Including Therapeutics.EULAR evidence based recommendations for gout. Part II: management. Report of a Task Force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT).Ann Rheum Dis2006;65:1312–24., , , , , , et al, for the