Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Correlation and agreement between physical and ultrasound examination after a training session dedicated to the standardization of synovitis assessment in rheumatoid arthritis patients

Correlation and agreement between physical and ultrasound examination after a training session... Objectives: Assessing disease activity in rheumatoid arthritis (RA) patients requires comprehensive quantification of tender and swollen joints. We aimed to evaluate the correlation and agreement between rheumatologists after a training session dedicated to the standardization of synovitis assessment and compare its performance with a refer‑ ence imaging modality such as musculoskeletal ultrasonography (MSUS). Methods: In this cross‑sectional study, a total of 28 and 10 joints in RA patients were evaluated by physical exami‑ nation and ultrasound (US), respectively. After participating in a training session, individual joint assessment for tenderness and swelling was performed by three rheumatologists. MSUS examination was performed separately by an experimented radiologist in a standardized manner, evaluating findings according to the Outcome Measures in Rheumatology Clinical Trial (OMERACT ) guidelines. Results: A total of 80 RA patients were included, with a mean Disease Activity Score based on 28 joints (DAS28)‑ ESR of 4.02. The interobserver overall agreement and concordance rate in a total of 2240 joints assessed was 81.7% (k = 0.449, p < 0.0001) for tender joints and 66% (k = 0.227, p < 0.0001) for swollen joints. The overall concordance rate was fair (Fleiss’ kappa = 0.21, p = 0.027) with an overall agreement of 67.18% yet, more joints were found to be swollen by the US assessment, compared to the physical examination (43% vs 39%). Conclusion: In our study population, joint tenderness showed better interobserver agreement, correlation, and concordance rate than joint swelling. When comparing the US assessment to the physical examination, a fair overall concordance rate supports the need for the implementation of training sessions dedicated to standardization in rheu‑ matology clinics. Keywords: Physical examination, Rheumatoid arthritis, Synovitis, Ultrasonography Introduction Rheumatoid arthritis (RA) is a chronic systemic auto- *Correspondence: gquintanal@unal.edu.co immune disease characterized by hyperplasia and Reumavance Group, Rheumatology Section, Department of Internal Medicine, Fundación Santa Fe de Bogotá University Hospital, Carrera 7 inflammation of synovial tissue with subsequent bone No. 117‑15, 110111 Bogota, Colombia erosion and loss of joint space with a noteworthy Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Quintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 2 of 7 association on functionality and quality of life [1]. In Materials and methods order to assess disease activity in RA patients, there is Patients not an individual parameter to be used. It is thus nec- Patients were recruited from an outpatient clinic based in essary to achieve a comprehensive approach through the Rheumatology Department at Fundación Santa Fe de the use of several individual clinical and or laboratory Bogotá University Hospital, Colombia. Stratified random parameters and to develop quantitative indexes to be sampling was conducted, selecting patients with different used in daily clinical practice by rheumatology health disease activity states (according to the most recent clini- professionals [2–4]. cal record) from a previously established ongoing cohort Among the most commonly used methods, the of 820 patients. The sample size was calculated for the Disease Activity Score (DAS), and its modified ver- desired correlation coefficient of 0.6, a population corre - sion, DAS28, are based on a tender and swollen joint lation coefficient of 0.8, a power of 0.8, and a confidence count, combined with other parameters such as a interval of 0.95. All those who were invited to participate patient global health assessment [4, 5]. Joint tender- fulfilled the 2010 ACR/European League Against Rheu - ness assessments estimate the patient’s response to matism (EULAR) classification criteria [10] and were at potentially painful stimulation. On the other hand, least 18  years old. Those who had a history of trauma, joint swelling assessment measures synovial inflamma- septic arthritis, joint replacement or synovectomy, joint tion or effusion recognized by fluid displacement. The deformity, and or crystal arthropathy were excluded. examination technique should include exerting contin- The following data were registered at baseline: age, uous and direct pressure on an affected joint with the gender, treatment, erythrocyte sedimentation rate (ESR), thumb and index fingers until the examiner’s nail bed C-reactive protein (CRP), rheumatoid factor (RF), global turns pale; this compares with a pressure of approxi- health assessed by the patient (GH), Clinical Disease mately 4 kg/cm [6, 7]. Activity Index (CDAI), and Simplified Disease Activity Although tender and swollen joint counts are con- Index (SDAI). sidered fundamental parameters to estimate disease activity, its assessment is not as straightforward as Training session dedicated to standardization assumed, thus exhibiting certain potential disad- Our planned training session was structured to focus on vantages. Amongst the prevailing concerns are poor standardization of the tenderness and swelling assess- reproducibility and substantial interobserver variabil- ment of 28 joints (bilateral shoulders, elbows, wrists, ity [6, 8]. One of the reasonable explanations could be knees, metacarpophalangeal (MCP), and interphalangeal the training and clinical experience gap among practi- (PIP)). tioners; therefore, finding a feasible solution through Three rheumatologists, with 10, 13, and 15  years standardization. of experience in clinical examination of RA patients, Historically, training sessions focused on standardi- attended three sessions (separated by 1–2  weeks). Each zation of all DAS28 parameters assessment have been session was divided into three steps: (1) individual joint proposed, and once applied, have shown to provoke a assessment was performed by each rheumatologist substantial reduction regarding the variation between (blinded to both clinical and other rheumatologist’s data), examiners [6–8]. The entire medical and nursing staff (2) 20-min discussion on practice observations, difficul - who will be assessing RA patients should be trained ties, limitations, and facilitating factors during the physi- regularly (at least once every year) and should be cal examination, in order to reach agreement on uniform trained together, allowing discussion about current criteria and technique, and (3) joint individual reassess- standard procedures. There has been proposed as well ment based on the unified criteria. On the third session, the use of ultrasonography to demonstrate active syno- disease activity indexes (DAS28-ESR, SDAI, and CDAI) vitis in case of disagreement [6]. were calculated individually by each rheumatologist. Prognostic factors, treatment decisions, monitoring, and complications are defined based on disease activ- Ultrasound assessment ity, thus consequently on a reliable tender and swol- Twenty minutes after the last training session’s third step, len joint count for each RA patient [9]. The aim of this each patient was instructed to proceed to the ultrasound study was to evaluate the correlation and agreement (US) assessment room. The US examination was per - between rheumatologists after a training session dedi- formed by a radiologist with 15  years of experience and cated to the standardization of synovitis assessment training in musculoskeletal radiology (blinded to physical and compare its performance with a reference imag- examinations’ data) on 10 joints (bilateral wrists, and 2nd ing modality such as musculoskeletal ultrasonography to 5th MCPs), using a GE (General Electric) LOGICQ (MSUS). E ultrasound machine with a 6–13  Hz multifrequency Q uintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 3 of 7 linear transducer. Findings were systematically graded Table 1 Clinical and demographic characteristics for all patients according to the Outcome Measures in Rheumatology Characteristics Values (n = 80) Clinical Trial (OMERACT) guidelines, evaluating syno- Age, years 55.3 (8.6) vitis as synovial hypertrophy defined by an abnormal Female, n (%) 68 (85) nondisplaceable and poorly compressible hypoechoic Disease duration (years) 11.56 (7.93) intraarticular tissue [11]. PIPs joins were considered as Positive RF, n (%) 71 (88) potential confounders due to the eventual overlapping of ESR (mm/h) 19.32 (5.6) tenosynovitis, therefore were not assessed. CRP (mg/dL) 5.6 (12) Synovitis grading was conducted based on a scoring DAS28‑ESR 4.02 (1.12) method initially introduced by Szkudlarek et  al., widely SDAI 24.4 (8.59) used for studies of this kind [12–14] and currently sup- CDAI 18.7 (9.01) ported by the EULAR-OMERACT ultrasound task- Number of DMARDs force [15, 16] (0 = absence of synovial thickening, 1 1, n (%) 15 (18%) (mild) = minimal synovial thickening, filling the angle ≥ 2, n (%) 58 (72%) between the periarticular bones, without bulging over bDMARD use, n (%) 16 (20) the line linking tops of the corresponding bones, 2 (mod- Oral glucocorticoid use, n (%) 27 (33) erate) = synovial thickening bulging over the line link- TJC‑28, median 9 ing tops of the periarticular bones but without extension along the bone diaphysis, 3 (severe) = severe synovial SJC‑28, median 5 thickening bulging over the line linking tops of the peri- Values are the mean (SD) unless otherwise indicated articular bones and with extension to at least one bone RF, rheumatoid factor; ESR, erythrocyte sedimentation rate; CRP, C‑reactive protein; DAS28‑ESR, Disease Activity Score based on 28 joint count and ESR; diaphysis). Normal distances between bone and joint SDAI, Simplified Disease Activity Index; CDAI, Clinical Disease Activity Index; capsule were acknowledged based on average population DMARDs, disease‑modified anti‑rheumatic drug; bDMARD, biological disease ‑ values proposed by Schmidt et al. [17]. modified anti‑rheumatic drug; TJC, tender joint count; SJC, swollen joint count Statistical analysis Statistical analysis was performed using STATA soft- ware, version 15.0. Descriptive analysis was presented for of them were women (85%), mean age was 55.3  years, continuous variables with central tendency measures as with 71 (88%) being seropositive for RF. The mean mean and standard deviation (SD) or as the median and (± SD) DAS28-ESR, SDAI, and CDAI were 4.02 ± 1.12 interquartile range (IQR) for normally or nonnormally (moderate activity), 24.4 ± 8.59, and 18.7 ± 9.01, respec- distributed data, respectively. For dichotomous vari- tively. All our patients received disease-modifying anti- ables, data were presented with percentages and absolute rheumatic drugs (DMARDs) treatment, corresponding values. to mono, ≥ 2 DMARD, and biological (b) DMARD Interobserver agreement and concordance were cal- therapy in 18%, 72%, and 20% of cases. Twenty-seven culated through Cohen’s kappa (between two observ- (33%) patients were receiving oral glucocorticoid ther- ers considering all the possible pairs, i.e., Observer A apy (daily dose of 5–10 mg (prednisone equivalent)). vs B, observer A vs C, observer B vs C), Fleiss’ kappa (between the three observers), and percentage of an overall agreement (percentage of observed exact agree- Interobserver correlation and agreement ments). The relative strength of agreement was described A total of 2240 joints (i.e., 28 joints in 80 patients) according to the following ranges of kappa (k) coeffi - were evaluated through physical examination. Com- cients: < 0.00 = poor, 0.00–0.20 = slight, 0.21–0.40 = fair, pared with tenderness, swelling assessment showed 0.41–0.60 = moderate, 0.61–0.80 = substantial, and 0.81– a slight strength of agreement with lower percentage 1.00 = almost perfect [18]. Linear correlation coefficients of overall agreement and concordance rates (66% and (Pearson correlation coefficient) were also calculated for Fleiss kappa = 0.227; 95% CI 0.203–0.250, p < 0.0001 vs tender and swollen joint counts. 81.7% and Fleiss kappa = 0.449; 95% CI 0.403–0.493, p < 0.0001) (see Table 2). Results The linear correlation coefficient (Spearman’s corre - Patient characteristics lation coefficient) was calculated, taking all the possi - The demographic and clinical characteristics of the 80 ble pairs (observer A–B, observer A–C, observer B–C), selected patients are illustrated in Table 1. The majority thus having an overall correlation coefficient of 0.81 and 0.431 for swollen and tender joints, respectively. Quintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 4 of 7 Table 2 Interobserver agreement and concordance for physical examination (PE) tenderness and swelling assessment Joint Inter-observer agreement and concordance PE tenderness assessment PE swelling assessment Fleiss kappa 95% CI p value Count Fleiss kappa 95% CI p value Count Right shoulder 0.521 0.494–0.652 < 0.0001 83 0.249 0.124–0.322 0.0001 27 Left shoulder 0.446 0.198–0.546 < 0.0001 64 0.134 0.092–0.184 0.018 23 Right elbow 0.4 0.308–0.537 < 0.0001 71 0.146 0.092–0.287 0.012 69 Left elbow 0.403 0.258–0.446 < 0.0001 64 0.245 0.166–0.392 0.0001 79 Right wrist 0.392 0.263–0.548 < 0.0001 78 0.121 0.007–0.168 0.030 123 Left wrist 0.325 0.261–0.419 < 0.0001 72 0.042 − 0.192 to 0.234 0.259 145 Right MCP1 0.394 0.326–0.551 < 0.0001 50 − 0.166 − 0.178 to 0.151 0.995 134 Right MCP2 0.298 0.225–0.362 < 0.0001 53 0.13 0.109–0.140 0.023 158 Right MCP3 0.282 0.153–0.323 < 0.0001 49 0.218 0.070–0.292 0.0004 109 Right MCP4 0.297 0.235–0.379 < 0.0001 33 0.397 0.376–0.476 < 0.0001 53 Right MCP5 0.379 0.236–0.477 < 0.0001 36 0.306 0.272–0.387 < 0.0001 63 Left MCP1 0.304 0.187–0.394 < 0.0001 44 0.094 0.024–0.148 0.073 95 Left MCP2 0.444 0.391–0.619 < 0.0001 36 0.262 0.133–0.314 < 0.0001 110 Left MCP3 0.382 0.323–0.453 < 0.0001 41 0.27 0.268–0.334 < 0.0001 69 Left MCP4 0.315 0.086–0.375 < 0.0001 32 0.177 0.059–0.232 0.003 45 Left MCP5 0.353 0.183–0.569 < 0.0001 26 0.217 0.172–0.266 0.0004 71 Right PIP1 0.299 0.050–0.367 < 0.0001 22 0.128 0.087–0.227 0.023 53 Right PIP2 0.353 0.310–0.379 < 0.0001 28 0.042 0.015–0.094 0.258 95 Right PIP3 0.351 0.312–0.486 < 0.0001 30 − 0.017 − 0.102 to 0.113 0.603 118 Right PIP4 0.232 0.111–0.250 < 0.0001 28 0.074 0.037–0.117 0.126 82 Right PIP5 0.396 0.294–0.431 < 0.0001 26 0.245 0.091–0.318 0.244 48 Left PIP1 0.165 0.018–0.304 < 0.0001 21 0.02 − 0.203 to 0.242 0.376 39 Left PIP2 0.167 0.085–0.299 < 0.0001 24 − 0.061 − 0.236 to 0.2 0.827 50 Left PIP3 0.298 0.196–0.368 < 0.0001 35 0.077 − 0.071 to 0.219 0.116 101 Left PIP4 0.185 0.018–0.267 < 0.0001 31 0.121 0.056–0.197 0.030 74 Left PIP5 0.339 0.270–0.573 < 0.0001 18 0.113 0.041–0.267 0.040 46 Right knee 0.476 0.318–0.530 < 0.0001 48 0.428 0.295–0.525 < 0.0001 58 Left knee 0.379 0.350–0.471 < 0.0001 57 0.214 0.038–0.362 0.0005 62 Overall agreement (%) 0.449 (81.7%) 0.403–0.493 < 0.0001 0.227 (66%) 0.203–0.250 < 0.0001 MCP, metacarpophalangeal; PIP, proximal interphalangeal Agreement and concordance between physical and US Table 3 Agreement and concordance between physical and US examination examination swelling assessment What is interesting in Table  3 is that, from the then Joint Overall Fleiss kappa p value joints assessed by both methods, there has been a agreement (%) slight relative increase in the number of swollen joints Right wrist 62.5 0.241 0.007 through US examination (345 of 800 joints assessed Left wrist 63.7 0.213 0.022 by US (43%)), when compared to physical examination Right MCP 2 60.0 0.195 0.040 (946 of 2400 joints assessed by our three rheumatolo- Right MCP 3 66.2 0.264 0.009 gist (39%)). Concordance strength of agreement was fair Right MCP 4 69.6 0.197 0.039 and ranged from Fleiss kappa 0.168–0.264. Likewise, Right MCP 5 70.9 0.244 0.014 the k coefficients between physical and US examina - Left MCP 2 62.5 0.244 0.011 tion in the left MCP5 joint (k = 0.168, p 0.06) followed Left MCP 3 67.5 0.228 0.019 by left MCP4 (k = 0.177, p 0.053) joint stand out as Left MCP 4 77.5 0.177 0.053 the lowest values. By contrast, k coefficient in wrist Left MCP 5 71.2 0.168 0.06 was slightly higher (k = 0.241 (p 0.007) and k = 0.213 MCP, metacarpophalangeal; PIP, proximal interphalangeal (p 0.023)). The percentage of overall agreement and Q uintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 5 of 7 are similar to ours, showing a more robust correlation concordance rate was 67.18% and k = 0.210; (p 0.027), regarding tender joints compared to that of swollen respectively. joints. When compared to MUSC, as previously stated, agree- Discussion ment and concordance were slightly lower for swelling As explained in the introduction, it is clear that training assessment. This discrepancy could be attributed to the sessions focused on the standardization of joint assess- mean age of our patients (55 years); it seems possible that ment techniques play a pivotal role in understanding the older the patient, the more frequent synovial thick- the reliability of quantitative indexes daily used in RA ening and incipient joint deformities, thus being con- patients. We aimed to evaluate the correlation and agree- founding for not only MSUS but also for PE assessment. ment between rheumatologists after such a training Moreover, it is important to consider what has been session and compare its performance with a reference widely held by recent reports, that is, the trend towards imaging modality, namely, MSUS, to find a reasonable higher agreement and concordance rates regarding swol- approach on behalf of the inevitable high intra- and inter- len joints count [22, 23]. These data must be interpreted observer variability described in the literature. with caution because of the use of the OMERACT syno- Contrary to expectations, despite the fact that all three vitis definition and its implications of considering even participating rheumatologists attended the proposed a minimal amount of intraarticular tissue as an abnor- training sessions, interobserver variability among them mal finding of synovial hypertrophy, thereby involving a was still present. The wide range of perceived concord - potential overestimation. ance rates suggests that the assessment of some individ- It could be argued that the average age and the duration ual joints may be particularly challenging on their own, and activity of the disease could contribute to the differ - thereby showing specific difficulties during standardiza - ences between physical examination and US; neverthe- tion, with possible attribution to an underlying long and less, those patients with joint deformity were excluded. It comprehensive learning process. is relevant to bear in mind that, in those populations with In terms of overall interobserver agreement and con- a poorly controlled- and a longer- disease, joint assess- cordance rates, as well as of overall correlation, our ment either by physical examination or by ultrasound findings propose that tenderness assessment was more remains a challenge when considering joint surface irreg- homogenous than swelling assessment. A possible expla- ularities. It supports the pivotal role of conventional radi- nation for these results may be the fact that joint tender- ography as the first choice for the evaluation of structural ness is inferred solely by the patient’s response, while that changes such as erosions. of joint swelling by physicians’ technique and interpreta- The limitations of this study include firstly the absence tion of findings. of power Doppler ultrasonographic assessment due to The importance of training sessions focused on stand - timing issues on behalf of the vast number of evaluated ardization was first stated by Scott et  al. in 1996 [19], joints; prior studies have noted the role of power Doppler whose findings suggested a considerably increased sensi - US in detecting subclinical and residual synovitis when tivity of measurement for both tender and swollen joints assessing synovial vascularity, although, it was beyond and a reduction in the mean coefficient of variation for the scope of our work to evaluate such conditions. Sec- the number of swollen joints (82% vs 59%) after a 60-min ondly, we did not count on another radiologist with suf- training session based on the EULAR handbook for joint ficient experience in MSUS to be considered as a second evaluation. Unlike Scott et  al., on a multicentric cohort evaluator; additionally, the US evaluation performed by a study evaluating standardization based on the afore- radiologist and not an articular US rheumatologist might mentioned EULAR handbook, Grunke et  al. [20] stated be considered as a source of uncertainty; nonetheless, that even when consistency and variability significantly the vast experience, as well as the specific MSUS train - improved, the mean number of tender, as well as swol- ing of our radiologist, support the internal validity of the len joints decreased. Nonetheless, a reference imaging assessments. Very little was found in the literature on the modality was not used. question of performance differences between those sce - In addition, clinical experience plays an essential part narios (radiologist vs. rheumatologist), remaining as an as recently proposed by El-Hadidi et  al. [21] where after intriguing issue for future research, especially for low and a consensus on joint assessment, the interobserver agree- middle-income countries where training for such medi- ment was calculated to compare an experienced rheuma- cal sub-specialties is not as frequent. tology professor (30  years of experience) with a young Thirdly, the absence of early RA patients, as well as of Rheumatology fellow (3  years of training). Although a patients in clinical remission, and finally, that we did not high correlation between professor and candidate was report pretraining session joint count results in order to described, specific results on joint assessment correlation Quintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 6 of 7 Competing interests propose a variability improvement. The latter considera - This research did not receive any specific grant from funding agencies in the tion must be handled with care when considering that public, commercial, or not‑for ‑profit sectors. We agree to bear the applicable this decision was based on the fact that pretraining ses- publication charges if their manuscript is accepted for publication. All authors declare that they have no competing interest. sion correlation and agreement scores were certainly variable, close to the null value, and even negative. Tak- Author details ing this into consideration, the gain of the stated train- Reumavance Group, Rheumatology Section, Department of Internal Medicine, Fundación Santa Fe de Bogotá University Hospital, Carrera 7 No. ing session was to revert such tendency and yet turn it to 117‑15, 110111 Bogota, Colombia. Department of Internal Medicine, School positive values, even though the kappa coefficients were of Medicine, Universidad Nacional de Colombia, Bogota, Colombia. Depar t‑ in the range of slight to moderate. It is worth mentioning ment of Internal Medicine, School of Medicine, Universidad de Los Andes, Bogota, Colombia. Department of Radiology, School of Medicine, Universidad that when adding more than two observers to the calcu- Nacional de Colombia, Bogota, Colombia. lations, the slightest discrepancy substantially lowers the statistical estimator; thus, an adjusted threshold must be Received: 3 June 2021 Accepted: 5 November 2021 taken when interpreting such results. Finally, a reason- able approach could be to hypothesize that the increase in the total swollen joint count through US examination References would have a substantial implication in the categorization 1. McInnes IB, Schett G. The pathogenesis of rheumatoid arthritis. N Engl J of the activity of the disease; nevertheless, even when it Med. 2011;365:2205–19. was beyond our scope, we trust that future research on 2. Backhaus M, Kamradt T, Sandrock D, Loreck D, Fritz J, Wolf KJ, et al. Arthritis of the finger joints: a comprehensive approach comparing con‑ this behalf will rise as a natural progression of this work. ventional radiography, scintigraphy, ultrasound, and contrast‑ enhanced magnetic resonance imaging. Arthritis Rheum. 1999;42:1232–45. Conclusions 3. Naredo E, Bonilla G, Gamero F, Uson J, Carmona L, Laffon A. Assessment of inflammatory activity in rheumatoid arthritis: a comparative study of Taken together, our findings indicate that joint tender - clinical evaluation with grey scale and power Doppler ultrasonography. ness assessment had a better interobserver agreement Ann Rheum Dis. 2005;64:375–81. and concordance rate, moreover a higher overall corre- 4. Van der Heijde D, Van’t Hof M, Van Riel P, et al. Development of a disease activity score based on judgment in clinical practice by rheumatologists. lation coefficient than joint swelling. The evidence from J Rheumatol. 1993;20:579–81. our results suggests a fair overall concordance rate when 5. Prevoo MLL, Van’t Hof MA, Kuper HH, et al. Modified disease activity comparing the US examination to the physical examina- scores that include twenty‑ eight‑joint counts. Development and valida‑ tion in a prospective longitudinal study of patients with rheumatoid tion, thus supporting the need for the implementation of arthritis. Arthritis Rheum. 1995;38(1):44–8. training sessions dedicated to standardization in rheuma- 6. Porter D, Gadsby K, Thompson P, White J, McClinton C, Oliver S. DAS28 tology clinics to guarantee reliability in identifying tender and rheumatoid arthritis: the need for standardization. Musculoskelet Care. 2011;9(4):222–7. and swollen joints within rheumatologists. 7. Scott IC, Scott DL. Joint counts in inflammatory arthritis. Clin Exp Rheu‑ matol. 2014;32(Suppl. 85):S7–12. Acknowledgements 8. Sokka T, Pincus T. Joint counts to assess rheumatoid arthritis for clinical Not applicable. research and usual clinical care: advantages and limitations. Rheum Dis Clin N Am. 2009;35(4):713–22. Authors’ contributions 9. Smolen JS, Landewé RBM, Bijlsma JWJ, et al. EULAR recommendations for GQ‑L: Conceptualization, methodology, formal analysis, investigation, writing‑ the management of rheumatoid arthritis with synthetic and biological original draft preparation, writing‑review and editing. KM ‑ C: Writing‑ original disease‑modifying antirheumatic drugs: 2019 update. Ann Rheumatic draft preparation, writing‑review and editing. JBF‑S: writing‑ original draft Dis. 2020;79:685–99. preparation, writing‑review and editing. PM ‑P: Conceptualization, methodol‑ 10. Aletaha D, Neogi T, Silman AJ, Funovits J, Felson DT, Bingham CO, et al. ogy, investigation. PC‑A: Conceptualization, methodology, investigation. EC: 2010 Rheumatoid arthritis classification criteria: an American College Investigation. All authors read and approved the final manuscript. of Rheumatology‑European League Against Rheumatism collaborative initiative. Ann Rheum Dis. 2010;69:1580–8. Funding 11. Wakefield RJ, Balint PV, Szkudlarek M, Filippucci E, Backhaus M, D’Agostino This research did not receive any specific grant from funding agencies in the MA, et al. OMERACT 7 special interest group. Musculoskeletal ultrasound public, commercial, or not‑for ‑profit sectors. including definitions for ultrasonography pathology. J Rheumatol. 2005;32:2485–7. Availability of data and materials 12. Szkudlarek M, Wakefield R, Backhaus M, Terslev L. The discriminatory The data underlying this article will be shared on reasonable request to the capacity of ultrasound in rheumatoid arthritis: active vs inactive, early vs corresponding author. advanced, and more. Rheumatology. 2012;51(suppl 7):vii6–9. 13. Wakefield RJ, Green MJ, Marzo ‑ Ortega H, Conaghan PG, Gibbon WW, Declarations McGonagle D, et al. Should oligoarthritis be reclassified? Ultrasound reveals a high prevalence of subclinical disease. Ann Rheum Dis. Ethics approval and consent to participate 2004;63:382–5. This study was conducted with the approval of our Local Research Ethics 14. Szkudlarek M, Court‑Payen M, Jacobsen S, Klarlund M, Thomsen H, Øster ‑ Committee in accordance with the Declaration of Helsinki. Every patient gave gaard M. Interobserver agreement in ultrasonography of the finger and written and spoken informed consent. toe joints in rheumatoid arthritis. Arthritis Rheum. 2003;48(4):955–62. 15. Bruyn GA, Iagnocco A, Naredo E, Balint PV, Gutierrez M, Hammer HB, Consent for publication et al. OMERACT definitions for ultrasonographic pathologies and All authors approved the final version of this manuscript to be published. Q uintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 7 of 7 elementary lesions of rheumatic disorders 15 years on. J Rheumatol. 21. El‑Hadidi K, Gamal S, Saad S. Inter ‑ observer agreement of standard joint 2019;46(10):1388–93. https:// doi. org/ 10. 3899/ jrheum. 181095. count examination and disease global assessment in a cohort of Egyp‑ 16. Terslev L, Naredo E, Aegerter P, et al. Scoring ultrasound synovitis in rheu‑ tian Rheumatoid Arthritis patients. Reumatol Clín. 2019;15(5):258–63. matoid arthritis: a EULAR‑ OMERACT ultrasound taskforce‑Part 2: reliability 22. Hirata A, Ogura T, Hayashi N, Takenaka S, Ito H, Mizushina K, et al. and application to multiple joints of a standardised consensus‑based Concordance of patient‑reported joint symptoms, physician‑ examined scoring system. RMD Open. 2017;3:e000427. https:// doi. org/ 10. 1136/ arthritic signs, and ultrasound‑ detected synovitis in rheumatoid arthritis. rmdop en‑ 2016‑ 000427. Arthritis Care Res. 2017;69(6):801–6. 17. Schmidt WA, Schmidt H, Schicke B, Gromnica‑Ihle E. Standard refer ‑ 23. Salaffi F, Filippucci E, Carotti M, Naredo E, Meenagh G, Ciapetti A, et al. ence values for musculoskeletal ultrasonography. Ann Rheum Dis. Inter‑ observer agreement of standard joint counts in early rheumatoid 2004;63:988–94. arthritis: a comparison with grey scale ultrasonography: a preliminary 18. Landis JR, Koch GG. The measurement of observer agreement for cat‑ study. Rheumatology. 2008;47:54–8. egorical data. Biometrics. 1977;33:159–74. 19. Scott DL, Choy EHS, Greeves A, Isenberg DM, Kassinor D, Rankin E, et al. Publisher’s Note Standardizing joint assessment in rheumatoid arthritis. Clin Rheumatol. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 1996;15(6):579–82. lished maps and institutional affiliations. 20. Grunke M, Antoni C, Kavanaugh A, Hildebrand V, Dechant C, Schett G, et al. Standardization of joint examination technique leads to a signifi‑ cant decrease in variability among different examiners. J Rheumatol. 2010;37(4):860–4. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Advances in Rheumatology Springer Journals

Correlation and agreement between physical and ultrasound examination after a training session dedicated to the standardization of synovitis assessment in rheumatoid arthritis patients

Download PDF
7 pages

Loading next page...
 
/lp/springer-journals/correlation-and-agreement-between-physical-and-ultrasound-examination-XdYnbLrCNB

References (53)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2523-3106
DOI
10.1186/s42358-021-00227-y
Publisher site
See Article on Publisher Site

Abstract

Objectives: Assessing disease activity in rheumatoid arthritis (RA) patients requires comprehensive quantification of tender and swollen joints. We aimed to evaluate the correlation and agreement between rheumatologists after a training session dedicated to the standardization of synovitis assessment and compare its performance with a refer‑ ence imaging modality such as musculoskeletal ultrasonography (MSUS). Methods: In this cross‑sectional study, a total of 28 and 10 joints in RA patients were evaluated by physical exami‑ nation and ultrasound (US), respectively. After participating in a training session, individual joint assessment for tenderness and swelling was performed by three rheumatologists. MSUS examination was performed separately by an experimented radiologist in a standardized manner, evaluating findings according to the Outcome Measures in Rheumatology Clinical Trial (OMERACT ) guidelines. Results: A total of 80 RA patients were included, with a mean Disease Activity Score based on 28 joints (DAS28)‑ ESR of 4.02. The interobserver overall agreement and concordance rate in a total of 2240 joints assessed was 81.7% (k = 0.449, p < 0.0001) for tender joints and 66% (k = 0.227, p < 0.0001) for swollen joints. The overall concordance rate was fair (Fleiss’ kappa = 0.21, p = 0.027) with an overall agreement of 67.18% yet, more joints were found to be swollen by the US assessment, compared to the physical examination (43% vs 39%). Conclusion: In our study population, joint tenderness showed better interobserver agreement, correlation, and concordance rate than joint swelling. When comparing the US assessment to the physical examination, a fair overall concordance rate supports the need for the implementation of training sessions dedicated to standardization in rheu‑ matology clinics. Keywords: Physical examination, Rheumatoid arthritis, Synovitis, Ultrasonography Introduction Rheumatoid arthritis (RA) is a chronic systemic auto- *Correspondence: gquintanal@unal.edu.co immune disease characterized by hyperplasia and Reumavance Group, Rheumatology Section, Department of Internal Medicine, Fundación Santa Fe de Bogotá University Hospital, Carrera 7 inflammation of synovial tissue with subsequent bone No. 117‑15, 110111 Bogota, Colombia erosion and loss of joint space with a noteworthy Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Quintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 2 of 7 association on functionality and quality of life [1]. In Materials and methods order to assess disease activity in RA patients, there is Patients not an individual parameter to be used. It is thus nec- Patients were recruited from an outpatient clinic based in essary to achieve a comprehensive approach through the Rheumatology Department at Fundación Santa Fe de the use of several individual clinical and or laboratory Bogotá University Hospital, Colombia. Stratified random parameters and to develop quantitative indexes to be sampling was conducted, selecting patients with different used in daily clinical practice by rheumatology health disease activity states (according to the most recent clini- professionals [2–4]. cal record) from a previously established ongoing cohort Among the most commonly used methods, the of 820 patients. The sample size was calculated for the Disease Activity Score (DAS), and its modified ver- desired correlation coefficient of 0.6, a population corre - sion, DAS28, are based on a tender and swollen joint lation coefficient of 0.8, a power of 0.8, and a confidence count, combined with other parameters such as a interval of 0.95. All those who were invited to participate patient global health assessment [4, 5]. Joint tender- fulfilled the 2010 ACR/European League Against Rheu - ness assessments estimate the patient’s response to matism (EULAR) classification criteria [10] and were at potentially painful stimulation. On the other hand, least 18  years old. Those who had a history of trauma, joint swelling assessment measures synovial inflamma- septic arthritis, joint replacement or synovectomy, joint tion or effusion recognized by fluid displacement. The deformity, and or crystal arthropathy were excluded. examination technique should include exerting contin- The following data were registered at baseline: age, uous and direct pressure on an affected joint with the gender, treatment, erythrocyte sedimentation rate (ESR), thumb and index fingers until the examiner’s nail bed C-reactive protein (CRP), rheumatoid factor (RF), global turns pale; this compares with a pressure of approxi- health assessed by the patient (GH), Clinical Disease mately 4 kg/cm [6, 7]. Activity Index (CDAI), and Simplified Disease Activity Although tender and swollen joint counts are con- Index (SDAI). sidered fundamental parameters to estimate disease activity, its assessment is not as straightforward as Training session dedicated to standardization assumed, thus exhibiting certain potential disad- Our planned training session was structured to focus on vantages. Amongst the prevailing concerns are poor standardization of the tenderness and swelling assess- reproducibility and substantial interobserver variabil- ment of 28 joints (bilateral shoulders, elbows, wrists, ity [6, 8]. One of the reasonable explanations could be knees, metacarpophalangeal (MCP), and interphalangeal the training and clinical experience gap among practi- (PIP)). tioners; therefore, finding a feasible solution through Three rheumatologists, with 10, 13, and 15  years standardization. of experience in clinical examination of RA patients, Historically, training sessions focused on standardi- attended three sessions (separated by 1–2  weeks). Each zation of all DAS28 parameters assessment have been session was divided into three steps: (1) individual joint proposed, and once applied, have shown to provoke a assessment was performed by each rheumatologist substantial reduction regarding the variation between (blinded to both clinical and other rheumatologist’s data), examiners [6–8]. The entire medical and nursing staff (2) 20-min discussion on practice observations, difficul - who will be assessing RA patients should be trained ties, limitations, and facilitating factors during the physi- regularly (at least once every year) and should be cal examination, in order to reach agreement on uniform trained together, allowing discussion about current criteria and technique, and (3) joint individual reassess- standard procedures. There has been proposed as well ment based on the unified criteria. On the third session, the use of ultrasonography to demonstrate active syno- disease activity indexes (DAS28-ESR, SDAI, and CDAI) vitis in case of disagreement [6]. were calculated individually by each rheumatologist. Prognostic factors, treatment decisions, monitoring, and complications are defined based on disease activ- Ultrasound assessment ity, thus consequently on a reliable tender and swol- Twenty minutes after the last training session’s third step, len joint count for each RA patient [9]. The aim of this each patient was instructed to proceed to the ultrasound study was to evaluate the correlation and agreement (US) assessment room. The US examination was per - between rheumatologists after a training session dedi- formed by a radiologist with 15  years of experience and cated to the standardization of synovitis assessment training in musculoskeletal radiology (blinded to physical and compare its performance with a reference imag- examinations’ data) on 10 joints (bilateral wrists, and 2nd ing modality such as musculoskeletal ultrasonography to 5th MCPs), using a GE (General Electric) LOGICQ (MSUS). E ultrasound machine with a 6–13  Hz multifrequency Q uintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 3 of 7 linear transducer. Findings were systematically graded Table 1 Clinical and demographic characteristics for all patients according to the Outcome Measures in Rheumatology Characteristics Values (n = 80) Clinical Trial (OMERACT) guidelines, evaluating syno- Age, years 55.3 (8.6) vitis as synovial hypertrophy defined by an abnormal Female, n (%) 68 (85) nondisplaceable and poorly compressible hypoechoic Disease duration (years) 11.56 (7.93) intraarticular tissue [11]. PIPs joins were considered as Positive RF, n (%) 71 (88) potential confounders due to the eventual overlapping of ESR (mm/h) 19.32 (5.6) tenosynovitis, therefore were not assessed. CRP (mg/dL) 5.6 (12) Synovitis grading was conducted based on a scoring DAS28‑ESR 4.02 (1.12) method initially introduced by Szkudlarek et  al., widely SDAI 24.4 (8.59) used for studies of this kind [12–14] and currently sup- CDAI 18.7 (9.01) ported by the EULAR-OMERACT ultrasound task- Number of DMARDs force [15, 16] (0 = absence of synovial thickening, 1 1, n (%) 15 (18%) (mild) = minimal synovial thickening, filling the angle ≥ 2, n (%) 58 (72%) between the periarticular bones, without bulging over bDMARD use, n (%) 16 (20) the line linking tops of the corresponding bones, 2 (mod- Oral glucocorticoid use, n (%) 27 (33) erate) = synovial thickening bulging over the line link- TJC‑28, median 9 ing tops of the periarticular bones but without extension along the bone diaphysis, 3 (severe) = severe synovial SJC‑28, median 5 thickening bulging over the line linking tops of the peri- Values are the mean (SD) unless otherwise indicated articular bones and with extension to at least one bone RF, rheumatoid factor; ESR, erythrocyte sedimentation rate; CRP, C‑reactive protein; DAS28‑ESR, Disease Activity Score based on 28 joint count and ESR; diaphysis). Normal distances between bone and joint SDAI, Simplified Disease Activity Index; CDAI, Clinical Disease Activity Index; capsule were acknowledged based on average population DMARDs, disease‑modified anti‑rheumatic drug; bDMARD, biological disease ‑ values proposed by Schmidt et al. [17]. modified anti‑rheumatic drug; TJC, tender joint count; SJC, swollen joint count Statistical analysis Statistical analysis was performed using STATA soft- ware, version 15.0. Descriptive analysis was presented for of them were women (85%), mean age was 55.3  years, continuous variables with central tendency measures as with 71 (88%) being seropositive for RF. The mean mean and standard deviation (SD) or as the median and (± SD) DAS28-ESR, SDAI, and CDAI were 4.02 ± 1.12 interquartile range (IQR) for normally or nonnormally (moderate activity), 24.4 ± 8.59, and 18.7 ± 9.01, respec- distributed data, respectively. For dichotomous vari- tively. All our patients received disease-modifying anti- ables, data were presented with percentages and absolute rheumatic drugs (DMARDs) treatment, corresponding values. to mono, ≥ 2 DMARD, and biological (b) DMARD Interobserver agreement and concordance were cal- therapy in 18%, 72%, and 20% of cases. Twenty-seven culated through Cohen’s kappa (between two observ- (33%) patients were receiving oral glucocorticoid ther- ers considering all the possible pairs, i.e., Observer A apy (daily dose of 5–10 mg (prednisone equivalent)). vs B, observer A vs C, observer B vs C), Fleiss’ kappa (between the three observers), and percentage of an overall agreement (percentage of observed exact agree- Interobserver correlation and agreement ments). The relative strength of agreement was described A total of 2240 joints (i.e., 28 joints in 80 patients) according to the following ranges of kappa (k) coeffi - were evaluated through physical examination. Com- cients: < 0.00 = poor, 0.00–0.20 = slight, 0.21–0.40 = fair, pared with tenderness, swelling assessment showed 0.41–0.60 = moderate, 0.61–0.80 = substantial, and 0.81– a slight strength of agreement with lower percentage 1.00 = almost perfect [18]. Linear correlation coefficients of overall agreement and concordance rates (66% and (Pearson correlation coefficient) were also calculated for Fleiss kappa = 0.227; 95% CI 0.203–0.250, p < 0.0001 vs tender and swollen joint counts. 81.7% and Fleiss kappa = 0.449; 95% CI 0.403–0.493, p < 0.0001) (see Table 2). Results The linear correlation coefficient (Spearman’s corre - Patient characteristics lation coefficient) was calculated, taking all the possi - The demographic and clinical characteristics of the 80 ble pairs (observer A–B, observer A–C, observer B–C), selected patients are illustrated in Table 1. The majority thus having an overall correlation coefficient of 0.81 and 0.431 for swollen and tender joints, respectively. Quintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 4 of 7 Table 2 Interobserver agreement and concordance for physical examination (PE) tenderness and swelling assessment Joint Inter-observer agreement and concordance PE tenderness assessment PE swelling assessment Fleiss kappa 95% CI p value Count Fleiss kappa 95% CI p value Count Right shoulder 0.521 0.494–0.652 < 0.0001 83 0.249 0.124–0.322 0.0001 27 Left shoulder 0.446 0.198–0.546 < 0.0001 64 0.134 0.092–0.184 0.018 23 Right elbow 0.4 0.308–0.537 < 0.0001 71 0.146 0.092–0.287 0.012 69 Left elbow 0.403 0.258–0.446 < 0.0001 64 0.245 0.166–0.392 0.0001 79 Right wrist 0.392 0.263–0.548 < 0.0001 78 0.121 0.007–0.168 0.030 123 Left wrist 0.325 0.261–0.419 < 0.0001 72 0.042 − 0.192 to 0.234 0.259 145 Right MCP1 0.394 0.326–0.551 < 0.0001 50 − 0.166 − 0.178 to 0.151 0.995 134 Right MCP2 0.298 0.225–0.362 < 0.0001 53 0.13 0.109–0.140 0.023 158 Right MCP3 0.282 0.153–0.323 < 0.0001 49 0.218 0.070–0.292 0.0004 109 Right MCP4 0.297 0.235–0.379 < 0.0001 33 0.397 0.376–0.476 < 0.0001 53 Right MCP5 0.379 0.236–0.477 < 0.0001 36 0.306 0.272–0.387 < 0.0001 63 Left MCP1 0.304 0.187–0.394 < 0.0001 44 0.094 0.024–0.148 0.073 95 Left MCP2 0.444 0.391–0.619 < 0.0001 36 0.262 0.133–0.314 < 0.0001 110 Left MCP3 0.382 0.323–0.453 < 0.0001 41 0.27 0.268–0.334 < 0.0001 69 Left MCP4 0.315 0.086–0.375 < 0.0001 32 0.177 0.059–0.232 0.003 45 Left MCP5 0.353 0.183–0.569 < 0.0001 26 0.217 0.172–0.266 0.0004 71 Right PIP1 0.299 0.050–0.367 < 0.0001 22 0.128 0.087–0.227 0.023 53 Right PIP2 0.353 0.310–0.379 < 0.0001 28 0.042 0.015–0.094 0.258 95 Right PIP3 0.351 0.312–0.486 < 0.0001 30 − 0.017 − 0.102 to 0.113 0.603 118 Right PIP4 0.232 0.111–0.250 < 0.0001 28 0.074 0.037–0.117 0.126 82 Right PIP5 0.396 0.294–0.431 < 0.0001 26 0.245 0.091–0.318 0.244 48 Left PIP1 0.165 0.018–0.304 < 0.0001 21 0.02 − 0.203 to 0.242 0.376 39 Left PIP2 0.167 0.085–0.299 < 0.0001 24 − 0.061 − 0.236 to 0.2 0.827 50 Left PIP3 0.298 0.196–0.368 < 0.0001 35 0.077 − 0.071 to 0.219 0.116 101 Left PIP4 0.185 0.018–0.267 < 0.0001 31 0.121 0.056–0.197 0.030 74 Left PIP5 0.339 0.270–0.573 < 0.0001 18 0.113 0.041–0.267 0.040 46 Right knee 0.476 0.318–0.530 < 0.0001 48 0.428 0.295–0.525 < 0.0001 58 Left knee 0.379 0.350–0.471 < 0.0001 57 0.214 0.038–0.362 0.0005 62 Overall agreement (%) 0.449 (81.7%) 0.403–0.493 < 0.0001 0.227 (66%) 0.203–0.250 < 0.0001 MCP, metacarpophalangeal; PIP, proximal interphalangeal Agreement and concordance between physical and US Table 3 Agreement and concordance between physical and US examination examination swelling assessment What is interesting in Table  3 is that, from the then Joint Overall Fleiss kappa p value joints assessed by both methods, there has been a agreement (%) slight relative increase in the number of swollen joints Right wrist 62.5 0.241 0.007 through US examination (345 of 800 joints assessed Left wrist 63.7 0.213 0.022 by US (43%)), when compared to physical examination Right MCP 2 60.0 0.195 0.040 (946 of 2400 joints assessed by our three rheumatolo- Right MCP 3 66.2 0.264 0.009 gist (39%)). Concordance strength of agreement was fair Right MCP 4 69.6 0.197 0.039 and ranged from Fleiss kappa 0.168–0.264. Likewise, Right MCP 5 70.9 0.244 0.014 the k coefficients between physical and US examina - Left MCP 2 62.5 0.244 0.011 tion in the left MCP5 joint (k = 0.168, p 0.06) followed Left MCP 3 67.5 0.228 0.019 by left MCP4 (k = 0.177, p 0.053) joint stand out as Left MCP 4 77.5 0.177 0.053 the lowest values. By contrast, k coefficient in wrist Left MCP 5 71.2 0.168 0.06 was slightly higher (k = 0.241 (p 0.007) and k = 0.213 MCP, metacarpophalangeal; PIP, proximal interphalangeal (p 0.023)). The percentage of overall agreement and Q uintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 5 of 7 are similar to ours, showing a more robust correlation concordance rate was 67.18% and k = 0.210; (p 0.027), regarding tender joints compared to that of swollen respectively. joints. When compared to MUSC, as previously stated, agree- Discussion ment and concordance were slightly lower for swelling As explained in the introduction, it is clear that training assessment. This discrepancy could be attributed to the sessions focused on the standardization of joint assess- mean age of our patients (55 years); it seems possible that ment techniques play a pivotal role in understanding the older the patient, the more frequent synovial thick- the reliability of quantitative indexes daily used in RA ening and incipient joint deformities, thus being con- patients. We aimed to evaluate the correlation and agree- founding for not only MSUS but also for PE assessment. ment between rheumatologists after such a training Moreover, it is important to consider what has been session and compare its performance with a reference widely held by recent reports, that is, the trend towards imaging modality, namely, MSUS, to find a reasonable higher agreement and concordance rates regarding swol- approach on behalf of the inevitable high intra- and inter- len joints count [22, 23]. These data must be interpreted observer variability described in the literature. with caution because of the use of the OMERACT syno- Contrary to expectations, despite the fact that all three vitis definition and its implications of considering even participating rheumatologists attended the proposed a minimal amount of intraarticular tissue as an abnor- training sessions, interobserver variability among them mal finding of synovial hypertrophy, thereby involving a was still present. The wide range of perceived concord - potential overestimation. ance rates suggests that the assessment of some individ- It could be argued that the average age and the duration ual joints may be particularly challenging on their own, and activity of the disease could contribute to the differ - thereby showing specific difficulties during standardiza - ences between physical examination and US; neverthe- tion, with possible attribution to an underlying long and less, those patients with joint deformity were excluded. It comprehensive learning process. is relevant to bear in mind that, in those populations with In terms of overall interobserver agreement and con- a poorly controlled- and a longer- disease, joint assess- cordance rates, as well as of overall correlation, our ment either by physical examination or by ultrasound findings propose that tenderness assessment was more remains a challenge when considering joint surface irreg- homogenous than swelling assessment. A possible expla- ularities. It supports the pivotal role of conventional radi- nation for these results may be the fact that joint tender- ography as the first choice for the evaluation of structural ness is inferred solely by the patient’s response, while that changes such as erosions. of joint swelling by physicians’ technique and interpreta- The limitations of this study include firstly the absence tion of findings. of power Doppler ultrasonographic assessment due to The importance of training sessions focused on stand - timing issues on behalf of the vast number of evaluated ardization was first stated by Scott et  al. in 1996 [19], joints; prior studies have noted the role of power Doppler whose findings suggested a considerably increased sensi - US in detecting subclinical and residual synovitis when tivity of measurement for both tender and swollen joints assessing synovial vascularity, although, it was beyond and a reduction in the mean coefficient of variation for the scope of our work to evaluate such conditions. Sec- the number of swollen joints (82% vs 59%) after a 60-min ondly, we did not count on another radiologist with suf- training session based on the EULAR handbook for joint ficient experience in MSUS to be considered as a second evaluation. Unlike Scott et  al., on a multicentric cohort evaluator; additionally, the US evaluation performed by a study evaluating standardization based on the afore- radiologist and not an articular US rheumatologist might mentioned EULAR handbook, Grunke et  al. [20] stated be considered as a source of uncertainty; nonetheless, that even when consistency and variability significantly the vast experience, as well as the specific MSUS train - improved, the mean number of tender, as well as swol- ing of our radiologist, support the internal validity of the len joints decreased. Nonetheless, a reference imaging assessments. Very little was found in the literature on the modality was not used. question of performance differences between those sce - In addition, clinical experience plays an essential part narios (radiologist vs. rheumatologist), remaining as an as recently proposed by El-Hadidi et  al. [21] where after intriguing issue for future research, especially for low and a consensus on joint assessment, the interobserver agree- middle-income countries where training for such medi- ment was calculated to compare an experienced rheuma- cal sub-specialties is not as frequent. tology professor (30  years of experience) with a young Thirdly, the absence of early RA patients, as well as of Rheumatology fellow (3  years of training). Although a patients in clinical remission, and finally, that we did not high correlation between professor and candidate was report pretraining session joint count results in order to described, specific results on joint assessment correlation Quintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 6 of 7 Competing interests propose a variability improvement. The latter considera - This research did not receive any specific grant from funding agencies in the tion must be handled with care when considering that public, commercial, or not‑for ‑profit sectors. We agree to bear the applicable this decision was based on the fact that pretraining ses- publication charges if their manuscript is accepted for publication. All authors declare that they have no competing interest. sion correlation and agreement scores were certainly variable, close to the null value, and even negative. Tak- Author details ing this into consideration, the gain of the stated train- Reumavance Group, Rheumatology Section, Department of Internal Medicine, Fundación Santa Fe de Bogotá University Hospital, Carrera 7 No. ing session was to revert such tendency and yet turn it to 117‑15, 110111 Bogota, Colombia. Department of Internal Medicine, School positive values, even though the kappa coefficients were of Medicine, Universidad Nacional de Colombia, Bogota, Colombia. Depar t‑ in the range of slight to moderate. It is worth mentioning ment of Internal Medicine, School of Medicine, Universidad de Los Andes, Bogota, Colombia. Department of Radiology, School of Medicine, Universidad that when adding more than two observers to the calcu- Nacional de Colombia, Bogota, Colombia. lations, the slightest discrepancy substantially lowers the statistical estimator; thus, an adjusted threshold must be Received: 3 June 2021 Accepted: 5 November 2021 taken when interpreting such results. Finally, a reason- able approach could be to hypothesize that the increase in the total swollen joint count through US examination References would have a substantial implication in the categorization 1. McInnes IB, Schett G. The pathogenesis of rheumatoid arthritis. N Engl J of the activity of the disease; nevertheless, even when it Med. 2011;365:2205–19. was beyond our scope, we trust that future research on 2. Backhaus M, Kamradt T, Sandrock D, Loreck D, Fritz J, Wolf KJ, et al. Arthritis of the finger joints: a comprehensive approach comparing con‑ this behalf will rise as a natural progression of this work. ventional radiography, scintigraphy, ultrasound, and contrast‑ enhanced magnetic resonance imaging. Arthritis Rheum. 1999;42:1232–45. Conclusions 3. Naredo E, Bonilla G, Gamero F, Uson J, Carmona L, Laffon A. Assessment of inflammatory activity in rheumatoid arthritis: a comparative study of Taken together, our findings indicate that joint tender - clinical evaluation with grey scale and power Doppler ultrasonography. ness assessment had a better interobserver agreement Ann Rheum Dis. 2005;64:375–81. and concordance rate, moreover a higher overall corre- 4. Van der Heijde D, Van’t Hof M, Van Riel P, et al. Development of a disease activity score based on judgment in clinical practice by rheumatologists. lation coefficient than joint swelling. The evidence from J Rheumatol. 1993;20:579–81. our results suggests a fair overall concordance rate when 5. Prevoo MLL, Van’t Hof MA, Kuper HH, et al. Modified disease activity comparing the US examination to the physical examina- scores that include twenty‑ eight‑joint counts. Development and valida‑ tion in a prospective longitudinal study of patients with rheumatoid tion, thus supporting the need for the implementation of arthritis. Arthritis Rheum. 1995;38(1):44–8. training sessions dedicated to standardization in rheuma- 6. Porter D, Gadsby K, Thompson P, White J, McClinton C, Oliver S. DAS28 tology clinics to guarantee reliability in identifying tender and rheumatoid arthritis: the need for standardization. Musculoskelet Care. 2011;9(4):222–7. and swollen joints within rheumatologists. 7. Scott IC, Scott DL. Joint counts in inflammatory arthritis. Clin Exp Rheu‑ matol. 2014;32(Suppl. 85):S7–12. Acknowledgements 8. Sokka T, Pincus T. Joint counts to assess rheumatoid arthritis for clinical Not applicable. research and usual clinical care: advantages and limitations. Rheum Dis Clin N Am. 2009;35(4):713–22. Authors’ contributions 9. Smolen JS, Landewé RBM, Bijlsma JWJ, et al. EULAR recommendations for GQ‑L: Conceptualization, methodology, formal analysis, investigation, writing‑ the management of rheumatoid arthritis with synthetic and biological original draft preparation, writing‑review and editing. KM ‑ C: Writing‑ original disease‑modifying antirheumatic drugs: 2019 update. Ann Rheumatic draft preparation, writing‑review and editing. JBF‑S: writing‑ original draft Dis. 2020;79:685–99. preparation, writing‑review and editing. PM ‑P: Conceptualization, methodol‑ 10. Aletaha D, Neogi T, Silman AJ, Funovits J, Felson DT, Bingham CO, et al. ogy, investigation. PC‑A: Conceptualization, methodology, investigation. EC: 2010 Rheumatoid arthritis classification criteria: an American College Investigation. All authors read and approved the final manuscript. of Rheumatology‑European League Against Rheumatism collaborative initiative. Ann Rheum Dis. 2010;69:1580–8. Funding 11. Wakefield RJ, Balint PV, Szkudlarek M, Filippucci E, Backhaus M, D’Agostino This research did not receive any specific grant from funding agencies in the MA, et al. OMERACT 7 special interest group. Musculoskeletal ultrasound public, commercial, or not‑for ‑profit sectors. including definitions for ultrasonography pathology. J Rheumatol. 2005;32:2485–7. Availability of data and materials 12. Szkudlarek M, Wakefield R, Backhaus M, Terslev L. The discriminatory The data underlying this article will be shared on reasonable request to the capacity of ultrasound in rheumatoid arthritis: active vs inactive, early vs corresponding author. advanced, and more. Rheumatology. 2012;51(suppl 7):vii6–9. 13. Wakefield RJ, Green MJ, Marzo ‑ Ortega H, Conaghan PG, Gibbon WW, Declarations McGonagle D, et al. Should oligoarthritis be reclassified? Ultrasound reveals a high prevalence of subclinical disease. Ann Rheum Dis. Ethics approval and consent to participate 2004;63:382–5. This study was conducted with the approval of our Local Research Ethics 14. Szkudlarek M, Court‑Payen M, Jacobsen S, Klarlund M, Thomsen H, Øster ‑ Committee in accordance with the Declaration of Helsinki. Every patient gave gaard M. Interobserver agreement in ultrasonography of the finger and written and spoken informed consent. toe joints in rheumatoid arthritis. Arthritis Rheum. 2003;48(4):955–62. 15. Bruyn GA, Iagnocco A, Naredo E, Balint PV, Gutierrez M, Hammer HB, Consent for publication et al. OMERACT definitions for ultrasonographic pathologies and All authors approved the final version of this manuscript to be published. Q uintana‑López et al. Advances in Rheumatology (2021) 61:68 Page 7 of 7 elementary lesions of rheumatic disorders 15 years on. J Rheumatol. 21. El‑Hadidi K, Gamal S, Saad S. Inter ‑ observer agreement of standard joint 2019;46(10):1388–93. https:// doi. org/ 10. 3899/ jrheum. 181095. count examination and disease global assessment in a cohort of Egyp‑ 16. Terslev L, Naredo E, Aegerter P, et al. Scoring ultrasound synovitis in rheu‑ tian Rheumatoid Arthritis patients. Reumatol Clín. 2019;15(5):258–63. matoid arthritis: a EULAR‑ OMERACT ultrasound taskforce‑Part 2: reliability 22. Hirata A, Ogura T, Hayashi N, Takenaka S, Ito H, Mizushina K, et al. and application to multiple joints of a standardised consensus‑based Concordance of patient‑reported joint symptoms, physician‑ examined scoring system. RMD Open. 2017;3:e000427. https:// doi. org/ 10. 1136/ arthritic signs, and ultrasound‑ detected synovitis in rheumatoid arthritis. rmdop en‑ 2016‑ 000427. Arthritis Care Res. 2017;69(6):801–6. 17. Schmidt WA, Schmidt H, Schicke B, Gromnica‑Ihle E. Standard refer ‑ 23. Salaffi F, Filippucci E, Carotti M, Naredo E, Meenagh G, Ciapetti A, et al. ence values for musculoskeletal ultrasonography. Ann Rheum Dis. Inter‑ observer agreement of standard joint counts in early rheumatoid 2004;63:988–94. arthritis: a comparison with grey scale ultrasonography: a preliminary 18. Landis JR, Koch GG. The measurement of observer agreement for cat‑ study. Rheumatology. 2008;47:54–8. egorical data. Biometrics. 1977;33:159–74. 19. Scott DL, Choy EHS, Greeves A, Isenberg DM, Kassinor D, Rankin E, et al. Publisher’s Note Standardizing joint assessment in rheumatoid arthritis. Clin Rheumatol. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 1996;15(6):579–82. lished maps and institutional affiliations. 20. Grunke M, Antoni C, Kavanaugh A, Hildebrand V, Dechant C, Schett G, et al. Standardization of joint examination technique leads to a signifi‑ cant decrease in variability among different examiners. J Rheumatol. 2010;37(4):860–4. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Advances in Rheumatology – Springer Journals

Published: Nov 18, 2021

Keywords: Physical examination; Rheumatoid arthritis; Synovitis; Ultrasonography

There are no references for this article.