The OMERACT Ultrasound Group: A Report from the OMERACT 2016 Meeting and Perspectives

Further development in RA

The US WG group has validated US in RA by defining elementary lesions such as tenosynovitis and synovitis, including its components^3,4, and developed and validated a consensus-based scoring system for synovitis and tenosynovitis for clinical trials^5,6,7,8,9,10 with a good sensitivity to change¹¹. However, US elementary lesions such as synovial effusion (SE) and synovial hypertrophy (SH; with or without Doppler activity as a measurable sign of inflammatory activity) may be seen in healthy controls (HC)^12,13. Therefore a threshold is needed for separating normal findings from pathology, thereby creating a definition for what may be perceived as minimal disease criteria. This will influence the definition of early disease onset, the optimal response to therapy, and the identification of US remission. The subgroup working on defining minimal disease criteria for RA presented data comparing US findings obtained in HC with US findings obtained from an early arthritis (EA) cohort with the aim of identifying which US elementary components are present in the joints of HC, their prevalence, and their preferred location, and how these differ from patients with EA. This comparison represents the first step in identifying the US findings clinically important in EA and to describe at joint level which US lesions should be considered pathological.

In the conducted study, US findings of SH, SE, and power Doppler (PD) in joints of HC and patients with EA were systematically documented, and the distribution and grading of the findings at joint level were compared to reassess the threshold of abnormality in regions of overlap.

The participants were recruited from 2 centers: the healthy subjects from Ambroise Pare Hospital, Boulogne-Billancourt, France, and the patients with EA from City Hospital and Queen Elizabeth Hospital, Birmingham, UK. The healthy subjects were excluded if they had previous or present signs of joint disease¹³. Patients with EA who were naive to disease-modifying antirheumatic drugs (DMARD) were included if they had ≥ 1 clinically swollen joint due to inflammatory arthritis as judged by a physician and symptom duration ≤ 3 months. DMARD-naive patients with EA were excluded if joint symptoms were solely attributed to osteoarthritis (OA). US was performed on the wrist and metacarpophalangeal 1–5 joints (MCP), proximal interphalangeal 1–5 joints (PIP), and metatarsophalangeal 2–5 joints (MTP), and SH and PD were scored semiquantitatively (0–3) using the consensus-based European League Against Rheumatism-OMERACT scoring system⁹; SE was scored as a binary variable.

The demographics results of the 2 cohorts are shown in Table 1.

In HC, 7% had at least 1 joint abnormality versus 30% in the EA group. In HC, the abnormalities were seen in the wrist, MCP 1–4, and MTP 2–4, whereas SH and PD could be seen in all types of examined joints in the EA cohort, but SE was rare. In HC, the findings of SH were predominantly grade 1, although grade 2 and grade 3 could also rarely be found. In the EA cohort, all grades were seen, with grade 3 being most frequent in the MCP. For PD, only grade 1 was seen in HC, whereas grades 1–3 were seen in the EA cohort.

The next step is to determine a cutoff level and to study a group of healthy elderly participants, because the grade of synovial hypertrophy-like changes appears to be age-related¹⁴.

Validating US in pediatrics

The increasing need for the use of US in pediatric rheumatology led to the formation of the pediatric subgroup. To test validity and to improve the applicability of US in JIA, in 2011–2012 the subgroup investigated the use of US among pediatric rheumatologists¹⁵ and performed a systematic literature review¹⁶, which highlighted face and content validity of US for detecting synovitis in JIA with higher sensitivity than clinical examination.

Then the subgroup defined and validated the components of healthy pediatric joints in a multistep consensus process involving a panel of international experts on musculoskeletal ultrasound (MSUS) in children.

In the first step, a group of experts joined a Web-based consensus process to develop definitions for the following components of the pediatric joint: hyaline cartilage, secondary ossification center, joint capsule, synovial membrane, and cortical ossified bone. A definition reaching ≥ 80% of agreement on a Likert scale from 1–5 was accepted¹⁷.

In the second step, in a face-to-face meeting, a subgroup of these experts produced additional definitions for 2 joint components that were considered relevant in physiological vascularization of a pediatric joint, i.e., fat pad and physis, which were not described in detail in the previous definitions of pediatric joint components¹⁸.

In the third step, the applicability of the previous and new definitions was tested in a live exercise involving healthy children. Following standardized image acquisition and machine-setting protocols, 4 joints (i.e., wrist, second MCP, knee, and ankle) were examined in 4 different age groups (toddler and preschool ages 2–4 yrs, young children ages 5–8 yrs, preadolescent ages 9–12 yrs, and teenager ages 13–16 yrs). Using κ statistics, the intraobserver agreement for the applicability of all definitions ranged from 0.44–1, and for the interobserver agreement ranged from 0.33–1, with highest agreement for the wrist and lowest for MCP 2. Thereafter, the 2 new definitions from the second step were agreed on through a Delphi process among a wider group of pediatric MSUS experts (manuscript in preparation).

Further, the group defined age-related findings, i.e., physiological vascularization and ossification grade in healthy children. In a live exercise with healthy children, it was shown that physiological vascularization can be detected with up to 3 solitary PD signals within the normal joint owing to physiological vascularization localized predominantly in the fat pad, the epiphysis, the physis, and the short bone cartilage. Additionally, the group developed an ossification grade definition in children¹⁹.

Using the same methodology, the subgroup moved to define and validate the US elementary lesions in JIA and developed preliminary definitions for synovitis as presented in Table 2, in which synovitis can be based on B-mode findings alone, but not solely on the Doppler²⁰. Taking this synovitis definition and the previously investigated physiological vascularization into account, the group developed a semiquantitative scoring system (0–3) for synovitis for both B-mode and Doppler mode through a Delphi process and a face-to-face meeting (manuscript in preparation). The scoring system was tested in a JIA patient-based exercise conducted in 2016. Scoring of synovitis is applicable only if synovial hypertrophy has been detected on B-mode. Doppler scoring differs from the Doppler scoring applied in adults⁹, in which Doppler grade 2 and grade 3 are related to Doppler activity below or above 50% of the SH, respectively, whereas in children the cutoff between Doppler grade 2 and grade 3 is 30%.

The next step is to test the reliability of the scoring system and its sensitivity to change.

Current perspectives

At the OMERACT 2016 meeting, new data were presented from the vasculitis subgroup, the RA in the foot and ankle subgroup, the gout subgroup, and the calcium pyrophosphate disease (CPPD) subgroup. The vasculitis and CPPD subgroups had performed systematic literature reviews and identified proposed definitions for elementary lesion in giant cell arteritis and in CPPD. After circulating these in Delphi exercises, agreements for the definitions of elementary lesions in vasculitis and CPPD were obtained.

Considerable progress has been made in the already existing subgroups for OA in the hand and foot, cartilage damage in RA, synovial biopsies, bone erosions and vessel channels, dactylitis, and psoriatic arthritis. The US group has also moved to new areas in rheumatology such as Sjögren syndrome and scleroderma, and initiated validation procedures.