This study included all children (age 2 to 18 years) diagnosed as CD as per the modified ESPGHAN criteria [2] from outpatient clinic or impatient services in SMS Medical College, Jaipur from May 2014 to April 2016.

We divided these children into two groups: group 1 constituted children newly diagnosed as CD and were on gluten-containing diet whereas group 2 constituted those already diagnosed with CD, and on GFD for more than 6 months. Children who had any connective tissue disorder, concomitant rheumatic fever or history of injury to joints in last two months were excluded from the study. Informed consent was obtained from parents of the enrolled children. The study was approved by the ethical committee of the institution.

All patients were evaluated by complete medical history, including history of musculoskeletal symptoms and other co-morbidities; and detailed physical exami-nation, including musculoskeletal assessment. Ultrasonographic assessment of bilateral hip, knee and ankle joints was performed, using a Hitachii Hivision Preirus machine, equipped with a 9–14 MHz linear array transducer, operating at 14 MHz frequency (gain 50%). Ultra-sonography evaluation of joints was performed by a single radiologist experienced in pediatric musculo-skeletal Ultrasonography who was blinded to patient’s laboratory and clinical features. The radiologist looked for joint effusion, synovial hypertrophy or power Doppler signal (evidences of synovitis) in the above mentioned joints.

Statistical analysis: 　Chi-square test was used to compare frequencies between two groups. P<0.05 was considered statistically significant.

A total of 120 children (mean (SD) age 6.7 (3.9)y; 63 males) were enrolled in the study. Group1 and Group 2 had 60 children each. No significant difference was observed between the two groups regarding the demographic profile and clinical presentation of CD (Table I).

TABLE I  Baseline Characteristics of Study Participants 

Ultrasonographic assessment showed presence of at least one abnormality in 19 (31.7%) CD patients in group 1 whereas only 2 (3.3%) of group 2 patients had USG abnormalities (P< 0.001).

The most frequently involved joint was the knee where 11 children of group 1 had USG abnormalities. (Table II). The finding in knee joint in group1 were: joint effusion in 7, synovial hypertrophy in 2, and joint effusion along with synovitis in 2 children. In group 2, both patient had joint effusion. Findings at hip joint included joint effusion in 3, synovial hypertrophy in one, and combined hypertrophy and effusion in one child. Two children in group 1 had evidence of joint effusion in ankle, and only one patient had multiple joint (hip and knee) involvement. Joint effusion was mild in all the cases.

TABLE II	Joint Ulatrasonography Findings in Children with Celiac Disease
	      

Majority of the patients with ultrasonographic evidence of joint abnormalities were asymptomatic suggesting subclinical synovitis. Clinically only 3 (5%) patients of group 1 had joint pain and limitation of joint movement at presentation while none of group 2 patients had similar symptoms. The mean (SD) age of cases having arthralgia at the time of presentation was significantly higher than asymptomatic cases [11.7 (7.5) vs 6.0 (3.3) years; P=0.01].

In this study, ultrasonographic evidence of joint abnormalities were present in approximately one-third of children newly diagnosed with CD. This frequency was significantly higher than in children with CD on GFD for more than six months, suggesting that GFD may lead to improvement in joint abnormalities associated with CD. In our study joint effusion was the most common abnormality followed by synovial hypertrophy. Knee joint was the most frequently involved joint followed by hip and ankle.

Small sample size, lack of inclusion of healthy controls and cross-sectional nature of the study were few limitations of this study.

Adelizzi, et al. [12] first described the association of arthritis and CD following which there were many similar case reports [5,6]. Both arthritis and small bowel mucosal changes have been reported to improve with a gluten-free diet [13,14]. Lubrano, et al. [15] studied 200 adult patients with CD, and found that arthritis was present in 26% of patients. Lagnocco, et al. [8] in 2014, for the first time studied children with CD for joint involvement using ultrasonography. They found that 50% of their newly diagnosed CD children had abnormalities. The pattern of joint involvement was similar to our study.

Subclinical joint effusion seems to be a relatively frequent finding seen in newly-diagnosed CD patients. Ultrasonography should be considered as a useful imaging tool for identifying CD patients with joint changes that have not yet manifested clinically. Early diagnosis and treatment may prevent subclinical synovitis to manifest clinically in later life. These findings also suggest that GFD has the potential to improve joint manifestations seen in CD.

Contributors: KG: concept and design of study, drafted the manuscript; PA: collected and analysed the data; RKG: analysis of data and literature search; SS: intellectual input in final drafting and overall supervision.

Funding: None; Competing interest: None stated.