Background/Purpose: Carpal tunnel syndrome (CTS) is the most common nerve compression syndrome and a common extra-articular manifestation of rheumatoid arthritis (RA). Different causes of CTS are known, among them inflammatory and non-inflammatory pathologies. Electroneurography (ENG) of the median nerve, the method of choice to diagnose CTS, measures impairment of nerve conduction velocity without explaining its underlying cause. However, because the electrical stimulation is often not well tolerated, ENG results may come out inconclusive. Using greyscale ultrasonography (GS-US) provides anatomic information including a structural representation of the carpal tunnel. The aim of this study is to investigate the performance of nerve GS-US in the diagnosis of CTS in patients with RA.

Methods: Consecutive patients with active RA under suspicion of CTS presenting to a large rheumatologic center were included. Both hands were examined by an experienced neurologist including ENG and a GS-US (ML linear probe with 6-15 Hz) of the median nerve. An established grading system for ENG (1), and an established system for GS-US based on cut-offs for the nerve cross sectional area (CSA) [mild: 0,11-0,13cm², moderate: 0,14-0,15 cm², severe: > 0,15 cm² CTS (2)] were used. In addition, the Boston Carpal Tunnel Syndrome Questionnaire (BCTSQ) was used to assess CTS symptoms (3).

Results: Both hands of 58 patients with active RA (n=116) and clinical suspicion of CTS (in 38 cases bilaterally) were included. After clinical examination, CTS was suspicious in 96 hands (82.8%), and 59 of all hands had a final diagnosis of CTS (50.9%). Of the latter, 43 hands (72.9%) had a positive ENG and 16 (27.1%) a positive GS-US finding only, while 30 hands (50.8%) were positive in both examinations.

There was a good correlation of the cross-sectional area (CSA) as well as the CSA-ratio to the ENG findings: the larger the CSA, the more severe was the CTS as assessed by ENG (Spearman’s rho=0.554; p< 0.001). The more severe the GS-US findings of CTS were, the more definite were the distal motor latency (Spearman's rho=0.554; p< 0.001) and sensible nerve conduction velocity of the median nerve (Spearman's rho=-0.5411; p< 0.001).

In the 46 hands positive in GS-US, tenosynovial hypertrophy of the flexor tendons was detected in 19 hands (41.3%), 7 of which (36.8%) also showed an additional cystic mass. In these 19 patients, clinical complains were more severely present than in patients with non-inflammatory CTS, as assessed by the BCTSQ with a total score of 68.8±13.4 vs. 59.3±13.7, respectively (p=0.007).

Conclusion: In patients with active RA and clinical complains of CTS, ultrasound examinations provide additional information about inflammation which is helpful for a diagnosis of CTS. Thus, ENG and nerve GS-US should be used complementary for a diagnostic workup of CTS in RA patients with a suspicion of CTS. Power-Doppler may further improve the diagnostic performance of GS-US.

1. Padua L et al. Acta Neurol Scand 1997; 96:211–217

2. El Miedany et al., Rheumatology (Oxford). 2004 Jul; 43(7):887-895

3. Levine DW et al. J Bone Joint Surg Am 1993; 75: 1585-1592

Figure: BCTSQ scores in patients with diagnosis of CTS and absence or presence of RA-related tenosynovial hypertrophy

ACR Meeting Abstracts - https://acrabstracts.org/abstract/ultrasonography-of-the-median-nerve-in-patients-with-rheumatoid-arthritis-under-suspicion-of-carpal-tunnel-syndrome/