OBJECTIVES

To assess the value of adding the ultradistal level to other more thoroughly studied levels of the carpal tunnel when measuring the cross-sectional area (CSA) of the median nerve (MN) by ultrasound (US) in diagnosing patients with primary carpal tunnel syndrome (CTS).

METHODS

Patients clinically diagnosed with primary CTS and healthy controls were included. The MN-CSA was measured by US at three wrist levels: proximal, distal and ultradistal. The best cut-off to differentiate cases and controls was determined for the CSA and for the difference between levels of the same wrist. The performance of different definitions for US-CTS compared with the clinical diagnosis of CTS was evaluated: (1) CSA above cut-off at each level; (2) CSA-difference above cut-off at each level; (3) ≥1 level with CSA above cut-off and (4) ≥1 CSA-difference above cut-off. Definition 3, excluding the ultradistal level, and combinations of definitions were also tested.

RESULTS

In total, 219 patients and 39 controls were included. The CSA was higher in patients (10.5-16.8 mm) than controls (6.2-7.6 mm). The difference between groups was maximal at the ultradistal level (right: 10.1 mm; left: 8.3 mm). The CSA cut-offs were 11 mm, 9 mm and 10 mm at the right, and 10 mm, 8 mm and 10 mm at the left, for the proximal, distal and ultradistal levels, respectively. Definition 3 yielded the best balance between sensitivity (98%) and specificity (95%) (right hand). Removing the ultradistal level from definition 3 decreased sensitivity to 90%, maintaining the same specificity.

CONCLUSIONS

Adding the ultradistal level improves the performance of US for diagnosing CTS. We suggest adding it in clinical practice when investigating CTS.