Ultrasonographic reference sizes of the median and ulnar nerves and the cervical nerve roots in healthy Japanese adults.

The objective of this study was to identify, for practical use, ultrasonographic reference values for nerve sizes at multiple sites, including entrapment and non-entrapment sites along the median and ulnar nerves and among the cervical nerve roots. We verified reliable sites and site-based differences between the reference values. In addition, we found associations between the reference nerve sizes and several physical characteristics (gender, dominant hand, age, height, weight, body mass index [BMI] and wrist circumference). Nerves were measured bilaterally at 26 sites or levels in 60 healthy Japanese adults (29 males; age, 35.4 ± 9.7 y; BMI, 22.3 ± 3.6 kg/m(2); wrist circumference, 16.0 ± 1.3 cm on the right side and 15.9 ± 1.2 cm on the left side). The mean reference nerve sizes were 5.6-9.1 mm(2) along the median nerve, 4.1-6.7 mm(2) along the ulnar nerve and 2.14-3.39 mm among the cervical nerve roots. Multifactorial regression analyses revealed that the physical characteristics most strongly associated with nerve size were age, BMI and wrist circumference at the entrapment sites (F = 7.6, p < 0.01, at the pisiform bone level of the carpal tunnel; F = 15.1, p < 0.001, at the level of Guyon's canal), as well as wrist circumference and gender at the non-entrapment sites (F = 70.6, p < 0.001, along the median nerve; F = 24.7, p < 0.001, along the ulnar nerve). Our results suggest that the factors with the greatest influence on nerve size differed between entrapment and non-entrapment sites. Site-based differences in nerve size were determined using one-way analyses of variance (p < 0.001). Intra- and inter-observer reliability was highest for the median nerve, at both the distal wrist crease and mid-humerus; at the arterial split along the ulnar nerve; and at the fifth cervical nerve root level. No systematic error was indicated by Bland-Altman analysis; the coefficients of variation were 5.5%-9.2% for intra-observer reliability and 7.1%-8.7% for inter-observer reliability.