Carpal tunnel syndrome assessment with diffusion tensor imaging: Value of fractional anisotropy and apparent diffusion coefficient
- 336 Downloads
To quantitatively assess carpal tunnel syndrome (CTS) with DTI by evaluating two approaches to determine cut-off values.
In forty patients with CTS diagnosis confirmed by nerve conduction studies (NCs) and 14 healthy subjects (mean age 58.54 and 57.8 years), cross-sectional area (CSA), apparent diffusion coefficient (ADC) and fractional anisotropy (FA) at single and multiple levels with intraobserver agreement were evaluated.
Maximum and mean CSA and FA showed significant differences between healthy subjects and patients (12.85 mm2 vs. 28.18 mm2, p < 0.001, and 0.613 vs. 0.524, p=0.007, respectively) (10.12 mm2 vs. 19.9 mm2, p<0.001 and 0.617 vs. 0.54, p=0.003, respectively), but not maximum and mean ADC (p > 0.05). For cut-off values, mean and maximum CSA showed the same sensitivity and specificity (93.3 %). However, mean FA showed better sensitivity than maximum FA (82.6 % vs. 73.9 %), but lower specificity (66.7 % vs. 80 %), and significant correlation for maximum CSA, 97 % (p < 0.01), with good correlation for maximum ADC and FA, 84.5 % (p < 0.01) and 62 % (p=0.056), respectively.
CSA and FA showed significant differences between healthy subjects and patients. Single measurement at maximum CSA is suitable for FA determination.
• DTI showed that FA is stronger than ADC for CTS diagnosis.
• Single- and multiple-level approaches were compared to determine FA and ADC.
• Single-level evaluation at the thickest MN cross-sectional area is sufficient.
KeywordsMedian nerve Carpal tunnel syndrome Magnetic resonance imaging Functional magnetic resonance imaging Diffusion tensor imaging
Apparent diffusion coefficient
Carpal tunnel syndrome
Diffusion tensor imaging
Magnetic resonance imaging
Nerve conduction studies
Region of interest
Compliance with ethical standards
The scientific guarantor of this publication is Andrea S. Klauser.
Conflict of interest
The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.
The authors state that this work has not received any funding.
Statistics and biometry
One of the authors has significant statistical expertise (Dr. Christian Kremser).
Written informed consent was obtained from all participants in this study.
Institutional Review Board approval was obtained.
Study subjects or cohorts overlap
No study subjects or cohorts have been previously reported.
Diagnostic or prognostic study.
Performed at one institution
- 19.Team RDC. R (2011): A Language and Environment for Statistical Computing. R Foundation for Statistical Computing.Google Scholar