# Body surface area–adjusted median nerve cross-sectional area and multimodal ultrasound improve diagnosis of carpal tunnel syndrome

## TL;DR

This study shows that combining ultrasound techniques with body surface area adjustments improves the accuracy of diagnosing carpal tunnel syndrome.

## Contribution

The study introduces a BSA-adjusted Z-score for median nerve cross-sectional area and combines it with SMI and SWE for better CTS diagnosis.

## Key findings

- BSA-based Z-score improved specificity compared to fixed CSA cutoffs for CTS diagnosis.
- SMI and SWE parameters were independent predictors of CTS in multivariable analysis.
- The combined SMI + SWE model achieved the highest diagnostic accuracy with 83% sensitivity and 90% specificity.

## Abstract

To evaluate the diagnostic performance of high-frequency ultrasound combined with Superb Microvascular Imaging (SMI) and Shear Wave Elastography (SWE) for carpal tunnel syndrome (CTS), and to develop an individualized diagnostic approach using a body surface area (BSA)–adjusted median nerve CSA at the pisiform level.

This retrospective study included 47 wrists with carpal tunnel syndrome (CTS) and 94 control wrists. Median nerve cross-sectional area (CSA) was measured at four anatomical sites. Superb Microvascular Imaging (SMI) and Shear Wave Elastography (SWE) were used to assess intraneural vascularity and stiffness, respectively. A linear regression model was developed to estimate the expected CSA at the pisiform level based on body surface area (BSA), and a BSA-based Z-score was calculated accordingly. Receiver operating characteristic (ROC) analyses were performed to compare the diagnostic performance of (i) a fixed CSA cutoff at the pisiform level, (ii) the BSA-based Z-score, and (iii) a combined SMI + SWE logistic regression model.

Ultrasound parameters differed significantly between the CTS and control groups (P < 0.05). The BSA-based Z-score derived from the CSA at the pisiform level yielded an AUC of 0.924 (95% CI 0.879–0.969) and improved specificity (83%; 95% CI 0.738–0.899) compared with the fixed CSA cutoff (75%; 95% CI 0.644–0.829). In multivariable analysis, SMI- and SWE-derived parameters remained independent predictors of CTS (P < 0.001). The combined SMI + SWE logistic regression model demonstrated the best diagnostic performance (AUC 0.944; 95% CI 0.906–0.982), with 83% sensitivity (95% CI 0.692–0.924) and 90% specificity (95% CI 0.826–0.955).

High-frequency ultrasound combined with Superb Microvascular Imaging (SMI) and Shear Wave Elastography (SWE) enables accurate, noninvasive evaluation of CTS. A BSA-based CSA Z-score improves specificity in CSA-based diagnosis, and integrating SMI and SWE further enhances overall diagnostic performance.