To evaluate the accuracy of ultrasonography (US) in the diagnosis of carpal tunnel syndrome (CTS) in patients with a bifid median nerve on the basis of cross-sectional area (CSA) measurements of the median nerve at the level of the carpal tunnel (CSAc), with additional measurements obtained more proximally (CSAp) at the level of the pronator quadratus muscle. This HIPAA-compliant study was approved by the local institutional review board; informed oral and written consent were obtained. Fifty-three wrists in 49 consecutive patients with a bifid median nerve and CTS symptoms and 28 wrists in 27 healthy volunteers with a bifid median nerve were examined by using US. Two independent US examiners who were blinded to prior test results measured median nerve CSA at two levels, CSAc and CSAp. The difference between CSAc and CSAp (ΔCSA) was calculated for each wrist. Receiver operating characteristic (ROC) analysis was performed. The study population included 17 men and 32 women (mean age, 55.1 years; age range, 24-78 years). The control population included 13 men and 14 women (mean age, 52.6 years; age range, 24-86 years). Mean CSAc was approximately 5 mm(2) greater in patients with CTS than in healthy volunteers (P < .0001), while mean ΔCSA was 5.8-5.9 mm(2) greater in patients with CTS (P < .0001). A CSAc threshold of 12 mm(2) provided sensitivity and specificity of 84.9% and 46.5%, respectively, while a ΔCSA threshold of 4 mm(2) provided sensitivity and specificity of 92.5% and 94.6%, respectively. ROC analysis demonstrated a significant advantage of ΔCSA (area under ROC curve [A(z)] = 0.95-0.96) compared with CSAc (A(z) = 0.84-0.85) for the diagnosis of CTS (P < .003). The use of a ΔCSA parameter improves the diagnostic accuracy of US for the presence of CTS in patients with a bifid median nerve.