Background  Conus medullaris (CM) refers to the enlarged structure at the end of the spinal cord, and its position is very important to the diagnosis of tethered spinal cord (TC). At present, there are many methods to locate CM during prenatal ultrasound examination, but these methods are not accurate and effective in practical application.

  Objective  To explore a new method to locate CM through prenatal ultrasound and evaluate its diagnostic performance in fetal TC.

  Methods  The data of fetuses undergoing prenatal ultrasound examinations with normal findings from October 2020 to October 2021 in the Department of Ultrasound Diagnosis of the First Medical Center of Chinese PLA General Hospital were retrospectively collected. On the transabdominal ultrasound image, the distance between the end of the CM and the most distal ossification center of the spine was measured by trace method and the measured value was recorded as the CD value of trace method. Linear regression analysis on the CD value of trace method and gestational age was performed to obtain the linear regression equation. The fetuses who were diagnosed as TC by MRI or pathology from March 2015 to December 2021 in our department were retrospectively enrolled as the case group; and the fetuses who were diagnosed with spinal lesions without TC by MRI or pathology from September 2014 to August 2021 were enrolled as control group. The diagnostic capacity of the CD value of trace method for fetal TC was explored.

  Results  We enrolled 1 172 normal fetus. The normal ranges of the CD value of trace method during each gestational age were evaluated. The linear regression analysis of the CD value of trace method and GA was performed to obtain the linear regression equation: the CD value of trace method =0.037GA-2.275 (R²=0.885, P＜0.001). The case group included 26 fetuses, and the control group included 23 fetuses. When the CD value of trace method was lower than the 5th percentile, the fetus was diagnosed as TC. The sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of this method were 100% (26/26), 82.61% (19/23), 86.67% (26/30), 100% (19/19), 91.84% (45/49), respectively. The correlation test was carried out on the CD value of trace method and the gold standard method, indicating that the results of the two methods were obviously correlated( P＜0.001). The superiority test of the two methods was conducted, and the difference was not significant (P=0.134), which meant the gold standard method had no significant advantage over the CD value of trace method.

  Conclusion  The CD value of trace method can illustrate the location of the CM accurately and intuitively, which helps to quickly determine whether there is a low position of the CM during the prenatal ultrasound examination, and is of great significance to the detection of the fetal TC.