基于产前3.0 T TrueFISP MRI胎儿股骨生长动力学特征的观察

Observation of growth dynamic characteristics of fetal femur based on prenatal 3.0 T MR TrueFISP sequence

目的探讨基于产前3.0 T真稳态进动梯度回波序列(TrueFISP)MRI胎儿股骨生长动力学特征。方法纳入2018年1月—2021年12月青岛大学附属泰安市中心医院进行产前MR检查的66例孕妇的临床资料。孕妇年龄20~46(30.7±6.4)岁,孕25^(+1)~38^(+6)(32.7±3.5)周。使用3.0 T MR对胎儿股骨行产前TrueFISP扫描,在矢状面图像上测量胎儿股骨干长度,股骨近端和远端软骨骨骺的最大前后径和高度;在冠状面图像上测量股骨近端和远端软骨骨骺的最大横径。观察指标:(1)观察胎儿股骨MRI解剖特征;(2)观察胎儿股骨干长度及近端、远端软骨骨骺横径、前后径、高度与孕周的相关性。计算各项影像学指标在10种回归模型(线性、对数曲线、反函数曲线、二次曲线、三次曲线、复合曲线、幂函数曲线、S形曲线、生长曲线和指数曲线回归)中的决定系数(R2)值。根据最大R2值确定各项观察指标的最适回归模型,并获取相应的最佳生长动力学回归方程。结果(1)MR TrueFISP图像显示66例胎儿股骨解剖特征:孕28周之前,胎儿股骨干呈低信号,骨皮质与骨髓腔分界不清;孕28周之后,胎儿股骨干骨皮质呈低信号,骨髓腔呈条状较高信号,骨髓腔与骨皮质分界清楚。孕25~39周,胎儿股骨干上方为软骨性股骨头及颈、大转子,呈高信号,与股骨干上端分界清楚;股骨干下方为股骨远端软骨性骨骺,呈高信号,与股骨干下端分界清楚。(2)在10种回归模型中,胎儿股骨各项影像学指标均随着孕周的增加呈曲线增长。胎儿股骨干长度以三次曲线模型为最适回归模型,R2值0.882,曲线拟合优度高,最佳生长动力学回归方程为Y_(股骨干长度)=-21.879+2.972×X_(孕周)-4.251×10^(-3)×X^(3)_(孕周)。胎儿股骨近端软骨骨骺高度以三次曲线模型为最适回归模型,R2值0.642,曲线拟合优度一般,最佳生长动力学回归方程为Y_(股骨近端软骨骨骺高度)=-16.720+1.357×X孕周-3.227×10^(-3)×X^(3)_(孕周)。胎儿股骨远端软骨骨骺前后径以幂函数曲线模型为最适回归模型,R2值0.489,曲线拟合优度低,最佳生长动力学回归方程为Ŷ股骨远端软骨骨骺前后径=0.887×X0.861孕周。胎儿股骨近端软骨骨骺横径、前后径以及远端软骨骨骺横径和高度均同时遵从复合曲线、生长曲线和指数曲线模型,R^(2)值分别为0.568、0.452、0.549和0.387,曲线拟合优度低,其指数曲线模型的最佳生长动力学回归方程分别为Y_(股骨近端软骨骨骺横径)=6.961×exp(0.028×X孕周)、Y_(股骨近端软骨骨骺前后径)=7.301×exp(0.025×X_(孕周))、Y_(股骨远端软骨骨骺横径)=6.114×exp(0.035×X_(孕周))和Y_(股骨远端软骨骨骺高度)=8.727×exp(0.021×X_(孕周))。结论基于产前3.0 T的TrueFISP MRI,胎儿股骨各影像学指标的生长动力学特征符合曲线回归模型,可为产前评估胎儿骨骼发育状况提供重要依据。

Objective This study aims to investigate the growth dynamic characteristics of the fetal femur based on true fast imaging with steady-state precession(True FISP)images of prenatal magnetic resonance imaging(MRI).Methods The clinical data of 66 pregnant women who underwent an MRI examination in the Affiliated Taian City Central Hospital of Qingdao University,Qingdao,China,from January 2018 to December 2021 were included.The pregnant women were 20-46(30.7±6.4)years old,and the gestational age was 25^(+1)-38^(+6)(32.7±3.5)weeks.True FISP sequence scans was performed on the fetal femurs using 3.0T MR.The length of the femoral shaft and the maximum anteroposterior and high diameters of proximal and distal femoral cartilage epiphysis were measured on sagittal images.The maximum transverse diameters of the proximal and distal epiphyses of the femur were measured on coronal images.Observed indicators were(1)the MRI anatomical characteristics of the fetal femur and(2)the correlation between fetal femoral shaft length,transverse diameter,anteroposterior diameter,and high diameter of the proximal and distal femoral cartilage epiphysis and gestational age.The coefficients of determination(R2)of each imaging index in 10 regression models(linear,logarithmic,inverse,quadratic,cubic,composite,power,S-shaped,growth,and exponential curves)were calculated.According to the maximum R^(2)value,the optimal regression model of each observation index was determined,and the corresponding optimal regression equation of growth dynamics was obtained.Results(1)Anatomical features of 66 fetal femurs shown by MRI TrueFISP images:fetal femoral shaft showed low signal and the boundary between bone cortex and bone marrow cavity was unclear before 28 weeks of gestation.The femoral shaft cortex showed a low signal and the bone marrow cavity showed a strip high signal,with a clear boundary between bone marrow cavity and bone cortex,after 28 weeks of gestation.The cartilaginous femoral head,neck,and greater trochanter above the femoral shaft,showing high signal and clear boundary with the upper end of the femoral shaft,were observed from 25 to 39 weeks of gestation.Below the fetal femoral shaft is the cartilaginous epiphysis of the distal femur,which is hypersignaled and demarcated from the lower end of the femoral shaft.(2)In the 10 regression models,all imaging indicators of the fetal femur showed a curve increase with the increase of gestational age.The cubic curve model was the most suitable regression model for fetal femoral shaft length,with an R^(2)value of 0.882 and a good curve fitting.The optimal growth dynamics regression equation was Y_(femoral shaft length)=-21.879+2.972×Xgestational age-4.251×10^(−3)×X3 gestational age.The cubic curve model was used as the optimal regression model for the height diameter of the fetal proximal femur cartilage epiphysis,with an R2 value of 0.642 and a general curve fitting.The optimal growth dynamic regression equation was Ŷproximal femur cartilage epiphysis height=-16.720+1.357×Xgestational age-3.227×10^(−3)×X3 gestational age.The power curve model was used as the optimal regression model with an R^(2) value of 0.489 and a low curve fitting.The optimal growth dynamics regression equation was Ŷanteroposterior diameter of distal femoral cartilage epiphysis=0.887×X0.861 gestational age.The transverse,anterior,and posterior diameters of the proximal femoral cartilage epiphyseal and the transverse and high diameters of the distal femoral cartilage epiphyseal all simultaneously followed the composite curve,growth curve,and exponential curve model with R2 values of 0.568,0.452,0.549,and 0.387,respectively.Their curve fittings were low,and the optimal growth dynamics regression equations of the exponential curve model were as follows:Ŷtransverse diameter of proximal femoral cartilage epiphysis=6.961×exp(0.028×Xgestational age),Ŷanteroposterior diameter of proximal femoral cartilage epiphysis=7.301×exp(0.025×Xgestational age),Ŷtransverse diameter of distal femoral cartilage epiphysis=6.114×exp(0.035×Xgestational age),and Ŷheight diameter of distal femoral cartilage epiphysis=8.727×exp(0.021×Xgestational age),respectively.Conclusion Based on the TrueFISP images of prenatal MRI,the growth dynamics characteristics of each imaging index of fetal femur conforms to the curving regression model,which can provide an important basis for prenatal evaluation of fetal skeletal development.