磁共振多回波Dixon技术对肥胖儿童肝脏脂肪含量的定量评估

Quantitative Evaluation of Liver Fat Content in Obese Children via Magnetic Resonance Multi-Echo Dixon Technique

目的 探讨磁共振多回波Dixon技术对肥胖儿童肝脏脂肪含量定量分析的价值。资料与方法 前瞻性收集2019年12月—2020年12月在湖南省儿童医院诊断为肥胖的73例儿童,根据非酒精性脂肪性肝病诊疗指南分为3组:肥胖A组29例为超声未提示脂肪肝并谷丙转氨酶(ALT)、谷草转氨酶(AST)正常;肥胖B组21例为超声提示脂肪肝并ALT、AST正常;肥胖C组23例为超声提示脂肪肝并ALT或AST异常,选择年龄、性别相匹配的健康儿童27例作为对照组。测量肥胖相关形态指标(体重、身高、腰围、体重指数)及实验室指标[ALT、AST、AST/ALT、尿酸、血清甘油三酯、高密度脂蛋白、低密度脂蛋白、空腹血糖、空腹胰岛素、胰岛素抵抗指数(HOMA-IR)],行肝脏多回波Dixon序列扫描获得定量参数质子密度脂肪分数(PDFF)。分析PDFF与肥胖相关各形态指标及实验室指标的相关性,比较各组间PDFF差异及PDFF对不同组肝脏脂肪浸润程度的鉴别。结果 PDFF与体重指数(r=0.718)、腰围(r=0.740)、ALT(r=0.752)、AST(r=0.584)、AST/ALT(r=-0.737)、尿酸(r=0.607)、血清甘油三酯(r=0.384)、高密度脂蛋白(r=-0.426)、空腹胰岛素(r=0.666)、HOMA-IR(r=0.657)均相关(P均<0.01)。对照组、肥胖A组、肥胖B组、肥胖C组的PDFF值分别为(2.14±0.97)%、(3.18±1.90)%、(9.36±7.59)%、(21.49±11.20)%,各组比较差异有统计学意义(F=46.032,P<0.01)。受试者工作特征曲线显示PDFF鉴别对照组和肥胖A组、肥胖A组和肥胖B组、肥胖B组和肥胖C组的敏感度分别为0.690、0.769、0.826,特异度分别为0.667、0.793、0.810,曲线下面积分别为0.710、0.831、0.826。结论 多回波Dixon技术可以早期发现肥胖儿童肝脏脂肪浸润,其定量参数PDFF与肝脏脂肪浸润程度密切相关,是一个敏感、可靠的定量评估肝脏脂肪含量的影像学指标。

Purpose To explore the value of the multi-echo Dixon technique of magnetic resonance in a quantitative analysis of the liver fat content in obese children. Materials and Methods Seventy-three children diagnosed obesity in Hunan Children’s Hospital from December 2019 to December 2020 were prospectively collected and divided into three groups: children in obese group A(29 cases) were diagnosed by ultrasonography as not having a fatty liver with normal aspartate transaminase(AST) and alanine aminotransferase(ALT). In obese group B(21 cases), the ultrasound indicated a fatty liver with normal AST and ALT. In obese group C(23 cases), the ultrasound indicated fatty liver with abnormal AST or ALT. Twenty-seven age-and gender-matched healthy children were selected as the control group. Obesity-related morphological indicators(including weight, height, abdomen circumference and body mass index) and laboratory index(ALT, AST, AST/ALT, uric acid, triglycerides, high-density lipoprotein, low-density lipoprotein, fasting blood-glucose, fasting insulin and homeostasis model assessment of insulin resistance) were measured and obtained. A liver multi-echo Dixon sequence scan was performed to obtain the quantitative parameters of the patient’s proton density fat fraction(PDFF) in order to analyze the correlations between PDFF and clinical obesity-related indicators, and the differences of PDFF among the different groups were compared. Results PDFF was significantly correlated with body mass index(r=0.718), waist circumference(r=0.740), ALT(r=0.752), AST(r=0.584),AST/ALT(r=-0.737), uric acid(r=0.607), triglycerides(r=0.384), high-density lipoprotein(r=-0.426), fasting insulin(r=0.666) and homeostasis model assessment of insulin resistance(r=0.657), all P<0.01. The mean PDFF values were as follows: control group(2.14±0.97)%;obesity group A(3.18±1.90)%;obesity group B(9.36±7.59)%;and obesity group C(21.49±11.20)%, respectively, and there were significant differences in mean PDFF values among these four groups(F=46.032, P<0.05). The receiver operator characteristic curve(ROC) showed that the sensitivity/specificity of PDFF can distinguish the control group from obesity group A, obesity group A from obesity group B, obesity group B from obesity group C. The sensitivity/specificity was 0.690/0.667, 0.769/0.793 and 0.826/0.810, respectively, and the area under the curve(AUC) was 0.710, 0.831 and 0.826, respectively. Conclusion The multi-echo Dixon technique can be used for early detecting the liver fat infiltration in obese children, and its quantitative parameter PDFF is closely related to the degree of liver fat infiltration, which is a sensitive and reliable imaging marker for quantitatively evaluating the liver fat content.