Bril Fernando, McPhaul Michael J, Kalavalapalli Srilaxmi, Lomonaco Romina, Barb Diana, Gray Meagan E, Shiffman Dov, Rowland Charles M, Cusi Kenneth
Internal Medicine, Department of Medicine, University of Alabama at Birmingham, Birmingham, AL 35294, USA.
Division of Endocrinology, Diabetes and Metabolism, University of Florida, Gainesville, FL 32610, USA.
J Clin Endocrinol Metab. 2021 Oct 21;106(11):e4360-e4371. doi: 10.1210/clinem/dgab417.
Patients with nonalcoholic fatty liver disease (NAFLD) are characterized by insulin resistance and hyperinsulinism. However, insulin resistance measurements have not been shown to be good diagnostic tools to predict NAFLD in prior studies.
We aimed to assess a newly validated method to measure intact molecules of insulin by mass spectrometry to predict NAFLD.
Patients underwent a 2-hour oral glucose tolerance test (OGTT), a liver magnetic resonance spectroscopy (1H-MRS), and a percutaneous liver biopsy if they had a diagnosis of NAFLD. Mass spectrometry was used to measure intact molecules of insulin and C-peptide.
A total of 180 patients were recruited (67% male; 52 ± 11 years of age; body mass index [BMI] 33.2 ± 5.7 kg/m2; 46% with diabetes and 65% with NAFLD). Intact fasting insulin was higher in patients with NAFLD, irrespective of diabetes status. Patients with NAFLD without diabetes showed ~4-fold increase in insulin secretion during the OGTT compared with all other subgroups (P = 0.008). Fasting intact insulin measurements predicted NAFLD in patients without diabetes (area under the receiver operating characteristic curve [AUC] of 0.90 [0.84-0.96]). This was significantly better than measuring insulin by radioimmunoassay (AUC 0.80 [0.71-0.89]; P = 0.007). Intact fasting insulin was better than other clinical variables (eg, aspartate transaminase, triglycerides, high-density lipoprotein, glucose, HbA1c, and BMI) to predict NAFLD. When combined with alanine transaminase (ALT) (intact insulin × ALT), it detected NAFLD with AUC 0.94 (0.89-0.99) and positive and negative predictive values of 93% and 88%, respectively. This newly described approach was significantly better than previously validated noninvasive scores such as NAFLD-LFS (P = 0.009), HSI (P < 0.001), and TyG index (P = 0.039).
In patients without diabetes, accurate measurement of fasting intact insulin levels by mass spectrometry constitutes an easy and noninvasive strategy to predict presence of NAFLD.
非酒精性脂肪性肝病（NAFLD）患者的特征是胰岛素抵抗和高胰岛素血症。然而，在先前的研究中，胰岛素抵抗测量并未被证明是预测NAFLD的良好诊断工具。
我们旨在评估一种新验证的通过质谱法测量胰岛素完整分子以预测NAFLD的方法。
患者接受了2小时口服葡萄糖耐量试验（OGTT）、肝脏磁共振波谱分析（1H-MRS），如果被诊断为NAFLD，则进行经皮肝活检。质谱法用于测量胰岛素和C肽的完整分子。
共招募了180名患者（67%为男性；年龄52±11岁；体重指数[BMI]33.2±5.7kg/m2；46%患有糖尿病，65%患有NAFLD）。无论糖尿病状态如何，NAFLD患者的空腹胰岛素完整分子水平更高。与所有其他亚组相比，无糖尿病的NAFLD患者在OGTT期间的胰岛素分泌增加了约4倍（P=0.008）。空腹胰岛素完整分子测量可预测无糖尿病患者的NAFLD（受试者操作特征曲线下面积[AUC]为0.90[0.84-0.96]）。这明显优于通过放射免疫测定法测量胰岛素（AUC为0.80[0.71-0.89]；P=0.007）。空腹胰岛素完整分子在预测NAFLD方面优于其他临床变量（如天冬氨酸转氨酶、甘油三酯、高密度脂蛋白、血糖、糖化血红蛋白和BMI）。当与丙氨酸转氨酶（ALT）（胰岛素完整分子×ALT）结合时，其检测NAFLD的AUC为0.94（0.89-0.99），阳性和阴性预测值分别为93%和88%。这种新描述的方法明显优于先前验证的非侵入性评分，如NAFLD-LFS（P=0.009）、HSI（P<0.001）和TyG指数（P=0.039）。
在无糖尿病患者中，通过质谱法准确测量空腹胰岛素完整分子水平构成了一种简单且非侵入性的预测NAFLD存在的策略。
