内脏脂肪指数对多囊卵巢综合征患者胰岛素抵抗的预测价值

The Predictive Value of Visceral Adiposity Index for Insulin Resistance in Patients with Polycystic Ovary Syndrome

摘要: 目的：探究能否在多囊卵巢综合征(polycystic ovary syndrome, PCOS)人群中应用内脏脂肪指数(visceral adiposity index, VAI)预测胰岛素抵抗(insulin resistance, IR)。方法：本回顾性队列研究连续纳入2023年7月至2024年9月于重庆医科大学附属第二医院妇科就诊，符合入组标准的PCOS患者244例，根据是否存在IR分为IR组及非IR组，比较两组VAI和中国人内脏脂肪指数(Chinese visceral adiposity index, CVAI)水平。应用Logistic回归分析CVAI与PCOS患者患IR的相关性。受试者工作特征(receiver operating characteristic, ROC)曲线确定VAI和CVAI预测PCOS患者IR的截断值。混淆矩阵评价在PCOS患者中应用CVAI预测IR的效能。根据CVAI预测PCOS患者IR的截断值分为高CVAI组及低CVAI组，比较两组IR患病率及临床特征。结果：IR组CVAI (64.45 vs. 15.13)显著高于非IR组，差异有统计学意义(P < 0.05)。Logistic回归分析表明，CVAI (OR = 1.058, 95%CI 1.039~1.079, P < 0.001)为PCOS患者发生IR的独立危险因素(P < 0.05)。ROC曲线和混淆矩阵结果显示，CVAI预测PCOS患者IR的截断值为42.73，曲线下面积为0.889，约登指数为0.651，灵敏度为81.20%，特异度为83.90%，准确率为82.79%，精确率为78.10%，召回率为81.19%。高CVAI组IR患病率(78.10% vs. 13.67%)高于低CVAI组，差异有统计学意义(P < 0.05)。高CVAI组腰臀比(0.90 vs. 0.82)、游离雄激素指数(7.00% vs. 3.68%)、空腹血糖(4.95 mmol/L vs. 4.68 mmol/L)、空腹胰岛素(17.70 uU/mL vs. 7.28 uU/mL)和低密度脂蛋白胆固醇(2.82 mmol/L vs. 2.47 mmol/L)高于低CVAI组，差异有统计学意义(P < 0.05)。高CVAI组性激素结合球蛋白水平(24.34 nmol/L vs. 45.71 nmol/L)、黄体生成素(9.64 mIU/ml vs. 12.19 mIU/ml)、黄体生成素卵泡刺激素比值(1.70 vs. 1.88)低于低CVAI组，差异有统计学意义(P < 0.05)。睾酮(51.42 ng/dL vs. 50.11 ng/dL)、硫酸脱氢表雄酮(278.60 ng/mL vs. 268.20 ng/mL)、卵泡刺激素(5.67 mIU/ml vs. 6.12 mIU/ml)、总胆固醇(4.74 mmol/L vs. 4.53 mmol/L)在两组间的差异无统计学意义(P > 0.05)。结论：在PCOS人群中应用CVAI来预测IR具有一定的可行性，课题组将进一步开展前瞻性研究进行验证。

Abstract: Objective: To explore the feasibility of applying the visceral adiposity index (VAI) as a predictive marker for insulin resistance (IR) in individuals with polycystic ovary syndrome (PCOS). Methods: This retrospective cohort study consecutively enrolled 244 PCOS patients who visited the Department of Gynecology at the Second Affiliated Hospital of Chongqing Medical University between July 2023 and September 2024 and met the inclusion criteria. The participants were stratified into IR and non-IR groups based on insulin resistance status, with subsequent comparison of visceral adiposity index (VAI) and Chinese visceral adiposity index (CVAI) levels between the two groups. We further compared the CVAI levels between the two groups. Logistic regression analysis was applied to assess the correlation between CVAI and the risk of IR in PCOS patients. The receiver operating characteristic (ROC) curve determines the cutoff value of the CVAI for predicting IR in patients with PCOS. The confusion matrix evaluates the efficiency of CVAI in predicting IR among PCOS patients. Based on the CVAI cutoff value, participants were divided into high CVAI and low CVAI groups, and the prevalence of IR and clinical characteristics were compared between the two groups. Results: The CVAI in the IR group (64.45 vs. 15.13) was significantly higher than that in the non-IR group, with a statistically significant difference (P < 0.05). Logistic regression analysis indicated that CVAI (OR = 1.058, 95%CI 1.039~1.079, P < 0.001) is an independent risk factor for the development of IR in PCOS patients (P < 0.05). ROC curve analysis revealed that the cutoff value of CVAI for predicting IR in PCOS patients was 42.73, with an area under the curve of 0.889, a Youden index of 0.651, a sensitivity of 81.20%, and a specificity of 83.90%. In PCOS patients, the accuracy of CVAI for predicting IR was 82.38%, with a precision of 77.38% and a recall of 81.25%. The high-CVAI group exhibited significantly higher prevalence of insulin resistance (78.10% vs. 13.67%, P < 0.05), along with elevated waist-to-hip ratio (0.90 vs. 0.82), free androgen index (7.00% vs. 3.68%), fasting plasma glucose (4.95 mmol/L vs. 4.68 mmol/L), fasting insulin (17.70 μIU/mL vs. 7.28 μIU/mL), and LDL-cholesterol levels (2.82 mmol/L vs. 2.47 mmol/L) compared to the low-CVAI group (all P < 0.05). The high-CVAI group demonstrated significantly lower levels of sex hormone-binding globulin (24.34 vs 45.71 nmol/L), luteinizing hormone (9.64 vs. 12.19 mIU/mL), and LH/FSH ratio (1.70 vs 1.88) compared with the low-CVAI group (all P < 0.05). No statistically significant differences were observed between groups for testosterone (51.42 vs. 50.11 ng/dL), dehydroepiandrosterone sulfate (278.60 vs. 268.20 ng/mL), follicle-stimulating hormone (5.67 vs. 6.12 mIU/mL), or total cholesterol (4.74 vs. 4.53 mmol/L; all P > 0.05). Conclusion: The use of CVAI for predicting IR in patients with PCOS demonstrates certain feasibility, and our research team will further conduct prospective studies to validate these findings.