Relationship Between Serum β2-MG，1，5-AG，GLP-1 and Islet Function in Diabetic Patients and Their Predictive Value for Comorbid Cognitive Impairment
-
摘要:
目的 探究血清β2-微球蛋白（β2-microglobulin，β2-MG）、1，5-脱水葡萄糖醇（1，5-Anhydroglucitol，1，5-AG）、胰高血糖素样肽-1（glucagon-like peptide-1，GLP-1）与糖尿病（diabetes mellitus，DM）患者胰岛功能的关系，并基于上述指标探讨其对DM患者合并认知功能障碍（cognition impairment，CI）的预测价值。 方法 前瞻性选择2020年8月至2024年8月四川省人民医院金牛医院收治的206例2型糖尿病（type 2 diabetes，T2DM）患者作为研究对象，纳入T2DM组，根据是否合并CI分为CI组（n = 62）和N-CI组（n = 144）。另选取同期206例于该院进行体检的健康志愿者作为对照（NC组）。检测血清β2-MG、1，5-AG、GLP-1水平；采用Pearson法分析血清β2-MG、1，5-AG、GLP-1水平与胰岛功能指标相关性；多因素Logistic回归分析T2DM患者合并CI的影响因素；ROC曲线评估血清β2-MG、1，5-AG、GLP-1水平对T2DM患者合并CI的预测价值。 结果 与NC组相比，T2DM组血清FPG、2hPG、HbA1c、FINS、HOMA-IR、β2-MG水平升高，HOMA-β、1，5-AG、GLP-1降低（P < 0.05）。血清β2-MG水平与FPG、2hPG、HbA1c、FINS、HOMA-IR呈正相关，与HOMA-β呈负相关（P < 0.05），1，5-AG、GLP-1与其相反（P < 0.05）。与N-CI组相比，CI组受教育年限及血清1，5-AG、GLP-1水平降低，T2DM病程、β2-MG水平升高（P < 0.05）。校正受教育年限、Fazekas总分、T2DM病程后，β2-MG、1，5-AG、GLP-1仍是T2DM患者合并CI的影响因素（P < 0.05）。血清β2-MG、1，5-AG、GLP-1水平联合预测T2DM患者合并CI的AUC（0.968）显著高于单一指标预测的AUC（0.817、0.845、0.840）（P < 0.05）。内部验证显示模型拟合良好（Hosmer-Lemeshow检验：χ2 = 4.554，P = 0.804），预测概率与实际结果一致。DCA曲线显示，联合检测模型在0.04～0.96的阈值概率区间内提供更高的净获益。 结论 T2DM患者血清β2-MG水平升高，1，5-AG、GLP-1水平降低，与胰岛功能及CI密切相关，联合检测对T2DM患者合并CI的预测效能较高。 Abstract:Objective To investigate the relationship between serum β2-microglobulin （β2-microglobulin, β2-MG）, 1, 5-anhydro-D-sorbitol （1, 5-Anhydroglucitol, 1, 5-AG）, glucagon-like peptide-1 （glucagon-like peptide-1, GLP-1） and pancreatic islet function in patients with diabetes mellitus （diabetes mellitus, DM）, and to explore the predictive value of these indicators for comorbid cognitive impairment （cognition impairment, CI） in DM patients. Methods A total of 206 patients with type 2 diabetes mellitus （type 2 diabetes, T2DM） admitted to Jin Niu Hospital of Sichuan Provincial People's Hospital from August 2020 to August 2024 were prospectively selected as the T2DM group. According to the presence of CI, they were further divided into a CI group （n = 62） and a non-CI （N-CI） group （n = 144）. Another 206 healthy volunteers undergoing physical examinations during the same period were selected as the normal control （NC） group. Serum levels of β2-MG, 1, 5-AG, and GLP-1 were measured. Pearson correlation analysis was used to assess the correlation between serum levels of β2-MG, 1, 5-AG, GLP-1 and pancreatic islet function indicators. Multivariate logistic regression was applied to identify influencing factors for comorbid CI in T2DM patients. The predictive value of serum β2-MG, 1, 5-AG, and GLP-1 levels for comorbid CI in T2DM patients were evaluated using receiver operating characteristic （ROC） curves. Results Compared with the NC group, the T2DM group showed increased levels of FPG, 2hPG, HbA1c, FINS, HOMA-IR, and β2-MG, and decreased levels of HOMA-β, 1, 5-AG, and GLP-1 （P < 0.05）. Serum β2-MG levels were positively correlated with FPG, 2hPG, HbA1c, FINS, and HOMA-IR, and negatively correlated with HOMA-β （P < 0.05）. while 1, 5-AG and GLP-1 showed opposite correlations （P < 0.05）. Compared with the N-CI group, the CI group had fewer years of education and lower serum levels of 1, 5-AG and GLP-1, but longer T2DM duration and higher β2-MG levels （P < 0.05）. After adjusting for years of education, Fazekas' total score and T2DM duration, β2-MG, 1, 5-AG, and GLP-1 remained independent influencing factors for comorbid CI in T2DM patients （P < 0.05）. The area under the curve （AUC） for the combined prediction of comorbid CI in T2DM patients using serum β2-MG, 1, 5-AG, and GLP-1 levels （0.968） was significantly higher than that of any single indicator （0.817, 0.845, 0.840） （P < 0.05）. Internal validation indicated good model fit （Hosmer-Lemeshow test: χ2 = 4.554, P = 0.804）, with predicted probabilities consistent with actual outcomes. The DCA curve showed that the combined detection model provided higher net benefit within the threshold probability range of 0.04 to 0.96. Conclusion T2DM patients exhibit elevated serum β2-MG levels and decreased 1, 5-AG and GLP-1 levels, which are closely associated with islet function and CI. The combined detection of these markers demonstrates high predictive efficacy for comorbid CI in T2DM patients. -
图 2 内部验证预测模型的校准曲线
Figure 2. Calibration Curve of the Prediction Model from Internal Validation
表 1 NC组、T2DM组血清β2-MG、1，5-AG、GLP-1水平及胰岛功能指标比较（$ \bar x \pm s $）
Table 1. Comparison of serum β2-MG，1，5-AG，GLP-1 levels and pancreatic function indicators between the NC group and the T2DM group（$ \bar x \pm s $）
指标 NC组（n = 206） T2DM组（n = 206） t P 年龄（岁） 58.95 ± 8.33 57.66 ± 7.75 1.627 0.104 性别 男 105（50.97） 111（53.88） 0.350 0.554 女 101（49.03） 95（46.12） FPG（mmol/L） 5.21 ± 0.53 8.42 ± 2.45 18.380 <0.001* 2hPG（mmol/L） 6.69 ± 1.03 10.76 ± 3.84 14.693 <0.001* HbA1c（%） 5.52 ± 0.65 7.67 ± 1.71 16.868 <0.001* FINS（mIU/L） 8.01 ± 2.43 12.77 ± 3.55 15.881 <0.001* HOMA-IR 1.75 ± 0.23 5.14 ± 1.30 36.855 <0.001* HOMA-β 95.64 ± 15.01 42.12 ± 10.37 42.105 <0.001* β2-MG（mg/L） 1.83 ± 0.42 4.04 ± 0.65 40.987 <0.001* 1，5-AG（mmol/L） 143.14 ± 33.81 52.98 ± 10.53 36.543 <0.001* GLP-1（pmol/L） 35.10 ± 6.58 22.25 ± 3.34 24.994 <0.001* *P < 0.05。 
下载: 导出CSV
表 2 T2DM组患者血清β2-MG、1，5-AG、GLP-1水平与胰岛功能指标相关性
Table 2. Correlation between serum β2-MG，1，5-AG，GLP-1 levels and pancreatic function indicators in the T2DM group
临床指标 β2-MG 1，5-AG GLP-1 r P r P r P FPG 0.433 0.007* −0.464 0.004* −0.519 <0.001* 2hPG 0.398 0.046* −0.635 <0.001* −0.486 0.001* HbA1c 0.617 <0.001* −0.489 0.001* −0.558 <0.001* FINS 0.405 0.032* −0.388 0.048* −0.407 0.031* HOMA-IR 0.482 0.002* −0.598 <0.001* −0.467 0.003* HOMA-β −0.514 <0.001* 0.411 0.012* 0.502 <0.001* *P < 0.05。
下载: 导出CSV
表 3 T2DM患者合并CI的单因素分析[（$ \bar x \pm s $）/n（%）]
Table 3. Univariate analysis of comorbid CI in T2DM patients[（$ \bar x \pm s $）/n（%）]
指标 N-CI组（n = 144） CI组（n = 62） χ2/t P 年龄（岁） 57.46 ± 7.56 58.14 ± 7.90 0.584 0.560 性别 男 81（56.25） 30（48.39） 1.078 0.299 女 63（43.75） 32（51.61） BMI（kg/m2） 23.58 ± 3.05 23.87 ± 3.33 0.609 0.543 吸烟史 50（34.72） 28（45.16） 2.007 0.157 饮酒史 40（27.78） 22（35.48） 1.223 0.269 HDL-C（mmol/L） 1.19 ± 0.43 1.32 ± 0.45 1.963 0.051 LDL-C（mmol/L） 3.15 ± 0.40 3.24 ± 0.47 1.404 0.162 TG（mmol/L） 1.59 ± 0.27 1.63 ± 0.25 0.997 0.320 TC（mmol/L） 4.66 ± 0.75 4.85 ± 0.84 1.608 0.109 受教育年限（年） 11.25 ± 3.65 8.95 ± 2.11 4.635 <0.001* T2DM病程（年） 5.83 ± 1.65 8.21 ± 2.01 8.875 <0.001* 高血压 23（15.97） 14（22.58） 1.284 0.257 高脂血症 22（15.28） 15（24.19） 2.338 0.126 T2DM家族史 37（25.69） 20（32.26） 0.933 0.334 PSQI（分） 5.83 ± 1.84 6.36 ± 2.08 1.822 0.070 HAMD评分（分） 8.42 ± 2.15 8.97 ± 2.82 1.528 0.128 Fazekas总分（分） 1.28 ± 0.41 2.07 ± 0.58 11.128 0.000 FPG（mmol/L） 8.32 ± 1.72 8.64 ± 1.53 1.265 0.207 2hPG（mmol/L） 10.63 ± 2.65 11.07 ± 2.71 1.086 0.279 HbA1c（%） 7.60 ± 2.11 7.82 ± 2.32 0.666 0.506 FINS（mIU/L） 12.72 ± 3.08 12.90 ± 3.17 0.381 0.703 HOMA-IR 5.05 ± 1.07 5.35 ± 1.20 1.779 0.077 HOMA-β 42.38 ± 6.18 41.51 ± 6.54 0.911 0.364 β2-MG（mg/L） 3.68 ± 0.94 4.89 ± 0.91 8.555 <0.001* 1，5-AG（mmol/L） 56.17 ± 7.75 45.57 ± 8.33 8.802 <0.001* GLP-1（pmol/L） 23.52 ± 3.85 19.31 ± 3.01 7.658 <0.001* *P < 0.05。
下载: 导出CSV
表 4 T2DM患者合并CI的多因素Logistic回归分析
Table 4. Multivariate Logistic regression analysis of comorbid CI in T2DM patients
影响因素 β SE Wald χ2 OR 95%CI P 模型1 Fazekas总分 0.898 0.426 4.453 2.457 1.066～5.663 0.035* 受教育年限 −0.228 0.084 7.377 0.796 0.675～0.938 0.007* 病程 0.745 0.225 10.957 2.106 1.355～3.273 0.001* β2-MG 1.129 0.287 15.470 3.092 1.762～5.427 <0.001* 1，5-AG −0.435 0.154 7.994 0.647 0.478～0.852 0.005* GLP-1 −0.435 0.127 11.754 0.647 0.504～0.830 <0.001* 常量 −3.321 0.892 12.559 0.032 — <0.001* 模型2 β2-MG 0.671 0.246 7.449 1.957 1.208～3.169 0.006* 1，5-AG −0.345 0.134 6.641 0.708 0.544～0.921 0.010* GLP-1 −0.260 0.104 6.253 0.771 0.629～0.945 0.012* 常量 −2.845 0.372 43.876 0.087 — <0.001* *P < 0.05；模型1未校正；模型2校正受教育年限、Fazekas总分、T2DM病程。
下载: 导出CSV
表 5 血清β2-MG、1，5-AG、GLP-1水平对T2DM患者合并CI的预测价值
Table 5. Predictive value of serum β2-MG，1，5-AG and GLP-1 levels for comorbid CI in T2DM patients
指标 截断值 敏感性（%） 特异性（%） Youden指数 AUC 95%CI β2-MG 3.87 87.10 65.28 0.524 0.817 0.757～0.867 1，5-AG 53.62 91.94 63.19 0.551 0.845 0.788～0.891 GLP-1 21.91 96.77 56.94 0.537 0.840 0.782～0.887 联合 0.45 91.94 92.36 0.814 0.968 0.933～0.987 Zβ2-MG-联合/P 5.121/<0.001* Z1，5-AG-联合/P 4.979/<0.001* ZGLP-1-联合/P 5.284/<0.001* *P < 0.05。
下载: 导出CSV
-
[1] Yu X, He H, Wen J, et al. Diabetes-related cognitive impairment: Mechanisms, symptoms, and treatments[J]. Open Med, 2025, 20(1): 20241091. doi: 10.1515/med-2024-1091 [2] Zhang Y, Peng G, Leng W, et al. Association between serum β2-microglobulin and left ventricular hypertrophy in patients with type 2 diabetes mellitus: A cross-sectional study[J]. J Diabetes, 2024, 16(8): e13599. doi: 10.1111/1753-0407.13599 [3] Xu H, Chen R, Hou X, et al. The clinical potential of 1, 5-anhydroglucitol as biomarker in diabetes mellitus[J]. Front Endocrinol, 2024, 15: 1471577. doi: 10.3389/fendo.2024.1471577 [4] 谭爱春, 张骞, 程杰, 等. 血清1, 5-脱水葡萄糖醇水平与2型糖尿病患者胰岛素抵抗和微血管并发症的关系[J]. 中华预防医学杂志, 2024, 58(6): 875-882. doi: 10.3760/cma.j.cn112150-20240219-00129 [5] 吴梦梦, 周迪夷. GLP-1受体激动剂对2型糖尿病患者胰岛β细胞保护机制的研究进展[J]. 浙江医学, 2023, 45(23): 2568-2571. doi: 10.12056/j.issn.1006-2785.2023.45.23.2023-691 [6] Hölscher C. Glucagon-like peptide-1 class drugs show clear protective effects in Parkinson’ s and Alzheimer’ s disease clinical trials: A revolution in the making?[J]. Neuropharmacology, 2024, 253: 109952. doi: 10.1016/j.neuropharm.2024.109952 [7] 高永祥, 张晋昕. Logistic回归分析的样本量确定[J]. 循证医学, 2018, 18(2): 122-124. [8] 陈红, 李小明. 老年2型糖尿病认知功能损害及影响因素探究[J]. 中国医师杂志, 2019, 21(11): 1736-1738. doi: 10.3760/cma.j.issn.1008-1372.2019.11.038 [9] 中华医学会, 中华医学会杂志社, 中华医学会全科医学分会, 等. 2型糖尿病基层诊疗指南(实践版·2019)[J]. 中华全科医师杂志, 2019, 18(9): 810-818. [10] Smyth C. The Pittsburgh sleep quality index (PSQI)[J]. Insight J Am Soc Ophthalmic Regist Nurses, 2000, 25(3): 97-98. [11] Leucht S, Fennema H, Engel R, et al. What does the HAMD mean?[J]. J Affect Disord, 2013, 148(2-3): 243-248. doi: 10.1016/j.jad.2012.12.001 [12] Cedres N, Ferreira D, Machado A, et al. Predicting Fazekas scores from automatic segmentations of white matter signal abnormalities[J]. Aging, 2020, 12(1): 894-901. doi: 10.18632/aging.102662 [13] Kang J M, Cho Y S, Park S, et al. Montreal cognitive assessment reflects cognitive reserve[J]. BMC Geriatr, 2018, 18(1): 261. doi: 10.1186/s12877-018-0951-8 [14] Garfield V, Farmaki A E, Eastwood S V, et al. HbA1c and brain health across the entire glycaemic spectrum[J]. Diabetes Obes Metab, 2021, 23(5): 1140-1149. doi: 10.1111/dom.14321 [15] Yang Y, Zhao J J, Yu X F. Expert consensus on cognitive dysfunction in diabetes[J]. Curr Med Sci, 2022, 42(2): 286-303. doi: 10.1007/s11596-022-2549-9 [16] Lu H H, Zhou Y, Chen C, et al. Meta-analysis of the effect of exercise intervention on cognitive function in elderly patients with type 2 diabetes mellitus[J]. BMC Geriatr, 2024, 24(1): 770. doi: 10.1186/s12877-024-05352-z [17] Shen Y, Chen J J, Yao W B, et al. Predictive value of serum β2-microglobulin in cardiac valve calcification in maintenance hemodialysis patients[J]. J Thorac Dis, 2023, 15(9): 4914-4924. doi: 10.21037/jtd-23-1185 [18] Gong S, Ma R, Zhu T, et al. Elevated serum beta-2 microglobulin level predicts short-term poor prognosis of patients with de novo acute Omicron variant COVID-19 infection[J]. Front Cell Infect Microbiol, 2023, 13: 1204326. doi: 10.3389/fcimb.2023.1204326 [19] 司璐, 薛丽娜, 马胜南, 等. 血清LPA、AngII及β2-MG在老年2型糖尿病患者病情监控中的作用及其与胰岛素抵抗的相关性分析[J]. 现代生物医学进展, 2024, 24(2): 386-390. [20] Yang H, Wu Q, Zheng J. Expression and clinical significance of 25-hydroxyvitamin D, insulin-like growth factor 1, and beta-2 microglobulin in cognitive dysfunction after ischemic stroke in the elderly[J]. Neuroreport, 2025, 36(3): 127-134. doi: 10.1097/WNR.0000000000002128 [21] 詹江萍, 周建平, 陈小芳, 等. β2-MG、25(OH)D、Hcy对老年糖尿病患者合并认知功能障碍的评估价值[J]. 世界最新医学信息文摘(连续型电子期刊), 2024, 24(25): 169-172. doi: 10.3969/j.issn.1671-3141.2024.025.043 [22] Migała M, Chałubińska-Fendler J, Zielińska M. 1, 5-anhydroglucitol as a marker of acute hyperglycemia in cardiovascular events[J]. Rev Diabet Stud, 2022, 18(2): 68-75. doi: 10.1900/RDS.2022.18.68 [23] 李雪艳, 廖经忠, 黄康康, 等. 血清1, 5-AG在2型糖尿病早期诊断中的应用价值[J]. 现代医学, 2023, 51(9): 1220-1226. doi: 10.3969/j.issn.1671-7562.2023.09.006 [24] 王丽娜, 贾新菊, 郭玉卿, 等. 老年2型糖尿病患者1, 5-脱水葡萄糖醇水平与轻度认知功能障碍的相关性[J]. 中华行为医学与脑科学杂志, 2024, 33(7): 618-623. doi: 10.3760/cma.j.cn371468-20231127-00264 [25] Bellia C, Lombardo M, Meloni M, et al. Diabetes and cognitive decline[J]. Adv Clin Chem, 2022, 108: 37-71. doi: 10.1016/b978-0-12-407824-6.00035-5 [26] García-Casares N, González-González G, de la Cruz-Cosme C, et al. Effects of GLP-1 receptor agonists on neurological complications of diabetes[J]. Rev Endocr Metab Disord, 2023, 24(4): 655-672. doi: 10.1007/s11154-023-09807-3 [27] Panou T, Gouveri E, Popovic D S, et al. The therapeutic potential of dipeptidyl peptidase 4 inhibitors and glucagon-like peptide-1 receptor agonists in diabetic peripheral neuropathy[J]. Diabetes Ther, 2025, 16(6): 1077-1105. doi: 10.1007/s13300-025-01712-z [28] 牛兆霞, 赵玉洁, 李玲玲. 血清Aβ1-42、GLP-1与2型糖尿病患者认知功能障碍的相关性及其危险因素分析[J]. 临床和实验医学杂志, 2024, 23(1): 32-36. doi: 10.3969/j.issn.1671-4695.2024.01.009 [29] 李琪, 王真, 任天成. 2型糖尿病患者血清GLP-1、Aβ1-42、MCP-1水平与认知功能障碍的相关性[J]. 分子诊断与治疗杂志, 2024, 16(3): 442-445, 449. [30] Zheng J, Xie Y, Ren L, et al. GLP-1 improves the supportive ability of astrocytes to neurons by promoting aerobic glycolysis in Alzheimer’s disease[J]. Mol Metab, 2021, 47: 101180. doi: 10.1016/j.molmet.2021.101180 -
点击查看大图
图(3) / 表(5)
计量
- 文章访问数: 30
- HTML全文浏览量: 17
- PDF下载量: 11
- 被引次数: 0
