Construction and Validation of a Predictive Model of Serum miR-30d-5p,miR-146a-5p,and Treg/Th17 Ratio for Predicting 90-day Prognosis in Elderly Patients with Sepsis
-
摘要:
目的 探讨血清miR-30d-5p、miR-146a-5p及调节性T细胞/辅助性T细胞17(Treg/Th17)比值与老年脓毒症患者预后的关系,并构建及验证列线图预测模型。 方法 采用前瞻性观察研究方法,纳入2019年1月至2023年12月于连云港市第二人民医院接受治疗的老年脓毒症患者382例。按7∶3比例随机分为建模组(n = 268)和验证组(n = 114)。建模组根据90 d生存情况分为生存组(n = 182)与死亡组(n = 86)。采集患者入院24 h内血清及外周血,采用qRT-PCR检测血清miRNAs水平,流式细胞术检测Treg/Th17比值。多因素Logistic回归分析筛选独立危险因素,构建列线图模型;采用ROC曲线、校准曲线及决策曲线分析(decision curve analysis,DCA)验证模型效能。 结果 死亡组血清miR-30d-5p、miR-146a-5p水平均显著低于生存组,Treg/Th17比值、APACHE II评分显著高于生存组(P < 0.05)。多因素Logistic回归分析显示,低水平miR-30d-5p、低水平miR-146a-5p、高Treg/Th17比值及高APACHE II评分是老年脓毒症患者90 d死亡的独立危险因素。基于上述指标构建的列线图模型,建模组AUC为0.886 (95%CI: 0.842~0.930),验证组AUC为0.862 (95%CI: 0.795~0.929)。建模组Hosmer-Lemeshow检验 P = 0.652,验证组 P = 0.584,校准曲线显示预测概率与实际发生率一致性良好。DCA曲线证实模型具有较高的临床净获益。 结论 血清miR-30d-5p、miR-146a-5p降低及Treg/Th17比值升高与老年脓毒症不良预后密切相关,基于此构建的预测模型具有良好的区分度和准确度,可为临床评估提供参考。 -
关键词:
- 老年脓毒症 /
- miR-30d-5p /
- miR-146a-5p /
- Treg/Th17比值 /
- 预后 /
- 预测模型
Abstract:Objective To investigate the relationship between serum microRNA-30d-5p (miR-30d-5p), microRNA-146a-5p (miR-146a-5p), regulatory T cell/helper T cell 17 (Treg/Th17) ratio and the prognosis of elderly patients with sepsis, and to construct and verify a nomogram prediction model. Methods A prospective observational study was conducted, enrolling 382 elderly sepsis patients treated at the Second People's Hospital of Lianyungang from January 2019 to December 2023. Patients were randomly divided in a 7∶3 ratio into a modeling group (n = 268) and a validation group (n = 114). The modeling group was stratified into a survival group (n = 182) and a mortality group (n = 86) based on 90-day survival status. Serum and peripheral blood samples were collected within 24 hours of admission. Serum miRNA levels were detected using qRT-PCR, and the Treg/Th17 ratio was determined by flow cytometry. Multivariate Logistic regression analysis was used to identiry independent risk factors and construct a nomogram model. The model performance was validated using ROC curves, calibration curves, and decision curve analysis (DCA). Results Serum miR-30d-5p and miR-146a-5p levels were significantly lower in the mortality group compared to the survival group, while the Treg/Th17 ratio and APACHE II score were significantly higher (P < 0.05). Multivariate Logistic regression analysis showed that low miR-30d-5p levels, low miR-146a-5p levels, high Treg/Th17 ratio, and high APACHE II score were independent risk factors for 90-day mortality in elderly sepsis patients. The nomogram model constructed based on these indicators showed an AUC of 0.886 (95%CI: 0.842 ~ 0.930) in the modeling group and 0.862 (95%CI: 0.795~ 0.929) in the validation group. The Hosmer-Lemeshow test yielded P = 0.652 in the modeling group and P = 0.584 in the validation group. The calibration curve showed good consistency between predicted probability and actual incidence rate. Conclusion Decreased serum miR-30d-5p and miR-146a-5p levels, and increased Treg/Th17 ratio are closely associated with adverse prognosis in elderly sepsis patients. The prediction model constructed based on these parameters demonstrates good discriminative ability and accuracy, providing a valuable reference for clinical assessment. -
Key words:
- Elderly sepsis /
- miR-30d-5p /
- miR-146a-5p /
- Treg/Th17 ratio /
- Prognosis /
- Predictive model
-
表 1 建模组生存组与死亡组临床特征及指标比较[($ \bar x \pm s $)/n(%)/[M(Q1,Q3)]
Table 1. Comparison of clinical characteristics and indicators between survival group and mortality group [($ \bar x \pm s $)/n(%)/[M(Q1,Q3)]
临床特征/指标 生存组 (n=182) 死亡组 (n=86) t / Z / χ2 P 年龄 (岁) 74.22 ± 6.50 75.78 ± 7.09 1.824 0.069 性别 (男/女) 102 (56.0) / 80 (44.0) 49 (57.0) / 37 (43.0) 0.035 0.851 体重指数 (BMI,kg/m2) 23.47 ± 3.11 23.12 ± 3.53 0.897 0.370 合并症 高血压 95 (52.2) 48 (55.8) 0.258 0.611 糖尿病 60 (33.0) 31 (36.0) 0.251 0.616 冠心病 55 (30.2) 28 (32.6) 0.147 0.702 慢性肾脏病 38 (20.9) 20 (23.3) 0.178 0.673 感染部位 1.152 0.764 肺部 98 (53.8) 48 (55.8) 腹腔 45 (24.7) 20 (23.3) 泌尿系 21 (11.5) 11 (12.8) 其他 18 (9.9) 7 (8.1) 发病至ICU时间 (h) 10.52 ± 4.78 11.23 ± 5.31 1.099 0.273 APACHE II 评分 (分) 16.51 ± 4.23 23.11 ± 5.33 10.452 <0.001* SOFA 评分 (分) 7.23 ± 2.13 10.51 ± 2.79 9.873 <0.001* 乳酸 (Lactate,mmol/L) 2.00 (1.50,2.80) 4.50 (3.20,6.50) 10.130 <0.001* 降钙素原 (PCT,ng/mL) 3.80 (2.50,6.20) 15.00 (10.00,25.00) 12.551 <0.001* 血管活性药物使用 75 (41.2) 68 (79.1) 30.210 <0.001* miR-30d-5p (相对表达) 1.85 ± 0.42 0.92 ± 0.25 18.560 <0.001* miR-146a-5p (相对表达) 2.10 ± 0.55 1.15 ± 0.30 14.231 <0.001* Treg (%) 4.21 ± 1.10 6.85 ± 1.52 15.340 <0.001* Th17 (%) 2.15 ± 0.65 1.42 ± 0.40 9.850 <0.001* Treg/Th17 比值 2.12 ± 0.75 5.15 ± 1.30 22.410 <0.001* *P < 0.05。 表 2 老年脓毒症患者90 d预后的多因素Logistic回归分析
Table 2. Multivariate Logistic regression analysis of 90-day prognosis of elderly patients with sepsis
变量 β SE Wald χ2 P OR (95%CI) APACHE II评分 0.185 0.062 8.903 0.003* 1.203 (1.066~1.358) Treg/Th17 比值 0.652 0.150 18.895 <0.001* 1.919 (1.430~2.576) miR-30d-5p −1.250 0.320 15.258 <0.001* 0.286 (0.153~0.536) miR-146a-5p −0.980 0.280 12.250 <0.001* 0.375 (0.217~0.649) Constant −1.520 0.850 3.190 0.074 - *P < 0.05。 表 3 联合预测模型与各单一指标预测效能的比较
Table 3. Comparison of predictive performance between the combined model and individual predictors
预测指标/模型 组别 AUC (95%CI) 最佳截断值 敏感度 (%) 特异度 (%) P (vs. 联合模型) 联合预测模型 建模组 0.886 (0.842~0.930) 0.315 (总分) 84.9 81.3 - 验证组 0.862 (0.795~0.929) 0.315 (总分) 82.1 78.5 - APACHE II 评分 建模组 0.758 (0.690~0.826) 19.5 73.3 70.9 <0.001* 验证组 0.745 (0.651~0.839) 19.5 71.8 69.2 0.018* Treg/Th17 比值 建模组 0.795 (0.731~0.859) 3.55 77.9 74.2 0.003* 验证组 0.782 (0.693~0.871) 3.55 75.0 72.4 0.035* miR-30d-5p 建模组 0.812 (0.750~0.874) 1.18 80.2 75.8 0.011* 验证组 0.798 (0.710~0.886) 1.18 78.6 74.1 0.076 miR-146a-5p 建模组 0.789 (0.724~0.854) 1.45 76.7 73.1 0.005* 验证组 0.775 (0.684~0.866) 1.45 74.3 71.0 0.041* P值通过DeLong检验比较联合模型与各单一指标AUC值得出;*P < 0.05。 -
[1] Gao X, Cai S, Li X, et al. Sepsis-induced immunosuppression: Mechanisms, biomarkers and immunotherapy[J]. Front Immunol, 2025, 16: 1577105. [2] 李程锦, 石松菁, 陈湘平, 等. 老年脓毒症患者血清氨基末端脑钠肽前体升高的影响因素分析[J]. 中外医疗, 2023, 42(17): 1-6+12. [3] 曹成龙, 马向丽, 刘贻晶, 等. 乳酸化修饰在脓毒症作用机制中的研究进展[J]. 中华急诊医学杂志, 2024, 33(4): 584-590. doi: 10.3760/cma.j.issn.1671-0282.2024.04.023 [4] Li J, Jia Q, Yang L, et al. Sepsis-associated encephalopathy: Mechanisms, diagnosis, and treatments update[J]. Int J Biol Sci, 2025, 21(7): 3214-3228. doi: 10.7150/ijbs.102234 [5] Royo M, Joseph-Mullol B, Sandoval S, et al. Integrative miRNA-mRNA profiling uncovers mechanisms of belimumab action in systemic lupus erythematosus[J]. Front Immunol, 2025, 16: 1553971. doi: 10.3389/fimmu.2025.1553971 [6] Jennings VA, Rumbold-Hall R, Migneco G, et al. Enhancing oncolytic virotherapy by extracellular vesicle mediated microRNA reprograming of the tumour microenvironment[J]. Front Immunol, 2024, 15: 1500570. doi: 10.3389/fimmu.2024.1500570 [7] Deng R, Cui X, Zhang R, et al. Pathologic function and therapeutic potential of extracellular vesicle miRNA in sepsis[J]. Front Pharmacol, 2024, 15: 1452276. doi: 10.3389/fphar.2024.1452276 [8] Oda S, Matsumoto H, Togami Y, et al. mRNA–miRNA integration analysis of T-cell exhaustion in sepsis from community-acquired pneumonia[J]. Acute Med Surg, 2025, 12(1): e70054. [9] Zhao Y, Zhu R, Hu X. Diagnostic capacity of miRNAs in neonatal sepsis: A systematic review and meta-analysis[J]. J Matern Fetal Neonatal Med, 2024, 37(1): 2345850. doi: 10.1080/14767058.2024.2345850 [10] Qin M, Zhu J, Xing L, et al. Adipose-derived exosomes ameliorate skeletal muscle atrophy via miR-146a-5p/IGF-1R signaling[J]. J Nanobiotechnol, 2024, 22(1): 754. doi: 10.1186/s12951-024-02983-7 [11] Qin M, Wang Y, Wang Z, et al. Adipose-derived small extracellular vesicle miR-146a-5p targets Fbx32 to regulate mitochondrial autophagy and delay aging in skeletal muscle[J]. J Nanobiotechnol, 2025, 23(1): 287. doi: 10.1186/s12951-025-03367-1 [12] 黄伟. 《第三版脓毒症与感染性休克定义国际共识》解读[J]. 中国实用内科杂志, 2016, 36(11): 959-962. [13] Laczynski D J, Gallop J, Sicard G A, et al. Benchmarking a center of excellence in vascular surgery: Using acute physiology and chronic health evaluation II to validate outcomes in a tertiary care institute[J]. Vasc Endovascular Surg, 2023, 57(8): 856-862. doi: 10.1177/15385744231183744 [14] Yamakawa K, Okamoto K, Seki Y, et al. Clinical practice guidelines for management of disseminated intravascular coagulation in Japan 2024. Part 1: Sepsis[J]. Int J Hematol, 2025, 121(5): 592-604. doi: 10.1007/s12185-024-03896-9 [15] Gray A J, Oatey K, Grahamslaw J, et al. Albumin versus balanced crystalloid for the early resuscitation of sepsis: An open parallel-group randomized feasibility trial- the ABC-sepsis trial[J]. Crit Care Med, 2024, 52(10): 1520-1532. doi: 10.1097/CCM.0000000000006348 [16] Gu C, Liu Y, Lv J, et al. Kurarinone regulates Th17/Treg balance and ameliorates autoimmune uveitis via Rac1 inhibition[J]. J Adv Res, 2025, 69: 381-398. doi: 10.1016/j.jare.2024.03.013 [17] Ito J T, Alves L H V, de Mendonça Oliveira L, et al. Effect of exercise training on modulating the TH17/TREG imbalance in individuals with severe COPD: A randomized controlled trial[J]. Pulmonology, 2025, 31(1): 2441069. doi: 10.1080/25310429.2024.2441069 [18] 郑慧萍, 刘颖, 王念慈, 等. 老年脓毒症休克伴免疫抑制病人早期使用小剂量糖皮质激素对预后的影响[J]. 实用老年医学, 2025, 39(4): 382-386. doi: 10.3969/j.issn.1003-9198.2025.04.012 [19] More M P, Saha P, Roy S, et al. Underlying metabolic syndrome exacerbates Vibrio vulnificus-induced acute kidney injury via systemic Th17/Treg dysregulation[J]. Am J Physiol Ren Physiol, 2025, 329(5): F627-F644. doi: 10.1152/ajprenal.00073.2025 [20] 何峻, 伍松柏, 吕爱莲, 等. 老年脓毒症合并免疫抑制患者免疫调理治疗策略[J]. 实用医学杂志, 2021, 37(6): 718-721. [21] 樊琪, 董宁, 吴瑶, 等. 脓毒症老年小鼠树突状细胞高尔基体应激和自噬水平变化及其对免疫功能的影响[J]. 中国急救医学, 2023, 43(11): 868-874. doi: 10.3969/j.issn.1002-1949.2023.11.004 [22] Zhang Z, Zou X, Zhang R, et al. Human umbilical cord mesenchymal stem cell-derived exosomal miR-146a-5p reduces microglial-mediated neuroinflammation via suppression of the IRAK1/TRAF6 signaling pathway after ischemic stroke[J]. Aging, 2021, 13(2): 3060-3079. doi: 10.18632/aging.202466 [23] 黄光仙, 王怡洁. 脓毒症免疫抑制机制及免疫治疗的研究进展[J]. 昆明医科大学学报, 2025, 46(5): 1-11. doi: 10.12259/j.issn.2095-610X.S20250501 [24] 蔡涛, 刘莲, 李兰, 等. 保肾排毒汤对脓毒症急性肾损伤大鼠血清中肾素-血管紧张素-醛固酮系统及vWF、SEPCR的调控作用[J]. 中国老年学杂志, 2024, 44(12): 2990-2993. [25] 宋苏沛, 姜莹, 拾晴, 等. 生脉散联合清瘟败毒饮对脓毒症(气阴两虚证)老年患者免疫功能的影响[J]. 中国中医急症, 2025, 34(4): 642-645. doi: 10.3969/j.issn.1004-745X.2025.04.017 [26] Zhou W, Li C, Yun H, et al. Long non-coding RNA EPB41L4A-AS1 as a biomarker of sepsis alleviates inflammatory response by targeting miR-146a-5p[J]. Eur J Med Res, 2025, 30(1): 728. doi: 10.1186/s40001-025-02991-9 [27] Franconi F, Lodde V, Capobianco G, et al. Effects of maternal smoking on inflammation, autophagy/mitophagy, and miRNAs in endothelial cells: Influence of newborn sex[J]. Eur J Pharmacol, 2025, 998: 177648. doi: 10.1016/j.ejphar.2025.177648 [28] Zhang H, Zheng W, Li D, et al. miR-146a-5p promotes chondrocyte apoptosis and inhibits autophagy of osteoarthritis by targeting NUMB[J]. Cartilage, 2021, 13(2_suppl): 1467S-1477S. doi: 10.1177/19476035211023550 -
下载: