神经炎症相关因子在老年脊柱骨折患者术后谵妄发生预测中的价值及预测模型构建

汪海钢; 李艳玲; 李跃兵; 刘璐; 黄余亮; 王新华; 杨世伟

doi:10.16766/j.cnki.issn.1674-4152.004201

神经炎症相关因子在老年脊柱骨折患者术后谵妄发生预测中的价值及预测模型构建

doi: 10.16766/j.cnki.issn.1674-4152.004201

1.
浙江中医药大学附属第二医院麻醉科，浙江杭州 310012
2.
浙江中医药大学附属第二医院推拿科，浙江杭州 310012
3.
嘉兴市中医院麻醉科, 浙江嘉兴 31400

基金项目:

浙江省中医药科技计划项目 2024ZR173

详细信息

通讯作者:
李艳玲，E-mail：13588453050@163.com

中图分类号: R683.2
计量
- 文章访问数: 1
- HTML全文浏览量: 0
- PDF下载量: 0
- 被引次数: 0
出版历程
- 收稿日期: 2025-02-11

The value of neuroinflammation-related factors in predicting the occurrence of postoperative delirium in elderly patients with spinal fractures and the construction of predictive models

1.
Department of Anesthesiology, the Second Affiliated Hospital of Zhejiang Chinese Medicine University, Hangzhou, Zhejiang 310012, China

摘要

摘要: 目的为探究神经炎症相关因子与老年脊柱骨折患者谵妄发生的关系，纳入老年脊柱骨折患者测量其基线神经炎症相关因子水平，并据此构建风险预测模型。方法纳入2022年1月—2024年6月浙江中医药大学附属第二医院426例老年脊柱骨折手术患者，根据7∶3的比例将患者分为训练集(298例)与验证集(128例)。根据患者术后谵妄发生情况分组(发生组和未发生组)，探讨老年脊柱骨折术后谵妄发生的影响因素以构建预测模型，并分析其预测效能。结果多因素logistic回归分析显示，年龄、手术时间、白介素-6(IL-6)、IL-1β、C-反应蛋白(CRP)、中枢神经特异蛋白(S100β)均为老年脊柱骨折手术患者术后谵妄发生的独立影响因素(P < 0.05)；ROC曲线分析显示，基于各项影响因素构建的模型预测老年脊柱骨折手术患者术后谵妄发生的曲线下面积为0.881，灵敏度为85.71%、特异度为79.75%，约登指数为0.655。验证集验证显示，该预测模型的曲线下面积为0.858(95% CI: 0.782~0.916)，灵敏度为80.65%，特异度为76.47%，约登指数为0.571；决策曲线分析显示训练集和验证集的阈值概率分别为1%和2%，提示该预测模型具有较高的净收益。结论年龄、手术时间、IL-6、IL-1β、CRP、S100β均为老年脊柱骨折患者谵妄发生的影响因素，据此构建的预测模型具有较高的预测效能。
- 老年脊柱骨折 /
- 术后谵妄 /
- 神经炎症 /
- 风险预测模型 /
- 决策曲线
Abstract: Objective To explore the relationship between neuroinflammation neuroinflammation-related factors and delirium in elderly patients suffering from spinal fractures, this study constituted an investigation into the measurement of baseline neuroinflammation-related factors in elderly patients with spinal fractures. The construction of a risk prediction model was then based on the findings. Methods From January 2022 to June 2024, a total of 426 elderly patients afflicted with spinal fractures were enrolled in the Second Affiliated Hospital of Zhejiang Chinese Medicine University. According to the ratio of 7∶3, the patients were divided into a training set (n=298) and a verification set (n=128). According to the incidence of postoperative delirium, the patients were categorized into two groups (the occurrence group and the non-occurrence group) to explore the influencing factors of delirium. Meanwhile, a prediction model was also constructed. Results Multivariate logistic regression analysis showed that age, operation time, interleukin-6 (IL-6), IL-1β, C-reactive protein (CRP), and central nervous system-specific protein (S100β) were all independent influencing factors for the occurrence of postoperative delirium in elderly patients undergoing spinal fracture surgery (P < 0.05). ROC curve analysis indicated that the area under the curve (AUC) of the model constructed based on the above influencing factors for predicting postoperative delirium in elderly patients undergoing spinal fracture surgery was 0.881, with a sensitivity of 85.71%, a specificity of 79.75%, and a Youden index of 0.655. Validation by the validation set showed that the AUC of this prediction model was 0.858 (95% CI: 0.782-0.916), with a sensitivity of 80.65%, a specificity of 76.47%, and a Youden index of 0.571. Decision curve analysis revealed that the threshold probabilities of the training set and the validation set were 1% and 2% respectively, suggesting that this prediction model had a relatively high net benefit. Conclusion Age, operation time, IL-6, IL-1β, CRP and S100β as the influencing factors of delirium in elderly patients with spinal fracture, and the prediction model based on these factors has high prediction efficiency.
- Senile spinal fracture /
- Postoperative delirium /
- Neuroinflammation /
- Risk prediction model /
- Decision curve
利益冲突 无

HTML全文

图 1 训练集预测模型的ROC曲线

Figure 1. ROC curve of training set prediction model

下载: 全尺寸图片幻灯片

图 2 验证集预测模型的ROC曲线

Figure 2. Prediction and analysis of recurrence risk of validation set

下载: 全尺寸图片幻灯片

图 3 训练集预测模型的决策曲线

Figure 3. Decision curve of training set prediction model

下载: 全尺寸图片幻灯片

图 4 验证集预测模型的决策曲线

Figure 4. Decision curve of validation set prediction model

下载: 全尺寸图片幻灯片

表 1 训练集和验证集老年脊柱骨折手术患者一般资料比较

Table 1. Comparison of general data of patients in training set and verification set

项目	训练集(n=298)	验证集(n=128)	统计量	P值
年龄(x ±s，岁)	68.12±6.50	69.23±7.88	1.402^a	0.162
性别[例(%)]			1.901^b	0.168
男性	182(61.07)	69(53.91)
女性	116(38.93)	59(46.09)
BMI(x ±s)	25.23±2.42	24.96±2.28	1.100^a	0.272
受教育年限[例(%)]			2.101^b	0.147
≤12	201(67.45)	77(60.16)
＞12年	97(32.55)	51(39.84)
基础疾病[例(%)]
高血压	95(31.88)	31(24.22)	2.5234^b	0.112
糖尿病	89(29.87)	33(25.78)	0.731^b	0.393
高脂血症	61(20.47)	29(22.66)	0.257^b	0.612
冠心病	43(14.43)	19(14.84)	0.012^b	0.911
慢性阻塞性肺疾病	24(8.05)	6(4.69)	1.550^b	0.213
美国麻醉医师协会分级[例(%)]			3.206^b	0.201
Ⅰ级	157(52.68)	59(46.09)
Ⅱ级	113(37.92)	50(39.06)
Ⅲ级	28(9.40)	19(14.84)
手术部位[例(%)]			1.419^b	0.234
颈椎	135(45.30)	50(39.06)
腰椎	163(54.70)	78(60.94)
手术时间(x ±s，min)	177.83±23.64	173.76±26.52	1.499^a	0.135
术中低血压[例(%)]	9(3.02)	3(2.34)	0.150^b	0.699
IL-6(x ±s，pg/mL)	23.00±5.72	23.71±4.81	1.317^a	0.189
IL-1β(x ±s，pg/mL)	3.65±1.07	3.58±1.14	0.592^a	0.555
CRP(x ±s，pg/mL)	5.17±1.89	5.04±1.81	0.671^a	0.503
S100β(x ±s，pg/mL)	59.17±15.14	62.14±16.08	1.778^a	0.077
注：^a为t值，^b为χ²值。

下载: 导出CSV

表 2 训练集发生组与未发生组老年脊柱骨折手术患者一般资料比较

Table 2. Comparison of general data of elderly patients undergoing spinal fracture surgery between the occurrence group and the non-occurrence group in the training set

项目	发生组(n=56)	未发生组(n=242)	统计量	P值
年龄(x ±s，岁)	71.82±5.34	66.33±5.50	6.894^a	< 0.001
性别[例(%)]			0.299^b	0.584
男性	36(64.29)	146(60.33)
女性	20(35.71)	96(39.67)
BMI(x ±s)	24.97±2.11	25.55±2.44	1.642^a	0.102
受教育年限[例(%)]			0.497^b	0.481
≤12年	40(71.43)	161(66.53)
＞12年	16(28.57)	81(33.47)
基础疾病[例(%)]
高血压	17(30.36)	78(32.23)	0.074^b	0.786
糖尿病	20(35.71)	66(28.51)	1.126^b	0.289
高脂血症	13(223.21)	48(19.83)	0.319^b	0.572
冠心病	10(17.86)	33(13.64)	0.656^b	0.418
慢性阻塞性肺疾病	6(10.71)	18(7.44)	0.659^b	0.417
美国麻醉医师协会分级[例(%)]			8.641^b	0.013
Ⅰ级	25(44.64)	132(54.55)
Ⅱ级	20(35.71)	93(38.43)
Ⅲ级	11(19.64)	17(7.02)
手术部位[例(%)]			1.170^b	0.275
颈椎	29(51.79)	106(43.80)
腰椎	27(48.21)	136(56.20)
手术时间(x ±s，min)	179.37±19.71	168.22±16.71	3.920^a	< 0.001
术中低血压[例(%)]	5(8.93)	4(1.65)	8.219^b	0.004
IL-6(x ±s，pg/mL)	24.62±1.81	23.18±1.17	5.685^a	< 0.001
IL-1β(x ±s，pg/mL)	3.89±1.11	3.56±0.98	2.048^a	0.044
CRP(x ±s，pg/mL)	5.75±1.72	4.95±1.51	3.206^a	0.002
S100β(x ±s，pg/mL)	65.45±12.35	58.91±13.26	3.521^a	0.001
注：^a为t值，^b为χ²值。

下载: 导出CSV

表 3 老年脊柱骨折手术患者术后谵妄风险因素的logistic回归分析

Table 3. Logistic regression analysis of postoperative delirium risk factors in elderly patients undergoing spinal fracture surgery

变量	B	SE	Waldχ²	P值	OR值	95% CI
年龄	0.171	0.037	21.501	< 0.001	1.186	1.104~1.275
美国麻醉医师协会分级	0.349	0.278	1.577	0.209	1.418	0.822~2.444
手术时间	0.023	0.011	4.083	0.043	1.023	1.001~1.046
术中低血压	1.623	0.883	3.378	0.066	5.067	0.898~28.592
IL-6	0.600	0.144	17.428	< 0.001	1.822	1.375~2.415
IL-1β	0.483	0.198	5.937	0.015	1.621	1.099~2.389
CRP	0.268	0.119	5.017	0.025	1.307	1.034~1.652
S100β	0.057	0.016	13.466	< 0.001	1.059	1.027~1.091
注：赋值方法如下，术后是否发生谵妄(因变量)，是=1，否=0；年龄、手术时间、IL-6、IL-1β、CRP、S100β均以实际值赋值；ASA分级，Ⅰ级=0，Ⅱ级=1，Ⅲ级=2；术中低血压，未发生=0，发生=1。

下载: 导出CSV

表 4 结肠息肉患者术后复发风险因素的预测效能

Table 4. Predictive efficacy of risk factors for postoperative recurrence in patients with colonic polyps

模型	曲线下面积(95% CI)	约登指数	敏感性(%)	特异性(%)
训练集预测模型	0.881(0.838~0.915)	0.655	85.71	79.75
验证集预测模型	0.858(0.782~0.916)	0.571	80.65	76.47

下载: 导出CSV

参考文献(22)

[1]	中国老年医学学会麻醉学分会. 中国老年患者术后谵妄防治专家共识[J]. 国际麻醉学与复苏杂志, 2023, 44(1): 1-27. Chinese Geriatrics Society. Expert consensus on prevention and treatment of postoperative delirium in elderly patients in China[J]. International Journal of Anesthesiology and Resuscitation, 2023, 44(1): 1-27.
[2]	WATNE L O, POLLMANN C T, NEERLAND B E, et al. Cerebrospinal fluid quinolinic acid is strongly associated with delirium and mortality in hip-fracture patients[J]. J Clin Invest, 2023, 133(2): e163472. DOI: 10.1172/JCI163472.
[3]	ZHANG S, TAO X J, DING S, et al. Associations between postoperative cognitive dysfunction, serum Interleukin-6 and postoperative delirium among patients after coronary artery bypass grafting: a mediation analysis[J]. Nurs Crit Care, 2024, 29(6): 1245-1252. doi: 10.1111/nicc.13081
[4]	HU J, ZHANG Y, HUANG C, et al. Interleukin-6 trans-signalling in hippocampal CA1 neurones mediates perioperative neurocognitive disorders in mice[J]. Br J Anaesth, 2022, 129(6): 923-936. doi: 10.1016/j.bja.2022.08.019
[5]	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. GAO Y X, ZHANG J X. Determination of Sample Size in Logistic Regression Analysis[J]. The Journal of Evidence-Based Medicine, 2018, 18(2): 122-124.
[6]	ALDECOA C, BETTELLI G, BILOTTA F, et al. European Society of Anaesthesiology evidence-based and consensus-based guideline on postoperative delirium[J]. Eur J Anaesthesiol, 2017, 34(4): 192-214. doi: 10.1097/EJA.0000000000000594
[7]	RASHID M H, SPARROW N A, ANWAR F, et al. Interleukin-6 mediates delirium-like phenotypes in a murine model of urinary tract infection[J]. J Neuroinflammation, 2021, 18(1): 247. DOI: 10.1186/s12974-021-02304-x.
[8]	SPARROW N A, ANWAR F, COVARRUBIAS A E, et al. IL-6 Inhibition reduces neuronal injury in a murine model of ventilator-induced lung Injury[J]. Am J Respir Cell Mol Biol, 2021, 65(4): 403-412. doi: 10.1165/rcmb.2021-0072OC
[9]	QIAN H, GAO F, WU X, et al. Activation of the CD200/CD200R1 axis attenuates neuroinflammation and improves postoperative cognitive dysfunction via the PI3K/Akt/NF-κB signaling pathway in aged mice[J]. Inflamm Res, 2023, 72(12): 2127-2144. doi: 10.1007/s00011-023-01804-1
[10]	BARRETO CHANG O L, MAZE M. Defining the role of Interleukin-6 for the development of perioperative neurocognitive disorders: evidence from clinical and preclinical studies[J]. Front Aging Neurosci, 2022, 14: 1097606. DOI: 10.3389/fnagi.2022.1097606.
[11]	XIAO M Z, LIU C X, ZHOU L G, et al. Postoperative delirium, neuroinflammation, and influencing factors of postoperative delirium: a review[J]. Medicine (Baltimore), 2023, 102(8): e32991. DOI: 10.1097/MD.0000000000032991.
[12]	LEE H J, HWANG D S, WANG S K, et al. Early assessment of delirium in elderly patients after hip surgery[J]. Psychiatry Investig, 2011, 8(4): 340-347. doi: 10.4306/pi.2011.8.4.340
[13]	SUN Y, PENG H P, WU T T. Postoperative C-reactive protein predicts postoperative delirium in colorectal cancer following surgery[J]. Clin Interv Aging, 2023, 18: 559-570. doi: 10.2147/CIA.S387117
[14]	DODSWORTH B T, REEVE K, FALCO L, et al. Development and validation of an international preoperative risk assessment model for postoperative delirium[J]. Age Ageing, 2023, 52(6): afad086. DOI: 10.1093/ageing/afad086.
[15]	MA X, MEI X, TANG T, et al. Preoperative homocysteine modifies the association between postoperative C-reactive protein and postoperative delirium[J]. Front Aging Neurosci, 2022, 14: 963421. DOI: 10.3389/fnagi.2022.963421.
[16]	RUHNAU J, MULLER J, NOWAK S, et al. Serum biomarkers of a pro-Neuroinflammatory state may define the pre-operative risk for postoperative delirium in spine surgery[J]. Int J Mol Sci, 2023, 24(12): 10335. DOI: 10.3390/ijms241210335.
[17]	ZHOU Y, MA Y, YU C, et al. Detection analysis of perioperative plasma and CSF reveals risk biomarkers of postoperative delirium of parkinson's disease patients undergoing deep brain stimulation of the subthalamic nuclei[J]. Clin Interv Aging, 2022, 17(1): 1739-1749.
[18]	TAYLOR J, PARKER M, CASEY C P, et al. Postoperative delirium and changes in the blood-brain barrier, neuroinflammation, and cerebrospinal fluid lactate: a prospective cohort study[J]. Br J Anaesth, 2022, 129(2): 219-230. doi: 10.1016/j.bja.2022.01.005
[19]	KURUP M T, SARKAR S, VERMA R, et al. Comparative evaluation of intraoperative dexmedetomidine versus lidocaine for reducing postoperative cognitive decline in the elderly: a prospective randomized controlled trial[J]. Anaesthesiol Intensive Ther, 2023, 55(5): 349-357. doi: 10.5114/ait.2023.134251
[20]	TANG C, LI Y, LAI Y. Intraoperative dexmedetomidine for prevention of postoperative cognitive dysfunction and delirium in elderly patients with lobectomy: a propensity score-matched, retrospective study[J]. Int J Gen Med, 2024, 17: 2673-2680. doi: 10.2147/IJGM.S456762
[21]	葛宇, 于海洋, 梁成民, 等. 椎体后凸成形术治疗骨质疏松椎体压缩性骨折影响后凸畸形改善的相关因素[J]. 中华全科医学, 2023, 21(11): 1827-1829, 1927. doi: 10.16766/j.cnki.issn.1674-4152.003234 GE Y, YU H Y, LIANG C M, et al. The related factors of kyphoplasty for osteoporotic vertebral compression fractures affecting the improvement of kyphosis deformity[J]. Chinese Journal of General Practice, 2023, 21(11): 1827-1829, 1927. doi: 10.16766/j.cnki.issn.1674-4152.003234
[22]	王金伙, 陶强, 康逸群, 等. 目标导向液体疗法对老年全髋关节置换手术患者围术期免疫炎性反应及早期认知功能的影响[J]. 西部医学, 2022, 34(8): 1147-1151, 1156. WANG J H, TAO Q, KANG Y Q, et al. Effects of target-directed fluid therapy on perioperative immune inflammatory response and early cognitive function in elderly patients undergoing total hip arthroplasty[J]. Med J West Chin, 2022, 34(8): 1147-1151, 1156.