重症急性胰腺炎继发持续性炎症-免疫抑制-分解代谢综合征的影响因素及预测模型构建

李婵; 梁志海; 唐国都

doi:10.3969/j.issn.1001-5256.2023.06.019

重症急性胰腺炎继发持续性炎症-免疫抑制-分解代谢综合征的影响因素及预测模型构建

DOI: 10.3969/j.issn.1001-5256.2023.06.019

广西医科大学第一附属医院消化内科，南宁 530021

基金项目:

国家自然科学基金 (81970558)

伦理学声明：本研究方案经由广西医科大学第一附属医院医学伦理委员会审批，批号：2023-E107-01。

利益冲突声明：本文不存在任何利益冲突。

作者贡献声明：李婵负责课题设计，收集数据，资料分析，撰写论文；梁志海、唐国都负责拟定写作思路，指导撰写文章并最后定稿。

详细信息

通信作者:
唐国都，tguodu02@126.com (ORCID：0000-0002-2849-3244)

计量
- 文章访问数: 206
- HTML全文浏览量: 69
- PDF下载量: 28
- 被引次数: 0
出版历程
- 收稿日期: 2022-10-24
- 录用日期: 2022-11-28
- 出版日期: 2023-06-20

Influencing factors for persistent inflammation, immunosuppression, and catabolism syndrome in patients with severe acute pancreatitis and establishment of a predictive model

Department of Gastroenterology, The First Affiliated Hospital of Guangxi Medical University, Nanning 530021, China

Research funding:

National Natural Science Foundation of China (81970558)

More Information

Corresponding author: TANG Guodu, tguodu02@126.com (ORCID：0000-0002-2849-3244)

摘要

摘要: 目的探讨重症急性胰腺炎(SAP)继发持续性炎症-免疫抑制-分解代谢综合征(PICS)的影响因素，并构建预测模型。方法回顾性分析2012年5月—2022年5月广西医科大学第一附属医院163例因SAP收入重症医学科、急诊重症监护室患者的临床资料。依据PICS诊断标准分成2组：PICS组(65例，SAP发生PICS患者)和非PICS组(98例，SAP未发生PICS患者)。符合正态分布的计量资料2组间比较采用成组t检验；不符合正态分布的计量资料2组间比较采用Mann-Whitney U秩和检验。计数资料2组间比较采用χ²检验或Fisher确切概率法。计算方差膨胀因子、相关系数矩阵热图评估变量间多重共线性，采用Lasso回归及多因素Logistic回归筛选出独立危险因素，构建列线图预测模型。采用受试者工作特征曲线、校准曲线及Hosmer-Lemeshow拟合优度检验对模型进行内部验证；采用临床决策曲线评估模型的临床实用性。结果单因素分析结果显示，平均动脉压、血红蛋白、红细胞压积(HCT)、中性粒细胞与淋巴细胞比值、血小板与淋巴细胞比值(PLR)、尿素、肌酐、格拉斯哥昏迷评分(GCS)、APACHE Ⅱ、SOFA、机械通气、急性呼吸窘迫综合征、急性肾损伤(AKI)、急性肝损伤、低血容量性休克、脓毒症、腹腔高压、腹腔出血、多器官功能障碍综合征在PICS组和非PICS组间比较差异均有统计学意义(P值均＜0.05)。Lasso回归筛选的预测变量包括PLR、HCT、APACHE Ⅱ、SOFA、机械通气、AKI、低血容量性休克、腹腔高压。多因素Logistic回归显示，PLR、机械通气、AKI、低血容量性休克是SAP发生PICS的独立危险因素(OR分别为1.006、4.324、3.432、6.910，P值均＜0.05)。将上述因素进行模型拟合，经bootstrap内部验证列线图模型曲线下面积为0.874(95%CI：0.822~0.925)，校准曲线接近参考曲线，Hosmer-Lemeshow拟合优度检验表明该模型具有良好的拟合度(χ²=8.895，P=0.351)。临床决策曲线分析显示预测模型具有良好的临床实用性。结论 PLR、机械通气、AKI、低血容量性休克是SAP继发PICS的独立危险因素，构建的列线图模型具有良好的区分度、校准度和临床实用性。
- 胰腺炎 /
- 持续性炎症-免疫抑制-分解代谢综合征 /
- 危险因素
Abstract: Objective To investigate the influencing factors for persistent inflammation, immunosuppression, and catabolism syndrome (PICS) in patients with severe acute pancreatitis(SAP), and to establish a predictive model. Methods A retrospective analysis was performed for the clinical data of 163 patients who were admitted to the intensive care unit and the emergency intensive care unit due to SAP in The First Affiliated Hospital of Guangxi Medical University from May 2012 to May 2022, and according to the diagnostic criteria for PICS, these patients were divided into PICS group (65 SAP patients with PICS) and non-PICS group (98 SAP patients without PICS). The independent-samples t test was used for comparison of normally distributed continuous data between two groups, and the Mann-Whitney U rank sum test was used for comparison of non-normally distributed continuous data between two groups; the chi-square test or the Fisher's exact test was used for comparison of categorical data between two groups. Variance inflation factor and correlation matrix heatmap were used to evaluate multicollinearity between variables, and Lasso regression and multivariate logistic regression were used to identify independent risk factors and establish a nomogram predictive model. The receiver operating characteristic (ROC) curve, the calibration curve, and the Hosmer-Lemeshow goodness-of-fit test were used for the internal validation of the model, and the decision curve was used to evaluate the clinical practicability of the model. Results The univariate analysis showed that there were significant differences between the PICS group and the non-PICS group in mean arterial pressure, hemoglobin, hematocrit (HCT), neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio (PLR), blood urea nitrogen, creatinine, Glasgow coma score, Acute Physiology and Chronic Health Evaluation Ⅱ (APACHE Ⅱ) score, Sequential Organ Failure Assessment (SOFA) score, mechanical ventilation, acute respiratory distress syndrome, acute kidney injury (AKI), acute liver injury, hypovolemic shock, sepsis, intra-abdominal hypertension, intra-abdominal hemorrhage, and multiple organ dysfunction syndrome (all P < 0.05). The Lasso regression analysis showed that related predictive variables included PLR, HCT, APACHE Ⅱ, SOFA, mechanical ventilation, AKI, hypovolemic shock, and intra-abdominal hypertension, and the multivariate logistic regression analysis showed that PLR (odds ratio [OR]=1.006, P < 0.05), mechanical ventilation (OR=4.324, P < 0.05), AKI (OR=3.432, P < 0.05), and hypovolemic shock (OR= 6.910, P < 0.05) were independent risk factors for PICS in patients with SAP. Model fitting was performed for the above factors, and bootstrap internal validation showed that the nomogram model had an area under the ROC curve of 0.874 (95% confidence interval: 0.822-0.925); the calibration curve of the model was close to the reference curve, and the Hosmer-Lemeshow goodness-of-fit test showed that the model was well fitted (χ²=8.895, P=0.351). The decision curve analysis showed that the predictive model had good clinical practicability. Conclusion PLR, mechanical ventilation, AKI, and hypovolemic shock are independent risk factors for PICS in patients with SAP, and the nomogram model established has good discriminatory ability, calibration, and clinical practicability.
- Pancreatitis /
- Persistent Inflammation, Immunosuppression and Catabolism Syndrome /
- Risk Factors

HTML全文

图 1 多重共线性检验

Figure 1. Test for multicollinearity

下载: 全尺寸图片幻灯片

图 2 Lasso回归进行临床特征筛选

注：a，19个临床特征的系数曲线；b，Lasso回归10倍交叉验证选择临床特征。

Figure 2. Selection of potential predictors of PICS in patients with SAP by the Lasso regression

下载: 全尺寸图片幻灯片

图 3 SAP患者发生PICS风险预测模型列线图

Figure 3. Nomogram of the risk predictive model for PICS in SAP patients

下载: 全尺寸图片幻灯片

图 4 SAP患者发生PICS风险预测模型的ROC曲线

Figure 4. ROC curve of the risk predictive model for PICS in SAP patients

下载: 全尺寸图片幻灯片

图 5 SAP患者发生PICS风险预测模型的校准曲线

Figure 5. Calibration curve of the risk predictive model for PICS in SAP patients

下载: 全尺寸图片幻灯片

图 6 SAP患者发生PICS风险预测模型的决策曲线

Figure 6. Decision curve of the risk predictive model for PICS in SAP patients

下载: 全尺寸图片幻灯片

表 1 2组患者一般资料比较

Table 1. Comparison of general data between the two groups

项目	PICS组(n=65)	非PICS组(n=98)	统计值	P值
年龄(岁)	48.00(39.50~59.00)	47.50(37.00~59.25)	Z=-0.005	0.996
性别[例(%)]			χ²=0.916	0.339
男	52(80.0)	72(73.5)
女	13(20.0)	26(26.5)
BMI(kg/m²)	23.94(22.00~27.55)	24.98(22.49~27.76)	Z=-0.668	0.504
既往史[例(%)]
高血压	14(34.7)	34(20.9)	χ²=3.255	0.071
糖尿病	8(12.3)	15(15.3)	χ²=0.290	0.590
腹部疾病	27(41.5)	52(53.1)	χ²=1.897	0.168
SAP病因[例(%)]
胆源性	5(7.7)	11(1.1)	χ²=0.551	0.458
高脂血症性	17(26.2)	34(34.7)	χ²=1.326	0.250
酒精性	12(18.5)	18(18.4)	χ²＜0.001	0.988
创伤性	2(3.1)	3(3.1)	χ²＜0.001	0.995
病因不明	29(44.6)	32(32.7)	χ²=2.388	0.122
体温(℃)	37.60(37.00~38.00)	37.50(36.78~38.10)	Z=-0.553	0.580
心率(次/min)	119.43±23.02	112.69±21.48	t=1.881	0.062
呼吸(次/min)	24.00(20.00~33.00)	26.00(21.50~32.00)	Z=-1.064	0.287
平均动脉压(mmHg)	98.33(80.67~111.67)	104.50(91.59~113.08)	Z=-2.207	0.027
预后
开放饮食时间(d)	8.00(3.50~13.50)	7.00(3.00~11.00)	Z=-1.288	0.198
病死率[例(%)]	26(40.0)	16(16.3)	χ²=11.450	0.001
住院时间(d)	30.00(22.00~51.00)	22.00(17.00~27.25)	Z=-4.666	＜0.001
总费用(万元)	22.23(12.10~39.31)	8.18(6.06~11.86)	Z=-6.893	＜0.001

下载: 导出CSV

表 2 2组患者PICS指标比较

Table 2. Comparison of PICS factors between the two groups

指标	PICS组(n=65)	非PICS组(n=98)	Z值	P值
ICU住院时间(d)	22.00(16.00~42.50)	7.00(4.00~11.00)	-8.862	＜0.001
CRP(mg/L)	180.39(120.84~192.00)	190.67(109.04~192.00)	-0.265	0.791
Lym(×10⁹/L)	0.60(0.49~0.75)	1.07(0.72~1.39)	-6.547	＜0.001
前白蛋白(g/L)	83.30(52.25~104.15)	110.90(76.73~155.35)	-3.488	＜0.001
白蛋白(g/L)	28.30(26.10~31.25)	30.80(27.58~34.78)	-3.517	＜0.001

下载: 导出CSV

表 3 2组患者实验室指标和并发症比较

Table 3. Comparison of laboratory data and complication between the two groups

指标	PICS组(n=65)	非PICS组(n=98)	统计值	P值
WBC(×10⁹/L)	14.46(8.94~18.69)	14.72(10.45~20.09)	Z=-0.857	0.391
Hb(g/L)	87.70(75.05~112.70)	106.60(83.68~126.13)	Z=-2.532	0.011
血小板计数(×10⁹/L)	176.00(117.15~237.20)	196.35(141.93~272.40)	Z=-1.366	0.172
Neu(×10⁹/L)	13.22(8.00~17.00)	12.18(8.38~17.45)	Z=-0.095	0.924
NLR	21.69(11.47~30.30)	12.35(7.41~23.13)	Z=-3.383	0.001
PLR	261.85(198.42~455.46)	207.28(118.19~315.79)	Z=-3.210	0.001
HCT	0.26(0.22~0.36)	0.33(0.26~0.39)	Z=-2.649	0.008
PH	7.44(7.36~7.48)	7.43(7.38~7.48)	Z=-0.061	0.951
PO₂(mmHg)	102.00(78.05~136.05)	88.50(71.00~126.03)	Z=-1.161	0.246
PCO₂(mmHg)	34.10(30.40~39.40)	35.55(30.00~39.00)	Z=-0.278	0.781
氧合指数	253.00(198.00~332.90)	234.50(184.25~326.50)	Z=-0.552	0.581
血Na(mmol/L)	138.10(133.95~143.20)	137.00(133.80~140.95)	Z=-0.727	0.467
血K(mmol/L)	4.10(3.80~4.52)	4.00(3.59~4.31)	Z=-1.466	0.143
TBil(μmol/L)	26.30(13.30~47.30)	20.55(12.48~35.03)	Z=-1.469	0.142
DBil(μmol/L)	13.30(6.25~28.20)	9.30(5.00~18.75)	Z=-1.737	0.082
BUN(mmol/L)	10.36(6.32~17.51)	6.43(4.15~10.50)	Z=-3.250	0.001
SCr(μmol/L)	152.00(71.50~293.50)	73.50(55.75~141.00)	Z=-2.915	0.004
血淀粉酶(U/L)	76.00(44.00~306.50)	103.00(53.75~389.75)	Z=-0.935	0.350
GCS评分	15.00(13.00~15.00)	15.00(15.00~15.00)	Z=-2.973	0.003
APACHE Ⅱ	13.00(10.00~16.00)	11.00(7.75~14.00)	Z=-3.173	0.002
SOFA	6.00(4.00~10.00)	4.00(2.00~6.00)	Z=-4.309	＜0.001
并发症[例(%)]
机械通气	54(83.1)	34(34.7)	χ²=17.785	＜0.001
ARDS	54(83.1)	49(50.0)	χ²=17.785	＜0.001
AKI	40(61.5)	23(23.5)	χ²=44.521	＜0.001
急性肝损伤	16(24.6)	12(12.2)	χ²=105.631	＜0.001
低血容量性休克	15(23.1)	3(3.1)	χ²=108.741	＜0.001
脓毒症	37(56.9)	27(27.6)	χ²=50.971	＜0.001
肺部感染	63(96.9)	90(91.8)	χ²=3.053	0.158
腹腔高压	34(52.3)	22(22.4)	χ²=57.715	＜0.001
腹腔出血	12(18.5)	6(6.1)	χ²=118.409	＜0.001
肠瘘	2(3.1)	2(2.0)	χ²=0.175	0.524
MODS	49(75.4)	36(36.7)	χ²=26.749	＜0.001

下载: 导出CSV

表 4 SAP患者发生PICS多因素Logistic回归分析

Table 4. Multivariate logistic regression analysis of PICS in patients with SAP

变量	β值	P值	OR	95%CI
PLR	0.006	＜0.001	1.006	1.003~1.009
HCT	-2.810	0.256	0.060	0.000~7.768
APACHE Ⅱ	0.076	0.124	1.079	0.982~1.195
SOFA	0.053	0.458	1.055	0.913~1.216
机械通气	1.464	0.002	4.324	1.723~11.423
AKI	1.233	0.018	3.432	1.253~9.854
低血容量性休克	1.933	0.013	6.910	1.685~39.068
腹腔高压	0.743	0.107	2.102	0.852~5.245

下载: 导出CSV

参考文献(18)

[1]	VANZANT EL, LOPEZ CM, OZRAZGAT-BASLANTI T, et al. Persistent inflammation, immunosuppression, and catabolism syndrome after severe blunt trauma[J]. J Trauma Acute Care Surg, 2014, 76(1): 21-29. DOI: 10.1097/TA.0b013e3182ab1ab5.
[2]	GENTILE LF, CUENCA AG, EFRON PA, et al. Persistent inflammation and immunosuppression: a common syndrome and new horizon for surgical intensive care[J]. J Trauma Acute Care Surg, 2012, 72(6): 1491-1501. DOI: 10.1097/TA.0b013e318256e000.
[3]	MIRA JC, GENTILE LF, MATHIAS BJ, et al. Sepsis pathophysiology, chronic critical illness, and persistent inflammation-immunosuppression and catabolism syndrome[J]. Crit Care Med, 2017, 45(2): 253-262. DOI: 10.1097/CCM.0000000000002074.
[4]	NAKAMURA K, OGURA K, NAKANO H, et al. C-reactive protein clustering to clarify persistent inflammation, immunosuppression and catabolism syndrome[J]. Intensive Care Med, 2020, 46(3): 437-443. DOI: 10.1007/s00134-019-05851-3.
[5]	EFRON PA, MOHR AM, BIHORAC A, et al. Persistent inflammation, immunosuppression, and catabolism and the development of chronic critical illness after surgery[J]. Surgery, 2018, 164(2): 178-184. DOI: 10.1016/j.surg.2018.04.011.
[6]	XU D, XUN J. Persistentinflammation-immunosuppression-catabolism syndrome in severe acute pancreatitis[J]. J Chin Pract Diagn Ther, 2018, 32(7): 725-728. DOI: 10.13507/j.issn.1674-3474.2018.07.030. 徐冬, 荀江. 重症急性胰腺炎并持续性炎症-免疫抑制-分解代谢综合征研究进展[J]. 中华实用诊断与治疗杂志, 2018, 32(7): 725-728. DOI: 10.13507/j.issn.1674-3474.2018.07.030.
[7]	YANG R, DU LL, ZHANG ZD, et al. Clinical observation of severe patients with persistent inflammation immunosuppression catabolism syndrome[J]. China J Emerg Resuscitation Disaster Med, 2021, 16(8): 919-921. DOI: 10.3969/j.issn.1673-6966.2021.08.021. 杨蓉, 杜玲玲, 章志丹, 等. 重症患者持续性炎症-免疫抑制-分解代谢综合征的临床观察研究[J]. 中国急救复苏与灾害医学杂志, 2021, 16(8): 919-921. DOI: 10.3969/j.issn.1673-6966.2021.08.021.
[8]	KOUPENOVA M, CLANCY L, CORKREY HA, et al. Circulating platelets as mediators of immunity, inflammation, and thrombosis[J]. Circ Res, 2018, 122(2): 337-351. DOI: 10.1161/circresaha.117.310795.
[9]	ZHOU H, MEI X, HE X, et al. Severity stratification and prognostic prediction of patients with acute pancreatitis at early phase: A retrospective study[J]. Medicine (Baltimore), 2019, 98(16): e15275. DOI: 10.1097/md.0000000000015275.
[10]	HAWKINS RB, RAYMOND SL, STORTZ JA, et al. Chronic critical illness and the persistent inflammation, immunosuppression, and catabolism syndrome[J]. Front Immunol, 2018, 9: 1511. DOI: 10.3389/fimmu.2018.01511.
[11]	CHEN MY, CHEN MX, WANG MX, et al. Risk factors for acute kidney injury in severe acute pancreatitis: a Meta-analysis[J]. Chin Gen Pract, 2022, 25(30): 3834-3842. DOI: 10.12114/j.issn.1007-9572.2022.0452. 陈美颖, 陈木欣, 王明欣, 等. 重症急性胰腺炎患者并发急性肾损伤危险因素的Meta分析[J]. 中国全科医学, 2022, 25(30): 3834-3842. DOI: 10.12114/j.issn.1007-9572.2022.0452.
[12]	WANG X, ZHUANG X, WEI R, et al. Protective effects of Acanthopanax vs. Ulinastatin against severe acute pancreatitis-induced brain injury in rats[J]. Int Immunopharmacol, 2015, 24(2): 285-298. DOI: 10.1016/j.intimp.2014.12.020.
[13]	WANG Y, QI WJ, ZENG YW, et al. Mechanism of action of magnolol in the treatment of acute lung injury in a rat model of severe acute pancreatitis[J]. J Clin Hepatol, 2020, 36(12): 2782-2787. DOI: 10.3969/j.issn.1001-5256.2020.12.028. 王燕, 齐文杰, 曾亚薇, 等. 厚朴酚治疗重症急性胰腺炎大鼠模型并发急性肺损伤的作用机制[J]. 临床肝胆病杂志, 2020, 36(12): 2782-2787. DOI: 10.3969/j.issn.1001-5256.2020.12.028.
[14]	WANG X, WANG XK, CEN RF, et al. Influence of Xuebijing injection on clinical effects and cellular immune function of patients with severe acute pancreatitis[J]. Chin J New Drugs Clin Rem, 2015, 34(6): 476-479. DOI: 10.14109/j.cnki.xyylc.2015.06.016. 王雪, 王先坤, 岑荣飞, 等. 血必净注射液对重症急性胰腺炎疗效及患者细胞免疫功能的影响[J]. 中国新药与临床杂志, 2015, 34(6): 476-479. DOI: 10.14109/j.cnki.xyylc.2015.06.016.
[15]	WANG P, LI L, LAN D, et al. Effects of hemofiltration therapy on biochemical indexes, liver and kidney function and inflammatory factors in patients with severe acute pancreatitis complicated with abdominal compartment syndrome[J]. J Clin Exp Med, 2022, 21(4): 384-388. DOI: 10.3969/j.issn.1671-4695.2022.04.014. 王萍, 李乐, 兰东, 等. 血液滤过治疗对重症急性胰腺炎合并腹腔间隔室综合征患者生化指标、肝肾功能及炎症因子水平的影响[J]. 临床和实验医学杂志, 2022, 21(4): 384-388. DOI: 10.3969/j.issn.1671-4695.2022.04.014.
[16]	SU HY, MO ZX, CHEN R, et al. Effect of thymosin α1 on immunity, metabolism and prognosis in elderly patients with sepsis followed by persistent inflammation, immunosuppression, and catabolism syndrome[J]. J Pract Med, 2018, 34(1): 119-123. DOI: 10.3969/j.issn.1006-5725.2018.01.030. 苏和毅, 莫泽珣, 陈蕊, 等. 胸腺肽α1干预治疗对脓毒症后的PICS老年患者免疫、代谢功能及预后的影响[J]. 实用医学杂志, 2018, 34(1): 119-123. DOI: 10.3969/j.issn.1006-5725.2018.01.030.
[17]	LUBBERS T, KOX M, DE HAAN JJ, et al. Continuous administration of enteral lipid- and protein-rich nutrition limits inflammation in a human endotoxemia model[J]. Crit Care Med, 2013, 41(5): 1258-1265. DOI: 10.1097/CCM.0b013e31827c0a17.
[18]	ZHU X, PRIBIS JP, RODRIGUEZ PC, et al. The central role of arginine catabolism in T-cell dysfunction and increased susceptibility to infection after physical injury[J]. Ann Surg, 2014, 259(1): 171-178. DOI: 10.1097/SLA.0b013e31828611f8.