Identification of factors related to thrombocytopenia in patients with sepsis based on Bayesian network model

doi:10.3877/cma.j.issn.1674-0785.2022.11.015

Abstract

Abstract:

Objective

To identify the factors related to thrombocytopenia in patients with sepsis, and to construct a Bayesian network model of thrombocytopenia in those patients to explore the network relationship between thrombocytopenia and its related factors and to reflect the extent of influence of various factors on thrombocytopenia in patients with sepsis by network model reasoning.

Methods

Ninety-eight patients with sepsis admitted to the intensive care unit (ICU) of Shanghai Sixth People's Hospital from January 2019 to December 2020 were selected. Among them, there were 53 males and 45 females, with an age of (59.37±4.28) years. The occurrence of thrombocytopenia in all patients during ICU stay was statistically analyzed and the patients were divided into either an occurrence group or a non-occurrence group according to the occurrence of thrombocytopenia or not. Baseline data questionnaire was designed to collect baseline data of the two groups. Logistic regression analysis was used to screen the influencing factors of thrombocytopenia in patients with sepsis, and the relationship between each influencing factor and thrombocytopenia in patients with sepsis was analyzed using the Bayesian network model.

Results

Among the 98 patients with sepsis, 33 had thrombocytopenia (33.67%). Fungal infection, septic shock, interleukin-6 (IL-6) level, and maximum amplitude of thromboelastogram (MA) differed significantly between the two groups (P＜0.05 for all), but there was no statistical significant difference in other data between the two groups (P＞0.05 for all). Logistic regression analysis demonstrated that fungal infection (odds ratio [OR]=7.185, 95% confidence interval [CI]:1.168-44.184), septic shock (OR=4.024, 95%CI:1.081-14.983), and overexpression of serum IL-6 (OR=9.360, 95%CI:2.283-38.379) were risk factors for thrombocytopenia in patients with sepsis (P＜0.05 for all), while elevation of MA value (OR=0.814, 95%CI:0.734-0.902) was a protective factor for thrombocytopenia (P＜0.05). Directed acyclic Bayesian network structure graph showed that fungal infection, septic shock, IL-6, and MA value were associated with thrombocytopenia in patients with sepsis.

Conclusion

The occurrence of thrombocytopenia in patients with sepsis may be related to fungal infection, septic shock, high IL-6 level, and low MA value

Key words: Sepsis, Thrombocytopenia, Bayesian network, Interleukin 6, Platelet count

Yiqing Tong, Jianming Zhang, Xingxing He, Yimu Fu, Gang Zhao, Qiming Feng. Identification of factors related to thrombocytopenia in patients with sepsis based on Bayesian network model[J]. Chinese Journal of Clinicians(Electronic Edition), 2022, 16(11): 1120-1125.

/ / Recommend

Add to citation manager EndNote|Ris|BibTeX

URL: https://zhlcyszz.cma-cmc.com.cn/EN/10.3877/cma.j.issn.1674-0785.2022.11.015

https://zhlcyszz.cma-cmc.com.cn/EN/Y2022/V16/I11/1120

Figures/Tables 6

项目	发生组（n=33）	未发生组（n=65）	χ²/t值	P值
性别［例（%）］			0.004	0.948
男	18（54.6）	35（53.9）
女	15（45.5）	30（46.2）
年龄［例（%）］			0.046	0.831
≥60岁	17（51.5）	32（49.2）
＜60岁	16（48.5）	33（50.8）
感染部位［例（%）］
肺部感染	21（63.6）	43（66.2）	0.063	0.969
腹腔感染	7（21.2）	13（20.0）
泌尿系统感	5（15.2）	9（13.9）
真菌性感染［例（%）］			16.867	＜0.001
是	13（39.4）	4（6.2）
否	20（60.6）	61（93.8）
SOFA（ $x ¯$ ±s，分）	3.39±0.92	3.37±0.86	0.106	0.916
APACHEⅡ（ $x ¯$ ±s，分）	21.38±5.32	22.07±6.13	0.550	0.584
感染性休克［例（%）］			7.685	0.006
有	18（54.6）	17（26.2）
无	15（45.5）	48（73.9）
使用抗血小板药物［例（%）］			0.722	0.396
是	7（21.2）	19（29.2）
否	26（78.8）	46（70.8）
使用抗凝药物［例（%）］			0.520	0.471
是	6（18.2）	16（24.6）
否	27（81.8）	49（75.4）
CRRT治疗［例（%）］			0.008	0.929
是	13（39.4）	25（38.5）
否	20（60.6）	40（61.5）

项目	发生组（n=33）	未发生组（n=65）	χ²/t值	P值
性别［例（%）］			0.004	0.948
男	18（54.6）	35（53.9）
女	15（45.5）	30（46.2）
年龄［例（%）］			0.046	0.831
≥60岁	17（51.5）	32（49.2）
＜60岁	16（48.5）	33（50.8）
感染部位［例（%）］
肺部感染	21（63.6）	43（66.2）	0.063	0.969
腹腔感染	7（21.2）	13（20.0）
泌尿系统感	5（15.2）	9（13.9）
真菌性感染［例（%）］			16.867	＜0.001
是	13（39.4）	4（6.2）
否	20（60.6）	61（93.8）
SOFA（ $x ¯$ ±s，分）	3.39±0.92	3.37±0.86	0.106	0.916
APACHEⅡ（ $x ¯$ ±s，分）	21.38±5.32	22.07±6.13	0.550	0.584
感染性休克［例（%）］			7.685	0.006
有	18（54.6）	17（26.2）
无	15（45.5）	48（73.9）
使用抗血小板药物［例（%）］			0.722	0.396
是	7（21.2）	19（29.2）
否	26（78.8）	46（70.8）
使用抗凝药物［例（%）］			0.520	0.471
是	6（18.2）	16（24.6）
否	27（81.8）	49（75.4）
CRRT治疗［例（%）］			0.008	0.929
是	13（39.4）	25（38.5）
否	20（60.6）	40（61.5）

项目	发生组（n=33）	未发生组（n=65）	t/χ²值	P值
IL-6（ng/L）	121.48±18.45	105.26±17.23	4.003	＜0.001
PCT（μg/L）	1.19±0.31	1.07±0.30	1.851	0.067
CRP（mg/L）	123.45±21.73	125.17±22.68	0.360	0.720
APTT（s）	29.32±4.27	28.69±4.18	0.700	0.486
PT（s）	12.47±2.18	11.98±2.40	0.984	0.327
FIB（g/L）	2.95±0.89	2.73±0.85	1.192	0.236
D-D（mg/L）	4.65±1.03	4.39±1.10	1.129	0.262
WBC（×10⁹/L）	13.60±2.17	12.95±2.60	1.234	0.220
MA值（min）	57.30±8.23	66.95±6.79	6.183	＜0.001
K值（min）	1.32±0.45	1.28±0.34	0.492	0.624
R值（min）	5.10±1.68	4.87±1.46	0.700	0.486

指标	B	标准误	WaId χ²值	P值	OR值（95%CI）
常量	5.224	3.040	2.954	0.086	-
真菌性感染	1.972	0.927	4.528	0.033	7.185（1.168~44.184）
感染性休克	1.392	0.671	4.309	0.038	4.024（1.081~14.983）
IL-6	2.236	0.720	9.650	0.002	9.360（2.283~38.379）
PCT	1.933	1.015	3.626	0.057	6.912（0.945~50.548）
MA值	-0.206	0.053	15.271	＜0.001	0.814（0.734~0.902）

指标	AUC	95%CI	最佳截断值	特异性	敏感度
IL-6	0.743	0.636~0.849	113.590 μg/L	0.708	0.758
MA值	0.825	0.735~0.16	62.435 min	0.800	0.727

自变量	变量说明	赋值
真菌性感染	分类变量	“1”=是，“0”=否
感染性休克	分类变量	“1”=是，“0”=否
IL-6	分类变量	“1”=≥113.590 ng/L，“0”=＜113.590 ng/L
MA值	分类变量	“1”=≥62.435 min，“0”=＜62.435 min

父节点				血小板减少症
真菌性感染	感染性休克	IL-6≥113.590 ng/L	MA值＜62.435 min	发生	未发生
是	是	是	是	0.667	0.333
是	是	是	否	1.000	0.000
是	是	否	是	0.333	0.667
是	否	是	是	0.500	0.500
否	是	是	是	0.750	0.250
否	否	是	是	0.375	0.625
否	是	否	是	0.071	0.929
否	是	是	否	1.000	0.000
是	是	否	否	0.765	0.235
是	否	否	是	0.333	0.667
是	否	是	否	1.000	0.000
是	否	否	是	0.333	0.667
否	否	否	否	0.400	0.600
否	否	否	是	0.028	0.972
否	否	是	否	0.500	0.500
否	是	否	否	0.800	0.200

References 18

1	Li MF, Li XL, Fan KL, et al. Platelet desialylation is a novel mechanism and a therapeutic target in thrombocytopenia during sepsis: an open-label, multicenter, randomized controlled trial [J]. J Hematol Oncol, 2017, 10(1):104.
2	Koyama K, Katayama S, Muronoi T, et al. Time course of immature platelet count and its relation to thrombocytopenia and mortality in patients with sepsis [J]. PLoS One, 2018, 13(1):e0192064.
3	周南君, 汤展宏. 血小板减少症对脓毒症患者预后的影响:系统回顾和Meta分析 [J].中国实用内科杂志, 2021, 41(3):238-242,256.
4	Bedet A, Razazi K, Boissier F, et al. Mechanisms of thrombocytopenia during septic shock: a multiplex cluster analysis of endogenous sepsis mediators [J]. Shock, 2018, 49(6):641-648.
5	江玲芝, 汪明珊, 王金柱, 等. 脓毒症合并急性肾损伤患者继发血小板减少症的危险因素分析 [J].中华临床感染病杂志, 2017, 10(3):187-193.
6	中国医师协会急诊医师分会, 中国研究型医院学会休克与脓毒症专业委员会. 中国脓毒症/脓毒性休克急诊治疗指南(2018) [J].中国急救医学, 2018, 38(9):741-756.
7	中华医学会血液学分会止血与血栓学组. 成人原发免疫性血小板减少症诊断与治疗中国专家共识(2016年版) [J].中华血液学杂志, 2016, 37(2):89-93.
8	Guirgis FW, Puskarich MA, Smotherman C, et al. Development of a simple sequential organ failure assessment score for risk assessment of emergency department patients with sepsis [J]. J Intensive Care Med, 2020, 35(3):270-278.
9	谢铎文, 潘景业. 急性生理学与慢性健康状况评分系统:1978-2010 [J].中国中西医结合急救杂志, 2010, 17(6):378-381.
10	廖秋霞, 席修明. 脓毒症患者发生血小板减少的危险因素及其对临床结局的影响 [J].内科急危重症杂志, 2020, 26(4):310-314.
11	Zhou X, Tang G. Some doubts on the meta-analysis of the clinical significance of thrombocytopenia complicating sepsis [J]. J Infect, 2019, 79(3):277-287.
12	陈静, 龚宗炼, 葛颖, 等. 脓毒症患者感染病原学特点及血液指标的监测分析 [J].中华医院感染学杂志, 2019, 29(17):2571-2574.
13	Harbi GA, Chaari A. Platelets parameters in septic shock: clinical usefulness and prognostic value [J]. Blood Coagul Fibrinolysis, 2020, 31(7):421-425.
14	Fogagnolo A, Taccone FS, Campo G, et al. Impaired platelet reactivity in patients with septic shock: a proof-of-concept study [J]. Platelets, 2020, 31(5):652-660.
15	谢然, 陈礼文, 张浩, 等. IL-6评估发热伴血小板减少综合征患者病情及预后研究 [J].安徽医科大学学报, 2021, 56(9):1475-1479.
16	李健, 许华, 王兵, 等. 白细胞介素-18在脓毒症血小板减少症中的表达及临床意义 [J].实用医学杂志, 2018, 34(02):205-208.
17	崔利丹, 金志鹏, 王琪, 等. 血栓弹力图在脓毒症及脓毒性休克患儿并弥散性血管内凝血病情评估中的价值 [J].中华实用儿科临床杂志, 2017, 32(18):1398-1401.
18	Ninan KF, Iyadurai R, Varghese JK, et al. Thromboelastograph: a prognostic marker in sepsis with organ dysfunction without overt bleeding [J]. J Crit Care, 2021, 65:177-183.

[1]	Yuanyuan Han, Sabeiti Reziya, Zhijie Mao, Aierken Mufunayi, Chen Lu, Xiaohong Sang, Mulati Aerman, Li Zhang. Effect of combined blood purification on serum inflammatory cytokines levels and clinical prognosis in patients with sepsis [J]. Chinese Journal of Critical Care Medicine(Electronic Edition), 2023, 16(04): 272-278.
[2]	Jianbiao Meng, Geng Zhang, Yanna Jiao. Exploratory study on early intestinal microecological changes in patients with sepsis-induced myocardial dysfunction [J]. Chinese Journal of Critical Care Medicine(Electronic Edition), 2023, 16(04): 279-285.
[3]	Yu Chen, Fang Feng, Lu Zhang, Jian Liu. Bioinformatics analysis to screen key pathogenic genes in septic cardiomyopathy [J]. Chinese Journal of Critical Care Medicine(Electronic Edition), 2023, 16(04): 286-291.
[4]	Xiaoyan Zhang, Dongqiong Xiao, Hu Gao, Lin Chen, Fajuan Tang, Xihong Li. The protective effect and mechanism of overexpression of transcription factor 12 on the cerebral cortex of rats with sepsis-associated encephalopathy [J]. Chinese Journal of Obstetrics & Gynecology and Pediatrics(Electronic Edition), 2023, 19(05): 540-549.
[5]	Xu Wei, Ge Zhang, Jinlin Wu. Monitoring and treatment of neonatal sepsis-induced coagulopathy [J]. Chinese Journal of Obstetrics & Gynecology and Pediatrics(Electronic Edition), 2023, 19(04): 379-386.
[6]	Mei Yang, Chun Zhou, Aihong Zhao, Qin Wang. Construction of a risk nomograph model for plastic bronchitis caused by refractory mycoplasma pneumoniae pneumonia in children [J]. Chinese Journal of Experimental and Clinical Infectious Diseases(Electronic Edition), 2023, 17(04): 274-281.
[7]	Xiaodan Wang, Yuan Wang, Xiangyu Cui, Xiaolei Ren. Analysis of pathogenic bacteria resistance of drug and high risk factors of death in urogenic sepsis after endoscopic surgery for upper urinary tract stones [J]. Chinese Journal of Endourology(Electronic Edition), 2023, 17(06): 611-615.
[8]	Xuecheng Wu, Yuanwei Li, Wuxiong Yuan, Jiansong Wang, Yongzhong Shi, Qiang Lu, Zhuo Li, Jia Chen, Zhe Liu, Yili Teng, Zhiyong Gao. Diagnostic value of inflammatory mediators profile combined with procalcitonin in urogenic sepsis [J]. Chinese Journal of Endourology(Electronic Edition), 2023, 17(05): 476-480.
[9]	Shangwen Dou, Huan Deng, Bangfeng Liu, Gaoyuanzhi Yue, Huacai Zhu, Yongda Liu. Role of preoperative urine culture re-examination in the prediction of infectious complications related to mini-percutaneous nephrolithotomy [J]. Chinese Journal of Endourology(Electronic Edition), 2023, 17(04): 361-366.
[10]	Wenhui Liu, Tao Wu, Xi Zhang. Advances in mesenchymal stem cells combined with thrombopoietin receptor agonists in platelet recovery after allogeneic hematopoietic stem cell transplantation [J]. Chinese Journal of Cell and Stem Cell(Electronic Edition), 2023, 13(04): 242-246.
[11]	Ruanxin Miao, Xi Qiao. Role of Toll-like receptor in sepsis-induced acute kidney injury [J]. Chinese Journal of Kidney Disease Investigation(Electronic Edition), 2023, 12(04): 210-214.
[12]	Shiming Li, Wei Huang, Ling Liu. Mechanism of high mobility group box protein 1 mediated sepsis-induced coagulopathy and treatment progress [J]. Chinese Journal of Critical Care & Intensive Care Medicine(Electronic Edition), 2023, 09(03): 269-273.
[13]	Chao Gao, Jie Chao, Haibo Qiu. Emerging roles of transcription factor T-bet of Th17 in immune imbalance of sepsis [J]. Chinese Journal of Critical Care & Intensive Care Medicine(Electronic Edition), 2023, 09(03): 280-285.
[14]	Rui Tan, Jing Wang, Jiangquan Yu, Ruiqiang Zheng. Progress in understanding of role of high density lipoprotein, apolipoprotein A-I, and serum amyloid A in sepsis [J]. Chinese Journal of Clinicians(Electronic Edition), 2023, 17(06): 749-753.
[15]	Xing Cai, Ruiqiang Zheng. Progress in application of heparin-binding protein in sepsis [J]. Chinese Journal of Clinicians(Electronic Edition), 2023, 17(04): 487-490.

Please choose a citation manager

Content to export