切换至 "中华医学电子期刊资源库"

中华普通外科学文献(电子版) ›› 2024, Vol. 18 ›› Issue (01) : 44 -50. doi: 10.3877/cma.j.issn.1674-0793.2024.01.008

论著

基于LASSO-Cox回归分析的非轻症急性胰腺炎死亡风险列线图预测模型的建立和临床应用效果分析
王晓梅, 刘冰(), 马丽琼, 卢祖静, 苗建军   
  1. 075000 张家口,陆军第八十一集团军医院重症医学科
    075000 张家口,陆军第八十一集团军医院普通外科
  • 收稿日期:2023-05-29 出版日期:2024-02-01
  • 通信作者: 刘冰

Development and analysis of clinical application effect of a prognostic nomogram based on LASSO-Cox regression in patients with non-mild acute pancreatitis

Xiaomei Wang, Bing Liu(), Liqiong Ma, Zujing Lu, Jianjun Miao   

  1. Department of Intensive Care Unit, the 81st Group Army Hospital of PLA, Zhangjiakou 075000, China
    Department of General Surgery, the 81st Group Army Hospital of PLA, Zhangjiakou 075000, China
  • Received:2023-05-29 Published:2024-02-01
  • Corresponding author: Bing Liu
引用本文:

王晓梅, 刘冰, 马丽琼, 卢祖静, 苗建军. 基于LASSO-Cox回归分析的非轻症急性胰腺炎死亡风险列线图预测模型的建立和临床应用效果分析[J/OL]. 中华普通外科学文献(电子版), 2024, 18(01): 44-50.

Xiaomei Wang, Bing Liu, Liqiong Ma, Zujing Lu, Jianjun Miao. Development and analysis of clinical application effect of a prognostic nomogram based on LASSO-Cox regression in patients with non-mild acute pancreatitis[J/OL]. Chinese Archives of General Surgery(Electronic Edition), 2024, 18(01): 44-50.

目的

构建非轻症急性胰腺炎(NMAP)患者死亡风险的列线图预测模型,并验证其预测效能和临床应用价值,同时分析其对于其他评分系统的优势。

方法

纳入大型重症监护数据库MIMIC-Ⅲ中的606例NMAP患者临床资料,按7∶3比例随机分为训练集和验证集。采用LASSO-Cox回归分析构建NMAP患者死亡风险列线图预测模型,并通过受试者工作特征(ROC)曲线、校准曲线以及决策曲线分析(DCA)对列线图模型进行评估。然后比较列线图模型与急性胰腺炎严重程度床边指数(BISAP)、序贯器官衰竭评分(SOFA)、快速序贯器官功能衰竭评分(qSOFA)、急性生理评分Ⅲ(APS Ⅲ)及牛津急性疾病严重程度评分(OASIS)对NMAP患者死亡风险的预测效能。

结果

LASSO-Cox回归分析结果表明,年龄以及入院24 h内的收缩压、红细胞分布宽度(RDW)、血清白蛋白、尿素氮(BUN)、总胆红素和国际标准化比值(INR)是与NMAP患者死亡风险相关的独立危险因素(P<0.05),以此建立的列线图预后模型预测训练集NMAP患者14、30、60、90 d内死亡风险的ROC曲线下面积(AUC)分别为0.76(95% CI:0.67~0.83)、0.79(95% CI:0.72~0.83)、0.83(95% CI:0.77~0.87)、0.83(95% CI:0.78~0.88),验证集对应AUC分别为0.85(95% CI:0.76~0.94)、0.83(95% CI:0.76~0.91)、0.86(95% CI:0.79~0.93)、0.87(95% CI:0.81~0.93)。校准曲线显示训练集和验证集模型预测概率与实际生存率之间均具有较好的一致性,DCA曲线显示列线图模型阈值在0.2~0.8时具有明显的临床净获益。时依ROC曲线显示列线图模型的预测效能优于BISAP、SOFA、qSOFA、APSⅢ及OASIS评分(均P<0.05)。

结论

基于年龄以及入院24 h内的收缩压、RDW、血清白蛋白、BUN、总胆红素和INR建立的列线图预测模型简单便捷,可早期评估NMAP患者的死亡风险,且准确性较高。

Objective

To construct a nomogram prediction model for early prediction of mortality risk in patients with non-mild acute pancreatitis (NMAP) and analyze its clinical application effect and advantages over other scoring systems.

Methods

Clinical data of 606 patients with NMAP from the large medical information mart for intensive careⅢ database (MIMIC-Ⅲ ) were selected. The patients were randomly divided into training and validation sets in a 7∶3 ratio. LASSO-Cox regression analysis was performed to construct a nomogram prediction model for mortality risk in NMAP patients. The model’s performance was assessed through receiver operating characteristic (ROC) curve, calibration curves, and decision curve analysis (DCA). Additionally, the predictive efficacy of the nomogram model was compared with BISAP, SOFA, qSOFA, APS Ⅲ, and OASIS scores.

Results

LASSO-Cox regression analysis identified age, systolic blood pressure within 24 hours of admission, red blood cell distribution width (RDW), serum albumin, blood urea nitrogen (BUN), total bilirubin, and international normalized ratio (INR) as independent risk factors associated with mortality in NMAP patients (P<0.05). A nomogram prognostic model was developed based on these factors. The area under the curve (AUC) for the nomogram model was 0.76 (95% CI: 0.67-0.83), 0.79 (95% CI: 0.72-0.83), 0.83 (95% CI: 0.77-0.87), and 0.83 (95% CI: 0.78-0.88), respectively, for predicting mortality at 14, 30, 60, and 90 days in NMAP patients. The validation set demonstrated AUC values of 0.85 (95% CI: 0.76-0.94), 0.83 (95% CI: 0.76-0.91), 0.86 (95% CI: 0.79-0.93), and 0.87 (95% CI: 0.81-0.93), respectively. Calibration curves indicated excellent agreement between predicted and observed probabilities of mortality in both the training and validation sets. The DCA curve indicated that the nomogram had significantly positive net benefit when the threshold probability ranged from approximately 0.2 to 0.8. The ROC curve revealed superior prediction efficiency of the nomogram model compared to BISAP, SOFA, qSOFA, APSⅢ, and OASIS scores (P<0.05).

Conclusion

The nomogram model, incorporating age, systolic blood pressure within 24 hours of admission, RDW, serum albumin, BUN, total bilirubin, and INR offers a simple and convenient tool for accurate prediction of death risk in NMAP patients early.

表1 选自MIMIC-Ⅲ数据库中606例NMAP患者的临床资料特征
指标 总体(606例) 训练集(425例) 验证集(181例) 统计值 P
年龄(岁)a 59.4 (46.1, 72.3) 60.1(46.2, 73.0) 57.9 (45.2, 71.4) -1.097 0.273
性别b       0.363 0.547
347 (57.3) 242 (56.9) 105 (58.0)    
259 (42.7) 183 (43.1) 76 (42.0)    
病因b       0.211 0.976
胆源性 230 (38.0) 161 (37.8) 69 (38.1)    
脂源性 57 (9.4) 39 (9.2) 18 (9.9)    
酒精性 115 (19.0) 81 (19.1) 34 (18.8)    
其他 204 (33.6) 144 (33.9) 60 (33.2)    
入院24 h内生命体征a
体温(℃) 37.7(37.2, 38.4) 37.7(37.1, 38.3) 37.8 (37.3, 38.6) -0.327 0.743
收缩压(mmHg) 93 (82, 107) 93 (81, 107) 92 (82, 106) -0.054 0.957
呼吸(次/分) 29 (24, 34) 29 (24, 34) 29 (25, 34) -0.553 0.580
心率(次/分) 115 (98, 129) 114 (97, 128) 117 (102, 131) -1.260 0.208
入院24 h内血液指标a        
白细胞(×109/L) 14 (9, 19) 14 (9, 19) 13 (10, 19) -0.117 0.907
红细胞(×1012/L) 3.97(3.42, 4.53) 3.99 (3.41, 4.54) 3.96 (3.50, 4.49) -0.066 0.947
血小板(×109/L) 220 (156, 303) 221(156, 304) 214 (157, 297) -0.526 0.599
血红蛋白(g/L) 12.20 (10.70, 13.90) 12.20 (10.70, 13.90) 12.20 (10.80, 14.20) -0.061 0.951
红细胞比容(%) 36 (32, 41) 36 (32, 41) 36 (32, 42) -0.045 0.964
RDW(%) 14.80 (13.90, 16.30) 14.80 (13.90, 16.60) 14.80 (13.80, 16.00) -0.596 0.551
中性粒细胞-淋巴细胞比值 9.2 (5.1, 16.7) 9.0 (4.9, 15.3) 9.6 (5.5, 16.9) -1.013 0.331
血小板-淋巴细胞比 186 (106, 302) 188 (108, 303) 183 (106, 291) -0.645 0.519
单核细胞-淋巴细胞比 2.35 (1.33, 4.00) 2.31 (1.32, 4.00) 2.50 (1.39, 4.10) -0.871 0.384
淀粉酶(U/L) 213 (84, 654) 216 (85, 636) 193 (79, 713) -1.279 0.201
脂肪酶(U/L) 246 (69, 1 163) 255 (69, 1 096) 244 (69, 1 336) -0.039 0.969
离子钙(mmol/L) 8.40 (7.80, 9.00) 8.40 (7.80, 9.10) 8.40 (7.80, 8.90) -1.200 0.230
白蛋白(g/L) 31.0 (27.0, 38.0) 31.0 (27.0, 38.0) 31.0 (27.0, 37.0) -1.245 0.213
肌酐(μmol/L) 123.76 (79.56, 229.84) 123.76 (79.56, 247.52) 114.92 (79.56, 229.84) -0.319 0.750
BUN(mmol/L) 10.00 (6.07, 17.49) 10.35 (6.07, 18.92) 9.28 (5.71, 15.00) -0.908 0.364
丙氨酸转氨酶(U/L) 48 (24, 149) 48 (23, 137) 48 (24, 176) -0.239 0.811
天冬氨酸转氨酶(U/L) 69 (33, 197) 69 (33, 181) 69 (32, 217) -0.635 0.526
总胆红素(μmol/L) 18.81(8.55, 51.3) 17.10(8.55, 46.17) 20.52(10.26, 56.43) -0.623 0.533
葡萄糖(mmol/L) 8.77(6.72, 12.38) 8.55(6.72, 11.77) 9.27(6.94, 13.76) -0.347 0.729
INR 1.30 (1.20, 1.70) 1.30 (1.20, 1.70) 1.30 (1.20, 1.60) -1.176 0.239
阴离子间隙(mmol/L) 17 (14, 21) 17 (14, 21) 17 (14, 21) -1.537 0.124
碳酸氢盐(mmol/L) 24.0 (20.0, 27.0) 24.0 (20.0, 27.0) 24.0 (20.0, 27.0) -0.298 0.353
合并症b          
心力衰竭 165 (27.2) 117 (27.5) 48 (26.5) 0.550 0.459
慢性肺病 97 (16.0) 69 (16.2) 28 (15.5) 0.228 0.633
高血压 332 (54.8) 229 (53.9) 103 (56.9) 0.847 0.357
糖尿病 140 (23.1) 94 (22.1) 46 (25.4) 0.777 0.378
肾功能不全 111 (18.3) 80 (18.8) 31 (17.1) 0.244 0.621
慢性肝病 156 (25.7) 109 (25.6) 47 (26.0) 0.532 0.466
图1 基于LASSO回归的NMAP患者死亡预测因子筛选A为log(λ)与LASSO回归系数的关系;B为十折交叉验证后log(λ)与MSE的关系,虚线a代表MSE最小时最优调和系数,此时λ=0.019,共筛选出20个变量;虚线b代表MSE1个SE内最优调和系数,此时λ=0.064,共筛选出7个变量。NMAP为非轻症急性胰腺炎
表2 基于LASSO的NMAP患者死亡风险多因素Cox回归分析
图2 非轻症急性胰腺炎患者死亡风险列线图预测模型
图3 列线图模型预测非轻症急性胰腺炎患者不同时间内死亡风险的ROC曲线 A为训练集;B为验证集;AUC为曲线下面积
图4 列线图模型预测非轻症急性胰腺炎患者90 d内死亡风险的校准曲线 A为训练集;B为验证集
图5 列线图模型预测非轻症急性胰腺炎患者死亡风险的决策曲线
图6 列线图模型及其他评分系统预测非轻症急性胰腺炎患者死亡风险的时依ROC曲线比较 急性胰腺炎严重程度床边指数(BISAP);快速序贯器官衰竭评分(qSOFA),序贯器官衰竭评分(SOFA);为急性生理评分(APS Ⅲ);牛津急性疾病严重程度评分(OASIS);AUC为曲线下面积
[1]
Mederos MA, Reber HA, Girgis MD. Acute pancreatitis[J]. JAMA, 2021, 325(4): 382.
[2]
中华医学会急诊分会, 京津冀急诊急救联盟, 北京医学会急诊分会, 等. 急性胰腺炎急诊诊断及治疗专家共识[J]. 临床肝胆病杂志, 2021, 37(5): 1034-1041.
[3]
Tee YS, Fang HY, Kuo IM, et al. Serial evaluation of the SOFA score is reliable for predicting mortality in acute severe pancreatitis[J]. Medicine (Baltimore), 2018, 97(7): e9654.
[4]
Wu Q, Wang J, Qin M, et al. Accuracy of conventional and novel scoring systems in predicting severity and outcomes of acute pancreatitis: A retrospective study[J]. Lipids Health Dis, 2021, 20(1): 41.
[5]
Liu F, Yao J, Liu C, et al. Construction and validation of machine learing models for sepsis prediction in patients with acute pancreatitis[J]. BMC Surg, 2023, 23(1): 267.
[6]
刘晓宁, 刘冰, 马丽琼, 等. 中性粒细胞-淋巴细胞比值在新亚特兰大分类标准下对重症急性胰腺炎早期预测价值的Meta分析和系统评价[J/CD]. 中华普通外科学文献(电子版), 2022, 16(1): 67-73.
[7]
Xu X, Ai F, Huang M. Deceased serum bilirubin and albumin levels in the assessment of severity and mortality in patients with acute pancreatitis[J]. Int J Med Sci, 2020, 17(17): 2685-2695.
[8]
Cheng T, Liu BF, Han TY, et al. Efficiency of red cell distribution width in predicting severity and mortality of patients with acute pancreatitis: A protocol for systematic review and meta-analysis[J]. Medicine, 2021, 100(6): e24658.
[9]
Liu Q, Zheng HL, Wu MM, et al. Association between lactate-to-albumin ratio and 28-days all-cause mortality in patients with acute pancreatitis: A retrospective analysis of the MIMIC-IV database[J]. Front Immunol, 2022, 13: 1076121.
[10]
Johnson AE, Pollard TJ, Shen L, et al. MIMIC-Ⅲ, a freely accessible critical care database[J]. Sci Data, 2016, 3(1): 160035.
[11]
Banks PA, Bollen TL, Dervenis C, et al. Classification of acute pancreatitis-2012: revision of the Atlanta classification and definitions by international consensus[J]. Gut, 2013, 62(1): 102-111.
[12]
Zhu K, Lin H, Yang X, et al. An in-hospital mortality risk model for elderly patients undergoing cardiac valvular surgery based on LASSO-Logistic regression and machine learning[J]. J Cardiovasc Dev Dis, 2023, 10(2): 87.
[13]
Li Y, Lu F, Yin Y. Applying logistic LASSO regression for the diagnosis of atypical Crohn’s disease[J]. Sci Rep, 2022, 12(1): 11340.
[14]
中国医疗保健国际交流促进会急诊医学分会, 脓毒症预防与阻断联盟. 重症急性胰腺炎预防与阻断急诊专家共识[J]. 中国急救医学, 2022, 42(5): 369-379.
[15]
Vasudevan S, Goswami P, Sonika U, et al. Comparison of various scoring systems and biochemical markers in predicting the outcome in acute pancreatitis[J]. Pancreas, 2018, 47(1): 65-71.
[16]
Li C, Ren Q, Wang Z, et al. Early prediction of in-hospital mortality in acute pancreatitis: A retrospective observational cohort study based on a large multicentre critical care database[J]. BMJ Open, 2020, 10(12): e041893.
[17]
Han D, Xu F, Li C, et al. A novel nomogram for predicting survival in patients with severe acute pancreatitis: An analysis based on the large MIMIC-III clinical database[J]. Emerg Med Int, 2021, 2021: 9190908.
[18]
Zhu XG, Jiang JM, Li YX, et al. Development and validation of a nomogram for predicting survival in patients with acute pancreatitis[J]. World J Emerg Med, 2023, 14(1): 44-48.
[19]
Pan LN, Pan SA, Lei BH, et al. Development of a novel nomogram incorporating red blood cell distribution width-albumin ratio for the prediction of 30-day mortality in acute pancreatitis patients[J]. Emerg Med Int, 2022, 2022: 1573931.
[20]
Cao Y, Su Y, Guo C, et al. Albumin level is associated with short-term and long-term outcomes in sepsis patients admitted in the ICU: A large public database retrospective research[J]. Clin Epidemiol, 2023, 15: 263-273.
[21]
Thongprayoon C, Cheungpasitporn W, Chewcharat A, et al. Risk of acute respiratory failure among hospitalized patients with various admission serum albumin levels: A cohort study[J]. Medicine (Baltimore), 2020, 99(9): e19352.
[22]
Wu BU, Johannes RS, Sun X, et al. Early changes in blood urea nitrogen predict mortality in acute pancreatitis[J]. Gastroenterology, 2009, 137(1): 129-135.
[23]
Yang Z, Dong L, Zhang Y, et al. Prediction of severe acute pancreatitis using a decision tree model based on the revised atlanta classification of acute pancreatitis[J]. PLoS One, 2015, 10(11): e0143486.
[24]
Jiang X, Su Z, Wang Y, et al. Prognostic nomogram for acute pancreatitis patients: An analysis of publicly electronic healthcare records in intensive care unit[J]. J Crit Care, 2019, 50: 213-220.
[25]
Xu X, Ai F, Huang M. Deceased serum bilirubin and albumin levels in the assessment of severity and mortality in patients with acute pancreatitis[J]. Int J Med Sci, 2020, 17(17): 2685-2695.
[26]
Zhang FX, Li ZL, Zhang ZD, et al. Prognostic value of red blood cell distribution width for severe acute pancreatitis[J]. World J Gastroenterol, 2019, 25(32): 4739-4748.
[1] 许杰, 李亚俊, 韩军伟. 两种入路下腹腔镜根治性全胃切除术治疗超重胃癌的效果比较[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(01): 19-22.
[2] 高杰红, 黎平平, 齐婧, 代引海. ETFA和CD34在乳腺癌中的表达及与临床病理参数和预后的关系研究[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(01): 64-67.
[3] 李代勤, 刘佩杰. 动态增强磁共振评估中晚期低位直肠癌同步放化疗后疗效及预后的价值[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(01): 100-103.
[4] 梁孟杰, 朱欢欢, 王行舟, 江航, 艾世超, 孙锋, 宋鹏, 王萌, 刘颂, 夏雪峰, 杜峻峰, 傅双, 陆晓峰, 沈晓菲, 管文贤. 联合免疫治疗的胃癌转化治疗患者预后及术后并发症分析[J/OL]. 中华普外科手术学杂志(电子版), 2024, 18(06): 619-623.
[5] 张志兆, 王睿, 郜苹苹, 王成方, 王成, 齐晓伟. DNMT3B与乳腺癌预后的关系及其生物学机制[J/OL]. 中华普外科手术学杂志(电子版), 2024, 18(06): 624-629.
[6] 屈勤芳, 束方莲. 盆腔器官脱垂患者盆底重建手术后压力性尿失禁发生的影响因素及列线图预测模型构建[J/OL]. 中华腔镜泌尿外科杂志(电子版), 2024, 18(06): 606-612.
[7] 韩加刚, 王振军. 梗阻性左半结肠癌的治疗策略[J/OL]. 中华结直肠疾病电子杂志, 2024, 13(06): 450-458.
[8] 董佳, 王坤, 张莉. 预后营养指数结合免疫球蛋白、血糖及甲胎蛋白对HBV 相关慢加急性肝衰竭患者治疗后预后不良的预测价值[J/OL]. 中华消化病与影像杂志(电子版), 2024, 14(06): 555-559.
[9] 刘郁, 段绍斌, 丁志翔, 史志涛. miR-34a-5p 在结肠癌患者的表达及其与临床特征及预后的相关性研究[J/OL]. 中华消化病与影像杂志(电子版), 2024, 14(06): 485-490.
[10] 陈倩倩, 袁晨, 刘基, 尹婷婷. 多层螺旋CT 参数、癌胚抗原、错配修复基因及病理指标对结直肠癌预后的影响[J/OL]. 中华消化病与影像杂志(电子版), 2024, 14(06): 507-511.
[11] 曾明芬, 王艳. 急性胰腺炎合并脂肪肝患者CT 与彩色多普勒超声诊断参数与其病情和预后的关联性研究[J/OL]. 中华消化病与影像杂志(电子版), 2024, 14(06): 531-535.
[12] 沈炎, 张俊峰, 唐春芳. 预后营养指数结合血清降钙素原、胱抑素C及视黄醇结合蛋白对急性胰腺炎并发急性肾损伤的预测价值[J/OL]. 中华消化病与影像杂志(电子版), 2024, 14(06): 536-540.
[13] 王景明, 王磊, 许小多, 邢文强, 张兆岩, 黄伟敏. 腰椎椎旁肌的研究进展[J/OL]. 中华临床医师杂志(电子版), 2024, 18(09): 846-852.
[14] 韦巧玲, 黄妍, 赵昌, 宋庆峰, 陈祖毅, 黄莹, 蒙嫦, 黄靖. 肝癌微波消融术后中重度疼痛风险预测列线图模型构建及验证[J/OL]. 中华临床医师杂志(电子版), 2024, 18(08): 715-721.
[15] 颜世锐, 熊辉. 感染性心内膜炎合并急性肾损伤患者的危险因素探索及死亡风险预测[J/OL]. 中华临床医师杂志(电子版), 2024, 18(07): 618-624.
阅读次数
全文


摘要