[Risk factors for malnutrition in ulcerative colitis complicated with pyoderma gangrenosum and construction of a lasso regression-based prediction model].

Q3 Medicine

南方医科大学学报杂志 Pub Date : 2025-03-20 DOI:10.12122/j.issn.1673-4254.2025.03.09

Lin Shen, Cuihao Song, Congmin Wang, Xi Gao, Junhong An, Chengxin Li, Bin Liang, Xia Li

{"title":"[Risk factors for malnutrition in ulcerative colitis complicated with pyoderma gangrenosum and construction of a lasso regression-based prediction model].","authors":"Lin Shen, Cuihao Song, Congmin Wang, Xi Gao, Junhong An, Chengxin Li, Bin Liang, Xia Li","doi":"10.12122/j.issn.1673-4254.2025.03.09","DOIUrl":null,"url":null,"abstract":"Objectives: To explore the risk factors for malnutrition in patients with ulcerative colitis complicated with pyoderma gangrenosum and establish a nutritional risk prediction model for these patients.Methods: A total of 277 patients with ulcerative colitis complicated with pyoderma gangrenosum treated from 2019 to 2024 were divided into malnutrition group (n=185) and normal nutrition group (n=92) according to whether malnutrition occurred. The data of 25 potential related factors pertaining to general demography, living and eating habits, and disease-related data were compared between the two groups. Lasso regression was used to screen the risk factors, and a nomogram model was established based on the screened factors and its prediction performance was assessed.Results: The patients in the malnutrition group and normal nutrition group showed significant differences in 21 factors including gender, age, education level, BMI, place of residence, course of disease, and SAS language score (P<0.05). Lasso regression analysis identified 6 factors associated with malnutrition in these patients, namely the duration of ulcerative colitis, activity of ulcerative colitis, duration of pyoderma gangrenosum, number of chronic diseases, SAS score, and sleep quality. The nomogram prediction model established based on these 6 factors had an AUC of 0.992 (95% CI: 0.984-1.000) for predicting malnutrition in these patients, and its application in 14 clinical cases achieved an accuracy rate of 100%.Conclusions: The duration of ulcerative colitis, activity of colitis, duration of pyoderma gangrenosum, number of chronic diseases, anxiety, and sleep quality are closely related with malnutrition in patients with ulcerative colitis complicated by pyoderma gangrenosum, and the nomogram prediction model based on these factors can provide assistance for predicting malnutrition in these patients.","PeriodicalId":18962,"journal":{"name":"南方医科大学学报杂志","volume":"45 3","pages":"514-521"},"PeriodicalIF":0.0000,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11955897/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"南方医科大学学报杂志","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.12122/j.issn.1673-4254.2025.03.09","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Medicine","Score":null,"Total":0}

引用次数: 0

Abstract

Objectives: To explore the risk factors for malnutrition in patients with ulcerative colitis complicated with pyoderma gangrenosum and establish a nutritional risk prediction model for these patients.

Methods: A total of 277 patients with ulcerative colitis complicated with pyoderma gangrenosum treated from 2019 to 2024 were divided into malnutrition group (n=185) and normal nutrition group (n=92) according to whether malnutrition occurred. The data of 25 potential related factors pertaining to general demography, living and eating habits, and disease-related data were compared between the two groups. Lasso regression was used to screen the risk factors, and a nomogram model was established based on the screened factors and its prediction performance was assessed.

Results: The patients in the malnutrition group and normal nutrition group showed significant differences in 21 factors including gender, age, education level, BMI, place of residence, course of disease, and SAS language score (P<0.05). Lasso regression analysis identified 6 factors associated with malnutrition in these patients, namely the duration of ulcerative colitis, activity of ulcerative colitis, duration of pyoderma gangrenosum, number of chronic diseases, SAS score, and sleep quality. The nomogram prediction model established based on these 6 factors had an AUC of 0.992 (95% CI: 0.984-1.000) for predicting malnutrition in these patients, and its application in 14 clinical cases achieved an accuracy rate of 100%.

Conclusions: The duration of ulcerative colitis, activity of colitis, duration of pyoderma gangrenosum, number of chronic diseases, anxiety, and sleep quality are closely related with malnutrition in patients with ulcerative colitis complicated by pyoderma gangrenosum, and the nomogram prediction model based on these factors can provide assistance for predicting malnutrition in these patients.

查看原文本刊更多论文

溃疡性结肠炎合并坏疽性脓皮病营养不良的危险因素及基于套索回归预测模型的构建

目的：探讨溃疡性结肠炎合并坏疽性脓皮病患者营养不良的危险因素，建立溃疡性结肠炎合并坏疽性脓皮病患者营养风险预测模型。方法：选取2019 ~ 2024年收治的277例溃疡性结肠炎合并坏疽性脓皮病患者，根据是否发生营养不良分为营养不良组（185例）和正常营养组（92例）。比较两组的一般人口学、生活和饮食习惯、疾病相关数据等25项潜在相关因素的数据。采用Lasso回归筛选危险因素，根据筛选的因素建立nomogram模型，并对其预测性能进行评价。结果：营养不良组患者与正常营养组患者在性别、年龄、文化程度、BMI、居住地、病程、SAS语言评分（PCI: 0.984 ~ 1.000）等21项预测营养不良的指标上存在显著差异，在14例临床病例中应用其预测营养不良的准确率为100%。结论：溃疡性结肠炎病程、结肠炎活动度、坏疽性脓皮病病程、慢性疾病数量、焦虑程度、睡眠质量与溃疡性结肠炎合并坏疽性脓皮病患者营养不良密切相关，基于这些因素的nomogram预测模型可为预测溃疡性结肠炎合并坏疽性脓皮病患者营养不良提供帮助。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊