Predictive value of three nutritional indexes for disease activity in patients with inflammatory bowel disease

Malnutrition is prevalent in patients with inflammatory bowel disease (IBD); however, its ability to predict the disease activity in IBD remains unexplored. Therefore, this study aimed to explore the association between malnutrition and disease activity in IBD. In this retrospective study, we enrolled 1006 patients diagnosed with IBD from the First Affiliated Hospital of Wenzhou Medical University from 2011 to 2022. Malnutrition was assessed based on the prognostic nutritional index (PNI), geriatric nutritional risk index (GNRI), and controlling nutritional status (CONUT) scores. Logistic regression analyses were performed to identify predictors for disease activity. Restricted cubic spline analysis was performed to evaluate the possible nonlinear relations, and subgroup analysis was performed to explore potential interactions. Additionally, prediction performances were compared through receiver operating characteristic curves, net reclassification improvement, and integrated discrimination improvement. The prevalence of malnutrition calculated by the PNI, GNRI, and CONUT scores in IBD was 16.9%, 72.1%, and 75.6%, respectively and significant correlations were observed among them. Multivariate logistic regression analysis showed that PNI, GNRI, and CONUT were independent risk factors for disease activity, and no significant nonlinear relationship was observed between disease activity and all three indexes. No statistically significant interactive effect was found in nearly all the subgroups. GNRI showed the highest predictive value compared with PNI and CONUT. Additionally, combining any of the three indexes improved the ability of C-reactive protein to predict IBD activity. All three nutritional indexes evaluated malnutrition to be an independent risk factor for IBD activity.

炎症性肠病（inflammatory bowel disease, IBD）患者普遍存在营养不良的情况，但其在预测IBD疾病活动度方面的价值仍未得到探索。因此，本研究旨在探讨营养不良与IBD疾病活动度之间的关联。本研究为回顾性研究，纳入了2011年至2022年温州医科大学附属第一医院收治的1006例IBD确诊患者。营养不良的评估基于预后营养指数（prognostic nutritional index, PNI）、老年营养风险指数（geriatric nutritional risk index, GNRI）以及营养控制状况评分（controlling nutritional status, CONUT）。采用Logistic回归分析识别疾病活动度的预测因素，限制性立方样条分析用于评估潜在的非线性关联，亚组分析则用于探索潜在的交互作用。此外，通过受试者工作特征曲线（receiver operating characteristic curve, ROC）、净重新分类改善指数（net reclassification improvement, NRI）以及综合判别改善指数（integrated discrimination improvement, IDI）比较了各项预测模型的预测性能。基于PNI、GNRI与CONUT评分计算得出的IBD患者营养不良患病率分别为16.9%、72.1%与75.6%，且三者间存在显著相关性。多因素Logistic回归分析显示，PNI、GNRI与CONUT均为疾病活动度的独立危险因素，且三者与疾病活动度之间均未发现显著的非线性关联。几乎所有亚组中均未发现具有统计学意义的交互效应。与PNI和CONUT相比，GNRI展现出最高的预测价值。此外，将三项营养指标中的任意一项与C反应蛋白（C-reactive protein, CRP）结合，均可提升其预测IBD活动度的能力。本研究证实，三项营养指标均将营养不良评估为IBD活动度的独立危险因素。