Construction and validation of nomogram model for prognosis of gastritis patients based on baseline data and...

Gastritis, a global inflammatory disorder, progresses from symptomatic discomfort to potentially malignant changes. Existing staging systems (e.g., OLGA) focus on cancer risk but ignore modifiable factors like inflammation markers and Helicobacter pylori infection. We developed a Nomogram model based on baseline data, inflammatory markers and infectious pathogens for predicting the prognosis of gastritis patients and validating it.

Retrospectively collect the clinical data of patients diagnosed with gastritis, including baseline characteristics, inflammatory markers, and pathogenic infection test results. Univariate and multivariate analyses were performed to identify independent risk factors associated with the prognosis of gastritis patients, based on which a Nomogram prediction model was constructed. The model's accuracy, calibration, and discriminative ability were internally validated using the concordance index (C-index), calibration curve, and the area under the receiver operating characteristic curve (AUC).

Among the 185 patients in the training set, 43 (23.24%) had poor treatment outcomes, while in the validation set of 79 patients, 18 (22.78%) exhibited poor treatment outcomes. No statistically significant differences were observed between the training and validation sets in terms of the incidence of poor treatment outcomes, baseline characteristics, or inflammatory and infectious markers parameters (p > 0.05). Univariate analysis revealed significant differences (p < 0.05) between the poor-outcome and favorable-outcome groups in dietary score, white blood cell count, neutrophil percentage, lymphocyte percentage, C-reactive protein (CRP) level, erythrocyte sedimentation rate (ESR), serum albumin level, and H. pylori infection status. Multivariate logistic regression analysis identified dietary score, neutrophil proportion, CRP, ESR, serum albumin level, and H. pylori infection as independent risk factors for poor endoscopic treatment outcome (p < 0.05). Subsequently, a nomogram prediction model was constructed. The model demonstrated good calibration and fit between predicted and actual outcomes in both the training and validation sets. ROC curve analysis showed that the nomogram model achieved AUC values of 0.808 in the training set and 0.800 in the validation set for predicting gastritis prognosis.

The Nomogram model constructed in this study based on baseline data, inflammation indicators and infectious pathogens can effectively predict the prognosis of patients with gastritis, which can provide a powerful reference for clinical individualized treatment decision-making.