Scholarlink

<p>The preoperative differentiation of small gastric gastrointestinal stromal tumors (GISTs) from other submucosal lesions remains clinically challenging due to overlapping endoscopic ultrasound (EUS) features. This study aimed to develop and validate a novel risk-scoring system integrating EUS imaging and clinical characteristics to improve preoperative diagnostic accuracy for small gastric GISTs versus non-GIST submucosal tumors. We retrospectively analyzed 1303 patients with gastric submucosal tumors (SMTs), who were divided into a training cohort (<i>n</i> = 670), an internal validation cohort (<i>n</i> = 287), and an external validation cohort (<i>n</i> = 346). Clinical and EUS characteristics were compared between small GISTs and non-GIST SMTs. Independent predictive factors identified through multivariate logistic regression were used to construct a risk-scoring model. The model’s diagnostic performance was assessed using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, positive and negative predictive values (PPV, NPV), and accuracy. Small gastric GISTs accounted for 614 cases (47.1%). Four independent predictors were identified: tumor location in the cardia/fundus (1 points), origin within the muscularis propria (2 points), hypoechoic echogenicity (1 point), and extraluminal growth pattern (2 points). The scoring system demonstrated strong discriminatory power, with AUCs of 0.776 (95%CI: 0.727–0.826) in the internal and 0.844 (95%CI: 0.805–0.884) in the external validation cohorts. Sensitivities were 0.867 (95%CI: 0.800–0.918) and 0.828 (95%CI: 0.767–0.878), and NPVs were 0.827 and 0.785, respectively. Risk stratification revealed that in the internal cohort, 17.3% of lesions were low-risk (0–2 points), 52.8% intermediate-risk (3–4 points), and 96.2% high-risk (5–6 points). Corresponding proportions in the external cohort were 21.5%, 73.9%, and 95.0%. The proposed EUS-based risk-scoring system shows robust performance in preoperatively diagnosing small gastric GISTs. It holds promise for improving clinical decision-making and optimizing treatment strategies by enabling non-invasive and accurate risk stratification.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A novel risk-scoring system based on endoscopic ultrasound and clinical characteristics for the preoperative diagnosis of small gastric gastrointestinal stromal tumors

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月24日

查看全文
  • Luojie Liu,
  • Yunfu Feng,
  • Sijie Zheng,
  • Bin He,
  • Chao Ma,
  • Zhibing Wang,
  • Xiaodan Xu,
  • Jian Chen,
  • Ying Shi

摘要

The preoperative differentiation of small gastric gastrointestinal stromal tumors (GISTs) from other submucosal lesions remains clinically challenging due to overlapping endoscopic ultrasound (EUS) features. This study aimed to develop and validate a novel risk-scoring system integrating EUS imaging and clinical characteristics to improve preoperative diagnostic accuracy for small gastric GISTs versus non-GIST submucosal tumors. We retrospectively analyzed 1303 patients with gastric submucosal tumors (SMTs), who were divided into a training cohort (n = 670), an internal validation cohort (n = 287), and an external validation cohort (n = 346). Clinical and EUS characteristics were compared between small GISTs and non-GIST SMTs. Independent predictive factors identified through multivariate logistic regression were used to construct a risk-scoring model. The model’s diagnostic performance was assessed using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, positive and negative predictive values (PPV, NPV), and accuracy. Small gastric GISTs accounted for 614 cases (47.1%). Four independent predictors were identified: tumor location in the cardia/fundus (1 points), origin within the muscularis propria (2 points), hypoechoic echogenicity (1 point), and extraluminal growth pattern (2 points). The scoring system demonstrated strong discriminatory power, with AUCs of 0.776 (95%CI: 0.727–0.826) in the internal and 0.844 (95%CI: 0.805–0.884) in the external validation cohorts. Sensitivities were 0.867 (95%CI: 0.800–0.918) and 0.828 (95%CI: 0.767–0.878), and NPVs were 0.827 and 0.785, respectively. Risk stratification revealed that in the internal cohort, 17.3% of lesions were low-risk (0–2 points), 52.8% intermediate-risk (3–4 points), and 96.2% high-risk (5–6 points). Corresponding proportions in the external cohort were 21.5%, 73.9%, and 95.0%. The proposed EUS-based risk-scoring system shows robust performance in preoperatively diagnosing small gastric GISTs. It holds promise for improving clinical decision-making and optimizing treatment strategies by enabling non-invasive and accurate risk stratification.