探讨口服造影剂胃超声检查在胃部病变筛查中的应用价值。

纳入2015年1月至2018年12月于中山大学附属第八医院体检行胃超声检查同时取得胃镜检查结果者3573例。超声观察胃腔、胃壁及胃周结构，并测量胃壁厚度。分析胃超声检查与胃镜诊断结果，绘制胃壁厚度评估胃病变的ROC曲线，并计算ROC曲线下面积、最佳截断值及诊断敏感度和特异度。采用线性加权Kappa检验对胃壁厚度评估胃部病变的结果与胃镜检查结果进行一致性分析。

所有纳入对象经术后病理证实胃癌20例。胃超声检查筛查出18例胃癌，均经手术病理证实为胃腺癌。超声检查漏诊2例，误诊2例。胃镜漏诊1例。以手术病理、胃镜活检病理结果及胃镜结合临床随访结果为诊断标准，绘制胃壁厚度评估正常胃壁与病变胃壁的ROC曲线。ROC曲线下面积为0.946（95%CI：0.930~0.962），最佳截断值为5.5 mm，其诊断敏感度和特异度分别为97.5%和98.3%。绘制胃壁厚度评估肿块胃壁与非肿块胃壁的ROC曲线。ROC曲线下面积为0.829（95% CI：0.758~0.900），最佳截断值为9.5 mm，其诊断敏感度和特异度分别为79.7%和75.9%。基于ROC曲线阈值分析结果，以胃壁厚度为依据诊断不同病变，胃壁厚度＜6 mm判定为正常或非糜烂性胃炎，6 mm≤胃壁厚度≤9 mm判定为糜烂性胃炎，胃壁厚度＞9 mm判定为肿块。对胃壁厚度评估结果与胃镜检查结果进行一致性分析，Kappa值为0.792（P＜0.001）。

胃壁厚度评估胃部病变与胃镜诊断结果一致性较好，胃超声检查作为一种简便、快捷的胃疾病筛查手段，对于胃癌的筛查及其他胃部疾病的初筛及随诊具有重要意义。

To probe into the value of gastric wall thickness measured by oral contrast trans abdominal gastric ultrasonography in evaluating gastric lesions.

From January 2015 to December 2018, 3573 patients underwent gastroscopic examination and gastric ultrasonography at the Eighth Affiliated Hospital of Sun Yat-Sen University. The stomach cavity, gastric wall, and perigastric structure were observed by gastric ultrasonography, and the thickness of the gastric wall was measured. The results of gastric ultrasonography and gastroscopy were analyzed. The ROC curve of gastric wall thickness was plotted and the area under the ROC curve, the best cutoff value, sensitivity, and specificity were calculated. Consistency analysis was performed on the results of gastric wall thickness measured by gastric ultrasonography and gastroscopy in the diagnosis of gastric lesions using linear weighted Kappa test.

Twenty patients were confirmed with gastric cancer by postoperative pathology. Gastric ultrasonography screening revealed 18 cases of gastric cancer, which were proved to be gastric adenocarcinoma by pathology. Gastric ultrasonography missed and misdiagnosed two cases of gastric cancer each. Gastroscopy missed one case. The diagnostic criteria were surgical pathology, pathological results of gastroscopic biopsy, and combined gastroscopic clinical follow-up results. The ROC curves of normal and diseased gastric wall thickness were plotted according to the pathological results of gastroscopic biopsy and clinical follow-up results. The area under the ROC curve was 0.946 (95% CI: 0.930-0.962), the best cutoff value was 5.5 mm, and the diagnostic sensitivity and specificity were 97.5% and 98.3%, respectively. Gastric mass and non-mass were judged according to the results of gastroscopy and pathology, and the ROC curves of gastric mass and non-mass were plotted. The area under the ROC curve was 0.829 (95% CI: 0.758-0.900), the best cutoff value was 9.5 mm, and the diagnostic sensitivity and specificity were 79.7% and 75.9%, respectively. Based on the results of ROC curve analysis, different lesions were diagnosed according to the thickness of the gastric wall. Gastric wall thickness＜6 mm was determined as normal or non-erosive gastritis, 6 mm≤gastric wall thickness≤9 mm was determined as erosive gastritis, and gastric wall thickness＞9 mm was determined as mass. The consistency between the results of gastric wall thickness evaluation and endoscopic examination was analyzed. The Kappa value was 0.792 (P＜0.001).

Evaluation of gastric wall thickness is consistent with the results of gastroscopy. Gastric ultrasonography, as a simple and rapid screening method for gastric diseases, is of great significance for screening gastric cancer and other gastric diseases. It is worth popularizing in people with routine physical examination.