论文部分内容阅读
背景/目的:胃排空延迟(DGE)常见于有胃食管反流性疾病(GERD)患者,对其需进一步地加以研究。本研究对由pH监测所诊断的餐后及空腹相对食管暴露是否能预测DGE进行了评估。方法:30例GERD患者接受了长程食管pH值监测并按照闪烁成像分入DEG 组或非DGE组。采用餐后与空腹的反流指数比值、餐后长时间反流事件的相对频率、最长反流事件的分布来评价相对食管暴露。对这些参数预测DGE的作用进行了评估,采用ROC(受试者操作特征)曲线来选取连续变量的界值并采用Logistic回归模型来计算概率。结果:餐后与空腹的反流指数比值比餐后长时间反流事件的相对频率的ROC曲线下面积要大,在餐后空腹比值为1 时其敏感性和特异性有良好的均衡,餐后空腹比值大于1时,即餐后反流指数大于空腹,其敏感性为93%、阴性预测值为91%。餐后发生最长反流事件的特异性为94%、阳性预测值为89%。餐后返流指数较高加上餐后发生最长反流事件诊断DGE的概率为94%;空腹反流指数较高加上空腹发生最长反流事件诊断非DGE的概率为95%。这些组合代表了本组资料的60%。结论:餐后空腹反流指数比值和最长反流事件分布似乎能准确
Background / Aim: Delayed gastric emptying (DGE) is common in patients with gastroesophageal reflux disease (GERD) and needs further investigation. The present study evaluated whether DGE could be predicted by postprandial and fasting relative esophageal exposure diagnosed by pH monitoring. Methods: Thirty patients with GERD underwent long-term esophageal pH monitoring and were assigned to DEG or non-DGE group according to scintigraphy. Relative esophageal exposure was assessed using the ratio of postprandial reflux index, the relative frequency of postprandial long-term reflux events, and the distribution of longest reflux events. The effect of these parameters on the prediction of DGE was evaluated, the ROC (Subject Operating Characteristic) curve was used to select the cutoff for continuous variables and the logistic regression model was used to calculate the probability. RESULTS: The ratio of postprandial and fasting reflux index was greater than the area under the ROC curve of relative frequency of postprandial long-term reflux events, with a good balance of sensitivity and specificity after a fasting ratio of 1 After the fasting ratio is greater than 1, the postprandial reflux index greater than fasting, the sensitivity was 93%, the negative predictive value of 91%. The longest reflux event after meal was 94% specific, with a positive predictive value of 89%. The high rate of postprandial reflux index plus the longest post-meal reflux event diagnosis of DGE was 94%; the high fasting reflux index plus the probability of non-DGE diagnosis of longest fasting reflux was 95%. These combinations represent 60% of the information in this group. CONCLUSIONS: The postprandial fasting reflux index ratio and the distribution of longest reflux events seem to be accurate