计算机视觉食管曲张静脉压力检测技术的初步研究

Objective To evaluate the clinical reliability and feasibility of computerized endoscopic balloon manometry in vitro and in vivo, in measurement of pressure of esophageal varices. Methods Computerized endoscopic balloon manometry was used to measure the pressure of variceal model with different diameter (3 mm, 6 mm and 8 mm) and intraluminal pressures (ranging from 8 to 36 mm Hg), and the findings were compared with actual pressures. The technique was also applied in 23 patients with liver cirrhosis and esophageal varices, and its correlation with hepatic venous pressure gradient and other factors related with varices bleeding. Results The study in vitro showed that the measured intraluminal pressure was correlated significantly with the actual value ( r ≥ 0. 993, P ＜ 0. 001 ) without obvious measurement bias(95% CI = -0.13 cm H2O to 0. 33 cm H2O). The measurement in 23 patients were success with little variation coefficient (r≥0. 998) between repeated procedures. Regression analysis showed a good correlation between variceal pressure and hepatic venous pressure gradient (r=0. 858, P ＜ 0. 001 ). A higher variceal pressure was strongly associated with presence of previous bleeding episodes, vascular diameter and presence of red color signs, but did not correlate with the parameter of Child-Pugh classification ( t = 0. 31, P =0. 76). Conclusion Computerized endoscopic balloon manometry is reliable and feasible to examine esophageal variceal pressure, and is very likely to be a valuable clinical index for variceal bleeding.

A new computerized endoscopic balloon manometry to detect esophageal variceal pressure: experimental validation and clinical feasibility