Non-optical, wire-guided echoscope (Esophagoprobe) markedly increases the rate of complete staging in patients with esophageal carcinoma

Background: EUS is highly accurate for staging patients with esophageal carcinoma and correlates with clinical outcome. Unfortunately, many patients present with malignant strictures that do not permit passage of standard echoendoscopes. Staging from a point proximal to the stricture may not detect vascular involvement and cannot assess for celiace adenopathy. Dilation immediately prior to EUS is associated with a high rate of perforation (up to 25%). A new echoendoscope (Olympus MH-908) has smaller in diameter (7.9 mm), has a tapered tip and can be passed over a Savary guidewire. We investigated the efficacy and safety of this device to perform complete staging. Methods: We retrospectively reviewed all cases of EUS staging of esophageal carcinoma at our institution. Data were recorded regarding the ability to successfully traverse the esophagus and examine the celiac axis. Cases performed after neoadjuvant treatment were excluded. Success rates before the availability of the esophagoprobe were compared with those after its availability. Results: We reviewed 116 consecutive EUS examinations that were performed for pre-therapeutic staging of esophageal carcinoma. 98 of these examinations were performed prior to the availability of the esophagoprobe and 18 after. Data regarding successful echoendoscope passage and staging is shown below. Without the MH-908 With the MH-908 Complete exam 64 of 98 (65%) 18 of 18 (100%) p=0.003 Since it became available at this institution, the esophagoprobe has been used in 11 of 18 cases with no complications noted (specifically, no cases of suspected perforation). Four cases of suspected aortic invasion were identified. Conclusions: Use of the new non-optical, wire-guided, taper-tip dedicated echoendoscope eliminated the occurrence of incomplete esophageal staging in our practice. This compares to previous incomplete exam rate of 35%. These results were statistically significant.