Download PDF
Endoscopy 2002; 34(6): 461-463
DOI: 10.1055/s-2002-31996
Original Article

© Georg Thieme Verlag Stuttgart · New York

Length of Esophageal Cancer and Degree of Luminal Stenosis during Upper Endoscopy Predict T stage by Endoscopic Ultrasound

M.  S.  Bhutani 1 , C.  J.  Barde 2 , R.  J.  Markert 2 , N.  Gopalswamy 2
  • 1 Center for Endoscopic Ultrasound, University of Texas Medical Branch, Galveston, Texas, USA
  • 2 Wright State University/Veterans’ Affairs Medical Center, Dayton, Ohio, USA
Further Information

Publication History

1 December 2000

22 January 2002

Publication Date:
04 June 2002 (online)

Also available at
   Buy Article Permissions and Reprints
Zoom Image

Background and Study Aims: Endoscopic ultrasonography (EUS) is considered to be the most accurate modality for T staging of esophageal cancer. This study attempted to determine whether endoscopic features such as the length and degree of luminal stenosis in esophageal cancer can predict the T stage on EUS.
Patients and Methods: Thirty-five patients with newly diagnosed esophageal adenocarcinoma or squamous-cell carcinoma undergoing EUS prior to initiation of any treatment were included in the study. The length of the tumor was assessed prospectively during esophagogastroduodenoscopy (EGD) before EUS in 22 patients. Radial EUS was then performed in these patients. The other 13 patients had sufficient luminal stenosis to prevent complete advancement of the echo endoscope through the tumor. In these 13 patients, the length of the esophageal cancer was not examined, but the T and N stage up to the level of maximum advancement of the echo endoscope through the tumor were noted.
Results: All 13 patients with luminal stenosis had at least a T3 (n = 12) or T4 (n = 1) lesion up to the level of maximum advancement of the echo endoscope. Among the 22 patients in whom the length of the esophageal cancer was measured, the mean length in the 13 patients with a T1 or T2 lesion on EUS was 2.6 cm. The mean length in the nine patients with T3 esophageal cancer was 7.1 cm. The difference in the mean length of T1 or T2 lesions (2.6 cm) was significantly different (P < 0.001) from the mean length of T3 lesions (7.1 cm). Using a clinical diagnostic testing approach, when ≥ 5 cm length was used as a criteria for diagnosing T3 lesions, the sensitivity was 89 %, specificity 92 %, positive predictive value 89 %, and negative predictive value 92 %. There was also a suggestion of increased chances of lymph-node metastases with increasing length of esophageal cancer.
Conclusions: In esophageal carcinoma, endoscopic features such as the length of the cancer and the degree of luminal stenosis correlate with T staging on EUS. Esophageal cancers that are ≥ 5 cm in length, or are sufficiently stenotic to prevent passage of an endoscope, are much more likely to be T3 or higher-stage lesions, while those that are < 5 cm in length have a greater chance (92 %) of being T1 or T2.