Endoscopic Ultrasonographic Evaluation of Hemodynamics in Esophageal Variceal Patients Related to Early Variceal Relapse

Abstract

Purpose: We evaluated retrospectively the hemodynamics of esophageal varices before and after endoscopic injection sclerotherapy (EIS) and investigated the influence of hemodynamics on variceal recurrence as revealed by endoscopic color Doppler ultrasonography (ECDUS). Methods: Patients were divided according to those with early recurrence (Group A, N = 16) and those without recurrence over the long-term (Group B, N = 12), and those without recurrence between 1 and 3 years (Group C, N = 281). All 309 cases were examined by ECDUS before EIS and 3–5 months after EIS. We monitored the color flow images of blood vessels of paraesophageal veins (PEV), perforating veins (PFV), and cardiac intramural veins (CV) with ECDUS. PEV were classified as either shallow PEV or deep PEV, and PFV were classified as either the inflowing or the outflowing type or mixed type. Results: Before EIS, the detection rates of blood flow in Group A were 16/16 (100%) for PEV, 12/16 (75.0%) for CV, and 13/16 (81.3%) for PFV. PFV were inflowing type in 10 of 13 cases. The detection rates of blood flow in Group B were 12/12 (100%) for PEV, 9/12 (75.0%) for CV, and 6/12 (50.0%) for PFV. PFV were inflowing type in 3 cases and outflowing type in one case, and mixed type in 2 cases. The detection rates of blood flow in Group C were 281/281 (100%) for PEV, 219/281 (77.9%) for CV, and 131/281 (46.7%) for PFV. PFV were inflowing type in 69 cases and outflowing type in 31 cases, and mixed type in 31 cases. The detection rates by ECDUS of PFV and inflowing PFV in Group A were significantly higher than those in Group B and C before EIS. After EIS, the detection rates of blood flow in Group A were 16/16 (100%) for PEV, 9/16 (56.3%) for CV, and 5/16 (31.3%) for PFV. PFV were the inflowing type in all 5 cases. The detection rates of blood flow in Group B were 11/12 (91.7%) for PEV, zero/12 (0%) for CV, and 2/12 (16.7%) for PFV. PFV were of the outflowing type in all 2 cases. The detection rates of blood flow in Group C were 251/281 (89.3%) for PEV, 33/281 (11.7%) for CV, and 6/281 (2.1%) for PFV. PFV were of the outflowing type in all 6 cases. The detection rates by ECDUS of CV, PFV, and inflowing PFV in Group A were significantly higher than those in Group B and C after EIS. Conclusion: ECDUS enables prediction of early recurrence of esophageal varices based on hemodynamic findings before and after EIS.