|
|
Back to Annual Meeting Program
Hiatal Hernia, Barrett's Esophagus and Long-Term Symptom Control After Laparoscopic Fundoplication for Gastresophageal Reflux
Joumanah Hafez*1,2, Johannes Lenglinger2, Friedrich Wrba3, Marcus Hudec4, Christiane Wischin2, Johannes Miholic2 1Department of ENT, Martin Luther University Halle, Medical Faculty, Halle/Saale, Germany; 2Department of Surgery, Medical University Vienna, Vienna, Austria; 3Department of Pathology, Medical University Vienna, Vienna, Austria; 4Department of Scientific Computing, University of Vienna, Vienna, Austria
Objective: To determine the long-term symptom control after laparoscopic fundoplication for gastroesophageal reflux disease (GERD), and possible prognostic factors. Design, setting and patients: A cohort of 271 patients, operated at a university hospital from 1996 through 2002, was eligible for evaluation after a median interval of 102 months (range, 12 to 158). Main outcome measures: The time between operation and recurrence of reflux symptoms (i.e. time to treatment failure) served as the end point for statistical analysis. Putative risk factors for symptom recurrence were analyzed by univariate analysis and by using Cox's multiple hazards regression. Results: According to Kaplan-Meier estimates the rate of reflux symptom recurrence was 15% after 108 months, 11% in cases without intestinal metaplasia, but 43% in patients with long-segment (33cm) Barrett's esophagus (BE; p<0.0001). Reflux symptoms recurred in 22% of cases with a hiatal hernia (HH) 33 cm before operation, but only in 7% with smaller or absent HH (p=0.005). Multivariate analysis revealed a relative risk of 6.6 (CI 3.0) for long-segment BE and of 3.0 (CI 1.7) for HH 33 cm. A strong statistical interaction was found between HH 33 cm and long-segment BE: the small group (n=18) of cases exhibiting both risk factors revealed an exaggerated recurrence rate of 72% at 108 months. Conclusions: Laparoscopic fundoplication for symptomatic GERD provided a long-lasting abolition of reflux symptoms in 231 of 271 (85%) patients. HH 33 cm and long-segment BE materialized as independent prognostic factors favoring recurrence.
Univariate analysis of putative risk factors for symptom recurrence Variable | n | % recurrence after 36 months | % recurrence after 108 months | p-Value (log-rank test) | All patients | 271 | 7% | 15% | | Mode of fundoplication | | | | 0.49 | | Total (Nissen) n=197 | 7% | 16% | | | Partial (Toupet) n=74 | 7% | 14% | | Intestinal metaplasia | | | | <0.001 | | Absent n=199 | 3% | 11% | | | Short segment BE n=43 | 12% | 16% | | | Long segment BE n=29 | 28% | 43% | | DeMeester's score | <50 n=209 | 5% | 10% | 0.001 | | >/= 50 n=62 | 15% | 28% | | Hiatal hernia | < 3cm n=93 | 5% | 7% | 0.005 | | >3cm n=153 | 10% | 22% | | Contraction amplitudes | <62mmHg n=131 | 10% | 19% | 0.07 | | >62 n=138 | 6% | 12% | |
Gender, age, LES pressure and operation time period were not significant. Multivariate analysis (Cox’s multiple hazards model) of prognostic factors for time to symptom recurrence. Variable | Risk Ratio | Lower CL | Upper CL | p-Value | HH>/= 3cm | 3.8 | 1.7 | 10.2 | <0.001 | BE>/= 3cm | 6.6 | 3.1 | 13.5 | <0.001 | Nissen vs. Toupet | 1.0 | 0.7 | 1.5 | 0.91 | Propensity score | 0.97 | 0.6 | 1.2 | 0.85 |
Back to Annual Meeting Program
|