Univariate and Multivariate Analyses of Preoperative Factors Influencing Symptomatic Outcomes of Transoral Fundoplication

Reginald CW Bell, MD, FACS, Mark A Fox, MD, FACS, William E Barnes, MD, FACS, Peter G Mavrelis, MD, Robert W Sewell, MD, FACS, Bart J Carter, MD, FACS, Glenn M Ihde, MD, Karim S Trad, MD, FACS, David Dargis, DO, Kevin M Hoddinott, MD, FACS, Katherine D Freeman, MSN, NP, Tanja Gunsberger, DO, Mark G Hausmann, MD, FACS, Brian DaCosta Gill, MD, FACS, Erik Wilson, MD, FACS. SurgOne Foregut Institute, Englewood, Colorado; Crossville Medical Group, PA, Crossville, Tennessee; Livingston Hospital and Healthcare Services, Inc. CAH, Salem, Kentucky; Internal Medicine Associates, Merrillville, Indiana.

OBJECTIVE: Identification of preoperative factors that predict symptomatic outcomes of transoral fundoplication (TF) in chronic gastroesophageal reflux disease (GERD) patients refractory to proton pump inhibitor (PPI) therapy.

METHODS: Univariate and multivariate logistic regression analyses were performed on data from 158 consecutive patients who underwent TF with the EsophyX device between January 2010 and June 2012 in 14 community centers. Variables included age, gender, body mass index, GERD duration, PPI therapy duration, presence of hiatal hernia , esophagitis, Hill grade, quality of life scores (QOL) on PPIs, % total time pH < 4 and DeMeester score on 48-hour pH monitoring. Covariates with P values ≤ 0.25 on univariate analyses underwent backwards stepwise multivariate regression analyses. P value ≤ 0.05 was considered significant. Symptomatic outcomes were grouped as follows: (1) Successful, completely off PPIs and ≥ 50% reduction of a Gastroesophageal Reflux Disease Health-related Quality of Life (GERD-HRQL) score or normalization of Reflux Symptom Index (RSI) score (≤ 13); (2) Partially successful, on PPIs and ≥ 50% reduction in GERD-HRQL score or normalization of RSI score; (3) Poor, reoperation or < 50% reduction in GERD-HRQL or abnormal RSI score. Analyses were performed separately on patients with typical and atypical symptoms before TF.

RESULTS:  All patients suffered from typical GERD symptoms. Additionally, 78% (124/158) of patients suffered from atypical symptoms (RSI score > 13). Median age was 59 (range 19-80) years; 71% (112/158) were female. Six percent (10/158) with recurrent GERD symptoms refractory to PPI therapy underwent revisional procedure (9 laparoscopic Nissen, 1 TF). Globally, at a median follow-up of 22 (10-43) months, 70% of patients (111/158) had successful or partially successful outcomes; 30% (47/158) had poor outcomes. Seventy-seven percent (121/158) of patients were off daily PPIs; 70% (110/158) were completely off PPIs. The median GERD-HRQL score improved from 28 (10-50) before TF on PPIs to 5 (0-45), P < 0.001. In 124 patients with atypical symptoms, the median RSI score improved from 26 (14-45) before TIF on PPIs to 5.5 (0-41), P < 0.001.
In patients with typical symptoms, univariate analyses revealed 4 preoperative factors predictive of successful outcomes: age ≥ 50 [odds ratio (OR) = 2.4; 95% confidence interval (CI) = 1.2- 4.8, P = 0.014]; GERD-HRQL score ≥ 15 on PPIs (OR=6.0, CI = 1.2-29.4, P = 0.026, RSI score ≥ 13 on PPIs (OR = 2.4, CI = 1.1-5.2, P = 0.027) and GERSS score ≥ 18 on PPIs (OR = 2.6, CI = 1.2-5.8, P = 0.018). Female gender approached closely the significance level (OR = 2.0, CI = 1.0-4.0, P = 0.053). Age and GERD-HRQL score on PPIs remained significant predictors at the multivariate level. In patients with atypical symptoms, the only factor associated with successful outcomes was GERD-HRQL score ≥ 15 on PPIs (OR=9.9, CI = 0.9-4.6, P = 0.036).

CONCLUSIONS: Elevated preoperative QOL scores and age ≥ 50 were most closely associated with successful outcome of TF in this analysis; objective measures were not. Further studies are needed to confirm these findings.

View Poster

« Return to SAGES 2014 abstract archive