Preoperative Predictors of Symptom Improvement After Laparoscopic Fundoplication

BACKGROUND AND RATIONALE Gastroesophageal reflux disease affects an estimated 10-20% of adults in Western populations, with a rising prevalence elsewhere. Most patients are controlled with lifestyle measures and proton-pump inhibitor (PPI) therapy, but a substantial subgroup has medically refractory disease, troublesome regurgitation, PPI dependence with concerns about long-term use, or a large hiatal hernia. For these patients, laparoscopic fundoplication remains the most durable and widely performed antireflux operation, and it normalizes esophageal acid exposure in the majority of appropriately selected patients.

Published series nonetheless report that a meaningful minority, commonly 10-30%, experience persistent or recurrent symptoms, troublesome side effects such as dysphagia or gas-bloat, or dissatisfaction with the outcome. Suboptimal results are frequently attributable not to technical failure but to patient selection. Operating on patients whose symptoms are not driven by acid reflux, for example those with functional heartburn or esophageal hypersensitivity, predictably yields poor symptomatic benefit despite a technically sound wrap.

Several preoperative factors have been individually associated with symptomatic outcome. Objective pathological acid exposure, a positive symptom-reflux association, predominance of typical symptoms, and a good preoperative response to acid-suppressive therapy have each been linked to favorable results. Atypical or extra-esophageal symptom predominance, normal acid exposure, high psychological burden, obesity, and certain motility abnormalities have been associated with less benefit. Much of this evidence derives from heterogeneous retrospective series with differing outcome definitions, and the independent contribution of each factor, when modeled together and adjusted for confounding within a contemporary diagnostic framework (Lyon Consensus; Chicago Classification version 4.0), remains incompletely defined.

OBJECTIVES Primary objective: to identify the independent preoperative predictors of clinically significant symptom improvement 12 months after laparoscopic fundoplication.

Secondary objectives: to estimate the proportion of patients achieving clinically significant symptom improvement at 12 months; to identify preoperative predictors of secondary outcomes including discontinuation of antisecretory therapy, resolution of regurgitation, and postoperative dysphagia and gas-bloat symptoms; to develop and internally validate a multivariable prediction model and an optional simplified risk score; and to examine the concordance between subjective symptom improvement and objective reflux control in the subgroup undergoing postoperative physiological retesting.

Primary hypothesis: pathological esophageal acid exposure, a positive symptom-reflux association, predominance of typical symptoms, and a good preoperative PPI response are each independently associated with higher odds of symptom improvement. Secondary hypothesis: high baseline psychological burden, atypical or extra-esophageal symptom predominance, normal acid exposure, and obesity are independently associated with lower odds of improvement.

DESIGN AND SETTING Prospective observational cohort study conducted in accordance with the STROBE statement and, for the prediction-model component, the TRIPOD reporting guideline. Consecutive patients meeting the eligibility criteria are enrolled at preoperative assessment and followed for 12 months after surgery. There is no randomization and no experimental allocation; the type of fundoplication is determined by a prespecified motility-based algorithm reflecting routine clinical practice. The study is conducted at the Department of Surgery, Minia University Hospitals, Faculty of Medicine, Minia University, a tertiary referral center for foregut surgery, with the option of additional collaborating centers applying the same protocol.

CANDIDATE PREOPERATIVE PREDICTORS Ten prespecified preoperative variables are recorded for every participant, with age and sex forced into the multivariable model as covariates: (1) pathological acid exposure, defined as acid exposure time greater than 6% on off-PPI 24-hour pH-impedance monitoring and/or a DeMeester score greater than 14.7, categorized as pathological versus borderline or normal; (2) symptom-reflux association, defined as symptom association probability at least 95% and/or symptom index at least 50% for the index symptom, categorized as positive versus negative; (3) symptom phenotype, categorized as predominantly typical (heartburn or regurgitation) versus atypical or extra-esophageal (cough, throat symptoms, non-cardiac chest pain); (4) PPI response, defined as at least 50% relief of the dominant symptom on standard- or double-dose PPI for at least 8 weeks, categorized as good versus poor; (5) esophageal body motility on high-resolution manometry using the Chicago Classification version 4.0, categorized as normal versus ineffective esophageal motility; (6) hiatal hernia size measured as axial hernia length on endoscopy, manometry or imaging, categorized as at least 3 cm versus less than 3 cm; (7) erosive esophagitis by Los Angeles grade at endoscopy, categorized as erosive (grades A to D) versus non-erosive; (8) body mass index from measured weight and height at baseline, categorized as at least 30 versus less than 30 kg/m2; (9) psychological burden, defined as a Hospital Anxiety and Depression Scale anxiety or depression subscale score of at least 8 and/or elevated somatization on the PHQ-15, categorized as high versus low; and (10) symptom duration, categorized as at least 5 years versus less than 5 years.

SURGICAL TECHNIQUE All operations are performed or directly supervised by experienced foregut surgeons using a standardized laparoscopic technique: complete hiatal dissection with preservation of the vagi, circumferential mobilization of the distal esophagus to achieve at least 2.5 cm of tension-free intra-abdominal length, posterior crural approximation, and division of short gastric vessels as required to construct a tension-free wrap. A complete (Nissen, 360-degree) fundoplication, fashioned as a short floppy 1.5-2 cm wrap calibrated over a 56-60 Fr bougie, is performed in patients with normal esophageal body motility. A posterior partial (Toupet, 270-degree) fundoplication is performed in patients with ineffective esophageal motility. Wrap type is recorded and included as a covariate and subgroup variable. Standardized antibiotic and venous thromboembolism prophylaxis, analgesia, and a stepwise postoperative diet are applied.

SCHEDULE OF ASSESSMENTS Assessments are performed at screening and baseline, at operation, and at 1, 3, 6 and 12 months postoperatively. Baseline assessment comprises informed consent and eligibility, demographics and history, upper gastrointestinal endoscopy, off-PPI 24-hour pH-impedance monitoring, high-resolution manometry, GERD-HRQL, GerdQ, HADS and PHQ-15, anthropometry, and antisecretory medication use. GERD-HRQL, GerdQ and medication use are repeated at 1, 3, 6 and 12 months; HADS and PHQ-15 and patient satisfaction at 6 and 12 months; anthropometry at 12 months. Adverse events are recorded at every postoperative contact. Postoperative endoscopy and pH-impedance are performed selectively, for persistent symptoms or as part of the objective-control substudy.

SAMPLE SIZE The primary analysis is a multivariable logistic regression predicting responder status. The number of candidate predictors in the final model is constrained by the number of events in the less frequent outcome category, applying a minimum of ten events per variable. Assuming a conservative non-responder proportion of 20% and a final model containing up to eight predictors, at least 80 non-responder events are required, corresponding to 400 patients with complete 12-month follow-up. Allowing for approximately 15% attrition, the target enrollment is 470 patients. With 400 evaluable patients, a predictor present in roughly 30% of the cohort is detectable with about 80% power at a two-sided alpha of 0.05 for an adjusted odds ratio of approximately 2.0, and the target satisfies published criteria for the minimum sample size required to develop a multivariable prediction model with acceptable shrinkage and precision.

STATISTICAL ANALYSIS Continuous variables are summarized as mean with standard deviation or median with interquartile range according to distribution, and categorical variables as counts and percentages. Responders and non-responders are compared using the t-test or Mann-Whitney U test and the chi-squared or Fisher exact test as appropriate. A multivariable binary logistic regression model is fitted with responder status as the dependent variable; candidate predictors are specified a priori, variables with a univariable association at p less than 0.10 and those of established clinical importance are entered up to the limit of eight predictors, and age and sex are forced into the model. Results are reported as adjusted odds ratios with 95% confidence intervals and two-sided p-values at an alpha of 0.05. Linearity of continuous predictors is assessed using restricted cubic splines where appropriate, and multicollinearity is checked using variance inflation factors. Discrimination is quantified by the C-statistic, calibration is assessed using a calibration plot and the Hosmer-Lemeshow test, and overall fit is summarized by the Nagelkerke R-squared. Internal validation uses bootstrap resampling with 1,000 replications to estimate optimism, derive a uniform shrinkage factor, and report an optimism-corrected C-statistic. A simplified integer risk score may be derived from the shrunken regression coefficients. If missingness exceeds 5% and is plausibly at random, multiple imputation by chained equations is used with results pooled by Rubin's rules, and a complete-case analysis is reported as a sensitivity analysis. Prespecified subgroup analyses include complete versus partial wrap and typical versus atypical symptom phenotype. Secondary binary outcomes are analyzed with logistic regression, and time to symptomatic recurrence is explored using Kaplan-Meier methods and Cox regression. No interim analysis is planned. Analyses will be performed in R version 4.3 or later and/or IBM SPSS Statistics.

BIAS MINIMIZATION Selection bias is reduced by enrolling consecutive eligible patients and reporting participant flow per STROBE. Outcome ascertainment uses validated self-administered instruments with prespecified definitions, and outcome assessors are, where feasible, independent of the operating team. Assessors are trained on the operational definitions, and source-data verification is performed on a random sample of records. All predictors and the analysis plan are specified before data lock to avoid data-driven model selection.

ETHICS The study is conducted in accordance with the Declaration of Helsinki and applicable local regulations. Ethical approval was granted by the Medical Research Ethics Committee (Institutional Review Board) of the Faculty of Medicine, Minia University (approval number 2132:9/26, dated 02 September 2026). Written informed consent is obtained from every participant. Because the study is observational and surgery is performed on routine clinical indications, it carries no incremental procedural risk beyond standard care; the principal burden is completion of questionnaires.