Optimization of impedance-measured reflux events in GORD utilizing acid exposure time

Background: Combining impedance with pH monitoring improves the detection and characterization of gastro-oesophageal reflux (GOR), yet the two modalities frequently differ in GOR quantification. Ambulatory 24-h pH-impedance monitoring often reveals more significant oesophageal acid exposure than impedance-measured reflux activity in patients with symptomatic gastro-oesophageal reflux disease (GORD). The purpose of this study is to elucidate the discrepancies between these modalities by assessing the predictive accuracy of impedance compared to acid exposure standards.

Methods: A single-institution, retrospective review of sequential 24-h pH-impedance results of 72 patients with symptomatic GOR off anti-secretory therapy was conducted. Reflux events measured by impedance were stratified by patient position and compared to oesophageal acid exposure time (AET). Oesophageal AET limits for GORD detection were utilized as gold standards to generate serial receiver operator characteristics (ROC) curves to assess the sensitivity and specificity of current impedance GORD detection limits and identify optimized impedance standards based on area under the curve (AUC) analysis.

Results: Mean total AET time was 10.5% (± 9.9%), and 63.8% of patients had elevated AET. By impedance, median GOR frequency was 43 (IQR 21-68), and 22.2% exceeded conventional GOR frequency limits of normal. ROC curve analysis revealed the current impedance standard of > 73 GOR events has a sensitivity of 32.6% and specificity of 96.5% (AUC 0.74) for GORD detection. By AUC analysis, an impedance threshold of > 41 GOR events is optimal for GORD detection (sensitivity 69.6%, specificity 80.7%, AUC 0.83).

Conclusion: Conventional impedance standards for abnormal GOR frequency are weakly sensitive for the detection of GORD, providing a possible explanation to discrepancies in AET and impedance interpretation. Lowering impedance-measured GOR frequency limits to > 41 optimizes sensitivity and specificity while increasing congruence between pH and impedance metrics.