University of Missouri - Columbia Lexington, Kentucky
Lekan Akanbi, MD1, Mazen Elsheikh, MBBCh2, Ujas Patel, MD3, Stefanie Stephens, RN3, Kathleen Wong, RN3, Margaret A. Pylmale, DNP, RN-BC3, Scott Roth, MD3, Lisbeth Selby, MD3, Bahaaeldeen Ismail, MD, MSc3; 1University of Missouri - Columbia, Lexington, KY; 2Ain Shams University, Cairo, Al Qahirah, Egypt; 3University of Kentucky, Lexington, KY
Introduction: Previous studies showed adequate ability of baseline impedance measured during high-resolution impedance manometry (HRIM)to distinguish patients with gastroesophageal acid reflux disease (GERD) from controls. However, its role in erosive esophagitis or Barrett’s esophagus is uncertain. We investigated the accuracy of baseline impedance and a new metric; baseline total impedance (BTI) ratio in differentiating patients with erosive esophagitis or Barrett’s from those without. Methods: We retrospectively included patients with pH study proven GERD, who underwent an upper endoscopy and HRIM at the University of Kentucky between 7/2012 to 10/2019. Baseline total lower impedance was calculated using the manometry smart mouse tool as the mean impedance for the 3 cm above the lower esophageal sphincter, starting at the beginning of the baseline period till the end of the tenth protocol swallow (Figure 1). These steps were repeated for the 3 cm below the upper esophageal sphincter to measure the total upper impedance. The BTI ratio (upper: lower) was then calculated. The ability of these metrics to detect endoscopic changes was assessed using receiver operator curves (ROC). Results: We identified 22 cases with moderate to severe esophagitis (grade B, C or D) and/or Barrett’s and 29 controls. BTI ratio was higher in cases; median 2.64 (IQR 1.85-4.02) versus 1.71 (1.20-2.17); p=0.001 (Table 1). The ability to identify Barrett’s/esophagitis (grade B, C or D) was higher for BTI ratio (area under the curve [AUC] 0.762) than for the conventional baseline impedance method (AUC 0.646) (Figure 2A). A BTI ratio threshold of 2.37 had a moderate accuracy for differentiation with sensitivity 63.6%, specificity 82.8%, positive predictive value (PPV) 73.6% and negative predictive value (NPV) 75%.We performed a secondary analysis to assess the ability of BTI ratio to detect Barrett’s and/or severe esophagitis (grade C, D) (n=9), which showed a higher diagnostic accuracy (AUC 0.92) (Figure 2B). A threshold of 2.8 had a sensitivity and specificity of 88% with a PPV of 84.8% and NPV of 90.6%. Discussion: Our proposed BTI ratio outperformed the conventional baseline impedance method in distinguishing Barrett’s and/or esophagitis. It further showed higher capacity for identification of Barrett’s and/or severe esophagitis. This metric can serve as a non-invasive option for patients undergoing HRIM or when endoscopy is considered risky. Further studies with larger sample size are needed to validate the results.
Table 1: Baseline characteristics of patients with GERD categorized by presence of erosive esophagitis and/or Barrett’s esophagus
EE: Erosive Esophagitis, SD: Standard Deviation, IQR: Interquartile range, AET: Acid Exposure Time
Figure 1: Demonstration of measuring the total upper (A) and lower (B) baseline impedance using the smart mouse tool.
Figure 2A: Receiver operating characteristic curve of conventional baseline impedance and baseline total impedance ratio to discriminate esophagitis (B-D) and/or Barrett's from controls. Figure 2B: Receiver operating characteristic curve of baseline total impedance ratio to discriminate severe esophagitis (C-D) and/or Barrett's from controls.
Disclosures: Lekan Akanbi indicated no relevant financial relationships. Mazen Elsheikh indicated no relevant financial relationships. Ujas Patel indicated no relevant financial relationships. Stefanie Stephens indicated no relevant financial relationships. Kathleen Wong indicated no relevant financial relationships. Margaret Pylmale indicated no relevant financial relationships. Scott Roth indicated no relevant financial relationships. Lisbeth Selby indicated no relevant financial relationships. Bahaaeldeen Ismail indicated no relevant financial relationships.