This article is more than 5 years old. Information may no longer be current.
ASGE endorses real-time imaging-assisted endoscopic targeted biopsy for BE surveillance
Click Here to Manage Email Alerts
Targeted biopsies guided by advanced imaging techniques for detecting dysplasia during surveillance of patients with Barrett’s esophagus met the performance thresholds set by the ASGE PIVI, according to the results of a meta-analysis.
These findings led the ASGE Technology Committee to endorse the use of these advanced imaging techniques to guide targeted biopsies for BE surveillance, replacing the current protocols for random biopsy.
“The ASGE has identified endoscopic real-time imaging of BE as a key area for new endoscopic technologies and has outlined ... the performance thresholds for an imaging technology with targeted biopsies to eliminate the need for random biopsies during endoscopic surveillance of BE,” Barham K. Abu Dayyeh, MD, MPH, of the Mayo Clinic in Rochester, Minn., Nirav Thosani, MD, MHA, of the Ertan Digestive Disease Center of Excellence, Memorial Hermann Hospital, Texas Medical Center, Houston, and McGovern Medical School, UTHealth, and committee members wrote.
Barham K. Abu Dayyeh
Nirav Thosani
These thresholds required the imaging technology with targeted biopsy to have 80% specificity, a per-patient sensitivity of at least 90% and a negative predictive value of at least 98% for detecting high-grade dysplasia or early esophageal adenocarcinoma vs. standard protocols.
To determine whether a number of advanced imaging technologies meet the PIVI thresholds, Abu Dayyeh, Thosani and colleagues performed a systematic review and meta-analysis of literature published through August 10, 2015. They included 25 studies in the meta-analysis, evaluating chromoendoscopy, electronic chromoendoscopy using narrow-band imaging (NBI) with or without autofluorescence imaging, and confocal laser endomicroscopy (CLE).
For acetic acid chromoendoscopy, the pooled sensitivity was 96.6% (95% CI, 95-98), negative predictive value was 98.3% (95% CI, 94.8-99.4) and specificity was 84.6% (95% CI, 68.5-93.2). Methylene blue and indigo carmine chromoendoscopy did not meet the PIVI thresholds.
For electronic chromoendoscopy using NBI, the pooled sensitivity was 94.2% (95% CI, 82.6-98.2), negative predictive value was 97.5% (95% CI, 95.1-98.7) and specificity was 94.4% (95% CI, 80.5-98.6). Electronic chromoendoscopy with autofluorescence imaging did not meet the PIVI thresholds, and electronic chromoendoscopy using FICE or i-Scan were not evaluated in the meta-analysis due to limited data.
For endoscope-based CLE, the pooled sensitivity was 90.4% (95% CI, 71.9-97.2), negative predictive value was 98.3% (95% CI, 94.2-99.5) and specificity was 92.7% (95% CI, 87-96). However, the endoscope used in these studies is no longer available. Probe-based CLE did not meet the PIVI thresholds.
The committee concluded that their findings confirm “that the thresholds set by ASGE PIVI for real-time imaging-assisted endoscopic targeted biopsy during endoscopic surveillance of BE have been met by acetic acid chromoendoscopy, [NBI], and eCLE, at least by endoscopists with expertise in advanced imaging techniques at referral centers. The ASGE Technology Committee therefore endorses the use of these modalities during surveillance of nondysplastic BE for obtaining targeted biopsy specimens by endoscopists proficient in these techniques.”
Although autofluorescence imaging and pCLE showed promise, they have not met the PIVI thresholds, and further research is needed on i-Scan, FICE, optical coherence tomography, high-resolution microendoscopy, endocytoscopy and spectroscopy, they added. – by Adam Leitenberger
Disclosure: Abu Dayyeh and Thosani report no relevant financial disclosures. Please see the full study for a list of all other committee members’ relevant financial disclosures.
Click Here to Manage Email Alerts