The reference surveillance method in patients with Barrett's esophagus is careful endoscopic observation, with targeted as well as random 4-quadrant biopsies.
Autofluorescence endoscopy may make it easier to locate neoplasia.
Dr Borovicka and colleagues from Switzerland elucidated the diagnostic accuracy of surveillance with Autofluorescence endoscopy-guided plus 4-quadrant biopsies vs the conventional approach.
The research team enrolled a total of 187 of 200 consecutive Barrett's esophagus patients.
Of these patients, 73% were male with a mean age 67 years, who underwent endoscopy for Barrett's esophagus in 4 study centers.
In phase 1, patients were randomly assigned to either-targeted biopsy followed by 4-quadrant biopsies or conventional endoscopic surveillance with 4-quadrant biopsies.
The team excluded patients with early cancer or high-grade dysplasia, who underwent endoscopic or surgical treatment.
| Autofluorescence endoscopy yielded high-grade dysplasia rates of 12%|
Patients who declined to participate in phase 2 of the study were also excluded.
The research team reported that 130 patients remained.
These patients were examined again with the alternative method after a mean of 10 weeks, using the same methods described.
The team's main study parameter was the detection of early cancer, adenocarcinoma or high-grade dysplasia.
The secondary aim in the second phase was the additional value of the Autofluorescence endoscopy-guided approach after conventional surveillance.
The researchers derived test accuracy measures from the first study phase.
The research team found that the Autofluorescence endoscopy yielded adenocarcinoma/high-grade dysplasia at rates of 12%.
The team noted that conventional approaches yielded adenocarcinoma/high-grade dysplasia rates of 5% on a per-patient basis.
With Autofluorescence endoscopy, previously unrecognized adenocarcinoma or high-grade dysplasia lesions were identified in 4% of the patients.
The researchers observed new lesions in 1% with the conventional approach.
Of the 19 adenocarcinoma/high-grade dysplasia lesions detected during Autofluorescence endoscopy in the first study phase, 8 were visualized.
A further 11 were only detected using untargeted 4-quadrant biopsies, giving a sensitivity of 42%.
Of the 766 biopsies classified at histology as being nonneoplastic, 58 appeared suspicious.
The specificity was 92%, with a positive predictive value of 12%, and a negative predictive value 99%.
In the second phase using Autofluorescence endoscopy, the team detected 2 further lesions in addition to the initial alternative approach in 3% of cases.
Conventional endoscopy identified only 2% of lesions.
Dr Borovicka's team concluded, “In this referral Barrett's esophagus population with a higher prevalence of neoplastic lesions, the Autofluorescence endoscopy -guided approach improved the diagnostic yield for neoplasia.”
“In comparison with the conventional approach using 4-quadrant biopsies.”
“However, Autofluorescence endoscopy alone was not suitable for replacing the standard 4-quadrant biopsy protocol.”