Table 4

Diagnostic yield and relevance of random biopsies during follow-up (FU)

FindingCumulative incidenceRelevance
Patient rate
%
(95% CI) n/N*
Endoscopy rate
%
(95% CI) n/N†
Biopsy rate
%
(95% CI) n/N
FU‡;
Years
Median (P25–P75)
FU‡,
N endoscopies
Median (P25–P75)
Reproduced
%
(95% CI) n/N
Progression to LGD/HGD/EAC
%
(95% CI) n/N
Progression to HGD/EAC
%
(95% CI) n/N
NSE random biopsies
 Buried IM2.7
(1.5 to 4.8)
(10/376)
1.1
(0.6 to 2.0)
(10/924)
0.1
(0.1 to 0.2)
(10/8588)
4 (4 to 5)4 (4 to 5)0
(0 to 3.4)
(0/10)
0
(0 to 3.4)
(0/10)
0
(0 to 3.4)
(0/10)
Cardia random biopsies
 IM13.8
(11.5 to 15.5)
(150/1121)
7.2
(6.3 to 8.3)
(198/2733)
NA3 (2 to 4)3 (2 to 4)33.3
(25.8 to 41.8)
(43/129)§
2.3
(0.1 to 7.2)
(3/129)
0
(0 to 2.9)
(0/129)
 LGD0.81
(0.42 to 1.5)
(9/11 121)
0.73
(0.46 to 1.15)
(20/2733)
NA2 (2 to 5)2 (2 to 4)75.0
(35.6 to 95.6)
(6/8)¶
NA0
(0 to 40.2)
(0/8)
  • The diagnostic yield of random biopsies from NSE and cardia and long-term follow-up of abnormal findings.

  • *N = patients with at least 1 endoscopy with sampling from NSE or cardia.

  • †N = Endoscopies with sampling from NSE or cardia.

  • ‡Median FU after detection of buried BE; IM; of LGD.

  • §N=patients with IM in the cardia; either at end of treatment (n=78) or during FU (n=72). Patients with treatment (n=9) or no FU (n=12) were excluded.

  • ¶A single patient underwent additional RFA and was not included for the FU analysis.

  • **Adjusted for potential confounders age, gender, length of BE, worst pathology at baseline, reflux stenosis, incident lesion.

  • BE, Barrett’s oesophagus; IM, intestinal metaplasia; LGD, low-grade dysplasia; NSE, neosquamous epithelium; RBx, random biopsies; RFA, radiofrequency ablation.