Barrett’s oesophagus

Intestinal metaplasia of the distal oesophagus replacing normal squamous epithelium — a premalignant condition driven by chronic GERD. Management hinges on the presence and grade of dysplasia: non-dysplastic Barrett’s enters surveillance, low-grade dysplasia (LGD) confirmed by expert pathology warrants endoscopic eradication therapy (EET), and high-grade dysplasia (HGD) or intramucosal carcinoma mandates EET. Oesophagectomy is reserved for submucosal invasion or EET failure.

diagnosis

EGD with biopsies — salmon-coloured mucosa extending proximal to the gastro-oesophageal junction (GOJ)
Seattle protocol: 4-quadrant biopsies every 1–2 cm throughout the Barrett’s segment, plus targeted biopsies of any visible lesion
Histological requirement: intestinal metaplasia with goblet cells on biopsy (North American definition; BSG accepts columnar metaplasia without goblet cells)

Prague classification

Standardised reporting of Barrett’s extent:

C = circumferential extent (cm above GOJ)
M = maximum extent (including tongues)
Example: C3M5 = 3 cm circumferential, 5 cm maximal

Short-segment Barrett’s: <3 cm. Long-segment: ≥3 cm — higher dysplasia and cancer risk.

dysplasia grading and confirmation

All dysplasia must be confirmed by a second expert GI pathologist before management decisions.

grade	definition	action
non-dysplastic (NDBE)	intestinal metaplasia, no dysplasia	surveillance
indefinite for dysplasia	reactive changes vs true dysplasia unclear	optimise PPI × 3–6 months → repeat EGD
low-grade dysplasia (LGD)	confirmed by expert review	EET (preferred) or surveillance every 6 months
high-grade dysplasia (HGD)	confirmed by expert review	EET
intramucosal carcinoma (T1a)	confined to mucosa	EET (endoscopic resection ± ablation)

LGD is over-diagnosed

Community pathology over-calls LGD in up to 73% of cases. Expert review frequently downgrades to NDBE or indefinite — always obtain second opinion before committing to eradication.

surveillance intervals

All patients on PPI therapy during surveillance. Intervals assume no visible lesions and confirmed dysplasia grade.

grade	interval
NDBE, <3 cm	EGD every 5 years
NDBE, ≥3 cm	EGD every 3 years
indefinite for dysplasia	repeat EGD at 3–6 months after PPI optimisation
LGD (confirmed, not treated)	every 6 months
post-EET (complete eradication)	3 months, 6 months, then annually for ≥2 years

when to exit surveillance

Limited life expectancy or comorbidities precluding EET or surgery — surveillance without therapeutic intent has no benefit. Have the conversation early.

endoscopic eradication therapy (EET)

indications

Confirmed LGD (after expert pathology review) — SURF (2014) showed 25% absolute risk reduction for progression to HGD/cancer vs surveillance
HGD
T1a intramucosal adenocarcinoma (well/moderately differentiated, no lymphovascular invasion)

techniques

modality	role
endoscopic mucosal resection (EMR)	first-line for visible lesions — provides histological staging (T-stage, differentiation, LVI, margins)
radiofrequency ablation (RFA)	flat Barrett’s eradication after EMR of visible lesions; AIM dysplasia (2009) — 81% complete eradication of dysplasia at 1 year
endoscopic submucosal dissection (ESD)	en-bloc resection of larger lesions (>15 mm) or when EMR cannot achieve complete resection
cryotherapy	alternative ablation for residual Barrett’s post-EMR if RFA not feasible

Sequence for nodular Barrett’s with dysplasia: EMR of visible lesion first (staging) → RFA of remaining flat Barrett’s segment → surveillance.

when EET is not enough

Refer for oesophagectomy:

Submucosal invasion (T1b) — lymph node metastasis risk 15–25%
Poorly differentiated histology
Lymphovascular invasion on EMR/ESD specimen
Positive deep margin that cannot be re-resected

risk factors for progression

Annual progression rate NDBE → adenocarcinoma: ~0.3–0.5%/year. Higher with:

Long-segment Barrett’s (≥3 cm)
Confirmed LGD (0.5–1.3%/year → HGD/cancer)
Persistent oesophagitis despite PPI
Smoking, central obesity, male sex, older age

common traps

Skipping expert pathology review for dysplasia — community over-diagnosis of LGD drives unnecessary procedures
Biopsying only visible lesions and ignoring the Seattle protocol — flat dysplasia is easily missed without systematic sampling
Stopping PPI after eradication — Barrett’s recurrence rates increase without ongoing acid suppression; continue indefinitely post-EET
Assuming T1a is always endoscopically manageable — poorly differentiated or LVI-positive T1a needs surgical evaluation
Surveillance in patients who would never be candidates for treatment — reassess goals of care before repeating EGDs

Barrett's oesophagus