Aims
Volume expansion following reimbursement of advanced procedures typically raises concerns about quality dilution, particularly when techniques disseminate beyond specialized centers. This prospective Belgian registry investigated whether national reimbursement for radiofrequency ablation (RFA) of Barrett's esophagus affected treatment outcomes, comparing two 7-year periods before and after reimbursement implementation (2010–2016 vs 2017–2023) across 12 designated expert centres. We assessed changes in case volume, patient characteristics, treatment patterns, efficacy, and safety.
Methods
Patients were allocated according to their first RFA procedure into pre-reimbursement (2010–2016) and post-reimbursement (2017–2023) cohorts. Variables included demographics, Prague C/M classification, worst pre-RFA histology, pre-RFA endoscopic resection (ER), number of RFA sessions, efficacy outcomes (complete eradication of intestinal metaplasia [CE-IM], complete eradication of dysplasia [CE-D]), and adverse events. Efficacy required ≥1 surveillance biopsy within 12 months after final RFA. Statistical comparisons used chi-square or Fisher's exact tests for categorical variables and Mann–Whitney U tests for continuous variables.
Results
RFA case volume increased 71% following reimbursement (345 vs 589 patients, p<0.001). Post-reimbursement patients presented with shorter Barrett's segments (Prague C: 3 vs 2, p=0.039; M: 5 vs 4, p=0.031) and lower HGIN prevalence (62% vs 55%, p=0.024). Contrary to concerns about quality dilution, efficacy improved significantly (CE-IM: 83.1% vs 87.5%, p=0.048), with a strong trend for CE-D (94.0% vs 96.5%, p=0.079), despite higher volumes. Median RFA sessions remained unchanged (2, p=0.052). Safety profiles were excellent and equivalent: bleeding (5% vs 6%, p=0.62), stenosis (9% vs 8%, p=0.71), with zero RFA-related perforations in either period.
|
|
Pre (2010–2016) |
Post (2017–2023) |
p-value |
|
Patients |
345 |
589 |
<0.001 |
|
Median age (years) |
66 |
65 |
0.41 |
|
Sex (male) |
72% |
72% |
0.92 |
|
Prague C (median, IQR, min–max) |
3 (0–6), 0–15 |
2 (0–5), 0–18 |
0.039 |
|
Prague M (median, IQR, min–max) |
5 (2–8), 1–17 |
4 (2–7), 0–20 |
0.031 |
|
NDBE |
4% |
5% |
0.44 |
|
LGD |
7% |
9% |
0.28 |
|
HGIN |
62% |
55% |
0.024 |
|
EAC pT1a |
20% |
22% |
0.57 |
|
EAC pT1b |
7% |
9% |
0.33 |
|
Any ER |
58% |
61% |
0.55 |
|
EMR |
55% |
56% |
0.77 |
|
ESD |
3% |
5% |
0.18 |
|
Median RFA sessions (IQR, min–max) |
2 (1–3), 1–5 |
2 (1–3), 1–5 |
0.052 |
|
CE-IM |
83.1% |
87.5% |
0.048 |
|
CE-D |
94.0% |
96.5% |
0.079 |
|
Bleeding |
5% |
6% |
0.62 |
|
Stenosis requiring dilation |
9% |
8% |
0.71 |
|
Perforation |
0 after RFA; 4 after ER |
0 after RFA; 4 after ER |
1.00 |
Conclusions
RFA efficacy improved despite a 71% post-reimbursement volume increase, likely driven by strict centralisation within accredited expert centres and earlier-stage disease at presentation. These findings challenge assumptions about volume–quality trade-offs and show that centralised delivery can expand access while improving outcomes.