Impact of oesophagitis classification in evaluating healing of erosive
oesophagitis after therapy with proton pump inhibitors: a pooled analysis.
Author(s): Yaghoobi M, Padol S, Yuan Y, Hunt RH.
Affiliation(s): Department of Medicine, University of Toronto, Toronto, Ontario, Canada.
Publication date & source: 2010, Eur J Gastroenterol Hepatol. , 22(5):583-90
BACKGROUND AND AIMS: The results of clinical trials with proton pump inhibitors
(PPIs) are usually based on the Hetzel-Dent (HD), Savary-Miller (SM), or Los
Angeles (LA) classifications to describe the severity and assess the healing of
erosive oesophagitis. However, it is not known whether these classifications are
comparable. The aim of this study was to review systematically the literature to
compare the healing rates of erosive oesophagitis with PPIs in clinical trials
assessed by the HD, SM, or LA classifications.
METHODS: A recursive, English language literature search in PubMed and Cochrane
databases to December 2006 was performed. Double-blind randomized control trials
comparing a PPI with another PPI, an H2-RA or placebo using endoscopic assessment
of the healing of oesophagitis by the HD, SM or LA, or their modified
classifications at 4 or 8 weeks, were included in the study. The healing rates on
treatment with the same PPI(s), and same endoscopic grade(s) were pooled and
compared between different classifications using Fisher's exact test or chi2 test
where appropriate.
RESULTS: Forty-seven studies from 965 potential citations met inclusion criteria.
Seventy-eight PPI arms were identified, with 27 using HD, 29 using SM, and 22
using LA for five marketed PPIs. There was insufficient data for rabeprazole and
esomeprazole (week 4 only) to compare because they were evaluated by only one
classification. When data from all PPIs were pooled, regardless of baseline
oesophagitis grades, the LA healing rate was significantly higher than SM and HD
at both 4 and 8 weeks (74, 71, and 68% at 4 weeks and 89, 84, and 83% at 8 weeks,
respectively). The distribution of different grades in study population was
available only for pantoprazole where it was not significantly different between
LA and SM subgroups. When analyzing data for PPI and dose, the LA classification
showed a higher healing rate for omeprazole 20 mg/day and pantoprazole 40 mg/day
(significant at 8 weeks), whereas healing by SM classification was significantly
higher for omeprazole 40 mg/day (no data for LA) and lansoprazole 30 mg/day at 4
and 8 weeks. The healing rate by individual oesophagitis grade was not always
available or robust enough for meaningful analysis. However, a difference between
classifications remained.
CONCLUSION: There is a significant, but not always consistent, difference in
oesophagitis healing rates with the same PPI(s) reported by the LA, SM, or HD
classifications. The possible difference between grading classifications should
be considered when interpreting or comparing healing rates for oesophagitis from
different studies.
|