P104 Clinical outcomes in severe oesophagitis

Introduction

Patients with Los Angeles (LA) grades C and D reflux oesophagitis are collectively considered to suffer from severe forms of oesophagitis. We aimed to determine if there are differences in clinical outcomes between those with grade C and grade D oesophagitis at index oesophago-gastro-duodenoscopy (OGD1).

Methods

Patients in NHS Lothian with grades C or D reflux oesophagitis at OGD1 between 01/01/14-31/12/15 were identified. Univariate analysis was performed to identify factors significantly associated with grade.

Results

In total, 964 patients had severe oesophagitis, 61.7% (n=595) grade C and 38.3% (n=369) grade D. Median age was 64.0 years; 54.6% (n=526) of patients were male. Patients with grade D oesophagitis were more likely to be a current (p=0.002) or ex-smoker (p=0.013), have a history of alcohol excess (p=0.050), have a stricture at OGD1 (p=0.024), present with a gastrointestinal bleed (p=<0.001), be inpatients (p=<0.001), live in a more deprived deprivation decile (p=0.001), be deceased at 6 (p=0.001) and 12 (p=<0.001) months post OGD1 and at time of notes follow-up (p=<0.001). 39.7% (n=236) with grade C and 40.4% (n=149) with grade D had a follow-up OGD (OGD2) specifically for healing of oesophagitis (overall median time to OGD2 10.1 weeks). Differences between grades were not statistically significant for Barrett’s oesophagus, dysplasia or oesophageal cancer at either OGD1 or OGD2, and there was no significant difference regarding diagnosis of stricture on OGD2. (Table 1).

Conclusions

Patients with grade D oesophagitis at OGD1 were not at greater risk of Barrett’s oesophagus, dysplasia or oesophageal cancer. However, the higher mortality rates in grade D patients indicate that they may have increased levels of frailty and co-morbidities. As current British Society of Gastroenterology guidance recommends repeat endoscopy in all with grade D, these data may have implications in the decision to perform further endoscopic procedures in this cohort.