Introduction Post-mortem studies suggest that chronic pancreatitis is present in 6–12% of the population, yet the diagnosis of chronic pancreatitis is infrequent. We hypothesised that previously undetected pancreatic exocrine insufficiency is seen in unselected patients referred to secondary care gastroenterology clinics.

Methods A multicentre retrospective analysis of all gastroenterology patients tested for faecal elastase (FEL-1) between 2009–13 was performed. In Sheffield and Middlesbrough a FEL-1 <200 μg/g was defined as abnormal. Demographics, indication, co-morbidities and response to enzyme supplementation were recorded. Additionally, the findings of abdominal imaging were recorded. Prevalence of low FEL-1 was compared between the two centres (Fishers exact test). Binary logistic regression was used to determine if comorbidities could predict pancreatic insufficiency.

Results 1887 patients (mean age 51.6, SD 16.91, 1144 females) were included. Sheffield’s group contained 1350 patients (mean age 49.1, SD 16.37, 857 females), and Middlesbrough’s 537 (mean age 57.9, SD 16.60, 287 female).

The most common indication to test FEL-1 was diarrhoea (n = 1252), followed by abdominal pain (n = 378) and weight loss (n = 125).

The overall prevalence of low FEL-1 was 11.4% (Sheffield 11.0% vs. Middlesbrough 22.9% p < 0.0001). 13.7% (n = 171/1252) of patients with diarrhoea as the predominant symptom had FEL-1 <200. Of those with abdominal pain and weight loss 12.4% (n = 47/378) and 27.2% (n = 35/125) had low FEL-1 respectively.

86.8% (n = 236) of patients with low FEL-1 had abdominal imaging, (MRI, CT or US). 50% of imaging was normal (n = 136), 33.1% (n = 90) demonstrated pancreatic pathology consistent with either chronic pancreatitis or malignancy.

Binary logistical regression showed FEL-1 <200 was strongly associated with excess alcohol intake, diabetes mellitus, intrinsic pancreatic disease (malignant or non-malignant) and HIV infection (p < 0.0001).

79% (n = 128) of patients treated with pancreatic enzyme supplementation subjectively reported benefit from therapy. 12.3% (n = 20) had no benefit and in 8.6% (n = 14) it was not possible to assess benefit from medical records.

Conclusion This is the largest study to report detection of exocrine pancreatic disease in unselected gastroenterology clinics. Exocrine pancreatic insufficiency is strongly associated with diabetes mellitus, intrinsic pancreatic disease, high alcohol intake and HIV. Creon provides symptomatic benefit for those with pancreatic insufficiency, but further work is needed to establish appropriate dosage of enzyme supplementation. Clinicians should have a low threshold for checking FEL-1.

Disclosure of Interest None Declared.