There are currently no early predictive biomarkers for severity of acute pancreatitis (AP) that would allow stratification of patients for potential early interventional therapies. Hyperglycaemia is frequently observed to accompany and contribute to severe AP. However, the underlying mechanism is multifactorial, including in the acute phase of injury, where elevated adrenaline, cortisol and glucagon and inflammatory cytokine-induced insulin resistance all contribute to hyperglycaemia. The investigators propose that the extent of collateral injury of pancreatic β-cells and consequent loss of insulin secretion during the course of acute pancreatitis (AP) underlies disease severity. The investigators will measure plasma C-peptide (as a reliable readout of endogenous insulin), with moment-to-moment glucose monitoring (using subcutaneous continuous glucose monitoring devices), and bacterial endotoxin (lipopolysaccharide (LPS) in a prospective cohort of 30 severe AP patient blood samples taken every 5 days for up to 5 weeks of hospitalization.

Acute pancreatitis (AP) is associated with metabolic dysregulation including dysglycaemia which may predict poor clinical outcomes. Continuous glucose monitoring (CGM) offers a novel method for assessing glycaemia continuously in real time. The percentage of time spent in normal glucose range (TIR; 3.9-10.0 mmol/L) may correlate with key outcomes, including length of hospital stay (LOS), need for critical care, and mortality. This study aims to evaluate whether TIR measured via CGM can serve as a predictive marker in AP management.

While hyperglycaemia is frequently observed to accompany and contribute to severe AP, the underlying mechanism is not fully understood and is likely to be multifactorial and change over the disease course. For example, in the acute phase of injury, the elevated adrenaline, cortisol and glucagon and inflammatory cytokine-induced insulin resistance contribute to hyperglycaemia. The investigators propose that later in the course, the extent of collateral injury of pancreatic β-cells and consequent loss of insulin secretion may also impact disease severity. This loss of insulin secretion leads to the loss of insulin-mediated pancreatic antimicrobial secretion into the gut leading to gut dysbiosis, inflammation, reduced barrier function, bacterial translocation and infected pancreatic necrosis, resulting in severe AP. This suggests that plasma insulin depletion (with extensive hyperglycaemia) may represent early predictive biomarkers and elevated plasma bacterial endotoxin may represent a late biomarker for disease severity.

This is a single-centre, single-arm, non-randomised, observational study of adults (18 years and older) admitted to Manchester Royal Infirmary with acute pancreatitis. The study aims to recruit 30 participants over 12 months. The study's primary objective is to determine the overall time spent in normal glucose range (TIR; 3.9-10.0 mmol/L) in patients admitted with acute pancreatitis. Secondary objectives include:

1. Evaluating the association between TIR and clinical severity of acute pancreatitis.
2. Evaluating the association between TIR and length of hospital stay.
3. Determining the relationship between changes in plasma c-peptide and bacterial endotoxin with clinical severity of acute pancreatitis.

Inclusion Criteria:

1. Age 18 years or over
2. Admission diagnosis of acute pancreatitis (based on Revised Atlanta Criteria)
3. Ability to provide informed consent in English

Exclusion Criteria:

1. Known diabetes mellitus
2. Use of insulin therapy before admission
3. Pregnancy
4. Contraindications to CGM (e.g., allergy to device adhesive)

The study will be conducted on the surgical and medical wards at Manchester Royal Infirmary, (Manchester University NHS Foundation Trust [MFT]). Adults with acute pancreatitis will be recruited following admission to hospital under the treating clinical team. Patients will be identified by a member of the usual clinical team based upon analysis of the acute surgical take on our electronic health record (HIVE), cross referencing abdominal pain and raised lipase and imaging when appropriate.

Anonymised primary and secondary endpoints data, following reasonable request.

12 months following study completion.

Following reasonable request to the researcher by email.