MO626: Risk of Type 2 Diabetes After Stress Hyperglycaemia in Surviving Neurological Patients Admitted to ICU: Assessment of Insulin Need in Pancreatic Donation

Abstract

Abstract BACKGROUND AND AIMS ‘Stress hyperglycaemia’ is a condition characterized by elevated blood glucose levels during stress or illness. It is often multifactorial and leads to transient insulin resistance and/or insulin deficiency in critical ill patients. Patients with stress hyperglycaemia could be potential pancreas donors, although there is some concern about the influence of donor insulin use on pancreas graft survival. Studies have reported that stress hyperglycaemia increases the risk toward the development of type 2 diabetes (T2D) on the long term, but the patient profiles were of those not usually considered for pancreas donation. The objective of our study was to assess the impact of insulin requirements during intensive care unit (ICU) in surviving patients who otherwise would be potentially eligible for pancreas donation. The risk of T2D at 3 years after discharge was assessed to understand the pancreas function recovery after stress hyperglycaemia. METHOD we developed a retrospective study including patients admitted in the ICU between March 2011 and December 2017 due to severe neurological acute conditions [subarachnoid haemorrhage (SAH), head trauma, non-SAH haemorrhage or stroke]. Inclusion criteria were as follows: severe neurological condition requiring need for intubation, age <56 years and survival at discharge. Exclusion criteria included: obesity (BMI ≥30), diagnosis of diabetes prior to admission, HVB, HVC or HIV positive and lost to follow-up <3 years. Baseline characteristics, length of hospital and ICU stay, peak AST/ALT and amilase/lipase, insulin use during admission as well as variables associated with hyperglycaemia (use of parenteral nutrition, infections, use of corticosteroids or propofol sedation) were collected from patient's electronic medical records of our institution. Primary endpoint was diagnosis of T2D at 3 years after discharge (HbA1C ≥6.5% or fasting plasma glucose ≥126 mg/dL). RESULTS A total of 76 patients were included, with insulin requirement during ICU admission reported in only 5 (6.6%). Median age was 45 years (IQR 35–52), 50% were males, median Charlson index was 0, and 25 (32.9%) had present or past smoking habit. Hypertension was present in 11 (14.5%) and dyslipidaemia in 7 (9.2%). The main admission cause was SAH in 45 patients (59.2%) followed by head trauma in 21 (27.6%). As for hyperglycaemic factors, during admission, 6 (7.9%) received parenteral nutrition, 48 (63%) had an infectious complication, 51 (67%) received corticosteroids and 49 (64.5%) were under propofol sedation. A total of 55 (72.4%) received noradrenaline. Median amylase, lipase, AST and ALT peak were 115 U/L (69.5–233), 70 U/L (34–222), 51 U/L (33.25–101.5) and 94 U/L (35.5–165), respectively. Median ICU and hospital stay were 10 (5–17.75) and 29 (19–42.75) days, respectively. Those who required insulin had longer ICU stay compared with the median of the overall cohort (25 versus 10 days); all of them received corticosteroids and were diagnosed with SAH. At 3 years of follow-up, only three patients (3.9%) were diagnosed with T2D, but none of them had received insulin during admission. One patient was diagnosed 7 months after discharge and the other two were diagnosed 1 year after discharge. Three patients died during the follow-up, but any of them also required insulin during admission. CONCLUSION Insulin requirement in patients with severe neurological acute conditions who could meet criteria for pancreas donation was less common than expected, despite large use of corticoids, infectious complications and propofol sedation as hyperglycaemic factors. Remarkably, only three patients were diagnosed of T2D during the follow-up and none of them received insulin during admission. Prospective studies are needed in order to understand pancreas recovery after stress hyperglycaemia.