Endocrine Abstracts

Aims: Our aim was to explore factors related to the acute presentation of diabetic keto-acidosis (DKA) in patients with Type 2 diabetes mellitus (T2D), and to identify any predictors for severity of DKA in T2D.

Methods: This was a retrospective study based on acute medical admissions presenting to the Royal Berkshire Hospital during 2008. Criteria for inclusion included a confirmed diagnosis of DKA (defined by a plasma glucose level >11.1 mmol/l, ketonuria and acidosis (pH <7.3, serum bicarbonate <15 mmol/l or base excess <−10)) on a background of T2D (defined by previous use of oral hypoglycaemic agents, age >30 years and BMI >30 kg/m²). Data ascertained included HbA1c, BMI, age, sex, serum pH, bicarbonate, base excess and lactate, plasma glucose and underlying precipitant for DKA. Analyses included Pearson correlations. Data are presented as mean (S.D.).

Results: We identified 18 patients who presented acutely with DKA on a background of T2D (ten men; eight women). Mean age was 52.4 years (12.6) and time between diagnosis of T2D and acute presentation of DKA was 10.1 years (12.5). Data from the time of acute admission with DKA were as follows: BMI 36.3 kg/m² (31.4); pH 7.13 (0.15); bicarbonate 9.9 mmol/l (3.2); plasma glucose 31.1 mmol/l (12.3); lactate 5.1 mmol/l (4.5); base excess −17.5 (13.5). Mean HbA1c (within 1 month of admission) was 11.2% (2.8). The precipitant for DKA was infection in nine patients (50%), insulin omission in four patients (22%) and alcohol-related in two patients (11%). There was a significant independent inverse correlation between HbA1c and pH at admission (r²=−0.52, P=0.05).

Conclusions: DKA is a relatively common occurrence in patients with T2D, which should be borne in mind during acute management of patients with T2D. The vast majority of patients with T2D presenting in DKA had poor glycaemic control, and HbA1c correlated significantly with severity of DKA.