Aim: To follow up patients without known diabetes, but with hyperglycaemia in hospital for diabetes at one year.
Methods: 159 patients with a random plasma glucose > or = 7.8 mmol/L recorded during hospital admission were sent a questionnaire and invited to have the following test one year following discharge: fasting plasma glucose, HbA1c and fasting lipid profile. Those with a fasting plasma glucose > or = 5.5 and < 7.0 mmol/L, and/or those with a HbA1c > or = 6.0%, were asked to have an oral glucose tolerance test. Those with a fasting plasma glucose > or = 7.0 mmol/L were defined as having diabetes.
Results: There were 88 full responses. Nineteen (21.6%) had diabetes and nine impaired glucose tolerance. Hb1Ac was > or = 6% in five subjects with a fasting plasma glucose < 5.5 mmol/L. Two had impaired glucose tolerance and one diabetes. If a random plasma glucose in-hospital of 10 mmol/L is used as a threshold for later testing, as suggested by previous studies, then 25% of those with an abnormal result would have been missed.
Conclusions: A high proportion of those with hyperglycaemia in hospital have diabetes or impaired glucose tolerance at one year. Initial testing with fasting plasma glucose and HbA1c avoided oral glucose tolerance test in 76% of cases. Use of HbA1c detected otherwise missed diabetes and impaired glucose tolerance. A random plasma glucose of > or = 7.8 mmol/L in hospital targets patients who should be tested for impaired glucose tolerance or diabetes following discharge.