Real gains but practical limitations to glycaemic control with insulin in type 2 diabetes

Background Insulin is increasingly used for type 2 diabetes when oral therapy is inadequate. We have examined the results in unselected patients in a UK hospital diabetes clinic. Methods Clinical database records from 1994 to 2002 were analysed for anthropomorphic data, blood pressure, HbA1C and plasma cholesterol in type 2 diabetic patients starting insulin therapy. Results In 335 patients, HbA 1C at four years correlated positively with HbA1C before starting insulin (r=0.31, p<0.01) and negatively correlated with age at the time of starting insulin (r=-0.19, p<0.01). In two cohorts changed to insulin therapy before and after publication of the results of the UKPDS (1998), HbA1C improved in the first year, with little subsequent change over four years, while weight continued to rise. The later cohort had a significantly greater reduction in HbA1C (p=0.001). Even in this group, only 21.6% of patients achieved HbA1C < 7.5%. Conclusions In a UK diabetes clinic, insulin therapy for type 2 diabetes improved glycaemic control in the first year, but not thereafter, while weight gain continued in subsequent years. Tighter targets may have promoted improved results.