Abstract
Forty-six diabetics treated with twice-daily insulin were seen every two weeks for six months in an intensive education programme aided by regular home urine glucose testing. Control was improved with a decrease in 24-hour urinary glucose excretion (median 138 mmol/24 h (24.8 g/24 h) falling to 70 mmol/24 h (12.6 g/24 h); p less than 0.002), glycosylated haemoglobin concentration (mean 11.4 +/- SD 2.3% falling to 10.4 +/- 1.5%; p less than 0.001), and Diastix score (median 3.0 falling to 1.3; p less than 0.001). There was no reported increase in hypoglycaemia. Thirty-eight of the diabetics proceeded to a nine-month randomised cross-over study of the effect on blood glucose control of monitoring urinary glucose or blood glucose measured visually or by a reflectance meter using appropriate reagent strips. No further improvement in control was observed after home blood glucose monitoring. Nevertheless, 29 out of 37 patients preferred blood to urine glucose monitoring. During both the education and cross-over studies there was evidence of an initial improvement in control followed by deterioration. This was independent of the monitoring method used in the cross-over period and may have been due to waning enthusiasm. Despite patient enthusiasm and other reports to the contrary, home blood glucose monitoring offered no improvement in control over intensive attention and conventional urine glucose monitoring.