Frequent self-monitoring of blood glucose had a significant effect on HbA1c only when combined with frequent on-line reporting, in a study from Canada.

A total of 229 patients with type 2 diabetes, treated with oral hypoglycaemic agents only and registered with the Internet-based Blood Glucose Monitoring System, were evaluated. The patients had similar baseline characteristics.

They were asked to test their blood glucose before meals, and to submit a standardised report to the endocrinologist every two weeks, which showed the mean glucose and standard deviation of values.

The endocrinologist was asked to reply to each report with a brief feedback of two to four sentences.

As well as the on-line advice, the patients received all the normal face-to-face care expected.

Data on 191 patients who produced results eligible for the study were analysed. The researchers then retrospectively reviewed their records for: HbA1c, frequency of self-monitoring of blood glucose and on-line communication with an endocrinologist during the preceding six months.

The analysis divided patients into two groups according to frequency of blood glucose testing. There were 73 patients in the infrequent testers group and 118 in the frequent testers. Infrequent testing was defined as once or less per day, and frequent testing as twice or more a day. There was no significant difference in other baseline characteristics between the frequent and infrequent testing groups. Reporting frequency was defined as the average number of reports per three months over the most recent six-month period.

There was a statistically significant difference in HbA1c between the group who tested frequently and also reported frequently when compared with those who only tested frequently but did not report frequently: 

7.4±0.2% (57±2 mmol/mol vs

7.9±0.4% (63± 5 mmol/mol).

The authors recommend that on-line management of self-monitoring of blood glucose should be incorporated into practice on the basis of their results. However, the average duration of diabetes in both groups was well over ten years, making the value of tight control of HbA1c of relatively less value than tight control of blood pressure and lipids. The difference in HbA1c demonstrated was relatively modest for significant input from both patients and healthcare professionals.

Previous studies of self-monitoring of blood glucose, such as ESMON and DIGEM which used intensive but conventional follow-up of larger populations and for longer duration not only failed to demonstrate improved HbA1c levels, but also suggested worse outcomes in terms of psychosocial wellbeing.^1,2

This study seems to me to confirm that even with internet review, self-monitoring of blood glucose has a limited role in improving glycaemic control in type 2 diabetes patients on oral hypoglycaemic agents. I feel that the main role of self-monitoring of blood glucose in this group is the detection of hypoglycaemia.

Dr Matthew Lockyer