The risk of hypoglycemia (low blood sugar) associated with the treatment of diabetes increases with age. Hypoglycemia is a common reason for admission to hospital for older patients with diabetes. Older patients are often unaware that their blood sugar is low and asymptomatic hypoglycemia, as assessed by continuous blood sugar monitoring, is frequent in the elderly. There is also evidence that older people with diabetes are more likely to develop cardiovascular events such a heart attack and more likely to die suddenly when compared to older people without diabetes. It is possible that low blood sugar levels contribute to the increased frequency of these events, but this possibility has never been studied. The purpose of this study is to assess how frequently low blood sugar occurs in older patients with diabetes and to see if low blood sugars adversely affect heart function in these patients.

Investigators propose a pilot study in 20 patients over age 70 who have type 2 diabetes and are being treated with insulin. These patients will be enrolled from the elderly diabetes clinic at Vancouver General Hospital. Patients will be asked to come to the Gerontology research laboratory in the Research Pavilion at Vancouver General Hospital. A trained research nurse will instrument each patient with an iPro2 glucose sensor (Medtronic Canada). These sensors reliably and continuously measure blood glucose for periods of up to 7 days. Briefly, the skin will be swabbed with a disinfectant and a small catheter will be inserted subcutaneously. This needle will be attached to a glucose sensor. Patients will wear this sensor for 6 days. Patients will measure their glucose using a glucometer 4 times each day during this period and will also be given a log book to keep track of glucose values. Patients will also be asked to record any symptoms of hypoglycemia. At the end of 6 days the sensor will be removed.

At the same time the glucose sensor is started, a trained research nurse will connect the patient to a single use CardioSTAT ECG recorder. The archived ECG waveforms will be downloaded for QT and T-wave alternans analysis. The heart rate and QT interval will be measured at baseline and the end of each interval from the digitized ECG.

Patients and relevant family will be provided with in person education regarding the function and use of the monitor, implications for bathing and sleep, and contact information for troubleshooting. Patients will be asked to change the leads at home twice during the 6 days of the study. Patients will be asked to keep a log book of any cardiac symptoms during the 6 days of the study, as well as their activities. The glucose and CardioSTAT monitor will undergo time synchronization to ensure ability to do correlative analysis.

This is a feasibility pilot to establish preliminary data for analysis. The CardioSTAT monitor will allow assessment of cardiac arrhythmias, myocardial ischemia and cardiovascular autonomic function. The results from the glucose sensor and the CardioSTAT monitor will be correlated and compared to each other to determine if hypoglycemia has an adverse effect on heart function. Hypoglycemia is likely to induce autonomic responses captured on the CardioSTAT monitor and concordant ischemic S-T segment changes if coronary disease is present. It may also detect any resultant arrhythmias, although ischemic arrhythmias are uncommon in 6 days of monitoring. If investigators determine that hypoglycemia has negative effects on cardiac function, investigators may able to design treatments that would prevent these effects from happening in the future.

Inclusion Criteria:

• 70 years old and over
• had type 2 diabetes for at least 5 years
• treated with insulin glargine.
• BMI of between 20 and 35 Kg/M2,
• A1c between 7 and 8.5 %.
• well controlled hypertension and hyperlipidemia.

Exclusion Criteria:

• cannot speak english or give informed consent, or cognitive impairment
• glomerular filtration rate (GFR) less than 40ml/min,
• poorly controlled Chronic Heart Failure
• active coronary artery disease
• active cerebrovascular disease although a past history of CAD or stroke will not results in exclusion.

Only 11 participants were enrolled and 1 withdrew due to skin irritation

Adverse event date were collected during study period of up to 7 days when glucose sensors were attached.