Exercise Physiology Study

Overview

People with type 1 diabetes often find exercise very difficult to manage, because of the high risk for low blood glucose levels. This can occur very quickly once exercise starts and presents many risks for subjects, such as severe symptoms, confusion, passing out, seizures, and even coma or death in very severe cases. Preventing low blood glucose levels during and after exercise is important because physical exercise is a key component of managing diabetes. It is often hard to correctly adjust insulin infusion rates or doses before exercise as the relationship between exercise and changes in glucose levels in those who have type 1 diabetes is still not fully understood. Therefore, the investigators propose this study to further our understanding in this area. This study is designed to help separate the effects of insulin from those of muscle work (non-insulin effects) on the changes in blood glucose levels during aerobic exercise. The main hypothesis is that the non-insulin effects occur quickly during exercise and account for the rapid change in blood glucose levels once aerobic exercise begins. These effects can be separated from the slower changes in insulin sensitivity that occur because of exercise, and which account for reduced insulin demand even after exercise has stopped. The investigators will investigate the effects of both moderate and intense aerobic exercise at different levels of insulin in the body to help separate the insulin and non-insulin effects. The investigators wish to recruit 26 subjects to take part in this study. Subjects will be randomly divided into two groups, with 13 in each group. Group 1 will undergo moderate aerobic exercise, while group 2 will undergo intense aerobic exercise. Each subject will repeat the exercise study three times on three separate days at least 2 weeks apart, while having insulin infused at a low, a medium, and a high rate. Subjects will have an IV line placed in each arm, one for drawing blood relatively frequently during the study, and another for infusion of insulin, glucose, and a special glucose tracer (non-radioactive). Each study lasts about 9 hours. Information from this study will be used to help develop a mathematical model of how glucose changes during exercise in type 1 diabetes. Such a model of type 1 diabetes and exercise will be very useful for adjusting insulin doses in patients who use multiple daily injections of insulin, and can help to guide an automated insulin delivery system, such as the artificial pancreas.

Diabetes mellitus afflicts close to 10% of our population and 5% of those with diabetes have type 1, which is defined by an absolute deficiency of insulin. The need for managing diabetes is critical, given the economic burden of this disease, with over $175 billion dollars in direct health care costs, and almost another $70 billion in indirect costs for disability and work loss. The personal impact is equally as important for people with this disease, as diabetes mellitus is the leading cause of blindness, the need for kidney dialysis, and non-traumatic amputations in the United States. Type 2 diabetes is associated with reduced insulin sensitivity and the metabolic syndrome, and dietary modification and exercise are important components in the management of underlying insulin resistance. However, these lifestyle strategies are also important in type 1 diabetes for many reasons: 1) type 1 diabetes subjects now live into adulthood, when insulin resistance and obesity become factors for glycemic control, 2) latent autoimmune diabetes of adulthood (LADA) represents a "mixed" form of autoimmune diabetes where some type 2 diabetes characteristics such as insulin resistance can exist, and 3) dietary modification and exercise remain effective means for management of acute hyperglycemia and, in the longer term, HbA1c, potentially reducing the risk of microvascular complications. Therefore, the need for exercise is still evident in subjects with type 1 diabetes to maintain good glycemic control and to prevent complications from developing. However, exercise is challenging for people with T1D to manage. Exercise causes increased insulin sensitivity along with rapid uptake of glucose by muscle and other tissues within the body, leading to a sharp decline in glucose levels and hypoglycemia as shown by other groups as well as ours. Without adjustments in insulin for exercise, hypoglycemia is common in persons with type 1 diabetes. In a study of 48 individuals with T1D, with no adjustments to insulin, who exercise for 60 min at a moderate intensity, glucose levels dropped on average by 40%, with 52% of subjects falling to 70 mg/dL or below. Despite this clear need for insulin adjustments for exercise, there are no uniform recommendations on how to dose insulin around the time of exercise. In 2006, the DirecNet Study Group published a study on the impact of suspending basal insulin at the start of exercise in 40 children with type 1 diabetes on insulin pump therapy. This intervention significantly reduced hypoglycemia (from 43% to 16%), but much more commonly resulted in hyperglycemia (increased from 4% to 23%). Schiavon and Cobelli et al addressed this issue of how to best adjust insulin for exercise using in silico simulations. Adjusting insulin doses in the in silico environment decreased hypoglycemia from 88% to 16% of patients when a universal adjustment was applied, and to 4% when an individual adjustment was applied. The study described within this protocol is designed to disambiguate the impact of exercise on insulin and non-insulin mediated effects on glycemic control. To achieve this, the investigators will perform a series of stable glucose tracer studies in which subjects will be fasting for about 8 hours and will undergo aerobic exercise at a moderate and intense level for 45 minutes while insulin rates are clamped at a low (subject's basal rate), medium (basal x 1.5), and high (basal x 3) insulin infusion rate. Subject's basal rates will be obtained from injected basal insulin amounts, such as NPH/glargine/detemir, or basal rates in those who use insulin pumps and will be adjusted for the HbA1c, as described in the OHSU AP system. Di-deuterated glucose (6,6-2H2-glucose) which is not radioactive and which can be metabolized via usual pathways in the human body will be the stable tracer. Each subject per arm will undergo 3 10-hour studies while blood glucose, insulin, and glucagon levels are captured throughout the study, and catecholamine and fatty acid levels are captured during and just after the exercise period, as outlined below. Glucose tracer levels will be measured at OHSU through the Bioanalytical Shared Resource/PK core lab, and calculation of rate of appearance (Ra) and rate of disappearance (Rd) of glucose will be performed by our colleagues at McGill University using a non-steady state model of glucose dynamics. The data obtained from this study will inform an updated model of glucose regulation in type 1 diabetes, providing exercise as an input to the model, which will be utilized in a model predictive control (MPC) system for managing type 1 diabetes. Such a system can be used to deliver insulin and/or glucagon to manage glycemic changes during and outside of exercise.