This is a proposal to enter into a multicenter collaborative clinical trial to answer questions of extreme importance in the management of insulin-dependent (Type I) diabetes mellitus: 1. Can an intensive therapeutic intervention be mounted which will produce better diabetic """"""""control"""""""" than that seen in a similar group receiving conventional treatment in the same clinical setting? 2. How close to the normal metabolic state will the patients be who receive the intensive therapy? 3. Can this intervention be maintained over a prolonged period of time (e.g., issues of patient acceptibility and patient adherence, as well a long term physician and institutional commitment)? 4. What are the side effects of such an intervention, both medical and psychological? 5. Can evaluations be perfected which can be used as outcome criteria in a long-term study? Five patients in an intensive intervention group will receive insulin via continuous subcutaneous infusion with specific meal-related boluses. Blood glucose levels will be monitored three to four times a day. The results will be used to make rapid adjustments in insulin dose, diet or exercise. The frequency and """"""""intensity"""""""" of medical contacts will be increased in an attempt to achieve adherance to the program and optimize control. Control will be evaluated by a combination of weekly profile of around-the-clock glucose levels, glycosylated hemoglobin, and metabolic studies performed during brief research hospitalizations. The five patients in a control group will receive the best of conventional therapy, and will be monitored for control primarily by assessment of percent negative urines and measurement of glycosylated hemoglobin. Outcome measures for both groups will involve ophthalmologic (fundus photographs, fluorescein angiography and studies of retinal blood flow), neurologic (motor and sensory nerve function and renal evaluations.
