I have always asked doctors what is the ideal interval between taking an insulin injection and starting a meal. Because regular insulin does not lower blood glucose immediately after injection many physicians recommend an injection-meal interval (IMI).
By asking patients to inject well before beginning a meal, these physicians hope to compensate for the lag time between the injection of insulin and its onset of action.
Hubert Overmann, Lutz Heinemann tried to find out what physicians recommend to their patients with respect to the IMI when prescribing intensive insulin therapy. A total of 58 diabetologists were surveyed by means of a structured questionnaire.
A fixed IMI of 15 (0–30) min [median (range)] was recommended by 29% of the 58 diabetologists, and a flexible IMI was recommended by 71%. The minimal interval for the suggested flexible IMI was 0 min and the maximal interval 45 min (median 23 min).
The researchers compared these results with findings of 192 patients with type 1 diabetes from a population based study. In this study patients were asked by questionnaire about their daily life handling of the IMI.
Among the group of 134 patients reporting use of a flexible IMI, 62% used an IMI of ≤15 min, 16% one of 20–25 min, and 21% one of ≥30 min. There were 12 patients using a flexible IMI who adapted it so frequently that they could not state a typical interval. A total of 58 patients (30%) used a fixed IMI (67% used an IMI of ≤15 min, 7% one of 20–25 min, 26% one of ≥30 min).
The surveys showed that diabetologists advocating intensive insulin therapy usually recommend an IMI shorter than 30 min. The majority of patients (75%) with type 1 diabetes use an IMI of <30 min in daily life.
In sum, most doctors say the interval should be between 15 to 30 minutes. However, some recommend that a patient must wait till the action of the insulin is at its optimum (determined by taking a reading at 15-minute intervals) before starting a meal.
But such recommendations just cause more confusion. Still, for good BS control, an insulin-dependent diabetic would be better off knowing when to start eating.
I came across this oft-cited article by ME Lean, LL Ng, and BR Tennison and its summary in the British Medical Journal that sheds some light on the subject. Though no clear-cut answers are provided, it's good to know what researchers have discovered.
In a survey of 225 diabetics treated with insulin, 24 (10.6%) claimed never to have received advice concerning the interval between insulin injection and eating. Of the remainder, 67 (33%) admitted disregarding advice and using shorter intervals.
There was a significant (p less than 0.01) difference between the reported frequencies of clinical hypoglycaemia in patients using different intervals.
The effects on glucose control of intervals between insulin injection and breakfast of zero, 15, 30, and 45 minutes were studied for periods of one week in 11 patients with type I diabetes who were receiving twice daily injections of monocomponent porcine insulins and high fibre, high carbohydrate diets, using standard home blood glucose monitoring techniques to measure blood glucose concentrations each morning.
The delay of 45 minutes resulted in the lowest frequency of hypoglycaemia and the most acceptable pattern of glucose concentrations measured one and two hours after breakfast and before lunch. Combining results obtained at these three times, the mean increment in blood glucose concentration was smaller after allowing a delay of 45 minutes than after delays of zero (p less than 0.001), 15 (p less than 0.03), and 30 (NS) minutes.
A delay of 30 minutes resulted in smaller mean increments in blood glucose concentration than did delays of zero (p less than 0.001) and 15 (NS) minutes. These results suggest that this aspect of diabetic management may be neglected, with important consequences for blood glucose control.
An increase in delay between insulin injection and eating to 45 minutes would be a simple and safe way of improving blood glucose control in at least the 37% of the diabetic population surveyed in this study who currently allow less than 15 minutes.