Issue 137 Item 5 Bedtime Snacks Only For Diabetics with Low Nocturnal Glucose Le

Bedtime glucose levels should determine whether a bedtime snack is necessary for adults with insulin-dependent diabetes, say Canadian researchers. Investigators from three hospitals affiliated with McGill University, Montreal, Quebec, recruited 15 adults with type 1 or insulin-dependent diabetes as participants in a randomized, placebo-controlled, cross-over trial to determine the impact of four bedtime snack compositions on nocturnal glycemic control.

The adults were undergoing intensive insulin management using lispro insulin before meals and NPH, insulin at bedtime. The frequency of hypoglycemia below 72 mg/Dl or 4 mmol/L and morning hyperglycemia above 180mg/Dl or 10mmol/L was also assessed in their study.

Four strategies were used: a standard snack consisting of two starch and one protein exchange, two snacks that used raw cornstarch or pure protein to substitute for 15 grams of carbohydrate (one starch exchange), and no snack. The snacks were all equivalent in calories, fat and total available glucose.

Intravenous blood glucose sampling was undertaken hourly throughout the night.

The investigators found that the majority of nocturnal hypoglycemic events were seen in patients with low glycemic levels (<126mg/Dl or 7 mmol/L) at bedtime who did not eat a snack. Patients who ate either a standard or protein snack had no nocturnal hypoglycemic events, no matter the bedtime glycemic level.

In addition, no nocturnal hypoglycemic events were seen in any patient whose bedtime glycemic level was normal (>180 mg/Dl or 10 mmol/L), whether or not they ate a snack. But almost half the episodes of morning hyperglycemia occurred in patients with this bedtime level.

The researchers concluded that no bedtime snack was necessary for those whose bedtime glucose level was above 180mg.Dl or 10 mmol/L. Any snack is advisable at levels between 126 and 180mg/Dl or 7 and 10 mmol/L, and below 126mg/Dl or 7mmol/L a standard or protein snack is recommended. Diabetes Care Jan 2003;26:9-15