A New Approach to Carb Counting for AP Users
Two slices of toast equals 30 grams of carbs, minus 2 grams of fiber, that’s 28 grams of carbs. A pat of butter, no carbs! A tablespoon of strawberry jam? Thirteen grams of carbs. Two cups of fresh strawberries and pineapple, 30 grams of carbs.
Is it breakfast, or is it calculus class? It’s neither. It’s carb counting and it’s a reality faced by people with type 1 diabetes (T1D) every day.
Carb counting is a cornerstone of blood glucose management. It enables people with T1D to calculate their insulin needs for the day, allows them greater flexibility in their diet and greater control over their glycemic levels. It will be especially important for those planning to use artificial pancreas (AP) technologies to manage their disease, as the systems use estimates of carbs to be eaten during a meal to calculate the amount of insulin needed for pre-meal bolusing.
But carb counting can be as burdensome as it is beneficial. Carb counting makes many demands: having to adhere to a daily meal plan, memorizing the carb content of various foods, calculating a ratio of insulin needed to compensate for every bite of carbohydrates eaten, and balancing carb intake against exercise, illness, stress or lack of sleep. It can also be frustrating due to the amount of hidden carbohydrates in processed foods that can throw off the best estimates. In fact, a recent study published in Clinical Diabetes determined that the average accuracy score for insulin-dependent carb counters was just 59 percent.
Investigators at the Institut de Researches Cliniques de Montréal think they may have found a way to make carb counting easier for people with T1D who will be using AP technologies to manage their disease.
Endocrinologist Rémi Rabasa-Lhoret , M.D., Ph.D., and his team have developed a simplified alternative to precise carb counting that requires users to simply estimate whether their meal will be small/usual, medium or large in terms of carb content and give themselves a partial pre-meal insulin bolus based on that assessment. The algorithms running their AP systems will then administer insulin as needed to make up the difference based on its reading of their glycemic levels. The boluses the users will give themselves will be based on their current insulin-to-carb ratios.
In an early study of this bolusing method, participants using standard carb counting methods and those using the partial bolusing estimates achieved similar mean glucose levels, time spent in range, and number and severity of post-meal blood sugar swings. In fact, those who anticipated eating a large meal and bolusing accordingly achieved better post-meal blood glucose control than those who counted carbs.
Why it matters
Carb counting is a valuable tool in T1D management, but it can be boring and frustrating when even diligent counting fails to account for the proper dose of insulin needed, resulting in potentially dangerous highs and lows. The technique being refined by Dr. Rabasa-Lhoret and his colleagues could minimize both the pre-meal effort of carb counting and post-meal blood glucose excursions for those using AP systems, significantly improving both their quality of life and their glycemic control.