Workbook Examples of CGM’s and Type 1 management – Module E6B – Combe Grove Practical

These real-world CGM’s complement the CGM graphs presented in various parts of this course. They highlight aspects of T1 management based on one person’s experience, but the general principles apply to all. They apply equally to any dietary preference, not just ketogenic diets. The only difference between a ketogenic diet and a high carbohydrate diet is the amount of insulin required, being higher in volume on a high carbohydrate diet. 

There is an explanation and a graph, Think about what has happened before looking at the answer. Hopefully, these examples will boost your confidence in managing Type 1. 

Example 1

This is a nine-hour trace covering the sleep period. The glucose limits are set between 3 and 7 mmol/l to represent near physiological concentrations. This person had consumed a meal higher in protein content than ideal. the meal consisted of 10 g of carbs and 60g of protein. For a 75kg person, this is close to a day’s allowance in one meal. They had injected 4 units of rapid-acting insulin to cover this at the time of the meal. They had also injected 8 units of Tresiba basal (part of a twice-daily 16U per day long-acting regimen). Their IPR (insulin to protein ratio) is 1:20 and ICR ( Insulin to carb ratio) 1:10

Have a think about what happened here. What is the spike at 2200? What is causing the glucose rise through the fasting period at night?

The spike at 2200 was caused by 6g of glucose consumed for a sensation of hypo, ( note the low blood glucose). It is doubtful that the CGM was accurate when it gave a reading of 2. This person had hypo awareness at 2.8mmol/l reliably. They should have conducted a capillary check glucose test to calibrate the CGM if they felt it was inaccurate.

The steady upward trend of glucose throughout the night is due to the effect of gluconeogenesis due to protein. Without such a late meal with a high protein load, the Tresiba should have covered the glucose to give a flat trace.

Possible solution. The glucose range was physiological, so there was nothing to do in this case. A  glucose level of 8-10 mmol/l might have required a unit or so of insulin, depending on insulin sensitivity. In this example, there is no harm in waiting. There is a  possibility of precipitating hypoglycaemia at night. For better control in the future avoid late meals and split the daily protein evenly between meals. For one meal a day it is better to eat before 8 pm. Late meals upset nocturnal hormonal profiles.

Example 2  

This is a trace of an overnight basal insulin dose of 8 units. 

This is a simple case of too high a basal load. A steady glucose decline throughout the night and a hypoglycaemic picture in the morning. Note the lack of dawn effect here. The person is of an age where Growth Hormone is less dominant.

The solution to this problem is to reduce the basal dose by two units and re-assess the next night. There is no need for concern when setting basal insulin as everything happens slowly. 

Basal adjustment is often needed early in the change to a keto diet, then after a few weeks or even months when the insulin sensitivity improves. Sometimes, people notice a difference in insulin requirement with generally improved control for no apparent reason. This might last for a few days or weeks and then return to a more normal level for that individual.  There are many mysteries in Type 1 diabetes control, and not everything has an answer. A CGM is a superior monitor for discovering these subtleties. 

Example 3

This CGM is from someone on a time-delayed eating strategy. They would have injected 8 units of Tresiba in the morning as part of their twice-daily basal regimen and probably used 2 units of rapid insulin to cover and dawn effect. Insulin needs are often greater in the mornings, but not always. 

This trace shows a reduction of glucose level followed by a plateau. This is where endogenous glucagon production balances the effect of basal insulin. This is a satisfactory trace for keto in Type 1 and shows how insulin antagonists can be used for good control. This is near physiological.  You can see many of these traces from the blogs associated with the Zerofive100 Project.

Example 4

This is an overnight CGM. Think about what is causing the two peaks at 0400 and 0700 and then the gradually rising glucose level in the early morning. 

This is a case of hypo rescue during the night. The person would have woken to a CGM alarm. If they were not using a CGM, it is possible that the typical symptoms of hypo would wake them; sweating, headache, tachycardia, vivid dreams etc. The hypo was rescued with the same quantity of glucose, six grams, on each occasion. Some people need more. The gradual rise in glucose on waking is the dawn effect. One point of interest is the two different shapes of the hypo rescue gradients. The result is blunted in the second trace, and the dawn effect was operating soon after. This likely represents a change in insulin response as the hormones of the dawn effect became more dominant.

Example 6

This CGM is a trace of a one- meal-a-day (OMAD)(evening) plan. It is from early morning until after lunch. Think about what has happened here.

The trace from 0600-0800 represents the dawn effect. This would be corrected with 2 units of rapid-acting insulin, and basal insulin would be injected at the same time. The slow downward gradient of the action of the insulin is seen here. The rise in glucose around 12:30 is caused by 9g of glucose tablets. As soon as the effect of the tablets has worn off, the downward gradient of the basal insulin asserts itself again. Note the similarity of the gradients though this is not always the case. There is an absence of a plateau which suggests a phase of insulin sensitivity. If you were advising this patient, you would suggest a reduction in basal dose by around two units. Despite the change in glucose levels, this is still within the physiological range of glucose. Why not leave the downward trend at 12:00 and see what happens? That is a valid point. There is some flattening of the trace. It is likely that the person had a hypo sensation. Hypo sensations can happen over a range of glucose levels in each individual. It is important to treat the symptoms rather than the number.  

The important point is that the person recognises a hypo and acts on it


It has been shown throughout these many examples that there are often no ‘right’ answers but often what is required is a logical approach to dealing with the presenting problem. Because of the many variables that affect blood glucose, it is necessary to include these in the problem-solving process. Relying on the interaction between carbohydrates and insulin alone is not sufficient. Remember the old adage that the patient will tell you what is wrong with them if only you will listen. Good history taking and an appreciation of the many variables will enable you to help your patient to improve their diabetes control. It is a highly rewarding experience when a patient transitions to a keto lifestyle and you have sufficient information from this course to facilitate that journey. There will be scenarios that are perplexing but worry not. You are not alone. Here is a thread of tweets from people with Type 1 who are equally perplexed about the mysteries of their condition.