Info

Figure 10.3 Profile from a patient with type 1 diabetes treated with an insulin pump. The HbAlc is similar to that in Figure 10.2. but the dynamical properties are clearly different. Reproduced from Medtronic.

Figure 10.3 Profile from a patient with type 1 diabetes treated with an insulin pump. The HbAlc is similar to that in Figure 10.2. but the dynamical properties are clearly different. Reproduced from Medtronic.

• A combination of the two

• Take an occasional sporadic reading to check that the system has not moved 'too far out'

• Reserve self-monitoring for episodes of acute illness

This list suggests a distinction between retrospective and prospective approaches. Each has its benefits and limitations

Retrospective analysis

The traditional approach aims to identify patterns that are only evident when several days or more of data are gathered continuously. This may improve understanding of the dynamical behaviour. Software is available to enable sharing of the data between patient and clinician via the internet (see Figure 10.5). However, there are several limitations of this approach.

First of all, the profile may be 'complete,' but significant excursions may occur between monitoring (see Figure 10.4). Carbohydrate intake and insulin doses may be recorded much less consistently than the glucose data, and exercise is very difficult to quantify. Data may be uploaded to a personal computer to assist with processing and statistical evaluation, but more often the raw data are presented without any such tools. The human eye struggles to perceive patterns in numerical data, particularly when decimal places are used, which is why graphical displays may be very helpful.

Secondly, the assumption is often made that the profile has been gathered without the emerging results influencing the patient's behaviour, as it might if we were collecting it during a study on an animal, or during a 'blinded' monitoring exercise. But the patient's responses to the readings may be a powerful determinant of the dynamical behaviour. High fasting levels may arise through over-correction of nocturnal hypoglycaemia, for instance, but such details may not be recorded in the profile.

Despite these limitations, the following guidelines may help improve control based on retrospective analysis of the profile.

Figure 10.4 Increase in blood glucose after breakfast on each of the three monitoring days during a continuous glucose monitoring system (CGMS) recording. A self-monitoring schedule involving measurements four times a day (before each meal and at bedtime) would have missed these excursions. Reproduced from Medtronic.

measure the diurnal profile over a period of days. This approach is still required in difficult cases, including the more brittle patterns seen in children.

Self-monitoring technology has changed this situation, as a patient can now easily build such a profile through regular monitoring in the more natural environment of their usual daily activities (see Box 10.3).

Box 10.3 Retrospective and prospective approaches

The interpretation and analysis of self-monitored results should be tailored to the patient's needs and the type of diabetes and treatment regimen. There are a number of possible strategies. Patients may:

• Build a profile, in tabulated or graphical form, to examine retrospectively either alone, or in consultation with a practitioner

• Take readings purely to influence immediate actions prospectively, e.g. to detect and avoid imminent hypoglycaemia or as a precaution before driving

Figure 10.4 Increase in blood glucose after breakfast on each of the three monitoring days during a continuous glucose monitoring system (CGMS) recording. A self-monitoring schedule involving measurements four times a day (before each meal and at bedtime) would have missed these excursions. Reproduced from Medtronic.

Advice based on the retrospective approach (adapted from 5th edition of the ABC of Diabetes)

Unless there is good reason to alter doses acutely (such as imminent hypoglycaemia or intercurrent illness), for most patients it is best to keep doses fairly stable (with less than 10% change) from day to day. Patterns of variation are more important than single random glucose values.

Examine the profile and attempt to identify consistent, reproducible peaks and troughs.

To increase blood glucose in the troughs:

• Eat more carbohydrate at or before the times when blood glucose values are at their lowest, usually at mid-morning and at bed-time

• Reduce the dose of insulin before the trough

• Consider changing a short acting insulin to a rapid acting insulin analogue to avoid pre-meal hypoglycaemia

To decrease blood glucose in the peaks:

• Reduce carbohydrate intake at the meals that precede the peaks

• Increase the dose of insulin before the peak

These adjustments should be made with an awareness of the duration of action of the various types of insulin (see Chapter 8).

To decrease fasting hyperglycaemia:

• Increase the evening intermediate or long acting insulin

• If this causes nocturnal hypoglycaemia, consider splitting the pre-dinner insulin into two parts, with the short or rapid acting insulin before dinner and the longer acting insulin at bedtime. Alternatively, if the patient is taking an intermediate acting insulin in the evening, consider changing to a long acting insulin analogue such as detemir or glargine.

To reduce nocturnal hypoglycaemia:

• Reduce the dose of the evening intermediate insulin or long acting insulin analogue

• Advise the patient to take carbohydrate at bedtime. This is particularly important if the bedtime blood glucose level is <6.0 mmol/L

• Consider changing evening intermediate acting insulin to a long acting insulin analogue, to avoid nocturnal 'peaking' of insulin levels.

Software programmes have been designed to assist in dose adjustment to optimise insulin regimens. These include the CoPilot system (see Figure 10.5).

The prospective approach

In the past the retrospective approach was promoted as the only appropriate technique, but may seem like 'driving a car by looking through the rear mirror', to quote one patient. This is a particular problem if life-styles require flexibility and immediate needs are inconsistent from one day to the next. To continue the metaphor, patients are likely to want to use the front windscreen prospectively as well as the rear mirror retrospectively, but there are dangers if the patient is not sufficiently skilled at this.

0 0

Post a comment