The Variability of Blood Glucose in Type 1 Diabetes

Stefan Du Rietz  (e-mail: sdr (at) this domain)

Probability distribution of fasting blood glucose (BG) readings

Read more about the statistical method in my Introduction.

SDR is a man, 55 years old, with type 1 diabetes for 49 years. Fig. 1 shows the statistical distribution of his morning BG meter readings during 1¾ year, 1998-03-15 to 1999-12-14 (640 days).

How to read the graphs.

Figure 1. Cumulative probability distribution of SDR BG before breakfast (n = 638):
GM = 5.16 mmol/l  FV = 1.48  FV⁴ = 4.82

Evidently there is a great variability, although he also tests in the middle of the night (fig. 2) and then takes an extra injection of regular insulin if BG is too high or eats carbohydrates if BG is too low.

Fig. 2 shows – during the same time – the statistical distribution of his BG meter readings in the middle of the night, about 4 hours after the evening injection of NPH insulin.

How to read the graphs.

Figure 2. Cumulative probability distribution of SDR BG by night (n = 605):
GM = 5.46 mmol/l  FV = 1.45  FV⁴ = 4.39

The mean, GM, is marginally higher by night than before breakfast, while the factor of variation, FV, is marginally lower.

AKE is a woman, 27 years old, with type 1 diabetes for 10 years. Fig. 3 shows the statistical distribution of her morning BG meter readings during her second pregnancy (252 days).

How to read the graphs.

Figure 3. Cumulative probability distribution of AKE BG before breakfast (n = 252):
GM = 5.51 mmol/l  FV = 1.20  FV⁴ = 2.08

GM is about the same as for SDR while FV is much smaller.

Fig. 4 shows – during the same time – the statistical distribution of her BG meter readings at night, about 4 hours after the evening injection of NPH insulin.

How to read the graphs.

Figure 4. Cumulative probability distribution of AKE BG by night (n = 157):
GM = 5.54 mmol/l  FV = 1.45  FV⁴ = 4.45

Here, both GM and FV are the same as those for SDR.

JH is a woman, 27 years old, with type 1 diabetes for one year.

Fig. 5 shows the statistical distribution of her morning BG meter readings during this first year (365 days) after the diagnosis.

How to read the graphs.

Figure 5. Cumulative probability distribution of JH BG before breakfast (n = 360):
GM = 3.95 mmol/l  FV = 1.13  FV⁴ = 1.61

You can see how the remaining function of the b-cells limits the variations – probably to the same extent as in healthy people [1].

Discussion

Comparing the variations in morning BG of SDR, AKE and JH discloses the great importance during fasting of the remaining endogenous insulin production of AKE and JH. On the other hand, their BG variability after food intake (All BG) is much larger due to a larger and more varying insulin requirement, which is not possible to optimally supply with injections of constant insulin doses.

References:

1. Oddly enough, there are no published long-term data on this. Consequently, the normal variations in healthy people are actually unknown.
2. Kovatchev BP, Cox DJ, Gonder-Frederick LA, Clarke W: "Symmetrization of the blood glucose measurement scale and its applications." Diabetes Care 20: 1655-8, 1997. PubMed

© 1999-2024 Stefan Du Rietz.  Last modified: 2022-06-21 00:41 (MEST).