For example, take a look at Jenny’s evening basal test in Figure 6.7 .
She had lunch (and bolused) at noon, and then fasted until almost 11 p.m.
Her glucose was stable from 4 to 7 p.m., then rose at a rapid rate from 7 to 11 p.m.
Because Jenny is nine years old, we will consider adjusting her basal settings one hour before the observed rise in her glucose levels.
Her basal settings from 3 to 6 p.m. are verified.
Since her levels rose from 7 to 11 p.m., we will raise her basal rate from 6 to 10 p.m.
And because she is currently using a relatively low basal rate (0.25 unit/hour [u/hr] in the evening), we will increase her basal by 0.10, to 0.35 u/hr, based on the large rise that took place.
Figure 6.7: Nine-year-old Jenny experienced a large glucose rise during an evening basal test.
Basal rates need to be adjusted one to two hours prior to observed blood sugar changes.
Incidentally, even though most pumps allow basal rates to be adjusted on the half hour, it is best to set the start time for basal segments on the hour, rather than on the half hour (for example, 1:00 rather than 1:30). Remember, basal insulin is made up of tiny boluses of rapid-acting insulin every few minutes, and rapid insulin works over a three- to four-hour period. Basal insulin levels in the bloodstream don’t change dramatically all at once, so there is no benefit to making adjustments on the half hour.
Quality of Basal Adjustments
When setting up a twenty-four-hour basal program, our objective is to mimic normal pancreatic function as closely as possible.
A healthy pancreas secretes basal insulin in a pattern that repeats every twenty-four hours, based on the presence of hormones that influence the liver’s glucose output.
As a result, the pancreas produces more basal insulin at certain hours, less at others.
Normally, there is almost always one peak period and one valley in the twenty-four-hour period, not multiple peaks and valleys.
A basal program that includes multiple peaks and valleys is like a three-eyed fish: not impossible but not exactly normal.
In most cases, multiple peaks in the twenty-four-hour basal pattern mean that basal insulin is compensating for some other aspect of the insulin program that is not set up properly.
For example, consider the basal pattern in Figure 6.8 .
Figure 6.8: A “respectable” twenty-four-hour basal pattern
This pattern has one peak (between 5 and 8 a.m.) and one valley (1 to 10 p.m.). It has integrity, as far as basal programs go.
Now consider the pattern in Figure 6.9 .
Figure 6.9: A twenty-four-hour basal pattern with multiple peaks
This program has two peaks: a big peak from 5 to 10 a.m. and a smaller peak from 4 to 8 p.m. Given that the basal rate is 0.5 continuously from 2 p.m. until 7 a.m. except for those few hours in the late afternoon, there is a good chance that the 4 to 8 p.m. rates are set too high. Perhaps the basal peak in the late afternoon is compensating for an afternoon snack that is not covered with a sufficient bolus.
A basal program with multiple peaks is like a three-eyed fish: not impossible but not exactly normal.
Now let’s take a look at a couple of examples designed to put your basal fine-tuning skills to the test.
Annette is a teenager who hates just about everything, especially having diabetes.
The only thing she hates more is having high and low blood sugars that interfere with her life.
To test her morning basal rate, she agrees to skip breakfast and check her blood sugars hourly through the morning (her parents promised to increase the data limit on her phone if she followed through).
She is currently on a flat basal rate of 1.5 units per hour all day and night.
Here are the results.
What would you recommend?
7 a.m.: 184 mg/dl (10.2 mmol/l)
8 a.m.: 192 (10.7)
9 a.m.: 177 (9.8)
10 a.m.: 190 (10.6)
11 a.m.: 224 (12.4)
12 p.m.: 259 (14.4)
Annette’s blood sugar seems to hold reasonably steady from 7 a.m. to 10 a.m., so the 1.5 u/hr basal setting is verified from 6 to 9 a.m. (since she is a child, in more ways than one, we consider the basal settings one hour before the blood sugar checks). Her blood sugar rose significantly from 10 a.m. to 12 p.m., so she is going to need a fairly large basal increase (0.3 u/hr) between 9 a.m. and 11 a.m. Her new settings look like this:
midnight to 9 a.m.: 1.5
9 a.m. to 11 a.m.: 1.8
11 a.m. to 12 p.m.: 1.5
Since a retest is needed to evaluate the new settings, we combined a late-morning test with an early afternoon test; she will have an early breakfast (around 6 a.m.) and then fast until 3 p.m. so that we can evaluate how her basal rates are working from 10 a.m. until 3 p.m.
Paul is a real estate novelist who never had time for a wife (borrowed that line from Billy Joel, know which song?). He is fifty-five years old and has been wearing a pump for the past fifteen years but needs some fine-tuning of his settings. Here are his current basal rates:
midnight to 7 a.m.: 0.8 u/hr
7 a.m. to 1 p.m.: 0.6 u/hr
1 p.m. to 8 p.m.: 0.5 u/hr
8 p.m. to midnight: 0.6 u/hr
The first thing you might notice is that he has one peak (overnight) and one valley (in the middle of the day), so right away, his program passes the quality “sniff test.” Upon our recommendation, he runs an overnight basal test first, with a home-cooked dinner (and bolus) before 8 p.m., then no calories, boluses, or exercise until the following morning. The results can be found in Figure 6.10 . What would you recommend?
Figure 6.10: Paul’s overnight basal test
This basal test features a very important lesson.
Even though the wake-up reading is right on target, it does not necessarily mean that the basal setting is correct during the night.
The fact is that it is not.
Paul’s glucose level is slightly elevated but holds steady from midnight until 2 a.m., proving that his basal settings are correct from 10 p.m. until 12 a.m. (two hours prior to the observed pattern, since he is an adult).
The glucose then starts to decline until 6 a.m., when he wakes up and has breakfast.
The decline from 2 to 6 a.m. will require a basal reduction from 12 to 4 a.m., by 0.1 u/hr (since it was a modest drop and he is on modest basal settings).
This gives us a basal pattern that looks like this
12 a.m. to 4 a.m.: 0.7 u/hr
4 a.m. to 7 a.m.: 0.8 u/hr
7 a.m. to 1 p.m.: 0.6 u/hr
1 p.m. to 8 p.m.: 0.5 u/hr
8 p.m. to midnight: 0.6 u/hr
Uh oh, see the problem? By keeping basal at 0.8 from 4 to 7 a.m., we have created an extra peak in his twenty-four-hour basal pattern. Since we know that he needs less basal insulin during the night and that his needs are even lower in the morning, it is reasonable to extend the 0.7 u/hr rate until 7 a.m., giving us this pattern:
midnight to 7 a.m.: 0.7 u/hr
7 a.m. to 1 p.m.: 0.6 u/hr
1 p.m. to 8 p.m.: 0.5 u/hr
8 p.m. to midnight: 0.6 u/hr
For Paul’s next test, we can start the testing at 2 a.m. (since he was steady until then on the previous basal test) and continue through the morning. In other words, he may have an evening snack and bolus as late as 10 p.m., and then fast until lunchtime. This will allow him to evaluate his basal rates from 2 a.m. until about noon.
Normal fasting blood sugars do not necessarily mean that the overnight basal doses are correct.
Idiosyncrasies About Basal Testing
Life is never simple when you have diabetes. Basal insulin levels and basal testing don’t always follow textbook rules. There are a few quirks that you should be aware of.
For starters, never test your basal rates the first day or night when you start on your pump or new injection program. You need at least twenty-four hours for previously injected long-acting insulin to clear from your body.
Anytime the blood glucose level is above 180 mg/dl (10 mmol/l), the kidneys will channel some sugar into the urine. This may produce a slight decrease in the blood sugar concentration. Thus, when performing a basal test with elevated blood sugar, a slight drop-off in the blood sugar is to be expected and does not necessarily mean that the basal setting is too high.
In some people, a blood sugar level that is borderline low or dropping quickly will cause a hormonal response that induces a blood sugar rise. Again, this does not necessarily mean that the basal setting is too low.
After finishing a fasting test, don’t be surprised to see a significant spike in the blood sugar right after eating again. It is natural for food to digest extra quickly after a prolonged fast.
If your family includes multiple blood relatives who require insulin, their basal needs will probably be quite similar. The actual doses may be different, but the timing and magnitude of peaks and valleys will likely match up. We can’t explain why, but this is what we tend to see.
In terms of overall basal patterns, people who are still producing some of their own insulin tend to have flatter basal profiles than those who are truly insulin-dependent. For those with type 2 diabetes or LADA as well as type 1s still in a honeymoon phase, the pancreas will adjust its own insulin secretion to offset some of the peaks and valleys in the liver’s glucose secretion. For those who produce little or none of their own insulin, the difference between the peak and valley basal needs can be significant.
Beyond Basal Basics
For kids, establishing a basal program that works properly is very important.
But don’t feel like you have to repeat an entire set of basal tests every time your child grows a little bit.
That would be cruel and unusual punishment for both of you.
Your child’s pattern (timing and magnitude of peaks and valleys) should stay the same throughout their growth years.
When it becomes obvious that they need more basal insulin (because of a persistent overnight rise, for example), it is usually fine to increase all of the basal settings throughout the day and night.
However, following the transition from childhood to adulthood, it may be necessary to revisit and retest the basal settings since hormone levels tend to shift in both time and magnitude.
An additional note about insulin pumps: all pumps can store settings for more than one complete twenty-four-hour program. Secondary basal programs can be useful during periods of heightened insulin need (such as sick days, travel days, or prior to menstruation) or decreased insulin need (such as the start of your period or days filled with extra physical activity). Use of alternate basal programs will be discussed in more detail in