I’m going to open the hood on basic diabetes management so you can see the nuts and bolts of how a type 1 diabetic uses blood sugar measurement and insulin injections to manage their blood sugar.

This explanation of diabetes and insulin therapy covers the basics of a Diabetic’s How To:

1. Establishing a basal rate of insulin
2. Establishing a ratio of insulin-to-carbohydrate
3. Learning to judge carbohydrate content of food and the effect of what you eat on your blood sugar
4. Learning to correct high blood sugar with a bolus of insulin

Two tools: blood sugar and insulin

The two most basic tools of the diabetic are blood sugar measurement, and injected insulin.

A glucometer is a device which uses a test strip to read a sample of blood (taken from poking a fingertip with a lancet) and generates a number, which we call “Blood Glucose” (BG) or blood sugar. There are two systems of units used. Americans use “milligrams per deciliter” or mg/dL, and the rest of the world uses “millimoles per liter” or mmol/L. We can translate easily between the two systems using the constant, 18. One mnol/L = 18 mg/dL.

The diabetic must inject insulin to move sugar, which is flowing in the bloodstream after being released from digestion in the intestines, through cell membranes so it can be used in cell metabolism, i.e. for the body’s energy.

Normal blood sugar

We use the glucometer to measure how much sugar is in the bloodstream. A non-diabetic naturally produces insulin to maintain a balanced blood sugar level of about 70-100 mg/dL or 4.0-5.5 mmol/L. The type 1 diabetic will attempt to inject an amount of insulin to simulate this healthy blood sugar level.

Thus, the thumbnail sketch you probably already knew. But this does not scratch the surface of what an endocrinologist must apprehend to begin a diabetic on blood sugar management. Now the technology begins, and the therapy is dictated by the characteristics of the medicine: by recombinant-DNA synthetic insulins and by insulin pumps, tiny devices used to precisely measure and infuse insulin into the body as a replacement for injections through needles.

I will describe two broad methods: first, injected insulins, and second, the insulin pump. These are different because the insulin pump enables the use of only one kind of insulin, infused in a constant stream into the body, mimicking the pancreas’ natural secretion of insulin.

Two types: long-acting and fast-acting insulins

Injected insulin, however, must use two types of insulin. (For advanced type 2 diabetics, and some type 1 diabetics who are on a poor system of therapy, a “hybrid” insulin may be used which combines the two types of insulin into one. This may be appropriate for some type 2 diabetics who still have some pancreatic function and some natural insulin, but is horriby imprecise for a type 1 diabetic.) The first insulin is a “long-acting” insulin; you can imagine a time-released insulin which offers a baseline amount of insulin over a long-period of time. Some work on a 12-hour scale, some on a 24-hour, and deliveries are generally imperfect and include time scales in between.

The second insulin is a “fast-acting” or “short-acting” insulin. Upon injection, this insulin begins acting within 15-30 minutes, maximizes its action around 45-75 minutes, and is done acting wtihin 120 minutes, assuming normal delivery mechanism (sometimes, for instance, a greasy meal can slow blood sugar digestion, or a pocket of scar tissue can delay insulin absorption.) The different kinds of “fast-acting” insulin have different action schedules, giving us the broad ranges I indicated above.

The normal human pancreas maintains a baseline level of insulin in the bloodstream to maintain ongoing glucose metabolism. This baseline is approximated by the use of the long-acting or slow insulins. Hence, they are injected on a regular schedule every 12 or 24 hours — or even every 8 hours, as in my own case.

When a person eats, the bloodstream absorbs glucose from the intestines, causing a sudden rise in blood sugar. The normal human pancreas releases a hopefully accurate amount of insulin to accommodate this surge of sugar, moving it out of the bloodstream and into the cells.

When the type 1 diabetic eats, he must inject a “bolus” of fast-acting insulin to approximate the natural action of the pancreas, attempting to match the amount of sugar that has been added to the bloodstream.

Calculation time

Now we can see the essential, active process of a diabetic’s insulin therapy. The diabetic must choose how much slow insulin to use as a baseline, and how much fast insulin to take when eating food, drinks or snacks.

Basal testing

The diabetic must perform a ‘basal test’ to determine how much slow insulin must be taken to match the body’s ongoing, maintenance level of glucose metabolism. To do this, he must avoid the dramatic changes in blood sugar which result from eating food, which means, he must fast and observe how his blood sugar changes as a function of only his slow-acting insulin.

Of course, the diabetic must start with some level of slow-acting insulin. The endocrinologist will set a level based on what patients who are similar in age, body composition and weight are using. When I started injecting insulin in 1993, I was placed on 16 untis of NPH (an older, slow-acting insulin) daily. I soon increased this to 20 units of NPH daily and later to 22-24 units of NPH daily. It’s common for doctors to start with a conservative (probably too small) amount of insulin, as an overdose of insulin is more dangerous than insufficient insulin.

The process of basal testing must continue until the diabetic can inject his slow-acting insulin, on a normal delivery schedule, and observe stable blood sugar readings while skipping meals. These fasting periods should be observed at all times of day, as blood sugars and thus insulin needs tend to vary, sometimes significantly, at different times of the day.

With successful basal testing concluded, the diabetic has a prescribed amount of slow-acting insulin that should be taken on a regular schedule.

The insulin bolus: so a diabetic can eat

Fast-acting insulin enables a diabetic to eat. The diabetic must, over months and years, learn how his body’s blood sugar level is affected by the foods he eats, and must calculate an amount of fast-acting insulin to balance the blood sugar changes that result from digestion. For example: I want to eat a snack of an apple with two tablespoons of peanut butter and a glass of milk. How much insulin must I inject to maintain a healthy blood sugar level?

(I follow a modified version of the paleo diet: paleo plus dairy. I do not eat sugars at all, and an absolute minimum of grains and starches. My diet consists primarily of meat and vegetables, dairy, nuts and seeds and a little fruit. Coffee is my sin, and if there’s an afterlife, I’ll return to chocolate, breakfast cereals, oatmeal chocolate chip cookies, blini, chocolate, waffles, chocolate, pizza, chocolate, Jamba Juice, chocolate and and and and…)

A diabetic can calculate the insulin he needs to take by using a ratio of insulin-to-carboyhdrates. This is the theoretical model employed by endocrinologists and taught to all diabetics. For example, my snack of apple, peanut butter and milk includes about 15 grams of fructose from the apple, about 7 grams of natural carbohydrate and 2 grams of added sugar (sucrose) in the peanut butter, and about 16 grams of lactose in my glass of milk (~12 ounces or 330 ml). Thus, a total of 40 grams carbohydrate in this snack, which I must take an appropriate amount of insuiln for.

I decide the amount of insulin based on the ratio of how many units of insulin I need to cover a certain amount of carbohydrates. A very typical ratio is 1:15 — a unit of insulin, for 15 grams CHO. (A ‘unit’ of insulin, by the way, is 0.01 milliliters of insulin. I believe this was established in practice by Eli Lilly, the first producers of pharmaceuitcal insulin.)

What determines your ratio is your insulin sensitivity, which can vary greatly from one individual to the next, and will also vary for the same person depending on time of day, sleep schedule, stress levels, physical activity, and other factors about which the diabetic is forever speculating and worrying.

I just now ate that snack, and I took 4 units of insulin to cover it. I’m physically inactive today, writing this article. I’ve been suffering some reduced insulin sensitivity recently, perhaps due to overtraining combined with insufficient sleep (thus raised cortisol levels).

The implicit insulin:CHO ratio is 1:10. But I didn’t start from an insulin:CHO ratio, did I? I seemed to jump straight to “5 units,” and only calculate my insulin:CHO ratio as an afterthought. Hmmmm. What gives? Let’s look at a second method of deciding how much insulin to bolus.

Theory and practice — triangulating the insulin bolus

Above, I wrote that the units-to-grams-carbohydrate ratio is the theoretical model employed by endocrinologists and taught to all diabetics. There is a second method, and as a diabetic’s experience increases, the second method will replace the first. It is simply: experience observing the blood sugar results of using specific amounts of insulin for specific meals. Over time, the diabetic eats the same meals and snacks and learns how to modify calculated insulin boluses to get better results.

To do this, the diabetic must measure blood sugar before the meal, take insulin at an appropriate time (usually 0-15 minutes prior to eating, using a modern recombinant-DNA technology synethetic insulin like Humalog or Novolog), and control for other factors like waking in the morning (increased cortisol –> decreased insulin sensitivity), physical activity, etc. Then the diabetic must measure blood sugar again about 90-120 minutes after the meal or snack, to observe the results.

I learned very early that endocrinologists would insist to me, that based on my measured ratios of 1:15 insulin:CHO, a meal like a double-double hamburger from In-N-Out includng a medium fries and diet Coke should only require about 5-6 units of insulin. In fact, for me this meal requires 8 units of insulin. The diabetic must learn from experience and modify his insulin therapy, or he will not achieve successful results. I illustrated with only one specific example, but in fact the seeming “exceptions” are more common than the norm, in most diabetics’ experience. (Now that I am on the paleo diet, I would not consume the fries, the diet Coke or the hamburger buns in the above meal.)

The corrective bolus

The diabetic’s experience of life consists in constantly struggling to balance his blood sugar. From waking in the morning, sudden release of cortisol drives increasing blood sugars. Over the next two-three hours, as insulin sensitivity increases to normal, extra insulin injected to counter the “dawn phenomenon” of cortisol may suddenly drive the blood sugar low. If the diabetic doesn’t sense or measure the low BG and take corrective action, he will crash.

Scott Hanselman has written a good analogy insulin therapy as comparable to flying a plane: trying to fly a plane at controlled altitude without crashing or stalling. Many people quickly say “it’s a tightrope,” the slightest inattention and one falls off. The airplane captures this truth: without active management and the feedback of one’s instruments, one cannot maintain level flight.

Invariably, the diabetic experiences unexpected increases in blood sugar. The diabetic may feel that his blood sugar is too high. The experience includes a very vague feeling of being muddy, hot, slow, and sometimes is very unpleasant; it may be symptoms like sudden thirst, blurry eyes, needing to urinate. Or checking his blood sugar, the diabetic gets a reading like “200 mg/dL” or “8 mmol/L”. A corrective bolus is in order: immediately inject insulin to reduce the blood sugar, returning it to a target level of 70-100 mg/dL or 4-5.5 mmol/L.

Two numbers: the target blood sugar level, and the insuln-to-blood-sugar-drop ratio.

To manage the correction of high blood sugars, the diabetic needs to know what effect an injection of insulin will have in decreasing his blood sugar. In my experience using fast-acting insulins (like Humalog and Novolog (US) or Novorapid (EU)), one unit drops my BG by 20 points of mg/dL. So, if I had a blood sugar of 200 mg/dL (more than twice my target blood sugar), I would take a bolus of 6 units of insulin, which should drop it to 80 mg/dL. As I’m currently using the mmol/L blood sugar measurement, my target blood sugar is about 5 mmol/L. If I measure a BG of 9.0, I’ll take 4 units of insulin, expecting this to return me to 5.0 mmol/L. This is equivalent to a ratio of 1:1 by the mmol/L scale, or 1:18 on the mg/dL scale.

The insulin pump

The two-pronged explanation above, first examining basal rates and then boluses, which both cover food eaten and correct high blood sugars, are based on how we adapted the technology of slow-acting insulin to the metabolic needs of the human body.

The exact same needs can be handled differently if we have a device, like an artificial pancreas, that can continually deliver a controlled stream of insulin into the body. That’s what the insulin pump does. It only infuses short-acting insulin, which simplifies matters. The infusion is maintained by a tube which is always in place, called a ‘canula’ (not a catheter). The canula is placed ’sub-cutaneously’, meaning under the skin. The pump is not like an IV, the canula does not touch the veins or bloodstream directly. The canula’s insertion is called an “infusion site” and must be changed frequently; usually every two days.

The “pumper” or diabetic using an insulin pump can assert a much more precise level of control over his insulin delivery and thus over his blood sugar. Smaller amounts of insulin can be delivered. The hassle of a new injection for every adjustment is eliminated; the pumper only needs to type in a number and press a button to execute.

Hyperglycemia

Without insulin therapy, or with insufficient insulin therapy, the diabetic’s blood sugar rises, out of control, and the diabetic can slip into a diabetic coma resulting from severe hyperglycemia. Over the longer term, chronic hyperglycemia causes compromised circulation, impeded by excessive glucose in the bloodstream. This leads to many pathologies, including neuropathy, retinopathy (blindness), nephropathy (to renal failure), macroangiopathy (high blood pressure, diabetic feet, etc). These comprise a strong motivation for the insulin-dependent diabetic to take his insulin and keep his blood sugar low!

Hypoglycemia

However, too much insulin is equally and perhaps even more dangerous. Too much insulin has an immediate effect as the blood sugar drops low and the brain has insufficient glucose. Called ‘insulin shock,’ this is acute hypoglycemia. It can result in death, but is quite common in occurrence and generally involves standard symptoms: shakiness (from adrenaline), irritability, drowsiness, unresponsiveness, hunger, cognitive impairment, and reportedly nausea.

Over time, even only a few weeks or months, a diabetic who is chronically going hypoglycemic can become desensitized to the symptoms. If this happens, the diabetic may have no way of knowing that he’s “going low”. Without testing blood sugar, he may have no way of knowing the impairment and risk he’s facing. The diabetic may go lower still and experience far more severe symptoms, including severe cognitive impairment, seen in loss of judgment, loss of fine motor skills, loss of balance, visual or auditory hallucinations, paranoia, seizures, and a wide range of other problems resulting from the brain’s loss of function.

Fitness and nutrition for the diabetic

While this article mentions some of the grave consequences of diabetes, and describes the process of calculating insulin therapy, I don’t believe it communicates just how difficult managing one’s diabetes can be.

In my own experience, fitness has been an invaluable tool in coping with diabetes. My fitness training, which used to consist primarily of running, has kept me motivated and frequently worked as an anti-depressant. It has always worked to keep my insulin sensitivity higher, allowing me to take less insulin.

However, what I cannot fail to advocate is a more recent discovery I’ve made: the role of nutrition. I have known for many years that I ought to be choosing the foods I eat by their insulin requirements. A fascinating correlation is the insulin demands of a food, and how recently that food was incorporated into the human diet.

The foods highest in glycemic index, and thus most shocking to the system in their demand for insulin, in their tendency to be stored as body fat, in their contribution to obesity and other metabolic disorders, are also the foods that were most recently introduced to the human diet via a few technological changes.

First, agriculture introduced grains, starches, glutens into the diet. This was probably only 40,000 years ago, and very recently relative to the evolution of our digestive organs and thus the factors for which their function and performance were adapted. More recently still are a host of other dietary evils, the driving forces behind which are largely economic: low cost makes sweet foods, based on sugars, kept widely available at low cost with preservatives, omnipresent in our diet and delivering tremendous short-term satisfaction for a fraction of the cost of other foods. Their dominance and ubuiquity in our modern diet are impossible to fight without a clear understanding of just how harmful they are to us.

I believe the paleo model, understanding that our bodies function best using the foods for which they were evolved, is the best guidance for selecting healthy foods. And an athletic, performance-based fitness both motivates one to eat well and tests one’s nutrition for its ability to fuel great performance and keep one feeling energetic and aware.