Why hypoglycemia happens in type 1 diabetes

The underlying reason is that injected insulin works regardless of the glucose level. In a person without diabetes, the first defense against a falling glucose level is switching off insulin secretion. It starts while glucose is still within the normal range, at around 80–85 mg/dL (4.4–4.7 mmol/L) [1]. Insulin already delivered by a pen or a pump cannot be switched off, because it is already in the body. In other words, the first line of defense is missing from the start.

This is where the imbalance comes from. Hypoglycemia happens when the insulin on board exceeds what the body needs at that moment, and that need keeps changing. A meal smaller than the one you counted for, unplanned physical activity or a dose that is slightly too large all produce the same result. It is not a failure of character, but the limit of a treatment that tries to replace an organ. Hypoglycemia is the main factor that limits glycemic control in type 1 diabetes [2]. It is a problem of the treatment, not of the disease itself.

In a person without diabetes, the defense against a falling glucose level works in three lines, each starting at a slightly lower glucose level:

• Switching off insulin secretion — the first line, which starts while glucose is still within the normal range [1].
• Releasing glucagon — the hormone that draws glucose out of the liver.
• Activating the sympathetic nervous system — together with the release of adrenaline from the adrenal glands.

The symptoms do not come from adrenaline acting as a hormone. They are produced mostly by the sympathetic nerves, which release two substances locally. The first is acetylcholine, which causes sweating, sudden hunger and tingling [3]. The second is noradrenaline, which causes tremor, palpitations and pallor. Adrenaline in the blood adds to these effects, but the signs appear even without that hormone [4].

In type 1 diabetes the first line is missing, and the second one fades until it too is sometimes absent. Insulin that has already been given cannot be switched off. The glucagon response to hypoglycemia is usually lost within the first years of the disease [2]. What remains is the sympathetic response, which is itself blunted by recent hypoglycemia. This is one of the reasons hypoglycemia is far more common in type 1 diabetes than in other forms of diabetes.

Glucagon is the hormone opposite to insulin and it is produced by the alpha cells of the pancreas. Its job is to tell the liver to release glucose when the glucose level falls. It is the second line of defense, not the first. The first one remains switching off insulin secretion, a step that is impossible when insulin comes only from outside. In type 1 diabetes the alpha cells are still there, and their defect is selective. They no longer respond to a falling glucose level [5]. They do keep releasing glucagon after a protein-rich meal or during physical activity.

The accepted explanation is the loss of the local signal sent by their neighbors, the beta cells. When the beta cells disappear, so does the internal command of the pancreatic islets (islets of Langerhans), which used to contribute to glucagon release [5]. Glucagon still works when it is given as a medicine, because the liver responds to it normally. That is why treatment from outside remains effective, even though your own glucagon is no longer released in time.

The main cause is not how long you have had diabetes, but the hypoglycemia of the past days and weeks. A single episode of hypoglycemia lowers the sympathetic response to the next one, sometimes as early as the following day [6]. The threshold at which symptoms appear moves down, sometimes below the threshold at which thinking is already affected.

A vicious circle sets in. You feel less, so you correct later, you end up lower, and the body gets used to it even more. The result of this loop has a medical name, hypoglycemia-associated autonomic failure [7]. There is only one way to break the circle, by consistently avoiding hypoglycemia for a few weeks. It is a temporary change of targets, made together with your doctor and not on your own.

It is the medical name of the vicious circle described above, known in the literature by the abbreviation HAAF (hypoglycemia-associated autonomic failure). The term says that the autonomic nervous system, the one that sets off the alarm, no longer reacts in time. This is not the classic autonomic neuropathy, a complication of long-standing diabetes. It is a functional disorder, triggered by recent hypoglycemia and largely reversible [7].

Two things are lost together, the response of the defense hormones and the symptoms that warn you. The practical consequence is a much higher risk of severe hypoglycemia. People who no longer feel their lows have about six times more severe episodes than the others [8]. Recovery is uneven. About 3–4 weeks of strict avoidance of hypoglycemia restore the perception of symptoms. The hormonal response is still deficient after three months [9]. You get the warning back, but not the automatic hormonal defense. This possible problem is assessed at least once a year, at the routine review [10].

It can, but only up to a point. The liver keeps a store of glucose in the form of glycogen and releases it on command from glucagon or adrenaline. It also makes new glucose from other raw materials (mostly amino acids), a process called gluconeogenesis. The kidney can make glucose from other materials too, but it uses most of what it makes for its own needs [11].

The glucose store in the liver does have some important limits. The insulin available in the blood directly brakes glucose release from the liver, so it works against this line of defense. Glycogen stores fall after prolonged physical activity and after fasting. Alcohol blocks the making of new glucose, although not the release from the stores [12]. It therefore becomes dangerous above all when the stores are already used up.

No. Once it reaches the subcutaneous tissue, insulin forms a depot from which it is absorbed gradually. There is no antidote and it cannot be taken back. A pump can stop insulin delivery in the very next second, but it can do nothing about the insulin that has already entered the tissue.

Two practical rules follow. The first is that hypoglycemia with insulin on board (subcutaneously) is treated with carbohydrates, not by waiting [10]. The second is that a dose that is too large will act all the way to the end, so it has to be covered gradually with food and watched. Automated systems reduce the risk a great deal, but they cannot take back the insulin that has already been delivered into the subcutaneous tissue either [13].

Yes. It happens often. The calculation can be flawless, but insulin absorption varies from one day to the next. The injection site matters, and so do skin temperature, a hot shower, rubbing the area or physical activity that uses the muscle underneath. An injection that lands in the muscle by mistake is absorbed much faster and is a known cause of seemingly unexplained hypoglycemia [14].

The thickened areas that appear where you inject insulin into the same spot every time have a similar effect. People with such lipohypertrophy raise their doses step by step for as long as they keep injecting there [15]. The same dose moved afterwards into an area of normal skin is absorbed more efficiently and can cause hypoglycemia. Correct rotation of sites remains the simplest preventive measure [14]. Food adds its own part. A meal high in fat slows gastric emptying, so glucose reaches the blood later than the insulin does. The result is a lower glucose level at first and a higher one afterwards, at four hours, with the same dose [16].

Yes. Some episodes remain unexplained even if you go through everything carefully. Variable absorption, a small error in estimating carbohydrates and the exercise from the day before can add up without any of them standing out [17].

It matters not to turn every episode into an investigation. A single mild episode, corrected quickly, does not need an explanation at any cost. What counts are the patterns, meaning the same hour, the same activity or the same meal. Patterns almost always have a cause that can be corrected [10].

The brain runs almost exclusively on glucose [1]. The local glycogen store in the glial (support) cells covers only a few minutes. That is why, when the glucose level falls a long way, brain function deteriorates for a while. Thinking slows down, vision blurs and even simple decisions become difficult.

Cognitive function recovers more slowly than the glucose level does. Reaction time stays altered for up to 40–75 minutes after glucose has returned to normal [18]. In the vast majority of cases everything is reversible and the brain comes back completely. The real concern is repeated severe episodes, above all in children under 6, whose brain is still developing [19]. At that age the glucose targets have to be chosen carefully, and the priority has to be preventing severe hypoglycemia [20].