| My fiancée was close to death from a missed diagnosis of Type 1 diabetes. We were extremely lucky that she actually got a diagnosis; A nurse at the oncology ward had a hunch. (Overburdened medical personnel miss diagnoses.) Her HbA1c long-term blood sugar measurement was literally off the charts. Off the chart on the wall on the hospital, right through the red “DANGER” section. She had lost 17% of her bodyweight and half of her hair. We have learned the following: You want a continuous glucose monitor (CGM), and an insulin pump, with rapid-acting insulin. You do not want long-acting insulin. Especially not insulin glargine. It is a dangerous form of insulin: It is injected as a blob of insulin (“bolus”) under the skin. It’s supposed to stay in the place and slowly fritter and diffuse into the bloodstream. This effect is due to crystallization of the modified insulin amino acid chain in the pH level inside the body. It’s clever. However, if insulin glargine happens to go into a blood vessel, it works pretty much instantly. Then it is not long-acting at all. The subcutaneous environment is… living tissue. There are capillaries. It can and will go into directly into the bloodstream at some point. For some reason. At injection time, or due to bumps and jostles and physical pressure. Then there are no brakes on it. 24 hours’ worth of active insulin can go into the bloodstream. Then every cell in the body is signaled hard to pull glucose from the blood – all the flesh in the body rips the sugar out of your blood, leaving too little for the brain. People start losing consciousness in minutes, csn pass out, and may die. Insulin degludec is another long-acting insulin, one which is acceptable in this regard. The bolus stays glommed together due to polymerization of the insulin so it’s more robust. It also has molecular brakes on it kind of – if it does go into the bloodstream, it attaches to albumin in blood plasma and doesn’t become active all at once. However, using rapid-acting insulin in a pump is strictly better because this replicates the insulin oscillation of the pancreas. “[The pancreas’] basal insulin level is not stable. It oscillates with a regular period of 3-6 min. After a meal the amplitude of these oscillations increases but the periodicity remains constant. The oscillations are believed to be important for insulin sensitivity by preventing downregulation of insulin receptors in target cells. Such downregulation underlies insulin resistance, which is common in type 2 diabetes.”—https://en.wikipedia.org/wiki/Insulin_oscillation The insulin pumps work this way. They pump the insulin with a rhythm. It’s easier to control the blood sugar and you need less insulin. It’s crucially different and much better for long-term health and quality of life. The most important things are a good blood sugar sensor (continuous glucose meter or CGM), and a good pump with a good user interface. We haven’t tried one of the closed-loop automatically controlled pumps that reads from the sensor and autoadjusts, but it’s clear from the problems people have with badly designed pump that a good “semiautomatic” pump is better than a poor and unpredictable sensor-feedback-controlled automatic one. These have issues like sometimes cutting out without giving clear warning if blood sugar is high, and the blood sugar target range can’t be adjusted low enough (liability issues afaik) so people tend to stay at too high blood sugar levels which do long-term harm. I do assume that a good automatic pump is magic. Also, glucose metabolism and inflammation are intimately related, which ties into the vascular system. There is a lot more actionable science out on this than most doctors are aware of. (Overburdened medical personnel are unable to keep abreast of scientific developments.) |