I've always said that there must be a hundred reasons for blood glucose levels to vary, and that we understand about fifty of them. I think we can now say that another reason has been identified.
A recent article in Diabetes Care, Changes in Altitude Cause Unintended Insulin Delivery From Insulin Pumps, investigated changes in insulin pumps' delivery rates during air travel. The authors state that they had found 50 children and adults using insulin pumps who reported hypoglycemia after beginning commercial air travel. Come to think of it, I have also had hypoglycemia during long flights, but I always associated the change in blood glucose to the inevitable hassles of flight days, with meal timing being off, and brisk walking through huge and confusing airport terminals to catch close connections, and I hadn't previously associated my hypos to the decrease in air pressure.
The authors, with assistance from airline and engineering experts, found that insulin pumps deliver unexpected extra insulin when at high altitude conditions. The high altitude conditions could be mimicked in a hypobaric chamber, and were actually studied during a jet airplane flight. They used 10 pumps, 5 from one manufacturer and 5 from another, and loaded the insulin pumps at sea level. Whether in the hypobaric chamber or in flight, decreased air pressure caused the pumps to deliver extra insulin; the flight caused even more insulin delivery than the chamber (1.37 units of extra insulin for one brand and 1.01 units for the other). And approximately the reverse was seen when air pressure was subsequently increased: insulin was sucked back into the pumps (but less than the outflow: 0.87 and 0.58 units for the two brands of pumps).
The explanation for the extra insulin delivery is gas bubbles: some bubbles that form when air pressure is reduced, and others that already existed, but both of which expand as pressure reduces, thus forcing insulin out of the pumps' storage chambers. The more insulin in the pump, the more bubbles. Then, as the airplane descends and air pressure increases, the bubbles redissolve and insulin is sucked back.
The research was very clear on this, but the logical question of whether this does in fact result in early-flight hypoglycemia and subsequent hyperglycemia when landing wasn't studied. They hedge on this: "Excess insulin delivered during ascent in an airplane may cause hypoglycemia 1-2 h later. Hyperglycemia may follow due to decreased insulin deliver during descent. Whether abnormalities in insulin delivery during ﬂight cause clinical effects depend on factors such as insulin sensitivity, current glycemic control, food intake, and pump settings, among others."
But it sure does sound logical. In pump patients with ultra-tight diabetes control, it's one more factor that can influence BG levels.
(The authors also present lengthy, overly complex, and somewhat confusing recommendations for pumpers when flying, but that's another story. My plan will simply be to cut back on my pump's basal rate before takeoff, and resume normal insulin delivery when the plane lands.)