Hypoglycemia in a pilot with insulin-treated diabetes

A 38-year-old pilot, with type 1 diabetes since adolescence, is at the controls of a short commercial flight. During descent, the continuous glucose monitor alarm goes off and shows a low value with a downward arrow. The copilot notices that he responds slowly to control calls and mixes up words when reading the checklist. He says he is fine. Shortly before, he had eaten less than usual at lunch because of a delay. The scene is fictional and was written for teaching.

This text uses the scenario as a setting for clinical reasoning. The trap is waiting for tremor and sweating before thinking of low glucose. In people with long-standing diabetes, confusion can come first, and the person themselves is the last to notice. The cockpit adds two doubts: does the monitor measure well when pressure changes, and is the insulin pump delivering what it should? The text is not an aviation authority protocol. If you play the case of the day on NEXO, you will recognize the challenge: the most important clue may come from the person beside you.

Key points

• Pilots with insulin-treated diabetes fly in some countries under protocols with measurements before and during flight.
• In people who have had type 1 diabetes for years, the glucagon response to low glucose is usually lost.
• Repeated hypoglycemia reduces warning signs, and confusion can appear before tremor.
• Slow speech and mixing up words during descent call for a measured glucose before any other hypothesis.
• In a conscious person, oral glucose is the preferred treatment, with a new measurement after a few minutes.
• Bubbles in the pump cartridge can change insulin delivery during climb and descent.

Why this is a reasoning scenario

The case brings together a common emergency, a person who may not recognize their own condition, and a place where the decision must be quick. Low glucose can cause confusion, loss of consciousness, seizures and even death, and how each person reacts depends on how many times they have had hypoglycemia before (Seaquist et al., 2013). In the cockpit, the time between noticing and acting counts.

The series on aviation medicine has already covered the assessment of sudden incapacitation risk from heart disease and how medications and substances affect performance. Here the focus is a condition the pilot knows well and treats every day, and that can still surprise.

What the scene teaches

The pilot's clearest symptom is something he does not notice. The one who notices the slow speech is the copilot. Reasoning must give weight to what others observe, not only to what the patient reports. The post on pretest probability in diagnosis shows why the person's group changes the probability before any test: in a pilot who uses insulin and ate little, hypoglycemia is the first hypothesis.

How pilots on insulin fly today

For a long time, insulin use prevented work as a commercial pilot. In 2002, Canada was the first country to allow selected people with insulin-treated diabetes to fly commercial aircraft (Meçani et al., 2026c). In 2019, the United States aviation authority created a special certification protocol that relies on continuous glucose monitor data (Meçani et al., 2026b). An Italian review on air traffic controllers recalls that hypoglycemia is the main incapacitation risk for crew and that the reassuring results of the European protocol for insulin-treated pilots opened the way for updated requirements (Verde et al., 2025).

The European protocol adopted by the United Kingdom, Ireland and Austria calls for glucose measurements before the flight, in the 30 minutes before takeoff, every hour during the flight and in the 30 minutes before landing, all checked by the copilot and read aloud. The values are classified in a traffic light system: green from 5.0 to 15.0 mmol/L, amber from 4.0 to 4.9 and from 15.1 to 20.0 mmol/L, and red below 4.0 or above 20.0 mmol/L, which requires immediate action (Garden et al., 2025). A systematic review describes the same system in mg/dL (Meçani et al., 2026a).

What the protocol data show

In the observational study with certified pilots, among those using an insulin pump, more than 99 out of every 100 measurements taken on duty fell in the green range, and no in-flight measurement fell in the hypoglycemia range (Garden et al., 2025). The point for reasoning is that the protocol assumes measuring and acting early. The pilot in this case is in a situation where that process failed: his glucose fell, and he did not react on his own.

The post about fitness to fly discusses how the doctor assesses who boards. The logic of the pilots' protocol is similar, but repeated every hour: fitness is checked during the flight, not only before it.

Pathophysiology: when the defense fails

In a person without diabetes, a drop in glucose reduces insulin release, increases glucagon release and activates adrenaline. In type 1 diabetes, insulin comes from outside and keeps acting even when glucose falls. The review by McCrimmon and Sherwin recalls a second problem: within up to 5 years of diagnosis, almost all people with type 1 diabetes stop producing an adequate glucagon response to hypoglycemia (McCrimmon and Sherwin, 2010). Glucagon is the main fast-acting hormone against low glucose, and without it recovery becomes much slower.

The loss of warning signs

The third link is adrenaline. The joint review by the American Diabetes Association and the Endocrine Society describes autonomic failure associated with hypoglycemia: glucose defense fails because the fall in insulin is missing, the rise in glucagon is missing and adrenaline rises less, and the lack of hypoglycemia awareness comes from a reduced sympathetic response (Seaquist et al., 2013). This failure is usually caused by recent hypoglycemia and is at least partly reversible when new episodes are avoided.

The autonomic signs are tremor, sweating, palpitations and hunger. When they are missing, or when episodes recur hours or days apart, lack of awareness is likely (Seaquist et al., 2013). For the pilot in this case, that means the first manifestation may be slowed thinking, which he himself has difficulty recognizing.

The threshold moves

The same review explains why it is hard to fix a single number that defines hypoglycemia. The threshold at which symptoms appear falls after recent hypoglycemias and rises in people with poorly controlled diabetes and few episodes (Seaquist et al., 2013). That is why the authors propose an alert value, not a disease cutoff. A pilot with tight control and frequent episodes may be confused at a glucose level that, in another person, would still cause only hunger.

How the picture presents in the cockpit

The proposed alert value is a glucose at or below 70 mg/dL, or 3.9 mmol/L, measured by capillary blood glucose or by continuous monitor (Seaquist et al., 2013). The international consensus on time in range classifies readings from 3.0 to 3.8 mmol/L as level 1 below target and readings below 3.0 mmol/L as level 2 (Battelino et al., 2019). Severe hypoglycemia is defined by the need for help from another person to receive carbohydrate, glucagon or another measure (Seaquist et al., 2013).

In the cockpit, this definition carries special weight. If the copilot needs to intervene, the episode is already severe by definition, even if the pilot can still speak and move. The discussion about severity stops depending on the number on the monitor and starts depending on the ability to act alone.

Signs that count

In the scene, the signs that count are slow speech, word swapping, delayed response to calls and insisting that everything is fine. He also ate less than usual, which takes glucose out of the system while insulin keeps acting. The monitor alarm with a down arrow adds the trend: glucose is falling, not just low.

Differential diagnosis

Slow speech in the cockpit has other causes, and some call for very different management. The exercise is to keep the alternatives open while glucose is measured, without letting them delay the most likely treatment.

Hypothesis	Clues in favor	Clues against
Hypoglycemia	Insulin, smaller meal, monitor alarm, slow speech	Normal glucose confirmed on capillary measurement
Hypoxia	Altitude, other people with symptoms, cabin alarm	Rapid improvement with glucose; descent with a pressurized cabin
Stroke	Weakness on one side, crooked face, persistent slurred speech	Complete improvement after glucose
Fatigue	Long duty day, short sleep, diffuse slowness	Monitor alarm and measured low glucose

Hypobaric hypoxia also causes slowness and errors of judgment that the person does not notice. The practical difference is that hypoxia affects the environment, and other crew members may have similar symptoms, while hypoglycemia is individual and corrects with glucose.

Stroke and hypoglycemia

Stroke is the alternative that weighs most. Hypoglycemia can mimic a focal neurological deficit, and that is why glucose must be measured in anyone with sudden neurological change. If speech and strength return to normal after glucose, the picture points to hypoglycemia. If anything persists, stroke returns to the top of the list, and landing at the nearest airport with resources becomes the priority.

Fatigue and disorientation

Fatigue and spatial disorientation enter the differential of any error in the cockpit. They rarely explain a low glucose alarm, but they can coexist with it. Meal and sleep times change with the schedule, and that also changes the need for insulin. The post on differential diagnosis with clues shows how to keep alternatives open without stalling the decision.

The monitor and cabin pressure

The question many ask is whether the continuous monitor measures well during pressure changes. A study in a hypobaric chamber compared the continuous monitor with capillary blood glucose in people with type 1 diabetes and found a very strong correlation between the two, on the ground and in flight simulation (Fan et al., 2025). The authors conclude that the two methods perform comparably and that all the devices tested served for care decisions during the simulation.

Data from commercial flights add a nuance. In a study with people with type 1 diabetes using several models at the same time, two systems showed higher and consistent accuracy in all phases of flight, and two others showed more variation (Meçani et al., 2026b). The authors conclude that modern monitors can support diabetes care in aviation, but that performance differs between systems.

What to do with the alarm

For the reasoning of the case, the alarm is a strong clue, not a verdict. The systematic review records that the monitors showed small signal variation under hypobaric conditions (Meçani et al., 2026a). If there is time and someone to help, capillary blood glucose confirms the value. If the pilot is confused, oral glucose comes before confirmation, because treating a hypoglycemia that does not exist causes little harm, and failing to treat one that does can cause a lot.

The consensus on time in range defines the bands used to read the monitor data and organizes the reading into time in range, above and below it (Battelino et al., 2019). In the review after the episode, time below range in the previous weeks helps to understand whether the pilot had been having repeated hypoglycemias.

The insulin pump on the way up and the way down

The second problem with pressure is insulin delivery. In a hypobaric chamber, the drop in pressure caused bubbles to form and grow inside the insulin cartridge. Full cartridges delivered 0.60 U of insulin more during a 20-minute ascent and 0.51 U less during the descent, compared with ground level (Garden et al., 2025). In a simulated rapid decompression, the excess was 5.6 U (Garden et al., 2025).

The explanation is physics. The systematic review notes that the phenomenon is explained by Boyle's law, according to which the volume of a gas increases when pressure falls: as cabin pressure rises and falls, air bubbles in the pump or the tubing expand and push insulin even though the pump has not failed (Meçani et al., 2026a).

What happened to the pilots

In the observational part of the same study, pilots who used a pump had fewer values in the low red range and fewer out-of-range values in flight than those who used multiple injections, and they kept tighter control (Garden et al., 2025). In other words, the physical effect exists, but in practice the pump coexisted with good control. Another study on commercial flights assessed automated insulin delivery systems and found control that was maintained and safe, similar to ground level, with no level 2 hypoglycemia (Meçani et al., 2026c).

For the pilot in this case, the question about the pump comes after treating the glucose. Hypoglycemia during descent fits with less food and active insulin, and the pump tends to deliver less, not more, at that stage (Garden et al., 2025). The most likely cause remains the smaller meal.

Management on board

The first measure is flight safety: the copilot takes control. Then comes treatment. The American Diabetes Association standards of care recommend 15 to 20 g of glucose as the preferred treatment for a conscious person with hypoglycemia, although any carbohydrate containing glucose can be used (American Diabetes Association, 2013). If the measurement taken 15 minutes later still shows hypoglycemia, treatment should be repeated, according to the same document. When glucose returns to normal, the person should eat a meal or a snack to keep the episode from coming back.

When the person cannot swallow

Oral glucose assumes someone who is awake and able to swallow safely. The same standards recommend that glucagon be prescribed for everyone at significant risk of severe hypoglycemia and that caregivers or family members learn to give it, since administration is within reach of laypeople (American Diabetes Association, 2013). This text does not give a glucagon dose. Management follows the flight kit and the medical guidance available.

After the episode

A pilot who had a hypoglycemia that required help is no longer pilot in command on that flight. The team records the time, the monitor value, the capillary measurement, the amount of glucose given and the response. The standards of care recommend that impaired awareness of hypoglycemia or one or more severe episodes lead to a review of treatment (American Diabetes Association, 2013). In the pilots' protocol, this record also guides the conversation with the medical examiner about certification.

Fear of losing the license can lead to hiding symptoms. That is why an environment in which a pilot reports hypoglycemia without fear is safer than one in which the pilot hides it.

Step-by-step reasoning in the case

The first step is to name the risk: a pilot with type 1 diabetes for many years, using insulin, with a smaller meal than usual. The baseline rate of hypoglycemia in this context is high, and the history of long-standing disease suggests loss of the glucagon response (McCrimmon and Sherwin, 2010).

The second step is to give weight to what the copilot sees. Slurred speech and word mix-ups count for more than the pilot's statement that he is fine, because impaired awareness is part of the picture.

The third step is to treat before finishing the investigation. Oral glucose is fast, cheap and safe, and a capillary reading can be taken right after. The new reading at 15 minutes tells you whether the treatment was enough (American Diabetes Association, 2013).

The fourth step is to review the differential with the response. If speech returns to normal, hypoglycemia explains the picture. If something persists, stroke and hypoxia come back onto the list, and the decision about landing changes. The post on diagnostic pause describes this moment of checking before closing the case.

The fifth step is to look back. The smaller meal, the lunchtime insulin, the pump and the monitor history from the previous weeks help explain why glucose fell and why he did not notice.

One last point is reviewing your own reasoning. In the cockpit, this includes asking whether the explanation you chose explains all the signs, and not just the most striking one.

Common errors

The first error is waiting for tremor and sweating. In people with long-standing diabetes and frequent hypoglycemia, these signs may be absent, and confusion comes first (Seaquist et al., 2013).

The second error is accepting the word of a confused person. The pilot who says he is fine may be exactly the one who does not perceive hypoglycemia.

The third error is dismissing the alarm out of distrust of the device at altitude. Studies show that the monitor and capillary glucose behave comparably under pressure changes (Fan et al., 2025).

The fourth error is stopping after the first glucose reading. The repeated reading and the next meal are part of the treatment (American Diabetes Association, 2013).

The fifth error is blaming the pump before looking at the meal. The physics of bubbles is real, but the most common cause of hypoglycemia remains too much insulin for the food available.

The sixth error is closing the differential too early. The post on signs, symptoms and history shows how to bring together the account and what is observed. If the response to glucose is incomplete, stroke needs to come back to the table.

How NEXO helps train this reasoning

To use the case of the day as training, try three habits. Before the first guess, ask which reversible cause would explain everything. With each clue, see whether it came from the patient or from those around them. After the result, note which serious hypothesis you ruled out and with which clue.

Three exercises with cases

Look for the reversible cause. For each case of confusion, write which simple test would change management within minutes.

Separate account and observation. List what the patient says and what others see, and compare the weight of each.

Plan with the response. Write what you would do if the patient improved and what you would do if they did not.

The post on case-based learning in clinical practice shows how cases help fix this habit.

The limits on using the game are described on the NEXO economy page. If you research the teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information under curation and under creators, and institutions can learn about the proposal on the partnerships page.

Frequently asked questions

Can pilots with insulin-treated diabetes fly?

Yes. Canada was the first country to allow it, in 2002 (Meçani et al., 2026c), and the European protocol requires glucose readings before and during the flight (Garden et al., 2025).

Can confusion come before tremor in hypoglycemia?

Yes. Recent hypoglycemia reduces the sympathetic response and the warning signs (Seaquist et al., 2013).

Does the continuous monitor work with pressure changes?

Yes. In a hypobaric chamber, the monitor performed comparably to capillary glucose (Fan et al., 2025).

Does cabin pressure change insulin pump delivery?

Yes. Bubbles in the cartridge led to over-delivery on ascent and under-delivery on descent in simulation (Garden et al., 2025).

Is oral glucose the preferred treatment in a conscious person?

Yes. Standards of care indicate oral glucose and a repeat measurement after a few minutes (American Diabetes Association, 2013).

Final checklist

• I considered hypoglycemia in any pilot who uses insulin and changes behavior.
• I gave weight to what the copilot observed.
• I handed over control before treating.
• I gave oral glucose to the conscious person.
• I measured glucose again after treatment.
• I offered a meal after correction.
• I kept stroke, hypoxia and fatigue in the differential.
• I checked the pump and the monitor history after the episode.
• I recorded values, times and responses.
• I referred the person for reassessment of treatment and certification.

Want to practice the reasoning with progressive clues? Play the case of the day.

References

1. American Diabetes Association. Standards of medical care in diabetes--2013. Diabetes Care. 2013. PMID 23264422.
2. Battelino T, Danne T, Bergenstal RM, et al. Clinical Targets for Continuous Glucose Monitoring Data Interpretation: Recommendations From the International Consensus on Time in Range. Diabetes Care. 2019. PMID 31177185.
3. Fan KS, Manoli A, Baumann PM, et al. Comparison of continuous glucose monitoring with self-monitoring of blood glucose in type 1 diabetes in the changing atmospheric pressures in aviation: a hypobaric flight simulation. Diabetologia. 2025. PMID 39888427.
4. Garden GL, Fan KS, Paterson M, et al. Effects of atmospheric pressure change during flight on insulin pump delivery and glycaemic control of pilots with insulin-treated diabetes: an in vitro simulation and a retrospective observational real-world study. Diabetologia. 2025. PMID 39496965.
5. McCrimmon RJ, Sherwin RS. Hypoglycemia in type 1 diabetes. Diabetes. 2010. PMID 20876723.
6. Meçani R, Basta S, Baumann PM, et al. In-flight glycaemic control using automated insulin delivery systems: A within-subject comparative pilot study. Diabetes Obes Metab. 2026. PMID 41424231.
7. Meçani R, Basta S, Baumann PM, et al. Performance of continuous glucose monitoring systems (CGMs) during commercial flights in T1DM: A within-subject comparative pilot study. Diabetes Obes Metab. 2026. PMID 41480660.
8. Meçani R, Muka T, Cigler M, et al. The use of diabetes technology under aviation conditions: A systematic review of human and in-vitro studies and future considerations. Diabet Med. 2026. PMID 41578663.
9. Seaquist ER, Anderson J, Childs B, et al. Hypoglycemia and diabetes: a report of a workgroup of the American Diabetes Association and the Endocrine Society. Diabetes Care. 2013. PMID 23589542.
10. Verde P, Piccardi L, Gentile S, et al. Work-Related Stress and Glucose Regulation in Air Traffic Control Officers: Implications for Medical Certification. Biomedicines. 2025. PMID 41007688.