• Hypoxaemia
• nosebleeds
• oxygen
• stroke
• thrombosis
• pulmonary embolism
• rare lung diseases
• asthma
• bronchiectasis
• COPD pharmacology
• cystic fibrosis

Individuals with pulmonary arteriovenous malformations (PAVMs)1 and hereditary haemorrhagic telangiectasia (HHT)2 commonly have low oxygen saturations and anaemia, parameters that are used in the general population to indicate medical fitness to fly3–5 (and online supplementary references). There are very limited published data on flight tolerance for HHT/PAVM patients.

Using the retrospective study methodology reported in full in the online supplementary material, we received 159 replies from 308 questionnaires sent out to individuals with definite HHT (response rate 51.6%). The average age at the time of reply was 55 years (range 18–90), 12 respondents had not flown and two (pilot and cabin crew) reported more than 10 000 flights. The remaining 145 HHT-affected respondents (96 (66%) with PAVMs) reported 18 943 flight hours over 3950 flights (online supplementary table 1).

The majority (111/145; 77% (95% CI 69.6% to 83.5%)) reported no in-flight or postflight complications. Six (4.1% (0.86% to 7.4%)) reported dyspnoea, two had a deep vein thrombosis and one suffered an ischaemic stroke while flying. The most common in-flight complications were HHT-related nosebleeds (epistaxis). Complications were more frequent during long-haul flights. Many participants listed flights over several decades, but none reported an increase in the frequency or severity of complications as they got older.

For participants with PAVMs who had not reported in-flight dyspnoea, there was a wide range in arterial oxygen saturation (SaO₂) levels (figure 1A). There was no difference in median SaO₂ between those who reported in-flight dyspnoea and those who did not (figure 1B). Flights where dyspnoea was reported did not correspond to times when SaO₂ were lowest for that particular individual (figure 1C). Similarly, there appeared to be no relationship between dyspnoea and either haemoglobin or serum iron (online supplementary figure 1). There was also no relationship between thrombotic complications and oxygen saturations/haemoglobin (online supplementary figure 2) or between in-flight nosebleeds and basal nosebleeds frequency (if at least once per month) or haemoglobin (online supplementary figure 3).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Sea level oxygen saturation in 96 participants who flew with pulmonary arteriovenous malformations (PAVMs). (A) Earliest and most recent arterial oxygen saturation (SaO₂) values for PAVM patients who did not report in-flight dyspnoea (improvements were the result of PAVM embolisation). (B) Mean erect oxygen saturations (SaO₂) at sea level for individuals who reported in-flight dyspnoea and those who did not. Horizontal bars denote medians. There was also no difference in earliest or latest SaO₂ (data not shown). (C) Serial SaO₂ in participants who reported dyspnoea over periods of 1–17 years (median 7.5). Circles indicate periods in which flights were reported to cause dyspnoea. The flight causing dyspnoea for participant 107 was the only long-haul flight taken by that individual.

In conclusion, and as discussed in more detail in the online supplementary material, the principal findings of this study were that flying appears safe for the majority of individuals with PAVMs and HHT despite abnormal oxygen saturations and haemoglobin concentrations. With the exception of nosebleeds, complications, when they occurred, were usually self-limiting. It was difficult to predict who will experience complications, with the best predictor appearing to be previous flight experience. The findings are surprising, and raise difficulties in recommendations for in-flight oxygen and prophylaxis of venous thromboemboli.