Professor John Royle
I thought I should start right at the beginning and explain to the non-medical audience exactly what a thrombosis is. Blood is normally fluid. When it coagulates it forms a clot. You have all seen examples of that when you scratch or cut yourself. Coagulation stops the bleeding from continuing indefinitely. When a clot forms inside a blood vessel it is called a thrombosis.
In the legs, there are two sets of veins -- the superficial surface veins, which when enlarged may become varicose veins, and the deep veins of the leg. The deep veins cannot be seen or felt. They conduct the blood back towards the heart with the assistance of multiple valves. A thrombosis in a superficial vein is called an SVT or superficial venous thrombosis. A thrombosis in a deep vein is called a deep venous thrombosis or a DVT. If a piece of clot breaks off from the inside of the vein in the leg where it formed, it will travel to the heart, and then on to the lungs. This transported clot is called an embolus and when it reaches the lungs, it is called a pulmonary embolus or PE.
It is of interest that the ancient Greeks knew quite a lot about varicose veins. There is a tablet in the museum in Athens, which was taken from the ruins of the Asklepeion, which shows a man’s leg on which there are varicose veins. Hippocrates, who taught in the medical school on Cos in the fifth and fourth centuries BC, knew of the relationship between venous disease and ulceration of the leg. He advised the application of “a double compress, wetted with wine”, when treating an ulcer associated with varicose veins.1 Today the essential part of treatment of a venous ulcer is compression.
In the 19th century, a German pathologist Rudolph Virchow enunciated the triad of requirements for thrombosis, which are still the basis of treatment and research today.2 The triad is:
A. venous stasis, (i.e. sluggish or slow flow of blood)
B. vessel wall abnormality,
C. hypercoagulability (i.e. increased tendency for blood to clot).
In 1940, Dr Keith Simpson was the assistant lecturer in forensic pathology at Guy’s Hospital, London. (He later became quite a famous forensic pathologist). Simpson described an increase in deaths from pulmonary emboli in autopsies done by the coroner’s pathologist on people with sudden death. In September/October 1939 there were four deaths due to pulmonary embolism. In September/ October 1940, during the London “blitz”, there were 24 deaths from pulmonary embolism. 21 of these deaths had occurred after a night or a succession of nights in an air raid shelter, sleeping on a deck chair or similar seat. This was the first account that prolonged sitting could cause a DVT.2 Simpson advocated replacement of the deck chairs by bunks, which was done, and the incidence of pulmonary embolism fell again.
In 1954 John Homans (whose name is well known to the medical audience because of Homans’ sign for DVT) reported five cases of DVT after prolonged sitting.4 Two of these cases were of DVT following air travel. These were the first reported cases of DVT after air travel. The first case was a doctor who flew from Boston to Venezuela and back (14 hours each way). On reaching home, an hour or two after leaving the plane, he noticed a painful calf. A DVT was diagnosed and he was treated with warfarin.
Moving on 20 years to 1974, the President of the United States of America, Richard Nixon, travelled to Salzburg, then to the Middle East, and then to the Soviet Union. Nixon’s personal physician accompanied him. After the first leg of the journey, to Salzburg, the physician confirmed that Nixon had a DVT and advised against further travel. Nixon ignored that advice and travelled on to Egypt, where he had a further clot. He then went on to the Soviet Union and had yet another one. After his resignation in August 1974, he continued to experience problems with swelling of the left leg. In October 1974, Nixon underwent an operation in which a Miles clamp was placed on his left common iliac vein to prevent pulmonary embolism. He nearly died of this surgery because of bleeding complications. Because of his serious condition in hospital, he couldn’t testify at the Watergate cover-up trial in Washington. The references for these details are interesting -- from five issues of the Los Angeles Times.5,6,7,8,& 9
Over the next 20 years sporadic reports of thrombosis related to air travel appeared in the literature. One author, Cruickshank, dubbed the syndrome -- the “economy class syndrome”.10 Professor Reg Lord from Sydney reported 45 cases at the 21st World Congress of the International Society for Cardiovascular Surgery in Lisbon, Portugal in 1993. He gave a further paper at an International meeting in Hawaii later that same year. In Hawaii I spoke after him reporting that two of my friends had died of pulmonary embolism after air travel. One was a medical colleague, who flew from Singapore to Auckland and dropped dead almost immediately after he alighted from the plane. My main purpose in mentioning the Hawaiian meeting is because of my own personal experiences. The day after the meeting, my wife Pam and I hired a car and drove around the big island all day. The temperature was in the high 90s. We flew back to Honolulu, and I had a beer while waiting for the Qantas flight home. Once aboard I curled up in a seat in a corner, in economy class, and I went to sleep. After arriving home in Melbourne next morning, I noticed that my urine was very strong - I was obviously dehydrated. I went over to Heidelberg to visit my rooms, and then crossed the street to visit the Austin Hospital opposite. I noticed a pain in the left calf. I knew what it was and went straight to the vascular laboratory where scans showed a DVT. In summary, I had had a long flight, was dehydrated, had some alcohol, was in the cramped position in economy class, and was immobile - asleep. I subsequently had a second DVT (not after air travel), and now wear elastic stockings permanently to control swelling in my leg.
Subsequent to that Hawaiian meeting, Bo Eklof, a Hawaiian surgeon, and his colleagues analysed the records of all cases of pulmonary embolism and DVT at the Straub Clinic in Hawaii from 1988 to 1994.11 He found that 44 patients had developed their clots after air travel. 24 patients had a DVT alone, five patients had a pulmonary embolus alone, and 11 had both. He persuaded his colleagues at the Military Hospital in Hawaii to do a similar retrospective study. They found 36 patients over the same period of time. Thus, there were 80 patients in Hawaii, over a period of six years, who had been hospitalized for DVT/PE after air travel. This is an incidence of 13 to 14 patients per year. Of course, it takes a long flight to get to Hawaii. Thus the incidence in Hawaii is higher than in any other American state. In Australia, you need a long flight to go anywhere overseas, hence the higher incidence in Australia than in many other countries. The French also studied the incidence of DVT/PE after flights to and from Reunion Island, which is a long flight, and they found a significant incidence of DVT/PE, (46 cases in 1 year).12
Lord, Eklof, the Military Hospital in Hawaii, and the Reunion Island study all found that the majority of patients with DVT/PE after air travel had some disposing conditions for DVT, and I will discuss these later.
- What about the incidence in the normal ordinary flying population? Two good studies have been done. At Charles de Gaulle airport in Paris, France, all patients who on arrival require urgent medical care, are taken to the Hopital Avicenne. An analysis was done of all patients arriving alive at Charles de Gaulle airport, who were taken to the Hopital Avicenne with suspected pulmonary embolism between 1993 and 2000.13 Of 170 patients suspected of pulmonary embolism, two thirds(114) did not have it and one third (56) did. They analysed these 56 patients with respect to the distance travelled before arrival. Over the same period of time 135 million passengers arrived at Charles de Gaulle airport. From the airport statistics, they were able to determine the number of kilometres passengers had travelled to arrive in Paris. The graph (figure 1) summarizes an analysis of the results. The number of cases of pulmonary embolism per million arrivals is shown. Among those with pulmonary embolism, there were no patients who had travelled less than 2500 km. The risk of pulmonary embolism under 5000 km travel was very small. Thereafter the incidence rose to 4.77 per million if the distance travelled was greater than 10,000 km.
Figure 1: Charles de Gaulle Airport Study
Severe Pulmonary Embolism after Air Travel
- A similar study was done at Madrid airport.14 Once again all passengers requiring urgent medical treatment are taken to the same hospital (Ramon y Cajal Hospital). A similar result was obtained. This is shown in the accompanying graph (figure 2). The graph shows that no pulmonary embolism was found with less than six hours travel, but the number was 1.65 per million with travel lasting more than eight hours. This is not dissimilar to the second-highest group in the French study. Madrid airport only receives 30% of the passengers that arrive at Charles de Gaulle airport, and there are fewer patients travelling the longer distances before arrival. However, the same trend emerges -- the longer the travel, the greater the risk.
Figure 2: Madrid-Barajas Airport Study
Pulmonary Embolism after Air Travel
- Neither of these studies gives any idea of the number of passengers who develop symptoms a day or two after arrival (like I did). Thus the true incidence of pulmonary embolism would obviously be higher than indicated by these studies.
- The academic purists would say that you need a double blind randomized prospective study to solve the problems presented. A randomized study was done by Scurr from London.15 Scurr was interested to study passengers who did not have any predisposing factors to DVT. He advertised in the daily press, and at travel agencies, for volunteers. All volunteers had to be over the age of 50. Any prospective traveller who was suffering from cardiac disease, respiratory disease, malignancy, recent injury, a past history of DVT or any serious medical condition, was excluded from the study. All volunteers had to consent to have an ultrasound performed before travel, the travel had to include a flight of more than eight hours, and after the return flight (again of more than eight hours), they had to have a second ultrasound. Blood was taken for analysis of known thrombophilic (blood clotting) factors before the initial flight and again on return. The passengers were divided into two groups, half were instructed to wear elastic stockings on all flights, and the other half did not wear elastic stockings at all.
- Of the 115 passengers in the stocking group, no one had a DVT. Of the 115 patients in the control group without stockings there were 12 passengers who had a DVT. This is an incidence of 10%. In the stocking group 4 passengers had a superficial venous thrombosis. In each case, the SVT occurred at knee level, where the upper end of the stocking impinged on some varicose veins. There were no superficial venous thromboses in the control group. There were no extensive serious DVT’s and there were no pulmonary emboli in either group. We should note that in all passengers there were two flights (i.e. there and back). We should therefore perhaps consider the incidence to be half of 10% (5%).
- Since that time, an international group from Italy, London and Australia (the Australian representative is Ken Myers) have done a series of studies published as LONFLIT 1, 2, 3 and 4.16, 17, 18 & 19 These studies have been prospective studies, with controls, testing various pharmacological agents and elastic stockings. They found that the incidence of DVT’s in the control groups was similar to that found by Scurr, 4% - 6%. They showed that low-risk passengers are at very low risk. The LONFLIT 3 study.18 showed a significant reduction in DVT in high risk passengers given an injection of low molecular weight Heparin before the flight. In the LONFLIT 4 study,19 there were two groups. One group flew from London to New York (seven to eight hours). The other group flew from London to Phoenix (11 to 12 hours). It was on the longer flights that problems occurred. Clearly, the longer the flight, the greater the risk. The LONFLIT consortium have decided that factors putting patients at higher risk include: -
- A past history of DVT or SVT
- Coagulation disorders.
- Severe obesity.
- Taller than 190 cm.
- Heavier than 90 kg.
- Bone and joint problems restricting mobility.
- Cancer within the previous two years.
- Varicose veins.
- Clinical cardiovascular disease.
- Clinical disease requiring medical treatment.
Although this list does not include age, other studies such as a study done on arrivals in Western Australia20 have clearly indicated that this is a factor.
Now let’s look at the medical reasons for this. Why do thromboses occur? I have already described Virchow’s triad of stasis, vessel wall abnormality, and hypercoagulability.
- STASIS. In the LONFLIT list of risk factors, stasis figures in many of them. Mobility is difficult in passengers who are very obese or who have severe arthritis.
- VESSEL WALL ABNORMALITY. Previous DVT and SVT leave residual abnormalities in the vessel wall. I will consider some experimental work in a moment.
- HYPERCOAGULABILITY. Blood contains red blood cells, white blood cells, small particles called platelets, and many different kinds of proteins. Some of these proteins are involved in the clotting process. The way the red blood cells, the platelets and the proteins inter-react to form a clot is very complex, and I do not intend to try and explain this. However there are two readily identifiable genetic blood problems which give an increased tendency for blood to clot. These are: -
a. Factor V Leiden gene mutation.
b. Prothrombin G20210A gene mutation.
Scurr took blood from all the participants in his study. 11 of 200 had factor V Leiden abnormality, which is in accord with the generally accepted community incidence of 5%. Four passengers had the prothrombin abnormality, and one passenger had both the prothrombin and the Leiden factor V abnormalities. This passenger, who had both genetic abnormalities, had an SVT. Two passengers with DVT were factor V Leiden positive. Other retrospective studies have found these genetic abnormalities in a significant number of patients.
Do healthy people get serious DVT’s? Yes, but it is quite uncommon, although Scurr found minor DVT’s in 10%. Most of the retrospective studies found that one or more of the high risk factors already listed were present. (Lord found 76% had one risk factor, Eklof 68%, the Hawaiian Military Hospital, 44% and the Reunion Island study 82%.)
An airline study showed that 0.25% of pilots got a DVT over a 10 year period, an incidence of one in 400. A pilot friend of mine, when I told him I was giving this paper in June, said two of his pilot colleagues had had severe DVT’s. One was off work for six months as a result of his DVT.
The only ”normal” person that I ever treated for a DVT was a young woman of 27. She was a member of the Australian women’s cycling team. She competed in France and then returned to Australia next day. She was almost certainly dehydrated. She may have been on the contraceptive pill, which some authors have labelled a “risk” factor. I doubt that the contraceptive pill is a “risk” factor. If the contraceptive pill were a real predisposing factor, when you consider the number of young women who travel, we would have hundreds of thousands of young women with DVT‘s after air travel.
Let us consider injury. We have considerable evidence from the orthopaedic literature, that injury to the legs predisposes to DVT. A study in the United States of America, some years ago showed that 17% of patients with an injury to the ankle, sufficient to require a plaster or plaster slab but who did not require hospitalisation, developed a DVT after trauma. Steve Waugh is the best example of this. He tore a calf muscle in a match in England, returned to Australia soon after, and had a DVT. Bob Carmichael, a leading Australian tennis player, was one of the five patients in Bo Eklof’s list in Hawaii. He had a serious knee injury when competing in the Volvo Masters tournament in the central United States. He was returning to Australia when he suffered a DVT and a pulmonary embolus on the way. He had to interrupt his flight in Hawaii, where he was treated by Bo Eklof. Injury increases the stickiness of platelets, which makes blood more likely to clot. It also results in reduced mobility, which also makes the blood more likely to clot.
Is there anything on a plane which may predispose to clots? We have hundreds of thousands of people travelling long distances by bus and by car. There is not an epidemic of DVT’s in them, although an occasional case is seen, as was reported by Homans in 1954. The most obvious difference is the degree of mobility, or lack of mobility. However on a plane there is also dehydration -- the air circulated through the plane is dry, and therefore every passenger is steadily dried out. The air breathed out is saturated with water vapour, and the air breathed in has none. Alcohol, tea, and coffee all tend to produce dehydration. Dehydration, ”thickens the blood” and hence predisposes to thrombosis.
On a plane there is hypoxia. The oxygen level in the blood as you sit listening to me would be about 95 mmHg. When depressurized to 2500 m or 8000 feet (the normal on a plane in flight), the oxygen level would be 56 mmHg in healthy passengers.21 In passengers with heart disease or respiratory disease it would be lower still. This is called hypobaric hypoxia. Quite a number of experiments have been done on blood coagulation factors under conditions of hypobaric hypoxia. 20 volunteers had blood taken before, during, and after a return flight from Vienna to Washington. They showed the blood changes which predispose to thrombosis.22 In another experiment 71 healthy volunteers had blood taken before and after three separate activities. All participants were exposed to an 8 hour flight, an 8 hour movie marathon and 8 hours of normal activity at home. Each exposure was separated by a 2 week interval. Blood changes predisposing to thrombosis were seen in the flight study only.23
The second of Virchow’s triad was vessel wall abnormality. I have already mentioned the changes after DVT and SVT. Endothelial cells are the cells which line blood vessels, arteries and veins. Animal experiments have shown damage to endothelial cells can be caused by hypobaric hypoxia.24 Cultured human umbilical vein wall cells show similar endothelial cell damage with hypobaric hypoxia.25 These changes are more pronounced on reoxygenation. It has not been possible to look at the cells lining vein walls in passengers before and after long flights but perhaps these experiments explain why a lot of DVT’s and pulmonary emboli occur after landing, rather than during the flight. I could go on and describe further studies, but that would bore you.
In a passenger over the age of 50, after a long flight:-
The risk of a minor DVT is one in 20.
Some of these minor DVT’s have a more serious DVT.
Some of these serious DVT’s have a really major DVT or a pulmonary embolus. I can’t tell you the percentage risk of this. The figures I quoted at the beginning about Paris and Madrid airports are just a fraction of the number of passengers admitted to hospital in the two weeks after completion of air travel.20
Now comes the crunch, and what you are all waiting for. As individuals, what do we do?
On short flights e.g. in Australia or New Zealand no special precautions are necessary. However when we fly overseas some precautions are advisable for everyone.
- In young and healthy people, the risk is very low. They should drink plenty of non-alcoholic fluids and move the legs around from time to time. At refuelling stops, walk around the terminal for 40 to 50 minutes instead of sitting and drinking coffee or beer. Nothing more is necessary.
- In those people with mild or very moderate risk, e.g. anyone who is healthy but over 50, drink plenty of non-alcoholic fluids and move the legs around as above, but in addition wear below knee elastic stockings. Walking around the terminal at stops is more important in this group.
- In those people in the high risk category, anticoagulation with warfarin or heparin is needed. The LONFLIT 3 study showed that low molecular weight heparin (LMW Heparin) was very effective in high risk passengers.18 Warfarin takes some time for the dose to be stabilized. Therefore, by far the easiest option is LMW heparin. This is not difficult to administer. The drug comes ready to use in loaded syringes, with the dose already worked out. The passenger just needs to take the rubber bung from off the needle, plunge the needle into some fat near the navel and squeeze the syringe. When the drug is obtained from the chemist, it comes in a box with 10 loaded syringes. You don’t need that many, and therefore you can practise, using one or two syringes. The easiest way of obtaining practise is to inject the drug into an orange. During my time in practice I had many patients who required heparin before long flights and they did not find it difficult to self administer the drug.
- If in doubt about which category you might fit into, you should see your doctor and discuss it with him/her. You will need to see the doctor anyway if you are in group 3.
I am often asked about aspirin. Theoretically, as it reduces the stickiness of platelets, it should be effective. However there are no studies that have shown aspirin to have any beneficial effect in the prevention of DVT. The LONFLIT3 study did not show any benefit from taking aspirin.18 I would not advise anyone against taking aspirin but if you are in category 3 you need heparin.
I am also asked about travelling economy class or upgrading to business class or first class. There is no evidence that thrombosis is more likely in economy class than in business class or first class.26
IN SUMMARY remember:-
- Drink water,
- Move around,
- Elastic stockings if over 50, and
- Heparin injections if at high risk.
I have given you an overview of the problem, and I hope I have provided you with some suggestions on what you should do personally, next time you fly.
- Adams F. (1849) The Genuine works of Hippocrates translated from the Greek. London Sydenham Society. Reprinted by Williams and Wilkins, Baltimore 1939. Page 339.
- Virchow R. (1846) Die Verstopfung der Lungenarterie und ihre Folgen. Beitr. Exp. Path. Physiol. Berlin, Traube 2, 1.
- Simpson, K.(1940) Shelter deaths from pulmonary embolism. Lancet Dec. 14 Page 744.
- Homans J. (1954) Thrombosis of the deep leg veins due to prolonged sitting. New Eng. J. of Medicine Page 148.
- Foley T.J. (1974) Nixon could have died of clot. Los Angeles Times July 5 Sec.1, Page 1.
- Los Angeles Times. (1974) September 21, Sec.1, Page 1.
- Los Angeles Times. (1974) September 26, Sec.1, Page 3.
- Belcher J., Ruhlow J. (1974) Nixon in hospital, may face surgery. October 24, Sec. 1, Page 1.
- Alexander G., Belcher J. (1974) Nixon treated for shock, is critical. October 30, Sec., Page 1.
- Cruickshank J.M., Gorlin R., Jennett B. (1988) Air travel and thrombotic episodes: The economy class syndrome. Lancet 2, Page 497.
- Wong H.P.W., Eklof B., Kistner R.L., Masuda E.M. & Sonntag B.(1994)Air travel-related “Coach class thrombosis”- A Hawaiian syndrome? SPHF Proceedings Vol. 58, Page 35.
- Paganin F., Bourde A., Yvin J.L., et al. (2003) Venous thromboembolism in passengers following a 12-hour flight: a case-control study Aviation Space & Environ Med.Vol.74 Page 1277.
- Lapostolle F., Surget V., Borron S.W., et al. (2001) Severe pulmonary embolism associated with air travel. New Eng. J. of Med. Vol.345,No 11, Page 779.
- Peres-Rodriguez E., Jimenez D., Diaz G., et al. (2003) Incidence of air travel-related pulmonary embolism at the Madrid-Barajas airport. Arch. Int. Med. Vol. 163, Page 2766.
- Scurr J.H., Machin S.J., Nailey-King,S. et al. (2001) Frequency and prevention of symptomless deep-vein thrombosis in long-haul flights: a randomized trial. Lancet Vol. 357, Page 1485.
- Belcaro G., Geroulakis G., Nicolaides A.N., et al. (2001) Venous thromboembolism from air travel. The LONFLIT Study. Angiology Vol.52 Page 369.
- Belcaro G., Geroulakis G., Nicolaides A.N., et al. (2001) Maladie veineuse thrombo-embolique apres vol aerien. Les etudes LONFLIT. Angeiologie 53, Page 17.
- Belcaro G., Geroulakis,G., Nicolaides A.N., et al. (2002) Venous thromboembolism from air travel: The LONFLIT3 study. Prevention with aspirin vs low-molecular-weight heparin (LMWH) in high-risk subjects: A randomized trial. Angiology Vol.53 Page 1.
- Cesarone M.R., Belcaro G., Nicolaides A.N., et al. (2003) The LONFLIT4-Concorde deep venous thrombosis and edema study: Prevention with travel stockings Angiology Vol 54 Page 143.
- Becker N.G., Salim A., Kelman C.W. (2006) Air travel and the risk of deep vein thrombosis. ANZ J. of Public Health Vol.30 Page5.
- Mortazavi A., Eisenberg M.J., Langleben D. et al. (2003) Altitude-related hypoxia: risk assessment and management for passengers on commercial aircraft. Aviat. Space Environ Med.Vol.74 Page 922.
- Schobersberger W., Fries D., Mittermayr M., et al. (2002)Changes of biochemical markers and functional tests for clot formation during long-haul flights. Thromb. Res.Vol.108, Page 19.
- Schreijer A.J.M., Cannegieter S.C., Meijers,J.C.M., et al. (2006) Activation of coagulation system during air travel: a crossover study. Lancet Vol. 367 Page 832.
- Staunton M., Drexler C., Dulitz M.G., et al. (1999) Effects of hypoxia-reoxygenation on microvascular endothelial function in the rat hippocampal slice. Anesthesiology Vol.91 Page 1462.
- Collard C.D., Vakeva A., Bukusoglu C., et al. (1997) Reoxygenation of hypoxic human umbilical vein endothelial cells activates the classic complement pathway. Circulation Vol.96 Page 326
- Gallus A.S. (2005) Travel, venous thromboembolism and thrombophilia. Seminars in Thrombosis and Hemostasis Vol.31 Page 90.
Copyright 2007. Greek/Australian International Legal and Medical Conference.
For more information contact Jenny Crofts at email@example.com