European Journal of Vascular & Endovascular Surgery
Volume 31, Issue 2 , Pages 187-199, February 2006

Venous Thromboembolic Complications Following Air Travel: What's the Quantitative Risk? A Literature Review

  • K.R. Aryal
    • ,
  • H. Al-khaffaf

      Affiliations

    • Corresponding Author InformationCorresponding author. H. Al-khaffaf, Department of Vascular Surgery, 1st Floor Pennine House, Burnley General Hospital, Burnley, Lancashire BB10 2PQ, UK.

Department of Vascular Surgery, Burnley General Hospital, Burnley, Lancashire BB10 2PQ, UK

Accepted 18 August 2005. published online 17 October 2005.

Article Outline

Abstract 

Objectives

To quantify the risk of venous thromboembolism (VTE) following air travel and assess methods of prevention.

Design

Review of literature.

Methods

We reviewed Pubmed, Medline, Embase and Cochrane Databases for studies that have assessed the risk of VTE associated with air travel.

Results

There is an association of VTE with air travel with pooled odds ratio of 1.59 (confidence interval 1.04–2.43) from three case control studies and relative risk of 2.93 (confidence interval 1.5–5.58) from two controlled cohort studies. The incidence of symptomatic pulmonary embolism (PE) is extremely low but there is substantial increase when the distance travelled is more than 5000mile (1.5 PE per million passengers) or time of flight is more than 8h duration (2.57 PE per million passengers). The quantitative risk of lower limb venous thrombosis in high-risk subjects is 5% per flight and 1.6% per flight for low risk subjects following long haul flights. All six randomised trials to test the below knee compression stockings with ankle pressures of 14–30mmHg have shown reduction in lower limb venous thrombosis.

Conclusion

VTE is more common in those with additional risk factors when the risk is about 5% per air travel for long haul flights. Class I or II below knee compression stockings are effective in the prevention of lower limb venous thrombosis.

Keywords: Deep vein thrombosis, Pulmonary embolism, Venous thromboembolism, Air travel, Long haul flight, Lower limb venous thrombosis

 

Back to Article Outline

1. Introduction 

Over the last few years there has been a lot of interest by the media, air-line industry and public health institutions in the problem of venous thromboembolism (VTE) following long haul flights. In 2001 the Guardian reported the death of a 27-year-old British woman as a result of pulmonary embolism after long flight from Australia to London.1 The World Health Organisation (WHO) in a press release in Geneva after consultation with international experts and airline industry in March 2001 stated that there is probably some association between air-travel and venous thrombosis but the evidence is weak and further research is needed to quantify the risk.2, 3

The relationship between prolonged sitting and pulmonary embolism (PE) was reported as early as 1940. There were many cases of PE in Britons forced to remain seated in cramped conditions for hours during the Second World War.4 Homans first reported a case of deep vein thrombosis (DVT) after a 14-h flight in a physician in 1954.5 Since then various studies, first in the form of case reports and case control studies and more recently in the form of randomized controlled trials (RCTS) have been published to support the association between air-travel and VTE.

The aim of this review is to quantify the risk of VTE following air-travel. We also wanted to assess the methods of prevention of VTE following long haul flights.

Back to Article Outline

2. Methods 

Literature search was carried out using key words and medical subject heading (MeSH) terms ‘deep vein thrombosis’, ‘venous thromboembolism’, ‘pulmonary embolism’, ‘air travel’, ‘economy class syndrome’ and ‘long haul flight’ on Pubmed, Medline, Embase and Cochrane databases limited to human subjects (in November 2004). Full text of all relevant papers in English language was retrieved. Further papers identified from the reference list of major articles were cross-referenced by hand.

2.1. Inclusion criteria 

Thrombosis of deep, superficial or muscle veins of the lower limbs and PE have been used to mean VTE. Thrombosis of the cerebral and subclavian veins has not been considered in this review although there may be some association between these and air travel. Different types of thromboembolism have been specified as far as possible.

Flights with duration of more than 6–7h have been considered as long haul flights. These times are spent inside the aeroplane and do not include time spent in transit or stop-over. For final analysis only case control studies, controlled cohort, randomized trials and incidence studies have been included. Case reports and case series have been excluded. Level of evidence obtained from the studies has been stated according to recommendations for grading evidence by Scottish Intercollegiate Guidelines Network Grading Review Group.6

2.2. Statistics 

Pooled estimates of odds ratios from case control studies and pooled estimates of relative risks from controlled cohort studies were expressed as fixed-effect model according to Mantel and Haenszel (due to absence of heterogeneity in the studies), using Statsdirect Package software.

Back to Article Outline

3. Results 

Forty-three studies relevant to this review were found 18 of which met the inclusion criteria. Of these, five were case control studies, two controlled cohort and eight randomized controlled trial studies (three of them with two trials in each study) and three incidence studies for symptomatic pulmonary embolism. The remaining studies although quoted here in the text have not been included for final analysis because they did not have much strength in terms of the level of evidence.

3.1. Incidence of deep vein thrombosis in the general population 

The reported incidence of deep vein thrombosis (DVT) in the general population is around 1–2 person per 1000 population per year. One to two percent of those who have DVT may develop PE many of which may be fatal.7, 8, 9

3.2. Difficulty in recruiting subjects for venous thromboembolism studies in long haul flights 

Meaningful comparison between the general population and those who are taking long haul flight is difficult. People taking long haul flights are healthier and have lesser risk than the general population and, therefore, are not a true sample for comparison. In addition to that, many high-risk patients may not take air travel due to health reasons. The total number of recruits necessary for meaningful comparison is large. A correspondence letter in the Lancet suggested that 2500 airline passengers would have to be studied to confirm a significant increase in the DVT compared with controls.10 This sample size was based on the power calculation with the 1 per 1000 incidence of DVT in the general population and extrapolating it on the 24-h flight.10 Furthermore, performance of necessary investigations for assessment of VTE needs considerable resource and time. D-Dimer assay, though highly sensitive, has a false positive rate for other conditions undermining its specificity.11, 12 Although, venography has been considered as gold standard by many authors, it is invasive and may not be acceptable to many recruits.13 Therefore, all studies have used duplex scan for the diagnosis of lower limb venous thrombosis. Although it is well known that duplex scan has a sensitivity and specificity of 97% for proximal DVT, the sensitivity decreases for distal DVT when compared with venography.12, 14 Another problem faced in such studies is the number of drop outs, even after the subjects have agreed to take part in the study, due to flight connection problems and, therefore, analysis carried out on the remaining subjects may not be a true indicator of the overall subjects taking part in the study. The timing of assessment is an equally difficult issue and the formation of clot may be many days after the passengers leave the flight. Many airlines may claim that thrombosis detected after the flight may have been present before the passenger boarded the flight. Ideally, all studies should have duplex scan before and after the flight. Despite these limitations there have been numerous studies related to air travel and VTE.

3.3. Pathogenesis of thromboembolism in long haul flights 

Virchow's classic triad15 of venous stasis, vessel wall damage and hypercoagulable state of blood remains true for the formation of thrombus. Prolonged sitting causes venous stasis and is the triggering factor. Significant fluid retention (>1l) in the lower leg was noted in long haul air travellers which may promote venous compression16 and another study model showed significant increase in the ankle, calf and leg circumference after 6h simulated flight.17 The pressure by the edge of the seat to the back of the leg, exaggerated by crossing of the legs in the narrow cramped condition of the economy class seats may damage the endothelium of the leg veins. There have been reports of an increase in platelet count and platelet packing after 6h of simulated flight creating hypercoagulable state.17 Additional factors such as dehydration due to alcohol intake, low fluid intake and dry atmosphere of the aeroplane cabin increases the risk of haemoconcentration.18 Decreased oxygen tension and low ambient pressure in the aeroplane have been shown to impair fibrinolysis and to induce activation of coagulation.19, 20

3.4. Venous thromboembolism following long haul flights: stratification of risk 

Various factors influence the formation of VTE. Air passengers with additional risk factors are likely to be at higher risk. Passengers with out any known clinical disease and not on any medical treatment are low risk subjects. High risk individuals include those with previous history of DVT, hypercoagulable states such as factor V Leiden mutation, obesity, limitation of mobility due to bone or joint problems, neoplastic disease with in the previous 2 years, hormone replacement therapy, diabetes, hypertension, cardiac failure, other cardiovascular disease, large varicose veins and recent trauma.21, 22

3.5. Case reports and case series 

Until mid to late 90s, most published literature was in the form of case reports or case series. Symington and Stack coined the term ‘Economy class syndrome’ after looking at 182 cohorts of pulmonary embolism patients. They found out that out of eight passengers in various modes of prolonged travel three had travelled between Britain and India or Portugal in the economy class of the aeroplane.23 This was further strengthened by Cruickshank et al. (one of the subjects being himself) who reported six cases of pulmonary embolism after an air travel.24 In a post mortem examination of 61 aeroplane sudden deaths referred to the coroner arriving at Heathrow airport, Sarvesvaren found that 11 (18%) were due to pulmonary embolism and all had flights of more than 6h duration.25 Mercer and Brown reviewed the records of 134 pulmonary embolism cases and checked for a history of air travel in 66 of them. Thirty-three (50%) of those 66 had history of air travel for 4h or more and inferred that air travel was the most common risk factor for VTE.26 In Honolulu, Hawaii out of 254 patients admitted with DVT and or PE, 44 (17.3%) had developed their symptoms during or after air travel.27

3.6. Case control studies 

Five case control studies have been published. In these studies, comparisons were made between patients having VTE with those not having VTE and history of long distance travel (in two studies) or air travel (in three studies) was recorded (Table 1).

Table 1. Venous thrombo-embolism and air travel, case control studies
AuthorYearType of studyLevel of evidencePatient group (No.)Control group (No.)Travel historyNumber of patient vs control for history of travelOdds ratio (95% confidence interval)Pooled estimate of odds ratio (95% confidence interval)
Ferrari E. et al.281999Prospective case controlII−With VTED* (160)No VTED* (160)>4h journey39 vs 123.98 (1.9–8.4)2.6 (1.79–3.79)
Samama M.M. et al.292000Prospective case controlII−With DVT† (494)With out DVT† (494)Long distance travel62 vs 312.35 (1.45–3.80)
Kraaijenhagen R.A. et al.302000Prospective case controlII−With DVT† (186)No DVT† (602)Air travel history4 vs 131 (0.3–1.4)1.59 (1.04–2.43)
Arya R. et al.312002Prospective case controlII−With DVT (185)Without DVT (383)Air travel >8h14 vs 221.3 (0.6–2.8)
Martinelli I. et al.322003Prospective case controlII−With VTED (210)No VTED (210)Air travel history31 vs 162.1 (1.1–4.0)

*VTED, venous thromboembolic disease.

†DVT, deep venous thrombosis.

Ferrari et al. compared the venous thromboembolic patients in their cardiology department with other patients with out venous thromboembolism (160 patients in each group). They found that recent journey of more than 4h duration (whatever means of transport) was found in 39 cases of thromboembolic group and 12 cases of control group with an odds ratio of 3.98. However, in this study only nine of 39 patients had travelled by air, with a mean duration of 5±0.7h. Of the remaining patients, 28 travelled by car and two by train.28 Samama et al. compared 494 DVT case patients and 494 control patients with influenzal or rhinopharyngeal syndrome. All these were medical patients (those who did not have surgery or plaster of Paris application). There was a history of long distance travel in 62 (12.5%) of the DVT group and 31 (6.3%) of the control population with and odds ratio of 2.35.29 From these two studies there appears to be strong association between travel and VTE (pooled estimate of odds ratio 2.6, 95% CI 1.79–3.79) (Table 1). Although in these studies all subjects did not travel by aeroplane, it emphasizes the point that prolonged sitting in cramped conditions for long periods increases the risk of VTE.

Kraaijenhagen et al.30 took a cohort of 788 suspected cases of DVT and compared 186 of confirmed cases of DVT with 602 controls without DVT and found out a history of air travel in four of DVT group and 13 of non-DVT group with and odds ratio of 1. There was dissimilar number of patients in the two groups and most patients travelled for less than 5h duration.30 Arya et al. from a cohort of 568 cases of suspected DVT compared 185 confirmed cases of DVT with 383 cases without DVT in terms of air travel. There was no significant link between air travel and DVT in this cohort with an odds ratio of 1.3. However, when one additional risk factor was present the risk of DVT was increased in air travellers (odds ration 3.0 confidence interval 1.1–8.2).31 A more recent study by Martinelli et al.32 showed that history of air travel was more common in patients with VTE compared with controls with odds ratio of 2.1. From the pooled estimate of odds ratio for these three studies there appears to be stronger association between VTE and air travel (odds ratio 1.59) (Table 1).

3.7. Controlled cohort studies 

Two controlled cohort studies have been published in which air passengers were compared with control group of non-passengers for presence of VTE. Schwarz et al. in a prospective controlled pilot study examined 160 air passengers with 160 age and sex matched non-travelling volunteers and assessed lower limb venous thrombosis on compression ultrasound, which was present in 2.5% of air passengers and 0.6% of non-travelling volunteers.33 In another larger study by the same group comparing 964 passengers with 1213 non-passengers, there were 2.8% thrombotic complications (DVT and muscle vein thrombosis) in the passenger group compared with 1% in the control group.34 This 1% in the control group is high and it may be because large number (35.7%) of them had varicose veins and many had additional risk factors including elevated factor VIII level in 30%. Pooled estimate of the relative risk from these two studies gives relative risk of 2.93 with strong association between air travel and VTE (Table 2).

Table 2. Lower limb venous thrombosis and air travel: results from the prospective controlled cohort studies
AuthorYearType of studyLevel of evidencePatient group (No.)Control group (No.)Measured outcomeNumber of patient vs control for measured outcomeRelative risk (95% confidence interval)Pooled estimate of relative risk (95% confidence interval)
Schwarz T. et al.332002Controlled cohort pilot studyII+Long haul flight passengers >8h (160)Non-travelling volunteers (160)Isolated calf muscle vein thrombosis4 vs 14 (0.6–26.4)2.93 (1.5–5.58)
Schwarz T. et al.342003Controlled cohortII+Long haul flight passenger >8h (964)Non-passenger (1213)Lower limb venous thrombotic events27 vs 122.83 (1.46–5.49)

3.8. Incidence of symptomatic pulmonary embolism in air travellers 

Three studies looking at symptomatic cases of PE in the referral hospitals from three busy airports of Paris, Madrid-Barajas and Sidney have been published. One looked at the distance traveled and the other two at duration of the flight (Table 3). There was substantial increase in PE incidence when the distance travelled is more than 5000km or of more than 8h duration. In Paris, the overall incidence was only 0.39 person per million passengers but once the distance travelled was more than 5000km the incidence was 1.5 per million passengers. This increased further for distances above 10,000km (4.77 pulmonary embolism per million passengers).35 Similarly in Madrid-Barajas the incidence was 1.65 per million for duration of more than 8h.36 In Sydney airport where all the analysed flights were of more than 9h duration the incidence of PE embolism was 1.65 per million passengers.37

Table 3. Incidence of pulmonary embolism from three busy airports of Paris, Madrid and Sydney after long haul flight
AuthorYear of publicationTotal no. of passengersTotal cases of pulmonary embolismOverall incidence of PERisk according to distance travelled or flight durationRemarks
Lower incidenceHigher incidence
Lapostolle F. et al.352001135.29 million560.41/million passengers0.01/million for those <5000km1.5/million if distance >5000km
Perez-Rodriguez E. et al.36200341 million160.39/million passengers0.25/million for flights 6–8h1.65/million for flights more than 8hAll flights more than 6h
Hertzberg S.R. et al.3720036.58 million172.57/million passengers 2.57/million for flights more than 9hAll flights more than 9h

The incidence of pulmonary embolism appears to be low based on the above figures. However, all these are cases of symptomatic pulmonary embolism and this may only indicate the tip of the iceberg. All asymptomatic cases, those who had DVT with out pulmonary embolism, those who had symptoms after some days following the flight or those who did not attend because of the connection problems have been missed.

3.9. Incidence of lower limb venous thrombosis from recent randomised or controlled studies 

LONg FLighT Thrombosis (LONFLIT) study was designed to evaluate the incidence of deep vein thrombosis after long haul flights and preventative measures. It published its first results in 2001 and showed that in high-risk subjects the incidence may be as high as 5% but in low risk subjects the incidence was zero.22

3.10. Passengers with additional risk factors 

All subjects with additional risk factors have higher risk of thrombosis. McQuillan et al.38 showed that the most common risk factors were oestrogen, previous history of deep vein thrombosis and factor V Leiden. At least one additional risk factor was present in more than 80% of the patients who developed VTE with in 30 days of travel. Similarly, in another study by Paganin et al.39 of 46 cases of VTE, 38 (82%) had additional risk factors. History of previous VTE, recent trauma, presence of varicose veins, obesity, immobility during flight and cardiac disease was significantly higher in the thrombotic group compared with controls. In another study by Hosoi et al. of 101 patients with DVT 15 had history of travel of whom 13 (87%) had additional risk factors.40

In all five studies in high risk subjects in LONFLIT series the incidence of lower limb venous thrombosis for those not receiving any prophylaxis ranged from 4.5 to 7.6% (overall 5%) per long haul flight22, 41, 42, 43 (Table 4). Although there was no risk stratification in Scurr's study44 the 10% incidence of DVT per long haul flight in the control group was high and it could be because most subjects were more than 50 years of age and median duration of travel being 24h (? High-risk).

Table 4. Quantitative risk for lower limb venous thrombosis in air travellers with out any prophylaxis
No.AuthorYearTotal no. of subjectsTotal no. of subjects with thrombosisPercentage of venous thrombosisLevel of evidenceFlight durationAssessment tool for thrombosis
1Scur J.H. et al.442001116 control1210I+> 8hDuplex ultrasound
2Belcaro G. et al.22 (LONFLIT 1)2001389 high risk13 DVT*4.9II+12.4hDuplex ultrasound
6 SVT†
355 low riskNone0
3Belcaro G. et al.22 (LONFLIT 2)2001422 control, high risk19 DVT*4.5I+12.4hDuplex ultrasound
4Caserone M.R. et al.41 (LONFLIT 3)200282 high risk control44.82I+?Duplex ultrasound
5Belcaro G. et al.45 (LONFLIT 4 Concorde oedema SSL) I+
Part 12002179 low, med risk control4 DVT*2.2 7–8hDuplex ultrasound
2 SVT†1.1
Part 22002143 low, med risk, control3 DVT*2.1 11–12hDuplex ultrasound
3 SVT†2.1
6Caserone M.R.46 (LONFLIT 4 ECOTRAS) I+
Part 1200398 low medium risk controlNone0 7–8hDuplex ultrasound
Part 2200371 low risk controlNone0 11–12hDuplex Ultrasound
7Caserone M.R.47 (LONFLIT 4 Concorde DVT and oedema study) I+
Part 1200374 low medium risk controlNone0 7–8hDuplex ultrasound
Part 2200366 low medium risk control2 DVT3 11–12hDuplex ultrasound
2 SVT3
8Cesarone M.R.42 (LONFLT-FLITE)200392 high risk control5 DVT5.4I+7–8h
2 SVT2.2
9Belcaro G.43 (LONFLIT 5 JAP)2003102 high risk6 DVT5.8I+11–12h
Total 1087 high risk555
986 low risk161.6
116 risk not known1210

*DVT, deep vein thrombosis.

†SVT, superficial vein thrombosis.

3.11. Passengers with no (low) risk factor 

Out of six studies involving the low medium risk groups in LONFLIT studies three did not have any lower limb venous thrombosis. Three studies showed it to be 3.3, 4.2 and 6%, respectively45, 46, 47 (Table 4). Pooled data from these six studies shows the risk to be 1.6% per long haul flight.

In the New Zealand Air Traveller's Thrombosis (NZATT) study 878 passengers with flight duration of at least 10h with low to moderate risk were included. All had D-Dimer estimation. Only 112 cases with raised D-Dimer underwent compression ultrasonography. There were only nine cases of venous thromboembolism (four pulmonary embolism and remaining were deep vein thrombosis).48 But 17% of the subjects were using compression stockings and 31% were using aspirin. The use of D-Dimer as a screening tool in this study may have missed many patients with normal D-Dimers. This assay though highly sensitive in suspected cases of VTE does not appear to have reliable predictability for flight related DVT. None of 12 ultrasound detected DVTs in Scurr's paper had raised D-Dimer.44 Of 27 DVT or isolated calf muscle vein thrombosis, only 11 had raised D-Dimers in Schwarz's paper.34 There was no significant difference in D-Dimer level between ultrasound detected DVT and those without DVT in two of the LONFLIT studies.42, 43 The remaining LONFLIT studies do not mention any thing about D-Dimer assay.22, 41, 45, 46, 47

3.12. LONg FLIghT (LONFLIT) studies 

LONg FlighT Thrombosis (LONFLIT) study was originally set up to detect the incidence of lower limb venous thrombosis following long haul flights. LONFLIT 1 looked at the incidence in the low risk and high-risk groups.22 LONFLIT 2 study evaluated the role of below knee graduated elastic compression stockings in high-risk subjects.22 The LONFLIT 3 study41 evaluated aspirin and low molecular weight heparin (LMWH) in high-risk subjects for reduction of DVT.

The LONFLIT 4 study45, 46, 47 has three different publications relating to the thrombosis and air travel. All were in low and medium risk groups. In the first study45 Flight socks, Scholl, UK below knee compression stocking with ankle pressure of 14–17mmHg were used. There was a reduction in DVT and oedema in the stocking group. In the second paper45 Sigvaris, Traveno, Ganzoni, below knee graduated compression stocking with ankle pressure of 12–18mmHg were used. In both part 1 and part 2 there was no DVT in either the control or the stocking group but there was a reduction in oedema in the stocking group. The third of the LONFLIT 4 papers47 looked at Kendall travel socks (Tyco Healthcare, Mansfield, USA) with ankle pressure of 20–30mmHg and showed reduction both in DVT and oedema in the stocking group.

LONFLIT-FLITE study42 looked at the effect of profibrinolytic FLITE tablets (pinokinase) on high-risk subjects and in LONFLIT 5 JAP study,43 Flight sock, Scholl, UK was used with ankle pressure of 14–17mmHg. Both studies showed reduction in DVT.

3.13. Prevention of thrombo-embolic complications in long haul flights 

Some of the simple measures should be advised to passengers. These include drinking plenty of fluids (one glass of water at least every 2h) but not much alcohol to prevent dehydration. Tight clothing should be avoided. They should be advised not to keep their bag in front of seats to limit stretching the legs and should stretch their legs at least every hour for 2min. They should be encouraged to be mobile (3min every hour).22

Of the 11 randomised trails published so far to see the efficacy of various measures nine were graduated below knee compression stockings with ankle pressures ranging from 14 to 30mmHg.22, 43, 44, 45, 46, 47 Among the remaining two, one tested LMWH and aspirin41 and the other tested profibrinolytic agent pinokinase called FLITE41 (Table 3).

Among the studies that have tested stockings there was a reduction in lower limb venous thrombosis in six trials to 0.24 and 0.98% in two instances22, 43 and 0% in the remaining four trials.44, 45, 47 Three trials had 0% lower limb venous thrombosis in the control group and there was no advantage in this respect45, 46 but there was reduction in leg oedema in these subjects.

LMWH single dose injection before travel reduced the incidence from 4.2 to 0%.41 However, the incidence of DVT in subjects taking aspirin in this trail was 3.6%. The dose of aspirin used rather high (400mg once daily for 3 days) and 13% in this group experienced gastrointestinal side effects.41 Another trial looking at the effect of profibrinolytic agent FLITE showed reduction of lower limb venous thrombosis from 7.6 to 0%.43

Based on these studies, below knee graduated compression stockings class I or II seem to reduce the incidence of DVT and oedema in long haul flights. Low molecular weight heparin is probably useful and so is antifibrinolytic agent but more research is needed. Aspirin has not been useful and it caused gastro intestinal symptoms in stead (Table 5).

Table 5. Prevention of thrombo-embolic complications, results from randomised trials
No.AuthorType of studyLevel of evidenceProphylaxis measureAssessment tool for thrombosisNumber of subjectsNo. of thrombosis subjects (%)P valueRemarksDrop outs
ControlIntervenedControlIntervened
1Scur J.H. et al.44Randomised trialI+Class I below knee compression stocking pr 20–30mmHgDuplex scan116115 12 (10%)0 (0%) All asymptomatic
2Belcaro G. et al.22 (LONFLIT 2)Randomised trialI+Below knee stocking ankle pr 25mmHgDuplex scan422411 19 (4.5%)1 (0.24%) <0.02All high risks? No mention
3Cesarone M.R. et al.41 (LONFLIT 3)Randomised trialI−Aspirin, LMWH*Duplex scan8384824 (4.82%)3 (3.6%)0<0.02All high risk17, 16, 18 respectively
4Belcaro G. et al.45 (LONFLIT 4 Concorde oedema SSL) I+
Part 1Randomised trial Below knee flight socks 14–17mmHgDuplex scan179179 6 (3.35%)0 <0.05Low, med risk flight duration 7–8h5 and 9
Part 2Randomised trial Below knee flight socks 14–17mmHgDuplex scan136135 6 (4.2%)0 <0.005Low, med risk, flight duration 11–12h6 and 8
5Cesarone M.R. et al.46 (LONFLIT 4) EcoTraS
Part 1Randomised trialI+Below knee Traveno stockings 12–18mmHgDuplex scan9897 00 Low, med risk, 7–8h10, 6
Part 2Randomised trialI+As in part 1Duplex scan7175 00 low, med risk, 11–12h11, 8
6Cesarone M.R. et al.47 (LONFLIT 4 Concorde DVT† and oedema) I+
Part 1Randomised trial Travel socks 20–30mmHgDuplex scan7272 00 Low and med risk, flight 7–8h2 and 4
Part 2Randomised trial As in part 1Duplex scan6664 40 <0.05Low, med risk flight 11–12h0, 2
7Cesarone M.R.42 (LONFLT-FLITE)Randomised trialI+FLITE tab profibrinolytic agentDuplex scan9294 7 (7.6%)0 <0.05High risk11, 7
8Belcaro G.43 (LONFLIT 5 JAP)Randomised trialI+Scholl (UK) flight socks 14–17mmHgDuplex scan102103 6 (5.8%)1 (0.98%) <0.0025High risk12, 7

*LMWH, low molecular weight heparin.

†DVT, deep vein thrombosis.

3.14. Economy class vs business class traveller and thrombo-embolic complications 

BEST (Business vs Economy class Syndrome as a cause of venous Thrombosis) study49 from South Africa tried to address this issue with 180 passengers in the business class and 719 in the economy class. Elevated D-Dimers were present in 12% business class and 7% economy class passengers, which was measured in all subjects. However, duplex ultrasound was optional and was only performed in 86-business class and 348 economy class subjects none of who had venous thrombosis. The limitation of this study was that there were dissimilar number of passengers in the two groups because they could not continue to recruit their business class passengers due to cost constraints to reach similar number to economy class passengers and duplex scan was not done in all.

3.15. Symptomatic vs asymtomatic deep vein thrombosis 

From the published studies to date the majority of lower limb venous thrombosis were asymptomatic (Table 6). But in all these studies the passengers had duplex scan to rule out thrombi before they boarded the flight. Repeat scan was performed to evaluate the presence of thrombi on completion of the flight and further scans were not carried out. The papers describing symptomatic VTE following air-travel have not had duplex evaluation before the travel and it raises the question if they had preexisting thrombi before they boarded the flight.

Table 6. Symptomatic vs asymptomatic lower limb venous thrombosis
AuthorTotal no. of thrombosisAsymtomatic (No.)Symptomatic (No.)
Schwarz T.334 ICMVT*40
Schwarz T.3420 ICMVT*191
7 DVT†52
Scurr J.H.4412 DVT†120
Cesarone M.R.417 (control and aspirin group combined)43
Belcaro G.45 (LONFLIT 4 Concorde oedema SSL)
Part 1660
Part 2651
Cesarone M.R.47 (LONFLIT Concorde DVT and oedema study)
Part 22 DVT†20
Cesarone M.R.42 (LONFLIT-FLITE)770
Belcaro G.43 (LONFLIT 5 JAP)6 DVT†60

*ICMVT, isolated calf muscle venous thrombosis.

†DVT, deep venous thrombosis.

3.16. Site of lower limb venous thrombosis 

Table 7 illustrates the site of lower limb venous thrombosis from the available papers. Most of the DVT were in the calf segment. There were a few muscle vein thrombosis. The actual importance of muscle vein thrombosis is still unclear although some authors have claimed that it is a marker of DVT.34

Table 7. Site of lower limb venous thrombosis
AuthorTotal of subjectsTotal no. of lower limb venous thrombosisSite of deep vein thrombosisRemarks
Scur J.H. et al.4411612 DVTAll calf
Belcaro G. et al.22 (LONFLIT 1)389 high risk13 DVT5 proximal (femoral)
6 SVT8 distal (popliteal)
Belcaro G. et al.22 (LONFLIT 2)422 high risk19 DVTSite not mentionedFemoral and popliteal vein scanned
Caserone M.R. et al.41 (LONFLIT 3)82 high risk4 DVTSite not mentionedFemoral and popliteal vein scanned
Belcaro G. et al.45
Part 1179 low, med risk4 DVT2 popliteal
2 SVT2 proximal tibial
Part 2143 low, med risk3 DVT3 popliteal
Cesarone M.R. et al.47
Part 266 low medium risk2 DVT2 popliteal
2 SVT
Cesarone M.R.42 (LONFLT-FLITE)92 high risk5 DVTSite not mentionedScanned veins fem, pop or tibial
2 SVT
Belcaro G.43 (LONFLIT 5 JAP)102 high risk6 DVT3 superficial femoral
2 popliteal
1 soleal
Schwarz T. et al.331604 ICMVTAll soleal muscle vein
Schwarz T. et al.3496420 ICMVT17 soleus muscle vein
7 DVT3 gastrocnemius muscle vein
5 peroneal or tibial posterior veins
2 popliteal vein

DVT, deep vein thrombosis; SVT, superficial vein thrombosis; ICMVT, isolated calf muscle vein thrombosis.

3.17. Aisle seat vs non-aisle seat traveller and occurrence of venous thrombosis 

Some of the LONFLIT studies have reported higher incidence of DVT in central or window seat passengers than aisle seat passengers (Table 8) implying that aisle seats had protective effect in relation to the development of venous thrombosis.22, 41, 45

Table 8. Incidence of lower limb venous thrombosis in non-aisle seat passengers from LONFLIT studies
StudyAuthorYearTotal no. of thrombosisTotal no. of thrombosis in non-aisle seat
LONFLIT 1Belcaro G. et al.2220011918
LONFLIT 2Belcaro G. et al.22200119 DVT*All 19
LONFLIT 3Cesarone M.R. et al.4120027 DVT*6
LONFLIT 4 Concorde oedema SSL
Part 1Belcaro G. et al.45200266
Part 2Belcaro G. et al.45200264
LONFLIT 4 Concorde DVT oedema
Part 2Cesarone M.R.4720034Not mentioned
LONFLI-FLITECesarone M.R.4220037Not mentioned
LONFLIT 5 JAPBelcaro G.4320037Not mentioned

*DVT, deep vein thrombosis.

Is the female sex more prone to thrombo-embolic complications after long haul flights?

There was no difference in the development of DVT in male and female sexes in LONFLIT 3, LONFLIT 4 Concorde oedema-SSL study both part 1 and 2. Among females, three of seven females in LONFLIT 3 and all six in part 1 and 2 of LONFLIT 4 Concorde oedema-SSL study were taking oral contraceptive pills.40, 44 Another study has shown that the use of oral contraceptive with air travel increases the chance of thrombosis by 14 fold.32

3.18. Timing of deep vein thrombosis following air travel 

The risk of VTE has been shown to be highest with in 2 weeks of a long haul flight.50 In this study, Kelman et al. looked at 153 Australian citizens admitted to Hospitals with VTE with in 100 days of arrival on an international flight. Forty-six events occurred with in 14 days of the flight. The observed number of cases with in 2 weeks of arrival significantly exceeded the number expected under the assumption of uniform distribution over the 0–100 days after arrival.

Back to Article Outline

4. Conclusion 

There appears to be an association between long haul flights and VTE. The incidence of lower limb venous thrombosis in low risk passengers is about 1.6% while in those with additional risk factors is about 5%. Below knee graduated compression stocking with ankle pressure ranging from 14 to 30mmHg (class 1 or 2) have been shown to be effective in reducing lower limb venous thrombosis (level of evidence 1+). They are also effective in reducing oedema of the legs in low risk subjects. Their role in the prevention of pulmonary embolism needs further research. Similarly, the role of low molecular weight heparin and profibrinolytic agents either singly or in combination with stockings also needs further research. The idea that aisle seat passengers are more protected than the window or middle seat passengers is rather interesting and may be a reflection of the fact that these patients are freer to move in the aisle than other passengers. The advice that passengers should drink plenty of fluid, be mobile in the aeroplane, stretch their legs and avoid tight constricting clothes is based on common sense and indirect conclusion rather than direct evidence and also needs further research.

Back to Article Outline

Acknowledgements 

The authors would like to thank to Mr M. Joshi, MSc, M. Math(open), Cstat, Research support statistician, Medical statistics Unit Fylde College, Lancaster University, Lancaster LA1 4YF for his help with statistical analysis.

Back to Article Outline

References 

  1. Perry K. Blood clot kills woman after flight. Guardian. 2001;
  2. Scientific experts and airline industry call for further research of venous thrombosis. Press Release WHO/11; 2001.
  3. Mendis S, Derek Y, Alwan A. Air travel and venous thromboembolism. Bull World Health Organ. 2002;80:403–406
  4. Simpson K. Shelter deaths from pulmonary embolism. Lancet. 1940;2:744
  5. Homans J. Thrombosis of the deep leg veins due to prolonged sitting. N Engl J Med. 1954;250:148–149
  6. Harbour R, Miller J. A new system for grading recommendation in evidence based guidelines. BMJ. 2001;323:334–336
  7. Nordström M, Lindblad B, Bergqvist D, Kjellstrom T. A prospective study of the incidence of deep vein thrombosis with defined urban population. J Intern Med. 1992;232:155–160
  8. Anderson FA, Wheeler HB, Goldberg RJ, Hosmer DW, Patwardhan NA, Jovanovic B, et al. A population based perspective of the hospital incidence and case fatality rates of deep vein thrombosis and pulmonary embolism. The Worcester DVT study. Arch Intern Med. 1991;151:933–938
  9. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet. 1999;353:1167–1173
  10. Burnand KG, McGuinness CL, Smith A. Deep-vein thrombosis in long-haul flights. Lancet. 2001;358:837
  11. Frost SD, Brotman DJ, Michota FA. Rational use of D-Dimer measurement to exclude acute venous thromboembolic disease. Mayo Clin Proc. 2003;78:1385–1391
  12. Tovey C, Wyatt S. Diagnosis, investigation and management of deep vein thrombosis. BMJ. 2003;326:1180–1184
  13. De Valois JC, van Schaik CC, Verzijlbergen F, van Ramshorst B, Eikelboom BC, Meuwissen OJA. Contrast venography: from gold standard to ‘golden back up’ in clinically suspected deep vein thrombosis. Eur J Radiol. 1999;11:131–137
  14. Kearon C, Julian JA, Newman TE, Ginsberg JS. Noninvassive diagnosis of deep venous thrombosis. Ann Intern Med. 1998;128:663–677
  15. Virchow R. Collected works on scientific medicine. Frankfurt: Meidinger; 1856;
  16. Landgraff H, Vanselow B, Schulte HD, Mulmann MV, Bergan L. Economy class syndrome: rheology, fluid balance and lower limb oedema during a simulated 12-h long distance flight. Aviat Space Environ Med. 1994;65:930–935
  17. Hollingsworth SJ, Dialysis M, Barker SGE. Long haul flight and deep vein thrombosis: a model to help investigate the benefit of aspirin and below-knee compression stockings. Eur J Vasc Endovasc Surg. 2001;22:456–562
  18. Carruthers M, Arguelles AE, Mosovich A. Man in transit: biochemical and physiological changes during intercontinental flights. Lancet. 1976;1:977–981
  19. Gerter JP, Perry L, L'Italien G, Chung-Welch N, Cambria RP, Orkin R, et al. Ambient oxygen tension modulates endothelial fibrinolysis. J Vasc Surg. 1993;18:939–946
  20. Bendz B, Rostrup M, Sevre K, Anderson TO, Sandset PM. Association between acute hypobaric hypoxia and activation of coagulation in human beings. Lancet. 2000;356:1657–1658
  21. Belcaro G, Stansby G, Nicolaides AN. The venous clinic. London: Imperial College Press; 1999;
  22. Belcaro G, Geroulakos G, Nicolaides AN, Myers KA, Winford M. Venous thromboembolism from air travel the LONFLIT study. Angiology. 2001;52:369–374
  23. Symington IS, Stack BH. Pulmonary thromboembolism after travel. Br J Dis Chest. 1977;71:138–140
  24. Cruickshank JM, Gorlin R, Jennett B. Air travel and thrombotic episodes: the economy class syndrome. Lancet. 1988;2:497–498
  25. Sarvesvaran R. Sudden natural deaths associated with commercial air travel. Med Sci Law. 1986;26:35–38
  26. Mercer A, Brown JD. Venous thromboembolism associated with air travel: a report of 33 patients. Aviat Space Environ Med. 1998;69:154–157
  27. Eklof B, Kistner RL, Masuda EM, Sonntag BV, Wong HP. Venous thromboembolism in association with prolonged air travel. Dermatol Surg. 1996;22:637–641
  28. Ferrari E, Chevallier T, Chapelier A, Baydouy M. Travel as risk factor for venous thromboembolic disease: a case control study. Chest. 1999;115:440–444
  29. Samama MM. An epidemiologic study of risk factors for deep vein thrombosis in medical outpatients: the Sirius study. Arch Intern Med. 2000;160:3415–3420
  30. Kraaijenhagen RA, Haverkamp D, Koopman MMW, Prandoni P, Piovella F, Buller HR. Travel and risk of venous thrombosis. Lancet. 2000;356:1492–1493
  31. Arya R, Barnes JA, Hossain U, Patel RK, Cohen AT. Long-haul flights and deep vein thrombosis: a significant risk only when additional factors are also present. Br J Haematol. 2002;116:653–654
  32. Martinelli I, Taioli E, Battaglioli T, Podda GM, Passamonti SM, Pedotti P, et al. Risk of venous thromboembolism after air travel: interaction with thrombophilia and oral contraceptives. Arch Intern Med. 2003;163:2771–2774
  33. Schwarz T, Langenberg K, Oettler W, Halbritter K, Beyer J, Siegert G, et al. Deep vein and isolated calf muscle vein thrombosis following long-haul flights: Pilot study. Blood Coagul Fibrinolysis. 2002;13:755–757
  34. Schwarz T, Siegert G, Oettler W, Halbritter K, Beyer J, Frommhold R, et al. Venous thrombosis after long haul flights. Arch Intern Med. 2003;163:2759–2764
  35. Lapostolle F, Surget V, Borron SW, Desmaizieres M, Sordelet D, Lapandry C, et al. Severe pulmonary embolism associated with air travel. N Engl J Med. 2001;345:779–783
  36. Pérez-Rodríguez E, Jiménez D, Diaz G, Perez-Walton I, Luque M, Guillen C, et al. Incidence of air travel-related pulmonary embolism at the Madrid-Barajas airport. Arch Intern Med. 2003;163:2766–2770
  37. Hertzberg SR, Roy S, Hollis G, Brieger D, Chan A, Walsh W. Acute symptomatic pulmonary embolism associated with long haul air travel to Sydney. Vasc Med. 2003;8:21–23
  38. McQuillan AD, Eikelboom JW, Baker RI. Venous thromboembolism in travellers: can we identify those at risk?. Blood Coagul Fibrinolysis. 2003;14:671–675
  39. Paganin F, Bourde A, Yvin JL, Genin R, Guijarro JL, Bourdin A, et al. Venous thromboembolism in passengers following a 12-h flight: a case–control study. Aviat Space Environ Med. 2003;74:1277–1280
  40. Hosoi Y, Geroulakos G, Belcaro G, Sutton S. Characteristics of deep vein thrombosis associated with prolonged travel. Eur J Vasc Endovasc Surg. 2002;24:235–238
  41. Cesarone MR, Belcaro G, Nicolaides AN, Incandela L, De S, Geroulakos G, et al. Venous thrombosis from air travel: the LONFLIT3 study. Prevention with aspirin vs low-molecular weight heparin (LMWH) in high risk subjects: a randomised trial. Angiology. 2002;53:1–6
  42. Cesarone MR, Belcaro G, Nicolaides AN, Ricci A, Geroulakos G, Ippolito E, et al. Prevention of venous thrombosis in long-haul flights with Flite Tabs: the LONFLIT-FLITE randomized, controlled trial. Angiology. 2003;54:531–539
  43. Belcaro G, Cesarone MR, Nicolaides AN, Ricci A, Geroulakos G, Shah SS, et al. Prevention of venous thrombosis with elastic stockings during long-haul flights: the LONFLIT 5 JAP study. Clin Appl Thromb Hemost. 2003;9:197–201
  44. Scurr JH, Machin SJ, Bailey-King S, Mackie IJ, McDonald S, Smith PD. Frequency and prevention of symptomless deep-vein thrombosis in long-haul flights: a randomised trial. Lancet. 2001;357:1485–1489
  45. Belcaro G, Cesarone MR, Shah SS, Nicolaides AN, Geroulakos G, Ippolito E, et al. Prevention of oedema, flight microangiopathy and venous thrombosis in long haul flights with elastic stockings. A randomised trial. The LONFLIT 4 Concorde oedema-SSL study. Angiology. 2002;53:635–645
  46. Cesarone MR, Belcaro G, Nicolaides AN, Geroulakos G, Lennox A, Myers KA, et al. The LONFLIT4-Concorde-Sigvaris Trveno stockings in long flights (EcoTraS) study: a randomised trial. Angiology. 2003;54:1–9
  47. Cesarone MR, Belcaro G, Errichi BM, Nicolaides AN, Geroulakos G, Ippolito E, et al. The LONFLIT4-Concorde deep venous thrombosis and oedema study: prevention with travel stockings. Angiology. 2003;54:143–154
  48. Hughes RJ, Hopkins RJ, Hill S, Weatherall M, Van de Water N, Nowitz M, et al. Frequency of venous thromboembolism in low to moderate risk long distance air travellers: the New Zealand Air Traveller's Thrombosis (NZATT) study. Lancet. 2003;362:2039–2044
  49. Jacobson BF, Munster M, Smith A, Burnand KG, Carter A, Abdool-Carrim AT, et al. The BEST study—a prospective study to compare business class versus economy class air travel as a cause of thrombosis. S Afr Med J. 2003;93:522–528
  50. Kelman CW, Kortt MA, Becker NG, Li Z, Mathews JD, Guest CS, et al. Deep vein thrombosis and air travel record linkage study. BMJ. 2003;327:1072–1075

PII: S1078-5884(05)00541-1

doi:10.1016/j.ejvs.2005.08.025

European Journal of Vascular & Endovascular Surgery
Volume 31, Issue 2 , Pages 187-199, February 2006

Article Tools