Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

DVTs and all that

Systematic review
Risk factors for DVTs
Travel as a risk factor
Flying and aspirin for DVT prevention
We get much information about the dangers of deep vein thrombosis (DVT) with air travel, about not drinking too much alcohol, keeping hydrated, taking a walk or doing exercise, and taking an aspirin before we fly. Some evidence helps. One systematic review looks at the incidence of DVT in the general population [1], while another [4] calculates the NNT of aspirin to prevent one DVT while flying.

Systematic review

Several databases were searched to 2001 for articles on the incidence of DVT and thromboembolism. Studies reported all diagnosed patients in a defined general population in a developed country. DVT diagnoses had to be confirmed by clinical tests (like a scan) or a satisfactory validation study of the accuracy of the diagnosis performed. The age range had to be specified, so incidence rates could be calculated per 100,000 person years. For the purposes of the review, DVT cases had to be new and not to be recurrent, combined with pulmonary embolism, and due to any cause.


Finally accepted were nine studies from a combined population of about 19 million persons published since 1976. Most studies were conducted in Sweden or the USA.

Incidence rates adjusted to include only new DVTs, due to all causes, and for all ages of the population are shown in Figure 1. Most studies clustered around an incidence of 50 per 100,000. DVT occurred rarely below 20 years, but increased with age (Figure 2), so that in over 70s the rate was 200 per 100,000. Incidence was about the same in men and women.

Figure 1: Individual studies of DVT incidence in the general population

Figure 2: Age and DVT incidence

The causes of DVT were attributed to cancer or previous hospital admission, for about a quarter to a third of cases for each cause. About 40% of cases of DVT had no known cause.

Risk factors for DVTs

A reasonably large case-control study in France [2] examined 636 patients presenting with DVT and paired them by age and sex with a control group of patients presenting with influenza or rhinopharyngeal syndrome. DVT had to be documented by at least one objective test. Risk factors were classified as intrinsic or permanent, and triggering or transient.

In 988 patients who had not undergone surgery or had a plaster cast on the lower extremity during the preceding three weeks, a number of intrinsic and triggering factors were associated with a higher risk of DVT (Table 1). Odds ratios of about 3 or above were found for a previous history of DVT or pulmonary embolism, venous insufficiency and chronic heart failure, with pregnancy, violent effort or muscular trauma, deterioration in general condition and being confined to bed or armchair as the main triggers.

Table 1: Risk factors for DVTs

Risk factor

Case patients

Control patients

Odds ratio

Intrinsic factors

History of DVT or embolism 21 2.4 16
Venous insufficiency 70 41 4.5
Chronic heart failure 10 4.5 2.9
BMI more than 30 15 7.0 2.4
Standing more than 6 hours/day 39 32 1.9
More than 3 pregnancies 17 9.8 1.7

Triggering factors

Pregnancy 2.4 0.3 11
Violent effort or muscular trauma 7.9 1.0 7.6
Deterioration in general condition 6.3 1.2 5.8
Immobilisation 8.0 2.0 5.6
Long distance travel 13 6.3 2.4
Infectious disease 19 13.0 2.0

Travel as a risk factor

Another case-control study from France examined this issue [3]. It used all patients admitted for DVT or pulmonary embolism from 1992-1995, with control patients those admitted for a an event other than these. Since it was a cardiology department, these were mostly chest pain, hypertension and syncope. There were 160 cases and controls.

Cases and controls were well matched for age (mean 66 years), but there were more women in the cases (Table 2). More of the cases were obese (not defined in this paper), had a history of venous thromboembolism, or had made a journey lasting more than four hours in the preceding four weeks.

Table 2: Significant associations with DVT

Risk factor

Case patients

Control patients

Relative risk
(95% CI)

Female 48 34 1.4 (1.1 to 1.9)
History of DVT 12 4.4 2.7 (1.2 to 6.3)
Obesity 34 20 1.7 (1.2 to 2.5)
Recent travel 24 7.5 3.3 (1.8 to 6.0)

Of the 39 cases who had made a long journey, 28 made it by car, nine by plane and two by train. The mean journey length was about six hours for each mode of travel, and the average time between journey and occurrence of symptoms was 13 days, but with much variation. For 29 of the 39 cases there was no other circumstance or disease to explain the event.

Flying and aspirin for DVT prevention

Is there any way to assess the benefit of taking aspirin to prevent DVT in a long distance flight? One study has attempted to do that [4].

It took the risk of DVT as being about 20 per 100,000 travellers for one long distance journey a year, based on a British Parliamentary Select Committee estimate of the risk as being 0-40 per 100,000 travellers. Other literature suggests this to be a reasonable figure, but the authors used a range of estimates of 10-40 per 100,000 to represent different levels of risk, like age.

The potential benefit of aspirin was a risk reduction of 29% for 160 mg of aspirin daily for 35 days from a 13,000 patient study of aspirin after hip fractures.

Table 3 shows the results in terms of numbers of patients needed to be treated with aspirin to prevent one of them having a travel-related DVT. Estimates range from 8,600 at high risk, to 34,000 for low risk.

Table 3: Numbers needed to treat with aspirin to prevent one DVT, at different levels of risk

Per 100,000 travellers or travellers treated with aspirin

Estimated risk of travel-related DVT 10 20 30 40
Number of travel-related DVTs prevented by aspirin 3 6 9 12
Number of travel-related DVTs occurring despite aspirin treatment 7 14 21 28
NNT with aspirin to prevent one additional travel-related DVT 34,000 17,000 11,000 8,600


DVT is uncommon, and especially uncommon in people without recent cancer or previous hospital admission. With these conditions it is 50 per 100,000 per year, and without them it is about 20 per 100,000 per year. Being older, having certain medical conditions, and having a BMI over 30 increase the risk. Long journeys also increase the risk. The risk is still about 1 in 5,000 in the general population, and that sort of risk is attributed to long distance air travel. To put it in perspective, the risk of being killed on the roads in the UK is about 1 in 17,000 each year.

Aspirin, or stockings, may help, but lots of people have to take aspirin for one to be helped. We don't know what dose, and for how long, is effective. We don't know how effective it would be, and we don't have a good handle on how potential harms relate to potential benefits.

Despite a low risk to the individual, there are a lot of travellers by car, rail and air. At peak levels, Heathrow airport handles about 250,000 travellers a day, many on long flights. Even a low incidence of travel-related DVT should impact on the overall incidence of DVT. But studying benefits of any intervention will not be easy given the enormous size needed for any study. In the meantime the advice we are given for air travel might be applied to other forms of travel or behaviour.


  1. FJ Fowkes et al. Incidence of diagnosed deep vein thrombosis in the general population: systematic review. European Journal of Vascular and Endovascular Surgery 2003 25: 1-5.
  2. M Samama et al. An epidemiologic study of risk factors for deep vein thrombosis in medical outpatients. Archives of Internal Medicine 2000 160: 3415-3420.
  3. E Ferrari et al. Travel as a risk factor for venous thromboembolic disease: a case-control study. Chest 1999 115: 440-444.
  4. YK Loke, S Derry. Air travel and venous thrombosis: how much help might an aspirin be? Medscape General Medicine 2002 4 (3) ( ).
previous or next story in this issue