Risk factors for venous thromboembolism
Background Many risk factors for venous thromboembolism have been identified, but two particular exposures - the use of combined oral contraceptives and long-distance air travel - have generated considerable concern in recent years. In contrast, a possible link between venous thromboembolism and a third exposure - the use of psychotropic drugs - was first raised in the 1950s, but has received surprisingly little attention. Information about all three exposures and the risk of fatal events is limited. These risks were examined in three inter-related national population-based studies. Methods The underlying study population included all men and women aged 15 - 59 years who died in New Zealand between 1990 and 2000, for whom the underlying cause of death was pulmonary embolism. The potential associations between fatal pulmonary embolism and the use of oral contraceptives and psychotropic drugs were explored in a general practice records-based case-control study. Non-users were the reference category for all analyses. Contraceptive supply data were used to estimate the absolute risk of death from pulmonary embolism in users of oral contraceptives. A second case-control study, in which computer-assisted telephone interviews were undertaken with the next of kin of cases who had been resident in New Zealand, and with sex and age-matched controls randomly selected from the electoral roll, investigated the possible association between long-distance air travel and fatal pulmonary embolism. Finally, the absolute risk of dying from pulmonary embolism following a long-distance flight was estimated in a descriptive study based on official migration data and deaths in recent air travellers. Results The adjusted odds ratio for use of any oral contraceptive in the three months before the index date (the onset of the fatal episode) was 13.1 (95% CI 4.4- 39.0). The odds ratio for formulations containing desogestrel and gestodene was about three times higher than the point estimate for levonorgestrel products; preparations containing cyproterone acetate appeared to carry the highest risk. The estimated absolute risk of fatal pulmonary embolism in current users of oral contraceptives was 10.5 (95% CI 6.2 - 16.6) per million woman-years. The adjusted odds ratio for current use of any antipsychotic was 13.3 (95% CI 2.3 - 76.3). Low-potency antipsychotics carried a 20-fold increase in risk; thioridazine was the main drug involved. Antidepressant use was also associated with a significantly increased risk (adjusted odds ratio 4.9 [95% CI 1.1 - 22.5]). Compared with non-travellers, people who had undertaken a flight of more than eight hours' duration in the preceding four weeks were eight times more likely to die from pulmonary embolism (odds ratio 7.9 [95% CI 1.1 - 55.1]). The absolute risk of fatal pulmonary embolism following air travel of more than eight hours was 1.3 (95% CI 0.4 - 3.0) per million arrivals. Conclusions The present research was the first to have estimated the relative risks of fatal pulmonary embolism in relation to three exposures: oral contraceptive use in a population in which preparations containing desogestrel and gestodene preparations were widely used, conventional antipsychotics, and long-distance air travel. The findings were consistent with previous, and subsequent, studies of non-fatal events. Increased risks of fatal pulmonary embolism in users of antidepressants, and in people with an intellectual disability, have not been described previously and warrant further investigation. Referral and diagnostic biases are very unlikely in these studies of fatal events, and other types of bias and possible confounding are considered unlikely explanations for the findings.
Advisor: Skegg, David; Paul, Charlotte
Degree Name: Doctor of Philosophy
Degree Discipline: Preventive and Social Medicine
Publisher: University of Otago
Research Type: Thesis
Hardcopy of thesis in two volumes, digitized copy in one Pdf.