Third-generation oral contraceptives and venous thrombosis

Occlusive vascular diseases in oral contraceptive users. Epidemiology, pathology and mechanisms

Despite being an unprecedented departure from normal physiology, the combined oral contraceptive is not only highly effective, but it also has a remarkably good safety record. Concerns over safety persist, though, particularly with regard to venous thromboembolism (VTE), stroke and myocardial infarction (MI). Epidemiological studies consistently show an increase in risk of VTE, but the results are more contentious with regard to arterial diseases. Despite 40 years of research, the mechanisms behind these adverse effects are not understood. In this review, we integrate information from published studies of the epidemiology and pathology of the occlusive vascular diseases and their risk factors to identify likely explanations for pathogenesis in oral contraceptive users. Oral contraceptives induce both prothrombotic and fibrinolytic changes in haemostatic factors and an imbalance in haemostasis is likely to be important in oral contraceptive-induced VTE. The complexity of the changes involved and the difficulty of ascribing clinical significance has meant that uncertainty persists. A seriously under-researched area concerns vascular changes in oral contraceptive users. Histologically, endothelial and intimal proliferation have been identified in women exposed to high plasma estrogen concentrations and these lesions are associated with thrombotic occlusion. Other structural changes may result in increased vascular permeability, loss of vascular tone and venous stasis. With regard to arterial disease risk, epidemiological information relating to dose effects and joint effects with other risk factors, and studies of pathology and changes in risk factors, suggests that oral contraceptive use per se does not cause arterial disease. It can, nevertheless, synergise very powerfully with subclinical endothelial damage to promote arterial occlusion. Accordingly, the prothrombotic effects of the oral contraceptive estrogen intervene in a cycle of endothelial damage and repair which would otherwise remain clinically silent or would ultimately progress - in, for example, the presence of cigarette smoking or hypertension - to atherosclerosis. Future work in this area should focus on modification of the effects of established risk factors by oral contraceptive use rather than modification of the supposed risk of oral contraceptive use by established risk factors. Attempts to understand vascular occlusion in oral contraceptive users in terms of the general features of VTE or with reference to atherosclerosis may be limiting, and future work needs to acknowledge that such occlusions may have unique features. Unequivocal identification of the mechanisms involved would contribute considerably to the alleviation of fears over vascular disease and to the development of even safer formulations.

Validation of the diagnosis of venous thromboembolism in general practice database studies

AIMS

The study was conducted to determine whether the method for selecting cases of venous thromboembolism (VTE) from general practice databases significantly affected the findings of an epidemiological study.

METHODS

Cases of VTE were identified from the UK General Practice Research Database (GPRD) by searching for codes for deep vein thrombosis (DVT) and pulmonary embolism (PE). These had to be supported by evidence of anticoagulation and be exposed to a combined oral contraceptive (COC) at the time of the event. Additional information about the event was sought from general practitioners who were requested to complete a questionnaire and to supply anonymised copies of hospital letters and discharge summaries.

RESULTS

Of the 285 cases identified from the GPRD, additional information was available for 177 VTE events. This information showed that 84% of those events were supported by hospital investigations or a death certificate. Using only verified cases, rather than all GPRD identified events, did not alter the results of the epidemiological study.

CONCLUSIONS

The GPRD provides information of sufficiently high quality to allow valid epidemiological research of VTE events. Excluding cases without a database record of hospital admission would lead to valid events being overlooked, and an under-estimate of the disease incidence.

Third- and second-generation oral contraceptives are associated with similar risk estimates for venous thromboembolism

A scientific discussion between 1995 and 1999 addressed the question whether second- and third-generation oral contraceptives (OCs) were associated with different risks of venous thromboembolism (VTE). Results from three epidemiological studies became available in the course of 1995, in which such differences were observed. Although it was unclear at that time whether these observations reflected causality or were induced by bias and/or confounding, some regulatory bodies in Europe restricted the indication for use of third-generation oral contraceptives. Immediate media attention generated a pill scare in those, but also other, countries. Indications for the influence of bias were observed in the initial studies of 1995 and further substantiated in subsequent utilization and prescribing surveys. The most important bias seemed to be related to differences in age and duration of use between third- and second-generation OC users. A number of new studies as well as new analyses in two of the 1995 databases included measures to limit the effect of the identified biases/confounders. These studies observed similar risks of venous thromboembolism with second- and third-generation oral contraceptives. Two other recently published studies did not or could not include the same level of control for confounding and reported similar results as the 1995 studies, thus reconfirming the relevance of the identified confounders. Population data show that the massive switch in the UK from third-generation OCs to second-generation OCs in 1995 has not resulted in a reduction of the incidence of VTE in OC users after 1995, illustrating that the risk of VTE is not determined by the type of low-dose pill used. In addition, data from trend analyses, spontaneous reporting and studies addressing hemostatic mechanisms in pill-users also do not support a potential difference in risk of VTE between users of second- and third-generation OCs.

Venous thromboembolic disease and combined oral contraceptives: A re-analysis of the MediPlus database

In October 1995 the Committee on Safety of Medicines advised UK doctors and pharmacists that oral contraceptives containing desogestrel and gestodene were associated with double the risk of venous thromboembolic events (VTE) compared to pills containing other progestogens. In 1997 data was analysed from the MediPlus database of UK general practitioner records, which reported odds ratios for desogestrel and gestodene lower than that for levonorgestrel. Here the results of a more stringent nested case control analysis on the MediPlus database are reported. The study was larger and cases were verified. A crude incidence of idiopathic VTE was found amongst users of combined oral contraceptives of 4.6 per 10 000 exposed women years. Using levonorgestrel 150 microg + ethinyloestradiol 30 microg as reference, non-significant odds ratios of 1.1 (0.5-2.6) for desogestrel 150 microg + ethinyloestradiol 30 microg and 1.1 (0.5-2.4) for gestodene 75 microg + ethinyloestradiol 30 microg were found. The results of this study show no significant difference in risk between different formulations of combined oral contraceptive.

Newer oral contraceptives and the risk of venous thromboembolism

Research on the relationship between venous thromboembolism and the progestagen content of combined oral contraceptives has pointed to an increase in risk associated with products containing desogestrel and gestodene. Although many biases must have been at play in these nonexperimental studies, the errors that have been suggested and examined are not of a sufficient magnitude to account for the observed results. The most plausible explanation of the available data is that combined oral contraceptives containing desogestrel and gestodene carry a very small risk of venous thromboembolism, which exceeds the even smaller risk carried by products containing levonorgestrel. The position of norgestimate is uncertain.