Elsevier

The Lancet

Volume 348, Issue 9033, 12 October 1996, Pages 977-980
The Lancet

Articles
Risk of venous thromboembolism in users of hormone replacement therapy

https://doi.org/10.1016/S0140-6736(96)07113-9Get rights and content

Summary

Background

The association between current use of oral contraceptives and increased risk of venous thromboembolism (VTE) has been firmly established. Although data-sheets for hormone replacement therapy (HRT) carry similar warnings as regards VTE, evidence of an association is inconclusive. We carried out a hospital-based case-control study to investigate whether current use of HRT is associated with VTE.

Methods

We screened all women aged 45–64 years admitted to hospitals in the area of the Oxford Regional Health Authority with a suspected diagnosis of VTE between February, 1993, and December, 1994. We recruited 81 cases of idiopathic VTE and 146 hospital controls with disorders of eyes, skin, ears, respiratory and alimentary tracts, kidneys, bones, and joints, and trauma; controls were matched to cases for age-group and date and district of admission. To increase the study power, an additional 22 cases of idiopathic VTE and 32 hospital controls admitted before February, 1993, were recruited retrospectively. Participants were questioned about medical and gynaecological history, use of oral contraceptives and HRT, use of other drugs within the previous 3 months, and lifestyle and socioeconomic characteristics. Detailed diagnostic data were extracted from the notes of eligible cases. Matched analyses, adjusted for body-mass index, socioeconomic group, and history of varicose veins, were undertaken by conditional logistic regression.

Findings

44 (42·7%) cases and 44 (24·7%) controls were current users of HRT. The adjusted odds ratio for VTE in current users of HRT compared with non-users (never-users and past users combined) was 3·5 (95% CI 1·8–7·0; p<0·001). No association was found with past use, and risk appeared to be highest among short-term current users (adjusted likelihood ratio test of trend in odds ratios across different durations of current use, p=0·011).

Interpretation

Current HRT use is associated with risk of VTE. The increased risk may be concentrated in new users. The number of extra cases appears to be only about one in 5000 users per year. These findings need to be weighed against the probable benefits of long-term treatment, including reductions in risks of osteoporotic fracture and coronary heart disease, and the probable modest increase in risk of breast cancer.

Introduction

Most studies linking the use of oestrogen to increased risk of venous thromboembolism (VTE) have been carried out in young women in relation to use of oral contraceptives.1 Hormone replacement therapy (HRT) is not generally believed to carry a similar risk in postmenopausal women. Although none of the studies to address this issue has found a significant increase in VTE associated with HRT,2, 3, 4, 5 each lacked power to detect important risks. The British National Formulary6 states that the evidence of an increased thrombotic risk associated with HRT is questionable, but lists active thrombophlebitis or thromboembolic disorders as contraindications to treatment. Similarly, although a working party of the UK Royal College of Obstetricians and Gynaecologists on prophylaxis against thromboembolism concluded that there was insufficient evidence to suggest that HRT is associated with an increased risk of VTE, their report acknowledged the need for further research.7

Our study was undertaken to investigate a possible association between current use of HRT and the risk of idiopathic deep-vein thrombosis (DVT) and pulmonary embolism (PE). The protocol was based on that of a case-control study to examine the association between oral-contraceptive use and VTE in young women.8

Section snippets

Cases

Cases were recruited between February, 1993, and December, 1994, from hospitals in the area of the Oxford Regional Health Authority (as defined before April, 1994) by twice-weekly screening of all relevant wards. Eligible cases were women aged 45–64 years with a suspected diagnosis of PE, DVT, or both. Women with a history of PE, DVT, stroke, or myocardial infarction, and those with a history within the previous 6 weeks of surgery, pregnancy, trauma, or illness necessitating bed rest for longer

Results

108 cases (69 with DVT and 39 with PE) and 232 controls satisfied all inclusion criteria. The proportions of cases with definite, probable, possible, and other diagnostic certainty ratings were 73%, 20%, 4%, and 3% for DVT, and 33%, 51%, 10%, and 5% for PE. Cases in the 'other' category were excluded from all further analyses. Participants without matching cases or controls were also excluded. 103 cases and 178 controls remained; of these, 22 cases and 32 controls had been recruited

Discussion

The results of our study and those of Jick and colleagues12 suggest a possible causal relation between current HRT use and idiopathic VTE. This interpretation is supported by results from a study of exogenous hormones in relation to risk of PE by Grodstein and colleagues.13 In our study, risk of VTE seemed to be highest among short-term users, whereas no association was seen with past use. Although the magnitude of estimated risk was greater among users of higher-dose preparations than among

References (21)

There are more references available in the full text version of this article.

Cited by (0)

View full text