HRT review finds increased risk of blood clots and stroke

Tuesday March 10 2015

"Women on HRT pills should be aware that there is a small chance of an increased risk of blood clots and possibly stroke," BBC News reports.

This story is based on an update of a review on the effects of hormone replacement therapy (HRT) on the risk of heart and blood vessel diseases (cardiovascular diseases).

The updated results supported their previous findings that HRT did not reduce the risk of deaths from any cause or from cardiovascular causes during follow-up. However, it did increase the risk of stroke and blood clots (for example deep vein thrombosis – DVT).

Researchers estimated that for every 1,000 women taking HRT, an extra six would experience stroke and an extra eight would experience a blood clot.

The review is robust and the trials of good quality. Still, there are some important points to note. The trials did include different doses and types of oral HRT given over different time periods, but results may not be representative of each of these individually. Also, the women were, on average, about 60 years old at the start of the trials, so results may not be representative of women who start taking HRT at a younger age.

If you are receiving HRT and are concerned, or are considering it, you should discuss your individual risk factors (such as a family history of clotting or stroke), as well as potential benefits, with your GP.

Where did the story come from?

The study was carried out by researchers from the University of Oxford and other research centres in the UK and Spain. They were carrying out the review for the Cochrane Collaboration – an international independent network of researchers and professionals, as well as patients and carers, which produces and updates a library of systematic reviews on a wide range of healthcare questions. The Cochrane Collaboration does not accept commercial or conflicted funding.

The study in question was published in the Cochrane Library – an online resource that is free to access.

The UK media generally covered this story well, giving balanced viewpoints from the authors of the review.

What they don’t mention is that the previous version of the review had similar findings, so the results are not unexpected.

What kind of research was this?

This was a systematic review of randomised controlled trials (RCTs), assessing whether HRT affects postmenopausal women’s risks of cardiovascular (heart) disease. Observational studies had suggested that women taking HRT were at lower risk of death or heart disease events during follow up. However, later RCTs had contradicted these findings. Some research has suggested that HRT might only reduce cardiovascular risk if it is started soon after the menopause starts.

A systematic review is the best way of assessing what the existing research says about any given question. They aim to use transparent, rigorous and unbiased methods to identify as much of the relevant evidence as possible, to assess its quality, and to analyse and interpret their findings.

Over time, new research evidence is published, so Cochrane reviews are regularly updated to incorporate new evidence and see if conclusions change as a result. The current publication updated the previous version of this review from 2013. The previous version found that HRT did not reduce the risk of heart problems, but did increase risk of stroke and blood clotting events.

What did the research involve?

The researchers searched multiple literature databases to identify RCTs comparing the effects of HRT versus a dummy pill (placebo) or no treatment. They selected those RCTs that met their inclusion criteria, assessed their quality, and pooled their results in a meta-analysis.

The researchers only included RCTs in women followed up for at least six months. HRT had to be given orally – trials of HRT patches, creams etc. were excluded. HRT could contain oestrogen alone or oestrogen plus a progestogen.

The main outcomes the researchers were interested in were death from any cause, death from a heart-related (cardiovascular) cause, non-fatal heart attack, stroke or chest pain (angina). They were also interested in blood clots, either in the lungs (pulmonary embolism) or DVT.

The results of the trials were analysed using standard meta-analytical techniques.

What were the basic results?

The searches identified six relevant RCTs published since the review was last published. This brought the total number of RCTs to 19, featuring 40,410 postmenopausal women. Nine RCTs included relatively healthy women, the majority of whom were not known to have heart disease, and 10 RCTs included women with known heart disease. The RCTs assessed various types and doses of HRT, for different lengths of time (seven months to 10 years). The RCTs were generally of good quality.

Meta-analysis of the trials found that HRT did not affect women’s risk of death from any cause or death from cardiovascular disease during follow up, or of non-fatal heart attacks, compared to placebo or no treatment. Similar results were obtained if trials in women with and without heart disease were analysed separately.

HRT increased risk of stroke compared to placebo or no treatment – with an extra six women per 1,000 experiencing stroke with HRT. On average, across the studies, about 31 women per 1,000 taking HRT experienced stroke during follow-up, compared to 25 women per 1,000 not taking HRT. This meant that for every 165 women taking HRT, there would be one extra stroke over an average of about four years. The overall quality of the evidence on this outcome was rated as high.

HRT also increased risk of blood clot (venous thromboembolism) – with an extra eight women per 1,000, on average, experiencing clots with HRT. On average across the studies, about 19 women per 1,000 taking HRT experienced clots during follow-up, compared to 10 women per 1,000 not taking HRT. The overall quality of the evidence on this outcome was rated as moderate.

If the women were split by when they started taking HRT, results in the group of women who started taking HRT more than 10 years after menopause started were similar to the overall results above.

However, HRT reduced risk of death during follow-up in the women who started taking it less than 10 years after menopause started (relative risk (RR) 0.70, 95% confidence interval (CI) 0.52 to 0.95).

HRT also reduced risk of death from heart disease or non-fatal heart attack in these women (RR 0.52, 95% CI 0.29 to 0.96). The risk of stroke was not significantly affected by HRT in this group, but the risk of blood clots was still increased (RR 1.74, 95% CI 1.11 to 2.73). These analyses did not include as many women as the overall analyses (only about 8,000-9,000), and relied largely on one trial.

How did the researchers interpret the results?

The researchers concluded that their findings
"provide strong evidence that treatment with hormone therapy in post-menopausal women overall … has little if any benefit and causes an increase in the risk of stroke and venous thromboembolic events".

Conclusion

This updated Cochrane review has found that oral HRT increases risk of stroke and blood clots, and does not appear to reduce overall risk of cardiovascular disease or death during follow-up.

More exploratory analyses suggested that HRT might reduce risk of death from heart disease or non-fatal heart attack if it was started within 10 years of menopause, but this finding needs further confirmation.

The review was carried out using robust methods and the trials were of good quality. Its findings are in line with the previous version of the review, and also with other reviews.

There are some points to note:

  • This review specifically looked at the effects of HRT on the risk of heart and vascular disease. It did not assess the effects of HRT on women’s menopausal symptoms or quality of life.
  • Although there were 19 trials included, three large trials (two testing combined HRT) had the greatest impact on the analyses. The authors note that there is debate about whether the results of these trials apply to all types of HRT, and all women receiving HRT.
  • The trials did include different doses and types of HRT, but results may not be representative of each of these individually. The review did not assess non-oral HRT; therefore, results may not apply to this form of HRT.
  • The women were, on average, about 60 years old at the start of the trials. Many women would start HRT at a younger age than this, soon after the start of the menopause, to counteract menopausal symptoms.

Overall, the review supports previous findings.

When prescribing any medicine, it is important to consider the balance of benefits and harms for the individual. If you are receiving HRT and are concerned, or are considering it, you may find it useful to discuss individual risk factors with your GP, such as whether you have an increased risk of developing a blood clot or stroke. This will help you to weigh your risks against the benefits HRT can bring, especially if your menopausal symptoms are particularly severe.

The UK MHRA notes that HRT does relieve menopausal symptoms, and suggests that for all women taking HRT, the lowest effective dose should be used for the shortest time.

Analysis by Bazian
Edited by NHS Choices