November 27, 2024

expert reaction to study on HRT tablets and increased heart disease/blood clot risk

A study published in the BMJ looks at HRT tablets and increased risk of heart disease and blood clots. 

Dr Paula Briggs, Consultant in Sexual and Reproductive Health, Liverpool Women’s NHS Foundation Trust & Immediate Past Chair of the British Menopause Society, said:

“Randomised controlled trials are the gold standard – they match participant demographics to ensure that outcomes are genuinely related to the intervention studied, irrespective of confounders, which for menopausal women include age, time since last menstrual period, body mass index, blood pressure, fasting lipid profile, family history and a personal history of diabetes.

“Observational studies are inherently flawed, due to an inability to confirm patient compliance with medication and challenges relating to an accurate recording of patient outcomes. This also applies to studies using an emulated target trial design and this study does not build on the existing database associated with a more robust research methodology.

“Registry database studies are inevitably influenced by confounders including the demographics of the population included and logistic regression analysis was not applied to this research to adjust for confounders.

“The data provided in this study is not new information and should not influence clinical practice in the UK or elsewhere. It is important to acknowledge that the risk of a cardiovascular event increases with age in all women, irrespective of use of HRT. However transdermal estrogen minimises the risk of VTE by avoiding first pass activation of D-dimers in the liver.

“This publication refers to a variety of different menopausal hormone therapies, including oral menopausal hormone therapy containing both estrogen with and without a progestogen, oral estrogen with local progestogen, tibolone, a unique oral formulation, which stimulates hormone receptors (gonadomimetic), transdermal therapy containing estrogen, with and without a progestogen, and no HRT.

“It is also important to consider biological plausibility. With the wealth of research data that is available for the arterial and metabolic action of HRT, particularly oestrogen, any claims that HRT is causing cardiovascular harm would need to demonstrate the biological basis for this.

Does the press release accurately reflect the science?

“Yes, it confirms what was already known.  The press release confirms that oral HRT is associated with an increase in the risk of venous thromboembolism.  We cannot confirm from this study that any of the preparations are associated with an increase in cardiovascular events as essential information in relation to patient demographics is missing.

Is this good quality research?  Are the conclusions backed up by solid data?

“Observational studies and in particular, database studies, have a place, particularly in areas where prescribing is increasing and trends may, as a result become apparent. It is however important to stress, that this study does not provide novel data.

How does this work fit with the existing evidence?

“It confirms findings from previous studies but does not provide new evidence.

Have the authors accounted for confounders?  Are there important limitations to be aware of?

“There are major confounders which have not been accounted for, including BMI as the most significant example.

What are the implications in the real world?  Is there any overspeculation?  

“No change in clinical practice.  Women should have a robust baseline assessment and only those with a low baseline risk of CVD/VTE should be offered oral HRT, with supportive information, informing them of the increase in risk. This is important as some women do not respond to transdermal delivery routes.”

What are the findings? What does the research show?

“Nothing new.

Oral oestrogen is associated with an increase in the risk of VTE.

“This is not new information but highlights the need for concomitant lifestyle interventions when considering use of HRT for midlife women (who also have age related risk factors). The research confirms that oral HRT is associated with an increase in the risk of VTE.”

Dr Stephen Burgess, Group Leader at the MRC Biostatistics Unit, University of Cambridge, said:

“While the reporting of this research in the press release is technically accurate, it may be confusing to a casual reader. The “trials” conducted here are not clinical trials in the true sense of the term, rather they are observational comparisons. Additionally, these are not 138 separate trials, but instead 138 comparisons of the same dataset. This is a legitimate analysis strategy that is recommended for the analysis of such data. However, there is no randomization of trial participants, as one might expect in a clinical trial. Hence, while the analysis strategy has some advantages over that of a randomized clinical trial (such as having a large sample and being more representative of “real world” practice), the comparison of individuals prescribed versus not prescribed HRT is not a randomized comparison, and so is more susceptible to the influence of confounding factors. The investigators have accounted for these factors as well as they can; but some uncertainty remains as to whether observed differences in risk are truly due to HRT treatment in a causal sense.

“Despite this intrinsic limitation, this research provides helpful evidence that agrees with previous findings (including those from randomized trials) that HRT confers additional risk of clotting diseases: including risk of heart attack, stroke, and venous thromboembolism. This research adds to that evidence by suggesting that different HRT treatments may affect these various diseases to a greater or lesser extent, although no treatment is risk free. Patients should discuss and weigh the benefits of HRT versus the potential risks with their healthcare professionals.”

Rachel Richardson, Methods Support Unit Lead, Evidence Production and Methods Directorate, Cochrane, said:

“While the authors have used an innovative design to examine the effects of HRT, this is still an observational study, and the influence of potential confounding variables cannot be ruled out. For example, the authors were not able to take BMI into account. It may be that doctors were more likely to prescribe particular types of HRT to women with obesity, and obesity is also a risk factor for heart disease.

“Another problem with observational studies is that researchers are not always able to get all the data that they need.  Sources of information are often incomplete, or not designed to provide the data in the exact format needed. The authors of this study used information from the Swedish national patient register and the cause of death register to identify women who developed heart disease. However, the national patient register only identifies women who are diagnosed in hospital and women diagnosed in primary care, or who were not formally diagnosed, would not be picked up.

“Finally, it’s important to note that although this study included over 900,000 women, the authors were trying to identify outcomes that are rare. This means that their estimates of the differences between HRT and control often have wide margins of error. For example, the estimate for the effect of tibolone versus control on developing ischaemic heart disease does show an increased risk with tibolone, however the wide margin of error means that there could be no difference.”

‘Contemporary menopausal hormone therapy and risk of cardiovascular disease: Swedish nationwide register based emulated target trial’ by Therese Johansson et al. was published in The BMJ at 23:30 UK time on Wednesday 27th November. 

DOI: 10.1136/bmj-2023-078784

Declared interests

Dr Stephen Burgess: I have no relevant conflict of interest to declare.

Rachel Richardson: I have no interests to declare.

For all other experts, no reply to our request for DOIs was received.