Use of postmenopausal hormone therapy and risk of Alzheimer’s disease in Finland: nationwide case-control study
BMJ 2019; 364 doi: https://doi.org/10.1136/bmj.l665 (Published 06 March 2019) Cite this as: BMJ 2019;364:l665Linked Editorial
Menopausal hormone therapy and cognition
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
We read the article by Savolainen-peltonen et al with interest and congratulate the authors for their valuable contribution on the discussion on the risks of postmenopausal hormone therapy use. It was surprising to notice, though, that the authors cited our previous research on the topic (references 12 and 13 in Savolainen-Peltonen et al). as ‘supporting the protective effect of hormone therapy on Alzheimer’s disease’.
The only protective association observed in references 12 and 13 of Savolainen-Peltonen et al, was a decreased risk of AD among long-term users (observed only for self-reported use in reference 13), and possible explanations for this finding, including healthy user bias, were discussed in both articles.
In reference 13, a cohort study with self-reported hormone therapy use complemented with register-based hormone therapy use from the same sources as Savolainen-Peltonen et al, the main finding was that ‘Postmenopausal estrogen use was not associated with AD risk in register-based or self-reported data (hazard ratio/95% confidence interval 0.92/0.68–1.2, 0.99/0.75–1.3, respectively)’.
Reference 12, the MEDALZ study was based on the same data sources as the current paper by Savolainen-Peltonen et al (including 55% of the same women; AD cases from 2005-2011). In that paper, we summarised our results as follows: ‘use of systemic estrogen and progestogen was associated with an increased risk of AD, with ORs (95% CI) of 1.10 (1.06-1.12) and 1.13 (1.10-1.17) respectively’. Despite some differences in the exposure assessment, the adjusted OR, 95% CI for estrogen use in our study (1.10, 1.06 to 1.12) was nearly identical to that observed in the study by Savolainen-Peltonen et al (OR, 95% CI 1.09, 1.05 to 1.14). Therefore, it was puzzling to see that these findings were interpreted as supportive for protective effect.
References (numbering corresponds to that in the Savolainen-Peltonen et al)
12. Imtiaz B, Taipale H, Tanskanen A, et al. Risk of Alzheimer’s disease among users of postmenopausal hormone therapy: A nationwide case-control study. Maturitas2017;98:7-13. doi:10.1016/j.maturitas.2017.01.002 pmid:28274328
13Imtiaz B, Tuppurainen M, Rikkonen T, et al. Postmenopausal hormone therapy and Alzheimer disease: A prospective cohort study. Neurology2017;88:1062-8. doi:10.1212/WNL.0000000000003696 pmid:28202700
Competing interests: No competing interests
I am impressed by the results of this paper by Savolainen-Peltonen and others. However, I am not convinced by the relation between usage of hormone therapy and onset of Alzheimer’s disease (AD). How long is the period between prescription of hormone therapy and diagnosis of AD? Patients with AD are known to show preclinical signs before diagnosis of AD. I wonder if hormone therapy starts the mechanism of AD pathophysiology or worsens the AD state.
Competing interests: No competing interests
The Finish case-control study should be interpreted with caution. The study identified 84,739 cases of Alzheimer’s disease (AD) from 1999 and 2013 matched to 84,739 controls by age and hospital location from the entire population. Thus, based on this approach, exposure probability for two control subjects one with ten years of data and one with two years of data in the database can be very different. The control subject with ten years of data has a higher chance of being prescribed hormone replacement therapy (HRT) than the second subject leading to biased odds ratios. Controls should be selected in such a way that takes into account their person time contribution in the population.
Moreover, HRT prescribing trends have varied in the last twenty years mainly due to publication of studies that have highlighted serious adverse events with HRT. One land mark study published in 2002 by The Women’s Health Initiative Investigators showed an increase in the risk of breast cancer and cardiovascular disease with HRT use which prompted a decrease in HRT prescriptions that might have lasted for a number of years. Thus, calendar time bias might have led to an increase in the odds ratio if more unexposed controls had been selected during this period.
References:
1. Writing Group for the Women's Health Initiative Investigators. Risks and Benefits of Estrogen Plus Progestin in Healthy Postmenopausal Women: Principal Results From the Women's Health Initiative Randomized Controlled Trial. JAMA. 2002;288(3):321–333.
Competing interests: No competing interests
The observed association between Alzheimer's disease and hormone replacement therapy (HRT) using cases and age-matched controls is a natural consequence of the protective effects of HRT against cardiovascular disease (CVD). Women carrying apolipoprotein E4 alleles are at higher risk of cardiovascular disease and also of Alzheimer's disease. If they take HRT then their mortality from CVD is reduced and they are more likely to survive to develop Alzheimer's disease. Putting it another way, a cohort of women who take HRT and survive to old age will carry apolipoprotein E4 alleles at higher frequency than survivors who have not taken HRT. Hence they will be at higher risk of Alzheimer's disease. A similar mechanism explained what appeared to be a protective effect of smoking against Alzheimer's disease, which was accounted for by the fact that apolipoprotein E4 carriers who smoked tended to die young from CVD.
Competing interests: No competing interests
The almost universal use of contraceptive “Pill” progestogens and oestrogens in young women causes biochemical perturbations and often medical complications. At the menopause falling levels of sex steroid hormones can unmask symptoms. although the menopause is a long-evolved normal event, menopausal symptoms made worse by previous contraceptive hormone exposures are “treated” with further progestogens and oestrogens. This has led to the rise of HRT.
The Finnish post menopausal case-control hormone therapy (HT) and Alzheimer’s disease study found use of both systemic combined and oestrogen only (estradiol) HT was associated with a 9 to17% increased risk of Alzheimer’s disease. The mean age at initiation of systemic HT was 52 years. In women aged < 60 years at HT initiation, the risk increases were associated with HT exposure over 10 years. The authors think that HT users should be informed of the risk of Alzheimer’s disease with prolonged use.1
“Women need to be informed about the dangers of Hormone Replacement Therapy” is one of our most accessed publications according to Research Gate. It was about the epidemic increases in breast and ovarian cancers and vascular diseases with HRT.2
The menopause in healthy women is physiological but menopausal symptoms are like a lighthouse which flashes signals of adverse reactions to foods and chemicals. In my experience undiagnosed essential nutrient deficiencies, infections or increases in toxic DNA adducts are common. Both hormonal contraceptives and HT can lower white cell zinc levels, red cell magnesium levels and raise serum copper levels which can reduce copper stores and impair superoxide dismutase function. Zinc is essential for brain function and necessary for hundreds of enzymes. Zinc and magnesium deficiencies can block omega-3 and omega-6 essential fatty acid pathways. Zinc and magnesium deficiencies also increase headaches, vasomotor symptoms like hot flushes, mood changes and osteoprosis.3-5 HT use can cause addiction with severe withdrawal symptoms when discontinued.6
HRT promoters concentrate on giving different types and doses of progestogenic and oestrogenic hormones rather than investigating common biochemical abnormalities.
1 Savolainen-Peltonen H, Rahkola-Soisalo P, Hoti F, Vattulainen P, Gissler M, Ylikorkala O, Mikkola TS. Use of postmenopausal hormone therapy and risk of Alzheimer’s disease in Finland: nationwide case-control study. BMJ 2019; 364 doi: https://doi.org/10.1136/bmj.l665 (Published 06 March 2019) BMJ 2019;364:l665.
2 Price EH, Little HK, Grant ECG. Steel CM. Women need to be warned about dangers of hormone replacement therapy. BMJ Clinical Research 1997;314:376-77.
3 Grant ECG, Howard JM, Davies S, Chasty H, Hornsby B, Galbraith J. Zinc deficiency in children with dyslexia: concentrations of zinc and other minerals in sweat and hair. BMJ 1989;296:607-609.
4 Grant ECG. The pill, hormone replacement therapy, vascular and mood over-reactivity, and mineral imbalance. J Nutr Environ Med 1998;8:105-116.
5 McLaren Howard J, Grant ECG, Davies S. Hormone replacement therapy and osteoporosis: bone enzymes and nutrient imbalances. J Nutr Environ Med 1998;8:129-138.
6 White M, Grant ECG. Addiction to oestrogen and progesterone. J Nutr Environ Med 1998;8 :117-120.
Competing interests: No competing interests
Sir,
The recent study by Savolainen-Peltonen and colleagues highlights again the potential impact of menopause and hormone replacement therapy (HRT) on the risk of developing Alzheimer’s disease (AD). The authors report an increased risk of AD after prolonged HRT use regardless of the timing of onset of HRT, contrary to previous observational studies and trials.
Despite the authors’ findings, we would like to raise several issues which we consider important in this context. In particular, the distribution of cases on HRT with AD across the ages was strongly skewed in terms of sample size towards the younger cases. Specifically, the <50-54 group was 3 times larger than all the remaining age brackets combined. The authors report no statistical risk differences across age brackets, however Figure 1 shows clearly that the risk varies considerably with 60-64 year olds showing the highest risk, whereas younger (<60) showed no increased risk. This would strongly suggest that there is a ‘critical’ window’ effect in that the timing of the start of HRT may be crucial for the risk of developing or protection from AD but the skewed nature of the data might have overshadowed such an effect.
Further, the authors mention that apolipoprotein E (APOE) is highly prevalent in the Finnish population and APOE has been shown to interact with HRT to potentially increase AD risk. This raises the question as to whether the admixture of APOE4 vs. APOE3 cases might have led to the result, which would suggest that a more personalised approach to HRT would be needed as a function of underlying APOE genotype. Despite the impressive follow-up of the sample and its linkage to the Finnish health system, the authors highlight that they did not have HRT prescription data for under 50 year olds. In this regard, future investigation using birth cohorts might be particularly relevant as they allow establishment of the lifelong risk for AD as a function of HRT use.
We feel that this paper is an important part in the link between menopause and AD and the influence of HRT. However the applicability of its findings are limited, and do not demonstrate causality. There is an urgent need for research into factors that influence perimenopausal and early postmenopausal AD, including pre-existing cardiovascular disease, the interaction of genotype with risks from HRT use and finally whether women with pre-existing, asymptomatic early AD are those that are more prone to seek HRT during the early menopausal years.
Edward Morris
Consultant Gynaecologist, Norfolk & Norwich University Hospital, Norwich, UK
Michael Hornberger
Professor of Applied Dementia Research, Norwich Medical School, Norwich, UK
On behalf of:
British Menopause Society Medical Advisory Council
Spracklen House, Dukes Place, Marlow SL7 2QH
References
Savolainen-Peltonen Hanna, Rahkola-Soisalo Päivi, Hoti Fabian, Vattulainen Pia, Gissler Mika, Ylikorkala Olavi et al. Use of postmenopausal hormone therapy and risk of Alzheimer’s disease in Finland: nationwide case-control study BMJ 2019; 364 :l665
Competing interests: EM has received speakers fees, honoraria and research funding from Gedeon Richter, Abbott, Besins, Pfizer, Mylan, Novo Nordisk, Cook Medical, Chugai Pharma and Kebomed. He is a trustee of the British Menopause Society/Womens Health Concern and a trustee of the Royal College of Obstetricians and Gynaecologists MH has no disclosures
ENDOCRINES , ALZHEIMER"S DISEASE and OLD TIMER'S ADVICE
==================================================
When pharmacology was in its infancy, physician teachers of vintage age ( = OLD TIMERS ) would advice their novices, " Do No Harm ".
In the light of the notes of Savolainen-Peltonen et al ( vide BMJ 2019:364;l665 ), consideration has to be made of the possible deleterious effects of prescribing estrogenic hormones to females at a particular age. What if the recipient is one's own sister ? !
Competing interests: No competing interests
GPs and gynaecologists have a near impossible task of keeping up to date and counselling women who might be suffering from the often appalling symptoms of the menopausal syndrome that could be assuaged by HRT. This new paper is frankly unhelpful and misleading. The inevitable headlines in the media have added to the problem. Bad news about HRT is always picked by the press.
There are two serious problems with this paper. As always the relative risks look frightening yet in absolute terms we are looking at a number of 18 per 10,000 women on HRT.
The other problem, as hinted at by the authors, is the problem of selection bias in a case control study. The early symptoms of dementia might easily be confused by one of the common symptoms of the menopause, confusion or lapses of memory. Such a tiny difference in absolute risk could easily be explained away by this bias. Many women frightened off taking HRT may not live long enough to suffer from Alzheimer’s disease, dying prematurely from cardio-vascular and cerebra-vascular disease or the complications from pathological fractures. The BMJ now has a responsibility of repairing the damage.
Competing interests: No competing interests
We were disappointed to read the media reporting about this study1. There is still much confusion and uncertainty about the benefits of HRT which include lowering future risk of cardiovascular disease and osteoporosis2,3. Many healthcare professionals are still concerned about prescribing HRT as they are not aware of the current guidelines on the diagnosis and management of the menopause. Research undertaken by the Primary Care Women's Health Forum (PCWHF) has demonstrated this.4
A recent survey of 2904 women revealed that 43% (1,152) were either disappointed or frustrated with the menopause care they had received and 66% (736) had been offered antidepressants instead of HRT which goes against current guidance. Women are struggling to receive evidence based treatment for their menopausal symptoms which often lead to them having worsening quality of lives. Many women increase weight and reduce their exercise because of their symptoms so are increasing their future risk of obesity, cardiovascular disease and osteoporosis; conditions which are already costing the NHS billions of pounds.
Research has shown that low levels of estrogen may lead to cognitive decline; there is an increased risk of dementia in women who have undergone a surgical hysterectomy at an early age5. The results of this study, which is not a randomised controlled trial, are clear that in absolute terms, 9 to 18 additional cases of Alzheimer’s disease a year will be detected in 10 000 women aged 70-80. This number is so small and needs to be considered in context. Some of those women in this study taking HRT might have been prescribed it for prodromal symptoms of dementia such as memory dysfunction. Of note, the authors clearly state that their previous research has shown a marked decrease in risk of death from vascular dementia in those on HRT6.
Women who start HRT within 10 years of their menopause and continue to take it in the long term have a lower risk of all-cause mortality, including from cancer7.
There are different types of HRT. This study did not differentiate between oral or transdermal oestrogen nor did women take micronised progesterone8. Transdermal oestrogen and micronised progesterone are optimal types of HRT for women.
In addition, there is increasing evidence to support the use of testosterone in menopausal women as part of their hormone replacement therapy. Testosterone replacement in women has been shown to exhibit neuroprotective effects within the brain, including protection against oxidative stress, serum deprivation-induced apoptosis and soluble β-amyloid toxicity9.
It is so important that women are given accurate, evidence-based information so they can have an individualised consultation regarding their perimenopausal and menopausal symptoms.
References
1. BMJ 2019;364:l665
2. NICE guideline NG23 – Menopause:diagnosis and management 2015
3. 2016 IMS Recommendations on women’s midlife health and menopause hormone therapy. Climacteric 2016;19(2):109-50
4. BJFM 2018, 12-22
5. Neurology 2007; 69,1074–1083.
6. J Clin Endocrinol Metab 2017;102:870-7.
7. JAMA. 2017 Sep 12;318(10):927-938
7. Br J Gen Pract. 2018;68(675):499-500
8. Front Neuroendocrinol 2009 ; 30 : 239 – 58
Competing interests: No competing interests
Re: Use of postmenopausal hormone therapy and risk of Alzheimer’s disease in Finland: nationwide case-control study
Alzheimer’s Disease (AD) is important for UK women as it is now the commonest cause of death, having surpassed heart disease, which remains the main cause of death in men [1]. While it is well established that the hippocampus is the seat of the patho-physiology of AD, there are no definitive interventions to prevent or treat this awful pandemic, and any publication on AD will attract public attention and has the potential to modify behavior. We are therefore concerned that in their recent paper in the BMJ, Savolainen-Peltonen, H et al (2018) [2] seem to make claims that are misleading to the scientific community and the public. The objective of their study was clearly specified: to compare the use of hormone therapy between Finnish postmenopausal women with and without a diagnosis of AD. Yet they have ended up reporting that the use of systemic hormone therapy was associated with a 9-17% increased risk of AD. The design of a retrospective case-control study such as this allows the authors to conclude on the association between AD and exposure to hormone treatment in their past, but any claim on the odds or risk of AD is misleading, and raises epidemiological suspicion of the other findings.
The odds, as used in horse racing and betting, and odds ratio, as used in science, are completely different from each other. Each is completely different from measures of effect used to measure risk of an event like AD, such as relative risk, risk ratio or risk rate. To clarify, in science one can only measure the risk of developing an event. Although there have been numerous publications in the past, it has always been senseless to measure the risk of using HRT, since in reality what is often being measured is exposure. Especially in a retrospective analysis, it makes more sense to measure the odds or odds ratio of using HRT in women who have a disease such as AD. Thus Tables 2 and 3 that report on the odds ratio of AD are scientifically misleading because they can only measure the exposure, that is, the odds ratio of exposure to HRT.
Moreover Figure 1, which correctly reports on the odds ratio of exposure, shows that there is no evidence that the age of initiation of HRT has any effect between women with and without AD because all the confidence intervals include 1.0.
Unfortunately, again the paper misleads by reporting on the risk of AD.
This publication [2] simply helps to sow anxiety in a public that has long been subjected to alarmist messages about HRT.
1. Office of National Statistics, 2016.
https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarri...
hs/bulletins/deathsregisteredinenglandandwalesseriesdr/2015.
2. Savolainen-Peltonen Hanna, Rahkola-Soisalo Päivi, Hoti Fabian, Vattulainen Pia, Gissler Mika, Ylikorkala Olavi et al. Use of postmenopausal hormone therapy and risk of Alzheimer’s disease in Finland: nationwide case-control study BMJ 2019; 364:l 1665
Competing interests: No competing interests