A note before the story. The patients you meet in this newsletter are composites — portraits drawn from patients I have cared for over years, with identifying details changed and the physiology kept honest. Composites let me teach the truth of an exam-room encounter without breaking the trust of the women who lived them. Nothing in this newsletter is medical advice for any individual reader.
The first time I met Jennifer she was 51, 8 months out from her last menstrual period, and sitting in my office because her gynecologist and her primary care doctor could not agree. On the intake sheet she had written my mother had a stent at 68, my father had a bypass at 62, my sister is on a statin, and I am now sleeping 4 hours a night. As my mouse hovered over her appointment slot the reason-for-visit window popped open — clearance for hormone therapy. I could hear the whole story in that sentence before I opened the chart.
The numbers she brought with her were from her primary care’s office and they told a familiar story. At 46 her blood pressure had been 118 over 74 and her LDL cholesterol had been 90. She ran 5 miles four mornings a week. Her dietary inventory was reassuring. For 4 years the numbers had said her prevention plan was working.
Then the ovaries had gone quiet.
The office cuff on the day she came to see me read 138 over 88. What would her home log say? The LDL her primary care doctor had drawn 3 weeks earlier was 130. Nothing in her life had changed except a hormonal signal that had been steady for 30 years.
Her gynecologist had already offered her transdermal estradiol and oral micronized progesterone.
Her primary care doctor had said what a careful primary care doctor learned to say in 2005: not with your family history.
She came to me because both were experienced, both were careful, and both could not be right.
The visit at the intersection of hormone therapy and the heart is a two-decisions visit, not a one-decision visit, and it is the visit I want to walk you through week after week in this newsletter. I want to teach you the version of it I use with Jennifer and with the women who ask her questions in my office. I want to teach you the mechanism at the level the evidence actually supports. I also want to share the art of how I deliver it. I want to teach you what the trial the whole conversation goes back to actually studied. And I want to give you the language to bring back to your own visit.
Before menopause, estrogen is associated with vascular effects involving endothelial function, vascular tone, and lipid metabolism. Menopause is often accompanied by changes in blood pressure, body composition, lipids, and vascular function. Those changes deserve attention, even when diet and exercise have not changed.
Think of your arteries like the pipes inside the walls of a house — the pipes that carry the oxygen and nutrients every organ needs to stay alive. But unlike the pipes in your kitchen, these pipes are alive. They are connected to each other. They talk with each other and with the blood moving through them. And they are lined, on the inside, with a single layer of tiny tiles — one cell thick, laid edge to edge from the aorta all the way out to the smallest capillary. That lining is the endothelium. It is the surface that decides how the pipes behave (and where) — whether it relaxes or tightens, whether it stays smooth or gets sticky, whether the blood that touches it clots or keeps moving. For 30 years of a woman’s reproductive life, circulating estrogen is one of the signals those tiles are listening to. When ovarian estrogen falls at menopause, one of the signals they had been listening to changes. Some women feel it fast. Some do not. The pipes are still there. The tiles are still there. What comes off is one of the signals that had been talking to them, quietly, for 30 years, while she was busy living the rest of her life.
But biology alone does not tell us that starting hormone therapy will prevent heart attacks. That is the sentence the whole rest of this article is built around. Mechanism is not outcome. And the trial that stopped in 2002 is the reason we know it is not.
The Women’s Health Initiative — the trial that shaped how a generation of doctors was taught to think about hormone therapy — was designed and enrolled in the 1990s and stopped its combination-therapy arm early in July 2002. The mean age at randomization in that arm was 63. Only about 3 percent of the women were in their early 50s. Two-thirds were 60 or older. The average woman in the WHI was more than a decade past her last period the day she took her first pill.
That is not a criticism of the WHI. It is an observation about who was in the room. The women in Jennifer’s shoes today are not, on average, 63. They are 48, and 51, and 54. They are inside the first decade after their last period — what the field now calls the window.
In 2007 the WHI investigators went back and stratified the trial by age at initiation. The story that came out of that reanalysis is more careful than the way it has been repeated on the internet since. WHI reanalyses suggested that absolute risks were lower, and the overall balance looked more favorable, among younger women and those closer to menopause. But the trial did not establish hormone therapy as a treatment to prevent coronary disease. The 50-to-59 subgroup coronary heart disease hazard ratio was 0.93, with a confidence interval that crossed 1. Timing matters. But timing did not prove cardiovascular protection.
Timing matters because the balance of benefits and risks of starting systemic therapy is generally more favorable earlier in menopause than when therapy begins after age 60 or more than 10 years after menopause. Later initiation carries higher absolute risks of coronary events, stroke, venous thromboembolism, and dementia than earlier initiation, in part because baseline risks are higher with age and time since menopause.
Major societies have integrated that finding into how they recommend the conversation happen. NAMS. ACOG. The Endocrine Society. The European Menopause and Andropause Society. All of them incorporate age, time since menopause, route, dose, and individual risk into shared decision-making. None of them recommend systemic hormone therapy solely to prevent cardiovascular disease. What all of them do is treat symptomatic women inside the window as candidates for a serious individualized conversation about symptom treatment, and treat systemic therapy started well outside the window as a higher-risk decision that requires a specific reason.
Jennifer’s gynecologist was reading that body of work.
Jennifer’s primary care doctor was reading the summer of 2002.
The gap between those two conversations is the gap this newsletter exists to close. Not by arguing against the primary care doctor — he was, in 2005, reading the most careful summary available. By offering the better mechanism, which is that a woman in Jennifer’s window is a candidate for a serious individualized conversation with her clinicians about symptom treatment and a separate, load-bearing cardiovascular prevention plan.
Before I said anything about whether a trial of systemic therapy was reasonable, I wanted to see her risk picture the way I look at everyone’s risk picture in this seat. Her intake and her exam did not turn up a contraindication. What I still needed were 4 numbers her chart did not yet have. The first was a coronary artery calcium score. The second was a high-sensitivity C-reactive protein. The third was an ApoB — the newer marker I like better than LDL alone because it actually counts the atherogenic particles in her blood rather than the cholesterol riding inside them. The fourth was a lipoprotein(a), the inherited particle every woman with her family history deserves to have measured once in her life.
The first was a coronary artery calcium score. A CAC score is a non-contrast CT scan of the heart that measures calcified plaque (cured spackle) in the coronary arteries. In a woman inside the window, in front of a hormone-therapy decision, it answers a specific question. How much plaque has 51 years of biology, family history, and the recent metabolic shift actually laid down in the pipes? A CAC of 0 in a woman her age is a strong signal of low near-term cardiovascular risk — in the Multi-Ethnic Study of Atherosclerosis the coronary event rate in participants with a CAC of 0 was 0.8 per 1,000 person-years, and the warranty period of a zero score in a woman her age is estimated at roughly 5 to 7 years (MESA, Circulation 2016; MESA, JACC: Cardiovascular Imaging 2021). A CAC in the hundreds moves the risk-benefit conversation for systemic therapy in the other direction — in MESA, participants with a CAC above 100 had a coronary event rate of 20.2 per 1,000 person-years, roughly a 25-fold increase (MESA, Circulation 2016).
The second was a high-sensitivity C-reactive protein. hs-CRP is a downstream marker of systemic inflammation. The categories I use are the ones the AHA and CDC put on the table in 2003 and have not moved since: below 1 mg/L is low cardiovascular risk, 1 to 3 is average, above 3 is high (Pearson et al., Circulation 2003; CDC/AHA workshop follow-up, Circulation 2004). An hs-CRP above 10 mg/L is not read as vascular risk at all until it is repeated 2 weeks later and the person is examined for infection or another acute inflammatory cause (Pearson et al., Circulation 2003). The evidence that hs-CRP identifies a group who benefits from primary prevention comes principally from JUPITER, in which apparently healthy men and women with LDL cholesterol under 130 mg/dL and hs-CRP of 2.0 mg/L or more had a 44 percent reduction in first vascular events on rosuvastatin (Ridker et al., New England Journal of Medicine 2008). hs-CRP does not by itself decide whether hormone therapy is on the table. It changes how carefully I want to watch the tiles in the years ahead.
The third was an ApoB. A standard lipid panel gives you cholesterol concentration — the cargo. Hash marks up the test tube, if you will. Is that test tube filled with particles of sand or particles of gravel? ApoB counts the particles doing the carrying, one apolipoprotein B molecule per atherogenic particle. When cholesterol and particle count agree, either one tells you what you need to know. When they disagree — a normal LDL with a rising ApoB is the classic postmenopausal pattern — the particle count is the number that predicts risk. Her ApoB came back at 102. Once ApoB is on the table, the natural next question is what kind of particles those are. An NMR lipoprofile answers that. NMR gives you the average LDL particle diameter in nanometers, the total number of LDL particles circulating (LDL-P), and — the number I care most about — the count of small dense LDL particles, the subfraction that penetrates the endothelial tiles most easily. For Jennifer, the NMR came back with an average LDL particle size of 21.8 nm — Pattern A — a total LDL-P of 1,279 nmol/L, and a small LDL-P of 283 nmol/L. That is a cleaner picture than her ApoB and her LDL alone would have predicted: the particle count is modest, the particles themselves are the larger, more buoyant kind, and the small dense subfraction — the species that drives the highest per-particle risk — sits at roughly a fifth of her total LDL-P. In prospective cohort data from ARIC, small dense LDL-cholesterol in the highest quartile carried roughly a 50 percent higher risk of coronary heart disease independent of standard risk factors, and predicted risk even in people whose standard LDL-C looked low — the exact scenario a postmenopausal woman with a rising ApoB often finds herself in (Hoogeveen et al., ARIC, Arterioscler Thromb Vasc Biol 2014). LDL particle number by NMR has been shown to add prognostic information beyond LDL-C in the Framingham Offspring cohort as well (Cromwell et al., J Clin Lipidol 2007), and in women specifically in the Women’s Health Study (Blake et al., Circulation 2002). ApoB tells me how big the spackle bucket is. A CAC tells me how much cured spackle is already in the wall (what about uncured plaque? — that is a longer discussion coming in a future week). NMR tells me what is in the bucket.
The fourth was a lipoprotein(a) (a sticky spackle). Lp(a) is a genetically determined atherogenic particle that is fixed for life, largely unmoved by diet, exercise, or standard statin therapy, and a strong independent risk factor for coronary disease when elevated. Because it is inherited, every adult deserves it measured once — and a woman whose mother had a stent and whose father had a bypass deserves it measured this year. Her Lp(a) came back at 34 nmol/L, on the lower end of the normal range. That is one inherited channel of risk we can now stop worrying about.
Jennifer’s CAC came back at 0. Her hs-CRP was 0.7. Her ApoB was 102, with an NMR showing an average LDL particle size of 21.8 nm — Pattern A — a total LDL-P of 1,279 nmol/L, and a small LDL-P of 283 nmol/L. Her Lp(a) was 34 nmol/L. Her home cuff ran 138 over 88 and her LDL was 130, but the pipes themselves had not yet accumulated visible plaque, the inherited channel was quiet, the particles she was carrying were the larger and more buoyant kind, and her inflammatory tone was low. That is a picture in which the near-term risk of a time-limited trial of systemic therapy for symptom treatment is lower than the picture would be in a woman with a CAC of 340, Pattern B particles, an Lp(a) of 180, and an hs-CRP of 4. It is not zero risk. It is a picture in which the balance of benefits and risks moves toward reasonable.
Those 4 numbers are also the numbers we would watch forward whether or not hormone therapy was in the plan. In my practice, when someone in Jennifer’s seat asks about hormone therapy, I am also, in the same visit, asking a bigger question about her pipes.
None of these 4 numbers tell me whether hormone therapy will protect her heart — no test does, because that is not what hormone therapy does. What they tell me is whether the near-term risks of a time-limited trial of systemic therapy for her symptoms are acceptable given her pipes today.
So we made 2 decisions that day, not 1.
The first was about her symptoms. A time-limited trial of systemic therapy was reasonable for symptom treatment. She and her gynecologist would work out the dose, the duration, and the follow-up. Transdermal estradiol is often favored when vascular or VTE risk is a concern because it avoids first-pass hepatic metabolism and may have a more favorable thrombotic profile than the oral form. Oral micronized progesterone is commonly selected because of its tolerability, its endometrial-protection role in a woman with a uterus, and — in a woman sleeping 4 hours — its bedtime sedating effect.
The second was about her prevention plan. That plan did not depend on the first decision. A 2-week home BP log using the protocol I teach in Lower Your Blood Pressure: Help Your Doctor Reduce Your Medication (Cordate Press; Kindle available now, paperback October 20, 2026): validated upper-arm cuff, sit quietly for 5 minutes, take 3 readings in a row and record the lowest, morning and evening. A repeat ApoB at 6 months. A repeat NMR at 12 months. A repeat CAC in 5 years, sooner if the picture changed. Her Lp(a) result went into the chart as a lifetime baseline. A 3-month follow-up on my calendar. Pharmacologic decisions about pressure and lipids deferred to that visit, once the log was in hand.
Hormone therapy was not a substitute for cardiovascular prevention. It never has been. In my practice the visit that gets both decisions right is the visit that treats them as 2 decisions, not 1.
At 3 months her home log ran 122 over 78 — some of that a return of her overall stress load with her sleep restored, some of it what happens when a person starts checking their pressure daily and paying attention. Her sleep had come back to 6 and a half hours. Her hot flashes were down from 6 to 1. Her repeat LDL, at 6 months, was 106 (ApoB 91), a change we interpreted cautiously rather than attributing to one intervention. Her family history and her lipid numbers still required an independent prevention plan. We talked about it at that visit and made the second set of decisions with the second set of numbers.
Her symptoms were better controlled. We had a structured plan to monitor treatment and manage her cardiovascular risk factors. Jennifer was, I told her, going to be watched more carefully than most women her age, because her family had asked us to.
If a woman like Jennifer were in your kitchen tonight, telling you her pressure has climbed 20 points in a year and her sleep is at 4 hours and her gynecologist has offered her a patch and her primary care doctor has said no — there are 3 lines she may be hearing that I want to name for you, because they are the reliable sign that the conversation across the desk has not fully caught up with the current evidence.
You cannot take hormones because your mother had a heart attack. A family history of premature coronary disease is not, by itself, an automatic ban on hormone therapy. It is a reason to assess cardiovascular risk carefully and decide whether systemic therapy is appropriate for symptom treatment. It is also a reason to build a serious prevention plan that would matter whether or not she chose hormone therapy.
The WHI showed hormones cause breast cancer. In WHI, the combined regimen of conjugated equine estrogen plus medroxyprogesterone acetate was associated with an increased breast-cancer incidence; estrogen alone in women with prior hysterectomy had different findings. Risks vary by formulation, progestogen, duration, and patient factors. For the specific WHI combined regimen, the excess was small in absolute terms — often communicated as several additional cases per 10,000 women per year — but it should not be treated as a direct estimate for every modern regimen.
You have to stop at 5 years. There is no universal automatic stop date at 5 years. Duration should be individualized, with periodic reassessment of symptoms, benefits, risks, dose, route, and alternatives.
If you hear any of those 3 lines, none of them is an emergency and none of them is a fight. The move is not to argue harder. The move is to bring the actual clinical question back into the room — given my age, my time since my last period, my cardiovascular risk factors, and my symptom burden, what would a serious individualized conversation look like? — and to ask what your clinician thinks about it. In my practice most of the time the question, once it is in the room, does its own work.
If you are reading this and you are 47 or 51 or 54 and your blood pressure is drifting or your sleep is at 4 hours or your LDL or ApoB has moved for reasons your diet cannot explain, the visit you deserve is the one Jennifer and I built. Two decisions, not one. A symptom-treatment conversation you and your gynecologist, primary care physician, or another clinician comfortable with the current menopause evidence can have. A separate prevention plan for pressure, lipids, and family history that would be the same plan whether or not hormone therapy was on the table. A follow-up on the calendar. A home log. A repeat lipid panel with ApoB. A named next step if it is not working.
I will keep writing about this. Next Tuesday I want to walk you into the WHI itself — what it actually studied, who was in the room, and what the 2 decades since have moved into the shared-decision language of every current major society statement. The week after that, what a real hormone-therapy visit looks like when it is done well. The week after that, we widen the lens, because hormone therapy is 1 of several conversations the window opens, and I want you to know which system moved first for you.
If a friend sent you this and you would like next Tuesday’s post in your inbox, you can subscribe below.
Jennifer is doing well. She still runs. She is still 4 years out from the age at which her father had his bypass. She and I both know that. Her symptoms are better controlled. Her prevention plan is on my calendar. We are watching the numbers together.
Sameer Azhak, MD, FACC is a preventive and integrative cardiologist in Wayne, New Jersey. He writes The Menopause Heart on the cardiovascular half of the menopausal transition. His companion Substack, The Integrative Cardiologist, covers broader preventive cardiology. His books are published by Cordate Press. The blood pressure protocol referenced above comes from his book Lower Your Blood Pressure: Help Your Doctor Reduce Your Medication, available on Amazon Kindle now and in paperback October 20, 2026. This newsletter is educational and is not medical advice for any individual reader.