50-80% Of Alzheimer's Patients Also Have A Condition Called Caa That Most Doctors Don't Check For

TL;DR: 50-80% of Alzheimer's patients also have cerebral amyloid angiopathy (CAA), which is a condition that makes brain blood vessels fragile and prone to hemorrhage. But most routine MRIs don't check for it. If your loved one has a dementia diagnosis, ask whether their MRI included SWI or GRE sequences. If they have CAA, it changes what medications are safe and what treatments are available. Details below.

Disclaimer: I'm not a doctor, I'm just a caregiver sharing research. Everything in this post should be discussed with a qualified neurologist before acting on it.

If you're reading this then you or your loved one probably has a dementia diagnosis. Alzheimer's disease is by far the most common cause of dementia and accounts for about 60-80% of dementia cases.

But Alzheimer's often doesn't come alone...Autopsy studies have found something called cerebral amyloid angiopathy (CAA) in roughly 50-80% of dementia/Alzheimer's cases. Doctors can diagnose probable CAA using the right MRI sequences which are designed to detect tiny areas of bleeding in the brain. These blood-sensitive sequences are not always included in the routine MRI workup for dementia, especially early on. So many people with CAA never know they have it unless the right MRI is ordered.

CAA is a separate progressive neurological disease. Some people only have CAA, and some have both Alzheimer's and CAA. It's caused by amyloid depositing in brain blood vessel walls, rather than forming plaques like in typical Alzheimer's. This condition makes the blood vessels extremely fragile and prone to bleeding, which explains the need for special MRI sequences used to detect blood. And here's the thing that nobody explains clearly enough: Alzheimer's is a slow disease where you lose function over years, but CAA can cause a sudden brain hemorrhage if a weakened vessel ruptures so it’s a much more acute and immediate danger.

On top of that, the current FDA-approved anti-amyloid therapies, lecanemab and donanemab, can cause ARIA (amyloid-related imaging abnormalities), which includes brain swelling and bleeding. By the way - this post is not meant to talk anyone out of those drugs. It is a major breakthrough that they exist, and for many patients they are absolutely worth it. The problem is that CAA can greatly increase hemorrhage risk on these therapies and often makes treatment or trial participation unsafe or impossible. So if you have this condition and are lucky enough to find out before suffering from a massive brain hemorrhage and subsequently getting the right MRI ordered, you're kind of out of luck. You have one or two terrible diseases but can't take the only FDA approved disease modifying drugs available, and 99% of trials slam the door shut in your face because you’re considered too fragile to be a worthwhile participant.

My mom was prescribed lecanemab by a neurologist in our city, but he'd only ordered a standard MRI. We sought a second opinion at an academic medical center out of state, and they required the more sophisticated MRI before prescribing anti-amyloid therapy. That MRI found 9 microhemorrhages and bilateral cortical superficial siderosis. If we hadn't gotten that second opinion, my mom would be on a drug that could have caused a catastrophic brain bleed.

Reasons why I'm writing this:

1. So anyone with a Dementia or Alzheimer's diagnosis will ask whether their MRI includes the proper sequences to detect CAA (GRE or SWI).
2. To share my months of research into CAA safety and interventions, because they do exist.

1. Audit every single medication and supplement for bleeding risk

This is the first thing you should do before adding anything new: Make sure nothing your loved one is already taking is quietly increasing their risk of a brain bleed.

SSRIs are the big one nobody talks about. Prozac, Zoloft, Paxil, Lexapro - these are prescribed constantly for depression in dementia patients, and for good reason. But SSRIs inhibit platelet serotonin reuptake, which means they have antiplatelet effects. My mom was on 60mg of fluoxetine (Prozac) but nobody flagged this as a bleeding concern until we started digging into CAA ourselves, and she's now tapering under her neurologist's supervision. If your family member has CAA and is on an SSRI, raise this with their doctor. It doesn't mean stop immediately, but it's a conversation that needs to happen.

Coming off an SSRI doesn’t mean no more mood support because there are antidepressants that don't carry antiplatelet risk. Bupropion (Wellbutrin) and mirtazapine (Remeron) both work through different mechanisms and don't touch platelet serotonin. Do NOT let anyone just swap in a different SSRI like sertraline or escitalopram because they all carry the same bleeding risk. This is a conversation to have with the prescribing doctor, not something to do on your own.

NSAIDs - ibuprofen, naproxen, Advil, Aleve. People (especially elderly patients) take these constantly for arthritis and general aches but they all have antiplatelet effects. With CAA, every unnecessary antiplatelet exposure is a roll of the dice.

Aspirin - a lot of older adults are on daily low-dose aspirin for heart protection. With CAA the hemorrhage risk may outweigh the cardiac benefit. This needs to be discussed with their doctor.

Blood thinners - warfarin, Eliquis, Xarelto. If your family member is on these for AFib or blood clot history then the risk calculation changes completely with a CAA diagnosis. This doesn't mean they should just stop because some people genuinely need anticoagulation, but the neurologist and cardiologist need to be talking to each other about it. I have my mother wearing a medical alert bracelet now that explains she should NOT be given any blood thinners unless there are no other options.

Supplements people don't think about like high-dose fish oil, vitamin E, and ginkgo biloba all have mild antiplatelet properties. These are things people buy over the counter without a second thought, but with CAA even mild antiplatelet effects matter when the vessels are already fragile.

Statins - controversial, but needs to be discussed. The SPARCL trial subgroup and Biffi et al. (2011) found statin use may be associated with increased lobar hemorrhage risk aka exactly the type CAA causes. If your family member is on a statin and has CAA please raise it with their neurologist.

The point is: go through every single pill bottle in your loved one's medicine cabinet and ask their neurologist "is this safe with CAA?" You might be surprised how many things no one thought to question.

2. Minocycline - the single most important intervention we've found

Minocycline is a generic antibiotic ($10/month) that inhibits MMP-2 and MMP-9, which are the enzymes that chew up the vessel walls in CAA and cause them to rupture. It doesn't clear amyloid but it does stabilize the vessels so they don't bleed.

The data: A retrospective cohort from MGH (the leading CAA research center in the world) looked at 16 patients with aggressive CAA who were started on minocycline. Their rate of brain hemorrhages dropped from 2.18 events per patient per year to 0.46 - an 80% reduction. (Bax et al., 2024, JAHA). A proper randomized trial called BATMAN is now underway to confirm this.

This is the only intervention we've found with direct human CAA-specific data showing hemorrhage reduction. If your family member has CAA then ask their neurologist about this. The paper is from Dr. Viswanathan's group at MGH, which runs the leading CAA research program in the country under Dr. Steven Greenberg.

3. Get the heart right - blood pressure AND heart rate both matter

High blood pressure is the single most evidence-based modifiable risk factor for CAA hemorrhage. If your family member has CAA then their BP needs to be tightly controlled with a target around 130/80. This isn’t aggressively low because that starves the brain of blood flow. but consistently controlled with minimal variability because BP spikes are what rupture fragile vessels. This is standard stroke guideline stuff but it bears repeating because it's the one thing with the strongest evidence behind it.

What most people don't realize is that BP variability can be just as dangerous as high average BP. A single spike from a high-sodium meal, a stressful argument, or straining on the toilet can hit 180, 200+ systolic for a few minutes, and with CAA that might be all it takes to rupture a fragile vessel. The average number on the monitor at the doctor's office doesn't tell the whole story so you need to be tracking at home.

Here's how to do it right: take 10 seated readings over 2-3 weeks at random times of day. Sit for 5 minutes first, feet flat, arm supported. Don't just check in the morning.. get readings after meals, after stress, at night. Log them and bring the log to the doctor. One reading at a clinic visit is almost useless for CAA management because you need the pattern.

On the other end: low heart rate (bradycardia) is a different problem that nobody connects to brain disease. My mom's resting heart rate is in the 40s and for months this was treated as a separate cardiology issue, but it's not. The brain clears amyloid waste through pathways driven by arterial pulsatility aka fewer heartbeats means less clearance force. Low cardiac output also means less blood and oxygen reaching the brain, and the orthostatic dizziness from bradycardia means fall risk, which in a CAA patient means a serious hemorrhage risk.

If your family member has CAA, make sure both their blood pressure AND heart rate are being actively managed. Push for cardiology evaluation if either is off.

4. Exercise - important but watch out

150+ minutes per week of aerobic exercise and 2x/week strength training. This is one of the best-supported interventions for slowing cognitive decline. But for CAA patients there's a critical safety ceiling: no heavy lifting with breath-holding (Valsalva). Heavy straining can cause dangerous acute blood pressure spikes that are asking for a rupture. So keep exercise to moderate intensity, keep breathing, and never strain to failure.

5. Fall prevention - the most dangerous acute event for a CAA patient

For most elderly people, a fall means a broken bone. For a CAA patient, a fall that hits the head can trigger a brain hemorrhage because vessels are already fragile and prone to rupture. This makes fall prevention one of the highest-priority safety interventions for anyone with CAA, and it's one that most neurologists won't bring up because they're focused on the disease and not the home environment.

Ask the PCP or neurologist for a physical therapy referral specifically targeting balance and gait. At home: remove loose rugs, install grab bars in the bathroom, make sure hallways and stairs are well-lit, and address anything causing dizziness whether that's medications (blood pressure drugs, sedatives, antidepressants), orthostatic hypotension, or untreated vision problems. Non-slip footwear matters. If your family member is unsteady on their feet, this is not a "we'll get to it" problem ok? Treat it as urgent.

6. If your loved one has cortical superficial siderosis - ask about vessel wall MRI

This is newer research and most neurologists won't bring it up. Cortical superficial siderosis (cSS) is the strongest predictor of future brain hemorrhage in CAA. My mom has it bilaterally, and for a long time the assumption was that unless you meet full criteria for "CAA-related inflammation" (CAA-ri) then there's no active inflammation to treat.

A 2026 study (Arndt et al., Annals of Clinical and Translational Neurology) looked at 15 CAA patients with cSS using post-contrast vessel wall MRI - another specialized imaging sequence that most standard MRIs don't include. Only 27% of these patients met the formal criteria for CAA-ri, but 93% showed vessel wall enhancement or sulcal hyperintensities near the siderosis sites, meaning almost all of them had active inflammation that standard imaging was missing.

7 of those patients who got follow-up imaging after corticosteroid treatment showed regression of the inflammation, and a separate matched cohort study from the same group found that corticosteroid therapy was associated with longer time before the next hemorrhage (p=0.041) and longer time before any cerebrovascular event (p=0.037).

Caveat: this was a small study, single center, the matched cohort data is retrospective and not yet peer-reviewed, and steroids carry real risks in elderly patients (bone loss, immunosuppression, glucose problems, muscle wasting). But if your loved one has CAA with cortical superficial siderosis, it's worth asking their neurologist whether a vessel wall MRI would be informative. If it shows active inflammation, there might be something treatable that everyone assumed wasn't there.

What's in the pipeline

• Mivelsiran/ALN-APP (Alnylam/Regeneron): This is the only drug in clinical trials right now that was developed specifically for CAA. It's an RNA interference therapy that shuts off amyloid production at the source by silencing the APP gene. The Phase 2 trial (cAPPricorn-1, NCT06393712) just closed enrollment at MGH, NYU Langone, and other sites. If your loved one has CAA then sign up for the CAA Research Recruitment Registry (CAAR3) at angiopathy.org to be contacted for future trials.
• Minocycline BATMAN trial: Randomized controlled trial of minocycline specifically for CAA. If this confirms the cohort data, it could change the standard of care.

What to avoid

• Anti-amyloid antibodies (lecanemab, donanemab, aducanumab) - CAA can substantially increase ARIA-related bleeding risk and often changes eligibility or risk tolerance
• NSAIDs (ibuprofen, naproxen) - antiplatelet/bleeding risk
• Anticoagulants and antiplatelets unless absolutely necessary - discuss risk/benefit with neurologist
• High-dose aspirin
• Heavy resistance exercise with breath-holding

Happy to answer questions.