Reviewed by the Help Dementia Editorial Team — our editors review every article for accuracy against guidance from the National Institute on Aging, the Alzheimer’s Association, and peer-reviewed sources.
Certain Alzheimer’s drugs designed to clear amyloid from the brain—particularly aducanumab (Aduhelm) and lecanemab (Leqembi)—can trigger dangerous side effects visible only on MRI scans. These effects, called amyloid-related imaging abnormalities (ARIA), include brain swelling and microhemorrhages that may cause no symptoms at first but can lead to serious complications if left undetected. Patients taking these medications require regular MRI scans to catch these changes early, before they cause cognitive decline, confusion, or permanent brain damage.
The stakes are high because ARIA doesn’t announce itself through obvious symptoms. A patient might feel fine while their brain shows significant swelling or multiple tiny bleeds on an MRI image. Without monitoring, these changes can progress silently. This is why medical guidelines now mandate MRI screening before starting these drugs and periodic imaging afterward—it’s the only reliable way to know whether the medication is helping clear amyloid without simultaneously damaging brain tissue.
Table of Contents
- What Are Amyloid-Related Imaging Abnormalities and Why Do They Occur?
- The Different Types of Imaging Abnormalities and Their Severity Spectrum
- How MRI Detects These Changes Before Symptoms Appear
- Practical Logistics and Burden of Monitoring Schedules
- Interpreting Results and Managing Incidental Findings
- Which Patients Face Highest Risk of ARIA?
- Real-World Monitoring Scenarios and Stopping Rules
- Frequently Asked Questions
What Are Amyloid-Related Imaging Abnormalities and Why Do They Occur?
ARIA represents the brain’s inflammatory response to aggressive amyloid clearance. When anti-amyloid monoclonal antibodies like lecanemab remove amyloid plaques, the immune system sometimes overreacts, causing edema (swelling) in the brain parenchyma or microhemorrhages in small vessels. These abnormalities appear as bright signals on MRI (T2-weighted and FLAIR sequences for swelling, susceptibility-weighted imaging for microhemorrhages) and are distinct from normal aging changes.
The mechanism reflects a fundamental trade-off: the drug works by triggering immune cells to attack amyloid, but this immune activation occasionally damages surrounding brain tissue. A patient on lecanemab might show ARIA-E (edema) in the temporal or parietal lobes while experiencing only mild headache or no symptoms at all. In contrast, a patient whose drug is stopped before ARIA develops has already missed the window to benefit from amyloid clearance, making early detection crucial for balancing treatment benefit against risk.
The Different Types of Imaging Abnormalities and Their Severity Spectrum
ARIA-E (amyloid-related imaging abnormalities—edema) can range from small, asymptomatic patches to extensive brain swelling visible even to untrained eyes on an mri scan. Some patients show microedema that resolves within weeks without treatment, while others develop symptomatic ARIA-E with headache, confusion, or gait disturbance that requires hospitalization and corticosteroids. The clinical trials for lecanemab reported ARIA-E in roughly 12% of treated patients versus 1-2% in placebo groups, but symptomatic cases requiring medical intervention occurred in fewer than 3% overall.
ARIA-H (microhemorrhages) can be even more concerning because microhemorrhages in critical brain regions—such as the brainstem or deep white matter—carry higher risk of stroke or cognitive worsening than microhemorrhages in the cortex. A patient may have dozens of microhemorrhages on MRI without any acute symptoms, yet the presence of multiple microhemorrhages is a signal to pause or stop the drug. This is why baseline MRI before starting treatment is essential: it documents the patient’s preexisting microhemorrhage burden so new ones can be distinguished from old ones during monitoring.
How MRI Detects These Changes Before Symptoms Appear
Conventional MRI sequences reveal ARIA with high sensitivity. T2 FLAIR sequences are particularly effective at detecting edema because fluid (swelling) appears bright against the darker background of normal brain tissue. A radiologist reviewing the scan can identify edema in the hippocampus, cortex, or white matter and assess its extent. Susceptibility-weighted imaging (SWI) or gradient echo sequences highlight microhemorrhages as dark spots, distinguishing them from artifacts or normal vascular calcification.
The timing of MRI scans matters. Most treatment protocols call for baseline MRI before the first dose, then repeat imaging at 2-4 weeks, again at 8-12 weeks, and periodically thereafter depending on the drug and patient risk factors. For a 72-year-old woman starting lecanemab, a baseline scan shows no edema or microhemorrhages; her 4-week follow-up scan reveals mild edema in the left temporal lobe—asymptomatic, but it signals that her immune response is accelerated. Her neurologist might adjust her infusion schedule or monitor more closely, preventing progression to symptomatic ARIA.
Practical Logistics and Burden of Monitoring Schedules
MRI monitoring for Alzheimer’s drug therapy introduces real logistical demands on patients and caregivers. Patients must travel to imaging centers, lie still in a loud machine for 30-45 minutes, and repeat this process multiple times over weeks or months. For patients with cognitive decline, anxiety, or claustrophobia, even one MRI can be stressful. For a caregiver juggling medical appointments, work, and family responsibilities, scheduling multiple imaging sessions alongside infusion appointments and neurologist visits becomes a genuine burden.
Cost is another limitation, though insurance typically covers MRI scans ordered for medication monitoring. A single brain MRI costs between $1,000 and $3,000 depending on facility and region; monitoring protocols require at least 2-4 scans in the first 3 months. For uninsured or underinsured patients, this financial barrier may delay or prevent access to necessary monitoring, leaving them and their doctors uncertain whether the treatment is safe. Some healthcare systems have addressed this by consolidating MRI appointments or using shorter, streamlined protocols, but access disparities remain.
Interpreting Results and Managing Incidental Findings
MRI scans ordered for ARIA monitoring sometimes reveal unrelated brain abnormalities—old microinfarcts, white matter disease, brain atrophy, or small tumors—that complicate the clinical picture. These incidental findings require follow-up imaging or specialist consultation, adding time and expense. A 68-year-old man undergoing baseline MRI before starting lecanemab may learn that his scan shows a 5-millimeter pituitary lesion unrelated to Alzheimer’s disease, prompting endocrinology referral and additional tests.
Radiologists interpreting ARIA monitoring scans must distinguish new abnormalities from preexisting ones, which is why baseline imaging is critical. Without a baseline, it’s impossible to know whether microhemorrhages shown on the follow-up scan are new (attributable to the drug) or chronic (preexisting). A radiologist comparing two scans must use consistent imaging protocols and carefully note the location and number of lesions. Disagreement between radiologists, or between scans performed at different facilities with different MRI machines, can cause confusion about whether the patient’s ARIA is progressing or stable.
Which Patients Face Highest Risk of ARIA?
Certain patient characteristics increase ARIA risk. Advanced age, presence of the APOE4 genetic variant, baseline microhemorrhage burden, and amyloid positivity (the drug doesn’t work well in amyloid-negative patients anyway, so they’re typically excluded) all correlate with higher ARIA incidence. Additionally, patients on anticoagulation or antiplatelet drugs for heart disease or prior stroke face elevated risk of microhemorrhages during anti-amyloid therapy.
A 78-year-old woman with two copies of APOE4 and three baseline microhemorrhages on MRI enters an anti-amyloid drug trial; her ARIA risk may be 20-30%, higher than a 65-year-old with no APOE4 copies and a clean baseline MRI. Her neurologist must decide whether the potential cognitive benefit justifies the heightened imaging abnormality risk, and intensive monitoring becomes even more important. Informed consent conversations with such patients must explicitly address ARIA risk and the need for frequent MRI surveillance.
Real-World Monitoring Scenarios and Stopping Rules
In clinical practice, neurologists stop or pause anti-amyloid therapy when MRI reveals ARIA that is either symptomatic or rapidly progressive. A patient on lecanemab who develops headache, confusion, or gait problems alongside MRI evidence of edema typically receives a pause in infusions, high-dose corticosteroids, and close neurological observation. Some patients resume the drug at a slower infusion rate after edema resolves; others discontinue it permanently.
The decision depends on symptom severity, the extent of brain changes, and the individual patient’s response to intervention. In clinical trials, roughly 40% of patients who develop symptomatic ARIA required hospitalization, intravenous corticosteroids, and intensive monitoring. Once ARIA resolves (which typically takes weeks to months), some patients have resumed anti-amyloid therapy without recurrence, while others have not. The uncertainty is part of why monitoring is so critical—it provides the data needed to make real-time treatment adjustments and protect patients from preventable harm.
- —
Frequently Asked Questions
How often do patients need MRI scans while on anti-amyloid drugs?
Most protocols require baseline imaging before starting treatment, then follow-up scans at 2-4 weeks, 8-12 weeks, and periodically thereafter. Patients at higher risk or those who develop early signs may need more frequent imaging.
Can patients with pacemakers or metal implants get MRI scans?
Traditional MRI is contraindicated with most pacemakers and some metal implants. Patients with such devices may be ineligible for anti-amyloid therapy, or alternative imaging (such as CT with specialized protocols) may be used with neurologist approval, though MRI remains the gold standard.
What happens if ARIA-edema is found on an MRI scan?
Depending on whether the patient has symptoms, the neurologist may pause infusions, prescribe corticosteroids, and repeat imaging in 1-2 weeks. Asymptomatic ARIA may be monitored closely without intervention; symptomatic ARIA usually requires treatment and possible discontinuation of the drug.
Are there drugs that can prevent ARIA from developing?
No preventive medication reliably stops ARIA. Slower infusion rates and careful patient selection based on age and genetic risk factors are the main strategies to reduce incidence.
Can ARIA cause permanent brain damage?
Most ARIA resolves with medical management, and long-term cognitive consequences of reversible ARIA are not yet fully understood. However, severe or untreated ARIA carries risk of lasting cognitive decline or stroke.
What if a patient has already had a stroke—can they still take anti-amyloid drugs?
Patients with prior stroke carry elevated ARIA risk, particularly ARIA-H (microhemorrhages). Close monitoring and shared decision-making with the neurologist are essential; some such patients are excluded from therapy.
