Does Anesthesia Affect or Worsen Dementia?

The honest answer is that decades of research have not produced a clear verdict. Some large population studies find that people who undergo anesthesia and surgery develop dementia at higher rates, while other well-designed studies find no increased risk at all. The confusion stems partly from the difficulty of separating what anesthesia does to the brain from what surgery, hospitalization, and the underlying illness do. What is increasingly clear is that short-term cognitive problems after surgery are real and common, especially in older adults, and that certain perioperative factors can raise or lower the chances of those problems becoming lasting.

What Actually Happens to Thinking After Surgery

Two distinct conditions muddy the conversation. The first is postoperative delirium, an acute episode of confusion, disorientation, and fluctuating awareness that typically appears within hours to days after surgery. It is diagnosed through bedside observation and tends to resolve, though it can last weeks in severe cases. The second is postoperative cognitive dysfunction, a subtler decline in memory, attention, or processing speed that shows up only on formal testing and can persist for weeks or months. These two conditions share some risk factors but differ in how they look, how they are detected, and likely how they arise.

Both are frequently lumped together in media coverage under the umbrella of “anesthesia causing brain problems,” which makes the picture seem simpler than it is. A review in Progress in Neuro-Psychopharmacology and Biological Psychiatry noted the serious concern that either postoperative delirium or cognitive dysfunction could lead to dementia, while emphasizing that they are distinct syndromes requiring different diagnostic approaches.¹ A separate review in Minerva Anestesiologica described them as sometimes reported on the same continuum but unlikely to share the same underlying biology, with delirium being acute in onset and postoperative cognitive dysfunction being subtler and longer lasting.²

Postoperative delirium, in particular, is strikingly common among older surgical patients. A large analysis of over 5.5 million inpatient admissions for major noncardiac surgery found that roughly 3.6% of patients developed postoperative delirium. Those who did were nearly three times more likely to die within 30 days and about four times more likely to be discharged somewhere other than home.³ So even if the link to long-term dementia remains debated, these short-term cognitive disturbances carry real consequences of their own.

What Population Studies Actually Show

The epidemiological evidence is genuinely contradictory, and researchers have spent years trying to reconcile it. A Taiwanese population-based study found that patients who underwent anesthesia and surgery had roughly double the risk of later dementia compared with controls, with a hazard ratio of about 1.99. The elevated risk held across different anesthesia types, including general, regional, and intravenous.⁴ A separate nationwide cohort study found that exposure to general anesthesia was associated with roughly a 29% increase in dementia risk after adjusting for age, sex, healthcare visits, and other illnesses, and that more exposures and longer cumulative anesthesia time pushed the risk higher.⁵

But studies like these face a fundamental problem: people who have surgery are not identical to people who do not. They tend to be sicker, more likely to have cardiovascular disease or diabetes, and more likely to experience the kind of physiological stress that itself could promote cognitive decline. A prospective study in the Journal of the American Geriatrics Society tackled this head-on. It followed older adults over time and found that high-risk surgery with general anesthesia was not associated with a greater risk of dementia or Alzheimer’s disease. People who had other types of surgery with general anesthesia actually showed a lower risk of dementia than those with no anesthesia history at all.⁶ That finding does not prove anesthesia is protective. It more likely reflects the “healthy user” effect: people well enough to undergo elective surgery and recover from it may be healthier overall than those who never have surgery.

The Twin Study That Helps Untangle Cause and Effect

One of the most clever approaches to this question came from a Danish twin study. Researchers identified pairs of identical and fraternal twins where one twin had undergone major surgery and the other had not, then compared their cognitive scores. If anesthesia truly damaged the brain, you would expect the surgery-exposed twin to consistently score lower. Instead, the surgery-exposed twin had the lower cognitive score in exactly 49% of pairs, which is effectively a coin flip. The result held whether the twins were identical (sharing all their genes) or fraternal.⁷

This matters because identical twins share virtually all genetic and early-life environmental factors. If you strip away those confounders, the apparent link between surgery and worse cognition disappears. The implication is that a lot of what looks like anesthesia-related cognitive decline in population studies may actually be explained by the pre-existing characteristics of people who end up needing surgery. Their genetics, their chronic diseases, and their baseline brain health were already pointing toward cognitive decline before they ever entered an operating room.

Does the Type of Anesthesia Matter?

Even if the broad question of “does anesthesia cause dementia” remains unresolved, a more nuanced question has gotten traction: does the specific drug or technique matter? A large propensity-matched study of older adults with hip fractures compared three approaches and found meaningful differences. Patients who received inhalational general anesthesia (breathing in anesthetic gases) had higher rates of subsequent dementia than those who received total intravenous general anesthesia, who in turn had higher rates than those who received regional anesthesia like a spinal block.⁸

A meta-analysis comparing sevoflurane (a common inhaled anesthetic) with propofol (the standard intravenous agent) found that overall rates of postoperative cognitive dysfunction did not differ significantly between the two. However, in the first several hours after surgery, patients who received sevoflurane scored lower on cognitive screening tests. By one to three days out, those differences either vanished or even reversed.⁹ A separate meta-analysis of elderly noncardiac surgical patients found that on postoperative days two through six, propofol anesthesia was associated with a lower incidence of cognitive dysfunction and lower levels of inflammatory markers compared with inhaled anesthesia.¹⁰

A randomized trial directly comparing isoflurane and propofol told a somewhat different story: there was no significant difference in memory, executive function, or delirium rates between the two groups. The strongest predictor of postoperative cognitive decline in that study was not the anesthetic used but the patient’s pre-surgical cognitive function.¹¹ So the picture is still blurry. There are hints that inhaled agents may be somewhat harder on the brain in the short term, but the evidence is not strong enough to make broad clinical recommendations based solely on anesthetic type.

How Anesthetics Could Theoretically Harm the Brain

Laboratory research has identified several plausible mechanisms by which anesthetics might promote the kind of brain changes seen in Alzheimer’s disease. Common inhaled agents like isoflurane, sevoflurane, and desflurane have been shown in cell and animal studies to increase production and accumulation of beta-amyloid protein and to trigger hyperphosphorylation of tau proteins, both hallmarks of Alzheimer’s pathology.¹² Separate work showed that halothane and isoflurane reduce the concentration of amyloid-beta needed to start forming toxic clumps, essentially lowering the threshold for the protein to begin aggregating.¹³

Another pathway involves the blood-brain barrier, the tightly regulated boundary that normally keeps inflammatory molecules in the bloodstream from entering the brain. Animal research has shown that the combination of anesthesia and surgical stress activates an enzyme that damages this barrier, allowing peripheral inflammation to flood into the central nervous system and trigger neuroinflammation, synaptic dysfunction, and delirium-like behavior.¹⁴ Evidence also suggests that blood-brain barrier dysfunction and neuroinflammation together play a role in cognitive complications after anesthesia.¹⁵

The catch is that virtually all of this mechanistic work comes from cell cultures or animal models, often using anesthetic doses and durations that exceed what humans typically receive. Whether these effects translate meaningfully to a 70-year-old having a two-hour knee replacement under carefully monitored anesthesia is an open question. The lab findings tell us anesthetics can do worrying things to neurons in a dish. They do not tell us those things happen at clinically relevant levels in a living human brain.

Pre-Existing Cognitive Problems Change the Equation

One pattern that does emerge consistently is that people who already have some degree of cognitive impairment before surgery are at substantially higher risk for trouble afterward. A study examining elderly surgical patients found that those with pre-existing mild cognitive impairment or dementia experienced postoperative delirium at more than three times the rate of cognitively normal patients. But the finding ran in the other direction too: elderly patients who were cognitively normal before surgery but then developed postoperative delirium were more likely to later be diagnosed with mild cognitive impairment or dementia.¹⁶

This creates a chicken-and-egg problem that haunts the entire field. Did the delirium unmask a disease process that was already quietly developing? Or did the delirium itself, through inflammation and metabolic disruption, push the brain across a threshold into persistent decline? Researchers suspect both things happen in different patients, which is part of why the population-level statistics are so hard to interpret. An expert consensus review noted that clinicians often fail to screen patients for pre-existing cognitive impairment before surgery, meaning many people go into the operating room with unrecognized vulnerabilities.¹⁷

Anesthesia Depth and Brain Monitoring

How deeply unconscious you are during surgery may matter more than the specific drug used. When anesthesia is pushed too deep, the brain’s electrical activity can fall into a pattern called burst suppression, where periods of complete electrical silence alternate with short bursts of activity. Research suggests that time spent in burst suppression and the pattern of brain activity during emergence from anesthesia may predict postoperative delirium, though studies have produced conflicting results about how strong the connection is.¹⁸

Intraoperative hypotension, where blood pressure drops significantly during surgery, has also drawn attention as a potential contributor to brain injury. A retrospective study of older high-risk surgical patients found an association between the duration of low blood pressure episodes and postoperative delirium, though the relationship weakened and lost statistical significance after additional adjustments were made.¹⁹ The practical takeaway is that maintaining appropriate blood pressure and avoiding unnecessarily deep anesthesia are reasonable goals, even if the evidence linking these factors to long-term dementia specifically remains incomplete.

Reducing Cognitive Risk Around Surgery

Because you cannot always avoid surgery, the more actionable question for most people is what can be done to protect the brain during the perioperative period. Several strategies have accumulated enough evidence to be taken seriously.

Dexmedetomidine, a sedative that works differently from traditional anesthetics, has shown promise. A systematic review found that its use in adults over 65 undergoing noncardiac surgery could prevent postoperative delirium, partly by reducing inflammatory markers and partly by decreasing the toxic effects of other anesthetic drugs.²⁰ A meta-analysis of randomized controlled trials in patients over 60 undergoing regional anesthesia found that intravenous dexmedetomidine during surgery cut the risk of postoperative delirium by about half and the risk of postoperative cognitive dysfunction by roughly 65%.²¹

Non-drug approaches also appear effective. A network meta-analysis of non-pharmacological interventions for preventing postoperative delirium in hospitalized older adults found that multicomponent programs, combining elements like early mobility, sleep hygiene, reorientation, and cognitive engagement, were associated with significantly lower odds of developing delirium compared with standard care alone.²² Simple things like making sure you have your glasses and hearing aids, keeping a normal sleep-wake cycle, staying hydrated, managing pain without over-relying on opioids, and having familiar people around can all contribute.

An international expert consensus under the American Society of Anesthesiologists’ Perioperative Brain Health Initiative emphasized six practical actions: educating patients and families about cognitive risk, screening for cognitive impairment before surgery, implementing non-pharmacological delirium prevention, optimizing pain control, using brain monitoring during anesthesia, and avoiding antipsychotic medications for delirium management.¹⁷ Preoperative cognitive screening tools exist for this purpose, and their use is growing, though it remains inconsistent across hospitals.²³

What About Children and Anesthesia?

The concern about anesthesia and brain development in young children runs on a separate track from the geriatric research, though it often gets folded into the same public worry. Animal studies have consistently shown that exposure to common anesthetics during critical windows of brain development can trigger widespread neuronal cell death and long-term learning and memory impairment. This has been demonstrated across species from rodents to primates, with both inhaled agents and intravenous drugs like propofol and ketamine.²⁴

The animal data prompted the U.S. Food and Drug Administration in 2016 to add a warning to anesthetic labels about potential effects on brain development in children under three. Human studies, however, have been more reassuring. Large observational studies and one major randomized trial have generally not found clinically meaningful differences in neurodevelopmental outcomes between children who had brief, single exposures to anesthesia and those who did not. The concern has not been dismissed, but the consensus has settled into a cautious position: brief, necessary procedures should not be delayed out of fear of anesthesia, while elective procedures in very young children should be considered carefully. This is a genuinely different biological question from the one in older adults, because the developing brain and the aging brain are vulnerable in different ways and for different reasons.

When the Evidence Is This Mixed, What Should You Do?

If you or a family member is facing surgery and worrying about cognitive effects, the evidence supports a few practical steps. Ask your surgeon and anesthesiologist about cognitive screening before the procedure, especially if the patient is over 65 or has noticed memory changes. Discuss whether regional anesthesia is an option, since the data, while not conclusive, consistently trends toward lower cognitive complication rates with regional techniques. Mention any concerns about brain health so the anesthesia team can consider strategies like brain-activity monitoring and appropriate drug selection. After surgery, prioritize the basics that reduce delirium risk: adequate sleep, pain management, hydration, familiar surroundings, and early mobility.

What the evidence does not support is canceling or indefinitely postponing necessary surgery because of dementia fears. The medical conditions that require surgery carry their own cognitive risks through pain, immobility, infection, and systemic inflammation. The cleanest take on the data is this: surgery and anesthesia together impose a transient stress on the brain that most people recover from fully, that some people (especially those with pre-existing cognitive vulnerability) recover from incompletely, and that may accelerate a disease process already underway in a subset of patients. Whether anesthesia itself, divorced from the surgery and the illness that led to it, independently causes dementia is something the current evidence still cannot confirm or rule out.