Mitochondrial Dysfunction: An Underlying Driver of Autism and What May Help

Apr 18 2026

(Illustration by Lumi Liu)

For decades, autism has been understood as a condition of brain wiring. However, a growing body of research points to a more fundamental problem: The brain may simply not be getting enough energy to develop the way it should.

Up to 80 percent of children with autism show signs of mitochondrial dysfunction.

In mitochondrial dysfunction, the mitochondria—the power plants inside cells—cannot produce enough energy for the body, which can create significant problems for a child’s development.

The link between mitochondrial dysfunction and autism was first proposed in 1985, when clinicians reported seeing four autism patients who also had lactic acidosis—a buildup of lactate in the body which occurs when it cannot efficiently create energy—suggesting that mitochondrial health may have been impaired.

Since 1985, bigger and more elaborate studies have only shown that this problem may be more prevalent in autism than previously thought.

The Peculiar Role

The brain expends 20 percent of the body’s energy, making it uniquely vulnerable when mitochondrial activity is impaired. Animal studies have shown that mice with dysfunctional mitochondria develop autism-like symptoms, with or without apparent brain changes.

It is still unclear if mitochondrial dysfunction is a cause or a feature of autism. Many of the genetic and environmental factors that contribute to autism also impair mitochondrial health.

Genetically, genes that drive mitochondrial dysfunction have been directly linked to autism. At the environmental level, exposure to early antibiotics, infections, air pollution, nutritional deficiencies, and autoimmune diseases, all of which can raise the risk of developing both conditions, according to neurologist Richard Frye.

“The mitochondria [are] very, very sensitive to environmental changes,” Frye told The Epoch Times. His research suggests that some of this stress can start prenatally and remain after birth.

Children with autism also often have higher levels of oxidative stress. Glutathione, which neutralizes oxidative stress, needs energy from the mitochondria to be made.

“So if the mitochondria [aren’t] working, you’re going to make less glutathione and not be able to control that oxidative stress,” Frye said.

Inflammation is another common feature of autism, and mitochondrial dysfunction can drive that too. Anti-inflammatory cells require mitochondria to be activated, whereas inflammatory cells don’t, Frye said.

The Prevalence

Among children with autism, about 5 percent have classic mitochondrial disease, caused by genetic changes.

Most children with autism, however, have mitochondrial dysfunction not traceable to a gene.

Children who have autism and mitochondrial dysfunction often show mitochondrial-related symptoms such as fatigue, developmental delay, muscle weakness, and nerve pains.

One unique factor is that mitochondrial dysfunction is often linked to regression-type autism, where children appear to be developing typically, but in early childhood start regressing and lose their skills.

This is in line with medical understanding of mitochondrial dysfunction—most patients may not show symptoms for years before they suddenly manifest.

The most comprehensive research comes from Robert Burrier and colleagues who analyzed blood samples from more than 1,000 young children, comparing metabolic markers in those with autism to those without.

Their study, also known as the Children’s Autism Metabolome Project, identified seven different metabolic profiles in children with autism, all of which showed abnormal pathways that may be implicated in mitochondrial dysfunction.

Treatment of Mitochondrial Dysfunction

Supplements are the most common treatments for mitochondrial dysfunction in part because the condition affects every person differently, and supplements provide a broader treatment when targeted drugs are often unavailable.

There’s no magic pill. “Even for genetically confirmed mitochondrial disease, we only have two FDA treatments, so we have very little definitive treatment for these patients.” Dr. Frances Kendall, a geneticist, told The Epoch Times.

A common treatment is a mitochondrial cocktail. The contents vary depending on a child’s individual needs, but common ingredients include CoQ10, levocarnitine, and alpha-lipoic acid. All three of these compounds are naturally produced in the body and play important roles in normal mitochondrial function.

One study administered these three supplements and found that children with autism generally improved in at least one metabolic health marker. Some also improved their behavioral scores.

Clinical trials of levocarnitine, an amino acid derivative used to treat carnitine deficiency, and CoQ10 individually have shown improvements in behavior and communication. A single case study showed that levocarnitine supplementation stopped regression in a 4-year-old boy.

Other nutrients included in this cocktail include B-group vitamins. B1, B2, and B3 are directly involved in energy production in mitochondria, while B6 helps prevent oxidative stress and supports energy production. Clinical trials in which vitamin B6 was supplemented with magnesium showed improvements in behavior and language.

Children with autism often have low levels of B9 and B12. These two nutrients help maintain mitochondrial DNA health and support the production of more mitochondria.

Creatine, which improves muscle function and cellular energy, is sometimes also included.

Ketogenic diets have shown promise in clinical trials for improving mitochondrial function, and exercise may encourage the production of more mitochondria, which can be helpful for some individuals, Kendall said.

[series_posts_list][/series_posts_list]

No Standard Diagnosis

While established clinical practices exist to diagnose mitochondrial dysfunction, they are not routinely used in clinics to determine whether it is playing a role in a child’s autism.

Clinical research into autism diagnostic biomarkers is both expensive and time-consuming.

Burrier, who holds a PhD in biochemistry and is the principal investigator of the Children’s Autism Metabolome Project, said the group’s research identified as many as 153 different metabolic biomarkers that may be involved in autism.

However, this research needs to be verified, perhaps one group of metabolites at a time, to make sure that their findings can be applied to both the diagnosis and therapeutic intervention.

“It’s a little unfortunate. Biomarker research is currently great academic research. However, from a commercial perspective, diagnostics have little value if there are no targeted therapeutics. Much of the current thinking on autism therapeutics also turns to commonly available supplements, which likewise have little commercial value,” Burrier told The Epoch Times.

Though mitochondrial dysfunction is a fairly niche area of research, Frye hopes there will be growing interest in the future for better treatments for children and families.

Marina Zhang

Marina Zhang is a health reporter for The Epoch Times. She covers both health news and in-depth features on emerging health issues. Marina holds a bachelor's degree in biomedicine from the University of Melbourne. Contact her at marina.zhang@epochtimes.com.