Aripiprazole

Limits on what this shows

• Key negative/null findings: None specific to mitochondrial structure. A related 2024 study in the same Fao liver-cell line (see Oxidative stress below) found aripiprazole reduced, not increased, stress-kinase activation and improved cell survival under an added oxidative challenge, which complicates any simple "mitochondrial harm" narrative and is logged as a genuine counter-signal rather than omitted.
• At-risk subgroups: None established in humans. Shaju et al. 2025 (BMJ Open, PMID 41093357) describes a planned randomized controlled trial protocol to evaluate atypical antipsychotic-induced mitochondrial dysfunction in schizophrenia patients; this is a protocol only, with no results yet, and is flagged as ongoing research rather than used as a data source.
• Counter-evidence on record: No study directly falsifying the Complex I finding was located. The major limitation on record across all four sources is that every measurement is preclinical (Drosophila, embryonic mouse neurons, rat hepatoma cell line), and dose/concentration translation to sustained human therapeutic exposure is not established by any of them.

Regulatory status

Not a labeled effect; the FDA label does not mention mitochondrial function.

Population evidence

Hardy et al. 2023 (Biol Direct, PMID 37528429, DOI 10.1186/s13062-023-00375-9): aripiprazole, brexpiprazole, and cariprazine directly inhibit Complex I at the ubiquinone-binding channel at clinically relevant concentrations, with greater bioenergetic impairment in ventral-midbrain than forebrain embryonic mouse neurons, and chronic exposure in adult Drosophila caused mitochondrial structural damage and locomotor dysfunction. Pirc Marolt et al. 2023 (Antioxidants, PMID 38001783, DOI 10.3390/antiox12111930): a therapeutic-range 6 μM aripiprazole concentration produced mitochondrial hyperpolarisation within 5 minutes, sustained through long-term exposure, in rat Fao hepatoma cells. Hurcomb et al. 2025 (Dis Model Mech, PMID 40126029, DOI 10.1242/dmm.052180): oral aripiprazole reduced mitochondrial membrane potential and activated the JNK apoptosis pathway in Drosophila intestinal epithelial cells, impairing gut function and faecal output; the effect was suppressed by antioxidants. Panizzutti et al. 2023/2024 (Eur Arch Psychiatry Clin Neurosci, PMID 38072867, DOI 10.1007/s00406-023-01727-2) included aripiprazole among antipsychotics studied for cellular energy-metabolism effects; this review confirmed aripiprazole was in the studied panel but could not retrieve the full abstract's quantitative results (see access note).

Mechanism

Direct off-target inhibition of respiratory Complex I at the ubiquinone-binding channel (named, Tier B mechanistic, Hardy et al. 2023), plus a separate, not-yet-mechanistically-resolved hyperpolarisation and ROS signal in liver and gut epithelial cells.

Framing caution

This is laboratory and animal mechanistic evidence from four independent research efforts, not a clinical finding in aripiprazole-treated people. It should not be presented to consumers as an established mitochondrial-toxicity risk, and it should not be merged with the Oxidative stress or ATP production entries below even though it shares sources with both; see the overlap map.

Limits on what this shows

• Key negative/null findings: Kramar et al. 2022 and 2024 are themselves the negative/protective finding: aripiprazole did not simply worsen oxidative markers under an added stressor in the liver-cell system, and in the 2024 head-to-head it behaved differently from, and less harmfully than, olanzapine on several measures.
• At-risk subgroups: Not established in humans; all evidence is cell or Drosophila models.
• Counter-evidence on record: The bidirectional finding within Kramar et al. 2024 itself (protective on MAPK/survival, harmful on apoptosis/superoxide/mitophagy under the same stress test) is recorded here explicitly rather than resolved into a single direction, consistent with this protocol's rule against forcing a real, mixed signal into a cleaner story than the data support.

Regulatory status

Not a labeled effect in either direction.

Population evidence

Pirc Marolt et al. 2023 (PMID 38001783): superoxide increased significantly at all timepoints in Fao cells exposed to 6 μM aripiprazole; hydrogen peroxide production was significantly higher specifically after long-term exposure; glutathione showed a downward trend. Hurcomb et al. 2025 (PMID 40126029): ROS elevated in Drosophila intestinal cells, suppressed by antioxidant co-treatment. Against that, Kramar et al. 2022 (Int J Mol Sci, PMID 35955425), "Antipsychotic Drug Aripiprazole Protects Liver Cells from Oxidative Stress," found aripiprazole protective against an externally applied oxidative challenge in the same Fao cell system. Kramar et al. 2024 (Int J Mol Sci, PMID 39456900, DOI 10.3390/ijms252011119) directly compared aripiprazole (framed as "metabolically neutral") against olanzapine (framed as "obesogenic") under hydrogen-peroxide challenge: aripiprazole-treated cells showed reduced MAPK and STAT3 activation and better survival than olanzapine-treated cells, but also higher apoptotic rates, increased superoxide production, and enhanced mitophagy under that same stress condition, an explicitly bidirectional result within one paper.

Mechanism

Not a single named biochemical mechanism; proposed via direct molecular effects on glutathione availability and on MAPK/STAT3 stress-signaling pathways (Tier B, Kramar et al. 2024).

Framing caution

This is a mixed preclinical signal, not a settled harmful or protective effect. It must not be presented to consumers as either a clear antioxidant-depletion risk or a reassurance that aripiprazole is antioxidant-protective; no human oxidative-stress biomarker data in aripiprazole-treated patients was located.

Limits on what this shows

• Key negative/null findings: Risperidone, the third antipsychotic in the same cohort, did not show the same copper decrease, which argues against a simple pooled antipsychotic-class artifact and supports a drug-specific, or at least not universal, signal. No separate copper effect size for aripiprazole alone could be confirmed, so the possibility that the pooled clozapine-and-aripiprazole result is driven mainly by clozapine cannot be excluded.
• At-risk subgroups: None established beyond the general schizophrenia inpatient cohort studied. Separately, aripiprazole is sometimes used to treat the neuropsychiatric presentation of Wilson's disease, a genetic copper-overload disorder (two case reports located, PMID 41531725 and PMID 27490186); these describe aripiprazole as a treatment for psychosis arising from a pre-existing copper-handling disorder, not aripiprazole causing a copper effect, and are noted here only as differential-diagnosis context, not as depletion evidence.
• Counter-evidence on record: None located that directly contradicts the direction of the Chen et al. finding. An earlier pass of this review, working only through the Crossref API under heavy rate-limiting elsewhere, had returned a clean negative result for one narrow phrasing, "aripiprazole copper ceruloplasmin", which this broader re-search supersedes; that narrower phrasing simply had not captured the trace-element cohort study located this session.

Regulatory status

Not a labeled effect; copper is not mentioned in the FDA label sections reviewed for aripiprazole.

Population evidence

Chen et al. 2018 (Biol Trace Elem Res, PMID 28470477, DOI 10.1007/s12011-017-1039-6): a case-control study of 165 schizophrenia inpatients versus 614 healthy controls measured serum magnesium, copper, calcium, phosphorus, iron, and zinc, then followed the patient group prospectively (average 3.8 weeks) through antipsychotic treatment with risperidone, clozapine, or aripiprazole, using repeated-measures ANOVA. Serum copper was significantly decreased after clozapine and aripiprazole treatment specifically (P < 0.05); phosphorus rose after all three drugs including aripiprazole, and zinc fell in several patient subgroups. The abstract and results report the copper finding jointly for clozapine and aripiprazole rather than as two separately quantified effect sizes, and no numeric mean serum copper value or percent change specific to the aripiprazole subgroup alone was retrievable this session.

Mechanism

Not established by the authors. No named biochemical pathway links aripiprazole to copper handling; the finding is an observed association in treated patients, not a demonstrated chelation, absorption, or excretion mechanism.

Framing caution

This is a single human study with a real, statistically significant, drug-named finding, not two independent confirmations, and the effect size for aripiprazole alone is not separately quantified in the abstract. It should be presented as a real but bounded signal worth clinical awareness, not as an established, quantified depletion magnitude, and not as grounds for a specific supplementation dose.

Limits on what this shows

• Key negative/null findings: Not retrievable this session; exact effect sizes and any null sub-findings within the five class-level papers could not be confirmed because the databases that would host their full text were rate-limited throughout this review.
• At-risk subgroups: Patients with baseline iron-deficient states starting aripiprazole, particularly at the highest-akathisia-risk dosing context identified in the label (adjunctive use in major depressive disorder, 25% incidence).
• Counter-evidence on record: None located; the direction-of-effect check required by this protocol was applied explicitly, this is a risk-factor-for-an-outcome association (low iron may increase susceptibility to a drug side effect), not evidence that aripiprazole itself lowers iron status, and it is written up as such rather than as a depletion claim.

Regulatory status

Akathisia frequencies are FDA-labeled (via RxList's transcription of the Abilify prescribing information): 8% on aripiprazole versus 4% on placebo in adult schizophrenia trials; 13% versus 4% in adult bipolar-mania monotherapy trials; 25% versus 4% in major-depressive-disorder adjunctive-therapy trials; and, in pediatric bipolar-mania trials (ages 10 to 17), 8.2% (10 mg) and 11.1% (30 mg) versus 2.1% on placebo. No iron or ferritin finding is itself in the aripiprazole label.

Population evidence

Pall, Williams, and Blake 1986 (Lancet, DOI 10.1016/s0140-6736(86)92790-x), a one-page letter, first proposed the iron-akathisia link. Subsequent studies corroborated it in typical/early-generation antipsychotic cohorts: Horiguchi 1991 (Acta Psychiatr Scand, DOI 10.1111/j.1600-0447.1991.tb03148.x), "Low serum iron in patients with neuroleptic-induced akathisia and dystonia"; Soni et al. 1993 (Br J Psychiatry, DOI 10.1192/bjp.163.5.669), "Serum Iron Abnormalities in Neuroleptic-Induced Akathisia in Schizophrenic Patients"; Jo et al. 1997 (Ewha Med J, DOI 10.12771/emj.1997.20.2.239); and Hofmann et al. 2000 (Psychiatry Research, DOI 10.1016/s0165-1781(00)00115-3, confirmed via bibliographic record as a real, indexed, 37-citation paper). All of these predate aripiprazole's 2002 US approval and studied typical or early second-generation neuroleptics, not aripiprazole; this is Tier C class extrapolation. Full abstract text (exact effect sizes) could not be retrieved for any of these five sources this session (see access note); only bibliographic metadata (title, journal, year, and for Hofmann et al., citation count) was confirmed.

Mechanism

Proposed role for iron as a cofactor in central dopaminergic pathways implicated in akathisia; not confirmed for aripiprazole specifically, and not established as a depletion mechanism at all, only as a susceptibility factor.

Framing caution

This is not a claim that aripiprazole depletes iron and must never be presented as a reason to supplement iron preemptively. Serum iron and ferritin studies are a reasonable clinical work-up if treatment-emergent akathisia occurs, not a routine or preventive recommendation, and any iron repletion should follow a confirmed deficiency, given the real overload risk of empiric supplementation.

Limits on what this shows

• Key negative/null findings: Cussotto et al. found that despite significant taxonomic shifts, aripiprazole's effect on ileal permeability was not causally attributable to the microbiome change, arguing against a simple composition-drives-permeability narrative. Leão et al.'s short-chain-fatty-acid findings for aripiprazole did not reach statistical significance despite the diversity change. The two studies that measured diversity, Cussotto in rats and Leão in vitro, both found aripiprazole increased it, while Ait Chait's ex vivo human-colon model found aripiprazole decreased specific beneficial phyla and increased Proteobacteria, a pattern often associated with dysbiosis in other contexts; this direction conflict is recorded here rather than resolved into one narrative.
• At-risk subgroups: None established in humans. Leão et al.'s model specifically probed sex-hormone interactions and reported aripiprazole's effect only in the female-derived fermentation model, so whether the same direction holds in a male-derived model was not confirmed by that paper.
• Counter-evidence on record: No human clinical cohort measuring gut microbiome composition specifically in aripiprazole-treated patients was located. One 2026 human study, Zhang et al. (J Affect Disord 414, DOI 10.1016/j.jad.2026.122389, PMID 42604694), a prospective follow-up of gut microbiota and metabolic changes during antipsychotic-lithium treatment in acute bipolar mania, lists aripiprazole among the therapies used in its treated cohort by indexed subject heading, but its full abstract and quantitative results were not retrievable this session; it is logged here as a confirmed, real, human study worth a dedicated follow-up pass, not as supporting or contradicting evidence for either direction.

Regulatory status

Not a labeled effect; the FDA label does not address gut microbiome composition.

Population evidence

Cussotto et al. 2019 (Psychopharmacology 236(5):1671-1685, DOI 10.1007/s00213-018-5006-5): chronic oral aripiprazole administration in rats significantly increased microbial species richness and diversity, and increased the genera Clostridium, Peptoclostridium, and Intestinibacter and the family Christensenellaceae, versus untreated controls; aripiprazole-treated animals also showed increased ileal permeability, though the authors found the microbiome-composition and permeability changes were not causally linked. Ait Chait et al. 2023 (Int J Mol Sci 24(8):7326, DOI 10.3390/ijms24087326): in a 48-hour ex vivo fermentation model of the human colon using donor fecal samples, aripiprazole significantly decreased the phyla Firmicutes and Actinobacteria while increasing Proteobacteria, reduced the families Lachnospiraceae, Lactobacillaceae, and Erysipelotrichaceae, and lowered the short-chain fatty acids butyrate, propionate, and acetate; a probiotic combination (Lacticaseibacillus rhamnosus HA-114 and Bifidobacterium longum R0175) largely reversed these changes. Leão et al. 2025 (Microbiology Spectrum, DOI 10.1128/spectrum.02350-25), using an in vitro colonic-fermentation model adapted from the PolyFermS system with donor fecal samples from both sexes, found aripiprazole treatment increased microbial diversity and altered short-chain-fatty-acid profiles in the female-derived model, though the SCFA changes did not reach statistical significance (P > 0.05).

Mechanism

Not resolved to a single named pathway across the three papers. Proposed routes include direct antimicrobial or growth-modulating activity of aripiprazole on specific bacterial taxa (Ait Chait et al.) and a broader ecological shift in richness and diversity following chronic systemic exposure in vivo (Cussotto et al.); none of the three isolate a specific molecular mechanism.

Framing caution

This is preclinical and ex vivo mechanistic evidence from three independent groups, not a human clinical dysbiosis finding, and the direction of effect, increased diversity versus decreased specific beneficial phyla, is genuinely mixed across systems. It must not be presented to consumers as an established microbiome-harm risk in either direction, and it is distinct from the Hurcomb et al. 2025 Drosophila gut-epithelial-cell toxicity finding logged under Mitochondrial burden and Oxidative stress above, which measured cellular damage, not microbial composition; see the overlap map.

Limits on what this shows

• Key negative/null findings: The Pirc Marolt liver-cell finding is itself a partial null/counter-finding against a simple "ATP depletion" story: total ATP production was not shown to fall in that tissue, only its source shifted.
• At-risk subgroups: Not established in humans; the Hardy et al. finding of greater ventral-midbrain than forebrain vulnerability is a preclinical, region-specific signal only.
• Counter-evidence on record: No human ATP-production or cellular-energy study in aripiprazole-treated patients was located. Consumer "low energy" or fatigue claims sometimes associated with antipsychotics generally are not supported by any human ATP measurement found in this review for aripiprazole specifically.

Regulatory status

Not a labeled effect.

Population evidence

Hardy et al. 2023 (PMID 37528429): direct Complex I inhibition drove ATP depletion and cellular toxicity in embryonic mouse neurons, with greater effect in ventral-midbrain neurons. Pirc Marolt et al. 2023 (PMID 38001783): in 6 μM aripiprazole-treated Fao cells, a "significant glycolytic ATP production rate increase of up to 20% of total ATP production" was observed, interpreted as compensatory glycolysis offsetting reduced mitochondrial (oxidative) ATP synthesis; the same study reported diminished pentose-phosphate-pathway activity and lower NADPH. Panizzutti et al. 2023/2024 (PMID 38072867) included aripiprazole among antipsychotics assessed for cellular energy-metabolism effects; full quantitative results were not retrievable this session.

Mechanism

Complex I inhibition impairs oxidative ATP synthesis in neurons (Hardy et al.); the liver-cell model instead shows a compensatory shift toward glycolytic ATP production (Pirc Marolt et al.), consistent with, but not identical to, the same underlying mitochondrial stress.

Framing caution

Reported as a distinct entry from Mitochondrial burden per this project's rule against merging them, but both rest partly on the same Hardy et al. and Pirc Marolt et al. sources; see the overlap map. The tissue-dependent, partly compensatory direction must not be simplified into a blanket "aripiprazole depletes ATP" claim.

Limits on what this shows

• Key negative/null findings: The FDA label and the consumer sources checked (RxList, WebMD) contain no reported hemoglobin, hematocrit, or clinical-anemia signal for aripiprazole in humans; the drug's own labeling is silent on erythrocyte effects specifically.
• At-risk subgroups: None established; this is a fixed-concentration in vitro finding, not a clinical population result.
• Counter-evidence on record: Olanzapine, run as a comparator drug in the same experiment, did not produce the same haemolysis signal at the concentration tested, which the authors present as evidence the effect is aripiprazole-specific rather than a general antipsychotic or vehicle artifact.

Regulatory status

Not addressed in the FDA label; the label documents leukopenia, neutropenia, and rare agranulocytosis (white-cell effects) and rare thrombocytopenia (platelet effect), which are class-level antipsychotic hematologic warnings, not red-cell or oxygen-delivery findings, and are tracked separately in the monitoring panel rather than merged into this entry.

Population evidence

Pirc Marolt et al. 2023 (PMID 38001783): residual blood from 25 healthy donors (9 male, 16 female, normal complete blood counts) was obtained, and aripiprazole was added ex vivo at a 6 μM final concentration for a 2-hour, 37°C incubation. Destruction of approximately 1% of erythrocytes was detected "only in the presence of ARI," versus DMSO and olanzapine controls. This is an in vitro, drug-added-to-donor-blood design, not blood drawn from patients treated with aripiprazole.

Mechanism

Not fully elucidated by the authors; proposed to relate to the same membrane/redox perturbation observed in the paired liver-cell experiments in the same paper (see overlap map).

Framing caution

Roughly 1% in vitro haemolysis at a fixed laboratory concentration and a short incubation is not evidence of clinical anemia and must not be marketed as such. It must also not be conflated with the drug's separately labeled leukopenia/neutropenia/thrombocytopenia warnings, which concern different blood-cell lineages.