Emerging Paradigms in Sevoflurane Anesthesia: Mitochondrial Genome Sensitivity

Mitochondrial Genome Sensitivity

Anesthesiology · Seminar week 32 · released September 14, 2026 · includes a discussion video

This seminar will delve into the hypersensitivity triggers during anesthesia involving sevoflurane linked to specific mitochondrial DNA variants (m.11232 in the MTND4 gene), as…

Learning Objectives

• Explain how volatile anesthetics interact with mitochondrial complex I and distinguish anesthetic sensitivity from mitochondrial toxicity.
• Interpret the evidence linking homoplasmic MT-ND4 m.11232T>C to severe neurologic injury after sevoflurane exposure.
• Differentiate mitochondrial pharmacogenetic injury from malignant hyperthermia, anaphylaxis, residual neuromuscular blockade, and ordinary delayed emergence.
• Apply maternal-family history and targeted mitochondrial testing to preoperative risk assessment.
• Design a regional, intravenous, or propofol-sparing anesthetic for a patient with known or suspected susceptibility.
• Recognize the limitations of current penetrance estimates, variant classification, and ancestry-based screening.
• Integrate genetic findings into durable perioperative documentation, family counseling, and postoperative surveillance.

Overview of Mitochondrial Involvement in Anesthetic Reactions

%%FIG0%% Mitochondria are not the sole molecular target of general anesthesia: sevoflurane-induced hypnosis, amnesia, immobility, and autonomic suppression arise largely from effects on neuronal ion channels and distributed neural networks. Mitochondria nevertheless influence how much anesthetic produces those effects and whether exposure remains reversible or becomes injurious. This distinction is central. A patient may be pharmacodynamically sensitive—requiring an unexpectedly low end-tidal concentration for unconsciousness—without sustaining cellular injury. Conversely, a genetically susceptible patient may tolerate apparently stable intraoperative physiology while anesthetic inhibition of oxidative phosphorylation produces delayed neurologic deterioration.

Complex I, or NADH:ubiquinone oxidoreductase, accepts electrons from NADH, transfers them toward ubiquinone, and couples that transfer to proton movement across the inner mitochondrial membrane. The resulting electrochemical gradient drives ATP synthesis. MT-ND4 is one of seven complex I membrane-arm subunits encoded by mitochondrial DNA. ND4 does not form the NADH-binding catalytic site; rather, it participates in the conformational machinery coupling electron transfer to proton translocation. High-demand tissues—including cerebral cortex, basal ganglia, myocardium, respiratory muscle, and peripheral nerve—have limited tolerance for a sudden fall in ATP production.

Volatile anesthetics inhibit complex I-linked respiration in experimental systems at concentrations relevant to anesthesia. Sevoflurane and isoflurane can reduce NADH-linked oxygen consumption, perturb membrane potential, increase reactive oxygen species, and under some conditions drive ATP synthase in reverse, consuming rather than generating ATP (PMID: 39326543). These findings do not mean routine sevoflurane causes mitochondrial disease. They show that complex I constitutes a biologically plausible vulnerability point when baseline respiratory reserve is genetically reduced.

Human data predated recognition of the m.11232T>C phenotype. In a multicenter study of 91 children undergoing diagnostic muscle biopsy, patients with biochemical complex I deficiency reached a bispectral index of 60 at a median end-tidal sevoflurane concentration of 0.98%, compared with 2.2% in children without mitochondrial abnormalities. Complex I activity correlated with the concentration required for hypnosis, although differences were less pronounced during surgical maintenance and no serious perioperative complications occurred (Hsieh et al., PMID: 33591116). Hypnotic sensitivity therefore does not automatically predict neurotoxicity, and the concentration needed to prevent movement is not interchangeable with the concentration producing unconsciousness.

The variant-specific evidence emerged in 2026. Sanz-Pons and colleagues investigated seven previously healthy patients with severe acute neurologic deterioration after predominantly minor procedures. All shared a mitochondrial haplotype containing homoplasmic NC_012920.1:m.11232T>C in MT-ND4, producing p.Leu158Pro. Patient-derived fibroblasts and transmitochondrial cybrids showed profound suppression—approximately 90% in reported assays—of complex I-dependent oxygen consumption during clinically relevant sevoflurane exposure, while complex II/III-linked respiration was relatively preserved. Pure propofol reduced respiration dose-dependently but did not produce the same genotype-associated differential effect (PMID: 42117732).

Teaching Point: In this seminar, “sevoflurane hypersensitivity” means a pharmacodynamic and bioenergetic susceptibility, not an IgE-mediated allergy.

The proposed mechanism is that replacement of helix-compatible leucine with proline at ND4 residue 158 destabilizes conformational coupling. The carrier may remain clinically normal until sevoflurane further suppresses complex I. ATP depletion then impairs ion pumps, promotes cytotoxic edema and excitotoxicity, increases lactate generation, and preferentially injures metabolically demanding deep-gray structures.

Nuance: Causality is compelling but not completely isolated. The experimental cybrids carried the entire D1b-related mitochondrial haplotype, comprising numerous linked variants, rather than an engineered m.11232T>C-only genome. An MT-CYB control did not reproduce the phenotype, but mitonuclear interactions and other linked variants remain possible modifiers. There is no knock-in animal model, prospective exposure trial, penetrance estimate, or validated safe sevoflurane threshold.

ClinGen subsequently classified m.11232T>C as likely pathogenic specifically for volatile-anesthetic hypersensitivity after reviewing 11 unrelated affected individuals, rarity, segregation, and functional data. It remains a variant of uncertain significance as a cause of primary mitochondrial disease in the absence of anesthetic exposure. That condition-specific distinction prevents a positive result from being misinterpreted as proof of a chronic multisystem disorder.

Audience Poll: Which is the more actionable intraoperative clue: a normal lactate, a normal blood pressure, or unexpectedly profound EEG suppression at a low end-tidal sevoflurane concentration?

Clinical Case Studies Involving the m.11232 MTND4 Variant

%%FIG1%% The peer-reviewed evidence is a small, clinically ascertained cohort rather than a randomized trial. In the seven-patient Sanz-Pons series, severe neurologic deterioration followed general anesthesia for mostly routine procedures. All seven received sevoflurane; six also received propofol boluses and fentanyl, so the clinical histories alone cannot pharmacologically isolate sevoflurane. A subset of the same individuals tolerated intravenous-only anesthesia, strengthening—but not proving—the volatile-specific signal. Complete mitochondrial sequencing identified a shared D1b-related haplotype with homoplasmic m.11232T>C, while exome sequencing in three patients found no consistent nuclear candidate (PMID: 42117732).

The Chilean Ministry of Health independently reported adults and children with delayed awakening, severe neurologic injury, and death following apparently uncomplicated anesthesia. Neuroimaging involved basal ganglia, cerebellum, substantia nigra, brainstem, and other deep-gray structures (PMID: 41705724). These cases substantially overlap with the molecular cohort and must not be counted as an independent second series. A separate published adult report described a previously healthy 20-year-old Venezuelan woman who developed hyperlactatemia, metabolic acidosis, and severe encephalopathy after propofol induction followed by sevoflurane at approximately 0.7–1 MAC with opioids. MRI demonstrated basal ganglia, brainstem, and cerebellar injury; recovery required intensive neurocritical care and decompressive posterior-fossa surgery. The adult case demonstrates that the phenotype is not confined to young children.

A characteristic danger is the mismatch between an apparently routine intraoperative course and catastrophic emergence. The first sign may be failure to awaken, loss of previously present motor responses, abnormal tone, seizures, or evolving acidosis. Normal oxygen saturation and blood pressure do not exclude cellular energy failure, and a single normal lactate cannot exclude compartmentalized cerebral bioenergetic injury.

MUST ACT: Unexpected delayed emergence after volatile anesthesia requires immediate discontinuation of the volatile agent, high-flow oxygen, confirmation of adequate ventilation and circulation, and transition to an intravenous technique if anesthesia must continue. Secure the airway, inspect the breathing circuit, verify drug doses and infusion history, and obtain quantitative neuromuscular monitoring before attributing the event to genetics.

The differential must remain broad. Residual propofol, opioid, benzodiazepine, or neuromuscular block is more common than mitochondrial injury. Check temperature, glucose, arterial or venous blood gas, electrolytes, ionized calcium, lactate, renal and hepatic indices, creatine kinase, and urine output. Consider hypoxemia, hypercapnia, hypotension, hypoglycemia, embolic stroke, intracranial hemorrhage, nonconvulsive status epilepticus, sepsis, and local-anesthetic toxicity. Use naloxone, flumazenil, or neuromuscular reversal only when the corresponding diagnosis is credible; reversal does not replace ventilatory support.

Malignant hyperthermia presents differently: progressive end-tidal carbon dioxide elevation despite increased ventilation, inappropriate tachycardia, rigidity, mixed acidosis, hyperkalemia, and later hyperthermia, rhabdomyolysis, or disseminated intravascular coagulation. If that pattern is present, stop volatile anesthetics and succinylcholine, hyperventilate with 100% oxygen, and administer dantrolene 2–2.5 mg/kg promptly, repeating until hypermetabolism is controlled. Do not wait for genetic testing. Mitochondrial disease alone does not establish malignant-hyperthermia susceptibility, and dantrolene has no demonstrated treatment role for isolated MT-ND4-related neurologic injury.

Perioperative anaphylaxis is suggested by abrupt hypotension, bronchospasm, edema, flushing, or urticaria, although skin findings may be absent. Treat immediately with epinephrine and crystalloid according to the institutional algorithm and obtain acute and baseline tryptase. Butyrylcholinesterase deficiency instead produces isolated prolonged flaccid paralysis after succinylcholine or mivacurium, with absent or weak train-of-four responses; treatment is sedation, ventilation, and time.

Framework: For unexplained postoperative coma, work simultaneously through four domains: residual drugs and paralysis; respiratory, metabolic, and temperature abnormalities; primary neurologic injury or seizure; and uncommon anesthetic syndromes such as malignant hyperthermia, anaphylaxis, or mitochondrial toxicity.

If neurologic recovery remains abnormal, activate stroke and neurocritical-care pathways, obtain CT/CTA when indicated, proceed early to diffusion-weighted MRI, and use EEG to exclude nonconvulsive seizures. Involve neurology, metabolic medicine, genetics, and intensive care. There is no validated mitochondrial antidote; treatment centers on stopping exposure, restoring perfusion and oxygen delivery, preventing hypo- or hyperglycemia, correcting clinically important metabolic derangements, controlling seizures, and avoiding secondary fever or hypothermia.

Decision Point: A prior uneventful anesthetic does not “clear” a patient. Exposure concentration, duration, age, illness, fasting, and metabolic reserve may differ, and the penetrance of m.11232T>C remains unknown.

Pharmacogenetics and Its Role in Anesthetic Hypersensitivity

%%FIG2%% Pharmacogenetics links a specific inherited variant to drug response; pharmacogenomics considers multiple genes, pathways, and clinical variables across the medication regimen. The perioperative environment is unusually demanding because a patient may encounter hypnotics, analgesics, antiemetics, antibiotics, vasopressors, and neuromuscular blockers for the first time within minutes. An actionable result must therefore answer three questions: Is the association analytically valid? Does it predict a clinically important phenotype? Does an effective alternative exist?

MT-ND4 m.11232T>C represents an emerging pharmacodynamic association. Mitochondrial DNA is maternally inherited, and affected cases have generally carried the variant at or near homoplasmy. Healthy mothers carrying the same mitochondrial background demonstrate that the genotype may be silent without the relevant exposure. The variant is likely pathogenic for volatile-anesthetic hypersensitivity, but that classification does not establish complete penetrance, quantify risk for an individual carrier, or prove that every halogenated agent carries equal risk.

Nuance: The 2021 Borden pharmacogenomics appraisal, PMID: 34376788, predates discovery of this MT-ND4 association. It provides a useful framework for perioperative clinical decision support but is not evidence for m.11232T>C. The variant-specific citation is Sanz-Pons et al., PMID: 42117732.

Comparison with mature drug–gene pairs illustrates what is—and is not—known. Pathogenic RYR1 or CACNA1S variants can establish presumptive malignant-hyperthermia susceptibility; CPIC recommends avoiding all potent volatile anesthetics and succinylcholine and using a non-triggering technique (PMID: 30499100). A negative gene panel does not exclude malignant hyperthermia, however, because not every causative variant is known; caffeine-halothane or in-vitro contracture testing remains important when the clinical history is convincing.

Reduced-function BCHE variants cause prolonged metabolism of succinylcholine and mivacurium. This is prolonged neuromuscular blockade, not anesthetic hypersensitivity or malignant hyperthermia. Plasma enzyme activity and dibucaine number can identify the phenotype, while genotyping distinguishes inherited deficiency from acquired reduction due to pregnancy, liver disease, malnutrition, critical illness, or medication effects. A known affected patient should receive an alternative such as rocuronium with quantitative monitoring and an appropriate reversal plan.

CYP2D6 provides another clinically actionable example. Codeine requires CYP2D6 conversion to morphine, and tramadol requires conversion to O-desmethyltramadol. CPIC recommends avoiding both drugs in ultrarapid metabolizers because of toxicity and in poor metabolizers because of inadequate analgesia; morphine, hydromorphone, fentanyl, or nonopioid strategies are not dependent on this bioactivation pathway (PMID: 33387367). By contrast, current evidence does not support routine OPRM1- or COMT-guided opioid dosing. Similarly, CYP2B6 and UGT1A9 associations with propofol pharmacokinetics have not produced a broadly accepted genotype-based induction algorithm; age, frailty, hemodynamics, comedications, and EEG response remain more useful at the bedside.

Teaching Point: A genetic association becomes clinically useful only when it changes the anesthetic, dose, monitoring, or disposition. A long panel of variants of uncertain significance can create noise without improving safety.

The ImPreSS perioperative implementation study illustrates the workflow problem. During its 71-patient pilot, clinicians accessed available pharmacogenomic results before or during 58% of operations, and truly high-risk drug–gene encounters were uncommon (PMID: 35213469; NCT03729180). Borden and colleagues identified credible evidence across numerous perioperative medications but emphasized the need for prospective clinical-utility testing and embedded decision support (PMID: 34376788). A PDF buried in the chart is not actionable during induction.

For mitochondrial results, interpretation must include the exact HGVS designation, tissue tested, homoplasmy or heteroplasmy level, laboratory methodology, phenotype-specific classification, and report date. Direct-to-consumer panels and routine exomes may omit or inadequately cover mitochondrial DNA. A negative targeted m.11232T>C result substantially lowers risk attributable to this allele but does not exclude other mitochondrial disorders or every cause of anesthetic sensitivity.

Audience Poll: Would you change an anesthetic for an MT-ND4 variant reported simply as a VUS, or would you first require the exact allele, phenotype-specific classification, maternal pedigree, and laboratory confirmation?

Integrating Genetic Information into Anesthetic Planning

%%FIG3%% Integration begins with phenotype and pedigree, not a sequencing report. Ask about prior delayed awakening, postoperative weakness, unexplained ICU admission, seizures, metabolic acidosis, rhabdomyolysis, fever or rigidity, severe neurologic deterioration, and deaths after minor procedures. Construct the maternal lineage explicitly: the relevant question is direct maternal Venezuelan ancestry, not nationality alone. A person with only paternal Venezuelan ancestry does not inherit that father’s mitochondrial genome. Ancestry should be asked sensitively and never inferred from appearance, surname, or language.

Framework: Classify the patient into one of four pathways: confirmed m.11232T>C; direct maternal risk with testing pending or unavailable; negative targeted testing but a concerning phenotype; or neither genetic nor clinical risk. Urgency determines whether evaluation can precede surgery, but it does not justify avoidable exposure.

For elective surgery, refer a potentially affected patient to clinical genetics or a mitochondrial specialist. The laboratory order should explicitly request NC_012920.1:m.11232T>C in MT-ND4; targeted mtDNA analysis is sufficient for this specific question, while complete mitochondrial sequencing is more appropriate when the patient has a broader multisystem phenotype. Because reported affected patients are homoplasmic, blood testing should generally detect the risk allele. Obtain informed consent that addresses maternal relatives and possible incidental findings. A positive woman may transmit mitochondrial DNA to all children; a positive man does not transmit it to his children.

Document the result as a phenotype-specific safety alert: “MT-ND4 m.11232T>C—risk of severe neurologic injury with volatile anesthetics; avoid volatile agents pending specialist guidance.” A generic “sevoflurane allergy” may be visible but obscures the mechanism and may fail to alert clinicians ordering isoflurane or desflurane. Include the original report, classification date, prior exposure history, recommended alternatives, and postoperative observation plan.

Decision Point: If urgent surgery must proceed before results return, treat a patient with direct maternal ancestry and a compelling personal or family history as potentially susceptible. Do not postpone time-critical care solely for genetic testing.

Regional or local anesthesia is preferred when it can provide surgical conditions without deep sedation. When general anesthesia is necessary, use a volatile-free technique and titrate every hypnotic rather than applying a fixed “normal” dose. One adult approach is incremental propofol—approximately 0.5 mg/kg aliquots to loss of response, often totaling 1–2 mg/kg in a stable patient—followed by the lowest effective infusion, commonly 50–150 micrograms/kg/min, with remifentanil approximately 0.05–0.2 micrograms/kg/min. Dexmedetomidine 0.2–0.7 micrograms/kg/h without a loading dose, regional blockade, ketamine, and multimodal analgesia may reduce propofol exposure. These are reference ranges, not MT-ND4-specific prescriptions; children require age-appropriate pharmacokinetic dosing and pediatric-anesthesia expertise.

Place processed EEG before induction, inspect the raw waveform and suppression ratio, and generally target an index compatible with surgical anesthesia—often 40–60—while avoiding burst suppression. Ketamine, electromyographic artifact, age, hypothermia, and neurologic disease can distort a proprietary index. TIVA also introduces awareness risk from IV disconnection or pump failure, so keep the cannula visible, use dedicated pumps and anti-reflux protection, and confirm drug delivery whenever the EEG or hemodynamics change unexpectedly.

If neuromuscular blockade is required, a nondepolarizer such as rocuronium can be used with quantitative train-of-four monitoring. In significant myopathy, response may be prolonged; do not extubate by elapsed time or visual twitch assessment. Succinylcholine is not specifically contraindicated by m.11232T>C, but it should be avoided when malignant hyperthermia remains possible or when myopathy creates hyperkalemic-rhabdomyolysis risk.

General mitochondrial-disease precautions should be individualized. Minimize fasting, schedule early, maintain euvolemia, normoxia, normocapnia, and normothermia, and monitor glucose. Patients with established mitochondrial disease may need dextrose-containing fluid to prevent catabolism, except during ketogenic therapy, pyruvate-metabolism disorders, or a documented adverse response to glucose (PMIDs: 25503498 and 28749475). An asymptomatic m.11232T>C carrier does not automatically require high-rate dextrose. Avoid unmonitored hyperglycemia and interpret lactate as a trend rather than a binary screen.

MUST ACT: After anesthesia, continue observation until ventilation, strength, speech, behavior, and cognition have returned to documented baseline. New delayed awakening, dystonia, weakness, seizure, or acidosis warrants immediate neurologic and metabolic evaluation rather than routine discharge.

Guidelines for Screening and Avoidance Strategies with Volatile Anesthetics

%%FIG4%% Guidance changed rapidly in 2026. ASA and the Society for Pediatric Anesthesia advised sensitive screening for maternal Venezuelan ancestry, genetics consultation and explicit m.11232T>C testing when feasible, avoidance of volatile anesthetics in unresolved high-risk patients, regional anesthesia or TIVA, processed EEG to avoid excessive depth and burst suppression, and postoperative observation until neurocognitive baseline. European, Canadian, Chilean, Spanish, Portuguese, and Venezuelan organizations issued similar precautionary statements. On July 2, 2026, the FDA announced an investigation into severe neurologic outcomes following sevoflurane exposure and advised consideration of intravenous or regional alternatives in potentially susceptible patients.

These are interim risk-mitigation statements, not evidence from comparative anesthetic trials. Routine population-wide genetic screening is not currently recommended. Targeted assessment is reasonable for a patient with direct maternal Venezuelan ancestry, a personal history of otherwise unexplained postoperative neurologic deterioration, or a maternal-family cluster of severe anesthetic outcomes. Children and adults may be affected, and absence of preexisting mitochondrial symptoms is not reassuring.

MUST ACT: A confirmed carrier—or a clinically high-risk patient whose result is unavailable—should not receive sevoflurane. Because comparative safety data for desflurane, isoflurane, and other halogenated agents are absent and complex I inhibition may be a class effect, current precautionary practice is to avoid all volatile anesthetics rather than substitute one volatile for another.

This position is narrower than saying that every mitochondrial disorder mandates volatile avoidance. Before recognition of m.11232T>C, case series showed that many patients with genetically confirmed mitochondrial disease tolerated short sevoflurane, isoflurane, or propofol anesthetics, with no clear relationship between agent and complication (PMID: 27885500). Consensus guidance therefore favored careful titration rather than a universal prohibition. The new MT-ND4 phenotype is a specific emerging exception characterized by catastrophic neurologic outcomes and compelling ex-vivo complex I sensitivity.

Regional anesthesia, neuraxial anesthesia, peripheral nerve blocks, and local infiltration avoid general-anesthetic exposure when appropriate, although local-anesthetic systemic toxicity remains possible at excessive plasma concentrations. For general anesthesia, options include cautious short-duration propofol-based TIVA, ketamine, etomidate, dexmedetomidine, midazolam or another short-acting benzodiazepine, short-acting opioids, and multimodal regional analgesia. The Sanz-Pons experiments found no variant-specific differential response to propofol, and some reported patients tolerated intravenous-only anesthesia; neither observation proves that prolonged propofol is harmless.

Propofol impairs several mitochondrial processes and prolonged or high-dose administration can produce propofol infusion syndrome, characterized by otherwise unexplained lactic or high-anion-gap acidosis, rhabdomyolysis, hyperkalemia, bradyarrhythmia or Brugada-like ECG changes, cardiac failure, lipemia, and renal injury. Avoid prolonged ICU propofol sedation in known mitochondrial disease. For short TIVA, use the lowest effective dose, carbohydrate and perfusion support when appropriate, processed EEG, and serial metabolic assessment in longer or physiologically stressful cases. Stop propofol immediately if a PRIS pattern develops.

A volatile-free plan requires prevention of accidental administration, but m.11232T>C is not malignant hyperthermia. Use a vapor-free workstation or a reliably prepared circuit when feasible, disable or remove vaporizers, and conduct a team briefing. There is no validated residual-parts-per-million threshold for this phenotype, and prophylactic dantrolene is not indicated. Activated-charcoal filters and formal malignant-hyperthermia machine preparation are required when MH susceptibility independently remains possible, not merely because an MT-ND4 result is present.

Nuance: Avoidance is not the whole protocol. Excessive propofol, hypotension, hypoxemia, hypocapnia or hypercapnia, hypoglycemia, hypothermia, shivering, prolonged fasting, pain, and postoperative vomiting can all reduce energy supply or increase metabolic demand. A “nonvolatile” anesthetic that ignores physiology is not mitochondrial protection.

A previous normal sevoflurane anesthetic does not override a positive result. Conversely, a verified negative targeted result should not be interpreted as excluding MH, BCHE deficiency, anaphylaxis, another mitochondrial variant, or ordinary causes of delayed emergence. When testing cannot be completed before an urgent procedure, a regional or carefully titrated intravenous plan is generally less consequential than exposing a plausible carrier while waiting for definitive penetrance data.

Audience Poll: For an asymptomatic patient with direct maternal Venezuelan ancestry and no test result, which is the more defensible plan for a short procedure: delay necessary care, proceed with sevoflurane because prior anesthesia was normal, or use an available regional/volatile-free technique with enhanced monitoring?

Future Directions in Pharmacogenomics in Anesthesiology

%%FIG5%% The immediate research priority is not another association report; it is a denominator. Registries must capture carriers with adverse events, carriers with uneventful anesthetics, exact end-tidal concentrations, exposure duration, age, fasting, illness, comedications, EEG data, lactate trends, imaging, and neurologic outcomes. Without exposed controls, neither penetrance nor absolute risk can be estimated. Population studies across Venezuela and Indigenous American D1 lineages should determine carrier frequency without treating nationality as a genotype.

Mechanistic studies must separate m.11232T>C from its mitochondrial background. Isogenic transmitochondrial models carrying only p.Leu158Pro, followed by neuronal and cardiomyocyte studies, could test whether the allele itself is sufficient or requires linked variants or a particular nuclear genome. Candidate endpoints include complex I assembly, NADH/NAD+ balance, oxygen consumption, ATP concentration, membrane potential, reactive oxygen species, calcium handling, and reversibility after washout. Head-to-head testing should include sevoflurane, isoflurane, desflurane, propofol formulations, ketamine, etomidate, dexmedetomidine, remimazolam, and commonly coadministered drugs. No prospective trial should deliberately expose a known carrier to sevoflurane merely to establish causality.

Framework: Future prediction will require three layers: genotype and haplotype; exposure intensity and duration; and the physiologic context that determines metabolic reserve. Genotype alone is unlikely to explain every outcome.

Biomarker development should focus on signals available before irreversible injury. Processed EEG and burst suppression may reveal unusual hypnotic sensitivity, but their ability to predict mitochondrial toxicity is unproven. Cerebral near-infrared spectroscopy can trend regional oxygenation but does not measure ATP generation. Blood lactate may remain normal despite localized cerebral failure. Candidate research biomarkers include acylcarnitines, lactate-to-pyruvate ratio, circulating cell-free mitochondrial DNA, neurofilament light, and glial fibrillary acidic protein. None currently justifies delaying cessation of a suspected trigger while awaiting results.

Rapid clinical testing is another priority. A targeted assay that returns within hours could be useful for urgent surgery, but it must preserve confirmatory quality and detect homoplasmy, heteroplasmy, and sample errors. Laboratories need structured reporting that distinguishes “likely pathogenic for volatile-anesthetic hypersensitivity” from “uncertain for primary mitochondrial disease.” Reclassification should automatically update the electronic record rather than depend on a patient rediscovering an old PDF.

Implementation science may prove as important as sequencing. The ImPreSS trial demonstrated that preoperative results can be integrated into perioperative decision support, yet clinicians accessed them inconsistently and high-risk encounters were uncommon (PMID: 35213469). The broader PREPARE implementation trial found that a multigene prescribing strategy reduced clinically relevant adverse drug reactions by roughly 30%, but it was not anesthesia-specific and did not validate MT-ND4 screening (PMID: 36739136). Future perioperative systems should display a concise phenotype, prohibited agents, acceptable alternatives, evidence level, and specialist contact when the relevant drug is ordered.

Decision Point: An interruptive alert is justified when the consequence is catastrophic and the alternative is straightforward; it becomes harmful when it fires for every VUS or fails to distinguish MT-ND4 risk from RYR1-associated malignant hyperthermia.

Equity is inseparable from implementation. “Maternal Venezuelan ancestry” is a temporary risk-enrichment tool derived from the cases recognized to date, not a biologic diagnosis and not evidence that Venezuelan patients as a group are unsafe to anesthetize. Screening must avoid stigmatization, denial of surgery, or false reassurance in carriers outside the initially recognized population. Maternal pedigree questions should be voluntary, clinically explained, and paired with access to alternatives and confirmatory testing.

The field should ultimately move from ancestry-triggered avoidance toward genotype-informed, exposure-specific recommendations. That transition will require transparent registries, international case definitions, maternal cascade studies, cost-effectiveness analyses, and continuously updated decision support. Until then, the defensible posture is precision with humility: recognize the signal, avoid a plausible trigger when alternatives exist, and state clearly what remains unknown.

Audience Poll: What evidence would change your practice most—a reliable rapid assay, an exposure-adjusted penetrance estimate, an isogenic functional model, or a validated volatile-free protocol with prospective outcomes?

Clinical Case: Volatile-Free Planning for a Suspected Carrier

This is a composite teaching case, not an additional published m.11232T>C report.

A 45-year-old woman is scheduled for elective laparoscopic cholecystectomy. She is neurologically normal, exercises regularly, and has normal glucose, bicarbonate, creatinine, ECG, and functional capacity. During the preanesthetic interview, she reports that her mother was born in Carabobo, Venezuela, and that a maternal cousin had profound delayed awakening and permanent motor impairment after a brief dental anesthetic. The records are unavailable. She herself received intravenous sedation for endoscopy without difficulty but has never received a volatile anesthetic.

Audience Poll: Does her normal phenotype lower the risk enough to proceed with sevoflurane?

The family history is compatible with direct maternal transmission and exposure-dependent disease. It is not diagnostic: the cousin could have experienced hypoxic injury, overdose, anaphylaxis, malignant hyperthermia, or an unrelated neurologic event. Because surgery is elective, the anesthesiologist requests genetics consultation and clinical-grade targeted mtDNA testing specifying MT-ND4 m.11232T>C. Testing identifies the variant at homoplasmy and reports it as likely pathogenic for volatile-anesthetic hypersensitivity. The patient is counseled that this result predicts an anesthetic risk, not inevitable chronic mitochondrial disease, and that penetrance remains unknown.

The operation proceeds with a volatile-free workstation, standard monitors, continuous core temperature, processed EEG, and quantitative neuromuscular monitoring. The IV is visible and dedicated to TIVA. After preoxygenation, propofol is administered in 0.5 mg/kg increments rather than as a rapid fixed bolus, with remifentanil started at a low infusion rate. Rocuronium provides intubating conditions. Maintenance uses the lowest propofol concentration maintaining an appropriate raw EEG pattern and index, supplemented by remifentanil, port-site local anesthesia, acetaminophen, and a small dexmedetomidine infusion. Sevoflurane, desflurane, isoflurane, and succinylcholine are not used.

Glucose, ventilation, temperature, blood pressure, and perfusion remain normal. The EEG shows no burst suppression. Quantitative monitoring confirms adequate reversal before awake extubation. In recovery, the team assesses orientation, speech, limb strength, coordination, and behavior against the documented baseline rather than relying only on an Aldrete score. She resumes oral carbohydrate and is discharged only after sustained neurologic recovery.

Teaching Point: The decisive information was not nationality alone. It was direct maternal lineage, a compatible maternal-family event, confirmatory testing, and the availability of a clinically reasonable alternative.

If the same patient instead presented with acute cholecystitis before testing could return, urgent source control should not be withheld. She would be managed as potentially susceptible with the same volatile-free approach. If she developed rising end-tidal carbon dioxide, rigidity, hyperkalemia, or rapid hyperthermia, the team would treat malignant hyperthermia immediately despite the mitochondrial diagnosis. If she simply failed to awaken, the team would stop anesthetic delivery, confirm ventilation and neuromuscular recovery, check glucose and acid-base status, and initiate early neurologic evaluation rather than assuming that every delayed emergence represents MT-ND4 toxicity.

MUST ACT: Provide the patient with a copy of the result and anesthetic plan, enter a durable EHR alert, and recommend genetics-guided counseling for maternal relatives. A successful volatile-free anesthetic does not justify deleting the alert.

Tonight on Shift

• Ask specifically about prior neurologic deterioration after anesthesia, maternal-family anesthetic events, and direct maternal Venezuelan ancestry.
• For confirmed or unresolved high-risk patients, avoid all volatile anesthetics; favor regional anesthesia or carefully titrated TIVA with processed EEG.
• Treat delayed emergence as a diagnosis, not a label: verify ventilation, glucose, temperature, drug effect, quantitative neuromuscular recovery, seizure, and stroke.
• Distinguish MT-ND4 injury from malignant hyperthermia, anaphylaxis, and BCHE deficiency; give dantrolene only when the clinical syndrome supports MH.
• Minimize fasting and physiologic stress, maintain perfusion and normothermia, and observe until cognition, strength, ventilation, and behavior return to baseline.
• Document the exact variant and phenotype-specific risk, involve genetics and neurology when indicated, counsel maternal relatives, and report suspected events through patient-safety and pharmacovigilance systems.

Read this seminar as Markdown · All seminars · Lecture library · Question bank