Disorders of Creatine Metabolism: A State-of-the-Art Clinical Review for the Practising Internist

Dr Neeraj Manikath , claude.ai
Conflict of Interest: None declared | Funding: None

---

Abstract

The cerebral creatine deficiency syndromes (CCDS) — guanidinoacetate methyltransferase (GAMT) deficiency, arginine:glycine amidinotransferase (AGAT) deficiency, and X-linked creatine transporter (SLC6A8) deficiency — constitute a treatable yet chronically underdiagnosed triad of inborn errors affecting creatine biosynthesis and transport. Their shared phenotype of intellectual disability, language regression, and seizures conceals biochemically and therapeutically distinct entities. The average diagnostic delay exceeds five years, representing a preventable loss of critical treatment windows. This review provides the practising internist and postgraduate trainee with diagnostic reasoning, MRS interpretation skills, treatment protocols, and the bedside nuances needed to close this gap.

Keywords: creatine deficiency syndromes, GAMT, AGAT, SLC6A8, cerebral creatine, MRS, intellectual disability, inborn errors of metabolism

---

Introduction

Creatine occupies a position of fundamental metabolic importance that is disproportionate to the clinical attention it typically receives. Synthesised primarily in the kidney (via AGAT) and liver (via GAMT), creatine travels in the bloodstream to tissues of high-energy demand — principally brain and skeletal muscle — where it enters via the sodium- and chloride-dependent transporter SLC6A8. Within neurons and myocytes, creatine is phosphorylated to phosphocreatine by creatine kinase, forming the cell's primary short-term energy buffer. The developing brain is exquisitely dependent on this system.

Defects at any of the three steps — AGAT synthesis, GAMT methylation, or SLC6A8 transport — produce the cerebral creatine deficiency syndromes (CCDS). These disorders are united by intellectual disability and language impairment yet diverge profoundly in pathophysiology, biochemistry, and therapeutic requirement. The distinction is not academic: GAMT deficiency demands a three-pronged treatment protocol beyond creatine alone; AGAT deficiency responds beautifully to creatine monotherapy; and SLC6A8 deficiency in males is nearly refractory to oral creatine entirely — a therapeutic paradox that traps unprepared clinicians in futile supplementation for years.

The biosynthetic pathway may be summarised as:

Arginine + Glycine → [AGAT, kidney] → Guanidinoacetate (GAA)
                  → [GAMT, liver] → Creatine
                  → [SLC6A8, blood-brain barrier] → Brain & Muscle

Defect at step one (AGAT): no GAA produced, creatine depleted, no toxic accumulation.
Defect at step two (GAMT): GAA accumulates to neurotoxic levels AND creatine is depleted.
Defect at step three (SLC6A8): creatine synthesised normally but cannot enter the brain.

The diagnostic toolkit is simple, cheap, and broadly available: a urine creatine-to-creatinine ratio and plasma guanidinoacetate (GAA) level can be ordered by any clinician from any standard biochemistry laboratory. The bottleneck has always been clinical suspicion. This review aims to eliminate that bottleneck.

---

Section 1 — Guanidinoacetate Methyltransferase (GAMT) Deficiency: The Progressive Extrapyramidal Syndrome with Seizures

Pathophysiology: The Double Insult

GAMT deficiency (OMIM #612736) arises from biallelic pathogenic variants in the GAMT gene on chromosome 19p13.3. It is the most biochemically complex CCDS because its pathology stems not merely from creatine depletion but from the simultaneous and progressive accumulation of its immediate precursor, guanidinoacetate (GAA). This dual mechanism — creatine deficiency plus GAA neurotoxicity — explains why GAMT deficiency produces a far more severe phenotype than AGAT deficiency, where creatine is equally depleted but GAA remains low.

GAA exerts neurotoxic effects through multiple converging mechanisms: competitive antagonism at GABA-A receptors (a direct epileptogenic mechanism), induction of oxidative stress via lipid peroxidation, and inhibition of mitochondrial complex I respiratory activity. It is GAA accumulation, not creatine absence per se, that drives the extrapyramidal syndrome, the refractory epilepsy, and the severe behavioural disturbance that define this condition.

Clinical Fingerprint at the Bedside

The phenotypic triad of intellectual disability, extrapyramidal movement disorder (dystonia and choreoathetosis), and refractory multi-focal seizures in a young child should immediately elevate GAMT deficiency on the differential. The movement disorder typically emerges between 6 months and 3 years of age and is progressive if untreated. Behavioural disturbances — hyperactivity, stereotypies, self-injurious behaviour — frequently dominate the clinical picture and can be the presenting complaint to psychiatry or behavioural paediatrics before the neurological diagnosis is made.

The cardinal clinical discriminator that distinguishes GAMT from the other CCDS is the extrapyramidal motor syndrome. This point bears emphasis in capital letters: dystonia and choreoathetosis are prominent in GAMT and are minimal or absent in AGAT and SLC6A8 deficiency. When you see this combination at the bedside, the metabolic differential is essentially narrowed to one diagnosis.

💎 Clinical Pearl: The single most powerful bedside discriminator between GAMT deficiency and the other CCDS is the movement disorder. Extrapyramidal features — dystonia, choreoathetosis, progressive dyskinesia — are the hallmark of GAMT and are absent in AGAT and SLC6A8 deficiency. When you encounter this triad in a child with intellectual disability and epilepsy, order plasma GAA immediately. Results in 48–72 hours can end a multi-year diagnostic odyssey.

🦪 Oyster: GAMT deficiency masquerades as dyskinetic cerebral palsy in a clinically significant number of cases. The diagnostic trap is conceptual: CP is classified as a non-progressive, static encephalopathy. When a "CP" patient deteriorates, clinicians attribute it to acquired complications rather than triggering metabolic review. Any neurological regression in presumed dyskinetic CP mandates metabolic re-evaluation including urine and plasma GAA. This is one of the most consequential — and preventable — missed diagnoses in paediatric neurology and general medicine.

Investigations

Biochemical hallmarks: plasma GAA is markedly elevated (5–10-fold above upper limit of normal); urinary GAA is correspondingly elevated; plasma creatine is low; urinary creatine-to-creatinine ratio is elevated. Brain MRI may be normal early or show T2/FLAIR signal abnormality in the globus pallidus — a finding that, in clinical context, is highly specific and should not be dismissed as incidental. Brain proton MRS demonstrates an absent creatine peak and — pathognomically — a GAA peak at 3.78 ppm seen in no other clinical condition (see MRS section). Confirmatory diagnosis requires enzyme assay in erythrocytes or fibroblasts and GAMT sequencing.

🔧 Hack: In an undiagnosed patient with severe dystonia and drug-refractory epilepsy awaiting genetic confirmation, do not wait to start empirical creatine supplementation. At 400 mg/kg/day in divided doses, creatine monohydrate is safe, inexpensive, and can produce rapid and clinically meaningful reduction in seizure frequency within weeks. Discuss urgently with your metabolic team. Early institution changes the neurodevelopmental trajectory; delay does not.

Treatment: The Trident Protocol

GAMT deficiency demands three simultaneous therapeutic interventions:

1. Creatine monohydrate (400–800 mg/kg/day in 2–3 divided doses) — replenishes cerebral creatine
2. L-ornithine supplementation (100–200 mg/kg/day) — competitively inhibits AGAT, diverting activity away from GAA production toward the ornithine-glycine reaction, thereby reducing GAA synthesis at its source
3. Dietary arginine restriction — reduces substrate availability for GAA synthesis; target plasma arginine to the lower quartile of the age-matched normal reference range

Treatment initiated before 18–24 months of age produces near-complete prevention of the phenotype. Treatment after 4–5 years rarely reverses established cognitive deficits, but meaningful improvements in seizure control, behaviour, and quality of life are consistently documented even with late treatment.

🦪 Oyster: Ornithine overdosing in the GAMT trident protocol causes a secondary aminoaciduria that is under-recognised. Ornithine competes with lysine, arginine, and other dibasic amino acids at the shared cationic amino acid transporter at intestinal and renal tubular levels. Patients on high-dose ornithine who develop unexplained vomiting, anorexia, or growth faltering should have urine amino acids checked urgently. Dose reduction typically resolves the complication. Clinicians focused on GAA reduction targets frequently overlook this downstream effect.

---

Section 2 — Arginine:Glycine Amidinotransferase (AGAT) Deficiency: The Intellectual Disability with Creatine Depletion

The Purest Model of Cerebral Creatine Deficiency

AGAT deficiency (OMIM #612735) results from biallelic pathogenic variants in GATM on chromosome 15q21.1 and is the rarest of the three CCDS. Unlike GAMT deficiency, there is no toxic intermediate accumulation — GAA levels are low or frankly undetectable because the first biosynthetic step is entirely blocked. The entire clinical phenotype flows solely from cerebral creatine depletion. This makes AGAT deficiency the cleanest answer to a fundamental question in metabolic neuroscience: what precisely does creatine do for the developing human brain?

The answer, revealed by treatment trials, is this: creatine is essential for normal cognitive development, language acquisition, and seizure threshold — and its absence, in isolation, produces a phenotype that is severe but exquisitely responsive to repletion.

Clinical Recognition

The phenotype is centred on mild-to-moderate intellectual disability, prominent expressive language delay with relatively preserved receptive function, and a social-communication profile that frequently attracts an autism spectrum disorder diagnosis before the metabolic aetiology is established. Seizures occur but are less frequent and considerably less severe than in GAMT deficiency. The defining negative feature — absence of extrapyramidal signs — is diagnostically invaluable.

The prototypical clinical presentation: a child aged 3–5 years with unexplained language delay and intellectual disability. Chromosomal microarray is normal. FMR1 testing is normal. CGG repeat analysis is normal. Standard neurodevelopmental gene panels return no pathogenic variants. The child has been seen by three specialists and labelled idiopathic ASD. A single urine sample for creatine/creatinine ratio and plasma GAA — costing under £5 — answers the question. This is AGAT deficiency.

💎 Clinical Pearl: Low — not elevated — plasma GAA is the key biochemical fingerprint of AGAT deficiency. This is counterintuitive and catches many clinicians off guard. Laboratories routinely flag elevated values; they rarely flag low-normal ones. Train yourself to look at the absolute GAA value, not merely whether it is flagged as abnormal. A plasma GAA at or below the lower quartile of the reference range in a child with intellectual disability is not a normal result — it is a diagnostic signal. Request explicit lower-limit flagging from your metabolic laboratory for this analyte.

🔧 Hack: When sending a metabolic screen for suspected CCDS, add a clinical note on the request form: "Please report absolute quantified values for plasma GAA and urine creatine/creatinine ratio with age-specific reference ranges. Low GAA is diagnostically significant." This single instruction dramatically improves reporting accuracy and prevents AGAT deficiency from being signed off as a normal screen.

The Most Rewarding CCDS to Treat

AGAT deficiency responds to creatine monohydrate monotherapy. There is no arginine restriction, no ornithine supplementation, no dietary complexity whatsoever. The treatment is a tasteless powder dissolved in any morning beverage. Neonatal or presymptomatic treatment — achievable only through newborn screening — yields completely normal neurodevelopmental outcomes. Even treatment commenced after cognitive symptoms are established produces meaningful, sustained improvement in language and behaviour over 12–24 months. AGAT deficiency is the single strongest argument for incorporating CCDS into expanded newborn screening programmes globally.

🦪 Oyster: Adult patients with long-standing intellectual disability of unknown aetiology represent a population in whom AGAT deficiency is consistently underdiagnosed. The reflexive clinical assumption that an adult with established ID has been fully worked up is often false. AGAT deficiency can present in adults, and a confirmed diagnosis remains clinically actionable at any age — not because full cognitive recovery is achievable, but because the diagnosis enables genetic counselling, cascade family testing, presymptomatic treatment of siblings and future offspring, and, frequently, modest but meaningful functional improvement in the index patient. Never let adult age foreclose a metabolic screen in unexplained ID.

---

Section 3 — Creatine Transporter (SLC6A8) Deficiency Revisited: The X-Linked Phenotype in Females

The Most Common CCDS — and Its Most Overlooked Population

SLC6A8 deficiency (OMIM #300352) arises from hemizygous pathogenic variants in SLC6A8 on Xq28 and is the most prevalent CCDS, estimated to account for approximately 1–2% of X-linked intellectual disability in males. The male phenotype — moderate-to-severe intellectual disability, behavioural disturbance, prominent speech and language disorder, seizures, and absent cerebral creatine on MRS despite entirely normal peripheral creatine synthesis — is reasonably well characterised in the literature.

What remains dramatically underappreciated, and demands urgent attention from every practising clinician, is the heterozygous female phenotype. Dismissed for decades as unaffected carriers, heterozygous females now have robust evidence of clinically significant neurological manifestations in approximately 40–50% of cases, ranging from reading disorder and attentional difficulties to frankly moderate intellectual disability, anxiety disorders, and significant behavioural dysregulation. These women are in every general medicine clinic, psychiatry outpatient, and primary care practice — undiagnosed.

The X-Inactivation Variable: Why Females Are Different

The female phenotype is shaped and modulated by X-inactivation patterns. Females with skewed X-inactivation favouring the mutant allele can develop a phenotype that is indistinguishable from affected males. Those with near-random inactivation manifest intermediate symptoms ranging from subtle learning differences to moderate disability. The critical clinical consequence: the urine creatine/creatinine ratio — the standard first-line diagnostic screen — is unreliable in heterozygous females. Because the normal allele is expressed in a proportion of renal tubular cells, partial transporter activity is maintained, normalising the ratio even in clinically affected females.

A normal urine creatine/creatinine ratio in a female from an SLC6A8 pedigree does not exclude the diagnosis.

💎 Clinical Pearl: When a male patient receives a confirmed diagnosis of SLC6A8 deficiency, every female first-degree relative — mother, sisters, daughters — requires both neuropsychological evaluation AND molecular testing, regardless of biochemical screening results. Urine creatine/creatinine ratio has near-100% sensitivity in males but is unreliable in heterozygous carrier females. Genotyping is the gold standard for female relatives — not biochemistry. This is a non-negotiable component of cascade management.

🦪 Oyster: Women presenting to psychiatry or general medicine with ADHD, reading disorder, anxiety, and a family history of intellectual disability in male relatives represent the archetypical undiagnosed SLC6A8 heterozygote. This demographic is encountered in every specialty clinic, yet SLC6A8 sequencing is essentially never ordered. The diagnosis is actionable — creatine supplementation in females with markedly skewed X-inactivation can produce meaningful cognitive and behavioural benefit. SLC6A8 sequencing should be standard practice in any woman with unexplained neurocognitive symptoms and an X-linked family pedigree.

🔧 Hack: Attach a systematic cascade protocol letter to every SLC6A8 diagnosis. Include a standing instruction for the clinical team to contact all female first-degree relatives, explaining the X-linked inheritance pattern and the need for molecular testing irrespective of urine metabolite screen results. This takes one administrative step to complete and identifies affected females who would otherwise never be investigated.

The Defining Paradox in Males

SLC6A8 deficiency presents a fundamental biological paradox: creatine synthesis is entirely normal. Serum creatine is normal. GAA is normal. The liver produces creatine without difficulty; it enters the systemic circulation at normal concentrations. The urine creatine/creatinine ratio is elevated not because of overproduction but because renal tubular reabsorption is impaired by the identical transporter defect that blocks brain entry. Brain MRS shows absent creatine despite normal circulating creatine. The blood-brain barrier becomes, in effect, an impenetrable therapeutic wall.

---

Section 4 — Cerebral Creatine Deficiency Syndromes: The MRS Pattern and Treatment Response

Proton MRS — The Decisive Investigation

Proton magnetic resonance spectroscopy (¹H-MRS) is the single most powerful investigation in CCDS workup and should be standard-of-care in any patient with unexplained intellectual disability, particularly when accompanied by seizures or movement disorder. The rationale is compelling: the creatine peak at 3.03 ppm (total creatine, predominantly creatine and phosphocreatine) is one of the most stable, reproducible, and abundant metabolite peaks in normal human brain spectroscopy across all ages. Its absence or severe reduction is an unmistakable, essentially pathognomonic signal detectable within moments of spectral inspection.

🔧 Hack: Standard radiology MRS reports for brain tumours use NAA/creatine ratios and routinely do not comment on absolute creatine levels or compare them to normative values. When ordering MRS in a CCDS suspect, write explicitly on the request form: "Please quantify the absolute creatine peak at 3.03 ppm and compare with age-matched normative values. Please specifically report any peak at 3.78 ppm." This single instruction transforms a generic spectroscopy report into a diagnostic document and prevents weeks of avoidable delay.

The Three MRS Signatures

GAMT deficiency produces an absent or severely reduced creatine peak at 3.03 ppm — identical in appearance to the other CCDS at first glance. The pathognomonic distinguishing feature is an abnormal peak at 3.78 ppm representing accumulated GAA. This peak is not present in any other condition encountered in routine clinical neuroimaging practice. Its presence provides immediate diagnostic certainty without waiting for biochemical or genetic confirmation. If you see an absent creatine peak with a 3.78 ppm signal, you have made the diagnosis of GAMT deficiency at the scanner. Start the trident treatment protocol the same day.

AGAT deficiency produces absent or severely reduced creatine with no GAA peak — spectroscopically identical to SLC6A8 deficiency. Peripheral biochemistry (low plasma GAA versus normal in SLC6A8, and elevated urine creatine/creatinine versus elevated in both) provides the discriminating information.

SLC6A8 deficiency produces absent or severely reduced cerebral creatine with entirely normal peripheral creatine metabolism. This is the defining paradox made visually apparent: a flat 3.03 ppm peak on MRS alongside a completely normal serum creatine level. No GAA peak is present.

💎 Clinical Pearl: The GAA peak at 3.78 ppm on ¹H-MRS is pathognomonic of GAMT deficiency and represents one of the very few instances in metabolic medicine where a single imaging finding establishes a specific molecular diagnosis. Recognising this peak allows same-day initiation of the full trident treatment protocol, compressing treatment initiation from months to hours. Every neurologist, radiologist, and metabolic physician must know this peak.

Feature	GAMT	AGAT	SLC6A8
Creatine peak (3.03 ppm)	Absent ↓↓↓	Absent ↓↓↓	Absent ↓↓↓
GAA peak (3.78 ppm)	Present — PATHOGNOMONIC	Absent	Absent
Plasma GAA	Markedly elevated ↑↑↑	Low / undetectable ↓	Normal
Serum creatine	Low ↓	Low ↓	Normal
Urine Cr/Crtn ratio	Elevated ↑	Elevated ↑	Markedly elevated ↑↑↑
Response to oral Cr	Partial (needs trident)	Excellent	Minimal in males

MRS as a Longitudinal Treatment Monitor

Serial MRS is invaluable for assessing treatment response. In GAMT and AGAT deficiency, oral creatine supplementation progressively restores the cerebral creatine peak to near-normal over 3–12 months, and this MRS normalisation correlates directly with clinical improvement in seizure frequency, language acquisition, and behavioural measures. The MRS response curve effectively validates treatment adequacy and dose sufficiency.

In SLC6A8 deficiency, the creatine peak remains absent or severely reduced despite oral supplementation — the scanner shows no improvement. This MRS non-response is mechanistically expected and is not a sign of poor adherence or inadequate dosing. It is the most important finding to communicate clearly and compassionately to families who are waiting for clinical improvement.

🦪 Oyster: If a patient with an apparent diagnosis of GAMT or AGAT deficiency shows no MRS improvement in cerebral creatine after 12 months of documented adequate supplementation, revisit the diagnosis and strongly consider SLC6A8 deficiency. Initial biochemical differentiation can be imprecise if plasma GAA was borderline, the urine creatine/creatinine was not optimally timed, or the molecular result was awaited but acted upon presumptively. MRS non-response is the functional stress test that forces diagnostic re-evaluation.

---

Section 5 — Creatine Supplementation Protocols: Dosing, Monitoring, and Resistance Mechanisms

Standard Dosing Across the CCDS

The foundation of treatment across all three CCDS is oral creatine monohydrate. Dosing recommendations:

Population	Starting Dose	Target Maintenance	Divided Doses
Children (GAMT/AGAT)	0.3 g/kg/day	400–800 mg/kg/day	2–3 times daily
Adults (all CCDS)	Loading: 0.3 g/kg/day × 5–7 days	3–5 g/day	Twice daily
SLC6A8 females	0.3 g/kg/day	400 mg/kg/day	Twice daily

Creatine monohydrate powder dissolved in any beverage is bioequivalent to encapsulated formulations and substantially cheaper — an important consideration for lifelong treatment. Creatine ethyl ester, creatine hydrochloride, and buffered creatine formulations are heavily marketed but carry no evidence of superiority in CCDS and lack the evidence base of the monohydrate. Do not be swayed by sports supplement marketing when prescribing for metabolic disease.

🔧 Hack: Creatine monohydrate is essentially tasteless and dissolves completely in warm or cold beverage. Splitting the daily dose into morning and evening administrations maintains more stable serum and tissue creatine concentrations than single daily bolus dosing. For children, dissolving the morning dose in fruit juice at breakfast and the evening dose in warm milk achieves near-perfect adherence with zero palatability issues. Many families report better compliance with creatine than with any pharmaceutical agent they have been prescribed.

The GAMT Trident Protocol: Detailed Management

GAMT deficiency demands simultaneous pursuit of all three therapeutic arms. Creatine monotherapy in GAMT is insufficient — it replenishes creatine but does nothing to lower the toxic GAA that continues to accumulate.

Arm 1 — Creatine monohydrate (400–800 mg/kg/day, divided 2–3 times): Replenishes cerebral creatine pool. MRS normalisation expected at 3–12 months.

Arm 2 — L-ornithine (100–200 mg/kg/day, divided 2–3 times): Competitively inhibits AGAT, diverting it from the arginine-glycine reaction to the ornithine-glycine reaction, directly reducing GAA biosynthesis at source. Target: plasma GAA normalisation within 6 months.

Arm 3 — Dietary arginine restriction: Reduces substrate availability for residual AGAT activity. Target plasma arginine to lower quartile of age-matched normal range. Dietitian involvement is essential. Avoid arginine deficiency — monitor plasma amino acid profile 3-monthly.

Monitoring in GAMT: Plasma GAA and arginine 3-monthly; plasma creatine 3-monthly; urine amino acids (to exclude ornithine-induced aminoaciduria) quarterly; formal neuropsychological assessment annually; brain MRS at baseline, 12 months, then every 3 years.

SLC6A8 Deficiency: The Therapeutic Frontier

The fundamental therapeutic challenge in SLC6A8 deficiency is biological: the transporter defect prevents creatine crossing the blood-brain barrier regardless of peripheral creatine concentration. Several bypass strategies have been explored:

Creatine precursor supplementation (arginine + glycine): Supplies GAA biosynthesis substrates in excess, hoping that small amounts of locally synthesised creatine within neurons — via residual or alternative transport mechanisms — might partially restore the cerebral pool. Human trial data show modest and inconsistent benefit.

Cyclocreatine: A phosphocreatine analogue that can be transported into the brain via alternative carriers independent of SLC6A8. Demonstrates promising efficacy in mouse models. Early-phase human trials are underway. Currently investigational.

DHA (docosahexaenoic acid) supplementation: Marginal cognitive and behavioural benefit observed in observational studies, presumed to operate through non-creatine-dependent neuroprotective mechanisms. Low-risk as adjunct therapy while awaiting definitive treatments.

💎 Clinical Pearl: Do not perpetuate futile oral creatine supplementation indefinitely in males with confirmed SLC6A8 deficiency. The absence of clinical response is not a compliance failure — it is mechanistically inevitable. After 12 months of supplementation with no MRS evidence of cerebral creatine improvement and no clinical change, have an honest, compassionate conversation with the family. Redirect therapeutic energies toward robust educational support, speech-language therapy, intensive behavioural intervention, and consideration of enrolment in clinical trials of cyclocreatine or other novel agents. This is not therapeutic nihilism; it is precision medicine applied with integrity.

The Resistance Mechanism Ladder

When a patient fails to respond to creatine supplementation, a structured diagnostic ladder prevents premature therapeutic surrender and missed re-diagnoses:

Step 1 — Confirm adherence: Measure plasma creatine and 24-hour urine creatinine. Both should be elevated with adequate supplementation. Failure to rise suggests non-adherence or malabsorption.

Step 2 — Exclude malabsorption: Creatine absorption is impaired by concurrent gastrointestinal pathology. Consider a gastric emptying study or formulation change (powder to liquid suspension) in patients with GI comorbidities.

Step 3 — Re-examine the molecular diagnosis: Could this patient have SLC6A8 deficiency miscategorised as GAMT or AGAT based on borderline biochemistry or delayed molecular results? Brain MRS non-response is the functional stress test. Re-sequence if necessary.

Step 4 — Assess GAA reduction adequacy (GAMT only): Persistent plasma GAA elevation despite ornithine supplementation suggests dose inadequacy, poor adherence to arginine restriction, or dietary arginine excess that requires quantification via a 3-day diet diary analysed by a metabolic dietitian.

🔧 Hack: In outpatient metabolic clinics, a spot urine creatine/creatinine ratio at every review visit serves as a rapid and inexpensive proxy for both treatment adherence and diagnostic categorisation in CCDS. In AGAT and GAMT patients on treatment, the ratio should decrease toward normal as cerebral creatine replenishment occurs. In SLC6A8 patients, it remains elevated regardless of supplementation — a persistent elevated ratio in a GAMT/AGAT patient after 12 months should trigger adherence reassessment before attributing treatment failure to the diagnosis.

Monitoring Framework: What Every Follow-Up Visit Must Include

Parameter	Frequency	Target / Action
Plasma creatine	3-monthly (yr 1), 6-monthly thereafter	Low-normal; dose-adjust if subtherapeutic
Plasma GAA (GAMT only)	3-monthly	Normalise to age-matched range
Plasma arginine (GAMT only)	3-monthly	Lower quartile of normal; avoid deficiency
Urine creatine/creatinine	3-monthly	Trending toward normal (GAMT/AGAT)
Urine amino acids (GAMT only)	3-monthly	Exclude ornithine-induced aminoaciduria
eGFR / serum creatinine	Annually	Baseline surveillance; normal renal function not a concern at therapeutic doses
Neuropsychological assessment	Annually (children)	Track cognitive and language trajectory
Brain ¹H-MRS	Baseline, 12 months, 3-yearly	Confirm cerebral creatine repletion; MRS non-response re-opens differential

A note on renal safety: Creatine monohydrate at therapeutic doses in patients with normal baseline renal function has not been associated with nephrotoxicity in any rigorous long-term clinical study. Annual eGFR monitoring is appropriate surveillance but should not generate anxiety or drive dose reduction in the absence of objective renal impairment.

---

Conclusion: Closing the Diagnostic Gap

The cerebral creatine deficiency syndromes sit at a clinically important but under-attended intersection of neurology, metabolism, and general medicine. They are rare enough to be unfamiliar, yet common enough in aggregate to appear in every busy intellectual disability clinic, epilepsy unit, and neurodevelopmental service. The diagnostic toolkit is simple, cheap, and broadly available. The bottleneck has always been clinical suspicion.

With GAMT and AGAT deficiency, early diagnosis and targeted treatment transforms a devastating progressive neurological disorder into a condition compatible with near-normal cognitive development. With SLC6A8 deficiency, timely diagnosis redirects therapeutic effort from futile supplementation toward supportive intervention, family counselling, and clinical trial access. In every case, the cascade diagnosis of affected relatives — particularly the heterozygous females so consistently overlooked in SLC6A8 pedigrees — extends the diagnostic dividend far beyond the index patient.

The internist who adds CCDS to their differential diagnostic repertoire, who interrogates every brain MRS report for the creatine peak and the 3.78 ppm GAA signal, and who reflexively orders a urine creatine/creatinine ratio and plasma GAA in any unexplained intellectual disability will make diagnoses that are genuinely, measurably, and sometimes completely life-altering. That is the standard to which this review aspires to elevate its readership.

---

Ten Clinical Take-Home Points

1. GAMT deficiency causes both creatine depletion AND toxic GAA accumulation — two distinct, simultaneous pathological mechanisms requiring a trident treatment approach
2. Extrapyramidal features (dystonia, choreoathetosis) at the bedside strongly discriminate GAMT from AGAT and SLC6A8 deficiency
3. The GAA peak at 3.78 ppm on ¹H-MRS is pathognomonic for GAMT deficiency — a diagnosis you can make at the scanner
4. AGAT deficiency responds to creatine monohydrate monotherapy — the most gratifying CCDS to diagnose and treat
5. Low (not elevated) plasma GAA is the biochemical fingerprint of AGAT deficiency — train yourself to look at absolute values
6. Urine creatine/creatinine ratio is unreliable for diagnosing heterozygous SLC6A8 females — molecular testing is mandatory
7. Up to 50% of SLC6A8 heterozygous females have clinically significant neurocognitive symptoms — they are in every clinic
8. Oral creatine is mechanistically futile in SLC6A8 males — MRS non-response is not a compliance failure, it is expected
9. GAMT management requires creatine + ornithine + arginine restriction; ornithine overdosing causes secondary aminoaciduria
10. Average diagnostic delay exceeds 5 years — a simple urine creatine/creatinine ratio and plasma GAA can close this gap today

---

References

1. Mercimek-Mahmutoglu S, Salomons GS. Creatine deficiency syndromes. In: Adam MP, et al., eds. GeneReviews [Internet]. Seattle: University of Washington; 2009 (Updated 2015). NCBI Bookshelf NBK3794.

2. Schulze A. Creatine deficiency syndromes. Mol Cell Biochem. 2003;244(1–2):143–150.

3. Stöckler S, Schutz PW, Salomons GS. Cerebral creatine deficiency syndromes: clinical aspects, treatment and pathophysiology. Subcell Biochem. 2007;46:149–166.

4. Battini R, Alessandrì MG, Leuzzi V, et al. Arginine:glycine amidinotransferase (AGAT) deficiency in a newborn: early treatment can prevent phenotypic expression of the disease. J Pediatr. 2006;148(6):828–830.

5. Schulze A, Battini R. Pre-symptomatic treatment of creatine biosynthesis defects. Subcell Biochem. 2007;46:167–181.

6. Sijens PE, Verbruggen KT, Meiners LC, et al. ¹H chemical shift imaging in guanidinoacetate methyltransferase deficiency: revealing abnormal creatine and guanidinoacetate metabolism. Magn Reson Med. 2005;53(6):1316–1321.

7. Braissant O, Henry H, Béard E, Uldry J. Creatine deficiency in the brain: animal models and implications for human diseases. Amino Acids. 2011;40(5):1357–1365.

8. Salomons GS, van Dooren SJM, Verhoeven NM, et al. X-linked creatine-transporter gene (SLC6A8) defect: a new creatine-deficiency syndrome. Am J Hum Genet. 2001;68(6):1497–1500.

9. van de Kamp JM, Mancini GMS, Salomons GS. X-linked creatine transporter deficiency: clinical aspects and pathophysiology. J Inherit Metab Dis. 2014;37(5):715–733.

10. Nasrallah F, Feki M, Kaabachi N. Creatine and creatine deficiency syndromes: biochemical and clinical aspects. Pediatr Neurol. 2010;42(3):163–171.

11. Item CB, Stöckler-Ipsiroglu S, Stromberger C, et al. Arginine:glycine amidinotransferase deficiency: the third inborn error of creatine metabolism in humans. Am J Hum Genet. 2001;69(5):1127–1133.

12. Valayannopoulos V, Boddaert N, Chabli A, et al. Treatment by oral creatine, L-arginine and L-glycine in six French patients with creatine transporter defect. J Inherit Metab Dis. 2012;35(1):151–157.

13. Leuzzi V, Mastrangelo M, Battini R, Cioni G. Inborn errors of creatine metabolism and epilepsy. Epilepsia. 2013;54(2):217–227.

14. Stromberger C, Bodamer OA, Stöckler-Ipsiroglu S. Clinical characteristics and diagnostic clues in inborn errors of creatine metabolism. J Inherit Metab Dis. 2003;26(2–3):299–308.

15. Newmeyer A, Cecil KM, Schapiro M, Clark JF, Degrauw TJ. Incidence of brain creatine transporter deficiency in males with developmental delay referred for brain MRI. J Dev Behav Pediatr. 2005;26(4):276–282.

16. Dunbar M, Jaggumantri S, Sargent M, Stockler-Ipsiroglu S, van Karnebeek CD. Treatment of X-linked creatine transporter (SLC6A8) deficiency: systematic review of the literature and three new cases. Mol Genet Metab. 2014;112(4):259–274.

17. Mercimek-Andrews S, Salomons GS. Guanidinoacetate methyltransferase deficiency. Updated 2015. In: Adam MP et al., eds. GeneReviews. NCBI Bookshelf NBK1227.

---

This article is for educational purposes. Clinical decisions should be made in consultation with qualified metabolic specialists.
Word count (main text): approximately 3,050 words

Internal Medicine | Neurometabolic Disease The Disorders of Mitochondrial Dynamics

 REVIEW ARTICLE

Internal Medicine | Neurometabolic Disease

 

The Disorders of Mitochondrial Dynamics:

A Clinician's Guide to Bedside Recognition, Precision Diagnosis,

and Emerging Therapeutic Horizons

Dr Neeraj Manikath , claude.ai

 

Correspondence: Department of Internal Medicine & Neurometabolic Disease

J Adv Intern Med. 2025;14(3):e1-e18. DOI: 10.xxxx/jaim.2025.mit-dyn

Received: January 2025 | Accepted: March 2025 | Published: April 2025

 

ABSTRACT

Mitochondrial dynamics—the continuous cycle of organelle fusion and fission—is governed by a family of dynamin-related GTPases whose mutations produce a spectrum of neurological and multisystem disorders. Five gene-disease pairs dominate the clinical landscape: OPA1 mutations causing Dominant Optic Atrophy 'Plus'; MFN2 mutations causing Charcot-Marie-Tooth Type 2A with optic and central nervous system involvement; DNM1L (DRP1) mutations causing a severe neonatal/infantile encephalopathy; GDAP1 mutations causing mixed axonal-demyelinating neuropathy with vocal cord paresis; and MFF mutations causing encephalopathy with basal ganglia involvement. Collectively, these conditions remain profoundly under-recognised, with diagnostic delays of 5–20 years being the norm rather than the exception. This review provides the internist and neurologist with a bedside-oriented framework for clinical recognition, a structured diagnostic algorithm, a practical approach to the expanding therapeutic landscape, and a series of clinical pearls, oysters, and bedside hacks distilled from expert clinical experience. We address nuances including the misdiagnosis traps, genotype-phenotype variability, the role of mtDNA instability as a secondary feature, and the emerging therapeutic pipeline. A particular emphasis is placed on the pattern-recognition skills that allow early diagnosis before confirmatory genetic testing, which can reduce the diagnostic odyssey from years to months.

KEY WORDS: OPA1; MFN2; DRP1; DNM1L; GDAP1; MFF; mitochondrial fusion; mitochondrial fission; Dominant Optic Atrophy Plus; Charcot-Marie-Tooth Type 2A; mitochondrial encephalopathy; precision medicine; neurometabolic disease

 

1. Introduction and Pathophysiological Framework

Mitochondria are not static organelles. They exist within cells as dynamic, interconnected networks, constantly undergoing cycles of fusion—the merging of two mitochondria into one—and fission—the division of a single mitochondrion into two daughter organelles. This mitochondrial dynamic equilibrium, far from being a morphological curiosity, is fundamental to mitochondrial quality control, calcium buffering, reactive oxygen species handling, apoptosis regulation, and the faithful distribution of mitochondrial DNA (mtDNA) during cell division. When the molecular machinery governing this equilibrium is disrupted by germline mutations, the consequences range from isolated optic neuropathy to devastating neonatal multi-organ failure.

The core fusion machinery comprises the outer mitochondrial membrane (OMM) GTPases Mitofusin-1 (MFN1, gene MFN1) and Mitofusin-2 (MFN2, gene MFN2), and the inner mitochondrial membrane (IMM) GTPase Optic Atrophy protein 1 (OPA1, gene OPA1). Fission is mediated primarily by Dynamin-Related Protein 1 (DRP1, gene DNM1L), recruited to the OMM by receptor/adapter proteins including MFF (Mitochondrial Fission Factor, gene MFF), FIS1, MiD49, and MiD51. GDAP1 (Ganglioside-Induced Differentiation-Associated Protein 1) localises to the OMM and regulates both fission and interaction with the ER-mitochondria tethering complex.

A useful clinical heuristic is the fusion-defect phenotype versus the fission-defect phenotype. Fusion defects (OPA1, MFN2) tend to produce conditions characterised by progressive neurodegeneration, axonal neuropathy, optic atrophy, and secondary mtDNA depletion or multiple deletions. Fission defects (DRP1/DNM1L, MFF) tend to produce more severe, earlier-onset encephalopathies with brain malformations and lactic acidosis, reflecting the inability to isolate and eliminate damaged mitochondria by the mitophagy pathway. GDAP1, straddling both processes, produces a distinctive mixed neuropathy phenotype.

 

⭑ CLINICAL PEARL: Fusion vs. Fission at the Bedside

When you encounter a patient with progressive optic atrophy + peripheral neuropathy, think FUSION defect (OPA1, MFN2). When you encounter a neonate or infant with severe encephalopathy, microcephaly, and lactic acidosis, think FISSION defect (DRP1/DNM1L, MFF). This single distinction will focus your genetic work-up and prevent expensive shotgun testing.

 

2. OPA1 Mutations: Dominant Optic Atrophy and the 'Plus' Phenotype

2.1 Genetics and Epidemiology

OPA1 encodes a dynamin-related GTPase critical for IMM fusion and cristae remodelling. With a prevalence of approximately 1 in 30,000, autosomal dominant optic atrophy (DOA) caused by OPA1 mutations (OMIM #165500) represents the most common inherited optic neuropathy, exceeding the prevalence of Leber's hereditary optic neuropathy (LHON). Over 500 pathogenic variants are documented; approximately 60% cause haploinsufficiency through frameshift, nonsense, or splice-site mutations, while approximately 40% are missense variants, many clustering in the GTPase domain.

2.2 Clinical Spectrum: From Isolated to 'Plus'

The classical presentation is bilateral, insidious visual loss beginning in the first two decades of life, with centrocaecal scotomata, loss of colour vision, and temporal disc pallor—the optic disc pallor characteristically affects the temporal segment first, corresponding to the maculopapillary bundle. Visual acuity ranges from near-normal (6/9) to severe impairment (counting fingers), but rarely reaches no-perception-of-light, a useful discriminator from LHON which can be more acute and severe.

Approximately 20% of OPA1 patients—particularly those harbouring missense variants in the GTPase or dynamin central (GED) domains—develop extra-ocular features constituting the DOA-plus syndrome: sensorineural hearing loss (SNHL), cerebellar ataxia, peripheral neuropathy (predominantly axonal), myopathy with exercise intolerance, parkinsonism, and rarely multiple sclerosis-like demyelination. Ptosis and progressive external ophthalmoplegia (PEO) occur in a subset, overlapping with the classic mitochondrial PEO phenotype secondary to OPA1-driven mtDNA instability.

 

◆ CLINICAL OYSTER: The Ophthalmology-Neurology Disconnect

Many OPA1 patients are followed exclusively in ophthalmology for years before the 'plus' features are recognised. The pearl: any patient with established DOA who develops new neurological symptoms—unexplained fatigue, exercise intolerance, hearing loss, imbalance—should be formally reassessed for DOA-plus. Conversely, any adult with 'idiopathic axonal neuropathy + optic atrophy' should have OPA1 sequenced before being labelled cryptogenic.

 

⭑ CLINICAL PEARL: The Temporal Pallor Sign

In a young patient with unexplained visual decline, dial your ophthalmoscope to the optic disc and look specifically at the temporal half. Temporal pallor in isolation, with preserved nasal margin colour, is the signature of DOA (as opposed to the global pallor of optic neuritis sequelae or the pseudopapilloedema of drusen). If you can see this pattern, you have a provisional diagnosis before any test is ordered.

 

2.3 Secondary mtDNA Instability: A Critical Nuance

A conceptually important feature of OPA1 mutations is the induction of secondary mtDNA multiple deletions in post-mitotic tissues such as skeletal muscle. This occurs because OPA1-driven IMM fusion is required for the complementation and quality control of mtDNA. When OPA1 function is reduced, mtDNA segments become isolated in fragmented mitochondria and accumulate deletions. This secondary mtDNA instability explains the PEO phenotype and can produce muscle biopsy findings of ragged-red fibres and COX-negative fibres—findings which, if encountered without a prior genetic diagnosis, may misdirect the clinician toward primary mtDNA diseases.

 

◉ BEDSIDE HACK

Muscle Biopsy Misdirection in OPA1 Disease: If a patient's muscle biopsy shows ragged-red fibres and multiple mtDNA deletions but NO single large-scale mtDNA deletion and NO pathogenic mtDNA point mutation, always request OPA1 sequencing as part of the nuclear gene panel. Failing to do so will leave the diagnosis in limbo for years. The formula: multiple mtDNA deletions + optic atrophy = nuclear gene screen first.

 

3. MFN2 Mutations: Charcot-Marie-Tooth Type 2A and Beyond

3.1 Genetics

MFN2 (Mitofusin-2) is the most abundant mitofusin in the nervous system and mediates OMM fusion, ER-mitochondria tethering, and axonal mitochondrial trafficking. MFN2 mutations cause autosomal dominant CMT type 2A (CMT2A, OMIM #609260), accounting for approximately 20% of all CMT2 cases—making it the most common genetic cause of axonal CMT. De novo mutations account for a significant minority, and rare recessive cases exist. The MFN2 coiled-coil region (heptad repeat domains HR1 and HR2) and the GTPase domain are mutational hotspots.

3.2 Phenotypic Range and the Central Nervous System Involvement

The classical CMT2A phenotype is a length-dependent, predominantly motor axonal neuropathy with onset typically in the first decade. Foot deformity (pes cavus, hammer toes), distal weakness and wasting, and reduced or absent deep tendon reflexes are cardinal features. The clinical severity ranges from severe, with wheelchair dependency in childhood, to remarkably mild with preserved ambulation into the sixth decade.

The critical insight for the internist is that MFN2-CMT2A is not confined to the peripheral nervous system. A clinically significant subset develops optic atrophy (CMT2A with optic atrophy, formerly called HMSN VI), and a further subset shows white matter abnormalities on brain MRI, pyramidal tract signs, and cerebellar involvement. This CNS involvement, when present, creates a complex multisystem phenotype that overlaps with hereditary spastic paraplegia and leukodystrophy and is frequently misdiagnosed.

 

◆ CLINICAL OYSTER: MFN2 White Matter Disease Masquerading as MS

MFN2 mutations can produce periventricular and subcortical white matter T2-hyperintensities on MRI that are mistaken for multiple sclerosis, particularly in young women presenting with gait difficulty. The differentiating feature: the neuropathy in MFN2 is a pure axonal pattern (low CMAP amplitudes, near-normal conduction velocities, no conduction block)—a pattern incompatible with primary demyelinating MS-related neuropathy. Check nerve conduction studies before attributing white matter lesions to MS in any young patient with an axonal neuropathy.

 

3.3 Axonal Mitochondrial Transport: The Pathomechanistic Key

MFN2 anchors mitochondria to the MIRO-TRAK complex in axons. Loss of MFN2 function disrupts anterograde and retrograde axonal mitochondrial transport, leading to a paucity of mitochondria at distal axonal terminals—the presumed mechanism for the length-dependent neurodegeneration. This is functionally distinct from the pure fusion defect, and explains why MFN2 disease can be more severe than MFN1 disease despite both mediating OMM fusion.

 

◉ BEDSIDE HACK

The NCS Signature of MFN2: In a patient with CMT phenotype, if the nerve conduction studies show markedly reduced CMAP amplitudes with only mildly reduced conduction velocity (e.g., median motor CV > 38 m/s, CMAP < 1mV), this axonal pattern—especially with early onset and severe degree—has a >70% likelihood of being MFN2 in reported series. Send MFN2 first in this pattern rather than waiting for panel results. Time saved = critical counselling earlier.

 

⭑ CLINICAL PEARL: Examining the Parents in CMT2A

Because penetrance of MFN2 mutations is variable, the affected parent may be minimally symptomatic. In clinical practice, examining the parents for subtle foot deformity, distal wasting, and performing a brief neurological exam including deep tendon reflex assessment and vibration sense testing at the medial malleolus can reveal the inheritance pattern before genetic results are available. A positive family examination is itself clinically diagnostic and enables cascade testing.

 

4. DNM1L (DRP1) Mutations: Severe Neonatal/Infantile Encephalopathy

4.1 Genetics and Pathomechanism

Dynamin-Related Protein 1 (DRP1), encoded by DNM1L, is the master regulator of mitochondrial fission. DRP1 is a cytosolic protein that is recruited to the OMM by adaptor proteins (MFF, FIS1, MiD49/51), where it oligomerises into a helical ring structure and constricts the membrane to complete fission. DNM1L mutations (OMIM #614388) cause autosomal dominant (usually de novo) or rarely recessive severe encephalopathy with impaired neuronal mitochondrial distribution. The mutations are predominantly missense, affecting the GTPase domain (reducing GTPase activity) or the GED/stalk domain (disrupting oligomerisation), resulting in fission failure and the appearance of hyperfused, elongated mitochondrial networks.

4.2 Clinical Phenotype: The Encephalopathy of Fission Failure

DRP1 deficiency presents in the neonatal period or early infancy with a devastating phenotype: severe hypotonia, seizures (often refractory), microcephaly (which may be progressive), lactic acidosis, hyperammonaemia, and elevated plasma very-long-chain fatty acids in some cases—reflecting involvement of the peroxisomal pathway due to MFF-dependent peroxisomal fission. Brain MRI shows simplified gyration, abnormal myelination, optic atrophy, and cerebellar hypoplasia. Most affected infants do not survive beyond the first years of life.

 

⭑ CLINICAL PEARL: The Peroxisomal Clue in DRP1 Disease

DRP1 is required for both mitochondrial AND peroxisomal fission—peroxisomes also use MFF and DRP1 for division. Consequently, some DRP1-deficient patients have elevated very-long-chain fatty acids (VLCFAs) and elevated pipecolic acid, mimicking a peroxisomal biogenesis disorder. When you encounter a neonate with encephalopathy, elevated VLCFAs, AND lactic acidosis, do not assume a primary peroxisomal disorder—request DNM1L sequencing. The combination of both organellar signatures is pathognomonic of fission machinery failure.

 

4.3 Critical Management Pitfall: Valproate

Valproate (valproic acid, VPA) is a potent inhibitor of mitochondrial beta-oxidation and OXPHOS and is absolutely contraindicated in all primary mitochondrial disorders. In DRP1 deficiency, where mitochondrial function is already critically compromised, VPA can precipitate acute liver failure and accelerated neurological deterioration. Seizures in these patients should be managed with mitochondrially-safe agents such as levetiracetam, lamotrigine, or phenobarbital.

 

◉ BEDSIDE HACK

The AVOID List for Mitochondrial Dynamic Disorders: Commit to memory the drugs to avoid in all patients with confirmed or suspected mitochondrial dynamic disorders: (1) Valproate—mitochondrial toxin, risk of acute liver failure; (2) Metformin—inhibits complex I, may precipitate lactic acidosis particularly in OPA1 and MFN2; (3) Statins—inhibit CoQ10 synthesis and may worsen myopathy; (4) Aminoglycosides—may worsen SNHL in OPA1-plus; (5) Linezolid—mitochondrial protein synthesis inhibitor; (6) Nucleoside reverse transcriptase inhibitors—deplete mtDNA. Documenting these as allergies/contraindications in the electronic health record protects patients from inadvertent prescribing by other clinicians.

 

5. GDAP1 Mutations: Axonal and Demyelinating Neuropathy with Vocal Cord Paresis

5.1 Genetics and Unique Pathobiology

GDAP1 (Ganglioside-Induced Differentiation-Associated Protein 1) is an atypical member of the glutathione S-transferase (GST) superfamily, anchored to the OMM by a single transmembrane domain. Its functions include regulation of mitochondrial fission, maintenance of the ER-mitochondria calcium transfer interface, and modulation of reactive oxygen species. GDAP1 mutations (OMIM #214400, #607706, #607831) cause autosomal recessive CMT4A (severe, demyelinating) or CMT2K (axonal, milder) or the rare autosomal dominant CMTDIB form. The recessive forms are particularly prevalent in Mediterranean populations (Spain, Tunisia, Turkey) due to founder effects.

5.2 The Distinguishing Clinical Feature: Vocal Cord Paresis

The feature that should immediately prompt GDAP1 testing is the combination of peripheral neuropathy with hoarseness or dysphonia due to vocal cord paresis. This extra-somatic manifestation—involvement of the recurrent laryngeal nerve-innervated laryngeal muscles—is present in approximately 20–30% of GDAP1 patients with recessive mutations and is rarely seen in other CMT subtypes. Associated features include diaphragmatic paresis (causing orthopnoea and sleep-disordered breathing), and respiratory muscle weakness requiring NIV.

 

◆ CLINICAL OYSTER: Vocal Cord Paresis in CMT: Think GDAP1 First

When a patient with hereditary neuropathy develops new hoarseness, the reflex is to search for a space-occupying lesion compressing the recurrent laryngeal nerve. While this must be excluded, the clinician who recognises 'hereditary neuropathy + new hoarseness' as a phenotypic signature of GDAP1 disease will save the patient a diagnostic odyssey. The hoarseness in GDAP1 is progressive, bilateral, and accompanied by respiratory muscle involvement. Laryngoscopy showing bilateral cord paresis without a structural cause = GDAP1 until proven otherwise.

 

5.3 Mixed Electrophysiological Pattern

A critical bedside and electrophysiological nuance: GDAP1 recessive mutations produce a mixed axonal-demyelinating neuropathy pattern on nerve conduction studies. Conduction velocities are intermediate (between the severe slowing of CMT1 and the preserved velocities of CMT2), with reduced amplitudes. This intermediate NCS pattern—which mimics CMT1B (MPZ mutations), CMTX, or hereditary neuropathy with liability to pressure palsies (HNPP)—can be clarified by recognising the co-occurrence of vocal cord paresis, early and severe onset, and consanguinity.

 

⭑ CLINICAL PEARL: Respiratory Monitoring in GDAP1 Disease

All confirmed GDAP1 patients, particularly those with recessive severe (CMT4A) mutations, require annual respiratory function testing: FVC sitting and lying (supine FVC drop > 10% = diaphragmatic weakness), overnight oximetry, and laryngoscopy. The threshold for nocturnal NIV initiation should be low: FVC < 60% or symptomatic nocturnal hypoventilation. Coordinated care with respiratory medicine is mandatory, not optional.

 

6. Mitochondrial Fission Factor (MFF) Deficiency: Encephalopathy with Basal Ganglia Involvement

6.1 Genetics and Clinical Features

MFF (Mitochondrial Fission Factor), encoded by the MFF gene on chromosome 2q36.1, is an OMM adaptor protein that recruits DRP1 to fission sites. Autosomal recessive MFF deficiency (OMIM #617086) causes a progressive encephalopathy with prominent basal ganglia involvement, spastic paraplegia, and neuroimaging findings of basal ganglia T2 hyperintensity and restricted diffusion resembling Leigh syndrome or striatal necrosis. The phenotype can also include optic atrophy and peripheral neuropathy, making it phenotypically overlapping with OPA1 and DRP1 diseases.

6.2 Neuroimaging as a Diagnostic Clue

The MRI signature of MFF deficiency provides a distinctive clue: bilateral symmetrical basal ganglia abnormalities (particularly the putamen and globus pallidus), cortical atrophy disproportionate to age, and cerebral white matter signal change. Magnetic resonance spectroscopy (MRS) may show elevated lactate peaks and reduced NAA. This imaging pattern, when combined with the clinical context of progressive spastic paraplegia in a child from a consanguineous family, should immediately prompt MFF sequencing.

 

⭑ CLINICAL PEARL: The Basal Ganglia-Metabolic Disease Matrix

When you encounter symmetric basal ganglia signal change on MRI, the differential should always include metabolic causes alongside structural: (1) Leigh syndrome (mitochondrial complex deficiency); (2) Biotin-thiamine responsive basal ganglia disease (SLC19A3 mutations—treatable!); (3) Organic acidurias (methylmalonic, propionic); (4) Wilson disease; (5) Mitochondrial fission defects (DRP1, MFF). Symmetry is the key—asymmetric lesions favour structural/vascular causes. In MFF disease, the basal ganglia involvement occurs against a background of global developmental delay and lactic acidosis, differentiating it from the reversible encephalopathy of biotin-responsive disease.

 

◉ BEDSIDE HACK

The Biotin Empirical Trial in Basal Ganglia Encephalopathy: In any child with subacute encephalopathy and bilateral basal ganglia lesions of uncertain aetiology, initiate high-dose biotin (5-10 mg/kg/day) and thiamine (100-300 mg/day) empirically while awaiting metabolic and genetic results. This is the treatment for biotin-thiamine responsive basal ganglia disease (BTBGD) and has dramatic, life-saving benefit if the diagnosis is correct. If the condition is MFF deficiency, this empirical treatment carries no harm and does not delay accurate diagnosis.

 

7. A Structured Diagnostic Approach

The diagnostic odyssey in these conditions—averaging 7–12 years across registries—reflects both disease rarity and the multi-specialty nature of the phenotypes. The following framework, built on pattern recognition and stepwise testing, can compress this odyssey to 2–4 months in most cases.

 

Table 1: Step-by-Step Diagnostic Algorithm for Suspected Mitochondrial Dynamic Disorder

Step	Action	Clinical Tip
1. Pattern Recognition	Identify multi-system neurological syndrome: optic atrophy + neuropathy + hearing loss + myopathy	The 'MOAN' mnemonic: Myopathy, Optic Atrophy, Ataxia, Neuropathy – any 2 = screen for mitochondrial dynamic disorder
2. Family History	3-generation pedigree with penetrance analysis; examine parents for subclinical features	Examine the 'asymptomatic' parent – fundoscopy may reveal pallor; nerve conduction may be abnormal
3. Metabolic Screen	Fasting plasma lactate + pyruvate ratio, plasma amino acids, urine organic acids, CK, LFTs	Lactate may be NORMAL at rest in OPA1 and MFN2 – exercise lactate or CSF lactate is more sensitive
4. Neuroimaging	Brain MRI with DWI, FLAIR, MRS; spinal cord imaging if indicated	Look for white matter signal in MFN2; basal ganglia T2 hyperintensity in MFF; optic nerve atrophy on MRI orbits
5. Electrophysiology	Nerve conduction studies + EMG; VEP; BAER; ERG if indicated	MFN2: pure axonal neuropathy (low amplitude, normal velocity). GDAP1: mixed axonal+demyelinating pattern
6. Functional Testing	Skin/muscle biopsy for respiratory chain enzymology and fibroblast mitochondrial morphology assays	Fibroblast cultures show fragmented mitochondria in DRP1 deficiency and elongated tubules in OPA1/MFN2 loss
7. Genetic Confirmation	Gene panel (OPA1, MFN2, DNM1L, GDAP1, MFF) or whole exome sequencing	Panel first if phenotype is specific; WES if panel negative – novel variants need functional validation
8. Cascade Testing	Test first-degree relatives once proband is confirmed; check penetrance and genotype-phenotype correlation	In OPA1: penetrance is ~88% for optic atrophy alone, but much lower for 'plus' features – counsel accordingly

 

8. Clinical and Molecular Comparison of the Five Disorders

 

Table 2: Comparative Clinical, Molecular, and Therapeutic Features

Gene / Disease	Inheritance	Cardinal Features	Key Test	Metabolic Clue	Therapy Direction
OPA1 / DOA Plus	AD	Optic atrophy, SNHL, ataxia, myopathy, parkinsonism	OPA1 sequencing + deletion	Plasma lactate (mild↑), CK	Idebenone (trials); CoQ10
MFN2 / CMT2A	AD (rarely AR)	Axonal neuropathy, optic atrophy, cerebellar ataxia, CNS white matter	MFN2 sequencing	Electrophysiology: axonal	Supportive; NAD+ precursors
DRP1 / EMPF1	AD/de novo	Severe encephalopathy, microcephaly, lactic acidosis, early death	DNM1L sequencing	Lactate↑↑, pyruvate↑	Supportive; avoid valproate
GDAP1 / CMT2K/4A	AR (AD rare)	Axonal+demyelinating neuropathy, vocal cord paresis, hoarseness	GDAP1 sequencing	EMG: mixed neuropathy	Supportive; vocal cord MGT
MFF / EMPF2	AR	Encephalopathy, basal ganglia involvement, spastic paraplegia	MFF sequencing	Lactate, MRI basal ganglia signal	Supportive; biotin trials

 

9. Therapeutic Landscape: Current and Emerging

9.1 Supportive and Symptomatic Management

For all five disorders, comprehensive multidisciplinary care remains the cornerstone of management. This includes regular ophthalmological surveillance (visual acuity, visual fields, OCT nerve fibre layer analysis) in OPA1 and MFN2 disease; neurological monitoring with serial neuropsychological assessment; audiological surveillance and hearing aid prescription when indicated; physiotherapy and orthopaedic management for neuropathy-related deformity; and respiratory surveillance in GDAP1 disease. Genetic counselling of affected individuals and at-risk family members is an ethical imperative.

9.2 Disease-Modifying Approaches

Idebenone, a short-chain benzoquinone with complex I bypass activity, is approved in several jurisdictions for LHON and has been used in DOA/OPA1 disease in open-label settings. Evidence from the LHON experience and mechanistic rationale support its use in OPA1 disease, though randomised controlled trial data specifically in DOA-plus are lacking. CoQ10 supplementation (10–30 mg/kg/day) is widely prescribed, with biological plausibility but limited trial evidence; it is generally well-tolerated and reasonable as an adjunct.

NAD+ precursor supplementation (nicotinamide riboside, NR, or nicotinamide mononucleotide, NMN) has strong mechanistic rationale in MFN2 disease, where impaired mitochondrial dynamics reduces NAD+ bioavailability and SIRT1/3-dependent mitochondrial biogenesis. Early phase clinical trials in CMT2A are underway. Exercise therapy, counter-intuitively, has evidence for benefit in mitochondrial myopathy (including endurance training at 60–70% VO2max), though high-intensity exercise causing rhabdomyolysis must be avoided.

 

◆ CLINICAL OYSTER: The Exercise Paradox in Mitochondrial Disease

The common clinical advice to restrict exercise in mitochondrial disease is now recognised as harmful. Aerobic exercise training at moderate intensity increases mitochondrial biogenesis, improves respiratory chain function, and reduces the heteroplasmy of mtDNA mutations in muscle. In OPA1 and MFN2 disease, tailored aerobic exercise (cycling, swimming at 60-75% peak heart rate for 30-45 min, 4-5 days/week) is now considered standard of care and should be prescribed formally rather than left to patient discretion. Resistance exercise has additional benefit for counteracting disuse atrophy.

 

9.3 Emerging Therapeutic Strategies

Gene therapy for OPA1 haploinsufficiency is the most advanced in the pipeline. Intravitreal AAV2-OPA1 injection has shown photoreceptor and retinal ganglion cell preservation in murine models, and phase I/II trials are in preparation. For MFN2 disease, AAV-mediated MFN2 gene replacement has restored mitochondrial transport and reduced neuropathy severity in mouse models. The challenge for systemic delivery remains achieving adequate transduction in distal axons.

Pharmacological enhancement of residual fusion activity is under investigation. Leflunomide, historically used as a DHODH inhibitor in rheumatoid arthritis, has been identified as an MFN activator and has shown benefit in CMT2A mouse models, providing a potential drug repurposing opportunity. Small molecule DRP1 inhibitors (e.g., Mdivi-1 and its derivatives) paradoxically show benefit in DRP1-gain-of-function models but would be harmful in DRP1 loss-of-function disease—illustrating the importance of molecular diagnosis before therapeutic application.

 

◉ BEDSIDE HACK

Precision Prescribing in Mitochondrial Dynamic Disorders—The Gene-Drug Matrix: Before prescribing any putative mitochondrial supplement, map the treatment to the molecular defect. CoQ10 is most rational in complex III/I defects (not clearly beneficial in fusion/fission disorders). Idebenone targets the visual pathway—relevant in OPA1 and MFN2 with optic atrophy, not in pure fission defects. NAD+ precursors target the Sirtuin axis—most rational in MFN2 disease where mitochondrial trafficking failure starves axon terminals of NAD+-generating capacity. Thiamine and riboflavin address specific co-factor deficiencies—irrelevant in structural GTPase defects. Matching the mechanism to the molecule avoids the 'mitochondrial cocktail' prescribing that provides no benefit and false reassurance.

 

10. Perioperative and High-Risk Clinical Scenarios

10.1 Anaesthetic Considerations

Patients with confirmed mitochondrial dynamic disorders require anaesthetic planning that minimises metabolic stress and avoids mitochondrial toxins. Key principles include: avoidance of prolonged fasting (maintain glucose-containing IV infusion perioperatively); avoidance of propofol infusion syndrome risk by limiting propofol infusions, particularly in children (though single-dose induction is generally safe); preference for regional anaesthesia over general anaesthesia where feasible; monitoring for lactic acidosis with serial blood gas analysis intra- and post-operatively; and avoidance of succinylcholine (risk of hyperkalaemia due to rhabdomyolysis in myopathic patients).

10.2 Pregnancy Management

Pregnancy in OPA1 and MFN2 disease is generally well-tolerated, as these are autosomal dominant conditions with germline penetrance. Preconception counselling must include accurate recurrence risk assessment (50% for autosomal dominant, 25% for recessive). Pre-implantation genetic diagnosis (PGD) is available and should be discussed. The metabolic demands of pregnancy may unmask or worsen subclinical disease manifestations, particularly the myopathy and exercise intolerance of OPA1-plus. Gestational diabetes screening is important given the intersection of mitochondrial function with insulin signalling.

 

⭑ CLINICAL PEARL: The Preconception Counselling Imperative

For every woman of childbearing age diagnosed with an autosomal dominant mitochondrial dynamic disorder (OPA1, MFN2), preconception counselling is not optional—it is a fundamental duty of care. The counselling must cover: (1) 50% recurrence risk per pregnancy; (2) highly variable penetrance (the child may be minimally affected or severely affected—neither can be predicted); (3) availability of PGD to select unaffected embryos; (4) prenatal diagnosis by chorionic villus sampling or amniocentesis; (5) the option of gamete donation. Document this counselling in the medical record.

 

11. Natural History and Prognostic Indicators

The natural history of these disorders is highly variable and genotype-phenotype correlations, while imperfect, provide useful prognostic guidance. In OPA1 disease, haploinsufficiency variants (truncating/frameshift) have lower risk of 'plus' features compared to missense variants in the GTPase domain. In MFN2 disease, younger age of onset and CMT2A score > 15 at 10 years predict more severe disability at 30 years. In DRP1/DNM1L disease, survival beyond the first decade is rare in the classical neonatal presentation, though milder late-onset cases exist.

Key biomarkers under development include plasma neurofilament light chain (NfL), which reflects acute axonal injury and shows promise as a disease activity and progression biomarker in CMT2A and OPA1-plus; retinal nerve fibre layer (RNFL) thickness measured by OCT, which correlates with visual prognosis and disease duration in DOA; and plasma GDF-15, which is elevated in mitochondrial disorders broadly and may serve as a diagnostic and therapeutic response biomarker.

 

⭑ CLINICAL PEARL: OCT as a Monitoring Tool in OPA1 Disease

Optical coherence tomography (OCT) of the retinal nerve fibre layer (RNFL) is inexpensive, non-invasive, and provides a quantitative, reproducible measure of retinal ganglion cell and axon loss in OPA1 disease. Annual OCT enables early detection of acceleration in neurodegeneration—a trigger to escalate therapy or enrol in a trial. The RNFL thickness correlates with visual acuity and visual field loss and should be part of the mandatory annual review for every OPA1 patient.

 

 

12. Conclusion: The Informed Clinician as the Rarest Resource

The disorders of mitochondrial dynamics represent a fascinating intersection of cell biology, clinical neurology, and internal medicine. Their diagnosis requires a clinician who can synthesise disparate findings—visual decline, a foot deformity noticed incidentally, a complaint of hoarseness, an abnormal MRI—into a coherent phenotypic narrative pointing to a specific molecular defect. The diagnostic delay these patients endure is not a failure of technology but a failure of clinical pattern recognition.

The post-graduate and consultant reader who assimilates the clinical pearls in this review—the temporal pallor sign, the fusion-fission heuristic, the vocal cord paresis signature, the peroxisomal clue in DRP1 disease, the AVOID list for mitochondrial toxins, the exercise prescription paradox—will be equipped to make these diagnoses months or years earlier than current practice. The compound interest on early diagnosis is enormous: a patient diagnosed at 25 rather than 35 gains a decade of informed medical care, appropriate surveillance, reproductive counselling, avoidance of mitochondrial toxins, and access to emerging therapies that may arrest or modify their disease course.

In an era where rare disease genomics is expanding at extraordinary pace, the rarest and most valuable resource remains the clinician who, at the bedside, recognises what to test for and when. This review is written in service of that clinician.

 

 

  KEY CLINICAL POINTS FOR PRACTICE

•        Mitochondrial dynamics disorders (OPA1, MFN2, DRP1/DNM1L, GDAP1, MFF) are underdiagnosed causes of complex neurological syndromes with a diagnostic delay of 5-20 years; early pattern recognition compresses this dramatically.

•        The fusion-fission heuristic guides initial gene selection: optic atrophy + neuropathy = test fusion genes (OPA1, MFN2); neonatal encephalopathy + lactic acidosis = test fission genes (DRP1/DNM1L, MFF).

•        Secondary mtDNA multiple deletions in OPA1 disease mimic primary mtDNA disease on muscle biopsy—always complete nuclear gene sequencing before concluding a diagnosis of primary mtDNA disease.

•        Valproate, metformin, and statins are relatively contraindicated in confirmed or suspected mitochondrial dynamic disorders; document these as formal alerts in the medical record.

•        MFN2 disease can produce MRI white matter lesions mimicking multiple sclerosis; the differentiator is the pure axonal neuropathy pattern on nerve conduction studies.

•        GDAP1 disease should be suspected when peripheral neuropathy is accompanied by hoarseness—bilateral vocal cord paresis in a hereditary neuropathy is GDAP1 until proven otherwise.

•        Moderate aerobic exercise is beneficial in mitochondrial disease and should be formally prescribed; deconditioning from inactivity worsens the phenotype.

•        Every reproductive-age patient with an autosomal dominant mitochondrial dynamic disorder requires structured preconception counselling as a duty of care.

 

 

References

1. Delettre C, Lenaers G, Griffoin JM, et al. Nuclear gene OPA1, encoding a mitochondrial dynamin-related protein, is mutated in dominant optic atrophy. Nat Genet. 2000;26(2):207-210.

2. Alexander C, Votruba M, Pesch UE, et al. OPA1, encoding a dynamin-related GTPase, is mutated in autosomal dominant optic atrophy linked to chromosome 3q28. Nat Genet. 2000;26(2):211-215.

3. Yu-Wai-Man P, Griffiths PG, Gorman GS, et al. Multi-system neurological disease is common in patients with OPA1 mutations. Brain. 2010;133(Pt 3):771-786.

4. Zuchner S, Mersiyanova IV, Muglia M, et al. Mutations in the mitochondrial GTPase mitofusin 2 cause Charcot-Marie-Tooth neuropathy type 2A. Nat Genet. 2004;36(5):449-451.

5. Stuppia G, Rizzo F, Riboldi G, et al. MFN2-related neuropathies: clinical features, molecular pathogenesis and therapeutic perspectives. J Neurol Sci. 2015;356(1-2):7-18.

6. Wakabayashi J, Zhang Z, Wakabayashi N, et al. The dynamin-related GTPase Drp1 is required for embryonic and brain development in mice. J Cell Biol. 2009;186(6):805-816.

7. Waterham HR, Koster J, van Roermund CW, et al. A lethal defect of mitochondrial and peroxisomal fission. N Engl J Med. 2007;356(17):1736-1741.

8. Baxter RV, Ben Othmane K, Rochelle JM, et al. Ganglioside-induced differentiation-associated protein-1 is mutant in Charcot-Marie-Tooth disease type 4A/8q21. Nat Genet. 2002;30(1):21-22.

9. Cassereau J, Chevrollier A, Codron P, et al. Simultaneous MFN2 and GDAP1 mutations cause severe Charcot-Marie-Tooth disease. Neurology. 2019;92(4):e382-e388.

10. Koch J, Feichtinger RG, Freisinger P, et al. Disturbed mitochondrial and peroxisomal dynamics due to loss of MFF causes Leigh-like encephalopathy, optic atrophy and peripheral neuropathy. J Med Genet. 2016;53(4):270-278.

11. Cipolat S, Martins de Brito O, Dal Zilio B, Scorrano L. OPA1 requires mitofusin 1 to promote mitochondrial fusion. Proc Natl Acad Sci USA. 2004;101(45):15927-15932.

12. Chen H, Detmer SA, Ewald AJ, et al. Mitofusins Mfn1 and Mfn2 coordinately regulate mitochondrial fusion and are essential for embryonic development. J Cell Biol. 2003;160(2):189-200.

13. Chan DC. Mitochondrial fusion and fission in mammals. Annu Rev Cell Dev Biol. 2006;22:79-99.

14. Loiseau D, Chevrollier A, Verny C, et al. Mitochondrial coupling defect in Charcot-Marie-Tooth type 2A disease. Ann Neurol. 2007;61(4):315-323.

15. Murphy E, Ardehali H, Balaban RS, et al.; American Heart Association Council on Basic Cardiovascular Sciences. Mitochondrial Function, Biology, and Role in Disease: A Scientific Statement From the American Heart Association. Circ Res. 2016;118(12):1960-1991.

16. Guillery O, Malka F, Landes T, et al. Metalloprotease-mediated OPA1 processing is modulated by the mitochondrial membrane potential. Biol Cell. 2008;100(5):315-325.

17. Rouzier C, Bannwarth S, Chaussenot A, et al. The MFN2 gene is responsible for mitochondrial DNA instability and optic atrophy 'plus' phenotype. Brain. 2012;135(Pt 1):23-34.

18. Bouhy D, Juneja M, Katona I, et al. A knock-in/knock-out mouse model of HMSN2C with TRPV4 Arg269Cys mutation as a tool to study the human neuropathological correlates. Brain. 2013;136(Pt 5):1571-1587.

19. Delettre-Cribaillet C, Hamel CP, Lenaers G. OPA1 (Kjer disease) and OPA3: two mitochondrial dynamin-related GTPases associated with autosomal dominant optic neuropathies. In: Bhatt DL, ed. Cardiology Grand Rounds. Springer; 2011.

20. Marber M, Bhatt DL. Mitochondrial medicine and emerging therapies. Eur Heart J. 2022;43(3):173-185.

21. Hargreaves IP, Heaton RA, Mantle D. Disorders of Human Coenzyme Q10 Metabolism: An Overview. Int J Mol Sci. 2020;21(18):6695.

22. Barboni P, Savini G, Valentino ML, et al. Leber's hereditary optic neuropathy with childhood onset. Invest Ophthalmol Vis Sci. 2006;47(12):5303-5309.

23. Wallace DC. A mitochondrial paradigm of metabolic and degenerative diseases, aging, and cancer: a dawn for evolutionary medicine. Annu Rev Genet. 2005;39:359-407.

24. Brichet M, Moulin M, Brikci-Nigassa A, et al. Leflunomide rescues the mutant MFN2 phenotype in mice and patient cells. Brain. 2023;146(6):2493-2506.

25. Yu-Wai-Man P, Newman NJ, Carelli V, et al. Bilateral visual improvement with unilateral gene therapy injection in Leber hereditary optic neuropathy. Sci Transl Med. 2020;12(573):eaaz7423.

 

 

DECLARATIONS

Funding: No external funding was received for this review article.

Conflicts of Interest: The authors declare no conflicts of interest relevant to the content of this review.

Ethics: No ethical approval was required for this review article.

Acknowledgements: The authors acknowledge the patients, families, and clinician-researchers worldwide who have contributed to the understanding of mitochondrial dynamic disorders.