Consensus Guidelines for Diagnosis and Management of Pyruvate Dehydrogenase Complex Deficiency Nandaki Keshavan, Julia Neugebauer, Marcello Bellusci, Enrico Bertini, Garry Brown, Niklas Darin, Suzanne DeBrosse, Lucy Drexler, Stefan Drexler, Gregory M. Enns, Rebecca D. Ganetzky, Kelly Gilbert, Austin Larson, Rebecca Legi, April N. Lehman, Camile Newby, Manuel Schiff, Rachel Skeath, Peter W. Stacpoole, Sarah Thompson, Emma Watt, Saskia B. Wortmann, Jirair K. Bedoyan, Shamima Rahman Abstract Primary pyruvate dehydrogenase complex deficiency (PDCD) comprises a group of monogenic disorders caused by pathogenic variants in genes encoding subunits of, or regulatory components affecting, the pyruvate dehydrogenase complex. The clinical phenotype spans a broad continuum, ranging from early onset congenital lactic acidosis to infantile or childhood onset global developmental delay with epilepsy, through to more attenuated adult‐onset neurological presentations. Reports from patients and advocacy groups indicate substantial variability in clinical management across both emergency and outpatient settings and between centres internationally. This variability underscores the need for systematic evaluation of the evidence base and the development of harmonised, consensus‐driven clinical guidelines to standardise care and improve outcomes. An international consortium of experts from Europe and North America, including metabolic physicians, neurologists, dietitians, geneticists and patient representatives, was convened. The group undertook a structured review of the literature and developed guideline statements addressing disease classification, recognition, diagnostic evaluation, dietary and non‐dietary management, surveillance for complications, genetic counselling and transition to adult services. Each recommendation was assigned a GRADE rating reflecting strength and quality of evidence. Consensus was achieved using a Delphi methodology. In total, 199 recommendations reached consensus and constitute the core of these guidelines. These recommendations provide a framework for consistent, high‐quality, multidisciplinary care. The process also identified key evidence gaps, highlighting priorities for future research and the ongoing need to develop effective disease‐modifying therapies. [JIMD-49-0-g003.webp] Introduction The pyruvate dehydrogenase complex (PDC) catalyses the conversion of pyruvate to acetyl‐CoA in mitochondria. This key reaction links glycolysis with the tricarboxylic acid (TCA) cycle and electron transport chain. PDC is fundamentally important in the brain where most energy generated is derived from glucose. Elsewhere, PDC plays a central role in regulating the balance of substrate utilisation and in the nucleus provides acetyl‐CoA for acetylation reactions. PDC is composed of multiple copies of three catalytic components, pyruvate dehydrogenase (E1), dihydrolipoamide acetyltransferase (E2), dihydrolipoamide dehydrogenase (E3) and one structural component, E3 binding protein (E3BP). Each catalytic component has a cofactor: thiamine diphosphate (ThDP) for E1, lipoic acid for E2 and flavin adenine dinucleotide (FAD) for E3 [, , ]. The first step of the reaction cycle is the oxidative decarboxylation of pyruvate by E1 with ThDP. CO2 is released and the remaining 2‐carbon unit is transferred via an ethylidene‐ThDP intermediate to the E2 enzyme where it reductively acetylates a lipoamide cofactor. E2 then catalyses the transfer of the acetyl group to CoA, leaving the thiols of the lipoamide cofactor in their reduced state. These are subsequently re‐oxidised by E3 with the reduction of its FAD cofactor. The cycle is completed by the re‐oxidation of the FAD with NAD+, yielding acetyl‐CoA and NADH as the final products (Figure ). E3 is not unique to PDC but is shared with other α‐ketoacid dehydrogenase complexes, α‐ketoglutarate dehydrogenase complex and branched chain α‐ketoacid dehydrogenase complex, which also share their overall molecular architecture and reaction mechanism. PDC activity is subject to close control in different tissues under different nutritional conditions and disease states [, , ]. Its activity is regulated by specific kinases (PDKs) and phosphatases (PDPs) which phosphorylate and dephosphorylate the E1α subunit resulting in inactivation and reactivation of the complex, respectively. While pyruvate inhibits PDK thereby activating PDC, increased NADH/NAD+ and acetyl‐CoA/CoA ratios activate PDK thereby acting as negative regulators. As a result of these mechanisms, the overall activity of the complex is responsive to dynamic alterations in the relative concentrations of key intermediates and metabolic circumstances [, , ]. Primary PDC deficiency (PDCD) can result from pathogenic variants in any one of the genes encoding subunits of the complex; however, most cases are due to defects in PDHA1, the X‐linked gene encoding the E1α subunit. There are several reports of pathogenic variants in PDHX, which encodes E3BP [, , , , ], but only a few cases of defects in PDHB [, , , ] and DLAT [, ], the genes encoding E1β and E2, respectively. Pathogenic variants in DLD, the gene encoding E3, are found relatively frequently in some ethnic groups [, ] and result in a combined deficiency of all α‐ketoacid dehydrogenase complexes containing this subunit, frequently presenting with clinical features that differ from those seen in other forms of PDCD. Pathogenic variants in PDP1 are also reported to cause PDCD [, , ]. Currently there are no international guidelines for diagnosing and managing patients with PDCD. Feedback from patient advocacy groups has highlighted that this has resulted in inequitable care, confusion among non‐experts and risk of clinical error surrounding specific aspects of care, especially emergency and perioperative management and use of ketogenic diet. This initiative aims to close this gap by formulating a comprehensive guideline which will delineate best practice, minimise potential ambiguities and standardise clinical care. Methods Scope of Guideline This guideline pertains to primary PDCD caused by pathogenic variants in the following genes: PDHA1, PDHB, DLAT, PDHX and PDP1. DLD deficiency will not be considered in the current guidelines as it differs substantially from the other primary deficiencies both in terms of its clinical and biochemical presentation and its management. Secondary deficiencies of PDCD, for example, those related to biosynthesis of lipoic acid and ThDP, will also not be considered. Guideline Development This guideline has been developed by combining consensus opinion of experts in the field and evidence obtained through appraisal of the published literature. The Delphi method was used to develop consensus guidelines for recognition, diagnosis and management of primary PDCD, together with Grading of Recommendations, Assessment, Development and Evaluation (GRADE) methodology to evaluate the strength of recommendations as detailed by the International Guideline Harmonization group (https://www.ighg.org/international‐guideline‐harmonization‐group/methods/grading‐system/). Before and during meetings, guideline group members received training in standardised literature evaluation according to GRADE methodology to assess the quality of evidence and determine the strength of recommendation for each statement. The guideline working group consisted of 19 physicians (metabolic specialists, biochemical geneticists, neurologists, biochemists), three metabolic dietitians and five patient advocacy group representatives. At the outset, the patient groups were asked to formulate questions and queries that they would like the guideline to address. They were also involved throughout the guideline’s development in a consultative capacity, including input into conceptualisation and wording of guideline statements, but did not participate in voting during the Delphi process. Among the 21 voting members, the gender distribution was 10 females and 11 males. Geographic representation was equally balanced between the United States and Europe. Eight working groups were formed, each tasked with defining key questions in one of the following subjects: (1) biochemical and biophysical characteristics of PDC; (2) clinical features of PDCD; (3) diagnostic approaches; (4) management approaches; (5) dietary management; (6) surveillance for disease complications; (7) genetic counselling; and (8) transition to adult services. Monthly virtual meetings and additional interim subgroup meetings occurred over a 12 month period from 14 October 2024. Literature Review A literature review on PDCD was performed in January 2025 in the following databases: PubMed, Google Scholar, Cochrane Library and ClinicalTrials.gov. Publications were excluded if they were not relevant to the predefined key questions or were not written in English. A total of 377 peer reviewed publications were identified as relevant to the topic. Details of search terms and the PRISMA flow chart are provided in the . Consensus Procedure The initial voting process was carried out remotely between September and November 2025 online using a survey‐based platform (Qualtrics, https://login.qualtrics.com) set up at UCL, UK. Voting was conducted using a structured Delphi methodology. Each guideline statement and its corresponding GRADE score was voted upon using a 5‐point Likert scale (1: absolutely disagree; 2: disagree; 3: neutral; 4: agree; 5: absolutely agree). ‘Strong consensus’ was pre‐defined as > 70% of scores being ≥ 4 or ≤ 2 with a mean score ≥ 4 or ≤ 2. If either condition was met, then the statement was deemed to have reached ‘consensus’. If neither parameter was met, the statement was considered to lack consensus. Survey data were compiled and analysed to determine the level of consensus for each statement. A face‐to‐face meeting of the whole guideline group took place in Cambridge, UK on 12–14 November 2025. All guideline statements not reaching consensus, and those with suggested amendments or having at least one member indicate absolute disagreement, were discussed, including level of evidence and optimal wording of the statements. Changes to the statements were made and voting repeated using Mentimeter (https://menti.com) during the face‐to‐face guideline group meeting in November 2025 and online using Qualtrics in January–February 2026. Recommendations Raw data from voting within the Delphi process and the degree of consensus for each statement are included in the . Clinical Features of PDCD To aid clinical recognition of PDCD, a literature search aimed to rigorously evidence the clinical spectrum of disease, categorised by age of onset (prenatal, neonatal, infantile/early childhood, older children, adolescents and adults). These features are summarised in guideline statements C1–C28 in Table . Prenatal disease onset occurs frequently in PDCD, affecting approximately 50% of individuals [, ], identified on routine fetal ultrasound. When PDCD manifests before 26 weeks of gestation, essential fetal neurodevelopmental processes are disrupted, including neuronal proliferation, migration and organisation. In contrast, energy failure occurring later in gestation tends to result in discrete brain lesions. As a result, most individuals with prenatal‐onset PDCD present with a combination of developmental abnormalities and discrete brain lesions [, , , , , , ]. These include microcephaly, neuronal migration abnormalities such as pachygyria, polymicrogyria, periventricular nodular heterotopias, corpus callosum hypoplasia or agenesis, cerebellar and brainstem hypoplasia [, , , ]. Neuroimaging may demonstrate decreased brain volume, ventriculomegaly or periventricular cysts [, , , ]. Craniofacial dysmorphism is a frequent feature in PDCD [, , , , , , ]. The features often resemble those seen in fetal alcohol syndrome, a similarity that may be attributed to the effects of acetaldehyde, a metabolite of alcohol which is elevated in cases of alcohol exposure and is known to be a potent inhibitor of the pyruvate dehydrogenase complex (PDC) [, ]. These craniofacial features include microcephaly, a narrow head with frontal bossing, a broad nasal bridge, an upturned nose with flared nostrils, a long and smooth philtrum, thin lips, low‐set ears, and retrognathism [, , , , , , ]. Intrauterine growth restriction and low birth weight are frequently observed in PDCD [, , , ]. Neonatal clinical presentation is characterised by hypotonia, lactic acidosis, respiratory distress, encephalopathy, lethargy, seizures, feeding difficulty and failure to thrive [, , , , , , , , , , , , , , , , , , , , , ]. Many neonates demonstrate evidence of prenatal disease onset. Neonatal presentation is described in PDHA1, PDHB, PDHX and PDP1 deficiencies but rarely in DLAT deficiency. Infantile/childhood disease presentation is characterised by significant variability in symptoms and disease severity. Clinical features include hypotonia, developmental delay, feeding difficulties, microcephaly, spasticity and optic atrophy [, , , ]. Patients may experience paroxysmal symptoms including breakthrough seizures, ataxia, stroke‐like episodes, somnolence, intermittent weakness and dystonia. Rarely, individuals present later in childhood with intermittent ataxia, paroxysmal dystonia or dyskinesia or other atypical clinical findings such as alternating hemiplegia or episodic limb paralysis. In older children and adults, PDCD may manifest for the first time with ataxia and/or peripheral neuropathy [, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ]. Peripheral neuropathy takes various forms in PDCD, most often axonal, but may include recurrent demyelination or Guillain‐Barre‐like syndrome resembling chronic inflammatory demyelinating polyneuropathy [, , , , , , , , , , , , , , , ]. Other late‐onset features include dystonia, typically episodic [, , , , , , , , , , , , , , , , , , ], movement disorders such as chorea, parkinsonism and tremor [, , , , , , , , , ], episodic weakness (often precipitated by carbohydrates or intercurrent illness), exercise intolerance, ptosis, stroke‐like episodes, Leigh syndrome spectrum neuroimaging findings, dysarthria, focal seizures, migraines, dementia, psychiatric symptoms, ophthalmoplegia, diplopia, loss of vision and optic neuropathy [, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ]. Diagnostic Testing Guideline statements D1–D11 refer to diagnostic testing (Table ). Biochemical Testing We recommend measuring lactate and pyruvate and calculating the lactate/pyruvate ratio which is typically low to normal (10–20), but can be > 20 in patients who are acutely ill, or be spuriously elevated in blood following difficult venepuncture or because of pyruvate degradation during suboptimal sample handling [, ]. Lactate and pyruvate elevation is more frequent in CSF than in blood [, ]. Some patients with PDCD may have normal blood and CSF lactate and pyruvate. We recommend that plasma amino acids and urine organic acids be measured; these tests are more sensitive post‐prandially after a carbohydrate‐containing meal and less sensitive after fasting. Alanine may be elevated (reflecting transamination of elevated pyruvate) but ratios of alanine to other amino acids, such as alanine:leucine (≥ 4.0), proline:leucine (≥ 3.0), (alanine+proline):leucine (≥ 6.5), (alanine+proline):(leucine+lysine) (≥ 2.5) and alanine:lysine (≥ 3.0) may be elevated even when alanine is normal [, , ]. Specific amino acid ratios are highly sensitive but not specific for PDCD. Urine organic acids in PDCD typically show elevations of lactate and pyruvate, but non‐specific elevations in some organic acids and tricarboxylic acid intermediates may also be seen, depending on the genetic aetiology of PDCD [, , ]. Carbohydrate loading tests are not recommended. However, in healthcare contexts where genetic testing is either too costly or unavailable, a carbohydrate loading test could be considered, although the rigour needed to obtain and interpret a meaningful result is challenging. Enzymatic Testing PDC enzyme activity is typically measured by quantifying 14CO2 release from [1‐14C]‐pyruvate in cell or tissue homogenates [, ]. Cultured skin fibroblasts are the patient cells of choice for assessing PDC enzymatic activity, but lymphocytes and skeletal muscle may be used. PDC activity assays in cultured skin fibroblasts are highly sensitive for detecting PDHA1‐related disease, with sensitivities of 97% in females and 91% in males, outperforming activity assays in lymphocytes and muscle []. However, normal enzyme activity can be seen in female patients, related to skewed X‐inactivation. Lymphocyte‐ and muscle‐based testing lack sensitivity: 36% and 58%, respectively, for identifying known PDC‐deficient females with pathogenic PDHA1 mutations []. Protein studies, such as immunoblotting or proteomics, may identify missing components, but often this is only available on a research basis []. The working group carefully considered the role of measuring PDC activity in diagnostic testing and concluded after much discussion that it should not be routinely recommended. The working group acknowledged that most centres in the United States and Europe do not offer PDC enzymology presently and historically very few laboratories in the world are able to carry out the test reliably to the required accredited standard. The assay is also no longer available in several countries owing to limited availability of the labelled substrate. Furthermore, the test is not 100% specific since low PDC activity may be seen in secondary causes of PDCD, which are outside the scope of this guideline. For the diagnosis of PDP1 deficiency, PDP1 enzymology is challenging to perform, requires specific assay conditions and is complex to interpret owing to the presence of isoenzymes. Molecular testing is therefore the test of choice to confirm primary PDCD. Enzymatic testing is an adjuvant methodology that can be used if genetic testing is inconclusive. Assessment of Thiamine Responsiveness in PDHA1 Disease In the past, PDC enzyme activity was used to assess thiamine responsiveness in vitro by identifying a dose‐dependent response of PDC activity to ThDP at both low and high ThDP concentrations. The working group discussed that this testing is no longer available to an accredited standard in most centres. Furthermore, it was noted that in vitro evidence of thiamine responsiveness frequently does not correlate with clinical thiamine responsiveness. Clinicians in the working group agreed they would not withhold thiamine from patients while they are awaiting results of in vitro responsiveness testing or indeed in patients for which in vitro testing suggests non‐responsiveness to thiamine. Molecular genetic testing can identify variants known to be thiamine‐responsive including those that affect residues needed for thiamine binding in E1α (Figure ), but other variants may also show thiamine responsiveness. PDC enzyme activity may be less sensitive in patients with thiamine‐responsive variants if performed in a thiamine‐rich medium, a limitation that should be considered when evaluating variants of uncertain significance. Genetic Testing Molecular genetic diagnostic testing is the test of choice regardless of availability or results of biochemical investigations. However, genetic and biochemical testing are often undertaken in parallel, and we support either a genetics‐in‐parallel approach or a genetics‐first approach. The choice of molecular testing—targeted genetic panel or broader genomic approach—depends on clinical suspicion, financial availability and local turnaround times. Variants in PDHA1 have variable expressivity, creating challenges in predicting phenotype from genotype. Prognostication should not be based on genotype alone []. X Inactivation and Somatic Mosaicism Studies X inactivation studies are not usually needed in diagnostic testing but may be useful in interpreting sequencing data that indicates mosaicism between wild‐type and mutated alleles. X inactivation studies usually use peripheral tissue (fibroblasts or lymphocytes) and may not correlate with other tissues such as the brain, the predominantly affected organ in PDCD. Cultured cells from females with PDHA1‐related PDCD past passage p10 may develop selection bias towards cells that express only the wild‐type allele. X inactivation patterns ≥ 70:30 bias in favour of the wild‐type allele are associated with normal enzyme activity [, , ]. Conventional X inactivation assays can detect skewing but typically cannot determine which X chromosome (mutant or wild‐type) is preferentially active without further molecular data [, , ]. Genetic Counselling Guideline statements G1–G4 refer to genetic counselling (Table ). PDHA1‐related PDCD is inherited in an X‐linked manner whereas pathogenic variants in PDHB, DLAT, PDP1 and PDHX exhibit autosomal recessive inheritance. Molecular testing is important to determine the specific subtype of PDCD and for genetic counselling regarding mode of inheritance and recurrence risk in subsequent pregnancies. Mothers of children with PDHA1 variants, the most common genetic aetiology of PDCD, should be offered genetic testing to determine whether the variant is maternally inherited or de novo. Most affected males (60%–63%) have de novo variants [, ] but a significant minority are maternally inherited. Most affected females (85%–95%) have de novo variants [, ] and a smaller minority of pathogenic variants are maternally inherited. If the proband's variant is not detected in the mother, recurrence risk is low but above the general population risk because of possible germline mosaicism. Management Guideline statements about management of PDCD are subdivided into general (M1–M19, Table ), metabolic drug (M20–M29, Table ), emergency (M30–M46, Table ), perioperative (M47–M52, Table ) and dietetic management (M53–M66, Table ). General Management See Table for general management recommendations (M1–M19). Patients should be managed within the context of a multidisciplinary team which combines the expertise of a metabolic disease expert or neurologist, dietitian for patients on a ketogenic diet and other allied healthcare professionals and community services depending on individual needs. Depending on their individual problems, patients with PDCD will need multiple therapies to manage symptoms, but care should be taken when commencing new medications to ensure that they do not contain carbohydrates in patients on a ketogenic diet to prevent disruption of ketosis [, , ]. We recommend that existing published guidelines [] are followed for management of seizures in patients with PDCD. However, topiramate should be avoided as first‐line treatment since it may cause acidosis, and propofol should be used cautiously in the management of status epilepticus owing to risk of propofol infusion syndrome due to inhibition of fatty acid oxidation []. Patients and families should be provided access to palliative care services for advice on managing symptoms, and for family support, especially at advanced disease stages. Metabolic Drug Management Guideline statements regarding metabolic drug management are listed in Table . We recommend thiamine supplementation (150–1000 mg/day in three divided doses) in patients with PDCD. Thiamine is reported to improve lactic acidosis, muscle weakness, paroxysmal episodes, movement disorders and may help stabilise the disease, regardless of whether there is a thiamine‐responsive genetic variant [, , , , , , , , , , , , , , , , , , ]. Patients with thiamine‐responsive PDCD variants (Figure ) often show paroxysmal symptoms and a marked clinical improvement after thiamine treatment, although the supporting evidence remains limited [, , , , , , , , , , , , , , , , , , ]. In other patients, thiamine may still be beneficial, considering its potential effect in stabilising the PDC enzyme. Patients with free carnitine deficiency secondary to ketogenic diet therapy should receive supplementation [, , , , ]. Bicarbonate or citrate may be used in management of metabolic acidosis [, , , , ]. Dicholoroacetate is an investigational drug with known safety and tolerability which activates PDC by inhibiting PDK, thereby lowering lactate levels [, , , , ]. A recently completed clinical trial aimed to assess efficacy of DCA in management of PDCD (NCT02616484), with results yet to be reported. Emergency Management Patients with PDCD may experience acute metabolic decompensations during intercurrent illnesses or other metabolic stressors such as fasting, insufficient food intake, inappropriately high carbohydrate intake, anaesthesia and surgery, even after initiation of disease‐specific treatment [, , , , , ]. Patients should receive an emergency management letter to be taken to the emergency department during acute illnesses. Key biochemical abnormalities include metabolic acidosis, elevated lactate and ketones above the patient's usual baseline. Figure summarises the acute management of patients with PDCD during metabolic decompensation, including necessary blood investigations. Where IV hydration is needed, we recommend saline‐based fluids with no or low dextrose content since high glucose delivery will exacerbate metabolic acidosis in PDCD and negate therapeutic ketosis in patients on KD. Lactate‐containing fluids are contraindicated. Dextrose‐containing fluids may be used with caution in specific situations, including hypoglycaemia < 2.5 mmol (45 mg/dL) and in prolonged IV fluid use, with regular blood glucose and ketone monitoring. Patients with metabolic acidosis (pH < 7.20 and/or bicarbonate < 10 mmol/L and/or base deficit >10 mmol/L) who are acutely unwell should receive IV or enteral bicarbonate/citrate supplementation. Patients with hyperketosis > 5 mmol/L in the context of acute metabolic decompensation should be managed with IV hydration using normal saline or electrolyte‐balanced crystalloids that lack dextrose to decrease ketone levels. Once adequate hydration has been achieved, careful use of low dextrose infusion may be needed if hyperketosis persists. In all cases, the underlying medical cause for deterioration should be investigated and treated without delay. Patients presenting with severe acute decompensation and/or acute neurological manifestations, including but not limited to Leigh syndrome spectrum lesions, leukoencephalopathy, dystonia, ataxia and epilepsy, should be admitted to an intensive care unit. Perioperative Management Table lists guidance statements about perioperative management. Patients with PDCD are at increased risk of anaesthesia‐related complications. Anaesthesia can exacerbate respiratory symptoms and trigger respiratory failure, requiring careful monitoring throughout the procedure [, , , , , , ]. We recommend slow titration of all anaesthetics, avoidance of neuromuscular blockade in PDCD patients with muscle involvement and advise against prolonged use of propofol which may cause lactic acidosis and interfere with ketogenesis [, , , , , , ]. Duration of preoperative fasting should be minimised, and regular perioperative monitoring of acid–base balance, glucose, ketone bodies and lactate is essential. Perioperative IV fluid considerations should be as described above for emergency management. Patients not receiving enteral feeds for more than 24 h should be commenced on parenteral nutrition and be restarted on their usual enteral feed as soon as possible, especially if on a ketogenic diet. Dietetic Management Ketogenic diet (KD) was first reported to be beneficial in PDCD in 1976 []. KD shifts the body's primary energy source from carbohydrates to fats, promoting the production of ketone bodies, which can enter the TCA cycle and serve as an alternative fuel []. Ketone bodies can cross the blood–brain barrier, providing the brain with an alternative energy source to glucose. Currently, no formal guidelines exist for the use of KD in PDCD, and research is limited due to the rarity of the condition. The KD is characterised by a high fat content, controlled protein intake to support growth, and restricted carbohydrate consumption. The goal is to induce and maintain a state of ketosis, whereby ketone bodies become the predominant energy substrate []. In some of the literature, the term KD was used to describe diets that were low in carbohydrate, but with high protein content (up to 30% energy). These do not align with current definitions of KD. The high protein content means protein will be directed through gluconeogenesis and glycolysis to produce pyruvate and consequently lactate; these diets will not drive the production of ketones. There are four main types of KD, as follows. Classical Ketogenic Diet The classical KD is based on long‐chain triglycerides and is calculated using a ratio of grams of fat to grams of combined protein and carbohydrate []. Commonly used ratios include 4:1 and 3:1. Protein is provided at levels sufficient to support growth and development, while carbohydrate intake is strictly limited to promote ketosis []. Medium Chain Triglyceride Ketogenic Diet The medium chain triglyceride (MCT) KD modifies the classical approach by incorporating MCTs as the primary fat source []. MCTs are more efficient at generating ketone bodies compared to long‐chain triglycerides (LCTs), allowing for greater flexibility in protein and carbohydrate intake, which can improve the diet's palatability []. Originally formulated with 60% MCT, the diet has since been adjusted to 30% to reduce gastrointestinal side effects associated with high MCT intake []. The typical macronutrient distribution is: 30%–60% MCT, 15%–45% LCT, 10% protein and 15% carbohydrate []. Modified Ketogenic Diet Due to the restrictive nature of classical and MCT KDs, adherence can be challenging, particularly for older children. The Modified Ketogenic Diet (MKD) offers a more flexible alternative, with a fat to protein+carbohydrate ratio of approximately 1:1. It has demonstrated efficacy in managing epilepsy in children []. The MKD allows for increased protein and carbohydrate intake, making it more appealing and easier to implement in daily meal planning []. Low Glycaemic Index Diet The Low Glycaemic Index Diet (LGID) was developed following observations that children on the KD maintained stable blood glucose levels []. This stability was hypothesised to contribute to the therapeutic effects of the KD []. The LGID permits carbohydrate intake of 40–60 g per day, provided the carbohydrates have a glycaemic index below 50 []. Side Effects Side effects of KD include renal calculi, hypercholesterolaemia, gastrointestinal symptoms (e.g., vomiting, constipation and diarrhoea), and poor growth, particularly in children on more restrictive versions of the diet. These potential adverse effects highlight the importance of close clinical monitoring and multidisciplinary support when initiating and maintaining KD []. Outcomes The evidence base for outcomes of KD management of PDCD is limited to small, descriptive cohorts or single case reports and uncontrolled observational studies [, , , , , , , , , , , , , , , , ]. Assessments of outcomes in the literature were largely subjective rather than objectively measured, reflecting a low level of evidence and highlighting the need for more rigorous controlled studies. The lack of controlled studies makes it difficult to attribute outcomes to KD rather than the natural disease course within advances in general clinical management []. Outcomes reported in the literature include improved biochemistry (with mixed reports regarding lactate lowering effects), neurocognitive development including social interaction and communication, gross and fine motor development, sleep, movement disorder, ataxia, seizures and reversal of brain lesions detected on MRI. There is some evidence that KD may help in decreasing the severity and frequency of paroxysmal neurological episodes. Recommendations for use of KD in PDCD are summarised in Table . We recommend that KD should be used in the management of primary PDCD caused by PDHA1, PDHB, DLAT, PDHX and PDP1 variants, given the shared underlying pathophysiology. KD should be commenced as soon as possible after confirming a genetic diagnosis of PDCD and excluding disorders in which KD is contraindicated (e.g., fatty acid oxidation disorders, DLD and ECHS1 deficiency). The choice of KD, formulation and route of administration should be tailored to the individual's clinical condition, and their and the family's ability to adhere to the diet. Target ketone levels are individualised to the patient, based on symptom control. Disease Monitoring and Surveillance Guideline recommendations S1–S78 about disease monitoring in PDCD are listed in Table and summarised in Table . They are grouped into the following domains: multidisciplinary care, laboratory tests, gastroenterology, psychomotor delay/intellectual disability/cognitive impairment, neuropsychiatric and mental health, seizures, disorders of movement and tone (ataxia, dystonia, choreoathetosis, parkinsonism, spasticity, tremor), peripheral neuropathy, neuroimaging, cardiology, respiratory function, orthopaedics/bone health, kidney/renal function, dentistry/drooling, hearing, ophthalmology, sleep and pain. For many of these domains, the recommendations are not specific to PDCD, but feedback from patient advocates emphasised the need to provide guidance in these areas for PDCD. Transition to Adult Services Table lists recommendations concerning transition to adult services (T1–T12). Transition is not merely transfer of care from paediatric to adult services, but a process in which patients are empowered to make decisions for themselves wherever possible. There is considerable variability in cognitive function and psychosocial ability of patients with PDCD, and this should be considered at the outset of the transition process, with the aim of involving patients as much as possible in decisions, thereby promoting autonomy. The working group recognised that the transition period is associated with increased risk since care is transferred from a paediatric team that is highly knowledgeable about the patient's medical history and needs to an adult service that is comparatively unfamiliar with the individual. Planning transition should begin at least 2 years before the intended transfer of care to adult services. It is important that transition is undertaken specialist‐to‐specialist, ideally including a joint in‐person appointment that enables thorough handover of the patient's medical history, medications, diet plan and emergency plan to the adult team. Transition arrangements from paediatric to adult services are highly geographically and service dependent. The working group recognised that in many centres such as those in the USA where patients are managed by a specialist geneticist with expertise in metabolic disorders and a scope of practice that is not defined by age, patients with PDCD are taken care of primarily by a single specialist team throughout their lives and therefore a formalised transition process does not occur. However, it is still important to ensure that patients and families are supported in accessing adult services in the community and given advice on how to access acute medical services. Discussion For rare diseases, finding high‐quality evidence to support clinical decision‐making is often limited by low patient numbers, publication bias, few randomised controlled trials and insufficient detail in published literature regarding how patients have been managed. It is therefore important to draw upon the experience of clinical and scientific experts to assess the available evidence systematically. By establishing a globally recognised clinical guideline for PDCD, we intend to break down geographical boundaries, promote equitable use of healthcare resources and provide a baseline of clinical guidance that can be revised in future to keep pace with novel therapeutic discoveries and clinical trial advances. When developing this guideline, we ensured that we included the voice of patients from conceptualisation through to finalisation of guideline statements. During the process, especially at the face‐to‐face workshop, participants were reminded of the importance of ensuring that the guideline is written in a manner that is accessible to patients, families and to both expert and non‐expert clinicians. It is anticipated that during acute illnesses, patients and families may take this clinical guideline to the emergency department, where the medical team may be unfamiliar with the management of PDCD and may need to institute immediate emergency management before being able to consult with the patient's specialist team. The guideline is also designed to be a resource for non‐emergency management of patients reviewed in outpatient departments with a focus on holistic multidisciplinary care. Guideline statements have been written noting that they will be referred to by doctors, dietitians and other allied healthcare professionals from different paediatric and adult specialties, reflecting the multisystemic clinical presentation of PDCD. Conclusion A comprehensive first clinical guideline for PDCD is presented, based on appraisal of evidence and identification of international best practice. It is hoped that this guideline will facilitate disease recognition and standardise management with the aim of improving outcomes in patients with PDCD. In future, the working group will evaluate new evidence periodically and update the recommendations accordingly.