Clinical manifestations

Presentation and course

The age of pyruvate dehydrogenase complex deficiency presentation ranges from fetal, neonatal, or infantile to late-onset in children and adults. The clinical presentation of pyruvate dehydrogenase complex deficiency is highly variable and ranges from fatal congenital lactic acidosis and congenital brain abnormalities, including ventriculomegaly (35% to 85%), corpus callosum abnormalities (15% to 55%) and Leigh syndrome (12% to 25%), to global developmental delay and failure to thrive, to relatively mild or intermittent ataxia, exercise-induced episodic paroxysmal dystonia or ataxia, Guillan Barre-like syndrome, or peripheral neuropathy with normal cognitive function and long survival (64; 65; 59; 32; 17; 53; 15; 71; 67). Muscular hypotonia (46% to 89%), developmental delay (57% to 83%), epilepsy (16% to 57%), and microcephaly (49%) are other common findings in patients with pyruvate dehydrogenase complex deficiency (64; 65; 32; 17; 53). Visual and eye movement impairments, including cortical blindness at one extreme to intermittent exotropia or esotropia at the other, as well as hearing impairments, have been reported. Age at diagnosis appears to be the best predictor of survival and cognitive outcome in those affected with pyruvate dehydrogenase complex deficiency (17). About 4% of patients with pyruvate dehydrogenase complex deficiency have normal brain imaging (CT or MRI) findings (53). Several vignettes highlight the clinical overlap and unique features of the various subclasses of pyruvate dehydrogenase complex deficiency (Table 1).

Typical neonatal or infantile presentation of pyruvate dehydrogenase complex deficiency is characterized by hypothermia, respiratory distress or failure, vomiting, metabolic and lactic acidosis, delayed motor development, poor linear growth or weight gain, intellectual disability, speech delay, and neurologic findings (eg, lethargy, apathy, hypotonia, pyramidal and extrapyramidal signs, ataxia, or seizures). Brain anomalies can be noted. Blood or CSF lactate and pyruvate are elevated with normal lactate to pyruvate (L/P) ratio (reference range: 10 to 20), differentiating primary pyruvate dehydrogenase complex deficiency from primary mitochondrial disorders, such as those due to electron transport chain dysfunction where L/P ratios are generally elevated. Fasting, intercurrent illness, excessive exercise, or other metabolic stressors may result in metabolic decompensation with a variable degree of metabolic and lactic acidosis and movement anomalies requiring emergent attention and management. The specific management approach depends on the patient and their specific disease. In most cases, polytherapy with ketogenic diet or intralipid, thiamine supplementation, dichloroacetate, bicarbonate or buffering agents for metabolic acidosis, riboflavin supplementation, L-carnitine supplementation if on ketogenic diet, or antiepileptic medications for seizure control constitute the management approach for individuals with pyruvate dehydrogenase complex deficiency. Sugar-free versions of supplements and medications are necessary for individuals on ketogenic diet to maintain adequate ketosis. Molecular genetic confirmation of the pyruvate dehydrogenase complex deficiency diagnosis is necessary for best management choices and outcome. Suspected cases of pyruvate dehydrogenase complex deficiency with enzymatically confirmed low pyruvate dehydrogenase complex activity by cell- or tissue-based assays should also have confirmatory molecular genetic testing completed before the start of high-fat interventions, such as ketogenic diet or intralipid. Use of ketogenic diet or intralipid therapy in newborns and infants with pyruvate dehydrogenase complex deficiency subclass other than primary-specific pyruvate dehydrogenase complex deficiency may be ineffective or even fatal (see Table 1) (20; 05; 72). Management approaches and emergency protocol (for primary-specific pyruvate dehydrogenase complex deficiency) for families are detailed in the Management section of this article.

Table 1. Newborn and Infant Pyruvate Dehydrogenase Complex Deficiency Presentation Vignettes: Diagnostic Approaches and Intervention Impacts

Prognosis and complications

The age at diagnosis appears to be the best predictor of survival and cognitive outcome in those affected with pyruvate dehydrogenase complex deficiency (17); thus, early identification and diagnosis of pyruvate dehydrogenase complex deficiency are crucial (07; 87). Females have better survival, but surviving females are more severely affected than males (17; 53). Death among severe newborn and infantile patients with pyruvate dehydrogenase complex deficiency are generally about 40%, 60%, and 90% for those younger than 3 months, younger than 1 year, and younger than 4 years of age, respectively (17).

Neuroimaging findings

Notable brain anomalies in individuals with pyruvate dehydrogenase complex deficiency determined by CT or MRI include cerebral atrophy or hypoplasia (common); leukoencephalopathy, ventriculomegaly and ventricular asymmetry (common); periventricular cystic lesions or intraventricular septations (occasional); corpus callosum anomalies, including partial to complete agenesis (common); absent septum pellucidum, bilateral basal ganglia, and midbrain or brainstem T2-weighted hyperintense lesions or cystic cavitations (common); Leigh-like lesions with predominant involvement of globus pallidus and bilateral stroke-like lesions; mega cisterna magna; hypoplastic cerebellum or cerebellar vermian anomalies, lesions or absence; and optic nerve hypoplasia (03; 01; 17; 53; 66; 67).

1H MRS findings can include increased lactate peaks in sampled frontal white matter, basal ganglia, and midbrain structures.

Biological basis

Etiology and pathogenesis

Pyruvate formed from cytosolic glycolysis crosses the outer mitochondrial membrane, probably via a voltage-dependent anion channel, and then is actively transported across the inner mitochondrial membrane by the mitochondrial pyruvate carrier. Oxidative decarboxylation of cytosolic malate by malic enzyme also produces pyruvate. Once within the mitochondrial matrix, pyruvate can be oxidized into acetyl-CoA by pyruvate dehydrogenase complex or carboxylate to oxaloacetate by pyruvate carboxylase, the latter being an early step in gluconeogenesis. Functional defect in pyruvate dehydrogenase complex results in elevated pyruvate within the cytoplasm, where it can be converted to alanine via a transamination reaction by alanine transaminase or to lactate by lactate dehydrogenase. Elevated lactate inhibits the first step in proline metabolism by proline oxidase, resulting in proline elevation (38). Lactate production yields less than one tenth of the ATP derived from complete glucose oxidation via the tricarboxylic acid cycle. Disease-causing variants in the primary genes associated with low pyruvate dehydrogenase complex activity result in pyruvate dehydrogenase complex deficiency by interfering with pyruvate flux and energy production from carbohydrate oxidation with resultant lactic acidemia, hyperpyruvicemia, hyperalaninemia, and hyperprolinemia aggravated with consumption of carbohydrate. Unlike in primary mitochondrial disorders where the blood lactate to pyruvate (L/P) ratio is high (usually greater than 25), in pyruvate dehydrogenase complex deficiency, the L/P ratio both in blood and CSF is usually normal, ranging from 10 to 20. Defective pyruvate dehydrogenase complex deficiency with reduced acetyl-CoA production results in decreased production of reducing equivalents in the form of NADH and, thus, decreased ATP through the mitochondrial respiratory chain and, consequently, an energy deficit.

Focus on primary-specific pyruvate dehydrogenase complex genes. Loss of primary pyruvate dehydrogenase complex enzymatic function may occur from loss of mRNA expression, subunit instability, or loss or reduction of functional activity of various primary pyruvate dehydrogenase complex-associated gene products (Table 2), which could, in part, be due to partial or complete deletion of the specific gene. Most cases of pyruvate dehydrogenase complex deficiency (at least 75%) are due to pathogenic or likely pathogenic variants in PDHA1, with over 130 nonduplicate disease-causing variants reported (42; 63; 59; 32; 19). Males and females with somatic mosaicism in PDHA1 have been reported (18). E1α null variants are likely lethal as none have been identified in males, except in a mosaic state (18). A few large rearrangements have been identified, such as a large intragenic 4.2 kb deletion covering part of intron 5 to part of intron 9 (g.10,145_14,371 del 4,227) (11). Most cases of pyruvate dehydrogenase complex deficiency due to PDHA1 (about 75%) are due to de novo disease-causing variants, whereas about 25% are due to inheritance from a carrier “asymptomatic” mother. About half of all nonduplicate E1α disease-causing missense variants (DMVs) are found within exons 5 to 7 (19). About 85% of E1α DMVs are solvent-inaccessible (“buried”), and about 30% of buried E1α DMVs are involved in subunit-subunit interface contact (19). Arginine (R) replacements constitute about 40% of all the E1α DMVs (72; 19). Recurrent replacements of mature proteins R234, R273, and R349 (aka, p.R263, p.R302, and p.R378) together predominate among all arginine replacements (53; 19). Certain E1α DMVs show sex predominance. For example, mature E1αR273 replacements to histidine (p.R302H) or cysteine (p.R302C) show female predominance (95%) (72; 19). About 10% of E1α DMVs are due to recurrent mature R349 replacements (mature R349H or R349C; p.R378H or p.R378C, respectively), exhibiting an average 20% to 25% of control mean in vitro fibroblast-based pyruvate dehydrogenase complex activity (72; 19). Mature E1α R349 replacements disrupt proper subunit-subunit interface contact and cooperative networks with other subunit residues and form a pocket in 3D E1 structure (37; 19; 24). Molecular dynamic simulations show that E1α DMVs influence the E1 active sites allosterically by altering the information flow and cooperative networks (24).

Disease-causing variants in PDHB are a rare cause of pyruvate dehydrogenase complex deficiency (52; 76; 19). An individual has been reported with dysregulation of the ubiquitin-proteasome system resulting in E1β instability and, thus, pyruvate dehydrogenase complex deficiency but without a molecular defect in PDHB (26).

Disease-causing variants in other primary-specific genes, such as DLAT (28; 47), PDP1 (43; 14; 06), and PDHX (12; 46; 33) are reported. A founder PDHX p.R446* variant is known in the European Roma population (in Bulgaria, Romania, Slovakia, and Hungary) (33).

Activating mutations in the pyruvate dehydrogenase kinase isoenzyme 3 (PDK3) gene, another primary-specific gene, causes X-linked dominant Charcot-Marie Tooth disease (CMTX6). The underlying pathogenic cause of peripheral neuropathy in these individuals is reduced pyruvate flux due to hyperphosphorylation of pyruvate dehydrogenase complex (36; 56).

Primary-specific pyruvate dehydrogenase complex gene variants constitute 80% to 90% of all cases of pyruvate dehydrogenase complex deficiency (76; 72).

Epidemiology

Two studies found the incidence of pyruvate dehydrogenase complex deficiency to be about 1 in 40,000 live births annually in the United States (07; 87). The study from Ohio was a prospective study completed in a 1-year period, with newborn screening dried blood spot data of alanine and proline included (07); the second study from Pennsylvania was a retrospective study that reviewed 11 years of medical records at a major medical center (87). Certain amino acid ratios, such as Ala/Leu, Ala/Lys, Pro/Leu, (Ala+Pro)/Leu, or (Ala+Pro)/(Leu+Lys) or their combinations, are highly sensitive (90% or greater) but not specific (75% to 85%) for identifying patients with pyruvate dehydrogenase complex deficiency (07; 87). Second-tier molecular testing of dried blood spots for primary-specific pyruvate dehydrogenase complex deficiency genes (or at least PDHA1) would increase newborn screening specificity for this disorder and exclude other disorders, such as other primary mitochondrial disorders not on newborn screening. The estimated prevalence of pyruvate dehydrogenase complex deficiency in the United States is about 2,000 cases, but only about 50% of those are confirmed diagnostically (SAOL Therapeutics data, December 2023).

Pyruvate dehydrogenase complex deficiency due to chromosome X-linked PDHA1 constitutes at least 75% of affected cases, with about 60% of those due to missense substitutions (17; 53; 76; 05; 07; 19). Pyruvate dehydrogenase complex deficiency due to primary-specific genes (PDHA1, PDHB, DLAT, PDHX, PDP1, and PDK3) constitute 80% to 90% of all cases with pyruvate dehydrogenase complex deficiency (76; 72).

Prevention

Pyruvate dehydrogenase complex deficiency is not preventable. However, early diagnosis and early interventions with known or novel therapies are likely to lead to improved developmental and cognitive outcomes and quality of life with fewer seizures, less hospitalization, and more prolonged survival. Many of the known interventions still leave patients with significant systemic disease and developmental disability.

Differential diagnosis

Associated or underlying disorders

Patients with pyruvate dehydrogenase complex deficiency are subclassified into at least three different groups (primary-specific, primary-generalized, and secondary pyruvate dehydrogenase complex deficiency) with important clinical consequences (05; 72). Although pathogenic or likely pathogenic variants in more than 38 genes cause functional (enzymatic) pyruvate dehydrogenase complex deficiency, the overwhelming majority (80% to 90%) of affected individuals are subclassified into primary-specific pyruvate dehydrogenase complex deficiency (due to variants in either PDHA1, PDHB, DLAT, PDHX, PDP1, or PDK3), with at least 75% due to PDHA1 pathogenic or likely pathogenic variants. This fact has important and practical diagnostic and therapeutic implications and is crucial for the prospect of future implementation of pyruvate dehydrogenase complex deficiency newborn screening. Impairment (or lack) of cofactors (eg, thiamine pyrophosphate), substrate (eg, pyruvate), complex component modifications (eg, by lipoylation or glutathionylation), and pyruvate dehydrogenase complex regulators (eg, pyruvate dehydrogenase phosphates and their regular and PD kinases) also result in pyruvate dehydrogenase complex deficiency. A list of known, possible, and potential molecular etiologies of impaired human pyruvate oxidation is shown in Table 2.

Table 2. Subclassification of Pyruvate Dehydrogenase Complex Deficiency and Other Pyruvate Metabolism-Related Disorders (currently known, possible, and potential molecular etiologies of impaired pyruvate oxidation)

Acquired disorders

Acquired disorders in the differential diagnosis of primary pyruvate dehydrogenase complex deficiency include:

Dietary thiamine deficiency. Dietary thiamine deficiency may develop in children who are critically ill for other reasons, resulting in an acquired decrease in pyruvate dehydrogenase complex activity and biochemical findings of elevated lactate and pyruvate (with normal ratio) similar to individuals with primary pyruvate dehydrogenase complex deficiency (48).

Fetal alcohol syndrome. Facial features characteristic of fetal alcohol syndrome include a short nose, long and smooth philtrum, and thin upper lip as well as primary microcephaly; such features are similar to those seen in pyruvate dehydrogenase complex deficiency. Furthermore, alcohol-induced CNS insult results in midline anomalies, such as agenesis of the corpus callosum, also commonly seen in pyruvate dehydrogenase complex deficiency. High acetaldehyde concentrations result from chronic alcoholism due to enhanced ethanol oxidation by induction of cytochrome P-450 2E1 (CYP 2E1) and concomitant reduction in aldehyde dehydrogenase activity. Hard and colleagues show that acetaldehyde impairs pyruvate dehydrogenase complex activity by a mixed inhibition-type mechanism, implicating acquired decreased pyruvate dehydrogenase complex activity in the developing fetus exposed to alcohol and, consequently, fetal alcohol syndrome (25).

Diagnostic workup

Key tests and procedures

Establishing the diagnosis. A diagnostic algorithm for pyruvate dehydrogenase complex deficiency has been recommended, and the reader is referred to Figure 2 in the 2017 article by Shin and colleagues (72).

Quantitative analysis of plasma amino acids, urine organic acids, and blood lactate and pyruvate are the most important laboratory tests for the initial recognition of pyruvate dehydrogenase complex deficiency in an individual. Metabolic and lactic acidosis, as well as keto acidosis, are initially observed. In contrast to pyruvate carboxylase deficiency, fasting hypoglycemia and hyperammonemia are not expected features of pyruvate dehydrogenase complex deficiency. Blood lactate, pyruvate, alanine, and proline can be intermittently normal when measured after an overnight fast but characteristically increased after a dietary intake of carbohydrates. Specific plasma amino acid ratios, such as Ala/Leu, Ala/Lys, Pro/Leu, (Ala+Pro)/Leu, and (Ala+Pro)/(Leu+Lys) and their combinations, are highly sensitive (90% or greater, depending the specific ratio or combination ratios used) but only 75% to 85% specific for identifying individuals with pyruvate dehydrogenase complex deficiency (07; 87). Often, plasma amino acids may be reported normal, but the specific amino acid ratios (noted above) could suggest pyruvate dehydrogenase complex deficiency.

Functional (enzymatic) diagnosis of pyruvate dehydrogenase complex deficiency is established in a proband with supportive metabolic analyte findings, such as elevated blood lactate and pyruvate with normal lactate to pyruvate (L/P) ratio, elevated blood alanine and often proline, Ala/Leu ≥4.0, Ala/Lys ≥3.0, Pro/Leu ≥3.0, (Ala+Pro)/Leu ≥6.5, and (Ala+Pro)/(Leu+Lys) ≥2.5, and typical brain anomalies, by measuring low pyruvate dehydrogenase complex enzyme activity in cultured fibroblast cells from skin biopsy, WBC lymphocytes, or skeletal muscle tissue (Table1) (72). Pyruvate dehydrogenase complex activity can be measured at low and high thiamine pyrophosphate concentrations to identify thiamine-responsive pyruvate dehydrogenase complex deficiency. It is important to know the limitations of clinical enzymatic testing (ie, its sensitivity and variability with patient sex, cell or tissue type used, and passage number used in assays) because this has important diagnostic and management implications (72).

A molecular genetic diagnosis of pyruvate dehydrogenase complex deficiency is established in a proband with supportive metabolic analyte findings and typical brain anomalies, by identifying a single pathogenic variant (or likely pathogenic variant) in an X-linked gene such as PDHA1 associated with pyruvate dehydrogenase complex deficiency (as heterozygous or hemizygous in a female or male, respectively) or biallelic pathogenic or likely pathogenic variants in other pyruvate dehydrogenase complex deficiency-associated genes by molecular genetic testing (Table 2).

Notably, the turnaround time in receiving test results varies among laboratories and should be considered when determining the order and type of testing. Frequently, both types of testing are ordered together or sequentially, especially if one of the test results is ambiguous or insufficient to establish pyruvate dehydrogenase complex deficiency diagnosis.

Biochemical diagnosis. Assaying pyruvate dehydrogenase complex in cultured fibroblasts in cases where the underlying genetic etiology is PDHA1 (most common) is highly sensitive irrespective of sex; 97% (95% confidence interval [CI]: 90%-100%) and 91% (95% CI: 82%-100%) in females and males, respectively (72). In contrast to fibroblast-based testing, lymphocyte- and muscle-based testing are not sensitive (36% [95% CI: 11%-61%, p=0.0003] and 58% [95% CI: 30%-86%, p=0.014], respectively) for identifying known pyruvate dehydrogenase complex-deficient females with pathogenic PDHA1 variants (72). The sensitivity and variability of pyruvate dehydrogenase complex activity testing using cultured fibroblasts, WBC lymphocytes, or skeletal muscle tissue for other pyruvate dehydrogenase complex deficiency-associated genes remain to be thoroughly evaluated.

Molecular diagnosis. Per ACMG/AMP variant interpretation guidelines, the terms “pathogenic variant” and “likely pathogenic variant” are synonymous in a clinical setting, meaning that both are considered diagnostic and can be used for clinical decision-making (62).

Molecular genetic testing approaches can include a combination of gene-targeted testing (single gene testing, multigene panel) and comprehensive genomic testing (exome sequencing, genome sequencing). Gene-targeted testing requires that the clinician determine which genes are likely involved, whereas comprehensive genomic testing does not.

If primary lactic acidosis and pyruvate dehydrogenase complex deficiency are suspected based on biochemical (elevated lactate and pyruvate with normal L/P ratio and specific amino acid ratios and their combinations) and clinical phenotype (brain anomalies and neurodevelopmental findings), a multigene panel that includes PDHA1 and other pyruvate dehydrogenase complex deficiency-associated genes of interest (such as the panel noted in legend of Table 2) may be more likely to identify the genetic cause of the condition while limiting the identification of variants of uncertain significance and pathogenic variants in genes that do not explain the underlying phenotype. However, comprehensive exome or genome testing does not require the clinician to determine which gene is likely involved. Therefore, exome or genome sequencing can be used. Although most pathogenic variants in pyruvate dehydrogenase complex deficiency-associated genes to date are within the coding regions and are likely to be identified on exome sequencing, pathogenic splicing variants outside the canonical splice junction, deep intronic variants, as well as structural variants resulting in pyruvate dehydrogenase complex deficiency, could be identified by genome sequencing. Ordering expedited exome or genome sequencing is necessary in cases of critically ill newborns or infants.

Classes of pathogenic variants. Although truncating or nonsense variants in pyruvate dehydrogenase complex deficiency-associated genes are more often associated with severe pyruvate dehydrogenase complex deficiency phenotype, the clinical phenotype of individuals with missense, splice site, or deep intronic variants in those same genes may vary from mild to severe presentations. Individuals with thiamine-responsive PDHA1 variants tend to more often have a mild to moderate phenotype, likely depending on disease onset and start and choices of management interventions after diagnosis is established.

Management

There is no cure for pyruvate dehydrogenase complex deficiency. In most cases, polytherapy is the management approach for individuals with pyruvate dehydrogenase complex deficiency. However, the specific approach depends on the patient and their specific disease. Molecular genetic confirmation of the pyruvate dehydrogenase complex deficiency diagnosis is necessary for best management choices and outcomes.

Ketogenic diet. Life-long use of ketogenic diet (and avoidance of high carbohydrate diets) is the main therapeutic intervention for primary-specific pyruvate dehydrogenase complex deficiency, with positive outcomes noted in the areas of epilepsy, ataxia, sleep disturbance, speech and language development, social functioning, and frequency of hospitalizations (89; 17; 68; 73; 16). Although ketogenic diet therapy has never been evaluated in a clinical trial (and probably will not be in the future), its use has the basis of rationale in the oxidation of beta-hydroxybutyrate and acetoacetate (ketone bodies), which cross the brain barrier and do not depend on pyruvate dehydrogenase complex for oxidation. Oxidation of the ketone bodies provides the brain with an alternative fuel source to generate acetyl-CoA used for maintaining the mitochondrial pools of tricarboxylic acid cycle intermediates and production of reducing equivalents for subsequent ATP production through the mitochondrial respiratory chain, thus restoring the energy deficit characteristic of pyruvate dehydrogenase complex deficiency. Diffusion of ketone bodies across the blood-brain barrier is facilitated by the monocarboxylic transporters.

The proportions of fat, carbohydrate, and protein in ketogenic diet vary considerably in actual practice (eg, 4:1 or 2:1 fat-to-carbohydrate and protein ratio), with recommendations for minimal (less than 10%) energy from carbohydrates (89). The recommended ketogenic diet ratio varies from as low as a ketogenic diet 1:1 fat-to-carbohydrate and protein ratio ("modified" ketogenic diet or equivalent to a modified Atkins diet) to as high as a ketogenic diet 4:1 ratio or slightly more. No studies have definitively demonstrated the superiority of any particular ketogenic diet ratio. Tolerance of ketogenic diet varies between individuals; thus, treatment must be individualized. The recommended amount of dietary fat can vary widely, from approximately 55% to 80%, with variable proportions of unsaturated and saturated fats. The best outcomes are associated with maintaining plasma or serum beta-hydroxybutyrate concentration at 3.0 to 3.5 mEq/L (73). Ketosis is infrequently monitored by measurement of blood beta-hydroxybutyrate or beta-hydroxybutyrate/acetoacetate ratio, but these are helpful measures for maintaining ketosis. Additionally, long-term ketogenic diet use can be detrimental to lipid metabolism with regard to cholesterol or lipid profiles (23). Poor cholesterol/lipid profiles in patients on the ketogenic diet presumably can lead to a higher risk of cardiovascular disease. A less restrictive partial ketogenic diet has been used on a select cohort of patients with DLD-E3 deficiency (categorized as primary-generalized pyruvate dehydrogenase complex deficiency) due to a homozygous DLD p.D479V variant; the less restrictive diet showed improved survival but no significant improvement in quality of life (81). Certain ketogenic diet regimens can be deficient in vitamins and calcium; therefore, they should be supplemented with carbohydrate-free preparations as needed.

Because of the delayed diagnosis of individuals with pyruvate dehydrogenase complex deficiency (17; 73; 87), the efficacy of early perinatal intervention with ketogenic diet has not been systematically investigated. In utero brain anomalies in fetuses later assessed to have primary-specific pyruvate dehydrogenase complex deficiency due to PDHA1, PDHX, and DLAT are well documented (74; 50; 60; 66), but the use of ketogenic diet in a pregnant mother with an affected fetus with known pyruvate dehydrogenase complex deficiency has not been reported. Although the use of beta-hydroxybutyrate in infants has been found to be safe and effective for multiple acyl-CoA dehydrogenase deficiency (85; 86), the utility of beta-hydroxybutyrate (whether as a dietary supplement during inpatient management) for rapid induction of ketosis in newborns or infants with pyruvate dehydrogenase complex deficiency has not been tried.

Thiamine (vitamin B)*. Supplementation with high-dose thiamine (500 to 2000 mg/day) is most commonly used. Several reports of thiamine-responsive PDHA1 variants are known (Table 3) (49; 70; 17; 13; 15; 84; 19). Certain pathogenic PDHA1 variants impacting E1α regions that are either not directly participating in thiamine pyrophosphate binding or have low affinity for thiamine pyrophosphate are considered thiamine-responsive (Table 3) (49; 70; 32; 17; 13; 15; 84; 54; 55; 19). In contrast, the PDHA1 precursor p.R119W variant, where mature R90 directly binds thiamine pyrophosphate and is catalytically crucial, is not responsive to thiamine, at least in vitro in cultured fibroblasts (32). Thiamine use has not been reported to have adverse side effects; consequently, it is commonly administered to patients with pyruvate dehydrogenase complex deficiency, irrespective of responsiveness or efficacy.

Table 3. Thiamine-Responsive Pathogenic Variants of E1α *

Dichloroacetate. Dichloroacetate is a structural analogue of pyruvate and an activator of pyruvate dehydrogenase complex that has been recognized to have potential clinical benefit and has been used in clinical trials. The mechanism of dichloroacetate activation of pyruvate dehydrogenase complex and, thus, lowering of blood lactate was shown to be inhibition of pyruvate dehydrogenase kinases, with different binding sites (78). As an inhibitor of the E1 kinase, dichloroacetate maintains any residual E1 activity of mutant pyruvate dehydrogenase complex in its maximal active (dephosphorylated) form. Two prior clinical trials of dichloroacetate, one controlled, the other open-label, in children with congenital lactic acidosis and adults with various mitochondrial disorders, including pyruvate dehydrogenase complex deficiency, showed a reduction in CSF or blood lactate but did not show significant overall clinical benefit (04; 77). Dichloroacetate has been associated with toxic neuropathy in subjects with other mitochondrial disorders, such as MELAS (35). A difference between children and adults and an individual association with genetic variations have been shown to affect the rate of catabolism of dichloroacetate (and the risk of peripheral neuropathy) (35), which has been proposed as a basis for pharmacokinetic monitoring and management of safe clinical use of dichloroacetate (78). Stimulation of pyruvate dehydrogenase complex with dichloroacetate enhances tricarboxylic acid cycle anabolic bioenergetics in monocytes (90). Furthermore, dichloroacetate metabolism correlates with the expression of the GSTZ1 protein (40). The lower expression of GSTZ1 in Whites, Hispanics, and Asians who possess a specific lysine (K) carrier haplotype of GSTZ1 results in lower enzymatic activity and slower metabolism of dichloroacetate compared with those who possess the non-K carrier haplotype (40). There are three nonsynonymous functional single nucleotide polymorphisms in human GSTZ1 [rs7975 G > A (E32K), rs7972 G > A (G42R), and rs1046428 C > T (T82 M)] that give rise to five major haplotypes in order of population frequency haplotypes, EGT (GSTZ1C, wild type 50%); KGT (GSTZ1B, 28%); EGM (GSTZ1D, 15%); KRT (GSTZ1A, 7%); and KGM (GSTZ1F, 0.4%) (39). Patients with at least one EGT haplotype (EGT carrier) metabolized dichloroacetate faster than EGT noncarriers. The most common dosage of dichloroacetate used is about 25 mg/kg/day divided in two doses for patients with the GSTZ1 EGT carrier haplotype (“fast metabolizer”) and about half this dosage for those GSTZ1 EGT noncarriers (“slow metabolizer”) (39). A controlled Phase III trial of dichloroacetate in children with pyruvate dehydrogenase complex deficiency (ClinicalTrials.gov ID#: NCT02616484), using a validated parental observation of child behavior scale as the primary outcome measure is undergoing result analysis and FDA review.

Lipoic acid. Studies have suggested that exogenous lipoic acid, although beneficial as an antioxidant with very low toxicity, is not utilized for mitochondrial lipoylation (45). Thus, the utility of lipoic acid supplementation in pyruvate dehydrogenase complex deficiency is unclear.

L-carnitine (if on the ketogenic diet). Supplementation with L-carnitine (carbohydrate or sugar-free) is utilized to protect against secondary carnitine deficiency, which is associated with long-term use of the ketogenic diet.

Bicarbonate or buffering agents for metabolic acidosis (due to elevated lactate or ketosis) at 0.5 to 2.5 mEq/kg/day is necessary for individuals with pyruvate dehydrogenase complex deficiency, particularly those on ketogenic diet. Examples of buffering agents include Bicitra (sodium citrate/citric acid) where 1 mL = 1 mEq bicarbonate, Cytra-K (potassium citrate/citric acid) where one packet = 30 mEq bicarbonate with 0.063 g CHO/packet, or a mix of both for sodium or potassium balance.

Any standard antiseizure therapy or other medications or supplements (such as riboflavin) should be carbohydrate- or sugar-free if the patient is on the ketogenic diet. Sleep disturbances, including difficulty falling asleep, are common in pyruvate dehydrogenase complex deficiency, and melatonin (sugar-free) has proven useful.

Medical therapy for the underlying intercurrent illness should be instituted concurrent to treatment for the inborn error of metabolism.

Outcomes

The efficacy of ketogenic diet intervention depends on the pyruvate dehydrogenase complex deficiency disease phenotype and the attainment and maintenance of ketosis (73). The most favorable developmental outcomes are associated with early implementation of the ketogenic diet (89) and are most pronounced in individuals with relatively mild presentations of pyruvate dehydrogenase complex deficiency, such as childhood-onset ataxia (73). Ketogenic diet is most beneficial in individuals with a milder pre-treatment disease course, disease onset after the neonatal period, higher baseline developmental functioning, and absence of structural brain anomalies (73). However, ketogenic diet can be ineffective in patients with severe brain damage in utero or at birth and may be harmful (or even lethal) in cases where pyruvate dehydrogenase complex deficiency is associated with more general impairment of formation of acetyl-CoA (eg, fatty acid beta-oxidation, branched-chain amino acids metabolism defects or E3 deficiency) or oxidation of acetyl-CoA (eg, tricarboxylic acid cycle defects, such as SUCLA2 deficiency or other mitochondrial dysfunction) (20; 05; 31). Ketogenic diet can be difficult to tolerate long-term, and nausea is reported with ketogenic diet use, which has been suggested as a manifestation of excess ketosis (88). Ketogenic diet still leaves patients, particularly those with severe disease and brain anomalies, with significant systemic disease and developmental disability.

Individuals on ketogenic diet should be provided bicarbonate or buffering agents for metabolic acidosis secondary to ketosis and should have scheduled thyroid function, iron, total iron binding capacity and ferritin, vitamins (including vitamin D 25-hydroxy battery), zinc, selenium, and free and total plasma carnitine concentrations evaluated and supplemented as needed for best outcomes. Carnitine deficiency can be seen during the course of ketogenic diet treatment, but it is rarely symptomatic. Selenium deficiency was found in 20% of patients on a ketogenic diet, and this deficiency may not be evident for several months (09). Furthermore, because selenium deficiency can alter normal cardiac function (02), obtaining an initial echocardiogram, electrocardiography, and selenium concentration are advised before initiation of ketogenic diet.

Among more than 130 known nonduplicate PDHA1 pathogenic variants, less than 20 (Table 3) are thiamine-responsive. The efficacy of high-dose thiamine supplementation (usually administered in a sugar-free form) is variable and depends on the individual’s thiamine-responsive PDHA1 pathogenic variant (49; 70; 32; 17; 13; 15; 84; 54; 55). Despite the use of nonphysiologic high-dose thiamine (500 to 2000 mg/day), only up to two thirds maximal in vitro enzymatic efficacy is usually achieved. Consequently, a low-strength ketogenic diet (modified Atkins diet to ketogenic diet 2.5:1) is usually administrated in conjunction with high-dose thiamine for better metabolic balance and more efficacious clinical outcomes and quality of life.

Special considerations

Pregnancy

Once the primary pyruvate dehydrogenase complex deficiency-causing pathogenic variants are identified in an affected family member, molecular genetic prenatal testing for a pregnancy at increased risk and preimplantation genetic testing are possible. There could be differences in perspective among medical professionals and within families regarding the use of prenatal testing, so discussion of these issues may be helpful. Prenatal fetal brain anomalies have been observed, such as cerebellar hypoplasia and ventriculomegaly and ventricular asymmetry, periventricular cystic lesions, corpus callosum anomalies (thin or absent), absent septum pellucidum, basal ganglia and midbrain anomalies and cystic cavitations, mega cisterna magna, hypoplastic cerebellum, cerebellum vermian anomalies or absence, optic nerve hypoplasia at different stages of gestational age (50; 57; 83; 22; author personal experience).

Anesthesia

As with all mitochondrial disorders, propofol should be used with caution or avoided and should not be used for an extended period, such as for prolonged sedation. Similarly, known mitochondrial toxins, such as valproic acid (divalproex sodium) and derivatives, should be avoided. Fentanyl or remifentanil have minimal mitochondrial effects (30). Remifentanil, an opioid with a fast offset because it is rapidly metabolized by serum cholinesterases, has proven useful. The sedative benzodiazepines, as well as the CNS alpha-adrenergic agonist dexmedetomidine, have also proven useful for mitochondrial diseases, although their clearance may be slowed and effects prolonged (61; 30). Analgesic doses of ketamine generally are well tolerated (61; 30).

For pyruvate dehydrogenase complex deficiency patients with malignant hyperthermia, a nontriggering anesthetic consisting of bolus of low-dose propofol for induction, ketamine boluses, benzodiazepine-like midazolam boluses, dexmedetomidine hydrochloride, and remifentanil infusions, +/- muscle relaxation as needed have proven useful as well.