Clinical manifestations

Presentation and course

Primary carnitine transporter deficiency, now known as the “high-affinity carnitine transporter defect”, is a specific entity characterized by carnitine-responsive cardiomyopathy with or without weakness, hypoglycemic hypoketotic encephalopathy, and failure to thrive, with low plasma and tissue concentrations of carnitine (generally less than 5% of normal), lipid storage in muscle, heart, and liver, and severe renal leak of carnitine with absence of an abnormal dicarboxylic aciduria. Exclusion of other defects in fatty acid oxidation is essential. Currently, more than 50 cases of the primary carnitine transporter defect have been documented in the literature. In these cases, either the defect in carnitine uptake has been directly demonstrated in cultured skin fibroblasts (89; 155; 119; 32; 31; 142; 40; 133; 10; 17; 06; 18; 104), or mutations have been identified in the high-affinity carnitine transporter gene, OCTN2 (85; 62; 67; 87). Three other cases warrant consideration on the basis of indirect evidence, including the following symptoms:

	(1) Low serum and tissue carnitine concentrations.
	(2) Cardiomyopathy.
	(3) Rapid and dramatic response to carnitine therapy (or if no carnitine was given, a positive family history of a similarly affected sibling with carnitine-responsive cardiomyopathy).
	(4) Absence of abnormal dicarboxylic aciduria (seen in patients with secondary deficiency due to defects in beta-oxidation).
	(5) A lack of repletion of carnitine tissue stores despite oral carnitine therapy, implying a transporter defect (16; 148).

There appear to be two major clinical presentations of the high-affinity carnitine transporter defect, namely one of earlier onset acute recurrent episodes of hypoglycemic encephalopathy, and a second characterized by slightly later onset slowly progressive hypertrophic and dilative cardiomyopathy with myopathy. However, it should be emphasized that these two presentations are not necessarily mutually exclusive either within the same family or in the same patient.

In 22 cases with a demonstrated transporter defect in fibroblasts (40; 133), the most common initial presenting feature was progressive cardiomyopathy (hypertrophic or dilatative), which occurred in 13 of the patients between 1 and 7 years of age (median age was 3 years). Patients are normal at birth and may appear healthy for several years before the development of the characteristic cardiomyopathy that may be rapidly fatal unless carnitine therapy is instituted. Hypoglycemia was the initial presentation in 10 patients, occurring between 3 months and 2.5 years of age (median age was 1.5 years). Three of these 10 patients had evidence of cardiomyopathy at the time of the initial hypoglycemic episode. In a family, a boy presented with hypoglycemia at 8 months, whereas his brother presented with cardiomyopathy at 7 years of age. In another family where there was consanguinity, a 2-year-old boy presented with hypoglycemic coma; his two brothers had died at the age of 10 months from an identical rapidly progressive illness. Autopsy of one brother revealed lipid accumulation in the liver, a "globoid hypertrophic heart," and low total carnitine concentrations in heart and liver (40). As both presentations have occurred in a single family, Stanley and colleagues have suggested that the differences in clinical manifestations do not reflect different underlying genetic defects, but rather that the circumstances promoting fasting stress or decreased dietary intake of carnitine may predispose a sufferer to acute catabolic episodes, prior to the development of the cardiomyopathy (133). Skeletal muscle weakness was described in six patients at initial presentation and was associated with cardiomyopathy in three patients and with hypoglycemia in one patient. All patients developed a cardiomyopathy if followed long enough, and it is likely that many had muscle weakness, a symptom that may not have been specifically noted because it was mild or of lesser significance in comparison to the severity of the cardiomyopathy (133). A 13-year-old girl with presyncope and palpitations was found to have a shortened rate-corrected QT interval with significant left ventricular dysfunction due to a homozygous mutation in the SLC22A5 gene resulting in an in-frame deletion mutation affecting the OCTN2 protein (47). Following treatment with long-term, high-dose oral carnitine supplementation, her QT interval returned to within normal range with significant improvement of left ventricular function.

In the patients with hypoglycemia, a defect in ketogenesis was noted in 9 of 10 children, at least during the acute catabolic episodes. Hepatic fatty acid oxidation was noted to be intact in one patient after recovery from the acute episode and in a toddler who presented with cardiomyopathy at 3.5 years of age.

Within a few weeks, there was a dramatic improvement in the cardiomyopathy and myopathy of all patients treated with high-dose oral carnitine supplementation; within a few months of therapy, there was a reduction of heart size toward normal. For example, one 4-year-old girl suffered with end-stage cardiac failure, marked left ventricular dilatation, and severe mitral regurgitation; however, within 4 weeks of carnitine therapy, there was an inversion of the former characteristically peaked T-waves in the anterolateral precordial leads on ECG, and a markedly increased exercise tolerance. Within 7 months of therapy, the patient had a dramatic reduction in heart size from 240% to 130% of predicted normal left ventricular volumes. By 6.5 years of age, all cardiac medications could be discontinued and by 9 years of age, she was a competitive swimmer at school (142). Furthermore, three children with significant failure to thrive prior to therapy had a marked improvement in growth after therapy (142). A variegate anemia has also been noted in several cases where the response to carnitine therapy was not striking (142). In contrast, one boy who presented with severe cardiomyopathy and severe anemia improved with carnitine therapy (12). In one patient, the impairment in fasting ketogenesis was corrected with carnitine therapy (146). Eighteen of 19 patients on carnitine therapy for periods of 1 to 10 years continue to be healthy (133). One of these 19 patients is moderately impaired with weakness and cardiomyopathy at the age of 20, which is thought to be due to poor compliance with carnitine therapy. In another follow-up study of three older patients diagnosed nearly a generation ago who, combined, were treated for more than 50 patient years, the patients continued to respond to carnitine therapy and remained well except for the irreversible sequelae of the pretreatment illnesses (13). Peripheral neuropathy has also been reported in a 3-year-old child with the carnitine transporter defect (83). Attention deficit/hyperactivity disorder responding to carnitine therapy has been described as an associated feature in OCTN2 deficiency (63).

Plasma total carnitine concentrations were markedly decreased in all 22 patients; 19 had values less than 5 µmol/L (normal is 40 to 60 µmol/L) and two had values between 5 and 10 µmol/L. The ratio of esterified to free carnitine tended to be normal. Muscle total carnitine concentrations in 14 patients ranged from 0.05% to 20% of the normal mean; six patients had values below 1%, five patients had values of 1% to 5%, two patients had values of 6% to 10%, and one patient had a value of 20% of the control mean. Liver total carnitine was 5% of normal in one patient (146) and was 6% of normal in the autopsied liver of the undiagnosed brother of a boy with a demonstrated transporter defect in cultured skin fibroblasts (40). In this same boy, the heart total carnitine at autopsy was 2% of the control mean. In a 4-year-old girl with a demonstrated carnitine uptake defect in fibroblasts, the carnitine concentration in heart was less than 1% of controls (32; 31). In all patients tested, the urinary organic acid profiles were normal or showed only modest elevations of medium-chain dicarboxylic acids, within the range found in fasting children. Following carnitine therapy, the plasma carnitine concentrations approached the normal range (varying from 9 to 60 µmol/L) but were infrequently elevated due to impaired renal conservation. Out of 12 patients, one patient had plasma concentrations that varied from 9 to 28 µmol/L, six patients had concentrations of 30 to 39 µmol/L, and five patients had concentrations of 40 to 60 µmol/L. Impaired renal conservation of carnitine has been reported in several patients (119; 32; 146; 31; 133). In one patient, over the 3 days following acute withdrawal of oral carnitine, there was a rapid decrease in plasma total carnitine concentrations from 18 to 3.3 µmol/L, whereas the fractional excretion of free carnitine ranged between 74% and 230% of the filtered load (normal is less than 5%). A patient was detected by newborn screening with a total carnitine concentration of 67% of the normal value (24). At 1 year of age, after interruption of carnitine supplementation for a 4-week period, the total carnitine had dropped to 12.7 µmol/L (normal is 25 to 65 µmol/L) with a free of 10.4 µmol/L. This infant was homozygous for an OCTN2 gene mutation (p.P46S). The authors emphasized that neonates with primary carnitine deficiency might present with relatively high levels of total carnitine due to placental carnitine transfer and drew attention to the importance of regular follow-up and genetic diagnosis in patients with nonclassical presentations.

The documented muscle carnitine deficiency was not corrected with oral carnitine therapy. The muscle carnitine concentrations increased by 2.5-fold to 6-fold of the pretreatment values; however, the absolute carnitine concentrations only reached 0.7% to 13% of control values, suggesting that the transport defect might also be present in muscle. Defective muscle carnitine uptake has now been directly confirmed in myoblast culture from a child with the fibroblast-confirmed carnitine transporter defect (104). In contrast, the liver carnitine concentrations in one patient increased from 55 nmol/g before therapy to 740 nmol/g after therapy (normal is 900 to 1800 nmol/g) (146), suggesting that the deficiency in liver was corrected by raising plasma carnitine concentrations toward normal.

Asymptomatic mothers with primary carnitine deficiency were identified by low carnitine concentrations in their infants on newborn screening (116). Carnitine transport was significantly reduced in fibroblasts obtained from all individuals with primary carnitine deficiency but was significantly higher in the fibroblasts of asymptomatic women than in those of the symptomatic individuals (p< 0.01). DNA sequencing indicated an increased frequency of nonsense mutations in symptomatic individuals (p< 0.001). Cells from asymptomatic women had higher mean concentrations of residual carnitine transport activity compared to that of symptomatic individuals due to the presence of at least one missense mutation (116); this provides a genotype-phenotype correlation and explanation for the asymptomatic state of the mothers. A woman in her early twenties presented with syncopal episodes caused by ventricular tachycardia and a prolonged QT interval and was found to have primary carnitine deficiency due to two mutations in the SCL22A5 gene (23). The arrhythmias were poorly controlled by pharmacologic therapy, and a defibrillator was installed. After diagnosis and treatment with high-dose L-carnitine, no further syncopal episodes have occurred, and the QT interval returned to normal. As a precaution, low-dose metoprolol therapy and a defibrillator are still in place.

In a nationwide screening program in the Faroe Islands, the prevalence of primary carnitine deficiency was found to be the highest reported in the world (1 out of 300) (107). Genetic analysis was performed in all individuals with a blood free carnitine of less than 7 µmol/L, and genetic analysis revealed homozygosity or compound heterozygosity for mutations in the SLC22A5 gene. This program identified 76 Faroese adults (ages 15 to 80 years). All patients identified were either asymptomatic or had minor symptoms when diagnosed (106). A review of medical records among the included patients revealed that 37 patients had previously been admitted to the national hospital in the Faroe Islands prior to diagnosis of primary carnitine deficiency. A female patient, aged 34 years, survived ventricular fibrillation following exposure to pivalic acid, which is known to lower carnitine levels. A male patient, aged 29 years, had several previous admissions with hypoglycemia and seizures and was diagnosed with epilepsy, and a 28-year-old female had a minor myocardial infarction. Echocardiography, including LVEF, global longitudinal strain, and dimensions, was normal apart from left ventricular hypertrophy with normal systolic function in one young male. Symptoms such as fatigue were reported in 43% with a reduction to 12% (p< 0.01) following initiation of L-carnitine supplementation. Of note, 82% reported participation in sports, of which 52% were on a competitive level. ECGs showed limited changes; eight patients were identified with atrial fibrillation, bundle branch block, T-wave inversion, or ST junction depression, and the rest had minor or no changes. Twenty-four-hour cardiac telemetry showed no ventricular arrhythmias. The mean plasma free carnitine increased from 6.1 µmol/L to 15.1 µmol/L (p< 0.01) within 50 days of L-carnitine supplementation (mean dosage of 46 mg/kg/day).

The obligate heterozygote parents have been essentially clinically asymptomatic and may have normal or moderately reduced plasma total carnitine concentrations, which presumably reflects some impairment in renal reabsorption. The plasma carnitine concentrations in the mothers ranged from 20 to 29 µmol/L in eight of the women, and from 30 to 39 µmol/L in four of the women. In the fathers, concentrations ranged from 20 to 29 µmol/L in three of the men, from 30 to 39 µmol/L in eight of the men, and from 40 to 60 µmol/L in one of the men (142; 133).

Prognosis and complications

The primary complications would arise from a lack of recognition of the disorder, with a progression of the clinical symptomatology including the cardiomyopathy, myopathy, and failure to thrive, as well as the risk of sudden infant death in the presence of fatty acid oxidation stressors such as fasting, cold exposure, stress, and infection. In the case of late recognition and treatment of hypoketotic, hypoglycemic encephalopathy, this could result in significant injury to the central nervous system with secondary cognitive delay and a seizure disorder.

Clinical vignette

A 9-year-old girl was born to unrelated asymptomatic parents of East Indian and Caucasian heritage (142). Her sister presented with congenital hypotonia and muscle atrophy, and at 1 year of age was found to have progressive cardiomyopathy with severe mitral regurgitation. This sibling also had failure to thrive and progressive motor deterioration and died at 3.5 years of age; fatty infiltration of the liver was noted at necropsy.

The index patient was born a footling breech. At 1 month of age, a cardiac murmur was noted. By 14 months, her gross motor milestones reached a plateau, and over the next 2 years, she developed slowly progressive weakness and recurrent respiratory infections. At 3 years of age, her clinical findings included congestive heart failure with moderate-to-severe mitral regurgitation. Marked left ventricular dilatation was present on cardiac echogram. The ECG demonstrated peaked T waves in the anterolateral precordial leads. She was started on digoxin and diuretics. At 4 years and 3 months of age, she was found to have low carnitine concentrations in serum and in muscle. The total muscle carnitine was less than 5% of control values. A muscle biopsy showed lipid storage, but beta-oxidation enzymes were normal. Serum creatine kinase was mildly elevated. Serum glucose was normal, as were urinary organic acids. There was no abnormal dicarboxylic aciduria. Oral carnitine was instituted at 1 g three times daily. Within 1 week, appetite, affect, and cardiac function had dramatically improved. By 4 weeks, she had gained weight and had markedly increased exercise tolerance. The ECG also showed inversion of the formerly peaked T waves. The predicted left ventricular end systolic and end diastolic dimensions markedly improved from 240% and 205% of normal, respectively, to 175% and 160% of normal. By 4 months, she had good exercise tolerance, and by 7 months of therapy, the predicted left ventricular end systolic and end diastolic dimensions had decreased to 150% and 130% of normal, respectively. At 5 years of age, she had regained gross motor milestones. Her digoxin and diuretics were discontinued by 6.5 years of age. At 9 years of age, she had an IQ of 140 and was a competitive swimmer at school. On examination, she had a mildly myopathic facies, mild hypotonia, and persistent, although improving, fine motor delay. Her electromyogram was normal. Motor nerve conduction studies revealed borderline abnormalities. Her serum carnitine concentrations fluctuated between borderline low and normal (total serum carnitine of 25 to 60 µmol/L and free of 17 to 33 µmol/L; controls total 51.5 ± 11.6 µmol/L and free 40.0 ± 9.5 µmol/L), and she continued to have decreased renal reabsorption of carnitine. Her father's serum carnitine concentrations were normal (total 60.9 µmol/L; free 54.7 µmol/L), and her mother's serum carnitine concentrations were low normal (total 35.7 µmol/L; free 27.8 µmol/L).

Carnitine uptake studies in the cultured skin fibroblasts of the patient showed minimal carnitine uptake throughout the entire range of physiologic concentrations of carnitine, precluding the calculation of Michaelis constant and maximal velocity values. At a carnitine concentration of 5 µmol/L (Michaelis constant value), the mean rate of uptake was 2% of the control value. Carnitine uptake studies in the cultured skin fibroblasts of both parents revealed normal Michaelis constant values of 5.0 µmol/L (controls are 5.5 ± 0.58 µmol/L) but reduced maximal rates of carnitine uptake with a maximal velocity of 1.33 pmol/min per mg protein (controls are 3.4 ± 0.36), which was 40% of control values. The normal Michaelis constant values for carnitine substrate concentration and the reduced maximal velocity values for carnitine uptake in the heterozygote parents suggested the presence of a reduced number of normally functioning carnitine transporters. The family history of a previously affected sibling, as well as the demonstrated heterozygosity in both parents on carnitine uptake studies, supported an autosomal recessive pattern of inheritance.

Biological basis

The pathophysiology of OCTN2 deficiency relates to the insufficient uptake of carnitine into tissues by plasmalemmal OCTN2, leading to tissue carnitine deficiency and, thereby, to defective long-chain fatty acid oxidation and to the inappropriate renal losses of free carnitine due to the renal reabsorption defect of free carnitine related to the defective renal OCTN2.

Etiology and pathogenesis

Theoretical causes of primary carnitine deficiency as suggested by Rebouche and Engel (113) include the following:

Defective biosynthesis combined with reduced dietary intake. In omnivores, approximately 75% of carnitine stores are obtained from the diet, and approximately 25% of stores are endogenously synthesized. In strict vegetarians, endogenous carnitine synthesis provides over 90% of total available carnitine (108). All human tissues studied (skeletal muscle, heart, liver, kidney, and brain) are capable of the biosynthesis of carnitine from the essential amino acids methionine and lysine to carnitine's immediate precursor, gamma-butyrobetaine (111). The final conversion of gamma-butyrobetaine to L-carnitine by gamma-butyrobetaine hydroxylase can only be done in the liver, kidney, and brain in humans (30; 111). Thus, gamma-butyrobetaine must be exported to these tissues for final conversion to L-carnitine, and then L-carnitine in its final form can be taken up by all tissues.

Of interest in a study of vegetarians given intravenous L-carnitine infusions, it was found that there was a reduction in skeletal muscle carnitine transport by OCTN2 as demonstrated by a 33% reduction (p< 0.05) in muscle OCTN2 mRNA and 37% reduction (p=0.09) in muscle OCTN2 protein compared to nonvegetarian controls (135). It was concluded that vegetarians have a lower muscle total carnitine and reduced capacity to transport carnitine into muscle than nonvegetarians, possibly because of reduced muscle OCTN2 content.

Defective intestinal absorption. Under normal conditions (ie, in omnivores), about 70% to 80% of dietary carnitine is absorbed (109). Colonic OCTN2 gene expression has been shown to be upregulated by peroxisome proliferator-activated receptor gamma in humans (22).

Defective transport. The defective transport of carnitine affects the normal uptake or release of carnitine from tissues.

Renal loss due to decreased tubular reabsorption or increased excretion of carnitine. At normal physiological concentrations in plasma, more than 90% of filtered carnitine is reabsorbed by the kidney (115).

Increased degradation. No evidence for defective carnitine biosynthesis (110) or defective absorption or excessive degradation (114) was found in several patients with "systemic" carnitine deficiency; however, all of these patients were subsequently found to have medium-chain acyl-CoA dehydrogenase deficiency. In mammals, carnitine degradation is not of any quantitative importance and is primarily accomplished by bacteria in the gut (09).

In 1988, Eriksson and colleagues and Treem and associates offered definitive evidence for a defect in the plasma membrane uptake of carnitine in cultured skin fibroblasts, muscle, kidney, and presumably heart (32; 146). In 1997, Pons and coworkers confirmed defective muscle carnitine uptake in myoblast culture from a child with a fibroblast-proven carnitine transporter defect (104).

Carnitine (beta-hydroxy-gamma-trimethylaminobutyric acid), a water-soluble quaternary amine, has several important intracellular functions (09; 07), which are listed below:

	(1) It serves as an essential cofactor for mitochondrial fatty acid oxidation by transferring long-chain fatty acids as acylcarnitine esters across the inner mitochondrial membrane.
	(2) It facilitates branched-chain alpha-keto acid oxidation.
	(3) It shuttles acyl-moieties that have been chain-shortened by beta-oxidation out of peroxisomes in the liver.
	(4) It modulates the intramitochondrial acyl-CoA and CoA sulfhydryl ratio in mammalian cells.
	(5) It traps potentially toxic acyl-CoA metabolites that may increase excessively during acute metabolic crises through esterification. These metabolites may impair the citric acid cycle, gluconeogenesis, urea cycle, and fatty acid oxidation.

In omnivores, approximately 75% of carnitine sources come from the diet and approximately 25% come from endogenous synthesis (108). Major dietary sources of carnitine are meat, poultry, fish, and dairy products (114). Under normal conditions, about 70% to 80% of dietary carnitine is absorbed in omnivores (109). In strict vegetarians endogenous carnitine synthesis provides greater than 90% of total available carnitine (108). Human milk and most milk-based formulas contain adequate amounts of carnitine to sustain early growth and development. In term infants who are fed unsupplemented soy protein-based formulas, plasma carnitine concentrations were markedly lower than in carnitine-supplemented infants (94). However, no deficits in growth or development were identified in term infants consuming essentially carnitine-free formulas over at least the first 4 months of life (99), despite evidence suggesting greater utilization of alternative extramitochondrial pathways for oxidation of fatty acids in the unsupplemented group. Skeletal muscle contains over 90% of total body carnitine in humans, and the concentration of skeletal muscle carnitine is approximately 70 times higher than in plasma (108). The plasma carnitine concentrations are regulated largely by the renal threshold, which is approximately 40 µmol/L (29). Active transport of carnitine across the proximal renal tubule minimizes the urinary loss. At normal physiological concentrations in plasma, more than 90% of the filtered carnitine is reabsorbed by the kidney (115). Urinary loss and growth requirements define the need for carnitine, as it does not undergo significant metabolic degradation (25). All human tissues studied (skeletal, muscle, heart, liver, kidney, and brain) are capable of the biosynthesis of carnitine from the essential amino acids (methionine and lysine) to carnitine's immediate precursor, gamma-butyrobetaine (111). The final conversion of gamma-butyrobetaine to L-carnitine by gamma-butyrobetaine hydroxylase can only be done in liver, kidney, and brain in humans (30; 111). Thus, gamma-butyrobetaine must be exported to these tissues for final conversion to L-carnitine, and then L-carnitine in its final form can be taken up by all tissues. The hepatic gamma-butyrobetaine hydroxylase is developmentally regulated, being about 25% of adult enzymatic activity at birth (100).

Children with the plasmalemmal high-affinity carnitine transporter defect are characterized by low (less than 5%) serum and tissue (muscle, heart, and liver) concentrations of carnitine, and by decreased renal reabsorption of carnitine (32; 146; 142). Impairment of fatty acid oxidation is suggested by deficient ketogenesis during fasting and by the accumulation of lipid in tissues. The tissues most severely affected, namely heart, muscle, liver, and kidneys, derive much of their energy from long-chain fatty acid oxidation, and carnitine is essential for this process. The normal acyl-CoA dehydrogenase activities in the cultured skin fibroblasts of one patient (146), and the absence of significant dicarboxylic aciduria in all patients (142; 133), indicate that the carnitine deficiency in these children is not secondary to other defects of fatty acid oxidation known to be associated with secondary carnitine deficiency. The rapid and dramatic clinical improvement in cardiac function, strength, and growth following high-dose oral carnitine supplementation also suggests that the defect is due to primary carnitine deficiency. In contrast, in patients with secondary carnitine deficiency, it is still controversial whether carnitine therapy is of any benefit in either the acute or chronic situation (131). Furthermore, increasing evidence suggests that carnitine therapy may be potentially harmful in intramitochondrial defects of long-chain fatty acid oxidation, through the excessive accumulation of long-chain acylcarnitines. These long-chain acylcarnitines may be arrhythmogenic (20) and have been shown to have detergent properties on isolated canine myocytic sarcolemmal membranes, and to potentiate free radical-induced lipid membrane peroxidative injury in ischemia (82).

Primary carnitine deficiency is known to be due to a defect in the active transport of carnitine across the plasma membrane, namely in the specific high-affinity carrier-mediated carnitine transporter OCTN2; studies of carnitine uptake in vitro in cultured skin fibroblasts from patients with primary carnitine-responsive cardiomyopathy support this concept (146; 31; 142). Under normal conditions, the carnitine concentration in tissues (other than brain) is 20-fold to 70-fold higher than in plasma and parallels the capacity of the tissue to metabolize fatty acids. Human tissue concentrations are as follows: heart concentrations (3500 to 6000 nmol/g) are greater than muscle; muscle concentrations (2000 to 4600 nmol/g) are greater than liver; liver concentrations (1000 to 1900 nmol/g) are greater than brain; finally, brain concentrations are 200 to 500 nmol/g (131). Therefore, uptake occurs across a large concentration gradient, and is maintained by a transport system that is generally held to be sodium-gradient-dependent as well as energy-dependent (09; 115; 07). Because carnitine is not significantly degraded in the body, the kidney is capable of adjusting to wide variations in dietary carnitine, as the renal threshold for carnitine is 40 µmol/L, and this concentration is identical to the normal serum concentration (29). The carnitine deficiency associated with the transporter defect has two components: (1) the renal reabsorption of carnitine is impaired, leading to extremely low plasma carnitine concentrations; (2) tissues that share the defect, such as muscle and fibroblasts, are unable to concentrate carnitine, and, therefore, intracellular carnitine concentrations increase little when carnitine therapy is instituted to increase plasma concentrations to normal. This is supported by the lack of repletion of muscle carnitine concentrations in several patients who were biopsied again after high-dose carnitine therapy. A defect in muscle carnitine uptake has now been directly demonstrated in myoblast culture from a child with the fibroblast-proven carnitine transporter defect (104). In the case of Treem and colleagues, the muscle carnitine concentration increased from 0.4% to 2.6% of normal (146), and in the case of Tein and colleagues, the muscle carnitine concentration increased from 5% to 13% of normal (142). In the first patient, there was a reduction in fat droplets in type I muscle fibers from 20% to 5%; in the second patient, there was a complete resolution of the lipid storage. The second patient also regained normal strength, suggesting that the amount of intracellular carnitine required for efficient long-chain fatty acid oxidation for muscular function is low (eg, greater than 5% normal).

The therapeutic effect of carnitine administration may result from the flooding of an unaffected lower-affinity carnitine transporter, thereby bypassing the specific high-affinity carnitine transporter. This is consistent with the data in fibroblasts showing that in the absence of high-affinity carnitine uptake, intracellular concentrations of carnitine passively follow the extracellular concentrations (133). In contrast to muscle, liver carnitine concentrations rose dramatically from 5% to 55% of normal in one patient after oral supplementation (146), suggesting that the carnitine depletion in liver was due to low serum carnitine concentrations.

The heart carnitine concentrations were also low (below 1%) in a 4-year-old girl with a demonstrated uptake defect in fibroblasts (32; 31), and 2% of control mean in the autopsied heart of an undiagnosed brother of a boy with demonstrated transporter defect (40). Whether the severe cardiomyopathy is due to low serum carnitine or to a specific transporter defect is unclear, as no patient has had repeat endocardial biopsies following carnitine therapy. In an OCTN2 defective human-induced pluripotent stem cell lines model, carrying a full OCTN2 knockout and a homozygous OCTN2 (N32S) loss-of-function mutation, which demonstrated lower force development and resting length in engineered heart tissue compared with isogenic control, genome-wide analysis and pharmacological inhibitor experiments have identified ferroptosis, an iron- and lipid-dependent cell death pathway associated with fibroblast activation, as a novel primary carnitine deficiency cardiomyopathy disease mechanism (80).

The clinical and in vitro findings in patients with a carnitine uptake defect are compatible with a disorder of the plasma membrane transport system for carnitine that is predominantly expressed by muscle, heart, kidney, and fibroblasts, but not by liver (32; 146; 142). Furthermore, the Michaelis constant values for carnitine uptake in cultured human heart cells (4.8 ± 2.2 µmol/L) (04) and muscle (1.90 ± 1.38 µmol/L) (112) are similar to the Michaelis constant observed in cultured human skin fibroblasts of 3.24 ± 0.58 µmol/L (146) to 5.5 ± 0.58 µmol/L (142), and different from the Michaelis constant values observed in human liver (500 µmol/L) and brain (greater than 1000 µmol/L) (07).

Stanley and colleagues suggest that the carnitine uptake defect appears to have more severe consequences for muscle and heart than for other tissues, such as liver or fibroblasts, and that heart and muscle may have a higher requirement for carnitine than other tissues (133). They point out that the Michaelis constant for carnitine of carnitine palmitoyltransferase 1,I (1 of the key rate-limiting enzymes in long-chain fatty acid oxidation) in cardiac and skeletal muscle has been reported to be 5-fold to 10-fold higher than for the hepatic carnitine palmitoyltransferase 1,I isoform (86). Of interest, it has been shown that muscle contraction induced by electrical stimulation of rat hindlimb muscles facilitates carnitine uptake in the stimulated skeletal muscle, possibly via contraction-induced translocation of its specific transporter OCTN2 to the plasma membrane (38). In studies of L-carnitine homeostasis during normal postnatal development in the wild-type rat, it has been shown that postnatal increases in heart L-carnitine concentrations are significantly correlated to postnatal increases in heart OCTN2 expression (77). Of interest, in a randomized, placebo-controlled study of healthy young adults, the ingestion of high-dose caffeine stimulated plasma carnitine clearance during hypercarnitinemia, which the investigators proposed was likely caused by increased Na+/K+ ATPase pump activity leading to muscle carnitine loading (156).

In the studies of carnitine uptake in cultured skin fibroblasts from the affected patients, there is minimal or no uptake of carnitine throughout the entire range of physiological concentrations of carnitine, thereby precluding the possibility of calculating the Michaelis constant and maximal velocity values. At a carnitine concentration of 5 µmol/L, the mean rate of uptake in four patients was 2% of that of controls (142). All of the parents studied of these four affected children showed intermediate maximal rates for carnitine uptake, ranging from 13% to 44% of control maximal velocity values (0.44 to 1.50 pol/min vs. 3.4 pmol/min per mg fibroblast protein), whereas Michaelis constant values were normal. Similarly, Stanley and colleagues demonstrated a maximal velocity value in parents that was 40% of control subjects, but normal Michaelis constant values (133). The finding of normal Michaelis constant values coexisting with reduced maximal velocity values in all of the parents' fibroblasts suggests the presence of a reduced number of normally functioning carnitine transporters and supports an autosomal recessive pattern of inheritance. Although the parents of affected patients with the carnitine uptake defect are asymptomatic, the carnitine uptake defect is partially expressed. The parents may have normal or modestly reduced plasma carnitine concentrations; this presumably reflects some impairment in renal reabsorption. The plasma carnitine concentrations in the mothers ranged from 20 to 29 µmol/L in eight women and from 30 to 39 µmol/L in four women; in the fathers, concentrations ranged from 20 to 29 µmol/L in three men, from 30 to 39 µmol/L in eight men, and from 40 to 60 µmol/L in one man (142; 133). The fibroblast carnitine concentration was reduced to 55% of controls in the mother of an affected child (32). In studies performed by Stanley and colleagues on the effect of the carnitine uptake defect on the steady-state concentrations of intracellular carnitine in fibroblasts equilibrated with different concentrations of carnitine for 48 hours, control subjects demonstrated a steep increase in intracellular carnitine with extracellular carnitine concentrations in the range of Michaelis constant for carnitine uptake (133). Beyond 10 µmol/L, intracellular carnitine increased linearly with extracellular carnitine, probably reflecting passive diffusion. In contrast, in patient fibroblasts, intracellular carnitine only increased linearly with extracellular carnitine. In the fibroblasts from parents, intracellular fibroblast carnitine concentrations were intermediate between patients and control subjects and followed a curvilinear relationship to extracellular carnitine similar to control subjects. Tissue concentrations of carnitine have not been measured in parents, but it is likely that muscle carnitine concentrations would be lower in heterozygotes, both because of their lower plasma concentrations and because of their partial defect in muscle uptake.

There have been major advances in the molecular characterization of the high-affinity carnitine transporter. Wu and colleagues cloned a full-length cDNA for OCTN2, which is a member of the organic cation transporter family, from a human placental trophoblast cell line (162). The OCTN2 cDNA encodes for a 557-amino acid protein with a predicted molecular mass of 63 kDa. The organic cation transporters function primarily in the elimination of cationic drugs (and other xenobiotics) in tissues such as the kidney, intestine, liver, and presumably, the placenta. Tamai and colleagues found that OCTN2 from a human kidney cDNA library shared high homology with OCTN1, and identified OCTN2 as a physiologically important, high affinity, sodium-dependent carnitine transporter in humans (138). They demonstrated that OCTN2 is strongly expressed in the kidney, skeletal muscle, heart, and placenta in adults on Northern blot analysis. Homology search of the Genbank database indicates that the human gene (30 kb) (SLC22A5) coding for OCTN2 has been sequenced in its entirety as a part of the Human Genome Project and consists of 10 exons and 9 introns (162). Shoji and colleagues demonstrated tight linkage between the primary systemic carnitine deficiency disease allele and D5S436 on 5q (129). Haplotype analysis revealed that the responsible genetic locus lay between D5S658 and D5S434. The closest microsatellite marker D5S436 was located at 5q31.1. This region on 5q is syntenic with the murine juvenile visceral steatosis (murine model for primary systemic carnitine deficiency) gene located on chromosome 11 of the mouse (92).

Lamhonwah and Tein studied the expression of OCTN2 in cultured fibroblasts and lymphoblasts in two unrelated patients who had previously documented carnitine uptake defects (142; 69). In both patients, they found truncating mutations in the cDNA, and the abnormal transcripts showed a partial cDNA deletion of nucleotides 255-1649, resulting in a predicted truncated null protein of 92 amino acids. Both patients were compound heterozygotes. In one patient the second mutant allele revealed a 19-bp insertion between nucleotides 874 and 875, resulting in a frameshift yielding a predicted truncated protein of 284 amino acids, whereas in the second patient, the second mutant allele had a deletion of nucleotides 875 through 1046, resulting in a predicted truncated protein of 237 amino acids. These three frameshift mutations in the OCTN2 cDNA resulted in a premature stop codon and truncated protein that would make the carnitine transporter nonfunctional and would account for the negligible uptake on kinetic studies. Nezu and colleagues identified four mutations in three pedigrees with the carnitine uptake defect (91). Affected individuals in one family were homozygous for the deletion of a 113-bp region containing the start codon. In the second pedigree, the affected individual was shown to be a compound heterozygote for two mutations that caused a frameshift and a premature stop codon. In an affected individual belonging to a third family, there was a homozygous splice-site mutation also resulting in a premature stop codon. Wang and colleagues reported different mutations in two unrelated patients (160). The first patient was homozygous (and both parents heterozygous) for a single base pair substitution converting the codon for Arg-282 to a stop codon (R282X). The second patient was a compound heterozygote for a paternal 1-bp insertion that produced a stop codon (Y401X), and a maternal 1-bp deletion that produced a frameshift, creating a subsequent stop codon (458X). These mutations decreased the levels of mature OCTN2 mRNA and resulted in nonfunctional proteins. The R282X mutation has also been reported by Burwinkel and colleagues, and by Vaz and colleagues (11; 153). Other reported mutations include R169Q (58; 91; 141), R169W (159), M179L (58), Y211C (153), G242V (159), W283R (85), W283C (58), A301D (159), W351R (159), V446F (85), E452K (158), S467C (58), and P478L (141). Lamhonwah and colleagues reported on 11 mutations (delF23, N32S and one 11-bp duplication in exon 1; R169W in exon 3; a donor splice mutation (IVS3+1 G> A) in intron 3; frameshift mutations in exons 5 and 6; Y401X in exon 7; T440M, T468R, and S470F in exon 8) in 11 children (67). The carnitine uptake (at Km of 5 µM) in the cultured skin fibroblasts of these affected children ranged from 1% to 20% of normal controls. There was no correlation between residual uptake and severity of the clinical presentation suggesting that the wide phenotypic variability is likely related to exogenous stressors exacerbating carnitine deficiency. Most importantly, strict compliance with carnitine from birth appeared to prevent the development of the phenotype. A homozygous deletion of 17081C of the OCTN2 gene, resulting in a frameshift at R282D and leading ultimately to a premature stop codon (V295X) in the OCTN2 transporter, has been associated with two affected homozygous cardiomyopathic children and three homozygous sudden deaths, two of which corresponded to the classic sudden infant death syndrome (SIDS) phenotype (87). Furthermore, a truncating R254X mutation, previously described as a founder mutation in the Chinese population, has now been reported in a Saudi Arabian kindred, suggesting that it may be a recurrent mutation or an ancient founder mutation (68).

Most missense mutations identified in individuals with OCTN2 deficiency affect putative transmembrane or intracellular domains of the transporter. Exceptions are the substitutions P46S and R83L, which are located in the extracellular loop close to the putative glycosylation sites (N57, N64, N91) of OCTN2. It has been shown that the mutations P46S and R83L impair physiological glycosylation of OCTN2 (34). It has further been demonstrated that glycosylation does not affect maturation of OCTN2 transporters to the plasma membrane, but the three asparagines that are normally glycosylated are located in a region important for substrate recognition and turnover. It is estimated that over 100 mutations have been reported and that the c.136C>T (p.P46S) mutation is a frequent mutation (81). In a series of 20 Chinese patients with OCTN2 deficiency, 18 different mutations were identified, the most frequent mutation being c.760C>T (p.R254X), which occurred in 25% of these patients (46). Homozygous patients with R254X were late-onset cases that presented with dilated cardiomyopathy and muscle weakness after 1 year of age.

In a study of 140 subjects in whom carnitine transport was 20% or less of normal in fibroblasts, mutations in the coding region of the SLC22A5 gene could not be identified in about 16% of the alleles causing primary carnitine deficiency (37). Prediction algorithms failed to determine the functional effects of amino acid substitutions in the OCTN2 protein in approximately 20% of cases. It was, therefore, concluded that functional studies in fibroblasts remain the best strategy to confirm or exclude a diagnosis of primary carnitine deficiency (37). In a 2019 study from the Netherlands, a novel 5’UTR c.-149 G>A variant, which introduces a functional upstream out-of-frame translation initiation codon that suppresses translation from the wild-type ATG of SLC22A5 resulting in reduced OCTN2 protein and lower OCTN2 transport activity, was identified in 57 of 235 individuals with OCTN2 deficiency (33). The allele frequency was calculated to be 24.2%, making this variant the most frequent cause of primary carnitine deficiency in this cohort in the Netherlands.

In a study of 150 OCTN2 variants identified in diverse ancestral populations, uptake assays with 14C-carnitine revealed that 105 variants (70%) significantly reduced transport of carnitine compared to wild-type OCTN2, and 37 variants (25%) severely reduced function to less than 20% of normal (59). All ancestral populations harbored loss-of-function variants; 62% of green fluorescent protein (GFP)–tagged variants impaired OCTN2 localization to the plasma membrane of human embryonic kidney (HEK293T) cells, and subcellular localization significantly associated with function, revealing a major loss-of-function mechanism for OCTN2 deficiency. These investigators designed a model to classify variants as functional (> 20% function) or loss-of-function (< 20% function) to improve interpretation of variants for clinical decision-making in the diagnosis and treatment of OCTN2 deficiency.

OCTN2 belongs to a family of organic cation transporters, which are important in the elimination of cationic drugs (and other xenobiotics) in tissues such as kidney, intestine, placenta, and liver (105; 150; 165). OCTN2 expression has been demonstrated on the brush-border membrane of differentiated Caco-2 cells (27). OCTN2 has also been shown to be functionally involved in the transfer of acetyl-L-carnitine across the blood-brain barrier in mice (50). OCTN2 immunoreactivity has been detected in rat astrocytes by immunocytochemical staining, and inhibition of OCTN2 expression by RNA interference significantly inhibited L-3H-carnitine and acetyl-L-3H-carnitine uptake into astrocytes, suggesting that OCTN2 is functionally expressed in rat astrocytes and is responsible for L-carnitine and acetyl-L-carnitine uptake into these cells (51). OCTN2 is also expressed in human sperm with a Km of 3.39 ± 1.16 µmol/L (163). It was localized to the basolateral membrane of primary-cultured rat epididymal epithelial cells as the first step of permeation from blood to spermatozoa (55). It was also detected in primary-cultured rat Sertoli cells, which constitute part of the blood-testis barrier (54). L-carnitine transport in human placental brush-border membranes has been shown to be mediated by OCTN2 with an apparent Km of 11.09 ± 1.32 µmol/L (61). Grube and colleagues have demonstrated that OCTN2 is localized in the apical membrane of syncytiotrophoblasts, suggesting an important role in the uptake of carnitine during fetal development (44). L-carnitine deficiency due to a defect in OCTN2 in a mouse model has been shown to lead to reduced placental concentrations of L-carnitine to less than 10% of normal and to embryonic lethality (127). Localization of OCTN2 by in situ hybridization, laser microdissection, and immunofluorescence microscopy revealed expression of OCTN2 mainly in endothelial cells of the human heart (43). In human skeletal muscle, insulin has been shown to stimulate L-carnitine accumulation during hypocarnitinemia, which is associated with an increase in OCTN2 transcription (134).

Of great interest has been the association of OCTN1/2 variants within the IBD5 locus on disease susceptibility and growth parameters in inflammatory bowel disease (ulcerative colitis and Crohn disease) (93; 118; 157; 101). Of note, the OCTN2-deficient mouse (juvenile visceral steatosis mouse) spontaneously develops intestinal villous atrophy, breakdown, and inflammation with intense lymphocytic and macrophage infiltration, leading to ulcer formation and gut perforation (128). In a murine model of colonic inflammation, L-carnitine was shown to display immunosuppressive properties with abrogation of both innate (interleukin-1beta and IL-6 production) and adaptive (T cell proliferation in draining lymph nodes) immune responses, thereby decreasing intestinal inflammation (36). In further studies, severe carnitine deficiency in Octn2-/- newborn mice has been shown to lead to severe gut and immune phenotypes with widespread villous atrophy, increased apoptosis of enterocytes, and a pro-inflammatory response with gut injury, suggesting that carnitine plays a major role in neonatal Octn2-/- mouse gut development and differentiation (130).

Although the mechanism is not clear, single nucleotide polymorphisms of OCTN1 and OCTN2 genes are also associated with increased incidences of rheumatoid arthritis and asthma (137). In a Korean population of 193 patients with Crohn disease and 281 healthy control individuals, a common promoter haplotype of OCTN2 was found, which regulates the transcriptional rate of OCTN2 and influences the clinical course of Crohn disease (101).

OCTN2 expression has also been demonstrated in the A-cells but not the B-cells of mouse pancreas by in situ hybridization and immunohistochemistry with the anti-OCTN2 antibody (53). OCTN2-mediated carnitine uptake has been demonstrated in a human proximal tubule cell line (Caki-1) with an apical expression pattern of OCTN2 (42). OCTN2 expression has also been demonstrated in the human nasal epithelium mainly on the apical side of the nasal cells (125).

OCTN2 belongs to a subfamily of organic cation/carnitine transporters. Other members of this family that transport carnitine with different affinities are OCTN1, which has been characterized in humans and mice (140; 139) and subsequently OCTN3 in humans (64; 163). OCTN1 has a low affinity for carnitine, and OCTN3 is an intermediate-affinity carnitine transporter. We have demonstrated OCTN3 to be localized to human peroxisomal membranes (64) and OCTN1 to be expressed in human mitochondria (70) where they likely play pivotal roles in the intracellular transport of carnitine and acylcarnitines between peroxisomes and mitochondria. Thus, OCTN1 and OCTN3 would be well positioned to play crucial roles in maintaining intracellular carnitine homeostasis along with plasmalemmal OCTN2. In murine brain, OCTN1, -2, and -3 are expressed in many regions throughout the central nervous system with a pattern suggestive of roles in modulating cerebral bioenergetics and in acetylcholine generation for neurotransmission in olfactory, satiety, limbic, memory, motor, and sensory functions. They are also strongly expressed in the choroid plexus cells forming the blood-brain barrier. This distribution may play a role in the pattern of neurologic injury that occurs in hOCTN2 deficiency during catabolic episodes of hypoglycemic hypoketotic encephalopathy, and which may manifest with cognitive impairment, hypotonia, and seizures (65). OCTN2 has also been shown to be involved in the transport of acetyl-L-carnitine from the blood to the retina across the inner blood-retinal barrier (136). Functional expression of OCTN2 has also been demonstrated in the human brain capillary endothelial cell line hCMEC/D3, which is a human blood-brain barrier model (98). By taking advantage of the specific expression of OCNT2 on both brain capillary endothelial cells and glioma cells, L-carnitine-mediated cellular recognition and internalization via OCTN2 has been shown to facilitate significantly the transcytosis of nanoparticles across the blood-brain barrier and the uptake of nanoparticles in glioma cells, resulting in improved antiglioma efficacy (60).

In adult murine heart, all three transporters showed strong expression in cardiomyocytes, lamina fibrosa of cardiac valves, great arteries and intermuscular arterioles, and a striking differential expression in the vagal innervated sinoatrial and atrioventricular nodes (71). The expression pattern suggested potential roles for the transporters in modulating myocardial bioenergetics, valvular function, and acetylcholine generation for parasympathetic vagal innervation of the cardiac conduction system. OCTN1 and OCTN2 expression also occurs in human corneal and conjunctival epithelial cells (41). Furthermore, all three transporters are expressed in normal control human sperm (163; 56). Spermatozoan maturation, motility, and fertility are, in part, dependent on the progressive increase in epididymal and spermatozoal carnitine, critical for mitochondrial fatty acid oxidation, as the sperm pass from the caput to the cauda of the epididymis. The juvenile visceral steatosis mouse, which arises from a point mutation in the mOCTN2 gene, presents with infertility due to obstructive azoospermia, which is highly responsive to L-carnitine therapy with restoration of fertility in male mice (164). Identification of individuals with defective sperm carnitine transport may provide potentially treatable etiologies of male infertility, responsive to L-carnitine supplementation.

Given the roles of human OCTN2 in the transport of acylcarnitines and other cationic compounds, in combination with the fact that it is inhibited by a wide variety of xenobiotics and shows wide tissue distribution, these data suggest that human OCTN2 is of considerable pharmacological and toxicological importance. Because certain of these drugs and carnitine seem to compete for the same substrate-binding site on OCTN2, it is likely that mutations causing the high-affinity carnitine transporter defect are also associated with the loss of ability to transport these drugs, leading to increased renal clearance and, hence, to decreased systemic half-life and therapeutic efficacy. In human OCTN2-transfected HEK293 cells, L-[3H]-carnitine uptake has been shown to be significantly inhibited by a large number of xenobiotics, including lipophilic organic cations (quinidine, verapamil, and emetine) and zwitterionic compounds such as beta-lactam antibiotics (cephaloridine) that are in extensive clinical therapeutic use. The anionic compounds valproic acid and probenecid were moderate inhibitors (96; 39; 95). Certain compounds (eg, tetraethylammonium, quinidine, verapamil, cephaloridine, valproic acid) were shown to be directly transported by human OCTN2 in transfected cells. Both pivaloylcarnitine and valproylcarnitine have been shown to inhibit L-carnitine reabsorption in the perfused rat kidney, and their concentration-dependent inhibition was also observed for hOCTN2 (97). Although loss-of-function mutations in OCTN2 may be rare, common variants of OCTN2 found in healthy populations may contribute to variation in the disposition of carnitine and some clinically used drugs (151).

Etoposide has been associated with significant urinary loss of carnitine in mice and in patients with cancer and has been shown to be a competitive inhibitor of hOCTN2-mediated carnitine uptake, which may contribute to its treatment-related toxicities (49). In an oxazaphosphorines-induced acute cardiomyopathic rat model, study data suggested that oxazaphosphorine therapy decreased myocardial carnitine content following the inhibition of OCTN2 mRNA and protein expression in cardiac tissues and increased urinary loss of carnitine secondary to the inhibition of OCTN2 mRNA and protein expression in proximal renal tubules (120). Carnitine supplementation was shown to attenuate cyclophosphamide-induced inhibition of OCTN2 mRNA and protein expression in heart and renal tissues (120).

In keratinocytes retrotransduced with HPV16E6 and E7, both OCTN2 mRNA and protein levels were reduced (122). A similar downregulation of OCTN2 mRNA level was observed in a naturally HPV16-infected cancer cell line, CaSki, harboring several copies of HPV16 whole genome. The treatment of keratinocytes retrotransduced with HPV16 E6 and E7 with 5-aza-cytidine rescued the OCTN2 expression, indicating that the mechanism of downregulation was linked to DNA methylation (122). Low levels of mRNA expression of OCTN2 were also found in several nonvirus-related epithelial cancer cell lines. The treatment of these cell lines with 5-aza-cytidine was again shown to rescue the expression of OCTN2. These data demonstrated for the first time that the OCTN2 transporter is generally downregulated in virus and nonvirus-mediated epithelial cancers, probably via methylation of its promoter region (122).

Epidemiology

Since the demonstration of a cellular defect in the uptake of carnitine in the cultured skin fibroblasts of children with the primary carnitine transporter defect in 1988 (32; 146), a total of 26 fibroblast-proven cases have been described in the literature (142; 40; 133; 10; 17; 18; 104). Three other highly probable cases can be added between 1980 and 1988, during the time period when fibroblast uptake studies were not available (16; 148). Because the clinical recognition of this disorder is only now slowly increasing (primarily in tertiary care referral metabolic centers), and because there has been no widespread application of screening programs for the primary carnitine transporter defect, and as only a few centers can measure carnitine uptake in cultured skin fibroblasts at the present time, the precise determination of the incidence and prevalence of the disease is difficult. However, a number of centers now screen patients with myopathy, cardiomyopathy, and hypoglycemic encephalopathy with plasma carnitine determinations that should provide preliminary evidence for this defect. Therefore, given the relatively small numbers of patients accumulated over the past 6 years, one might speculate that the incidence of this disorder may be in keeping with the group of rare autosomal recessive diseases, the incidence of which is usually estimated to be between 1 in 50,000 people and 1 in 100,000 people. The ethnic backgrounds of the 18 families (20 patients) described by Stanley and colleagues included European, African descent, North African Arab, Asian Indian, Mexican, and Chinese, suggesting that the defect is widely distributed genetically (133). Consanguinity was present in five of the families studied by Stanley and colleagues, and in one of the families described by Garavaglia and colleagues (40). In nine of these families, there was a history of sibling deaths with illnesses compatible with the carnitine transporter defect. Plasma total carnitine concentrations were low in all 12 mothers and in 11 of 12 fathers who were tested. All of the studied parents of the affected children showed intermediate maximal rates for carnitine uptake, ranging from 13% to 44% of control maximal velocity values, supporting the autosomal recessive pattern of inheritance.

In a population-screening study, Koizumi and associates demonstrated that the overall prevalence of heterozygotes is 1% in the Akita prefecture in Japan, giving an estimated incidence of primary carnitine transporter deficiency as 1 in 40,000 births in this population (58). Multiple logistic analyses showed that mutations were a larger independent risk factor than aging and hypertension for echocardiographic abnormalities in carriers. Among carriers greater than 20 years of age, without other confounding diagnoses such as hypertension, about 33% were found to exceed the 95% upper limit on echocardiographic parameters. One carrier had clinically apparent cardiac hypertrophy and cardiomegaly. It was suggested that a low primary carnitine phenotype and aging or environmental factors may act synergistically to lower the efficiency of cardiac ATP production. This may suggest that aging carriers could be at risk for cardiac decompensation, particularly in the presence of multiple risk factors, including medications that compete with carnitine at the high affinity carnitine transporter (OCTN2) site.

The incidence of primary carnitine transporter deficiency has been estimated to be 1 in 37,000 newborns in Australia (161) and 1 in 142,000 in the United States (145), with the highest incidence of 1 in 300 reported in the Faroe Islands, an archipelago that has remained geographically isolated (106). It has been suggested that newborn screening may miss cases of the high affinity carnitine transporter defect, as carnitine is transferred from the mother to the child via the placenta and concentrations of carnitine may be normal in an infant if the sample is collected shortly after birth (78). The carnitine concentrations may then decline over time. It has, therefore, been suggested that there be two screenings, the first within approximately 48 hours after birth and the second at 7 to 21 days of life (78).

In a newborn screening program in Quanzhou, China, 364,545 newborns were evaluated by tandem mass spectrometry (MS/MS), and 36 newborns and five mothers were diagnosed with OCTN2 deficiency, which resulted in an incidence of OCTN2 deficiency in children of 1 in 10,126 in the Quanzhou area (75). Five novel variants were found by next generation sequencing. In a newborn screening program in the Zhejiang province of China in which 3.4 million newborns were screened by tandem mass spectrometry followed by molecular genetic analysis in suspected positive patients, 113 newborns were diagnosed with OCTN2 deficiency in addition to affected 63 mothers. The incidence of OCTN2 deficiency in newborns and mothers in Zhejiang was 1 in 30,182 and 1 in 54,137, respectively (76). Thirty-seven distinct variants were identified in SLC22A5, of which 10 were novel. In a newborn screening study of dried blood spots analyzed through tandem mass spectrometry (MS/MS) in Shanghai, China, between January 2003 and December 2021, 40 of 1,247,274 newborns were diagnosed with OCTN2 deficiency, with an incidence of approximately 1:32,200 (15). Twenty-three different variants were identified in the SLC22A5 gene, including eight novel variants, and almost all screened OCTN2-deficient patients were asymptomatic when they were treated with high-dose oral L-carnitine supplementation immediately after diagnosis at a median age of 30 days.

In a study, the frequency of mutations in the SLC22A5 gene encoding OCTN2 was determined in 324 individuals with cardiomyopathy and compared to that in a normal population (02). The frequency of variants affecting carnitine transport was 0.61% (2 out of 324) in patients with cardiomyopathy, which was not significantly different from the frequency of 1.11% (3/324) in the general population. This suggests that heterozygosity for OCTN2 deficiency is not more frequent in patients with unselected types of cardiomyopathy.

Surprisingly, 11 asymptomatic affected mothers of affected children have been identified (74) suggesting that the expression of the disease may be influenced by multifactorial or epigenetic factors. Patients exhibiting symptoms as adults tend to have at least one missense mutation resulting in residual activity (78).

Prevention

Primary carnitine transporter deficiency is an autosomal recessive disorder, and, thus, genetic counseling to the family is critical. Carnitine uptake studies in cultured skin fibroblasts are important for diagnosis and screening of siblings of affected patients at risk, as well as for establishing the carrier heterozygote state in siblings and parents. Early identification of presymptomatic homozygote siblings with institution of carnitine supplementation may decrease immediate and long-term morbidity, as well as mortality related to hypoglycemic coma, cardiomyopathy, myopathy, and sudden infant death. At present, no prenatal diagnostic techniques are available.

Once the diagnosis is confirmed, the institution of high-dose oral carnitine supplementation (100 mg/kg per day in four divided daily doses) will largely reverse the clinical manifestations of the disorder. The avoidance of known fatty acid oxidation stressors such as prolonged fasting, cold exposure with shivering thermogenesis, prolonged exercise, and infection is important in the reduction of acute catabolic crises. Furthermore, in the event of vomiting and fasting with infection, it is important to provide intravenous fluids with an adequate carbohydrate supply (eg, 10% dextrose) to avert a fatty acid oxidation crisis.

Differential diagnosis

Though the above defects bear the closest clinical and biochemical similarity in presentation to the primary carnitine transporter defect, other possibilities to be considered within the differential diagnosis of infantile or early childhood-onset hypertrophic cardiomyopathies with hypotonia would include the following:

(1) The mitochondrial encephalomyopathies should be considered, particularly cytochrome oxidase deficiency or mtDNA mutations that may also have a secondary carnitine deficiency. These disorders could be distinguished by other associated clinical features such as sensorineural hearing loss, myoclonus, stroke-like episodes and ataxia, elevated serum and cerebrospinal fluid lactates, the possible presence of ragged-red fibers and subsarcolemmal mitochondrial accumulations on muscle biopsy, and a possible Leigh disease-like picture on CT scan of the head. In the case of mtDNA mutations, there would also be in an inheritance pattern consistent with a maternal mitochondrial pattern of transmission.

(2) Barth syndrome (or benign X-linked cardiomyopathy with cyclic neutropenia and low plasma carnitine) should be considered (05; 52). This disorder can be distinguished by its X-linked pattern of inheritance and cyclic neutropenia, as well as by the presence of 3-methyl-glutaconic aciduria on organic acid screening. Furthermore, if the children are supported during the first years of life, there is a gradual spontaneous improvement or resolution of the cardiomyopathy by puberty.

(3) Pompe disease, or acid maltase deficiency (type II glycogen storage disease), should be considered. This disorder also presents in the first few months of life, but can be distinguished by marked cognitive delay, a large tongue, marked hypotonia related to the anterior horn cell disease, glycogen storage myopathy, and death within the first 1 to 2 years of life.

Confusing conditions

The primary carnitine transporter defect must be distinguished from other genetic disorders of fatty acid oxidation wherein carnitine deficiency is secondary. Several features may be helpful, and are listed below (131):

(1) Plasma and tissue carnitine deficiency is more severe (less than 5% of control values) in the primary transporter defect than in the secondary carnitine deficiency disorders due to intramitochondrial beta-oxidation defects, such as medium-chain acyl-CoA dehydrogenase deficiency or long-chain acyl-CoA dehydrogenase deficiency, in which the carnitine concentrations are usually approximately 25% to 50% of normal.

(2) The ratio of esterified carnitine to total plasma carnitine is usually normal (eg, 25% to 30%) in the primary transporter defect, whereas there is an increase in the esterified carnitine fraction (greater than 40%) in the secondary carnitine deficiency disorders due to increased conjugation of the acyl-CoA esters that accumulate proximal to the metabolic block, with free carnitine.

(3) The abnormal increases in urinary dicarboxylic acids that are seen during fasting with intramitochondrial beta-oxidation defects are not seen in the primary transporter defect. This finding may reflect the high level of block in the pathway of fatty acid oxidation (133), as an abnormal dicarboxylic aciduria is also not seen with hepatic form of carnitine palmitoyltransferase I deficiency (08).

(4) In contrast to most of the other known fatty acid oxidation defects, the most common clinical manifestation in the transporter defect is cardiomyopathy, rather than hypoketotic and hypoglycemic encephalopathy. Of the other fatty acid oxidation disorders, the defects that may also present with cardiomyopathy include deficiencies of carnitine and acylcarnitine translocase, long-chain and very long-chain acyl-CoA dehydrogenase, short-chain L-3-hydroxyacyl-CoA dehydrogenase, long-chain L-3-hydroxyacyl-CoA dehydrogenase or trifunctional protein, and the severe infantile form of carnitine palmitoyltransferase II.

Diagnostic workup

	• Initial investigations should include serum carnitine concentrations, which should be very low and should demonstrate an inappropriate increased fractional excretion of urinary free carnitine.
	• Molecular genetic analysis should confirm biallelic pathogenic mutations in SLC22A5, including the 5’ untranslated region (UTR).
	• If biallelic mutations in SLC22A5 are not found and there is strong clinical and biochemical evidence for OCTN2 deficiency, diagnosis can be confirmed by demonstrating markedly impaired carnitine uptake in cultured skin fibroblasts.

A physician should be suspicious of a defect in the carnitine transporter if there is clinical history of early normal development, followed by the onset of a slowly progressive hypertrophic or dilatative cardiomyopathy and myopathy beginning in the first years of life, with acute episodes of hypoketotic, hypoglycemic coma induced by fasting. During the initial presentation of hypoglycemic coma, it is important to concomitantly measure the urine ketones and determine the serum free fatty acid to ketone ratio, in order to establish the defect in ketogenesis. (The normal serum free fatty acid to ketone ratio is 1 to 1. If this ratio exceeds 2.5 to 1, this implies a defect in ketogenesis.) Cardiac echogram would demonstrate a hypertrophic or dilative cardiomyopathy. The best initial screening investigation would be the total and free plasma carnitine concentrations; these would be low, generally less than 5% of control values, with a normal esterified carnitine fraction (25% to 30%). This is in contrast to the secondary carnitine deficiency disorders, where the plasma carnitine concentrations are usually 25% to 50% of control values with an increased esterified carnitine fraction of greater than 40%. An early morning urinary organic acid screen, collected particularly after normal overnight fasting, should show no evidence of an abnormal dicarboxylic aciduria. An abnormal dicarboxylic aciduria is characteristic of the intramitochondrial beta-oxidation disorders. There should also be evidence of inappropriate carnitine excretion in the urine despite the low plasma carnitine concentrations, as this would indicate an impairment in the renal tubular reabsorption of carnitine; the normal renal threshold is 40 µmol/L. If a muscle or endocardial biopsy is obtained, there should be evidence of lipid storage, particularly in the case of muscle in the type I fibers, and the tissue carnitine concentrations should be low (eg, less than 5% of control values).

Following the initial screening investigations, a defect in the plasmalemmal uptake of carnitine can then be directly confirmed with carnitine uptake studies performed in vitro in cultured skin fibroblasts obtained from a skin biopsy of the patient. These studies should indicate negligible uptake of carnitine throughout the physiological range (0.1 to 50 µmol/L) of carnitine substrate concentrations (eg, 2% of control uptake at 5 µmol/L of carnitine), thereby precluding the calculation of Michaelis constant and maximal velocity values.

In a study of 140 subjects with carnitine deficiency who underwent carnitine uptake studies in fibroblasts and were shown to have less than 20% of normal carnitine transport and who then underwent sequencing of the 10 exons and flanking regions of the SLC22A5 gene, which was performed in 95 out of 140 subjects identified, causative variants were identified in 84% of alleles (37). Thus, mutations in the coding region of the SLC22A5 gene could not be identified in about 16% of the alleles causing primary carnitine deficiency. It was, therefore, concluded that functional studies in cultures fibroblasts remained the best strategy to confirm or exclude a diagnosis of primary carnitine deficiency (37).

The obligate heterozygote state of the parents can be best confirmed by carnitine uptake studies in vitro in the cultured skin fibroblasts of the parents, as the plasma carnitine concentrations are variable in the heterozygotes and may be slightly reduced or normal (142). Therefore, though plasma carnitine concentrations in the parents are a useful and simple screen, in vitro cultured fibroblast uptake studies provide more definitive evidence of heterozygosity and demonstrate normal Michaelis constant values but reduced maximal velocity values of 13% to 44% of control (142; 133). It is also important to screen the siblings in the family, initially with plasma carnitine determinations and then with cultured skin fibroblast carnitine uptake studies to determine whether they are homozygote or heterozygote for the transporter defect because this will affect their therapeutic management.

Retrospective quantitation of acylcarnitine profiles from stored filter cards of dried neonatal blood spots using electrospray ionization-tandem mass spectrometry can reliably diagnose OCTN2 deficiency (35). However, appropriate correction for sample decay during storage must be made because free carnitine increases during storage but can be reliably quantitated under standardized derivatization conditions. Furthermore, examining acylcarnitine profiles can supplement free carnitine concentrations as a discriminating marker.

Newborn screening programs for primary carnitine deficiency can identify affected patients at risk for this condition prior to irreversible damage (78).

Management

	• Mainstay of therapy is the institution of high-dose oral L-carnitine supplementation in a dose of at least 100 mg/kg/day divided in four daily doses, which is a lifelong requirement.
	• Higher doses may be required during acute stressors such as infectious illnesses and may need to be given intravenously in the event of vomiting.
	• Compliance with daily carnitine supplementation is essential to prevent morbidity and mortality.

On diagnosis of the carnitine transporter defect, the mainstay of therapy is the institution of high-dose oral carnitine supplementation in a dose of 100 mg/kg per day in four divided daily doses; this therapy will be a lifelong requirement, and should result in a dramatic improvement in symptoms within a few weeks, and a reduction of heart size toward normal within a few months of therapy, as evidenced by cardiac echogram (142; 133). On the return of adequate cardiac function, the traditional anti-failure medications can usually be slowly weaned. The requirement for compliance with high-dose oral L-carnitine supplementation for life is critical as highlighted by a case in which L-carnitine therapy, which had reversed the childhood-onset cardiomyopathy, was discontinued by an affected individual at 13 years of age (21). At the age of 29, he presented with heart failure and atrial fibrillation and required admission to an intensive care unit where he was treated with intravenous carnitine supplementation along with other anti-failure medications, which led to improved cardiac function within 2 weeks. On resumption of his continuous oral carnitine supplements, his left ventricular ejection fraction normalized, and the following 8 years on therapy were uneventful. The authors emphasize the importance of continuous metabolic monitoring, the potential reversibility of cardiomyopathy in an adult patient, and the potential risks in the period of transition from pediatric to adult care.

Once the child is receiving carnitine therapy, the defect in long-chain fatty acid oxidation is generally corrected and ketogenesis should be intact, given the normalized liver concentrations of carnitine. However, it may still be prudent, as in the other long-chain fatty acid oxidation disorders, to maintain a high-carbohydrate (60% to 65% of calories) with low-fat (25% to 30% of calories) diet (modified according to age of child and in consultation with your nutrition consultants) and to consider supplementation of the diet, as needed, with medium-chain triglyceride oil that the patient can metabolize. This would provide the remaining tissues with readily utilizable energy substrates because these tissues (eg, skeletal muscle, heart, kidney) express the transporter defect and tissue carnitine deficiency is only partially corrected by carnitine supplementation. The child should have three regular meals per day with three snacks between meals, including most importantly a bedtime snack to reduce overnight fasting.

It is also critically important to avoid known precipitating factors for fatty acid oxidation stress and catabolic crises, such as fasting, cold exposure with shivering thermogenesis, prolonged exercise (eg, greater than 30 minutes continuously), stress, and infection. In the event of infection with vomiting and fasting, it is important to institute intravenous fluids (including 10% dextrose) to provide the patient with a continuous available carbohydrate source in order to avert a fatty acid oxidation crisis. It may also be clinically judicious to avoid certain medications (such as valproic acid) that may exacerbate the plasma (90; 72) as well as tissue carnitine deficiency (126). Valproic acid therapy has been associated with microvesicular lipid accumulation and mitochondrial abnormalities in skeletal muscle (88). Multiple mechanisms for valproic acid-associated secondary carnitine deficiency may be involved, including a dose-related inhibition of plasmalemmal carnitine uptake as demonstrated in vitro in cultured control human skin fibroblasts by Tein and colleagues (143). Furthermore, an increased risk for valproic acid-associated impairment of carnitine uptake has been demonstrated in the fibroblasts of heterozygotes for the plasmalemmal transporter defect (144).

Low concentrations of plasma carnitine have also been seen in infants on long-term pivampicillin therapy. These patients excrete large amounts of pivaloylcarnitine in the urine, and the plasma carnitine concentrations gradually return to normal after cessation of the pivampicillin therapy (48). In a study of seven healthy adult volunteers given 7 to 8 weeks of pivmecillinam, the median total muscle carnitine concentration decreased (46% of pretreatment concentrations), and there was a transient reduction in the left ventricular mass (01).

Similarly, treatment with doxorubicin in a rat model resulted in a significant and dose-dependent decrease in mRNA expression of OCTN2 and H-FABP (heart fatty acid binding protein), total carnitine and ATP in cardiac tissues, and a significant increase in cardiac enzymes (121), which must, therefore, be monitored in children receiving this as chemotherapy. Interestingly, carnitine supplementation restored doxorubicin-induced inhibition of gene expression of H-FABP and OCTN2 to normal and decreased myocardial carnitine and ATP to control concentrations.

In individuals who are heterozygotes for the transporter defect and who already have modestly reduced plasma carnitine concentrations, it will be important (in order not to contribute to further plasma and potentially tissue carnitine deficiency) to ensure adequate dietary carnitine, as it accounts for approximately 75% of total body carnitine stores in omnivores (108). Dietary deficiencies with reduced plasma carnitine concentrations may occur in strict vegetarians, in infants receiving unsupplemented soy-protein formulas, and in individuals receiving long-term total parenteral nutrition without carnitine supplementation (108). Preterm infants are particularly susceptible to carnitine deficiency if the dietary sources are inadequate, such as in carnitine-free total parenteral nutrition (103), because of their markedly reduced tissue stores (123) and their reduced biosynthetic capacities for endogenous carnitine synthesis (25% of adult hepatic gamma-butyrobetaine hydroxylase activity at birth) (100). Carnitine deprivation in neonatal piglets resulted in low carnitine status with lower rates of hepatic palmitate oxidation and lipid deposition in liver and skeletal muscle as well as a higher incidence of muscle weakness and cardiac failure than in carnitine-supplemented piglets (102).

On diagnosis of a patient with the primary carnitine transporter defect, it is important to provide genetic counseling to the family regarding the autosomal recessive pattern of inheritance and the 25% recurrence risk for future pregnancies. In addition, it is important to screen the presymptomatic siblings and institute carnitine supplementation in the confirmed homozygotes in order to decrease immediate and long-term morbidity and mortality and avert sudden infant death. This is especially important given the frequent history of previously unexplained sibling deaths in these families (142; 133).

Of interest, in a study using wild-type mice, PPAR alpha was shown to mediate transcriptional upregulation of Octn2 and Octn3 in tissues as well as the enzymes involved in hepatic carnitine biosynthesis (57).

Outcomes

On review of the 20 patients described by Stanley and colleagues, a patient has died from complications of intestinal adhesions following placement of a feeding gastrostomy tube (133). Eighteen of the other 19 patients continue to be healthy on carnitine therapy after periods of 1 year to 10 years. Only a single patient is moderately impaired with weakness and cardiomyopathy at the age of 20, but this is thought to be due to poor compliance with the carnitine therapy. The long-term prognosis into adulthood is unknown at the present time, as the first child recognized to have this disorder was started on carnitine therapy in 1981. As these individuals continue to be followed, this information should become available in the future.

Special considerations

Pregnancy

In 1996, Christodoulou and colleagues reported the first attempt at prenatal diagnosis of the carnitine transporter defect in a fetus at high risk for the disorder (18). The proband's sister, born of unrelated Chinese parents, had been previously diagnosed to have the carnitine transporter defect in cultured skin fibroblasts. Analysis of cultured chorionic villus, obtained at 11 weeks' gestation and performed at 17 weeks' gestation (after prolonged culture because of slow growth of cells) predicted that the fetus was not affected, but might be heterozygous for the carnitine transporter defect. However, chromosome 15 satellite DNA markers showed no paternal contribution, suggesting that the chorionic villus cells assayed were of predominantly maternal origin. Subsequent assay of cultured amniocytes predicted that the fetus would be affected, given a carnitine uptake value that was 0.5% of control values. This was confirmed in the newborn period by the demonstration of a dramatic decline of plasma carnitine in the first 2 weeks of life (proband total 5 µmol/L, free 3 µmol/L; controls total 35 to 65 µmol/L, free 30 to 60 µmol/L). It was concluded that prenatal diagnosis of the carnitine transporter defect is possible, but where results depend on extended culture of chorionic villus, molecular studies should be performed to confirm genetic contributions from both parents.

The plasma carnitine concentrations in normal infants are lower in the first week than later in life (124), and this may be related to decreased maternal concentrations (84). Of note, as females heterozygous for the carnitine transporter defect usually have low plasma carnitine concentrations (132), and because individuals who are homozygous for the carnitine transporter defect may initially present with hypoglycemia in infancy (133), Christodoulou and colleagues elected to treat the mother of the above-described proband with carnitine supplementation during the second half of her pregnancy (18). The subsequent cord blood plasma carnitine in the proband was just above the upper limit of normal for age. It was postulated that prenatal carnitine therapy in this patient minimized the likelihood of the development of clinical features in the neonatal period. It was, therefore, suggested that prenatal carnitine therapy be considered for those pregnancies at high risk of having the carnitine transporter defect. This may be particularly important in heterozygote mothers who are also vegetarian, in whom low serum carnitine concentrations may be further reduced by a limited exogenous intake of carnitine.

In murine studies, it has been demonstrated that there is a dynamic upregulation of OCTN2 in pregnant and lactating mammary glands, which likely provides the suckling infant with adequate carnitine for the rapid postnatal upregulation of fatty acid oxidation and ketogenesis; this is critical for cerebral energy metabolism during fasting hypoglycemia (66). This could be another limitation in carrier mothers. L-carnitine deficiency due to a defect in OCTN2 in a mouse model has been shown to lead to reduced placental concentrations of L-carnitine to less than 10% of normal and to embryonic lethality (127).

Normative data for newborn whole blood carnitine concentrations determined by electrospray tandem mass spectrometry have been established in a study of 24,644 newborns at 1.85 ± 0.95 days of age (14). The entire cohort was stratified according to total carnitine values into a middle total carnitine group representing 90% of the population and lower and upper total carnitine groups representing 5% of the population, respectively. Normative data were derived from the middle total carnitine group of full-term infants (N=19,595). Total carnitine was 72.42 ± 20.75 µmol/L, free carnitine was 44.94 ± 14.99 µmol/L and acylcarnitine was 27.48 ± 8.05 µmol/L. Interestingly, in controlled analyses, prematurity was not associated with total carnitine concentrations, whereas low birth weight (< 2500 g) and male sex were significantly associated with higher total carnitine concentrations. The association of low birth weight with the higher total carnitine values may be related to decreased tissue carnitine uptake. The sex effect may be related to hormonal influences on carnitine metabolism.

A newborn screening study of both newborns and mothers conducted in Taipei from January 2001 to July 2009 demonstrated an incidence of the OCTN2 defect of 1 in 67,000 in newborns and a prevalence of 1 in 33,000 in mothers (73). The six mothers identified with the OCTN2 defect, one with cardiomyopathy, were placed on carnitine therapy at the time of diagnosis. The one with cardiomyopathy showed improvement in cardiac function after treatment, emphasizing the importance of treatment.

It is important to note that there can be false positives and false negatives on newborn screening, which relate to maternal plasma carnitine concentrations. Newborn screening has identified maternal OCTN2 deficiency, often in asymptomatic women. In a 2019 Newborn Screening Ontario study of a kindred with OCTN2 deficiency due to a 5’UTR mutation (SLC22A5; NM_003060:c.-149G> A), three newborns were positive on newborn screening and were found to be unaffected themselves; however, their mothers (sisters) were affected. There were also two affected children born to an affected male and his heterozygous wife who were false negatives on newborn screening but had increased fractional excretion of free urinary carnitine (154). Plasma carnitine was decreased, and fractional excretion of free urinary carnitine was increased in all affected individuals. The two false negatives were homozygous siblings born to a heterozygous mother who had normal plasma free carnitine concentrations. The elevated renal fractional excretion of carnitine appeared to be a more sensitive and specific test in these infants. Treatment of homozygous mothers with 2 to 6 gm of carnitine daily throughout pregnancy was sufficient to avoid false positive newborn screening of their five subsequent infants.

In a qualitative study on the perspectives of 12 mothers who had been diagnosed with primary carnitine deficiency through newborn screening of their child, the psychological burden was considered as limited, and most believed that the benefits of knowing about primary carnitine deficiency outweighed the disadvantages of feelings of anxiety or uncertainty (152).

Anesthesia

Preoperative assessment of the patient should include evaluation of the neurologic status as well as the cardiopulmonary systems and biochemical status of the patient, eg, serum glucose and carnitine. Serum carnitine concentrations should be optimized preoperatively. The patient should receive the usual daily dose of carnitine on the morning of operation and should continue to receive the regular carnitine dosages by intravenous route if necessary. During any operative procedure, the most critical aspect will be the maintenance of stable blood glucose concentrations before, during, and after the operation in order to avert precipitating a fatty acid oxidation crisis. If the patient is lethargic or drowsy, serum glucose should be measured immediately. It may be worthwhile to initiate the intravenous fluids (including 10% dextrose) on the evening before surgery to eliminate any period of fasting prior to the operative procedure. For induction, it may be worthwhile to use a nondepolarizing muscle relaxant and avoid succinylcholine, because of the unknown nature of the interaction of succinylcholine and of the myopathy (117). Intraoperatively, cold exposure should be avoided and sterile conditions maintained to decrease the risk of infection. Furthermore, it will be important to measure serum glucose concentrations at regular intervals throughout the procedure to ensure adequate metabolic substrate availability and prevent potentially catastrophic hypoglycemic episodes.

Postoperatively, it may be prudent to give an antiemetic to prevent vomiting. Following the procedure, particularly if the operation was initiated under emergency conditions where the catabolic stress with catecholamine release may stimulate peripheral lipolysis, it may be wise to continue the intravenous glucose therapy for as long as necessary until the patient is well over the acute phase of the procedure and is able to tolerate oral fluids and feedings well. The patient should be carefully followed for the possibility of postoperative infection; if infection occurs, it should be quickly and aggressively treated.

If a patient is referred for emergency operation while in a metabolic crisis, the patient must be adequately rehydrated with careful maintenance of serum glucose. In addition, serum glucose, electrolytes, acid-base status, carnitine, creatine kinase, ammonia, liver function, and renal function tests must be evaluated and corrected if necessary.