Clinical manifestations

Presentation and course

In the classical form of maple syrup urine disease, infants appear well at birth. Symptoms begin after 3 to 5 days and progress rapidly to death within 2 to 4 weeks if branched-chain amino acid restriction is not implemented. Early manifestations include feeding difficulties, irregular respirations, progressive loss of the Moro reflex, and apnea. Severe hypoglycemia and mild hyperammonemia may occur, as may pancreatitis. These infants may develop convulsions, opisthotonos, pedaling movements of the legs, and generalized muscular rigidity with or without intermittent flaccidity. Death may occur following the development of cerebral edema. At autopsy, cortical atrophy is seen on CT or MRI scan. The myelin is usually hypodense, which is thought to be due to failure of myelinization. Localization of edema to the cerebellum and posterior capsule may be pathognomonic (14).

If treatment is initiated within the first week of life, cerebral edema is reversible and the prognosis is good (29). Because respiratory arrest may occur in the neonatal period, care in a tertiary neonatal intensive care unit is recommended. The metabolic derangement and its reversibility in the brain can now be followed using diffusion-weighted imaging and magnetic resonance spectroscopy (41).

Although approximately 80% of patients with maple syrup urine disease present with the classical form and very low residual enzymatic activity (2% or less), most other patients present with intermittent disease and have higher enzymatic activity (5% or more). These patients have a normal psychomotor development but have acute metabolic attacks during catabolic stresses in infancy or childhood (characterized by elevated BCAA, ketonuria, ataxia, and lethargy). Rare patients present with the intermediate form; they have developmental delays but decompensate after the neonatal period. Note that the ex vivo enzymatic activities can be quite different from in vivo measurements of 13C-leucine oxidation (77), illustrating their limited predictive values.

Less severe variants of maple syrup urine disease have symptoms occurring later in childhood following high protein intake or when a catabolic state is produced by fever, surgery, or prolonged fasting. Patients may be averse to protein on diet history and present with developmental and growth delays. Clinical manifestations may occur in early or late childhood and are characterized by intermittent episodes of ataxia, ketoacidosis, and failure to thrive. Intellectual disability and psychomotor delays are sequelae if left undiagnosed and untreated. "Intermittent," "intermediate," and "thiamine-responsive" forms are clinical classifications that are now recognized to overlap with one another and probably reflect the large number of private mutations identified in the genes for branched-chain alpha-ketoacid dehydrogenase and their various effects on dysfunction of this multienzyme complex. The degree of impaired branched-chain alpha-ketoacid dehydrogenase and the amount of environmental stress on the patient determine the rapidity, age of onset, and intensity of symptoms.

Patients from 5 families have been described with the thiamine responsive form of maple syrup urine disease (19). The pre-requisites for thiamine responsiveness seem to be an intact E1 subunit, at least 1 allele encoding a full-length E2 subunit (albeit containing a missense mutation), and significant intact cell residual activity. Dietary restriction of leucine is still needed in most thiamine-responsive forms.

In type 3 maple syrup urine disease, mutations are found in the E3 subunit. The E3 subunit (dihydrolipoamide dehydrogenase) is shared with the pyruvate dehydrogenase complex and alpha-ketoglutarate dehydrogenase complex. BCKAs and alpha-ketoglutarate are elevated in the urine along with elevated plasma concentrations of lactate, pyruvate, BCAA and alanine. About 20 patients have been reported, and the phenotype is characterized by a progressive encephalopathy with metabolic decompensations beginning in infancy. Renal tubulopathy and cardiomyopathy can occur.

Diagnosis for all forms is best made before irreversible clinical damage is produced through population-based newborn screening (61). In the symptomatic child, the smell of burnt sugar or maple syrup may be manifest in urine, sweat, or ear wax. Cerumen is an excellent concentrator of the lipid-soluble, sweet-smelling derivative of isoleucine (ie, sotolone), and a smell of the otoscope may make a diagnosis. Bedside screening of urine using the dinitrophenylhydrazine (DNPH) reaction is useful though this is not now routine practice. Diagnosis is made by quantitative and qualitative analysis of plasma and urine L-leucine, L-isoleucine, alloisoleucine, and L-valine. Low plasma glutamine can be noted. Gas chromatography of urine will reveal the branched-chain alpha-ketoacids (alpha-ketoisocaproate, alpha-keto-beta methylvaleric, and alpha-ketoisovaleric acids) as well as their hydroxy derivatives when patients are in poor dietary control. Molecular testing or enzymatic analysis of cultured cells can confirm the diagnosis and enable more rational therapeutic decisions, prognosis, and genotyping of other family members (23; 24). Whole-body leucine oxidation of 13C labeled leucine is rarely used except for research protocols (27).

Prognosis and complications

Outcome is not easily predictable (47; 39; 102; 59; 40; 84; 85; 91). However, if the newborn is diagnosed before apnea or secondary central nervous system damage has occurred and is maintained on an appropriate diet of calories, protein, and carbohydrate, prognosis is good (59). Complications include intercurrent infections and over-restricted dietary protein and calories with consequent developmental, growth, and hematological consequences. At the other extreme of complications is recurrent branched-chain alpha-ketoacidosis due to noncompliance or inadequate nutritional control with excessive intake of branched-chain amino acids. If the infant has been apneic with cerebral edema and hypoxemia for a period of time, outcome is guarded and may include irreversible corticospinal and cerebellar dysfunction. Studies have demonstrated an association between both early control of plasma leucine levels and long-term plasma leucine levels and intellectual outcome in patients with maple syrup urine disease (47; 39; 40; 60; 20; 17). A more recent study of 21 French patients with maple syrup urine disease (mean age of 8.7 years) who were treated early and who had mean leucine levels of about 200 uMol/L demonstrated an average full-scale IQ in the normal range, with a subset of patients having higher verbal IQ as compared to performance IQ (12). Interestingly, attention issues were reported by the caregivers of most patients in this cohort (12). In addition, an increased prevalence of neuropsychiatric comorbidities, including mood disorders and anxiety, has been reported in individuals with maple syrup urine disease (60; 01).

Clinical vignette

The first child of a nonconsanguineous couple with a noncontributory family history was born at term without complications after an uneventful pregnancy. On the third day of life and 3 hours after his last breastfeeding, during which he was irritable, the child would not take the breast and was drowsy. He progressively slipped in a comatose state with pedaling movements of the legs and intermittent apnea requiring intubation.

Clinical examination revealed generalized hypertonia, a bulging fontanel, and urine with a burnt sugar smell. DNPH testing of the urine done at the bedside was positive for alpha-ketoacids (cloudy appearance of urine on 1:1 mixture in a clear tube with a DNPH solution consisting of 0.1% 2,4-dinitrophenylhydrazine in 2N HCl). Laboratory investigations revealed a mildly elevated ammonia and severe hypoglycemia. Plasma amino acids showed elevations of leucine, isoleucine, and valine, along with the pathognomonic finding of alloisoleucine.

The infant was managed by hemodialysis and TPN with restriction of leucine but adequate supplementation of valine and isoleucine. The neurologic status and metabolic abnormalities quickly improved; he was switched to enteral feeding and he was weaned from ventilation and dialysis within 24 hours. He was discharged home 5 days later after the parents were taught about his new diet and given instructions on management of minor and major decompensations.

Biological basis

Etiology and pathogenesis

Maple syrup urine disease is caused by defects in the branched-chain alpha-ketoacid dehydrogenase complex proteins. These are nuclear-encoded mitochondrial proteins. Mutations have been identified in 4 proteins of the complex: E1-alpha and E1-beta subunits composing the dimeric E1 or branched-chain alpha-ketoacid decarboxylase, the branched-chain dihydrolipoamide acyltransferase (E2), and lipoamide oxidoreductase (E3). Several cofactors are involved in the overall reaction, including thiamine pyrophosphate, lipoamide covalently bound to E2, coenzyme A, and flavin and nicotinamide adenine dinucleotides (FAD and NAD). The overall reaction catalyzes the decarboxylation of the branched-chain alpha-ketoacid, acyl-transfer with the thiamine cofactor by E1 to lipoic acid bound to E2, production and release of the branched-chain acyl CoA, reoxidation of lipoic acid by the E3 flavoprotein, and production of NADH and H+ from NAD+ to reduce FAD. The reaction is regulated through inactivation by phosphorylation of 2 serine residues on E1-alpha through a kinase and activation through their phosphatase PPM1K.

A homozygous mutation in PPM1K has been associated with a mild form of maple syrup urine disease (65).

Loading studies in maple syrup urine disease patients revealed that leucine (and alpha-isocaproate, leucine’s ketoacid present in equimolar concentration with leucine) increases were associated with the most severe neurologic symptoms, especially if levels were above 1 mM (89). It was shown as early as 1967 that alpha-isocaproate inhibits the Krebs cycle at concentrations found in maple syrup urine disease patients (25). A plethora of subsequent studies have suggested that the pathophysiology of maple syrup urine disease is complex. Research in the pathophysiology of maple syrup urine disease revolves around:

The mouse model with an iMSUD phenotype provides further evidence for the first 2 pathophysiological mechanisms and is a useful model for potential therapeutic studies (106). There is also a natural bovine model, but dietary therapy fails to allow survival of this model.

Epidemiology

Maple syrup urine disease is panethnic and has an incidence between 1 in 185,000 and 1 in 215,000 (18). The Y394N mutation in the E1-alpha gene, however, is common in the Old Order Mennonite population, with an incidence of classic maple syrup urine disease of 1 in 358 (71).

Prevention

Newborn screening for maple syrup urine disease is now most commonly done by tandem mass spectrometry from dried blood spots and measures whole-blood concentration ratios of leucine and isoleucine to alanine and phenylalanine. ACT sheets are available from the ACMG websites.

Differential diagnosis

Confusing conditions

The signs and symptoms of maple syrup urine disease are similar to several other inherited disorders producing accumulation of organic acids in the brain. These include, among others, propionic and methylmalonic acidemia. Occasionally, ammonia levels are increased but not to the levels seen in disorders of the urea cycle. Secondary increases in valine can be seen with prolonged fasting and ketosis. The fragrant smell that helps in clinical diagnosis may not appear in the newborn period, so biochemical analysis of blood and urine for branched-chain amino acids and their alpha-ketoacid products is a necessary screen requiring final diagnosis through enzymatic and molecular confirmation. Elevations in the 3 branched-chain amino acids with low or normal levels of branched-chain alpha-ketoacids in the urine have been detected in individuals with biallelic variants in BCAT2, which encodes the mitochondrial branched-chain aminotransferase (49). Moreover, unlike in maple syrup urine disease, L-alloisoleucine is not typically detected in the plasma of individuals with BCAT2 deficiency (49).

Diagnostic workup

The first pathognomonic sign of maple syrup urine disease may be the sweet-smelling diaper or cerumen at the tip of the otoscope. A bedside diagnosis could be made using the dinitrophenylhydrazine reaction on a few milliliters of urine. On about 1.0 mL of heparinized blood, quantitation of plasma amino acids will lead to a diagnosis. Normal, early-childhood ranges of L-leucine, L-isoleucine, and L-valine concentrations in plasma are 29 to 151 µM, 20 to 96 µM, and 67 to 299 µM, respectively. In classic maple syrup urine disease, plasma leucine concentration will be greater than 400 µM at 48 hours and often may be well above 2000 µM. L-alloisoleucine is a pathognomonic amino acid in maple syrup urine disease and results from the racemization of l-isoleucine during transamination (52). In some chromatographic assays, alloisoleucine co-elutes with isoleucine, thus, preventing its detection and quantification. When the plasma leucine concentration is above 400 µM, its alpha-ketoacid derivative will overcome the renal threshold and appear in urine. Evaluation of urinary organic acids by gas chromatography and mass spectrometry will virtually diagnose maple syrup urine disease and enable immediate dietary intervention. Starvation alone may increase branched-chain amino acids in plasma in normal children but not to concentrations that produce increased urinary branched-chain alpha-ketoacids. Further diagnoses should include enzyme analysis in dermal fibroblasts or transformed lymphoblasts cultured from the patient. These cells or freshly isolated leukocytes can also be used for mutational analysis available in several laboratories (see GeneTests website for most recent approved testing laboratories).

Management

Management of clinically stable infants and children. Management is directed at nutritional therapy to meet the specific needs of the infant in maintaining optimum growth and development. The leucine tolerance as well as valine and isoleucine requirements decrease with age and should be adjusted according to individual patients. Monitoring of amino acids can be performed by the use of dried blood spots at home. Protein and energy requirements also decrease with age, and the diet should be adjusted accordingly. Growth and psychomotor development should be monitored. Essential amino acid deficiency can lead to anemia, acrodermatitis, hair loss, and growth failure amongst other complications and should, thus, be given in sufficient amounts and monitored (73). Micronutrients, fatty acids, and mineral intake should also be adequate (04; 54; 93). Use of LNAA-supplemented formula has been advocated based on clinical data and a mouse model (93). Pharmacological amounts of thiamine (8 to 10 mg/kg per day) should be added but require at least 3 weeks to produce an effect (27). It is important to emphasize the lifelong nature and need for monitoring diet records, growth and development, and biochemical parameters in patients with maple syrup urine disease. Acute and chronic management of maple syrup urine disease should be performed with the help of a metabolic physician, a specialized nutritionist, and, in severe decompensations, an intensivist. Several more detailed works exist on the subject (03; 10; 02; 91; 93; 30) and nutritional management guidelines have been published (32).

Carnitine supplementation. In maple syrup urine disease patients, carnityl adducts of the branched-chain alpha-ketoacids are of minor therapeutic importance in producing an alternative pathway for their excretion. If recurrent ketoacidosis depletes free carnitine, supplementation may be indicated. Also, carnitine has improved the oxidative stress in a mouse model of maple syrup urine disease (50) and may improve oxidative stress in patients with the disorder (57; 58).

Liver transplantation. Liver transplantation has become an increasingly common therapeutic approach in patients with maple syrup urine disease. A case series described outcomes in 54 patients who had liver transplantation in the United States, with 98% patient survival and 96% graft survival (53). Of the 54 patients, the authors described the clinical course of 37 patients who received liver transplants at their own center, and all 37 patients had increased leucine tolerance posttransplantation (53). Even after transplantation, hyperleucinosis with or without encephalopathy has been reported in rare cases in the setting of intercurrent illness, and thus, precautions should be taken in the setting of illness even after liver transplantation (38). Pre- and post-transplantation IQ scores were similar in the majority of 14 patients who were tested (81). However, liver transplantation is associated with improved long-term metabolic stability and reduces the risk for future metabolic decompensations and, thus, likely prevents worsening of cognitive outcome (28).

Management of minor intercurrent illnesses. If the child has a minor intercurrent illness and no obvious neurologic deterioration, he could be followed at home by administration of a high-calorie BCAA-free formula and monitoring of urine with ketone detection sticks or DNPH test if the parents have been appropriately trained to perform the test. A retrospective chart review suggests that the use of leucine-free formula can reduce the plasma leucine level by approximately 50% from the initial level in a 24-hour period in the setting of hyperleucinemia (79).

Management of severe metabolic decompensation.

Prevent catabolism and promote anabolism. Treat infection or precipitating illness and use antiemetics if the patient is nauseous (vomiting can worsen ketoacidosis and complicate enteral feeding). It is of utmost importance to promote anabolism during an acute attack by immediately providing sufficient calories and BCAA-free essential and nonessential amino acids. Depending on the status of the child, one can use orogastric perfusion of BCAA-free formula with excess calories (150 to 170 KCal/kg per day) and adequate protein (3.0 g/kg per day). An intravenous BCAA-free solution has been utilized as an alternative to enteral formula in individuals who cannot tolerate enteral feeds (05). Isoleucine and valine should be supplemented because they will rapidly become limiting for protein synthesis. Alanine, a gluconeogenic amino acid, is a co-substrate for the BCAA aminotransferases and can also be supplemented. Glutamine is used more rapidly in times of stress, and the glutamate-glutamine-GABA cycle is disrupted by leucine; thus, glutamine can also be supplemented (63; 103). Intravenous dextrose with normal saline and 20 meq KCl is recommended initially. Hypoglycemia can be severe in decompensated maple syrup urine disease, but if glucose rises too much with the therapy, insulin can be initiated.

Detoxify blood with hemodialysis. In cases of severe decompensations, eg, with cerebral edema, coma or cardiovascular compromise, extracorporeal removal therapy (ECRT) might become necessary. Several physicians advocate promoting anabolism instead of resorting to ECRT (63; 59). Peritoneal dialysis and exchange transfusions were used early on, but the leucine clearance rate is relatively slow and complications are more frequent than with other techniques. The effectiveness of hemodialysis in maple syrup urine disease has been reported by several groups (75; 76; 95; 44; 78; 72; 68). Continuous venovenous hemodialysis was compared with nutritional support and was shown to be much faster at reducing leucine levels (43; 72). Kinetic modeling can predict the efficiency of continuous hemodialysis (42), and intermittent hemodialysis was shown to be even faster than continuous hemodialysis (68).

Monitor and minimize brain swelling. Monitor signs such as the level of consciousness, hyperreflexia, papilledema, vomiting, headache, hypertension, heart rate deceleration, and abnormal respiratory pattern. Cerebral edema is prevented by rapidly correcting leucine levels, maintaining normal serum sodium levels, correcting acidosis by the use of bicarbonate if necessary, avoiding rapid osmolarity changes, and minimizing stress. If cerebral edema occurs, the use of furosemide, mannitol, and hypertonic saline should be considered with an intensivist.

Experimental therapies. Antioxidants such as vitamin E or C have been suggested to decrease the oxidative stress damage (99; 74). Carbamylglutamate, an analog of N-acetylglutamate that can stimulate the early steps of the urea cycle, has also been shown to have some marginal efficacy at helping to drop ammonia levels during the decompensation of a patient (46). Liver cell transplantation with 3% correction of hepatocytes increases growth and lifespan in mice (88). Similarly, transdermal hepatic injections of human amnion epithelial cells starting in the neonatal period resulted in doubling of BCKDC enzyme activity and reduction in BCAA levels in the serum and brain of mice (87). Subcutaneous transplantation of adipose tissue from wild-type mice into 2 different mouse models for maple syrup urine disease resulted in decreased BCAA levels compared to untransplanted mice (104). Gene therapy has not yet been reported in experimental models of maple syrup urine disease. Norleucine competes with leucine for brain entry at the blood-brain barrier (94) and has been shown to be effective in treating mice with iMSUD (106). It also appears that norleucine inhibits transamination of leucine to alpha-ketoisocaproate (106; 105). Because phenylbutyrate inhibits BCKDC kinase and increases the complex’s activity in vitro and in vivo, a clinical trial in patients with maple syrup urine disease is underway (15). Interestingly, a rapid reduction in leucine level was observed in the setting of an acute metabolic decompensation in a patient using intravenous sodium phenylacetate/sodium benzoate (51). In addition, a study demonstrated reduction of leucine with oral sodium phenylbutyrate in the setting of acute leucinosis (107). This result suggests that oral sodium phenylbutyrate might be considered for reducing leucine levels in individuals who are not appropriate for extracorporeal removal of leucine or in small centers where this expertise is not available although more controlled studies are likely needed to definitively demonstrate efficacy. Finally, a structurally-based design approach has been used to design a novel inhibitor (S-CPP) of the BCKDC that results in decreased plasma BCAAs after drug delivery (97). Preclinical studies of metformin in the iMSUD mouse model demonstrated a reduction in the levels of leucine and its corresponding ketoacid in skeletal muscle and serum and associated improvements in mitochondrial function (90). Further studies are needed in order to determine whether metformin may be a useful therapeutic option for individuals with maple syrup urine disease.

Gene therapy is also a growing area of investigation in animal models of MSUD. To date, attempts at using liver-directed gene therapy using liver specific promoters in murine models have been suboptimal, with better results achieved using vectors targeting extra-hepatic tissue, such as muscle (36; 69).

Special considerations

Pregnancy

In a first report of a maple syrup urine disease patient giving birth to a normal offspring, the mother required large amounts of branched-chain amino acid intake during pregnancy, presumably to provide these essential nutrients to her fetus. After delivery, her requirements for L-leucine, L-isoleucine, and L-valine dropped considerably (98). Subsequently, another pregnancy was reported with a similar increased leucine tolerance during pregnancy followed by a short drop after delivery (37).

Anesthesia

It is important to prevent endogenous protein catabolism and consequent branched-chain alpha-ketoacidosis caused by prolonged fasting before, during, and after surgery. Maintenance of adequate calories and BCAA-restricted protein intake may require specialized hyperalimentation fluids.