3-Hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) lyase deficiency in Saudi Arabia

J. Inher. Metab. Dis. 14 (1991) 174 188 © SSIEM and Kluwer AcademicPublishers. Printed in the Netherlands

3-Hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) Lyase Deficiency in Saudi Arabia P.T. OZAND 1, A. AL AQEEL1'2, G. GASCON1, J. BRISMAR3, E. THOMAS'* and H. GLEISPACH5 Department of t Paediatrics, 3Radiology, 4Nutrition and 5Biological and Medical Research, King Faisal Specialist fIospital Research Centre, Riyadh 11211, Saudi Arabia," 2Riyadh Armed Forces Hospital, Riyadh S,mmary" Deficiency of 3-hydroxy-3-methylglutaryl-CoA lyase has been studied in 11 Saudi infants. The diagnosis was established by the measurement of enzyme activity in lymphocytes, in fibroblasts and, in seven patients, by the gas chromatography/mass spectrometer pattern of excreted organic acids in the urine. In seven infants the disease caused a devastating acidotic attack within the first day of life, while in two the crisis occurred by the third day of life. In two infants from one family the disease appeared later in infancy. The clinical presentation of an acidotic attack is lethargy, hyperpnoea, tachypnoea and seizures, either at birth (two infants), following first feeding (in five infants), or following vomiting or refusal of food in later infancy. The acidotic attacks recurred later in life following minor illness or refusal to eat. The acidosis of this enzyme deficiency progresses rapidly, leading to cardiopulmonary arrest and death within hours of onset unless treated promptly. In four surviving infants diagnosed and treated early, development is normal. Magnetic resonance and computerized tomography brain scans in these infants, however, show white matter lesions and mild atrophy.

INTRODUCTION Recent reviews by Wysocki and Haehnel (1986) and Gibson and colleagues (1988a, 1988b, 1990) cover 23 patients with 3-hydroxy-3-methylglutaryl-CoA tyase deficiency (HMG-CoA lyase deficiency, EC 4.1.3.4). The clinical presentaton of the disease in these previoius reports differs from that in our 11 Saudi Arabian patients studied during the past four years. Although the disease is characterized by repeated acidotic attacks, our patients presented at a younger age and with more severe attacks. We believe that this disease remains underdiagnosed in our area because its early onset with severe organic acidaemia is not appreciated. In the neonatal period the disease is fatal unless promptly treated. In four children who are alive and currently followed, MS received 7.8.90 Accepted 23.11.90 174

HMG-CoA Lyase Deficiency

175

the outcome has been satisfactory despite numerous episodes of repeated acidotic attacks and evidence of white matter disease. PATIENTS All patients in this study were delivered in other hospitals. Five infants were referred to the Inborn Errors of Metabolism Service, King Faisal Specialist Hospital and Research Centre for treatment. We received biological material such as urine for gas chromatography/mass spectrometry (GC/MS), blood for lymphocytes, and skin biopsy for fibroblast studies from six infants who died before they could be transferred. A brief summary of the 11 patients with HMG-CoA lyase deficiency follows.

Group 1: Neonates from other hospitals Patients [6: These patients manifested a similar clinical picture. Two infants were born prematurely, were small for gestationat age, and were delivered following obstetric signs of fetal distress. Both had lactic acidosis (pH 6.8--7.0) and hypoglycaemia at birth. Despite adequate alkali and glucose administration they died within 72 hours. The urine GC/MS revealed the presence of large amounts of organic acids present in a pattern suggesting HMG-CoA lyase deficiency. Fibroblasts grown from post mortem skin biopsies indicated the presence of HMG-CoA lyase deficiency. Four infants developed severe lactic acidosis with hypoglycaemia following the first breastfeed. Intervention was delayed for several hours as the severity of the condition was not appreciated, and they died of irreversible lactic acidosis within 96 hours. HMG-CoA lyase activity was deficient in the fibroblasts grown from post mortem skin biopsies. The family history of one patient revealed three neonatal deaths after similar symptoms. Group 2: Three patients who survived neonatal lactic acidosis Patient 7: This child was the product of a first cousin marriage. She was delivered, following fetal bradycardia and late decelerations, with meconium staining. She developed cyanosis, vomiting and lethargy following her first feed and was rehospitalized at 3 days. In the intensive care unit she improved after intravenous (i.v.) infusion of glucose and alkalinizing agents. On resuming milk feeds the next day, she developed severe lactic acidosis, hypoglycaemia and coma. The urine GC/MS indicated HMG, 3-methylglutaric, 3-methytgtutaconic and 3-OH-isovaleric acids and 3-methytcrotonylglycine. The serum amino acids analysis indicated a 50% increase in glutamic acid, alanine, serine and glycine. She was transferred to our service at 2 weeks of age and was placed on a low fat (30%) and low leucine (87 mg/kg/day) diet. At this time her blood pH, ammonia and lactate were normal. She did well but the parents dropped out of the follow-up and discontinued the diet and medications. At the age of 6 months, following refusal to eat, she arrived at the emergency room with convulsions, coma, hepatomegaly of 7 cm, increased deep tendon reflexes and bilateral sustained ankle clonus. Glucose was 0.55mmol/L pH 7.13, BE --13 mmol/L, H C O 3 9.6retool/L, lactate 10mmol/L, pyruvate 0.4mmol/L, ammonia 105#mol/L, GPT 175IU/L, and GOT 365 IU/L. Seizures were controlled with phenobarbital. Her J. Inher. Metab. Dis. 14 (1991)

t76

Ozand et al.

acidosis and hypoglycaemia were corrected by alkali and glucose administration and L-carnitine at 100 mg/kg/day. She developed hypokalaemia which was corrected by gradual infusion of potassium over three days. She left the intensive care unit after three days, at which time hepatomegaly had decreased to 2-3 cm and bilateral ankle clonus had disappeared. She is 3 years old at present and has suffered four more acidotic attacks requiring short term hospitalizations, each attack provoked by neglect, minor infection or simply refusal to eat. Her psychomotor development has been assessed on a yearly basis using the Bayley Infant Development Scales initially, and, from 2 years of age, the Leiter International Performance Scales. She is developing normally. Her growth is progressing at the fifth percentile (her parents have short stature). She has no neurological sequelae. Both her lymphocytes and fibrobtasts lack HMG-CoA lyase activity. At the age of 3 years she is showing moderate HMG-uria. The computerized tomography (CT) scan of the brain at 2 weeks of age showed a prominent interhemispheric tissue and possible white matter abnormalities round the frontal horns. At the age of 2.5 years a magnetic resonance (MR) study demonstrated normal sized ventricles and sulci but confirmed the presence of white matter changes round the frontal horns, and a multitude of 1-2ram lesions, mainly 'in the subcortical supratentorial white matter (Figure 1). Following an uneventful pregnancy and delivery, this girl became irritable at 3 days of age. She was taken to a local hospital where improvement followed i.v. fluid administration. After formula feeding she developed lethargy, at which time an organic acid disorder was suspected. A GC/MS analysis of the urine indicated large amounts of HMG, 3-methylglutaric acid, 3-OH-isovaleric acid, and 3-methylcrotonylglycine. An immediate lymphocyte study for HMG-CoA lyase revealed the deficiency and she was transferred to our hospital for further diagnostic studies and treatment. Deficient HMG-CoA lyase activity was later confirmed in cultured skin fibroblasts. At the time of referral she was 40 days old, her liver was 5-6 cm below the costal margin, and she showed mild spastic quadriparesis with exaggerated reflexes. Blood ammonia, lactate and pH were normal. A C T scan of the brain showed quite marked prominence of the interhemispheric tissue but was otherwise normal. (Figure 2A). She was placed on a low leucine (80mg/kg/day), low fat (20%) diet and received polycitra at 4 mEq/kg/day, and L-carnitine at 100 mg/kg/day. She was discharged in good condition. The parents dropped out of the clinic and discontinued both the diet and the medications. She was admitted outside Riyadh with severe lactic acidosis at the age of 5 months. The physician contacted our service and administered appropriate treatment. After two weeks she returned to our clinic showing limb spasticity and bilateral sustained ankle clonus, which later improved. Assessment using the Bayley Scale of Infant Development at the age of 8 months showed normal development in all areas. The EEG was normal. At the age of 14 months she developed a minor upper respiratory tract infection and refused to feed. She became lethargic 12 hours later, with a pH of 7.37. One hour later she became dusky. Repeat blood gases showed pH 7.23, HCO3 14.3mmol/L, BE --12.5 mmol/L and dextrostix < 40rag%. She went into Patient 8:

J. Inher. Metab. Dis. 14 (1991)

'H M G - C o A

177

L y a s e Deficiency

Figure 1 (Patient 7) A: (left) The CT scan of the brain at 2 weeks shows widening of interhemispheric fissure and some possible white matter abnormalities around the frontal horns. B: (right) MR of the brain at 2.5 years (1.5T; 2000/40) verifies the continued presence of white matter changes around the frontal horns and also (on the other sections) shows scattered lesions within the subcorfical white matter. shock with poor peripheral circulation and cold extremities, and 30min later, at which time the capillary blood pH was 6.90 and blood glucose 0 mg%, she suffered a cardiopulmonary arrest, and was resuscitated by the administration of large amounts of glucose and bicarbonate. Following a generalized convulsion treated with 20 mg/kg of phenobarbital i.v., she was admitted to the intensive care unit, where frequent episodes of hypoglycaemia occurred despite i.v. administration of 12 15mg glucose/kg/min. She regained consciousness 6 hours later and was extubated; 36 hours later she was sitting in bed playing. Although she showed increased deep tendon reflexes and bilateral sustained ankle clonus initially, she had no detectable neurological signs on discharge six days later. Both her physicians and parents became aware of her labile clinical condition. She presented with lactic acidosis eight times between 14 and 36 months of age, each time following minor infections, vomiting or refusal to eat. Each time the pH ranged between 7.t 7.3. Each time she was hospitalized briefly for observation and administration of i.v. fluids. She never experienced a devastating episode again. She visits the clinic monthly and is an apparently normal child at the age of 3 years. Her growth is 3 SD below, but parallel to normal percentiles. She continues to excrete large amounts of H M G in her urine. Yearly psychomotor evaluations indicate normal

J. Inher. Metab. Dis.

14 (1991)

178

Ozand et al.

B

A

C Figure 2 (Patient 8) A: CT scan of the brain at 1.5 months shows quite marked prominence of interhemispheric fissure. No white matter disease and normal sized ventric!es. B: CT scan of the brain at 22 months shows normal ventricle, sulci and fissures, ttigher slices show questionable white matter disease bilaterally in centrum semiovale. C: MR of the brain at 32 months (1.5T; 2000/80) demonstrates one high T2-intensity lesion immediately frontal to the right frontal horn and a few small lesions bilaterally in the subcortical white matter.

J, Inher. Metab. Dis. 14 (1991)

179

H M G - C o A Lyase Deficiency

development. Her EEG is normal. A C T scan at 22 months (Figure 2B) was normal except for questionable white matter disease in the centrum semiovale; a recent MR scan of the brain shows only one white matter lesion adjacent to the right frontal horn (Figure 2C). This baby girl was normal at birth and was breastfed. Six hours later she developed a devastating metabolic acidosis, pH 6.8, BE - 2 5 mmol/L, lactate 20 mmol/L, and glucose 0.2 mmol/L. The neonatologist immediately corrected her acidosis and hypoglycaemia. At the age of 24 hours her arterial blood pH was 7.2, and lactate 13.5mmol/L. A urine sample submitted for GC/MS analysis revealed massive HMG-uria and she was transferred to our service at the age of 7 days. Family history revealed that the parents were first cousins and had lost a baby girl shortly after birth with similar complaints. Five maternal aunts had also died with a similar disease. At the time of referral, physical and neurological examinations were entirely normal. The Brazelton Neonatal Assessment Scale indicated no abnormalities. The admission blood glucose was 5.8mmol/L, lactate 2.7mmol/L, pyruvate 100~tmol/L, and ammonia 77 #mol/L. She was placed on a low fat (30%) and leucine restricted formula (87mg/kg/day), with the addition of 4mEq/kg/day of polycitra, and L-carnitine at 100mg/kg/day. Her EEG was normal. She is followed as an outpatient once or twice a month. At present she is 9 months of age and is growing along the fifth percentile. She shows axial hypotonia and exaggerated deep tendon reflexes. Her parents refuse to co-operate for further neuroradiological and neurophysiological studies. She continues to show mild compensated metabolic acidosis and is excreting large amounts of HMG in the urine. Despite poor compliance the baby has not experienced another attack of severe acidosis. Patient 9:

Group 3: Two patients who did not experience a neonatal acidotic episode Patient 10: This baby boy was the product of a first cousin marriage in which a girl (patient 11) had previously died with HMG-CoA lyase deficiency. He was the product of a normal full term pregnancy and delivery and was breasffed at birth without any complications. He was diagnosed at the age of 2 months when a urine sample analysed by GC/MS revealed the characteristic organic acids of HMG-CoA lyase deficiency. At that time his paediatric and neurological examinations were within normal limits, blood lactate values ranged between 1.8-2.4retool/L, and ammonia values between 44-81 #mol/L. Blood amino acid profile, liver enzymes and EEG were normal. A C T study of the brain at 5 months showed slightly enlarged ventricles and prominent frontal sulci (Figure 3A). He was placed on a leucinerestricted (80 mg leucine/kg/day), low fat (30%) diet, with polycitra at 4 mEq/kg/day and L-carnitine at 100mg/kg/day. He experienced his first acidotic attack at the age of 6 months when he developed herpangina and refused to eat. He presented in shock with poor peripheral circulation and cold extremities, with pH 7.29, HCO- 14 retool/L, BE -- 11 retool/L, and glucose 2.2mmol/L. Because i.v. administration could not be started he was initially given glucose and bicarbonate intraosseously, and was discharged home within 48 hours. He experienced similar attacks following minor infections, repeated vomiting and

J. Inher. Metab. Dis. 14 (1991)

180

A

Ozand et al.

B

Figure 3 (Patient 10) A: CT scan of the brain at 5 months shows slightly enlarged ventricles and prominent suld in the frontal area. B: MR of the brain at 21 months (1.5T; 1800/80) demonstrates prominent white matter disease in the frontal and occipital regions with scattered changes throughout the subcortical white matter. Central myelination normal. Normal sized ventricles and sulci. refusal to eat at the ages of 8, 11, 11.5, t2, 13, 14 and 16 months. At present at 21 months of age, he shows normal development for an infant of 18 months on the Bayley Infant Development Scales and is growing on the fifth percentile. He has compensated metabolic acidosis and is excreting H M G in his urine. MR of the brain at 21 months revealed white matter changes in the frontal occipital regions and also scattered minor high T2 lesions throughout the subcortical white matter (Figure 3B). This baby girl was the sister of patient 10. She was delivered and followed at another hospital and developed her first acidotic attack at the age of 2 months following gastroenteritis which spontaneously improved. During a similar episode at the age of 3 months she was found to have hepatomegaly and was investigated for glycogen storage disease, galactosaemia and renal tubular acidosis. A liver biopsy indicated diffuse fatty infiltration. The liver enzymes and neutral amino acids in serum were raised 1.5-2.0 times. At the age of 6 months she presented with a severe acidotic coma, convulsions, pH 6.9, ammonia 835mmoI/L, glucose 0.9 mmol/L, and urate 1.0mmol/L. Her blood pressure was 220/120mm. Her liver edge was 5-6 cm below the costal margin, and she appeared to be blind. A fructose tolerance test was normal. Her acidosis improved gradually, and she was referred to our service at the age of 10 months for further studies. Patient 11:

J. Inher. Metab. Dis. t4 (1991)

H M G-CoA L yase Deficiency

181

At the time of referral her liver edge was 5-6 cm below the costal margin and firm. She showed bilateral cortical fisting postures and continuous myoclonic jerks. She had no visual pursuit and did not respond to visual threat. Her admission blood pH, lactate, glucose, ammonia, pyruvate and amino acids were normal. An E E G showed episodic slowing anteriorly, and independent sharp wave foci at the left and right occipital regions. A brain CT scan showed marked widening of the ventricular system and of sulci and fissures (Figure 4). At that time GC/MS was not available, and H P L C organic acid analysis of the urine revealed several organic acid peaks which could not be identified. She was placed on a branched chain amino acid restricted diet (MSUD formula) and was followed as an outpatient. She was admitted one month later with status epilepticus. She did not respond to lorazepam, though her dystonic posture improved somewhat upon administration of baclophen. She remained in a decerebrate, vegetative state for 2 months, and died at 19 months of age. The diagnosis was established post mortem when her cultured skin fibroblasts revealed deficiency of H M G - C o A tyase activity. No GC/MS studies of her urine are available.

Figure 4 (Patient 11) CT scan of the brain at 10 months shows marked widening of the ventricles and of sulci fissures supratentorially. No white matter disease demonstrated. J. Inher. Metab. Dis. 14 (1991)

182

Ozand et aI.

METHODS

G C / M S analysis': Urine samples, collected during acute acidosis and between attacks, were processed for quantitative measurement of organic acids as described by Hoffman and colleagues (1989) and Sweetman (1990), using a Hewlett Packard instrument (Model 5890 with mass detector 5970 and computer 9144/79559B/59824A). Enzyme measurements: Lymphocytes or fibroblasts grown from skin biopsies were used for the determination of HMG-CoA lyase activity (EC 4.1.3.4). In this report only those results detected in cultured fibroblasts will be listed. Control measurements were performed using fibroblasts from neonates, grown either from foreskins from newborns or from infants of both sexes with unrelated diseases. Twelve of these cell lines were established from babies who died within the first week of life from other causes. The fibroblasts are available from the Saudi Mutant Cell Culture Repository. HMG-CoA lyase activity was measured using a modification of the method of Gibson and colleagues (1990). Briefly, fibroblasts were harvested and washed three times with MEM and frozen. The pellet was homogenized in 0.3 ml of 20 mmol/L Tris-HC1, pH 8.0, and the protein content determined. The assay mixture contained Tris-HC1 100retool/L, MgC1/ 10mmol/L, dithiothreitol 5mmol/L, HMG-CoA (PL Biochemicals, 96% pure) 0.7retool/L, [3-14C]HMG-CoA (Amersham, UK, 52mCi/mmol) 0.5 x 106dpm/ml, pH 8.1 and protein 1 mg/ml, in a total volume of 0.2 ml, in triplicate. Following incubation for 30 rain at 37°C, the [3 - 14C]acetoacetate produced was converted into the D6nig6s salt of acetone (Reed et al., 1984). For this purpose, 0.2 ml of freshly prepared 100 mmol/L acetone and 3 ml of 5.3 N H2SO4 which contained 5.5 g% HgSO4 was added to each tube. The mixture was sealed and heated in boiling water for 30 min, chilled on ice, and kept at 4°C overnight. The D6nig6s salt was collected by centrifuging the tubes at 4°C for 10rain. The precipitates were washed with 1 ml of cold distilled water twice, and were finally dissolved in 2ml of 4N HC1 and counted with 10rot of Optifluor. The counts in the blank tubes were < 2% of the total counts present and 10-20% conversion of [3-14C-]HMG-CoA to acetoacetate was observed in healthy fibroblasts under the experimental conditions. The conversion was linear up to 90mins and for 0.251.5mg/ml protein concentrations. The results are expressed as nmol HMG-CoA converted in acetoacetate/min per mg protein. The Student t test was used to calculate the statistical significance. In addition, HMG-CoA lyase activity was assayed independently by Drs Gibson and Nyhan by a modified radiochemical assay in five fibroblast cultures (Gibson e~ al., 1990) to reconfirm the enzyme deficiency. RESULTS The clinical and biochemical presentation of HMG-CoA lyase deficiency in the 11 infants is presented in Table 1. All patients were the products of consanguineous marriages, and except for two children from one family, the disease was manifested by 3 days of life. At the time of acidotic attack they had significantly elevated lactate, J. Inher. Metab. Dis. 14 (1991)

Yes

Yes Yes, two girls and one boy died neonatally with the same disease

4F

5F 6F

First cousin

First cousin, one girl and five maternal aunts died neonatally with the same disease

8F

9F

As in patient 10

llF

na: not available

First cousin, full sibling of patient 11

10 M

Group 3:

First cousin

7F

Group 2:

Yes Yes Yes

1M 2M 3M

Group 1:

Consanguinity Coma, shock Coma, shock Coma

Presenting symptomatoIogy

2months

6 months

6 h after feeding

3 days

3 days

6.8/-25.0

7 . 2 3 / - 12.5

7 . 1 3 / - 13.0

7.0/na 6.9/na

6 . 9 5 / - 19.7

6.8-7.0/na 7.0 6.8 na

na/na

10.0/0.5

na

10/na

10/0.4

ha/ha > 20.0/na

11.3/na

37.9/40.2 ha/ha > 20.0/na

835

85

na

na

105

na 300

na

239 na > 500

0.9

2.2

0.2

2.2

0.55

< 1.0 < 1.0

< 1.0

< 1.0

1.1

Biochemical findings during acidosis pH/ Lactate (mmol/L)/ Ammonia Glycaemia base excess pyruvate (mmol/L) (/~mol/L) (retool/L)

Refusal to eat as 7.29/-11.0 a result of herpangina. Irritability, shock Vomiting, shock 6.9/na

Lethargy, vomiting, cyanosis Irritability, lethargy Vomiting, lethargy, convulsions

Cardiopulmonary arrest 1 day after feeding Coma, shock 6 h after first feed Cardiopulmonary arrest

6 h after first feed 6 h after first feed Birth, pematurely small for gestational age 2 h after birth

Age at first attack

Patients and clinical presentation of HMG-CoA lyase deficiency

Patient no. and sex

Table~l

184

Ozand et al.

hypoglycaemia, and hyperammonaemia. The absence of ketonuria despite the severe acidosis was notable. The diagnosis and progression of the disease is outlined in Table 2. The disease was confirmed by deficiency of HMG-CoA lyase activity in cultured fibroblasts. In seven infants the enzyme activity was also shown to be deficient in lymphocytes. The GC/MS revealed elevated urinary excretion of lactate, HMG, 3-methylglutaric acid, 3-methylglutaconic acid, 3-hydroxyisovaleric acid, and 3-methylcrotonylglycine, 10 to 10000 times the normal level in seven patients from whom urine samples were available. The HMG-uria persists with compensated metabolic acidosis between attacks. The age of onset of disease in our patients was lower than previously reported in the literature (Table 3). The activity of HMG-CoA lyase in fibroblasts as measured by us is listed in Table 4. Drs Nyhan and Gibson confirmed the deficiency in five of these children through a different procedure; the results are included in Table 4. We obtained serial neuroimaging studies of the brain in four patients (Figures 1-4). DISCUSSION HMG-CoA tyase deficiency in Saudi Arabia is a neonatal-onset organic acidaemia with lactic acidotic attacks in later infancy (Tables 1 and 2). Early death occurs if the condition is not recognized. Patients 1-6 were delivered in outside hospitals and died neonatally. Patients 7-10 were also delivered outside the Riyadh Centre, but there was prompt suspicion of an organic acidaemia. They survived and remain in good health. Patient 11 had irreversible brain damage at the time of referral and later died. Three of these latter five patients (7-11) experienced 3-11 acidotic attcks during the period of follow-up, all of which required hospitalization, demonstrating the continuing morbidity. While approximately two-thirds of our Saudi patients had manifested the disease during the first 24 hours, only one out of 23 previous patients did so (Table 3). Nine out of 11 Saudi patients, (the exception being two children from one family), had expressed the disease by 3 days of life, compared to less than one-third of total cases from previous reports. A review of the literature suggests that the majority of earlyonset patients originate from the Middle East and North Afl'ica (Sovik et aI., 1984; Zoghbi et al., 1986; Dasouki et aL, 1987; Duran et aL, 1979; Schutgens et at, 1979; Leonard et al., 1979). The degree of HMG-CoA lyase deficiency did not correlate with the age and severity of the disease at onset, nor with the number of repeated episodes (compare Table 4 to Tables 1 and 2). The disease invariably leads to a rapidly progressive acidosis which has also been noted by others (Zoghbi et aI., 1986; Schutgens et aL, 1979). A good example was patient 8, whose blood pH dropped from 7.37 to 6.90 within 90 minutes. Despite her cardiopulmonary arrest she recovered fully within 6 hours. Except for patient 9, acidotic episodes recurred at least once every 2-3 months (Table 2), despite parental compliance with the diet and medications in most instances. In previous reports such frequent recurrences have not been noted. In four reports no further acidotic attacks occurred (Leonard et al., 1979; Robinson et al., 1980; d. Inher. Metab. Dis. t4 (1991)

7~

Yes

Yes

NA

11

NA:Not analysed

Yes NA

Yes

Yes

Yes

Yes

Yes

Yes NA Yes NA NA Yes

Yes

Yes

Yes

Yes

Yes

Yes Yes Yes Yes Yes Yes

Enzyme deficient in Lymphocytes Fibrobtasts

NA Yes Yes Yes NA NA

Group 3 l0

9

2 3 4 5 6 Group 2 7

1

Group 1

GC/MS confirmed

13/19 months

9/18months

0/9months

11/36 months

6/36 months

Number of attacks/ chronologicalage

Diagnosis, progression and outcome of HMG-CoA lyase deficiency

Patient no.

Table 2

72 h 96 h 72 h 72 h 72 h 96 h

Normal developmental assessment, pronounced white matter disease periventricularly in MR of the brain Blind, deaf at 10 months with severe brain atrophy at the time of referral, status epilepticus, then oculogyric crisis at 11 months with decerebration, died at home

Normal developmental assessment, mild changes in MR of brain Normal developmental assessment, 1-2 high T2 intensity areas in MR of the brain Developmental delay of 4-5/9 months last visit

Died: Died: Died: Died: Died: Died:

Outcome and assessment

Ozand et aI.

186

Table 3 studies

The age of onset of HMG-CoA lyase deficiency in Saudi Arabia as compared to other

Age at onset Present study (% of case) Previous studies Gibson et al., 1998b Gibson et aL, 1990 (% of cases)

Age at the first acidotic attack 1-52 weeks More than 1 year

0 24 h

1-7 days

7 64

2 18

2 18

0

6a 2 35

10 2 52

ib

4

2 9

Ethnic origin of patients: "United Arab Emirates, Saudi Arabia, Pakistan, Morocco, Portugal; b Hispanic

Table 4

HMG-CoA lyase activity in fibroblasts of patients

Patients

By the procedure in methodology

As assayed by Gibson (1990)

(nmol/min per nag protein) Control patients a (Mean ± SD, n) (Range of activity) Previous controls (Mean _+ SD, n) (Range of activity) Simultaneous controls (Mean _ SD, n) Range of activity) 1 2 3 4 5 6 7 8 9 10 11 Father (nos 10 and 11) Mother (nos 10 and 11)

5.98 ___2.64 (n = 36) (2.13-18.65)

-

-

4.81 +_ 0.59 (n = 7) (3.94-5.70)

0.22 0.53 0.31 0.09 0.02 0.01 0.80 0.61 0.10 0.29 0.41 3.45 3.08

2.79 ± 0.97 (n = 3) (1.77-3.71) 0.36 0.60 0.55 0.46 0.81 -

a Control cells included fibroblasts from 12 newborns who died from unrelated disorders. All fibroblasts were harvested at early passage for enzyme determinations. Statistical significance as compared to controls was p < 0.005 for both determinations by the procedure in the methodology as assayed by Drs Gibson and Nyhan (t990). The activity in the fibroblasts of parents was not different from controls. L i s s o n et aI., 1981; D u r a n et al., 1979). I n six o t h e r r e p o r t s the a t t a c k s r e c u r r e d o n l y o n c e o r twice ( Z o g h b i et al., 1986; S c h u t g e n s et at., 1979; M a r k l o v a et al., 1987; G i b s o n et al., 1988a, 1988b, a n d 1990). O n l y in t w o p a t i e n t s of F r e n c h - C a n a d i a n o r i g i n d i d s u c h f r e q u e n t episodes o c c u r ( G i b s o n et al., 1988a). T h e u n u s u a l severity of the S a u d i cases m a y be d u e to o t h e r genetic factors, since o u r p a t i e n t s were

J. Inher. Metab. Dis. 14 (1991)

HMG-CoA

Lyase Deficiency

187

products of consanguineous marriages (Table 1). Only 7 out of the 23 previously reported patients were from consanguineous marriages (Gibson et al., 1988a, 1988b) and among those were all the infants from the Middle East. The diagnosis of HMG-CoA lyase deficiency can be established either by GC/MS analysis of urine or by enzyme assay in cells. Since the disease is characterized by the absence of significant ketonuria, it should be the primary diagnosis in any neonate or infant with a rapidly progressive acidotic coma, lactic acidosis, moderate hyperammonaemia and hypoglycaemia, but no ketonuria. GC/MS will reveal the characteristic pattern of HMG-CoA lyase deficiency both during the acidosis and between attacks. When a GC/MS analysis is not readily available, enzyme assay in lymphocytes will establish the disease within 24 hours. The activity in lymphocytes is 50-100% higher than in fibroblasts, and a concomitant assay of propionyl-CoA carboxylase, or another activity to ascertain the viability of lymphocytes should assure no false negative determinations. The radioactive methods described or other currently available procedures (Table 4) (Gibson et al., 1990) are satisfactory for this purpose. In all patients the diagnosis must be definitely confirmed by analysing enzyme activity in cultured skin fibroblasts. In four infants who are under current follow-up, both the neurological and psychometric assessments indicate normal development. These children received a leucine restricted (80-87mg/kg/day), low fat (20-30%) diet, to which was added polycitra as an alkalinizing agent (4 m E @ g / d a y ) (being more palatable than sodium bicarbonate), and L-carnitine at 100-200 mg/kg/day. Similar management has been advised by others (Dasouki et at., 1987; Gibson et at., 1988a, 1988b). Growth in our patients remained at the fifth percentile or below, with the appropriate growth velocity. It is difficult to conclude whether this failure to thrive is secondary to the chronic compensated metabolic acidosis, to iatrogenically created dietary deficiency, or to the disease itself. Despite apparent clinical normality, neuroimaging studies indicate residual encephalopathy (Figures 1-4). Initially some prominence of ventricles, sulci or fissures was observed in all patients. These changes became milder later, but loci of white matter abnormalities, especially at the frontal horns, were later found in all cases. In patient 11, who was decerebrate at the time of referral, atrophy of the brain was severe and no conclusions could be reached regarding concomitant white matter abnormalities (Figure 4). Only one previous report mentions the neural imaging in HMG-CoA lyase deficiency (Lisson et aI., 1981) and it notes the white matter changes. There are no previous reports ofleukoencephalopathy in CT or MR of the brain in normally developing patients with this disease. It is therefore difficult to conclude whether ours is a universal finding and whether it will have any clinical consequences later. ACKNOWLEDGEMENTS

The authors are grateful to Drs Gasudraz Ahmed of King Fahad National Guard Hospital, Riyadh, Saudi Arabia, K. M. Gibson of K. H. Courtwright and J. W. Summers Metabolic Disease Centre, Baylor Research Foundation, Baylor University Medical Centre, Dallas, USA, and W. L. Nyhan, Department of Paediatrics, University of California, San Diego, USA for their support. This study was supported by a grant provided by Sheikh Rafiq A1 Hariri (no. 85-0030). We would like to thank Mrs Ann Baez Giangreco and Mrs Lilia Fernandez for manuscript preparation.

J. Inher. Metab. Dis. 14 (1991)

188

Ozand et al.

REFERENCES

Dasouki, M., Buchanan, D., Mercer, N., Gibson, K. M. and Thoene, J. 3-Hydroxy-3methylglutaric aciduria: response to carinitine therapy and fat and leucine restriction. J. Inher. Metabl. Dis. i0 (1987) t42-146 Duran, M., Schutgens, R. B. H., Ketel, A., Heymans, H., Berntessen, W. J., Ketting, D. and Wadman, S. K. 3-Hydroxy-3-methylglutaryl coenzyme A lyase deficiency: postnatal management following prenatal diagnosis by analysis of maternal urine. J. Pediatr. 95 (1979) 1004-1007 Faull, K., Botton, P., Halpern, B., Hammond, J., Danks, D. M., Hahnet, R. R., Wilkinson, S. P., Wysocki, S. J. and Masters, P. L. Patient with defect in leucine metabolism. N. Engl. J. Med. 294 (1976) 1013 Gibson, K. M., Breuer, J., Kaiser, K., Nyhan, W. L., McCoy, E. E., Ferreira, P., Greene, C. L., Blitzer, M. G., Shapira, E., Reverte, F., Conde, C., Bagnell, P. and Cole, D. E. C. 3Hydroxy-3-methylglutarylcoenzyme A lyase deficiency: report of five new patients. J. Inher. Metab. Dis. 11 (1988a) 76-87 Gibson, K. M., Breuer, J. and Nyhan, W. L. 3-Hydroxy-3-methylglutaryl coenzyme A lyase deficiency: review of 18 reported patients. Eur. J. Pediatr. 148 (1988b) 180-186 Gibson, K. M., Lee, C. L., Kamail, V., Johnston, K., Beaudet, A. L., Craigen, W. J., Schwartz, R., Tsai, M. Y. and Tuchman, M. 3-Hydroxy-3-methylglutaryl coenzyme A lyase deficiency as detected by radiochemical assay in cell extracts by thin layer chromatography and identification of two new cases. Clin. Chem. 36 (1990) 297-303 Hoffmann, G., Aramaki, S., Blum-Hoffman, E., Nyhan, W. L. and Sweetman, L. Quantitative analysis for organic acids in biological samples: batch isolation followed by gas chromatographic-mass spectrometric analysis. Clin. Chem. 35 (1989) 587-595 Leonard, J. V., Seakins, J. W. T. and Griffin, N. K. Beta-hydroxy-beta-methylglutaric acidura presenting as Reye's syndrome. Lancet 1 (1979) 680 Lisson, G., Leupold, D., Bechinger, D. and Wallesch, C. CT findings in a case of deficiency of 3-hydroxy-3-methylglutaryl CoA lyase. Neuroradiology 22 (1981) 99-101 Marklova, E., Verner, P., Pehal, F., Bratova, M. and Polak, J. Case report: A new case of 3hydroxy-3-methylglutaryl coenzyme A lyase deficiency. J. Inher. Metab. Dis. 10 (1987) 399 Norman, E. J., Denton, M. D. and Berry, H. K. Gas chromatographic/mass spectrometric detection of 3-hydroxy-3-methylglutaryl CoA lyase deficiency in double first cousins. CIin. Chem. 28 (1982) 137-140 Reed, W. D., Baab, P. J., Hawkins, R. L. and Ozand, P. T. A double isotope method for the measurement of ketone-body turnover in the rat. Biochem. J. 219 (1984) 15-24 Robinson, B. H., Oei, J., Sherwood, G. W., Slyper, A. H., Heininger, J. and Mamer, O. Hydroxymethylglutaryl CoA lyase deficiency: features resembling Reye syndrome. Neurology 30 (1980) 714-718 Schutgens, R. B. H., Heymans, H., Ketal, A. and Veder, H. A. Lethal hypoglycemia in a child with a deficiency of 3-hydroxy-3-methylglutaryt coenzyme A lyase. J. Pediatr. 94 (1979) 89-91 Sovik, O., Sweetman, L., Gibson, K. M. and Nyhan, W. L. Genetic complementation analysis of 3-hydroxy-3-methy!glutaryl coenzyme A lyase deficiency in cultured fibroblasts. Am. J. Hum. Genet. 36 (i984) 791-801 Wysocki, S. J. and Haehnel, R. 3-Hydroxy-3-methylglutaryl coenzyme A lyase deficiency: a review. J. Iner. Metab. Dis. 9 (1986) 225-233 Zoghbi, H. Y., Spence, J. E., Beaudet, A. L., O'Brien, W. E., Goodman, C. J. and Gibson, K. M. Atypical presentation and neuropathological studies in 3-hydroxy-3-methylglutarylCoA lyase deficiency. Ann. Neurol. 20 (1986) 367-369

J. lnher. Metab. Dis. 14 (1991)