Abstract
Here, we report an individual, eventually diagnosed with HMG-CoA synthase deficiency, who presented with a cyclic vomiting phenotype. HMG-CoA synthase deficiency is a rare disorder affecting ketone body synthesis in which affected individuals typically present at a young age with hypoketotic hypoglycemia, lethargy, encephalopathy, and hepatomegaly, usually triggered by catabolism (e.g., infection or prolonged fasting). This individual presented with recurrent episodes of vomiting and lethargy, often associated with hypoglycemia or hyperglycemia, at 3 years of age. Metabolic labs revealed nonspecific abnormalities in her urine organic acids (showing mild elevation of dicarboxylic acids with relatively low excretion of ketones) and a normal acylcarnitine profile. Given her clinical presentation, as well as a normal upper gastrointestinal series, esophagogastroduodenoscopy with biopsies, and abdominal ultrasound, she was diagnosed with cyclic vomiting syndrome at 3 years of age. Molecular testing completed at 7 years of age revealed a previously reported pathogenic sequence variant (c.1016+1G>A) and a novel likely pathogenic deletion (1.57 kB deletion, including exon 1) within HMGCS2 consistent with HMG-CoA synthase deficiency. This individual’s presentation, mimicking cyclic vomiting syndrome, widens the clinical spectrum of HMG-CoA synthase deficiency. In addition, this case highlights the importance of molecular genetic testing in such presentations, as this rare disorder lacks specific metabolic markers.
Keywords: HMG-CoA synthase, mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase deficiency, cyclic vomiting syndrome, metabolic mimics of cyclic vomiting
Introduction
Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase deficiency, also known as HMG-CoA synthase deficiency, is a very rare inborn error of metabolism affecting ketone body synthesis. The disorder is autosomal recessive and caused by biallelic pathogenic variants in HMGCS2, which encodes mitochondrial HMG-CoA synthase—an enzyme that catalyzes the rate-limiting step of ketogenesis.1 -3 Deficiency of this enzyme leads to impaired ability to generate ketones, and thus maintain normal glucose levels during fasting. To date, this disorder has been reported in fewer than 80 individuals worldwide. 4 Affected individuals typically present at a young age, often in infancy, with vomiting, lethargy, respiratory distress, encephalopathy, and hepatomegaly. Episodes are usually triggered by catabolism (e.g., infection, stress, or prolonged fasting). Nonspecific laboratory abnormalities include hypoketotic hypoglycemia, metabolic acidosis, elevated transaminases, and dicarboxylic aciduria. Less common abnormalities reported include seizures, hepatic steatosis, hypertriglyceridemia, hypocalcemia, hypophosphatemia, hyperuricemia, hyperammonemia, and elevated lactate.4 -7 The majority of patients have presented with metabolic crises prior to 3 years of age; however, asymptomatic individuals have been reported which indicates that the full phenotypic spectrum of this condition is still evolving.6,8
Given the importance of ketones as a fuel source, inability to generate sufficient energy during times of catabolic stress can lead to devastating complications including life-threatening respiratory distress, coma, and death. Treatment relies on prompt initiation of intravenous (IV) dextrose-containing fluids with age-appropriate electrolytes during periods of catabolism to avoid metabolic decompensation. As the condition was first identified fewer than 20 years ago, long-term outcome data are limited. However, prompt diagnosis and initiation of supportive and symptomatic care has led the majority of patients to recover from their metabolic crises without any developmental delay or neurologic sequelae.
Here, we describe an individual presenting with cyclic vomiting who was diagnosed with HMG-CoA synthase deficiency at 7 years of life, further expanding the phenotypic spectrum of the disorder.
Case Presentation
A female infant was born full-term to nonconsanguineous parents after an uncomplicated pregnancy and delivery. Birth parameters were appropriate, but she had difficulty gaining weight, dropping to the 4%ile for weight by 2 months of age. As part of her failure-to-thrive (FTT) evaluation, she was seen by Cardiology and had a reassuring electrocardiogram and echocardiogram. She continued to be monitored by her Pediatrician for FTT of unknown etiology, which improved by 2 years of age. She was described by her parents to be a “spitty baby.”
In her second year of life, the patient began to have recurrent episodes of fatigue, non-bloody-non-bilious vomiting, nausea, anorexia, and abdominal pain. Episodes lasted 12 to 24 hours, typically occurred every 4-to-6 weeks, and she returned to her normal state of health in between episodes. Parents reported that episodes tended to occur when she was tired or “stressed,” but noted that positive emotion seemed to trigger her episodes as well. Symptoms typically improved with time and sleep. The patient completed a trial of ranitidine for suspected gastroesophageal reflux disease, without improvement in symptoms.
At 3 years of age, the patient presented to medical care during one of her episodes and was reported to be lethargic. Laboratory evaluation was notable for a mild leukocytosis (WBC 20.9, reference range 6-16 K/uL) and a urinalysis with 2+ ketones. Her blood glucose was 78 mg/dL (reference range unavailable) prior to initiation of IV fluids. She was admitted to the hospital for monitoring; IV fluids were titrated as oral intake improved. A renal ultrasound was normal. Given her clinical presentation (prodrome of decreased energy, followed by recurrent episodes of vomiting with return to baseline in between episodes), the Gastroenterology (GI) team suspected cyclic vomiting syndrome, but recommended continued outpatient evaluation and prompt evaluation, including metabolic labs prior to initiation of IV fluids, during her next episode.
One month later, the patient re-presented to the hospital with vomiting and decreased energy. Upon arrival, her blood glucose was low at 53 (70—99 mg/dL), with normal electrolytes, lactate, insulin, and cortisol. She was admitted for further monitoring, IV hydration, and a cyclic vomiting workup. Her glucose corrected with juice, IV fluids, and a regular diet. An upper gastrointestinal series (UGI) was limited, but sufficient to determine no malrotation. An esophagogastroduodenoscopy (EGD) with biopsies was normal. Two days later, she returned to baseline, tolerating oral intake and was discharged home. Metabolic labs resulted shortly after discharge—her urine organic acids, obtained after initiation of IV fluids, contained abnormalities that can be seen in long chain fatty acid oxidation disorders, but the subsequent plasma acylcarnitine profile resulted normal (Table 1).
Table 1.
Summary of Biochemical Lab Evaluation in This Patient.
Age (clinical status) | Plasma amino acids | Acylcarnitine profile | Carnitine levels | Urine organic acids |
---|---|---|---|---|
3 years (sick, after initiation of IV fluids) a | Low concentration of several amino acids suggesting low protein intake | In this sample the excretion of dicarboxylic acids, saturated and unsaturated, and long chain (C10-C14) 3-hydroxydicarboxylic acids, saturated and unsaturated, was elevated, with relatively low excretion of ketones (3-OH-butyric acid). This pattern can be seen in Long Chain 3-Hydroxy-Acyl-CoA/Trifunctional Protein (LCHAD/TFP) deficiency, an inherited disorder of fatty acid oxidation. Would evaluate plasma acylcarnitine profile and plasma free 3-hydroxy-fatty acid profile to confirm this finding. Would also consider DNA testing. Genetic and metabolic evaluations are also recommended since this is a potentially life-threatening disease. | ||
3 years (well) b | Normal | |||
7 years (well) b | This pattern of amino acids does not suggest a specific inherited metabolic defect | Normal | Normal (free 38 umol/L, total 43 umol/L) | Elevated acetoacetate is seen in ketosis (acetoacetate: 3 [ref 0-1 mmol/mol creatinine]) |
7 years (sick, prior to initiation of IV fluids) b | This pattern of amino acids does not suggest a specific inherited metabolic defect | This pattern of mildly elevated acylcarnitines is not consistent with a specific disorder. | Normal (free 21 umol/L, total 34 umol/L) | Elevated dicarboxylic and 3-hydroxy-dicarboxylic acids are seen in defects of fatty acid oxidation, in ketosis and in patients on diets containing medium chain triglycerides. In particular, elevations of unsaturated dicarboxylic species can be found in long chain fatty acid oxidation disorders. The patient’s recent plasma acylcarnitine profile was not suggestive of any of these disorders. If clinically indicated, consider repeat urine organic acid and plasma acylcarnitine analyses |
8 years (well) c | Normal | Normal (free 32 umol/L, total 36 umol/L) | Pattern of organic acid excretion appears normal in this sample. | |
8 years (sick, collected ~ 1 hour after IV fluids) a | In this sample the excretion of medium and long chain dicarboxylic acids, hydroxylated and nonhydroxylated, was increased. In addition, peaks with a mass spectrum corresponding to 5-hydroxyhex-2-enoic and 3-hydroxyhex-4-enoic acids and 4-hydroxy-6-methyl-2-pyrone were identified. The presence of these analytes has been described in patients with 3-hydroxy-3-methylglutaryl-CoA synthase deficiency, an inherited disorder of ketone synthesis. Would repeat this study and evaluate plasma acylcarnitines to confirm or exclude this possibility. Genetic and metabolic evaluations are recommended. | |||
8 years (sick, prior to initiation of IV fluids) a | Low-normal (free 14 umol/L, total 33) | 1) Mild ketonuria suggesting catabolic state. 2) In this sample the excretion of medium and long chain dicarboxylic acids, hydroxylated and nonhydroxylated, was increased. In addition, peaks with a mass spectrum corresponding to 5-hydroxyhex-2-enoic and 3-hydroxyhex-4-enoic acids and 4-hydroxy-6-methyl-2-pyrone were identified. The presence of these analytes has been described in patients with 3-hydroxy-3-methylglutaryl-CoA synthase deficiency, an inherited disorder of ketone synthesis. Would evaluate plasma acylcarnitines to confirm or exclude this possibility. Genetic and metabolic evaluations are recommended. |
||
8 years (well) c | This pattern of organic acids does not suggest a specific inherited metabolic defect |
Labs completed at Associated Regional and University Pathologists (ARUP) in Salt Lake City, UT. Reference ranges for total carnitine 31-78 umol/L, free carnitine 22-63 umol/L, esterified carnitine 3-38 umol/L, and ester/free ratio carnitine 0.1-0.9.
Labs completed at Quest Diagnostics Biochemical Genetics Laboratory in San Juan Capistrano, CA. Reference ranges for total carnitine 28-59 umol/L, free carnitine 19-51 umol/L, carnitine esters 3-16 umol/L, esterified/free ratio 0.09-0.49.
Labs completed at Stanford Biochemical Genetics Laboratory in Palo Alto, CA. Reference ranges for total carnitine 20.0-71.0 umol/L, free carnitine 18.0-58.0 umol/L, acyl/free ratio 0.1-0.4.
Three months later, the patient presented to the emergency department with recurrent vomiting. However, this presentation was notable for hyperglycemia (294 on a home glucometer) and upon arrival to the hospital, was elevated to 208 mg/dL (ref 70-100). A full electrolyte panel 1 hour later showed improved blood glucose of 150 mg/dL (ref 70-100) without intervention and normal electrolytes. Given her clinical presentation and normal abdominal ultrasound, UGI, EGD, and metabolic work up, the patient was diagnosed with cyclic vomiting syndrome by GI in the outpatient setting and offered cyproheptadine, which was deferred by the family. The patient was referred to Endocrinology given her documented hypo- and hyperglycemia; repeat laboratory evaluation, including HbA1c, glutamic acid decarboxylase antibody, IgG islet cell cytoplasmic antibody, and insulin antibody, were unremarkable.
In the subsequent years, the patient presented to the hospital multiple times with vomiting and altered glucose regulation. In 1 presentation, her home blood glucose was 60 when the vomiting episode began; however, 1 hour after tolerating oral nutrition, her glucose had risen to 200, and then 410, on the home glucometer—prompting family to take the patient to the hospital. Upon arrival, her glucose was 253 mg/dL (ref 70-100) and downtrended overnight to 218, 133, 83, 120, and 102 with no intervention. Endocrinology was reassured with her evaluation to-date, including a normal cortisol, HbA1c, and negative testing for islet cell autoantibodies.
At 7 years of age, the patient was evaluated by Metabolic Genetics for her history of vomiting, hypoglycemia, and occasional refractory hyperglycemia. Recommendations included repeat metabolic labs (drawn during well and sick states), mitochondrial sequencing, and a hypoglycemia gene panel. Urine organic acids, during the well-state, revealed a mildly elevated acetoacetate (3 mcg/mL, with ref range 0-1) as can be seen in ketosis (Table 1). Labs were repeated 1 month later, during an episode of fatigue, emesis, and hypoglycemia (home glucose 50). Her urinalysis had 1+ ketones and urine organic acids revealed elevated dicarboxylic and 2-hydroxydicarboxylic acids (Table 1). Molecular testing, performed at Prevention Genetics, a CLIA accredited clinical DNA testing laboratory, revealed a previously reported, pathogenic, maternally inherited, sequence variant, NM_005518.3(HMGCS2): c.1016+1G>A, and a novel 1.57 kB, likely pathogenic, paternally inherited, deletion (chr1:120,310,549-120,312,114, GRCh37/hg19) in HMGCS2. These findings, in conjunction with her clinical presentation, were supportive of a diagnosis of HMG-CoA synthase deficiency.
The patient is now closely followed by a biochemical genetics service. Given her glucose fluctuations, which can range from 32 mg/dL to greater than 500 mg/dL within minutes, she also continues to follow with Endocrinology. Her hyperglycemia seems primarily reactive and postprandial and even an appropriate oral treatment for hypoglycemia with 8-10 grams of rapid acting carbohydrates can lead to a blood sugar well over 400 mg/dL. She is currently monitoring glucoses at home with a continuous glucose monitoring system (Libre 3) with confirmation via glucometer if needed. She is now 10 years old and continues to be developmentally appropriate.
Aside from the individual treatment implications, identification of a molecular diagnosis for this patient has led to informed recurrence risk and prenatal counseling for her parents who are both confirmed carriers. Targeted testing was performed for patient’s younger sister given her history of an episode of borderline hypoglycemia and emesis to ensure this diagnosis was not missed. She was found to be a carrier of the pathogenic maternally inherited sequence variant. The sister recovered from this brief illness well and can elect further prenatal counseling in the future along with other family members.
Discussion
We have reported an individual who presented with recurrent episodes of vomiting and lethargy who was diagnosed with cyclic vomiting syndrome at 3 years of age. She was ultimately diagnosed with mitochondrial HMG-CoA synthase deficiency at 7 years of age. This patient’s presentation, mimicking cyclic vomiting syndrome and with altered glucose regulation, highlights the wide phenotypic spectrum of this condition. In addition, we highlight how the lack of reliable biochemical markers for this rare disorder of ketone body synthesis can lead to delays in diagnosis.
The pathophysiology of cyclic vomiting syndrome is not well understood. The condition is characterized by recurrent, episodic vomiting with a return to baseline in between episodes. However, other etiologies of acute and chronic vomiting must be excluded prior to diagnosis. The North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN) has published consensus guidelines in which all of the following criteria must be met prior to making a diagnosis of cyclic vomiting syndrome (CVS) (Table 2). 9 To aid in the evaluation and exclusion of other organic etiologies of recurrent vomiting, NASPGHAN also outlines a standard diagnostic approach. 9 For example, if attacks are precipitated by fasting, intercurrent illness, or high protein meals, then additional labwork is recommended prior to IV fluids including electrolytes, glucose, urine ketones, lactate, ammonia, plasma amino acids, urine organic acids, plasma carnitines, and acylcarnitine profile. 9
Table 2.
Consensus Diagnosis Criteria of Cyclic Vomiting.
1. At least 5 attacks in any interval, or a minimum of 3 attacks during a 6-mo period |
2. Episodic attacks of intense nausea and vomiting lasting 1 hour—10 days and occurring at least 1 wk apart |
3. Stereotypical pattern and symptoms in the individual patient |
4. Vomiting during attacks occurs at least 4 times per hour for at least 1 hour |
5. Return to baseline health b/w episodes |
6. Not attributed to another disorder |
This patient received a comprehensive evaluation in accordance with these guidelines. She had a normal UGI series, esophagogastroduodenoscopy with biopsies, and abdominal ultrasound which provided Gastroenterology reassurance against an acute or obstructive process. Given her glucose abnormalities, she was connected with Endocrinology, but her evaluation was unremarkable, including a normal cortisol, hemoglobin A1c, and negative islet autoantibodies. Her initial metabolic labs revealed nonspecific abnormalities in her urine organic acids, showing mild elevation of dicarboxylic acids with relatively low excretion of ketones. These abnormalities can be seen in fatty acid oxidation disorders, but her acylcarnitine profile was normal. Given she was presumed to have met the standard guidelines, she was diagnosed with CVS at 3 years of age (Table 2). In retrospect, this patient did not meet the sixth CVS diagnostic criterion (“not attributed to another disorder”), as she was ultimately diagnosed with an underlying metabolic condition; however, this was not appreciated at the time of her presentation given the lack of reliable biochemical markers.
In their review of metabolic conditions that present similarly to cyclic vomiting, Gelfand et al 10 acknowledged the difficulty clinicians experience when trying to determine whether every child with cyclic vomiting should receive a metabolic work-up and how extensive such testing should be. Many inborn errors of metabolism have biochemical lab abnormalities that are readily identified on standard biochemical labs and clinical evaluation. 10 For instance, a fatty acid oxidation disorder would be considered in an individual with recurrent vomiting episodes and hypoketotic hypoglycemia, metabolic acidosis, elevated transaminases, elevated CK, and elevated lactate. As this patient demonstrates, normal results or nonspecific abnormalities on standard biochemical labs do not rule out the presence of an underlying metabolic disorder.
Unlike fatty acid oxidation disorders, which typically have characteristic abnormalities, HMG-CoA synthase deficiency does not have reliable biochemical markers on plasma acylcarnitine analysis. Dicarboxylic aciduria is supportive, but not specific as this can also be present in fatty acid oxidation disorders.6,11 -13 Individuals are typically hypoketotic, as was noted on this patient’s initial urine organic acids at 3 years of age (Table 1); however, at other points, this patient had ketones in her urinalysis—highlighting the variability of this feature. In addition, some authors have proposed that increased urine 4-hydroxy-6-methyl-2-pyrone (4-HMP) may be associated with the condition; however, this analyte was not detected in many other reported patients and only noted in 2 of the 7 urine organic acid analyses from this patient (Table 1).4,12,13 Given specific biochemical markers for HMG-CoA synthase deficiency have yet to be identified, if clinicians have a high clinical suspicion for this disorder and/or are conducting a broad evaluation of individuals presenting with a cyclic vomiting phenotype, we recommend proceeding with molecular testing.
This patient’s late-onset presentation, accompanied by glucose dysregulation, widens the phenotypic spectrum of the disorder. While she has presented with hypoglycemia several times, she was euglycemic during her initial hospitalization and actually hyperglycemic after oral nutrition in subsequent admissions. Although atypical, there have been 3 patients with HMG-CoA synthase deficiency who did not present with hypoglycemia—emphasizing the importance of considering this condition even in euglycemic patients.5,13,14 Two patients with HMG-CoA synthase deficiency have been reported with refractory hyperglycemia though the pathophysiology leading to this glucose dysregulation remains unclear.6,7
Conclusion
We have described an individual who presented with a cyclic vomiting phenotype at 3 years of age who was ultimately diagnosed with HMG-CoA synthase deficiency at 7 years of age. Her clinical presentation widens the phenotypic spectrum of the disorder and highlights how additional evaluation is needed to understand the glucose dysregulation observed in this condition. Unlike several other metabolic disorders that can mimic cyclic vomiting syndrome, HMG-CoA synthase deficiency does not have reliable markers on routine biochemical labs. Thus, if there is a high clinical suspicion or if hypoglycemia is documented, we recommend evaluating for this life-threatening, but treatable, condition with molecular testing.
Acknowledgments
We thank the patient and her family for their participation. We thank Denise Z Salazar, PhD, FACMGG, and Quest Diagnostics lab for their discussion of the biochemical labs of this patient.
Footnotes
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethics Approval: Our institution does not require IRB approval for reporting individual cases.
Informed Consent: Verbal informed consent was obtained from a legally authorized representative (parent) and anonymized health information to be published in this article.
ORCID iD: Annie D. Niehaus
https://orcid.org/0000-0001-9425-3073
References
- 1. Thompson GN, Hsu BY, Pitt JJ, et al. Fasting hypoketotic coma in a child with deficiency of mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase. N Engl J Med. 1997;337(17):1203-1207. [DOI] [PubMed] [Google Scholar]
- 2. Fukao T, Mitchell G, Sass JO, et al. Ketone body metabolism and its defects. J Inherit Metab Dis. 2014;37:541-551. [DOI] [PubMed] [Google Scholar]
- 3. Newman JC, Verdin E. Ketone bodies as signaling metabolites. Trends Endocrinol Metab. 2014;25(1):42-52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Wu S, Shen L, Chen Q, et al. Clinical, biochemical, molecular, and outcome features of mitochondrial 3-hydroxy-3-methylglutaryl-coa synthase deficiency in 10 Chinese patients. Front Genet. 2022;12:816779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Conlon TA, Fitzsimons PE, Borovickova I, et al. Hypoglycemia is not a defining feature of metabolic crisis in mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase deficiency: further evidence of specific biochemical markers which may aid diagnosis. JIMD Rep. 2020;55(1):26-31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kılıç M, Dorum S, Topak A, Yazıcı MU, Ezgu FS, Coskun T. Expanding the clinical spectrum of mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase deficiency with Turkish cases harboring novel HMGCS2 gene mutations and literature review. Am J Med Genet A. 2020;182(7):1608-1614. [DOI] [PubMed] [Google Scholar]
- 7. Rojnueangnit K, Maneechai P, Thaweekul P, et al. Expanding phenotypic and mutational spectra of mitochondrial HMG-CoA synthase deficiency. Eur J Med Genet. 2020;63(12):104086. [DOI] [PubMed] [Google Scholar]
- 8. Zhang P, Hu X, Guo R, et al. Novel HMGCS2 pathogenic variants in a Chinese family with mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase deficiency. Pediatr Investig. 2019;3(2):86-90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Li BU, Lefevre F, Chelimsky GG, et al. North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition consensus statement on the diagnosis and management of cyclic vomiting syndrome. J Pediatr Gastroenterol Nutr. 2008;47(3):379-393. [DOI] [PubMed] [Google Scholar]
- 10. Gelfand AA, Gallagher RC. Cyclic vomiting syndrome versus inborn errors of metabolism: a review with clinical recommendations. Headache. 2016;56(1):215-221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Aledo R, Zschocke J, Pié J, et al. Genetic basis of mitochondrial HMG-CoA synthase deficiency. Hum Genet. 2001;109(1):19-23. [DOI] [PubMed] [Google Scholar]
- 12. Pitt JJ, Peters H, Boneh A, et al. Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase deficiency: urinary organic acid profiles and expanded spectrum of mutations. J Inherit Metab Dis. 2015;38(3):459-466. [DOI] [PubMed] [Google Scholar]
- 13. Wang Q, Yang YL, Liu M, et al. Clinical, biochemical, molecular and therapeutic characteristics of four new patients of mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase deficiency. Clin Chim Acta. 2020;509:83-90. [DOI] [PubMed] [Google Scholar]
- 14. Lee T, Takami Y, Yamada K, et al. A Japanese case of mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase deficiency who presented with severe metabolic acidosis and fatty liver without hypoglycemia. JIMD Rep. 2019;48(1):19-25. [DOI] [PMC free article] [PubMed] [Google Scholar]