A male patient came to our attention in 1981, when he was 3 years and 4 months old, due to acute cholestatic jaundice and hypertransaminasemia (aspartate aminotransferase 10X, alanine aminotransferase 10X, γ-glutamyltransferase 87 U/L). He had recently been diagnosed as having glycogen storage disease type I (GSD I) on the basis of clinical, laboratory, and histological findings.
When we first saw him, the clinical examination revealed a doll-like face, an enlarged abdomen with massive hepatosplenomegaly (the liver was palpable 3 cm beyond the transverse umbilical line), and his height was below the third percentile. Serum analysis showed mild fasting hypoglycemia (values of 40–45 mg/dL, without increase in glycemia after glucagon load following an overnight fast), hyperlipidemia (triglycerides 251 mg/dL, total cholesterol 268 mg/dL), mild hyperuricemia (5.8 mg/dL), hyperlactacidemia (3.2 mmol/L), and low platelet aggregation. Alkaline phosphatase, aldolase, creatinphosphokinase, and the electrocardiography were normal. An ultrasound examination of the abdomen confirmed hepatosplenomegaly with diffuse hyperechogenicity of the liver; no kidney abnormalities were present.
The enzymatic determination of glucose-6-phosphatase was not performed, relying only on the liver biopsy that showed massive glycogen storage. Therefore, the hypertransaminasemia and jaundice were interpreted as a concomitant acute viral hepatitis. Dietary management was begun with frequent meals, nocturnal enteral feeding by nasogastric tube to prevent hypoglycemia, and restricting fructose and galactose intake to avoid acidosis, and a rapid improvement in the clinical findings and serum values occurred.
During the follow-up, hypertransaminasemia (2X) persisted, which prompted a repeat liver biopsy; however, the findings were similar to the previous biopsy and the presence of limit dextrin was not noted. Then the patient no longer returned for any follow-up visit and all contact was lost.
The patient returned to our clinic at the age of 31, when he attended for genetic counseling, because he was concerned that his son could carry the same disease. At that time, the patient was completely asymptomatic, no organomegaly was present, and the abdominal ultrasound was normal, although despite lack of compliance with any recommended dietary measures. A mild hypertransaminasemia (aspartate aminotransferase 1 5X, alanine aminotransferase 2X) and hyperlipidemia (triglycerides 170 mg/dL, total cholesterol 263 mg/dL) were still present, suggesting presence of residual liver disease. Considering his present clinical picture, the diagnosis was reevaluated. The main differences between GSD I and III are shown in Table 1.
GSD I and III are both autosomal recessive diseases (1). GSD I is caused by the absence or deficiency of glucoce-6-phosphatase activity in the liver, kidney, and intestinal mucosa. Two subtypes can be identified: type Ia, caused by a defect of the glucose-6-phosphatase enzyme; and type Ib, caused by a defect of the translocase that transports glucose-6-phosphate across the microsomal membrane. Both defects lead to inadequate hepatic conversion of glucose-6-phosphate to glucose through normal glycogenolysis and gluconeogenesis, making affected individuals susceptible to fasting hypoglycemia not responsive to glucagon infusion (2). Accumulation of glycogen and fat is seen in the liver and kidneys, and is responsible for hepatomegaly and renomegaly. Typically, untreated infants presented at age 3 to 4 months with hepatomegaly, hypoglycemic seizures, or both.
Despite marked hepatomegaly, the liver transaminase levels are usually normal or only slightly elevated. GSD Ib has additional features of recurrent bacterial infections, oral and intestinal mucosal ulcers, and inflammatory bowel disease caused by impaired neutrophil and monocyte function as well as chronic neutropenia after the first few years of life. By the second or third decade of life, most of the patients with GSD I exhibit hepatic adenomas that can hemorrhage, and malignant transformation of the adenomas has been described occasionally; renal disease too (proteinuria, hypertension, renal stones) occurs in this period of life.
GSD III is caused by a deficiency of glycogen debranching enzyme activity, which, together with phopshorylase, is responsible for a complete degradation of glycogen (3). As a result of defective activity, an abnormal glycogen with short outer branch chains (limit dextrin) accumulates and may be seen on liver biopsy, causing hepatomegaly and hypoglycemia as in GSD I. Moreover, unlike GSD I, muscle and cardiac involvement can be present in 85% of cases. Cardiomyopathy can become severe after the third or fourth decade of life (subtype IIIa) (4). In infancy and early childhood, the disease may be indistinguishable from GSD I because hepatomegaly, hypoglycemia, hyperlipidemia, and growth retardation are common; however, in contrast to GSD I, after puberty, liver size and hypoglycemia usually improve significantly, although the liver disease may progress to fibrosis and cirrhosis necessitating liver transplantation in some cases. Moreover, although elevation of liver transaminases and fasting ketosis can be prominent, blood lactate and uric acid concentrations are usually normal. The administration of glucagon 2 hours after a carbohydrate meal provokes a normal increase in blood glucose, but after an overnight fast, it may provoke no change in blood glycemia.
Our patient presented with platelet dysfunction, mild hyperlipidemia and hyperuricemia, hyperlactacidemia, and hypoglycemia not responsive to glucagon infusion as characteristics of GSD I. When the patient returned at the age of 31, the typical features of GSD III were prominent and included persistent transaminasemia, normal liver size, and milder biochemical abnormalities then expected in GSD I.
Because of the changed clinical picture, the diagnosis of GSD I was reevaluated. The diagnosis of type III glycogenosis was then proposed and we decided to perform mutation analysis, which confirmed a mutation of the AGL gene c.664 + 3A>G (IVS6 + 3A>G) in a homozygous state. This mutation, as demonstrated by Lucchiari et al (5,6), affects splicing of exon 6. Indeed, this nucleotide alteration is therefore capable of abolishing the canonical splice consensus site at the 5′ terminus of intron 6 and is predicted to result in an inframe deletion of exon 6. The same mutation has been described previously in 4 cases. Although no muscle disease was found in the described cases, muscle involvement is expected to manifest in the adulthood, as the mutation is likely to cause GSD IIIa. Cardiac disease was present in 2 of 4 reported cases. In our patient, neither muscle weakness nor cardiac problems were detected. Interestingly, all of the patients described with the c.664 + 3A>G mutation, including the case presented here, originate from southern Italy, thus suggesting a possible founder effect.
The present case underlines the difficulty in making accurate diagnosis of GSD I or III using purely clinical and basic biochemistry approaches. As the clinical management and prognosis for these conditions differ significantly, it is important to apply modern laboratory techniques to achieve robust diagnosis. This is the fifth reported case with c.664 + 3A>G mutation. Because all of the patients originate from the same area of Italy, there is likely to be a founder effect, and in all of the future cases with suspected GSD III, this mutation should be initially looked for.
1. Wolfsdorf JI, Weinstein DA. Glycogen storage diseases. Rev Endocr Metab Disord
2. Janecke AR, Mayatepek E, Utermann G. Molecular genetics of type 1 glycogen storage disease. Mol Genet Metab
3. Lucchiari S, Santoro D, Pagliarani S, et al. Clinical, biochemical and genetic features of glycogen debranching enzyme deficiency. Acta Myol
4. Arad M, Maron BJ, Gorham JM, et al. Glycogen storage disease presenting as hypertrophic cardiomyopathy. N Engl J Med
5. Lucchiari S, Donati MA, Parini R, et al. Molecular characterisation of GSD III subjects and identification of six novel mutations in AGL. Hum Mutat
6. Lucchiari S, Pagliarani S, Salani S, et al. Hepatic and neuromuscular forms of glycogenosis type III: nine mutations in AGL. Hum Mutat