Hepatic Glycogenoses Among Children—Clinical and Biochemical Characterization: Single-Center Study

Abstract

Background

Glycogen storage disease (GSD) is typified by early morning seizures. Absence of this results in delayed diagnosis, especially the non-GSD 1 group. Data are limited to few patients with unclear outcome.

Objectives

1. Study the common presentation and types of GSD. 2. Study the clinical and biochemical outcome. 3. Review genetic mutations.

Methods

Observational study from May 2016–April 2019 at metabolic clinic at our center.

Results

Total of 30 patients were diagnosed with GSD. Ten were excluded—Fanconi-Bickel (3) and <4 months follow-up (7). Data were analyzed for 20 patients (16 males). Mean age at presentation was 4.3 yrs. All had hepatomegaly, 90% had short stature, and 40% had early morning seizures. Mean follow-up was 22 months. There was a statistically significant improvement in metabolic parameters on treatment (mean)—fasting glucose from 50.4 to 79.5 mg/dl, SGPT from 416 to 199 U/L. Lipid profile showed reduction in triglycerides (318–225 mg/dl) but minimal increase in cholesterol (178–188 mg/dl). Mean weight centile improved from 14.1 to 20.3 and height centile from 2.3 to 7.9. Genetic testing confirmed types VI (3), III (3), IXa (1), IXc (1), and Ia (1). Liver biopsy confirmed GSD in 15/20. All were managed with uncooked corn starch. In addition, omega-3 fatty acid was used in 8/20 and high protein diet in 2 with GSD type III.

Conclusion

Awareness of GSD needs to improve among pediatricians and hepatologists. The most common symptoms are asymptomatic hepatomegaly and short stature. Dietary therapy with uncooked corn starch remains mainstay of treatment. Mixed hyperlipidemia is difficult to control despite good metabolic improvement. Role of omega-3 fatty acid needs to be explored further. Genetic mutation analysis can assist with tailoring treatment and should get precedence over liver biopsy.

Glycogen storage disorders (GSDs) represent a group of rare genetic disorders caused by deficiency of enzymes or transport protein involved in glycogenolysis pathway, which results in storage of glycogen.¹ These disorders typically affect liver and muscle—cardiac and skeletal. Among the many types of GSDs, hepatic involvement characterizes the types I, III, VI, and IX. There are physiological differences among the various types with Type I being the most severe as both gluconeogenesis and glycogenolysis are affected. Hence significant hypoglycemia with hyperlactatemia and hyperuricemia are seen with Type I. Type Ib in addition has neutropenia. Type III may affect both liver and muscle with liver involvement generally improving with age. Type VI and IX are heterogenous disorders with absence of hyperlactatemia and hyperuricemia.

Glycogen storage disorders arerare, and few Indian studies have been published and none outlining clinical and biochemical profile at follow-up.^{2, 3, 4, 5}

The common understanding among clinicians is that if there is no history of early morning seizures, GSD is an unlikely diagnosis. In addition, many patients are advised liver transplant at the outset, rather than initiating dietary regime and monitoring progress before deciding on need for liver transplant.

Hence, a comprehensive Indian study on clinical characteristics, biochemical parameters, and follow-up data are very much needed. This cohort study was thus chosen to address this need.

Study methodology

This is a single center cohort study where all pediatric (0–18) patients seen in metabolic clinic from May 2016–April 2019 were included. Data were collected from the medical records and online patient result portal. Data were entered into Microsoft Excel and analyzed. Wilcoxon-sign rank test was used to establish statistical significance between the biochemical parameters at onset and follow-up, with P value < 0.05 considered significant. All the growth parameters were calculated based on World Health Organization growth chart centiles. The study was approved by the Institutional review board for Research.

Methods

First Visit

The standard of practice at our center involves all subjects with suspected GSD to undergo fasting blood test at first visit, including glucose, lactate, lipid profile, liver function test, uric acid, Creatine phosphokinase (CPK), complete blood count, 25 hydroxy vit D, and lactate. All of them also get an ultrasound abdomen done. Genetic analyses is offered to all at first visit and is done depending on the affordability of the family. Once preliminary reports are in favor of GSD, they are sent to the dietician. All are put on low fat, low Glycemic index (GI) diet with 3 meal and 3 snack plan. Importantly, all receive uncooked corn starch, 1.5 g/kg in cool skimmed milk 2–5 times/day, depending on clinician's recommendation. All with elevated CPK and thus suspected with GSD type IIIa are put on high protein diet of 3 gm/kg protein. All receive multivitamin and calcium supplementation.

Follow-up

All are advised follow-up 3 monthly in first year, 4 monthly in 2nd year, 6 monthly to yearly thereafter, depending on the proximity of their residence to our center. All are advised once a week fasting GRBS check at home to ensure level >70 mg/dl. If the level is lower, the family is advised contacting the medical team, and the dose of uncooked corn starch (UCCS) therapy is further increased to 2 gm/kg and rarely the frequency increased. At subsequent visit, if serum triglyceride remains >300 mg/despite good compliance to dietary therapy, low-dose omega-3 fatty acid, i.e. 1 capsule of Maxepa once daily (containing 170 mg eicasapentanoic acid and 115 mg docasapentanoic acid), is given for a minimum of 3 months to ensure triglyceride level remains below 300 mg/dl. Ultrasound abdomen is repeated once in 2 years. If suspecting GSD IIIa, ECHO is done at first visit, and for others, at least once within the first year of diagnosis. ECHO is repeated annually for those with cardiomyopathy.

Results

A total of 30 patients were diagnosed with GSD. Three had Fanconi-Bickel and 7 had follow-up < 4 months and hence excluded. Data were analyzed for 20 patients. All patients were followed up at the metabolic clinic. All were reviewed by the dietician and were initiated on 2–5 times/day of UCCS (1.5 g/kg/dose, mixed in cold skimmed milk/water) therapy.

• 1.Clinical Profile of GSD

Twenty children were studied. Geographic distribution was representative of our tertiary-care center referral pattern—7 patients from Tamil Nadu, 5 from Andhra Pradesh, 4 from Western India, 2 from North-east, and 2 from Bangladesh. 55% (11/20) were born of consanguineous parentage. Our cohort included one set of twins. Mean age at presentation was 4.3 ± 3.7 years, although mean age at symptom onset was 1.4 years. M:F ratio was 16:4.

Symptomatology included hepatomegaly in 100%, short stature (height < 5th centile World Health Organization growth chart) in 90%, motor delay in 50%, and early morning seizures in 40%.

On investigation, all had elevated transaminases and variable hyperlipidemia (Table 1). Elevated lactate (21.3 mmol/l and 3.6 mmol/l each, N range 0–2 mmol/l) and hyperuricemia (5.8 mg/dl and 9.1 mg/dl each, N range 3–6 mg/dl) was seen in 2 patients indicating GSD type I. Neutropenia (absolute neutrophil count of 675/mm³) was seen in one of these suggesting GSD Ib. CPK was elevated in 3 patients (6313, 1614, and 1433 u/l, N range 45–195 u/L), suggesting a diagnosis of GSD IIIa. One of these was confirmed on genetic testing.

Table 1.

Biochemical Follow-up (Mean 22 Months).

Parameter—mean (IQR)	At presentation	At follow-up	P value
Fasting glucose (mg/dl)	50.4 (36–64)	79.5 (60–88)	0.006
SGPT (U/l)	416 (170–551)	199 (64–267)	0.002
SGOT (U/l)	415 (80–653)	162 (60–176)	0.013
Se cholesterol (mg/dl)	178 (146–201)	188 (137–217)	0.179
Se triglycerides (mg/dl)	318 (185–380)	225 (130–294)	0.156

IQR: interquartile range

Wilcoxon-sign rank test was used for P value calculation, P < 0.05 significant.

The P values in bold are statistically significant (<0.05).

ECHO showed cardiomyopathy with concentric left ventricular hypertrophy and normal left ventricular (LV) function in 2 patients.

Ultrasound abdomen was done in all annually as per the recommendation.⁵ All had hepatomegaly and increased echogenicity of liver. One patient had a focal hypoechoic lesion of 14 mm in size in segment 5 of liver which has remained same over 22 month follow-up radiologically. His AFP levels are normal at 0.7 IU/ml (N range up to 5.5). Also at 41 month follow-up visit, he has remained well and is due to return with an MR abdomen imaging.

Diagnosis was established in 9/20 by genetic analysis. 15/20 patients underwent liver biopsy which was suggestive of GSD (details listed in Table 2).

• 2.Clinical follow-up

Table 2.

Clinical, Pathological, and Genetic Characteristics of Our Cohort.

Patient No	Age at presentation (years)	Follow-up (months)	Gender	Parental consanguinity	Genetic and mutation	Liver biopsy	Type of GSD
1	1.3	15	M	Yes	Exon 8 c. 698T > C, homozygous, PHKG2 gene, likely pathogenic	No	IXc
2	1	51	M	No	Intronic c 1620 + 1 G > A, homozygous, PYGL; pathogenic	Yes (done at another center)	VI
3	13.8	31	M	Yes	c.792 C > T, homozygous, G6PC: pathogenic	Yes (done at another center)	I
4	1.4	11	M	Yes	Exon 27 c.3637 C > T, homozygous, AGL, pathogenic	No	IIIa
5	2.8	4	M	No	Exon 10, c.928C > T, hemizygous, PHKA2, pathogenic	Yes—GSD with fibrosis	IX a
6	10.4	4	M	Yes	Exon 12, homozygous, AGL, pathogenic	Yes (done at another center) Fatty change in hepatocytes	III a
7#	6	7	M	Yes	Exon 16, c.1868 T > C, homozygous, PYGL, pathogenic	No	VI
8#	6	7	F	Yes	Exon 16, c.1868 T > C, homozygous, PYGL, pathogenic	No	VI
9	2	17	F	No	3816_3817delAG, Exon 28, homozygous, AGL, pathogenic	Yes (done at another center)	III a
10	3.3	6	M	No	Not done	Yes (done at another center)	Clinical diagnosis—GSD IIIa (as CPK elevated)
11	3	28	M	Yes	Not done	Yes—GSD, PAS positive removed by diastase, mild periportal and bridging fibrosis
12	1.3	39	M	Yes	Not done	Yes—GSD with portal and periportal fibrosis
13	2.8	29	F	No	Not done	Yes (done at another center)
14	10.9	15	M	Yes	Not done	Yes (done at another center)	Clinical diagnosis—GSD 1b as neutropenia +
15	1.8	6	M	No	Not done	Yes (done at another center)
16	1.1	4	M	Yes	Not done	Not done
17	2	14	M	No	Not done	Yes—GSD (PAS +, diastase sensitive) with mild portal and periportal bridging fibrosis
18	7	6	F	Yes	Not done	Yes (done at another center)
19	2.2	18	M	No	Not done	Yes (done at another center)
20	5.5	6	M	No	Not done	Yes—GSD (PAS+, diastase sensitive) with mild portal fibrosis

Mean follow-up period was 22 months (range 4–52).

There was clinical improvement in all on follow-up. Early morning seizures subsided in all on treatment. Growth was compromised in majority at treatment onset with 18/20 (90%) having short stature. The mean weight centile improved from 14.1 to 20.3 at follow-up. The mean height centile also improved on treatment from 2.3 to 7.9. The mean height velocity for the cohort on treatment was good at 8.1 cm/year (all were prepubertal).

• 3.Biochemical follow-up

The summary of biochemical data is depicted in Table 1. There was significant improvement in fasting glucose, SGPT, and serum glutamic oxaloacetic transaminase (SGOT) at follow-up. There was also a decline in triglycerides, although the difference was not significant. Marginal increase in serum cholesterol was noted.

We also checked all our patients for vitamin D levels annually, and the mean serum 25-hydroxy vitamin D was 35.6 ng/ml in our cohort.

• 4.Genetic analysis

This could be done only in 9/20 because of financial constraints. All the samples were sent to a standardized private lab externally for clinical exome analysis. The most common type in our cohort was type III and type VI (3 each) and 1 each of type Ia, IXa, and IXc. The mutations have been described in Table 2. This includes a set of twins positive for type VI GSD.

One of the challenging patients in our cohort was a boy referred at 13.8 years of age for liver transplant, indication being persistent hyperlactatemia and refractory hypoglycemia. Although mother was aware of corn starch, she was not aware of the exact dose or frequency of administration and had not received a diet plan despite being seen regularly by a local pediatrician from 1.5 years of age. He presented with a fasting (6 h) glucose of 9 mg/dl and lactate of 21.3 mmol/l. With intensive dietary treatment and 4 times a day of 2 gm/kg of UCCS with galactose-free milk, he has been doing well. At the last follow-up, 41 months into treatment, he had a fasting glucose value of 102 mg/dl and a fasting venous lactate of 3.8 mmol/l. His liver enzymes have shown a dramatic improvement at follow-up, SGOT 404 to 54 U/l and SGPT from 192 to 50 U/l. His genetic studies were suggestive of GSD I in keeping with his clinical presentation. This case reiterates the need for aggressive dietary regime, early referral to tertiary-care center and making caregivers aware of its significance to ensure good compliance.

Also of note are 3 patients with GSD IIIa who are on high protein 3 gm/kg diet apart from 2 to 3 times daily of UCCS.

Discussion

Glycogen storage disease is a rare inborn error of carbohydrate metabolism with an estimated incidence of 1 in 20,000–43,0000 live births.^{6, 7} There is scarcity of Indian data especially in regard to follow-up. There seems to be poor awareness of symptoms, and with a general view of liver transplant being the only option, it is common practice among clinicians to refer late to tertiary-care centers. Our study clearly demonstrates a lag of 3 years to seek medical help from a tertiary-care center.

Diet therapy is the mainstay of treatment with emphasis on avoidance of simple sugars and use of complex carbohydrates to ensure normoglycemia.^{7, 8} Thus, intermittent daytime and nocturnal UCCS remain the mainstay of therapy. In our cohort, there was significant improvement in clinical and metabolic parameters with the use of UCCS. UCCS is not expensive and freely available in all parts of India. Although nocturnal drip or gastrostomy feeds are not well accepted by Indian patients and long-acting waxy maize corn starch is not available in India, our study reiterates the importance of early dietary intervention with intermittent UCCS. Indeed the meta-analysis by Shah et al. demonstrated that both these approaches show similar results on normoglycemia maintenance and physical development.⁹ Although there may be role for fructose and galactose restriction in GSD type I, more emphasis has recently been placed on carbohydrate toxicity, perhaps its role in metabolic control in GSD IX and cardiomyopathy in GSD III.^{10, 11}

Hypertrophic cardiomyopathy in GSD III has been reported to become reversible with high protein diet.¹² In these patients, high protein and ketogenic diet has been described. In our cohort, the 2 patients with concentric left ventricular hypertrophy (LVH) had GSD type III (1 genetic study positive) and have been placed on high protein 3 gm/kg diet. Long-term follow-up is necessary to assess the clinical improvement on this diet.

To summarize the dietary therapy

• •Fasting plasma glucose to be maintained >70 mg/dl
• •Avoid fasting > 3–4 h (infants & children) and 5–6 h in adolescents
• •Frequent small feeds, avoid post prandial hyperglycemia
• •Avoid sucrose, fructose, and galactose containing common foods such as sugary fruits and dairy products in GSD type I
• •High protein diet in GSD type III
• •Consider ketogenic diet if significant cardiomyopathy
• •Supplement multivitamins, calcium, and vitamin D
• •Ensure compliance to UCCS therapy, establish safe fasting interval on UCCS if possible.

Mixed hyperlipidemia on the other hand remains difficult to treat. It has been reported in literature that hepatic adenoma development and progression may be related to poor metabolic control especially with the triglyceride level >500 mg/dl.^{13, 14} After initial dietary treatment, we have initiated low dose omega-3 fatty acid for our patients (8/20) who had elevated serum triglyceride > 300 mg/dl. We used a lower threshold as some of our patients are able to follow-up only at > 6 monthly intervals because of logistic reasons. Moreover, omega-3 fatty acid is considered a nutritional supplement by many with no side-effects of concern. This therapy has not been described in glycogen storage disease so far to the best of our knowledge. Overall efficacy remains to be authenticated after further studies. We have noted a decline in the triglyceride level in our cohort with this therapy. Only 1 patient was commenced on atorvastatin for hypercholesterolemia.

Growth in GSD is compromised and is of unclear etiology. However, catch-up growth has been reported with metabolic improvement. Our cohort demonstrated an improvement in height and weight centile on follow-up. All our patients received maintenance vitamin D supplementation, and our cohort had a mean serum 25 hydroxy vitamin D of 35.6 ng/ml. Hypovitaminosis D has been reported in as high as 61% of patients with GSD ¹⁴, and in fact low bone mass is associated with poor metabolic control.¹⁵ Thus, maintaining a healthy vitamin D level could have assisted with the catch-up growth in our cohort. Long-term follow-up is needed to know if these children reach their target height.

Long-term complications are increasingly being recognized now. In particular is the risk of hepatic adenoma and hepatocellular carcinoma in GSD I. This warrants liver ultrasound once in 12–24 months ⁵ in <18 years of age. It is extremely important for clinician to be aware that biologically hepatic adenomas and hepatocellular carcinoma (HCC) are different in GSD, and thus AFP (alpha fetoprotein) may remain normal despite malignant change. On the other hand, chronic iron deficiency anemia may be the surrogate marker.^{16, 17} Once a possible adenoma is identified 6 monthly, MR or computerised tomography (CT) is recommended. With scarcity of peripheral blood marker, imaging often results in the diagnosis of malignant transformation of hepatic adenomatosis as highlighted in one of the cases from India.¹⁸ Regression of hepatic adenoma has also been described with strict dietary therapy. One of our patients with GSD I has a stable hypoechoic lesion on 22 month follow-up with normal AFP and no features of iron deficiency anemia. He is being planned to have MRI abdomen with next review. Renal tubular dysfunction should also be screened for in the long term, and we do annual urinary microalbumin assessment which has remained normal in our cohort.

Overall prognosis in GSD Ia, III, VI, and IX remains good. A comprehensive review of hepatic glycogenosis very well describes this.¹⁵ GSD Ib remains difficult to treat because of neutropenia and associated periodontal and other infections as well as inflammatory bowel disease (IBD).

Liver transplant should be reserved for select few with poor metabolic control despite good compliance to dietary therapy. Hepatic adenomas with malignant transformation are the other common indication, although one must note that decrease in triglyceride levels itself have demonstrated regression of hepatic adenoma.¹⁶

In summary, our study analyzed data of 20 children (16 males) with GSD. Common perception of early morning seizures being the presenting complaint was noted in 40% of our cohort. All showed improvement in growth and biochemical parameters on dietary therapy. Ours is the first study to use omega-3 fatty acid therapy for elevated triglyceride level in GSD. Further studies need to be done to establish efficacy of this treatment. High protein diet was used in 2 patients with GSD type III with cardiomyopathy, both of whom have remained stable at follow-up.

GSD commonly present as asymptomatic hepatomegaly with majority having short stature. Dietary therapy with intermittent UCCS is the mainstay of treatment. Our study highlights the importance of genetic study as there is a significant role of high protein diet in GSD III. Fine tuning of metabolic control with careful use of ketone bodies and omega-3 fatty acid needs to be further studied.

Conflicts of interest

The authors have none to declare.