Hepatic Glycogenosis Resulting in Hyperammonemia in a Young Adult With Type 1 Diabetes Mellitus

Abstract

A 19-year-old male individual with poorly controlled type 1 diabetes mellitus (T1DM) of 16 years duration presented with diabetic ketoacidosis (DKA) and was found to have hepatomegaly with disproportionately elevated transaminases, hyperammonemia, and lactic acidosis, which persisted after DKA resolution. Autoimmune, infectious, metabolic, and toxic causes were excluded. Liver elastography demonstrated homogeneous, nonsteatotic, noncirrhotic enlargement. Histology revealed hepatocytes with pale or clear cytoplasm and glycogenated nuclei. Total clinical picture is consistent with Mauriac syndrome—a rare complication exclusively observed in people with T1DM with poor glycemic control, resulting in pubertal and growth delay, cushingoid facies, lactic acidosis, transaminitis, and hyperlipidemia. Although prevalence of suboptimal management of T1DM remains common, even with modern insulin delivery mechanisms, Mauriac syndrome remains rare and poorly characterized. We highlight a severe presentation of Mauriac syndrome and highlight the role of comprehensive assessment of liver dysfunction and imaging.

Introduction

Poorly controlled type 1 diabetes mellitus (T1DM) leads to a combination of hyperglycemia and iatrogenic hyperinsulinemia, which can induce hepatic glycogenesis with downstream clinical consequences in a developing pediatric patient. This syndrome, originally described by Mauriac in 1930, classically includes hepatomegaly, cushingoid facies, pubertal and growth delay [1, 2], and lactic acidosis [3, 4]. Although rare cases associated with variants in glycogen metabolism pathway have been described [5], no clear environmental or genetic predisposition is identified.

The central tenet of treatment for this glycogenosis is tight glycemic control that limits further glycogen storage and allows for hepatic recovery. In an era of continuous glucose monitoring (CGM) and insulin pumps as standard of care in pediatric diabetes management, this syndrome due to persistent hyperglycemia and cycles of hyperinsulinemia is less frequent, but it may become more prevalent as rates of T1DM continue to increase [6, 7]. Moreover, while suboptimal glycemic control is common, the presentation of Mauriac syndrome remains rare, suggesting unique individual susceptibility. Here, we present a particularly severe case of glycogenic hepatopathy in a 19-year-old with T1DM progressing to hyperammonemic acute liver injury requiring intervention by a pediatric gastroenterology specialist.

Case Presentation

A male patient aged 19 years and 11 months, with T1DM diagnosed at 3 years old, presented to the community emergency room with 7 days of progressive viral upper respiratory symptoms with 2 days of vomiting, palpitations, and ketonuria. Laboratory results (Table 1, Day 1) were consistent with diabetic ketoacidosis (DKA) with superimposed acute liver injury. Notably, the transaminase elevation was more severe than expected for the severity of the initial DKA presentation. Initial venous blood gas showed pH 7.2 (reference range, 7.31-7.41) and CO₂ of 9 mmol/L (reference range, 22-29 mmol/L), with β-hydroxybutyric acid 5.4 mmol/L (conventional: 56 mg/dL) (reference range, < 0.4 mmol/L [conventional: < 4 mg/dL]) with alanine aminotransferase (ALT) 627 U/L (reference range, 5-55 U/L) and aspartate aminotransferase (AST) 1024 U/L (reference range, 5-35 U/L). Transaminase levels had not been assessed since age 15 and were previously normal. Abdominal ultrasound demonstrated hepatomegaly with increased echogenicity. Severe elevation in liver enzymes persisted, and he transferred to our pediatric tertiary care center. His DKA resolved on day 2 with standard treatment, and he transitioned to multiple daily injections (MDI) with higher insulin dosing to achieve improved glycemic control.

Table 1.

Selected laboratory values with abnormal values in bold

Selected labs	Reference range (SI)	Day 1	Day 2 Am	Day 2 Pm	Day 3 at 0110	Day 3 Am	Day 3 Pm	Day 4 Am	Day 5 Am	Day 6 Am	Day 7 Am
AST	5-34 U/L	1024 U/L	2150 U/L	2363 U/L	2590 U/L	2807 U/L	ND	2983 U/L	904 U/L	345 U/L	183 U/L
ALT	5-55 U/L	627 U/L	795 U/L	826 U/L	924 U/L	933 U/L	ND	1198 U/L	908 U/L	675 U/L	448 U/L
ALP	40-150 U/L	ND	123 U/L	100 U/L	122 U/L	121 U/L	ND	140 U/L	147 U/L	141 U/L	123 U/L
GGT	12-64 U/L	ND	ND	ND	1531 U/L	ND	ND	1890U/L	2042U/L	2115 U/L	1692 U/L
Albumin	3.0-5.2 g/dL (0.3-0.52 g/L)	3.3 g/dL (0.33 g/L)	3.2 g/dL (0.32 g/L)	2.6 g/dL (0.26 g/L)	2.8 g/dL (0.28 g/L)	2.7 g/dL (0.27 g/L)	ND	2.6 g/dL (0.26 g/L)	2.7 g/dL (0.27 g/L)	2.9 g/dL (0.29 g/L)	2.7 g/dL (0.27 g/L)
Protein	6.0-8.3 g/dL (0.6-0.83 g/L)	ND	6.2 g/dL (0.62 g/L)	5.3 g/dL (0.53 g/L)	5.8 g/dL (0.58 g/dL)	5.5 g/dL (0.55 g/dL)	ND	5.5 g/dL (0.55 g/dL)	6.1 g/dL (0.61 g/L)	6.7 g/dL (0.67 g/L)	6.1 g/dL (0.61 g/L)
Ammonia	18.7-54.5 μg/dL (11-32 μmol/L)	ND	ND	138 μg/dL (81 μmol/L)	ND	186 μg/dL (109 μmol/L)	242 μg/dL (142 μmol/L)	131 μg/dL (77 μmol/L)	204 μg/dL (120 μmol/L)	119 μg/dL (70 μmol/L)	97 μg/dL (57 μmol/L)
Lactate	4.5-19.8 mg/dL (0.5-2.2 mmol/L)	ND	ND	59.5 mg/dL (6.6 mmol/L)	ND	21.6 mg/dL (2.4 mmol/L)	18 mg/dL (2 mmol/L)	ND	ND	ND	ND
CO2	22-29 mmol/L (mmol = mEq)	9 mmol/L	16 mmol/L	22 mmol/L	ND	21 mmol/L	ND	ND	ND	ND	ND
Anion gap	5-20 mmol/L (mmol = mEq)	33 mmol/L	20 mmol/L	5 mmol/L	ND	10 mmol/L	ND	ND	ND	ND	ND

Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, gamma-glutamyl transferase; ND, not done.

The patient primarily received outpatient care at our diabetes center. Several psychosocial barriers were identified, including major depression with prior suicide attempts, placement in foster care, delayed high-school graduation, past alcohol use disorder, and ongoing marijuana use. At admission, patient denied recently taking prescription medications aside from insulin, other recreational drugs, or dietary supplements. His baseline diabetes control was poor (Fig. 1A). The most recently prescribed MDI insulin dose was 1.3 units/kg/day. His glycated hemoglobin (HbA1c) before his admission was 11.5%; CGM demonstrated 19% time-in-range (reference range, 70-180 mg/dL [SI: 3.9-10 mmol/L]), 0% hypoglycemic, 11% hyperglycemic (181-250 mg/dL [SI: 10-13 mmol/L]), and 70% very hyperglycemic (> 250 mg/dL [SI: >13 mmol/L]). His HbA1c at admission was 13.4% (Fig. 1A). Despite this suboptimal control, he had not been admitted for DKA for almost 7 years.

Figure 1.

Longitudinal clinical features. A, Hemoglobin A1C in the 6 years preceding admission. B, Weight growth plotted on Centers of Disease Control and Prevention (CDC) 2000 growth chart. C, Height growth plotted on CDC 2000 growth chart. Black dots are from internal medical records and gray dots are from external medical records.

Physical examination revealed a short (160 cm), underweight male (body mass index [BMI] was 16.5 kg/m²) with marked nontender hepatomegaly 10 cm below costal margin, with Tanner 3 pubic hair, testes 10 to 12 mL, and absent facial hair. He did not have cushingoid facies, excess abdominal fat, nor abdominal distension beyond his markedly enlarged liver. Figure 1B and 1C show his growth trajectory plotted on US Centers of Disease Control and Prevention (CDC) 2000 growth curves.

Diagnostic Assessment

Infectious causes of acute hepatitis were excluded, including cytomegalovirus (CMV), Ebstein-Barr virus (EBV), and hepatitis viruses (HepA/B/C/E), adenovirus, as well as additional respiratory, gastrointestinal, and sexually transmitted infectious panels and stool culture panels (respiratory panel included adenovirus, coronaviruses, metapneumovirus, rhino/enterovirus, influenza A/B, parainfluenza, respiratory syncytial virus [RSV] A/B, Chlamydia pneumoniae, and Mycoplasma pneumoniae; sexually transmitted infectious panel included syphilis, HIV, Chlamydia, N. gonorrhoeae, and Trichomonas; hepatitis panel included HepA IgM, HepB surface antigen, HepB Core antibody (Ab) IgM, HepC Ab, HepE IgG, and DNA of CMV, EBV, and adenovirus; stool screen included Calprotectin, Campylobacter antigen, C. difficile toxin assay, and pathogen panel PCR for Campylobacter, Plesiomonas, Salmonella, Vibrio, Yersinia, E. coli, Cryptosporidium, Cyclospora, Entamoeba, Giardia, Adenovirus, Astrovirus, Norovirus, Rotavirus, and Sapovirus; autoimmune panel included LKM-1 IgG, Smooth muscle Ab, ANA screen, and total IgG).

Pediatric gastroenterology was consulted for additional evaluation and excluded additional causes of acute liver injury including acetaminophen overdose, copper metabolism anomaly, autoimmune hepatitis, alpha-1 anti-trypsin deficiency, thyroid disease, celiac disease, muscle injury, and primary metabolic diseases. Additional laboratory results revealed an elevated gamma-glutamyl transferase—consistent with prior reports in glycogenic hepatopathy—without plausible secondary sources, including kidneys, pancreas, spleen, heart, brain, and seminal vesicles. While his albumin was low (Table 1), he had a normal protime, international normalized ratio (INR), alkaline phosphatase, and direct bilirubin (0.3 mg/dL [SI: 5.1 µmol/L]) (reference range, 0.1-0.5 mg/dL [SI: 1.7-8.6 µmol/L]).

Magnetic resonance imaging (MRI) elastography (Fig. 2) and liver biopsy were performed. Elastography revealed normal hepatic stiffness, and pathology on 2 core samples showed “occasional glycogenated nuclei” with hepatocytes with clear or pale cytoplasm, which typically indicate glycogen accumulation in the setting of T1DM [8]. Otherwise, the samples demonstrated preserved lobular architecture without significant steatosis, fibrosis, cholestasis, or lobar inflammation, although minimal portal inflammation was noted. These findings were inconsistent with hepatitis and consistent with chronic metabolic changes, such as with Mauriac syndrome. Lactate was not measured at presentation but found to be elevated on day 2 of admission after resolution of DKA (Table 1), likely representing high anaerobic metabolism prior to admission without consistent insulin administration [3, 4]. The persistence of anaerobic metabolism likely exacerbated hepatopathy.

Figure 2.

MRI hepatic elastography with size annotation of liver span.

We further investigated the patient's delayed puberty with glycogenic hepatopathy. Gonadotropins levels were mid-pubertal (follicle stimulating hormone 1.22 IU/L, adult reference range 1.5-12 IU/L; luteinizing hormone 2.3 IU/L, adult reference range 1.4-15 IU/L), with a normal prolactin of 3.9 ng/mL (SI: 3.9 μg/L) (reference range, 2.0-18 ng/mL [SI: 2.0-18 μg/L]) and normal thyroid studies. His total testosterone level of 106 ng/dL (SI: 3.7 nmol/L) was below typical adult reference range (250-800 ng/dL [SI: 8.7-28 nmol/L] for Tanner 5) and concordant with mid-pubertal exam (reference range for Tanner 3, 21-719 ng/dL [SI: 0.7-25 nmol/L]). Bone age (BA) at admission was 14 years 9 months, demonstrating a delayed skeletal maturity by 5 years 3 months from the patient’s current age (19 years and 11 months). Prior BA at 18 years 1 month was 14 years. The patient's BA only advanced 9 months in the 22 months interval between the 2 x-rays, demonstrating slow skeletal maturation. Review of growth charts (Fig. 1B and 1C) demonstrates onset of growth failure at age 14, consistent with time of skeletal maturation arrest and possible onset of clinically significant glycogenic hepatopathy.

Treatment

The patient was started on continuous insulin infusion while monitoring electrolytes and was quickly transitioned to subcutaneous insulin. Pediatric gastroenterology evaluated his elevated ammonia at presentation. Hyperammonemia was initially suggestive of protein catabolism due to limited glucose metabolism, but repeat ammonia rose from 81 μg/L to 109 μg/L (SI: 48 to 64 μmol/L) (reference range, 19-54 μg/dL [SI: 11-32 μmol/L]) despite resolution of DKA. Rifaximin 400 mg was given twice daily, whereas lactulose 20 g was increased from daily to 2 times daily, then to 3 times daily to treat persistently rising ammonia to 142 μg/L (SI: 83 μmol/L) and as prophylaxis for hepatic encephalopathy. Additional liver function tests demonstrated a symmetrically elevated AST:ALT ratio, with hypoalbuminemia and normal coagulation function. Ultimately, his liver injury appeared to be attributable primarily to Mauriac syndrome with superimposed alcoholic liver disease (AST:ALT ratio > 2, though ethanol levels were not obtained) and hypoalbuminemia. Such significant liver dysfunction, especially with an asymmetric AST:ALT ratio has not been previously described in hepatic glycogenopathy. In this context, we propose that insulin treatment of DKA and hyperglycemia promoted hepatic glycogenesis in an impaired liver, further injuring the liver, limiting hepatic clearance of ammonia, and contributing to his clinical progression.

Outcome and Follow-Up

Patient was discharged on a MDI regimen of insulin of approximately 1 unit/kg/d as well as lactulose and rifaximin for persistent hyperammonemia. At discharge, his albumin was still low at 2.7 g/dL (SI: 0.27 g/L), and his ammonia (57 μmol/L [conventional: 97 μg/dL]) and transaminases (AST 183 U/L, ALT 48 U/L) remained elevated. At 10 days after discharge, he started having abdominal pain, increased diarrhea, decreased urine, and vomiting with progression to hematemesis. Community emergency department demonstrated that he was not in DKA at that time, with β-hydroxybutyric acid 0.5 mMol/L (conventional 5.2 mg/dL), his AST corrected to 25 U/L, ALT decreased markedly to 52 U/L, and his albumin corrected to 4.0 g/dL (SI: 0.4 g/L). His presentation was consistent with viral gastroenteritis.

Previously outlined social difficulties continue to negatively impact the patient's regular follow-up and health maintenance. He is still prescribed lactulose and rifaximin per gastroenterology recommendations because he has not completed the laboratory tests for ammonia and hepatic functions as recommended. Similarly, his CGM use remains inconsistent (50% of days) and his glycemic control remains poor (27% time in target glucose range, 1% in hypoglycemia, 32% in hyperglycemia, 40% in severe hypoglycemia); his most recent point-of-care HbA1c at age 20 years 7 months was 11.3%. Due to the patient's age, he will be transitioning to adult care shortly.

Discussion

Hepatic glycogenopathy (Mauriac syndrome), is a complication of longstanding and inadequately treated T1DM. Here, we highlight several acute complications of this condition. Continued increase in transaminases and ammonia after admission may have been precipitated by appropriate and increased insulin administration, temporarily inducing further hepatic glycogenesis in hepatocytes already congested with glycogen, thus causing liver damage. His acute liver injury required intervention for hyperammonemia with a dedicated gastroenterology team. To our knowledge, this severe acute manifestation of glycogenic hepatopathy from poorly managed T1DM has not been described in the literature and further adds to the spectrum of clinical presentation of this disease.

Although he developed a remarkably rare complication of T1DM, this patient's longstanding suboptimal glycemic management and poor control is not uncommon in adolescents. Numerous factors, including some present in our patient, such as unstable home situations, financial and physical limitations to care, and increasing autonomy in diabetes management contribute to well-described difficulties in achieving target glycemic control in adolescents. Despite prior notes of growth failure and delayed puberty, glycogenic hepatopathy was not screened or diagnosed until DKA presentation, indicating the value of considering this diagnosis in a broader population. This patient has deferred genetic testing for pathogenic variants in liver glycogen phosphorylase kinase, noted as a possible risk factor [5]. These authors assert that a genetic predisposition in addition to hyperglycemia are required to inhibit glycogenolysis to the extent of causing hepatopathy and growth delay.

Compared to other reports of glycogenic hepatopathy, several clinical features are unique. This patient did not have a distended abdomen nor cushingoid features, perhaps due to his low BMI compared to other case reports, possibly contributing to delayed recognition. Prior reports have suggested glycogenic hepatopathy may develop secondary to malnutrition, evidenced by his low BMI.

If progressive hepatic glycogenosis occurs, we must respond with enhanced case management and diabetes education because the mainstay of treatment is proper glycemic control to avoid further liver injury. Poor glycemic control required for hepatic glycogenic overload is more common with lower socioeconomic status and decreased access to care [9]. This patient still must navigate barriers for his care to ensure improved glycemic control, leading to further progression of his puberty and healing of the liver.

Learning Points

• Glycogenic hepatopathy is a diagnostic consideration in any patient with T1DM with growth failure, pubertal delay/arrest, and poor glycemic control. Annual screening for elevated transaminases should be considered.

• Glycogenic hepatopathy may cause acute liver injury, including hyperammonemia, which may necessitate hepatic specific intervention guided by gastroenterology as part of a multidisciplinary team approach.

• Treatment of co-occurring medical conditions, such as diabetic ketoacidosis, may precipitate acute decompensation with underlying glycogenic hepatopathy.

Contributors

All authors made individual contributions to authorship. C.B.D., A.L.G., and R.S.N. were involved in the diagnosis and management of the patient and manuscript submission. All authors reviewed and approved the final draft.

Abbreviations

Ab

antibody

ALT

alanine aminotransferase

AST

aspartate aminotransferase

BA

bone age

BMI

body mass index

CGM

continuous glucose monitoring

DKA

diabetic ketoacidosis

HbA1c

glycated hemoglobin

MDI

multiple daily injections

T1DM

type 1 diabetes mellitus

Funding

No public or commercial funding.

Disclosures

None declared.

Informed Patient Consent for Publication

Signed informed consent obtained directly from this adult patient.

Data Availability Statement

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.