Resting Energy Expenditure and Substrate Oxidation in Malnourished Patients With Type 1 Glycogenosis

Abstract

Context

Type 1a and 1b glycogenosis [glycogen storage disorder (GSD)1a, GSD1b] are rare diseases generally associated with malnutrition. Although abnormal substrate oxidation rates and elevated energy expenditures might contribute to malnutrition, this issue has not been investigated.

Objective

To investigate whether abnormal resting energy expenditure (REE) and substrate oxidation rate characterize patients with GSD1.

Design

Cross-sectional study

Setting

Outpatient referral center for rare diseases and laboratory of clinical nutrition at the University Hospital of Palermo

Patients

Five consecutive patients with GSD1 (4 type a, 1 type b; 3 men, 2 women; age range, 19 to 49 years)

Main Outcome Measures

The usual clinical procedures for patients with malnutrition, including REE and basal substrate oxidation rate (both indirect calorimetry), body composition (bioimpedance method), muscle strength (hand-grip test), and the usual laboratory tests, were performed.

Results

Malnutrition was clearly diagnosed in 2 patients (1 GSD1a and 1 GSD1b), with REE elevated in all five patients, and especially, in the two malnourished patients (+124% and +32.1% vs predictive values using Harris-Benedict equations). The two malnourished patients also exhibited lower basal protein oxidation rates (7.7% and 6.6%) than the nourished patients (range, 12.1% to 24.7%), with higher carbohydrate or lipid oxidation rates. Additionally, the two malnourished patients exhibited higher blood concentrations of lactic acid than the nourished patients.

Conclusions

According to data obtained from our small sample of patients with GSD1, elevated REEs seem to be a common characteristic that might contribute to malnutrition. Low basal protein oxidation rates and elevated blood lactic acid concentrations appear to be associated with malnutrition.

Glycogen storage disorders (GSDs) are a heterogeneous group of inborn errors of carbohydrate metabolism due to variants in genes encoding individual enzymes in the glycogen metabolic pathway (1). In the spectrum of GSD, GSD type 1 is the most frequently observed rare disease and is characterized by both defective glycogenolysis and gluconeogenesis with liver, kidney, and intestinal involvement. Two subtypes of this GSD have been distinguished: type 1a (GSD1a) or Von Gierke disease, which is characterized by glucose-6-phosphatase deficiency; and type 1b (GSD1b), which is due to a deficit of glucose-6-phosphate translocase. Glucose-6-phosphate translocase is a protein that transports glucose-6-phosphate across the microsomal membrane from the cytosol to the endoplasmic reticulum. Owing to insufficient hepatic conversion of glucose-6-phosphate into glucose through glycogenolysis and gluconeogenesis, hypoglycemia and increased blood lactate levels occur after a short period of fasting. Although almost all patients with GSD1 will be malnourished and lean, studies evaluating energy expenditure in this clinical condition are lacking. Resting energy expenditure (REE) is the main component of total daily energy expenditure (∼70% in adult sedentary people) (2). The REE represents the amount of energy expended by a person at rest, in postabsorptive fasting, and in thermoneutral conditions. It differs from the basal metabolic rate in that the latter is measured just after awakening in the morning. However, in practice, the REE and the basal metabolic rate will differ by <10%. Only one short study (3) has measured energy expenditure in 7 patients with GSD1. However, although a high REE was reported, no data were obtained regarding energy substrate oxidation and the possible nutritional effects.

In the present study, we measured REE and basal substrate oxidation rates in a small sample of patients with GSD1a to investigate the possible influences of their nutritional state.

Materials and Methods

Patients

From 2017 to 2018, one patient (patient 1) with GSD1b and four patients (patients 2, 3, 4, and 5) with GSD1a were referred by the outpatient Regional Center of Metabolic Rare Diseases to the Laboratory of Metabolism and Clinical Nutrition for nutritional evaluation. Both facilities are located in the University Hospital Policlinico “P. Giaccone” (Palermo, Italy). All measurements were obtained in the morning from 8:00 to 8:30 am after the last intake of corn flour (1 g/kg body weight) at 6:00 am (measurements taken over more prolonged periods of fasting are not possible because of hypoglycemia). The characteristics of the habitual diet were as follows: 60% to 70% of calories were from carbohydrates, 10% to 15% from protein (to provide the daily recommended intake), and the remaining from fat. In particular, patients consumed 1.7 to 2.5 g of cornstarch per 1 kg of body weight (ideal body weight) every 4 to 5 hours, plus one dose at bedtime. The patients periodically consumed iron supplements, folate, vitamin D, and vitamin B₁₂. All patients included in the present study had had normal thyroid hormone concentrations. The presence of hepatic lesions possibly related to adenomas or hepatocellular carcinoma were excluded by the ultrasound examination findings.

The study was conducted in accordance with the Declaration of Helsinki guidelines, and the institutional review board at the Dipartimento Biomedico di Medicina Interna e Specialistica of the University of Palermo (now Dipartimento di Promozione della Salute, Materno-Infantile, Medicina Interna e Specialistica di Eccellenza) approved the present study. All examinations and procedures performed were a part of the regular clinical procedures of the center for patients with nutritional disease. All the participants provided written informed consent before inclusion in the present study.

Anthropometric and clinical measurements

The height and body weight were measured with the participants lightly dressed and without shoes (Seca GmbH, Hamburg, Germany). The body mass index (BMI) was calculated as the body weight in kilograms divided by the height in square meters. The fat mass (percentage of body weight) and fat-free mass (FFM) were estimated using bioelectrical impedance, as previously described (4), using an 800-mA, 50-kHz, tetrapolar impedance plethysmography [bioelectrical impedance analysis (BIA); BIA-101 Anniversary, Akern Srl, Florence, Italy] to obtain body resistance (R; ohm), reactance (Xc, ohm), and phase angle [PA degrees = arctan (Xc/R) × (180/p)]. The use of crude BIA measures, such as the phase angle (PA), has received increasing attention as a plausible indicator of intra- and extracellular hydration and nutritional status (5). Also, skinfolds (subscapular, suprailiac, tricipital, bicipital) were measured using the Holtain caliper (Holtain Ltd., Ceosswell, UK), and the skinfold-derived fat mass was calculated according to the equations of Durnin and Womersley (6). The body circumference was measured at the umbilicus (waist circumference) and at the most prominent buttock level (hip circumference). The waist-to-hip ratio was used as an indirect index of body fat distribution. The grip strength was measured using a hydraulic hand dynamometer (model no. SH5001; Jamar, Saehan, Republic of Korea). Patients performed the test while sitting, with their shoulder adducted and forearm neutrally rotated, elbow flexed to 90°, and forearm and wrist in a neutral position. The patients were instructed to perform a maximal isometric contraction. The test was repeated within 15 to 20 seconds for each hand, and the average value (kg) of the three tests was used for the analysis (7). The systolic and diastolic arterial blood pressures were measured at 5-minute intervals in the seated position and performed twice using standardized procedures (Omron M6; Omron Health Care Co., Matsusaka, Mie, Japan).

Indirect calorimetry

The REE, respiratory quotient (RQ; VCO₂/VO₂; an indirect measure of the mixture of carbohydrate and lipid oxidation), and nonprotein RQ, which included the measurement of urinary nitrogen excretion during the previous 12 hours to calculate the protein, carbohydrate, and lipid oxidation rates (g/h or percentage of energy expenditure), were obtained using the indirect calorimetry method, as described previously (8, 9), and a ventilated hood system (Quark RMR; Cosmed, Rome, Italy). The device was equipped with an infrared analyzer for carbon dioxide measurement (VCO₂) and a zirconium cell analyzer for oxygen measurement (VO₂). The analyzers were calibrated before each test using gases with a known percentage of oxygen and carbon dioxide. In brief, the respiratory gas exchange was continuously measured for ∼1 hour. The data were obtained from ≥30 minutes of stable measurements, and the average intrasubject variability was 3.9% for the REE. The REE was calculated using the equation of Weir (10) and expressed both in absolute terms (kcal/24 h) and normalized for FFM size (kcal/kg FFM × 24 h). Concerning the substrate oxidation measurement (11), the urea nitrogen excreted in the urine during the previous 12 hours was assumed to be derived from protein oxidation. The grams of oxidized proteins were obtained by multiplying the amount (g) of ureic nitrogen by 6.25. Because the quantity of oxygen necessary to oxidize 1 g of protein and the amount of carbon dioxide produced are known, the nonprotein RQ is obtained by subtracting these estimated volumes from the measured volumes, which will exclusively reflect the relative proportions of lipid and carbohydrate oxidation. An RQ value of <0.70 might indicate that substantial neoglucogenic and ketogenic activity occurs. An RQ of >1 is indicative of lipogenesis starting from carbohydrates. In particular, increased production of lactate will produce carbon dioxide displacement from bicarbonates of the alkaline reserve to the exhaled component with a consequent increase in the RQ.

For each patient, the value of the measured REE was compared with the individual predicted values according to the equations of Mifflin et al. (12), World Health Organization et al. (13), and Grande and Keys (14).

Laboratory analysis

The fasting plasma glucose, total cholesterol, high-density lipoprotein cholesterol, triglycerides, uric acid, and creatinine concentrations were measured using standard clinical chemistry methods (Glucosio HK ultraviolet; Colesterolo tot. Mod P/D; Colesterolo HDL gen 3 mod P/917; Trigliceridi; Acido urico MOD P/917; Creatinina enzimatica; Roche Diagnostics, Monza, Italy). Basal insulin concentrations (Elecsys insulina; Roche Diagnostics) and HbA1c; (B-analyst HbA1c; Menarini Diagnostics, Florence, Italy) were also measured. The low-density lipoprotein cholesterol serum concentration was calculated using the Friedewald formula, and the estimated glomerular filtration rate was estimated using the Chronic Kidney Disease Epidemiology Collaboration equation. Insulin resistance was estimated using the homeostatic model assessment for insulin resistance (HOMA-IR) formula: fasting plasma insulin (mUI/L) × fasting plasma glucose (mmol/L)/22.5. Measurements of blood bicarbonate and lactate concentrations were obtained from venous samples using an amperometric method (ABL 800 FLEX; Radiometer, Copenhagen, Denmark).

Results

The physical, clinical, and laboratory characteristics of patients are presented in Tables 1 and 2. In particular, body size based on the BMI was within the normal range, with the exception of patients 1 and 5, who were underweight. The FFM was normal in all 5 patients; however, reduced muscle strength (Table 3) was observed in patients 1, 3, and 5. Patients 1 and 5 also had low blood concentrations of hemoglobin, high lactic acid concentrations, and low bicarbonate concentrations. Therefore, a condition of malnutrition was evident, especially in patients 1 and 5. High REE values were observed in all 5 patients (Table 4), especially in patients 1 and 5 (Table 5). The basal rates of substrate oxidation are presented in Table 4. Patient 1 exhibited a high lipid oxidation rate, patient 5 exhibited a high carbohydrate oxidation rate, and both patients had a low protein oxidation rate.

Table 1.

Physical and Clinical Characteristics of Patients

Characteristic	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5
Age, y	25	32	49	19	27
Sex	Male	Female	Female	Male	Male
Cigarette smoking	No	No	No	Yes	Yes
Body weight, kg	49.4	62	53.9	64.7	44.1
BMI, kg/m2	17.3	25.2	24.9	23.9	17.2
Blood pressure, mm Hg
Systolic	110	110	125	125	120
Diastolic	65	60	70	80	70
Heart rate, beats/min	68	75	68	86	82
Waist circumference, cm	76	92	84	91	72
Hip circumference, cm	85	103	97	92	84
Waist-to-hip ratio	0.89	0.89	0.87	0.99	0.86
BIA
Resistance, ohm	725	807	681	598	666
Reactance, ohm	91	66	58	66	71
PA, °	7.2	4.6	4.8	6.2	6.1
Fat mass, %	10.3	27.4	24.4	26.5	10.9
FFM, kg	44.3	45.0	40.7	47.6	39.3
Body skinfolds, mm
Subscapular	4	18	14	9	6
Suprailiac	3	13	7	9.5	3
Tricipital	6.5	10.5	4	8.5	7
Bicipital	7	5	4.5	7	5
Fat mass, %	4.3	27.4	24.4	14.6	8.5

Table 2.

Laboratory Measurements of Patients

Variable	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5
Blood concentration
Glucose, mg/dL	50	71	126	76	63
Cholesterol, mg/dL	104	208	244	195	112
HDL cholesterol, mg/dL	20	17	5	31	21
Triglycerides, mg/dL	284	452	551	387	198
LDL cholesterol, mg/dL	27	101	129	87	51
Uric acid, mg/dL	6	4.3	2.8	7.9	9.1
Insulin, μUI/mL	0.2	2.03	28.8	5.15	1.2
AST, U/L	25	38	18	48	16
ALT, U/L	38	35	13	58	23
γ-GT, U/L	62	122	19	62	27
Creatinine, mg/dL	0.49	0.45	0.62	0.50	0.98
Hemoglobin, g/dL	10.7	11.4	10.6	11.5	8.3
Lactic acid, mmol/L	8.8	3.4	6.3	5.6	7.6
HCO3−, mmol/L	16.1	21.7	21.3	21.3	18.0
HbA1c, %	4.5	4.5	4.1	4.5	4.6
HOMA-IR	0.02	0.36	8.96	0.97	0.20
Urinary ketone bodies, mg/dL	0	0	0	0	0
Urinary ureic nitrogen, g/24 h	9.1	9.7	14.2	9.3	4.5

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; γ-GT, γ -glutamyl transferase; HDL, high density lipoprotein; LDL, low-density lipoprotein.

Table 3.

Hand-Grip Test Results of Patients

Patient. No.	Right Hand, kg	Normal Value for Right Hand, kg	Left Hand, kg	Normal Value for Left Hand, kg
1	17	54.79 ± 10.43	18	50.12 ± 7.35
2	30	35.70 ± 8.71	28	30.84 ± 8.03
3	18	28.21 ± 6.85	14	25.40 ± 5.76
4	38	49 ± 11.1	32	42.18 ± 12.6
5	22	49 ± 11.1	22	42.18 ± 12.6

Data presented as median ± SD.

Table 4.

REE and Substrate Oxidation of Patients

Variable	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5
REE, kcal/24 h	3192	1600	1578	2116	1710
REE, kcal/kg/FFM/24 h	72	35.6	38.8	44.4	43.5
RQ	0.81	0.94	0.81	0.89	0.96
NPRQ	0.81	0.97	0.80	0.90	0.97
Basal oxidation
Carbohydrates, g/h	12.7	13.7	4.3	14.0	14.7
Carbohydrates, %	35.7	78.0	25.9	58.8	82.7
Lipids, g/h	8.3	0.4	3.6	2.6	0.9
Lipids, %	56.6	7.0	49.2	29.0	10.7
Proteins, g/h	2.5	2.5	4.1	2.4	1.2
Proteins, %	7.7	15.0	24.7	12.1	6.6

Abbreviation: NPRQ, nonprotein respiratory quotient.

Table 5.

Differences Between Measured and Predicted REE of Patients

	REE (kcal/24 h)
Patient No.	Measured	Mifflin et al. (12)	∆%	WHO et al. (13)	∆%	Grande and Keys (14)	∆%
1	3192	1432	+122.9	1436	+122.3	1382	+131.0
2	1600	1282	+24.8	1349	+18.6	1404	+14.0
3	1578	1055	+49.6	1284	+22.9	1270	+24.3
4	2116	1583	+33.7	1666	+27.0	1485	+42.5
5	1710	1313	+30.2	1356	+26.1	1226	+39.5

Abbreviations: ∆, percentage of difference of measured REE vs predicted value; WHO, World Health Organization.

Discussion

The incidence of GSD1 is ∼1 per every 100,000 births. The estimated incidence of GSD1b has been 1 case per 1 million births (1, 15). Therefore, owing to the rarity of the disease, we could include only a small group of five patients, a clear, but almost inevitable, limitation of the present study. The data we have presented suggest that malnutrition is a condition that could be present in patients with GSD1 (16, 17). Although only patients 1 and 5 were clearly underweight, patient 3, whose BMI and FFM were within the normal range, exhibited reduced muscle strength, indicative of malnutrition. The bioelectrical PA is known to inversely correlate with the ratio of extracellular/intracellular water, and low PA values have generally been found in malnourished and/or clinically compromised patients (5, 18). In contrast to what was expected, we found normal to high PA values in the two patients (patients 1 and 5) with advanced malnutrition. A possible explanation is that, as is known, each molecule of glycogen requires six molecules of water in depots (19); therefore, the high glycogen depots in FFM require more glycogen-associated intracellular water, which might have contributed to the high values of both FFM and PA we found. The high REE observed in all patients might have contributed to the development of malnutrition. The highest REE values were observed in those patients (patients 1 and 5) with severe malnutrition. Our results agree with the unique study reported by Feillet et al. (3). They had investigated the REE in seven adult patients with GSD1a, demonstrating values that were, on average, +116% ± 11% of the predictive value (3). A clear explanation for the high REE observed in all five patients in our study was not evident. As known, the FFM, and, above all, its metabolically active cellular fraction without the extracellular water component (i.e., the body cell mass) is the main determinant of the REE (20). However, our patients did not exhibit any increase in FFM. Furthermore, even when normalized for the amount of FFM, the REE still remained high. It would be necessary to observe that the FFM has different components with potentially different contributions to the REE. Different organs constitutive of FFM, including the heart, liver, and kidney, although representing an average of 5% to 6% of body weight, contribute ∼60% to the REE. In contrast, the muscles that provide ∼40% of body weight contribute ∼22% to the REE (21). Because the contribution of visceromegaly to the FFM is expected to be high in patients with glycogenosis (1), it cannot be excluded that the hypermetabolic status suggested by the elevated REE is a consequence of the high contribution from these organs. In particular, hepatomegaly is a characteristic of patients with GSD1 (16), and it has been calculated that the contribution of this organ might range from 200 to 400 kcal, according to its size (22, 23). Alternatively, neoglucogenic activity is increased in patients with GSD1, although it does not extend beyond glucose-6-phosphate production (1). In our patients, this metabolic tendency was suggested by the high serum lactate concentrations. It is well known that neoglucogenesis from alanine occurs with an energy cost and, possibly, contributes to the high value of REE observed (9). Although, to the best of our knowledge, no study has addressed this issue in patients with GSD1, indirect evidence is available to support this hypothesis. Very high concentrations of alanine were found in postabsorptive fasting conditions in patients with GSD1 compared with normal control individuals. Furthermore, the concentrations of alanine had significantly decreased after glucose intake (24). This behavior might suggest that a fasting gluconeogenesis attempt occurs, starting from alanine, which, clearly, cannot be completed owing to the lack of the glucose-6-phosphatase enzyme. However, in the previous intermediate steps, consumption of ATP occurs. In the mouse model of GSD1a, it was proved that increased production and mitochondrial accumulation of pyruvate, also from alanine and other tricarboxylic acid cycle intermediates, occurs (25). In contrast, pyruvate might flow into other energy-consuming processes. Another potential regulatory mechanism could be the increased activity of the ATP-consuming futile cycle (pyruvate to oxaloacetate to phosphoenolpyruvate to pyruvate), which contributes to attenuating gluconeogenic flux (26). Jones et al. (26) demonstrated an increased fraction of acetyl-CoA from pyruvate and increased pyruvate recycling fluxes, all ATP consuming pathways, in patients with GSD1a. Finally, even a minimal contribution to neoglucogenesis would seem to be possible at the muscular level (where glucose-6-phosphatase-β would act), although extremely insufficient to guarantee adequate endogenous glucose production (27). Both the RQ and the nonprotein RQ of the patients included in the present study were intermediate and comparable to those found in the healthy population (28). It is likely that two metabolic phenomena occur concomitantly with an opposite effect on the RQ, thus canceling out any global influence on the RQ. Thus, neoglucogenic activity would produce a low RQ, and the consumption of bicarbonates (as documented by the low blood concentrations in our patients) due to increased lactate production would lead to increased elimination of carbon dioxide with breathing, increasing the RQ. Although the evaluation of the RQ did not allow us to identify metabolic abnormalities in our patients with GSD1, the calculation of protein oxidation suggested low metabolic rates in the patients with advanced malnutrition (patients 1 and 5) that were compensated for by higher lipid or carbohydrate oxidation. Furthermore, in stable body weight conditions, the mixture of substrates oxidized will be influenced by the diet composition and the size of body fat stores (the greater the fat mass, the greater the lipid oxidation). Thus, in a negative energy balance, the body will oxidize more fats (and vice versa) (29). Our patients had a stable body weight (and, therefore, were in energy balance) and had limited fat stores. Thus, they oxidized an intermediate mixture of substrates or had a high carbohydrate oxidation rate as a result of their habitual diet composition.

The laboratory values obtained for our patients with GSD1 were largely expected (16) owing to the metabolic derangements (triglycerides in particular), liver complications (γ-glutamyl transferase, in particular), and malnutrition (hemoglobin, low creatinine values, urinary nitrogen). Also, the low HOMA-IR values were largely expected, because patients with GSD1 tend to have low glycemic values with, consequently, low insulin production. An exception was patient 3, who was the oldest patient of our five patients and had greater glycemic (with comparable HbA1c) and HOMA-IR values compared with the other patients. We do not have a clear explanation for these values; however, it has been reported that overtreatment with cornstarch can result in insulin resistance (16, 30).

An important limitation of our study was the use of BIA to determine the body composition, because no specific body composition equations are available for patients with GSD. The BIA method is based on the prediction of FFM from body water (4). However, in our patients, the total body water could have been altered by GSD, and thus, reducing the accuracy of the BIA. However, patients with GSD are usually very ill and poorly tolerate investigations that are not strictly necessary for their treatment. Therefore, only those investigations that were a part of the usual clinical practice at our center were performed, and a more accurate evaluation of body composition was not performed. Another important limitation of our study was the inadequate procedure for normalizing the REE by simply dividing its value by the FFM. More correctly, we should have considered the value of the positive intercept usually observed in the linear regression between the REE and FFM (31). However, our study included only five patients; thus, it was inappropriate to evaluate the correlation between the REE and FFM. However, we are confident that, despite this limitation, the results of our study can well support the hypothesis that patients with GSD1 will have a high REE because the individually measured absolute REE values were well above the predicted values obtained using different predictive equations (Table 5). Although the REE will generally be a high percentage of total energy expenditure (60% to 70%), this issue is completely unknown for patients with GSD1. Therefore, we could not exclude that the total energy expenditure might not be elevated in those with GSD1 and should be investigated using more sophisticated methods such as doubly labeled water (32).

In conclusion, the metabolic and nutritional aspects of GSD1 are extremely complex and require careful evaluation. As a result of hepatomegaly and hypoglycemia, patients with GSD1 can be characterized by a high REE, normal RQ, low to normal BMI, normal FFM with low muscle strength, and high lactate and low bicarbonate blood concentrations. Malnutrition can be associated with this clinical condition and contribute to limiting patients’ quality of life and life expectancy. To date, pharmacological therapeutic aids are almost nonexistent, and dietetic aids appear inadequate to guarantee satisfactory outcomes. A commitment by both the pharmaceutical and the food industries is desirable to ensure real progress in the treatment of this clinical condition.

Acknowledgments

Disclosure Summary: The authors have nothing to disclose.

Data Availability: Restrictions apply to the availability of the data generated or analyzed during the present study to preserve patient confidentiality or because they were used under license. The corresponding author will on request detail the restrictions and any conditions under which access to some data can be provided.

Abbreviations:

Abbreviations:
 
• BIA

  bioimpedance analysis

 
• BMI

  body mass index

 
• FFM

  fat-free mass

 
• GSD

  glycogen storage disorder

 
• HOMA-IR

  homeostatic model assessment for insulin resistance

 
• PA

  phase angle

 
• REE

  resting energy expenditure

 
• RQ

  respiratory quotient

Author notes

S.B. and D.N. contributed equally to this article.