In Vivo Whole-Body Resting Energy Expenditure and Insulin Action in Human Malignant Hyperthermia 

The study participants were 25 Swiss individuals on whom a previous MH diagnostic muscle biopsy had been performed between 1986 and 1996 because of a clinical episode compatible with MH or the existence of an MHS family member. The persons were contacted by mail and asked to enroll in the study on a voluntary basis. The participants were studied between 1995 and 1997 in random order. From the results of their IVCT (table 1), 10 patients had been diagnosed as MHN, 10 as MHS, and 5 as MHEh (IVCT results abnormal for halothane but normal with caffeine). As shown in table 1, the 25 persons were from 14 different families with no known MH genotype. Each person was otherwise healthy, took no medication, and had a normal blood pressure and fasting plasma glucose level. Only two and three persons in the MHN and MHS groups, respectively, had a first-degree hypertensive relative. No participant had a first-degree relative with diabetes other than one MHN person whose father had shown a transient hyperglycemic state while on temporary corticosteroid therapy.

No significant differences in physical parameters among MHN, MHS, and MHE persons was observed (table 1). The study protocol was approved by the Ethical Committee of the Faculty of Medicine of the University of Lausanne. All participants gave informed written consent before taking part in the study.

The muscle biopsies were carried out at the Basel University Hospital. All biopsy samples were taken from the vastus medialis muscle. IVCT was performed according to the protocol of the European Malignant Hyperthermia Group, 2,21and IVCT results of the first test for both halothane and caffeine were used. The time interval between the muscle biopsy and the muscle contracture test was about 60 min for all patients.

The participants were asked to avoid heavy physical exercise and tobacco and caffeine consumption for the 2 days preceding the study. The evening before the glucose clamp procedure (see Euglycemic Hyperinsulinemic Clamp), body composition was measured by the skinfold 22and bioimpedance 23methods. Briefly, the bicipial, tricipital, subscapular, and suprailiac skinfold thickness was measured in triplicate by the same investigator using a Harpenden caliper (CMS Weighing Equipment, Ltd., London, UK). Whole-body bioelectrical impedance was determined with a portable device built by the Electronic Department of the Lausanne University Hospital 24to measure the resistance and the reactance in triplicate at 0.5, 50, and 100 kHz after applying a 500-μA current from the right wrist to the right ankle. The arithmetical mean result of both techniques was used to calculate body FM and FFM. The participants then ate a standardized commercial meal (29% protein, 46% carbohydrate, 25% fat) providing 6.7 ± 1.3 kcal/kg body weight (mean ± SD). After a 10-h overnight fast, the participants underwent a hyperinsulinemic euglycemic clamp procedure as previously described. 25 

With the person lying comfortably, two intravenous lines were inserted under local anesthesia (lidocaine 1%): one in a vein of the antecubital fossa of the left arm (injection line) and one in a vein on the dorsal side of the right hand; the right hand was kept in a heated environment (50°C) throughout the test to arterialize the blood samples drawn from this line.

To determine the fasting endogenous glucose production rate, we began the test at time 0 by a primed (16.5 μM/kg), continuous (0.16 μM · kg⁻¹· min⁻¹) infusion of [6,6-²H]glucose until the start of the insulin infusion at time 150 min (fig. 1). The sterile [6,6-²H]glucose had been diluted to a 1% concentration in NaCl 0.9% under laminar flow and tested for sterility and apyrogenicity. Because we also mixed [6,6-²H]glucose with the exogenous glucose solution necessary to maintain euglycemia 26(see no. 3 of clamp description), we reduced the [6,6-²H]glucose infusion rate to 0.05 μM · kg⁻¹· min⁻¹during insulin infusion between time 150 and 270 min. The plasma deuterated glucose isotopic enrichment was measured in duplicate from samples drawn at 120, 135, 150 min during the fasting period and at 240, 255, and 270 min during the insulin-stimulated state.

The clamp (from time 150–270 min) consisted of: (1) a primed, continuous insulin (Actrapid HM, Novo, Bagsvaerd, Denmark) infusion (1 mU · kg⁻¹· min⁻¹); (2) a 4-mm/h KCl infusion to prevent hypokalemia; and (3) a variable infusion of 17.1% glucose (500 ml 20% unlabeled glucose and 90 ml 1%[6,6-²H]glucose, 0.857 mole % excess) as necessary to maintain the plasma glucose concentration at 5.6 mm for all persons. An amount of NaCl 0.9% replacing the volume of blood drawn (≈ 100 ml) was injected continuously during the clamp. The plasma glucose concentration was measured in duplicate 30, 15, and 1 min before insulin infusion (during fasting) and then every 5 min during insulin infusion. During insulin infusion, the mean glucose and insulin plasma concentrations between 240 and 270 min were (mean ± SE) 5.7 ± 0.1 mm and 76 ± 4 mU/l in the MHN group, 5.6 ± 0.1 mm and 79 ± 5 mU/l in the MHS group, and 5.7 ± 0.1 mm and 83 ± 5 mU/l in the MHE group, with no significant intergroup difference.

Respiratory exchange measurements, using a ventilated hood open-circuit indirect calorimeter, 27started at 90 min and lasted until the end of the test. The airflow through the hood was approximately 40 l/min and was measured at the outlet by a pneumotachograph connected to a differential manometer (model 47303A digital pneumotachograph, Hewlett-Packard, Palo Alto, CA). The rates of carbon dioxide production (VCO²) and of oxygen consumption (VO²) were constantly determined by an infrared carbon dioxide analyzer (Uras T2, full-scale 0–1%, Hartmann and Braun, Frankfurt, Germany) and a thermomagnetic oxygen analyzer (Magnos T2, full-scale 19–21%, Hartmann and Braun) in the gas mixture leaving the hood. For fasting and insulin-stimulated calorimetric measurements, we used the four to six consecutive, stable, 5-min periods occurring between 90 and 150 min and 240 and 270 min, respectively (fig. 1).

Urinary collections were carried out the night before (duration [mean ± SD]= 8.3 ± 1.3 h) and during the test (duration = 6.4 ± 0.5 h) for total urinary nitrogen production rate determination (corrected for the change in urea nitrogen pool size, 28as determined from plasma urea concentration samples drawn at time 120 and 150 min − baseline, and 240 and 270 min − under insulin stimulation). Protein oxidation rate (6.25 g protein/g urinary nitrogen) was calculated at baseline and during the insulin clamp procedure. The nonprotein respiratory quotient (VCO²/VO²) was then calculated and allowed the determination of the substrate oxidation rates and REE. 20 

The baseline endogenous glucose production rate was calculated from the plasma deuterated glucose isotopic enrichment measured between 120 and 150 min (mean ± SE: molar % excess = 1.41 ± 0.05%, 1.44 ± 0.07%, and 1.40 ± 0.03%, and coefficient of variation [CV]= 2.4 ± 1.6%, 2.9 ± 1.3%, and 2.3 ± 2.0% in the MHN, MHS, and MHE groups, respectively).

With no change in plasma glucose level, the total glucose disposal rate during insulin infusion equals the total isotopically determined glucose appearance rate assessed at the same time (molar % excess = 1.07 ± 0.04%, 1.08 ± 0.03%, and 1.03 ± 0.04%, and CV = 1.9 ± 1.3%, 1.7 ± 1.3%, and 1.8 ± 1.0% in the MHN, MHS, and MHE groups, respectively, at 240, 255, and 270 min in the insulin-stimulated state). When the glucose appearance was equal to or smaller than the exogenous glucose infusion rate, the endogenous glucose production rate was assumed to be fully suppressed, and we considered that total glucose disposal rate was equal to the exogenous glucose infusion rate (mean rate administered between time 240 and 270 min). When the glucose appearance was greater than the exogenous glucose infusion rate, endogenous glucose production rate was considered not fully suppressed, and the isotopically-determined glucose appearance rate represented total glucose disposal rate. The total glucose disposal rate was corrected for changes in plasma glucose concentration when necessary by using a volume of glucose distribution equal to the person’s extracellular water volume (20% of body weight). The difference between the total glucose disposal rate and the oxidative glucose disposal rate is the nonoxidative glucose disposal rate (“storage rate”). In humans, this rate mainly represents skeletal muscle glycogen deposition rate under insulin stimulation. 12 

We calculated the plasma glucose appearance rate using Steele’s steady-state equations, 29considering the low CVs observed in our isotopic enrichment and calorimetric measurements (in the MHN, MHS, and MHE groups, respectively: fasting VO²CV = 2.0 ± 0.3%, 2.0 ± 0.3%, 2.2 ± 0.5%; fasting VCO²CV = 2.4 ± 0.4%, 3.3 ± 0.5%, 3.1 ± 0.7%; insulin-stimulated VO²CV = 2.0 ± 0.2%, 2.1 ± 0.3%, and 2.5 ± 0.5%; and insulin-stimulated VCO²CV = 2.6 ± 0.3%, 3.0 ± 0.4%, 3.2 ± 0.5%).

Plasma glucose concentration was measured using the glucose oxidase method (Glucose Analyzer, Beckman Instruments, Fullerton, CA), urinary nitrogen was measured using the Kjeldahl method 30(N Autoanalyzer, Technicon, Tarrytown, NY) and plasma urea concentration was measured enzymatically (Urea Analyzer, Beckman Instruments). Plasma insulin concentrations were assessed in duplicate by radioimmunoassay, 31and the isotopic enrichment of plasma [6,6-²H]glucose was measured in a gas chromatograph–mass spectrometer (Hewlett Packard, Waldbronn Analytical Division, Waldbronn, Germany) as previously described. 25 

All data are presented as means ± SEM. Intergroup comparisons were performed using the Student unpaired t  test and intersituational comparisons were performed using the paired t  test. In the case of an abnormal distribution of one variable, the corresponding nonparametric test was used (Mann–Whitney rank sum test and Wilcoxon signed rank test). The Fisher exact test was used to compare the group gender ratios. Sample correlations are Pearson product–moment correlations. Multiple linear regression analysis was used to calculate predicted REE. A P  value < 0.05 was considered to indicate statistical significance.

No significant difference was found in plasma glucose and plasma insulin concentrations among the MHN, MHS, and MHE groups (glucose: 5.3 ± 0.1 mm, 5.2 ± 0.1 mm, and 5.2 ± 0.1 mm; insulin: 10 ± 1 mU/l, 8 ± 1 mU/l, and 8 ± 1 mU/l, respectively).

As shown in figure 2, total, oxidative, and nonoxidative glucose disposal rates were not significantly different among the MHN, MHS, and MHE groups, both in the fasting and insulin-stimulated conditions.

Fasting and insulin-stimulated lipid oxidation rates were 1.07 ± 0.08 and 0.44 ± 0.09 mg · kg⁻¹FFM · min⁻¹in the MHN group, 1.13 ± 0.06 and 0.46 ± 0.10 mg · kg⁻¹FFM · min⁻¹in the MHS group, and 0.87 ± 0.25 and 0.10 ± 0.03 mg · kg⁻¹FFM · min⁻¹in the MHE group. Fasting and insulin-stimulated protein oxidation rates amounted to 1.55 ± 0.15 and 1.29 ± 0.15 mg · kg⁻¹FFM · min⁻¹in the MHN group, 1.26 ± 0.15 and 1.33 ± 0.17 mg · kg⁻¹FFM · min⁻¹in the MHS group, and 1.50 ± 0.47 and 1.51 ± 0.24 mg · kg⁻¹FFM · min⁻¹in the MHE group. Again, there was no significant intergroup difference.

Fasting REE was 1413 ± 55 kcal/day in the MHN group, 1497 ± 61 kcal/day in the MHS group and 1440 ± 85 kcal/day in the MHE group; there was no significant difference among groups. Expressed per unit of FFM, fasting REE was 31.4 ± 1.3, 31.9 ± 1.0, and 28.9 ± 2.4 kcal · kg⁻¹FFM · day⁻¹in the MHN, MHS, and MHE groups, respectively; there was no significant difference among groups.

Under insulin stimulation, REE was 1427 ± 69 kcal/day, or 31.5 ± 1.0 kcal · kg⁻¹FFM · day⁻¹, in the MHN group; 1539 ± 59 kcal/day, or 32.8 ± 1 kcal · kg⁻¹FFM · day⁻¹, in the MHS group; and 1468 ± 100 kcal/day, or 29.4 ± 2 kcal · kg⁻¹FFM · day⁻¹, in the MHE group; there was no significant difference among groups.

Fat-free mass is the most important determinant of whole-body REE, 17as shown in figure 3, which depicts the relation between FFM and REE in the 10 MHN and 10 MHS persons in the current study.

In Ravussin’s model, 17,18body FM, age, and gender are considered, along with FFM, to explain interindividual REE variance. Using the same model, we determined the respective contribution of these four physical covariates to the REE that we measured in our 20-person population (10 MHN and 10 MHS). By multiple linear regression, we found that FFM, FM, age, and gender together explained 81 and 83% of the observed variance in fasting and insulin-stimulated REE, respectively. For each of the 10 MHN and 10 MHS persons, we calculated a value of predicted  REE using the following formula:

in which an arbitrary value of 0 was attributed to female subjects and a value of 1 was attributed to male subjects.

The difference between measured and predicted REE is the residual REE, which represents the part of daily energy expenditure not explained by its four main determinants, FFM, FM, age, and gender. 17The comparison of fasting and insulin-stimulated residual REE between MHN and MHS persons is shown in figure 4. MHS persons display a 6.4% higher residual REE (Δ kcal/insulin-stimulated REE^MHN= 91/1427) than MHN individuals under insulin stimulation. This small difference could not be found in the fasting condition.

The main finding of our study was that MHS and MHE persons, compared with MHN persons, do not exhibit any gross alteration of insulin action and energy expenditure. Despite in vivo  reported signs (higher glucose 7and insulin 8,9plasma concentrations) and biochemical muscle findings (elevated [Ca²⁺]^c11and myophosphorylase activity 10) compatible with a chronic insulin-resistant state of MHS individuals, we were unable to document evidence of insulin resistance in MHS persons in this study. Our results are in agreement with values of glucose total, oxidative, and nonoxidative disposal rates measured in normal control subjects 32under physiologic insulin stimulation (1 mU · kg⁻¹· min⁻¹). These parameters were similar in the 10 MHN and 10 MHS persons in the current study, thus showing an unimpaired in vivo  insulin action 12in MHS persons. In addition, unlike Rutberg et al.  7who found a 12% higher plasma glucose level in MHS persons, we measured normal fasting plasma glucose concentrations in our MHS persons, as did Campbell et al.  8Although a 133% greater fasting insulinemia was reported in MHS persons, 8we observed normal fasting plasma insulin concentrations in our MHS participants, which confirms the findings of other studies. 7 

The body composition data show no difference between MHS and MHN persons in terms of body weight, height, FM, and FMM (table 1). This is in agreement with previous results 7but in contrast to other data; using the skinfold method 22Campbell et al.  33found a slightly lower (4.6%) body percent fat in male, but not female, MHS humans compared with biopsied control subjects. Similarly, MHS pigs have a larger lean body mass (2–3%) than control pigs. 1This may be due to the genetic heterogeneity of human MH compared with MH in pigs which is associated with one single genetic mutation.

In agreement with the results of previously published studies, 7,8,14fasting whole-body REE was the same in our 10 MHS and 10 MHN persons. REE was also the same in both groups under insulin stimulation. Despite an increase in nonoxidative glucose disposal, which corresponds to a stimulation of muscle glycogen synthesis, REE under insulin stimulation was not increased, as reported by Thiébaud et al.  32with the same insulin infusion in control subjects. This might be due to an insulin-induced inhibition of gluconeogenesis in the liver, which is an energy-requiring process.

REE represents 60–70% of daily total energy expenditure in humans and can vary among individuals with the same body size and composition. 17Skeletal muscle accounts for as much as 40% of body weight, as much as 30% of REE, 17and explains as much as 50% of the whole-body REE variability in humans. 34The major energy-requiring process of calcium transport in the muscle cell is the SERCA, which maintains calcium homeostasis by pumping calcium from the cytosol back into the sarcoplasmic reticulum. According to Chinet et al. , 35the active Ca²⁺pumping by the mouse muscle sarcoplasmic reticulum might account for 25–40% of the total muscle energy expenditure in basal or near-basal conditions (i.e. , a possible calculated 7.5–12% of whole-body REE). HEK-293 cells, which do not spontaneously express RYR1, exhibit a twofold increase in their SERCA expression when transfected by RYR1 from human wild-type and MH-mutated–type above untransfected cells, 6as in an attempt to compensate for the leak of calcium through the RYR1 into the cytoplasm and maintain a normal resting [Ca²⁺]^c. If MH susceptibility implied a permanent leak of calcium out of its intracellular muscle stores, then the energy cost of a compensation mechanism maintaining calcium homeostasis in muscle might be reflected in whole-body REE. Whereas the absolute value of REE was the same in MHN and MHS persons, we found a slightly higher (6.4%) residual REE in MHS persons during insulin stimulation which was undetectable in the fasting condition.

The significantly greater residual REE observed in MHS persons under insulin stimulation might be related to a role played by calcium, which is incompletely understood in insulin-stimulated target tissues. In adipocytes, 36insulin exposure has been reported to increase cytosolic calcium concentration. In mouse skeletal muscle fibers, 37“near-membrane cytosolic” calcium concentration is dose-dependently elevated by insulin stimulation. This calcium increase might be enhanced in MHS muscle, leading to a corresponding extra amount of energy necessary for SERCA to maintain a normal cytosolic calcium concentration, such as directly measured in MHS human muscle cells. 38,39 

This study indicates that 10 MHN, 10 MHS, and 5 MHE persons with the same body composition have similar fasting plasma glucose and insulin concentrations, whole-body energy expenditure, and insulin sensitivity. In groups formed by 10 MHN and 10 MHS persons, we calculated a value of REE predicted by the determinants FFM, FM, age, and gender. The difference between the measured and the predicted REE, (i.e. , the residual REE) was slightly higher in MHS persons, but only during insulin stimulation. This finding suggests that insulin might challenge the energy balance of MHS muscle. Further studies are needed to investigate the role of MH mutations in muscle energy expenditure and its calcium homeostasis.

The authors thank Dr. L. Tappy, Dr. C. Bogardus, Dr. G. van Melle, and E. Temler for their assistance.