Glucose oxidation after a peritoneal and an oral glucose load in dialyzed patients

Kidney International, Vol. 45 (1994), pp. 1147—1152

Glucose oxidation after a peritoneal and an oral glucose load in dialyzed patients JACQUES DELARUE, CLAUDE MAINGOURD, FERNAND LAMISSE, MARIE-ANGE GARRIGUE, PHILIPPE BAGROS, and CHARLES COUET Clinique Médicale A et Laboratoire de Nutrition, Service de Dialyse Péritonéale, Service de Biochimie, Service de Néphrologie, Hopital Bretonneau, Tours, France

Glucose oxidation after a peritoneal and an oral glucose load In dialyzed

the oxidative metabolism and the thermogenic response in-

patients. Glucose oxidation and thermogenesis were studied after a duced by similar peritoneal and oral glucose load in CAPD peritoneal (P) and an oral (0) glucose load in nine chronically uremic patients undergoing continuous ambulatory peritoneal dialysis (CAPD) for 24.4 5.8 months. The 0 load (50 g) given was equivalent to the amount of glucose absorbed over six hours through the peritoneum of the subjects (51.7 3.3 g). Glucose oxidation and energy expenditure were obtained using indirect calorimetry in basal state and over the six hours following the glucose load. Glucose oxidation rate was higher from 60 to 180 minutes after 0 than after P (P < 0.05), with peak values of 3.85 0.28mg kg' min' and 2.80 0.17 mg kg . miir1 respectively (P < 0.05). Cumulated glucose oxidation over six hours was 53.6 0.6 versus 47.0 3.4 g after 0 and P respectively (NS). Glucose-induced thermogenesis was 8.7 2.9% versus 5 1.9% after

O and P, respectively (NS). The route of administration of glucose induces different kinetics of the glucose oxidation rate, but a similar amount of glucose absorbed either by the peritoneum or by the gut contributes in a similar extent to glucose and energy balance.

patients.

Methods

Patients Nine clinically stable, non-diabetic, chronically uremic patients (4 women and 5 men; 69.4 11.7 years old; 61.1 9.9 kg, Body Mass Index: 22.9 3.5 kg/m2) gave their written consent

for the participation to the study. Three patients were anuric. All the patients underwent CAPD for at least six months (9 to 55

months). None had suffered from peritonitis for at least six months. The clinical characteristics of the patients are presented in Table 1. The protocol was approved by the Ethical

Committee of our institution. The patients were studied twice, one to three weeks apart. In study I, glucose was given in peritoneum cavity; in study II, Continuous ambulatory peritoneal dialysis (CAPD) is a chronic renal replacement therapy in patients with end-stage glucose was given orally. Study I was initially performed in all renal disease [1]. This technique is based on the exchanges patients because it was necessary to determine the amount of between the blood and a peritoneal solution which is replaced glucose absorbed through the peritoneum. The patients' usual

many times a day through an implanted catheter. In this carbohydrate intake was at least 250 g a day including the

solution, glucose is used as the osmotic agent because of its low estimated amount of glucose brought by the dialysis solutions metabolic toxicity compared to other osmotic molecules [2]. (estimated to be 60% of glucose present in dialysis solutions). Most of the glucose present in the dialysis solution is absorbed The patients were instructed to eat a standardized meal and to

through the peritoneum [3] and contributes to glucose and to drain any dialysate from peritoneum 12 hours before each energy metabolism. The amount of glucose absorbed through study. All studies were conducted as follows: patients were the peritoneum is thought to have beneficial effects on the admitted after 12 hours fast in our Metabolic Unit between energy balance. It compensates for the low energy intake 07.00 and 08.00 a.m. They were placed at rest in a bed and secondary to the anorexia frequently observed in CAPD pa- maintained in the supine position throughout the experiments. tients [4]. The contribution of glucose to oxidative and energy A deep forearm vein was cannulated for blood sampling and metabolism after intraperitoneal administration may differ from was kept open by a saline solution (0.9% NaCI). Their peritothat observed after oral administration because of different neal cavity was emptied of any residual dialysate. In study I, glycemic and insulinemic responses [5] and because the gut— glucose was delivered into the peritoneum within 10 minutes, as which is by-passed when glucose is instilled in the peritone- a 2 liter 3.86% anhydrous dextrose solution (sodium lactate: um.—plays a specific role in the metabolic response to an oral 3.92 g/liter, NaCI: 3.67 g/liter, CaCl: 0.252 g/liter, MgC1: 0.152 glucose load [6]. The aim of the present study was to compare gfliter; Baxter, France). Blood samples were sequentially taken at times —30, 0, 30, 60, 90, 120, 150, 180, 210, 240, 300 and 360 minutes for the determination of glucose, lactate, non-esterified Received for publication April 19, 1993 and in revised form November 5, 1993 Accepted for publication November 9, 1993

fatty acids and insulin concentrations. Samples for plasma bicarbonate and urea level determinations were obtained at

© 1994 by the International Society of Nephrology

also sequentially sampled from 30 to 360 minutes for the

time 0 and time 360 minutes. The intraperitoneal solution was 1147

Delarue et a!: Glucose oxidation in dialyzed patients

1148

Table 1. Patients characteristics Duration of

Patient no. 1

2 3 4 5 6 7 8

9

Age

years

Sex

80 66 67 65 79 77 59 48 84

M M M M

F F

M

F F

Weight

kg 46 72 68 71 65 56 46 60 66

BMI kg/rn2 17.3

23.5 23.8 24.6 23.9 23.0 17.5

23.7 28,6

Primary renal disease Chronic glomerulonephritis Chronic glomerulonephritis Chronic interstitial nephritis Chronic glomerulonephritis Chronic pyelonephritis Chronic glomerulonephritis Chronic glomerulonephritis Chronic pyelonephritis Chronic interstitial nephritis

CAPD

Renal function

months

mI/mini] .73 m2

9 11 15

5.17 6.39 3.40

49

0

14

9.38

55 35

0 0

11

3.06

21

0.93

Renal function is determined as creatinine clearance + urea clearance divided by 2.

determination of glucose and urea concentrations. The patients were weighed at the beginning and at the end of the study. At the end of the study, the peritoneal cavity was emptied. The

peritoneal glucose load, that is, the difference between the

with pure N2 and CO2 concentration was increased by 1% and by 2% with pure CO2. For data acquisition, analogue voltage outputs of gas analyzers, temperature and humidity probes and of flowmeter were

amount of glucose initially administered and finally recovered processed by a computer (Macintosh II, Apple) via an anawas 51.7 3.3 g. During study II, each patient ingested a 50 g logue-digital converter (Mac addios, GW mc, Somerville, Masglucose load diluted in 200 ml water within five minutes. sachusetts, USA). Calculations of V02 and VCO2 were performed by using a scientific software (LabVIEW®, National Analytical procedures Instruments, Austin, Texas, USA). Data acquisition for each Glycemia was assayed using glucose oxidase method (Beck- parameter was performed every second and mean calculated man Glucose Analyzer II). L lactate was assayed enzymatically values were displayed every minute. Since the mass flowmeter (Lactate UV-System, Boehringer Mannheim, Meylan, France). was calibrated with fresh air, corrections were introduced Non-esterified fatty acid (NEFA) concentrations were deter- automatically to take into account the changes in water vapor mined by colorimetric method (Biochemica Test Combination, and in CO2 concentration. Boehringer Mannheim). Insulin was measured by radioimmuOxygen consumption (V02) and CO2 production (VCO2) noassay (Phadeseph insulin RIA, Pharmacia Diagnostics AB, were calculated using the classical equations, including the Uppsala, Sweden; normal basal insulin values: 70 to 140 pmoll Haldane correction for V02. V02 and VCO2 were measured liter). Urinary nitrogen was determined by the method of during 30 minutes in the basal state after a 30-minute adaptation Kjedahl [7]. Plasma and peritoneal urea was determined by period, then continuously until the end of the study. Indirect kinetic UV method on an automatized apparatus (Abbott Spec- calorimetry was performed in eight patients. One of the patients refused to wear the hood. trum, Rungis, France). Indirect calorimetry Indirect calorimetry was performed using an open-circuit, indirect calorimeter with a ventilated hood. Fresh atmospheric air was continuously drawn into the hood from the outside of the laboratory building. The flow rate at the outlet was measured by using a mass flowmeter (Setaram MF 2000-5000, Caluire, France). Humidity and temperature of the outfiowing air were measured by a thermometer-hygrometer (HMP 135 Y, Vaisala, Helsinki, Finland). A fraction of the outfiowing air was

continuously withdrawn by a pump, dried by an electric gas cooler (CGEK 4, Hartmann & Braun, Metz, France), filtered for dust, then analyzed for oxygen and carbon dioxide by differential analyzers (Magnetopneumatic 02 analyzer Magnos 4G and Infrared gas analyzer Uras 3G, Hartmann & Braun),

Calculations

The amount of glucose found in each dialysis solution was calculated as the product of the measured glucose concentration in the solution by the volume of the solution delivered into the peritoneal cavity. This volume was obtained by weighing

the dialysis bag before and after it was emptied, with the correction for the density of the solution. The amount of glucose absorbed through the peritoneum of the subjects (peritoneal glucose load) was calculated as the amount of glucose

instilled into the peritoneum minus the amount of glucose

remaining in the dialysate at time 360. The amount of glucose remaining in the dialysate was calculated as the product of the measured glucose concentration in dialysate by the volume of measuring differential levels of 02 and CO2 between the in- and the dialysate recovered from the peritoneal cavity at the end of the outfiowing air from the hood. The calibration procedure of the study (obtained by weighing the dialysate bag before and the analyzers was conducted as follows: the zero value of each after it was filled up). It should be stressed that this calculation analyzer was verified by flowing fresh air through the sample is of limited accuracy. Firstly, there was an intra-individual and reference gas lines simultaneously. The gain and the variation of peritoneal glucose absorption (CV: 10.5%; obtained linearity of the analyzers were checked using mixture of pure in 6 of the 9 subjects). Secondly, it was almost impossible to be air and N2 or CO2. The mixture was done by an accurate gas sure that the peritoneal cavity had been completely emptied at mixing pump (Digamix, WOsthoff Gmbh, Bochum, Germany). the end of the study. For these practical reasons, the oral Fresh air 02 concentration was decreased by 0.84 and by 1.88% glucose load was set at 50 g for each patient in study II.

Delarue et al: Glucose oxidation in dialyzed patients A

12

0

1149

B

10

cc ci)

cc

C.)

oE

100

** ** ** t**

6

Cl) C.)

>..

50

0

0

0 60 120 180 240 300 360

—60 0 60 120180240300360 Time, minutes

Time, minutes

600

C

3

*

D

**

400

=1

I:

*

200

0 —60 0 60 120180 240 300360

—60

Time, minutes

Time, minutes

800

E

*

600 U-

0 60 120180 240 300360

400

w

z

200

0

—60 0 60 120180240 300360 Time, minutes

Fig. 1. Evolution of glucose concentration in dialysate (A), glycemia

(B), insulinemia (C), lactatemia (D) and plasma nonesterfied fatty acids (E) following peritoneal (0) and oral (•) glucose load. * 0.05, ** P

< 0.01 for peritoneal vs. oral load.

rate. Protein oxidation was estimated from nitrogen excretion. In study I, nitrogen excretion was determined from the excreted unnary nitrogen and the urea excreted in the dialysate [9]. Total HCO3pool = 0.6 x W x sHCO3 nitrogen excretion was corrected for variation of body urea pool [10]. The rate of protein oxidation was assumed to be equal in where 0.6 is the estimated volume of bicarbonate distribution as study I and study II. Thus, the same rate of protein oxidation 60% of W, W is body weight, sHCO3 is serum HCO3 concen- was introduced in the calculation of substrates oxidation for tration (mmollliter). The bicarbonate pool did not vary from both studies. Glucose and lipid oxidation were obtained from time 0 and time 360 after the peritoneal load (824 246 vs. 771 NPRQ by using the tables of Lusk [11]. 243 mmol, respectively; mean SD) and after the oral load 229 vs. 832 207 mmol, respectively; mean SD). (879 Cumulated glucose balance Therefore no correction was introduced in the calculation of The cumulated glucose balance after the peritoneal load was vC02. calculated as the difference in the amount of glucose absorbed Substrate oxidation and oxidized over the six hours of the study. During study II, The non-protein respiratory quotient (NPRQ) was deter- glucose was assumed to be completely absorbed by the gut over mined from V02 and VCO2 and estimated protein oxidation the six hours [12]. Cumulated nonoxidative glucose disposal

Bicarbonate pooi Bicarbonate pool was calculated as follows [8]:

1150

1.0

Delarue et a!: Glucose oxidation in dialyzed patients

A

0.9

0 0.8

0.7

B

** **

0 60 120 180 240 300 360

5 .Q

4

0• a(a. O) o

2

*

1

0

0 60 120 180 240 300 360 Time, minutes

Time, minutes 1.6

C

**

1.00

D

E

1.2 C

0

0.8

0

0.4

0.

-J 0.0

0 60 120 180 240 300 360 Time, minutes

0.80

0 60 120 180 240 300 360 Time, minutes

Fig. 2. Evolution of respiratory quotient (A), glucose oxidation rate (B), lipid oxidation rate (C) and energy expenditure (D) following peritoneal (D) and oral (•) glucose load. * <0.05 for peritoneal vs. oral load.

was calculated by substracting from the glucose load the 1. Glucose concentration in dialysate decreased exponentially amount of glucose oxidized over the six hours and the glycosuria. Glucose-induced thermo genesis Glucose-induced thermogenesis was calculated as follows:

as a function of dwell time from 183 5 mri to 42 4 mM (Fig. 1A). Basal glycemia was similar in study I and study 11(5.1

0.3 vs. 4.8

0.3 m, respectively; Fig. lB). Glycemia in-

creased similarly in both studies until 150 minutes, with a peak value at 60 minutes (9.2 0.6 vs. 9.1 0.6 m study land II, energy expenditure (kcalmin) during the basal state was sub- respectively). Then the decrease was more progressive after the stracted from the mean energy expenditure (kcalmin) over the peritoneal load, glycemia remaining significantly higher (P < six hours following glucose administration, then divided by the 0.05) from 180 to 360 minutes in study II than in study I. Basal energy content of the glucose load (3.75 kcal/g). Using this insulinemia was similar in both studies (76 9 vs. 69 8 method of calculation, glucose-induced thermogenesis is ex- pmol/liter; study I and II, respectively; Fig. 1C). In study I, pressed as a percentage of the energy content of the glucose insulinemia peaked at 120 minutes (308 82 pmollliter). In load. study II, insulinemia peaked at time 60 minutes (428 41 pmol/liter). Insulinemia was higher from 60 to 90 minutes in Statistical analysis study II than in study I (P < 0.05). During the last three hours Results are expressed as mean SEM and statistical analysis insulinemia was slightly higher in study I than in study II (P < was performed by a statistical analysis software (Statview Il®, 0.05 at 210 mm). The results obtained for lactatemia are Casalabas Inc., California, USA). The Mann-Whitney U test or depicted in Figure 1D. No difference was observed between the Student's unpaired t-test were used where appropriated to study I and study II. The results obtained for NEFA are explore the statistical significance of the differerences observed depicted in Figure 1E. Basal plasma NEFA were not different between the two treatments. between the two loads (356 50 M and 382 66 sM; study I and II, respectively). A significant decrease in plasma NEFA Results concentration was observed in both studies (P < 0.01). Plasma Metabolites and hormone NEFA levels increased at the end of both studies but the levels The kinetics of glucose concentration in dialysate, glycemia, observed were higher in study II than in study I at 300 and 360 lactatemia, insulinemia and plasma NEFA are shown in Figure minutes (P < 0.05).

1151

Delarue et a!: Glucose oxidation in dialyzed patients

carbohydrate and energy intakes are considered in CAPD The mean respiratory quotient (RQ) was similar in both patients. Although the cumulated glucose oxidation and the 0.01 vs. 0.81 studies (0.83 0.02; study I and study II, cumulated glucose balance over six hours are similar with the two routes of glucose administration, the kinetics of the glucose respectively; Fig. 2A). It was significantly higher in study II oxidation rate are very different. The greater rate of glucose than in study I from 60 to 180 minutes. oxidation observed during the first three hours after the oral Respiratory quotient

load than after the intraperitoneal load may be explained by the higher insulinemic response inducing a greater rate of glucose disappearance from plasma and a greater rate of cellular utilization of glucose [17, 18]. The calculation of the rate of glucose and study II, respectively; Fig. 2B). The rate of glucose oxidation relies on the estimation of protein oxidation rate from oxidation was significantly higher in study II than in study I nitrogen excretion. In the present study, we assumed that the from 60 minutes to 180 minutes (P <0.05). nitrogen excretion of each patient was identical during the oral Cumulated glucose oxidation over the six hours was 47 3.4 load than during the peritoneal load. Such assumption appeared and 53.6 3.1 g in study I and study II, respectively (NS) and to be reasonable for the following reasons: (1) all patients were amounted for 91% and 107% of the loads, respectively (NS). clinically and biologically stable; (2) for each patient, the two Cumulated nonoxidative glucose disposal was 4.01 3.4 g in studies were performed within one to three weeks; (3) the study I and —3.6 3.1 g in study II. evaluation of nitrogen excretion is even more difficult in these

Glucose and lipid oxidation The mean rate of basal glucose oidation was similar in both studies (1.75 0.16 vs. 1.49 0.35 mg kg" min'; study I

The mean rate of basal lipid oxidation was similar in both

studies (1.01

patients than in healthy subjects; (4) each patient acted 0.16 and 1.10 0.16mg kg min'; study I uremic as his/her own control in our study. Even though error in the

and II, respectively; Fig. 2C). The decrease in lipid oxidation estimation of the rate of protein oxidation cannot be ruled out, was less pronounced at time 180 minutes (P < 0.05) in study I the kinetics of the RQ which do not depend on the rate of than in study II. Cumulated lipid oxidation was 18.2 17.6 2.2 g, respectively (NS).

2.3 g and

Glucose-induced thermo genesis

protein oxidation clearly indicate that the contribution of glucose to energy metabolism was much more pronounced during the first three hours following the oral load than the peritoneal load. The rate of glucose oxidation observed after the 50 g oral

Basal energy expenditure was similar for both studies (0.89 glucose load in our patients is similar to what has been observed 0.03 and 0.88 0.03 kcal . min1; study I and II, respectively; after a 50 g oral glucose load in healthy subjects [19]. Fig. 2D). The energy expenditure increased significantly from 0 The basal energy expenditure observed in our CAPD patients to 90 minutes in study II (P < 0.05), but not in study I. The is very close to that reported by Schneeweiss et at [20] in uremic

thermogenic response induced by the glucose load was not patients, particularly in chronically uremic patients with condifferent between the two studies (5 1.9 vs. 8.7 2.9%; servative treatment. The more abrupt increase in energy expenstudies I and II, respectively). diture observed alter the oral glucose load compared to the peritoneal load may be accounted for by the energy cost of Discussion The exponential decrease in glucose concentration observed in the dialysate has already been reported [13]. In our CAPD patients, 73.3 10.5% of the glucose delivered into the perito-

gastric emptying, glucose digestion and absorption on one hand [21] and by a greater rate of intracellular metabolism of glucose

on the other hand [17]. There was a 3.7% difference in the thermogenic response between the two routes of glucose adneal cavity was absorbed through the peritoneum. Such a ministration. The magnitude of this difference is close to the finding is in accordance with other data [14]. The slower decline theorical cost of intestinal absorption [20]. A similar result was

of glycemia observed after the peritoneal load during the last four hours of the study may be explained by the persistence of a glucose absorption through the peritoneum. Conversely, all the glucose is assumed to be completely absorbed within four hours following an oral load [12]. The higher insulin response observed between 30 and 120 minutes after the oral load than after the peritoneal load could be explained by a higher rate of glucose delivery in plasma and/or to other factors such as the cephalic phase of insulin release, the enteric release of CCK, GIP and glucagon like peptides (GLIP) which are bypassed during pentoneal administration of glucose [6, 15]. As to the plasma NEFA , their lower concentration observed at the end of the study after the peritoneal load may be related to the higher insulinemia [16]. Our study shows that the glucose delivered into the peritoneal cavity via a dialysis solution in CAPD patients is oxidized and contributes to energy metabolism. Indeed, the amount of glucose oxidized was similar over the six hour study whatever its route of administration. These data indicate that the perito-

neal source of glucose must be taken into account when

obtained by Vernet et al following enteral versus parenteral nutrient administration [22]. Our study shows that a similar cumulated glucose oxidation

and glucose balance are obtained alter peritoneal and oral administration of glucose in CAPD patients. Such a finding highlights the importance of taking into account the glucose present in peritoneal dialysis solution when estimating carbohydrate and energy intake in CAPD patients. Acknowledgments The technical help of Mrs. Catherine Girard and Mrs. Martine Objois is appreciated. We thank Professor Jean-Louis Bresson for his critical review of this manuscript.

Reprint requests to Jacques Delarue, Clinique Médicale A et Laboratoire de Nutrition, Hôpital Bretonneau, 37044-Tours Cedex, France. References 1. MONCRIEFF JW, POPOVICH RP: Continuous ambulatory peritoneal

dialysis best treatment for end-stage renal disease. Kidney mt 28 (Suppi 17):S23—S25, 1985

1152

Delarue et a!: Glucose oxidation in dialyzed patients

2. RUBIN J: Comments on dialysis solution, antibiotic transport, poisonings, and novel uses of peritoneal dialysis in Peritoneal Dialysis, edited by NOLPH KD, Dordrecht, Kiuwer Academic Publishers, 3.

B: Drainage volumes during continuous ambulatory peritoneal dialysis. ASAJO J 2:54-60, 1979 14. GRODSTEIN GP, BLUMENKRANTZ MJ, KOPPLE JD, MORAN JK,

1988, pp. 199—229

COBURN JW: Glucose absorption during continuous ambulatory

GRODSTEIN GP, BLUMENKRANTZ MJ, KOPPLE JD, MORAN JK,

COBURN JW: Glucose absorption during continuous ambulatory pentoneal dialysis. Kidney mt 19:564—567,

1981

4. LINDHOLM B, BERGSTROM J: Nutritional management of patients

undergoing peritoneal dialysis in Perironeal Dialysis, edited by NOLPH KD, Dordrecht, Kluwer Academic Publishers, 1988, pp.

peritoneal dialysis. Kidney mt 19:564—567, 1981 15. RASMUSSEN H, ZAWALICH KC, GANESAN S, CALLE R, ZAWALICH

WS: Physiology and pathophysiology of insulin secretion. Diabetes Care 13:655—666, 1990 16. GROOP LC, BONADONNARC, DELPRATO

5.

WIDEROE TE, SMEBY LC, MYIUNG OL: Plasma concentrations and transperitoneal transport of native insulin and C-peptide in patients on continuous ambulatory peritoneal dialysis. Kidney mt 25:82—87,

1984

6. 7.

J: Relationship between two stages of prandial insulin release in rats. Am J Physiol 235:E 103—111, 1978 HAWK PB: Kjeldahl method, in Practical Physiological Chemistry, LOUIS-SYLVESTRE

Toronto, Blakinston, 1947, pp. 814—822 8. LA GRECA G, BIASI0LI S, CHIARAMONTE S, DAy! M, FABRIS A, FER1AN! M, PISANI E, RoNco C, ZEN F: Acid-base balance on

peritoneal dialysis. Clin Nephrol 16:1—7, 1981 9. BLUMENKRANTZ MJ, KOPPLE JD, MORAN JK, GRODSTEIN GP,

COBURN JW: Nitrogen and urea metabolism during continuous ambulatory peritoneal dialysis. Kidney mt 20:78—82, 1981 10. JEQuIER E, ACHESON K, SCHUTZ Y: Assessment of energy expenditure and fuel utilization in man. Anna Rev Nutr 7:187—208, 1987 11. LUSK G: Animal calorimetry: Analysis of the oxidation of mixtures

of carbohydrate and fat. J Biol Chem 59:41-42, 1924 12. FERRANNINI E, WAHREN J, FELIG P, DEFRONZO RA: The role of

fractional glucose extraction in the regulation of splanchnic glucose metabolism in normal and diabetic man. Metabolism 29:28—35, 1980 13. RLJBIN

J,

NOLPH KD, PoPovIrch RP, MONCRIEFF JW, PROWANT

S,

RATHEISER K, ZYCK

K, FERRANNINI E, DEFRONZO RA: Glucose and free fatty acid metabolism in non-insulin-dependent diabetes mellitus. J Cliii in-

230—260

vest 84:205—213,

1989

17. FERRANNIN! E, LOCATELLI L, JEQUIER E, FELBER JP: Differential effects

of insulin and hyperglycaemia on intracellular glucose

disposition in humans. Metabolism 38:459—465, 1989 18. TAPPY L, RANDIN JP, FELBER JP, CHIOLERO R, SIMONSON DC, JEQuIER E, DEFRONZO RA: Comparison of thermogenic effect of fructose and glucose in normal humans. Am J Physiol 250:E7l8— E724, 1986 19.

MOERI R, GOLAY A, SCHUTZ Y, TEMLER E, JEQUIER E, FELBER

JP: Oxidative and non-oxidative glucose metabolism following graded dose of oral glucose in man. Diabete Metab 14:1—7, 1988 20. SCHNEEWEISS B, GRANINGER W, STOCKENHUBER F, DRUML W, FERENCI P, EIcHINGER S, GRIMM G, LAGONER AN, LENZ K:

Energy metabolism in acute and chronic renal failure. Am J Cliii NutrS2:596—60l, 1990 21. FLATT JP: The biochemistry of energy expenditure, in Advances in Obesity Research (vol II), edited by BRAY GA, Westport, Technomic, 1978, pp. 211—228 22. VERNET 0, CHRISTIN L, SCHUTZ Y, DANFORTH E, JEQUIER E:

Enteral versus parenteral nutrition: Comparison of energy metabolism in healthy subjects. Am J Physiol 250:E47—E54, 1986