Resting metabolic rate in chronic renal failure

Resting Metabolic Rate in Chronic Renal Failure Uwe Kuhlmann, MD, * Markus Schwickardi, MD, * Riidiger Trebst, MD, * and Harald Lange, M D *

Objective. A decrease in resting metabolic rate (RMR) in patients with chronic renal failure was assumed to occur because of the decreasing oxygen consumption of the kidneys, which in healthy subjects, accounts for 7.2% of RMR. Contrary to this assumption, RMR per body weight in end-stage renal disease was increased. Design and Methods: To test the impact of chronic renal failure on the RMR, direct bedside calorimetry was performed on 51 outpatients (age, 53.2 _+ 13.9 y; creatinine clearance, 6.9 to 52 mL/min). Twenty two of 51 patients were examined repeatedly (at the start of the study, after 3 months, and after 6 months) during declining kidney function. Results: In the total group, RMR per body weight (RMR/BW) was 100.0 -+ 4.96 kJ/kg/day and RMR per body surface area (RMR/BSA) was 4.582 _+ 0.181 kJ/min/1.73m 2. RMR/BW and RMR/BSA correlated significantly with creatinine clearance (n = 51, r = -.763, P < .001; n = 51, r = -.557, P < .001). In the follow-up group, creatinine clearance decreased from 27.5 _+ 9.5 mL/min initially, to 19.4 _+ 6.25 mL/min at 3 months, to 13.0 -+ 3.8 mL/min at 6 months (P < .001), while RMR/BW and RMR/BSA increased from 98.28 +- 6.3, to 101.64 _+ 5.46, to 105.42 -+ 6.3 kJ/kgBW/d (P < .005), respectively, and 4.41 +_ 0.126, to 4.578 -+ 0.168, to 4.704 _+ 0.168 kJ/min/1.73 m 2 (P <.05), respectively. Conclusion: Taking into account the reduced oxygen consumption of the shrinking kidneys, the normal RMR suggests an increased energy expenditure per body cell mass. The raising RMR in deteriorating excretory kidney function reflects the increasing energy expenditure in progressive chronic renal failure. © 2001 by the National Kidney Foundation, Inc.

N PATIENTS suffering from chronic renal failure (CRF), wasting and malnutrition are commonly observed. 1'2 In chronic malnutrition, a decrease in energy expenditure (EE) has to be expected. The oxygen consumption o f the kidneys in healthy subjects at rest accounts for 7.2% of the total metabolic rate. 3 Sclerosis of a parenchymatous Organ reduces the oxygen consumption of the tissue. There are no data available in the current literature on oxygen consumption of the kidney in patients with progressive CRF. Therefore, the replacement of high metabolic parenchyma of the normal kidney by low metaboric tissue of C R F is assumed to reduce the

I

*Center for Internal Medicine, Clinic of Nephrology, PhilippsUniversity of Marburg, Germany. Address reprint requests to H. Lange, MD, Centre of Internal Medicine, Clinic of Nephrology, Philipps-University of Marburg, Baldingerstr, 35033 Marburg, Germany. E-mail: nephro@ post. med. uni-marburg,de © 2001 by the National Kidney Foundation, Inc. 1051-2276/01/1104-0004535.00/0 doi: 10.1053/jren. 2001.26981

202

energy turnover. Hence, a decrease in total metabolic rate caused by malnutrition and the loss of kidney function must be supposed in patients with CRF. Contrary to this assumption, EE in end-stage renal disease (ESRD) turned out to be normal. 4's Therefore, taking into account the reduced blood supply o f the kidneys and the low renal oxygen consumption of <-5 to 10 mL/min, the normal resting metabolic rate (R.MR) suggests an increased EE per body cell mass. If this increased EE was caused by the progression of renal failure, an increasing EE should be expected with its progression. Few data are available in the current literature about EE in patients with CRF. It is assumed that absolute energy consumption is in the normal range, whereas energy intake is below the patients' actual requirements. This might contribute to altered morbidity and mortality in these patients. According to this assumption, dietary prescriptions for these patients are outlined in the literature. It was the aim of this study to test the impact of C R F on the R M R in cross-sectional Journal qf Renal :\'uttitio,, Vol 11, No 4 (October), 2001: pp 202-206

METABOLIC RATE IN CHRONIC RENAL FAILURE

203

and longitudinal studies o f patients with preESRD.

chronic infections, malignoma, or other severe diseases were excluded from this study, as well as those on /3-receptor blocking agents or thyroid hor-

Patients and Methods Patients

nlones.

There were 51 patients included in this study (31 men and 20 women, age 53.2 + 13.9 y). O f these patients, 22 had decreasing kidney function during follow-up and were retested twice (at 3 months and at 6 months). All patients had impaired kidney function with a creatinine clearance (Ccr) between 6 and 60 mL/min, blood urea concentrations between 12 and 22 mmol/L, and did not show uremic symptoms. N o patients suffered from acute inflammation of the kidneys or o f any other organ. Their hemoglobin concentrations were 10.8 + 0.92 g/dL, and none o f them received recombinant human erythropoietin. Kidney function had declined slowly for at least 3 months before the start of the study. The patients' mean height was 168.73 + 5.93 cm and the mean body weight was 68.03 + 6.96 kg. The underlying diseases were chronic glomemlonephritis (n = 26), obstructive nephropathy (n = 7), analgesic nephropathy (n = 3), urogenital tuberculosis (n = 1), polycystic kidney disease (n = 3), and diabetic nephropathy (n = 11). Patients with uncontrolled hypertension, insulin-dependent diabetes mellitus, thyroid dysfunction (abnom~ FT3/FT4), acute or

Laboratory Tests In this study, the tkMI( was defined as the heat loss of a human being resting comfortably for at least 30 minutes in a thermoneutral environment. The measurements were performed at 7 AM. Air pressure (1,003 -+ 4 hPa) and relative humidity (51% + 3%) were monitored. Air velocity was <0.2 m/s - t at 0.3 to 0.4 m above the skin surface. R o o m temperature was 26.4 ° -+ 0.5°C and was kept constant within 0.3°C at least 3 hours before the start of the R M R measurement and until the end o f the measurement. Physical activity was to be normal for 3 days before testing and the patients were asked to refrain from any unusual physical effort and to not have any food intake within 12 hours before testing. R M R was assessed by direct calorimetry by using a metabolic beam scale (Brookline Instruments, N e w York, NY) to determine evaporative heat loss and by high-sensitivity thermistors fixed to the patients' skin surface to determine dry heat loss caused by radiation, convection, and conduction. Skin and core temperatures were digitally registered and calculated by special software and equipment (Boehnig & KaUenbach, Dortmund,

Figure 1. The equipment for the procedure of direct calorimetry was constructed as a bed with the subject in a supine position on an insulating mattress. Dry heat loss was calculated from the signals of 7 thermistors that were attached to the skin, an eighth sensor inserted into the rectum, and a ninth sensor that simultaneously registered ambient temperature. A moveable electric coil of a beam scale, which converted the total relative mass loss to an electric signal, evaluated evaporative heat loss. An analogue-to-digital converter digitized the signals of the thermistors and of the scale. W'~h special software, an interfaced personal computer assisted in performing the measurement and in calculating heat losses caused by evaporation, radiation, convection, and conduction.

204

KUHLMANN ET AL

60 [min] 50

40

31r=

30

X

20

10

-25

-20

-15

-10

-0'5

O0 +15

+10 +15 4"20 +25

[g]

Figure 2. A recording of the total loss of mass versus time for 30 minutes. The subject's initial body mass was counterbalanced by the scale countermass and an additional 15 g standard calibration mass. The weighing test procedure was started at + 18 g. The subject's mass decreased from + 18 g to + 3 g within 30 minutes. After 4 minutes, and again after 23 minutes, 2 periods of mass calibration followed, whereby the standard mass of 15 g was removed for 5 minutes. The computer software automatically performed the graphic analyses of mass calibration and mass loss.

Germany) (Figs 1 and 2). The sum of both evaporative heat loss and dry heat loss is the total heat loss, which represents the R M R in the standardized conditions o f the testing laboratory. A reliable reproducibility o f RMP,, determination (0.3% 4. 3.6%) was achieved by this device at repeated R M R measurements within a 3-hour interval. 6 The original data computed by the calorimetry device are given in kcal/kg/day. According to the SI, the data o f the Lange et al6 study were converted in w / m a (1 watt = 86.4 kJ/d, 1 kJ = 0.239 kcal). Following our clinical practice, the data in this article are given in kJ/kg/d. T o determine kidney function, C o was measured by taking 24-hour urine collections and blood samples on the day of the R M R determination.

Statistical Analysis T o compare the samples, the Mann-Whitney rank-sum test was used for unpaired data and the Wilcoxon test was used for paired data. The correlation obtained from normal distributed data was analyzed by the product-moment correlation test. Significance was accepted as P < 0.05. Confidence intervals (CI) are given at alpha =

0.05. The study was carried out in accordance with the guidelines o f the Declaration of Helsinki and fulfdled the requirements o f good clinical practice.

Results Cross-Sectional Study The mean Ccr was 18.69 + 11.98 mL/min (CI, + 3.29), ranging between 6.9 and 52 mL/ min. The mean R M R was 100.0 + 4.96 kJ/kg/d (CI, 4. 1.36), which is equivalent to 6,787.2 4. 609 kJ/d (CI, 4- 184), or 4.582 4- 0.181 kJ/1.73 m2/min (CI, -+ 0.049). The correlation between R M R per body weight ( R M R / B W ) and Ccr was r = - . 7 6 3 (P < .001) (Fig 3). The correlation between R M R per body surface area (RMR/BSA) and Ccr was r = - . 5 5 7 (P < .001).

Longitudinal Study In the 22 patients who participated in the longitudinal study, CCr decreased significantly (P < .001) from 27.5 ± 9.5 mL/min (CI, 4. 3.97) initial/y, 19.4 4. 6.25 mL/min (CI, 4. 2.61) at 3 months, to 13.0 + 3.8 m L / m i n (CI: 4. 1.61)

205

METABOLIC RATE IN CHRONIC RENAL FAILURE Ccr (mllmin) 51 o 15 oO

oo o

o

25 35

o

[] o

[]

o~ °

~

o

~ o o o

°°

[] o

o

o []

[]

o

45 55 92,5

%

p < 0 001

I 97,5

102,5 107,5 RMR(kJIkglday)

112,5

Figure 3. Correlation of the results between RMR and glomerular filtration rate in 51 patients with CRF.

at 6 months. The mean values for R M R / B W were 98.28 + 6.3 kJ/kg/d (CI, + 2.64), 101.64 -+ 5.46 kJ/kg/d (CI, +-- 2.64; P < .005), and 105.42 + 6.3 kJ/kg/d (CI, + 2.58; P < .001), respectively (Fig 4). Calculated in relation to body surface area, the corresponding R M R values were 4.41 + 0.126 kJ/1.73m2/min (CI, + 0.05), 4.578 + 1.68 kJ/1.73m2/min (CI, + 0.70; P < .05), and 4.704 + 0.168 kJ/1.73m2/min (CI, + 0.07; P < .05), respectively. The anthropometric data for weight and BSA were of no significant difference in all groups (data not shown).

Discussion The R M R in patients with C R F was assumed to decline in accordance with the decreasing oxygen consumption of the shrinking kidneys. Contrary to this assumption and for unknown reasons, the EE of patients on dialysis was not decreased. 4 If the increased EE of uremic subjects was caused by the impact of uremia, a close relationship between the continuous deterioration of kidney function and the increasing R M R should be obtained. To compare total EE and kidney function, cross-sectional and longitudinal studies were performed by direct bedside calorimetry in 51 outpatients suffering from preESRD. In patients with CRF, increasing R M R / B W corresponded significantly with decreasing kidney function (Figs 3 and 4). In contrast, the R M R / B S A only increased slightly and the corresponding correlation between R M R / B S A and CCr was lower (r = --.557, P < .001). With declining body weight in CRF, lean body mass in relation to total body mass has been shown to rise.

Assuming the R M R rises with the progression of CRF, not only a correlation between renal function and R M R / B W but also with R M R / B S A has to be expected. Additionally, even slightly increasing R M R / B S A reflects the slightly increased EE per cell mass in contrast to the expected decline following the decreased renal oxygen consumption in progressive C R F , suggesting an increased metabolic rate in uremia. Similar results were obtained by recalculating earlier studies of indirect calorimetry for correlation of the degree of C R F with EE, expressed as kJ/kg/24 hr. 7 Although the final reason for the unexpected high R M R in chronic ESRD is still unknown, there are some alterations of the metabolism in uremia that could enhance the EE. The diminished oxygen capacity of anemia requires an increased cardiac output of the patient with uremia exercising like healthy subjects. 8"9 The correspondingly increased peripheral blood flow might contribute to the increased heat loss in ESRD. The reduced life span of red blood cells in uremia, and the resulting increase in erythrocyte synthesis and degradation, may lead to an increase in liver oxygen consumption. Metabolic acidosis and uremic toxins have been found to interfere with glycolytic metabohsm and oxygen consumption, subsequently increasing the EE of the muscle tissue. 1°-12 Hepatocyte gluconeogenic pathways and ureagenesis are affected by uremia, and an enhancement of adenosine triphosphate demand is assumed. 13 Furthermore, microcalorimetric studies show an increase of the heat production by human plasma in patients with uremia compared with that of normal plasma. 14 But nevertheless, the final reasons for the increased R M R in uremia remain to be elucidated.

p<0.001 start

p<0.001] 3rd month

6th month

35 30 25 20 15 10 • Ccrea (mllmin)

95 100 105 RMR(kJIkgBWlday)

110

Figure 4. Results of 22 patients with decreasing glomerular filtration rate and rising RMR initially, after 3 months, and after 6 months of observation.

KUHLMANN ET AL

906

References 1. Allman MA, Stewart PM, Tiller DJ, et al: Energy supplementation and the nutritional status of hemodialysis patients. A m J Clin Nutr 51:558-562, 1990 2. Bansal VK, Popli S, Pickering J, et al: Protein-calorie malnutrition and cutaneous anergy in hemodialysis maintained patients. A m J Clin Nutr 33:1608-1611, 1980 3. Ciba-Geigy: Wissenschaftliche TabeUen Geigy, Teilband K6rperfliissigkeiten, 8 Auflage, Basel, Switzerland, 1977, pp 225 4. Lange H, Krautwald E, Krautwald G, et al: The effect of extracorporeal haemodialysis on energy turnover. Proc Eur Dial Transplant Assoc 22:106-110, 1985 5. Schneeweiss B, Graninger W, Stockenhuber F, et al: Energy metabolism in acute and chronic renal failure. A m J Clin Nutr 52:596-601, 1990 6. Lange H, G~'ber T, Schwickardi M: A simplified procedure of direct calorimetry for bedside monitoring of the resting metabolic rate. EurJ Appl Physiol 71:58-64, 1995 7. Monteon FJ, Laidlaw SA, Shaib JK, et al: Energy expenditure in patients with chronic renal failure. Kidney Int 30:741747, 1986

8. Lange H, Janssen MT, ThiiroffJ, et al: Blood levels of lactate, pyruvate, azetazetate and beta-hydroxybutyrate in physically exercising patients with varying degrees of anemia undergoing chronic hemodialysis. Nieren- und Hochdruckkrankheiten 5:215-224, 1987 9. Jedicke H, Lange H: Endurance capacity of dialysis patients with renal anemia under erythropoietin administration. Nieren- und Hochdruckkrankheiten 12:690-696, 1991 10. Del Canale S, Fiaccadori E, Ronda N, et al: Muscle energy metabolism in uremia. Metabolism 35:981-983, 1986 11. Mitch WE, May RC, Maroni BJ, et al: Protein and amino acid metabolism in uremia: influence of metabolic acidosis. Kidney Int 27:205-207, 1989 (suppl) 12. Conjard A, Ferrier B, Martin M, et al: Effects o f chronic renal failure on enzymes of energy metabohsm in individual human muscle fibers. J Am Soc Nephrol 6:68-74, 1995 13. Cano N, Catelloni F, Fontaine E, et al: Isolated rat hepatocyte metabolism is affected by chronic renal failure. Kidney Int 47:1522-1527, 1995 14. Ljunggren L, Monti M, Thysell H, et al: Microcalorimetric studies on uraemic plasma. Scand J Chn Lab Invest 52:813817, 1992