Hormonal responses to fasting and refeeding in chronic renal failure ...

1 downloads 0 Views 327KB Size Report
Aug 10, 2004 - appetite-related hormones in response to fasting and refeeding ... with hypertensive nephrosclerosis, one with Alport's disease, two with focal ...
Am J Physiol Endocrinol Metab 288: E47–E55, 2005. First published August 10, 2004; doi:10.1152/ajpendo.00163.2004.

Hormonal responses to fasting and refeeding in chronic renal failure patients Sonali Deshmukh,1 Bradley G. Phillips,2 Thomas O’Dorisio,1 Michael J. Flanigan,1 and Victoria S. Lim1 1

Department of Medicine, College of Medicine, and 2College of Pharmacy, University of Iowa, Iowa City, Iowa

Submitted 8 April 2004; accepted in final form 5 August 2004

Deshmukh, Sonali, Bradley G. Phillips, Thomas O’Dorisio, Michael J. Flanigan, and Victoria S. Lim. Hormonal responses to fasting and refeeding in chronic renal failure patients. Am J Physiol Endocrinol Metab 288: E47–E55, 2005. First published August 10, 2004; doi:10.1152/ajpendo.00163.2004.—To study anorexia in chronic renal failure (CRF) patients, we measured appetite-related hormones in seven CRF patients and four controls. Plasma concentrations and fractional changes from baseline (values from day 1, 0800) are listed as control vs. CRF (means ⫾ SE). Leptin, although higher in CRF (5.6 ⫾ 1.7 and 34 ⫾ 17 ng/ml), was suppressed after fasting; decrements were ⫺51 ⫾ 9 and ⫺55 ⫾ 8%. Nocturnal surge present during feeding was abolished upon fasting in both groups. Neuropeptide Y (NPY) was elevated in CRF (72 ⫾ 12 vs. 304 ⫾ 28 pg/ml, P ⫽ 0.0002). NPY rhythm, reciprocal to that of leptin, was muted in CRF. Basal cortisol was similar in both groups (17 ⫾ 3 and 17 ⫾ 2 ␮g/dl). In the controls, cortisol peaked in the morning and declined in the evening. CRF showed blunted cortisol suppression. Decrements were ⫺61 ⫾ 3 and ⫺20 ⫾ 9% at 1800 on day 1 (P ⫽ 0.008) and ⫺61 ⫾ 8 and ⫺26 ⫾ 8% at 2000 on day 2 (P ⫽ 0.02). Basal ACTH (25 ⫾ 5 and 54 ⫾ 16 pg/ml) as well as diurnal pattern was not statistically different between the groups. Baseline insulin was 6 ⫾ 1 and 20 ⫾ 9 ␮U/ml. During fasting, insulin was suppressed to ⫺64 ⫾ 10 and ⫺51 ⫾ 9%, respectively. Upon refeeding, increments were 277 ⫾ 96 and 397 ⫾ 75%. Thus, in our CRF patients, anorexia was not due to excess leptin or deficient NPY. Impaired cortisol suppression should favor eating. Insulin suppression during fasting and secretion after feeding should enhance both eating and anabolism. The constant high NPY suggests increased tonic hypersecretion.

in CRF patients. The results were compared with those obtained from normal subjects. We wondered whether CRF patients may have too much leptin or perhaps their leptin is not suppressed with food deprivation. We considered that these patients might be deficient in neuropeptide Y (NPY), a potent orexigenic peptide that mediates the effect of leptin in the hypothalamus (11). Knowing that cortisol is essential for eating in mammals and that corticotropin-releasing hormone is a potent anorexigenic peptide (12, 30), we decided to profile cortisol and ACTH, the latter as a surrogate to corticotropinreleasing hormone. We also studied insulin, because serum insulin level tends to be higher in CRF patients, and insulin, in the central nervous system, suppresses eating (33). METHODS

in chronic renal failure (CRF) patients, and anorexia is likely an important contributing factor (3). It has been reported that food consumption declines as renal function deteriorates (14). When uremia becomes severe, anorexia is believed to be prevalent and is characterized by food aversion, early satiety, and, sometimes, nausea and vomiting. Despite the prevalence of these symptoms, their pathogenesis has never been explained. We hypothesized that uremia-related anorexia may result from hypothalamic dysfunction. Renal failure patients are known to have a number of hormonal abnormalities due to hypothalamic disturbances; examples include diminished pulsatile gonadotropin and growth hormone secretion, lack of positive feedback effect of estrogen on LH secretion, excessive prolactin production, and delayed puberty (18). Because eating is, by and large, a central nervous system-regulated event (11), we studied some of the appetite-related hormones in response to fasting and refeeding

Subjects. Seven CRF patients and four normal subjects were recruited. Table 1 lists the demography, nutritional status, and baseline hormonal profile of the two study groups. Table 2 includes the clinical parameters and modalities of treatment as well as letpin data individually in the seven CRF patients. The normal controls consisted of one woman and three men aged 32– 63 yr. The CRF group consisted of two predialysis, four hemodialysis, and one peritoneal dialysis (PD) patients. There were six men and one woman aged 31–56 yr. The two predialysis patients were initiated into maintenance dialysis because of high serum creatinine; they did not specifically complain of anorexia. Diagnoses of the CRF patients included two with hypertensive nephrosclerosis, one with Alport’s disease, two with focal segmental glomerulosclerosis, and one each with interstitial nephritis and immunotactoid glomerulopathy. No patients had diabetes or other concomitant systemic disease. All studies were conducted when the subjects were clinically stable and without intercurrent illness. Hemodialysis was performed three times weekly using CT190G dialyzers; KT/V ranged from 1.3 to 1.5. The single PD patient used the Fresenius PD plus system; he did five exchanges nocturnally plus one exchange during the daytime. His weekly KT/V was 2.75 and creatinine clearance 53 l/wk. Upon recruitment, subjects made one visit to the General Clinical Research Center (GCRC) dietitian for a dietary interview and anthropometric measurements. A subjective global assessment was also performed at that time. A menu was then created for each subject on the basis of his/her routine intake and preference. The reason for this dietary step was to ensure that the participants were eating regularly for at least 2 days before the study. Figure 1 is a time line representation of the experiment, showing that subjects ate the GCRCprepared diet on day ⫺1 and day 0. They were admitted to the GCRC in the afternoon of day 0, when a heparin lock was placed for blood draws and a snack was given at 2000. After this, food was withheld for 37 h until 0900 of day 2. During the fast, water was allowed ad libitum. Subjects were encouraged to maintain their routine physical activities as much as permitted by the schedule of the blood draws.

Address for reprint requests and other correspondence: V. S. Lim, Div. of Nephrology, Dept. of Medicine, T310, GH, Univ. of Iowa Hospitals, 200 Hawkins Dr., Iowa City, IA 52242 (E-mail: [email protected]).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

hormones; appetite; uremia

MALNUTRITION IS PREVALENT

http://www.ajpendo.org

0193-1849/05 $8.00 Copyright © 2005 the American Physiological Society

E47

E48

FASTING AND REFEEDING IN CRF PATIENTS

Table 1. Demography, nutritional status, and baseline serum hormone levels in study subjects Patient

Normal 1 2 3 4 Mean ⫾ SE CRF 1 2 3 4 5 6 7 Mean ⫾ SE P value

Age, yr/Sex

BMI

%Fat

Leptin, ng/ml

NPY, pg/ml

Cortisol, ␮g/dl

ACTH, pg/ml

Insulin, ␮U/ml

Glucose, mg/dl

32/M 53/M 63/F 40/F 47⫾7

25 25 21 24 23.8⫾1

18 21 25 19 20.8⫾1.6

6.2 3.6 10.0 2.5 5.6⫾1.7

61 63 108 56 72⫾12

10.2 19.9 23.4 12.9 16.6⫾3.1

22.9 36.9 28.9 10.7 24.9⫾5.5

5.2 5.9 8.9 5.4 6.4⫾0.9

106 89 90 92 94⫾4

31/M 36/M 41/F 56/M 51/M 36/M 52/M 42⫾4 0.51

29 21 27 20 26 22 35 25.7⫾2 0.50

26 14 32 17 17 11 29 20.9⫾3.1 0.98

66.7 4.7 124.5 3.6 5.1 4.7 29.0 34.1⫾17.4 0.259

381 255 403 297 297 184 309 304⫾28 0.0002

13.5 9.5 19.4 20.3 12.1 21.8 18.8 16.5⫾1.8 0.973

24.1 94.4 18.8 56.5 23.0 130.0 32.5 54.2⫾16.2 0.220

62.5 3.1 8.0 5.2 14.4 7.0 40.1 20.0⫾8.5 0.267

105 88 100 89 90 120 102 99⫾4 0.473

CRF, chronic renal failure; NPY, neuropeptide Y. Baseline hormone concentrations were obtained from day 1, 0800. CRF patients 1– 4 were on maintenance hemodialysis, 5 was on peritoneal dialysis, and 6 and 7 were studied a few days before maintenance hemodialysis.

Refeeding then followed and subjects were studied until 0800 of day 3. During refeeding, the menu remained constant, but portion sizes were ad libitum per subjects’ requests. The amount of food ingested before fasting and during refeeding was recorded. Blood was taken at 2000 on day 0, and at 0000, 0300, 0800, 1300, 1800, and 2000 on the other days. All blood samples were centrifuged immediately, and the plasma was frozen at ⫺60°C and thawed at the time of assay. Leptin, neuropeptide Y (NPY), cortisol, and ACTH were measured in all samples, whereas insulin and glucose were measured only in samples as indicated in Fig. 4. Visual analog scales (VASs) assessing appetite and hunger were obtained at designated times during waking hours as depicted in Fig. 5. Laboratory procedures. Leptin was measured by radioimmunoassay (RIA) using Linco kits (St. Charles, MO). The reagent has 100% specificity for human leptin. NPY was measured using RIA kits purchased from Peninsula Laboratories (San Carlos, CA). The antibody has a 100% cross-reactivity to human NPY; it has 0% and 3% cross-reactivity against human peptide YY (PYY) and human pancreatic polypeptide, respectively. Cortisol and insulin were measured by electrochemiluminescence immunoassay using the Elecsys system (Roche Diagnostics, Indianapolis, IN). ACTH was measured by a two-site chemiluminescence immunoassay using both a mouse monoclonal antibody and a goat polyclonal antibody. The former was

labeled with acridinium ester and the latter with biotin. Streptavidincoated magnetic particles were added, resulting in the formation of solid-phase particles. The solid-phase magnetic particles were then transported into the system luminometer and quantitated (Nichols Advantage). Appetite assessment. We assessed appetite, hunger, and satiety with VAS (27). We utilized five questions. How hungry do you feel? How full does your stomach feel? Do you feel nauseated? How strong is your desire to eat? How much food you think you can eat? Under each question is a line 10 cm in length (scale 0 –10) with words anchored at each end expressing the most negative sensation on the left, and the most positive one on the right. For example, if the question is “How hungry do you feel?” not hungry at all would be a score of 0 and extremely hungry a 10. Nutritional assessment. Anthropometry and subjective global assessment were used to assess the nutritional status of each participant. Anthropometry included height, weight, elbow breadth, midarm circumference, triceps skin fold, suprailiac skin fold, abdominal skin fold (vertical fold 2 cm lateral from the umbilicus), and waist and hip circumference. A trained nutritionist in the GCRC took these measurements. Percent body fat was estimated from these measurements. Subjective global assessment is a clinical technique used to assess nutritional status based on features of the history and physical exam-

Table 2. Diagnosis, treatment modality, clinical parameters, basal plasma leptin, and fractional changes in CRF patients during fasting and refeeding Fractional Changes, % Fasting Patient

Diagnosis

Rx

Serum Cr, mg/dl

1 3 7 2 4 5 6 Control*

HTN IN HTN FSGS Alport’s ITG FSGS

HD HD Pre HD HD PD Pre

14.4 9.2 13.3 11.9 8.3 13.5 16.5

Hb, g/dl

CRP, mg/dl

Serum Albumin, g/dl

Basal Leptin, ng/ml

11.0 13.4 10.3 10.5 14.6 10.6 9.0

⬍0.5 ⬍0.5 ⬍0.5 ⬍0.5 ⬍0.5 0.7 ⬍0.5

4.3 4.2 3.9 3.9 4.3 4.2 4.2

66.7 124.5 29.0 4.7 3.6 5.1 4.7

Feeding

Day 2, 0300

Day 2, 0900

Day 3, 0000

Day 3, 0300

⫺30 ⫺68 ⫺80 ⫺44 ⫺47 ⫺45 ⫺55 ⫺52⫾11

⫺29 ⫺78 ⫺84 ⫺45 ⫺49 ⫺41 ⫺56 ⫺51⫾9

⫺21 ⫺40 9 21 45 22 29 61⫾10

⫺6 ⫺34 30 36 69 67 43 82⫾14

Values for controls are means ⫾ SE. Diagnoses: HTN, hypertensive nephrosclerosis; IN, interstitial neprhitis; FSGS, focal segmental glomerulosclerosis; Alport’s, Alport’s disease; ITG, immunotactoid glomerulopathy. Rx, modality of treatment: HD, hemodialysis; Pre, predialysis; PD, peritoneal dialysis. Cr, creatinine; Hb, hemoglobin concentration; CRP, C-reative protein. Basal leptin levels were obtained on day 2, 0800. Fractional changes were derived from these baseline values. Patients 1, 3, and 7 had high basal leptin, and their nocturnal surge during feeding was markedly blunted. Patients 2, 4, 5, and 6 had low basal leptin, and their nocturnal surge was brisk, similar to those of controls, represented by *. Fractional decrement during fasting was comparable in all, controls as well as all CRF patients. AJP-Endocrinol Metab • VOL

288 • JANUARY 2005 •

www.ajpendo.org

E49

FASTING AND REFEEDING IN CRF PATIENTS

Fig. 1. Schematic representation of the experiment. On day ⫺1 and day 0, subjects ate a fixed General Clinical Research Center (GCRC)-prepared diet. On the afternoon of day 0, subjects were admitted to the GCRC, a heparin lock was placed, and a snack was given at 2000. Fasting started from 2000, day 0, until 0900 of day 2 (total 37 h). Refeeding extended from day 2, 0900, to day 3, 0800. Cross-hatched bars represent periods when subjects were not eating. During the study period, blood samples were obtained at 2000, 0000, 0300, 0800 (0900 on day 2), 1300, and 1800. Visual analog scales for assessing appetite were obtained during waking hours.

ination (8) The former includes weight change, dietary change, and gastrointestinal symptoms, and the latter includes evidence of subcutaneous fat loss, muscle wasting, and edema or ascites. The scores are A, well nourished; B, moderately malnourished; C, severely malnourished. RESULTS

Table 1 summarizes and compares the demography, nutritional status, and baseline hormonal profiles of the two study groups. All values are listed in the sequence of normal subjects and CRF patients, and statistically significant differences are listed in parenthesis. Age was 47 ⫾ 7 and 42 ⫾ 4 yr, body mass index (BMI) 23.8 ⫾ 1.0 and 25.7 ⫾ 2.0 kg/m2, and percent body fat 20.8 ⫾ 1.6 and 20.9 ⫾ 3.1%. Plasma leptin was 5.6 ⫾ 1.7 and 34.1 ⫾ 17.4 ng/ml, NPY was 72 ⫾ 12 and 304 ⫾ 28 pg/ml (P ⫽ 0.0002), cortisol was 16.6 ⫾ 3.1 and 16.5 ⫾ 1.8 ␮g/dl, and ACTH was 24.9 ⫾ 5.5 and 54.2 ⫾ 16.2 pg/ml. Plasma insulin was 6.4 ⫾ 0.9 and 20.0 ⫾ 8.5 ␮U/ml, and glucose was 94 ⫾ 4 and 99 ⫾ 4 mg/dl. Not listed in Table 1 are energy and protein intakes of the study subjects. Energy intake was 31.9 ⫾ 1.8 and 31.2 ⫾ 2.8 kcal䡠kg⫺1 䡠day⫺1 before fasting and 35.7 ⫾ 2.5 and 39.7 ⫾ 2.8 kcal䡠kg⫺1 䡠day⫺1 after the fast. Protein intake was 1.24 ⫾ 0.08 and 0.89 ⫾ 0.06 g䡠kg⫺1 䡠day⫺1 (P ⫽ 0.006) before the fast, higher in the normal subjects, and 1.48 ⫾ 0.16 and 1.19 ⫾ 0.11

g䡠kg⫺1 䡠day⫺1 after the fast. Although both study groups increased food intake postfasting, only in the CRF patients were the differences between pre- and postfast statistically significant; P values were 0.05 and 0.03, respectively, for calorie and protein intake. Although mean leptin, ACTH, and insulin values were higher in CRF patients, the differences did not reach statistical significance. This is because high values were present only in a few CRF patients. As shown in Tables 1 and 2, higher leptin and insulin values were found in three patients who have higher percent body fat. The highest leptin, 124, was found in a woman with 32% body fat. In the case of ACTH, all three high values were found in lean patients. In contrast, NPY was homogenously and markedly elevated in all CRF patients. Serum creatinine was 12.4 ⫾ 1.1 mg/dl and albumin 4.1 ⫾ 0.06 g/dl in the CRF patients. Leptin-NPY axis. Plasma leptin during the experimental periods are listed in the left half of Table 3. In normal subjects, on day 1, 0300, leptin peaked to 7.7 ⫾ 2.2 ng/ml; on the same day at 0800 (baseline), the value was 5.6 ⫾ 1.7 ng/ml. On day 3, 0300, leptin reached a height of 10.2 ⫾ 3.5 ng/ml. Leptin peaked at nocturnal times on both days. During the 37-h fast, nocturnal surge was totally obliterated, and plasma leptin

Table 3. Plasma leptin and NPY concentrations in normal subjects and CRF patients at different days/times of the experiment Leptin, ng/ml

NPY, pg/ml

Day/time

Normal

CRF

P

Normal

CRF

P

0/2000 1/0000 0300 0800* 1300 1800 2000 2/0000 0300 0900 1300 1800 2000 3/0000 0300 0800

5.3⫾1.4 7.1⫾2.2 7.7⫾2.2 5.6ⴞ1.7 4.3⫾1.1 3.3⫾0.7 2.9⫾0.6 2.5⫾0.4 2.3⫾0.4 2.5⫾0.6 3.2⫾0.8 5.4⫾2.3 5.8⫾2.1 9.0⫾3.0 10.2⫾3.5 6.6⫾1.7

24.9⫾12.1 36.4⫾17.7 32.3⫾15.5 34.1ⴞ17.4 24.0⫾11.7 22.7⫾12.1 19.3⫾9.8 15.6⫾7.9 14.4⫾7.5 12.7⫾6.7 10.5⫾4.9 15.5⫾6.6 19.6⫾8.3 25.9⫾10.6 30.3⫾12.0 33.6⫾14.7

0.26 0.25 0.27 0.26 0.24 0.27 0.25 0.25 0.26 0.29 0.31 0.29 0.25 0.28 0.25 0.21

54⫾11 51⫾8 55⫾5 72ⴞ12 62⫾14 47⫾7 39⫾4 28⫾8 29⫾8 56⫾15 60⫾15 60⫾16 53⫾16 44⫾11 49⫾11 73⫾7

309⫾29 286⫾29 295⫾26 304ⴞ28 309⫾19 288⫾20 267⫾23 242⫾24 234⫾23 262⫾23 286⫾29 260⫾27 262⫾36 245⫾31 249⫾28 289⫾23

0.0001 0.0002 ⬍0.0001 0.0002 ⬍0.0001 ⬍0.0001 ⬍0.0001 0.0001 0.0001 0.0001 0.0003 0.0007 0.0025 0.0011 0.0005 ⬍0.0001

Values are means ⫾ SE. Fasting extended from day 0, 2000 until day 2, 0900 (37 h) and refeeding from day 2, 0900 to day 3, 0800. *Baseline levels obtained at 0800, day 1. P, P values comparing normal with CRF subjects, derived from Student’s t-test. AJP-Endocrinol Metab • VOL

288 • JANUARY 2005 •

www.ajpendo.org

E50

FASTING AND REFEEDING IN CRF PATIENTS

dipped to a nadir of 2.3 ⫾ 0.4 ng/ml at 0300, day 2, and remained low at 2.5 ⫾ 0.6 ng/ml at 0900, day 2, when the fast ended. In the CRF patients, plasma leptin was higher than in the controls at all times during the experiment, but the difference never achieved statistical significance. Baseline leptin was 34.1 ⫾ 17.4 ng/ml. Nocturnal rise was less prominent compared with the controls. On day 1, plasma leptin increased slightly to 36.4 ⫾ 17.7 ng/ml at 0000, and on day 3 the highest nocturnal value was only 30.3 ⫾ 12.0 ng/ml at 0300. During the 37-h fast, plasma leptin did fall to 12.7 ⫾ 6.7 ng/ml at 0900 on day 2 and remained low, 10.5 ⫾ 4.9 ng/ml, at 1300 that same day. Figure 2A illustrates the fractional changes of plasma leptin at different experimental times compared with the baseline. Baseline values for all the peptides/hormones in this study were the values obtained on day 1, 0800. In the normal subjects, plasma leptin reached 47 ⫾ 24% above baseline at 0300, day 1. During the refeeding period, nocturnal surge reached levels of 61 ⫾ 10 and 82 ⫾ 14% above baseline, respectively, at 0000 and 0300, day 3. During fasting, nocturnal surge was totally abolished, and leptin dropped to a nadir of ⫺53 ⫾ 11% below baseline at 0300, day 2, and was ⫺51 ⫾ 9% of baseline at the end of the fast at 0900, day 2. In the CRF patients, fasting also effectively suppressed leptin; nadir at the end of the fast was ⫺55 ⫾ 6% at 0900, day 2. The nocturnal surge during regular feeding was, however, blunted. On day 1, the highest leptin increment was 13 ⫾ 11%, noted at 0300. On day 3, during refeeding, the increments were 9 ⫾ 11% at 0000

Fig. 2. Circadian variation of plasma leptin and neuropeptide Y (NPY) and their changes to fasting and refeeding. Fractional changes (%) of plasma leptin (A) and NPY (B) at designated times compared with baseline values obtained on day 1, 0800. Values are presented as means ⫾ SE. Solid lines and filled circles are derived from normal subjects, broken lines and open squares from patients with chronic renal failure (CRF). *P ⬍ 0.05 by Student’s t-test comparing the two study groups. Fasting, represented by the cross-hatched bar at the bottom of B began on day 0, 2000, and ended on day 2, 0900. Refeeding extended from day 2, 0900, to day 3, 0800. AJP-Endocrinol Metab • VOL

(P ⫽ 0.01) and 30 ⫾ 14% (P ⫽ 0.04) at 0300. As we scrutinized the data, it became evident that this blunted nocturnal surge found in CRF patients was present only in the three CRF patients who had high BMIs and whose basal leptin levels were high. Table 2 lists the CRF patients according to their basal serum leptin levels. Patients 1, 3, and 7 had high, whereas patients 2, 4, 5, and 6 had normal basal leptin levels. During feeding, nocturnal leptin surge was brisk and unequivocal in those patients with normal leptin but was blunted in those with high basal leptin. Fasting, on the other hand, suppressed leptin universally, whether basal levels were high or normal. Listed in the right half of Table 3 is plasma NPY at different times during the experiment. The values are about fivefold higher in the CRF patients at all times. In the normal subjects, higher values were found in the morning, between 0800 and 0900, the values were 72 ⫾ 12 pg/ml on day 1, 56 ⫾ 15 pg/ml on day 2, and 73 ⫾ 7 pg/ml on day 3. The nadir for these three days occurred at 0000, and the values were 51 ⫾ 8, 28 ⫾ 8, and 44 ⫾ 11 pg/ml, respectively. The nocturnal decline in NPY was exaggerated during fasting, 28 pg/ml on day 2 vs. 51 and 44 pg/ml on days 1 and 3, respectively. This NPY circadian rhythm was much muted in the CRF patients. Corresponding morning values for days 1, 2, and 3 were, respectively, 304 ⫾ 28, 262 ⫾ 23, and 289 ⫾ 23 pg/ml, and values at 0000 for the same 3 days were 286 ⫾ 29, 242 ⫾ 24, and 245 ⫾ 31 pg/ml. Figure 2B shows the fractional changes of plasma NPY at different experimental times compared with baseline. In the normal subjects, during feeding, NPY diurnal rhythm was opposite to that of leptin: lowest in the night and reaching peak values in the morning, from 0800 to 0900. The circadian rhythm of the CRF patients was qualitatively similar but muted. Fractional decrements of NPY in the controls and CRF patients were, respectively, ⫺29 ⫾ 7 and ⫺5 ⫾ 5% (P ⫽ 0.02) at 0000, day 1, and ⫺41 ⫾ 8 and ⫺18 ⫾ 8% at 0000, day 3. During fasting, NPY declined in both groups but more sharply in the controls; decrements were ⫺64 ⫾ 5 and ⫺19 ⫾ 6% (P ⫽ 0.0005) at 0000 and ⫺61 ⫾ 7 and ⫺22 ⫾ 3% (P ⫽ 0.0005) at 0300, day 2. Cortisol-ACTH axis. Table 4 details the plasma cortisol and ACTH response of the two study groups to experimental fasting and refeeding. At baseline, on day 1, 0800, plasma cortisol levels were similar in the two groups, 17 ⫾ 3 and 17 ⫾ 2 ␮g/dl. Cortisol levels were higher in the CRF patients during the following days/times: day 0, 2000, 4.3 ⫾ 0.4 vs. 10.8 ⫾ 1.2 ␮g/dl (P ⫽ 0.004); day 1, 1800, 6.4 ⫾ 1.1 vs. 12.3 ⫾ 0.7 ␮g/dl (P ⫽ 0.001), and 2000, 5.0 ⫾ 0.6 vs. 10.0 ⫾ 1.1 ␮g/dl (P ⫽ 0.01); day 2, 1800, 6.6 ⫾ 1.3 and 11.4 ⫾ 0.9 ␮g/dl (P ⫽ 0.01) and 2000, 6.0 ⫾ 0.9 and 11.8 ⫾ 1.5 ␮g/dl (P ⫽ 0.02). These data indicate that CRF patients had a blunted circadian suppression of serum cortisol. ACTH values were higher in CRF patients at all times, but only that of day 0, 2000, reached statistically significant difference: 9.8 ⫾ 1.1 and 27.8 ⫹ 5.5 pg/ml (P ⫽ 0.04). Particularly, higher values were present in patients 2, 4 and 6, who had relatively low body fat content (Table 1). The diurnal ACTH concentration pattern is similar in both study groups; levels tend to be higher in the morning hours and coincided with or shortly before the cortisol peaks. In the normal subjects, peak values were 24.9 ⫾ 5.5 pg/ml on day 1, 0800, 38.2 ⫾ 8.8 pg/ml on day 2, 0300, and 17.3 ⫾ 5.7 pg/ml on day

288 • JANUARY 2005 •

www.ajpendo.org

E51

FASTING AND REFEEDING IN CRF PATIENTS

Table 4. Plasma cortisol and ACTH concentrations in normal subjects and CRF patients at different days/times of the experiment Cortisol, ␮g/ml

ACTH, pg/ml

Day/time

Normal

CRF

P

Normal

CRF

P

0/2000 1/0000 0300 0800* 1300 1800 2000 2/0000 0300 0900 1300 1800 2000 3/0000 0300 0800

4.3⫾0.4 7.1⫾3.4 7.4⫾2.8 16.6ⴞ3.0 8.1⫾0.7 6.4⫾1.1 5.0⫾0.6 11.1⫾4.4 18.7⫾5.4 20.3⫾5.6 9.9⫾1.3 6.6⫾1.3 6.0⫾0.9 3.8⫾0.3 7.3⫾1.9 15.2⫾0.9

10.8⫾1.2 6.4⫾0.9 7.4⫾1.5 16.5ⴞ1.8 13.6⫾2.1 12.3⫾0.7 10.0⫾1.1 7.7⫾1.1 11.6⫾2.3 18.0⫾2.1 14.7⫾2.8 11.4⫾0.9 11.8⫾1.5 7.7⫾1.4 9.3⫾2.0 14.9⫾0.8

0.004† 0.81 0.99 0.97 0.09 0.001† 0.010† 0.35 0.20 0.66 0.25 0.012† 0.024† 0.076 0.53 0.86

9.8⫾1.1 11.2⫾1.8 17.4⫾2.7 24.9ⴞ5.5 17.1⫾3.0 11.9⫾2.3 10.3⫾1.8 18.8⫾3.7 38.2⫾8.8 18.9⫾3.5 11.9⫾2.2 11.6⫾2.3 10.2⫾2.0 10.1⫾1.4 17.3⫾5.7 13.7⫾3.2

27.8⫾5.5 19.6⫾5.1 27.1⫾5.4 54.2ⴞ16.2 36.7⫾9.7 32.5⫾8.4 24.2⫾5.1 26.3⫾7.7 28.2⫾4.8 50.7⫾18.4 31.7⫾6.8 27.2⫾7.3 25.7⫾6.7 20.2⫾5.0 30.6⫾4.9 42.3⫾14.1

0.040‡ 0.261 0.234 0.220 0.176 0.110 0.079 0.501 0.303 0.240 0.064 0.153 0.123 0.177 0.122 0.177

Values are means ⫾ SE. Fasting extended from day 0, 2000 until day 2, 0900 (37 h) and refeeding from day 2, 0900 to day 3, 0800. *Baseline levels obtained at 0800, day 1. P, P values comparing normal with CRF subjects, derived from Student’s t-test. †Higher cortisol (all in evening hours) and ‡higher ACTH in CRF patients.

3, 0300. Peak values in the CRF patients were 54.2 ⫾ 16.2, 50.7 ⫾ 18.4, and 42.3 ⫾ 14.1 pg/ml, respectively, on days 1, 2, and 3, all at 0800 – 0900. The peak tended to occur earlier, at 0300, in control subjects. Figure 3, A and B, illustrates the circadian rhythm of cortisol and ACTH. In the normal subjects, plasma cortisol was highest in the morning, 0800 – 0900, and lowest in the evening hours. Listed in the sequence of control vs. CRF patients, fractional decrements from baseline were ⫺70 ⫾ 7 and ⫺29 ⫾ 12% (P ⫽ 0.04) at 2000, day 0; ⫺61 ⫾ 3 and ⫺20 ⫾ 8% (P ⫽ 0.008) at 1800, day 1; ⫺59 ⫾ 8 and ⫺26 ⫾ 9% (P ⫽ 0.03) at 1800 and ⫺61 ⫾ 8 and ⫺26 ⫾ 8% (P ⫽ 0.02) at 2000, day 2. Evening cortisol suppression was blunted in CRF patients, and fasting did not alter the circadian cycle. As for ACTH, the peaks and troughs coincided with or preceded those of plasma cortisol. Diurnal rhythm was present and not different between controls and CRF patients with one exception, and that was on day 2, 0300, during fasting. At that time, plasma ACTH rose by 51 ⫾ 14% above baseline in the normal subjects and fell by ⫺34 ⫾ 14% in the CRF patients (P ⫽ 0.002). This apparent dramatic difference at that particular time was due to a coincidental combination of earlier onset of morning peak in the normal subjects at 0300 compared with 0900 in the CRF patients, combined with a higher baseline value in CRF subjects (Table 4). In the three CRF patients with very high baseline ACTH, their fractional diurnal changes from baseline were not different from those of the four patients with normal basal ACTH. Insulin-glucose axis. Table 5 lists the data on insulin and glucose during the experiment. Plasma insulin was higher in the CRF patients at all times, but none reached statistical significance. Insulin levels were highest in the two male patients with high leptin, but the female patient with very high leptin and increased body fat content had normal plasma insulin (Table 1). Fasting reduced plasma insulin from baseline of 6.4 ⫾ 0.9 to 1.2 ⫾ 0.3 ␮U/ml in normal subjects, and 20.0 ⫾ 8.5 to 6.8 ⫾ 3.2 ␮U/ml in CRF patients. Postfeeding, plasma insulin rose to 23.8 ⫾ 5.9 ␮U/ml in the controls and 85.5 ⫾ AJP-Endocrinol Metab • VOL

44.7 ␮U/ml in the CRF patients. Plasma glucose decreased from 94 ⫾ 4 to 77 ⫾ 7 mg/dl in the controls and from 99 ⫾ 4 to 87 ⫾ 5 mg/dl in the CRF patients during fasting and rose to 119 ⫾ 5 mg/dl in controls and 141 ⫾ 9 mg/dl in CRF patients during refeeding. Figure 4 illustrates insulin and glucose responses to fasting and feeding in the study groups. In normal controls, fractional

Fig. 3. Circadian variation of plasma cortisol and ACTH and their changes to fasting and refeeding. Fractional changes (%) of plasma cortisol (A) and ACTH (B) at designated times compared with baseline values obtained on day 1, 0800. Values are presented as means ⫾ SE. Solid lines and filled circles are derived from normal subjects; broken lines and open squares are from patients with CRF. *P ⬍ 0.05 by Student’s t-test comparing the two study groups. Fasting, represented by the cross-hatched bar at the bottom of B, began on day 0, 2000, and ended on day 2, 0900. Refeeding extended from day 2, 0900, to day 3, 0800. Shaded areas represent magnitude of cortisol and ACTH excess in CRF patients compared with normal subjects during the experiment.

288 • JANUARY 2005 •

www.ajpendo.org

E52

FASTING AND REFEEDING IN CRF PATIENTS

Table 5. Plasma insulin and glucose concentrations in normal subjects and CRF patients at different days/times of the experiment Insulin, ␮U/ml

Glucose, mg/dl

Day/time

Normal

CRF

P

Normal

CRF

P

1/0800* 2000 2/0900 1300 1800 2000 3/0800

6.4ⴞ0.9 1.2⫾0.3 2.3⫾0.8 12.3⫾4.0 19.0⫾4.9 23.8⫾5.9 7.0⫾1.1

20.0ⴞ8.5 12.4⫾8.0 6.8⫾3.2 21.4⫾7.1 37.9⫾10.0 85.5⫾44.7 20.0⫾8.6

0.27 0.33 0.33 0.39 0.21 0.33 0.29

94.3ⴞ4.0 79.8⫾4.7 77.0⫾6.8 107.8⫾10.4 107.0⫾2.7 118.8⫾5.1 97.3⫾4.9

99.1ⴞ4.3 83.6⫾3.9 87.1⫾4.8 104.0⫾7.6 119.4⫾11.1 140.9⫾8.8 101.4⫾4.4

0.47 0.56 0.25 0.78 0.43 0.11 0.56

Values are means ⫾ SE. Fasting extended from day 0, 2000 to day 2, 0900 (37 h) and refeeding from day 2, 0900 to day 3, 0800. *Baseline levels obtained at 0800, day 1. P, P values comparing normal with CRF subjects, derived from Student’s t-test.

decrements of plasma insulin during fasting were ⫺81 ⫾ 6 and ⫺64 ⫾ 0%, respectively, at 2000, day 1, and 0900, day 2, and corresponding values in the CRF patients were ⫺52 ⫾ 13 and ⫺51 ⫾ 9%. In the refeeding period, fractional increments of plasma insulin reached peaks of 277 ⫾ 96 and 397 ⫾ 75%, respectively, in the normal and CRF subjects at 2000, day 2. None of these oscillations was different between the two study groups. As for plasma glucose, the fractional decrements during fasting were ⫺19 ⫾ 4% in the normal subjects and ⫺12 ⫾ 4% in the CRF patients at 0900, day 2. Upon refeeding, fractional glucose increments were 26 ⫾ 6 and 44 ⫾ 12%, respectively, in the controls and the CRF patients at 2000, day 2. The responses were also similar in magnitude in the two study groups. To assess whether high intraindividual variation and overlapping values might mask actual differences in the hormones examined in the two groups, we calculated the intraindividual coefficients of variation (CV, %) of all the studied hormones. Due to the circadian variation and the physiological changes to feeding, we included only appropriate samples for the calculation of the CV. For leptin and NPY, the values of day 0, 2000, day 1, 0800, day 2, 2000, and day 3, 0800, were included; and for cortisol, ACTH, and insulin, the values of day 1, 0800, and day 3, 0800, were used. All values listed below are in the sequence of control vs. CRF patients: leptin, 19.6 ⫾ 6.0 and 21.1 ⫾ 3.9%; NPY, 26.5 ⫾ 7.4 and 16.9 ⫾ 3.3%; cortisol, 15.1 ⫾ 4.6 and 16.7 ⫾ 3.3%; ACTH, 41.1 ⫾ 12.2 and 24.8 ⫾ 8.0%; and insulin, 6.9 ⫾ 1.5 and 13.4 ⫾ 3.8%. The intraindividual CV is not high. Furthermore, none of the differences between the two groups reached statistical significance. Nutritional status and appetite assessment. The BMI and percent body fat were comparable in the two study groups (Table 1). Subjective global assessment identified all subjects to be well nourished; all participants had a score of A. There were no adverse gastrointestinal symptoms and no acute change in body weight. Physical examination did not detect any acute loss of fat or muscle wasting. All participants were fully functional in their usual activities. As for appetite assessment, the VAS showed that nausea was a rare occurrence. The sensation of feeling hungry and the desire to eat were comparable in both study groups. The sensation of fullness was appropriately reciprocal to that of hunger. Figure 5 illustrates the hunger scores of the two study groups during the entire experiment. The hunger score peaked at day 2, 0900 (the end of fasting) in both groups; the scores were 9.8 ⫾ 0.3 and 8.6 ⫾ 0.9, respectively, in the controls and the CRF patients. The AJP-Endocrinol Metab • VOL

hunger scores were similarly lowest for both groups at 2000, day 0; the scores were 1.0 ⫾ 0.6 and 3.4 ⫾ 0.6, respectively, in the normal subjects and the CRF patients. None of these values was different between the two groups. The desire-to-eat scores parallel those of the hunger scores in both groups. DISCUSSION

All of the hormones studied in this experiment demonstrate either striking changes to fasting and feeding or some patterns of circadian change. Baseline serum leptin was not statistically different between the two groups. In normal subjects, there was an unequivocal nocturnal leptin surge, and fasting profoundly suppressed leptin secretion. Sinha et al. (36) first reported nocturnal leptin rise in humans by measuring leptin in one 24-h cycle. Pratley et al. (26) did not detect any nocturnal rise when

Fig. 4. Insulin and glucose response to fasting and refeeding. Fractional changes (%) of plasma insulin (A) and glucose (B) at designated times compared with baseline values obtained on day 1, 0800. Values are presented as means ⫾ SE. Solid lines and filled circles are derived from normal subjects, broken lines and open squares from CRF patients. *P ⬍ 0.05 by Student’s t-test comparing the two groups. Fasting, represented by the cross-hatched bar at the bottom of the figure, began on day 0, 2000, and ended on day 2, 0900. Refeeding extended from day 2, 0900, to day 3, 0800.

288 • JANUARY 2005 •

www.ajpendo.org

FASTING AND REFEEDING IN CRF PATIENTS

Fig. 5. Visual analog scale illustrating “hunger” sensation. The scale on the y-axis is from 0 to 10, 0 being not hungry at all and 10 extremely hungry. Values are presented as means ⫾ SE. Solid lines and filled circles are derived from normal subjects, broken lines and open squares from CRF patients. *P ⬍ 0.05 by Student’s t-test comparing the 2 groups. Fasting, represented by the cross-hatched bar at the bottom of the figure, extended from day 0, 2000, to day 3, 0900.

they measured leptin during a 24-h fast. During a 22-h fast, Klein et al. (16) found a reduction in abdominal wall subcutaneous leptin production, and this was accompanied by a parallel decline in circulating leptin. In the present study, we measured leptin during a 56-h experiment consisting of prefast, fast, and refeeding periods. We not only confirmed the nocturnal surge but also found that fasting totally abolished this nocturnal rise. These findings suggest that the observed rise in serum leptin at night is a consequence of feeding during the day and not simply being linked to a diurnal rhythm. In rats, Saladin et al. (32) found ob gene expression to be related to feeding as well, and gene expression increased during the night when rats ate. When food was withheld, the nocturnal increase in ob mRNA was abolished. In humans, serum leptin increased at ⬃0000 h, 20 h after breakfast, and 4 h after the evening snack. In the rats, white adipose tissue ob gene increment also occurs at ⬃0000, 4 h after feeding initiation. Sinha et al. (36) stated that the discrepancy between humans and rats may be due to species difference. We hypothesized that the discrepancy in time of onset of leptin peak could be explained by the lag time between increase in ob gene expression in the white adipose tissue and the rise in circulating level. The former was measured in the rat and the latter in humans. Qualitatively, the leptin response to fasting and feeding was similar in the two study groups. Quantitatively, Fig. 2 indicates that the nocturnal surge during feeding was diminished in CRF patients, but, as shown in Table 2, this blunting was present only in the three subjects who had excessive body fat content and high basal leptin. The other four patients with normal basal leptin behaved exactly like the control subjects, showing an unequivocal nocturnal surge. We suspect that it is the increase in body fat, and not uremia, that led to the altered postfeeding AJP-Endocrinol Metab • VOL

E53

leptin metabolism in these three obese CRF patients. In the study of Sinha et al. (36), nocturnal leptin surge was of milder magnitude in obese subjects. Klein et al. (16) similarly found that leptin production from abdominal fat was less marked in obese compared with lean women. Several investigators reported that circulating leptin, factored for BMI, was increased in CRF patients (13, 34). It was suggested that the increased serum leptin is related to ongoing inflammation and may be contributing to anorexia (13). Our data do not show disproportionate elevation of leptin beyond that accounted for by body fat content. In addition, because leptin was so effectively suppressed during food deprivation, it is unlikely to be the cause of uremic anorexia. Circulating NPY levels were homogenously and markedly elevated in the CRF patients. This high level of NPY is unlikely to be due to assay problems, as the antibody used was 100% reactive to human NPY and had 0 and 3% crossreactivity to, respectively, human PYY and pancreatic polypeptide. The possibility of impaired renal excretion or degradation leading to high plasma concentration was considered but rejected for two reasons. First, despite the fact that kidney contributes to peptide hormone clearance 12– 40%, mostly by tubular uptake and degradation (15, 21, 34), not all peptides are increased in renal failure. This is because high plasma concentration should evoke a negative feedback decline in secretion. Second, there is a striking lack of parallel increments in peptide hormones in the end-stage renal disease (ESRD) population. When an increased level is noted, there is usually a component of increased production; decreased renal excretion or degradation plays a minor role. Examples to support this contention include the following. Elevation of leptin and insulin is found mostly in patients with high body fat, high parathyroid hormone in those with renal osteodystrophy, increased gonadotropins in postmenopausal women, and high TSH in hypothyroid subjects. More pertinently, kinetic study of prolactin, modestly elevated in CRF patients, illustrated unequivocally increased production (35). In control subjects, the circadian rhythm of NPY was reciprocal to that of leptin. It peaked between 0800 and 0900 when daily eating began and reached nadir values at ⬃0000 – 0400, when eating had ceased. This pattern is consistent with the current view that high leptin suppresses NPY. In the CRF patients, circadian NPY variation was remarkably muted; the diurnal curve was almost flat and circulating NPY levels were constantly elevated. (Table 3 and Fig. 2B). Of interest is the finding that fasting not only failed to increase NPY peaks but actually accentuated the 0000 – 0400 nadir (Fig. 2B). Because NPY is a potent orexigenic peptide, teleologically one would expect it to rise during fasting (28). In rodents, fasting does increase NPY gene expression in the central nervous system (5). However, NPY is not the only peptide responsible for increased eating during fast; other pathways stimulate eating equally well. The fact that parallel and redundant pathways exist in stimulating food acquisition is well illustrated in NPY knockout mice that maintain normal eating habit and body weight (9). NPY is not just an orexigenic peptide but has a far greater physiological role. It is a vasoactive peptide that is widely distributed in the central and peripheral nervous systems. In the central nervous system it inhibits sympathetic preganglionic neurons, leading to a reduction in thermogenesis of brown

288 • JANUARY 2005 •

www.ajpendo.org

E54

FASTING AND REFEEDING IN CRF PATIENTS

adipose tissue (4, 17). Centrally, NPY stimulates corticosterone secretion in rats, an effect mediated through the hypothalamus because corticotropin-releasing factor and ACTH secretion are both increased (30). In the periphery, NPY is coreleased with norepinephrine during sympathetic nerve stimulation; it enhances vasoconstriction and modulates cardiac function (23, 25). Zoccali et al. (38) recently reported elevated circulating NPY in CRF patients, showing a direct relationship between plasma NPY and norepinephrine and epinephrine. More importantly, elevated NPY predicts higher mortality from cardiovascular events. CRF is accompanied by a heightened sympathetic state, and the high serum NPY may reflect this (7). Glucocorticoid and insulin have been known for a long time to play important roles in eating and energy balance. The two hormones have reciprocal effects. In the central nervous system, glucocorticoid stimulates whereas insulin inhibits eating, acting through stimulation and inhibition of NPY, respectively (37). In contrast, in the periphery, glucocorticoid is catabolic, and insulin anabolic, with regard to energy storage. Adrenalectomy cures obesity of all origins, diet induced, hypothalamic disorder, or genetic aberration (29). This is due to a combination of cortisol deficiency and high corticotripin-releasing hormone secretion. Although cortisol stimulates eating, corticotropin-releasing hormone is an anorexigenic peptide (12). In our CRF patients, cortisol physiology was altered. There was a loss of evening-hour cortisol suppression, and ACTH tended to be higher; the latter suggests that the apparent hypercortisol state was hypothalamic in origin. The reason for the higher ACTH is not understood. It could be related to the high NPY state. In mammals, cortisol and NPY exhibit a positive feedback relationship (1). As for eating, the hypercortisol state tends to override the effect of high corticotropin-releasing hormone. With regard to insulin, CRF patients had higher plasma insulin levels at all times, but their responses to fasting and feeding were identical to those of normal subjects. Fasting resulted in insulin suppression, which should stimulate eating. More importantly, feeding led to a remarkable increase in insulin secretion, the latter making anabolism possible (19). A note should be made regarding the lower protein intake in the CRF patients, 0.89 g䡠kg⫺1 䡠day⫺1. This lower intake was due, in part, to imposed phosphorus restriction and is unlikely to cause neuroendocrine disturbance. In the non-Western world, most of the population consumes less than that amount of protein. It is important to emphasize that we (20) have shown repeatedly, with whole body in vivo amino acid kinetics, that this amount of protein intake, in the absence of any catabolic event, is perfectly adequate to maintain nitrogen and protein balance. This work has limitations, most important the small number of study subjects combined with the heterogeneous nature of the CRF group, consisting of predialysis, hemodialysis, and PD patients. Fortunately, the observed responses to the experimental maneuver were qualitatively similar in most subjects. Quantitatively, any difference noted appeared to be related to body fat mass and not uremia or the modality of treatment. Despite the limitations, this study demonstrated that the maneuver of fasting and refeeding provides a tool that can be used effectively for further study. In summary, new information derived from this study regarding eating and energy balance in CRF patients includes the AJP-Endocrinol Metab • VOL

following. 1) Anorexia, if present, was unlikely to be due to excess leptin or insufficient NPY. 2) Leptin response to fasting and eating were similar in the two study groups; fasting suppressed leptin and feeding resulted in a nocturnal surge. 3) There was a tonic hypersecretion of NPY in CRF patients, possibly related to heightened sympathetic state. 4) A hypercorticoid state was present in CRF patients, as was evident by impaired cortisol suppression in the evening hours; the high cortisol level should, however, favor eating. 5) Fasting effectively suppressed, and eating briskly stimulated, insulin secretion to a similar magnitude in controls and CRF patients; these responses favor eating when energy balance is low and anabolism when energy intake is abundant. 6) CRF patients responded appropriately to fasting, with hunger and increased desire to eat; despite the prevailing view that uremic patients are anorexic, we found that our CRF subjects responded to fasting with an appropriate sense of hunger and desire to eat, and, in fact, ate well after fasting. Another new observation, found in both study groups, was the failure of NPY to rise during fasting despite marked leptin suppression. The relatively normal pattern of the appetite-related hormones found in this study may be due to the fact that we screened our study patients vigorously to exclude those with concomitant illnesses and they are healthier than the general ESRD population. It should be emphasized that we have not evaluated the melanocortin pathways; increased ␣-melanocytestimulating hormone secretion or enhanced activity of the melanocortin-4 receptor binding inhibits eating and increases thermogenesis (10, 22). We also have not studied the ghrelin and PYY system; the former is secreted by the stomach to stimulate eating, and the latter is secreted by the large intestine and suppresses appetite (2, 24). We also have not assessed other orexigenic peptides, including the melanin-concentrating hormone, the agouti-related peptide, or the newly discovered orexins (6, 10, 31). ACKNOWLEDGMENTS We thank Dixie Ecklund and Phyllis Stumbo at the University of Iowa GCRC for their superb assistance, and we appreciate the time and energy input of the subjects who participated in this study. GRANTS This work was supported in part by the National Institutes of Health Grant RR-55, General Clinical Research Center, University of Iowa. REFERENCES 1. Akabayashi A, Levin N, Paez X, Alexander JT, and Leibowitz SF. Hypothalamic neuropeptide Y and its gene expression: relation to light/ dark cycle and circulating corticosterone. Mol Cell Neurosci 5: 210 –218, 1994. 2. Batterham RL, Cowley MA, Small CJ, Herzog H, Cohen MA, Dakin CL, Wren AM, Brynes AE, Low MJ, Ghatei MA, Cone RD, and Bloom SR. Gut hormone PYY3–36 physiologically inhibits food intake. Nature 418: 650 – 654, 2002. 3. Bergstrom J. Why are dialysis patients malnourished? Am J Kidney Dis 26: 229 –241, 1995. 4. Billington CJ, Briggs JE, Grace M, and Levine AS. Effects of intracerebroventricular injection of neuropeptide Y on energy metabolism. Am J Physiol Regul Integr Comp Physiol 260: R321–R327, 1991. 5. Brady LS, Smith MA, Gold PW, and Herkenham M. Altered expression of hypothalamic neuropeptide mRNAs in food-restricted and fooddeprived rats. Neuroendocrinology 52: 441– 447, 1990. 6. Chambers J, Ames RS, Bergsma D, Muir A, Fitzgerald LR, Hervieu G, Dytko GM, Foley JJ, Martin J, Liu WS, Park J, Ellis C, Ganguly S, Konchar S, Cluderay J, Leslie R, Wilson S, and Sarau HM.

288 • JANUARY 2005 •

www.ajpendo.org

FASTING AND REFEEDING IN CRF PATIENTS

7. 8. 9. 10. 11. 12. 13.

14. 15. 16. 17. 18.

19. 20. 21. 22. 23.

24.

Melanin-concentrating hormone is the cognate ligand for the orphan G-protein-coupled receptor SLC-1. Nature 400: 261–264, 1999. Converse RL Jr, Jacobsen TN, Toto RD, Jost CM, Cosentino F, Fouad-Tarazi F, and Victor RG. Sympathetic overactivity in patients with chronic renal failure. N Engl J Med 327: 1912–1918, 1992. Detsky AS, McLaughlin JR, Baker JP, Johnston N, Whittaker S, Mendelson RA, and Jeejeebhoy KN. What is subjective global assessment of nutritional status? J Parenter Enteral Nutr 11: 8 –13, 1987. Erickson JC, Clegg KE, and Palmiter RD. Sensitivity to leptin and susceptibility to seizures of mice lacking neuropeptide Y. Nature 381: 415– 418, 1996. Fan W, Boston BA, Kesterson RA, Hruby VJ, and Cone RD. Role of melanocortinergic neurons in feeding and the agouti obesity syndrome. Nature 385: 165–168, 1997. Friedman JM and Halaas JL. Leptin and the regulation of body weight in mammals. Nature 395: 763–770, 1998. Gardner JD, Rothwell NJ, and Luheshi GN. Leptin affects food intake via CRF-receptor-mediated pathways. Nat Neurosci 1: 103, 1998. Heimburger O, Lonnqvist F, Danielsson A, Nordenstrom J, and Stenvinkel P. Serum immunoreactive leptin concentration and its relation to the body fat content in chronic renal failure. J Am Soc Nephrol 8: 1423–1430, 1997. Ikizler TA, Greene JH, Wingard RL, Parker RA, and Hakim RM. Spontaneous dietary protein intake during progression of chronic renal failure. J Am Soc Nephrol 6: 1386 –1391, 1995. Katz AI and Rubenstein AH. Metabolim of proinsulin, insulin, and C-peptide in the rat. J Clin Invest 52: 1113–1121, 1973. Klein S, Horowitz JF, Landt M, Goodrick SJ, Mohamed-Ali V, and Coppack SW. Leptin production during early starvation in lean and obese women. Am J Physiol Endocrinol Metab 278: E280 –E284, 2000. Krukoff TL. NPY-containing nerve terminals and fibers in close association with peptidergic sympathetic preganglionic neurons. Neurosci Lett 77: 1– 4, 1987. Lim VS and Flanigan MJ. Organ and metabolic complications—abnormal endocrine function in chronic renal failure. In: Replacement of Renal Function by Dialysis (4th ed.), edited by Jacobs C, Kjellstrand CM, Koch KM, and Winchester JF. London: Kluwer Academic, 1996, p. 1236 –1242. Lim VS, Yarasheski KE, Crowley JR, Fangman J, and Flanigan MJ. Insulin is protein-anabolic in chronic renal failure patients. J Am Soc Nephrol 14: 2297–2304, 2003. Lim VS, Yarasheski KE, and Flanigan MJ. The effect of uraemia, acidosis, and dialysis treatment on protein metabolism: a longitudinal leucine kinetic study. Nephrol Dial Transplant 13: 1723–1730, 1998. Maack T, Johnson V, Kau ST, Figueiredo J, and Sigulem D. Renal filtration, transport, and metabolism of low-molecular-weight proteins: a review. Kidney Int 16: 251–270, 1979. Marks DL, Ling N, and Cone RD. Role of the central melanocortin system in cachexia. Cancer Res 61: 1432–1438, 2001. Morris MJ, Cox HS, Lambert GW, Kaye DM, Jennings GL, Meredith IT, and Esler MD. Region-specific neuropeptide Y overflows at rest and during sympathetic activation in humans. Hypertension 29: 137–143, 1997. Nakazato M, Murakami N, DateY, Kojima M, Matsuo H, Kangawa K, and Matsukura S. A role for ghrelin in the central regulation of feeding. Nature 409: 194 –198, 2001.

AJP-Endocrinol Metab • VOL

E55

25. Pedrazzini T, Seydoux J, Kunstner P, Aubert JF, Grouzmann E, Beermann F, and Brunner HR. Cardiovascular response, feeding behavior and locomotor activity in mice lacking the NPY Y1 receptor. Nat Med 4: 722–726, 1998. 26. Pratley RE, Nicolson M, Bogardus C, and Ravussin E. Plasma leptin responses to fasting in Pima Indians. Am J Physiol Endocrinol Metab 273: E644 –E649, 1997. 27. Raben A, Tagliabue A, and Astrup A. The reproducibility of subjective appetite scores. Br J Nutr 73: 517–530, 1995. 28. Sahu A, Kalra PS, and Kalra SP. Food deprivation and ingestion induce reciprocal changes in neuropeptide Y concentrations in the paraventricular nucleus. Peptides 9: 83– 86, 1988. 29. Sainsbury A, Cusin I, Rohner-Jeanrenaud F, and Jeanrenaud B. Adrenalectomy prevents the obesity syndrome produced by chronic central neuropeptide Y infusion in normal rats. Diabetes 46: 209 –214, 1997. 30. Sainsbury A, Rohner-Jeanrenaud F, Grouzmann E, and Jeanrenaud B. Acute intracerebroventricular administration of neuropeptide Y stimulates corticosterone output and feeding but not insulin output in normal rats. Neuroendocrinology 63: 318 –326, 1996. 31. Sakurai T, Amemiya A, Ishii M, Matsuzaki I, Chemelli RM, Tanaka H, William SC, Richardson JA, Kozlowski GP, Wilson S, Arch JRS, Buckingham RE, Haynes AC, Carr SA, Annan RS, McNulty DE, Liu WS, Terrett JA, Elshourbagy NA, Bergsma DJ, and Yanagisawa M. Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell 92: 573–585, 1998. 32. Saladin R, DeVos P, Guerre-Millo M, Leturque J, Girard B, Staels B, and Auwerx J. Transient increase in obese gene expression after food intake or insulin administration. Nature 377: 527–529, 1995. 33. Schwartz MW, Sipols AJ, Marks JL, Sanacora G, White JD, Scheurink A, Kahn SE, Baskin DG, Woods SC, Figlewicz DP, and Porte D JR. Inhibition of hypothalamic neuropeptide Y gene expression by insulin. Endocrinology 130: 3608 –3616, 1992. 34. Sharma K, Considine RV, Michael B, Dunn SR, Weisberg LS, Kurnik BRC, Kurnik PB, O’Connor J, Sinha M, and Caro JF. Plasma leptin is partly cleared by the kidney and is elevated in hemodialysis patients. Kidney Int 51: 1980 –1985, 1997. 35. Sievertsen GD, Lim VS, Nakawatase C, and Frohman LA. Metabolic clearance and secretion rates of human prolactin in normal subjects and in patients with chronic renal failure. J Clin Endocrinol Metab 50: 846 – 852, 1980. 36. Sinha MK, Ohannesian JP, Heiman ML, Kriauciunas A, Stephens TW, Magosin S, Marco C, and Caro JF. Nocturnal rise of leptin in lean, obese, and non-insulin-dependent diabetes mellitus subjects. J Clin Invest 97: 1344 –1347, 1996. 37. Strack AM, Sebastian RJ, Schwartz MW, and Dallman MF. Glucocorticoids and insulin: reciprocal signals for energy balance. Am J Physiol Regul Integr Comp Physiol 268: R142–R149, 1995. 38. Zoccali C, Mallamaci F, Tripepi G, Benedetto FA, Parlongo S, Cutrupi S, Iellamo D, Bonanno G, Rapisarda F, Fatuzzo P, Seminara G, Cataliotti A, and Malatino LS. Prospective study of neuropeptide Y as an adverse cardiovascular risk factor in end-stage renal disease. J Am Soc Nephrol 14: 2611–2617, 2003.

288 • JANUARY 2005 •

www.ajpendo.org