chronic renal disease: a growing problem

Kidney International, Vol. 64 (2003), pp. 1141–1151

NEPHROLOGY FORUM

Chronic renal disease: A growing problem

Principal discussant: Frederick Kaskel

Children’s Hospital at Montefiore, and Albert Einstein College of Medicine of Yeshiva University, Bronx, New York, USA

CASE PRESENTATIONA 13-year-old Hispanic adolescent had presented at the age

of 4 years with an initial diagnosis of minimal change nephroticsyndrome and growth failure. However, he was unresponsive totherapeutic courses of steroids, cyclophosphamide, and cyclo-sporine. Serial renal biopsies performed over 18 months even-tually revealed focal segmental glomerulosclerosis (FSGS) witha worsening chronicity index. Renal function gradually deterio-rated and end-stage renal disease (ESRD) ensued. He had anintractable nephrotic syndrome that warranted bilateral ne-phrectomy prior to renal transplantation. At age 7, he receiveda living related allograft, which was lost following delayed graftfunction, multiple acute rejection episodes, and recurrence ofFSGS within 4 weeks after transplantation. Immediate plasma-pheresis produced no remarkable clinical improvement.

After transplantation, the glomerular filtration rate (GFR)deteriorated from 57 mL/min/1.73 m2 to 22 mL/min/1.73 m2

within 6 months to 6.5 mL/min/1.73 m2 by 12 months. Hemodi-alysis (HD) was begun but was replaced with continuous cyclingperitoneal dialysis (CCPD) by 18 months post transplantation.He developed a cytomegalovirus infection within 6 monthsafter the transplant surgery. Other complications included ure-mic pericarditis, severe hypertension, septicemia, and a virus-induced hemorrhagic anemia, necessitating 10 hospitalizationsper year in the first 2 years. Serial nutritional evaluations re-vealed persistent growth failure with weight below the fifth, and

The Nephrology Forum is funded in part by grants from Amgen, Incor-porated; Merck & Co., Incorporated; Dialysis Clinic, Incorporated;and Bristol-Myers Squibb Company.

Key words: growth failure, recombinant human growth hormone, IGF-I,somatostatin, ghrelin, leptin.

2003 by the International Society of Nephrology

1141

height below the third, percentiles, respectively, for age andgender (Figs. 1 and 2). Recombinant human growth hormone(rhGH), given at an initial dose of 0.05 mg/kg/day at the ageof 8 years and increased to 0.06 mg/kg in 6 months, produceda prepubertal growth spurt.

Pubertal growth advanced to Tanner stage II at age 10 andstage III by age 12. The patient had improved tolerance forphysical exercise and greater involvement in sporting activities.The dialysis adequacy, as measured by Kt/V at the onset ofCCPD, was 1.78 and improved to 2.25 after 6 months of rhGHtherapy. Before therapy with rhGH, he received 0.5 �g/day ofcalcitriol (1,25 dihdroxycholecalciferol); after treatment was initi-ated, the dose was able to be reduced to 0.25 �g/day to controlhyperparathyroidism. Currently 13 years of age, he has a boneage of 12.5 years. His weight and height are in the 12.5 and20 percentiles, respectively, for age and gender. In spite of hischronic illness, he is a well-adjusted young adolescent withexceptional motivation and above-average school achievement.

DISCUSSION

Dr. Frederick Kaskel (Director of Pediatric Nephrol-ogy, Children’s Hospital at Montefiore, and Professor ofPediatrics, Albert Einstein College of Medicine of YeshivaUniversity, Bronx, New York): The normal kidney per-forms an array of essential metabolic and physiologicfunctions, including fluid and electrolyte regulation andmaintenance of acid-base balance, which are necessaryfor promoting normal homeostasis and thereby ensuringoptimal cellular and organ function. The renal system isinvolved in the production, processing, and secretion ofvarious hormonal substances, including 1,25 dihydroxy-cholecalciferol, erythropoietin, and prostaglandins. Thissystem is of paramount importance in excretion of en-dogenous nitrogen end products and toxic metabolitesof exogenous substances.

The metabolic consequences of poor renal functionare diverse and often multisystemic. The complexity of thephysiologic derangements in chronic renal disease (CRD)and the nonspecificity of CRD’s clinical characteristicsare some of the reasons for its widespread late diagnosis.Chronic renal disease is commonly under-recognizedand therefore under-treated in spite of the increasingavailability of therapeutic interventions that can slow itsprogression and/or ameliorate an offending comorbidity.

Nephrology Forum: Chronic renal disease and growth1142

Fig. 1. Growth curve for case presentation.

This case highlights the impact of metabolic dysfunc-tion in a growing child with CRD. In spite of the mile-stones achieved in our care for pediatric patients withCRD in the last few years, renal disease still constitutes asignificant source of biologic, physical, and psychosocialburdens to this population. Indeed, virtually every cellin the body is adversely affected by the unfavorablemetabolic milieu of chronic uremia. In this Forum I willhighlight some current concepts in growth failure as abiophysical consequence of a uremic endocrinopathyand nutritional alterations.

Growth failure

Growth failure is highly prevalent in children withCRD, and it is a significant cause of morbidity and mor-

tality [1]. Pediatric patients with CRD, including renaltransplant recipients, have a significant height deficit [2].At initiation of dialysis, 30% to 50% of patients haveheights below the third percentile for age; the final adultheight is compromised in approximately 50% of childrenwith prepubertal onset of renal disease [3]. The causesof impaired linear growth might include protein energymalnutrition, chronic metabolic acidosis, chronic ane-mia, recurrent infection, and endocrinopathy [4].

Children normally go through three phases of somaticgrowth and biologic maturation, each controlled pre-dominantly by a different regulatory mechanism [4]. Therapid growth in the first 2 years, influenced by favorablenutritional and metabolic profiles, accounts for one-thirdof the ultimate adult stature [4, 5]. The relatively stable

Nephrology Forum: Chronic renal disease and growth 1143

Fig. 2. Growth velocity for case presentation. Abbreviations are: NS, onset of nephrotic syndrome; HD, hemodialysis; RTX, renal transplant;GH, growth hormone.

growth in childhood is principally under hormonal con-trol, including the growth hormone/insulin-like growthfactor-I (GH/IGF-I) axis and thyrotropin [4]. Twentypercent of adult height is attained during puberty, whichis modulated both by growth and sex hormones [4–7].The degree of growth retardation in CRD is often deter-mined by the age of onset of renal disease. Infants withcongenital renal failure may suffer from intrauterinegrowth retardation [4, 5]. Further loss in height occursduring 0 to 3 months and 9 to 12 months; growth is relativelypreserved in mid-infancy. Early nutritional interventionand prevention of metabolic deficits of renal failure canpreclude the development of severe stunting in the first

2 years of life [5]. Conservative treatment of predialysispatients from birth to 3 years, that is, without the use ofrhGH, revealed a favorable growth rate, with heightvelocity at 22.2, 10.9, and 7.6 cm per year, for each year,respectively, all higher than the lower two standard devi-ation scores (SDS) [5]. During childhood, patients withCRD often grow at a rate equivalent to that achievedby the end of the second year [6]. The loss of renalfunction is the most important predictor of growth rate,even though it accounted for only 10% to 15% of varia-tion in growth [4]. In general, patients with GFR �25mL/min/1.73 m2 maintain growth rate along the initialtrajectory, while those with GFR �25 mL/min/1.73 m2

Fig. 3. The GH–IGF axis. Abbreviations are:ALS, acid labile subunit; GH, growth hor-mone; GHBP, GH binding protein; GHRH,GH releasing hormone; IGF, insulin-likegrowth factor; IGFBP, IGF binding protein(used with permission from Holt RIG: Fetalprogramming of the growth hormone-insulin-like growth factor axis. Trends EndocrinolMetab 13(9):392-397, 2002).

Fig. 4. Hormones that control eating. Leptinand insulin (lower portion of the figure) circu-late in the blood at concentrations proportion-ate to body-fat mass. They decrease appetite byinhibiting neurons (center portion of the figure)that produce the molecules neuropeptide Y(NPY) and agouti-related peptide (AgRP),while stimulating melanocortin-producing neu-rons in the arcuate-nucleus region of the hypo-thalamus, near the third ventricle of the brain.NPY and AgRP stimulate eating, and melano-cortins inhibit eating, via other neurons (topportion of the figure). Activation of NPY/AgRP-expressing neurons inhibits melano-cortin-producing neurons. The gastric hormoneghrelin stimulates appetite by activating theNPY/AgRP-expressing neurons. Peptide YY(3-36) (PYY3-36), released from the colon in-hibits these neurons and thereby decreasesappetite for up to 12 hours. PYY3-36 works inpart through the autoinhibitory NPY receptorY2R (used with permission from SchwartzMW, Morton GJ: Obesity: Keeping hunger atbay. Nature 418:595–597, 2002).


experience a loss in height of usually less than 0.2 SD/year [6]. Height velocity prior to the growth spurt is re-duced in children with CRD. The peak height velocityof puberty is delayed by about 2.5 years, with a reductionin its magnitude and total duration [8]. The pubertalheight gain is about 58% and 48% of that seen, respec-tively, in late-maturing boys and girls without CRD; thedecline in mean height for boys is �2.9 SD and for girls,–2.3 SD, respectively [8].

Chronic renal disease and its therapeutic modalitiescan delay puberty by an average of 2 years [7]. In spite ofnormal development of secondary sexual characteristics,subnormal testicular growth is common and is associatedwith germ cell deficiency, low sperm count, and perma-nent loss of reproductive capacity in patients with a pre-pubertal onset of renal failure [9]. Unlike adults, in thesepatients, the sexual defects can fail to correct even afterrenal transplantation [9].

Growth hormone is normally produced in abundantquantity by the pituitary somatotroph cells [10] (Fig. 3).It is basally secreted in a pulsatile manner, stimulated byhypothalamic GH releasing hormone (GHRH), inhibitedby somatostatin, and is under the negative feedback of theperipheral effectors GH and IGF-I [10]. Growth hormonereleasing hormone activates its receptor site, which in turninduces G protein for stimulation of cyclic adenosinemonophosphate (cAMP) synthesis and a consequenttranscription of GH [11]. The two human GH genes arelocated on chromosome 17 [11]. The hGH-N gene encodespituitary GH; its product undergoes a posttranslationalcleavage into two active GH with molecular weights of21.5 kD and 20 kD [12]. The 21.5 kD form constitutesapproximately 90% of the circulating GH. The somato-troph cells also possess receptors for GH releasing pep-tide (GHRP), called GH secretagogue receptor (GHS-R).The recent discovery of an endogenous secretagogue calledghrelin [13] adds to the availability of several exogenous(synthetic) GHRPs. Of the 5-somatostatin receptor sub-types (SSTR1 to SSTR5), only SSTR2 and SSTR5 mediateinhibition of GH secretion [14]. Growth hormone secre-tion also is influenced by nutritional states: it is stimu-lated by protein energy malnutrition, hypoglycemia, andhigh protein meals but is inhibited by hyperglycemia [10].

Growth hormone in the circulation stimulates synthe-sis of IGF-I in many tissues, most abundantly in the liver[10]. Receptors for GH are glycoproteins, some of whichare solubilized to produce growth hormone binding pro-tein (GHBP), a carrier protein that prolongs GH’s half-life by several hours [12]. Growth hormone is the mostpotent secretagogue for IGF-I, which mediates most, butnot all, of the action of GH. The IGF-Is are transportedin plasma bound to IGFBPs 1 to 6, but mostly to IGFBP-3,which occurs in the circulation as a 150 kD ternary com-plex bound to the IGF-I and its acid-labile subunit [15].Only about 1% of plasma IGF-I occurs in the free bio-

active form. The IGFs have much higher affinities forIGFBPs than do their receptors [16]. Furthermore, theyare protected from degradation by IGFBPs before theirrelease for receptor binding. The IGFBPs possibly re-duce the bioactivity of IGFs by a competitive inhibitionof the tissue receptor. On the other hand, they also mightfacilitate peripheral tissue action by a slow release of theligand, thereby preventing its premature degradation [16].

Insulin-like growth factor I also is synthesized in mes-enchymal cells in virtually all human tissues in an auto-crine (and/or paracrine) fashion, and some of it couldbe secreted into the systemic circulation [10]. A smallsingle-chain peptide with a molecular weight of 7650 kD,IGF-I belongs to the same family of genes as IGF-II andproinsulin [17]. Its physiologic activities are mediated viaIGF-I receptors. The IGF-I receptors are hetero-tetra-mers with 2 alpha and 2 beta subunits linked by disulfidebridges [18]. These receptors are widely distributed indifferent cell types, and they facilitate coordination ofsequential, proportionate, and symmetric body growth.The IGF activates the receptor by binding to its extracel-lular alpha subunit, which induces the transmembranebeta-unit counterpart for an autoactivation of tyrosine ki-nase [17], with a consequent phosphorylation of cytosolictyrosine residues and insulin receptor substrate (IRS-1and -2). The intracellular signaling cascade causes down-stream activation of the mitogen-activated protein 3(MAP-3) kinase pathway, thus stimulating cellular growth.Activation of IRS-1 also can stimulate the protein kinaseB pathway, which is important for protein synthesis, cellmotility, and inhibition of apoptosis [12].

Let’s compare the normal GH/IGF-I axis with that inpatients with CRD (Fig. 1). Growth hormone is secretedfrom the pituary gland under the control of the hypothal-amic hormones, somatostatin and GHRH, as well as themainly gastric ghrelin. GHRH and ghrelin bind to theirrespective receptors in the pituitary and stimulate GHsecretion. Somatostatin inhibits GH secretion. GH circu-lates, bound to GHBP, and acts through specific cell-surface receptors. Most of the anabolic actions of GH aremediated by IGF-I, which is produced in many differenttissues, with most circulating IGF-I being derived fromthe liver. IGF-I acts through the IGF-I receptor by auto-crine, paracrine, and classic endocrine mechanisms. IGF-Iis present in the circulation and extracellular space, al-most entirely bound to IGFBPs that coordinate and reg-ulate the biologic functions of the IGFs. Over 99% ofcirculating IGF-I is bound in a ternary complex compris-ing IGF-I, IGFBP-3, and an acid-labile subunit (ALS). Themajor source of circulating IGFBPs and ALS is the liver.IGF-I inhibits GHRH and GH secretion in a classicnegative feedback mechanism.

Growth hormone secretion by the hypothalamic-pitu-itary tract is poorly regulated in patients with CRD. Incontrast to healthy individuals, in these patients, glucose


infusion increases secretion of GH rather than inhibitingit [18]. Children with CRD can have increased (pulsatile)release of GH, the mechanism for which is unknown[18]. In addition, these patients have reduced renal clear-ance of GH that correlates with the loss of GFR. In spiteof an elevated serum GH level, the serum IGF-I concen-tration does not increase because its hepatic synthesis isreduced [10]. Furthermore, the bioavailability of IGF-Iis compromised because the levels of IGFBP-1, -2, -4,and -6 rise [9]. As a result of an increased proteolysis ofIGFBP-3, less IGF-I circulates as a ternary complex (150kD) [19]. Conversely, IGFBP-3 (29 kD) fragments in-crease, and these have less affinity for IGF-I. Conse-quently, effective delivery of IGF-I to its sites of actionis reduced [17–19]. There also can be end organ resis-tance to GH and IGF-I in chronic uremia. A study ofuremic rats showed a reduction in GH receptor densityin tibial growth plate [20] and low levels of mRNA forhepatic GH receptor [21]. Furthermore, chronic uremiais characterized by a defect in postreceptor signal trans-duction; tyrosine phosphorylation of JAK2, STAT5, andSTAT3 is impaired because of overexpression of sup-pressor of cytokine signaling-2 (SOCS-2) proteins [21].

Body weight and its composition are under strict con-trol by the hypothalamic system, which maintains a con-sistent “set point” by receiving feedback from peripheraleffectors [22] that can include leptin and ghrelin (Fig. 4).Ghrelin, an acylated peptide with a molecular weightof 3300 kD, is a recently discovered endogenous GHsecretagogue isolated from the stomach, hypothalamus,and systemic circulation of humans and rats [13, 23]. Itis produced by the gastric Gr-cells, which are morpholog-ically similar to the pancreatic alpha cells, and in lesseramount by the arcuate neurons of the hypothalamus. Itis more potent than GHRH, stimulating GH release in adose-dependent manner [24]. In addition to GH release,secretion of adrenocorticoid hormone (ACTH) and pro-lactin (PRL) also is increased, but ghrelin has no effect onluteinizing hormone (LH), follicle-stimulating hormone(FSH), and thyroid-stimulating hormone (TSH) [24].The range of its activity in humans is currently beingdefined, although recent studies suggest a role in theregulation of feeding behavior and body composition.A randomized, double-blind, crossover trial in humansshowed an increase in visual analog scores for appetite,total food intake, and energy consumption in study sub-jects [25]. However, unlike in rats, no effect on gastricemptying was seen, probably because of a relativelylower dose used in the human subjects. The mechanismfor ghrelin’s peripheral action could be inhibition of fattyacid oxidation and up-regulation of glycolysis, as sug-gested by an increase in respiratory quotient followingsystemic administration in rodents [23].

Evidence indicates that plasma ghrelin is generallyincreased by a negative energy balance but down-regu-

lated by a positive energy balance; plasma ghrelin corre-lates negatively with body mass index [26]. In addition,the plasma ghrelin concentration is higher in patients withanorexia nervosa but lower in obese subjects. Plasmaghrelin level falls following either oral or intravenousglucose administration [26]. Ghrelin mRNA is up-regu-lated in the stomachs of rodents after fasting and is re-stored to normal value after refeeding, although the in-tracellular signal for its release is unknown. The diurnalvariation of ghrelin secretion consists of a preprandialincrease and a decrease in level after feeding, followedby a nighttime peak at approximately 2:00 a.m. [26].Ghrelin therefore might play a role in the physiologicinitiation of food intake [23], and alteration of its func-tion could contribute to poor dietary intake and nutri-tional insufficiency in patients with CRD.

Leptin is a protein product of the obesity gene (ob)[27]. It is secreted by adipocytes, transported in plasmaby leptin-binding protein, and released into the centralnervous system. It decreases synthesis of anorexigenicneuropeptide Y (NPY) by activation of Jak/STAT sig-naling in agouti-related-protein (hypothalamic) neurons[28]. A highly potent anorectic, NPY promotes energyexpenditure and weight loss. Whereas ob/ob mice havegenetic traits of leptin deficiency, db/db mice have leptinreceptor insensitivity with a resultant hyperleptinemia[29]. Both models are characterized by gross obesity,high food intake, insulin hyporesponsiveness, and reducedenergy expenditure.

No substantiated data are available on the biologicrole of leptin in CRD. Studies are mostly of cross-sec-tional design. However, given the role of leptin in theregulation of nutrition and body weight in health, leptinmight play a significant role in the anorexia and weightloss of CRD. Serum leptin, mostly the free bioactiveform, is elevated in many patients with CRD [30]. Themechanism for its retention is unclear; data on renal clear-ance and/or metabolism of leptin are conflicting. How-ever, given its molecular weight of 16,000 kD [27], renalfiltration would be expected, and poor renal clearancewith low GFR has been documented [30]. Furthermore,interleukin-1 (IL-1) and tumor necrosis factor-� (TNF-�)cytokines, both commonly elevated in CRD patients, caninduce leptin release from adipocytes [31]. But leptin’srelationship with C-reactive proteins is poorly defined;the results of studies vary from a positive [32] to a nega-tive [33] correlation. Intradialysis parenteral nutritionelevates plasma GH and insulin, which synergisticallystimulate the release of leptin from peripheral tissues;this effect might be simulated by free IGF-I [34].

Unlike the appetite-stimulating peptide ghrelin, leptinmight play a negative regulatory role in food intake inthe body’s attempt to maintain an ideal nutritional statusand body composition. Leptin limits excessive weightgain, especially body fat mass [35]. Thus, a negative cor-


relation exists between leptin and serum albumin or leanbody mass [35]. On the other hand, an elevated serumleptin level also is found in obese patients [32], and itcorrelates with the body fat mass, presumably an attemptto down-regulate food intake. That a prospective studydiscerned a direct correlation between serum leptin/fatbody mass ratio with weight loss in HD patients suggestsa role for leptin in the loss of lean body mass [36].

Altered nutrition in CRD

Protein energy malnutrition is strongly associated withhigh morbidity and mortality rates in dialysis patients[37]. It is seen in 44% of patients at the initiation ofdialysis [38]. In addition, a direct correlation exists be-tween a progressive decline in GFR and deterioratingnutritional indices in most patients with CRD [38].Chronic renal disease might be associated with retentionof a yet-to-be identified “anorectic peptide,” probably amiddle molecule toxin [38] and/or leptin. Although dataon basal energy expenditure (BEE) seem to conflict(they are mostly cross-sectional in design), there is atendency for an elevated BEE in patients with advancedCRD but suppression of BEE in patients with chronicrenal insufficiency [39]. Thus a direct negative correla-tion exists between resting metabolic rate and GFR asmeasured by creatinine clearance [39].

In general, dietary energy intake is lower in patientswith CRD because of a spontaneous or therapeutic re-duction in protein intake and low calorie consumption.The decreased BEE in early CRD might represent anadaptive response to poor dietary energy consumption;the low protein intake in predialysis patients often causesreduced oxidation of amino acids, thus enhancing effi-cient utilization [40]. Furthermore, as I said earlier, cellu-lar metabolism is down-regulated in uremic states, withtendencies for reduced protein synthesis but increaseddegradation. In addition, patients with CRD have anadaptive low thyroid state, and correction often acceler-ates negative nitrogen balance and protein degradation[41]. These patients also have impaired peripheral con-version of T4 to T3 and decreased binding of T4 tothyroid-binding protein (TBG), but TSH level is withinnormal limits [41].

Early CRD brings a loss of lean body mass but apreservation of fat mass [42], probably due to an interplayof metabolic processes with lipogenic influences fromGH/IGF-I [10, 12] and also possibly from ghrelin [24–26].On the other hand, the loss of skeletal muscle mass mightresult from uremia per se, micro-inflammation, meta-bolic acidosis, nutritional insufficiency [38, 43, 44], andpossibly hyperleptinemia [36]. However, at the later stageof CRD, these adaptive responses can be overwhelmedby hypercatabolic factors (for example, infection, oxida-tive stress, cytokines, dialysis) with a consequent loss oflean and fat body masses [45]. An important factor in

the regulation of body composition is the metabolic con-sequence of uncoupling proteins (UCP) [46]. Upon acti-vation these proteins, exclusively expressed in brownadipose tissue, uncouple mitochondrial oxidative phos-phorylation, which possibly accounts for about one-fifthof BEE. Tumor necrosis factor might promote up-regula-tion of UCP-2 and UCP-3 genes in skeletal muscles ofuremic rats [31]. Although the role of UCP polymor-phism as a determinant of adverse nutritional outcomein CRD is yet to be elucidated, it could explain the vari-ability in biologic adaptation to metabolic stresses (includ-ing nutritional deficit) in patients with renal disease [47].

The combined effect of metabolic acidosis, chronicinflammation, and uremia likely increases protein degra-dation by activating the ubiquitin-proteasome pathway(UPP) [43], inducing branched-chain ketoacid dehydro-genase [44], increasing insulin resistance [16], and im-pairing functions of GH/IGF-I [15, 19]. In addition, meta-bolic acidosis is a potent inhibitor of hepatic albuminsynthesis [43]. Furthermore, the catabolic effect of meta-bolic acidosis is increased multiple-fold by glucocorticoidreleased in response to an intense inflammatory state[43, 44]. Conversely, in the absence of steroids, UPPoften is suppressed by an activated nuclear factor-�B(NF-�B), thereby limiting protein degradation [43, 44].

In conclusion, in spite of years of research, the bio-molecular characteristics of “uremic toxins” remain poorlyunderstood. However, there is little doubt about the pro-found adverse metabolic effects of uremia on biocellularactivities. The multisystemic dimension of its impact un-derlies the need for early and aggressive therapy, includ-ing renal replacement, optimal nutrition, and rhGH inthe pediatric population. The efficacy of available modal-ities of care can be enhanced by taking advantage ofour current understanding of the pathophysiology of thedisease. Thus, a need exists for combined rhGH therapy,which can include GHrP/ghrelin, IGF-I, and GH tomimic the physiologic state. We must intensify our effortsto reduce the “uremic load” with a more efficient andperhaps more frequent dialysis. The close interactionamong protein energy malnutrition, microinflammation,and cardiovascular diseases necessitates a search for amore effective therapy to break this vicious cycle. Further-more, studies are needed to define the biologic roles andtherapeutic potentials of leptin and ghrelin in CRD. Torealize these goals, concerted efforts by the nephrologycommunity, primary care physicians, and the public areneeded to improve early diagnosis of renal diseases. Theneed exists for a change in the public’s health behaviorconcerning CRD. The public must be motivated to seekroutine medical evaluation and screening, screening forrenal disease for those with a family history, and promptfollow-up when required. It has been estimated that asubstantial proportion of the American populace hasundiagnosed chronic renal insufficiency.


QUESTIONS AND ANSWERS

Dr. John T. Harrington (Division of Nephrology,Tufts-New England Medical Center, Boston, Massachu-setts): In the boy you presented today, GH wasn’t useduntil 1 year after renal transplanation. Can it be usedsooner? Is the delay in administering GH based on thepremise of “do no harm?” Or is it because adverse clini-cal effects have been seen post transplantation? Haveexperimental animals received GH from day 1 post trans-plant?

Dr. Kaskel: It appears that we withhold GH not onlybecause of concerns about rejection, but also because ofconcern for renal function. Experimental studies haveshown that GH in transgenic animals reduces renal func-tion [48]; this adverse effect was not seen in animals trans-genic for IGF-I. That might be one of the reasons whyGH is not being used right away after transplantation.

Dr. Harrington: If my recollection is correct, somechildren with partial lipodystrophy have progressive re-nal failure. What do we know about the peculiarities ofgrowth failure, ghrelins, or leptin in these patients withpartial lipodystrophy? Dr. Chesney, can you enlighten us?

Dr. Russell W. Chesney (Professor and Chairman,Le Bonheur Children’s Medical Center and University ofTennessee, Memphis, Tennessee): The partial lipodystro-phies are a heterogeneous group of disorders that includeloss of upper body subcutaneous adipose tissue, insulinresistance, hyperlipidemia, and diabetes mellitus [49].The typical autosomal-dominant form, which has no spe-cific renal involvement, is associated with mutations inthe nuclear laminin A, known as the familial partial lipo-dystrophy of Dunnigan (FPLD). Laminins are necessaryfor nuclear functioning, including bidirectional nuclearmembrane transcription factors. Plasma leptins are re-duced mainly in complete lipodystrophy in which lipo-atrophy is not confined to the upper body. In mousemodels of complete lipodystrophy, leptin administrationcan reverse diabetes and insulin resistance.

When partial lipodystropy is associated with progres-sive renal disease, the renal histologic finding is mesangio-capillary glomerulonephritis (MPGN) type II [50]. In par-tial lipodystrophy with type II MPGN, patients have aserum immunoglobulin termed complement 3 nephriticfactor (C3NeF) [51]. C3NeF is an autoantibody that bindsfactor H, which regulates the overwhelming activation ofcomplement. In its bound form, factor H cannot preventcomplement depletion and the development of membra-noproliferative glomerulonephritis (MPGN) [52]. Patientswith homozygous complement factor D deficiency alsocan show partial lipodystrophy with MPGN type II [52].

Dr. Sushil Sagar (Albert Einstein College of Medi-cine, Bronx, New York): Dr. Kaskel, you mentioned thatIGFBPs are elevated in CRD. Do we understand whatmakes the levels increase?

Dr. Kaskel: Experimental data in several animal mod-els of renal failure show that the elevated IGFBPs mightbe related to at least two factors. First, there is increasedliver mRNA for some binding proteins in spite of re-duced hepatic GH receptors [53]. Second, there is abuildup of the binding proteins because of poor renalclearance [54]. In other words, there is dysregulation ofboth synthesis and clearance.

Dr. Michael Goligorsky (Director, Division of Ne-phrology, New York Medical College, New York, NewYork): Since growth retardation is not uniformly ob-served in all children with CRD, I think it is possible tolearn a lot from the group that does not experiencegrowth failure. Has there ever been a side-to-side com-parison of the metabolic profiles of these two groups?

Dr. Kaskel: That is a very important question. Someof us have attempted to examine this question in amulticenter study funded by Genentech. I was workingwith Dr. Johnson and other members of our collabora-tive group looking at the GH/IGF-I axis and body com-position of these two groups of patients who were ondialysis. Unfortunately, we weren’t able to get that studycompleted. Valerie, do you have anything to add?

Dr. Valerie Johnson (Director, Division of PediatricNephrology, New York Weill Medical College of CornellUniversity, New York, New York): Body compositiondata from some European studies show no differencebetween these groups matched for age and gender [55].But nobody has looked at these groups carefully to deter-mine why they behave differently metabolically.

Dr. Fatai Bamgbola (NIH Clinical Research Fellow inPediatric Nephrology, Children’s Hospital at Montefiore,Bronx, New York): This phenomenon might be due togenetic differences among these patients. In Pima Indians,for instance, homozygous deletion of the UCP-2 genecontributes to a high prevalence of obesity [56]. Aspointed out by Dr. Kaskel, UCP-2 and UCP-3 are re-cently discovered genes, predominantly found in brownadipose tissue, which upon activation can lead to uncou-pling of mitochondrial oxidative phosphorylation andtherefore to increased basal energy expenditure. Thereis a well-designed study in patients on peritoneal dialysis,a unique population because they are constantly exposedto a highly glycemic milieu, in which increased adipositywas demonstrated in the subgroup of subjects with UCP-2homozygous gene deletion [47]. In addition, I believethe differences in biologic adaptation to metabolic stressalso can be related to factors that might include inductionof the UCP genes by TNF-� and differential activationof the ubiquitin-proteasome system, a major protein deg-radation pathway.

Dr. Chesney: Some recent studies have looked at poly-morphisms of promoter regions for TNF-� and trans-forming growth factor-� (TGF-�). Patients with TNF-�gene defects have more serious disease in certain clin-


ical conditions, such as Kawasaki disease [57]. This genedefect is also seen in adults with acute respiratory dis-tress syndrome, severe pneumonia, and sepsis [58]. Weneed to look at cytokine gene polymorphisms in patientswith CRD.

Dr. Goligorsky: Do children with CRD have hyper-homocysteinemia as adults do?

Dr. Chesney: Human studies have shown that adultswith hyperhomocysteinemia are more likely to developcardiovascular disease [59]. However, in patients withcardiovascular disease, instead of having hyperhomocys-teinemia, they have low plasma homocysteine levels.This discrepancy is probably related to the malnutritionassociated with cardiovascular disease.

Dr. Leigh Ettinger (Fellow in Pediatric Nephrology,Children’s Hospital at Montefiore, Bronx, New York): Doyou know of any study on the effect of growth hormonetherapy on cardiovascular morbidity and mortality inpatients with CRD?

Dr. Bamgbola: One study in adult patients with con-gestive heart failure showed that administration of GHimproved cardiovascular function [60].

Dr. Kaskel: Let me add that there is scanty evidenceconcerning GH’s cardiovascular effects in children, andthis issue needs to be investigated.

Dr. Sandra Blethen (Head of Endocrine Medicine,Genentech, Inc., San Francisco, California): Certainly inGH-deficient adults, particularly women, GH therapyimproves clinical outcomes. Also, a recent trial by Dr.Kaufman at the Los Angeles Children’s Hospital involv-ing GH-deficient children showed that the subjects whodid not receive GH had a very high homocysteine level(abstract; Krantz J et al, Pediatr Res 49:609, 2001).

Dr. Paul Saenger (Director of Pediatric Endocrinol-ogy, Children’s Hospital at Montefiore, Bronx, New York):Dr. Ettinger, your question addresses a possible adversecardiovascular effect of GH because it is a growth pro-moter. I don’t think we have any information on childrenwith GH deficiency. There are data on children withTurner’s syndrome and intrauterine growth retardationwho were treated with GH, using the same doses as inchildren with renal failure [61]. There were no adverseeffects on cardiovascular function as measured by cardiacultrasound in these patients.

Dr. Manju Chandra (Chief of Pediatric Nephrology,North Shore University Hospital, Long Island, New York):In infants with congenital renal malformation and renalfailure, what is the earliest time to start GH therapy?

Dr. Kaskel: We usually don’t start GH until the infantreaches 1 year of age. Dr. Blethen, would you like tocomment?

Dr. Blethen: I think the ideal approach is to providebasic nutrition and achieve metabolic correction. Dr.Fine showed that children less than 2 years of age dorespond to GH [61]. The main question is, once there is

evidence of growth failure, and you’ve done the best youcan with all the other weapons that you have, is thatthe best time to start GH? In our experience with GH-deficient children, the earlier you start, the better theoutcome in maximizing their growth.

Dr. Bamgbola: I’d like to mention the importance ofsupportive therapy. A German study of children less than3 years of age with congenital renal failure who weremanaged strictly on a conservative regimen (in a tertiaryhealth care institution, whereby adequate control of nu-trition and metabolic profiles was provided) found thatthe children sustained adequate growth in weight andstature, at least in the first 2 years of life [5]. It is thereforeof paramount importance to provide excellent support-ive care to maximize children’s growth before the ageof 2. Thereafter, optimal growth is equally dependenton the rate of GFR loss and endocrinologic factors, in-cluding GH and thyrotropin. During this latter phaseof somatic development, adjunct GH therapy is oftennecessary.

Dr. Harrington: One of the newer nutritional issuesin adult patients is carnitine deficiency. A whole industryis marketing carnitine lotion and pills to the populaceto enhance muscle strength and development. Are therestudies in children with CRD?

Dr. Kaskel: We screened 20 children on HD, and atleast 17 of them had carnitine deficiency. Studies in Europehave shown carnitine deficiency in children on dialysis[62]. Dr. Schreiber from Michigan, an expert on thistopic, believes that carnitine deficiency affects virtuallyevery system in the body [63]. This is a wide-open field,and we are about to study this problem in a more con-trolled fashion.

Dr. Goligorsky: In treating children with GH, do weuse a standard dose, or is the dose titrated accordingto certain parameters such as IGF-I level or somethingelse?

Dr. Kaskel: We use a starting dose of 0.05 mg/kg/dayat the onset and increase the dose as needed over time.

Dr. Blethen: In clinical trials, you have to keep theprotocol simple, so children are often treated with a fixeddose of GH. There is certainly a desire in the medicalcommunity to maintain the IGF-I level above averagevalues, but not excessively so. The problem with themeasurement of free IGF-I is that not everyone agreeson how to do it. There is a commercially available kit,but some people think it is not the best tool. I don’tbelieve there are sufficient solid data out there.

Dr. Johnson: Growth rate drops after the second tothird year of GH therapy. Do you think that the IGF-Ilevel has something to do with it?

Dr. Blethen: The so-called “waning effect of GH ther-apy” has been recognized for years. We still don’t under-stand it. I believe your observation is correct, becauseyou can overcome the waning effect in some instances


by bumping up the dose of the GH. Many years ago,Dr. Gertner did a study when GH was provided by thegovernment and many were given a fixed-dose regimen[64]. He persuaded one of the manufacturers to give himextra supplies, and he found that he could increase thegrowth rate by increasing the dose. I know he didn’tlook at IGF-I in that study, but even if he had, thedata would have been difficult to interpret given thelimitations of the IGF assays available then.

Dr. Chesney: I have one comment about GHBP inchildren with CRD. Unless one has a very precise anti-body that does not cross-react with other peptides, thepresence of many peptide compounds in uremic plasmaconfounds their measurement. This is certainly true withrespect to GH and its binding proteins. And it accountsfor the difficulty in measuring GHBP in uremic childrenand comparing it with measurements in other patientssuch as those with Turner’s syndrome or partial lipodys-trophy.

Reprint requests to Dr. Frederick Kaskel, Division of Nephrology,Children’s Hospital at Montefiore, 111 East 210th Street, Bronx, NewYork, New York, USA.E-mail: [email protected]

REFERENCES

1. Wong CS, Gipson DS, Gillen DL, et al: Anthropometric measuresand risk of death in children with end-stage renal disease. Am JKidney Dis 36:811–819, 2000

2. Seikaly M, Ho PL, Emmett L, Tejani A: 12th Annual Report ofthe North American Pediatric Renal Transplant Cooperative Study(NAPRTCS): Renal transplantation from 1987 through 1998. Pedi-atr Transplant 5:215–231, 2001

3. Rizzoni G, Broyer M, Brunner HP, et al: Combined report onregular hemodialysis and transplantation in Europe, 1985. ProcEDTA 23:55–83, 1986

4. Schaefer F, Mehls O: Endocrine and growth disturbances, inPediatric Nephrology, 4th ed, edited by Barratt MT, Avner ED,Harmon WE, Baltimore, Lippincott Williams and Wilkins, 1999,pp 1197–1230

5. Van Dyck M, Bilem N, Proesmans W: Conservative treatmentfor chronic renal failure from birth. Pediatr Nephrol 13:865–869,1999

6. Schaefer F, Wingen AM, Hennicke M, et al: Growth charts forpre-pubertal children with chronic renal failure due to congenitalrenal disorders. European Study Group for Nutritional Treatmentof Chronic Renal Failure in Childhood. Pediatr Nephrol 10:288–293, 1996

7. Scharer K: Study Group on Pubertal Development in ChronicRenal Failure. Growth and development of children with chronicrenal failure. Acta Pediatr Scand 366:90–92, 1990

8. Schaefer F, Seidel C, Binding A, et al: Pubertal growth in chronicrenal failure. Pediatr Res 28:5–10, 1990

9. Phadke A, MacKinnon K, Dossetor J: Male fertility in uremia:Restoration by renal allografts. CMAJ 102:607–608, 1970

10. Roelfsema V, Clark RG: The growth hormone and insulin-likegrowth factor axis: Its manipulation for the benefit of growth disor-ders in renal failure. J Am Soc Nephrol 12:1297–1306, 2001

11. Chen E, Liao Y, Smith D, et al: The human growth hormone locus:Nucleotide sequence, biology, and evolution. Genomics 4:479–497,1989

12. Goodman HM, Tai LR, Ray J, et al: Human growth hormonevariant produces insulin-like and lipolytic responses in rat adiposetissue. Endocrinology 129:1779–1783, 1991

13. Kojima M, Hosoda H, Date Y, et al: Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402:656–660, 1999

14. Shimon I, Taylor JE, Dong JZ, et al: Somatostatin receptor sub-type specificity in human fetal pituitary cultures: Differential roleof SSTR2 and SSTR5 for growth hormone, thyroid stimulatinghormone and prolactin regulation. J Clin Invest 99:789–812, 1997

15. Powell DR, Durham SK, Liu F, et al: The insulin-like growthfactor axis and growth in children with chronic renal failure: AReport of Southwest Pediatric Nephrology Study Group. J ClinEndocrinol Metab 83:1654–1661, 1998

16. Yakar S, Liu J, Le Roith D: The growth hormone/insulin-likegrowth factor system: Implications for organ growth and develop-ment. Pediatr Nephrol 14:544–549, 2000

17. Feld S, Hirschberg R: Growth hormone, the insulin-like growth-factor system, and the kidney. Endocrinol Rev 17:423–480, 1996

18. Rodger RS, Dewar JH, Turner SJ, et al: Anterior pituitary dys-function in patients with chronic renal failure treated by hemodialy-sis or continuous ambulatory peritoneal dialysis. Nephron 43:169–172, 1986

19. Haffner D, Schaefer F, Girard J, et al: Metabolic clearance ofrecombinant human growth hormone in health and in chronic renalfailure. J Clin Invest 93:1163–1171, 1994

20. Edmondson SR, Baker NL, Oh J, et al: Growth hormone receptorabundance in tibial growth plates of uremic rats: GH/IGF-I treat-ment. Kidney Int 58:62–70, 2000

21. Schaefer F, Chen Y, Tsao T, et al: Impaired JAK-STAT signaltransduction contributes to growth hormone resistance in chronicuremia. J Clin Invest 108:467–475, 2001

22. Schwartz MW, Woods SC, Porte D, Jr, et al: Central nervoussystem control of food intake. Nature 404:661–671, 2000

23. Bowers CY: Unnatural growth hormone-releasing peptide begetsnatural ghrelin. J Clin Endocrinol Metab 86:1464–1469, 2001

24. Takaya K, Ariyasu H, Kanamoto N, et al: Ghrelin strongly stimu-lates growth hormone (GH) release in humans. J Clin EndocrinolMetab 85:4908–4911, 2000

25. Wren AM, Seal LJ, Cohen MA, et al: Ghrelin enhances appetiteand increases food intake in humans. J Clin Endocrinol Metab86:5992–5995, 2001

26. Shiiya T, Nakazato M, Mizuta M, et al: Plasma ghrelin levels inlean and obese humans and the effect of glucose on ghrelin secre-tion. J Clin Endocrinol Metab 87:240–244, 2002

27. Zhang Y, Proenca R, Maffei M, et al: Positional cloning of themouse obese gene and its human homologue. Nature 372:425–432,1994

28. Caro JF, Sinha MK, Kolaczynski JW, et al: Decreased cerebrospi-nal fluid-leptin ratio in obesity: A possible mechanism of leptinresistance. Lancet 348:159–161, 1996

29. Halaas JL, Gajiwala KS, Maffei M, et al: Weight reductioneffect of the plasma protein encoded by the obese gene. Science269:543–546, 1995

30. Sharma K, Considine RV, Michael B, et al: Plasma leptin is partlycleared by the kidney and is elevated in hemodialysis patients.Kidney Int 51:1980–1985, 1997

31. Grunfeld C, Zhao C, Fuller J, et al: Endotoxins and cytokinesinduce expression of leptin, the ob gene products, in hamsters. JClin Invest 97:2152–2157, 1996

32. Heimburger O, Lonnqvist F, Danielsson A, et al: Serum immuno-reactive leptin concentration and its relation to the body fat contentin chronic renal failure. J Am Soc Nephrol 8:1423–1430, 1997

33. Don BR, Rosales LM, Levine NW, et al: Leptin is a negativeacute phase protein in chronic hemodialysis patients. Kidney Int59:1114–1120, 2001

34. Garibotto G, Barreca A, Sofia A, et al: Effects of growth hor-mone on leptin metabolism and energy expenditure in hemodialy-sis patients with protein-calorie malnutrition. J Am Soc Nephrol11:2106–2113, 2000

35. Young GA, Woodrow G, Kendall S, et al: Increased leptin/fatratio in patients with chronic renal failure: A cause of malnutrition?Nephrol Dial Transplant 12:2318–2323, 1997

36. Odamaki M, Furuya R, Yoneyama T, et al: Association of serumleptin concentration with weight loss in chronic hemodialysis pa-tients. Am J Kidney Dis 33:361–368, 1999

37. Chung SH, Lindholm B, Lee HB: Influence of initial nutritional


status on continuous ambulatory peritoneal dialysis patient sur-vival. Perit Dial Int 20:19–26, 2000

38. Kopple JD, Greene T, Chumlea WC, et al: Relationship betweennutritional status and the glomerular filtration rate: Results fromthe MDRD study. Kidney Int 57:1688–1703, 2000

39. Kuhlmann U, Schwickardi M, Trebst R, Lange H: Resting meta-bolic rate in chronic renal failure. J Ren Nutr 11:202–206, 2001

40. Tom K, Young VR, Chapman T, et al: Long-term adaptive re-sponses to dietary protein restriction in chronic renal failure. AmJ Physiol 268:E668–E677, 1995

41. Lim VS: Thyroid function in patients with chronic renal failure.Proceedings of the Second International Congress of Uremia Re-search, Nara, Japan, April 2001: Metabolic dysfunction in uremia.Am J Kidney Dis 38:580–584, 2001

42. O’Sullivan AJ, Lawson JA, Chan M, Kelly JJ: Body composi-tion and energy metabolism in chronic renal insufficiency. Am JKidney Dis 39:369–375, 2002

43. Mitch WE: Insights into the abnormalities of chronic renal diseaseattributed to malnutrition: Pathophysiology of chronic renal failureand complications. J Am Soc Nephrol 13:S22–S27, 2002

44. Ding X, Price SR, Bailey JL, Mitch WE: Cellular mechanismscontrolling protein degradation in catabolic states. Miner Electro-lyte Metab 23:194–197, 1997

45. Ikizler TA, Wingard RL, Sun M, et al: Increased energy expendi-ture in hemodialysis patients. J Am Soc Nephrol 7:2646–2653, 1996

46. Astrup A, Toubro S, Dalgaard LT, et al: Impact of the v/v 55polymorphism of the uncoupling protein 2 gene on 24-h energyexpenditure and substrate oxidation. Int J Obes Relat Metab Disord23:1030–1034, 1999

47. Nordfors L, Heimburger O, Lonnqvist F, et al: Fat tissue accumu-lation during peritoneal dialysis is associated with a polymorphismin uncoupling protein 2. Kidney Int 57:1713–1719, 2000

48. Doi T, Striker LJ, Gibson CC, et al: Glomerular lesions in micetransgenic for growth hormone and insulin-like growth factor-1.1. Relationship between increased glomerular size and mesangialsclerosis. Am J Pathol 137:541–552, 1990

49. Bahayana S, Hegele RA: The molecular basis of genetic lipodys-trophies. Clin Biochem 35:171–177, 2002

50. Shimomura I, Hammer RE, Ikemoto S, et al: Leptin reverses insulinresistance and diabetes mellitus in mice with congenital lipodystro-phy. Nature 401:73–76, 1999

51. Leanane P, Murphy G: Partial lipodystrophy and renal disease.Clin Exp Dermatol 25:605–607, 2000

52. Williams DG: C3 nephritic factor and mesangiocapillary glomeru-lonephritis. Pediatr Nephrol 11:96–98, 1997

53. Tonshoff B, Eden S, Weiser E, et al: Reduced hepatic growthhormone (GH) receptor gene expression and increased plasma GHbinding protein in experimental uremia. Kidney Int 45:1085–1092,1994

54. Houang M, Cabrol S, Perin L, et al: Insulin-like growth factor-1(IGF-1), insulin-like growth factor binding proteins (IGFBP) andinsulin-like growth factor type 1 receptor in children with variousstatus of chronic renal failure. Growth Horm IGF Res 10:332–341,2000

55. Boot AM, Nauta J, de Jong MCJW, et al: Bone mineral density,bone metabolism and body composition of children with chronicrenal failure, with and without growth hormone treatment. ClinEndocrinol 49:665–672, 1998

56. Walder K, Norman RA, Hanson RL, et al: Association betweenuncoupling protein polymorphisms (UCP2–UCP3) and energy me-tabolism/obesity in Pima Indians. Hum Mol Genet 7:1431–1435,1998

57. Quasney MW, Bronstein DE, Cantor RM, et al: Increased fre-quency of alleles associated with elevated tumor necrosis factor-alfa levels in children with Kawasaki disease. Pediatr Res 49:686–690, 2001

58. Freeman BD, Buchman TG: Gene in haystack: Tumor necrosisfactor polymorphisms and outcome in sepsis. Crit Care Med28:3090–3091, 2000

59. Kerins DM, Koury MJ, Cadevila A, et al: Plasma S-adenosylho-mocysteine is a more sensitive indicator of cardiovascular diseasethan plasma homocysteine. Am J Clin Nutr 74:723–729, 2001

60. Giustina A, Volterrani M, Manelli F, et al: Endocrine predictorsof acute hemodynamic effects of growth hormone in congestiveheart failure. Am Heart J 137:1035–1043, 1999

61. Fine RN, Stablein D, Cohen AH, et al: Recombinant humangrowth hormone post renal transplantation in children: A random-ized controlled study of the NAPRTCS. Kidney Int 62:688–696,2002

62. Mir S, Kantar M, Yalaz M, et al: Effect of hemodialysis oncarnitine levels in children with chronic renal failure. Pediatr Int44:70–73, 2002

63. Wang Y, Korman SH, Ye J, et al: Phenotype and genotype inprimary carnitine deficiency. Genet Med 3:387–392, 2001

64. Gertner JM, Tamborlane WV, Gianfredi SP, Genel M: Re-newed catch-up growth with increased replacement doses of humangrowth hormone. J Pediatr 110:425–428, 1987

chronic renal disease: a growing problem

Documents