UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Late effects of renal insufficiency in children. A national long-term follow-up study

Groothoff, J.W.

Link to publication

Citation for published version (APA):Groothoff, J. W. (2002). Late effects of renal insufficiency in children. A national long-term follow-up study.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 19 Oct 2020

8.. Severe bone disease and low bone failure minerall density after juvenile renal failure

Jaapp W Groothoff

Martinn Offringa

Berthee LF van Eck-Smit

Marikenn P Gruppen

Nicolee CAJ van de Kar

Ericc D Wolff

Marcc R Lilien

Jeann Claude Davin

Hugoo SA Hey mans

Friedoo W Dekker

KidneyKidney International, January 2003, in Press

115 5

8.8. Bone Disease

Abstract. .

Background:Background: Littl e is known about the late effects of juvenile end-stage renal diseasee (ESRD) on bone integrity. To establish clinical manifestations of metabolicc bone disease and Bone Mineral Density (BMD) in young adult patientss with juvenile ESRD, we performed a long-term outcome study.

Methods:Methods: A cohort was formed of all Dutch patients with onset of ESRD between 19722 and 1992 at age 0-14 years, born before 1979. Data were collected by revieww of medical charts, by current history, by physical examination, and by performingg Dual Energy X-ray Absorptiometry (DEXA) of the lumbar spine andd the femoral neck.

Results:Results: We retrieved clinical information in 247 out of 249 patients. Of all patients,, 61.4% had severe growth-retardation (<-2SD), 36.8% had clinical symptomss of bone disease, and 17.8% was disabled by bone disease. Growth-retardationn and clinical bone disease were associated with a long duration of dialysis.. DEXA was performed in 140 out of 187 living patients. Mean BMD SD correctedd for gender and age (Z score) of the lumbar spine was -2.12 (SD1.4), of thee femoral neck -1.77 (SD1.4). A low lean body mass was associated with a low lumbarr spine and a low femoral neck BMD; male gender, physical inactivity andd aseptic bone necrosis were associated with a low lumbar spine BMD.

Conclusion:Conclusion: Bone disease is a major clinical problem in young adults with pediatricc ESRD. Further follow-up is needed to establish the impact of the low bonee mineral densities found in these patients.

Introduction. .

Ass lif e expectancy has increased in children with chronic renal replacement therapyy (RRT), concern has arisen about its late complications. In adults, metabolicc bone disease is a common, often disabling, complication of end-stage renall disease (ESRD) [1-4]. Chronic acidosis and hypovitamin-D- and phosphate-inducedd secondary hyperparathyroidism can lead to disturbed bone modelingg with high or low bone turn-over in patients on chronic dialysis [5-8]. Inn addition, inactivity during chronic dialysis and corticosteroids, mainly given chronicallyy as anti-rejection therapy after transplantation, may lead to osteoporosiss [9-14]. In children with end-stage renal disease, malnutrition, reducedd bioactivity of Insulin-like growth-hormones and delayed puberty can aggravatee these effects, disturbing normal bone formation and growth [15-20]. Littl ee is known about the effects of ESRD of childhood on bone formation at adultt age, and about the extent of clinical bone disease in long term survivors of juvenilee ESRD.

116 6

8.8. Bone Disease

Wee conducted a national long-term follow up study in order to evaluate Late physical,, social and psychological Effects of Renal Insufficiency in Children (LERIC).. The cohort consisted of all Dutch children who had commenced RRT betweenn 1972 and 1992 and who had reached (or would have reached if still alive)) adulthood at the time of investigation. In this paper we report on the impactt on growth, the incidence of fractures, clinically apparent bone disease andd bone densities measured by DEXA in these young adults with pediatric onsett of ESRD.

Methods. .

StudyStudy design. The study was designed as a cohort study and consisted of a cross-sectionall and a retrospective part. The aim of the cross-sectional study was to establishh current health status and to measure bone mineral density of the lumbarr spine and femoral neck, in order to detect osteoporosis. The aim of the retrospectivee part of the study was to establish the extent of clinically apparent metabolicc bone disease since the start of RRT, and to evaluate the influence of a sett of predefined determinants on outcome parameters. The study covered the totall period of renal replacement therapy for each patient. The end of the study wass marked by the day of last chart review for potential non-participants in the cross-sectionall study and the day of the cross-sectional examination for participants.. The medical ethical committees of all participating centers approvedd the study.

FormationFormation of the cohort. The LERIC cohort comprises all Dutch patients who had startedd chronic RRT at age 0-14 years between 1972 and 1992, and who were bornn before 1979. Patients in whom renal function recovered within four monthss after commencing dialysis were excluded. Pre-emptively transplanted patientss were included. Patients who started RRT after 1991 were excluded in orderr to have at least a (potential) follow up period of 6 years. The National Dutchh Registry of patients on RRT (RENINE, Rotterdam, The Netherlands) submittedd a list of patients, based on these criteria. RENINE, founded in 1985, is thee Dutch source of the European Dialysis and Transplantation Association. Thee completeness of this database approach 100% as registration is compulsory forr reimbursement of RRT.

Retrospectivee registration of patients with RRT starting before 1985 was checkedd by comparing RENINE-data with the databases of all Dutch centers for pediatricc dialysis and kidney transplantation as well as the databases of all centerss for adult dialysis and transplantation.

Thee procedure of cohort formation was as follows: the list submitted by RENINEE consisted of only registry-numbers of patients, treatment-modality,

117 7

8.8. Bone Disease

andd the name of the last physician and hospital of treatment. All nephrologists inn the Netherlands received a list of those cohort-patients who were under their treatmentt or had been under their treatment at the time of death. The LERIC-teamm contacted all the physicians, asking them if the list sent by RENINE was consistentt with their registry-database. Patients were contacted by their own physicianss and invited to take part in this study.

DataData collection: Clinical bone disease. Between November 1998 and August 2000 memberss of the LERIC-team visited 37 hospitals in the Netherlands. They collectedd information about the occurrence of bone disease (pathological fractures,, deformities caused by metabolic bone disease, severe chronic bone pain,, aseptic bone necrosis and handicaps caused by bone disease) and determinantss with respect to these outcomes. The predefined variables were: totall duration of RRT and of hemodialysis, peritoneal dialysis and transplantation,, the age at onset of RRT, and the period of onset of RRT. All medicall charts of all patients, participants as well as non-participants in the cross-sectionall study, were reviewed. Emigrated patients were located and medicall information was obtained from their current physician. "Clinical bone disease"" was considered present in cases where at least one of the following conditionss occurred: deforming bone abnormalities, chronic pain related to the skeletall system, disabling bone abnormalities, aseptic bone necrosis and atraumaticc fractures. Disabling bone abnormalities were scored as "mild" , if the patientt was disabled, but main functions (walking, writing) were preserved. For instance,, patients who were not able to walk for more than 100 meters or in whomm motor functions were hindered by pain, all caused by metabolic bone disease,, were included in his category. Severe disabling bone abnormalities referredd to those conditions in which main motor functions were impaired by bonee disease (walking, standing up), or to conditions in which patients were chronicallyy dependent on special devices (e.g. cervical braces, external fixtures). Motorr disabilities caused by neurological disorders or extreme fatigue were excluded,, unless the neurological deficit was the result of metabolic bone diseasee (e.g. spinal cord lesion by spine fracture).

BoneBone mineral densities. Al l patients alive were invited once for examination in ourr hospital. Height was measured with a fixed stadiometer. Bone mineral densityy and lean body masses were established by Dual Energy X-ray Absorptiometryy (DEXA, QDR 4500A Hologic Inc., Waltham, Mass. USA). We assessedd bone mineral densities of the femoral neck and the lumbar spine at Ll -L4.. Bone mineral density (BMD) was defined as bone mineral content, divided byy the surface of the projected bone area, expressed in g/cm2.

Ass determinants with respect to bone mineral density we considered: physical inactivity,, gender, height, lean body mass, modality of RRT at time of investigation,, total duration of renal replacement therapy, total duration of dialysis,, age of onset of RRT, period of onset of renal replacement therapy (i.e.

118 8

8.8. Bone Disease

1972-19811 versus 1982-1992), the period between transplantation and bone densityy measurements, severe hyperparathyroidism, age of menarche in women,, the cumulative steroid dosage, aluminum-containing phosphate binders,, growth-hormone therapy, and a history of aseptic bone necrosis.

Mostt data on determinants were collected retrospectively as mentioned under "clinicall bone disease."

Otherr information, such as age of menarche in women and physical inactivity, wass collected during interview of the patient. We assessed information about physicall activity using parts of the MOS-SF 36 questionnaire (question number 3b:: "does your health now limi t you in moderate activities?" and question number3h:: "does your health limi t you from walking several blocks"), that appliess to a limited physical activity. The MOS-SF36 is a widely used tool for qualityy of life assessment [22]. It was assumed that true daily physical activity wass correlated with the subjectively scored ability of physical activity. "Physical inactivity"" was scored positive if both questions were answered with "yes".

Sincee the exact dosage of corticosteroids was not always documented reliably, wee scored the total time in days of which we were sure that the prescribed prednisonee or prednisolone dosage exceeded > 0.25 mg/kg per day. We also scoredd the total time of alternate day prescription of more than 0.25 mg/kg. Sincee all courses of "high-dose" methylprednisolone, given as anti-graft-rejectionn therapy, were prescribed according to well-defined protocols, we were ablee to express the total cumulative dosage of "high dose" methylprednisolone inn mg/kg. We estimated the cumulative dosage of corticosteroids by calculating thee total amount of predniso(lo)ne, assuming that the mean dosage of prednisonee would have been 0.5 mg/kg during the scored period. We added thee cumulative iv methylprednisolone dosage to the estimated oral steroid dosage.. When applicable, duration of growth hormone-therapy was established.. Severe hyperparathyroidism was scored "positive" if parathyroid hormonee had ever exceeded 3 times the normal value over a cumulative period off more than 12 months, or if a parathyroidectomy had been performed.

Heightt and clinical bone abnormalities in patients, who participated in the cross-sectionall study, were assessed by interview and physical examination. Thesee data were combined with data derived from the medical charts. Determinantss of bone disease of all the patients and data about clinically apparentt bone disease, growth, disabilities and atraumatic fractures of non-participantss and deceased patients were assessed from the medical charts.

ControlControl values: Control values of weight and height were derived from the Nationall Dutch Health Statistics (www.cbs.nl), of lean body mass from data of Boott et al. [23]. We compared BMD values with control values of European and Americann subjects, supplied by Hologic Inc.

119 9

8.8. Bone Disease

Tablee 1. Patient characteristics.

Patientss (n)

Malee (n)

Primaryy disease (n) Glomerulopathy y Obstructivee uropathy Congenitall renal malformation1

Metabolicc disease2

HUS S SLE E Other r

Agee at onset of RRT

Agee at time of investigation

Deceasedd (n)

Durationn of RRT

Durationn of dialysis

Durationn of transplantation

Startt RRT 1972-1981 (n)

249 9

136 6

90 0 70 0 44 4 8 8

18 8 5 5 14 4

10.6 6

29.3 3

63 3

16.0 0

4.1 1

11.7 7

151 1

(54.6%) )

(36.1%) ) (28.1%) ) (17.7%) ) (3.2%) ) (7.2%) ) (2.0%) ) (5.6%) )

(1.9-14.9)3 3

(20.7-41.7)̂ ^

(25.3%) )

(0.3-31.1)3 3

(0-25.6)3 3

(0-29.9)3 3

(60.6%) )

RRTRRT = renal replacement therapy; l renal dysplasia, cystic disease, Alport's Syndrome, nephronophtisis,nephronophtisis, congenital nephrotic syndrome; 2 oxalosis (3), cystinosis (5) 3 range; n= numbernumber of patients. Age and duration in years.

Tablee 2. Growth retardation and clinical manifestations of bone disease of all patients,, part icipants and non-participants of the cross-sectional study.

TotalTotal cohort: Participants: Non-participants: NN affected N affected N affected (n=247p(n=247p (n=140y (n=47p

Heightt <-2SD 153(61.9%) 83(59.3%) 31(66.0%)

Clinicall manifestations of bone disease 91 (36.8%) 51 (36.4%) 17(36.2%)

Deformitiess 63 (25.5%) 41 (29.3%) 9 (19.1%)

Pathologicall fractures 33 (13.4%) 16 (11.4%) 8 (17.0%)

Asepticc bone necrosis 32 (13.0%) 16 (11.4%) 8 (17.0%)

Mil dd disabling bone disease 26 (10.5%) 17 (12.1%)" 4 (8.5%)

Severee disabling bone disease 18 (7.3%) 8 (5.7%)* 3 (6.4%)

Invalidatingg bone disease (all) 44 (17.8%) 25 (17.9%)* 7 (14.9%) 11 Of two patients, who died in the early 1970's, no data on growth or bone abnormalities could bebe found; 2 participants in the cross-sectional study; 3 Living Non-participants in the cross-sectionalsectional study; 4 situation at time of cross-sectional investigation.

120 0

8.8. Bone Disease

StatisticalStatistical analysis. The relation between clinical bone disease and its determinantss was analyzed using the Student's t test for continuous variables withh a normal distribution, and with the chi-square test for nominal variables. AA comparison of nominal variables of participants and non-participants of the cross-sectionall study was performed by the chi-square-test. The relation betweenn Z scores of BMD and dichotomous determinants were analyzed by usingg the Student's t test, or if applicable the Mann-Whitney test. Pearson's correlationn was assessed between BMD Z scores of lumbar spine and femoral neckk and all potential determinants, except those who were only measured in subgroupss (i.e. menarche and the period between the last transplantation and momentt of investigation in transplanted patients). All significant determinants (sett at p<0.2) identified from these univariate analyses were studied with a linearr stepwise multiple regression using the F-statistics with p=0.05 as criterion forr selection. We used the regression coefficient relating lean body mass with bothh lumbar spine and femoral neck BMD and the difference in lean body mass betweenn patients and controls to calculate the predicted BMD and Z score, adjustedd for lean body mass.

Results. .

TheThe cohort. The RENINE database produced a list of 251 patients, who fulfilled thee inclusion criteria. Checking with the local databases of all participating centerss revealed the following information: one patient was mentioned twice andd three patients did not meet to the inclusion criteria. Reviewing the databasess of all dialysis and transplantation centers produced two extra patients.. According to RENINE, 7 out of 251 patients were lost to follow up. Six off these patients appeared to be alive and one patient had passed away. Thus thee definite cohort consisted of 249 patients. Of these, at time of the study, 63 patientss (25.3%) had died; one of them died 4 months after the cross-sectional investigation,, and was included as participant. No patients were lost to follow up. .

Off all LERIC patients we recorded a total of 3870 patient years, covering the wholee period from the onset of RRT until august 2000 or time of death, with 81 patientt years missing and with an average follow up of 15.5 years per patient. Off 2 patients (0.8%) who both died in 1974, no clinical data on bone disease couldd be found. The main characteristics of all the patients are listed in table 1. Off 187 patients alive 47 (25.1%) declined to participate in the cross-sectional study,, leaving 140 subjects. No significant differences were found in age, gender,, age of onset of RRT and therapy characteristics between participants andd non-participants of the cross-sectional study.

121 1

Cu u

o o

OJ J

D--fö ö i~ i~ <v <v

X, ,

O) )

T3 3 C C

01 1 C C O O

-Cl l

u u

!-0 0

^^ ^

O O

aii c

UU >«

U U

rt rt 01 1

1 1 G G 01 1 01 1

0) ) Xl l r; ; o o 03 3 (IJ J « «

co o 01 1

X» »

H H

n n o o CO O

C C CÖ Ö

Cu u en n C C rö ö Ü Ü

C C

t/ i i

03 3

«J J

TJ J

Q Q <-n n

QQ g ( N N

11 — v

VV Q * -- (LH

SS S a:: >H

o o Cv v

I - H H

co o ** *

**" " Cv , ,

O N N

r - H H

ON~" "

ïE--

ON N T — l l

en n

o o

m m CN N

CN N t v v

(N N N£> >

t v v O O

l O O

co o

ï?r r cv^ ^

ON N

O O I—1 1

LH" " QO, ,

H H

H H (Tl l 00 0 O O

i i

T - H H

T - H H

co o cv v

* *

cN N co" " o o

00 0

o o

f? ? ^ ^ o o

* *

orT T 2L--00 0

^ ^ CN N r-T T O O

CO O i—1 1

I—1 1

** *

CN N

T-H H

r-t r-t

r H H

O O

r H H

CO, ,

IN N O O r H H

ro" " m. . ND D O O r - i i

i n n

o o T-4 T-4

r H H l l

co o cp p

r H H

co o

Cv" " r H H

Cv v

co o

fT T LO, ,

rs i i CT^ CT^

CN" "

^ ^

CN N

o o 1 1

ÓC ÓC cN N i i

m m r H H

CN N i n n

\o o t v v

r H H

cv' ' r - i i

NO O

00 0

ö ö NO O

CN N

CO O tv ^ ^

CN N

"* * ^ ^

ON N Cv v

r H H

CN N T-H H

ro o CN N t v v

t t

r O O O O

m m Cv, ,

NO O r H H

CO O

"* * ON N

SS S c o o CN 1 1

T-H H

<* * CN N

O O CO, ,

NO O CN N

CN N

co o

d d r H H

ui i o o o o i n n r H H

l l

00 0 O O

ON N

d d co o

r H H i n n

00 0 Cv' '

3* * Ö Ö CN N

CO O i i

00 0 r^ r^

CN N Ln , ,

ON N

o\ \

Cv^ ^

00 0 NÓ Ó

Cv v ON N

O O

C/l l

e e <U <U

Tl l DH H

CO O

.s s tó tó tó tó

re e O) )

C C X X f --

cu u

X X U U

H H K K es s ffl ffl CO O

bt> > < <

j22 to

CN N t v . . ON N

SS c

fXX v^ 011 O)

CN N 00 0

t v v

CN N

33 r

00 0

ö ö

CO O

00 0 ON N

OJO O

c c

^ r r

vv ^

^ E - H H

ess " 3

SIM M LS.S S o o

§--c c o o K K

-<y y t n n

«« S £2

OO -ïi d^ vv O ON

"^^ 'S &v

55 ^ sf ^^ O , H-

-- f § -. sE- c

^^ JT « ^^ a "55 oo o II

§§ N2>K ^^ -c -S 22 .t: *H_ ss 42 2 p>> .2j II

8.8. Bone Disease

Forr logistical reasons, lumbar spine and femoral neck DEXA mass was not performedd in 4 out of 140 (2.9%) subjects. In two patients, DEXA measurement off the femoral neck was not possible because of a hip prosthesis.

ClinicalClinical bone disease. The mean height of all the patients was 160.6 cm (range 138.5-185).. Severe growth retardation (SD < -2.0) was seen in 61.4% of all LERIC-patients.. Clinical manifestations of bone disease had occurred in 36.8%, andd motor handicaps, caused by metabolic bone disease in 17.8 % of all patients.. Mil d disabilities were seen in 10.5%, severe disabilities in 7.3% of all patients.. No differences were seen in clinical bone disease, growth and motor handicapss between the total cohort, the participants and the non-participants in thee cross-sectional study (table 2).

Bothh clinical bone disease and severe growth retardation were associated with a longerr duration of RRT, a longer duration of dialysis and with an initiation of RRTT earlier in history. Patients with aseptic bone disease had on average a longerr duration of total RRT, a longer time with a well functioning renal graft, andd a start of RRT earlier in history, than those without aseptic bone necrosis. Severee bone disease was more apparent in patients with a history of severe hyperthyroidismm (table 3).

Thee occurrence of atraumatic fractures was associated with longer total dialysis duration.. No association was found between clinical manifestations of bone diseasee and the cumulative dosage of steroids. Patients with severe growth retardationn tended to have had more steroids than those with normal growth, a differencee that however was not statistically significant (table 3).

BoneBone mineral density. The mean standard deviation corrected for age and gender, off the BMD (Z score) of the lumbar spine was -2.12 (SD 1.4), of the femoral neck -1.777 (SD 1.4). For male patients the mean Z score of the BMD of the lumbar spinee was -2.38 (SD 1.4), and of the femoral neck -1.64 (SD 1.3). In female patientss the Z-scores of the BMD of the lumbar spine was -1.80 (SD 1.3), and of thee femoral neck -1.86 (SD 1.3). Gender specified BMD, height and weight are presentedd in table 4. AA lumbar spine Z score of less than -1SD, consistent with at minimum osteopenia,, was found in 111 (81.6%) of all 136 patients measured; 52 patients (38.2%)) had a lumbar spine Z-score of less than -2.5 SD. Of 134 patients 104 (77.6%)) had a femoral neck Z score of less than -1 SD and 51 (38.1%) had a femorall neck Z score of less than -2.5 SD. Of all 136 patients 121 (89.0%) had at leastt in one segment a BMD Z score of less than -1 SD and 71 (52.2%) had at leastt in one segment a BMD Z score of less than < -2.5 SD.

DeterminantsDeterminants of bone mineral density. No relation was found between age at commencementt of RRT, period of onset of RRT (1972-1981 versus 1982-1991), totall duration of RRT, total duration of dialysis, age of menarche in women and

123 3

8.8. Bone Disease

Tablee 4. Weight, height, lean body mass, and Bone Mineral Density (BMD) of lumbarr spine and femoral neck in male and female patients, compared to healthyy males and females.

Agee in yr. (range)

Meann height (cm)

Meann weight (kg)

Leann body mass (kg)

Meann BMD Lumbar spine

ZZ score BMD Lumbar spine1

Meann BMD femoral neck

ZZ score BMD femoral neck1

Meann age in yr. (range)

Meann height (cm)

Meann weight (kg)

Leann body mass (kg)

Meann BMD Lumbar spine

ZZ score mean BMD Lumbar spine1

Meann BMD femoral neck

ZZ score BMD femoral neck1

MaleMale patients (n=73)

20.7-41.7 7

164.77 (SD 9.6)

63.88 (SD13.5)

47.77 (SD7.8)

0.833 (SD 0.15)2

-2.388 (SD 1.4)2

0.766 (SD 0.14)3

-1.644 (SD 1.3)3

FemaleFemale patients (n=64)

20.88 - 40.6

156.11 (SD 8.9)

58.00 (SD13.4)

37.33 (SD5.6)

0.844 (SD 0.14)7

-1.800 (SD 1.3)7

0.700 (SD 0.13)8

-1.866 (SD 1.3)8

Controls Controls

20-40 0

182.99 (SD 8.0)4

80.44 (SD 9.1)4

59.11 (SD 7.5)5

1.099 (SD0.11)

0 0

0.911 (SD0.13)

0 0

Controls Controls

20-40 0

169.33 (SD 4.4)4

67.00 (SD 7.3)4

42.99 (SD 8.4)5

1.044 (SD 0.11)ö

0 0

1.155 (SD 0.14)6

0 0

MeanMean values and standard deviations (SD) are give, except for age. l Z score = Standard DeviationDeviation corrected for age and gender;2 n = 71;s n = 70;4 data from Centraal Bureau voor de StatistiekStatistiek (National Health Statistics: www.cbs.nl); 5 from Boot el al [23]; 6 data supplied by HologicHologic Inc; BMD in g/cm2;7 n = 63;8 n = 61.

124 4

8.8. Bone Disease

severee hyperparathyroidism on the one hand, versus bone mineral values on thee other hand. Significant lower mean Z scores of the lumbar spine were found inn male/ compared with female patients, in patients with physical inactivity, comparedd to those with normal activity, and in patients who received a renal transplantt less than 3 years ago, compared to those who lived for more than 3 yearss with a functioning renal graft at time of investigation. Mean differences in ZZ scores of lumbar spine and femoral neck BMD in patients categorized accordingg to potential determinants are given in table 5.

Multivariatee linear stepwise regression analysis revealed an association betweenn a low BMD Z score of the lumbar spine on one side, and male gender ((33 = -0.49, p<0.0001) and physical inactivity (P=-0.22, p=0.014) on the other side. AA low lean body mass was associated with a low lumbar spine Z-score ([3=0.46, p<0.0001),, as well as with a low femoral neck Z- score (P=0.41, p<0.0001) If the leann body mass was excluded from the analysis as variable, dialysis as modality off RRT at time of investigation was associated with a low femoral neck BMD Z scoree (B=-0.18,. p=0.03).

InIn the multivariate regression analysis height was not correlated with Z scores off both BMDs. Table 6 and 7 show the univariate and multivariate relations betweenn determinants and Z scores of the lumbar spine and femoral neck BMD. Thee predicted Z scores of the lumbar spine and femoral neck BMD of male and femalee patients, when adjusted for lean body mass are given in table 8.

Ass all our patients had received aluminum-containing phosphate binders at somee point, and the documentation of the dosage proved to be unreliable, furtherr analysis on these drugs could not be done. Growth-hormone therapy wass given to 43 (17.3%) patients. Since these were the first patients in the Netherlandss ever to receive growth hormone therapy on the indication of severee growth retardation caused by ESRD, we also excluded this drug from furtherr analysis for reasons of patient-bias. The mean cumulative period of prednisone-dosagee of more than 0.25 mg/kg was 2.63 years (SD 3.02). The meann dosage of methylprednisolone was 180 mg/kg. We estimated the overall meann dosage of prednisone to be 661 (SD 665) mg/kg over a mean cumulative periodd of 13.67 (SD 6.44) transplantation-years. Although the cumulative dosagee of corticosteroids was not correlated with BMD of either segment (data nott shown), patients with a history of aseptic bone necrosis had a significant lowerr mean Z score of the lumbar spine BMD (difference Z score = 0.91; p=0.006;; table 5). Patients who had ever used cyclosporine showed a lower lumbarr spine BMD Z score, which was, however, not significant compared to thosee on azathioprine (data not shown).

125 5

8.8. Bone Disease

Tablee 5. Differences in mean Bone Mineral Density (BMD) of the lumbar spine andd the femoral neck categorized according to several potential determinants.

Physical l inactivity y Lastt RRT

Startt RRT

Gender r

Txx as last RRT

Txx as last RRT

Asepticc bone necrosis s Menarche e

Hyperpara--Thyroidismc c

Yes Yes No o Dialysis s Tx x 1972-1981 1 1982-1992 2 Male e Female e <11 year Tx >11 year Tx <33 year Tx >33 year Tx Yes Yes No o >15thh year <16thh year Yes Yes No o

N N

33 3 103 3 29 9 107 7 74 4 62 2 73 3 63 3 4 4 104 4 11 1 96 6 16 6 120 0 32 2 29 9 63 3 72 2

BMDBMD Lumbar spinespine Z score (SD) (SD) -2.76-2.76 (1.2) -1.900 (1.4) -2.222 (1.4) -2.088 (1.4) -1.96(1.4) ) -2.288 (1.4) -2.377 (1.4) -1.888 (1.3) -2.911 (0.9) -2.055 (1.4) -2.600 (0.8) -2.011 (1.4) -2.911 (1.3) -2.000 (1.3) -1.788 (1.3) -2.066 (1.2) -2.322 (1.3) -1.911 (1.3)

Differ-Differ-ence ence

-0.86'' '

-0.14 4

0.32 2

-0.49̂ ^

-0.86 6

-0.59* *

-0.91'' '

0.28 8

-0.41 1

BMDBMD femoral neckneck Z score (SD)

-2.000 (1.4) -1.677 (1.3) -2.188 (1.2) -1.63-1.63 (1.3) -1.688 (1.4) -1.844 (1.2) -1.666 (1.3) -1.866 (1.3) -1.73-1.73 (1.9) -1.622 (1.3) -1.999 (1.3) -1.588 (1.3) -2.13-2.13 (1.3) -1.711 (1.3) -1.700 (1.2) -2.200 (1.3) -1.744 (1.3) -1.811 (1.3)

Differ-Differ-ence ence

-0.33 3

-0.552 2

0.16 6

0.2 2

-0.11 1

-0.41 1

-0.42 2

0.5 5

0.07 7

"" p<0.05; b p<0.01; c Severe Hyperparathyroidism = more than 12 months serum parathormon levellevel > 3 times normal value; RRT = renal replacement therapy; Tx = transplantation; SD = standardstandard deviation; Z score = Standard deviation corrected for age and gender.

Tablee 6. Pearson's correlations (R) between the lumbar spine and the femoral neckk Bone Mineral Densities (BMD) Z scores and determinants.

Physicall inactivity Dialysiss at time of investigation Startt RRT 1972-1981 vs. 1982-1992 Malee gender Asepticc bone necrosis Agee of menarche Bodyy height Leann body massd

ZZ score BMD lumbar spine:spine: R

-0.27* * * * 0.12 2 -0.2» » -0.21" "

0.15 5 0.16 6

ZZ score BMD femoral neck: R R

-0.12 2 -0.18" "

0.18 8 0.25b b 0.41' '

aa p<0.05;b p<0.01;c p<0.001;d assessed by DEXA; * not significant; RRT = renal replacement therapy.therapy. NOTE: Only characteristics significant at thep<0.2 level are shown.

126 6

8.8. Bone Disease

Discussion. .

ClinicalClinical bone disease. In this follow-up study in young adultswith ESRD since childhood,, we found that one third suffered from clinical manifestations of metabolicc bone disease. More than one out of six patients was disabled as a resultt of this bone disease and more than 60% was severely growth-retarded. Bothh growth-retardation and clinical bone disease were associated with a long durationn of dialysis, where as aseptic bone necrosis was associated with a long periodd of living with a functioning renal graft.

Growth-retardationn is a well-known complication of ESRD in children [15,19, 20].. Since we followed a group of patients who started RRT long before Growth Hormonee therapy became available, our findings on this matter are as expected. However,, the extent and severity of clinically manifest bone disease in these youngg patients surprised us. Although bone disease is a well-recognized problemm in ESRD, data on its clinical consequences in young adult patients with juvenilee ESRD are scarce. In their follow up study of 120 renal transplanted children,, Offner et al. report disabilities caused by bone disease in only 3 patientss [24]. However, data on this subject were gathered only by telephone-interview,, which in our experience can easily lead to underestimation of the disability.. For example, most of our patients who were unable to walk for more thann 100 meters or to climb a stairway, because of severe pain in the hips or knees,, first denied having any disability at all. Clinical manifestations of bone diseasee might therefore be a generally underestimated problem in young adult patientss with juvenile ESRD.

BoneBone mineral density. We found low bone mineral density values in nearly all patients.. More than 50% of them had a BMD Z score in at least one segment of lesss than -2.5, which is consistent with osteoporosis in otherwise healthy subjects.. The lumbar spine was more severely affected than the femoral neck. A loww BMD of the lumbar spine was associated with a low lean body mass, male gender,, physical inactivity and aseptic bone necrosis. A low BMD of the femorall neck was associated with a low lean body mass. The effect of lean body masss on BMD values was more apparent in male than in female patients.

Loww BMDs have been reported in dialysis and transplanted patients with adult onsett of ESRD [1,3-12]. The major structural bone abnormality found in dialysis patientss is a generalized thinning of cortical bone due to increased endocortical resorptionn [7]. After transplantation, an imbalance in bone remodeling characterizedd by osteoclast stimulation and osteoblast suppression, results in a reductionn of trabecular and cortical bone [9-11]. Further, the mineralization is retardedd in transplanted patients. Garlini et al showed that in adults the BMD approximatess normal value only after 10 years of successful transplantation [10]. .

127 7

8.8. Bone Disease

Tablee 7. Multivariate linear regression stepwise analysis to describe the relation betweenn the lumbar spine and femoral neck Bone Mineral Density (BMD) Z-scoress and its determinants: standardized coefficients (P).

ZZ score BMD lumbar spine: /? Z score BMD femoral neck: j3

Physicall inactivity -0.22b *

Malee gender -0.49c *

Leann body massd 0.46c 0.41c

aa p<0.05; h p<0.01; c p<0.001; d assessed by DEXA; * not significant. Excluding lean body massmass and height as determinants revealed dialysis at time of investigation negatively associated withwith BMD femoral neck Z score (f3=-0.18, p=0.03). Z score = Standard deviation corrected for ageage and gender. NOTE: Only characteristics significant at the p<0.05 level are shown.

Tablee 8. Predicted Z score of Bone Mineral Densities values, when adjusted for leann body mass.

FemaleFemale patients Male patients

Regressionn coefficient lumbar spine* 0.07 0.08

ZZ score BMD lumbar spine -1.80 (SD 1.3) -2.38 (SD 1.4)

Predictedd Z score BMD lumbar spine after correction -1.34 (SD 1.2) -1.48 (SD 1.2) forr lean body mass

Regressionn coefficient femoral neck* 0.05 0.1

ZZ score BMD femoral neck -1.86 (SD 1.3) -1.64 (SD 1.3)

Predictedd Z score BMD femoral neck after correction -1.60 (SD 1.2) -0.52 (SD 1.1) forr lean body mass

** regression coefficient between bone mineral density and lean body mass; SD ^standard deviation;deviation; BMD = bone mineral density; Z score = Standard deviation corrected for age and gender. gender.

128 8

8.8. Bone Disease

EtiologicalEtiological factors of low bone mineral density in renal disease. A low bone mineral densityy in ESRD results from a combination of specific and non-specific consequencess of renal disease. Hyperparathyroidism, the use of corticosteroids afterr transplantation, a low lean body mass, inactivity, and male gender are thoughtt to be the major causes of bone loss [2, 5-13, 18, 19-22]. The preferential losss of cortical bone in dialysis patients is thought to be caused by hyperparathyroidism.. Yet, normalization of parathormon plasma-levels does nott induce a reversal of this cortical thinning in dialysis patients [7]. This explainss why dialysis patients with a low turnover bone disease with osteomalaciaa who do not have apparent hyperparathyroidism, can still have severee cortical thinning. In these patients, the total bone loss is even greater thann in patients with osteitis fibrosa, as a result of concurrent cancellous bone losss [7]. Moreover, patients with a high turnover bone disease can have a normall or increased BMD as a result of an increase in yet abnormally structured osteoidd with amorphous calcium depositions [25]. It might also explain why in ourr patient group the total duration of apparent hyperparathyroidism was of noo apparent influence on the BMD.

Thee relation between lean body mass and bone mass is well established [17, 26-28].. The mechanism is believed to be partially hormonal and partially mechanical.. Muscle increases mechanical load on bones, which stimulated bone formation,, especially in weight-bearing bones. The fact that lean body mass correlatess better with bone mass than total body weight, supports the assumptionn of an additional hormonal factor. In our patients, the low lean body masss accounted for an important part of the low bone density values, especially inn males. However, even after correction for lean body mass, the decrease in BMDD of the lumbar spine was still more severe than in patients with adult onset off ESRD [10-12]. This counted for males as well as for females.

Thee fact that males were more affected in our study than females is in accordancee with the results of the study by Cueto-Manzano et al.[ll] . They investigatedd the extent of bone loss in long-term renal transplant patients and foundd male gender even to be the strongest predictive factor for a low bone masss [11]. It is unclear to what extent gender differences in loss of lean body masss played a role in their study. The effect of physical inactivity on bone loss iss well documented in the literature, as is the reversibility by exercise. Sodermann et al found increased BMD values of the total body, the lumbar spine andd all sites of the hip in adolescent female soccer players, compared to non-activee controls [14]. Slemenda et al showed that there was a direct relation betweenn the total hours of weight-bearing activity in children and increase in BMDD in the hip [13]. Spindler et al showed that also in hemodialysis patients increasee in femoral neck BMD was correlated with increase in muscular strengthh of the proximal muscles [26]. Our findings stress the importance of physicall activity in these patients, who tend to be more inactive than average.

129 9

8.8. Bone Disease

Corticosteroidss enhance resorption and reduce formation of mainly trabecular bone,, which is the primary component of the human vertebrae. Boot et al. foundd reduced lumbar spine BMD in 20 young adults who received a renal transplant.. This reduction was directly correlated to the cumulative steroid-dosagee [18]. Contrary to these results, we found no direct relation between corticosteroid-dosagee and BMD of the lumbar spine. One explanation could lie inn the fact that the mean steroid-dosage in their population was 1186 mg/kg, beingg nearly twice as much as used by our patients. On the other hand, some of ourr results indirectly suggest the influence of steroids on bone mass. Patients whoo were transplanted less than 3 years before the DEXA-examination showed aa 32% lower BMD of the lumbar spine, compared to patients who had lived for moree than 3 years with a renal graft, at time of investigation. This difference couldd reflect the more recent use of high steroid-dosage in the first group, implicatingg that this bone loss is reversible. This finding is compatible with data off Cueto-Manzano et al., who also found less reduction of bone mass in long-termm renal transplanted adult patients compared to short-term reports [11]. Otherr circumstantial evidence of the relation between steroids and bone density inn our study could be derived from the relation we found between a history of asepticc bone necrosis and low lumbar spine BMD.

TheThe impact of low Bone Mineral Density. ESRD in older patients goes with an increasedd risk of fractures. A direct relation between low BMD and increased riskk of fractures has been established in older dialysis patients, as well as in olderr renal transplanted patients [3,8,12]. Atsumi et al. found a double risk for vertebrall fractures for each standard deviation decrease in bone mineral density off the lumbar spine [8]. The prevalence of vertebral fractures in this study rangedd from 10% in the group aged 30-39 years to 34.5% in the age group 60-69 years.. In transplanted patients, a triple fracture rate has been reported [12].

Yet,, the interpretation of the low BMDs that we found in our patients is difficult .. Schönau has pointed out the problems of non-invasive bone analysis inn children and adolescents [25]. For instance, hyperparathyroidism in children cann lead to the formation of woven instead of normal structured lamellar osteoid,, mineralized with amorphous deposed calcium rather than as hydroxyy apatite [25]. In patients with these severe structural matrix disorders, thee BMD can be normal or even high. On the other hand, a low BMD in children andd adolescents could be either due to an insufficient mineralization (mineral perr volume) or to a reduced structural density (i.e. a decreased trabecular thickness)) as a result of inactivity or muscular dysfunction [25]. Whether low BMDD values in patients with ESRD since childhood implicate the same increasedd risk of atraumatic and potentially debilitating fractures, as in ESRD patientss with adult onset, is therefore at present speculative. No follow-up studiess on this subject have been published. However, there are some important argumentss in favor of this relationship. First, it is well established that bone masss later in life is determined by the peak bone mass acquired during

130 0

8.8. Bone Disease

adolescence,, and that consequently a low peak bone mass results in a higher riskk of osteoporosis [29, 30, 33]. Obviously, the mean peak bone mass in our patientt group was low. Second, it is also well demonstrated that patients with a loww bone mass as a result of a low lean body mass or inactivity are at risk higherr for fractures [31- 33]. Third, it is well established that malnutrition, inactivity,, and the use of cortocosteroids during childhood and growth-retardation,, all put patients at risk for osteoporosis [33]. A prolonged follow-up off the cohort is warranted to enact the actual impact of our findings on the incidencee of fractures.

LimitationsLimitations of the study. Evaluation of some potential determinants was sometimess cumbersome (e.g. hyperparathyroidism) or impossible (e.g. aluminum-containingg phosphate binders and Vitamin D preparations), due to thee retrospective character of that part of the study. Beside the limitation of bonee mass measurement using DEXA only, we could not perform DEXA-measurementss in 51 of 187 living patients of this cohort. Nevertheless, we believee that this group can be regarded as representative for the whole group, sincee no differences were found between the extent of clinically apparent bone disease,, and parameters with respect to bone mass of participants and non-participantss of the cross-sectional study.

Ass we discussed before, the value of DEXA as a measure of the true state of bonee in our patients is limited. New technologies, such as peripheral quantitativee CT (p-QCT) scanning and quantitative ultrasonography are promisingg in this respect. At present, however, no reliable (pediatric) reference dataa exist, limiting the ability of these tools to identify patients at risk for fracturee [33].

Conclusions. .

Metabolicc bone disease is an important health problem in young adults with juvenilee end-stage renal disease. Although the impact of the low BMDs that we havee found remains speculative, our data suggest that bone mass in these patientss is reduced to a level which puts them at risk for atraumatic and debilitatingg fractures. Further follow-up of these patients is warranted in order too establish the actual long-term implications of our findings, and the need for therapeuticc intervention in these patients.

References: : 1.. Lindberg JS, Moe SM: Osteoporosis in end-state renal disease. Seminars in Nephrology 19:115-

122,1999 9 2.. Parfitt AM: A structural approach to renal bone disease, ƒ Bone Miner Res 13:1213-1220,1998 3.. Alem AM, Sherrard DJ, Gillen DL, et al: Increased risk of hip fracture among patients with

end-stagee renal disease. Kidney Int 58:396-399,2000

131 1

8.8. Bone Disease

4.. Stein MS, Packham DK, Ebeling PR, et al: Prevalence and risk factors for osteopenia in dialysiss patients. Am ƒ Kidney Dis 28:515-522,1996

5.. Abdelhadi M, Nordenstrom J: Bone mineral recovery after parathyroidectomy in patients withh primary and renal hyperparathyroidism, ƒ Clin Endocrinol Metab 83:3845-3851,1998

6.. Amin S, La Valley MP, Simms RW, Felson DT: The role of vitamin D in corticosteroid-inducedd osteoporosis: a meta-analytic approach. Arthritis & Rheumatism 42:1740-1751,1999

7.7. Schober HC, Han ZH, Foldes AJ, et al.: Mineralized Bone Loss at Different Sites in Dialysis Patients:: Implications for Prevention, J Am Soc Nephrol 9:1225-1233,1998.

8.. Atsumi K, Kushida K, Yamazaki K, et al.: Risk factors for vertebral fractures in renal osteodystrophy.. Am J Kidney Dis 33:287-293,1999

9.. Weisinger JR, Carlini RG, Rojas E, Bellorin-Font E: Bone disease after renal transplantation. TransplantationTransplantation Proceedings 31:3033-3034,1999

10.. Garlini RG, Rojas E, Weisinger JR, et al: Bone Disease in Patients With Long-term Renal Transplantationn and Normal Renal Function. Am ƒ Kidney Dis 36,1:160-166, 2000

11.. Cueto-Manzano AM, Konel S, Hutchison AJ, et al: Bone loss in long-term renal transplantation:: histopathology and densitometry analysis. Kidney Int 55:2021-2029,1999

12.. Grotz WH, Mundinger FA, Gugel B, et al.: Bone fracture and osteodensitometry with dual energyy X-ray absorptiometry in kidney transplant recipients. Transplantation 58:912-915, 1994 4

13.. Slemenda CW, Mille r JZ, Hui SL, et al.: Role of physical activity in the development of skeletall mass in children, ƒ Bone Miner Res 6:1227-1233,1991

14.. Soderman K, Bergstrom E, Lorentzon R, Alfredson H: Bone mass and muscle strength in youngg female soccer players. Calaf Tissue Int 67:297-303, 2000

15.. Boot AM, Engels MA, Boerma GJ, et al: Changes in bone mineral density, body composition,, and lipid metabolism during growth hormone (GH) treatment in children withh GH deficiency, ƒ Clin Endocrinol Metab 82:2423-2428,1997

16.. Boot AM, Nauta J, de Jong MC, et al.: Bone mineral density, bone metabolism and body compositionn of children with chronic renal failure, with and without growth hormone treatment.. Clinical Endocrinology 49:665-672,1998

17.. Pietrobelli A, Faith MS, Wamg J, et al.: Association of lean tissue and fat mass with bone minerall content in children and adolescents. Obers Res 10: 10 (l):56-60, 2002

18.. Boot AM, Nauta J, Hokken-Koelega AC, et al: Renal transplantation and osteoporosis. Arch DisDis Child 72:502-506,1995

19.. Hanna JD, Krieg RJ, Jr., Scheinman JI, Chan JC: Effects of uremia on growth in children. SeminarsSeminars in Nephrology 16:230-241,1996

20.. Houang M, Cabrol S, Perin L, et al: Insulin-like growth factor-I (IGF-I), insulin-like growth factorr binding proteins (IGFBP) and insulin-like growth factor type I receptor in children withh various status of chronic renal failure. Growth Hormone & IGF Research 10:332-341, 2000

21.. Tsai KS, Cheng WC, Chen CK, et al: Effect of bone area on spine density in Chinese men andd women in Taiwan. Bone 21:547-551,1997

22.. McHorney CA, Ware JE, Raczek AE: The MOS 36-Item Short-Form Health Survey (SF-36): II .. Psychometric and clinical tests of validity in measuring physical and mental health constructs.. Medical Care 31:247-263,1993

23.. Boot AM, Bouquet J, de Ridder MAJ, et al: Determinants of body composition, measured by duall energy x-ray absorptiometry in Dutch children and adolescents. Am ƒ Clin Nutr 66: 232-238,1997 7

24.. Offner G, Latta K, Hoyer PF, et al: Kidney transplanted children come of age. Kidney Int 55:1509-1517,1999 9

25.. Briner VA, Thiel G, Monier-Fauger MC, et al: Prevention of cancellous bone loss but persistencee of renal bone disease despite normal 1,25 vitamin D levels two years after kidneyy transplantation. Transplantation 10:1393-1400,1995

26.. Macedo JM, Macedo CR, Elkis H, De Oliveira IR: Meta-analysis about efficacy of anti-resorptivee drugs in post-menopausal osteoporosis, ƒ Clin Pharm & TJterap 23:345-352,1998

132 2

8.8. Bone Disease

27.. Spindler A, Paz S, Berman A, et al.: Muscular strength and bone mineral density in haemodialysiss patients. Nephrol Dial Transplant 12:128-132,1997

28.. Aloia JF, Vaswani A, Ma R, Flaster E: To what extent is bone mass determined by fat-free or fatt mass? Am J Clin Nutr 61:1110-1114,1995

29.. Ellis KJ. Body composition of a young, multiethnic, male population. Am ƒ Clin Nutr 66: 1223-31,1997 7

30.. Ribot C, Tremollieres F, Pouilles JM. Late consequences of a low peak bone mass. Acta Paediatr SupplPaediatr Suppl 411:31-5,1995

31.. Hansen MA, Overgaard K, Riis BJ, Christiansen C. Role of peak bone mass and bone loss in postmenopausall osteoporosis: a 12-year study. BMJ 303:961 ,̂ 1991

32.. Pruzansky ME, Turano M, Luckey M, et al.: Low body weight as a risk factor for hip fracturee in both black and white women, ƒ Orthoped Res 7:192-7,1989

33.. Kelsey JL. Risk factors for osteoporosis and associated fractures. Public Health Rep 104:14-20, 1989 9

34.. Leonard MB, Zemel BS: Current Concepts in Pediatric Bone Disease. Pediatr Clin North Am 49(1)143-73,, 2002

133 3