Pediatr Nephrol (2006) 21: 457–462
DOI 10.1007/s00467-006-0025-6
EDITORIA L COMMENTARY
Frank Rauch
Watching bone cells at work: what we can see from bone biopsies
Received: 23 October 2005 / Revised: 5 December 2005 / Accepted: 9 December 2005 / Published online: 7 March 2006
# IPNA 2006
Abstract Histomorphometric analysis of iliac bone samples is a key tool for studying bone metabolism and, to a
lesser extent, bone mass and structure. Two types of bone
metabolic activity can be distinguished: modeling and
remodeling. Both processes are performed by the same
effector cells, osteoblasts and osteoclasts, but differ in the
way these cells are arranged. The main effect of remodeling is to renew bone, whereas modeling can lead to
rapid changes in bone shape, size and mass. Standard
histomorphometric analysis focuses on trabecular bone
and therefore mainly provides information on remodeling. Remodeling activity changes markedly with age
during development. This must be taken into account when
histomorphometry is used in the pediatric setting. Remodeling disorders encountered in the context of pediatric renal
bone disease include mineralization defects, as well as
abnormally high remodeling activity due to secondary
hyperparathyroidism or suppressed remodeling activity as
a consequence of over-treatment. Children and adolescents
with severe bone fragility should have a bone biopsy unless
the diagnosis is obvious from noninvasive examinations.
Histomorphometric analysis of transiliac bone biopsy
samples is especially valuable in clinical studies, as this
method provides safety and efficacy data that cannot be
obtained in any other way.
Keywords Bone histomorphometry . Children .
Modeling . Remodeling . Renal bone disease
Introduction
Ever since Harold Frost pioneered bone histomorphometry
in the early 1960s, this technique has been a key tool for
F. Rauch (*)
Genetics Unit, Shriners Hospital for Children,
1529 Cedar Avenue,
Montréal, Québec, H3G 1A6, Canada
e-mail: [email protected]
Tel.: +1-514-8425964
Fax: +1-514-8425581
studying bone metabolism and, to a lesser extent, bone
mass and structure. Histomorphometry of undecalcified
bone biopsy samples is a method used to directly obtain
quantitative information on bone tissue. When tetracycline
labeling is performed prior to biopsy, bone cell function
can be studied in vivo. Importantly for pediatric use, bone
histomorphometric results are not directly influenced by
the growth process. In contrast to some currently popular
indirect methods of bone analysis, histomorphometry
yields results with a known meaning.
Despite these advantages, bone histomorphometry is
underused in pediatrics. This may be partly due to the fact
that histomorphometry requires an invasive procedure to
obtain a bone sample, is labor-intensive, and needs special
equipment and expertise. Other reasons may include
overestimation of the utility of noninvasive bone diagnostics and lack of information about what bone histomorphometry does. The present contribution aims at making
histomorphometric reports more accessible to the general
reader by providing a brief introduction to the field. More
detailed information on pediatric histomorphometry is
available elsewhere [1].
Basic concepts of bone metabolism
Two types of bone metabolic activity can be distinguished:
modeling and remodeling. Both processes are performed
by the same effector cells, namely osteoblasts and
osteoclasts, but differ in the way that these cells are
arranged (Fig. 1). Remodeling consists of successive cycles
of bone resorption and formation on the same bone surface.
A group of osteoclasts digs a trench across the surface of a
trabecula (or a tunnel through cortical bone) which is then
refilled by a team of osteoblasts. The entire group of cells
involved in this process is named a remodeling unit or a
basic multicellular unit. The fact that osteoblast activity is
linked to previous osteoclast action has been named
“coupling” [1]. The difference between the amounts of
bone that are removed and added in one remodeling cycle
is called the “remodeling balance”. The remodeling
458
balance is typically close to zero so that there is no or little
net effect on the amount of bone. The main outcome of
remodeling is to renew bone tissue, which could be
important for preventing tissue damage accumulation [2].
Remodeling defects are thought to cause some of the most
frequently occurring bone disorders in adults, such as
postmenopausal osteoporosis. Consequently, this process
has been studied in great detail.
Bone metabolism in children and adolescents is much
more complex than in adults. Bones do not just undergo
remodeling, but also grow in length through endochondral
ossification and change their width and shape through
modeling [3, 4]. The big difference between bone modeling
and remodeling is that osteoclasts and osteoblasts are active
on different surfaces of a piece of bone in modeling
(Fig. 1). For example, the shaft of a long bone grows in
width when osteoblasts on the outer (periosteal) surface of
the bone deposit bone matrix and mineralize it [4]. At the
same time, osteoclasts may resorb bone from the inner
(endocortical) surface of the cortex, thus increasing the size
of the marrow cavity. In modeling, bone cells can be active
without interruption, whereas the actions of osteoclasts and
osteoblasts tend to cancel each other out in remodeling. For
this reason, modeling is a much more efficient process than
remodeling when it comes to achieving changes in bone
shape or mass.
Modeling is usually associated with an increase in bone
mass, as osteoclasts normally remove less bone tissue than
the osteoblasts deposit [3]. An example is the growth in
width of a long-bone shaft, where bone formation on the
periosteal surface usually outstrips bone resorption on the
Fig. 1 a Lateral view of a remodeling site in trabecular bone.
Osteoclasts in the front dig a trench across the bone surface, which is
then refilled by a team of osteoblasts. b Modeling site. Osteoblasts
and osteoclasts are located on opposite sides of a bone cortex. As
indicated by the arrows, osteoblasts add bone to the upper surface,
whereas osteoclasts remove bone from the lower surface. Thus, the
piece of bone in this example is moving upwards. The thickness of
the cortex will increase if osteoblasts add more bone than the
osteoclasts remove
endocortical surface, thus increasing cortical thickness
(Fig. 1). Modeling and remodeling are not just abstract
concepts, but can be observed directly in tetracyclinelabeled iliac bone samples (Fig. 2).
Published information on modeling in children and
adolescents is limited to a few small studies [5, 6]. These
preliminary results suggest that the bone formation activity
dedicated to modeling is often very high and independent
of concomitant remodeling activity. However, histomorphometric studies are typically limited to trabecular bone, a
location where most if not all of the bone metabolic activity
is thought to represent remodeling. Cortical surfaces,
where modeling activity is very high during development,
are usually not analyzed. Due to this paucity of data on the
modeling process, the following discussion focuses on
remodeling.
Iliac bone biopsy
In principle, histomorphometric analysis could be performed in any bone. In clinical pediatrics, however, the
utility of samples from nonstandard sites is limited at
Fig. 2 a Trabecular remodeling. Multiple short tetracycline doublelabels are present. b Cortical modeling. The picture shows part of a
bone cortex. The patient had received two courses of tetracycline
double-labeling within two years. The label from the older course is
seen on the right half of the picture, the newer (less intense) on the
left half. The bone surface at the time of biopsy is also indicated.
The bone between the older labels and the surface at the time of
biopsy (average distance 280 μm) was produced in the two years
between the first labeling course and the time of biopsy,
corresponding to a daily addition of 0.4 μm of bone. This is close
to the mineral apposition rate and therefore suggests that bone
formation has been going on uninterrupted for two years. In
comparison, bone formation in a single remodeling cycle lasts for
about 3–4 months and adds only about 40–50 μm of bone
459
present, because detailed reference data are only available
for the ilium [7]. Bone specimens for histomorphometric
evaluation should be horizontal, full-thickness biopsies of
the ilium from a site 2 cm posterior from the anterior
superior iliac spine. This should yield a sample containing
two cortices that are separated by a trabecular compartment
(Fig. 3). Vertical samples (from the iliac crest downwards,
also called the “Jamshidi approach”) are of questionable
utility because of the presence of the growth plate at the top
of the iliac crest. Turnover is very high and cortical
thickness is very low in bone tissue below the growth plate
and results are therefore not representative. Thus, the oftenused term “iliac crest biopsy” is a misnomer, as the iliac
crest should actually be avoided during the biopsy
procedure. A more accurate term would be “transiliac
biopsy”.
Transiliac biopsy specimens thus are useful for evaluating bone structure and metabolism. However, as growth
plate and primary spongiosa are not represented in the
sample, it does not provide direct information on the
process of endochondral ossification.
Histomorphometric parameters
Histomorphometrists use standardized terminology and
clear definitions that were established by a working group
of the American Society for Bone and Mineral Research
[8]. According to these definitions, “bone” is bone matrix,
whether it is mineralized or not. Unmineralized bone
matrix is called osteoid. The term “tissue” refers to both
bone and associated soft tissue, such as bone marrow.
Histomorphometric measurements are performed in twodimensional sections. Nevertheless, in order to stress the
three-dimensional nature of bone, the terminology committee favored a three-dimensional nomenclature for
reporting histomorphometric results [8]. Thus, what
Fig. 3 Section of an entire iliac biopsy specimen from a 10-year-old
boy without metabolic bone disease. In this section, core width is
7.5 mm, cortical width (the average length of the arrows indicated in
the two cortical compartments) is 1190 μm, and bone volume per
tissue volume in the trabecular compartment is 22.5%. Osteoid and
cellular structures cannot be identified at this magnification
appears as a line in a microscopic bone section is called a
surface, whereas what is visible as an area under the
microscope is referred to as a volume. This is done simply
by convention, and should not be mistaken as actual threedimensional measurements.
Histomorphometric parameters can be classified into
four categories (Table 1): structural parameters, static bone
formation parameters, dynamic formation parameters, and
bone resorption parameters. Structural parameters describe
the size and the amount of bone (Fig. 3). The outer size of a
transiliac biopsy specimen is called the core width, a
measure which reflects the thickness of the ilium. Cortical
width is the average width of the two both cortices. Bone
volume per tissue volume of trabecular bone represents the
proportion of the marrow cavity which is occupied by
bone. In trabecular bone, bone volume per tissue volume
can be schematically separated into two components:
trabecular thickness and trabecular number.
The group of static formation parameters comprises the
surface extent, thickness and relative amount of osteoid, as
well as the surface extent of osteoblasts (called the osteoid
surface per bone surface, osteoid thickness, osteoid volume
per bone volume and osteoblast surface per bone surface,
respectively; Table 1). Wall thickness reflects the amount
of bone that is created by an osteoblast team during a
remodeling event (Fig. 1a). Wall thickness should not be
confused with cortical thickness, which it does not have
any relationship to.
Dynamic bone formation parameters yield information
on in vivo bone cell function and can only be measured
when patients have received two courses of tetracycline
label prior to biopsy (Table 1). The two basic parameters
are the surface extent of mineralization activity (mineralizing surface per bone surface) and the speed of
mineralization in a direction perpendicular to the bone
surface (mineral apposition rate). The mineralization lag
time and bone formation rate per bone surface are derived
mathematically from these primary measures. It should be
noted that a high bone formation rate does not necessarily
lead to a net gain of bone. If the remodeling balance is zero,
the amount of bone will remain unchanged even if bone
formation rate is very high. The combination of a negative
remodeling balance and high bone formation rate will even
lead to rapid bone loss. Thus, the bone formation rate per
bone surface in trabecular bone indicates the activity of
bone turnover rather than the bone gain [2].
Bone resorption can only be quantified with static
parameters, which makes evaluating bone resorption the
least informative aspect of histomorphometric analysis. It
is possible to quantify the extent of bone surface that is
covered by osteoclasts or looks eroded (osteoclast surface
per bone surface and eroded surface per bone surface,
respectively), but it is not possible to tell from these
measures how much bone resorption is actually going on.
This may be an issue in the evaluation of renal bone
disease. In chronic renal failure, osteoclasts resorb bone
more slowly than normal, so that the extent of osteoclast
and eroded surfaces overestimates the rate of bone
resorption [9].
460
Table 1 The most commonly used histomorphometric parameters
Parameter
Abbreviation
Significance
Structural parameters
Core width (mm)
Cortical width (μm)
Bone volume / tissue volume (%)
C.Wi
Ct.Wi
BV/TV
Trabecular thickness (μm)
Trabecular number (/mm)
Tb.Th
Tb.N
Overall size of the biopsy specimen
Distance between periosteal and endocortical surfaces
Space taken up by mineralized and unmineralized bone relative to the total
size of a bone compartment
Self-explanatory
Number of trabeculae that a line through a trabecular compartment would hit
per millimeter of its length
Static formation parameters
Osteoid thickness (μm)
Osteoid surface / bone surface (%)
Osteoid volume / bone volume (%)
Osteoblast surface / bone surface (%)
Wall thickness (μm)
Dynamic formation parameters
Mineralizing surface / bone surface (%)
Mineral apposition rate (μm/d)
Mineralization lag time (d)
Bone formation rate / bone surface
(μm3×μm−2*y−1)
Static resorption parameters
Eroded surface / bone surface (%)
Osteoclast surface / bone surface (%)
O.Th
OS/BS
OV/BV
Ob.S/BS
W.Th
Distance between the surface of the osteoid seam and mineralized bone
Percentage of bone surface covered by osteoid
Percentage of bone volume consisting of unmineralized osteoid
Percentage of bone surface covered by osteoblasts
Mean thickness of bone tissue that has been deposited at a remodeling site
MS/BS
MAR
Mlt
BFR/BS
Percentage of bone surface showing mineralizing activity
Distance between two tetracycline labels divided by the length
of the labeling interval
Time interval between the deposition and mineralization of matrix
Amount of bone formed per year on a given bone surface
ES/BS
Oc.S/BS
Percentage of bone surface presenting a scalloped appearance
Percentage of bone surface covered by osteoclasts
Bone metabolism in children and adolescents
After the age of puberty, remodeling activity declines into
the much lower adult range [5].
The volume of trabecular iliac bone increases markedly
between 2 and 20 years of age [7]. This increase is entirely
explained by trabecular thickening, whereas there is no
change in trabecular number. Iliac trabeculae probably
become thicker during development because bone remodels with a positive balance [5]. It has been estimated that
during a remodeling cycle osteoblasts lay down about 5%
more bone than osteoclasts resorb. In other words, 95% of
the bone formation activity is required just to replace the
bone that was previously removed by the osteoclasts. Since
the difference between resorption and formation is very
small, a high remodeling activity is necessary to make
trabeculae noticeably thicker. Remodeling activity is
indeed elevated in young children, decreases until the age
of eight or nine years, and increases again during puberty.
Histomorphometry in pediatric renal bone disease
Metabolic bone problems arising in patients with renal
failure commonly revolve around two issues, bone
mineralization and bone remodeling activity [10]. Bone
mineralization can be impaired, and remodeling activity
can be too high or too low. Typical histomorphometric
findings in these and some other conditions are shown in
Table 2.
Let us first consider mineralization. Discussions of
mineralization disorders have become complicated in
recent years by the widespread misuse of the term
mineralization in the densitometric literature. Decreased
Table 2 Typical findings in key histomorphometric parameters in pediatric bone diseases. For explanations of abbreviations, see Table 1
Condition
C.Wi
Ct.Wi
BV/TV
O.Th
OS/BS
MAR
MS/BS
BFR/BS
Mlt
Severe osteomalacia (any etiology)
Hyperparathyroidism
Oversuppressed PTH secretion (“adynamic bone”)
Osteogenesis imperfecta
Idiopathic juvenile osteoporosis
Changes induced by bisphosphonate treatment
↔
↔
↔
↓
↔
↔
↔
↔
↔
↓
↔
↑↑
↔
↔
↔
↓↓
↓↓
↑
↑↑↑
↑
↔
↔
↓
↓
↑↑
↑
↓
↑
↓
↓
↓
↑
↔
↓
↓
↔
↓↓
↑
↓↓
↑
↓
↓↓
↓↓
↑↑
↓↓
↑
↓
↓↓↓
↑↑↑
↔
↔
↔
↔
↑↑
461
bone mineralization is often said to be present when low
bone mineral density is found. However, bone mineralization can only occur where bone matrix has been laid down
before. Most cases of low bone density are not related to a
problem with mineralization, but are due to insufficient
production of bone matrix or increased matrix removal
[11].
The physiological process of mineralization represents
the incorporation of minerals (calcium, phosphorus, and
others) into an organic matrix [11]. In the growing
skeleton, mineralization occurs in two different types of
tissue: growth plate cartilage and bone matrix. Rickets
refers to the changes caused by deficient mineralization at
the growth plate. Osteomalacia is the impaired mineralization of bone matrix. Rickets and osteomalacia usually
occur together as long as the growth plates are open; only
osteomalacia can occur after the growth plates have fused.
When the mineralization of bone matrix is disturbed,
unmineralized bone matrix accumulates, because osteoblasts continue to secrete osteoid for some time. In
histomorphometric terms, osteomalacia is defined as the
simultaneous occurrence of increased osteoid thickness
and increased mineralization lag time [12]. The criteria
used for abnormality in these parameters depend on the
source of reference data. Using the methodology established in my laboratory, an osteoid thickness above 9 μm
combined with a mineralization lag time longer than 25
days would be diagnostic of osteomalacia [13]. As osteoid
thickness and mineralization lag time do not vary between
two and twenty years of age, these criteria can be applied
throughout this age range [7].
It must be stressed that the diagnosis of osteomalacia can
only be made when both osteoid thickness and mineralization lag time are abnormally high. An isolated increase in
osteoid thickness can be caused by an increased rate of
osteoid production. An example of this is hyperparathyroidism [2]. An isolated increase in mineralization lag time
can be due to slow bone turnover, as occurs after
bisphosphonate therapy [14].
As mentioned earlier, renal bone disease is not only
associated with mineralization disorders, but also with
disturbances in bone remodeling activity. Bone remodeling
activity in patients with renal bone disease largely depends
on parathyroid hormone levels [2]. Hyperparathyroidism
increases remodeling, which means that a larger than
normal number of remodeling units are active on the bone
surface at any time. This leads to an increase in all
measurable components of the bone remodeling process.
The best parameter for measuring the cellular response to
hormonal and other signals is bone formation rate per bone
surface [2]. Bone formation rate per bone surface
sensitively reflects the effect of hyperparathyroidism on
bone turnover, as well as the effects of over-treatment [2].
In the latter case, a low bone formation rate per bone
surface indicates suppressed remodeling activity, a situation which has been dubbed “adynamic bone disease” [10].
In the field of renal bone disease, histomorphometric
findings have traditionally been used to classify patients
into the categories of hyperparathyroidism, osteomalacia,
mixed bone disease and adynamic bone disease [10]. As
noted by Parfitt, this categorization leads to a loss of
important information and makes it difficult to understand
the pathophysiology of the bone disorder in renal failure
patients [2]. Parfitt’s bleak view of the current use of bone
histomorphometry by nephrologists is compounded by his
observation that these diagnostic categories are based on
questionable parameters and arbitrary cut-off values.
Regarding pediatric histomorphometry, nephrologic studies often do not take into account the fact that bone
remodeling activity is dependent on age [7]. To improve
the utility of bone histomorphometry in renal bone disease,
Parfitt proposes a new system that avoids categorization
and calls for the use of better reference data [2, 15]. It
remains to be seen whether these proposals will improve
the use of histomorphometry in nephrologic studies. In any
case, some of this criticism was echoed in a recent report by
a group of experts in renal osteodystrophy [16].
The use of bone biopsies in pediatrics
Performing an iliac bone biopsy can provide diagnostic
clues about unclear bone fragility disorders. For example,
some forms of osteogenesis imperfecta can be diagnosed
on the basis of a characteristic histologic pattern [17].
Polyostotic fibrous dysplasia is sometimes difficult to
distinguish from osteogenesis imperfecta on clinical
grounds, but the diagnosis is usually quite obvious from
a bone histology viewpoint. This has therapeutic implications, as children with osteogenesis imperfecta usually
respond much better to bisphosphonate treatment than
patients with fibrous dysplasia [17, 18]. Thus, children
with multiple long-bone fractures or vertebral body
compressions without adequate trauma should have a
bone biopsy unless the diagnosis is obvious from
noninvasive examinations. Another indication for bone
biopsy is progressive bone deformity, which may sometimes arise without clear history of fractures.
A bone biopsy sample permits trabecular and cortical
bone structure, the mineralization process, bone lamellation (woven bone vs. lamellar bone, the appearance of
lamellae), the presence of calcified cartilage, the activity of
bone metabolism, and the appearance of bone cells to be
evaluated. All of this information is important in the
assessment of skeletal disease processes, but none of it is
reflected in “bone density”, whatever technique is used to
measure it. These considerations are particularly relevant in
the context of renal bone disease, which is a frequent but
understudied problem after juvenile renal failure [19]. The
skeletons of patients with kidney failure are affected by
some of the same risk factors that are also associated with
cardiovascular calcifications [20]. These include poorly
controlled secondary hyperparathyroidism, hyperphosphatemia, elevated calcium-phosphorus product, and possibly
the use of certain vitamin D analogs. If bone biopsies help
to improve the management of renal bone disease, the
benefit may therefore extend to extraskeletal organ
involvement as well.
462
When the only aim is to assess an individual patient, it is
not absolutely necessary to analyze the sample with quantitative histomorphometry. A qualitative evaluation of the
histological appearance may be sufficient in such cases.
However, a quantitative analysis is necessary in clinical
research settings, when numbers are needed to describe the
average effect of a disease or a treatment in a group of
patients. Histomorphometric evaluation of bone biopsy
samples should be a standard feature of studies that evaluate
experimental drugs to treat bone disorders in children and
adolescents. Current noninvasive methods of studying the
amount, distribution and metabolism of bone are fraught with
technical limitations and uncertainties regarding the interpretation of results. The availability of histomorphometric data
allows treatment effects to be judged in a rational way.
The most important argument for performing bone
biopsies in pediatric studies probably concerns patient
safety. This is especially true when children and adolescents
are treated with long-acting drugs, such as bisphosphonates.
Analysis of bone samples provides safety measures that
cannot be obtained in any other way. For example, bone
histologic studies have demonstrated that bisphosphonate
treatment in children can lead to accumulation of calcified
cartilage material in bone tissue, a disquieting finding that
calls for caution in the use of these drugs in growing patients
with minor skeletal symptoms [14, 21]. Thus, it is crucial to
include bone biopsies in study protocols in order to
document the efficacy of therapy as well as its safety.
Conclusions
Standard histomorphometric analysis of transiliac bone
biopsies mainly provides information on trabecular
remodeling. Assessment of cortical modeling processes is
feasible but has rarely been used until now. When
histomorphometric studies are performed in children and
adolescents, it is important to take the age-dependency of
many histomorphometric parameters into account. Children with multiple long-bone fractures or vertebral body
compressions that are not explained by adequate trauma
should have a bone biopsy unless the diagnosis is obvious
from noninvasive examinations. When new treatments of
bone disorders are studied, analysis of transiliac bone
biopsy samples provides safety and efficacy data that
cannot be obtained in any other way.
Acknowledgements Thanks go to Mark Lepik for preparing the
figures. The author is a Chercheur-Boursier Clinician of the Fonds de
la Recherche en Santé du Québec. This work was supported by the
Shriners of North America.
References
1. Rauch F (2003) Bone histomorphometry. In: Glorieux FH,
Pettifor J, Jueppner H (eds) Pediatric bone. Academic, San
Diego, CA, pp 359–374
2. Parfitt AM (2003) Renal bone disease: a new conceptual
framework for the interpretation of bone histomorphometry.
Curr Opin Nephrol Hypertens 12:387–403
3. Frost HM (1990) Skeletal structural adaptations to mechanical
usage (SATMU): 1. Redefining Wolff’s law: the bone modeling
problem. Anat Rec 226:403–413
4. Rauch F (2005) Bone growth in length and width: the yin and
yang of bone stability. J Musculoskelet Neuronal Interact
5:194–201
5. Parfitt AM, Travers R, Rauch F, Glorieux FH (2000) Structural
and cellular changes during bone growth in healthy children.
Bone 27:487–494
6. Rauch F, Travers R, Norman ME, Taylor A, Parfitt AM,
Glorieux FH (2002) The bone formation defect in idiopathic
juvenile osteoporosis is surface-specific. Bone 31:85–89
7. Glorieux FH, Travers R, Taylor A, Bowen JR, Rauch F,
Norman M, Parfitt AM (2000) Normative data for iliac bone
histomorphometry in growing children. Bone 26:103–109
8. Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H,
Meunier PJ, Ott SM, Recker RR (1987) Bone histomorphometry: standardization of nomenclature, symbols, and units.
Report of the ASBMR Histomorphometry Nomenclature
Committee. J Bone Miner Res 2:595–610
9. Jaworski ZF, Lok E, Wellington JL (1975) Impaired osteoclastic function and linear bone erosion rate in secondary hyperparathyroidism associated with chronic renal failure. Clin
Orthop Relat Res 107:298–310
10. Kuizon BD, Salusky IB (2003) Renal osteodystrophy: Pathogenesis, diagnosis, and treatment. In: Glorieux FH, Pettifor J,
Jueppner H (eds) Pediatric bone. Academic, San Diego, CA, pp
679–701
11. Rauch F, Schoenau E (2002) Skeletal development in premature infants: a review of bone physiology beyond nutritional
aspects. Arch Dis Child Fetal Neonatal Ed 86:F82–F85
12. Parfitt AM (1997) Vitamin D and the pathogenesis of rickets
and osteomalacia. In: Feldman D, Glorieux FH, Pike JW (eds)
Vitamin D. Academic, San Diego, CA, pp 645–662
13. Terpstra L, Rauch F, Plotkin H, Travers R, Glorieux FH (2002)
Bone mineralization in polyostotic fibrous dysplasia: histomorphometric analysis. J Bone Miner Res 17:1949–1953
14. Rauch F, Travers R, Plotkin H, Glorieux FH (2002) The effects
of intravenous pamidronate on the bone tissue of children and
adolescents with osteogenesis imperfecta. J Clin Invest
110:1293–1299
15. Parfitt AM, Qiu S, Rao DS (2004) The mineralization index—
A new approach to the histomorphometric appraisal of
osteomalacia. Bone 35:320–325
16. Martin KJ, Olgaard K, Coburn JW, Coen GM, Fukagawa M,
Langman C, Malluche HH, McCarthy JT, Massry SG, Mehls O,
Salusky IB, Silver JM, Smogorzewski MT, Slatopolsky EM,
McCann L (2004) Diagnosis, assessment, and treatment of
bone turnover abnormalities in renal osteodystrophy. Am J
Kidney Dis 43:558–565
17. Rauch F, Glorieux FH (2004) Osteogenesis imperfecta. Lancet
363:1377–1385
18. Plotkin H, Rauch F, Zeitlin L, Munns C, Travers R, Glorieux
FH (2003) Effect of pamidronate treatment in children with
polyostotic fibrous dysplasia of bone. J Clin Endocrinol Metab
88:4569–4575
19. Groothoff JW, Offringa M, Van Eck-Smit BLF, Gruppen MP,
Van De Kar NJ, Wolff ED, Lilien MR, Davin JC, Heymans
HSA, Dekker FW (2003) Severe bone disease and low bone
mineral density after juvenile renal failure. Kidney Int 63:
266–275
20. Sheth RD, Perez MD, Goldstein SL (2003) Cardiovascular
calcifications in pediatric patients receiving maintenance dialysis. Pediatr Nephrol 18:810–813
21. Whyte MP, Wenkert D, Clements KL, McAlister WH, Mumm
S (2003) Bisphosphonate-induced osteopetrosis. N Engl J Med
349:457–463