Osteoporos Int (2016) 27:3427–3437

DOI 10.1007/s00198-016-3723-3

REVIEW

Osteogenesis imperfecta in children and adolescents—new

developments in diagnosis and treatment
P. Trejo 1,2 & F. Rauch 1,2

Received: 7 June 2016 / Accepted: 25 July 2016 / Published online: 5 August 2016
# International Osteoporosis Foundation and National Osteoporosis Foundation 2016

Abstract Osteogenesis imperfecta (OI) is the most prevalent Introduction

heritable bone fragility disorder in children. It has been known
for three decades that the majority of individuals with OI have Osteogenesis imperfecta (OI) is a heritable disorder that is
mutations in COL1A1 or COL1A2, the two genes coding for mainly characterized by bone fragility, even though many
collagen type I alpha chains, but in the past 10 years defects in other organ systems can be involved. The typical clinical de-
at least 17 other genes have been linked to OI. Almost all scription of OI also includes blue or gray discoloration of the
individuals with a typical OI phenotype have a mutation in sclera and tooth abnormalities called dentinogenesis
one of the currently known genes. Regarding medical treat- imperfecta [1], but these extraskeletal features of OI are fre-
ment, intravenous bisphosphonate therapy is the most widely quently absent.
used medical approach. This has a marked effect on vertebra Even though OI has been recognized as a separate disorder
in growing children and can lead to vertebral reshaping after for more than a century, scientific and research interest in this
compression fractures, but there is little effect of bisphospho- disorder has increased markedly in the past decade. Several
nate therapy on the development of scoliosis. Bisphosphonate recent reviews have provided excellent overviews of the path-
treatment decreases long-bone fracture rates, but such frac- ophysiology and genetics of OI and related disorders [1–3].
tures are still frequent. Newer medications with anti- General introductions to OI are also available [4–7]. The pres-
resorptive and bone anabolic action are being investigated in ent review highlights recent developments in diagnosis and
an attempt to improve on the efficacy of bisphosphonates but treatment of OI that are of relevance to the clinical metabolic
the safety and efficacy of these new approaches in children bone disease specialist who is following young patients with
with OI is not yet established. OI.

Keywords Bisphosphonate . Bone fragility . Collagen . Classification

Fractures . Osteoblast . Osteogenesis imperfecta
The severity of bone fragility in OI varies widely. This is
reflected in the clinical Sillence classification from 1979 that
separated the severity spectrum of OI into four categories (OI
types I to IV) [8]. These OI types are still found in the 2015
This study was supported by the Shriners of North America and the Fonds
de recherche du Québec–Santé Nosology and Classification of Genetic Skeletal Disorders [9].
OI type I represents the mildest end of the spectrum with
* F. Rauch straight limbs (Bnon-deforming OI^), OI type II usually leads
[email protected] to death shortly after birth (Bperinatally lethal OI^), OI type III
is the most severe form of the disease in individuals surviving
1
Shriners Hospital for Children, 1003 Decarie, Montreal, Quebec, the neonatal period (Bprogressively deforming OI^), and OI
Canada H4A 0A9 type IV is characterized by a disease severity intermediate
2
McGill University, Montreal, Quebec, Canada between OI types I and III (Bmoderate severity OI^) [9].
3428 Osteoporos Int (2016) 27:3427–3437

Subsequent to the original Sillence classification, addition- forms of OI, but two genes (IFITM5, P4HB) are associated
al clinical OI types were delineated, based on distinctive phe- with dominant OI. Defects in one gene (PLS3) lead to X-
notypic characteristics (OI type V, hyperplastic callus forma- linked bone fragility.
tion; OI type VI, accumulation of unmineralized osteoid on These new genetic discoveries raise the issue of what genes
bone histology; OI type VII, rhizomelia) [10]. However, the should be deemed BOI-related genes^ and, indeed, what is the
clinical diagnosis of OI type VI is based on the evaluation of definition of OI. In contrast to some other congenital disor-
bone tissue [11], which often is not available for analysis, and ders, there are no agreed clinical criteria that define OI.
OI type VII has only been observed in an isolated community Fractures are the typical hallmark of OI but may result from
in Canada [12]. Consequently, five clinical types of OI are a variety of other genetic and non-genetic conditions as well
generally recognizable with routine diagnostic methods (his- [16, 17]. Most authors seem to agree that bone fragility lead-
tory, clinical examination, radiographs). ing to bone deformities should be diagnosed as OI unless there
More recently, genotypic OI types (designated as OI type are obvious signs of another genetic disorder [1–3].
VIII and OI types with higher numbers) have been listed in the In contrast, the Bcorrect^ classification of bone fragility
widely used Online Mendelian Inheritance of Man (OMIM; without bone deformities is not clear, unless there are Btypical
http://www.ncbi.nlm.nih.gov/omim/) database, where each extraskeletal features^ of OI, such as blue/gray sclera or
discovery of a new OI-associated gene has given rise to a dentinogenesis imperfecta, leading to a diagnosis of OI type
new OI type. However, admixing gene-based OI types to an I. However, typical extraskeletal features of OI can be absent
initially phenotypic classification is controversial [13]. The even when bone fragility is caused by Btypical^ OI mutations
clinically defined OI types are a convenient short-hand system in COL1A1 or COL1A2 [18]. When the constellation of bone
for describing a spectrum of complex phenotypes with a sin- fragility, white sclera, and white teeth is caused by mutations
gle number; this facilitates communication. In contrast, in genes other than COL1A1 or COL1A2, such as dominant
converting the name of an OI-related gene into an OI type mutations in WNT1, LRP5, or PLS3, terms such as Bidiopathic
number makes communication more difficult—it is simpler juvenile osteoporosis,^ Bjuvenile osteoporosis,^ or Bearly-on-
to just name the involved gene. Here we therefore use the set osteoporosis^ are used in the literature [2, 3], even though
phenotype-based classification proposed by the 2015 PLS3 mutations have also been classified among the causes of
Nosology and Classification of Genetic Skeletal Disorders OI type I [19] and are listed in the OI variant database. This is
[9] (Table 1). an area that requires further discussion among stakeholders.

Genes associated with OI

Genotype-phenotype correlations
It has been known for more than three decades that OI pheno-
types are most often caused by dominant alterations in one of More than 1500 different mutations have been associated with
the genes coding for collagen type I alpha chains COL1A1 OI and are listed in the OI variant database (http://www.le.ac.
(coding for the collagen type I alpha 1 chain, 1 (I)) or uk/ge/collagen/). Close to 90 % of these mutations affect
COL1A2 (coding for 2 (I)) [1]. The molecular diagnosis of COL1A1 or COL1A2. Establishing genotype-phenotype cor-
OI was initially performed by examining collagen type I pro- relations is notoriously difficult, but a few principles have
tein from skin fibroblasts [14] and later by Sanger sequencing emerged.
of COL1A1 and COL1A2. In the most detailed Sanger se- The most consistent genotype-phenotype correlation is that
quencing study to date, pathogenic COL1A1 or COL1A2 se- COL1A1 mutations leading to 1 (I) haploinsufficiency give
quence alterations were detected in 87 % of 142 children with rise to OI type I, with mild bone fragility, blue/gray sclera, and
a Btypical OI^ phenotype [15]. However, the diagnostic yield normal-looking teeth. The usual causes of 1 (I)
varied with the phenotypic group. Pathogenic variants were haploinsufficiency are COL1A1 stop or frameshift mutations
found in 94 % of individuals with OI type I, in 88 % of [20], but splice site mutations and deletions of the entire
patients with OI type III but in only 63 % of children with COL1A1 gene can have the same clinical outcome [21, 22].
OI type IV. Even though OI type I is also called Bmild OI,^ the bone
Starting in 2006, defects in at least 17 genes other than fragility is still substantial. Compared to the general popula-
COL1A1 and COL1A2 have been linked to OI phenotypes tion, children with COL1A1 haploinsufficiency mutations
[1] (Table 1). These genes are all expressed in osteoblasts, have an almost 100-fold increased rate of femur and tibia
and the majority of them are directly involved in collagen type fractures [23]. The majority of individuals with 1 (I)
I metabolism even though some of these genes seem to play a haploinsufficiency mutations have vertebral compression
role in other aspects of osteoblast function [1] (Table 1). fractures during childhood or adolescence and about a third
Defects in the newer OI-related genes usually lead to recessive develop scoliosis.
Osteoporos Int (2016) 27:3427–3437 3429

Table 1 Genes that have been associated with OI

Gene Protein Protein full name Clinical OI type Mutations (N) Comment

COL1A1 COL1A1 Collagen type I alpha 1 chain I, II, III, IV 848

COL1A2 COL1A2 Collagen type I alpha 2 chain I, II, III, IV 510
Collagen type I processing defects
CRTAP CRTAP Cartilage-associated protein III, IV 24
P3H1 P3H1 Prolyl-3-hydroxlase 1 III 49
PPIB CypB Cyclophyllin B III 10
FKBP10 FKBP65 FK506 binding protein, 65 kDa III, IV 25
SERPINH1 HSP47 Heat-shock protein 47 III, IV 2
PLOD2 LH2 Lysyl hydroxylase 2 III, IV 12
TMEM38B TMEM38B Transmembrane protein 38B IV 3
BMP1 BMP1 Bone morphogenetic protein 1 I, III, IV 11
SEC24D SEC24D SEC24D III, IV 3
SPARC SPARC Secreted protein, acidic, cysteine-rich IV 2
P4HB PDI Protein disulfide isomerase III 1 Cole-Carpenter Syndrome
Osteoblast signaling defects
SP7 SP7 Osterix; transcription factor Sp7 III 1
WNT1 WNT1 WNT1 IV 18
Other osteoblast genes, function to be determined
SERPINF1 PEDF Pigment-epithelium derived factor III, IV 21 When bone histology is
available: OI type VI
IFITM5 BRIL Bone-restricted Ifitm-like V 2
CREB3L1 OASIS Old astrocyte specifically induced II 1
substance
Gene defects with phenotypes similar to OI
WNT1 WNT1 WNT1 18 Dominant: juvenile osteoporosis
PLS3 Plastin-3 Plastin-3 7 X-linked osteoporosis
LRP5 LRP5 LDL receptor related protein 5 50# Recessive: osteoporosis
pseudoglioma syndrome;
dominant: juvenile osteoporosis

Mild, straight legs (OI type I); moderate, leg deformity (OI type IV); severe, leg deformity (OI type III); lethal, does not survive neonatal period (OI type II)
#
Information from the Human Gene Mutation Database (http://www.hgmd.cf.ac.uk/). All other data on the number of mutations is from the OI Variant
Databse (https://oi.gene.le.ac.uk)

Although OI type I is most often caused by 1 (I) from 60,706 unrelated individuals who are not diagnosed with
haploinsufficiency mutations, glycine substitutions in the triple genetic disorders. Of these, 43 have the types of COL1A1 or
helical domain of the 1 (I) or 2 (I) chain can also lead to an OI COL1A2 mutations that are generally thought to lead to OI (38
type I phenotype [15, 18]. In particular, glycine substitutions in individuals with glycine substitutions in the helical domains of
the first 125 residues of the 1 (I) or 2 (I) triple-helical domains 1 (I) or 2 (I); 5 individuals with COL1A1 frameshift muta-
often are associated with blue or gray sclera and normal-looking tions). Thus, the combined frequency of Btypical OI mutations^
teeth (i.e., absence of dentinogenesis imperfecta), similar to 1 (I) in the ExAc database (about 70 per 100,000 individuals)
haploinsufficiency mutations [15, 18]. However, some muta- is several fold greater than the prevalence of OI (about 10
tions affecting the first 90 amino acids are also associated with per 100,000 individuals), suggesting that the majority of
marked joint hyperlaxity [24]. Glycine substitutions in other individuals who have OI mutations are not diagnosed as
parts of the 1 (I) or 2 (I) triple helical domains often lead to having OI.
OI types II, III, or IV, but the phenotype associated with a given Mutations in genes other than COL1A1 and COL1A2 are
substitution is often hard to predict or to explain [25]. usually associated with a moderate to severe phenotype (OI
Insights from new sequencing databases raise the possibility type III, IV, or V). However, there are some exceptions. A
that many Btypical OI mutations^ do not cause an OI pheno- recessive BMP1 mutation affecting the polyadenylation signal
type. The ExAc database (version 0.3; http://exac. of one BMP1 transcript is associated with a mild disease
broadinstitute.org/) contains whole-exome sequencing results course that is similar to OI type I [26]. Mutations in PLS3 also
3430 Osteoporos Int (2016) 27:3427–3437

lead to a mild bone fragility phenotype that can be difficult to membrane at the forearm, but these features may not be pres-
distinguish from OI type I [27]. ent in young children [29, 30]. Finding the OI type V-specific
IFITM5 mutation alerts the clinician that the patient has a high
risk of OI type V complications, such as hyperplastic callus
Molecular diagnosis formation [31], radial head dislocation [32], and abnormalities
in the cranio-cervical junction [33].
The diagnosis of OI types, as proposed by the 2015 Nosology Triple-helical glycine substitutions caused by mutations in
and Classification of Genetic Skeletal Disorders, can usually exon 49 of COL1A2 have been found in children with OI who
be made using patient history, clinical examination, and radio- had intracranial hemorrhage and brachydactyly [34].
graphs. Molecular diagnosis by DNA sequence analysis is Mutations affecting the C-propeptide of 1 (I) are frequently
nevertheless useful to pinpoint the exact cause of OI. A typical associated with hip dysplasia [35]. Identifying bi-allelic
diagnostic workup is shown in Fig. 1. Molecular diagnosis not SERPINF1 mutations may influence the choice of medical
only provides precise information about recurrence risk (dom- treatment, as intravenous bisphosphonate therapy is probably
inant vs recessive OI) to affected individuals and their fami- less effective in the presence of SERPINF1 mutations than in
lies, but also allows identifying affected family members with OI due to other causes. Treatment with RANKL antibody
a high degree of certainty. This is particularly important in OI treatment has shown promising effects in some patients with
type I, where clinical signs of the disease can be subtle. SERPINF1 mutations [36, 37].
However, molecular diagnosis has a very low yield in infants The molecular diagnosis of OI at present is typically per-
that are evaluated for suspected child abuse and in whom formed by DNA sequence analysis of targeted gene panels
careful clinical examination has not revealed clinical features (Bnext-generation sequencing^) [38, 39]. The advantage of
of OI [17, 28]. these methodologies compared to traditional Sanger sequence
Molecular diagnosis can have direct consequences for the analysis is that all known OI-related genes can be analyzed in
clinical management of individual patients. For example, the a single analysis run, which reduces analysis time and cost. In
clinical diagnosis of OI type V is based on the presence of the experience of our molecular diagnosis laboratory, gene
hyperplastic callus or ossification of the interosseous panel analysis of currently known OI genes identifies
disease-causing mutations in at least 97 % of individuals
who have a clinical diagnosis of Btypical OI^ (n = 598) [40].

Treatment of OI

The therapeutic goals in OI vary with phenotype and mobility

status. Children with uncomplicated OI type I may have sim-
ilar levels of physical activity as their healthy peers [41].
Therefore, orthopedic and rehabilitation treatments in mild
OI are typically limited to fracture management. In this con-
text, medical follow-up chiefly serves to screen for complica-
tions, such as vertebral compression fractures, which in many
centers would prompt intravenous bisphosphonate treatment
[42–44]. In contrast, moderate to severe OI is often associated
with long-bone deformities, scoliosis, and reduced mobility,
creating the need for orthopedic and rehabilitation interven-
tions. Such interventions are essential for adequate care of
patients with moderate to severe OI but are beyond the scope
of the present contribution.
Ensuring that vitamin D serum levels remain Bsufficient^ is
a frequently cited goal in the supportive medical management
of OI [5, 45, 46], but the evidence base for any specific 25-
hydroxyvitamin D target level is thin. A bone
histomorphometric study on 71 children with OI found no
Fig. 1 Diagnostic workup for OI, as used at the authors’ institution.
When OI is suspected after clinical evaluation, sequence analysis of an
evidence that 25-hydroxyvitamin D levels in the range from
OI panel is performed using a next-generation sequencing method, as 13 to 103 nmol/L were associated with measures of bone
described [39] mineralization, metabolism, or mass [47]. However, two
Osteoporos Int (2016) 27:3427–3437 3431

cross-sectional retrospective studies observed a relationship

between vitamin D status and bone mineral density (BMD)
[48, 49]. Higher-level evidence was provided by a recent ran-
domized controlled trial on 60 children with OI that compared
two doses of vitamin D supplementation: 400 and 2000 inter-
national units [50]. The group receiving 2000 international
units of vitamin D had higher 25-hydroxyvitamin D serum
levels, but no differences in lumbar spine areal BMD or any
other outcome measures were detected. However, in the 12
patients with baseline 25-hydroxyvitamin D serum levels be-
low 50 nmol/L, lumbar spine areal BMD tended to increase
faster in the group receiving 2000 international units of vita-
min D. It is thus possible that this higher dose of vitamin D is
beneficial to patients with 25-hydroxyvitamin D serum levels
below 50 nmol/L, but further studies that specifically target
this group of patients are needed for verification.

Bisphosphonate treatment

Bisphosphonate therapy has been given to children with OI

for three decades [51] and this is by far the most widely used
medical treatment modality, even though many different pro-
tocols have been proposed. The bisphosphonate treatment ap-
proach used by the authors is shown in Fig. 2. Hundreds of
publications have reported on bisphosphonate treatment in OI
and concur that it leads to an increase in areal BMD at the
spine and several other skeletal sites [52–54]. This is not sur- Fig. 2 Bisphosphonate treatment approach as used by the authors to treat
children with OI who are 2 years of age or older. Intravenous zoledronate
prising, given that bone formation activity is very high during is administered at a dose that depends on lumbar spine areal BMD (LS-
growth, and even more elevated in severe OI, and is not aBMD) results. With this approach, children with less severe forms of OI
coupled to bone resorption in skeletal locations undergoing will receive a lower total exposure of zoledronate than children with more
bone modeling [55, 56]. Anti-osteoclast treatment will there- severe forms of OI. The reason for this is that children with less severe OI
reach the indicated LS-aBMD cutoffs more quickly. Note that the first
fore reliably increase bone mass as long as skeletal growth exposure to intravenous zoledronate occurs at a lower dose to minimize
continues. adverse events (in particular hypocalcemia and the Bacute phase
More controversial is the question of what clinically rele- reaction^) [75]. Once patients have reached final height, the treatment is
vant benefit children with OI derive from higher areal BMD. discontinued [107]. Long-term results of this approach have been de-
scribed [61]
Treatment outcomes such as fracture rate, mobility, and qual-
ity of life are difficult to assess in OI due to the rarity and
of finding patients who are willing to risk being randomized to
heterogeneity of the disorder, the developmental changes oc-
placebo, and the lack of equipoise about treatment benefits
curring during childhood, the multifactorial origin of fractures,
among clinicians who have experience with bisphosphonate
and the variability of concurrent orthopedic and rehabilitation
treatment in OI. Most likely therefore, clinicians will continue
treatments. Large long-term studies would be needed to ac-
to make treatment recommendations that are informed by less-
count for these confounding variables.
than-ideal evidence.
Recent systematic reviews have summarized the evidence
from randomized trials on bisphosphonates in OI [52–54].
These systematic reviews agree that the available evidence is Effect of bisphosphonates on the spine
insufficient to judge whether bisphosphonate therapy im-
proves outcomes other than areal BMD and that more studies Intravenous bisphosphonate therapy has a marked effect on
are needed. This conclusion reflects the fact that available the spine of growing children with OI. When vertebral com-
trials were underpowered to assess outcomes other than areal pression fractures are present at the start of therapy, these tend
BMD and had short treatment durations. However, the reality to reshape, i.e., vertebra gain a more normal shape through
is that large placebo-controlled long-term trials on growth (Fig. 3) [44, 57–60]. As reshaping of compressed ver-
bisphosphonates in OI are unlikely to be conducted. tebra is a growth-dependent process, the potential for
Obstacles to such trials include lack of funding, the difficulty correcting the shape of compressed vertebra depends on how
3432 Osteoporos Int (2016) 27:3427–3437

regained a normal shape by the time the children had finished

growing [61].
It is not clear that the positive spine effects of intravenous
bisphosphonate treatment are also achieved by oral
bisphosphonates. The largest trial on OI conducted to date
(n = 147 participants) did not find an effect of oral risedronate
on vertebral fractures [62]. A similarly sized randomized
placebo-controlled trial on oral alendronate (n = 139) also
did not observe improvements in vertebral shape [63].
However, it is possible that the observation intervals in these
studies on oral bisphosphonates were too short to detect ver-
tebral effects.
Intravenous bisphosphonate therapy, despite improve-
ments in the shape of individual vertebra, does not seem to
have a major effect on the development of scoliosis (Fig. 4).
Two recent studies who in aggregate examined about 750
individuals with OI concordantly concluded that intravenous
bisphosphonate treatment somewhat decreased the progres-
sion rate of scoliosis in OI type III, but there was no evidence
of a positive effect on scoliosis in OI types I and IV [64, 65].
More importantly, the prevalence of scoliosis at maturity was
not influenced by bisphosphonate treatment history in any OI
type. It thus appears that bisphosphonate therapy may slow the
progression of scoliosis in the most severely affected patients
but does not change the final outcome.

Fig. 3 Reshaping of vertebral bodies in a boy with OI type IV at Effects of bisphosphonates on long bones
12.0 years of age (a) and after 4 years of treatment with intravenous
zoledronate (b)
Both oral and intravenous bisphosphonate therapy seem to be
associated with a lower rate of long-bone fractures in children
much growth remains at the time of treatment inception. A with OI ([43, 61, 62, 66, 67]). Reported fracture rate reduc-
recent study in children with moderate to severe OI who tions are typically in the range of 30 to 60 %, which indicates
started intravenous bisphosphonate therapy before 5 years of some therapeutic efficacy but also implies that a large number
age found that the majority of compressed vertebra had of long-bone fractures continue to occur. Indeed, a recent

Fig. 4 Progression of spine curvature in a girl with OI type IV at the age age and continued until 15.4 years of age (initially pamidronate;
of 3.4 years (a), 11.9 years (b), 13.6 years (c), and 15.0 years (d). zoledronate after the age of 10.3 years). Spinal fusion surgery was
Treatment with intravenous bisphosphonates was started at 3.4 years of performed at the age of 14.4 years
Osteoporos Int (2016) 27:3427–3437 3433

study followed a group of 37 children with moderate to severe all children starting bisphosphonate treatment and to complete
OI for 15 years after the start of intravenous bisphosphonate any necessary dental work prior to the first bisphosphonate
therapy and documented a median of 6 femur and 5 tibia infusion [82].
fractures per patient [61]. Thus, long-bone fractures are clearly As many children with moderate to severe OI undergo
still a major problem despite bisphosphonate treatment. intramedullary rodding surgery to correct bone deformities,
Several factors may contribute to this persistence of high the interplay between bisphosphonate therapy and osteotomy
long-bone fracture rates during bisphosphonate treatment. healing is a clinically important topic. It had previously been
Children with OI have long-bone diaphyses with a very small found that intravenous pamidronate therapy was associated
bone cross - section, indicating decreased periosteal bone with an increased rate in delayed healing of osteotomy sites,
growth [68]. Total volumetric BMD of long-bone diaphyses but not of fracture sites [83]. A follow-up study was performed
is therefore abnormally high in OI, which is in marked contrast after changes in surgical technique as well in bisphosphonate
to the low BMD at metaphyseal sites [69]. Bisphosphonate infusion protocol (infusions were no longer given in the
treatment has no obvious effect on the cross-sectional size of 4 months following the osteotomy) had been implemented. A
long-bone shafts [61] and may have limited scope for increas- review of 261 rodding procedures on 110 patients showed that
ing BMD when total volumetric BMD is already elevated. delayed osteotomy healing still occurred but that the incidence
Material bone properties are also decreased in long-bone diaph- was markedly reduced under the modified approach [84].
yses of children with OI [70], and material properties of OI A number of recent case reports have suggested that atyp-
bone are not detectably altered by bisphosphonate therapy ical femur fractures occur in children and adults with OI who
[71]. Another factor contributing to long-bone fractures in had received bisphosphonate therapy [85–89]. The diagnostic
many children with moderate to severe OI are bone deformities. entity of atypical femur fractures was originally established to
These do not respond to any known medical intervention but describe a specific type of diaphyseal femur fractures in wom-
rather need to be corrected by surgical intervention. en with postmenopausal osteoporosis who had received anti-
Despite the continued occurrence of long-bone fractures, it resorptive treatment [90]. However, fracture epidemiology is
has long been noted that intravenous bisphosphonate treat- not as well characterized in OI as it is in postmenopausal
ment can improve mobility, especially when started in the first osteoporosis. It is therefore not clear that the type of femur
few years of life [44, 59]. Long-term follow-up suggests that fracture that would be deemed atypical in the context of post-
most children with OI type IV, but not those with OI type III, menopausal osteoporosis is actually atypical for individuals
achieve the ability to walk independently [72]. In individuals with OI. Indeed, transverse diaphyseal femur fractures (the
with OI type III, the functional goal typically is to achieve the key feature of atypical femur fractures [90]) have been among
ability to live independently despite restricted mobility [72]. the most common fractures in OI even before the bisphospho-
nate era [91]. Systematic studies are necessary to assess
Potential adverse events of bisphosphonates whether the prevalence of Batypical femur fractures^ is higher
in bisphosphonate-treated individuals.
Hypocalcemia, fever, and vomiting are well-known adverse
events during the first exposure to intravenous Drugs other than bisphosphonates
bisphosphonates such as pamidronate, ibandronate, or
zoledronate [73–77]. An apparently life-threatening systemic Even though most children with moderate to severe OI benefit
response to the first dose of zoledronic acid has been reported from bisphosphonate therapy, it is obvious that there is a lot of
in one child (with a diagnosis other than OI) [78]. Acute de- room for further improvement. Alternative medical ap-
terioration of respiratory status has also been reported in in- proaches are therefore being investigated.
fants and young children with OI who were receiving The options for anti-resorptive therapy are not limited to
pamidronate [79, 80]. It is an important responsibility of phy- bisphosphonates. Osteoclast inhibition can also be achieved
sicians administering bisphosphonates to report such severe with denosumab, a drug based on an antibody against
adverse events to drug safety systems such as MedWatch of RANKL. Denosumab is approved for the treatment of post-
the Food and Drug Administration (http://www.fda. menopausal osteoporosis and other skeletal disorders in
gov/Safety/MedWatch/). adults. A few case series have been published on the use of
Among the longer-term potential adverse events of denosumab in children with OI caused by SERPINF1 muta-
bisphosphonates, osteonecrosis of the jaw is a frequently men- tions [36, 37] and in patients with COL1A1 or COL1A2 mu-
tioned concern that has undergone intense scrutiny. However, tations [92]. In both situations, a decrease in bone metabolism
a recent systematic review did not identify any confirmed markers and an increase in areal BMD were observed.
occurrence in children with OI [81] and there seem to be no On a bone histological level, denosumab seems to have a
reported cases in adults with OI either. Some experts have similar effect on growing children as intravenous pamidronate
nevertheless recommended performing a dental review on [93, 94]. However, denosumab has a much shorter duration of
3434 Osteoporos Int (2016) 27:3427–3437

action than bisphosphonates, which can be seen as an advan- Conflicts of interest Pamela Trejo and Frank Rauch declare that they
have no conflict of interest.
tage because it allows better control of the duration of anti-
resorptive action. The reverse side of the shorter duration of
osteoclast inhibition is that potentially severe Brebound
hypercalcemia^ may occur when the anti-resorptive activity
References
of denosumab ceases [95, 96]. This has not been observed in
1. Forlino A, Marini JC (2016) Osteogenesis imperfecta. Lancet 387:
the reported children with OI who received denosumab, but
1657–1671
these children had all previously been treated with 2. Hendrickx G, Boudin E, Van Hul W (2015) A look behind the
bisphosphonates, which presumably prevents a sudden surge scenes: the risk and pathogenesis of primary osteoporosis. Nat
in osteoclast activity after denosumab treatment. Rev Rheumatol 11:462–474
As the genetic defect underlying OI primarily affects the 3. Rivadeneira F, Makitie O (2016) Osteoporosis and bone mass
disorders: from gene pathways to treatments. Trends Endocrinol
osteoblast, it is intuitively appealing to attempt therapy with Metab 27:262–281
stimulators of bone formation. A randomized controlled trial 4. Biggin A, Munns CF (2014) Osteogenesis imperfecta: diagnosis
on teriparatide in adults with OI showed increased BMD in and treatment. Curr Osteoporos Rep 12:279–288
mild OI and less effect in moderate to severe OI [97]. Another 5. Harrington J, Sochett E, Howard A (2014) Update on the evalua-
tion and treatment of osteogenesis imperfecta. Pediatr Clin North
approach to stimulate bone formation is through antibody- Am 61:1243–1257
mediated sclerostin inhibition. Sclerostin antibody treatment 6. Hoyer-Kuhn H, Netzer C, Semler O (2015) Osteogenesis
has shown promise in several OI mouse models with mild imperfecta: pathophysiology and treatment. Wien Med
bone involvement [98–100]. However, the beneficial effect Wochenschr 165:278–284
7. Shaker JL, Albert C, Fritz J, Harris G (2015) Recent developments
of sclerostin inhibition was less obvious in a mouse model
in osteogenesis imperfecta. F1000Res 4:681
of more severe dominant OI [101]. Clinical experience with 8. Sillence DO, Senn A, Danks DM (1979) Genetic heterogeneity in
sclerostin antibody treatment of OI has not yet been reported. osteogenesis imperfecta. J Med Genet 16:101–116
One potential issue with bone anabolic agents in OI is that 9. Bonafe L, Cormier-Daire V, Hall C et al (2015) Nosology and
bone formation rate is markedly increased in children with mod- classification of genetic skeletal disorders: 2015 revision. Am J
Med Genet A 167A:2869–2892
erate to severe OI prior to any bone-specific treatment [55, 102]. 10. Rauch F, Glorieux FH (2004) Osteogenesis imperfecta. Lancet
Somewhat counterintuitively, the low bone mass in OI is there- 363:1377–1385
fore not caused by a lack of bone formation activity but by high 11. Glorieux FH, Ward LM, Rauch F, Lalic L, Roughley PJ, Travers R
bone resorption. This may limit the effectiveness of bone ana- (2002) Osteogenesis imperfecta type VI: a form of brittle bone
disease with a mineralization defect. J Bone Miner Res 17:30–38
bolic treatments unless bone resorption is inhibited at the same
12. Ward LM, Rauch F, Travers R, Chabot G, Azouz EM, Lalic L,
time. A combination therapy using both an anabolic and an anti- Roughley PJ, Glorieux FH (2002) Osteogenesis imperfecta type
resorptive agent therefore appears intuitively appealing, similar VII: an autosomal recessive form of brittle bone disease. Bone 31:
to the situation in postmenopausal osteoporosis, where this ap- 12–18
proach has shown promising results [103]. 13. Van Dijk FS, Sillence DO (2014) Osteogenesis imperfecta: clini-
cal diagnosis, nomenclature and severity assessment. Am J Med
Genet A 164A:1470–1481
14. Bonadio J, Holbrook KA, Gelinas RE, Jacob J, Byers PH (1985)
Outlook Altered triple helical structure of type I procollagen in lethal peri-
natal osteogenesis imperfecta. J Biol Chem 260:1734–1742
15. Lindahl K, Astrom E, Rubin CJ, Grigelioniene G, Malmgren B,
The study of rare diseases often requires collection of data Ljunggren O, Kindmark A (2015) Genetic epidemiology, preva-
from large catchment areas and collaboration between centers. lence, and genotype-phenotype correlations in the Swedish popu-
In OI, several such collaboration are under way using regis- lation with osteogenesis imperfecta. Eur J Hum Genet 23:1042–
tries [104, 105], research networks supported by patient advo- 1050
16. Bronicki LM, Stevenson RE, Spranger JW (2015) Beyond osteo-
cacy organizations [106], or industry-sponsored pharmaceuti- genesis imperfecta: causes of fractures during infancy and child-
cal trials. Novel developments include the establishment of hood. Am J Med Genet C: Semin Med Genet 169:314–327
the Brittle Bone Disease Consortium (https://www. 17. Pepin MG, Byers PH (2015) What every clinical geneticist should
rarediseasesnetwork.org/cms/BBD) that is supported by the know about testing for osteogenesis imperfecta in suspected child
abuse cases. Am J Med Genet C: Semin Med Genet 169:307–313
National Institutes of Health in the USA and similar efforts 18. Rauch F, Lalic L, Roughley P, Glorieux FH (2010) Genotype-
in Europe. These collaborative efforts will help to accelerate phenotype correlations in nonlethal osteogenesis imperfecta
the development of novel therapeutic approaches and their caused by mutations in the helical domain of collagen type I.
application to clinical practice. Eur J Hum Genet 18:642–647
19. Valadares ER, Carneiro TB, Santos PM, Oliveira AC, Zabel B
(2014) What is new in genetics and osteogenesis imperfecta clas-
Acknowledgments F.R. received salary support from the Chercheur- sification? J Pediatr (Rio J) 90:536–541
Boursier Clinicien program of the Fonds de Recherche du Québec–Santé 20. Willing MC, Deschenes SP, Scott DA, Byers PH, Slayton RL,
and was supported by the Shriners of North America. Pitts SH, Arikat H, Roberts EJ (1994) Osteogenesis imperfecta
Osteoporos Int (2016) 27:3427–3437 3435

type I: molecular heterogeneity for COL1A1 null alleles of type I 37. Hoyer-Kuhn H, Netzer C, Koerber F, Schoenau E, Semler O
collagen. Am J Hum Genet 55:638–647 (2014) Two years’ experience with denosumab for children with
21. Schleit J, Bailey SS, Tran T, Chen D, Stowers S, Schwarze U, osteogenesis imperfecta type VI. Orphanet J Rare Dis 9:145
Byers PH (2015) Molecular outcome, prediction, and clinical con- 38. Sule G, Campeau PM, Zhang VW et al (2013) Next-generation
sequences of splice variants in COL1A1, which encodes the sequencing for disorders of low and high bone mineral density.
proalpha1(I) chains of type I procollagen. Hum Mutat 36:728–739 Osteoporos Int 24:2253–2259
22. Bardai G, Lemyre E, Moffatt P, Palomo T, Glorieux FH, Tung J, 39. Rauch F, Lalic L, Glorieux FH, Moffatt P, Roughley P (2014)
Ward L, Rauch F (2016) Osteogenesis imperfecta type I caused by Targeted sequencing of a pediatric metabolic bone gene panel
COL1A1 deletions. Calcif Tissue Int 98:76–84 using a desktop semiconductor next-generation sequencer. Calcif
23. Ben Amor IM, Roughley P, Glorieux FH, Rauch F (2013) Skeletal Tissue Int 95:323–331
clinical characteristics of osteogenesis imperfecta caused by 40. Bardai G, Moffatt P, Glorieux FH, Rauch F DNA sequence anal-
haploinsufficiency mutations in COL1A1. J Bone Miner Res 28: ysis in 598 individuals with a clinical diagnosis of osteogenesis
2001–2007 imperfecta: diagnostic yield and mutation spectrum..Osteoporos
24. Cabral WA, Makareeva E, Colige A, Letocha AD, Ty JM, Yeowell Int. In press
HN, Pals G, Leikin S, Marini JC (2005) Mutations near amino end 41. Pouliot-Laforte A, Veilleux LN, Rauch F, Lemay M (2015)
of alpha1(I) collagen cause combined osteogenesis imperfecta/ Physical activity in youth with osteogenesis imperfecta type I. J
Ehlers-Danlos syndrome by interference with N-propeptide pro- Musculoskelet Neuronal Interact 15:171–176
cessing. J Biol Chem 280:19259–19269 42. Glorieux FH (2007) Treatment of osteogenesis imperfecta: who,
25. Marini JC, Forlino A, Cabral WA et al (2007) Consortium for why, what? Horm Res 68(Suppl 5):8–11
osteogenesis imperfecta mutations in the helical domain of type 43. Lindahl K, Kindmark A, Rubin CJ, Malmgren B, Grigelioniene G,
I collagen: regions rich in lethal mutations align with collagen Soderhall S, Ljunggren O, Astrom E (2016) Decreased fracture
binding sites for integrins and proteoglycans. Hum Mutat 28: rate, pharmacogenetics and BMD response in 79 Swedish children
209–221 with osteogenesis imperfecta types I, III and IV treated with
26. Fahiminiya S, Al-Jallad H, Majewski J, Palomo T, Moffatt P, pamidronate. Bone 87:11–18
Roschger P, Klaushofer K, Glorieux FH, Rauch F (2015) A 44. Alcausin MB, Briody J, Pacey V, Ault J, McQuade M, Bridge C,
polyadenylation site variant causes transcript-specific BMP1 defi- Engelbert RH, Sillence DO, Munns CF (2013) Intravenous
ciency and frequent fractures in children. Hum Mol Genet 24: pamidronate treatment in children with moderate-to-severe osteo-
516–524 genesis imperfecta started under three years of age. Horm Res
27. van Dijk FS, Zillikens MC, Micha D et al (2013) PLS3 mutations Paediatr 79:333–340
in X-linked osteoporosis with fractures. N Engl J Med 369:1529– 45. Saraff V, Hogler W (2015) Osteoporosis in children: diagnosis and
1536 management. Eur J Endocrinol 173:R185–R197
28. Zarate YA, Clingenpeel R, Sellars EA, Tang X, Kaylor JA, 46. Thomas IH, DiMeglio LA (2016) Advances in the classification
Bosanko K, Linam LE, Byers PH (2016) COL1A1 and and treatment of osteogenesis imperfecta. Curr Osteoporos Rep
COL1A2 sequencing results in cohort of patients undergoing 14:1–9
evaluation for potential child abuse. Am J Med Genet A 47. Edouard T, Glorieux FH, Rauch F (2011) Relationship between
29. Rauch F, Moffatt P, Cheung M, Roughley P, Lalic L, Lund AM, vitamin D status and bone mineralization, mass, and metabolism
Ramirez N, Fahiminiya S, Majewski J, Glorieux FH (2013) in children with osteogenesis imperfecta: histomorphometric
Osteogenesis imperfecta type V: marked phenotypic variability study. J Bone Miner Res 26:2245–2251
despite the presence of the IFITM5 c.-14C>T mutation in all pa- 48. Edouard T, Glorieux FH, Rauch F (2011) Predictors and correlates
tients. J Med Genet 50:21–24 of vitamin D status in children and adolescents with osteogenesis
30. Grover M, Campeau PM, Lietman CD, Lu JT, Gibbs RA, imperfecta. J Clin Endocrinol Metab 96:3193–3198
Schlesinger AE, Lee BH (2013) Osteogenesis imperfecta without 49. Zambrano MB, Brizola E, Pinheiro B, Vanz AP, Mello ED, Felix
features of type V caused by a mutation in the IFITM5 gene. J TM (2016) Study of the determinants of vitamin D status in pedi-
Bone Miner Res 28:2333–2337 atric patients with osteogenesis imperfecta. J Am Coll Nutr 35:
31. Cheung MS, Glorieux FH, Rauch F (2007) Natural history of 339–345
hyperplastic callus formation in osteogenesis imperfecta type V. 50. Plante L, Veilleux LN, Glorieux FH, Weiler H, Rauch F (2016)
J Bone Miner Res 22:1181–1186 Effect of high-dose vitamin D supplementation on bone density in
32. Fassier AM, Rauch F, Aarabi M, Janelle C, Fassier F (2007) youth with osteogenesis imperfecta: a randomized controlled trial.
Radial head dislocation and subluxation in osteogenesis Bone 86:36–42
imperfecta. J Bone Joint Surg Am 89:2694–2704 51. Devogelaer JP, Malghem J, Maldague B, Nagant de Deuxchaisnes
33. Cheung MS, Arponen H, Roughley P, Azouz ME, Glorieux FH, C (1987) Radiological manifestations of bisphosphonate treatment
Waltimo-Siren J, Rauch F (2011) Cranial base abnormalities in with APD in a child suffering from osteogenesis imperfecta.
osteogenesis imperfecta: phenotypic and genotypic determinants. Skeletal Radiol 16:360–363
J Bone Miner Res 26:405–413 52. Dwan K, Phillipi CA, Steiner RD, Basel D (2014) Bisphosphonate
34. Faqeih E, Roughley P, Glorieux FH, Rauch F (2009) Osteogenesis therapy for osteogenesis imperfecta. Cochrane Database Syst Rev
imperfecta type III with intracranial hemorrhage and 7:Cd005088
brachydactyly associated with mutations in exon 49 of 53. Hald JD, Evangelou E, Langdahl BL, Ralston SH (2015)
COL1A2. Am J Med Genet A 149A:461–465 Bisphosphonates for the prevention of fractures in osteogenesis
35. Kishta W, Abduljabbar FH, Gdalevitch M, Rauch F, Hamdy R, imperfecta: meta-analysis of placebo-controlled trials. J Bone
Fassier F (2015) Hip dysplasia in children with osteogenesis Miner Res 30:929–933
imperfecta: association with collagen type I C-propeptide muta- 54. Rijks EB, Bongers BC, Vlemmix MJ, Boot AM, van Dijk AT,
tions. J Pediatr Orthop Sakkers RJ, van Brussel M (2015) Efficacy and safety of bisphos-
36. Semler O, Netzer C, Hoyer-Kuhn H, Becker J, Eysel P, Schoenau phonate therapy in children with osteogenesis imperfecta: a sys-
E (2012) First use of the RANKL antibody denosumab in osteo- tematic review. Horm Res Paediatr 84:26–42
genesis imperfecta type VI. J Musculoskelet Neuronal Interact 12: 55. Rauch F, Lalic L, Roughley P, Glorieux FH (2010) Relationship
183–188 between genotype and skeletal phenotype in children and
3436 Osteoporos Int (2016) 27:3427–3437

adolescents with osteogenesis imperfecta. J Bone Miner Res 25: 73. Robinson RE, Nahata MC, Hayes JR, Batisky DL, Bates CM,
1367–1374 Mahan JD (2004) Effectiveness of pretreatment in decreasing ad-
56. Rauch F, Travers R, Plotkin H, Glorieux FH (2002) The effects of verse events associated with pamidronate in children and adoles-
intravenous pamidronate on the bone tissue of children and ado- cents. Pharmacotherapy 24:195–197
lescents with osteogenesis imperfecta. J Clin Invest 110:1293– 74. Li M, Xia WB, Xing XP et al (2011) Benefit of infusions with
1299 ibandronate treatment in children with osteogenesis imperfecta.
57. Land C, Rauch F, Munns CF, Sahebjam S, Glorieux FH (2006) Chin Med J (Engl) 124:3049–3053
Vertebral morphometry in children and adolescents with osteogen- 75. Munns CF, Rajab MH, Hong J, Briody J, Hogler W, McQuade M,
esis imperfecta: effect of intravenous pamidronate treatment. Bone Little DG, Cowell CT (2007) Acute phase response and mineral
39:901–906 status following low dose intravenous zoledronic acid in children.
58. Semler O, Beccard R, Palmisano D, Demant A, Fricke O, Bone 41:366–370
Schoenau E, Koerber F (2011) Reshaping of vertebrae during 76. George S, Weber DR, Kaplan P, Hummel K, Monk HM, Levine
treatment with neridronate or pamidronate in children with osteo- MA (2015) Short-term safety of zoledronic acid in young patients
genesis imperfecta. Horm Res Paediatr 76:321–327 with bone disorders: an extensive institutional experience. J Clin
59. Astrom E, Jorulf H, Soderhall S (2007) Intravenous pamidronate Endocrinol Metab 100:4163–4171
treatment of infants with severe osteogenesis imperfecta. Arch Dis 77. Kumar C, Panigrahi I, Somasekhara Aradhya A, Meena BL,
Child 92:332–338 Khandelwal N (2016) Zoledronate for osteogenesis imperfecta:
60. Astrom E, Soderhall S (2002) Beneficial effect of long term intra- evaluation of safety profile in children. J Pediatr Endocrinol
venous bisphosphonate treatment of osteogenesis imperfecta. Metab
Arch Dis Child 86:356–364 78. Trivedi S, Al-Nofal A, Kumar S, Tripathi S, Kahoud RJ, Tebben
61. Palomo T, Fassier F, Ouellet J, Sato A, Montpetit K, Glorieux FH, PJ (2016) Severe non-infective systemic inflammatory response
Rauch F (2015) Intravenous bisphosphonate therapy of young syndrome, shock, and end-organ dysfunction after zoledronic acid
children with osteogenesis imperfecta: skeletal findings during administration in a child. Osteoporos Int
follow up throughout the growing years. J Bone Miner Res 30: 79. Munns CF, Rauch F, Mier RJ, Glorieux FH (2004) Respiratory
2150–2157 distress with pamidronate treatment in infants with severe osteo-
62. Bishop N, Adami S, Ahmed SF et al (2013) Risedronate in chil- genesis imperfecta. Bone 35:231–234
dren with osteogenesis imperfecta: a randomised, double-blind, 80. Olson JA (2014) Respiratory failure during infusion of
placebo-controlled trial. Lancet 382:1424–1432 pamidronate in a 3 year-old male with osteogenesis imperfecta:
a case report. J Pediatr Rehabil Med 7:155–158
63. Ward LM, Rauch F, Whyte MP et al (2011) Alendronate for the
81. Hennedige AA, Jayasinghe J, Khajeh J, Macfarlane TV (2013)
treatment of pediatric osteogenesis imperfecta: a randomized
Systematic review on the incidence of bisphosphonate related
placebo-controlled study. J Clin Endocrinol Metab 96:355–364
osteonecrosis of the jaw in children diagnosed with osteogenesis
64. Sato A, Ouellet J, Muneta T, Glorieux FH, Rauch F (2016)
imperfecta. J Oral Maxillofac Res 4:e1
Scoliosis in osteogenesis imperfecta caused by COL1A1/
82. Bhatt RN, Hibbert SA, Munns CF (2014) The use of
COL1A2 mutations—genotype-phenotype correlations and effect
bisphosphonates in children: review of the literature and guide-
of bisphosphonate treatment. Bone 86:53–57
lines for dental management. Aust Dent J 59:9–19
65. Anissipour AK, Hammerberg KW, Caudill A, Kostiuk T, Tarima 83. Munns CF, Rauch F, Zeitlin L, Fassier F, Glorieux FH (2004)
S, Zhao HS, Krzak JJ, Smith PA (2014) Behavior of scoliosis Delayed osteotomy but not fracture healing in pediatric osteogen-
during growth in children with osteogenesis imperfecta. J Bone esis imperfecta patients receiving pamidronate. J Bone Miner Res
Joint Surg Am 96:237–243 19:1779–1786
66. Sakkers R, Kok D, Engelbert R, van Dongen A, Jansen M, Pruijs 84. Anam EA, Rauch F, Glorieux FH, Fassier F, Hamdy R (2015)
H, Verbout A, Schweitzer D, Uiterwaal C (2004) Skeletal effects Osteotomy healing in children with osteogenesis imperfecta re-
and functional outcome with olpadronate in children with osteo- ceiving bisphosphonate treatment. J Bone Miner Res 30:1362–
genesis imperfecta: a 2-year randomised placebo-controlled study. 1368
Lancet 363:1427–1431 85. Meier RP, Ing Lorenzini K, Uebelhart B, Stern R, Peter RE,
67. Glorieux FH, Bishop NJ, Plotkin H, Chabot G, Lanoue G, Travers Rizzoli R (2012) Atypical femoral fracture following bisphospho-
R (1998) Cyclic administration of pamidronate in children with nate treatment in a woman with osteogenesis imperfecta—a case
severe osteogenesis imperfecta. N Engl J Med 339:947–952 report. Acta Orthop 83:548–550
68. Palomo T, Glorieux FH, Schoenau E, Rauch F (2016) Body com- 86. Manolopoulos KN, West A, Gittoes N (2013) The paradox of
position in children and adolescents with osteogenesis imperfecta. prevention—bilateral atypical subtrochanteric fractures due to
J Pediatr 169:232–237 bisphosphonates in osteogenesis imperfecta. J Clin Endocrinol
69. Rauch F, Land C, Cornibert S, Schoenau E, Glorieux FH (2005) Metab 98:871–872
High and low density in the same bone: a study on children and 87. Holm J, Eiken P, Hyldstrup L, Jensen JE (2014) Atypical femoral
adolescents with mild osteogenesis imperfecta. Bone 37:634–641 fracture in an osteogenesis imperfecta patient successfully treated
70. Albert C, Jameson J, Smith P, Harris G (2014) Reduced diaphy- with teriparatide. Endocr Pract 20:e187–e190
seal strength associated with high intracortical vascular porosity 88. Etxebarria-Foronda I, Carpintero P (2015) An atypical fracture in
within long bones of children with osteogenesis imperfecta. Bone male patient with osteogenesis imperfecta. Clin Cases Miner Bone
66:121–130 Metab 12:278–281
71. Weber M, Roschger P, Fratzl-Zelman N, Schoberl T, Rauch F, 89. Vasanwala RF, Sanghrajka A, Bishop NJ, Hogler W (2016)
Glorieux FH, Fratzl P, Klaushofer K (2006) Pamidronate does Recurrent proximal femur fractures in a teenager with osteogene-
not adversely affect bone intrinsic material properties in children sis imperfecta on continuous bisphosphonate therapy: are we
with osteogenesis imperfecta. Bone 39:616–622 overtreating? J Bone Miner Res
72. Montpetit K, Palomo T, Glorieux FH, Fassier F, Rauch F (2015) 90. Shane E, Burr D, Abrahamsen B et al (2014) Atypical
Multidisciplinary treatment of severe osteogenesis imperfecta: subtrochanteric and diaphyseal femoral fractures: second report
functional outcomes at skeletal maturity. Arch Phys Med of a Task Force of the American Society for Bone and Mineral
Rehabil 96:1834–1839 Research. J Bone Miner Res 29:1–23
Osteoporos Int (2016) 27:3427–3437 3437

91. Dent JA, Paterson CR (1991) Fractures in early childhood: osteo- 100. Grafe I, Alexander S, Yang T et al (2016) Sclerostin antibody
genesis imperfecta or child abuse? J Pediatr Orthop 11:184–186 treatment improves the bone phenotype of Crtap(−/−) mice, a
92. Hoyer-Kuhn H, Franklin J, Allo G, Kron M, Netzer C, Eysel P, model of recessive osteogenesis imperfecta. J Bone Miner Res
Hero B, Schoenau E, Semler O (2016) Safety and efficacy of 31:1030–1040
denosumab in children with osteogenesis imperfecta—a first pro- 101. Roschger A, Roschger P, Keplingter P, Klaushofer K, Abdullah S,
spective trial. J Musculoskelet Neuronal Interact 16:24–32 Kneissel M, Rauch F (2014) Effect of sclerostin antibody treat-
93. Wang HD, Boyce AM, Tsai JY, Gafni RI, Farley FA, Kasa-Vubu ment in a mouse model of severe osteogenesis imperfecta. Bone
JZ, Molinolo AA, Collins MT (2014) Effects of denosumab treat- 66:182–188
ment and discontinuation on human growth plates. J Clin 102. Rauch F, Travers R, Parfitt AM, Glorieux FH (2000) Static and
Endocrinol Metab 99:891–897 dynamic bone histomorphometry in children with osteogenesis
94. Rauch F, Travers R, Munns C, Glorieux FH (2004) Sclerotic imperfecta. Bone 26:581–589
metaphyseal lines in a child treated with pamidronate: 103. Reid IR (2015) Short-term and long-term effects of osteoporosis
histomorphometric analysis. J Bone Miner Res 19:1191–1193 therapies. Nat Rev Endocrinol 11:418–428
95. Boyce AM, Chong WH, Yao J et al (2012) Denosumab treatment 104. Tosi LL, Oetgen ME, Floor MK et al (2015) Initial report of the
for fibrous dysplasia. J Bone Miner Res 27:1462–1470 osteogenesis imperfecta adult natural history initiative. Orphanet J
96. Setsu N, Kobayashi E, Asano N, Yasui N, Kawamoto H, Kawai A, Rare Dis 10:146
Horiuchi K (2016) Severe hypercalcemia following denosumab
105. Yimgang DP, Shapiro JR (2016) Pregnancy outcomes in women
treatment in a juvenile patient. J Bone Miner Metab 34:118–122
with osteogenesis imperfecta. J Matern Fetal Neonatal Med 29:
97. Orwoll ES, Shapiro J, Veith S et al (2014) Evaluation of
2358–2362
teriparatide treatment in adults with osteogenesis imperfecta. J
Clin Invest 124:491–498 106. Patel RM, Nagamani SC, Cuthbertson D et al (2015) A cross-
98. Sinder BP, Eddy MM, Ominsky MS, Caird MS, Marini JC, sectional multicenter study of osteogenesis imperfecta in North
Kozloff KM (2013) Sclerostin antibody improves skeletal param- America—results from the linked clinical research centers. Clin
eters in a Brtl/+ mouse model of osteogenesis imperfecta. J Bone Genet 87:133–140
Miner Res 28:73–80 107. Rauch F, Munns C, Land C, Glorieux FH (2006) Pamidronate in
99. Jacobsen CM, Barber LA, Ayturk UM et al (2014) Targeting the children and adolescents with osteogenesis imperfecta: effect of
LRP5 pathway improves bone properties in a mouse model of treatment discontinuation. J Clin Endocrinol Metab 91:1268–1274
osteogenesis imperfecta. J Bone Miner Res 29:2297–2306