The Lesson of Temporary Brittle Bone Disease:
All Bones Are Not Created Equal
-Marvin E. Miller, M.D.
Wright State University School of Medicine
Department of Pediatrics
BONE Manuscript # 03-01-00010
Revision #2 – submitted on June 2, 2003
Accepted for publication: June 9, 2003
Address: Marvin E. Miller, M.D.
Children’s Medical Center
1 Children’s Plaza
Department of Medical Genetics
Dayton, Ohio 45404
Phone: (937) 641-5374
Fax: (937) 641-5325
e-mail:
[email protected]
Abstract
Temporary brittle bone disease (TBBD) is a recently described phenotype of multiple, unexplained fractures in the first year of life, and predominantly in the first 6 months of life. There is usually no other injury such as bruising, subdural hematomas, retinal hemorrhages, or other internal organ injury. The susceptibility to fracture is transient, and there are no other radiographic or biochemical abnormalities noted in the standard evaluation that might suggest an underlying cause. The child abuse and pediatric
radiology communities have, for the most part, been unwilling to accept this as a real condition, for they believe it is a ruse for child abuse. This review describes the experience of the author in evaluating infants with multiple unexplained fractures and the hypothesis that has emerged for explaining TBBD. The hypothesis is a prenatal application of the mechanostat/bone loading theory of bone formation and states that TBBD is caused by fetal immobilization which leads to fetal bone unloading and transient, relative osteopenia. Such susceptible infants can fracture with
routine handling and present with a pattern of fractures that is similar to that which has been thought to be highly specific for child abuse. The review presents: 1) the evidence that indicates that normal fetal movement is important for normal fetal bone strength, 2) a critique of the radiologic approach in the diagnosis of child abuse in infants with multiple unexplained fractures, 3) observations that would indicate that child abuse is unlikely in infants with TBBD, and 4) new approaches to the infant with
multiple unexplained fractures that would assist in accurate diagnosis.
KEY WORDS: temporary brittle bone disease
child abuse
bone loading
fetal immobilization
mechanostat
Utah paradigm
The Issue: Unexplained Fractures in Infancy and TBBD
The infant who presents with multiple unexplained fractures poses a diagnostic conundrum, and the final diagnosis will have lifelong social implications for the infant and the parents/caregivers. The interpretation of the plain X-ray is often the single most important determinant in the disposition of cases of infants with multiple unexplained fractures. The radiologic findings of metaphyseal fractures, posterior rib fractures, and multiple fractures at different ages of healing are thought to be highly specific for child abuse.1,21 When these radiologic findings are observed in the setting of 1) no antecedent history of trauma, 2) apparent normal bone density by plain radiograph, 3) no radiographic evidence of metabolic bone disease such as is seen in rickets, and 4) no biochemical evidence of bone disease with normal serum calcium, phosphorus, and vitamin D studies, and normal collagen studies from skin fibroblasts, then it is highly likely that the final diagnosis will be child abuse. This may occur even though the parents and other caregivers deny intentionally inflicting harm on the infant, and even though the risk profile of these individuals for committing
child abuse is low.5 With society’s ultimate goal of protecting helpless infants from potential harm, it is understandable that the medicolegal system has taken the approach of removing these infants from their parents and other caregivers until a complete evaluation of the situation can be done. The medicolegal approach to the infant with multiple unexplained fractures makes a very important assumption. Except for infants with rare genetic bone disorders associated with brittle bones, it is assumed that
apparently normal infants all have similar bone strength, and that if such an infant incurs multiple unexplained fractures, the forces to cause these fractures must have been significant and likely inflicted.
Until recently no one took exception to this approach. However, this is now a charged and hotly debated issue. In 1993 Colin Paterson described a series of infants with multiple unexplained fractures in which parents and caregivers denied intentional injury.39 The x-rays of these infants often had the radiologic features mentioned above that are thought to be pathognomonic of child abuse, and the infants had normal biochemical
evaluations that revealed no evidence of bone disorders such as osteogenesis imperfecta or rickets that might explain the fractures. Paterson called this entity “temporary brittle bone disease” (TBBD) and suggested it is a transient state of increased fracture susceptibility, predominantly during the first year of life, that most likely has a biological cause and not a social one. He has rightfully pointed out that it is difficult to envision child abuse in infants with up to 30 fractures at different stages of healing who have been seen by health care providers on multiple visits in the past and who have no evidence of bruising that one would think would be
present with such a multitude of fractures.
Paterson hypothesized that copper deficiency might be the cause of TBBD as infants with copper deficiency had previously been described with a phenotype similar to TBBD39. Copper deficiency can cause a brittle bone state that can mimic child abuse and was at one time caused by inadequate provision of copper in the diets of premature infants.40 However, Paterson was unable to demonstrate evidence of decreased serum copper concentrations
in infants with TBBD.
Paterson has had vocal critics who maintain that TBBD is merely a ruse for child abuse, that the reason the fractures disappear is that the infant is removed from the care of those inflicting the physical injuries, and that no biological explanation for TBBD has been demonstrated.2,7,21,47,50 Such cases often lead to legal proceedings in both civil and criminal courts where the defense uses the diagnosis of TBBD as their explanation for the fractures, and the prosecution maintains this is child abuse and felonious
assault. Are these deceptive parents who have maliciously designed a way of repeatedly injuring the bones of their child without leaving any telltale traces of injury to the skin? Or is the deception in our not appreciating some elusive biological cause of bone weakness in young infants that leads to these fractures? My own personal experience suggests the latter.29,30
Personal Experience
Since February, 1994, I have been referred infants with multiple,
unexplained fractures for clinical evaluation in which child abuse was suspected, but the parents and other caretakers denied intentional injury. In 1999 we described our initial experience in 33 such infants - 26 had a phenotype consistent with TBBD, 5 had evidence of child abuse, and 2 had osteogenesis imperfecta.29 The 26 infants with TBBD presented in a narrow window of time between 3-18 weeks of age. The criteria for the diagnosis of
TBBD was the following:
1) parents and caregivers denied wrongdoing
2) there were no apparent episodes of trauma to explain the fractures
3) there was no external skin injury such as bruising at the time of
presentation or at prior health care visits
4) evaluation revealed no other evidence of systemic findings that
would suggest child abuse – no retinal hemorrhages, no subdural hematomas,
and no visceral organ injury
5) radiographs show no evidence of metabolic bone disease such as rickets
6) laboratory studies evaluating for metabolic bone disease such as serum calcium and phosphorus and collagen analysis to evaluate for osteogenesis imperfecta are normal.
To obtain an objective measure of the bone strength of these infants, I measured bone density in TBBD infants using both radiographic absorptiometry and the highly sensitive method of peripheral, quantitative computed tomography. This was the first time formal bone density measurements were obtained in infants with multiple unexplained fractures, and the results were striking. When compared to controls of a similar age, the TBBD infants
had lower bone density.29 This observation was in spite of the fact that the plain radiographs of these infants with TBBD were interpreted by the radiologist as showing apparent normal bone density
The Hypothesis:
Fetal Immobilization Causes a Temporary Brittle Bone State
But if these infants with TBBD truly have a lower bone density, what is the basis of it? The clue came from the pregnancy histories.29 There was a striking association between TBBD and pregnancy histories of decreased fetal movement with intrauterine confinement. The causes of the intrauterine confinement included cephalopelvic disproportion, twinning, oligohydramnios, large maternal uterine fibroids, and maternal structural uterine anomalies.
There was also an increased frequency of deformations (structural birth defects of the musculoskeletal system such as clubfoot and dislocated hips) and short umbilical cords in the TBBD infants which is also consistent with decreased fetal movement and intrauterine confinement.15,28
The association of decreased fetal movement and intrauterine confinement with TBBD suggests a plausible explanation for TBBD, namely that intrauterine confinement with decreased fetal movement leads to reduced fetal bone loading with decreased fetal bone strength. Frost has recently proposed the mechanostat/mechanical loading theory of postnatal bone
formation which states that the primary determinant of bone strength is the load placed on the bone as shown in Figure 1.12,13 The load causes a strain in the bone which the bone is able to perceive through the mechanostat, and the bone responds to this by accordingly increasing or decreasing bone strength. The mechanostat is a feed-back system within the bone that can
receive the input of strain, evaluate and compare the strain to preset threshold values, and bring about appropriate signals to the effector cells, osteoblasts and osteoclasts. When there is an increased load placed on the bone, the mechanostat signals the biochemical machinery of the effector cells to increase bone strength. When there is a decreased load placed on the bone, the mechanostat signals the biochemical machinery of the effector
cells to decrease bone strength. Bone strength can be modified by changes in bone density and/or bone architecture. The cells that are signaled by the mechanostat to carry out the appropriate changes in bone density and architecture are the osteoblasts and osteoclasts.
There are two types of bone loading. The first is associated with the direct impact of bone against another object such as the increased load the leg bones realize during running. Individuals who participate in physical activities associated with impact such as soccer and running have greater bone strength than controls; whereas individuals who are placed in environments associated with decreased loading such as astronauts in the prolonged microgravity of space travel have decreased bone strength than
controls. The second is associated with the active and passive load the bone realizes from the muscles attached to it. The muscles that attach to a bone exert a small but continuing load on the bone even when the muscle is not actively moving the bone. Forceful muscle contractions promote bone strength. Weightlifters have greater bone strength than non-weightlifters and individuals with neuromuscular diseases such as Duchenne muscular
dystrophy or quadraplegics with longstanding immobilization have lower bone strength than controls.18
While the Utah paradigm has been applied to older children and adults, it also has relevance to the prenatal and immediate postnatal time periods. The application of this contemporary model of bone physiology to these time periods provides a simple and plausible explanation for TBBD, namely that fetal immobilization can lead to a transient state of bone brittleness in the early months of infancy. Several observations support this.
First, Rodriguez et al. has shown that infants with congenital neuromuscular disease in which there is both decreased fetal movement and decreased fetal muscle mass and function have diminished bone formation.44,45 Second, Rodriguez et al. have also described an experimental model of drug induced fetal immobilization in rats with tubocurarine which causes diminished bone formation and has been called the fetal akinesia deformation sequence.46 I have seen 3 infants with TBBD who were born to mothers who were taking similar drugs during the third trimester that cause decreased fetal movement.32 Noteworthy is that in the Rodriguez studies of both the human and rat, the fetal immobilization resulted in decreased bone diameter. In growing long bones most of the formation occurs on the outside of the bone (periosteum), while most of the resorption occurs on the inside of the bone (endosteum). Thus, the Rodriguez studies indicate that fetal immobilization causes reduced periosteal bone formation.
Third, the recently described MyoD/Myf5 double knockout mouse in which mice are bred who lack the important muscle proteins MyoD and Myf5 underscores the influence of fetal bone loading through fetal movement on intrauterine bone development.48 Compared to the wild type mouse, the MyoD/Myf5 double knockout mouse has: 1) smooth, featureless, and straight long bones, 2) thinner cortices by 19%, 3) decreased mineralization by 24%, and 4) absent ribs and medial parts of the clavicles. Osteoblast isolated from both the wildtype and the MyoD/Myf5 double knockout mouse were identical in their ability to respond to mechanical loading stimuli.
In the experimental animal and the human the response of bone to loading is dependent on the magnitude of the strain applied to the bone as well as the frequency of the strain (number of cycles/day).9,24 In the immature rat jump training with as little as 5 jumps/day for 8 weeks results in increased bone mass and strength in the tibia and femur.49 These observations suggest
that only a few peak loads/day are needed to promote bone formation in the long bones of growing animals.
If the above observations are applied to the fetus during intrauterine life, then it is apparent that there is considerable loading of the fetal skeleton through fetal movement. In normal human pregnancies fetal movement is consistently perceived by pregnant mothers beginning at about 16 weeks of gestation and is exuberant until term, the number of fetal movements perceived at 24-26 weeks gestation is about 54 movements/hour, and the fetus
is active some 9-18% of the time.43 Thus, there is a relatively long period of some 24 weeks of relatively high frequency bone loading of the fetal skeleton through movement against the resistance of amniotic fluid, kicking against the uterine wall, and from muscle contractions. This fetal activity is critical for normal fetal bone loading which leads to normal bone strength of the fetus and the newborn at the time of delivery. Attenuation of this fetal bone loading for whatever reason (intrauterine confinement, maternal use of medications during pregnancy that cause fetal
immobilization, or prematurity) will lead to a relatively weaker fetal and newborn skeleton that has a greater propensity to fracture with a given physical force.
The primary reason that TBBD has been overlooked as a distinct medical entity is that it is not associated with any measured biochemical abnormalities or radiographic abnormalities until there is significant lack of net gain in bone mass. This is similar to the pattern of biochemical and radiologic findings in disuse osteopenia.18
Revisiting the Radiologic Dogmas of Child Abuse
The plain radiograph has traditionally been used by radiologists and child abuse experts as a test which provides reliable information about bone strength. Bone strength is determined by bone density and bone architectural parameters. Typically, bone strength is given in terms of the load needed to cause the bone to fail (fracture). A failure load can be determined by standard mechanical equations for a bone loaded in compression, tension,
torsion, or bending. For example, the failure load for a bone of circular cross-section, loaded in bending (Bf), is given by the following equation:25
Bf = (I/Rp)sbs
Bf = failure load in bending
I = cross-sectional moment of inertia = (p/4)(Rp4 – Re4)
Rp = periosteal radius
Re = endosteal radius
sbs = bending ultimate stress coefficient (an empirically determined
constant of material property of the bone that is directly related to
bone density)
Bf = (p/4) (sbs) (Rp4 – Re4)/Rp
This equation indicate that the strength of a bone in bending is related to the bone density (through sbs), the periosteal radius, and the endosteal radius. However, the radiographic report on the standard set of skeletal X-rays taken for the purpose of a child abuse investigation for unexplained fractures typically provides none of these measurements. Bone density is directly related to sbs, but it is difficult for a plain radiograph to assess bone density, because of the insensitivity of the human eye to judge small changes in optical film density. Rp and Re can be measured in
radiographs directly; however, these measurements are typically not performed and reported. Moreover, small reductions in Rp that might be difficult to measure can have a great influence on Bf, because Bf is related to Rp raised to the fourth power. Yet, the diagnosis of child abuse is often made based on such normal appearing X-rays. Normal appearing X-rays do not necessarily mean normal strength bones, because there is typically no
objective evaluation of bone density or bone architectural parameters.
It is important to emphasize that the low bone density measurements found in TBBD infants had apparent normal bone density (whiteness) on the plain radiographs. For the eye to detect decreased bone whiteness on a plain radiograph, there must be greater than a 30% loss of bone density.8,22 Thus,
a 30% loss of bone density might not be appreciated as being abnormal on the plain radiograph, even though it would be associated with a marked increase in fracture risk.
If one accepts the fact that the plain radiograph is inadequate to detect the biomechanically induced osteopenia of TBBD, then it is understandable why infants with TBBD have the radiographic features thought to be highly specific for intentional injury - metaphyseal fractures, posterior rib fractures, and multiple fractures of different ages. These 3 findings could be expected from routine handling of an infant with brittle bones as a result of fetal/perinatal immobilization.
First, the metaphysis of the infant is the weakest structure of the
musculoskeletal system. It contains a large proportion of growing,
non-mineralized bone. The metaphysis is primarily trabecular bone, and the trabecular bone density of infants is significantly lower than that of older children or adults.33 The metaphysis will therefore be the first structure to yield if a twisting force is applied to an extremity of an infant. The twisting of an extremity in an infant who has normal-strength bones may require a force that would only be seen in nonaccidental injury. However, if the infant has an intrinsic bone disease with brittle bones, a metaphyseal
fracture of an extremity could result from routine handling of the infant such as in the changing of diapers or of clothing. Helfer et al. found metaphyseal fractures could be produced by passive range of motion exercises of the legs in 4 infants with presumed intrinsic bone disease, 3 of whom were premature.19 Grayev et al. described 8 infants with clubfeet who incurred metaphyseal fractures during physical therapy treatment of the clubfeet when their legs were passively manipulated and in which at least 3 had evidence of prenatal onset of decreased fetal movement.16 Clubfeet are most often caused by intrauterine confinement, and this observation indicates that the relatively insignificant forces associated with physical therapy may cause metaphyseal fractures in an at risk group of infants who experienced fetal immobilization.34 These combined observations indicate
that metaphyseal fractures are not pathognomonic of child abuse and decreased fetal movement and prematurity may predispose to such fractures.
Second, posterior rib fractures in child abuse are thought to result from thoracic compression in normal strength ribs which causes the rib to be levered over the fulcrum of its transverse process.21 However, if the ribs were weak from intrinsic bone disease and a routine compressive force was applied to the thoracic cage, such as in the normal picking up of the infant around the chest, a posterior rib fracture would also be the most likely type of rib fracture, since this is the biomechanically weakest segment of
the rib. Iatrogenic, posterior rib fractures have been described in large infants following vaginal delivery and in young infants after chest physiotherapy for respiratory disease.6,23
Third, the multiple fractures at different stages of healing also follows from this line of reasoning. Routine handling of the infant with biomechanically-induced osteopenia at multiple different times could produce any of a number of different types of fractures of various bones.
Physical Forces that Can Cause Fractures in TBBD
I believe fractures in an infant with TBBD might occur with physical forces from the routine handling of the infant. This includes the birth process, changing of clothes and diapers, picking the infant up around the chest, playing with the infant, using bicycle movements and exercises, and iatrogenic fractures from medical procedures/maneuvers such as performing a hip examination, holding for lumbar puncture and venipuncture, and performing physical therapy.19,16,34,23,6,16,31
No individual can reliably testify to the magnitude of forces needed to produce a fracture unless they know the strength of the bone. A plain radiograph cannot reliably assess bone strength. This is the flawed line of reasoning that has been used to incorrectly diagnose child abuse in some cases of infants with multiple unexplained fractures. In such cases a high reliance is usually placed on the radiologist’s interpretation that the bones have normal bone density, there is no evidence of metabolic bone
disease, and thus the only other reasonable explanation is that this is child abuse. The utilization of more objective bone density measurement technologies such as DEXA and peripheral quantitative computed tomography in the evaluation of infants with unexplained fractures can provide more meaningful information in trying to judge a cause of the fractures.42
However, even with these techniques there are limitations, if there are not appropriate age and size- matched controls.38
Prematurity and TBBD
Prematurity has a strong association with TBBD.39,29 It is well-known that premature infants are at increased risk to develop a temporary brittle bone state, and it has traditionally been thought that the primary cause of this was insufficient calcium and phosphate in the diet of the premature infant. However, there is emerging evidence that the bone disease of prematurity may
be more of a biomechanical issue than one of nutritional mineral
deficiency.35 First, premature infants fed formulas that contain higher concentrations of calcium and phosphate often have no increase in bone density.11 Second, there is increased bone resorption in the bone disease of prematurity which would indicate some other explanation for this condition, other than lack of mineral availability.4,36
It is suggested that this increased bone resorption in the markedly
premature infant compared to the term infant is secondary to decreased bone loading. During the last trimester the fetus is actively kicking and bouncing against the mother’s uterus. This fetal activity with associated muscle development is the primary determinant of fetal bone formation. The intrauterine loading of the fetal musculoskeletal system activates the mechanostat to increase bone strength by modulating bone formation and
resorption as shown in Figure 1. Fetal movement also promotes muscle growth which contributes to bone loading. Conversely, when an infant is born markedly premature, the infant is deprived of much of this musculoskeletal bone loading in utero. After birth, the markedly premature infant is often hypotonic and has a poverty of movements compared to the term infant.20
Thus, there is also postnatal modulation of the mechanostat to increase bone resorption in the very low birthweight (VLBW), premature infant compared to the term infant.
Moyer-Mileur et al. have shown that preterm infants who receive 5-10 minutes of daily physical activity realize a 76% greater gain in bone density by one month of life compared to control premature infants who receive no physical activity.37 This finding explains the observation that very sick premature infants who are on ventilators and require parenteral nutrition have a much greater frequency of bone fractures compared to premature infants who can take oral feeds and are not on ventilators.3,10 Healthy premature infants who are picked up and gently handled by caregivers for oral feedings and for nurture receive bone loading similar to what physical therapy might provide.
Noteworthy is that the pattern of fractures in VLBW, premature infants is similar to that in child abuse and in TBBD.3,10 Metaphyseal fractures, rib fractures, and fractures of various ages can be seen in the bone disease of prematurity. Because these infants are hospitalized in a newborn intensive care nursery (NICU), the diagnosis of child abuse is not made, but rather a
term is used to indicate an intrinsic bone disorder that has a biologic cause (bone disease of prematurity, rickets of prematurity, and others) and not a social one (child abuse and nonaccidental injury). However, if such an infant is discharged from the NICU and fractures are first noted following discharge from the NICU child abuse will certainly be considered. Such an infant would probably have the phenotype of TBBD. The fractures of the bone disease of prematurity usually occur at about 3 months of age, a similar time as when TBBD often presents.
The Transient Nature of TBBD The most perplexing issue of TBBD has been to explain its transient nature of fracture susceptibility during the first year of life. If prematurity and fetal immobilization related to intrauterine confinement are the primary determinants of TBBD, then it follows that the fracture susceptibility in the immediate postnatal period would be transient. Once the infant who was gestated in an environment of intrauterine confinement is born, the intrauterine confinement ceases, the mechanostat is activated for increasing bone strength, and bone strength begins to normalize. However, there is a window of time of approximately the first year of life, and especially in the first 6 months of life, when the infant’s bones are weaker and thus at increased risk to fracture. Likewise, the biomechanical effect of prematurity on bone formation is also transient as the premature infant begins to experience more bone loading later in the first year of life from
spontaneously moving and from being handled by caregivers.35
A markedly premature infant who was also confined, such as occurs in twinning, would thus have two significant factors which could decrease bone formation. Twins are over-represented in infants with multiple unexplained fractures who have a TBBD phenotype compared to their frequency in the general population.29,39
Evidence that TBBD is Not Child Abuse
Paterson has recently presented follow up of 96 infants who presented with multiple unexplained fractures and who were thought to have been abused but in whom he diagnosed TBBD.41 The future care of these infants was determined by civil proceedings and 65 infants were returned to their parents, 47 with conditions. Follow up of these 65 infants has been for a mean of 4.9 years (range 0-11 years). There has been no evidence of
subsequent child abuse in those infants returned to their parents. I have followed 3 such TBBD infants who were returned to their parents. Each child has been with their parents for about 5 years, and there has also been no evidence of child abuse in these 3 children.
Another compelling observation that TBBD is a distinct entity from child abuse is the conspicuous absence of significant bruising, swelling, or other severe internal injury. Physical trauma in most infants that causes a
fracture is strongly associated with bruising and other internal,
non-skeletal injury.27 When a child incurs a fracture caused by physical trauma, there is a probability that the traumatic event will also cause a bruise. Mathew et al. determined the frequency of bruising in children with fractures.26 The frequency of bruising was 8% at the time of presentation and 28% during the first week after the trauma. If there are multiple fractures, then each fracture is an independent event for showing bruising with a probability of 0.28 if observed during the first week after the fracture. Thus the likelihood that an infant who incurs a fracture will not have a bruise if observed during the week after the injury is .72. An infant
who incurs 15 fractures from alleged child abuse and has been evaluated during multiple health care visits would thus have a probability of (0.72)15 = .007 of not having a single bruise, a very unlikely scenario that should cause us to pause for alternative explanations. A more likely explanation for 15 fractures in the setting of no bruising and no internal injury is that there was no significant trauma to cause the fractures.
Garcia et al. found that in infants who have normal strength ribs and incurred multiple rib fractures from trauma (either from child abuse or from motor vehicle accidents), there is almost always internal thoracic injury.14 These combined observations about bruising and internal injury in the setting of rib fractures are indirect evidence for the existence of TBBD. When infants with multiple fractures have no other injuries, either bruising or internal injury, an alternative explanation of an intrinsic bone disorder
should be considered.
Finally, the presentation of infants with TBBD in the narrow time period which peaks at about 2 months of age is consistent with an etiology of fetal/perinatal onset. If TBBD were truly a guise for child abuse, it would be expected that there would be a similar frequency of TBBD in infants who were 6-24 months of age, and it is uncommon to find infants in this age group with TBBD.29
Table 1 compares characteristics of child abuse versus TBBD, and Table 2 summarizes the evidence that TBBD is not child abuse.
Conclusion
Except for individuals with rare genetic disorders associated with brittle bones such as osteogenesis imperfecta and for markedly premature infants who receive inadequate calcium and phosphate in their diets, it has been assumed that all neonates and infants are equal in their bone strength and propensity to fracture. This assumption is the foundation for concluding that infants with multiple unexplained fractures are victims of child abuse, until proven otherwise. However, the evidence presented herein suggests that all bones are not created equal” and that a previously unappreciated factor in understanding bone strength and fracture susceptibility in the infant is the prenatal and immediate postnatal bone loading on the skeleton. Situations which result in fetal immobilization such as intrauterine confinement and prenatal-onset neuromuscular diseases, and in relative immobilization in the immediate postnatal period such as in marked prematurity can produce a state of transient bone weakness.
Table 3 compares the 5 perinatal, brittle bone states which have been previously discussed: 1) the fetal akinesia deformation sequence in the rat (FADS), 2) the MyoD/Myf5 double knockout mouse (Null) 3) congenital neuromuscular diseases in the human (CNMD), 4) bone disease of prematurity in the human (BDP) and 5) temporary brittle bone disease in the human. Noteworthy is the similar types of fractures that can be seen in these conditions and the common underlying cause of fetal/perinatal immobilization
in all 5 conditions. FADS, Null, and CNMD are the most severe phenotypes of fetal immobilization, as they have evidence of osteopenia at birth by X-ray and histomorphometry. Infants with CNMD have fractures at birth, presumably from having intrauterine fractures or fractures from the delivery process. Infants with BDP or TBBD can have fractures at birth, but these more typically present in the first several months of life.
This hypothesis that fetal immobilization can lead to a transient state of bone fragility in the first months of life has been and will likely continue to be met with skepticism by the pediatric radiology and child abuse communities. However, when this hypothesis is considered in the context of our current knowledge of bone physiology as posited in the Utah paradigm, it is plausible and reconciles many of the observations about TBBD as presented
in Table 2.
If TBBD does exist, then the grim and sober reality of TBBD is that some parents have wrongly had their children taken from them, and some adults have been unjustly imprisoned for alleged acts of child abuse they did not commit. The time that these parents were forced to spend away from their children is irretrievable. Because the stakes in these cases are extraordinarily high with devastating, lifelong consequences, it is critical that those involved in child abuse allegations of infants with multiple unexplained fractures consider this hypothesis and test it with their own
experiences, both retrospectively and prospectively. The measurement of quantitative bone density and bone architecture parameters such as cortical bone width, periosteal diameter, and endosteal diameter, in which there are adequate age-matched controls for these measurements, may provide needed
information to further clarify this controversial issue.
REFERENCES
1. Ablin DS, Greenspan A, Reinhart M, and Grix A. Differentiation of child abuse from osteogenesis imperfecta. AJR 154:1035-1046; 1990.
2. Ablin DS and Sane SM. Non-accidental injury: confusion with temporary brittle bone disease and mild osteogenesis imperfecta. Pediatr Radiol 27:111-113; 1997.
3. Amir J, Katz K, Grunebaum M, Yosipovich Z, Wielunsky E. and Reisner H. Fractures in premature infants. J Pediatr Orthop 8:41-44; 1988.
4. Beyers N, Alheit B, Taljaard JF, Hall JM, Hough SF. High turnover
osteopenia in preterm infants. Bone 15:5-13;2; 1994.
5. Cadzow SP and Armstrong KL. Stressed parents with infants: reassessing physical abuse risk factors. Child Abuse and Neglect 23:845-853; 1999.
6. Chalumeau M, Foix-l’Helias, Scheinmann P, Zuani P, Gendrel D,
Ducon-le-Pointe. Rib fractures after chest physiotherapy for bronchiolitis or pneumonia in infants. Pediatr Radiol 32: 644-647; 2002.
7. Chapman S and Hall CM. Non-accidental injury or brittle bones. Pediatr Radiol 27:106-110; 1997.
8. Colbert C. The osseous system. Invest Rad 7: 223-232; 1972.
9. Cullen DM, Smith RT, and Akhter. Bone-loading response varies with strain magnitude and cycle number. J Applied Phsiol 91: 1971-1976; 2001.
10. Dabezies E and Warren PD. Fractures in very low birth weight infants with rickets. Clinical Orthopaedics and Related Research 335:233-239; 1997.
11. Farek J, Petersen S, Peitersen B, Michaelsen KF 2000 Diet and bone mineral content in term and premature infants. Pediatr Res 47:148-15; 2000.
12. Frost HM. Perspectives: A proposed general model of the mechanostat (suggestions from a new paradigm). Ana Rec 244:139-14; 1996.
13. Frost HM. Muscle, bone, and the Utah Paradigm: a 1999 overview. Med Sci Sports Exer 32: 911-917; 2000.
14. Garcia et al. Rib fractures in children: A marker of severe trauma. J Trauma 30:695-700; 1990.
15. Graham J. Smith’s Recognizable Patterns of Human Deformation. Second edition, WB Saunders, Philadelphia, 1988.
16. Grayev, AAM, Boal DKB, Wallach DW, and Segal LS. Metaphyseal fractures mimicking abuse during treatment for clubfoot. Pediatr Radiol 31:55-563; 2001.
17. Habert J, Haller JO. Iatrogenic vertebral body compression fractures in a premature infant caused by extreme flexion during positioning for a lumbar puncture. Pediatr Radiol 30:410-411; 2000.
18. Hangartner TH. Osteoporosis due to disue. Physical Medicine and Rehabilitation Clinics of North America 6:579-594; 1995.
19. Helfer RE, Scheurer SL, Alexander R, Reed J, and Slovis TL. Trauma to the bones of small infants from passive exercise: A factor in the etiology of child abuse. J Ped 104:47-50; 1984.
20. Kakebeeke TJ, von Siebenthal K Largo RH. Differences in movement quality at term among preterm and term infants. Biol. Neonate 71:367-378; 1997.
21. Kleinman PK. Diagnostic Imaging of Child Abuse. Second Edition. Mosby, St. Louis; 1998.
22. Lachman E. Osteoporosis: The potentialities and limitations of its roentgenological diagnosis. Am J Roentgenol 74:712-715; 1955.
23. Landman L, Homburg R, Sirota L, Dulizky F. Rib fractures as a cause of immediate neonatal tachypnea Eur J Pediatr 144:487-488; 1986.
24. Margulies JY, Simkin A, Leichter I, Bivas A, Steinberg R, Giladi M,
Stein M, Kashtan HMC. Effects of intense physical activity on the
bone-mineral content of the lower limbs of young adults. J Bone Joint Surg Am 68:1090-1093; 1986.
25. Martin RB, Burr DB, and Sharkey NA. in Skeletal Tissue Mechanics Springer, New York, pages 127-137; 1998.
26. Mathew MO, Ramamohan N, Bennet GC. Importance of bruising associated with paediatric fractures: prospective observational study. BMJ 317:1117-1118; 1998.
27. McMahon PM, Grossman W,Gaffney M, Stanitski C. Soft tissue injury as an indication of child abuse. Journal of Bone and Joint Surgery 77-A:1179-1183; 1995.
28. Miller ME, Higginbottom M and Smith DW. Short umbilical cord: Its origin and relevance. Pediatrics 67:618-621; 1981.
29. Miller ME and Hangartner TN. Temporary brittle bone disease: Association with decreased fetal movement and osteopenia. Calc Tissue Intl 64:137-143; 1999.
30. Miller ME. Temporary brittle bone disease: A real entity? Seminars in Perinatology 23:174-182; 1999.
31. Miller ME. Letter to the Editor Another perspective as to the cause of bone fractures in potential child abuse. Pediatr Radiol. 30:495-496; 2000.
32. Miller ME. Temporary brittle bone disease from intrauterine exposure to drugs that cause fetal immobilization. Calcif Tissue Int 70:359; 2002.
33. Miller ME and Hangartner TN. Cortical and trabecular bone density values in normal children measured by pQCT. Calcif Tissue Int 70:374; 2002.
34. Miller ME. Letter to the Editor. Fractures during physical therapy. Pediatric Radiol 32:536-37; 2002.
35. Miller ME. The bone disease of prematurity: A biomechanical perspective. Pediatr Res 53:10-15;2003.
DOI: 10.1203/01.PDR.0000039922.25000.17
36. Mora S, Weber G, Bellini A, Bianchi C, Chiumello G. Bone modeling alteration in premature infants. Arch Pediatr Adolesc Med 148:1215-1217; 1994.
37. Moyer-Mileur, Brunstetter V, McNaught TP, Gill G, Chan GM. Daily physical activity program increases bone mineralization and growth in preterm very low birth weight infants. Pediatrics 106:1088-1092; 2000.
38. Nelson DA and Koo WK. Interpretation of absorptiometric bone mass measuremetns in the growing skeleton: Issues and Limitations Calcif Tissue Intl 65:1-3; 1999.
39. Paterson CR, Burns J, and McAllion SJ. Osteogenesis imperfecta: The distinction from child abuse and the recognition of a variant form. Am J Med Gen 45:187-192; 1993.
40. Paterson CR and Burns J. Copper deficiency in infancy. J Clin Biochem Nutr 4:175-190; 1988.
41. Paterson CR and Monk EA. Long-term follow up of children thought to have had temporary brittle bone disease. Abstract. Presented at the First International Conference on Children’s Bone Health, 1999. Osteoporos Int 11(Supplement 4):S47-S48; 2000.
42. Rauch F and Schoenau E. Changes in bone density during childhood and adolescence: An approach based on bone’s biological organization. J Bone Miner Res 16:597-604; 2001.
43. Rayburn WF. Fetal body movement monitoring. Obstet and Gynecol clinics of North America. 17:95-109;1990.
44. Rodriguez JI, Palacios J, Garcia-Alix A, Pastor I, and Paniagua R.
Effects of immobilization on fetal bone development. A morphometric study in newborns with congenital neuromuscular diseases with Intrauterine onset. Calcif Tissue Int 43:335-339; 1988.
45. Rodriguez JI, Garcia-Alix A, Palacios, J, and Paniagua R. Changes in the long bones due to fetal immobility caused by neuromuscular disease. J. Bone and Joint Surg, 70-A:1052-1060; 1988.
46. Rodriguez JI, Palacios J, Ruiz A, Sanchez M, Alvarez I, Demiguel E. Morphological changes in long bone development in fetal akinesia deformation sequence: An experimental study in curarized rat fetuses. Teratology 45:213-221; 1992.
47. Shaw DG, Hall CM, and Carty H. [Letter to the Editor] Osteogenesis imperfecta: The distinction from child abuse and the recognition of a variant form. Am J Med Gen 56:110; 1995.
48. Skerry TM and Peet NM. MyoD/Myf5 null mice cannot move actively in utero and have thin weak long bones and no rib development. J Bone Miner Res 17:S170,1193; 2002.
49. Umemura Y, Ishiko T, Yamauchi Y, Kuronom, and Mashiko S. Five jumps per day increase bone mass and breaking force in rats. J Bone Miner Res 12:1480-1486; 1997.
50. Wynne J and Hobbs C (commentary) and Carty H (commentary) in response to article by Smith, R “Osteogenesis imperfecta, non-accidental injury, and temporary brittle bone disease.” Arch Dis Child 72:171-1764 ;1995.
Acknowledgments
The author is grateful to Dr. Thomas N. Hangartner and Shelley Miller for their suggestions and critical review of this manuscript and to the Childrens Medical Center Research Foundation.
Legend
Figure 1 illustrates the biomechanical model of bone formation which has been put forth by Frost known as the Utah Paradigm.12,13 While this model recognizes that genetic, nutrient, and hormonal factors are important in bone dynamics, this model presumes that in most situations biomechanical factors are the primary determinant of bone strength. The underlying principle of the Utah Paradigm is that bone has evolved an adaptive,
regulatory, feedback system to keep the strength of the bone in line with the loads placed on the bone. The centerpiece of this model is the mechanostat which resides within the bone and acts as the functional regulator of bone strength. Loads placed on the bone, no matter how small, cause a strain in the bone. Strain is the proportional change in length of the bone caused by a particular load, and its units are thus dimensionless. The mechanostat perceives an input signal of strain and compares it with preset values of strain threshold (ST) for either increasing bone strength (ST-IN) or decreasing bone strength (ST-DE). Bone strength is related to both bone density and bone architecture. If these STs are exceeded, then the mechanostat sends an output signal to the effector cells, osteoblasts and osteoclasts, to respond by appropriately modifying bone density and/or bone architecture, so that the strength of the bone is in line with the load on the bone.
If the bone is placed in a new environment where the load on the bone causes Strain > ST-IN bone, then the mechanostat will send an output signal to the effector cells to increase bone strength. If the bone is placed in a new environment where the load on the bone causes Strain < ST-DE bone, then the mechanostat will send an output signal to the effector cells to decrease bone strength. Bone strength remains unchanged if the load on the bone is
unchanged, or if a new load causes a strain in the bone that does not exceed the STs.
The Utah Paradigm has direct application to bone growth in the fetus and young infant. Bone loading is far greater in the fetus that experiences normal movement compared to the fetus in which movement is compromised and decreased. Normal fetal movement requires an intact neuromuscular system in which the intrauterine environment has normal space availability that allows for normal kicking and movement. Situations that can lead to decreased
fetal movement include intrauterine confinement, maternal use of medications that cause fetal immobilization, prematurity, and neuromuscular disorders that have a prenatal onset.
In growing bone such as occurs in the fetus and young infant, bone loading increases bone strength, in part, through an increase in periosteal thickness as a result of increased osteoblast activity in the periosteum. Bone strength may also be increased by increasing bone density.