A population study of fractures: what we can learn and what we cannot learn
Article first published online: 29 JUN 2013
© 2013 Mac Keith Press
Developmental Medicine & Child Neurology
Volume 55, Issue 9, pages 779–780, September 2013
How to Cite
Henderson, R. (2013), A population study of fractures: what we can learn and what we cannot learn. Developmental Medicine & Child Neurology, 55: 779–780. doi: 10.1111/dmcn.12173
- Issue published online: 8 AUG 2013
- Article first published online: 29 JUN 2013
It has been well established that children at the severe end of the cerebral palsy (CP) spectrum have skeletal fragility. Fractures in these children differ from those in typically developing children with regards to the usual mechanisms of injury and the most common sites of fracture. Clearly the bone is different, and different in multiple ways that render it more fragile: smaller size, thinner outer cortex, and diminished density.
What has not been as well studied is fracture history in children at the milder end of the CP spectrum. Uddenfeldt Wort et al. shed some light on this with their observational study of fracture events in a heterogeneous group of children with CP. It is encouraging that they found fracture rate, mechanisms, and locations to be similar to typically developing children. Behaviors and activities (and luck!) are the primary determinates of whether the bones of typically developing children or children with mild CP will fracture, not abnormal properties of their bone.
With the reassurance of reasonable bone health in more mildly affected children, focus returns towards skeletal fragility in the profoundly involved children. And here Uddenfeldt Wort et al. use observational associations in an effort to better understand the underlying pathophysiology, and go even further in recommending possible therapeutic interventions. In observational studies such as this and other similar studies, any factor which correlates with ‘severity’ of CP will be associated with increased risk of fracture due to less mobility, more surgeries/castings, seizure disorder, need for a gastrostomy, and diminished growth. It is tempting to assume direct cause and effect links among these factors, particularly when the association is consistent with what we think we know. Children with CP with a gastrostomy are more likely to have seizure disorder and atraumatic fractures than children with CP without a gastrostomy. Does a gastric tube contribute to seizure disorder or fractures?
Obviously, gastric tubes do not cause fractures or seizures, but aren't the bones of children with CP small because of malnutrition and osteopenic because of antiepileptic drugs and non-weight bearing? Certainly those statistical associations indicate direct cause and effect links. Or so we are quick to assume.
Even dramatic increases in weight with a gastric tube did not increase bone growth or density z-scores, and upper extremity bone size and density are both significantly less in the involved limb of children with hemiplegia.[3, 4] Given these findings, can we really be so sure that nutrition, or lack thereof, truly is a primary direct cause of skeletal fragility in children with quadriplegia? Also, there is increasing evidence that the newer anticonvulsant medications may not have an adverse effect on bone metabolism, so maybe this too is not the direct critical factor so often assumed.
In the study of Uddenfeldt Wort et al. the 27 children with quadriplegia who did not use a stander had a higher fracture rate than the 125 who did use one. So did the 27 have fractures that could have been prevented with a stander, or were these patients more severely involved and had contractures that prevented them from being in a stander? Stiff, flexed knees are one reason fractures in this population occur most commonly in the distal femur. By far the best quality study of the benefits of a standing program in this population found no increase in tibial bone density and concluded ‘such an intervention is unlikely to reduce the risk of lower limb fractures.
The fact is, we do not yet fully understand why many 14-year-olds with quadriplegia are no bigger than a 5-year-old and one consequence of this phenomenon is skeletal fragility. Overextending the conclusions one draws from simple observational studies, even if it ‘fits’ what we think we know, does not truly further our understanding of how to prevent or treat this problem.