Vitamin D deficiency versus non-accidental trauma: comment on “Rickets or abuse? A histologic comparison of rickets and child abuse-related fractures”

Abstract

We read with great interest the study by Kepron and Pollanen [1] concerning a histologic comparison of rickets and child abuse-related fractures. Recently, numerous studies have attempted to find a reliable method of differentiating vitamin D deficiency from traumatic causes of fracture in young children, with some studies concluding that vitamin D insufficiency was not associated with multiple fractures or the diagnosis of child abuse [2], and that fractures seen in overt rickets do not resemble non-accidental trauma [3]. Despite this, determining the underlying cause of fractures in young children is still a clinical dilemma, as the metaphyseal abnormalities of rachitic bones can be difficult to distinguish from classic metaphyseal lesions, which are highly specific for non-accidental trauma. Moreover, both diagnoses can co-exist. We recently studied the relationship between vitamin D deficiency and cause of fracture in young children. After IRB approval, we retrospectively reviewed all skeletal surveys performed over the last 10 years at our center. Of the 396 skeletal surveys performed, 99 were read to have possible non-accidental trauma. 25-hydroxyvitamin D levels were sent in 11 of these 99 cases, all in children under 1 year of age (Table 1). The prevalence of vitamin D deficiency was 12.5 % in the children with no clinical suspicion of vitamin D deficiency (95 % binomial CI 0.003–0.524, 1 of 8 cases), and was likely not related to the underlying cause of fracture in that single case. Three of the children were highly suspected of having vitamin D deficiency from their clinical course (prolonged NICU stay, or documented rickets of prematurity). Thus it appears that routine screening of vitamin D levels after suspected non-accidental trauma with fracture is not routinely useful in children for whom there is low clinical suspicion of vitamin D deficiency, although on occasion there is overlap in the diagnoses (1 of our 8 cases). While not currently accessible in the routine hospital setting, a histologic comparison of bone changes, as performed by Kepron and Pollanen [1], could be of great value in helping clinicians determine the underlying cause of fracture in infants and children.