Vertebral fractures assessment by morphometric X-ray absorptiometry (MXA) in children with Osteogenesis Imperfecta (OI)

Recurrent vertebral fractures are frequently seen in children with OI with resulting progressive deformity, loss of mobility and chronic bone pain. The purpose of this study was to evaluate the accuracy of MXA in comparison to visual semiquantitative (SQ) approach to identification of vertebral fractures in children with OI types I III, IV.. Subjects and methods: on 30 children (15 females) aged from 1-18 years , were acquired , on the same day, lateral images of the spine by conventional x-ray tube and by dual energy X-ray absorptiometry (DXA) using a QDR-4500 A densitometer (Hologic, Inc., Bedford, MA, U.S.A.). On spinal radiographs one pediatric radiologist performed visual SQ assessment to classify the type of vertebral fractures, as wedge, endplate, or compression, by severity of vertebral heights reduction , as mild (20-25%), moderate (25-40%) or severe (>40%). The SQ reading was performed without direct vertebral measurements and without knowledge of the morphometric results. On spinal DXA scans one skeletal radiologist performed the vertebral morphometry using a specified software to measure the vertebral bodies heights from T4 to L4. Six points were used to describe the shape of each vertebral body and from these points three vertebral heights were measured: anterior (Ha), middle (Hm), and posterior (Hp). The vertebral heights ratios identified three types of vertebral fractures: ha/hp (wedge) hm/hp (concavity) and hp/hpp (compression) if falled below the fracture threshold of 20%. Results: SQ and MXA agreed that 17 patients (58%) were fractured: 8 of them had fractures of grade 1, 5 of grade 2 and 4 of grade 3 ; SQ identified as fractured 137 (35%) of the 390 vertebrae analyzed; by MXA 23 vertebrae (5.9%) of four children were not visualized sufficiently for analysis , 131 (34%) of the 367 vertebrae analyzed were fractured. Conclusion: MXA disagree with SQ visual radiographic method particularly to identify mild deformities in the upper thoracic spine that sometimes is not sufficiently visualized on MXA images. Nevertheless this disadvantage of MXA has poor clinical relevance because of the lower prevalence of osteoporotic vertebral fractures in levels T4-T6. In fact MXA identified in this study correctly all fractured subjects. We conclude that MXA is a potentially useful, low-radiation technique in the monitoring of vertebral fractures in osteogenesis imperfecta.