Search

We have evaluated the applicability of a new Digital X-ray Radiogrammetry (DXR) system in a Dutch Caucasian pediatric population. For this study we enrolled 535 healthy participants who all signed an informed consent form. In addition, 20 children suffering from inflammatory bowel disease (IBD) and juvenile chronic arthritis (JCA) were enrolled. Radiographs of the left hand were obtained from all participants. From the healthy population a subset of children with a history of forearm fractures were separately analyzed. Measurements consisted of DXR (X-posure; Pronosco-Sectra, Linköping, Sweden). Five hundred thirty-five subjects were enrolled in the study. Twenty-two subjects (4.3%) were discontinued (age 3-10 years), all because of a nonrecognizable radiograph by the DXR system. The short-term coefficient of variation of DXR in this population was 0.59%. Significant differences in DXR-BMD between boys and girls for the ages of 11, 12, 16, 17, and 18 years were found. There were also significant differences in DXR-BMD between the sequential Tanner stages. For 88 subjects repeat radiographs were available (mean interval 1.8 years). In all cases an increase in DXR-BMD was seen. Girls with IBD, JCA, or a history of forearm fractures and boys with IBD showed a significantly lower DXR-BMD compared with healthy controls. We show that DXR is an applicable technique in children. Also, in a small subpopulation it is possible to discriminate children with a high risk of low BMD.

A large proportion of the variation in bone mass can be explained by genetic factors. We analyzed the G to T substitution in the Sp1 binding site in the first intron of the collagen type Ialpha1 (COLIA1) gene in relation to bone mass. The genotypes GG, GT, and TT were determined in 148 Caucasian children and young adults. We performed dual energy X-ray absorptiometry twice (mean follow-up time 4.4 years), and speed of sound (SOS) was assessed by tibial ultrasonometry at follow-up. Genotype distribution was 104 (70%) GG, 40 (27%) GT and 4 (3%) TT. Carriers of the T-allele had a 0.5 SDS (standard deviation score) decreased bone mineral content (BMC) of total body (P = 0.001), and a 0.4 SDS decreased bone mineral density (BMD) for both lumbar spine (P = 0.04) and total body (P = 0.05). The genotype effect on BMD and BMC decreased after adjustment for height or body mass index. When we calculated apparent BMD, these differences diminished to 0.1 SDS and were no longer significant. T-allele carriers had shorter stature (0.4 SDS; P = 0.04) and smaller bones (0.5 SDS lower width of the lumbar vertebral body; P = 0.01). The T-allele was also associated with lower SOS (P = 0.03), independent of BMD and BMC, and lower lean body mass. Similar associations were found at follow-up. The change in BMD and BMC SDS between the first and second measurement did not differ between the GG and GT&TT group. In conclusion, the COLIA1 polymorphism in children and young adults is associated with several bone characteristics. However, at least a part of the COLIA1 effect on bone mass may be related to differences in frame size.

There is a need for a reliable bone assessment technique in children. In this study, we compare an existing technique used in children, radiographic absorptiometry (RA), with a relatively novel technique, quantitative tibial ultrasonometry (QUS). In a prospective cohort study, we enrolled 290 girls (mean age 12.7 years) and 273 boys (mean age 12.4 years). Radiographs of the left hand and the left index finger were taken with an aluminium reference wedge within the field of exposure. Radiographic absorptiometry on the second middle phalanx at the mid-level (BMD50%) and proximal quarter (BMD25%) was performed with interactive software. Tibial QUS was performed using the SoundScan Compact. Multiple regression analysis showed that SOS correlated significantly with BMD25% for both boys (r = 0.65, P < 0.001) and girls (r = 0.59, P < 0.001), taking into account age and gender. The same applied for the correlation between speed of sound (SOS) and BMD50% in boys (r = 0.62, P < 0.001) and girls (r = 0.67, P < 0.001). Cubic regression between calendar age and BMD25% showed the best fit for both boys (r2 = 0.60) and girls (r2 = 0.60). For BMD50% a difference in regression was found between boys and girls. Quadratic regression gave a satisfactory fit for boys (r2 = 0.61 ) whereas for girls, a cubic relation was best (r2 = 0.59). Overall, there was a significant correlation between BMD25% and BMD50% for boys r = 0.89 and for girls r = 0.91 (both P < 0.001). Our data show a significant correlation between two different bone assessment techniques. In addition, these data suggest that both tibial ultrasonometry and RA are useful techniques in children.

Bone densitometry in children is a relatively new topic of interest within the field of osteoporosis. Bone densitometry techniques using an X-ray source have the disadvantage of radiation exposure. Also on some systems, motion artifacts are caused by long scan times. Tibial quantitative ultrasonometry (QUS) is ideally suited for children as it is radiation free and the interactive measurement provides real-time quality control. In this prospective study, we present data from 596 healthy children-309 girls, mean age 12.9 years (range 6.1-19.9), and 287 boys, mean age 12.3 years (range 6.1-19.6) from Rotterdam, The Netherlands. For all subjects, a short questionnaire regarding overall health was completed. To assess skeletal age, an X-ray of the left hand was taken and tibial QUS of the right tibia was performed using the SoundScan Compact. A statistically significant correlation was found between age and speed of sound (SOS)-r(2)(boys) = 0.52 and r(2)(girls) = 0.63 (both P < 0.001) and between skeletal age and SOS-r(2)(boys) = 0.56 and r(2)(girls) = 0. 63 (both P < 0.001). In boys, significant increase of mean SOS is seen between Tanner stages II and III and between IV and V. In girls there is a significant increase of mean SOS among all Tanner stages, except stages II and III. This is the first study to present normative tibial QUS data for Caucasian children and adolescents. In this study, normative data relative to skeletal age are also provided, facilitating the implementation of this technique in children with growth disorders showing dissociation between calendar and skeletal age.

Tibial quantitative ultrasonometry is a relatively novel technique in the field of bone sonometry, an emerging alternative to bone densitometry. The implementation of this technique in a pediatric population could prove valuable from a clinical as well as a research viewpoint. In clinical practice it is necessary to know the precision of this technique and the possible influence on measurements before implementation. This study presents the precision in a Caucasian pediatric population and the influence of measurement site, dexterity, brand of coupling gel, and temperature of coupling gel. To assess intra- and interobserver variance duplicate measurements, with repositioning, ultrasonometry was performed in 10 children over a short period of time. The observers were blinded for the results of the other observer and after each measurement the skin markings were removed. Intraobserver variance for operator one (MHL) and for observer two (SFGR) was CV 0.43%. The interobserver variance was CV 0.61%. Left midtibial and right midtibial speed of sound (SOS) measurements showed no significant differences. There were, however, significant differences in both boys and girls between right proximal versus right midtibial, right midtibial versus right distal, and right proximal versus right distal (for all P < 0.001). One-way analysis of variance (ANOVA) showed that neither the use of different coupling gels nor an increase in gel temperature had a significant influence on measurements. The results of our study show that tibial quantitative ultrasonography (QUS) is a highly reproducible technique in a Caucasian pediatric population.