A cross-sectional study design was used to compare bone strength in CPT children with NF1 and normal children without bone metabolic disease. The study was conducted at Hunan Children's Hospital, and institutional review board approval was obtained for this study. The guidelines of the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) were followed to report the cross-sectional study [17].
2.1. Participants
Thirty-seven children aged 3 to 10 years with a diagnosis of CPT who had not received any previous treatment and 40 children without metabolic bone disease were included in this study. Participants were excluded if they had any known bone disease, other chronic diseases, previous or current treatment that might affect bone metabolism (eg, celiac disease, thyroid disorder, systemic glucocorticoids), or if the child was unable to co-operate with the study protocol. All participants had provided informed consent from their parents.
2.2. Data collection
All data are collected by the same researcher (S.X.), ensuring that all measurements are taken in the same way. General information was collected, including name, age, sex, and body mass index (BMI). The ultrasonic bone densitometer (Pegasus Smart Medilink, French) was used to examine the bilateral calcaneus of the subjects. The probe was placed correctly and should be facing the heel to avoid errors in the measurement results. The measured data are automatically analyzed by the computer system. According to the measurement results, we can get the BUA, SOS values, and then referred to the methods to calculate the QUI, STI and BMDe values [18, 19]. To verify the reliability of the measurement, twenty subjects were randomly selected from the CPT and control groups and measured repeatedly by two authors (GY and SX) with an interval of 1 months to examine inter- and intra-rater agreements.
2.3. Observation index
Broadband ultrasound attenuation (BUA): ultrasound waves will be attenuated when passing through an elastic medium. During the measurement process, the transmitting probe of the bone densitometer emits ultrasound signals, which are received and amplified by the receiving probe after passing through the bone sample to obtain the corresponding transmitted signals. The transmitted signal is processed accordingly to obtain the attenuation curve. The attenuation is linearly related to the frequency. The slope of the attenuation curve is BUA. The thicker the bone, the greater the bone density, the greater the attenuation of the signal at the same frequency, and the greater the corresponding slope. The BUA value is related to the average density, size, spacing, orientation and scattering intensity of the bone trabeculae. Due to low bone calcium density, patients with osteoporosis have lower BUA values than normal individuals.
Speed of sound (SOS): the ultrasound emission time and signal reception time were recorded and the SOS value was obtained based on the width of the sample. It is an indicator to evaluate bone quantity and quality, bone condition and bone strength, and is proportional to bone volume.
Bone strength index (STI): according to the obtained results, STI = 0.67 × BUA + 0.28 × SOS - 420, which is an index obtained from the linear combination of BUA and SOS and mainly reflects the stiffness and rigidity of bone. It is a linear combination of BUA and SOS. Several studies have shown that it reflects both the mass and structural properties of cancellous bone. It has a better correlation and higher accuracy with bone mineral density. The index predicts the risk of osteoporotic fracture better than ultrasound velocity or attenuation parameters alone.
Quantitative ultrasound index (QUI): Statistical methods based on previous descriptions[18], QUI = 0.41 × (BUA + SOS) -571.
Bone mineral density estimation (BMDe) formula: Refer to Magkos' method[19] to convert BUA and SOS values into BMD values as follows: BMDe = 0.0025926 × (BUA + SOS) -3.687.
2.4. Statistical Analysis
All data were expressed as mean ± standard deviation and processed by STATA (version 13.0, Stata Corp LP, TX, USA). The intraclass correlation coefficient (ICC) was calculated to estimate intra- and inter-rater agreement. According to Landis' definition, an ICC of 0.6-0.8 was considered good agreement and greater than 0.8 was excellent agreement. Paired t-tests were used to determine within-group differences for various quantitative measures of quantitative ultrasound. For between-group analysis, t-tests were used to determine differences in various quantitative measures. Multivariate liner regression was used to investigate the relationship between differences in quantitative ultrasound measurements and age, BMI, NF1, and CPT Crawford type. P<0.05 was considered statistically significant.