Text

theses

Body composition (BC) assessments are used to monitor athletes’ overall health and track changes over time. These assessments are preferable compared to body weight alone or body mass index (BMI) because they provide additional information in regards to specific tissue quantities. However, the most accurate tools, such as air displacement plethysmography (ADP), are expensive and not feasible for assessing large groups of athletes, and field tools, such as bioelectrical impedance analysis (BIA) lack adequate validity and reliability needed to accurately track changes over time. Ultrasound (US) is proposed as a valid laboratory method that can also be used in field settings, as it is relatively inexpensive and portable. PURPOSE: To assess the validity of B-mode US (B-US) compared to ADP to determine BC in high-level adolescent and young-adult ballet dancers. A second study (Study 2) was conducted to further assess validity of different devices by comparing B-US, A-mode US (A-US), and BIA to ADP, as well as to a 3-compartment (3C) model. METHODS: Adolescent and young-adult vocational ballet dancers (N=48; Mage= 16.6 ± 1.6; MBMI=19.3 ±1.9) had their BC assessed by ADP and B-US. A subset of subjects (n=22) participated in Study 2, where BC was assessed via ADP, B-US, A-US, and BIA. Additionally, body density via ADP and total body water via BIA were used to calculate BC from Siri’s 3C equation for a reference measure. Pearson’s correlations were used to determine relationships between reference and experimental methods. Dependent t-tests were used to determine significant differences between methods for %BF, FFM, and FM. Bland-Altman plots were used to assess mean differences and identify the 95% LOA. Significance was set at P<0.05 for all measures. RESULTS: The primary study showed significant, strong correlations between B-US and ADP for %BF (females r=0.941; males r=0.773), FM (females r=0.943; males r=0.726), and FFM (females r=0.954; males r=0.985). Despite strong correlations among females, B-US significantly overestimated %BF and FM, and significantly underestimated FFM (P<0.05) compared to ADP. However, there were no differences between devices for any measure in males. In Study 2, there were no significant differences between devices for all measures in both males and females. There were strong correlations between ADP and both US devices, but lower correlations between ADP and BIA compared to US. Additionally, there were higher correlations between ADP and B-US than with A-US for %BF and FM in both males and females. As seen in the first study, FFM correlations were similarly high for both sexes across all devices. When comparing B-US and A-US to the 3C model, there were higher correlations for B-US than A-US for %BF in both sexes, and for FM in females only. Additionally, there were significant differences between A-US and 3C %BF, FM, and FFM results in males, but not in females. As seen in the previous analyses, there were higher correlations for FFM than there were for both %BF and FM for all methods in both males and females. CONCLUSION: B-US and A-US are valid BC assessment tools for adolescent and young-adult ballet dancers. It is possible that the combination of small errors in both FM and FFM measures resulted in larger %BF error, indicting these tools may be better suited for determining absolute FFM and FM rather than %BF. Overall, there was better agreement between the 3C model and both US devices in females compared to males, but there were higher correlations between 3C and B-US than 3C and A-US for both sexes. Future research should explore using a 4C reference method, as well as the use of different density models and prediction equations to improve %BF accuracy when using US.

ETD_10214

M.S.

Includes bibliographical references

doi:10.7282/t3-ekp7-j591

ETD graduate

The author owns the copyright to this work.