Author

Irina Shklyar

Date of Award

January 2014

Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)

Department

Medicine

First Advisor

Seward Rutkove

Second Advisor

Jonathan Goldstein

Subject Area(s)

Neurosciences, Medical imaging and radiology, Medicine

Abstract

Electrical impedance myography (EIM) and quantitative ultrasound (QUS) represent two rapid, non-invasive, quantitative, and child-friendly approaches to measure disease outcomes in clinical trials that

may surpass functional tests currently in use. In this thesis, we present a comparison of EIM and QUS data in six muscles from 25 DMD and 25 healthy subjects, aged 2-14 years. Quantitative ultrasound data can be

analyzed by measuring the brightness, or grayscale level (GSL), of the ultrasound image. To circumvent post-processing variability across manufacturers, for the first time, we introduce a new means of evaluating QUS--a quantitative backscatter analysis (QBA) on the raw radio frequency (RF) signal received by the ultrasound probe in DMD and control subjects. For EIM, either the phase at 50 kHz or the phase ratio (50 kHz/200 kHz) was used for measuring disease status. We found that taking the phase ratio substantially reduces the confounding effect of subcutaneous fat thickness on EIM (r=0.16, p=0.45) compared with the standard phase at 50 kHz (r=-0.72, p<0.0001) in averaged muscles. Using age as a surrogate for disease severity, we found that 50 kHz phase decreased with age in select muscles while GSL did not. The averaged phase correlated strongly with the North Star Ambulatory Assessment (NSAA) (rho=0.735, p=0.004), while GSL did not (rho=-0.442, p=0.115). Although the standard phase at 50 kHz averaged

values did not correlate with GSL, the phase ratio did (rho=-0.48, p=0.017). In DMD, age showed comparable correlation with "superficial" region of interest (ROI) QBA and GSL (average: rho=0.67,

p=0.0004 vs. rho=0.47, p=0.020), in contrast to "whole muscle" ROI, and the difference in rho values did not reach significance (average: z=1.00, p=0.16). Here, we show that "whole muscle" GSL detects disease pathology from an early age and remains stable over time; in contrast, DMD and control subjects begin with relatively similar EIM values and diverge with age. The moderate correlation between averaged EIM and QUS parameters indicate that the measures provide both overlapping and divergent information about dystrophic muscle. Our data confirm that EIM and QUS show promise as outcome measures that may aid the rapid identification of successful therapeutics currently in clinical pipelines.

Comments

This is an Open Access Thesis.

Share

COinS