This study shows that US is a valid and reliable tool to measure the diameter of the RF in stable, elderly CAD patients. It further shows that the diameter of RF, measured with US, is significantly correlated with different muscle strength parameters.
We found a high ICC for both test-retest of the measurement of the RF with US (0.97) as for the comparison of RF measurements with US and CT (0.92) in these older cardiac patients. This is in line with the study of Seymour et al.  who reported in COPD patients an ICC of 0.97 for test-retest reliability of US measurement of rectus femoris CSA and of 0.88 for validity of the US measurement of rectus femoris CSA compared with CT. Also Bemben et al.  and Kanehisa et al.  reported an ICC of 0.72 and 0.99 respectively for test-retest reliability of the CSA measurements using B-mode US technique in various age groups. Bemben et al. tested US and MRI reliability for muscle CSA of the RF at 15 cm above the patella and found no significant differences between both techniques in young subjects (age: 26 years). Similar results were found for the reliability of the US measurement of the vastus lateralis muscle (ICC between 0.997 and 0.999) and for the validity compared with MRI scans (ICC between 0.998 and 0.999) .
All patients included in this study were CAD patients without cardiac event during the last 9 months, who participated in at least one session of exercise training per week under supervision of a physiotherapist, and could therefore be considered to be still fairly active elderly.
The observed diameter of RF in our sample of cardiac patients is comparable to earlier findings. That is, Delaney et al.  showed a RF depth of 2.3 cm in resting position in healthy young males (mean age 24.6 years). Arts et al.  found a quadriceps diameter (thickness of rectus femoris + vastus intermedius) in males of 4.16 ± 1.02 cm (age range 17-90 years) whereas Nogueira et al.  found a RF diameter of 1.86 cm in 20 older men (age 69-76 years).
In addition we investigated the relation between muscle diameter, muscle force and muscle volume in this cohort. We found strong correlations (0.61-0.75) between muscle volume and diameter assessed with CT and all the muscle strength parameters. Muscle diameter assessed with US also significantly correlated (0.45-0.61) with all strength measures, although the correlations where somewhat less as compared to CT. Muscle volume of the mid-thigh region correlated significantly (r = 0.67, p = 0.001) with RF diameter (CT-technique), which was comparable to the results reported by Seymour et al.  in a healthy control group. RF diameter measured with US also significantly correlated with muscle volume (r = 0.49, p < 0.05). The reason for the less strong correlations when RF diameter was assessed by the US technique could be due to the higher variability of consecutive measurements in US (compression of the muscle tissue, deviation from perpendicular viewing). The more accurate determination of the middle of the femur using CT (visualization of the bone structure by means of the scout view) compared with surface determination of bony landmarks in US could also be an important factor.
Earlier, the US technique has shown to be a valid and reliable alternative to CT or MRI in studies comparing muscle RF CSA of COPD patients with healthy controls  and in a study to determine the effects of resistance training on muscle thickness or muscle volume in older men . In the latter study, an increase of 0.21 cm was found in the high velocity power training group, which is comparable with the MDD of 0.24 cm we found for test-retest reliability.