In the present study the legs without apparent symptoms

In the present study, the legs without apparent symptoms were all considered stage 0 because all patients who had undergone intrapelvic lymph node dissection and/or inflammation were considered to have more or less impaired lymph transport. In two patients with unilateral lymphedema, the cause was unclear so the contralateral asymptomatic leg was excluded from the study. Those with early and later stage II LE were separately assessed because the latter is supposedly affected by increasingly severe fibrosis. Images of legs representative of each ISL stage and LDS are provided in Figure 1.
Free-hand RTE with an ultrasound machine (HI VISION Preirus, Hitachi Aloka Medical, Ltd., Tokyo, Japan) was performed, as reported previously (Suehiro et al. 2014). Briefly, with the patient lying in the supine position, 7-mm-thick phantoms (Sonar Pad, Nippon BXI, Tokyo, Japan), which were each trimmed to 60 × 60 mm square, were placed on the skin at the middle of the inner thigh and the middle of the inner calf. The peak force of repeated rhythmic pkc inhibitor using an ultrasound probe was controlled to maintain phantom strains at 0.23%–0.47%, where the skin and subcutaneous tissue in normal legs could be optimally assessed. The region of interest (ROI) was set in the middle of the elastography window for all measurements. In the currently employed ultrasound machine, the ROI was automatically set as a circle. To monitor phantom strain, the ROI was set to include its full thickness. For measurement of skin strain, the ROI was set between the inferior margin of the entry echo and the dermal–hypodermal junction. For subcutaneous tissue, the ROI was set between the dermal–hypodermal junction and the superior margin of the deep muscular fascia. When the subcutaneous tissue was very thick, the lower margin was set as the lower limit of the elastography window.
After RTE, a B-mode scan of the skin and subcutaneous tissue was performed at the same points as the RTE using an ultrasound system (LOGIQ S6, GE Healthcare, Little Chalfont, Buckinghamshire, UK) with an 8- to 12-MHz linear transducer. Subcutaneous echogenicity (SEG) and echo-free space (SEFS) were graded respectively as previously described (Suehiro et al. 2013, 2014).

Results
Skin strain and subcutaneous tissue strain in the thigh and calf of patients with lymphedema, with respect to ISL stage, and in the thigh and calf of patients with LDS are illustrated in Figure 2. In the thigh, no significant differences in strain among ISL stages and LDS were observed either in subcutaneous tissue or in skin. In the subcutaneous tissue in the calf, again no significant differences in strain were observed among ISL stages, but the strain in LDS was significantly lower than that in ISL stages 0, II and late II. In calf skin, there was a significant decrease in strain in stage III compared with stages I and II, although there were no differences between stage I or II and stage 0. The strain in the calf with LDS was significantly lower than that in stage 0, I, II and late II lymphedema. On the other hand, SEG/SEFS grades in subcutaneous tissues of the thigh and calf increased with ISL stage (p < 0.001 for both) (Fig. 3a, b). When the images in the elastography window were closely observed, strain and echo-free space, in particular, seemed to be increased (Fig. 4a). It was then hypothesized that tissue strain might be affected not only by fibrotic changes, but also by fluid accumulation; therefore, the correlation between SEG/SEFS grade and subcutaneous tissue strain was studied. There was no correlation between SEG grade and subcutaneous tissue strain (Fig. 4b). There was no correlation between SEFS grade and subcutaneous tissue strain. Even when limited to stage II, there were no differences in subcutaneous tissue strain with respect to SEFS grade (Fig. 4c).
Discussion
It is known that skin and subcutaneous tissue fibrosis progresses in extremities with lymphedema (International Society of 2013), which should result in hardening of these tissues. However, as physicians and/or therapists may have already been aware, these tissues do not always feel harder than normal tissues, but may even feel softer, particularly in the early stages of lymphedema. We could not elucidate the decreased skin and subcutaneous tissue strain values, namely, increased hardness, in the legs with symptomatic lymphedema (stages I–III) compared with asymptomatic legs (stage 0). The order of strain according to the part of the leg, namely, subcutaneous tissue in the thigh > subcutaneous tissue in the calf > skin in the thigh > skin in the calf, was maintained in all stages of lymphedema, as previously reported (Suehiro et al. 2014). Moreover, skin and subcutaneous tissue strain values in calves with LDS were lower than those in legs with LE except for stage III; therefore, these measurements seemed reasonably reliable. We then considered that the skin and subcutaneous tissue strain in the legs with LE might not decrease until a very advanced stage. Mihara et al. (2011) assessed legs with lymphedema using RTE and also did not observe differences among ISL stages.