US with CDUS criteria that suggest malignancy are nonhomogeneous echotexture and architectural distortion due to infiltration of adjacent structures with angiogenesis with variable caliber of tumor vessels (Fig. 1.1). Benign tumors are more often homogeneous and regularly delineated and cause displacement rather than invasion of adjacent structures [56, 87]. However, there is considerable overlap between these two groups. Benign tumors such as skeletal muscle vascular malformation (formerly known as hemangioma), neurofibroma, and schwannoma can present with features of poor delineation and nonhomogeneous echotexture, while sarcomas on the other hand often demonstrate sharply defined margins due to pseudo-capsule formation [28]. If these sarcomas are small in size at the time of detection, the tumor necrosis that would result in nonhomogeneity on US may not yet have occurred. Some lesions are, however, also characterized by their specific location, e.g., subungual glomus tumor and branchial cyst [45], adding to the specificity of the diagnosis. CDUS and spectral Doppler analysis of soft tissue tumors is of limited value when differentiating benign from malignant tumors. If a multitubular-organized vascular pattern is present, the tumor is more likely to be benign. Flow characteristics are not specific enough to be applicable in clinical practice [54]. Resistive indices cannot be used to distinguish benign from malignant musculoskeletal soft tissue masses [71].

Experience with the potential value of strain and shear wave elastography in soft tissue tumors is still preliminary [86]. The mean strain ratios in strain elastography of malignant tumors were significantly higher than the mean strain ratios of benign tumors. There was no significant difference for strain histograms and visual scoring. Strain ratios may be used as an additional tool in soft tissue tumor evaluation, possibly minimizing the number of biopsies [108]. Quantitatively and qualitatively, there is no statistically significant association between shear wave velocity and malignancy. Currently, there is no clear additional role compared to B-mode imaging [98].

The best practical approach is to use ultrasound as a tool for selection of those lesions that are confidently diagnosed as benign. To prevent delay in diagnosis, all indetermined lesions should be submitted to further diagnostic work-up by MRI and imaging-guided CNB in that order [16]. Ultrasonographically definite benign lesions are homogeneous cystic lesions with strong posterior enhancement and sharp margins that do not show any vascular signal with CDUS (at highest sensitivity presets). Careful US examination combined with clinical correlation may suggest a specific diagnosis in the case of an epidermal inclusion cyst (Fig. 1.5), lipoma (Figs. 1.21 and 1.22), or, in the presence of a phlebolith, a skeletal muscle vascular malformation [87, 113]. Also in patients with an initial diagnosis of trauma with Morel-Lavallée, hematoma, or muscle tear, it is evident that an accurate history report is important. A hematoma does not arise spontaneously, except in patients with a coagulation disorder or treated with anticoagulation medication. Hematoma does not keep on growing and has the history of a direct or contusional trauma, mostly severe enough.

The trade-off for high-frequency, linear, musculoskeletal transducers is their limited depth of penetration and the small, static scan field. This is a disadvantage if the soft tissue swelling is large, localized deep in the flexor compartment of the calf, the proximal thigh, buttocks, or trunk or in case of obese patients. Extended field of view sonography (EFOVS) overcomes the disadvantage of a limited, standard field of view [131]. By generating a panoramic EFOVS image, size and anatomical spatial relationships of a soft tissue mass are better evaluated (Fig. 1.1a). EFOVS improves also communication of imaging findings to the referring clinician and contributes to increased reproducibility, which is important for follow-up of lesions [6, 26, 84, 132]. If EFOVS is unavailable, US is not the preferred imaging modality due to its lack of overview and penetration, and MRI should be used as examination of choice [5]. These characteristics of high-frequency transducer characteristics turn into benefits, however, when it comes to diagnosing very small lesions superficially located, lesions as is the case in the wrist, hand, and foot, and lesions of the skin and of peripheral neural origin. The scanning plane can be easily adjusted to the complex local anatomy of the hand, wrist, and foot as was discussed [42, 64, 138].

Other applications of US in soft tissue tumor imaging are US-guided interventional procedures, staging and grading of dermatologic lesions. Diagnostic procedures and therapeutic interventions that are guided by US are gaining in popularity in the musculoskeletal imaging, following the already more established use of this technique in mammography and the abdominal-genitourinary field.

Percutaneous interventions range from ganglion aspiration (18–22 gauge), fine-needle aspiration biopsy (FNAB) in suspected carcinoma metastasis, local recurrence of a soft tissue sarcoma, core needle biopsy (CNB) of extra- and intra-articular solid soft tissue masses, and preoperative needle wire localization of nonpalpable solid soft tissue and vascular tumors to aspiration and culture sampling of a fluid collection, percutaneous catheter drainage of subperiosteal abscess, and muscle biopsy in neuromuscular disease [14, 15, 20, 22, 25, 85, 101, 109, 141]. The procedures can be performed after US selection of the approach (site, depth, and needle angulation) and subsequent skin marking, or better, under real-time US guidance [25, 141].

In screening for nonpalpable subcutaneous metastases of melanoma and cancers of the lung, breast, colorectum, stomach, or ovary and for melanoma recurrences, 7.5 MHz linear array transducers can dynamically screen a wide area of the body [1, 38], but this procedure will probably be replaced by FDG-PET scan as a whole-body staging procedure with an overall sensitivity and specificity of 89 and 96 % [102, 114]. CT is obsolete as it underestimates the number of lesions [101].

Ultrahigh resolution US (20–30 MHz) is being used in imaging of nodular and infiltrative epidermal and dermal lesions. The width of the field of view is 1.2 cm, and the depth of penetration only 1–2 cm.

US using frequencies of 15 MHz or more can clearly define the skin layer morphology including changes in the epidermal thickness [124]. Although epidermal lesions are visible, accurate clinical assessment of their depth of extension is not possible without ultrahigh-frequency (> = 20 mHz) US. Ultrahigh-frequency US is a sensitive tool in lesion detection and delineation of the deep margin of skin lesions. In the majority of lesions, however, it cannot differentiate malignant from benign lesions and will not obviate further need for biopsy [39]. Even with these ultrahigh-frequency transducers, the normal epidermis cannot be visualized. Exceptions are the sole of the foot and the hypothenar area. Hypodermis can be visualized as a hyperechoic layer.

In the detection of local recurrences of soft tissue sarcoma, MRI and US appear to be equally sensitive. US, however, is the most cost-effective method in the detection of early local recurrences of soft tissue sarcomas and should therefore be used for initial routine follow-up and guided biopsies [4]. The presence of a non-elongated, hypoechoic mass is highly suspicious for local recurrence [23]. However, US may be inconclusive in the early postoperative period (3–6 months postoperatively), as inhomogeneous, hypoechoic masses may also represent hematoma, abscess, or granulation tissue (Fig. 1.4). US follow-up with comparison to a baseline study on MRI, both performed 4–6 weeks after surgery, can help to differentiate such cases. US-guided FNAB and/or CNB represents a possible alternative. MRI diagnosis of a soft tissue tumor recurrence in the immediate postoperative period or after irradiation can be extremely difficult due to the diffuse high signal intensity background in (fast) spin-echo T2-weighted images or post-contrast (fat-suppressed) spin-echo T1-weighted sequences. If the tumor recurs in poorly vascularized postoperative scar tissue, intravenous gadolinium administration may have little effect in terms of tumor enhancement and thereby increased conspicuousness [23, 67]. If scar tissue and recurrence cannot be differentiated, US-guided or CT-guided percutaneous biopsy should be considered [11, 23]. Under those circumstances the exam has prime prognostic and therapeutic value [141].

Obtaining an MR time slot may be a practical problem in most institutions if the imaging-guided biopsy has to be performed on short notice. In a US- or CT-guided procedure, this problem does not exist. In addition, the technical staff is limited to US operator.