Ultrasound in dermatology. Part II. Ultrasound of regional lymph node basins and subcutaneous tumours

ARTICLE

Ultrasound in dermatology is a non-invasive tool for the detection and verification of tumours in skin, subcutaneous tissues and in lymph node basins. Since the introduction of ultrasound examination in the routine practice of dermatology in the late seventies [1], ultrasound has evolved to become a standard diagnostic method, at least in Europe.

Subcutaneous or regional nodal metastases or benign tumours or tissue accumulation may not be palpable due to small size, distance from the skin surface or location in an area of postoperative or radiation fibrosis, which renders a physical examination on its own difficult. Palpation even by an experienced physician is also known to be an inaccurate technique for the assessment of lesions in head and neck cancer [2] as well as in scar tissue.

Ultrasound, on the contrary is able to give additional and reliable information about the exact position of a tumour, surrounding anatomical structures, the dimension of the lesion in two perpendicular diameters in mm, its echo pattern as well as the distance from the skin surface. High resolution ultrasound technique is not only able to detect small targets (< 3 mm), but also to predict difficulties in the surgical search for metastases due to a deep subcutaneous or intramuscular position, or a localization near structures (e.g. vessels) which could easily be traumatized.

Several invasive and non-invasive staging tools have been developed for the detection of regional disease or tumours in the soft tissue to support the often ineffective physical examination on its own. Among these techniques, ultrasound has been proven to have superior sensitivity to physical examination in the detection of regional metastases in melanoma patients [3-6] as well as in hematology and oncology patients [7] and it even facilitates the surgical management of such patients [8-10].

Based on technical equipment and clinical requirements, ultrasound used in dermatology is performed using frequencies between 7.5 and 15 MHz. Additionally duplex and colour flow analysis may complement grey-scale ultrasound by demonstrating tumour vascularity and characterising masses [11, 12].

In this part of the two CME articles on ultrasound we want to illustrate the most important technical details, procedural steps and the clinical use of high-frequency ultrasound in dermatology.

Technical details and devices

The physical phenomenon of transmission of ultrasound in tissue with medium-sized frequencies (7.5-15 MHz) does not substantially differ from high frequencies (> 20 MHz) as already described in part I (CME-article) [13]. Examination of intra- and subcutaneous lesions and lymph node basins is performed by scans which work exclusively in a so-called brightness-mode (B-mode or B-scan). Due to the rapid development in technology an appropriate presentation in real-time has been achieved based on a sequence of 30 pictures per second. Based on their excellent ability for high-resolution imaging above all in focused zones, a linear array transducer or curved array transducer proved to be diagnostically highly efficient. In these types of devices 128 or 256 crystals are arranged in a linear or convex manner on the transducer, which enables it to send out acoustic waves and to receive return echoes in groups, thus creating a homogenous result. In very superficially located lesions ultrasound can be performed with a gel stand-off pad for a better focus on the lesion without creating a difference in the impedance of the media. The pad is then removed to image the deeper subcutaneous tissues.

The enhancement of the ultrasound signals is due to the decrease in transmitted signals when passing through the different biological media by means of a so-called time gain compensation. Low intensity echoes returning from deeper subcutaneous regions are more highly enhanced.

For several years many ultrasound devices have offered the possibility of combining the grey-scale ultrasound (B-scan) scanning with the equipment of a high resolution colour coded duplex sonography, for the detection and characterisation of intranodal or intratumoral vascularisation.

Ultrasound-imaging and artefacts

The formation of an ultrasound image is mainly based on the phenomena of interfaces between the different media. According to the physical laws of acoustics, ultrasound waves are either reflected, deviated or absorbed at the interfaces.

Homogenous media are passed through without decreasing the output intensity. That means that almost none of the emitted energy will be reflected, thus resulting in echo-free zones. When passing the next acoustic interface, a so-called impedance difference occurs, which results in an increased reflection and will lead to a more hyperechoic imaging. Finally, there will be a mixture of so called "true image information" and artefacts, which can negatively interact the imaging on one hand (dorsal extinction behind air-bubbles) but can give also more information (dorsal enhancement behind cysts) (Fig. 1).

Anatomy of the skinin ultrasound

The sonomorphological correlate of normal skin is distinguished by a three-lamellar structure (Fig. 2). Despite regional variations in skin thickness it does not influence the ultrasound morphology [14].

A first hyperechoic entrance echo, which is created by the impedance between gel-stand-off pad and epidermis, is followed by an echo-free border. The latter corresponds to the cutis and is depicted in various sizes depending on the region. The border between cutis and subcutis again is imaged as a hyperechoic band.

The subcutaneous tissue itself is relatively free of echoes with inhomogeneously distributed internal echoes and distinct echoes of fibrous septa.

The next signal to follow is the fascia of the muscle imaged as a hyperechogenic band. Finally, the muscles present as a homogeneous and moderate hyperechoic structure which is said to be similar to the features of a feather in the longitudinal cross section. Bone and muscle are very different media resulting again in a highly discriminating impedance. Thus the bone surface is depicted very hyperechoically. The bone cannot be passed by ultrasound based on the reflection of the acoustic waves at the interface between muscle and bone and based on the nearly complete absorption within the bony material.

Ultrasound of regional lymph nodes

Because of its high prognostic impact and the possibility of total removal of metastases, staging of the regional nodal basin in regular intervals is particularly important in patients with high risk primary tumours [15]. Especially for patients presenting with primary melanoma, the sentinel lymph node dissection is becoming the most valuable prognostic standard [16-19]. For high-risk melanoma patients, who should be closely followed-up, techniques such as MRT scans, CT scans, PET or ultrasound B-scan are used to a greater or lesser extent. However, the first three methods may be too expensive and too invasive for routine follow-up, but are without doubt very useful for extended staging in the event of distant metastatic disease preceding further treatment decisions. By means of clinical follow-up, only palpable metastases are detected; these are then often already large in size or numerous. Recently ultrasound B-scan has emerged to be of value in the routine follow-up of melanoma patients based on their high potential to metastasise in lymph node basins. This is true for the detection of regional and non-palpable, in transit metastases. Studies have shown that this technique should be given preference over a physical examination alone [5, 6, 8, 20, 21]. Despite these data ultrasound B-scan unfortunately is neither widely accepted for routine melanoma follow-up nor frequently used for preoperative planning of surgical intervention.

The non-enlarged quiescent lymph node cannot be regularly depicted due to the echo-identical behaviour of the surrounding fatty tissue. Various stimuli such as inflammation or metastatic involvement lead to a typical relative enlargement of the lymph node. Those sonographic patterns are regarded as indicative for the underlying changes.

In the inguinal region, lymph nodes are generally detectable in nearly all human beings. This situation is quite different in the axillary region.

Lymph nodes are longitudinally shaped in B-scan, which show a central hyperechoic area (the cortex) and a narrow hypoechoic peripheral zone (the medulla), described as cockadephenomenon.

Mostly, those lesions are sharply delineated from the surrounding tissue. To differentiate between positive and negative lymph nodes the criteria described by Solbiati and Vasallo are used: the reactively enlarged, benign lymph node regularly shows a longitudinal and transverse axis ratio of more than 2. This ratio is called the Vasallo-Index [22, 23]. Criteria like central echoes and peripheral lack of echoes, are typical for a so-called regressively enlarged lymph node [24]. The hilum of these lymph nodes may be imaged as hyperechoic striae (Fig. 3) or as an echogenic eccentric area. The transitional stages are flowing. An actively inflamed lymph node changes its - initially - oval shape into a growing circular structure. This kind of lymph node occurs in patients with an acute exacerbation of an atopic dermatitis, an erysipelas or concomitant to an immunotherapy with interferons.

Compared to the regressively enlarged lymph nodes, those activated lymph nodes develop an increasing, hypoechoic, broad, peripheral zone sometimes ending in nearly echo-free round structures. The latter lesions cannot easily be distinguished from malignantly transformed lymph nodes. Here repetitive examination enables accurate estimates of changes in lymph node size and in echo-texture of masses. If the lesion regresses spontaneously or remains stable in size without sonographic, radiological or clinical evidence of diseases a benign lesion may be suggested. Similar changes in other lymph nodes of the same or another region may exclude malignancy. However, this is not a reliable criterion considering the fact that most of the lymph node basins in patients suffering from malignant lymphoma show typical alterations in shape and echo.

Additional duplex and colour flow analysis can complement grey scale ultrasound by giving more information about the hilum vascularity and by quantitatively and qualitatively assessing the internal colour flow [25, 26]. Sometimes hilum vessels can be depicted until they end tree-like in the periphery of the node (Fig. 4).

Although no absolutely reliable criteria specific for lymph node metastases have been defined, there exist some features which suggest the presence of metastases in ultrasound. The echogenicity of the nodal hilum is no longer characterised by the cockade architecture and the margins to the adjacent tissue are sharp. Suspicious lymph nodes can be scanned with a length/width ratio (Vasallo-Index) of less than 2 and without hyperechoic central striae. Balloon-shaped homogeneous structures hypoechoic to the adjacent soft tissue are regarded as indicative for metastatic changes. This hypoechoic image is responsible for the dorsal enhancement behind the lesion and is mostly due to a necrotic liquefaction (Fig. 5).

In colour Doppler sonography the typical lymph node metastasis reveals a peripheral perfusion pattern. Internal flow is seldom present in the hilum vessels [27, 28]. Even for an experienced "sonologist" the differentiation between actively inflamed lymph node and an early malignant lymph node involvement is sometimes difficult. Lesions presenting an asymmetrically enlarged periphery or focal intralymphonodal hypoechoic areas are suspicious for malignancy. Sonographical differentiation between various types of lymph node malignancies is not possible. The axillary lymph node metastasis of a mamma carcinoma or a malignant non-melanocytic cutaneous tumour show a hypoechogenicity, but may be somewhat more hyperechoic than a melanoma lymph node metastasis. However, the features are not exact enough to make a rule out of it. And it must be stressed that a histological diagnosis cannot be made by sonography alone.

Last but not least the most important argument against ultrasound technique is that the interpretation of the nodal architecture is a somewhat subjective instrumental investigation. It is very difficult to correlate one investigator's interpretation with a consensus in terms of collaborating individuals with an expertise in ultrasound. Unfortunately this is an area that has yet to have standardized parameters.

To achieve a definite diagnosis, ultrasound can be combined with other techniques such as the fine needle aspiration cytology (FNAC). A sonographically guided FNAC helps to establish a definite cytological diagnosis with a very high sensitivity of about 95% (and a specificity of a 100%) without causing any severe side effects and without a metastasising potential. By combination with ultrasound, a correct diagnosis can be reached even in those very small foci below 1 cm within a lymph node without decrease in sensitivity or specificity [6]. Additional molecular-biological techniques such as tyrosinase reverse transcriptase polymerase chain reaction from fine needle aspirates (FNA-PCR) are able to substitute for the lack of a well-trained cytologist [29, 30].

Ultrasound and melanoma

For several years now using ultrasound for preoperative staging has been recommended, for locoregional control of the disease as well as for postoperative monitoring in patients with malignant melanoma [3, 5, 6, 8, 30]. In a smaller subset of patients sonography was shown to be superior to palpation in the early detection of lymph node metastases [3, 5, 31]. The sensitivity of ultrasound was reported as between 90% and 100% [5, 7, 31-33]. In addition, ultrasound was able to discriminate between neoplastic and non-neoplastic lymph nodes from lymph node metastases. The sensitivity of clinical examination in detecting metastatic involvement, however, has been reported to range from 25% to 70% [5, 7, 31]. From these observations and based on the fact that ultrasound is cost-effective, non-invasive and well tolerated, a routine performance of ultrasound was recommended. Identification of lymph node metastases by ultrasound may help an early definition of subsets of patients at high risk for relapse, who may benefit from adjuvant therapies subsequent to excision of the metastases. Additionally performed ultrasound-guided FNAC enables an early verification of small melanoma metastases detected by ultrasound [9, 30, 34], thus facilitating a very timely and proper staging and enhancing precise treatment decisions. Rossi [17] stressed the fact that with combined ultrasound and FNAC all false positive cases at palpation or US alone were discovered. This way a large number of patients might be spared from surgical excision of clinically suspicious, yet benign lesions.

Ultrasound of different subcutaneous lesions

Because of their very heterogeneous but mostly reproducible echogenicity and relatively typical sonographic appearance, subcutaneous lesions of various origins can easily be discriminated from each other [35-37]. Although a definite diagnosis cannot be established by an ultrasound examination alone, the different behaviour in echogenicity and the concomitant artefacts allow for, or exclude, a distinct differential diagnosis. In combination with a medical history and clinical examination the ultrasound B-scan remains a helpful tool in the hands of the experienced examiner. Again, if ultrasound does not allow a reliable classification of the lesion, the ultrasound guided FNAC will contribute to the process of finding a diagnosis.

Nodal lesions

Solid tumours generally present with a homogeneous echo-feature. Malignant nodal lesions which primarily consist of tumour parenchyma distinguish themselves by their hypoechogenicity. With the portion of tumour stroma, the hyperechoic feature of the nodal lesion increases. Whereas cutaneous and subcutaneous metastases are imposing by an hypoechoic to echo-free appearance in ultrasound, the benign tumours being located in the subcutaneous tissue mostly are depicted as more hyperechoic. As an example of an often occurring accidental finding, we demonstrate the sonographical image of a lipoma in Figure 6. According to the size of the lipoma, cutis and subcutis are separated from each other. The echogenicity corresponds to that of the surrounding subcutaneous fatty tissue and is, in the case of a fibrolipoma, hyperechoic compared to the adjacent tissue, or even hypoechoic in case of an angiolipoma.

Differential diagnosis of hyperechoic lesions are fibromas, pilomatrixomas, or fibrous nodes. Pilomatrixomas often show, due to calcifications, subtumoral sound extinction, which are helpful in establishing the right diagnosis [38].

Cystic lesions

Cysts of different histological origin defined as vacuum filled with fluid are depicted as round to ovally shaped hypoechoic to echo-free structures, with the phenomenon of an distinct dorsal echo enhancement and lateral shadowing (Fig. 1). They show well-defined borders to the adjacent tissue.

Whereas the non-complicated cyst shows a clear echo-free appearance, the so-called complicated cyst reveals central echoes of various intensity due to its contents. In these cases the "cyst-typical" dorsal enhancement may be reduced. Such findings are reproducible e.g. in epidermal cysts (Fig. 7), in steatocystoma multiplex, in hematoma and seroma being in the status of organisation, in abcesses or in tumours and lymph nodes which are in various stages of necrotic liquefaction. With existing small and balloon-shaped seromas a metastasis has to be excluded by e.g. FNAC. In addition a seroma can be totally removed by a diagnostic puncture with specification of the sediment.

Angiomas

The most frequent type within angiomatous tumours in dermatology is the hemangioma cavernosum subcutaneum. According to the duration of the lesion and based on the degree of spontaneous regression and vascularity, hemangiomas are depicted more or less hyperechoic in ultrasound. A high vascularity is sonographically reflected by a dorsal enhancement. Often a combination consisting of hypoechoic (thrombotic) and hyperechoic portions with"dorsal extinction" (calcifications) occurs.

The colour Doppler sonography will assess the vascularity of hemangiomas (Fig. 8) and can be qualitatively and quantitatively assessed. Therefore, ultrasound is an excellent technique for the monitoring of hemangiomas following kryo-therapy or laser therapy.

Conclusion

Starting with some fundamental rules of the acoustics of high-frequency ultrasound, examples of typical sonographical patterns of benign and malignant lesions in dermatology are given.

Detection and characterisation of deeper cutaneous, subcutaneous and nodal structures require diagnostic imaging. Ultrasound is able to depict palpable and especially non-palpable lesions and allows for important information about size, echogenicity and anatomical position of the detected lesions prior to surgical removal. Ultrasound is highly effective in discriminating lymph node metastases and subcutaneous malignant lesions from non-specific innocent nodes. By including ultrasound in short term follow-up examinations of tumour patients, an early detection of tumour recurrences is possible. A combination of ultrasound and FNAC leads to a more precise diagnosis and unnecessary surgery might be avoided. Ultrasound can be used to observe lesions estimated as benign and to monitor therapeutic response over time. Ultrasound is relatively cost-effective, generally available and provides the possibility of scanning large areas of tissue without side effects.

References

1. Rukavina B, Mohar N. An approach of ultrasound diagnostic techniques of the skin and subcutaneous tissue. Dermatologica 1979; 158: 81-92.

2. Van den Brekel MW, Castelijns JA, Stel HV, Luth WJ, Valk J, van der Waal, Snow GB. Occult metastatic neck disease: detection with US and US-guided finde-needle aspiration cytology. Radiology 1991; 180: 457-61.

3. Prayer L, Winkelbauer H, Gritzmann N, Winkelbauer F, Helmer M, Pehamberger H. Sonography versus palpation in the detection of regional lymph-node metastases in patients with malignant melanoma. Eur J Cancer 1990; 26: 827-30.

4. Binder M, Kittler H, Steiner A, Dorffner R, Wolff K, Pehamberger H. Lymph node sonography versus palpation for detecting recurrent disease in patients with malignant melanoma. Eur J Cancer 1997; 33: 1805-8.

5. Rossi CR, Seno A, Vecchiato A, Foletto M, Tregnaghi A, De Candia A, Rubaltelli L, Montesco C, Lise M. The impact of ultrasound scanning in the staging and follow-up of patients with clinical stage I cutaneous melanoma. Eur J Cancer 1997; 33: 200-3.

6. Voit C, Mayer T, Proebstle TM, Weber L, Kron M, Krupienski M, Zeelen U, Sterry W, Schoengen A. Ultrasound-guided fine-needle aspiration cytology in the early detection of melanoma metastases. Cancer 2000; 25: 186-93.

7. Verbanck J, Vandewiele I, de Winter H, Tytgat J, van Aelst F, Tanghe W. Value of axillary ultrasonography and sonographically guided puncture of axillary nodes: a prospective study in 114 consecutive patients. J Clin Ultrasound 1997; 25: 53-6.

8. Nazarian LN, Alexander AA, Rawool NM Kurtz AB, Maguire HC, Mastrangelo MJ. Malignant melanoma: impact of superficial US on management. Radiology 1996; 199: 273-7.

9. Nieuwenhuis EJ, Colnot DR, Pijpers HJ, Castelijns JA, van Diest PJ, Brakenhoff RH, Snow GB, van den Brekel MW. Lymphoscintigraphy and ultrasound-guided fine needle aspiration cytology of sentinel lymph nodes in head and neck cancer patients. Recent Results Cancer Res 2000; 157: 206-17.

10. Voit C, Proebstle TM, Winter H, Kimmritz J, Sterry W, Schwürzer M. Presurgical ultrasound guided anchor-wire marking of soft tissue metastases in stage III melanoma patients. Dermatol Surg 2000 (in press).

11. Nazarian LN, Alexander AA, Kurtz AB, Capuzzi DM, Rawool NM, Gilbert KR, Mastrangelo MJ. Superficial melanoma metastases: appearances on gray-scale and color Doppler sonography. Am J Roentgenol 1998; 170: 459-63.

12. Alexander AA, Nazarian LN, Capuzzi DM, Rawool NM, Kurtz AB, Mastrangelo MJ. Color Doppler sonographic detection of tumor flow in superficial melanoma metastases: histologic correlation. J Ultrasound Med 1998; 17: 123-6.

13. Jemec GBE, Gniadecka M, Ulrich J. Ultrasound in dermatology. Part I. High frequency ultrasound. Eur J Dermatol 2000: 492-7.

14. Fornage BD, Deshayes JL. Ultrasound of normal skin. J Clin Ultrasound 1986; 14: 619-22.

15. Stadelmann WK, Rapaport DP, Soong SJ, Reintgen DS, Buzaid A, Balch CM. Prognostic clinical and pathological features. In: Balch CM, Houghton AN, Sober AJ, Soong SJ, eds. Cutaneous melanoma: clinical management and treatment results worldwide. St Louis: Quality Medical Publishing, 1998: 11-35.

16. Gershenwald JE, Thompson W, Mansfield PF, Lee FE, Colome MI, Tseng CH, Lee JJ, Balch CM, Reintgen DS, Ross MI. Multi-institutional melanoma lymphatic mapping experience: the prognostic value of sentinel lymph node status in 612 stage I or II melanoma patients. J Clin Oncol 1999; 17: 976-83.

17. Rossi CR, Scagnet B, Vecchiato A, Mocellin S, Pilati P, Foletto M, Zavagno G, Casara D, Montesco MC, Tregnaghi A, Rubaltelli L, Lise M. Sentinel node biopsy and ultrasound scanning in cutaneous melanoma: clinical and technical considerations. Eur J Cancer 2000; 36: 895-900.

18. Buzell RA, Zitelli JA. Favorable prognostic factors in recurrent and metastatic melanoma. J Am Acad Dermatol 1996; 34: 798-803.

19. Buzaid AC, Tinoco LA, Jendiroba D, Tu ZA, Lee JJ, Legha SS, Ross MI, Balch CM, Benjamin RS. Prognostic value of size of lymph node metastases in patients with cutaneous melanoma. J Clin Oncol 1995; 9: 2361-8.

20. Prayer L, Winkelbauer H, Gritzmann N, Winkelbauer F, Helmer M, Pehamberger H. Sonography versus palpation in the detection of regional lymph-node metastases in patients with malignant melanoma. Eur J Cancer 1990; 26: 827-30.

21. Binder M, Kittler H, Steiner A, Dorffner R, Wolff K, Pehamberger H. Lymph node sonography versus palpation for detecting recurrent disease in patients with malignant melanoma. Eur J Cancer 1997; 33: 1805-8.

22. Vasallo P, Wernecke K, Roos N, Peters PE. Differentiation of benign from malignant superficial lymphadenopathy: the role of high-resolution US. Radiology 1992; 183: 215-20.

23. Vasallo P, Edel G, Roos N, Naguib A, Peters PE. In vitro high-resolution ultrasonography of benign and malignant lymph nodes: a sonographic-pathologic correlation. Invest Radiol 1993; 28: 698-705.

24. Rubaltelli L, Proto E, Salmaso R, Bortoletto P, Candiani F, Cagol P. Sonography of abnormal lymph nodes in vitro: correlation of sonographic and histologic findings. Am J Radiol 1990; 155: 1241-4.

25. Tschammler A, Ott G, Schang T, Seelbach-Goebel B, Schwager K, Hahn D. Lymphadenopathy: differentiation of benign from malignant disease - color Doppler US assessment of intranodal angioarchitecture. Radiology 1998; 208: 117-23.

26. Dragoni F, Cartoni C, Pescarmona E, Chiarotti F, Puopolo M, Orsi E, Pignoloni P, de Gregoris C, Mandelli F. The role of high resolution pulsed and color Doppler ultrasound in the differential diagnosis of benign and malignant lymphadenopathy. Cancer 1999; 85: 2485-90.

27. Wu CH, Hsu MM, Chang YL, Hsieh FJ. Vascular pathology of malignant cervical lymphadenopathy. Cancer 1998; 83: 1189-96.

28. Moehrle M, Blum A, Rassner G, Juenger M. Lymph node metastases of cutaneous melanoma: diagnosis by B-scan and color Doppler sonography. J Am Acad Dermatol 1999; 41: 703-9.

29. Voit C, Schoengen A, Weber L, Proebstle T. Identification of melanoma metastases by tyrosinase-reverse transcription-polymerase chain reaction of fine needle aspirates. J Am Acad Dermatol 1998; 39: 1030-2.

30. Voit C, Schoengen A, Schwuerzer M, Weber L, Mayer T, Proebstle TM. Detection of regional melanoma metastases by ultrasound B-scan, cytology or tyrosinase RT-PCR of fine-needle aspirates. Br J Cancer 1999; 80: 1672-7.

31. Blum A, Schlagenhauff B, Stroebel W, Breuninger H, Rassner G, Garbe C. Ultrasound examination of regional lymph nodes significantly imporves early detection of locoregional metastases during the follow-up of patients with cutaneous melanoma. Cancer 2000; 88: 2534-9.

32. Merk G, Merk H, Ulrich J, Kühne KH, Willgeroth C. Observations and experiences in sonographic grading and staging of malignant melanoma. In: Altmeyer P, el-Gammal S, Hoffmann K, eds. Ultrasound in dermatology. Berlin: Springer, 1992: 130-6.

33. Uren RF, Howman-Giles R, Thompson JF, Shaw HM, Roberts JM, Bernard E, McCarthy WH. High-resolution ultrasound to diagnose melanoma metastases in patients with clinically palpable lymph nodes. Australas Radiol 1999; 43: 148-52.

34. Rodrigues LK, Leong SP, Ljung BM, Sagebiel RW, Burnside N, Hu TI, Ng BW, Miller JR, Kashani-Sabet M. Fine needle aspiration in the diagnosis of metastatic melanoma. J Am Acad Dermatol 2000; 42: 735-40.

35. Miyauchi S, Tada M, Miki Y. Echographic evaluation of nodular lesions of the skin. J Dermatol 1983; 10: 221-7.

36. Nessi R, Betti R, Bencini PL, Crosti C, Blanc M, Uslenghi C. Ultrasonography of nodular and infiltrative lesions of the skin and subcutaneous tissues. J Clin Ultrasound 1990; 18: 103-9.

37. Ulrich J, Gollnick H. Differential diagnosis of cutaneous and subcutaneous tumours assessed by 7.5 MHz ultrasonography. J Eur Acad Dermatol Venereol 1999; 12: 187-9.

38. Ulrich J, Wesarg I. High frequency ultrasound in the diagnostics of pilomatrixoma. Pediatr Dermatol 2000 (in press).