Ultrasound in dermatology – basic principles and applications

Ultrasound in dermatology – basic principles and applications

D. Rallan and C. C. Harland

Department of Dermatology, St Helier Hospital, Carshalton, Surrey

Summary Ultrasound is widely used in general clinical medicine for non-invasive internal

imaging. Over the last twenty years, technological advances have enabled the

application of high-resolution ultrasonic imaging to the skin. Equipment and hardware

is now available to produce cross-section images and three-dimensional

reconstructions of selected skin segments. Resolution in vivo is not comparable to

light microscopy but continues to improve with superior transducer designs. Skin

ultrasonography has been reliably employed as an imaging modality in experimental

designs, its quantifiable parameters being a distinct advantage. In particular, increased

water content of the upper dermis, as occurs in inflammatory conditions or as a result

of photodamage, can be demonstrated clearly as an echo-poor zone. Thus, the future of

high-resolution ultrasound (HRU) may reside in its experimental role in monitoring

inflammatory or photodamage processes in response to novel treatments. With regard

to skin tumours, HRU reliably measures tumour thickness and also holds promise as a

differentiator between seborrhoeic keratoses vs. melanoma and benign naevi vs.

melanoma. While largely an experimental tool, the potential as an accurate,

quantitative and reliable diagnostic and monitoring aid, merits further attention

with an emphasis on clinical outcome measures.

Introduction

The term ‘ultrasound’ refers to sound waves of a

frequency that is above the human hearing range. In

1794, two Italian zoologists, who showed that bats use

an acoustic orientation system, discovered the phenom-

enon of ultrasound. Almost 85 years later, the Curie

brothers demonstrated the piezo-electric effect of sound

generated from crystal.1

Ultrasound became established in general medical

imaging in 1950 but it was not until 1979 that its

usefulness in dermatology was recognized.2 Today, skin

ultrasound remains an experimental tool which has

received increasing attention alongside other noninva-

sive investigative methods. Herein, the principles of skin

ultrasonography are described and potential future uses

of high resolution ultrasound are discussed.

Doppler ultrasound is alluded to only briefly.

Basic physics

All sound requires a medium in which to propagate.

When traversing media with differing density, there is a

change in velocity. This change may cause refraction,

reflection or more usually a combination of both

(scatter). Sound is also absorbed by various structures

to varying degrees. Excess sound energy absorption can

cause cavitation or heat damage. At the appropriate

frequency, ultrasound has the potential to highlight

structural heterogeneity. Increasing the frequency

improves the resolution, thereby enhancing the ability

to discriminate between structures. A greater frequency,

however, reduces the depth penetration of the sound

wave.3

In general medical imaging, a frequency of 3.5–

7.5 MHz is used producing a resolution of 2–3 mm. For

Correspondence: D. Rallan, Department of Dermatology, St Helier Hospital,

Wrythe Lane, Carshalton, Surrey, SM5 1AA, UK.

Tel. ⁄ Fax: +44 20 8296 3479.

E-mail: [email protected]

Accepted for publication 8 July 2003

Experimental dermatology • Review article

632 � 2003 Blackwell Publishing Ltd • Clinical and Experimental Dermatology, 28, 632–638

skin imaging, a high frequency of 15 MHz and above is

used delivering a resolution of 60–120lm. This is

termed high resolution ultrasound (HRU).

Sound production and image generation

Figure 1 is a schematic example of a tranducer within a

scanner head used to generate A-, B-and C-scans which

will be discussed below.

A (amplitude) mode scanning

The most basic imaging relies on conversion of signals

into an amplitude value (Fig. 2). The radiofrequency

data is ‘enveloped’ to display amplitude data or a ‘peak’

on an oscilloscope. Each peak corresponds to an interface

between two structures of differing densities. The time

between the peaks on the oscilloscope enables calculation

of the distance between the corresponding interfaces and

therefore allows the determination of skin and tumour

thickness. The velocity of sound in the skin is taken as an

average of 1580 m ⁄ s. A-scanning is unidimensional.

B (brightness) mode scanning

If the transducer is moved along the skin surface,

several A-scans can be acquired. However, in this

instance, the returned signals are given a brightness

value or echogenicity on a grey scale instead of an

amplitude value. By means of dedicated software, this

Figure 1 A schematic representation of high-resolution ultra-

sound in operation; a transducer produces high-frequency sound.

A brief pulse is fired and the reflected sound is picked up by the

same transducer before the next pulse is fired (pulsed ultrasound).

The received signal is fed into a computer and can be processed in

several ways. An A-scan is unidimensional, and can only give

information at one line through the lesion; a B-scan is produced by

multiple A-scans in a vertical two dimensional plane. The

sequential back and forth movement of the transducer enables

acquisition of multiple B-scans and permits the construction of a

three-dimensional image.

Figure 2 Image generation for A- and

B- mode scanning. The raw data is elec-

tronically converted to either an amplitude

value or a brightness value to produce an

A- or a B- scan respectively.

Ultrasound in dermatology • D. Rallan and C. C. Harland

� 2003 Blackwell Publishing Ltd • Clinical and Experimental Dermatology, 28, 632–638 633

radiofrequency data is used to generate a two-dimen-

sional cross-sectional image (Fig. 2).

C (computed) mode scanning

B-scans are cross-sectional images in a plane perpen-

dicular to the skin surface as shown in Fig. 3. By

combining multiple B-scans, an image can be generated

which is in a plane parallel to the skin. An image in this

plane can only be computed (and not acquired directly).

Evolution and applications

Conventional ultrasound can be helpful in certain

specific clinical circumstances in dermatology. High

resolution A- and B- mode scanning, however, has

produced a host of experimental applications as listed in

Table 1. A detailed discussion of each is outside the

scope of this article. Some studies are used to demon-

strate basic concepts.

Skin thickness and A-mode scanning

In 1979, Alexander and Miller were the first to use

A-mode ultrasound for the determination of skin

thickness in 10 healthy volunteers.2 A comparison

with the gold standard of xeroradiography (a high

resolution method involving x-rays applied parallel to

the skin surface) showed a high correlation (r ¼ 0.99).

They concluded that ultrasound was safer, cheaper, and

more versatile as xerography was confined to the limbs.

It therefore rapidly became the preferred method for

determination of skin thickness, paving the way for

further experimental uses.

Amongst the first applications was the objective

assessment of skin atrophy caused by topical steroids.4

This study also confirmed the high correlation between

the two methods for both normal and corticosteroid

treated skin. Furthermore, Capewell et al. were the first

to use A-mode scanning to quantify dermal thinning in

asthmatic subjects on inhaled corticosteroids and oral

steroids.5 Skin thickness has been used as an index of

severity, and as a means of monitoring treatment

response, of plaque psoriasis.6,7 However the surface

scale interferes with the propagation of ultrasound and

must be moisturized prior to transducer application.

Allergic patch test reactions can be quantified accu-

rately when compared with traditional calliper meas-

urements.8,9 Skin thickness has also been found to vary

during the course of the day and this is thought to be

due to gravitational translocation of dermal fluid.10,11

Until 1983, skin ultrasound was used almost exclu-

sively for skin thickness determination. Coleman and

Lizzi were the first to investigate the potential of A-mode

scanning in tissue characterization.12 They used a

spectrum of frequencies and observed that ocular

malignant melanoma and metastatic carcinoma could

be reliably classified based on their acoustic reflectance

signatures. This technique of analysing acoustic spectra

Table 1 Examples of the potential applications of ultrasound

imaging in dermatology.

Applications

Common mode(s)

of scanning

Skin thickness determination2,4,5, A

Examination of lymph nodes in

skin disease21,23,26

B, C and duplex

sonography

Differentiation of soft tissue

tumours12,13,20

B and C

Staging and follow up of malignant

melanoma17,23,24

A, B and duplex

sonography

Examination of benign and malignant

epithelial tumours20,22

A, B and C

Examination and classification of

scleroderma morphoea31

B

Examination and assessment of

plaque psoriasis6,7

A and B

Assessment of dermal oedema10,11,19 A and B

Assessment of skin ageing and

photodamage18

A and B

Screening for necrotizing fasciitis32 Conventional sonography

Post-operative and inflammatory

abdominal wall abscess detection33

Conventional sonography

Figure 3 The difference between A-, B- and C-mode imaging. A

scans are unidimensional. Multiple A- scans combined produce a

two dimensional cross-section image (B-scan) as shown. C-scans

are also two dimensional but in a plane parallel to the skin surface

and are derived from multiple B-scans. Both B- and C-images can

be used for three dimensional reconstructions.



has not been evaluated fully for skin tumours. Some

features of conventional A-mode scanning had been

investigated with regard to small tumour characteri-

zation13 but B-mode technology now shows more

potential.14,15

Echogenicity and B-mode scanning

B-mode software technology became sufficiently devel-

oped by 1986 making it possible to generate cross-

sectional images. Consequently, it has become apparent

that tissue characteristics, in particular collagen fibres,

keratin and dermal water content, are the major

determinants of HRU images.15,16 Furthermore, dermal

collagen content, collagen type, collagen orientation and

collagen bundle size all have an effect on dermal

reflectivity ⁄ echogenicity.15 These factors determine the

number of collagen–matrix interfaces and hence the

extent of sound scatter. Thick keratin is highly reflective

and has the effect of producing an echolucent retro-

lesional shadow (Fig. 4). In contrast, dermal water is

poorly reflective and reduces dermal echogenicity.

Figure 4 B-scan of a seborrhoeic keratosis. The difference in

density between the gel and skin surface causes significant sound

reflection at the point of entry. This entry echo line (EEL), is

prominent in heavily keratinized tumours (e.g. seborrhoeic kera-

toses) which reflect almost all sound at the point of entry and

thereby cast a shadow in the dermis (posterior attenuation).

bbl – bubbles in gel.

300250

200150

100

50

00.5 1 1.5 2.5 3.52 3

0.5

1

1.5

2

2.520 40 60 80 100 120 140 160 180 200 220

(b)

300250

200150

100

50

00.5 1 1.5 2

1

1.5

2.0

2.520 40 60 80 100 120 140 160 180 200 220

(a)

A line (88) - through normal skin

Figure 5 (a) B-scan (top) from skin of a middle aged woman who is a longterm sunbed user. A malignant melanoma (MM) is shown. EEL –

entry echo line, SC – subcutis, B – bubble in gel. Note the distinct echopoor ’sub-epidermal low echogenic band (SLEB) indicative of

photodamage. An A-scan (bottom), is taken through the dashed line. (b) B-scan (top) and one A-scan (bottom) through a compound

naevus from a young male patient. EEL: Entry echo line, CN: compound naevus, B: bubbles in gel. Compared to Figure 5a, a wide echopoor

band beneath the EEL is not seen.



Descriptive studies on tissue characteristics of normal

skin and pathology have been well documented. De Rigal

et al. were one of the first groups to use quantitative

B-mode scanning in their study on photodamage.18

They observed and quantified a distinct subepidermal

low echogenic band (SLEB) in photodamaged skin

(Fig. 5). This was initially attributed to altered elastotic

collagen. Alternatively, such echopoor zones could be

equated to increased water content.16 Seidinari and

co-workers capitalized on this phenomenon by quanti-

fying dermal oedema in patch test reactions (Fig. 6).

The distribution of dermal oedema has now been

elucidated in several conditions: cardiac failure, post-

thrombotic venous disease and lymphoedema.19

B-mode scanning and skin tumours

During the last decade, the emphasis of research has

shifted towards the field of noninvasive assessment of

skin tumours.15,17,20–26 Tumour depth can be deter-

mined in vivo with good accuracy. In a study involving

144 melanomas, Gassenmaier et al. showed a strong

correlation (r ¼ 0.95) between sonometric and histo-

logic measurements of tumour thickness.25 Ultrasound

tends to overestimate tumour thickness. Tissue shrink-

age following excision and associated lymphocytic

infiltrates are thought to be the main causes of this

disparity.

Tumours are generally echo-poor. Most benign skin

tumours have more internal echoes compared to

malignant melanoma (Fig. 7). This difference may be

due to a relative paucity of collagen in invasive

melanoma. Quantification of features such as internal

shadows, has not yet been shown to have greater

accuracy than a subjective assessment of such images

by a consultant dermatologist.22 Furthermore, studies

which are currently in progress have yet to show

superior diagnostic accuracy for HRU vs. clinical

judgement at any level. Duplex sonography (B-scan

combined with Doppler ultrasound), shows good poten-

tial in this regard particularly in relation to lymph node

assessment.21,23,26

Figure 7 Benign naevus (left) compared to

malignant melanoma (right). Benign

tumours generally have more internal

echoes.

Figure 6 A typical ultrasound appearance of a positive allergy test

is shown (top) compared to normal skin (bottom). The oedema in

allergic and irritant reactions was found to occur mainly in the

papillary dermis. The degree of reduction in echogenicity could be

quantified, thus giving another parameter (apart from skin thick-

ness changes) for measuring patch test reactions. (Figure adapted

from Contact Dermatitis,1991; 24: 216-222).



Advantages and disadvantages of HRU imaging

Compared with xeroradiography, ultrasound is cheaper

and safer. Depending on the type of scan to be acquired,

the procedure can be quick. The noninvasive nature

enables accurate measurements to be made in vivo.

Prototype equipment is cumbersome (Fig. 8) and

certain sites (e.g. inner canthus) remain inaccessible.

The resolution in vivo is not comparable to that with

conventional light microscopy. There is at present, an

upper limit to the sound energy that can be transmitted

into the skin without causing damage.

Future potential

The ability to quantify photodamage and skin ageing is

perhaps to be the most promising application of skin

ultrasound in the foreseeable future. This is likely to be

driven by the booming ‘wellness industry’.27 With

improved tissue characterization, both qualitative and

quantitative, HRU may find a role as a diagnostic aid,

especially in conjunction with other noninvasive tech-

niques such as reflectance spectrophotometry and skin

patterning.28,29

The destructive properties of sound are already used

in other medical fields but have yet to be assessed in

skin tumour therapy. In order to achieve resolution

comparable to light microscopy, a frequency of

40–60 MHz is required.30 This is achievable ex vivo.

The development of epidermal cooling devices as used

for modern laser therapies will be necessary. However,

the ability to examine living tissue at a microscopic level

stands to introduce a whole new era of clinical

investigation and experimental designs for the future.

Acknowledgements

The department’s research is funded by Epsom and St

Helier NHS Trust Fellowship Scheme. We acknowledge

the collaboration and support provided by the Depart-

ment of Physics, Institute of Cancer Research, Royal

Marsden Hospital and especially from Dr Jeff Bamber

and Nigel Bush.

References

1 Curie J, Curie P. Development par pression d’electricite

polaire dans lees cristaux hemiedres a faces incliees.

CR Acad Sci 1881; 91: 294–5.

2 Alexander H, Miller DL. Determining skin thickness with

pulsed ultrasound. J Invest Dermatol 1979; 72: 17–9.

3 Bamber JC. Acoustic characteristics of biological media.

In: Cocker MJ, ed. Encyclopaedia of Acoustics. New York:

John Wiley and Sons, Inc., 1997: 1703–26.

4 Tan CY, Marks R, Payne P. Comparison of xeroradio-

graphic and ultrasound detection of corticosteroid induced

dermal thinning. J Invest Dermatol 1981; 76: 126–8.

5 Capewell S, Reynolds S, Shuttleworth D et al. Purpura and

dermal thinning associated with high dose inhaled corti-

costeroids. Br Med J 1990; 16: 1548–51.

6 Serup J. Non-invasive quantification of psoriasis plaques –

measurement of skin thickness with 15MHz pulsed

ultrasound. Clin Exp Dermatol 1984; 9: 502–8.

7 Gupta AK, Turnbull DH, Harsiewicz KA et al. The use of

high frequency ultrasound as a method of assessing the

severity of a plaque of psoriasis. Arch Dermatol 1996; 132:

658–62.

8 Serup J, Staburg B. Ultrasound for assessment of allergic

and irritant patch test reactions. Contact Dermatitis 1987;

17: 80–4.

9 Seidinari S, Di Nardo A, Pepe P et al. Ultrasound B-scan-

ning with image analysis for assessment of allergic patch

test reactions. Contact Dermatitis 1991; 24: 216–22.

10 Gniadecka M. Localisation of dermal oedema in lipo-

dermatoslcerosis: distribution, effects of posture and

compressive therapy evaluated by high frequency ultra-

sonography. Acta Derm Venereol 1995; 75: 120–4.

11 Tsukahara K, Takema Y, Moriwaki S et al. Dermal fluid

translocation is an important determinant of diurnal

variation in human skin thickness. Br J Dermatol 2001;

145: 590–6.

12 Coleman DJ, Lizzi FL. Computerised ultrasonic tissue

characterisation of ocular tumours. Am J Ophthalmol 1983;

96: 165–75.

Figure 8 The prototype equipment used by our group. Data

acquisition software and the scanner head (resting on patients

arm) could be condensed into a laptop with a toothbrush sized

probe in future models.



13 Edwards C, Al-Boosi MM, Marks R. The use of A-scan

ultrasound in the assessment of small skin tumours.

Br J Dermatol 1989; 121: 297–304.

14 Hughes BR, Black D, Srivastava A et al. Comparison of

techniques for the non-invasive assessment of skin

tumours. Clin Exp Dermatol 1987; 12: 108–11.

15 Harland CC, Bamber JC, Guesterson BA et al. High

frequency, high resolution ultrasound in the assessment

of skin tumours. Br J Dermatol 1993; 128: 525–32.

16 Gniadecka M, Quistorff B. Assessment of dermal water by

high frequency ultrasound: comparative studies with nuc-

lear magnetic resonance. Br J Dermatol 1996; 135: 218–24.

17 Mende U, Petzoldt W et al. Sonography: the ideal imaging

method for staging and follow-up of malignant melanoma.

In: Ultrasound in Dermatology (Altmeyer, P, El-Gammal, S,

Hoffman, K, eds), 1st edn. Berlin: Springer-Verlag 1992,

119–29.

18 De Rigal J, Escoffier C, Querleux B et al. Assessment of

ageing of the human skin by in vivo ultrasonic imaging.

J Invest Dermatol 1989; 93: 621–5.

19 Gniadecka M. Localisation of dermal oedema in lipoder-

matosclerosis, lymphoedema and cardiac insufficiency.

J Am Acad Dermatol 1996; 35: 37–41.

20 Hoffman K, El-Gammal S, Winkler K et al. Skin tumours in

high frequency ultrasound. In: Ultrasound in Dermatology

(Altmeyer P, El-Gammal S, Hoffman K, eds) 1st edn.

Berlin: Springer-Verlag, 1992, 181–201.

21 Schmid-Wendtner M, Partscht K, Korting HC et al.

Improved differentiation of benign and malignant

lympadenopathy in patients with cutaneous melanoma

by contrast-enhanced colour doppler sonography. Arch

Dermatol 2002; 138: 491–7.

22 Harland CC, Kale SG, Jackson P et al. Differentiation of

common benign pigmented lesions from melanoma by high

resolution ultrasound. Br J Dermatol 2000; 143: 281–9.

23 Prayer L, Winkelbauer H, Gitzmann N et al. Sonography

versus palpation in detection of regional lymph node

metastases in patients with malignant melanoma. Eur J

Cancer 1990; 26: 827–30.

24 Rossi CR, Seno A, Vecchiato A et al. Ultrasound guided

fine needle aspiration cytology in the early detection of

melanoma metastases. Cancer 2000; 25: 186–93.

25 Gassenmeier G, Kiesewetter F, Schell H. Value of high

frequency sonography in the determination of maximum

vertical tumour thickness in primary malignant melanoma

of the skin. In: Altmeyer, P, El-Gammal, S, Hoffman, K, eds,

Ultrasound in Dermatology, 1st edn. Berlin: Springer-Verlag,

1992, 221–7.

26 Tschammler A, Ott G, Seelbach-Goebel B et al. Lymhpa-

denopathy: differentiation of benign from malignant dis-

ease – colour doppler ultrasound assessment of intranodal

angioarchitecture. Radiology 1998; 208: 117–23.

27 Pilzer PZ. The next trillion. In: The Next Trillion, 1st edn.

(abridged version), Dallas: Video plus Inc 2001, 27–34.

28 Round AJ, Duller AWG, Fish PJ. Lesion classification using

skin patterning. Skin Res Technol 2000; 6: 183–92.

29 Wallace VP, Crawford DC, Mortimer PS et al. Spectropho-

tometric assessment of pigmented skin lesions: methods

and feature selection for evaluation of diagnostic

performance. Phys Med Biol 2000; 45: 735–51.

30 Foster FS, Pavlin CJ, Harasiewicz KA et al. Advances in

ultrasound biomicroscopy. Ultrasound Med Biol 2000; 26:

1–27.

31 Hoffman K, El-Gammal S et al. Examination of circum-

scribed scleroderma using 20-MHz B-scan ultrasound.

In: Altmeyer, P, El-Gammal, S, Hoffman, K, eds. Ultrasound

in Dermatology, 1st edn. Berlin: Springer-Verlag, 1992,

221–7. 231–43.

32 Yen ZS, Wang HP, Ma HM et al. Ultrasonographic

screening of clinically-suspected necrotizing fasciitis. Acad

Emerg Med 2002; 9: 1448–551.

33 Yeh HC, Rabinowitz JG. Ultrasonography and computer

tomography of inflammatory abdominal wall lesions.

Radiology 1982; 144: 859–63.



Documents

Ultrasound in dermatology – basic principles and applications