The Use of High-Frequency Ultrasound as a Methodof Assessing the Severity of a Plaque of PsoriasisAditya K. Gupta, MD, FRCPC; Daniel H. Turnbull, PhD; Kasia A. Harasiewicz, PEng, MSc;David T. Shum, MD, FRCPC; Gena N. Watteel, MSc; F. Stuart Foster, PhD; Daniel N. Sauder, MD, FRCPC

Background and Design: Ultrasound imaging, whileinitially developed to visualize internal organs, is now

being applied to image the skin. In this preliminary study,we used a high-frequency, 40-MHz ultrasound imagingsystem to provide high-resolution images in psoriasis andexamined the relationship between clinical and ultra-sound ratings in plaque-type psoriasis. The ultrasoundimage of a psoriatic plaque demonstrates a superficialechogenic band (band A), followed by a nonchogenic band(band B), and a deeper echogenic band (band C).

Results: In psoriatic plaques (N=145), the severity of thepsoriasis as assessed according to the degree of scaling, ery-thema, and thickness (SET score) correlated best with thewidth of band B (P<.001, r=0.86) and less well with the

width of bandsA (P<.001, r=0.59) and C (P<.001, r=0.44).For the treated psoriatic plaques (n=64), for which pairedreadings were available before and after therapy, changesin the SET scores correlated best with the change in thewidth of band B (P<.001, r=0.96) and less well with thechange in the width of bands A (P<.001, r=0.61) and C(P<.001, r=0.45). Ultrasound analyses and clinicalevaluation were performed by independent raters.

Conclusions: The data suggest that high-frequencyultrasound imaging may prove to be a noninvasivetechnique that can be used as an adjunct to the clinicalevaluation of the lesional severity of psoriatic plaques.(Arch Dermatol. 1996;132:658-662)

ULTRASOUND IMAGING is a

noninvasive, rapidly de¬veloping technology thatwas initially used to vi¬sualize internal organs. In

dermatology, ultrasound systems havebeen used to image cutaneous neo¬

plasms, eg, melanomas and basal cell car¬

cinomas, skin diseases such as psoriasisand scleroderma, and allergic contact, ir¬ritant, and atopic dermatitis.1"23 In the be¬ginning, amplitude scans, or A-scans, were

developed that provided one-dimen¬sional data. With A-mode scanning, theamplitude of the ultrasound wave is ex¬

pressed as a function of its distance fromthe transducer. The subsequent use of B-scans enabled two-dimensional or cross-

sectional images of psoriasis to be visual¬ized.24"33 With B-mode scanning, a two-dimensional image is synthesized from a

series of A-scans, each corresponding toa different position of the transducer. Inthis case, the amplitude of the ultra¬sound signal received from a given loca¬tion is displayed in gray scale, from low(black) to high (white).

The earlier B-scan imaging systemsused lower frequencies, between 2 and 20

MHz, to image psoriasis, and the resolu¬tion was limited to approximately 200 u.m

at best.24'29'34,35 With advances in high-frequency ultrasound technology, inves¬tigators have attempted to perform objec¬tive measurements on ultrasound imagesof psoriatic plaques. They have then triedto relate the results to clinical and histo¬logie findings.25"33-35 Recently, ultrasoundimaging systems operating at 40 to 50 MHzhave enabled improved visualization of cu¬

taneous structures.30'3133 In this study, we

investigated whether imaging in this fre¬quency range can be used to provide dataregarding the disease severity of a psori¬atic plaque.

RESULTS

Comparison between the ultrasoundimage and the corresponding histologiesection, both at the same magnification,showed that the initial echogenic band

See Patients and Methodson next page

From the Division ofDermatology, Departmentof Medicine (Drs Gupta andSauder), the Department ofImaging Research, SunnybrookHealth Science Center(Drs Turnbull and Foster andMs Harasiewicz), the Divisionof Medical Physics, TorontoSunnybrook Regional CancerCenter (Drs Turnbull andFoster and Ms Harasiewicz),and the Department of MedicalBiophysics (Dr Turnbull),University of Toronto,Toronto, Ontario, and theDepartments of Pathology(Dr Shum) and Statistics(Ms Watteel), Universityof Western Ontario, London.Dr Turnbull is now with theSkirball Institute ofBiomolecular Medicine andthe Department of Radiology,New York University MedicalCenter, New York.

at Mt Sinai School Of Medicine, on August 5, 2009 www.archdermatol.comDownloaded from

corresponded to the thickened stratum corneum and thesuperficial epidermis (designated band A) (Figure 1 andFigure 2). Beneath this band was a nonechogenic re¬

gion (designated band B). This represented the elon¬gated epidermal rete ridges and the edematous papillarydermal papillae with congested vessels. The reticular der¬mis beneath this zone corresponded to band C. The re¬

lationship between the histologie sections and the ultra¬sound images is approximate because some distortionoccurs during tissue preparation for microscopic exami¬nation and because it is difficult to obtain a histologiesection from precisely the same area that is imaged.

In the 31 patients, the plaques of psoriasis (N=145)were evaluated clinically and imaged using ultrasound.In 24 of these patients, a total of 64 plaques of psoriasiswere imaged before and 3 to 5 weeks after treatment with

a modified Goeckerman regimen incorporating topicalcorticosteroids, tars, anthralin, and UV-B light in the pso¬riasis day center. Ultrasound imaging was performed on

patients first thing in the morning, before any therapythat day for the psoriasis. We therefore obtained pairedpretherapy and posttherapy data from the 64 psoriaticplaques and used information from the 145 psoriaticplaques to determine the correlation between SET scoresand the widths of bands A, B, and C. The plaques cho¬sen for imaging were those 50 mm or greater in diam¬eter, with uniform scaling, erythema, and thicknessthroughout the plaque. In general, both clinical assess¬ment and imaging were performed at the center of theplaque, at various times during therapy. The use of high-frequency ultrasound to image plaques of psoriasis re¬

sulted in images that were reproducible when the same

PATIENTS AND METHODS

PATIENTS

Thirty-one patients with stable plaque-type psoriasis (22men and nine women; mean±SEM disease duration,14±2.0 years; mean±SEM age, 50±3.5 years) wereenrolled in this outpatient study. Psoriatic plaques were

imaged using 40-MHz ultrasound images, and a clinicalevaluation of the severity of each plaque was performed.The patients had no concomitant serious illness. Patientswho were lactating or pregnant were excluded from thestudy. All patients gave informed consent after a fulldiscussion of the details and procedures involved inthe study. The protocol and consent form were approvedby the institutional review board of the SunnybrookHealth Science Center, University of Toronto, Toronto,Ontario.

CLINICAL EVALUATION

The clinical severity of a psoriatic plaque was assessedaccording to the degree of scaling (desquamation), ery¬thema, and thickness (infiltration) (SET).36"39 In thisstudy, the severity of each individual psoriatic plaquewas graded according to the degree of SET, each compo¬nent of which was graded on a 7-point scale (0 indicatesthat the parameter was absent or clear; 0.5, very mild inseverity; 1, mild; 1.5, between mild and moderate; 2,moderate; 2.5, between moderate and severe; and 3,severe). The individual scores for each of the threeSET components were added to obtain the SET score(each component having a score ranging from 0 to 3).The SET score ranged from a minium of 0 (clinicallynormal or uninvolved skin) to a maximum of 9 (psoria¬sis with severe scaling, severe erythema, and severethickness).

To determine interrater agreement when different in¬vestigators assessed the same plaque of psoriasis, one phy¬sician (A.K.G.) and two nurses evaluated 50 different plaquesof psoriasis. For each plaque, the three evaluators as¬

signed a SET score independently, and the agreement be¬tween the three scores was then calculated.

ULTRASOUND IMAGING

In this study, a real-time B-scan ultrasound backscatter mi¬croscope operating at 40 MHz was used.33 It provided axial(depth) resolution of 30 u,m and lateral (horizontal) reso¬lution of 94 urn. The depth of penetration was 3 to 4 mm.The scanner used a focused transducer made of a copoly-mer of polyvinylidene difluoride and trifluoroethylene. Theultrasound signal received from the cutaneous tissues was

logarithmically amplified, demodulated, and passed througha custom scan converter that output the video images. Eachimage was produced by the transducer scanning over an8-mm linear path at a frame rate of five images per second.

Coupling to the skin was provided by a 30-mm-diameteracrylic cup filled with distilled water that was pressed gent¬ly against the skin over the area to be imaged. The individualwho imaged a psoriatic plaque and an adjacent area ofnormal-appearing skin 6 cm away was blinded to the results of theclinical evaluation. When imaging, the transducer was po¬sitioned perpendicular to the skin to avoid obliquity and toprevent errors during determination ofskin thickness. Echo¬genic bands generally comprised three layers, which we havelabeled A, B, and C, as illustrated schematically in Figure I.When indicated, the width of a band was measured at sev¬eral places and an average value was obtained. Electronic cali¬pers enabled the width of the bands to be measured accu¬

rately to the nearest 0.01 mm.

HISTOLOGIC EVALUATION

Punch biopsies (4 mm) were performed at the site of ul¬trasound imaging in 12 patients. The specimens were fixedin buffered formalin, and the sections were stained withhematoxylin-eosin. The ultrasound image was then com¬

pared with the corresponding histologie section.

STATISTICAL ANALYSIS

Nonparametric Spearman's rank correlation was used to ex¬amine the relationship between the SET scores and the widthofbands A, B, and C (Figure 1). Student's paired f test wasused to test the hypothesis that the mean difference betweenthe width ofband C in a psoriatic plaque (Cp) and in normal-appearing skin (C„) is equal to zero, ie, mean (Cp—Cn)=0.

at Mt Sinai School Of Medicine, on August 5, 2009 www.archdermatol.comDownloaded from

Figure 1. Line diagram depicting location of bands A, B, and C in ahigh-frequency ultrasound image of a plaque of psoriasis.

Figure 2. Comparison between histologie section and high-frequencyultrasound image of a plaque of psoriasis showing bands A, B, and C(both views are at the same magnification). Bar indicates 200 fim.

Figure 3. Top, Ultrasound image of a psoriatic plaque before treatment.Note the thickness of band B. Center, Ultrasound image when there hasbeen moderate, but not complete, clinical improvement of the psoriasis.Band B is still visible, although of decreased width compared with widthbefore therapy. Bottom, Ultrasound image of a nearby clinically uninvolvedarea of skin. One small division on the scale equals 100 p.m.

site in a particular plaque was imaged by independentinvestigators. Once the equipment had been set up, it tookapproximately 5 to 10 minutes to image a psoriatic plaque.

The dermatologie evaluations were performed by thesame investigator who graded the psoriatic plaques ac¬

cording to severity of SET (each component of SET wasrated from 0 to 3 on a 7-point scale). For each plaque,the SET scores were added, resulting in a score rangingfrom 0 to 9.

To assess the variability in SET scores when differ¬ent trained observers evaluate a given psoriatic plaque,50 plaques of psoriasis were each rated independentlyby three évaluators, and the interrator correlation coef¬ficient was found to be O.8.40

Ultrasound images of psoriatic plaques obtained be¬fore therapy were compared with the images seen aftertherapy and with the nearby clinically uninvolved skin. Twoexamples of psoriatic plaques that showed different re¬

sponses to treatment will be discussed in detail. In the firstpatient, the SET score of the psoriatic plaque before therapywas 7. Four weeks later, there was moderate to marked im¬provement, and the SET score had decreased to 2, with theplaque still demonstrating some scaling, erythema, andthickness. After treatment, both band B and band A haddecreased in thickness, although band B was still visible

(Figure 3, top and center). In clinically uninvolved skin,band B was not seen (Figure 3, bottom). In another pa¬tient, there was marked improvement of a psoriatic plaque,with a change in the SET score from 7.5 before therapy to0.5 after 4 weeks of treatment. In this patient, the ultra¬sound image showed that the band B seen before therapywas not visible when the psoriasis had practically cleared.

In some instances, the lower edge of band C was dif¬fuse, making it difficult to precisely determine the lowerborder of this band, particularly in psoriatic plaques im¬aged before the initiation of treatment. In such cases, thelower edge ofband C corresponded to the point at whichthere was a change from an echogenic to a less echo¬genic pattern. In untreated psoriatic plaques, the echo¬genic pattern of band C and the layer beneath this mightbe explained by the attenuation of signal that occurs as

the high-frequency ultrasound beam traverses the thick¬ened stratum corneum and epidermis. In contrast, bothin treated psoriatic plaques, which have a thinner stra-

at Mt Sinai School Of Medicine, on August 5, 2009 www.archdermatol.comDownloaded from

Figure 4. Relationship between the change in scores of scaling, erythema,and thickness (SET) and the change in width of band B in a plaque ofpsoriasis. B1-B2 indicates the difference in the width of band B betweentime points 1 and 2; SET1-SET2, the corresponding change in SET scoresbetween time points 1 and 2.

turn corneum and epidermis, and in normal-appearingskin, the attenuation of the ultrasound beam does notoccur to the same degree. The different degrees of at¬tenuation may play a role in explaining the appearanceof the lower border of band C.

The correlation between clinical evaluation of se¬

verity of a psoriasis plaque (SET score) and ultrasoundimaging (145 plaques) was investigated by measuring thethicknesses ofbands A (range, 48 to 496 |xm), B (range,0 to 832 |xm), and C (range, 0.77 to 2.112 mm) for eachplaque. Spearman's nonparametric rank-correlation analy¬sis demonstrated the best correlation between SET scoresand the width of band B (Bp) (P-C001, r=0.86), with sig¬nificant correlations also being obtained between SETscores and the widths ofbands A (Ap) (P<.001, r=0.59)and C (Cp) (P<.001, r=0.44).

The correlation between the change in SET score

(ASET) of a psoriatic plaque after 3 to 5 weeks oftreatment and the corresponding changes in ultra¬sound parameters for that plaque (n=64) demon¬strated that the best correlation existed between clini¬cal change (ASET) and the change in the thickness ofband B (AB) (Figure 4). The correlations were ASETvs AB (P<.001, r=0.96), ASET vs AA (P<.001, r=0.61),and ASET vs AC (P<.001, r=0.45). The data suggestthat for a given plaque, the change in the severity ofthe psoriasis is predicted most accurately by thechange in the thickness of band B.

Ultrasound images of psoriasis plaques and clini¬cally uninvolved skin 6 cm away (n=64) revealed a

thinner band A in normal-appearing skin (mean ± SEMthickness, 125±4 u,m) compared with the corre¬

sponding psoriatic plaque (179±6.6 u.m) using thepaired t test (P<.00l). This difference might beexplained by the lack of scale and hyperkeratosis ofuninvolved skin. Although band B was well formedand readily visible in psoriatic plaques, it was gener¬ally not observed in normal-appearing skin (27patients) or its width was very small (four patients).There was a significant difference in the mean width ofband C in psoriatic plaques (1.49±0.04 mm) and innearby normal-appearing skin (1.33±0.04 mm) bypaired £ test (P<.001).

COMMENT

In this study, we used high-frequency, B-mode, 40-MHzultrasound scanning to image plaques ofpsoriasis. In earlystudies, 15-MHz ultrasound A-scans or amplitude-modescans demonstrated that the thickness ofpsoriatic plaquesis increased compared with normal skin.34-35 When plaquesof psoriasis are viewed using ultrasound imaging, bandsof varying echogenicity have been observed.24"32-34-35 At 40MHz, ultrasound images of psoriasis exhibit a three-layerstructure similar to that reported by el-Gammal et al.31 Wefound that (1) band A corresponds to the scaly parakera-totic cell layers and the suprapapillary epidermis; (2) bandB represents the edematous and congested papillary der¬mis in the elongated epidermal rete ridges; and (3) bandC is associated with the underlying reticular dermis (Fig¬ures 1 and 2).

It has been suggested that the data from ultrasoundimages may reflect the severity of the psoriasis.24"33'35 Sono-grams obtained at the lower ultrasound frequencies maynot be of high enough quality to accurately assess thethickness of the echogenic bands. It is only recently thatthe development of higher-frequency systems has pro¬duced pictures with improved resolution, thereby en¬

abling us to image skin in greater detail.30"33 The 40-MHz high-frequency ultrasound machine, with its higherresolution, has enabled us to determine the size of thebands in a psoriatic plaque more precisely than we were

prevously able to do with the lower-frequency systems.Our data indicate that the width ofband B correlates bestwith the severity of psoriasis of a plaque as measured bythe SET score. Also, the change in disease activity of a

psoriatic plaque as measured by ASET correlates best withthe change in the thickness of the nonechogenic band B(Figure 4). To our knowledge, this is the first study inwhich the widths of the various bands visible in a plaqueof psoriasis were measured by means of high-frequency40-MHz ultrasound imaging and in which these find¬ings were correlated with the clinical severity of that pso¬riatic plaque, as rated by the degree of SET. These non-invasive data, which are not subject to clinical bias or

experience, could provide a useful adjunct to clinicalevaluation of the severity of the psoriatic activity and theefficacy of therapy for a given plaque.

The observation that band B corresponds to the ma¬

jor portion of the acanthotic epidermis and the superfi¬cial dermis is in agreement with the findings of el-Gammal et al.31 The width ofband B is not a direct measure

of thickness of a plaque of psoriasis; however, it reflectsthe thickness of the papillary dermis, which varies withthe degree of vascularity, edema, and inflammation. Thethickness of band B correlates with the clinical severityof a psoriatic plaque as assessed by the degree of SET.The normalization of the epidermis and dermis that oc¬curs with clinical improvement of a plaque of psoriasismay be responsible for the corresponding decrease in thewidth of the nonechogenic band B (Figure 3).

The use ofhigh-frequency, real-time ultrasound scan¬

ning has enabled the generation of high-resolution im¬ages of psoriatic plaques. Ultrasound imaging has the ad¬vantage of being a noninvasive, relatively inexpensivetechnology that is quick and quite easy to perform. We

at Mt Sinai School Of Medicine, on August 5, 2009 www.archdermatol.comDownloaded from

obtained reproducible images when different investiga¬tors used this method. Our findings suggest that furtberresearch is needed to evaluate the usefulness of high-frequency ultrasound imaging in rating disease activityin psoriasis. The ultrasound machine used in this studyis the initial prototype, and it is hoped that in the futurethe system will be more automated, enabling several pso¬riatic plaques to be imaged fairly rapidly.Accepted for publication January 17, 1996.

This study was supported in part by grants MA12164and 726 from the Medical Research Council of Canada,Ottawa, Ontario. Dr Foster thanks the National CancerInstitute of Canada, Toronto, through which he is sup¬ported as a Terry Fox scientist.

Reprints not available from the author.

REFERENCES

1. Perednia DA. What dermatologists should know about digital imaging. J AmAcad Dermatol. 1991;25:89-108.

2. Nessi R, Blanc M, Bosco M, et al. Skin ultrasound in dermatologic surgicalplanning. J Dermatol Surg Oncol. 1991;17:38-42.

3. Nessi R, Betti R, Bencini PL, Crosti C, Blanc M, Uslenghi C. Ultrasonographyof nodular and infiltrative lesions of the skin and subcutaneous tissues. J ClinUltrasound. 1990;18:103-109.

4. Kraus W, Schramm P, Hoede N. First experiences with a high-resolution ul-trasonic scanner in the diagnosis of malignant melanomas. Arch Dermatol Res.1983;275:235-238.

5. Shafir R, Itzchak Y, Heyman Z, Azizi E, Tsur H, Hiss J. Preoperative ultrasonicmeasurements of the thickness of cutaneous malignant melanoma. J Ultra-sound Med. 1984;3:205-208.

6. Edwards C, Al-Aboosi MM, Marks R. The use of A-scan ultrasound in the as-sessment of small skin tumors. Br J Dermatol. 1989;121:297-304.

7. Hoffmann K, el-Gammal S, Winkler K, Jung J, Pistorius K, Altmeyer P. Skintumours in high-frequency ultrasound. In: Altmeyer P, el-Gammal S, Hoff-mann K, eds. Ultrasound in Dermatology. New York, NY: Springer-Verlag NYInc; 1992:181-201.

8. Hoffmann K, Jung J, el-Gammal S, Altmeyer P. Malignant melanoma in 20\x=req-\MHz B-scan sonography. Dermatology. 1992;185:49-55.

9. Stiller MJ, Driller J, Shupack JL, Gropper C, Rorke MC, Lizzl F. Three-dimensional imaging for diagnostic ultrasound in dermatology. J Am Acad Der-matol. 1993;29:171-175.

10. Dines KA, Sheets PW, Brink JA, et al. High frequency ultrasonic imaging ofskin: experimental results. Ultrason Imaging. 1984;6:408-434.

11. Gropper C, Stiller MJ, Shupack JL, Driller J, Rorke M, Lizzl F. Diagnostic high\x=req-\resolution ultrasound in dermatology. Int J Dermatol. 1993;32:243-250.

12. Harland CC, Bamber JC, Gusterson BA, Mortimer PS. High frequency, highresolution B-scan ultrasound in the assessment of skin tumours. Br J Derma-tol. 1993;128:525-532.

13. Fornage BD, McGavran MH, Duvic M, Waldron CA. Imaging of the skin with20-MHz US. Radiology. 1993;189:69-76.

14. Riedl B, Hiller D, Stosiek N. B-scan ultrasound of regional lymph nodes in dif-ferent dermatological diseases. In: Atlmeyer P, el-Gammal S, Hoffmann K, eds.Ultrasound in Dermatology. New York, NY: Springer-Verlag NY Inc; 1992:87-92.

15. Gassenmaier G, Kiesewetter F, Schell H. Value of high-frequency sonographyin determination of vertical tumor thickness in primary malignant melanomaof the skin. In: Altmeyer P, el-Gammal S, Hoffmann K, eds. Ultrasound in Der-matology. New York, NY: Springer-Verlag NY Inc; 1992:221-227.

16. Brenner S, Ophir J, Weinraub Z. Thickness of basal cell epithelioma measuredpreoperatively by ultrasonography. Cutis. 1984;34:509.

17. Serup J. Localized scleroderma (morphoea): thickness of sclerotic plaques asmeasured by 15 MHz pulsed ultrasound. Acta Derm Venereol (Stockh). 1984;64:214-219.

18. Myers SL, Cohen JS, Sheet PW, Bies JR. B-mode ultrasound evaluation ofskin thickness in progressive systemic sclerosis. J Rheumatol. 1986;13:577\x=req-\580.

19. Hoffmann K, Gerbaulet U, el-Gammal S, Altmeyer P. 20-MHz B-mode ultra-sound in monitoring the course of localized scleroderma (morphea). Acta DermVenereol (Stockh). 1991;164:3-16.

20. Seidenari S, Di Nardo A. B scanning evaluation of allergic reactions with bi-nary transformation and image analysis. Acta Derm Venereol (Stockh). 1992;175:3-7.

21. Seidenari S, Di Nardo A. B scanning evaluation of irritant reactions with binarytransformation and image analysis. Acta Derm Venereol (Stockh). 1992;175:9-13.

22. Serup J. Characterization of contact dermatitis and atopy using bioengineeringtechniques: a survey. Acta Derm Venereol (Stockh). 1992;177:14-25.

23. Semple JL, Gupta AK, From L, et al. Preliminary evaluation of high-frequency(40-60 MHz) ultrasound imaging in the clinical assessment of cutaneous mela-noma. Ann Plast Surg. In press.

24. Sondergaard J, Serup J, Tikjob G. Ultrasonic A- and B-scanning in clinical andexperimental dermatology. Acta Derm Venereol (Stockh). 1985;65:76-82.

25. Hoffmann K, el-Gammal S, Schwarze H, Dirschka T, Altmeyer P. Examinationof psoriasis vulgaris using 20-MHz B-scan ultrasound. In: Altmeyer P, el\x=req-\Gammal S, Hoffmann K, eds. Ultrasound in Dermatology. New York, NY: Springer\x=req-\Verlag NY Inc; 1992:244-249.

26. Overgaard Olsen L, Serup J. High-frequency ultrasound scan for non-invasivecross-sectional imaging of psoriasis. Acta Derm Venereol (Stockh). 1993;73:185-187.

27. Querleux B, Leqeque JL, de Rigal J. Imagierie de la peau in vivo par \l=e'\chog-raphie ultrasonore. Ann Dermatol Venereol. 1988;115:980. Abstract.

28. Berson M, Vaillant L, Patat F, Pourcelot L. High-resolution real-time ultrasonicscanner. Ultrasound Med Biol. 1992;18:471-478.

29. Vaillant L, Berson M, Machet L, Collens IA, Pourcelot L, Lorette G. Ultrasoundimaging of psoriatic skin: a noninvasive technique to evaluate treatment ofpsoriasis. Int J Dermatol. 1994;33:786-790.

30. el-Gammal S, Hoffmann K, Auer T, et al. A 50-MHz high resolution ultrasoundimaging system for dermatology. In: Altmeyer P, el-Gammal S, Hoffmann K,eds. Ultrasound in Dermatology. New York, NY: Springer-Verlag NY Inc; 1992:297-322.

31. el-Gammal S, Auer T, Popp C, et al. Psoriasis vulgaris in 50-MHz B-scan ul-trasound: characteristic features of stratum corneum, epidermis and dermis.Acta Derm Venereol (Stockh). 1994;74:173-176.

32. Hoffmann K, Dirschka T, Schwartze H, et al. 20 MHz sonography, colorimetryand image analysis in the evaluation of psoriasis vulgaris. J Dermatol Sci. 1995;9:103-110.

33. Turnbull DH, Starkoski BG, Harasiewicz KA, et al. A 40-100 MHz B-scan ul-trasound backscatter microscope for skin imaging. Ultrasound Med Biol. 1995;21:79-88.

34. Serup J. Non-invasive quantification of psoriasis plaques: measurement of skinthickness with 15-mHz pulsed ultrasound. Clin Exp Dermatol. 1984;9:502\x=req-\508.

35. Hermann RC, Ellis CN, Fitting DW, Ho VC, Voorhees JJ. Measurement of epi-dermal thickness in normal skin and psoriasis with high-frequency ultra-sound. Skin Pharmacol. 1988;1:128-136.

36. Gupta AK, Matteson EL, Ellis CN, et al. Cyclosporine in the treatment of pso-riatic arthritis. Arch Dermatol. 1989;125:507-510.

37. Gupta AK, Ellis CN, Siegel MT, et al. Sulfasalazine improves psoriasis. ArchDermatol. 1990;126:487-493.

38. Marks R, Barton SP, Shuttleworth D, Finlay AY. Assessment of disease progressin psoriasis. Arch Dermatol. 1989;125:235-240.

39. Fredriksson T, Pettersson U. Severe psoriasis: oral therapy with a new reti-noid. Dermatologica. 1978;157:238-244.

40. Fleiss JL. The Design and Analysis of Clinical Experiments. New York, NY:John Wiley & Sons; 1986:1-32.

at Mt Sinai School Of Medicine, on August 5, 2009 www.archdermatol.comDownloaded from