|Year : 2017 | Volume
| Issue : 4 | Page : 203-209
Dermoscopy versus skin biopsy in diagnosis of suspicious skin lesions
Hassan Ibrahim1, Moustafa El-Taieb2, Ahmed Ahmed3, Radia Hamada4, Essam Nada5
1 Department of Dermatology, Venereology and Andrology, Qena Faculty of Medicine, South Valley University, Egypt
2 Department of Dermatology, Venereology and Andrology, Aswan Faculty of Medicine, Aswan University, Aswan, Egypt
3 Department of Pathology, Venereology and Andrology, Faculty of Medicine, Sohag University, Sohag, Egypt
4 Department of Dermatology, Qena General Hospital, Qena, Egypt
5 Department of Dermatology, Venereology and Andrology, Faculty of Medicine, Sohag University, Sohag, Egypt
|Date of Submission||30-Dec-2017|
|Date of Acceptance||28-Jul-2018|
|Date of Web Publication||19-Jul-2018|
Department of Dermatology, Venereology and Andrology, Quena Faculty of Medicine, South Valley University, 83523 Qena
Source of Support: None, Conflict of Interest: None
Background Dermoscopy is a simple and inexpensive diagnostic technique that permits the visualization of morphologic features that are not visible to the naked eye, forming a link between clinical dermatology and microscopic dermatopathology. For many years, skin biopsy was considered the only definite diagnostic tool that confirms or excludes the clinical diagnosis. Skin biopsies are invasive and have many adverse effects and precautions.
Objective To evaluate the accuracy of dermoscope in diagnosis of skin tumors and its correlation with clinical and pathological diagnosis.
Patients and methods Thirty-four patients who attended Dermatology Clinic at Qena University Hospital from January 2013 to December 2014 were recruited in a nonrandomized prospective study. The inclusion criteria were reported through the following: full history taking, such as (a) name, age, sex, duration of the lesion, onset, progress, and symptoms; (b) previous history of similar lesions or skin cancer; (c) family history of similar lesion or skin cancer; and (d) any recognized changes in the lesion in the past year such as change in size, consistency, hair growth, or bleeding; dermatologic examination, such as (I) type, site, size, shape, color, surface, and border of lesion have been also detected and reported by using of the three-point checklist as a method for differentiation between benign and suspicious lesions and (II) any specific manifestations as tenderness, bleeding, and recurrence have been detected; (III) digital photography has been performed using digital camera (Sony cyber-shot 16.1 mega pixels); dermoscopic findings by using dermoscope (HEINE BETA DELTA 20), and histopathological examination.
Results There was an excellent diagnostic reliability of dermoscopy compared with skin biopsy with interrater κ value of 0.859 (confidence interval: 0.734–0.984, P<0.001). The overall agreement between dermoscopical and histopathological diagnosis was recorded in 27/33 (81.8%) cases. The ability of dermoscopy to differentiate lesion categories was investigated. Nine of the 10 neoplastic lesions and 22 of the 23 non-neoplastic lesions were identified by dermoscopy [χ2(1)=24.2, P<0.001] with sensitivity and specificity rates of 90 and 95.7%, respectively, and positive and negative predictive values of 90 and 95.7%, respectively. Regarding differentiation benign from malignant skin lesions, dermoscopy identified 25 of the 26 benign lesions and identified all malignant skin lesions [χ2(1)=27.8, P<0.001]. The specificity and sensitivity were 96.2 and 100%, respectively, and the positive and negative predictive values were 100 and 87.5%, respectively.
Conclusion There was a good agreement between the dermoscopy and clinical diagnosis and also a good agreement between the dermoscopy and pathological diagnosis. So dermoscopy can be introduced as a routine diagnostic tool in dermatological examination and will be of a great aid in the accurate diagnosis of suspicious skin lesions before invasive skin biopsy. However, further studies with large sample size are needed later on.
Keywords: dermoscopy, skin biopsy, skin tumors
|How to cite this article:|
Ibrahim H, El-Taieb M, Ahmed A, Hamada R, Nada E. Dermoscopy versus skin biopsy in diagnosis of suspicious skin lesions. Al-Azhar Assiut Med J 2017;15:203-9
|How to cite this URL:|
Ibrahim H, El-Taieb M, Ahmed A, Hamada R, Nada E. Dermoscopy versus skin biopsy in diagnosis of suspicious skin lesions. Al-Azhar Assiut Med J [serial online] 2017 [cited 2019 Oct 15];15:203-9. Available from: http://www.azmj.eg.net/text.asp?2017/15/4/203/237136
| Introduction|| |
Epidermal skin tumors are classified broadly into tumors of the surface epithelial cells and tumors of epidermal appendages. Surface epithelial epidermal tumors are further subclassified into benign and malignant. Malignant tumors include basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), which together represent the main bulk of nonmelanoma skin cancers .
Tissue biopsy is an invasive technique that is generally important for diagnosis of any skin lesion, and it is specifically mandatory to confirm the diagnosis of malignant skin lesion. A biopsy cannot reasonably assure that the lesion is benign, and processing of samples requires an average of 72 h to establish the diagnosis of tissue specimens .
Dermoscopy is a simple and inexpensive diagnostic tool that permits the visualization of morphologic features that are not visible by the naked eye. Thus, dermoscopy forms a bridge that links between macroscopic clinical dermatology and microscopic dermatopathology .
In this study, we evaluated accuracy of dermoscope in diagnosis of skin tumors and its correlation with clinical and pathological diagnosis.
| Patients and methods|| |
Thirty-four patients who attended Dermatology Clinic at Qena University Hospital from January 2013 to December 2014, were recruited in a nonrandomized prospective study.
The history taking, general examination, and dermatological examination have been performed for all patients, such as type, site, size, shape, color, surface, and border of lesion, and any specific manifestations as tenderness, bleeding, and recurrence have been detected.
Digital photography had been performed using Digital camera (Sony cyber-shot 16.1 mega pixels; Japan), dermoscopic scanning was done using Heine delta 20 dermoscope (Heine series, Germany) (an attachment piece is used to connect the dermoscope to the digital camera), and histopathological examinations were done by the following: 3–5 punch biopsies have been taken from 29 examined skin lesions, and four lesions were taken by excisional biopsies and put in formalin 10%. Tissue biopsies were fixed in 10% formaldehyde, dehydrated in graded alcohol, cleared in xylene, and embedded in paraffin. Sections of 4 μm thick were prepared, deparaffinized in xylene (10 min), rehydrated in absolute, 95%, and 75% ethanol (2 min each) and washed in tap water. The sections were stained with hematoxylin (2 min), washed in tap water, stained with eosin (30 s), dipped in running water, dehydrated by dipping in graded alcohol, and washed in xylene.
Analysis of data was done by using statistical program for social science (SPSS, version 16; IBM SPSS) as description of quantitative variables as mean, SD, and range and qualitative variables as number and percentage. P value more than 0.05 was insignificant, P value of less than 0.05 was significant, and P value of less than 0.01 was highly significant. κ Test was used for detection of degree of agreement with the following interpretations: 0.0=no agreement, 0.3=poor agreement, 0.5=moderate agreement, 0.7=good agreement, 0.9=excellent agreement, and 1.0=perfect agreement.
The study was approved by the Research Ethics Committee of Qena Faculty of Medicine, South Valley University. An informed consent was obtained from all participants.
| Results|| |
All patients’ clinical characteristics are summarized in [Table 1]. Twenty-one of the investigated cases were females and 13 were males, and the age ranged between 16 and 74 years with a mean (SD) of 44.2 (16.7) years and a median of 40 years. All patients were complaining of suspicious skin lesions which were nodule, papule, plaque, or ulcer in 12, 3, 16, and 3 patients, respectively. All investigated lesions had a gradual onset and the majority had either a progressive course, reported in 27 cases, whereas stationary course was reported only in seven cases. Sixteen of the patients complained of face lesions, whereas lesions at forehead and scalp, trunk, and extremities were recorded in 9, 5, and 4 patients, respectively. Two of the investigated cases had recurrent lesions whereas seven patients had a history of at least one attack of bleeding from their lesions. On clinical examination, the size of the lesions ranged between 1 and 12 cm, with a mean (SD) value of 2.25 (2.35) and a median of 1.5 cm. The lesions were pigmented: brown-to black-colored in 22 patient’s, skin colored in 10 patients and either blue or white colored in one patient for each.
Dermoscopic diagnosis of the investigated cases varied greatly ([Table 2]). Twenty-two cases were reported as pigmented skin lesions, and the dermoscopic diagnosis of which was nevi, pigmented BCC, pigmented seborrheic keratosis (SK), smooth muscle hamartoma (Becker’s nevus), and pigmented discoid lupus erythematosus (DLE) in 9, 3, 7, 2 and 1 lesion, respectively. Cases diagnosed as nevi were further identified by dermoscopy as compound nevi in five cases, dermal nevi in two cases, and one case of each of congenital melanotic nevus and blue nevus. Based on dermoscopic criteria, 10 (29.4%) of the investigated cases were diagnosed as neoplastic lesions, of which eight were reported as malignant and two as benign. From the malignant cases, dermoscopic diagnosis was BCC in six cases and either SCC or Bowen’s disease in one case ([Figure 1] and [Figure 2]).
|Table 2 Dermoscopic and pathological diagnosis of the investigated cases|
Click here to view
|Figure 1 BCC: (a) clinical, picture, (b) Dermoscopy shows slate gray areas, arborizing blood vessels and maple leaf like structure & (c) Histopathologically, Dermis is infiltrated with sheets of malignant epithelial cells with basaloid features with peripheral palisading, focal pigmentation within and around the tumor sheets.|
Click here to view
|Figure 2 SCC: (a) clinical picture, (b) Dermoscopy shows Structurless white zone around central scale, ulceration, blood spots, irregular rounded blood vessels, blue whitish veil and black dots at periphery of lesion & (c) Histopathology of verrucous SCC showing: Verrucous growth of malignant epithelial cells of squamous origin, mild to moderate atypia, cell nests with central keratinization and tumor tissue infiltrates upper dermis.|
Click here to view
Final histopathological diagnosis has been established in 33 of the investigated cases, and the remaining case had been reported as crushed unsatisfactory biopsy material ([Table 2]). From the diagnosed lesions, 17 were pigmented and had been classified as nevi, pigmented BCC, pigmented SK, and pigmented granulomatous lesion in 7, 2, 7, and 1 case, respectively. Malignancy has been confirmed by biopsy in seven cases, of which five were BCC, one was SCC, and one was Bowen’s disease. Three of the lesions were benign tumors, including dermatofibroma, trichofolliculoma, and eccrine hidrocytoma; nine cases were diagnosed as seborrheic keratosis ([Figure 3]); and two cases were diagnosed as smooth muscle hamartoma.
|Figure 3 SK: (a) clinical picture, (b) Dermoscopy shows cobble stone appearance and milia like Cysts & (c) Histopathology showing: acanthosis, hyperkeratosis, increase pigmentation at basal layer and numerous keratin horn within the epidermis.|
Click here to view
The general diagnostic accuracy of dermoscopy was compared with the final histopathological diagnosis after exclusion of the single case reported as unsatisfactory biopsy. There was an excellent diagnostic reliability of dermoscopy compared with skin biopsy with interrater κ value of 0.859 (confidence interval: 0.734–0.984; P<0.001). The overall agreement between dermoscopical and histopathological diagnosis was recorded in 27/33 (81.8%) cases.
Based on histopathological evaluation, the investigated lesions were classified as neoplastic or non-neoplastic skin lesions in 10 and 23 cases, respectively, and as benign or malignant skin lesions in 26 and 7 cases, respectively. The ability of dermoscopy to differentiate these categories was investigated. Nine of the 10 neoplastic lesions and 22 of the 23 non-neoplastic lesions were identified by dermoscopy [χ2(1)=24.2, P<0.001], with sensitivity and specificity rates of 90 and 95.7%, respectively, and positive and negative predictive values of 90 and 95.7%, respectively. Regarding differentiation benign from malignant skin lesions, dermoscopy identified 25 of the 26 benign lesions and identified all malignant skin lesions [χ2(1)=27.8, P<0.001]. The specificity and sensitivity were 96.2 and 100%, respectively, and the positive and negative predictive values were 100 and 87.5%, respectively.
| Discussion|| |
A handheld instrument called a dermatoscope or dermoscope, which has a transilluminating light source and standard magnifying optics, is used to perform dermoscopy. The dermatoscope facilitates the visualization of subsurface skin structures that are not visible to the unaided eye. Dermoscopy helps to identify lesions that have a high likelihood of being malignant and to assist in differentiating them from benign lesions clinically mimicking these cancers. Colors and structures visible with dermoscopy are required for generating a correct diagnosis .
Dermatoscopy is an essential tool for any doctor who encounters and treats skin cancer, and the acquisition of a dermatoscope is the first step in a process that includes choosing a diagnostic method .
We found a good agreement between dermoscopic and clinical diagnosis in the first possibility (72.73%), excellent agreement between dermoscopic and clinical diagnosis in the first or second possibility (87.88%), and an excellent agreement between dermoscopic and clinical diagnosis in the first, second, or third possibility (90.91%).
Moreover, we found a good agreement between pathological and dermoscopic diagnosis in the first possibility (75.76%), a good agreement between pathological and dermoscopic diagnosis in the first or second possibility (84.85%), and a good agreement between pathological and dermoscopic diagnosis in the first, second, or third possibility (84.85%).
Agreement after exclusion of crushed material (unsatisfactory biopsy) showed a good agreement between pathological and dermoscopic diagnosis in the first possibility (80.65%), an excellent agreement between pathological and dermoscopic diagnosis in the first or second possibility (90.32%), and an excellent agreement between pathological and dermoscopic diagnosis in the first, second, or third possibility (90.32%).
In a previous study, there was a good to excellent agreement between dermoscopy and histopathology in diagnosis of BCC .
The diagnostic accuracy of experienced dermatologists for the clinical diagnosis of BCC does not exceed 70%, whereas the sensitivity and specificity of dermoscopy for BCC diagnosis range from 87 to 96% and 72 to 92%, respectively .
A very good correlation occurred for the diagnosis of epidermal nevi between histopathology with the dermoscopy .
Dermoscopy has been shown to increase the diagnostic accuracy for melanoma and to help differentiate melanoma from nevi. So dermoscopy is the perfect instrument to use during the evaluation of pigmented skin lesions in children, because it is painless and provides important information for the clinician that can assist in formulating appropriate management decisions .
Piccolo et al.  showed that there exists a perfect agreement between tele diagnosis (dermoscopy) and histopathology.
Dermoscopy is a valuable, noninvasive, widely used technique that has greatly improved the diagnostic accuracy of pigmented skin lesions and nonpigmented skin disorders, including skin cancers and inflammatory and infectious diseases. It allows in-vivo oobservation of the skin with visualization of morphological structures in the epidermis and papillary dermis, which are not discernible to the naked eye. With respect to skin cancers, dermoscopy has had a significant effect on the early diagnosis of melanoma and is an effective diagnostic tool for the clinical assessment of nonmelanoma skin cancers, such as BCC, Bowen’s disease (BD), actinic keratosis (AK), and SCC .
Zalaudek et al. ,, showed that glomerular vessels with a patchy distribution and scaly surface were found in 90% of all lesions of Bowen’s disease by dermoscopy. Warshaw et al. , demonstrated that both methods, clinical image tele evaluation and teledermoscopy, achieved excellent and equally high concordance rates (κ=0.84 for each) with the gold standard histopathology and showed equally high sensitivity with respect to differentiation of benign and malignant melanocytic and nonmelanocytic skin lesions.
Regarding the total accuracy of dermoscopic diagnosis, in our study, there was a good agreement between dermoscopy and clinical diagnosis (72.73%).
In our study, there was a good agreement between dermoscopy and pathological diagnosis (75.76%).
Cabrijan et al.  reported a 72.8% correlation between histopathology and dermatoscopy in diagnosing skin tumors.
| Conclusion|| |
There was a good agreement between the dermoscopy and clinical diagnosis and also a good agreement between the dermoscopy and pathological diagnosis. So dermoscopy can be introduced as a routine diagnostic tool in dermatological examination and will be of great aid in the accurate diagnosis of suspicious skin lesions before invasive skin biopsy.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Kirkham N. Tumors and cysts of the epidermis. In: Elder DE, Elenitsas R, Murphy GF, Johnson BJ, Xu X, editors. Lever’s histopathology of the skin. 10th ed. Philadelphia, PA: Lippincott Williams & Wilkins 2009. pp. 791-850.
John K. Malignant melanoma of the foot and ankle. Clin Podiatr Med Surg 2000; 17:347-361.
Zivkovic DT, Jovanovic D, Lazarevic V. Dermoscopy of melanoma. Acta Fac Med Naiss 2006; 23:53-56.
Marghoob AA, Usatine RP, Jaimes N. Current status of dermoscopy in the diagnosis of dermatologic disease. J Am Acad Dermatol 2013; 69:814-815.
Rosendahl C, Tschandl P, Cameron A, Kittler H. Diagnostic accuracy of dermatoscopy for melanocytic and nonmelanocytic pigmented lesions. J Am Acad Dermatol 2011; 64:1068-1073.
Guitera P, Pellacani G, Longo C, Seidenari S, Avramidis M, Menzies SW. In vivo reflectance confocal microscopy enhances secondary evaluation of melanocytic lesions. J Invest Dermatol 2009; 129:131-138.
Venturini M, Sala R, Gonzales S, Calzavara-Pinton PG. Reflectance confocal microscopy allows in vivo real-time noninvasive assessment of the outcome of methyl aminolaevulinate photodynamic therapy of basal cell carcinoma. Br J Dermatol 2013; 168:99-105.
Soyer HP, Kenet RO, Wolf IH, Kenet BJ, Cerroni L. Clinicopathological correlation of pigmented skin lesions using dermoscopy. Eur J Dermatol 2000; 10:22-28.
Malvehy J, Puig S, Braun R, Marghoob AA, Kopf AW. Handbook of dermoscopy. UK: Tylor and Francis; 2013.
Piccolo D, Smolle J, Wolf IH, Peris K, Hofmann-Wellenhof R, Dell’Eva G et al.
‘Face-to-face’ versus remote diagnosis of pigmented skin tumors: a teledermoscopic study. Arch Dermatol 1999; 135:1467-1471.
Fargnoli MC, Kostaki D, Pccioni A, MicantoniO T, Peris K. Dermoscopy in the diagnosis and management of non-melanoma skin cancers. Eur J Dermatol 2012; 22:456-463.
Zalaudek I, Argenziano G, Leinweber B. Dermoscopy of Bowen’s disease. Br J Dermatol 2004; 150:1112-1116.
Zalaudek I, Ferrara G, Di Stefani A, Argenziano G. Dermoscopy for challenging melanoma; how to raise the ‘red flag’ when melanoma clinically looks benign. Br J Dermatol 2005a; 153:200-202.
Zalaudek I, Leinweber B, Ferrara G, Soyer HP, Ruocco E, Argenziano G. Dermoscopy of fibroepithelioma of pinkus. J Am Acad Dermatol 2005b; 52:168-169.
Warshaw EM, Lederle FA, Grill JP. Accuracy of teledermatology for nonpigmented neoplasms. J Am Acad Dermatol 2009; 60:579-588.
Cabrijan L, Lipozencic J, Batinac T, Lenkovic M, Gruber F, Stanic Zgombic Z. Correlation between clinical-dermatoscopic and histopathologic diagnosis of skin tumors in our patients. Coll Antropol 2008; 32:195-197.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2]