Dermatoscopy in the early detection, control and surgical planning of basal cell carcinomas

Abstract

Introduction: Dermatoscopy can help practitioners analyze details that are imperceptible to the naked eye, such as basal cell carcinoma arboriform vascularization patterns, which can b e linked to the tumoral limit. Clinical examinations might yet fail in the early detection and demarcation of the extension of such lesions.
Objective: To study the use of dermatoscopy in basal cell carcinomas, aiming at early detection and delimitation of their extension. Method: Basal cell carcinomas (n = 123) were studied prospectively and not randomly, using dermatoscopy, at the author''''''''s private practice. Suspect areas, mainly the nose, underwent dermatoscopic scanning. If the vascular pattern was identified, the tumor was delimited by dermatoscopy, and the incision was carried out using that marking. Surgical margins were checked using conventional cuts, cross sections or micrographic surgery.
Results:The vast majority of the tumors (92%) were located in the face, of which 59% were not well delimited, 21% were well delimited, and 20% were clinically undetectable. Although the vascular pattern was not observed in 18% of the tumors, in cases with a positive identification, it correctly delimited the tumors in 84% of cases (of which 44% were verified with conventional sampling, 48% with micrographic surgery and 8% with cross sections).
Conclusion: Dermatoscopy is an important tool in the early detection and delimitation of the superficial extension of basal cell carcinomas, and is helpful in the surgical planning and clinical control of such lesions.

Keywords: DERMOSCOPY, CARCINOMA, BASAL CELL, MOHS SURGERY, BLOOD VESSELS, CAPILLARIES

INTRODUCTION

Although it is the most common cancer in humans, the clinical diagnosis of basal cell carcinoma (BCC) is not always straightforward. Presser and Taylor''''''''s 1987 study carried revealed only a 70% success rate in diagnosing the condition in an aca- demic setting when using only clinical criteria. 1 In addition to difficulties in diagnosis, it can also be hard to delimit BCC tumors on a purely clinical basis – especially those of predomi- nantly infiltrative growth, which require micrographic surgery since the safety margin concept is inadequate. 2

Dermatoscopy has already demonstrated its great value in the evaluation of cutaneous tumors; it is an indispensable tool in daily dermatological practice. 3 Initially used to examine pig- mented skin lesions, it is increasingly being applied to all types of cutaneous tumors. However, since most BCCs are not pig- mented, dermatoscopy can identify these tumors by their vascu- larization. In their study of the vascularization of cutaneous tumors, Kreusch and Koch described arboriform vessels in 95% of the examined BCCs. 4 Additionally, Kreusch suggests that the tumorous limits of BCCs can coincide with the limit observed in their vascularization; the capillaries of such tumors emerge in the periphery of the lesions, crossing over the lesion. 5 Since the vascular pattern in BCCs differs substantially from that of telangiectasias of normal skin – often visible even to the naked eye – the isolated detection of that type of vascularization through dermatoscopy can signal the presence of a lesion in its initial stages, which is practically imperceptible in a clinical examination. Taking into account that patients with a history of BCCs are 40% more likely to develop a new BCC than the population in general, 6 with lesions occurring mostly in the face (especially in the nose), 7 the early detection of new tumors, when they are not yet clinically visible, would be very helpful in treating those patients.

While dermatoscopy has already demonstrated its utility in evaluating pigmented lesions, there has been little attention devoted to its use in non-pigmented lesions – particularly in BCCs. This study demonstrates the importance of der- matoscopy in the early detection and delimitation of BCCs, which could have an enormous impact on their control and sur- gical planning.

METHODS

BCCs (n = 123) were examined using dermatoscopy in a prospective, open study carried out at the author''''''''s private prac- tice between September 2007 and July 2010. For tumors that could be identified clinically (i.e., without dermatoscopy), der- matoscopy was used to delimit the lesion in order to better plan the surgical treatment. The whole tumoral region, identified by the presence of the arboriform pattern, was marked. The inci- sion was carried out using this marking, with no additional safe- ty margin.

The decision to choose conventional or micrographic sur- gery was based more on the clinical situation than the accuracy of the dermatoscopic data. If micrographic surgery was indicated, the Munich method would be used so that the ratio of tumor/margin could be examined. In this method, the whole surgical piece – and not only the outer edge (surgical margin) – is studied using sequential parallel cuts carried out each 50-100 micra, extending from the bottom to the epidermal border of the specimen. In that way, even if the tumor has been totally extirpated without touching the surgical margin, it can still be seen in relation to the latter. If it was not possible to clinically identify the tumor, a dermatoscopic screening would be carried out in the suspected area, aiming to identify the characteristic vascular pattern. For patients with a history of BCCs who returned periodically for routine examinations, a dermatoscop- ic evaluation would be conducted all over the nose, even in the absence of clinical suspicion.

All cases were photographed both clinically and dermo- scopically. A stereoscopic dermatoscope of great magnification (up to 60x) was used (Kocher GmbH, Mössingen, Germany) (Figure 1). The pictures were captured with a Sony Cybershot DSC F717 camera with an Optiview adapter (Optiview Ltda., São Paulo, Brazil) or with a Fotofinder® (Fotofinder Systems GmbH, Germany).

RESULTS

Of the 123 tumors observed, 92% were located on the face. Of those, 59% were clinically poorly delimited, 21% were well delimited, and 20% were clinically imperceptible. Among the latter (a total of 25 tumors) only five were not located on the nose, and were discovered using dermatoscopic screening in an area indicated by the patient as suspicious for having already presented a small amount of bleeding or desquamation. Nonetheless, there were no clinical indications of a BCC in these other areas.

It was impossible to observe the vascular pattern in 22 (17.9%) of the 123 BCCs, even though a previous biopsy and the surgery itself demonstrated the histological presence of car- cinoma. Among those 22 tumors, only one was clinically well delimited, and in only two cases a small amount of pigment was detected in the dermatoscopy. Of the 101 BCCs with arbori- form vessels that were identifiable using dermatoscopy (82.1% of the total), all had their lateral limits demarcated by the der- matoscopic results; surgical margins were uncompromised in 84% of the cases. Of these 101 BCCs, 44 cases (44%) were ver- ified using conventional sampling, 9 (8%) through serial cuts and 48 (48%) with the Munich method of micrographic surgery.

When analyzing only tumors excised with micrographic surgery, 16 (33%) cases presented compromised margins in the first stage, and five cases did not demonstrate the presence of a tumor in the second stage. In eight cases, the tumor was present until the second stage. In three cases there was residual BCC until the third stage.

One false positive case was observed, with dermatoscopic findings of arboriform vessels and histopathologic findings of an intradermal melanocytic nevus associated with sebaceous hyperplasia in a lesion located in the nose.

DISCUSSION

In this study, the percentage of BCCs that presented an arboriform vascular pattern was well below the one found by Kreusch and Koch. 4 This can perhaps be explained by the different sampling methods used in the two sites where the study was carried out. The author''''''''s private practice receives a great number of referrals for micrographic surgery of lesions that have recurred, in general, one or more times. Usually, the tumor is either hidden by flaps or does not present the usual clinical characteristics of BCCs anymore, or has lost its more visible and diagnosable characteristics in a previous surgery. Patients come in search of a solution to the problem of a com- promised surgical margin, clinically presenting only a scar from a previous surgery. Four of the 25 cases of clinically impercepti- ble tumors had a similar history.

With the first dermatoscopic findings yielding negative results, the lesions were treated as cases where the surgical mar- gins had been compromised, with the absence of clinically per- ceptible symptoms except for the scar. Since the histopathologic examination sample that had yielded a report of a compromised margin suggested a possible false positive case (specimen without dye in the margin, artefacts originated in the handling of the sur- gical sample, etc), 2 it was decided that the case would be period- ically observed. Within six months to one year, still in the absence of clinical signs, the emergence of the vascular pattern could pro- gressively be perceived by observation (scanning examination) of the suspicious site, suggesting the need for micrographic surgery. Such cases were not recorded as yielding negative results from the dermatoscopic perspective, but as examples of the 25 clini- cally imperceptible tumors found using dermatoscopy. It is important to note that, unlike dermatoscopes that are more commonly used, the equipment used in this study generates stereoscopic images of great magnification. The thin and delicate arboriform vascularization detected by that equipment may sometimes pass unnoticed by dermatoscopic examinations with lower optical resolutions, which may hamper or even prevent the identification of a tumor that is still clinically imperceptible. 5 Even the captured photographic images can be difficult when compared to the clearer and more direct visualization available using the dermatoscope used in the present study.

In the studied sample, only 21% of the BCCs were clini- cally well delimited. Only one tumor did not present the char- acteristic arboriform vascular pattern, meaning that 79% of the sample was composed of tumors without reliable clinical char- acteristics for the correct delimitation, which was made exclu- sively using dermatoscopy. Unlike the markings described in some publications 8,9 or communications issued in medical meet- ings, 10 those carried out in this study were based almost exclu- sively on the interruption of the vascular pattern, which was detected using dermatoscopy (imperceptible in clinical exami- nations). This pattern was not perceptible in the pictures pub- lished in the literature. The observed dermatoscopic pattern from this study is aligned with that described by Kreusch and Koch 4 and more recently revised by Zalaudek, 11 which was the only criterion used to delimit the tumors – rather than the pres- ence of pigment, which was observed in only two cases. 5 Literature on the subject is still scarce, and the author is not aware of similar data published previously about using these vas- cular patterns for dermatoscopic delimitation of the lesions or even for early diagnosis. This may be due to the type of der- matoscope used (Figures 2and 3).

The micrographic surgery method adopted for the verifi- cation of the surgical margin is particularly useful, for the Munich method can be used to assess the ratio of tumor/mar- gin. Even with free surgical margins, the site where the tumor was located can be observed – which is not possible in periph- eral methods of micrographic surgery. 12 When analyzing only cases that used micrographic surgery, residual tumors could not be found in the second phase in five out of 16 cases that had margins compromised in the first phase. In this manner, it is sup- posed that, for those cases, having found a tumor in the margins might have characterized what is known as ''''''''coincident mar- gins''''''''. This paradox in the analysis of surgical margins has been described in another study, 2 however it was instrumental in demonstrating that the marking of the lateral margin using der- matoscopy was correct. Of the 48 cases that had micrographic surgery, only 11 (23%) presented margins that were compro- mised (of which eight cases presented two phases and only three presented three phases). Micrographic surgery was not indicat- ed in order to double check the dermatoscopy, but rather was influenced by the clinical situation, mainly recurrences, poorly delimited borders or compromised margins in the previous sur- gery. The 84% rate of negativity of the surgical margins can be challenged, because this was verified through micrographic sur- gery in 48% of the sample. In comparison, Caresana and Giardini obtained 98.5% of negativity in their cases, yet there was no verification through micrographic surgery. Additionally, they excluded sclerodermiform BCCs from their sample; all 200 tumors from their study were of the nodular type. 8

This study''''''''s sample contained 79% poorly delimited or clinically imperceptible tumors, with an 84% negativity rate ver- ified through micrographic surgery in almost half of these. It would be interesting to observe the real condition of the surgi- cal margins of all cases through micrographic surgery, studying primary and recurrent tumors in separate groups of patients. This is the only way to more accurately evaluate the precision of tumor delimitation using dermatoscopy. Therefore, although the information collected using dermatoscopy can help, it should not function as an indication as to whether micrographic surgery should be performed. Dermatoscopy works more as a propaedeutic element that is complementary to the patient''''''''s complete clinic picture.

The discovery of 25 cases of clinically imperceptible tumors, discovered only by scanning, calls attention to the speci- ficity of the arboriform vascular pattern as a predictive factor for BCCs. Since only five cases were not located on the nose, the study suggests that the dermatoscopic scan examination in that part of the body can reveal the presence of incipient tumors in patients with a history of several BCCs, especially because the nasal area is not extensive, and is the location of highest incidence for that type of lesion. Early detection allows more suc- cessful control and handling of BCCs (Figures 4 and 5).

CONCLUSIONS

Identifying the arboriform vascular pattern characteristic of BCCs using dermatoscopy is an important factor in the sur- gical planning and control of these tumors, for it contributes to their early detection, and is also helpful in determining the clin- ically obscure borders of the tumors.

References

1 . Presser SE,Taylor JR.Clinical diagnostic accuracy of basal cell carcinoma. J Am Acad Dermatol. 1987; 16(5 pt 1): 988-90.

2 . Kopke LFF, Bastos JCF, Andrade Filho JS, Gouvêa PS. Margem de segurança: um conceito antigo e relativo. An Bras Dermatol. 2005;80(3):279-86.

3 . Moscarella E,Catricalà C,Zalaudek I,Argenziano G.The dermatoscope as the dermatologist''s Stethoscope. Practical Dermatology. 2010: 34-8.

4 . Kreusch J, Koch F. Auflichtmikroskopische Charakterisierung von Gefässmustern in Hauttumoren.Hautarzt. 1996; 47(4): 264-72.

5 . Kopke, LFF. Dermatoscopia dos carcinomas basocelulares. In: Atlas de Dermatoscopia. Reeze G, Paschoal FM, Hirata S, eds. São Paulo: Lemar Editora, 2011."no prelo" .

6 . Marcil I, Stern RS. Risk of developing a subsequent nonmelanoma skin cancer in Patients with a history of nonmelanoma skin cancer – A critical review of the literature and Meta-analysis. Arch Dermatol. 2000; 136(12):1524-30.

7 . Lang PG Jr,Maize JG. Basal Cell Carcinoma. In Friedman R, Rigel D, Kopf A, Mathew N, Baker D, eds. Cancer of the skin. Philadelphia;London Toronto;Montreal Sidney; Tokyo: Saunders Company, 1991. p. 35-73.

8 . Caresana G, Giardini R. Dermoscopy-guided surgery in basal cell carcinoma. J Eur Acad Dermatol Venereol. 2010; 24(12): 1395-99.

9 . Vitaly Terushkin BS, Steven QW.Mohs Surgery for Basal Cell Carcinoma Assisted by Dermoscopy: Report of Two Cases. Dermatol Surg. 2009;35(12):2031–5.

10 . Sato MS, Suzuki HS, Rosas FM, Hammerschmidt M, Gomes da Silva LL, Serafini SZ. O impacto da dermatoscopia na cirurgia micrográfica de Mohs. Poster premiado durante o XXIII Congresso da Sociedade Brasileira de Cirurgia Dermatológica. Curitiba, 6-9 de abril de 2011.

11 . Zalaudek I, Kreusch J, Giacomel J, Ferrara G, Catricalà C, Argenziano G. How to diagnose nonpigmented skin tumors: A review of vascular structures seen with dermoscopy. Part II.Nonmelanocytic skin tumors. J Am Acad Dermatol. 2010; 63(3):377-86.

12 . Kopke LFF, Konz B. As diferenças fundamentais entre as variações da cirurgia micrográfica. An Bras Dermatol. 1994; 69:505-10.