What is the role of dermoscopy in melanoma diagnosis?

Author(s):

Professor Scott Menzies — Author
Dr Alex Chamberlain — Co-author
Professor H. Peter Soyer — Co-author
Associate Professor Pascale Guitera FACD PhD — Co-author
Cancer Council Australia Melanoma Guidelines Working Party — Co-author

In partnership with:

Melanoma Institute Australia

Cite this page

Menzies, S, Dr Alex Chamberlain, Soyer, P, Associate Professor Pascale Guitera FACD PhD, Cancer Council Australia Melanoma Guidelines Working Party. Clinical question:What is the role of dermoscopy in melanoma diagnosis? . In: Clinical practice guidelines for the diagnosis and management of melanoma. Sydney: Melanoma Institute Australia. [Version URL: https://wiki.cancer.org.au/australiawiki/index.php?oldid=186216, cited 2023 Jun 2]. Available from: https://wiki.cancer.org.au/australia/Guidelines:Melanoma.

Last modified:
17 May 2018 05:33:37

Background

Dermoscopy (dermatoscopy, surface microscopy, epiluminescence microscopy) is a technique that uses a hand-held magnifying device combined with either the application of a liquid between the transparent plate of the device and the skin, or the use of cross-polarised light. This technique allows the visualisation of diagnostic features of pigmented skin lesions that are not seen with the naked eye.^[1]^[2]^[3]^[4]

Back to top

Summary of systematic review results

Meta-analyses performed on studies in a variety of clinical and experimental settings have shown that using dermoscopy improves diagnostic accuracy for melanoma.^[5]^[6] From a meta-analysis of nine level II diagnostic studies subject to varying degrees of verification bias performed prospectively in a clinical setting^[7]^[8]^[9]^[10]^[11]^[12]^[13]^[14]^[15]^[16] the diagnostic accuracy for melanoma, as expressed by the relative diagnostic odds ratio, was 15.6 (95% CI 2.9–83.7) times higher for dermoscopy compared with naked eye (clinical) examination.^[17] Importantly, the meta-analysis was restricted to studies that directly compared the two methods within each study. Sensitivity of dermoscopy was 18% (95% CI 9%–27%; P=0.002) higher than for naked eye examination, but there was no evidence of an effect on specificity (9% higher for dermoscopy; P=0.18).^[17] Subsequent to this meta-analysis one level II study has been published in a primary care setting showing results consistent with the meta-analysis (42% increase in sensitivity and 5% increase in specificity with dermoscopy compared to naked eye).^[18] However, there was a significant improvement in the confidence of diagnosis of both true melanoma (17% increase) and true non-melanoma (16% increase) with dermoscopy. In a further randomized clinical trial in primary care of both pigmented and non-pigmented lesions the odds ratio for a correct diagnosis in the dermoscopy compared to naked eye group was 1.51 (95% CI:0.96-2.37, p=0.07). Again, consistent with the meta-analysis, the effect was greater for the diagnosis of melanoma (61.5% sensitivity using dermoscopy versus 22.2% for naked-eye).^[19]

Specificity can also be examined by its effect on excision rates of benign lesions, which was not addressed in the meta-analysis. Two such studies suggest reduced rates of excision of benign lesions using dermoscopy (reduced benign to malignant ratio of excised lesions and reduction of patients referred to biopsy) and provide indirect evidence for improved specificity in a specialist setting.^[8]^[9] The addition of dermoscopy to naked eye (clinical) examination has also been shown to reduce excisions of benign pigmented lesions in high-risk patients in a specialist setting^[20] and routinely managed pigmented lesions in primary care.^[18]^[19]

While there are fewer studies on dermoscopy in primary care (general practice), all five that were undertaken in this context (one study with both general practitioners and inexperienced specialists or trainees)^[21] show a consistently improved sensitivity for the diagnosis of melanoma or the identification of suspicious lesions requiring biopsy.^[7]^[18]^[19]^[21]^[22] It should be noted that all the studies cited were undertaken by clinicians with some training in dermoscopy (restricted to lectures or reading material in some studies). For this reason, and based on other evidence where lack of training can lead to a reduction of diagnostic accuracy^[23] some formal training in dermoscopy is required to achieve improvement in diagnostic accuracy.

Practice point
Dermoscopy can also identify diagnostic features in non-pigmented (amelanotic) lesions.

Back to top

Evidence summary and recommendations

Evidence summary	Level	References
From a meta-analysis of nine level II studies prospectively performed in a clinical setting, the diagnostic accuracy for melanoma, as expressed by the relative diagnostic odds ratio, was 15.6 times higher for dermoscopy compared with naked eye examination. Sensitivity of dermoscopy was 18% (95% CI 9%–27%; P=0.002) higher than for eye examination, but there was no evidence of an effect on specificity. Two subsequent level II studies showed results consistent with the larger meta-analysis.⁺	I, II	^[7], ^[8], ^[9], ^[10], ^[11], ^[12], ^[13], ^[14], ^[15], ^[16], ^[17], ^[18], ^[19]
Dermoscopy has been shown to reduce the benign:malignant ratio of excised melanocytic lesions and reduce the number of patients referred for biopsy in both specialists and primary care.⁺	II	^[8], ^[9], ^[18], ^[20]

⁺The studies were classified as III-2 according the NHMRC 2009 levels and grade of evidence. Using the Grade approach, the studies were then upgraded to level II if the only criteria not meeting level II was the pathologist was not blinded to clinical information of the patient/lesion since it is established that clinical information is required for an accurate pathological diagnosis of melanocytic lesions.