Archived Policies - Medicine
Optical Diagnostic Devices for Evaluating Skin Lesions Suspected of Malignancy
Dermatoscopy, using either direct inspection, digitization of images, or computer-assisted analysis of whole or partial body photography including, but not limited to reflectance confocal microscopy is considered experimental, investigational and/or unproven as a technique to evaluate or serially monitor pigmented skin lesions.
Computer-based optical imaging devices, e.g., multispectral digital skin lesion analysis, are considered experimental, investigational and/or unproven as a technique to evaluate or serially monitor pigmented skin lesions.
Dermatoscopy and computer-based optical imaging devices are considered experimental, investigational and/or unproven for defining peripheral margins of skin lesions suspected of malignancy prior to surgical excision.
NOTE: Whole body photography represents one component of dermatoscopy. CPT code 96904 may also be submitted to describe whole body photography without dermatoscopy.
There is interest in noninvasive devices that will improve the diagnosis of malignant skin lesions. One technique is dermatoscopy (dermoscopy, epiluminescence microscopy, in vivo cutaneous microscopy), which enables the clinician to perform direct microscopic examination of diagnostic features in pigmented skin lesions. Another approach is the use of computer-based light imaging systems. These techniques have the potential to improve diagnostic accuracy for suspicious skin lesions and may increase the detection rate of malignant skin lesions and/or reduce the rate of unnecessary biopsies.
Dermatoscopy, also known as dermoscopy, describes a family of noninvasive techniques that allow in vivo microscopic examination of skin lesions and is intended to help distinguish between benign and malignant pigmented skin lesions. The technique involves application of immersion oil to the skin, which eliminates light reflection from the skin surface and renders the stratum corneum transparent. Using a magnifying lens, the structures of the epidermis and epidermal-dermal junction can then be visualized. A handheld or stereomicroscope may be used for direct visual examination. Digitization of images, typically after initial visual assessment, permits storage and facilitates their retrieval, is often used for comparison purposes if a lesion is being followed over time.
A variety of dermatoscopic features have been identified that are suggestive of malignancy, including pseudopods, radial streaming, the pattern of the pigment network, and black dots. These features in combination with other standard assessment criteria of pigmented lesions, such as asymmetry; borders; and color, have been organized into algorithms to enhance the differential diagnosis of pigmented skin lesions. Dermatoscopic images may be assessed by direct visual examination or by review of standard or digitized photographs. Digitization of images, either surface or dermatoscopic images, may permit qualitative image enhancement for better visual perception and discrimination of certain features, or actual computer-assisted diagnosis.
Interpretation of dermatoscopy findings have evolved over time. Initially, lesions were evaluated using pattern analysis. More recently several algorithms were developed, including the asymmetry, border, color, and dermatoscopic (ABCD) structures rule of dermatoscopy, the 3-point and 7-point checklists of dermatoscopy by Argenziano, the Menzies method, and the color, architecture, symmetry, homogeneity (CASH) algorithm. (1) There remains a lack of consensus in the literature regarding the optimal dermatoscopic criteria for malignancy.
Dermatoscopy is also proposed in the serial assessment of lesions over time and for defining peripheral margins prior to surgical excision of skin tumors.
Computer-based optical diagnostic devices
A U.S. Food and Drug Administration (FDA)-approved multispectral digital skin lesion analysis (MSDSLA) device uses a handheld scanner to shine visible light on the suspicious lesion. The light is of 10 wavelengths, varying from blue (430 nm) and near infrared (950 nm). The light can penetrate up to 2.5 mm under the surface of the skin. The data acquired by the scanner are analyzed by a data processor; the characteristics of each lesion are evaluated using proprietary computer algorithms. Lesions are classified as positive (i.e., high degree of morphologic disorganization) or negative (i.e., low degree of morphologic disorganization) according to the algorithms. Positive lesions are recommended for biopsy. For negative lesions, other clinical factors are considered in the decision of whether or not to refer to biopsy. The FDA-approved system (see additional details in the Regulatory Status section) is intended only for suspicious pigmented lesions on intact skin and for use only by trained dermatologists.
Dermatoscopic devices cleared by the U.S. Food and Drug Administration (FDA) include:
• Episcope™ (Welch Allyn, Inc., Skaneateles Falls, NY) approved in 1995; intended use is to illuminate body surfaces and cavities during medical examination.
• Nevoscope™ (TRANSLITE, Sugar Land, TX) approved in 1996; intended use is to view skin lesions by either illumination or transillumination.
• Dermascope™ (American Diagnostic Corp., Hauppauge, NY) approved in 1999; intended use is to enlarge images for medical purposes.
• MoleMax™ (Derma Instruments, Austria) approved in 1999; intended use is to enlarge images for medical purposes.
• VivaScope® System (Lucid, Inc.) (reflectance confocal microscopy [RCM]) approved in 2008, intended use is to acquire, store, retrieve, display and transfer in vivo images of tissue, including blood, collagen and pigment, in exposed unstained epithelium and the supporting stroma for review by physicians to assist in forming a clinical judgment. (25)
One computer-based optical imaging device has been cleared by the FDA: MelaFind (MelaSciences, Inc. Irvington, NY) was approved in November 2011. Its intended use is to evaluate pigmented lesions with clinical or histological characteristics suggestive of melanoma. It is not intended for lesions with a diagnosis of melanoma or likely melanoma. MelaFind is intended for use only by physicians trained in the clinical diagnosis and management of skin cancer (i.e., dermatologists) and only those who have additionally successfully completed training on the MelaFind device. FDA documents further note:
“MelaFind is indicated only for use on lesions with a diameter between 2 mm and 22 mm, lesions that are accessible by the MelaFind imager, lesions that are sufficiently pigmented (i.e., not for use on non-pigmented or skin-colored lesions), lesions that do not contain a scar or fibrosis consistent with previous trauma, lesions where the skin is intact (i.e., non-ulcerated or non-bleeding lesions), lesions greater than 1 cm away from the eye, lesions which do not contain foreign matter, and lesions not on special anatomic sites (i.e., not for use on acral, palmar, plantar, mucosal, or subungual areas). MelaFind is not designed to detect pigmented non-melanoma skin cancers, so the dermatologist should rely on clinical experience to diagnose such lesions.”
As with any diagnostic tool, assessment of dermatoscopy involves a determination of its sensitivity, specificity, and positive and negative predictive values in different populations compared to a reference standard and whether the results of the diagnostic tests are ultimately used to benefit health outcomes. The reference standard for evaluation of pigmented skin lesions is excision with histologic diagnosis, in which, depending on the skill of the pathologist, sensitivity and specificity are considered near 100%. The relevant health outcome is early diagnosis of a malignancy. Clinically, dermatoscopy is used in combination with clinical assessment, either based on direct visual inspection or review of photographs. Therefore, the diagnostic performance of dermatoscopy combined with clinical assessment must be compared with clinical assessment alone and then compared to the reference standard of histology. There are 4 general clinical situations in which dermatoscopy might be of benefit.
1. When patients present with a lesion with a low pretest possibility of malignancy, dermatoscopy could potentially be used to determine which lesions did not require excision, i.e., a deselection process. In this clinical situation, the negative predictive value of dermatoscopy is the most relevant diagnostic parameter.
2. Some patients may present with multiple suspicious pigmented skin lesions such that excision of all or even some of them is not possible. In this clinical situation, a determination must be made which of the lesions is most clinically suspicious and requires excision. In this setting, the positive predictive value of dermatoscopy is the most relevant diagnostic parameter.
3. Serial assessment of lesions over time, as a technique to prompt excision when a lesion changes shape or color, is commonly performed in patients with multiple pigmented lesions or for lesions in locations difficult to excise. Serial conventional and digital photography has been used for this purpose. Both the positive and negative predictive values of results are relevant.
4. Use in defining peripheral borders of basal cell or squamous cell cancers to guide surgery. If dermatoscopy combined with clinical assessment is more accurate than clinical assessment alone in defining tumor borders, then it might be possible to excise the tumor with a narrower margin, thus preserving a larger amount of normal skin.
This policy was originally created in 2003 and was updated regularly with searches of the MEDLINE database. This section of the current policy has been substantially revised. Following is a summary of the key literature to date.
Dermatoscopy for Selecting or Deselecting Lesions for Excision
Diagnostic Accuracy Compared With Naked Eye Examination
A number of studies have reported on the diagnostic accuracy of dermatoscopy compared with clinical assessment, with histologic examination serving as the reference standard, and several meta-analyses have been published. In 2008, Vestergaard et al. reviewed the literature on the accuracy of dermatoscopy for the diagnosis of melanoma compared with naked eye examination. (2) Nine studies met the inclusion criteria; 2 were randomized controlled trials (RCTs) and the other 7 used a cross-sectional design. All were performed in an expert setting. There was variability across the studies for patient and lesion selections, naked eye criteria for melanoma, dermatoscopy criteria for melanoma, and follow-up. Hierarchical summary receiver operator curve (ROC) analysis was used to estimate the relative diagnostic accuracy for clinical examination with, and without, the use of dermatoscopy. The pooled relative diagnostic odds ratio (OR) for melanoma, for dermatoscopy compared with naked eye examination, was 15.6 (range, 2.9-83.7). Removal of 2 small outlier studies changed this to 9.0 (range, 1.5-54.6) but the odds of identifying melanoma remained higher with dermatoscopy. The authors concluded that dermatoscopy is more accurate than naked eye examination for the diagnosis of cutaneous melanoma in suspicious skin lesions performed in the clinical setting.
A 2009 meta-analysis by Rajpara et al. reviewed studies on dermatoscopy using a handheld dermatoscope, as well as studies on digital dermatoscopy with computer-aided diagnosis (CAD). (3) The latter technique was called artificial intelligence in the article.) The studies could be prospective or retrospective, evaluated dermatoscopy performed by experts, and used histology of excised lesions as the reference standard. Studies were not required to compare dermatoscopy with naked eye examination; thus, the study was not able to compare the diagnostic accuracy of dermatoscopy or digital dermatoscopy plus CAD to clinical examination. The investigators identified 30 studies; all but one of which were from Europe. A total of 9784 melanoma lesions were included in the review; of these, 8045 were analyzed by dermatoscopy and 2420 by CAD. The investigators conducted pooled analyses of studies, grouping them by the type of algorithm used for diagnosis (e.g., pattern analysis, asymmetry, border, color, dermatoscopic structures [ABCD] rule). The pooled sensitivity for dermatoscopy (30 analyses) was 88% (95% confidence interval [CI], 87% to 0.89%), and the pooled specificity was 86% (95% CI, 85% to 86%). For digital dermatoscopy with CAD, the pooled sensitivity was (12 analyses) 91% (95% CI, 88% to 93%), and the pooled specificity was 79% (95% CI, 77% to 81%). The pooled specificity of the CAD diagnosis was significantly lower than the dermatoscopy analysis; pooled sensitivities did not differ significantly. There were no significant differences in overall diagnostic performance of different algorithms. The authors noted that, whereas dermatoscopy has been used by trained clinicians in a practice setting, CAD has only been used in experimental settings using preselected lesions.
A representative review of recent studies follows.
In 2014, Koelink et al. conducted a cluster randomized trial on use of dermatoscopy in primary care. (4) Patients from 48 general practices in the Netherlands were randomized to an intervention group using dermatoscopy in addition to naked eye examination or a control group using only naked eye examination. Eligibility included being at least 18 years old and presenting with a suspicious skin lesion. All patients underwent routine naked eye examination and those in intervention practices were also examined with a dermatoscope. The reference standard, in hierarchical order, for diagnosis, was pathologic diagnosis for excised lesions, diagnosis by a dermatologist for referrals, or diagnosis by trial dermatologist. A total of 170 patients in 26 practices were included in the intervention group and 211 patients in 22 practices were included in the control group. Overall, the percentage of correctly diagnosed lesions was 98 of 194 (50.5%) in the intervention group and 90 of 222 (40.5%) in the control group. The OR of a correct diagnosis in the intervention group compared with the control group was 1.51 (95% CI, 0.96 to 2.37). Among melanoma lesions, the percentage of correct diagnoses was 8 of 13 (61.5%) in the intervention group and 2 of 9 (22.2%) in the control group (OR=5.52; 95% CI, 0.76 to 39.91). Neither of the differences was statistically significant. Because the study included relatively few melanoma lesions, there was a lack of precision in estimating the percentage of correct diagnoses.
Also in 2014, Unlu et al. published a comparison of dermatoscopic diagnostic algorithms and clinical assessment using histological diagnosis as the reference standard. (5) The study included 115 images of suspicious lesions. Three experienced dermatoscopists classified each of the lesions, in random order, as benign or malignant according to each of 4 algorithms. They were the ABCD rule, the 7-point checklist, 3-point checklist, and the color, architecture, symmetry, homogeneity (CASH) algorithm. The history and macroscopic images of the lesions were not provided to the dermatoscopists to avoid recall bias. According to histopathologic criteria, 24 lesions (20.9%) were classified as melanomas. A total of 18 (75%) of melanomas were correctly classified by clinical examination. In comparison, 22 (92%) of malignant lesions were correctly classified by the ABCD rule of dermatoscopy, 21 (88%) by the 7-point checklist, 19 (79%) by the 3-point checklist, and 22 (92%) by the CASH algorithm. All melanomas with a Breslow thickness of at least 0.75 mm were diagnosed correctly by the ABCD rule and the CASH algorithm. Overall, clinical examination had a sensitivity of 75% and specificity of 57%. The sensitivity and specificity of the dermatoscopic algorithms were 91.6% and 60.4% for the ABCD rule, 87.5% and 65.9% for the 7-point checklist, 79.1% and 62.6% for the 3-point checklist, and 91.6% and 64.8% for the CASH algorithm.
In 2011, De Giorgi et al. in Italy randomly selected 8 dermatologists who had attended a basic dermatoscopy course 6 months previously; none had extensive experience using dermatoscopy. (6) Each dermatologist was asked to examine separately clinical images only and then a combination of clinical images and dermatoscopic images of 200 melanocytic skin lesions (mean diameter <8.00 mm). All lesions had been histopathologically reviewed by a pathologist. Clinical images had been obtained with a digital camera, and dermatoscopy pictures were obtained using a dermatoscope. The dermatologists were asked to determine whether they thought the sample was a melanoma lesion (yes/no). Histopathologic diagnosis was used as the reference standard. Mean sensitivity was significantly increased when the clinician reviewed dermatoscopic images in addition to clinical images; specificity did not significantly change. Mean sensitivity and specificity of melanoma diagnosis using clinical image examination alone were 71.2% and 80.2%, respectively; using the combined examination, they were 84.1% and 80.2%, respectively. The authors pointed out, unlike actual clinical practice, dermatologists were not given information about the lesion history and were unable to examine other lesions from the same patient. In addition, while reviewing the dermatoscopy images, the dermatologists were also reviewing the clinical images for the second time.
A 2011 study by Rosendahl et al. analyzed a consecutive series of 463 pigmented lesions from a single center in Australia. (7) All lesions had been photographed, and dermatoscopic images had been taken prior to excision. Histopathology was used as the diagnostic reference standard. Lesions were categorized as benign or malignant; the latter category consisted of melanomas, basal cell carcinomas (BCCs), and squamous cell carcinomas (SCCs). The process of analysis consisted of presenting 2 clinical images of each lesion (overview, close-up) to a blinded reviewer who then made a diagnosis. The reviewer was then shown the dermatoscopic image and asked to give another diagnosis. Histopathologically, 246 of 463 (53.1%) of the lesions were melanocytic, and a total of 138 (30%) lesions were malignant. The reviewer’s diagnosis matched the histopathologic diagnosis in 320 (69.1%) of cases using clinical images alone and in 375 (80.1%) of cases using clinical images and dermatoscopic images. At a fixed specificity of 80%, the sensitivity was 70.5% without dermatoscopic images and 82.6% with dermatoscopic images. ROC area under the curve (AUC) analysis was also done to evaluate diagnostic accuracy. The AUC was significantly higher with dermatoscopy (0.89) than without dermatoscopy (0.83; p<0.001). When melanocytic and non?melanocytic lesions were examined separately, the difference in the AUC with and without dermatoscopy was statistically significant only for the melanocytic lesions (0.91 vs 0.84, respectively, p<0.001).
Section Summary: Diagnostic Accuracy of Selecting or Deselecting Lesions
Recent meta-analyses found that, overall; the diagnostic accuracy of dermatoscopy was higher than clinical assessment/naked eye examination. However, most published studies are retrospective, reported on the performance of clinicians who have extensive experience with dermatoscopic imaging, and were conducted outside of the United States. There is a lack of consensus about a standard approach to evaluating dermatoscopic images, although a 2009 meta-analysis and a 2014 study found several approaches may have similar diagnostic accuracy.
Impact on Patient Management or Health Outcomes
Several studies have evaluated the impact of dermatoscopy on patient management, including 2 RCTs. In 2006, an RCT by Argenziano et al. addressed whether dermatoscopy improves the accuracy of primary care physicians in triaging lesions suggestive of malignancy. (8) A total of 73 primary care physicians underwent a 1-day training course in dermatoscopy and were randomized to conduct examinations using naked eye examination only or naked eye examination plus dermatoscopy. Following the primary care evaluation, patients were reevaluated by dermatologists who were expert in melanoma and all lesions considered suggestive of skin cancer were excised. Over a 16-month period, 1345 patients were evaluated using naked eye examination and 1197 also underwent dermatoscopy. The primary study outcome was referral accuracy. Physicians in both groups referred a similar proportion of patients to a specialty clinic, 30.3% in the naked eye only group and 31.5% in the dermatoscopy group (p=0.787). In their reexaminations, dermatologists considered 6.3% of lesions in the naked eye only group and 6.4% in the dermatoscopy group to be suspicious for skin cancer. The positive predictive value (PPV) of primary care physician recommendations was low in both groups, 11.3% in the naked eye only group and 16.1% in the dermatoscopy group. However, the negative predictive value (NPV), the more clinically relevant outcome in this situation, was relatively high in both groups and was significantly higher in the dermatoscopy group than in the naked eye only group (98.1% vs 95.8%; p=0.004). According to histopathologic analysis of equivocal lesions, 23 malignant lesions were missed by naked eye examination alone versus 6 by dermatoscopy; the difference between groups was statistically significant (p=0.002).
In 2004, Carli et al. published an RCT that included 913 consecutive patients referred to a pigmented lesion clinic in Italy for evaluation of skin lesions. (9) A total of 302 participants were randomized to standard naked eye examination and 311 to naked eye examination with the possibility of dermatoscopy at the clinician’s discretion. In both groups, there was mandatory excision of equivocal lesions. (A third study arm involved the option of digital follow-up without immediate excision.) Examinations were done by experienced dermatologists with expertise in dermatoscopy. In the group that could use dermatoscopy, the number of lesions initially classified as suggestive or equivocal by naked eye examination was 158 of 311 (50.8%). After dermatoscopy, 28 of these 158 lesions (17.8%) were classified as suggestive or equivocal and were referred for excision. Thus, in the dermatoscopy group, a total of 28 of 311 (9.0%) lesions were referred for excision. The proportion of referrals was significantly lower than the naked eye only examination group, in which 47 of 302 (15.6%) lesions were referred for excision (p=0.013). Histologic analysis of excised lesions identified 3 melanomas in the naked eye examination only group and 2 in the combined examination group; the difference between groups was not statistically significant. No unexcised melanomas were identified in the 103 of 121 (85%) patients with clinically suspicious but dermatologically negative lesions who agreed to be reexamined several months later.
In addition to the RCTs, the 2011 study by De Giorgi et al. (previously described) addressed whether dermatoscopy leads to improved patient management. (6) The study asked dermatologists to decide whether they would recommend excision of lesions based on clinical images only or based on a combination of clinical images and dermatoscopic images. Dermatologists were told to simulate their practice setting and to attempt to minimize the number of negative lesions. Sensitivity and specificity were calculated based on whether any melanoma lesions would remain unexcised, with histopathologic findings as the reference standard. Mean sensitivity and specificity of the decision to excise using clinical image examination alone were 94.1% and 36.1%, respectively, and using the combined examination were 98.6% and 31.5%, respectively. Sensitivity was significantly higher when dermatologic images were available in addition to clinical images (p<0.003); there was no statistically significant difference in specificity.
Section Summary: Impact on Patient Management of Selecting or Deselecting Lesions
Several studies, including 2 RCTs, have evaluated the impact of dermatoscopy on patient management. One RCT found a significantly lower rate of excision recommendations when dermatologists had access to dermatoscopy compared with naked eye examination alone. Another RCT found that primary care physicians did not refer fewer patients to specialists when used dermatoscopy in addition to naked eye examination, but the NPV, a clinically relevant outcome, was significantly higher with dermatoscopy.
Dermatoscopy for Evaluation of Multiple Suspicious Pigmented Lesions
No studies were found that specifically addressed the issue of dermatoscopy for patients with multiple suspicious pigmented lesions to determine which lesions are most clinically suspicious and therefore require excision.
Dermatoscopy for Serial Assessments of Lesions
No prospective comparative studies were identified that compared outcomes after managing patients over time with and without dermatoscopy. A meta-analysis of data from noncomparative studies was published in 2013 by Salerni et al. (10) The authors identified 14 studies performed in a clinical setting. The studies included 5787 patients with a total of 52,739 lesions that were monitored using dermatoscopy (mean, 12 lesions per patient). Patients were followed for a mean of 30 months. During follow-up, the percentage of lesions excised per study ranged from 1.3% to 18.7%. A total of 4388 lesions were excised (8.3%). There were 383 melanomas detected (<1% of lesions that were being followed). Of the melanomas detected, 209 (55%) were in situ and 174 (45%) were invasive. The meta-analysis did not evaluate data on dermatoscopy compared with another technique for monitoring patients.
One study, published in 2009 by Menzies et al., compared an initial patient management decision with naked eye evaluation or dermatoscopy and then followed patients over time with short-term sequential digital dermatoscopy imaging (SDDI) (i.e., every 3 months). (11) The study was conducted in a general practice setting in Australia. Participating physicians were trained in the use of dermatoscopy with SDDI during a 2-hour workshop and online training. Seventy-four physicians completed the training, and 63 of these (85%) then assessed 374 lesions (median, 6 lesions per physician). Based on clinical assessment with the naked eye alone, all 374 lesions were assessed as requiring excision or referral. With dermatoscopy, lesions were triaged to 3 groups: 110 received immediate referral or excision, 192 were assigned to close follow-up with SDDI, and 72 were assigned to observation for change. The 192 SDDI lesions were reevaluated 3 months later. At that time, 46 lesions were referred/excised, 6 were triaged to continue SDDI, and 140 were triaged to standard observation. At the third visit (6 months after the initial visit), referral/excision was recommended for 2 of the 6 SDDI lesions, and the other 4 returned to standard care. In addition, 5 of the lesions previously recommended for observation were triaged to referral/excision. Thus, in this group of 374 lesions that would all have been recommended for referral/excision with clinical examination alone, the combination dermatoscopy plus SSDI intervention reduced the number of referrals/excisions by about half, to 163 (44%) of lesions. It is not known how many of the patients triaged to referral or excision would ultimately have had a biopsy.
Dermatoscopy for Defining Peripheral Margins of Cancerous Skin Lesions Before Surgery Impact on Patient Management or Health Outcomes
One RCT was identified that compared dermatoscopy with other methods of defining peripheral margins. (12) This was a 2013 trial published by Asilian and Momeni in which 60 patients with BCC in the head and neck area were randomized to naked eye examination (n=20), dermoscopy (n=20), or curettage (n=20) to determine the extent of tumor extension before Mohs micrographic surgery. In all patients, a 3-mm border was initially resected after the tumor margin was determined. If resection was found to be incomplete, patients received additional stages of Mohs surgery. The mean (SD) number of Mohs surgery resection stages, the study’s primary outcome, was 1.90 (0.55) in the curettage group, 1.55 (0.51) in the visual inspection group, and 1.65 (0.49) in the dermoscopy group. The difference between groups was not statistically significant (p=0.10). Health outcomes such as rates of recurrence or mortality rates were not reported.
A prospective nonrandomized study was published by Suzuki et al. in 2014. (13) The study included 44 patients with melanoma and indications for Mohs micrographic surgery. All patients were assessed with naked eye examination and had surgical margins demarcated in a blue or black marker. The first 21 patients referred for surgery received only this naked eye examination and the remaining 223 patients were also assessed using dermatoscopy (margins drawn in red marker). Outcomes did not differ significantly in the 2 groups (e.g., Mohs surgery required a similar number of stages).
Several studies conducted in Italy have evaluated dermatoscopy used to define peripheral borders of skin tumors to guide surgical excision. All were nonrandomized comparisons between clinical and dermatoscopic evaluation of suspected tumor margins. Most recently in 2012, Carducci et al. evaluated outcomes in 94 patients with a suspected clinical diagnosis of SCC. (14) Prior to surgery, margins in 46 patients were determined by clinical evaluation and margins in 48 patients were determined with digital dermatoscopy. A lateral margin of 4 to 6 mm was chosen for SCC not located on the scalp, ears, eyelids, nose, or lips. For lesions in those areas, margins of 6 to 10 mm were used. In the dermatoscopy group, clinical margins were first defined and outlined with a dermographic pencil. Then, dermatoscopy was performed, and the margins were redefined if pictures found that the margins were too near the pencil line. Histologic analysis of specimens was the reference standard. In the clinical evaluation group, 8 of 46 (17%) specimens showed incomplete margin excision compared with 3 of 48 (6%) in the digital dermatoscopy group. The difference between groups was statistically significant (p=0.015). The study was not randomized; the clinical evaluation group included patients who were evaluated before the introduction of digital dermatoscopy in that medical center.
In 2011, the Carducci research group published a similar study in patients with a suspected diagnosis of BCC of the head or neck. (15) A total of 84 patients were included. Lesions were examined either clinically or with digital dermatoscopy to determine margins. Surgical excision was undertaken with a 3-mm surgical margin. Margin involvement was found in 8 of 40 (20%) histologic specimens excised after clinical evaluation and 3 of 44 (7%) specimens excised after dermatoscopic detection of margins; this difference was statistically significant (p<0.007). Seven of the 11 (64%) specimens with margin involvement were nodular BCCs. Neither of the Carducci studies followed patients after surgical excision or reported health outcomes. Both studies used a digital Videocap dermatoscope, which has not been cleared for use in the United States.
In 2010, Caresana and Giardini studied 200 consecutive patients with BCC using 2-mm excision margins. (16) The margins were first marked using naked eye only, and then borders were confirmed using dermatoscopy. (The type of device used was not specified.) There was concordance in the peripheral margins drawn using the naked eye and dermatoscopy in 131 of 200 (66%) cases. In 69 cases, there was a larger margin with dermatoscopy, but this did not exceed 1 mm more than the clinical measurement in 55 (80%) of the 69 cases. According to histologic analysis, surgical excision using the 2-mm margin was found to be adequate in 197 of the 200 cases. After 10 to 30 months of follow-up, none of the 200 treated cases had signs or symptoms of recurrence. Because surgery was performed using the margins drawn with dermatoscopy in all cases, the study could not compare margins drawn using naked eye (clinical) assessment plus dermatoscopy with clinical assessment alone.
Section Summary: Defining Peripheral Margins
Only 1 published RCT comparing margins drawn with and without the aid of dermatoscopy, and that study does not report superior outcomes using dermatoscopy compared with visual inspection or curettage. This RCT and other available published studies provide limited information on health outcomes. The published studies are all conducted outside of the United States and at least 2 did not use U.S. Food and Drug Administration (FDA) ? approved devices.
Computer-Based Optical Diagnostic Device for Selecting or Deselecting Lesions for Excision
Diagnostic Accuracy Compared With Naked Eye Examination
One published prospective study identified evaluated the diagnostic performance of MelaFind, an FDA-approved computer-based optical diagnostic device. This 2011 industry-sponsored study by Monheit et al. included the data submitted to the FDA in the application for approval of the device. (17) The study included patients with at least 1 pigmented lesion scheduled for first-time biopsy. Lesions were between 2 mm and 22 mm in diameter. The following were exclusion criteria: anatomic site was not accessible to the device; lesion was not intact (e.g., open sores, ulcers, bleeding); lesion was on a palmar, plantar, or mucosal surface or under nails; lesion was in an area of visible scarring; and the lesion contained tattoo ink, splinter, or other foreign matter. In addition, lesions with a prebiopsy diagnosis of melanoma were excluded from the analysis. Histologic diagnosis was used as the reference standard.
A total of 1393 patients with 1831 lesions were enrolled. Of the 1831 lesions, 1632 (90%) were eligible and evaluable. There were 165 lesions not evaluable for various reasons including operator error and camera malfunction and ineligibility postenrollment related to scarring. Histologic analysis determined that 127 of 1632 lesions (7.8%) were melanoma. The sensitivity of MelaFind for recommending biopsy of melanomas was 98.2% (125/127 melanomas), with a 95% lower CI bound of 95.6%. The average specificity (averaged over clinicians) of MelaFind for melanoma was 9.5%. The accuracy of clinician diagnosis was determined by randomly selecting 25 melanoma cases and matching them with 25 nonmelanoma lesions. Clinicians were asked to classify the lesions into categories of melanoma: cannot rule out melanoma or not melanoma. The specificity of clinician diagnosis, as determined by the proportion of melanomas among the total number of lesions recommended for biopsy, was 3.7%, which was significantly lower than the specificity for MelaFind (p=0.02).
Using data from the industry-sponsored FDA-approval study, Wells et al. evaluated the diagnostic accuracy of MelaFind compared with the opinion of dermatologists. (18) A convenience sample of 39 dermatologists who had expressed interest in the MelaFind technology participated. The study was conducted over the Internet. A total of 47 lesions (23 malignant melanomas, 24 benign lesions) were randomly selected from the repository of lesions collected by MELA Sciences. Cases may have overlapped with the data used in the Monheit study, previously described. (17) Dermatologists were given images of the lesions taken before biopsy and case histories, but were not given MelaFind recommendations. Participants were asked whether they would recommend biopsy. MelaFind recommended biopsy of 22 of 23 melanoma lesions (sensitivity, 96%; lower limit of 95% CI, 83%). The average biopsy sensitivity for dermatologists was 80% (95% CI, 72% to 87%). Regarding specificity, MelaFind did not recommend biopsy for 2 of 24 benign lesions (specificity, 8%; 95% CI, 1% to 25%). In contrast, the biopsy specificity was 43% for dermatologists. In this study, the specificity of MelaFind was very low, i.e., findings suggested biopsy was needed for 22 of 24 benign lesions and the specificity of dermatologists’ reading was higher than in the Monheit study. Limitations in methodology include conducting the study over the Internet and clinician inability to view lesions. Also, clinicians participating may not have been representative of typical dermatologists, because this sample expressed interested in MelaFind and tested this technology in company-sponsored research.
Section Summary: Diagnostic Accuracy of Computer-Based Optical Devices
Only 1 published study has evaluated the accuracy of a computer-based optical diagnostic device. The study found that MelaFind was able to correctly identify 125 of 127 melanomas among evaluable samples; 10% of samples were not evaluable. One simulation study with a number of potential biases evaluated the potential impact on MelaFind on patient management decisions. The evidence is insufficient for evaluating the added benefit of using computer-based optical devices compared with clinical examination for selecting suspicious lesions for excision.
Impact on Patient Management or Health Outcomes
A 2012 study by Rigel et al. reported results of a simulation exercise with dermatologists attending an educational conference. (19) A total of 179 practicing dermatologists participated in the exercise. They were asked to evaluate lesions before and after receiving information from multispectral digital skin lesion analysis using the MelaFind device to indicate whether they would biopsy the lesion. There were 24 lesions, 5 known to be melanomas and 19 nonmelanoma pigmented lesions. Before information from the computer-based system, 13% of participants said they would biopsy all 5 lesions; this rose to 70% after evaluation by the MelaFind system. The authors reported that the average biopsy sensitivity for the 5 melanoma lesions was 69% before receiving information from MelaFind and 94% afterwards. In addition, the biopsy specificity was 54% before information from MelaFind and 40% afterwards. Exact numbers were not reported. Potential biases in this analysis include study design (simulation exercise) that may not reflect clinical practice and study location (the exercise occurred at a meeting where sponsorship was likely obvious). In addition, along with the information from MelaFind, participants were evaluating the lesion for the second time, and this reexamination might have affected biopsy recommendations.
Computer-Based Optical Imaging Devices for Serial Assessments of Lesions
No published studies were identified that addressed this topic.
Computer-Based Optical Imaging Devices for Defining Peripheral Margins of Cancerous Skin Lesions Before Surgery
No published studies were identified that addressed this topic.
Ongoing and Unpublished Clinical Trials
Some currently unpublished trials that might influence this review are listed in Table 1.
Table 1. Summary of Key Trials
VivaNet Study. A Multicenter Study of Confocal Reflectance Microscopy in Telemedicine (20)
Post-Approval Study of MelaFind (21)
NCT: national clinical trial
a Denotes industry-sponsored or cosponsored trial
Summary of Evidence
The evidence for dermatoscopy in patients who have lesions suspicious of melanoma includes a number of diagnostic accuracy studies and several meta-analyses. Relevant outcomes are overall survival, disease-specific survival, test accuracy, and change in disease status. The literature suggests that dermatoscopy is more accurate than naked eye examination when used in the expert clinical setting. The available evidence from prospective randomized controlled trials (RCTs) and other studies suggests that dermatoscopy used by specialists may lead to a decrease in the number of benign lesions excised and, when used by primary care physicians, may lead to fewer benign lesions being referred to specialists. The number of studies on the impact of dermatoscopy on patient management and clinical outcomes remains limited. The evidence is insufficient to determine the effects of the technology on health outcomes.
The evidence for computer-based optical diagnostic devices in patients who have lesions suspicious of melanoma includes several prospective diagnostic accuracy studies and a simulation study. Relevant outcomes are overall survival, disease-specific survival, test accuracy, and change in disease status. In the diagnostic accuracy study, 10% of samples were not evaluable and the simulation study had a number of potential biases. There are no studies comparing patient management decisions and health outcomes with and without these devices. The evidence is insufficient to determine the effects of the technology on health outcomes.
The evidence for dermatoscopy in patients who have pigmented lesions being monitored for suspicious changes consists of noncomparative studies. Relevant outcomes are overall survival, disease-specific survival, test accuracy, and change in disease status. The available does not clearly indicate that dermatoscopy results in better patient management decisions. The evidence is insufficient to determine the effects of the technology on health outcomes.
The evidence for computer-based optical diagnostic device in patients who have pigmented lesions being monitored for suspicious changes includes no published studies. Relevant outcomes are overall survival, disease-specific survival, test accuracy, and change in disease status. The evidence is insufficient to determine the effects of the technology on health outcomes.
The evidence for dermatoscopy and computer-based optical diagnostic devices in patients who have cancerous skin lesions referred for surgery includes 1 RCT and several observational studies. Relevant outcomes are overall survival, disease-specific survival, and treatment-related morbidity. The single RCT did not report superior outcomes using dermatoscopy compared with visual inspection or curettage. The published studies were all conducted outside of the United States and at least 2 did not use U.S. Food and Drug Administration?approved devices. None addressed computer-based optical devices. The evidence is insufficient to determine the effects of the technology on health outcomes.
Practice Guidelines and Position Statements
International Dermoscopy Society
In July 2007, the International Dermoscopy Society embarked on creating a consensus document for the standardization and recommended criteria necessary to effectively convey dermatoscopic findings to consulting physicians and colleagues. (22) The final items included in the document are as follows:
1. Pertinent personal and family history (recommended);
2. Clinical description of the lesion (recommended);
3. The 2-step method of dermatoscopy differentiating melanocytic from nonmelanocytic tumors (recommended);
4. The use of standardized terms to describe structures (recommended);
5. The dermatoscopic algorithm used (optional);
6. Information on the imaging equipment and magnification (recommended);
7. Clinical and dermatoscopic images of the tumor (recommended);
8. A diagnosis or differential diagnosis (recommended);
9. Decision concerning the management (recommended);
10. Specific comments for the pathologist when excision and histopathologic examination are recommended (optional).
National Comprehensive Cancer Network
The National Comprehensive Cancer Network melanoma guideline (23) does not mention dermatoscopy or dermoscopy. Biopsy is recommended for suspicious pigmented lesions.
American Academy of Dermatology
The American Academy of Dermatology 2011 guidelines for care and treatment of melanoma (24) do not mention dermatoscopy, e.g., in the discussion of determining surgical margins before surgery. The guidelines did not address evaluation of suspicious lesions.
Each benefit plan, summary plan description or contract defines which services are covered, which services are excluded, and which services are subject to dollar caps or other limitations, conditions or exclusions. Members and their providers have the responsibility for consulting the member's benefit plan, summary plan description or contract to determine if there are any exclusions or other benefit limitations applicable to this service or supply. If there is a discrepancy between a Medical Policy and a member's benefit plan, summary plan description or contract, the benefit plan, summary plan description or contract will govern.
Disclaimer for coding information on Medical Policies
Procedure and diagnosis codes on Medical Policy documents are included only as a general reference tool for each policy. They may not be all-inclusive.
The presence or absence of procedure, service, supply, device or diagnosis codes in a Medical Policy document has no relevance for determination of benefit coverage for members or reimbursement for providers. Only the written coverage position in a medical policy should be used for such determinations.
Benefit coverage determinations based on written Medical Policy coverage positions must include review of the member’s benefit contract or Summary Plan Description (SPD) for defined coverage vs. non-coverage, benefit exclusions, and benefit limitations such as dollar or duration caps.
The following codes may be applicable to this Medical policy and may not be all inclusive.
96904, 96931, 96932, 96933, 96934, 96935, 96936, 0400T, 0401T
ICD-9 Diagnosis Codes
Refer to the ICD-9-CM manual
ICD-9 Procedure Codes
Refer to the ICD-9-CM manual
ICD-10 Diagnosis Codes
Refer to the ICD-10-CM manual
ICD-10 Procedure Codes
Refer to the ICD-10-CM manual
The information contained in this section is for informational purposes only. HCSC makes no representation as to the accuracy of this information. It is not to be used for claims adjudication for HCSC Plans.
The Centers for Medicare and Medicaid Services (CMS) does not have a national Medicare coverage position. Coverage may be subject to local carrier discretion.
A national coverage position for Medicare may have been developed since this medical policy document was written. See Medicare's National Coverage at <http://www.cms.hhs.gov>.
1. Kardynal A, Olszewska M. Modern non-invasive diagnostic techniques in the detection of early cutaneous melanoma. J Dermatol Case Rep. Mar 31 2014; 8(1):1-8. PMID 24748903
2. Vestergaard ME, Macaskill P, Holt PE, et al. Dermoscopy compared with naked eye examination for the diagnosis of primary melanoma: a meta-analysis of studies performed in a clinical setting. Br J Dermatol. 2008; 159(3):669-676.
3. Rajpara SM, Botello AP, Townend J, et al. Systematic review of dermoscopy and digital dermoscopy/artificial intelligence for the diagnosis of melanoma. Br J Dermatol. 2009; 161(3):591-604.
4. Koelink CJ, Vermeulen KM, Kollen BJ, et al. Diagnostic accuracy and cost-effectiveness of dermoscopy in primary care: a cluster randomized clinical trial. J Eur Acad Dermatol Venereol. Nov 2014; 28(11):1442-1449. PMID 25493316
5. Unlu E, Akay BN, Erdem C. Comparison of dermatoscopic diagnostic algorithms based on calculation: The ABCD rule of dermatoscopy, the seven-point checklist, the three-point checklist and the CASH algorithm in dermatoscopic evaluation of melanocytic lesions. J Dermatol. Jul 2014; 41(7):598-603. PMID 24807635
6. De Giorgi V, Grazzini M, Rossari S, et al. Adding dermatoscopy to naked eye examination of equivocal melanocytic skin lesions: effect on intention to excise by general dermatologists. Clin Exp Dermatol. 2011; 36(3):255-259.
7. Rosendahl C, Tschandl P, Cameron A, et al. Diagnostic accuracy of dermatoscopy for melonocytic and nonmelanocytic pigmented lesions. J Am Acad Dermatol. 2011; 64(6):1068-1073.
8. Argenziano G, Puig S, Zalaudek I, et al. Dermoscopy improves accuracy of primary care physicians to triage lesions suggestive of skin cancer. J Clin Oncol. Apr 20 2006; 24(12):1877-1882. PMID 16622262
9. Carli P, de Giorgi V, Chiarugi A, et al. Addition of dermoscopy to conventional naked-eye examination in melanoma screening: a randomized study. J Am Acad Dermatol. May 2004; 50(5):683-689. PMID 15097950
10. Salerni G, Teran T, Puig S, et al. Meta-analysis of digital dermoscopy follow-up of melanocytic skin lesions: a study on behalf of the International Dermoscopy Society. J Eur Acad Dermatol Venereol. Jul 2013; 27(7):805- 814. PMID 23181611
11. Menzies SW, Emery J, Staples M, et al. Impact of dermoscopy and short-term sequential digital dermoscopy imaging for the management of pigmented lesions in primary care: a sequential intervention trial. Br J Dermatol. 2009; 161(6):1270-1277.
12. Asilian A, Momeni I. Comparison between examination with naked eye, curretage and dermoscopy in determining tumor extension before Mohs micrographic surgery. Adv Biomed Res. 2013; 2:2. PMID 23930247
13. Suzuki HS, Serafini SZ, Sato MS. Utility of dermoscopy for demarcation of surgical margins in Mohs micrographic surgery. An Bras Dermatol. Jan-Feb 2014; 89(1):38-43. PMID 24626646
14. Carducci M, Bozzetti M, de Marco G, et al. Preoperative margin detection by digital dermoscopy in the traditional surgical excision of cutaneous squamous cell carcinomas. J Dermatolog Treat. Apr 12 2013; 24(3):221-226. PMID 22390630
15. Carducci M, Bozzetti M, Foscolo AM, et al. Margin detection using digital dermatoscopy improves the performance of traditional surgical excision of basal cell carcinomas of the head and neck. Dermatol Surg. 2011; 37(2):280-285.
16. Caresana G, Giardini R. Dermoscopy-guided surgery in basal cell carcinoma. J Eur Acad Dermatol Venereol. 2010; 24(12):1395-1399.
17. Monheit G, Cognetta AB, Ferris L, et al. The performance of MelaFind: a prospective multicenter study. Arch Dermatol. Feb 2011; 147(2):188-194. PMID 20956633
18. Wells R, Gutkowicz-Krusin D, Veledar E, et al. Comparison of diagnostic and management sensitivity to melanoma between dermatologists and MelaFind: a pilot study. Arch Dermatol. Sep 2012; 148(9):1083-1084. PMID 22986873
19. Rigel DS, Roy M, Yoo J, et al. Impact of guidance from a computer-aided multispectral digital skin lesion analysis device on decision to biopsy lesions clinically suggestive of melanoma. Arch Dermatol. Apr 2012; 148(4):541-543. PMID 22351788
20. Sponsored by Lucid Inc. VivaNet Study: a multicenter study of confocal reflectance microscopy in telemedicine (NCT01385943). ClinicalTrials.gov. Accessed July 9, 2015.
21. Sponsored by MELA Sciences. Post-Approval Study of MelaFind (NCT01700114). Available at <http://www.clinicaltrials.gov>. Accessed July 9, 2015.
22. Malvehy J, Puig S, Argenziano G. Dermoscopy report: proposal for standardization. Results of a consensus meeting of the International Dermoscopy Society. J Am Acad Dermatol. 2007; 57(1):84-95.
23. National Comprehensive Cancer Network. Melanoma. Clinical practice guidelines in oncology, V3.2015. Available at: <http://www.nccn.org>. Accessed July 9, 2015.
24. American Acadey of Dermatology. Guidelines for the care and treatment of melanoma. 2011; <http://www.aad.org>. Accessed July 9, 2015.
25. VivaScope® System (K080788) Department of Health and Human Services, U.S. Food and Drug Administration Available at < http://www.accessdata.fda.gov> (accessed May 4, 2016).
26. Kadouch D.J., Schram, M. E., et al. In vivo confocal microscopy of basal cell carcinoma: a systematic review of diagnostic accuracy. J Eur Acad Dermatol Venereol. 2015 Oct; 29(10): 1890-7. PMID 26290493 Available at <http://www.ncbi.nlm.nih.gov> (accessed May 4, 2016).
27. Stevenson A.D., Mickan, S., et al. Systematic review of diagnostic accuracy of reflectance confocal microscopy for melanoma diagnosis in patients with clinically equivocal skin lesions. Dermatol Pract Concept. 2013 Oct 31; 3(4): 19-27. PMID 24282659 Available at <http://www.ncbi.nlm.nih.gov> (accessed May 4, 2016).
28. Optical Diagnostic Devices for Evaluating Skin Lesions Suspected of Malignancy (Archived). Chicago, Illinois: Blue Cross Blue Shield Association Medical policy Reference Manual. (2015 October) Medicine 2.01.42.
|7/1/2016||Document updated with literature review. Coverage unchanged.|
|1/1/2016||Document updated with literature review. In the Coverage section: including, but not limited to reflectance confocal microscopy was added to the following statement: Dermatoscopy, using either direct inspection, digitization of images, or computer-assisted analysis of whole or partial body photography including, but not limited to reflectance confocal microscopy is considered experimental, investigational and/or unproven as a technique to evaluate or serially monitor pigmented skin lesions.|
|4/1/2015||Reviewed. No Changes.|
|2/1/2014||Document title changed from Dermatoscopy. Document updated with literature review. A clarification was added to the first coverage statement to include indicating that dermatoscopy analysis of whole or partial body photography is considered experimental, investigational and unproven as a technique to evaluate or serially monitor pigmented skin lesions. An example was added to the second coverage statement of a computer-based optical imaging device e.g., multispectral digital skin lesion analysis. The following device was added to the third coverage statement: Computer-based optical imaging devices with additional clarification of the experimental investigational and unproven statement to include defining peripheral margins of skin lesions suspected of malignancy prior to surgical excision.|
|8/1/2011||Document updated with literature review. The coverage position was revised as follows: Dermatoscopy by any method (e.g., by direct inspection, digitization of images, ultraviolet photography or computer-assisted analysis is considered experimental, investigational and unproven for any indication, including but not limited to: evaluation or serial monitoring of pigmented skin lesions; or defining peripheral margins of basal cell carcinomas. Rationale completely revised.|
|5/1/2008||Revised/Updated Entire Document|
|3/15/2006||Revised/Updated Entire Document|
|8/15/2003||New Medical Document|
|Title:||Effective Date:||End Date:|
|Optical Diagnostic Devices for Evaluating Skin Lesions Suspected of Malignancy||07-01-2017||07-14-2018|
|Optical Diagnostic Devices for Evaluating Skin Lesions Suspected of Malignancy||05-15-2017||06-30-2017|
|Optical Diagnostic Devices for Evaluating Skin Lesions Suspected of Malignancy||07-01-2016||05-14-2017|
|Optical Diagnostic Devices for Evaluating Skin Lesions Suspected of Malignancy||01-01-2016||06-30-2016|
|Optical Diagnostic Devices for Evaluating Skin Lesions Suspected of Malignancy||04-01-2015||12-31-2015|
|Optical Diagnostic Devices for Evaluating Skin Lesions Suspected of Malignancy||02-01-2014||03-31-2015|