Blue Cross and Blue Shield of Montana (BCBSMT) considers ultrasonographic evaluation as a technique to assess photoaging or skin rejunvenation techniques cosmetic in nature.
BCBSMT considers ultrasonographic evaluation of skin lesions investigational.
Federal mandate prohibits denial of any drug, device or biological product fully approved by the FDA as investigational for the Federal Employee Program (FEP). In these instances coverage of these FDA-approved technologies are reviewed on the basis of medical necessity alone.
This policy does not address the potential use of ultrasonographic detection for subcutaneous lesions including lipomas, epidermal cysts or ganglions or for detecting regional lymph nodes and subcutaneous metastases in patients with melanoma.
Assessment of a diagnostic technology typically focuses on the following three parameters; 1) its technical performance; 2) diagnostic parameters (sensitivity, specificity, and positive and negative predictive values) in different populations of patients, such as those with inflammatory lesions or malignant melanoma; and 3) demonstration that the diagnostic information can be used to improve patient outcomes (clinical utility).
Ultrasonography is a longstanding technology, and its technical performance will not be considered further. The diagnostic performance of ultrasonography is typically compared to histological evaluation of specimens obtained by biopsy or excision, or clinical evaluation. If ultrasonography can accurately predict the thickness of melanotic lesions, it potentially could be used to assist in surgical planning and avoid repeated excisions. For example, the depth of recommended excision of melanoma is based on the thickness of the lesion (i.e., the Breslow index), as measured in the excisional biopsy; the recommended margin of excision is 1 cm for tumors with a Breslow index of 1 mm, 2 cm for those with an index between 1.01 and 4 mm, and 3 cm for those with an index over 4 mm. Therefore, some patients may have to undergo repeated excision if the original margin of excision is inadequate. For those with connective tissue diseases, such as scleroderma, progression of disease is typically assessed clinically. Ultrasonographic evaluation of the dermis could potentially provide a more objective and sensitive method of disease monitoring.
The policy was created with an initial literature search in 2003, at which time the literature was limited to studies evaluating diagnostic accuracy. The policy has been updated regularly with literature searches using MEDLINE®, most recently the literature was searched from July 2009 through August 2010. This section provides a summary of the literature.
A number of studies reporting diagnostic accuracy of ultrasound have been published in a variety of patient populations, primarily including patients with malignant melanoma, inflammatory lesions, or connective tissue disorders. Several studies examined the correlation of the thickness of melanotic lesions as assessed histologically and with ultrasonography. (1-10) In general, studies found a high degree of correlation, although some noted that the ultrasonographic assessment of the thickness of the lesion was often greater than that assessed histologically, perhaps due to shrinkage artifact in the histological specimen, or due to the inability of ultrasonography to distinguish an inflammatory reaction or normal nevus cells from malignant melanocytes. (6, 7) A 2009 systematic review by Machet and colleagues included 14 studies correlating high-resolution ultrasound with histological analysis in melanoma patients. The correlation coefficients in the studies ranged from 0.88 to 0.97 (median of 0.95). (11) Data on the ability of ultrasound thickness to predict adequate surgical margins were available from 7 of the studies, with a total of 860 lesions. The proportion of lesions in the individual studies that was well-classified by ultrasound ranged from 72% to 89%. In addition to the systematic review, Machet and colleagues conducted a prospective, single-center study in France that included 31 patients with suspected or confirmed primary cutaneous melanoma that had not been surgically removed. Average lesion thickness was 1.96 mm according to ultrasound and 1.95 mm by histology. The correlation between ultrasound and histological findings was 94% and it was possible to predict appropriate surgical margins in 84% of patients.
A study that included 57 patients assessed whether ultrasound biomicroscopy (50 MHz) could aid in the diagnosis of 8 non-melanocytic skin disorders. (12) There was a difference in thickness between lesional and normal skin for morphea, seborrheic keratosis, and psoriasis, and a correlation coefficient of 0.002 between the psoriatic area and severity index and ultrasound measurements. Dermal thickness could not be consistently assessed with ultrasound biomicroscopy for eczema, lichen planus, and seborrheic keratosis, and skin thickness was not different between port wine and normal skin.
A study published in 2009 investigated the optimal frequency of ultrasound machines for scanning thin melanocytic skin lesions (MSL). (13) The study included 37 patients with 50 suspicious MSL of maximal vertical tumor thickness less than 1 mm. Compared to histology, 100 MHz was more accurate than 20 MHz, although both overestimated tumor thickness (mean of 16 and 34 micrometer overestimation of tumor thickness, respectively). This study suggests that a higher frequency transducer may be more accurate than a 20-MHz transducer, which was used in many of the previously reported case series.
Several studies have evaluated the role of ultrasound in patient management among patients with skin lesions. A study by Jambusaria-Pahlajani and colleagues included 100 patients with biopsy-proven basal cell carcinoma or squamous cell carcinoma scheduled to undergo Mohs micrographic surgery. (14) Patients received a preoperative high-resolution (40 MHz) ultrasound scan after the surgeon initially drew a proposed surgical margin. The ultrasound technician identified any area of tumor that extended outside the proposed margin, and these areas were verified by histological examination. The sensitivity of ultrasound for correctly identifying areas of tumor extension beyond those proposed by the surgeon was low: 32% (95% CI: 15-54%). Ultrasound was more sensitive for the 43 larger tumors above the median of 1.74 sq cm than for the 41 smaller tumors (55% vs. 33%, respectively). The authors concluded that the sensitivity of high-frequency ultrasound was too low to be clinically useful. They noted, however, that the overall low sensitivity might be due in part from their decision to optimize the image of the dermis with greater resolution than the epidermis, thereby limiting the accuracy of imaging of the epidermis.
Wortsman and Wortsman conducted a retrospective single-center study in Chile. (15) The authors compared ultrasound diagnoses of 4338 skin lesions with clinical diagnosis, using histology as the reference standard. Of the 4338 lesions, 75 (2%) were malignant tumors and 677 (16%) were inflammatory or infectious lesions. (The majority of the skin lesions were benign nonvascular tumors such as enlarged lymph nodes and lipomas). All patients were referred to a department of radiology for further testing; specific reasons for referral were not provided. Clinicians did not have the ultrasound results available at the time of diagnosis, but they did have access to findings from laboratory tests. Ultrasound technicians were aware of the referring diagnosis. The referring diagnosis agreed with the histological diagnosis in 87% of the 75 malignant tumor cases and the addition of ultrasound findings increased the percentage to 91%. The referring diagnosis was correct in 77% of the inflammatory/infectious lesions and ultrasound increased this percentage to 99%. In both types of lesions, the increase in the proportion of correct diagnoses ultrasound was statistically significant (p<0.001). In 735 of the 4338 lesions (17%), including 3 malignant lesions, only ultrasound correctly identified the diagnosis. The authors said that the treatment plans were modified in all of these cases but did not provide details on the modifications. All ultrasound examinations were performed by the same physician which, although increasing the consistency of interpretation, may not be generalizable to findings by other clinicians. As noted above, the study was retrospective; prospective studies evaluating larger numbers of skin conditions relevant to this policy are needed.
An earlier study by Desei and colleagues compared the results from a 20 MHz scanner with clinically and histologically delineated margins for 50 superficial and nodular basal cell carcinomas, excluding those in locations difficult to image. (16) Ultrasound measurements were correlated (r >0.70) with clinical measurements prior to excision, with a mean 20% to 25% increase in size compared to naked eye measurements; 10% of ultrasound measurements showed clinical extension beyond the 4-mm margin. Limitations of this study include that the method for comparing ultrasound findings with preoperative clinical measurements was not clearly described and patient health outcomes were not evaluated.
The evidence is insufficient for determining whether the use of ultrasound leads to improved health outcomes in patients with skin lesions. No study identified examined whether the use of ultrasonography preoperatively resulted in improved health outcomes such as lower rates of disease recurrence or increased survival. Given the lack of sufficient high-quality evidence on the impact of ultrasound skin imaging on patient management, this technology is considered investigational. In addition, due to the cosmetic nature of the application, ultrasound skin imaging is considered not medically necessary to assess photoaging or skin rejuvenation techniques.
Technology Assessments, Guidelines and Position Statements
The National Comprehensive Cancer Network (NCCN) melanoma guideline does not mention use of ultrasonography for evaluating known or suspected melanomas. (17)
Medicare National Coverage
No National Coverage Determination found