The ABCD rule of dermoscopy was the first dermoscopy algorithm created to help differentiate benign from malignant melanocytic tumors. This algorithm, which was described by Stolz, was developed to quantitatively address the crucial question in dermoscopy of whether a melanocytic skin lesion under investigation is benign, suspicious (borderline), or malignant. Based only on four dermatoscopic criteria this method is relatively easy to learn and apply. The ABCD dermoscopy method has been extensively studied and it has been shown that it improves the diagnostic performance of clinicians evaluating pigmented skin lesions. It is the opinion of some that this method may be particularly well suited for clinicians with limited dermoscopy experience. The criteria that combine to create the ABCD rule of dermoscopy are asymmetry, border, color, and differential structures. To utilize these criteria, a scoring system was developed to calculate the total dermoscopy score (TDS) using a linear equation. With this TDS, a grading of lesions is possible with respect to the probability that the lesions under investigation are malignant.
In judging asymmetry, the lesions should be bisected by two 90degree axes that are positioned in such a manner as to bisect the lesion in its most symmetric plane, thereby yielding the lowest possible asymmetry score. It is important to acknowledge that the asymmetry criterion in the ABCD rule incorporates both contour asymmetry and the asymmetry in the distribution of dermoscopic colors and structures. In other words, asymmetry has to be evaluated by taking into account the distribution of colors and structures on either side of each axis, as well as the contour of the lesion. If asymmetry is absent with respect to both axes within the lesion then the asymmetry score is 0. If there is asymmetry in only one axis then the asymmetry score is 1. If there is asymmetry in both axes, then the asymmetry score is 2. It is important to realize that many lesions appearing clinically symmetric may be dermoscopically quite asymmetric by revealing asymmetric distribution of colors and structures.
The asymmetry score (0-2) is multiplied with 1.3 in order to calculate the A contribution to the Total Dermoscopy Score (TDS).
The evaluation of the border score is predicated upon whether there is a sharp, abrupt cutoff of pigment pattern at the periphery of the lesion or a gradual, indistinct cutoff. For the purpose of analysis, the lesions are divided into eight. The maximum border score of eight is given when the entire border (i.e., all eight segments) of the lesion reveal a sharp cutoff. If the border of the lesion in all eight quadrants has an indistinct cutoff then the minimum score of 0 is given. Minimum is 0 maximum is 8 and the factor 0,1 in order to obtain the total ABCD score.
In determining the color score, the following six colors are considered important. The presence of each color counts a 1 point and the factor is 0.5
- light brown
- dark brown
Four colors correspond with melanin distribution (i.e., light and dark brown reflect melanin localized mainly in the epidermis and/or superficial dermis, black represents melanin in the upper granular layer or stratum corneum or all layers of the epidermis, and blue-gray corresponds with melanin in the papillary dermis). White color corresponds with areas of regression and red color reflects the degree of inflammation or neovascularization. The color white is considered to be present only if the area in question is lighter (whiter) in color than the adjacent color of normal skin. The values for the color score range from 1 to 6. As was the case with asymmetry, the dermoscope reveals a wider range of colors than can be seen with the naked eye.
For evaluation of dermoscopic (also known as differential) structures, five main features are considered. For the presence of each structure a value of 1 is attributed. The multiplication factor is 0,5
- Structureless areas
- Pigment network
- Branched streaks (atypical network)
In general, melanomas display more dermoscopic structures as compared with nevi. Ninety percent of melanocytic nevi reveal three or fewer structural components. In contrast, 73% of melanomas reveal four or more structural components. It is important to be aware of a few caveats regarding the ABCD rule. First, dots and globules of any color (including red) are considered significant. Second, structureless areas are considered present if they encompass at least 10% of the lesion’s surface area. These structureless areas can be hypo- or hyperpigmented. Third, dots and branched streaks are counted if there are more than two of each, and globules are counted if there is at least one present within the lesion. combination of criteria: determination of total dermoscopy score
Calculation of the total dermoscopy score (TDS)
An accurate distinction between benign and malignant melanocytic lesions can be made if the individual scores of asymmetry, border, color, and dermoscopic structures (the components of the ABCD rule of dermoscopy) are multiplied by the coefficients 1.3, 0.1, 0.5, and 0.5, respectively, yielding a TDS.
To calculate a formula for a TDS that would reliably discriminate between benign and malignant melanocytic lesions, 157 histologically proven cases were equally and randomly divided between a training and test set. The reliability of the ABCD criteria and coefficients were tested on the test set of cases. At a threshold of 5.45, a diagnostic accuracy of 94% was obtained. The diagnostic accuracy measures the number of correct diagnoses of melanoma compared with all cases of melanoma and all cases falsely diagnosed as melanoma. In the training set, the specificity was 90% and the sensitivity was 100%, which indicates that all melanomas in the training set had a dermoscopy score greater than 5.45, and only approximately 10% of melanocytic nevi were falsely classified as malignant. In the test set, the diagnostic accuracy was 92%, sensitivity was 98%, and specificity was 90%, confirming the high diagnostic value of the ABCD rule of dermoscopy.
In an effort to maintain the highest sensitivity for diagnosing melanoma, the ABCD rule was further refined by classifying as suspicious for malignancy all lesions whose dermoscopy score falls between 4.75 and 5.45. These suspicious lesions should be excised or monitored very closely with a low threshold for excision should any changes develop.
As with any rule, there can be exceptions to the ABCD rule of dermoscopy. Melanocytic nevi with a lentiginous component may not be reliably classified via clinical or dermoscopic examination and these lesions often look suspicious under dermoscopy. False-positive TDS are also common when evaluating melanocytic nevi with globules or nevi with a papillomatous surface. In addition, Spitz/Reed nevus, congenital melanocytic nevus, nevus spilus, agminated nevus, recurrent melanocytic nevus, and ink spot lentigo can have a high TDS score. In these cases the typical “Gestalt” view and diagnosis is superior to the diagnosis based on the calculated TDS. Lesions located on the face, palms, soles, and mucosa, cannot reliably be analyzed using the ABCD rule. Lesions on these locations have their own unique set of diagnostic criteria.
Unfortunately, using the ABCD rule of dermoscopy does not guarantee that all melanomas will be correctly identified. In particular, it is not uncommon for amelanotic melanomas or deep nodular melanomas to have dermoscopy scores lower than 5.45. In these cases the presence of additional dermoscopic clues, such as regression structures, milky-red areas, and atypical blood vessels can assist in their diagnosis. These criteria are also helpful in the diagnosis of early, small, or so-called featureless melanomas.
In conclusion, one fact remains clear, namely, that three features are extremely important in differentiating melanoma from benign nevi. These three features are degree of symmetry/asymmetry, number of colors and number of dermoscopic structures present in the lesion. Benign nevi tend to be symmetric, having few colors and structures. In contrast, melanomas tend to be asymmetric, having many colors and structures.