TY - JOUR
T1 - To excise or not
T2 - Impact of MelaFind on German dermatologists' decisions to biopsy atypical lesions
AU - Hauschild, Axel
AU - Chen, Suephy C.
AU - Weichenthal, Michael
AU - Blum, Andreas
AU - King, Hadley C.
AU - Goldsmith, Jeff
AU - Scharfstein, Daniel
AU - Gutkowicz-Krusin, Dina
PY - 2014/7
Y1 - 2014/7
N2 - Objectives: To determine the biopsy sensitivity to melanoma of dermatologists in Germany and the impact of MelaFind® on their decisions to biopsy melanomas. Design Randomized two-armed online reader study presenting case information, clinical/dermatoscopic images of pigmented skin lesions and MelaFind results (Arm 2). Methods: Each participant was asked to review 130 pigmented skin lesions. Biopsy decisions of dermatologists without MelaFind versus MelaFind and dermatologists without MelaFind versus dermatologists with MelaFind were compared. Results: Dermatologists without MelaFind had average sensitivity to melanoma of 69.5 % and average specificity of 55.9 %. MelaFind had greater sensitivity than dermatologists alone (96.9 % vs. 69.5 %, one-sided p < 0.00001) and lower specificity (9.2 % vs. 55.9 %, one-sided p < 0.00001). Dermatologists with MelaFind had higher sensitivity than those without MelaFind (78 % vs. 69.5 %, one-sided p < 0.00001) and a lower specificity (45.8 % vs. 55.9 %, one-sided p < 0.00001). The number of dermatologists detecting over 90 % of melanomas increased from 3 of 101 without MelaFind to 22 of 101 with MelaFind (p = 0.00006) while specificity remained relatively equivalent (23 % vs. 21 %, p = 0.99). Conclusions: The MelaFind information, when incorporated into the final biopsy decision, can improve biopsy sensitivity with modest effect on biopsy specificity.
AB - Objectives: To determine the biopsy sensitivity to melanoma of dermatologists in Germany and the impact of MelaFind® on their decisions to biopsy melanomas. Design Randomized two-armed online reader study presenting case information, clinical/dermatoscopic images of pigmented skin lesions and MelaFind results (Arm 2). Methods: Each participant was asked to review 130 pigmented skin lesions. Biopsy decisions of dermatologists without MelaFind versus MelaFind and dermatologists without MelaFind versus dermatologists with MelaFind were compared. Results: Dermatologists without MelaFind had average sensitivity to melanoma of 69.5 % and average specificity of 55.9 %. MelaFind had greater sensitivity than dermatologists alone (96.9 % vs. 69.5 %, one-sided p < 0.00001) and lower specificity (9.2 % vs. 55.9 %, one-sided p < 0.00001). Dermatologists with MelaFind had higher sensitivity than those without MelaFind (78 % vs. 69.5 %, one-sided p < 0.00001) and a lower specificity (45.8 % vs. 55.9 %, one-sided p < 0.00001). The number of dermatologists detecting over 90 % of melanomas increased from 3 of 101 without MelaFind to 22 of 101 with MelaFind (p = 0.00006) while specificity remained relatively equivalent (23 % vs. 21 %, p = 0.99). Conclusions: The MelaFind information, when incorporated into the final biopsy decision, can improve biopsy sensitivity with modest effect on biopsy specificity.
UR - http://www.scopus.com/inward/record.url?scp=84903979245&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84903979245&partnerID=8YFLogxK
U2 - 10.1111/ddg.12362
DO - 10.1111/ddg.12362
M3 - Article
C2 - 24944011
AN - SCOPUS:84903979245
SN - 1610-0379
VL - 12
SP - 606
EP - 614
JO - JDDG - Journal of the German Society of Dermatology
JF - JDDG - Journal of the German Society of Dermatology
IS - 7
ER -