Sequential image analysis and measurement of pigmented lesions: false variations due to patient positioning

Clinical researchers are evaluating the utility of obtaining sequential images of pigmented lesions taken over time for purposes of comparison with the aim of detecting subtle changes suggestive of melanoma. Therefore, the image acquisition process is critical and will need to be strictly standardized before any firm conclusions can be drawn from analysis of sequential images. The influence of patient positioning on the accuracy of sequential image analysis has not been considered in most studies evaluating sequential images. In this experimental study, the influence of patient positioning on the size and shape of an inked circle placed on the skin was determined and measured. Inked circular marks (15 mm in diameter, area 176.71 mm2) were placed on the skin of the lumbar and suprascapular areas of 60 consecutive patients. The diameter and area of the 'circle' was measured with the patient in the prone position with head centred, prone position with head turned to the right, prone position with head turned left, and in the seated position. Statistical analysis was performed with Student's t-tests (paired data). We observed statistically significant differences in the shape, mean maximal diameter and area of the inked circular marks on both the suprascapular area and in the lumbar area after changes of patient positioning (P<0.001). To conclude, the position of the patient must be fixed and standardized during acquisition of sequential images, at least for lesions 15 mm in diameter or larger. Furthermore, it is our opinion that the methods used to control for patient positioning should be reported in the methodology section of studies that report on comparison of sequential images. Only then can we accurately compare sequential images and avoid 'false positive lesion enlargement' being categorized as a true change.