Value of pseudopod sign on high‐frequency ultrasound in predicting the pathological invasion of extramammary Paget's disease lesions

Abstract

Background Vertical invasion of extramammary Paget's disease (EMPD) is associated with poor prognosis. The usual vertical invasion route is directly downward or along the skin appendages. High-frequency ultrasound (HFUS) can be used to measure the EMPD lesion thickness, and visualize the pseudopod extensions due to skin appendage involvement. It is a non-invasive method for evaluating the extent of vertical invasion in EMPD. Objective To investigate the value of HFUS in predicting the extent of vertical invasion in EMPD.Methods In this retrospective study, 85 patients with EMPD were divided into two groups based on the pathology: invasive EMPD (iEMPD) group (n = 13) and in situ EMPD group (n = 72). The clinical and HFUS features of both the groups were analysed. The different types of pseudopodia morphology on HFUS were as follows: no pseudopodia, irregular bottom, small sphere, short strip, long strip, vase shape and nodular convex. These were further stratified into lowrisk and high-risk levels. ResultsThe clinical features were comparable between the two groups (P > 0.05). There were significant differences between the two groups in the HFUS features (lesion thickness, lesion shape, bottom shape, layer involvement, pseudopodia morphology and colour Doppler blood flow signal; all P < 0.05). The distribution of the pseudopodia morphology types in the in situ EMPD and iEMPD groups was as follows: no pseudopodia, 30/72 and 0/13; irregular bottom, 5/72 and 0/13; small sphere, 5/72 and 0/13; short strip, 21/72 and 0/13; long strip, 8/72 and 3/13; vase shape, 3/72 and 3/13; and nodular convex, 0/72 and 7/13 (P < 0.05 for all). The sensitivity and specificity of high-risk pseudopodia in identifying iEMPD were 100% and 84.7%, respectively.Conclusions High-frequency ultrasound provides morphological information regarding EMPD lesions. Risk stratification for pseudopodia can help to distinguish between iEMPD and in situ EMPD lesions.

Ultrasonography for the Diagnosis and Follow‐Up of Extramammary Paget Disease (EMPD)

“…The sensitivity, specificity, positive predictive value, and negative predictive value of high-risk pseudopodia in predicting invEMPD were 100, 84.7, 54.2, and 100%, respectively. 9 When skin biopsies of apocrine-rich areas from healthy volunteers were analyzed, it was seen that 19.4% of them presented skin appendages, which correlated sonographically with the presence of short strips in HFUS.…”

“…Inflammation may cause slight overestimation to the EMPD thickness, as it is seen hypoechoic as well. 9 Affected Layer (Epidermis, Dermis, or Subcutaneous) Ma et al 8 and Chen et al 6 (n = 17) found that layer involvement using US and the pathological results were consistent in 85.0 and 88.2% of the lesions and inconsistent in 15.0 and 11.8%, leading to κ coefficient of 0.625 and 0.746, respectively. They both used UBM.…”

“…A hyperechoic layer on the surface is seen as abnormal keratinization with various degrees of posterior acoustic shadowing. [6][7][8][9] These features are unspecific for EMPD diagnosis, but, compared with the surrounding healthy skin, it can guide on the extent of the lesion, as well as supply information on the invasion of the lesion (Figure 1).…”

“…It consists on the measurement from the surface to the maximum point of the hypoechoic band invasion, which includes epidermis and dermis in case it is involved too. Three assays have proved a good histological correlation with thickness measured with HFUS, [6][7][8][9] even though UBM seems more accurate. 6 Ma et al 8 (n = 60) found the optimal cutoff infiltration depth for predicting an invasive form was 1.55 mm.…”

“…A hyperechoic epidermal surface may be associated, but its presence has not been related to lesion invasion. 6,8,9 In fact, it is believed that it may be due to external factors such as scratching. 8…”