Ultrasonography for classifying lymphatic sclerosis types and deciding optimal sites for lymphatic-venous anastomosis in patients with lymphoedema,

We have previously categorised of degeneration of the collecting lymphatic vessels into four types: normal, ectasis, contraction and sclerosis type (NECST classification). Herein, we evaluated the collecting lymphatic vessels in lymphoedema-affected limbs using ultrasonography. In step 1, we investigated 110 lymphatic vessels from 25 patients with lymphoedema, who underwent lymphatic-venous anastomosis (LVA) following preoperative ultrasonography. We classified the lymphatic vessels using the NECST classification during intraoperative microscopic observation. Post-operatively, we evaluated the preoperative ultrasonographic images and identified the lymphatic vessels. In step 2, we investigated 79 lymphatic vessels from 17 patients. We performed ultrasonography and detected the lymphatic vessels preoperatively and compared the results with the intraoperative findings. This study is not blinded. In step 1, normal-type lymphatic vessels were observed as spicular and flat hypo-echoic lesions on ultrasonography. Ectasis-type lymphatic vessels appeared as a rounded hypo-echoic region and coloured on Doppler imaging once in 20-30 s. Contraction-type lymphatic vessels appeared as a small hypo-echoic region in the centre of the hyper-echoic ellipse. Sclerosis-type lymphatic vessels appeared as a hyper-echoic ellipse without lumen, similar to fibrotic tissues. In step 2, of 79 lymphatic vessels found intraoperatively, 65 (82.3%) were detected on ultrasonography and 37 (46.8%) were accurately diagnosed according to the NECST classification criteria preoperatively. All lymphatic vessels detected on ultrasonography were found intraoperatively. Collecting lymphatic vessels could be observed by ultrasonography in lymphoedema-affected limbs. Depending on the degree of collecting lymphatic vessel sclerosis-corresponding to the NECST classification-various findings such as spicular, rounded, hyper-echoic and similar to these were presented. Moreover, we can decide optimal sites for LVA preoperatively.