医学
吻合
淋巴系统
放射科
超声波
多普勒超声
核医学
多普勒效应
外科
病理
物理
天文
作者
Hyung Bae Kim,Sung Soo Jung,Min‐Jeong Cho,Nicolas Peirera,Changsik John Pak,Peter Hyun Suk Suh,Sang Hoon Lee,Joon Pio Hong
出处
期刊:Journal of Reconstructive Microsurgery
[Georg Thieme Verlag KG]
日期:2022-04-14
卷期号:39 (02): 092-101
被引量:11
标识
DOI:10.1055/s-0042-1745745
摘要
Despite the extensive use of various imaging modalities, there is limited literature on comparing the reliability between indocyanine green (ICG) lymphography, MR Lymphangiogram (MRL), and high frequency color Doppler ultrasound (HFCDU) to identify lymphatic vessels. In this study of 124 patients, the correlation between preoperative image findings to the actual lymphatic vessel leading to lymphovenous anastomosis (LVA) was evaluated. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and simple detection were calculated. Subgroup analysis was also performed according to the severity of lymphedema. Total of 328 LVAs were performed. The HFCDU overall had significantly higher sensitivity for identifying lymphatic vessels (99%) over MRL (83.5%) and ICG lymphography (82.3%)(p < 0.0001). Both ICG lymphography and HFCDU had 100% specificity and PPV. The NPV was 3.6%, 6.5% and 57.1% respectively for MRL, ICG lymphography, and HFCDU. All modalities showed high sensitivity for early stage 2 lymphedema while HFCDU showed a significantly higher sensitivity for late stage 2 (MRL:79.7%, ICG:83.1%, HFCDU:97%) and stage 3 (MRL:79.7%, ICG:79.7%, HFCDU:100%) over the other two modalities (p < 0.0001). This study demonstrated while all three modalities are able to provide good information, the sensitivity may alter as the severity of lymphedema progresses. The HFCDU will provide the best detection for lymphatic vessels throughout all stages of lymphedema. However, as each modality provides different and unique information, combining and evaluating the data according to the stage of lymphedema will be able to maximize the chance for a successful surgical outcome.
科研通智能强力驱动
Strongly Powered by AbleSci AI