Abstract: PURPOSE: To study the diagnostic significance of color Doppler ultrasound in deep facial and cervical lymphatic malformation and venous malformation by analyzing their color Doppler ultrasonographic characteristics. METHODS: The color Doppler ultrasonographic data of 29 cases of deep-seated facial and cervical lymphatic malformation and 32 cases of venous malformation which had been confirmed by surgical pathology or percutaneous puncture biopsy guided by DSA in Linyi Tumour Hospital from December 2012 to December 2013 were retrospectively analyzed. With final pathology or DSA as gold standard, specificity and sensitivity were calculated and compared with SPSS 18.0 software package. RESULTS: Of the 32 cases of venous malformation, there was no echo and interval by 2-dimensional ultrasound, hypoechoic thrombosis was found in 14 cases and high echo vein stone was found in 11 cases. There were artery flow around and PSV result was (13.98±7.93) cm/s, RI was (0.71±0.83) cm/s. In 30 cases vein blood flow signals were seen under probe pressure. Of the 29 cases, eight were single cystic lymphatic malformation, ten multiple and 11 mesh grid. There was no echo by 2-dimensional ultrasound but multiple hyperechoic interval was found in 21 cases. In 2 cases flow signals were seen under probe pressure which appeared as vein in PI. There were arterial blood flow signals in these intervals, PSV was 14.32±6.57 cm/s,RI was 0.68±0.79 cm/s. There was no significant difference between PSV results and RI results by Student′s t test （t=1.346, 1.474, P>0.05）, while there was significant difference (P<0.05) between the 2 groups in hypoechoic thrombosis (χ²=13.352), high echo vein stone (χ²=12.162), flow signals under probe pressure (χ²=46.015), occurrence of multiple hyperechoic interval (χ²=35.338) by Chi-square test (P<0.05). The sensitivity was 93.75% with probe pressure of blood flow signal to diagnose venous malformation, and the sensitivities of hypoechoic thrombosis (96.55%), hyperechoic vein stone (93.33%) and probe pressure (100%) were all high in diagnosis of venous malformation. The sensitivity and specificity of multiple hyperechoic interval to diagnose venous malformation was 72.41% and 100%, respectively. CONCLUSIONS: Color Doppler ultrasound is noninvasive, inexpensive and accurate in diagnosis of deep-seated facial and cervical lymphatic malformation and venous malformation. Multiple hyperechoic interval with arterial blood flow signals could support the diagnosis of lymphatic malformation, whereas hypoechoic thrombosis or high echo vein stone in the lesion and flow signals under probe pressure would help to confirm venous malformation.

Key words: Lymphatic malformation, Venous malformation, Color Doppler ultrasound, Differential diagnosis