The assessment of ovarian tumor angiogenesis: what does three-dimensional power Doppler add? A. Kurjak, S. Kupesic, B. Breyer, V. Sparac and S. Jukic* Department of Obstetrics and Gynecology and *Department of Pathology, Medical School University of Zagreb, ‘Sveti Duh’ Hospital, Zagreb, Croatia Key words: TUMOR ANGIOGENESIS, NEOVASCULARIZATION, THREE-DIMENSIONAL ULTRASOUND, POWER DOPPLER, VASCULAR GEOMETRY INTRODUCTION It has been known for over 25 years that the development of new blood vessels is necessary to sustain the growth, invasion and metastasis of tumors 1–3 . Angiogenesis is crucial for sustaining tumor growth, as it allows oxygena- tion and nutrient perfusion of the tumor as well as removal of waste products. Moreover, increased angiogenesis coin- cides with increased tumor cell entry into the circulation, and thus facilitates metastasis 4,5 . Cancer cells activate the quiescent vasculature to produce new blood vessels via an ‘angiogenetic switch’, often during the premalignant stages of tumor development. Data suggesting that control of angiogenesis is separate from control of cancer cell prolif- eration raise the possibility that drugs inhibiting angio- genesis could offer a treatment that is complementary to traditional chemotherapy, which directly targets tumor cells 1–3,6 . This exciting possibility has stimulated current research on tumor angiogenesis 7–10 . CONTRIBUTION OF TRANSVAGINAL COLOR DOPPLER Angiogenesis is a common phenomenon in malignant ovarian neoplasms, but the intensity of neovascularization may depend on the characteristics of the individual tumor 11 . Therefore, an incremental decrease of the imped- ance indices in the vessels of adnexal tumors may reflect the increase in angiogenesis and be an indication of the tumor’s malignant potential 12 . Animal models have shown that angiogenesis can be detected by Doppler ultrasound even in a small volume of malignant tumor (25 mg) 13 . These find- ings suggested that angiogenesis might be detected with current color Doppler ultrasound equipment even when the carcinoma is confined within the ovarian capsule or when it exhibits low malignant potential. Indeed, the study of several series has shown that color and pulsed Doppler sonography can depict ovarian carcinoma at stage I 14–16 . One study detected two of 18 stage-I ovarian carcinomas solely by the presence of an abnormal blood flow pattern in normal-sized ovaries 15 , whereas another study found three of 17 stage I cancers on the basis of abnormal blood flow 16 . However, in the latter study, two stage I tumors did not demonstrate flow and both were > 15 mm in size. It could be argued that these undiscovered malignant tumors had a low potential to induce an angiogenic response, or that the blood vessels were so small that they were impossible to depict with current color and pulsed Doppler equipment. The newly developed power or energy modes of color Doppler imaging permits the depiction of even smaller vessels, but, paradoxically, small intraparenchymal arteri- oles in benign and normal tissues may show a low imped- ance and a low-velocity blood flow pattern, giving rise to false-positive results. Nevertheless, the tendency towards progressive decrease in the vascular impedance from benign lesions to borderline, early and advanced malignan- cies has been reported 12 . This observation was supported circumstantially by an in vivo study which showed that there was a rise in vascularity with tumor progression in the melanocytic system, demonstrated by histopathology 17 . This in vivo information is in agreement with the concept that an increased vascular supply may facilitate tumori- genesis and aggressive biological behavior in a neoplastic system. Since the vascular impedance measured by color Doppler in the parenchymal vessels of an adenxal tumor will not always represent the microangiogenesis itself 18 , it is possible that the signal obtained by color and pulsed Doppler demonstrates the main neovascularized channel representing the summation of downstream resistance of Correspondence: Professor A. Kurjak, Department of Obstetrics and Gynecology, Medical School University of Zagreb, ‘Sveti Duh’ Hospital, Sveti Duh 64, HR-10000 Zagreb, Croatia Ultrasound Obstet Gynecol 1998;12:136–146 REVIEW Received 6–4–98 Accepted 2–6–98 136 98/075 AMA: First Proof