Faculty Bibliography

PURPOSE: To determine the role, accuracy, and selection criteria of ultrasonographic (US) guidance for biopsy for thoracic lesions. MATERIALS AND METHODS: Imaging-guided thoracic biopsies (n = 86) were performed in 84 consecutive patients. US guidance was used for lesions abutting the chest wall; computed tomographic (CT) guidance was used for all masses surrounded by aerated lung. Mass location and size, guidance modality, histologic results, procedure time, and complications were recorded. RESULTS: Thirty-four lesions (19 parenchymal, six pleural, six chest wall, three mediastinal) were amenable to US-guided biopsy. The mean mass diameter was 4.3 cm, the mean number of passes was 3.2, and the mean procedure time was 31.4 minutes. A histologic diagnosis was achieved in 31 (91%) patients, including all with small (< 2-cm) masses (n = 9). There was one case of pneumothorax. CT guidance was used in 52 (60%) of 86 cases. Lesions were parenchymal (n = 41), pleural (n = 1), and mediastinal and hilar (n = 10). The mean diameter was 2.9 cm, the mean number of passes was 2.3, and the mean procedure time was 45.2 minutes. A histologic diagnosis was achieved in 37 (71%) patients, including 18 of 27 with a small mass. Complications included pneumothorax (n = 21) and parenchymal hemorrhage (n = 2). CONCLUSION: US is an effective and safe alternative to CT for guidance at biopsy of masses abutting the chest wall. Real-time US visualization allows accurate needle placement, shorter procedure time, and performance in debilitated and less cooperative patients.

PURPOSE: To evaluate the accuracy of magnetic resonance (MR) and ultrasound (US) criteria for breast implant integrity. MATERIALS AND METHODS: One hundred twenty-two single-lumen silicone breast implants and 22 bilumen implants were evaluated with surface coil MR imaging and US and surgically removed. MR criteria for implant failure were a collapsed implant shell ("linguine sign"), foci of silicone outside the shell ("noose sign"), and extracapsular gel, US criteria were collapsed shell, low-level echoes within the gel, and "snowstorm" echoes of extracapsular silicone. RESULTS: Among single-lumen implants, MR imaging depicted 39 of 40 ruptures, 14 of 28 with minimal leakage; 49 of 54 intact implants were correctly interpreted. US depicted 26 of 40 ruptured implants, four of 28 with minimal leakage, and 30 of 54 intact implants. Among bilumen implants, MR imaging depicted four of five implants with rupture of both lumina and nine of 10 as intact; US depicted one rupture and helped identify two of 10 as intact. Mammography accurately depicted the status of 29 of 30 bilumen implants with MR imaging correlation. CONCLUSION: MR imaging depicts implant integrity more accurately than US; neither method reliably depicts minimal leakage with shell collapse. Mammography is useful in screening bilumen implant integrity.

PURPOSE: To determine if prominent vascularity and low pulsatility index (PI) and resistive index (RI) in endometrial arteries help differentiate carcinoma from benign lesions. MATERIALS AND METHODS: Forty-five postmenopausal women with at least 8-mm-thick endometrium at endovaginal ultrasound (US) underwent color duplex Doppler endovaginal US of the endometrium. PI and RI values were recorded from arterial waveforms generated in areas of increased vascularity. RESULTS: At color duplex Doppler endovaginal US, endometrial arterial flow was seen in 23 of 36 (64%) proved benign endometrial lesions. The range of PIs for benign lesions was 0.31-1.77 (mean, 0.72 +/- 0.33 [standard deviation]) and for RIs was 0.27-0.84 (mean, 0.48 +/- 0.13). Arterial vessels were seen in five of nine proved endometrial cancers (56%); the range of PIs was 0.42-1.17 (mean, 0.71 +/- 0.32) and for RIs was 0.34-0.69 (mean, 0.48 +/- 0.15). Differences in mean PI and RI for benign and malignant lesions were not significant. CONCLUSION: Low-impedance arterial flow is observed in various diseases of the endometrium, and there is an overlap between the PI and RI of benign and malignant lesions.

PURPOSE: To evaluate applications of three-dimensional (3D) ultrasound (US) in a clinical setting. MATERIALS AND METHODS: Sixty-two patients were examined with experimental 3D US transducers coupled to a commercially available US unit and a computer workstation. Images were acquired in either the sagittal or transverse plane and were reconstructed in two orthogonal planes, transverse and sagittal; the C-plane, parallel to the transducer surface; and a volumetric rotational transparency display. RESULTS: A variety of normal and diseased abdominal and superficial organs and selected fetal anatomy, and pathologic conditions were imaged. 3D US allowed display of anatomy and pathologic conditions in planes usually not possible with conventional two-dimensional (2D) US. CONCLUSION: Preliminary data suggest that 3D US may become a valuable clinical tool and adjunct to 2D US. 3D US allows depiction of normal and abnormal structures in previously unattainable planes, thus facilitating diagnosis and increasing operator diagnostic confidence.

PURPOSE: To characterize sonographic criteria for the detection of implant rupture. MATERIALS AND METHODS: One hundred nineteen symptomatic women (221 silicone implants) were evaluated with ultrasound (US) over 11 months to detect implant rupture. Fifty-nine breast prostheses in 31 women were surgically removed. RESULTS: Five sonographic findings associated with rupture were identified: echogenic noise, low-level homogeneous echoes within the implant, echogenic lines coursing within the implant, a contour bulge of the implant, and peri-implant fluid accumulations. The most useful finding in the detection of implant rupture was low-level homogeneous echoes, which were associated with 55% of ruptured implants and 16% of intact implants. This was statistically significant, achieving a 55% sensitivity and 84% specificity for the detection of implant rupture. Nine of 22 ruptured implants (41%) displayed none of the described criteria. CONCLUSION: US may be useful to detect implant rupture; however, a larger prospective study is necessary for adequate evaluation.

OBJECTIVE: The purpose of this study was to assess the blood flow characteristics of adnexal masses before surgical excision and to determine whether color flow Doppler sonography is useful for distinguishing benign from malignant masses. SUBJECTS AND METHODS: Thirty-one adnexal masses were evaluated with color flow Doppler transvaginal sonography. The pulsatility index and resistive index were calculated from the waveforms generated from blood flow within the ovary. Twenty-five lesions were benign and six were malignant on pathologic examination. Benign lesions included six endometriomas, six mesothelial cysts, three serous and one mucinous cystadenoma, three mature cystic teratomas, two hemorrhagic corpus luteum cysts, one cystadenofibroma, one sclerosing stromal cell tumor, one paratubal cyst, and one ovary that had undergone torsion with infarction. The malignant lesions consisted of three papillary serous cystadenocarcinomas, one granulosatheca cell tumor, one immature teratoma, and one metastasis of colon cancer to the ovaries. RESULTS: Benign tumors and cysts had a significantly higher pulsatility index (mean, 1.93 +/- 1.02; range, 0.23-3.99) and resistive index (mean, 0.77 +/- 0.22; range, 0.2-1.0) than did malignant tumors (pulsatility index: mean, 0.77 +/- 0.33; range, 0.31-1.09; resistive index: mean, 0.5 +/- 0.17; range, 0.27-0.67). However, some overlap in individual values for benign and malignant lesions was found. CONCLUSION: Our preliminary data suggest that high pulsatility and resistive indexes indicate benign adnexal processes; however, considerable overlap in pulsatility and resistive indexes between benign and malignant lesions was noted, and further work is needed before the validity of these factors is proved.

A correlative sonographic and histopathologic analysis was performed in 35 postmenopausal women with greater than 5-mm thickening of the endometrium at pelvic sonography. Women undergoing estrogen replacement were excluded from study. Four distinct sonographic patterns were encountered. Pattern 1 consisted of echogenic endometrium with small cysts (endometrial polyp with cystic hyperplasia [n = 9], atrophic endometrium with cystically dilated glands [n = 5], and atrophic endometrium [n = 3] at microscopic examination). Pattern 2 was homogeneous echogenic endometrium (proliferative endometrium [n = 3] and adenomyomatous polyp [n = 1]). Pattern 3 was irregular, inhomogeneous endometrium with ill-defined hypoechoic areas (endometrial carcinoma [n = 5], complex hyperplasia with atypia [n = 1], blood clots [n = 1], and atrophic endometrium with eosinophilic metaplasia [n = 1]). Pattern 4 was thin endometrium with fluid in the endometrial cavity (scant atrophic endometrium [n = 6]). Thus, an endometrial thickness of greater than 5 mm in postmenopausal women is associated with a variety of pathologic conditions. Subclassification of sonographic patterns may be helpful in differentiating benign cystic atrophy or cystic endometrial hyperplasia from malignant endometrial lesions.