| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Genitourinary Imaging |
1 From the Department of Radiology, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115 (C.M.C.T., S.G.S., D.L.B., B.J.M.); the Department of Health Care Policy, Harvard Medical School, Boston, Mass (K.H.Z., B.J.M.); and the Department of Radiology, Jefferson Medical College and Thomas Jefferson University Hospital, Philadelphia, Pa (A.B.K.). From the 1998 RSNA scientific assembly. Received February 16, 1999; revision requested April 5; revision received September 14; accepted September 24. C.M.C.T. was a GE-AUR Fellow. Supported in part by Public Health Service grant NIH-U01 CA9398-03, awarded by the National Cancer Institute, U.S. Department of Health and Human Services. Address correspondence to C.M.C.T. (e-mail: ctempany@bwh.harvard.edu).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: US, CT, and MR imaging were performed in 280 patients. Images were read by three radiologists from each of the five hospitals. Image analysis included determination of malignancy within the peritoneum (11 sites), lymph nodes (10 sites), and hepatic parenchyma. The standard of reference was based on surgical and histopathologic findings. Statistical methods used were receiver operating characteristic (ROC) curve analysis, pairwise comparison of areas under the ROC curves (Az), analysis of sensitivity and specificity pairs, and assessment of agreement between the degree of suspicion and standard of reference.
RESULTS: There were 118 patients with malignant tumors; 73 (62%) had stage III or IV disease. Metastases were found in the peritoneum in 70 (59%), nodes in 20 (17%), and liver in seven (6%) cases. In the peritoneum, MR imaging and CT (Az = 0.96 for both) were more accurate than US (Az = 0.86), especially in the subdiaphragmatic spaces and hepatic surfaces. MR imaging and CT were more sensitive than US (95%, 92%, and 69%, respectively) for peritoneal metastases. MR imaging was more accurate than CT for detection of lymph node metastases (Az = 0.76 vs 0.57, P = .04). In the liver, the Az values for the three modalities were 0.77–0.94.
CONCLUSION: CT and MR imaging are equally accurate, and either modality can be used to stage advanced ovarian cancer.
Index terms: Ovary, CT, 852.12113, 852.12115 • Ovary, MR, 852.121411, 852.121415, 852.12143 • Ovary, neoplasms, 852.32, 852.339 • Ovary, US, 852.12983, 852.12984 • Receiver operating characteristic (ROC) curve
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Ovarian cancer is staged surgically at laparotomy in accordance with the recommendations of the International Federation of Gynecology and Obstetrics (2,3). The survival rate is 93% for localized disease, in contrast with 55% and 25% for regional and advanced disease, respectively (1). It is unfortunate that only 24% of all cases are detected at the local stage and that up to 40% of cases are thought to be of a lower stage than they actually are at initial laparotomy (1,4–6). The detection of advanced disease prospectively needs to be improved for treatment planning—in particular, for the possible need to refer the patient to a cancer treatment center. This is of particular importance, since the availability of appropriate technical expertise for staging and debulking may require additional preparation or referral (5). Because all patients with advanced disease will require chemotherapy, they should be treated in a cancer center.
An accurate depiction of the sites of abnormal disease is important because it will help determine the sites at which biopsy will be performed at surgery. Cytoreduction may be attempted either with surgery or with chemotherapy and depends on the bulk and amount of metastatic disease present. It has been shown that there is improved survival with optimal cytoreduction or with removal of tumors larger than 2 cm (7). Thus, it is imperative to improve the preoperative image-based staging and, ultimately, the treatment of patients with advanced ovarian cancer.
This study was an analysis of data from the prospective, multiinstitutional Radiological Diagnostic Oncology Group (RDOG) study sponsored by the National Cancer Institute of the National Institutes of Health. The main goal of the RDOG study was to assess the accuracies of magnetic resonance (MR) imaging, computed tomography (CT), and gray-scale Doppler ultrasonography (US) in the diagnosis and staging of ovarian cancer. The primary analyses showed no significant differences among the three modalities for the overall diagnostic accuracy in any region of the pelvis or abdomen by using areas under receiver operating characteristic (ROC) curves (Az = 0.91 for all three modalities) (8). Because of the large amount of material analyzed and presented in the first article (8), the analyses did not cover in detail the important aspects of advanced disease. Thus, the purpose of this study was to compare and evaluate the relative roles of the three modalities in the diagnosis and staging of advanced disease (stages III and IV). In this study, we examined patients with advanced disease—specifically, in the peritoneum (11 sites), lymph nodes (10 sites), and liver. To our knowledge, prospective studies have not been performed previously to address this question or to compare all three imaging modalities.
The accuracies of CT and MR imaging for staging have been reported to range from 60% to 90% (4,9–15). Investigators in previous studies (16,17) have suggested that MR imaging may have a substantial role in the diagnosis of peritoneal disease. Investigators have also suggested that US is not useful for preoperative staging and recommended it for the initial imaging examination (4,18–21).
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Patient exclusion criteria included inability to give informed medical consent or ineligibility for full pelvic/abdominal surgery or for surgical exploration. Overall contraindications included pregnancy, having undergone prior pelvic/abdominal laparoscopic surgery, or having undergone surgery within 6 months of study entry. The standard contraindications to MR imaging were having a cardiac pacemaker or having intracranial vascular clips.
Imaging Techniques
State-of-the-art techniques were used at all five institutions. Standardized protocols, agreed on by the radiologists at the outset of study, were used in all cases. The technical specifications were as follows: US equipment with color and pulsed Doppler waveform capabilities, helical CT scanners, and 1.5-T MR imaging magnets with a phased-array multicoil and/or the body coil. The following is a brief description of the imaging protocols, with specific details on upper abdominal and pelvic imaging for the analysis of metastatic sites.
US Protocol
Women who had not begun menopause were scheduled, whenever possible, for examination within the first 8 days of the menstrual cycle. Women who had begun menopause were scheduled for examination at any time. In the pelvis, gray-scale US was used to evaluate masses and the ovaries for their sizes, echogenicities, and internal characteristics. Doppler US was not performed. In the pelvis and abdomen, all normal and diseased structures were sought, imaged, and recorded.
State-of-the-art, commercially available US equipment was used. The systems used were the UltraMark 9 HDI and HDI 3000 platforms (ATL-HDLa; Advanced Technologies Laboratories, Bothell, Wash) or the 128XP and 128XP/10 platforms (XP; Acuson, Mountain View, Calif). All machines had transabdominal and endovaginal probes with maximum frequencies of 2–5 MHz and 5–7 MHz, respectively. Most probes were curved or had annular arrays. All machines had color and pulsed Doppler capabilities.
CT Protocol
Somaton Plus or Plus-S (Siemens Medical Systems, Iselin NJ) or 9800 Advantage or HiSpeed Advantage (GE Medical Systems) scanners were used in either a dynamic or helical mode, with 1-second scanning times. The gastrointestinal tract was opacified and the pelvis was examined during peak arterial enhancement after the injection of 150 mL of 60% iodinated intravenous contrast material (ionic or nonionic). The abdomen was imaged by obtaining dynamic, nonhelical scans, with 5-mm collimation and with 5-mm-thick sections at 8–10-mm intervals. Imaging began in the pelvis and then moved up to the abdomen, with the patient in suspended respiration, from the symphysis pubis to the diaphragm.
MR Imaging Protocol
All sites used 1.5-T Signa magnets (GE Medical Systems, Milwaukee, Wis) with 10-mT/m gradient systems. All patients underwent a full examination of both the pelvis and the abdomen; as with CT, the pelvis was imaged first. The pelvis was imaged by using either a multicoil array or a body coil. A pelvic multicoil array was used whenever possible. However, the body coil alone was used when there was either a large mass (>15 cm) or a mass that extended superior to the L4-5 level. The abdomen was imaged in all cases by using the body coil. All patients were requested to fast for at least 3 hours prior to the study. A gadolinium chelate was administered intravenously. All patients received 1 mg of glucagon (Glucagen; Bedford Laboratories, Bedford, Ohio) intramuscularly just prior to imaging. The pelvis was examined with T2-weighted imaging followed by T1-weighted imaging, before and after intravenous administration of gadolinium chelate.
The remainder of the abdomen and pelvis above the mass was examined subsequently in the transverse plane with T2-weighted, fast spin-echo imaging and with transverse, fat-suppressed, T1-weighted, spin-echo imaging. For T1-weighted, spin-echo imaging, a repetition time of 400–600 msec and a minimum echo time were used. The field of view was set to fit the abdomen (usually 32–38 cm), and the matrix was 256 x 128–192, with two signals acquired. The section thickness was 8–10 mm, with a 2-mm intersection gap. T2-weighted, fast spin-echo imaging had the following parameters: 4,000/100–126, with an echo train length of 16. The field of view was set to fit the abdominal size, and the section thickness was 8–10 mm, with a 1.0–2.5-mm intersection gap. The matrix was 256 x 192 or 256, with two to four signals acquired. This sequence was repeated in the coronal or sagittal plane as indicated.
Surgical Protocol
Imaging results were available and were used by the surgeons to plan each patient's procedure. Patients underwent a complete resection of the primary tumor and, when indicated, a debulking procedure. Adequate exploration of all structures was attempted, with resection of known gross tumor and of associated lymph nodes, whenever possible, with en bloc excision. If tissues could not be resected, multiple biopsies were performed. Ascitic fluid was submitted for cytologic examination. The examination findings were recorded as being in one of the following three categories: no tumor or mass present, suspicious for tumor, or tumor present. Missing data were categorized with an explanation of area or site not examined, or organ or item absent. The extent of the tumor was also classified as either small (
2 cm) or large (>2 cm) or as not available or not evaluated.
Histopathologic Protocol
Pathologists from each institution performed their interpretations in a routine manner by using the revised World Health Organization histologic classification for ovarian neoplasm (22,23).
Image Analysis
Three radiologists from each of the five institutions interpreted the images prospectively, one radiologist for each modality. Each reader had access to clinical information and to the entry imaging study or studies. No reader had prior knowledge of the findings of the other protocol imaging studies or of either surgery or histopathologic examination.
Standardized forms were completed in each case. All forms required a detailed analysis of the pelvic mass and of the staging steps through the pelvis and abdomen. Malignancy was analyzed on the basis of a five-point rating scale of the degree of suspicion. A rating of 0 indicated that the findings were normal; 1, probably normal; 2, indeterminate; 3, probably abnormal; or 4, definitely abnormal.
All major anatomic areas within the pelvis and abdomen were evaluated, along with ascites or pleural effusion. The radiologist assigned a final stage for each patient on the basis of the modified International Federation of Gynecology and Obstetrics ovarian staging classification (2,3).
For the purpose of this study, we focused our analysis on the diagnosis of disease in three major areas—the peritoneum, lymph nodes, and liver. In the peritoneum, 11 individual sites were analyzed and then divided into six groups: (a) the anterior part of the abdomen, (b) the paracolic gutters (right and left), (c) the subdiaphragmatic spaces (right and left), (d) the mesentery (of the small bowel, transverse colon, and sigmoid colon), (e) the hepatic surface, and (f) the omenta (gastrocolic and infracolic).
Ten lymph node sites were evaluated and were divided into two groups: abdominal nodes (porta hepatis, splenic hilum, and paraaortic [right and left]) and iliac nodes (common [right and left], external [right and left], and internal [right and left]).
The liver also was evaluated for focal parenchymal lesions with all three modalities. This last area did not include lesions on the hepatic surface, which were included in the peritoneal sites. These lesions were rated by using the same five-point rating scale mentioned earlier on the basis of the degree of suspicion of malignancy.
Surgical and Histopathologic Correlation
Findings of US, CT, and MR imaging were used to plan each patient's surgical procedure. Patients underwent complete resection of the primary tumor if a debulking procedure was indicated. When the preoperative assessment resulted in detection or strong suspicion of abdominal spread, surgery commenced in the abdomen and continued into the pelvis. When only pelvic disease was suspected and only if malignancy was found at frozen section, surgery started in the pelvis and continued into the abdomen.
An attempt was made to obtain adequate information on all pelvic and abdominal structures with the resection of known gross tumor and of associated lymph nodes. The peritoneal cavity was explored carefully for the extent of disease. The hepatic and paraaortic nodes were palpated. When ascites was present, the amount was recorded and the fluid was aspirated, measured, and submitted for cytologic examination. Biopsy was performed (preferably with en bloc excision) on any suspicious areas, and specimens were submitted for histopathologic examination. If tissues could not be resected, multiple biopsies were performed. At the completion of each operation, a surgical data form and a detailed surgical report were dictated.
Pathologists from each institution performed their interpretation in a routine manner after gross description of the ovarian neoplasm and of other submitted tissues; all tissue blocks were analyzed microscopically. A minimum of one block per centimeter from the greatest tissue diameter was examined.
Standard of Reference
The standard of reference was determined by combining each patient's surgical and histopathologic findings. Histologic types were taken from the pathologists' forms for each patient to confirm the presence of cancer.
Data Collection
The standardized forms on which the patient's demographics, medical profiles, imaging findings, and surgical and histopathologic findings were recorded were maintained at the American College of Radiology, Reston, Va. These forms were reviewed for consistency prior to starting the final data analysis by the Statistical Center in the Department of Health Care Policy, Harvard Medical School, Boston, Mass.
Statistical Methods
The performances of US, CT, and MR imaging in the staging of advanced ovarian cancer in the three major anatomic areas were evaluated by using the maximum rating and the standard of reference across all sites in the respective areas. First, ROC curve analysis by using all possible clinical diagnostic criteria was conducted. The area under the curve (Az), determined by using a probit link, was a summary of the overall diagnostic accuracy (ROCFIT; Metz CE, Shen JH, Wang PL, et al, Department of Radiology, University of Chicago, Chicago, Ill, 1994). In addition, pairwise comparisons of Az according to modality for any two anatomic areas were made and were restricted to only common cases (CORROC2; Metz CE, Kronman HB, Department of Radiology, University of Chicago, Ill, 1994) (24). Similar analyses were repeated for the six subgroups with disease in the peritoneum and for the two subgroups with disease in the lymph nodes.
Next, based on the five-point rating scale, the most clinically meaningful cutoff point was defined as a threshold between ratings of 1 and 2. Thus, all findings rated as 0 or 1 were grouped and all findings rated as 2, 3, or 4 also were grouped to create a binary split. The resultant pair (sensitivity and specificity) was computed according to modality for each of the three anatomic areas. The agreement between the imaging results and the standard of reference then was evaluated by using a 
statistic.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
The results of the analysis of sensitivity and specificity pairs in Table 4 suggest that MR imaging and CT were more sensitive than US (95%, 92%, and 69%, respectively [MR vs US, P = .03; CT vs US, P = .05]) for the detection of peritoneal metastases. However, US was more specific than MR imaging and CT (93%, 80%, and 82%, respectively; P = .05 for both comparisons). The values in Table 5 showed moderate to fair agreement for all three modalities ( = 0.60–0.65).
|
|
2-cm) implants and 57 patients to have large (>2-cm) implants. In five patients, size data were not available. All three modalities had high sensitivities for the detection of the large implants—100% for MR imaging (43 of 43) or CT (34 of 34) and 80% (44 of 45) for US. Specificity data for small implants were limited because of small numbers but were as follows: US, 63% (five of eight); CT, 0% (zero of three); and MR imaging, 33% (two of six).
Lymph Nodes
Tables 1 and 2 showed that MR imaging (Az = 0.76) was superior to CT (Az = 0.57; P = .04) but that US (Az = 0.68) was not significantly different from the other two modalities for the accuracy of diagnosis of lymph node metastases. Furthermore, in the iliac nodes, MR imaging was superior to CT (Az = 0.93 and 0.46, respectively; P = .03) (Table 6). US was found to be more specific than MR imaging (93% and 84%, respectively; P = .04) (Table 4). In Table 5, poor agreement between modalities was observed ( = 0.16–0.22).
|
= 0.45, 0.26, and 0.24, respectively; P = .03 [US vs CT] and P = .02 [US vs MR imaging]) (Table 5). |
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Attendees of the National Institutes of Health consensus conference in 1995 recognized the need for innovative methods to treat patients with advanced ovarian cancer (26). Noninvasive methods such as CA-125 testing and imaging will play a major role in monitoring these new therapies for advanced ovarian cancer. Per the results of our study, examination with either MR imaging or CT can be recommended for staging in patients with advanced disease, especially for the detection of disease in the peritoneum. Both MR imaging and CT performed very well in detecting disease in the peritoneum and thus can be expected to perform well in monitoring treatment in these areas.
When we examine the results for disease in the peritoneum according to location, we note that all patients had disease in the omentum (70 of 70). The omentum can be evaluated well with either contrast material–enhanced CT or contrast-enhanced MR imaging. The next most common sites of disease were the subphrenic spaces, the mesentery (large and small bowel), the anterior part of the abdomen, and the paracolic gutters (45, 38, 37, and 35 patients, respectively). We did not differentiate the number of peritoneal cancers according to site. The results were computed on the basis of whether there was an implant detected in the peritoneal site of interest and on the basis of the ratings given by the readers by using any of the three modalities. We did not count the lesions individually.
In almost all of these locations, CT depicted slightly more disease than MR imaging; however, the differences were not statistically significant. Contrary to predictions, MR imaging performed very well in the abdomen despite the lack of an orally administered contrast agent for the bowel. The results for MR imaging may well be even higher when a good, reliable, orally administered contrast agent becomes widely available and is used. The Az values for the large- and small-bowel mesentery were 0.95 with CT and 0.93 with MR imaging; CT performed very well given that image acquisition in the upper abdomen was relatively delayed after the injection of intravenous contrast material. Most ovarian metastases to the peritoneum are of low attenuation or are cystic in appearance on CT scans (27). As an alternative, in serous carcinoma, the implants may be calcified (28). Thus, the timing of the intravenous contrast material bolus is not critical to their detection. As indicated by our results, delayed CT scanning is an excellent method for the detection of implants more than 2 cm in size.
When the results were evaluated on the basis of the size of the peritoneal implants, there were significantly more large implants. The critical size determination for treatment planning or for cytoreduction is 2 cm (7). Surgery can be performed in an attempt to debulk or to remove implants of 2 cm or greater. It has been shown that the 4-year survival rate is approximately 30% in patients with tumors debulked to less than 1–2 cm, compared with a long-term survival rate of less than 10% in patients with tumors that are debulked suboptimally (>2 cm) (25). The smaller tumors usually are best treated with chemotherapy. Thus, we thought it was important to determine first the size of the implants in our study and second, how well the imaging modalities performed on the basis of size.
Our findings indicate that both CT and MR imaging are excellent modalities for the detection of all large implants. Although US was not as accurate, it was able to depict about 44 (80%) of 55 of them. There were few patients with small (2-cm) implants in our study (n = 5); thus, it is difficult to make any conclusions as to the modality of choice.
It is, however, more important to prospectively identify the larger implants that can potentially be debulked or removed at surgery. As noted in the US methods, we used 2–5-MHz transabdominal transducers. We did not use high-frequency linear arrays, which might have been helpful to identify more superficial lesions in the liver, but there were not very many hepatic metastases in our study. Thus, it is unlikely that this difference would have significantly changed the overall accuracy of the study findings.
Twenty patients in this study were found to have metastatic lymph nodes at histopathologic sectioning. This was less than was expected—only 20 (27%) of 73 with malignancies of stage III or IV. The more typical rate is 41%–44% for stage III and 64%–87% for stage IV disease (29,30). The three modalities had relatively high specificities for lymph node metastases, which ranged from 93% for US to 84% for MR imaging, although all three modalities seemed to have only mediocre sensitivities (CT, 41%; MR imaging, 39%; and US, 32%). The number of metastatic nodes was relatively small, so it is difficult to draw conclusions. The higher number of positive nodes in the abdomen compared with the pelvis in our population was interesting. This can be explained in part by sampling, because more nodes were sampled in the abdomen than in the pelvis.
ROC analysis showed the overall performance of MR imaging to be the best, followed by US and then CT. The fact that MR imaging and US performed marginally better than CT may be explained by our technique. As ovarian cancer has a low predilection for spreading to the lymph nodes and to the liver, we thought we could modify the CT technique in favor of the pelvis over the abdomen. The contrast material bolus was followed from the pelvis up; thus, when the liver and upper abdomen were scanned, the bolus essentially was delayed. Thus, it is not surprising that CT was not as accurate as MR imaging or US.
All three modalities had similar accuracies for the diagnosis of hepatic metastases. The data were limited in this group, as there were very few patients with hepatic metastases. The similar accuracies were not unexpected, but it is useful to know that any of the three modalities can be used. Investigators in other studies (31,32) have evaluated the relative roles of CT and MR imaging in the detection of hepatic metastases. We did not address these issues in our study, as we had so few hepatic metastases and did not have a full histopathologic evaluation of the liver in every patient.
By using current techniques, such as those used in this study, either MR imaging or CT now can be used to evaluate the abdomen and pelvis in one examination. Both appear to be highly accurate for the staging of peritoneal disease. We did not compare the examinations on the basis of financial cost or of imaging time, nor did we evaluate the patient's individual preferences. Both MR imaging and CT have aspects that patients may find difficult, such as, for CT, the relative risk of receiving iodinated contrast material or of drinking orally administered contrast material or, for MR imaging, the feeling of claustrophobia. Some or all of these issues can be important when the optimal imaging modality is being selected for the patient. In many hospitals, the ease of scheduling also will be an important factor.
While CT and MR imaging did perform better in the analysis, the Az value for US also was very high. In our opinion, CT and MR imaging are equally accurate, and either modality can be used to stage advanced ovarian cancer. CT and MR imaging are more accurate than US, particularly in the subdiaphragm and along the hepatic surface. However, the accuracy of US was also high; US can be used to supplement CT or MR imaging, especially in the hepatic substance and in the lymph nodes. In conclusion, CT and MR imaging have similar high accuracies for the staging of ovarian cancer, and either modality can be used.
|
Acknowledgments |
|---|
We would like to acknowledge the research assistants who helped with patient care and data transfer: Bronwyn Head (Brigham and Women's Hospital), Lisa Herman (Brigham and Women's Hospital), Susan Calder (Brigham and Women's Hospital), Jane Z. Dumsha (Thomas Jefferson University Hospital), and Manette London, CCRC (University of Michigan Hospitals).
|
Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, C.M.C.T.; study concepts, C.M.C.T., K.H.Z.; study design, all authors; definition of intellectual content, C.M.C.T., K.H.Z.; literature research, C.M.C.T., K.H.Z.; clinical studies, C.M.C.T., D.L.B., S.G.S., A.B.K.; data acquisition, C.M.C.T., D.L.B., S.G.S., A.B.K.; data analysis, C.M.C.T., K.H.Z.; statistical analysis, K.H.Z.; manuscript preparation and editing, K.H.Z., C.M.C.T.; manuscript review, all authors.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  O. Akin, E. Sala, C. S. Moskowitz, N. Ishill, R. A. Soslow, D. S. Chi, and H. Hricak Perihepatic Metastases from Ovarian Cancer: Sensitivity and Specificity of CT for the Detection of Metastases with and Those without Liver Parenchymal Invasion Radiology, August 1, 2008; 248(2): 511 - 517. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. D. Yan, J. Sim, and D. L. Morris Selection of Patients with Colorectal Peritoneal Carcinomatosis for Cytoreductive Surgery and Perioperative Intraperitoneal Chemotherapy Ann. Surg. Oncol., June 1, 2007; 14(6): 1807 - 1817. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M V Pakkal and M Balogun Imaging of ovarian cancer Imaging, August 1, 2006; 18(1): 20 - 27. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J A Spencer A multidisciplinary approach to ovarian cancer at diagnosis Br. J. Radiol., October 1, 2005; 78(Special_Issue_2): S94 - S102. [Full Text] [PDF]
|
|
|
|
|
|
R. N. Low, B. Duggan, R. M. Barone, F. Saleh, and S. Y. T. Song Treated Ovarian Cancer: MR Imaging, Laparotomy Reassessment, and Serum CA-125 Values Compared with Clinical Outcome at 1 Year Radiology, June 1, 2005; 235(3): 918 - 926. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Goldberg-Zimring, B. Shalmon, K. H. Zou, H. Azhari, D. Nass, and A. Achiron Assessment of Multiple Sclerosis Lesions with Spherical Harmonics: Comparison of MR Imaging and Pathologic Findings Radiology, June 1, 2005; 235(3): 1036 - 1044. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Sironi, C. Messa, G. Mangili, B. Zangheri, G. Aletti, E. Garavaglia, R. Vigano, M. Picchio, G. Taccagni, A. D. Maschio, et al. Integrated FDG PET/CT in Patients with Persistent Ovarian Cancer: Correlation with Histologic Findings Radiology, November 1, 2004; 233(2): 433 - 440. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Gronlund, C. Hogdall, J. Hilden, S. A. Engelholm, E. V.S. Hogdall, and H. H. Hansen Should CA-125 Response Criteria Be Preferred to Response Evaluation Criteria in Solid Tumors (RECIST) for Prognostication During Second-Line Chemotherapy of Ovarian Carcinoma? J. Clin. Oncol., October 15, 2004; 22(20): 4051 - 4058. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Yoshida, T. Kurokawa, K. Kawahara, T. Tsuchida, H. Okazawa, Y. Fujibayashi, Y. Yonekura, and F. Kotsuji Incremental Benefits of FDG Positron Emission Tomography over CT Alone for the Preoperative Staging of Ovarian Cancer Am. J. Roentgenol., January 1, 2004; 182(1): 227 - 233. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. K. Pannu, R. E. Bristow, C. Cohade, E. K. Fishman, and R. L. Wahl PET-CT in Recurrent Ovarian Cancer: Initial Observations RadioGraphics, January 1, 2004; 24(1): 209 - 223. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Woodward, K. Hosseinzadeh, and J. S. Saenger From the Archives of the AFIP: Radiologic Staging of Ovarian Carcinoma with Pathologic Correlation RadioGraphics, January 1, 2004; 24(1): 225 - 246. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. H. Zou, J. R. Fielding, S. G. Silverman, and C. M. C. Tempany Hypothesis Testing I: Proportions Radiology, March 1, 2003; 226(3): 609 - 613. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. V. Coakley, P. H. Choi, C. A. Gougoutas, B. Pothuri, E. Venkatraman, D. Chi, A. Bergman, and H. Hricak Peritoneal Metastases: Detection with Spiral CT in Patients with Ovarian Cancer Radiology, May 1, 2002; 223(2): 495 - 499. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Rieber, K. Nussle, I. Stohr, D. Grab, S. Fenchel, R. Kreienberg, S. N. Reske, and H.-J. Brambs Preoperative Diagnosis of Ovarian Tumors with MR Imaging: Comparison with Transvaginal Sonography, Positron Emission Tomography, and Histologic Findings Am. J. Roentgenol., July 1, 2001; 177(1): 123 - 129. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
