|
|
||||||||
1 Department of Radiology, College of Medicine, The Catholic University of
Korea, Seocho-Ku, Seoul, South Korea.
2 Department of Radiology, University of Ulsan College of Medicine, Asan Medical
Center, 388-1 Poongnap-Dong Songpa-Ku, Seoul, South Korea.
3 Department of Gynecology, College of Medicine, The Catholic University of
Korea, 505 Banpo-Dong, Seocho-Ku, Seoul, South Korea.
Received September 25, 2001;
accepted after revision February 13, 2002.
Address correspondence to J. M. Lee.
Abstract
|
|
|---|
MATERIALS AND METHODS. We reviewed the records of 31 consecutive patients with pathologically proven epithelial carcinoma who underwent FDG PET 1 month before second-look surgery to assess recurrent tumor. Of these 31 patients, 21 patients also underwent CT 1 month before second-look surgery. The diagnostic accuracies of FDG PET (n = 31), CT (n = 21), and combined FDG PET and CT (n = 21) in detecting recurrent tumor were calculated and compared with each other using the Bennett's test in 21 patients who underwent both imaging studies. Detection rates of individual tumors relative to their sizes were compared between FDG PET and CT using the McNemar test.
RESULTS. The sensitivity, specificity, and accuracy of FDG PET, CT,
and combined FDG PET and CT for revealing recurrent ovarian cancer were 45.3%,
99.7%, 91.0%; 54.5%, 99.6%, 91.7%; 58.2%, 99.6%, 92.4%, respectively. We found
no statistically significant difference in the diagnostic accuracy of FDG PET,
CT, and combined FDG PET and CT (
2 < 5.991). Detection
rates of tumor nodules found on CT were significantly greater than those on
FDG PET when nodule size was 0.3-0.7 cm (p < 0.05).
CONCLUSION. FDG PET did not improve the overall diagnostic accuracy in detecting recurrent ovarian carcinoma compared with CT. Rather, FDG PET was inferior to CT in its ability to reveal small-tumor recurrence.
|
|
|---|
Clinical evaluation with tumor markers or physical examination is of little diagnostic value in the early detection and localization of recurrent ovarian tumor, leading to a reliance on imaging. CT is the most commonly used diagnostic technique before second-look surgery. However, CT is limited in revealing small lesions; even if the lesion is detected, CT cannot confirm the lesion as a tumor recurrence if it is too small. Recently, several reports have evaluated the usefulness of FDG positron emission tomography (PET) in revealing recurrent ovarian carcinoma [4,5,6,7,8,9,10,11]. However, to our knowledge, no reports show the diagnostic value of FDG PET for the detection of small tumor recurrence. The purpose of our study was to evaluate the value of FDG PET for diagnosis of recurrent ovarian carcinoma in comparison with CT and its ability to reveal small tumor recurrence.
|
|
|---|
FDG PET imaging was performed using an ECAT Exact 47 scanner (Siemens, Knoxville, TN) with a slice thickness of 5.1 and a zoom of 1.0 for whole-body coronal, sagittal, and axial images and a slice thickness of 3.5 and a zoom of 1.5 for regional pelvic axial images. Transmission scanning was performed using a germanium-68 rotating rod source during a 20-min acquisition for attenuation correction of regional pelvic images. Patients were injected IV with 370 MBq of FDG. After an uptake period of 45 min, whole-body emission scanning was initiated with six sequential images from the level of the middle auditory canal to the thigh. Emission scanning was then performed during a 30-min acquisition for attenuation-corrected regional pelvic images. Regional pelvic images were acquired as a set of sinograms and were reconstructed by a filtered back-projection method. To limit artifacts, the patients were asked to fast for 12 hr and to drink more than 500 mL of water. All patients received 5 mg of IV furosemide (Lasix; Han dok-Aventis pharma, Seoul, Korea) and were catheterized to reduce bladder activity. In addition, diazepam (10 mg by mouth) (Valium; Roche, Seoul, Korea) was used routinely to reduce FDG uptake in the skeletal muscles. The standardized uptake value was calculated from the amount of injected FDG, the patient's body weight, and the soft-tissue uptake of FDG in the regional images with the attenuation correction.
CT was performed using a Somatom Plus scanner (Siemens, Erlangen, Germany) with an 8- to 10-mm slice thickness and interval. A total of 100 mL of iopromide 62.3% (Ultravist 300; Schering, Berlin, Germany) was injected as a bolus at a rate of 2-3 mL/sec using a mechanical injector. Incremental or helical scanning from the diaphragm to the symphysis began 45 sec after starting the IV injection of contrast medium.
Two radiologists retrospectively interpreted the same CT scans together without the knowledge of histopathologic results or other imaging findings and came to a consensus on the CT findings. If recurrent tumor manifested as a nodule on CT, the size and site of each nodule were evaluated. Miliary peritoneal seeding was considered to be present when peritoneal thickening and contrast enhancement was visualized on CT.
A nuclear medicine specialist retrospectively interpreted the FDG PET scans. This specialist was also unaware of the histopathologic results or other imaging findings. Whole-body coronal, sagittal, axial, and regional axial black-and-white images were viewed on a computer screen. If a hypermetabolic lesion manifested as a nodule on FDG PET, the size and site of each nodule were evaluated. Estimation of lesion size on FDG PET was based on correlation with CT. If the nodule on the attenuation-corrected pelvic image was more than 2 cm in diameter, the standardized uptake value was measured [12]. If the nodule was less than 2 cm in diameter and it was on the outside of the pelvis, semiquantitative and visual analysis were used in the image interpretation for tumor recurrence [13]. On FDG PET, peritoneal seeding was considered present when FDG uptake was prominent on the peritoneal lining and on the surfaces of the solid organs. The presence of tumor recurrence was evaluated on CT and FDG PET using the following three-level confidence score: 0, absent; 1, equivocal; 2, present. On FDG PET, the presence of a hypermetabolic lesion with a standardized uptake value of more than 3 was considered positive for tumor recurrence (score 2), and suspected tumor recurrence with FDG uptake on visual analysis was considered an equivocal lesion (score 1). Scores 1 and 2 were considered positive for tumor recurrence on CT and FDG PET.
The sensitivity, specificity, accuracy, and positive and negative predictive values were evaluated for the 15 specific sites on FDG PET (n = 31), CT (n = 21), and combined FDG PET and CT (n = 21). These were compared with the results of second-look surgery. Furthermore, these diagnostic accuracies were compared with each other using Bennett's test [14] in 21 cases in which both FDG PET and CT were performed.
The detection rate of FDG PET and CT was evaluated for all tumor nodules confirmed at second-look surgery. To calculate the detection rates for a specific tumor size, we included the number of nodules in the denominator that were larger than the tumor size; the number of correctly detected nodules was the nominator. Detection rates of tumors relative to their sizes were calculated and compared between FDG PET and CT using the McNemar test. Tumor sizes measured at second-look surgery and standardized uptake values of nodules of more than 2 cm were also evaluated on FDG PET.
|
|
|---|
|
|
|
|
|
The overall lesion-based sensitivity, specificity, accuracy, positive
predictive value, and negative predictive value of FDG PET (n = 31),
CT (n = 21), and combined FDG PET and CT for revealing recurrent
ovarian carcinoma for 15 specific sites were 45.3%, 99.7%, 91.0%, 97.1%,
90.5%; 54.5%, 99.6%, 91.7%, 96.8%, 91.2%; 58.2%, 99.6%, 92.4%, 97.0%, 91.8%,
respectively (Table 1). When
lesion-based and patient-based diagnostic accuracies were compared among FDG
PET, CT, and combined FDG PET and CT in 21 patients who underwent both imaging
studies, no statistically significant difference was seen for diagnostic
accuracies (
2 < 5.991).
|
Fifty-one recurrent tumor nodules were confirmed at second-look surgery. The size of the tumor nodules ranged from 0.3 to 4.0 cm (mean, 1.1 cm). Detection rates of tumor relative to size on FDG PET and CT at 0.3, 0.4, 0.5, 0.6, and 0.7 cm were significantly greater on CT than those for FDG PET (p < 0.05) (Fig. 3). Although statistically insignificant, tumor detection rates for FDG PET were inferior to those on CT between 0.7 and 1.8 cm. The standardized uptake values of FDG from these measurable tumors of more than 2 cm in diameter were 3.7-16.5 (mean, 8.1).
|
|
|
|---|
There is a clear need for accurate, noninvasive diagnostic imaging to formulate an optimal therapeutic strategy for patients with residual or recurrent ovarian carcinoma and to monitor tumor response to therapy. Diverse imaging techniques have been used for this purpose. Among the imaging techniques, CT is most often used before second-look surgery. CT has considerable limitations revealing small peritoneal implants, which are the most common presentation of recurrent epithelial ovarian cancer. The overall sensitivity of CT for revealing residual tumor or tumor recurrence is reported to be in the range of 40-61% [23,24,25,26,27,28,29]. In addition, CT cannot definitely confirm tumor recurrence when only small nodular lesions are detected. To overcome these inherent limitations of CT, other diagnostic techniques, such as PET, CT with intraperitoneal administration of contrast material [30, 31], and radioscintigraphy using indium or various tumor markers, are alternative methods [32,33,34].
Several reports have suggested that FDG PET can reveal lesions otherwise missed on CT in recurrent ovarian carcinoma [4,5,6,7,8,9]. FDG PET is a form of computer-assisted imaging that produces images reflecting the biochemistry of tissues rather than their physical characteristics. It has the potential to reveal the biochemical differences between normal and malignant tissues in primary and metastatic malignancies. For these reasons, reassessment with FDG PET before second-look surgery in recurrent ovarian carcinoma is gaining acceptance because of the possibility for tumor confirmation when the conventional imaging findings are inconclusive. However, FDG PET also has limitations in revealing small lesions because the clarity of the technique is hampered by the metabolic activity of the tumor.
In our study, FDG PET did not improve overall diagnostic accuracy in revealing recurrent ovarian carcinoma compared with CT. Although patient-based diagnostic values of FDG PET for revealing recurrent ovarian cancer were comparable to those of other reports [4,5,6,7,8,9], the overall lesion-based diagnostic accuracy of FDG PET was much lower. In addition, FDG PET was inferior to CT in revealing small tumors in patients with recurrent ovarian carcinoma. Although in the report by Karlan et al. [6] the researchers could identify tumors of approximately 1.0 cm in diameter using FDG PET, we found detection rates for the nodules of less than 1.0 cm in diameter to be poorlower than 50% in our study. As a result, we suggest that FDG PET should not be used as a routine screening technique for the detection of recurrent tumor in patients with recurrent ovarian carcinoma in which disseminating small peritoneal seeding is common. Furthermore, FDG PET cannot be performed in patients with suboptimal disease in detecting tumor foci considering the same detection rates as those of CT for nodules that are more than 1.8 cm in diameter. However, additional analysis of sensitivity, specificity, and accuracy based on the combined FDG PET and CT subset in the 21 patients who had both studies showed a slight improvement in sensitivity without a corresponding decline in specificity. This finding suggests that an appropriate further study would reveal the marginal value of FDG PET in those patients with a negative or equivocal result on CT.
The principal limitation of this study was that the FDG PET technique might have limited the sensitivity of PET in our study. Specifically, because the transmission imaging was performed before the injection of FDG, the potential for patient motion was present. We made every effort to maintain the exact position before and after the FDG injection by marking the position on the patient using an Exact ECAT PET scanner. The use of measured attenuation correction rather than segmented attenuation correction increased the noisiness of the attenuation-corrected images. Also, we used only nonattenuation-corrected images outside the pelvis, which would be considered a limitation by many. Finally, the use of filtered back-projection as the reconstruction method rather than an iterative reconstruction method was another factor limiting the sensitivity of our PET technique. Therefore, further study with a new FDG PET technique might be warranted.
There are well-known problems with FDG PET. This technique is poor in determining the location of a lesion with only faintly revealed anatomic landmarks. To increase spatial resolution and allow localization of the tumor foci, other researchers have used a fusion image with anatomic imaging such as CT [6, 35, 36]. Another recognized problem with FDG PET is the false-positive and false-negative results. We experienced one false-positive result with an infected pseudocyst; other observers also reported false-positive results in patients with recurrent ovarian carcinoma on FDG PET [4, 5, 9]. Such results are unavoidable because FDG uptake is dependent on the metabolic rate of the imaged tissue. We experienced another false-positive result with FDG in the ureter on initial interpretation. Differentiation of physiologic colonic (bowel wall and lumen) and urine activity from tumor uptake of FDG is a difficult task. However, we could differentiate miliary tumor implants from bowel by tracing the physiologic FDG uptake on sequential FDG PET images. This task was not easy. Also, we experienced one false-negative result when FDG PET was performed 11 days after the completion of chemotherapy. The protocol for FDG PET at our institution is to perform the technique at least 3 weeks after the completion of chemotherapy to reduce false-negative results. When this patient underwent FDG PET 10 days later, recurrent tumor was detected.
In conclusion, FDG PET did not improve the overall diagnostic accuracy for detecting recurrent ovarian carcinoma compared with CT. Rather, FDG PET was inferior to CT in its ability to reveal small tumor recurrence.
Acknowledgments
We thank Yong Gyu Park for statistical analysis of the study; Dae Keun Lo,
Jane Lo, and Shi Jung Lo for assistance with manuscript preparation; and Gi
Jeong Cheon and Jung Mi Park for data analysis and interpretation of FDG PET
images.
|
|
|---|
This article has been cited by other articles:
![]() |
D. J. A. Margolis, J. M. Hoffman, R. J. Herfkens, R. B. Jeffrey, A. Quon, and S. S. Gambhir Molecular Imaging Techniques in Body Imaging Radiology, November 1, 2007; 245(2): 333 - 356. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Nakamoto, S. Sakamoto, T. Okada, M. Senda, T. Higashi, T. Saga, and K. Togashi Clinical Value of Manual Fusion of PET and CT Images in Patients with Suspected Recurrent Colorectal Cancer Am. J. Roentgenol., January 1, 2007; 188(1): 257 - 267. [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] |
||||
![]() |
N. Subhas, P. V. Patel, H. K. Pannu, H. A. Jacene, E. K. Fishman, and R. L. Wahl Imaging of Pelvic Malignancies with In-Line FDG PET-CT: Case Examples and Common Pitfalls of FDG PET RadioGraphics, July 1, 2005; 25(4): 1031 - 1043. [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] |
||||
![]() |
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] |
||||
![]() |
H. K. Pannu, R. E. Bristow, F. J. Montz, and E. K. Fishman Multidetector CT of Peritoneal Carcinomatosis from Ovarian Cancer RadioGraphics, May 1, 2003; 23(3): 687 - 701. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |