| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 All authors: Department of Thoracic Imaging, M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Unit 0371, Houston, TX 77030-4095.
Received April 26, 2005;
accepted after revision August 31, 2005.
Address correspondence to J. F. Bruzzi
(john.bruzzi{at}di.mdacc.tmc.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Images from all PET/CT studies performed consecutively between April and October 2003 on patients with non-small cell lung cancer were retrospectively reviewed. All abnormalities present on the CT component of the PET/CT scans that did not show abnormally increased 18F-FDG uptake were documented.
RESULTS. Three hundred twenty-one patients with non-small cell lung cancer (179 men, 142 women; mean age, 67 years; age range, 38-91 years) underwent initial staging (198/321 [62%]) or restaging (123/321 [38%]) PET/CT imaging during the study period. In 263 (82%) of the patients, CT showed 1,231 abnormalities that were not 18F-FDG avid. The abnormalities were located in the thorax (n = 650), abdomen and pelvis (n = 444), head and neck (n = 69), and bony skeleton (n = 68). In total, 298 (24%) of the abnormalities that were not 18F-FDG avid were located outside the range of a standard thoracic CT scan. The clinical importance of these abnormalities was classified as major (n = 48 [4%]), moderate (n = 465 [38%]), or minor (n = 718 [58%]). Four (1%) of the patients had findings of major clinical importance that did not show increased 18F-FDG uptake and were previously unsuspected.
CONCLUSION. Among patients with non-small cell lung cancer undergoing PET/CT, there is a high prevalence of CT abnormalities that do not show correlative 18F-FDG avidity but that may be clinically important.
Keywords: CT • 18F-FDG PET • incidental abnormalities • lung cancer • PET • PET/CT
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patients fasted for at least 6 hours before the PET/CT study. Blood glucose was measured 1 hour before injection of 18F-FDG; if the blood glucose level exceeded 200 mg/dL, the examination was deferred. Approximately 1 hour before scanning, patients received an injection of a mean of 15 mCi (555 MBq; range, 12-20 mCi, 444-740 MBq) of radioactive-labeled 18F-FDG. All scans were obtained on an integrated PET/CT scanner (Discovery ST-8, GE Healthcare). PET scans were acquired in the 2D mode for 3 minutes per bed position. PET images were reconstructed with standard vendor-provided reconstruction algorithms, which incorporated ordered subset expectation maximization. Attenuation correction of PET images was performed with attenuation data from the CT component of the examination. The manufacturer's software was used to correct emission data for scatter, random events, and dead-time losses.
The CT component of the study comprised an unenhanced MDCT examination from the base of the skull to the upper thighs (120 mA; 140 kVp; table speed, 13.5 mm/rotation). Axial CT images were reconstructed with a slice thickness of 3.75 mm.
For our study, axial CT images from all of the PET/CT studies were retrospectively reviewed on stand-alone PACS workstations (iSite, Stentor) and were correlated with findings from the accompanying PET and fused PET/CT images. Each case was reviewed independently by two diagnostic radiologists experienced in diagnostic CT and PET/CT interpretation. The radiologists documented the presence of CT abnormalities that did not show abnormal 18F-FDG uptake (defined as 18F-FDG accumulation in the lesion that did not exceed that of normal soft-tissue activity in the surrounding normal soft tissue, measured visually). In cases in which the two radiologists recorded different findings, the images were reviewed by a third radiologist, after which all three radiologists made a final decision about which lesions should be recorded. For each patient with an abnormality present on CT scans that did not show increased 18F-FDG uptake, details were recorded concerning the nature of the abnormality, the anatomic location, and any information from previous and subsequent imaging studies and from the medical notes. Note was also made of cases in which the abnormality was regarded as situated outside the range of a staging CT scan of the thorax—that is, the abnormality was considered extrathoracic when it was located either above the level of the thoracic inlet or below the level of the adrenal glands.
All recorded findings were classified as being of major, moderate, or minor significance, corresponding to definitions previously used in similar studies [2, 3]. A finding was considered of major importance when the CT appearance was highly suspicious for an abnormality relevant to the immediate treatment of the patient. Such abnormalities included additional cancers, metastatic lesions, and other abnormalities necessitating prompt clinical assessment or intervention as well as important comorbid conditions expected to seriously compromise the patient's prognosis.
Abnormalities were considered of moderate importance when the imaging appearances suggested clinically important disease but required further correlation with other clinical or imaging findings. Abnormalities of moderate importance included findings that may have required eventual medical or surgical treatment. Abnormalities were considered of minor importance when the imaging appearances did not suggest clinically important disease and did not require additional workup.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Of these patients, 263 (82%) had at least one finding present on the CT component of the studies that did not exhibit correlative 18F-FDG activity. These abnormalities constituted a total of 1,231 findings, of which 650 were located in the thorax, 444 in the abdomen or pelvis, 69 in the head and neck, and 68 in the bony skeleton (Tables 1, 2, 3 and Figs. 1A, 1B, 1C, 1D, 2A, 2B, 2C, 2D, 3A, 3B, 3C, 3D, 4A, 4B, and 4C). One hundred ninety (59%) of the patients had at least one abnormality located outside the range of a standard chest CT examination, accounting for a total of 298 such findings, a rate of 24%.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Findings of Major Importance
Forty-eight (4%) of the abnormalities in 43 (13%) of the patients were
considered of major importance, representing CT findings relevant to the
immediate treatment of the patient. These abnormalities were located in the
thorax (n = 29), abdomen or pelvis (n = 8), head and neck
(n = 5), and bony skeleton (n = 6). These findings are
detailed in Tables 1,
2,
3. Thirteen (1%) of the
findings of major importance that were extrathoracic in location would not
have been detected with a routine chest CT examination.
In 40 of the 43 patients, the abnormalities were already known from previous studies. However, in three (1%) of the patients, the additional abnormalities detected on the CT images represented metastatic disease (one case of left renal metastases and two cases of brain metastases) that had not been suspected before the PET/CT study. These abnormalities were found in one initial baseline study (one case of brain metastasis) and in two restaging studies (one case of brain metastasis and one of renal metastasis). None of these lesions showed 18F-FDG uptake higher than that of normal surrounding soft tissue. In all cases, the corresponding CT images were clearly abnormal. The presence of these metastatic lesions was subsequently confirmed with CT-guided biopsy (the case of renal metastases) (Figs. 1A, 1B, 1C, and 1D) and with MRI (the cases of brain metastasis) (Figs. 2A, 2B, 2C, and 2D). All three patients underwent management of the metastatic lesions (radiofrequency ablation of renal metastasis and radiosurgery of cerebral metastases). In addition, PET/CT depicted a previously unsuspected 6.7-cm abdominal aortic aneurysm in one patient undergoing initial baseline staging of non-small cell lung cancer. The patient underwent surgical repair of the aneurysm.
In total, four (1%) of the patients had additional abnormalities present on PET/CT scans that did not show elevated 18F-FDG uptake and were both clinically important and previously unsuspected. Three of the patients had metastases, and one patient had a large abdominal aortic aneurysm.
Findings of Moderate Importance
The 465 (38% of the total 1,231) non-18F-FDG-avid abnormalities
classified as being of moderate importance were found in 222 (69%) of the
patients. These abnormalities were located in the thorax (n = 313),
abdomen or pelvis (n = 145), and bony skeleton (n = 7).
Seventy-six (6%) of the findings were outside the range of a routine chest CT
scan.
Fifteen (1% of the total 1,231) of the findings initially considered of moderate importance on the basis of their initial CT appearance on PET/CT were found to be malignant on subsequent tests. This number included 10 metastatic pulmonary nodules that had enlarged by the time of follow-up imaging; three cytology-proven malignant pleural effusions, one biopsy-proven hepatocellular carcinoma; and one biopsy-proven prostatic carcinoma. All of these abnormalities were already known from previous imaging studies. Overall, 89 (7%) of the findings of major or moderate importance that were outside the range of a standard CT chest examination were found in 74 (23%) of the patients.
Findings of Minor Importance
The 718 (58% of the total 1,231) findings of minor importance were found in
262 (82%) of the patients. These abnormalities were located in the thorax
(n = 308), abdomen and pelvis (n = 291), head and neck
(n = 64), and bony skeleton (n = 55). Of these findings, 209
(17%) were outside the range of a routine chest CT scan. None of these
abnormalities was found to be of clinical significance on follow-up.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The importance of potential findings on the CT component of PET/CT is reflected in a recent white paper from the American College of Radiology [5] regarding interpretation of PET/CT studies. In the white paper it is recommended that the interpreting radiologist or nuclear physician report additional findings present on CT images, even if the examination is performed purely for anatomic localization only. There is much debate about who should interpret the studies, but it is clear that the CT component of the study cannot be ignored. Therefore, a comprehensive analysis of all CT images in addition to PET images is necessary, although laborious and often unrewarding because the study is routinely performed without IV contrast medium at most institutions. Furthermore, reimbursement rates for interpreting both CT and PET images together do not reflect the extra time required for interpretation of the CT images and the PET images.
Evidence in our study shows that careful evaluation of the CT component of integrated PET/CT scans may provide clinically important information. In our review of the cases of 321 patients with non-small cell lung cancer who underwent integrated PET/CT at our institution, 263 (82%) of the patients had additional findings on CT that did not exhibit abnormal 18F-FDG uptake. Most of these findings either were already known from previous imaging studies or were considered clinically unimportant. Forty-three (13%) of the patients, however, had findings considered of major clinical importance. In the cases of four (1%) of these patients, the additional findings were not previously known and prompted further investigation, yielding three additional cases of malignancy (one case of renal metastasis and two cases of cerebral metastases) and a 6.7-cm infrarenal abdominal aortic aneurysm. Although three of these four patients already had known metastatic disease elsewhere in the body, detection of the additional abnormalities changed clinical management by leading to lesion-specific treatment (radiofrequency ablation of renal metastasis, stereotactic-guided radiosurgery of cerebral metastasis, and surgical repair of aortic aneurysm, respectively).
Our classification of abnormalities as being of major, moderate, or minor clinical importance, although somewhat arbitrary, was based on methods described in similar studies [2, 3] and was used in an attempt to quantify the significance of abnormalities present on initial PET/CT images. Most of the additional findings detected in our study were classified as being of minor importance and included abnormalities such as gallstones, colonic diverticulosis, prostate enlargement, and renal calculi, which are extremely common in the general population and do not warrant additional investigation or follow-up. Early detection of such findings in patients already found to have malignant disease most likely has no immediate clinical benefit. However, we believe it is nevertheless important to document the presence of these clinically minor findings because, if symptoms due to these abnormalities develop, it is likely that the diagnosis will be reached more quickly if the presence of the abnormalities is known. Similarly, such documentation can become clinically relevant in the event of subsequent detection of the abnormalities with other imaging studies. In these cases, comparison with the previous PET/CT studies may become invaluable. Reporting the presence of such apparently minor abnormalities on PET/CT images is therefore in accordance with good clinical care, particularly when the patient is being billed for the CT component of the study in addition to the PET component.
The lack of apparent 18F-FDG accumulation by additional malignant lesions in our study may have occurred for a variety of reasons. In the cases of the two patients with brain metastases, it is already known that PET has relatively low sensitivity for cerebral metastatic lesions, mainly because of the limited spatial resolution of PET scanners and the intense metabolic activity already present throughout the rest of the cortex [6]. In certain cases, brain metastases appear as areas of reduced 18F-FDG uptake compared with the surrounding cortex [7]. It must also be remembered in interpretation of PET/CT studies for restaging that many patients have already undergone a variety of chemotherapy regimens and radiation therapy, which may alter the biology of tumor cells and affect their growth or their ability to take up 18F-FDG. In addition, it should be recognized that certain tumors with variable 18F-FDG uptake, such as prostate cancer, hepatocellular carcinoma, renal cell carcinoma, and osteoblastic bone metastases [8-13], may be present in patients with lung cancer and may be more readily detected owing to their abnormal appearance on CT images than owing to their PET appearance. Therefore, for a variety of reasons, important information may be present on the CT component of PET/CT studies that does not show abnormally elevated 18F-FDG uptake and that may be detected only through careful analysis of all CT images.
Many incidental findings identified on CT can be confidently characterized, including benign adrenal adenomas, bladder diverticulae, lipomas, and dermoid cysts. However, more problematic is the detection of indeterminate lesions that may require further imaging or biopsy before their innocuous nature can be determined. The presence and importance of these indeterminate abnormalities raise issues already being debated in reference to other cross-sectional imaging tools such as CT colonography [2, 3, 14, 15] and lung cancer screening [16]. Swensen et al. [16] reported a high incidence of additional CT findings in their study of lung cancer screening with CT. They identified 696 findings that were deemed clinically important, requiring further evaluation or intervention. Most of these abnormalities were found to be false-positive findings. The detection of additional abnormalities can be expected to be higher with PET/CT, which includes a CT examination of almost the entire body. The benefit of earlier detection of unexpected abnormalities must be carefully balanced against the increased cost of the examination and anxiety experienced by patients.
The foregoing factors pose challenges to the way PET/CT studies are interpreted [17]. It is clearly desirable for the interpreting physician to be able to discriminate clinically important findings that may affect clinical management from more innocuous incidental findings that require no further evaluation. Competence in both diagnostic cross-sectional imaging and PET is ideally required. However, at present these two specialties usually have independent training programs, and few radiologists and nuclear medicine physicians have dual board certification. There is still much discussion about the ideal way in which PET/CT studies should be interpreted, reported, and reimbursed. Clear consensus on such issues is lacking in the radiologic community [18]. Whether the CT images should be interpreted independently of the PET study or as an integral part of the PET/CT examination varies from institution to institution and depends on local issues, such as the way in which PET/CT studies are performed, departmental organization, and billing methods. At our institution, we believe a single report that includes information from both PET and CT images is ideal for communicating all relevant details to the clinician to help direct patient care.
Many additional abnormalities present on CT images from PET/CT studies do not show increased 18F-FDG uptake. Although only a relatively small percentage influence immediate clinical management, we believe these additional findings are of sufficient clinical importance to justify systematic review of the CT images from PET/CT studies and should be included in the final report.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. F. Bruzzi, R. F. Munden, M. T. Truong, E. M. Marom, B. S. Sabloff, G. W. Gladish, R. B. Iyer, T.-S. Pan, H. A. Macapinlac, and J. J. Erasmus PET/CT of Esophageal Cancer: Its Role in Clinical Management RadioGraphics, November 1, 2007; 27(6): 1635 - 1652. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
