| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1 Department of Radiology, Memorial Sloan-Kettering Cancer Center, 1275 York
Ave., New York, NY 10021.
2 Department of Epidemiology and Biostatistics, Memorial Sloan-Kettering Cancer
Center, New York, NY 10021.
3 Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY
10021.
4 Drexel University College of Medicine, Hahnemann Campus, Philadelphia, PA
19102.
Received September 23, 2003;
accepted after revision June 1, 2004.
Address correspondence to M. J. Gollub
(gollubm{at}mskcc.org).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. We observed that the addition of pelvic CT to 201 examinations of the chest and abdomen had a minimal effect on patient treatment. No pelvic examination changed the cancer stage, but three pelvic CT scans in three patients (3%) altered treatment.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To gain knowledge of the effects of this practice on clinical care, we undertook this retrospective study of the findings on the pelvic portion of CT scans involving the chest and abdominal regions. We are unaware of any other published series on pelvic CT findings in esophageal carcinoma.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The years 1997–2001 were queried. From the alphabetical list generated, 99 patients (201 scans) who met the previously mentioned criteria were selected and analyzed. The median number of scans per patient was one (range, 1–10). The specific reason for each CT scan was not examined in this retrospective study, but it was clear from the clinical information that most scans were obtained for initial staging or subsequent restaging of esophageal cancer.
Two radiologists reviewed the official reports available on the radiology information system and recorded the following data: initial versus follow-up scan, use of IV contrast material, findings on the chest and abdominal CT portion of the examination (negative or positive: presence of lymphadenopathy, bone metastases, lung lesions, intestinal masses, liver masses, or peritoneal implants), and findings on pelvic CT (negative or positive: including presence of lymphadenopathy, peritoneal implants, bone metastases, intestinal masses, or adnexal abnormalities). If findings on pelvic CT were positive, determination was made as to whether the disease was confined to the pelvis on the basis of findings in the chest and abdomen on the same scan. Pathology records were reviewed on all patients for histology of the primary tumor (e.g., squamous cell carcinoma [SCC], adenocarcinoma, or other), location of the tumor, and any biopsy, cytology, or surgery specimens in support of positive pelvic findings.
Computerized clinical records in the hospital information system were reviewed for up to 18 months for determination of treatment alterations based on any positive pelvic findings. Details regarding patient tumor stage varied among CT scans and were not consistently recorded. However, most patients had metastatic disease. Radiographs were only rereviewed when questions arose that could not be clarified by the report.
For this investigation, the inferiormost extent of an abdominal CT scan is the beginning of the iliac crest. Any abnormality seen on a CT image that included the pelvic bones was classified as pelvic. Lymphadenopathy was defined according to accepted criteria as nodes whose short axis measured greater than 10 mm.
Generalized estimating equations were used to test for differences in the proportions of scans showing metastases at different sites across types of histology [8]. Logistic regression models were used with an independent working covariance matrix to account for the multiple scans per patient. Analyses were conducted using Stata 7.0 for Windows (Stata Corporation).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Histology revealed 77 patients (156 scans) with adenocarcinoma: 19 patients (32 scans) with SCC, two patients with neuroendocrine carcinoma (11 scans), and one patient with poorly differentiated carcinoma (two scans). Forty-eight primary tumors were located at the esophagogastric junction (101 scans), 35 tumors at the lower third of the esophagus (55 scans), 12 tumors in the mid esophagus, and four tumors in the proximal esophagus.
Sixty-nine patients had 186 positive findings on abdominal examinations as follows: lymphadenopathy (n = 95), liver lesion (n = 39), lung lesion (n = 30), bone lesion (n = 7), intestinal lesion (n = 3), peritoneal implants (n = 9), and adrenal masses (n = 3) (Table 2). We found no significant difference between patients with adenocarcinoma and patients with SCC when comparing the proportion of scans showing liver, lung, or lymph node metastases in these two groups. Pelvic CT scans were positive for metastases or potential metastases in 31 (15%) of 201 scans and 20 (20%) of 99 patients. These included 30 cases of adenocarcinoma and one case of SCC (Table 3). The types of abnormalities associated with these positive pelvic CT scans were as follows: bone metastases (n = 10), lymphadenopathy (n = 10), peritoneal implants (n = 6), bowel wall thickening or mass (n = 2), and ovarian masslike enlargement (n = 3).
|
|
In 10 scans with findings of metastases in the pelvic bones, lower lumbar spine, or sacrum, four scans revealed that bone metastases were also present in the abdomen. Six scans in three patients revealed isolated pelvic bone metastases. In one patient with bone metastasis in the abdomen, no bone metastases were present in the pelvis. In all, in the 10 pelvic scans in which pelvic lymphadenopathy was present, lymphadenopathy was also present in the abdomen. In all six scans in which pelvic peritoneal implants were found, peritoneal implants were also found in the abdomen. In two other patients with implants in the abdomen, pelvic implants were not seen. In two patients with pelvic bowel abnormalities (including one with a sigmoid mass and one with rectal mural thickening), both patients had abdominal metastases. In three patients with ovarian masses, two had peritoneal implants in the abdomen. The third patient had isolated enlarged ovaries for which no follow-up was available. This patient had no abdominal metastatic disease.
Because it is a nonspecific finding, ascites was considered separately from potential metastatic disease and was not included in the tables. Peritoneal implants were required as evidence of peritoneal carcinomatosis. Twenty-seven patients had at least one scan showing ascites. The total number of scans was 31 (initial scans, n = 19; follow-up scans, n = 12), with ascites distributed as follows: abdominopelvic (n = 12), abdominal only (n = 2), and pelvic only (n = 17). No follow-up was available in 15 patients. In the remaining 12 patients, ascites resolved on follow-up CT in seven; remained stable in one; was associated with pancreatitis in one, anasarca in two, and peritoneal carcinomatosis (separately recorded as "implants") in one.
As such, isolated pelvic disease (three patients with bone metastases and one patient with an ovarian mass) was seen in four patients (4%) and seven scans (4%). One patient was asymptomatic but had an impending fracture requiring surgery (patient A). One patient's pain improved on the same chemotherapy regimen with which she was already being treated (patient B), and one patient had intractable pain requiring new radiation treatment (patient C). The patient with enlarged ovaries was asymptomatic but was lost to follow-up. An additional patient without isolated pelvic disease had an obstructed colon; the causative drop metastasis was seen only on pelvic images (patient D). This patient required surgical intestinal bypass. Therefore, in three patients (3/99, 3%; 95% confidence interval, 0.006–0.085), two with bone metastases (patients A and C), and one with a colon metastasis (patient D), treatment was affected by findings on pelvic CT.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patterns of spread of esophageal cancer have been well characterized. Both SCC and adenocarcinoma invade beyond the esophageal wall (T3–T4) to invade the mediastinum (trachea, pericardium, and aorta). Spread to cervical, thoracic, and upper abdominal lymph nodes is common and can skip contiguous stations [13]. Adenocarcinoma spreads hematogenously, particularly to the liver more often than does SCC [14]. In addition, according to at least one series, spread to abdominal lymph nodes occurs more frequently with adenocarcinoma compared with the usually more proximally located SCC [12]. Metastatic tumors to the pelvis can manifest radiologically in several forms, including solid peritoneal implants; malignant ascites; and nodal, skeletal, and gonadal metastases.
On the basis of autopsy studies, peritoneal spread of tumor in esophageal carcinoma occurs in 0–12% of cases [14, 15]. Wayman et al. [16] reported peritoneal spread in 10% of patients with esophagogastric junction tumors, primarily from adenocarcinoma. Whereas drop metastases to the pelvic cul-de-sac could theoretically occur and skip detection in the abdomen, this occurrence has not been reported in autopsy series [15, 17]. All our patients with peritoneal implants in the pelvis had simultaneous implants in the abdomen.
We chose to exclude ascites as potential metastatic disease. This exclusion may have limited our study by potentially underestimating the effect of detecting ascites limited to the pelvis in which no other disease was present (which would change stage and perhaps treatment if revealed to represent metastatic disease). In many instances ascites resolved, and in most instances, there was no follow-up. However, we believed that because of the nonspecificity of this condition, it was reasonable to exclude it.
Spread of esophageal carcinoma via lymphatics has been shown to occur both contiguously and in a skip pattern, with a worse prognosis in the latter [13]. Among the patients in our study, enlarged lymph nodes were not isolated to the pelvis in any case. Biopsy of nodes was not performed because in no case would it have changed the patient's stage. Metastases to pelvic organs or pelvic lymph nodes from esophageal cancer are not reported in autopsy series [15].
Blood-borne skeletal metastases represent the model for random disease spread throughout the body. Bone metastases may be frequent in esophageal cancer. In fact, one investigator found that micrometastases are detectable in up to 88% of rib and 15% of iliac bone-marrow aspirates [18]. Not surprisingly, all the proven pelvic-isolated metastases in our study were found in the bones. More important, in two of the three patients with isolated pelvic bone metastases, a change in treatment was required. Because of the weight-bearing nature of the pelvic bones, metastases here are likely to be clinically more important than in other skeletal locations.
Imaging strategies have changed with time and technologic advancement. Where available, PET or PET/CT (fusion scans) is being used more frequently in the staging of esophageal cancer and is reimbursable by the Center for Medicare and Medicaid Services. PET recently has been shown to reveal a greater number of bone metastases than does 99mTc bone scintigraphy and, more important, reveals areas with active tumor rather than simply bone turnover [19].
As other investigators have shown, pelvic CT is not always necessary, even with malignancies originating below the diaphragm. Giess et al. [20] found infrequent pathologic findings in the pelvis in patients with colon cancer originating in the extrapelvic portions of the colon. Drotman et al. [21] found little usefulness in scanning the pelvis in patients with breast cancer, in which most isolated metastases were also found in bones. The bone lesions had already been evident on routine bone scans [21]. Some investigators even suggested that there may not be a need for pelvic CT when searching for abscesses after surgery confined to the abdomen [22].
Our investigation revealed a rarity of distant metastases from esophageal carcinoma beyond the abdominal viscera. Even when pelvic metastases did occur, they were seldom isolated, never changed the tumor stage, and rarely affected management. Four patients deserve comment: The only patient who had no symptoms had an impending fracture due to an isolated pelvic bone metastasis and required surgery (patient A) and, thus, technically only one patient had a change in treatment solely based on pelvic CT findings. Two other patients (patients C and D) who were symptomatic had clarification of their symptoms using pelvic CT, which then led to proper treatment. In patient D, an obstructing sigmoid metastatic implant was detected, explaining obstipation thought to be a small-bowel obstruction. This patient underwent colon resection. Although this patient's CT scan of the pelvis altered treatment, the implant was not the only metastatic disease in the patient. In patient C, the cause of severe hip pain was shown to be due to metastatic disease, allowing proper intervention with radiation therapy. The fourth patient (patient B) had isolated pelvic bone metastasis; however, her pain was already responding to her ongoing chemotherapy and, as such, management was not affected.
These results suggest little role for the routine use of pelvic CT in esophageal cancer, especially in the current environment of greater radiation awareness and health care cost containment. However, a potential role exists in the search for bone metastases and is, in fact, being piloted in a research setting by at least one group on the basis of the O'Sullivan et al. findings of a high incidence of bone micrometastases [18]. Before the era of PET, this role was reserved for bone scintigraphy. In the current environment in which PET is available, bone metastases are quite sensitively detected, probably obviating the role of scintigraphy.
This study was retrospective and may have been biased. It is difficult to accurately assess retrospectively all the reasons that a particular CT was ordered; furthermore, the indications for a given scan were not formally assessed in this study. In addition, only reports, not scans, were systematically rereviewed. However, it was clear from our review of the clinical statements that most studies were obtained as part of an initial or restaging evaluation of tumor status or monitoring the effects of treatment.
As radiologic technology and advances in cancer care evolve, the diagnostic workup of patients may become more efficient and appropriate. Concerns regarding exposure to radiation, utilization of limited resources, and containment of health care costs continue to promote increased safety, efficacy, and efficiency in patient care. These issues have come to the forefront now more than ever with the ever-increasing volume of CT studies and the speed and ease with which images can be acquired with multidetector technology. In this spirit, we attempted to define whether it was clinically worthwhile in patient care to obtain a pelvic CT scan along with scans of the chest and abdomen. We found isolated pelvic metastases in 4% of patients, mainly confined to the skeletal system. A change in treatment was required for two of these patients and one other patient with an obstructing colon metastasis only seen in the pelvis, or 3% of patients overall. With the proliferation of PET scanners, especially at major cancer centers or with the use of nuclear scintigraphy, these metastases would have been detected easily. In instances in which findings of the pelvic portion of a whole-body PET scan are positive, it would make sense to add a diagnostic (i.e., oral and IV contrast agents) pelvic CT to the routine diagnostic CT of the chest and abdomen that is performed subsequently. As such, on the basis of our findings, we think that from the standpoint of routine care, there is minimal added benefit in the use of pelvic CT in patients with esophageal carcinoma. Further studies of cost–risk benefit are warranted to see if they influence the recommendation for or against using pelvic CT.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
