Role of Chest CT in Patients with Negative Chest X-Rays Referred for Hepatic Colorectal Metastases Stephen P. Povoski, MD, Yuman Fong, MD, Shirley C. Sgouros, MS, Nancy E. Kemeny, MD, Robert J. Downey, MD, and Leslie H. Blumgart, MD, FRCS
Background: Hepatic resection is the standard treatment for hepatic colorectal metastases. The lung represents the next most likely site, after the liver, of metastatic disease. Computed tomography (CT) of the chest is more sensitive than is chest x-ray in detecting metastatic lung lesions. However, the usefulness of chest CT in the evaluation of patients before hepatic resection remains uncertain. Methods: One hundred consecutive patients with negative chest x-rays and potentially resectable hepatic colorectal metastases underwent chest CT. Patients with CT findings suggestive of metastatic disease were subjected to thoracotomy or video-assisted thoracic surgery (VATS) before laparotomy and attempted hepatic resection. The operative findings and clinical course were analyzed. Results: Eleven of 100 patients had a positive chest CT. Four of these 11 patients had malignant lesions of the lung (three metastatic colorectal cancers and one primary lung cancer). There was no difference in median total hospital stay (8.5 days [range 7 to 13 days] vs. 8.0 days [range 3 to 49 days]), number of perioperative deaths (0 vs. 2 deaths), or long-term outcome between those patients with a positive chest CT undergoing thoracotomy/VATS and those patients with a negative chest CT. Overall, chest CT provided a positive yield of 4% and a positive predictive value of 36% for the detection of malignant lesions of the lung. Conclusions: Chest CT only minimally improved detection of malignant lesions of the lung over chest x-ray. Thoracotomy/VATS and wedge resection of lung nodules did not adversely affect outcome. The low positive yield and low positive predictive value of chest CT in the setting of a negative chest x-ray places in question the usefulness of routinely performing chest CT as part of the extent-of-disease work-up before hepatic resection. Key Words: Hepatic metastases--Colorectal--Computed tomography--Pulmonary metastases.
There are 131,000 new cases of colorectal cancer diagnosed per year in the United States. 1 Approximately 50% of these patients will develop recurrence within 5 years after treatment of their primary colorectal cancer, with the liver representing the site of recurrence in 40%
to 80% of cases. 2-6 Survival of patients with untreated hepatic colorectal metastases to the liver is poor, with median survival, depending on extent of disease, ranging from 4 to 18 months.V-11 Five-year survival is rare. However, during the last 15 to 20 years, hepatic resection has become widely accepted as an effective therapeutic approach for selected patients with liver metastases. Numerous large series have documented both the safety of the procedure, with an operative mortality of 5% or less, and prolongation of postresection survival, with 5-year survival reported in the range of 25% to 4 0 % . I2-32 Likewise, 10-year and 20-year survival rates as high as 28% and 18%, respectively, have been reported. 27 Determining the extent of disease is very important in
Received March 20, 1997; accepted September 2, 1997. From the Departments of Surgery (SPP, YF, SCS, RJD, LHB) and Medicine (NEK), Memorial Sloan-Kettering Cancer Center, New York, New York. Address correspondence and reprint requests to Yuman Fong, MD, Hepatobitiary Surgical Service, Dept. of Surgery, Memorial SloanKettering Cancer Center, 1275 York Ave., New York, NY 10021, USA. Presented at the 50th Annual Cancer Symposium of The Society of Surgical Oncology, Chicago, Illinois, March 20-23, 1997.
S. P. P O V O S K I E T AL.
selecting patients for resection of colorectal cancer metastasized to the liver. The lungs represent the most common extraabdominal site of metastases from colorectal cancer, but are rarely the only site of metastatic disease. 33'34 Therefore, radiographic evaluation of the chest plays an important role in patient selection. Traditionally, conventional chest x-ray has been used to detect pulmonary nodules. More recently, whole lung tomography has been shown to detect pulmonary metastases undetected by conventional chest x-rayY -37 During the past 15 to 20 years, computed tomography (CT) of the chest has been shown to be more sensitive than either conventional chest x-ray or whole lung tomography in detecting metastatic lung lesions in patients with extrathoracic malignancies. 3s-~5 However, the usefulness of chest CT in the extent-of-disease work-up done before hepatic resection in patients with metastatic colorectal cancer to the liver remains uncertain. The purpose of the present study was to evaluate the routine use of chest CT as part of the extent-of-disease work-up in this clinical setting.
MATERIALS AND METHODS Between May 1995 and August 1996, 100 consecutive patients evaluated by the Hepatobiliary Surgical Service at Memorial Sloan-Kettering Cancer Center with potentially resectable metastatic colorectal cancer to the liver and a negative chest x-ray (in both posterior-anterior and lateral projections) were identified by chart review. As standard of care at that time, all 100 patients also underwent chest CT as part of their extent-of-disease work-up. Patients with a negative chest CT were designated as group A. Patients with chest CT findings suggestive of pulmonary metastases were designated as group B. All patients in group B were subjected to thoracotomy or video-assisted thoracic surgery (VATS) before laparotomy and attempted hepatic resection. Thoracotomy or VATS was performed within 6 weeks of the chest CT being obtained. Resectability and the decision to proceed with hepatic resection were determined either at thoracotomy/VATS or at the time of laparotomy. Patients in group A were subdivided into groups of patients undergoing hepatic resection (group A1) and patients determined to be unresectable (group A2). Likewise, patients in group B were divided into subgroups of patients undergoing hepatic resection (group B 1) and patients determined to be unresectable (group B2). Unresectability at thoracotomy/VATS was defined as the presence of multiple, previously unrecognized pulmonary metastases that could not be completely resected and maintain an adequate pulmonary reserve. Unresectability at time of Ann Surg Oncol, VoL 5, No. l, 1998
laparotomy was defined as the presence of intraabdominal, extrahepatic metastatic disease (i.e., nodal or peritoneal metastases) or the presence of extensive bilobar hepatic metastases that would not allow salvage of at least two segments of the liver. The following parameters were recorded and analyzed: demographics (i.e., number of patients, median age, male:female ratio); CT/thoracotomy/VATS findings; resectability of liver metastases; perioperative outcome (i.e., total hospital stay, perioperative deaths); and long-term outcome (i.e., follow-up interval, recurrences, time to recurrence, disease status). Total hospital stay for patients with a negative chest CT was defined as the number of days each patient was hospitalized for laparotomy/hepatic resection. Total hospital stay for patients with a positive chest CT was defined as the sum of the number of days each patient was hospitalized for both laparotomy/hepatic resection and thoracotomy/VATS, whether these procedures were performed during a single or separate admissions to the hospital. Perioperative death was defined as death occurring within 30 days of the operation or during the same hospitalization. Data are expressed as median values (range). Age of patients, total hospital stay, follow-up interval, and time to recurrence were analyzed by Student's t-test. Perioperative deaths and recurrences were analyzed by Fisher's exact test.
RESULTS Demographics There were 89 patients in group A and 11 patients in group B. The median age for group A was 63 years (range 31 to 81 years), as compared to 62 years (range 39 to 81 years) for group B (P = .92). The male:female ratio for group A was 1.0:1.0 (45 males:44 females), as compared to 4.5:1 (9 males:2 females) for group B.
Thoracotomy/VATS Findings Patients in group B were subjected to thoracotomy or VATS and wedge resection of lung nodules before hepatic resection was contemplated (Table 1). Four of these 11 patients were found to have malignant lesions of the lungs. On pathologic examination, malignant lesions of the lungs ranged in size from 0.5 cm to 1.8 cm. Metastatic adenocarcinoma of colorectal origin was found in three of these four patients. One patient had a 0.7-cm deposit of metastatic adenocarcinoma in the right upper lobe (RUL). A second patient had 0.8-cm and 1.8-cm deposits of metastatic adenocarcinoma in the right middle lobe (RML) and a 0.8-cm deposit of metastatic
CHEST CT IN HEPATIC COLORECTAL METASTASES
T A B L E 1.
Patient 1 2 3 4 5 6 7 8 9 10 11
Malignant and benign lung lesions in patients with negative chest x-ray and positive chest CT
0.5 cm; RLL No gross lesion seen; RUL No gross lesion seen; LUL No gross lesion seen; LUL 0.2/0.4 cm; LLL 0.6 cm; RLL 0.5 cm/no gross lesion seen; RML No gross lesion seen; LLL a
CT, computed tomography; IPLN, intrapulmonary lymph node; L, left; LLL, left lower lobe; LUL, left upper lobe; R, right; RML, right middle lobe; RLL, right lower lobe; RUL, right upper lobe; VATS, video-assisted thoracic surgery. a All patients undergoing either thoracotomy or VATS underwent wedge resection of the lung nodules. b Number in parentheses is the number of microscopic lesions with each pathologic diagnosis. c Multiple subcentimeter lesions seen at R VATS, consistent with metastatic adenocarcinoma, but only one such lesion was wedge biopsied. d No lesion seen at L VATS.
adenocarcinoma in the right lower lobe (RLL). A third patient had multiple subcentimeter deposits of metastatic adenocarcinoma in the RUL. Finally, a 0.5-cm primary bronchoalveolar carcinoma was found in the RLL of one of the four patients with malignant lesions of the lungs. Seven of the 11 patients with a positive chest CT were found to have benign pathology of the lungs, including calcified granulomas, intrapulmonary lymph nodes, focal fibrosis, and bronchiolitis obliterans. These ranged in size from no gross lesion seen to 0.6 cm.
Resectability of Liver Metastases Hepatic resection was performed on 65 of 89 (73%) of group A, as compared to 8 of 11 (73%) of group B. Three of the four patients with malignant lesions of the lungs found at thoracotomy/VATS were treated by hepatic resection. All three of these patients underwent VATS/ thoracotomy and hepatic resection at the same operation. The fourth patient did not undergo hepatic resection, secondary to multiple, biopsy-proven, subcentimeter, metastatic colorectal tumor deposits seen at right VATS. Five of the seven patients with benign pathology of the lungs found at thoracotomy/VATS were treated by hepatic resection. Four of these five patients underwent thoracotomy/VATS and hepatic resection during the same operation. One of these five patients underwent thoracotomy/ VATS 14 days before undergoing hepatic resection. Two of the seven patients with benign pathology of the lungs found at thoracotomy/VATS were deemed unresectable at laparotomy because of the intraabdominal extent of disease. One of these two patients underwent thoracotomy/VATS and laparotomy at the same operation. The
other patient underwent thoracotomy/VATS 21 days before undergoing laparotomy.
Perioperative and Long-term Outcome Perioperative and long-term outcome are summarized in Table 2. The median total hospital stay for those patients undergoing hepatic resection was 8.0 days (range 3 to 49 days) for group A1, as compared to 8.5 days (range 7 to 13 days) for group B 1 (P = .79). The median total hospital stay for those patients who were determined to be unresectable was 5.0 days (range 1 to 15 days) for group A2, as compared to 5.0 days (range 3 to 8 days) for group B2 (P = .60). There were two perioperative deaths in group A1, as compared to no perioperative deaths in group B 1 (P = .79). There were no perioperative deaths in either group A2 or group B2. Overall, median follow-up is 11 months (range 3 to 19 months) for all patients in the study. Median follow-up is 12 months (range 4 to 19 months) for group A1, as compared to 9 months (range 7 to 19 months) for group B 1 (P = .79). Median follow-up is 9 months (range 3 to 19 months) for group A2, as compared to 13 months (range 10 to 15 months) for group B2 (P = .27). To date, 27 of 63 (43%) patients in group A1 have had recurrences, as compared to 3 of 8 (38%) of group B1 (P = .29). Nine of 63 (14%) patients in group A1 have had recurrences invoIving the lungs, as compared to 1 of 8 (13%) in group B1 (P = .40). The median time to recurrence for group A1 is 7 months (range 2 to 13 months), as compared to 5 months (range 3 to 8 months) for group B1 (P = .20). For group A1, 39 of 63 (62%) patients are currently Ann Surg Oncol, VoL 5, No. 1, 1998
S. P. P O V O S K I E T AL.
T A B L E 2,
Median total hospital stay (in days) Perioperative deaths Median follow-up (in months) Number of patients in follow-up Recurrences At any site Lung Lung/liver Lung/pelvis Liver Pelvis Chest wall Intraabdominal Median time to recurrence (in months) NED AWD DOD
AWD, alive with disease; CT, computed tomography; DOD, dead of disease; NED, alive with no evidence of disease. Patients with a negative chest CT undergoing hepatic resection. b Patients with a negative chest CT determined unresectable. c One patient in group A2 was lost to follow-up. d Patients with a positive chest CT undergoing hepatic resection. Patients with a positive chest CT determined unresectable. IGroup B2 includes one patient with multiple subcentimeter loci of metastatic adenocarcinoma in the RUL who was determined unresectable at VATS.
alive with no evidence of disease (NED), 18 of 63 (29%) are alive with disease (AWD), and 6 of 63 (10%) are dead of disease (DOD), as compared to group B 1, in which 5 of 8 (63%) patients are NED, 3 of 8 (38%) patients are AWD, and 0 of 8 (0%) patients are DOD. For group A2, 0 of 23 (0%) patients are currently NED, 16 of 23 (70%) patients are AWD, and 7 of 23 (30%) patients are DOD, as compared to group B2, in which 0 of 3 (0%) patients are NED, 2 of 3 (67%) patients are AWD, and 1 of 3 (33%) patients are DOD.
DISCUSSION Before hepatic resection was recognized as an acceptable treatment for hepatic colorectal metastases, the presence of hepatic metastases foretold a poor prognosis, with median survival for patients with untreated hepatic metastases ranging from 4 to 18 months. 7-11 However, over the past two decades, hepatic resection has become accepted as a potentially curative modality of treatment of patients with colorectal metastases to the liver. Numerous series in the literature document the safety of hepatic resection, as well as its efficacy in prolongation of survival, with 5-year survival reported in the range of 25% to 40%. 12-3a However, approximately 60% to 75% of patients surviving hepatic resection for metastatic colorectal cancer to the liver develop further recurfences, 22'46'47 with the liver being the most common site Ann Surg Oncol, Vol. 5, No. l, 1998
(65%) and the lung being the most common extrahepatic site (22%) of recurrent disease. 47 This high incidence of further recurrence probably is related to the presence of undetected hepatic or extrahepatic metastatic disease at the time of hepatic resection. Therefore, a thorough and complete extent-of-disease work-up is extremely important in the selection of appropriate candidates for hepatic resection. Radiographic evaluation of the chest has always played an important role in the selection of appropriate candidates for hepatic resection. Chest CT has been shown to be more sensitive than either conventional chest x-ray or whole lung tomography in detecting metastatic lung lesions in patients with extrathoracic malignancies) T M In a previous report by Chang et al., 43 25 patients with a history of extrathoracic malignancy and new lung nodules were evaluated. Conventional chest x-ray detected 21 lung nodules ranging in size from 0.4 cm to 4.0 cm (median 1.7 cm), all of which were found at operation. Whole lung tomography detected 38 lung nodules ranging in size from 0.3 cm to 1.9 cm (median 0.7 cm), with 32 of 38 detected at operation. Chest CT detected 69 lung nodules ranging in size from 0.2 cm to 4.2 cm (median 0.5 cm), with 47 of 69 detected at operation. Nineteen of 21 (90%) of the lung nodules visible on conventional chest x-ray proved metastatic. Twentyfive of 38 (66%) of the lung nodules visible on whole lung tomography were proved metastatic. Thirty-one of
CHEST CT IN HEPATIC COLORECTAL METASTASES
69 (45%) of the lung nodules visible on chest CT were proved metastatic. Additionally, of those lung nodules seen by chest CT alone, only 20% were proved metastatic. In the study by Chang et al., 43 therefore, chest CT appeared to be more sensitive, but less specific, than conventional chest x-ray in detecting malignant lung lesions in patients with a history of extrathoracic malignancy. Today, chest CT is used routinely at many institutions as part of the extent-of-disease work-up for determination of hepatic resectability in patients with metastatic colorectal cancer to the liver. However, no specific study has proven its utility in this particular setting. In our study, only four of 100 patients with negative chest x-rays and only four of 11 patients with chest CT findings suggestive of pulmonary metastases had thoracotomy/VATS-proven malignant lesions of the lung. This translates into a positive yield of chest CT of only 4% and a positive predictive value of chest CT of only 36%. Thus, chest CT only minimally improved detection of malignant lesions of the lung over conventional chest
x-ray. This suggests only limited value in routine performance of chest CT as part of the extent-of-disease workup before hepatic resection in patients with metastatic colorectal cancer to the liver. For patients undergoing hepatic resection, perioperatire and long-term outcomes were not statistically different for patients with a positive chest CT versus patients with a negative chest CT. These data suggest that thoracotomy/VATS and wedge resection of lung nodules did not negatively affect the outcome of these patients, testifying to the safety with which thoracotomy, VATS, and wedge resection of lung nodules can be performed. Nevertheless, there is no consensus with regard to the routine use of chest CT as a screening tool for the preoperative evaluation of patients with potentially resectable hepatic colorectal metastases who have a negative chest x-ray. Taking our group of 100 patients with a negative chest x-ray, at an estimated cost of $600 per chest CT, an additional $60,000 would have to be spent to perform routine chest CT. Likewise, at an estimated cost of $2500 per thoracotomy/VATS/wedge resection,
ScreeningCXR I NegativeCXR
II I NegativeChestCT
l HepaticResection I
[ HepaticResection ]
FIG. 1. Treatment algorithm for patients referred with potentially resectable hepatic colorectal metastases. InitialIy, a screening chest x-ray (CXR), in both posterior-anterior and lateral projections, is performed in all patients. Patients with a negative CXR proceed to laparotomy and attempted hepatic resection. Patients with CXR findings suggestive of pulmonary metastases are subjected to chest computed tomography (CT). Patients with a negative chest CT proceed to laparotomy and attempted hepatic resection. Patients with chest CT findings suggestive of pulmonary metastases are subjected to thoracotomy or video-assisted thoracic surgery (VATS). If lung pathology is determined to be benign, patients may proceed to laparotomy and attempted hepatic resection. If complete resection of lung metastases is possible, patients may proceed to laparotomy and attempted hepatic resection if clinically indicated. If complete resection of lung metastases is not possible, patients should not proceed to laparotomy and attempted hepatic resection.
Ann Surg Oncol, Vol. 5, No. 1, 1998
S. P. POVOSKI ET AL.
an additional $27,500 would have to be spent to further investigate the 11 suspicious pulmonary lesions seen on chest CT. This gives a total added cost of $87,500 spent to evaluate these 100 patients with a negative chest x-ray for appropriateness for hepatic resection. This added cost is substantial, especially when only four of 100 patients with a negative chest x-ray had thoracotomy/VATSproven malignant lesions of the lung. In the current atmosphere of cost containment and cost-effective medicine, it will likely become more difficult to justify routine chest CT screening of all patients with potentially resectable metastatic colorectal cancer to the liver when our current study demonstrates such a low positive yield and low positive predictive value of chest CT in this specific clinical setting. Increasing experience with lung resection for metastatic colorectal cancer has fostered a trend toward aggressive surgical resection of metastatic lung disease and has been shown to extend survival in patients with isolated lung metastases of colorectal origin. 48-53 Recently, two reports from Memorial Sloan-Kettering Cancer Center have addressed the usefulness of patients undergoing both liver and lung resections for metastases of colorectal origin. 54'55 Both reports have shown that selected patients with liver and lung metastases of colorectal origin should be considered for resection of both metastatic foci. This approach is safe and offers the only possibility of long-term survival, producing survival after the second operation that is essentially the same as would be expected after resection of a single recurrence. Based on these findings, the Hepatobiliary Surgical Service at Memorial Sloan-Kettering Cancer Center has elected to take an aggressive approach of performing both liver and lung resection in patients with potentially resectable metastatic colorectal cancer isolated to the liver and lungs. In summary, chest CT offers only minimal improvement over conventional chest x-ray in the detection of metastatic lesions of the lungs. Thoracotomy/VATS and wedge resection of lung nodules did not negatively affect outcome in patients undergoing hepatic resection. The low positive yield and low positive predictive value of chest CT, as well as the significant potential cost incurred by screening all patients with a chest CT and performing thoracotomy/VATS and wedge resection of lung nodules on all patients with a positive chest CT, places in question the utility of routinely performing chest CT as part of the extent-of-disease work-up done before hepatic resection. As a result of the findings of our paper, we propose that the investigation for possible metastatic disease to the lungs be limited initially to a screening chest x-ray. Only if the chest x-ray is abnormal Ann Surg Oncol, VoL 5, No. 1, 1998
should further investigations, as outlined in Figure 1, be performed. REFERENCES 1. Parker SL, Tong T, Bolden S, Wingo PA. Cancer statistics, 1997. CA Cancer J Ctin 1997;47:5-27. 2. Pestana C, Reitemeier RJ, Moertel CG, Judd ES, Dockerty MB. The natural history of carcinoma of the colon and rectum. Am J Surg 1964;108:826-9. 3. Bross IDJ, Viadana E, Pickren J. Do generalized metastases occur directly from the primary? J Chron Dis 1975;28:149-59. 4. Welch JP, Donaldson GA. The clinical correlation of an autopsy study of recurrent colorectal cancer. Ann Surg 1979;189:496-502. 5. Clarke DN, Jones PF, Needham CD. Outcome in colorectat carcinoma: seven-year study of a population. Br Med J 1980;280: 431-5. 6. Weiss L, Grundmann E, Torhorst J, et al. J Pathol 1986;150:195203. 7. Jaffe BM, Donegan WL, Watson F, Spratt JS. Factors influencing survival in patients with untreated hepatic metastases. Surg Gynecol Obstet 1968;127:1-11. 8. Bengmark S, Hafstrtm. The natural history of primary and secondary malignant tumors of the liver. I. The prognosis for patients with hepatic metastases from colonic and rectal carcinoma by laparotomy. Cancer 1969;23:198-202. 9. Oxley EM, Ellis H. Prognosis of carcinoma of the large bowel in the presence of liver metastases. Br J Surg 1969;56:149-52. 10. Wood CB, Gillis CR, Blumgart LH. A retrospective study of the natural history of patients with liver metastases from colorectal cancer. Clin Oncol 1976;2:285-8. 11. Bengtsson G, Carlsson G, Hafstrtm L, Jtnsson P-E. Natural history of patients with untreated liver metastases from colorectal cancer. Am J Surg 1981;141:586-9. 12. Wilson SM, Adson MA. Surgical treatment of hepatic metastases from colorectal cancers. Arch Surg 1976; 111:330-4. 13. Adson MA, van Heerden JA, Adson MH, Wagner JS, Ilstrup DM. Resection of hepatic metastases from colorectal cancer. Arch Surg 1984; t 19:647-51. 14. Fortner JG, Silva JS, Gotbey RB, Cox EB, Maclean BJ. Multivariate analysis of a personal series of 247 consecutive patients with liver metastases from cotorectal cancer. I. Treatment by hepatic resection. Ann Surg 1984;199:306-16. 15. Wagner JS, Adson MA, Van Heerden JA, Adson MH, Ilstrup DM. The natural history of hepatic metastases from colorectal cancer. A comparison with resective treatment. Ann Surg 1984;199:502-8. 16. Cady B, McDermott WV. Major hepatic resection for metachronous metastases from colon cancer. Ann Surg 1985;201:204-9. 17. Butler J, Attiyeh FF, Daly JM. Hepatic resection for metastases of the colon and rectum. Surg Gynecol Obstet 1986;162:109-13. 18. Bradpiece HA, Benjamin IS, Halevy A, Blumgart LH. Major hepatic resection for colorectal liver metastases. Br J Surg 1987;74: 324-6. 19. Nordlinger B, Parc R, Delva E, Quilichini M-A, Hannoun L, Huguet C. Hepatic resection for colorectal liver metastases. Influence on survival of preoperative factors and surgery for recurrences in 80 patients. Ann Surg 1987;205:256-63. 20. Former JG. Recurrence of colorectal caucer after hepatic resection. Am J Surg 1988;155:378-82. 21. Hughes KS, Rosenstein RB, Songhorabodi S, et al. Resection of the liver for colorectal carcinoma metastases. A multi-institutional study of long-term survivors. Dis Colon Rectum 1988;31:1-4. 22. Hughes KS, Simon R, Songhorabodi S, et al. Resection of the liver for colorectal carcinoma metastases: A malti-institutional study of indications for resection. Surgery 1988;103:278-88. 23. Steele G, Ravikumar TS. Resection of hepatic metastases from
CHEST CT IN HEPATIC COLORECTAL METASTASES
colorectal cancer. Biological perspectives. Ann Surg 1989;210: 127-38. 24. Schlag P, Hohenberger P, Herfarth C. Resection of liver metastases in colorectal cancer--competitive analysis of treatment results in synchronous verses metachronous metastases. Eur J Surg Oncol 1990; 16:360-5. 25. Vetto JT, Hughes KS, Rosenstein R, Sugarbaker PH. Morbidity and mortality of hepatic resection for metastatic colorectal carcinoma. Dis Colon Rectum 1990;33:408-13. 26. Doci R, Gennari L, Bignami P, Montalto F, Morabito A, Bozzetti F. One hundred patients with hepatic metastases from colorectal cancer treated by resection: analysis of prognostic determinants. Br J Surg 1991;78:797-801. 27. Scheele J, Stangl R, Altendorf-Hofmann A, Gatl FP. Indicators of prognosis after hepatic resection for colorectal secondaries. Surgery 1991;110:13-29. 28. Steele G, Bleday R, Mayer RJ, Lindblad A, Petrelli N, Weaver D. A prospective evaluation of hepatic resection for colorectal carcinoma metastases to the liver: gastrointestinal tumor study group protocol 6584. J Clin Oncol 1991;9:1105-12. 29. Rosen CB, Nagomey DM, Taswell HF, Helgeson SL, Ilstrup DM, Van Heerden JA, Adson MA. Perioperative blood transfusion and determinants of survival after liver resection for metastatic coinrectal carcinoma. Ann Surg 1992;216:493-505. 30_ Fong Y, Blumgart LH, Fortner JG, Brennan MF. Pancreatic or liver resection for malignant neoplasms is safe and effective in the elderly. Ann Surg 1995;222:426-437. 31. Fong Y, Kemeny N, Paty P, Blumgart LH, Cohen AM. Treatment of colorectal cancer: Hepatic metastasis. Semin Surg Oncol 1996; 12:219-52. 32. Fong Y, Cohen AM, Former JG, Morerro A, Prassad M, Blumgart LH, Brennan MF. Resection is safe and effective for hepatic colorectal metastases: an analysis of 456 consecutive cases. J Ctin Oncol 1997;15:938-46. 33. Gilbert JM, Jeffrey I, Evans M, Kark AE. Sites of recurrent tumours after "curative" colorectal surgery: implications for adjuvant therapy. Br J Surg 1984;71:203-5. 34. Putnam JB, Roth JA. Secondary tumors in the lung. In: Shields TW, Ed. General Thoracic Surgery. 4th ed. Baltimore: Williams & Wilkins, 1994:1334-52. 35. Moody DL, Edlich RF, Gedgaudas E. The roentgenologic identification of pulmonary metastases: evaluation of an operativelyproved series. Dis Chest 1967;51:306-10. 36. Didolkar MS, Cedermark BJ, Goel IP, Takita H, Moore RH. Accuracy of roentgenograms of the chest in metastases to the lungs. Surg Gynecol Obstet 1977;144:903-5. 37. Neifeld JP, Michaelis LL, Doppman JL. Suspected pulmonary metastases. Correlation of chest x-ray, whole lung tomograms, and operative findings. Cancer 1977;39:383-7. 38. Sagel S, Stanely RJ, Evens RG. Early clinical experience with motionless whole-body computed tomography. Radiology 1976; 119:321-30.
39. Kollins SA. Computed tomography of the pulmonary parenchyma and chest wall. Radiot Clin North Am 1977;15:297-308. 40. Mulm JR, Brown LR, Crowe JK. Detection of pulmonary nodules by computed tomography. A JR 1977; 128:267-70. 41. Muhn JR, Brown LR, Crowe JK. Use of computed tomography in the detection of pulmonary nodules. Mayo Clin Proc 1977;52: 345-8. 42. Schaner EG, Chang AE, Doppman JL, Conkle DM, Flye MW, Rosenberg SA. Comparison of computed and conventional whole lung tomography in detecting pulmonary nodules: a prospective radiologic-patholngic study. A JR 1978;131:51-4. 43. Chang AE, Schaner EG, Conkle DM, Flye MW, Doppman JL, Rosenberg SA. Evaluation of computed tomography in the detection of pulmonary metastases. A prospective study. Cancer 1979; 43:913-6. 44. McLoud TC, Wittenberg J, Ferrucci JT. Computed tomography of the thorax and standard radiographic evaluation of the chest: a comparative study. J Comput Assist Tomogr 1979;3:170-80. 45. Zerhouni EA, Stitik FP, Siegelman SS, et al. CT of the pulmonary nodule: a cooperative study. Radiology 1986;160:319-27. 46. Bozzetti F, BignamJ P, Morabito A, Doci R, Gennari L. Patterns of failure following surgical resection of colorectal cancer liver metastases. Rationale for a multimodal approach. Ann Surg 1987;205: 264-70. 47. Ekberg H, Tranberg K-G, Andersson R, Lundstedt C, H~igerstrand I, Ranstam J, Bengmark S. Pattern of recurrence in liver resection for colorectal secondaries. World J Surg 1987;11:541-7. 48. McCormack PM, Attiyeh FF. Resected pulmonary metastases from colorectal cancer. Dis Colon Rectum 1979;22:553-6. 49. Hughes ESR, McConchie IH, McDermott FF, Johnson WR, Price AB. Resection of lung metastases in large bowel cancer. Br J Surg 1982;69:410-2. 50. Wilking N, Petrelli NJ, Herrera L, Regal A-M, Mittelman A. Surgical resection of pulmonary metastases from colorectal adenocarcinoma. Dis Colon Rectum 1985;28:562-4. 51. Goya T, Miyazawa N, Kondo H, Tsuchiya R, Naruke T, Suemasu K. Surgical resection of pulmonary metastases from colorectal cancer. 10-year follow-up. Cancer 1989;64:1418-21. 52. Sauter ER, BoRon JS, Willis GW, Fair GH, Sardi A. Improved survival after pulmonary resection of metastatic colorectal carcinoma. J Surg OncoI 1990;43:135-8. 53. Mori M, Tomoda H, Ishida T, et al. Surgical resection of pulmonary metastases from colorectal adenocarcinoma. Arch Surg 1991; 126:1297-1302. 54. Smith JW, Fortner JG, Burt M. Resection of hepatic and pulmonary metastases from colorectal cancer. Surg Oncol 1992;1:399404. 55. Minnard EA, Fong Y, Weigel T, Blumgart LH, Butt M. Surgical resection for hepatic and pulmonary colorectal metastases. J Clin Oncol (in press).