- Email: [email protected]
Priorities for computed tomography and lymphography in the staging and initial management of Hodgkin's disease
ClinicalRadiology (1984) 35,447-449 © 1984 Royal Collegeof Radiologists
000%9260/84/329447502.00
Priorities for Computed Tomography and Lymphography in the Staging and Initial Management of Hodgkin's Disease R. A. SHIELS, J. STONE, D. V. ASH, S. C. CARTWRIGHT, H. J. CLOSE, T. S. WORTHY and P. J. ROBINSON*
Regional Radiotherapy Centre, Cookridge Hospital and *Diagnostic Imaging Unit, St James's University Hospital, Leeds
Thirty-five patients with Hodgkin's disease were staged with the aid of chest radiographs, bipedal lylnphograms and computed tomography (CT) scans. Computed tomographic findings altered management in only two patients (6%) by indicating enlargement of their radiotherapy fields. After lymphography, five patients (14%) were changed from Stage II (clinical and CT staging) to Stage III, so altering their management. Because either technique may show more extensive disease, CT and lymphography are complementary. Computed tomography should be performed initially. If it reveals no abnormality in the lymphogram area, lymphography, too, should be undertaken. Inverted Y fields are easier to visualise and design from lymphograms than from CT sections.
Both treatment and prognosis of patients with Hodgkin's disease vary with the extent of disease at diagnosis; staging is, therefore, important. Several investigators have reported on the place of computed tomography (CT) in the staging.process (Crowther et al., 1979; Ellert and Kreel, 1980; Lee and Balfe, 1982; Jonsson et al., 1983). In particular, it has been suggested that CT scans render chest tomography and bipedal lymphography unnecessary (Blackledge et al., 1981). We have attempted to assess whether CT (1) influenced the choice of treatment method, (2) caused radiotherapy fields to be changed from those which would have been chosen using lymphography, liver and spleen scintigraphy and chest radiography and (3) improved the detection of infradiaphragmatic disease.
at 24ram intervals from the suprasternal notch to the symphysis pubis were obtained using a rotating fanbeam machine (Philips T300). Half an hour before scanning, patients were given 800 ml of 4% Gastrografin (Schering AG). Intravenous contrast media were not used routinely. Nodes were regarded as abnormal if their smallest diameter measured over 10 mm in the mediastinum, abdomen and pelvis, or over 6ram in the retrocrural space. Patients usually had bipedal lymphography before CT; chest radiographs, mediastinal and hilar tomography were also carried out. RESULTS Comparison of Chest Radiography and CT See Table 2. In two of 10 cases where chest radiography and CT were both positive, CT demonstrated more extensive disease but did not alter management. In one of these patients, CT showed erosion of the sternum which was not detected on chest radiography and, in the other, CT demonstrated additional intrapulmonary lesions. In two patients with normal chest radiographs, CT showed mediastinal disease. One of these was already assessed as Stage IV for other reasons and so received chemotherapy, but the other was changed from Stage I to Stage II and required wider mantle radiotherapy fields. In another patient, CT showed no abnormality but tomography and bone scintigraphy revealed rib lesions. Although this patient was changed from Stage III to Stage IV, management by chemotherapy was unchanged. Comparison of Lymphography and CT
PATIENTS Between May 1980 and June 1983, 35 new patients with Hodgkin's disease were staged with the aid of chest radiography, bipedal lymphography and CT scans of thorax, abdomen and pelvis. Nine patients underwent laparotomy and splenectomy. Staging and histology are shown in Table 1. Of the 35 patients, 13 received radiotherapy, 10 radiotherapy plus chemotherapy and 12 chemotherapy alone.
See Table 3. In the five of nine patients where lymphography and CT were both positive, the latter demonstrated more extensive disease. In four of these, CT showed enlarged nodes in areas not opacified during lymphography, including the mesentery (two cases), splenic hilum and porta hepatis. In the fifth, CT showed hepatic lesions although scintigraphy
Table 1 - Staging and histology METHODS Computed tomography scans using 12mm sections Address for correspondence: Dr P. J. Robinson, Diagnostic ImagingUnit, St James'sUniversityHospital,BeckettStreet, Leeds LS9 7TF.
Initial stage IA IIA IIIA IVA Total
4 9 6 5
Histology IB IIB IIIB IVB 35
0 2 6 3
Lymphocytepredominant Nodular sclerotic Mixed cellularity Lymphocytedepleted
5 17 11 2 35
448
CLINICAL RADIOLOGY Table 4 - Comparison of laparotomy findings and CT scans (nine patients)
Table 2 - Comparison of chest radiography and CT scans
CT scans above diaphragm
Chest radiography + ve
-- ve
+ve -ve
12 23
10 1
2 22
Totals
35
11
24
Table 3 - Comparison of lymphography and CT scans
CT scans below diaphragm
Spleen + Nodes + Liver + -
Laparotomy
CT
3 6
2 (1 false - v e by CT) 7 (1 false +ve by CT)
2 7
1 (false +ve by CT) 8 (2 false - v e by CT)
0 9
0 9
Lymphograrns +VC
--ve
+ve -ve
11" 24
9 8
2* 16
Totals
35
17
18
*Two cases of splenomegaly.
showed only slight hepatomegaly. In no case was lymphography normal and CT abnormal in an area covered by the lymphogram. In two patients the lymphogram showed more extensive nodal disease below the diaphragm than did CT but this did not alter staging or management. In eight others, the lymphogram was abnormal when CT below the diaphragm was normal. In five of these cases, the lymphograms changed the patient's stage from Ii to III because of the demonstration of subdiaphragmatic disease. As a result, management was altered from radiation alone to chemotherapy or to a combination of these two. Comparison of Laparotomy Findings and CT See Table 4: CT demonstrated splenomegaly in two patients, confirmed at laparotomy, with positive histology in one. In the other patient, the enlarged spleen was histologically free of disease. A third patient had a spleen of normal size with histological involvement. One patient with enlarged nodes on CT was free of nodal disease histologically and two patients with normal-sized nodes had positive histology (Table 4). Effect of CT on Treatment Policy Of four patients in whom CT showed more extensive disease than did chest radiography, one was changed from Stage I to Stage II and required an increase in the size of the mantle field. One was changed from Stage III to Stage IIIE without affecting management and in the others neither stage nor management was affected. Of five patients in whom CT showed more extensive disease than was shown by lymphography below the diaphragm, the radiotherapy fields in one were enlarged as a result of CT demonstration of involved retrocaecal nodes, although staging (II) was not affected. In the others, management by chemotherapy or chemotherapy plus radiotherapy was unchanged. One patient, found on CT to have an enlarged spleen, might have been over-treated if laparotomy had not been performed, since subsequent histology showed that the spleen was not involved.
DISCUSSION Previous studies (Rostock et al., 1982, 1983) have found that introducing CT data modified the treatment of Hodgkin's disease in 32-60% of patients. In our series, the management of only two (6%) out of 35 patients was so changed, in both cases by enlarging the radiotherapy portals. Most of the modifications recorded by Rostock et al. (1982, 1983) arose from the demonstration by CT of previously undetected extranodal involvement (e.g. chest wall or pericardial invasion by a mediastinal mass). The lesser impact of CT in our series is partly explained by our more frequent use of chemotherapy or combination therapy, but we have also recorded a lower incidence of extranodal disease in our patient population. Without histological evidence obtained at laparotomy it is impossible in cases of discrepant results to decide which test is correct. The risks of overtreatment are probably smaller than the losses involved in under-treatment so there is a tendency to regard all discrepant results as positive. Certainly, we accepted either an abnormal CT scan or an abnormal lymphogram as evidence of disease below the diaphragm and treated accordingly rather than subjecting all patients to laparotomy. In fact, none of our eight patients with normal CT scans and abnormal lymphograms underwent laparotomy. Ellert and Kreel (1980) reported 25 patients with Hodgkin's disease staged by lymphography and CT. Two of 10 with negative CT scans had abnormal lymphograms in this series. In a larger series, reported by Best and Blackledge (1981), 60 patients underwent lymphography and CT between 1976 and 1979. No patients with normal CT scans had abnormal lymphograms. However, three of the abnormal CT scans showed abnormalities which were within the lymphogram area but were not detected by lymphography. Ten CT scans showed abnormalities outside the lymphogram area. Thus, these authors suggested that, in staging Hodgkin's disease, CT could replace lymphography. None of these series compared lymphography with laparotomy and histology. With any investigative technique , however, there are bound to be some false interpretations. Published series on large numbers of patients with Hodgkin's disease give lymphography a false positive rate of about 10% and a false negative rate of about 3% (Kademian and Wintaner, 1977; Sutcliffe et al., 1979).
CT AND LYMPHOGRAPHY IN HODGKIN'S DISEASE
Computed tomography may be compared with laparotomy findings. Ellert and Kreel (1980) reported on 22 patients who underwent CT and laparotomy. There was one false positive and one false negative CT scan. Best and Blackledge (1981) reported on 105 patients who underwent CT and laparotomy. There were 42 positive laparotomies, 10 true positive CT scans, one false positive CT scan and 31 false negative CT scans (30 of which missed splenic disease). Our own series also includes both false positive and false negative CT results. If patients are not all to undergo laparotomy, then their staging investigations should be as thorough as possible in order not to miss disease. Lymphography and CT are complementary procedures. Patients need not all undergo lymphography but those with normal CT scans in the lymphogram area should do so unless their stage is already so advanced (III or IV) that the demonstration of additional disease could not alter management (Lee and Balfe, 1982) or they are too frail. The design of inverted Y radiotherapy fields and of shielding blocks is greatly facilitated by a lymphogram and intravenous urogram whereas, with CT scanning alone, the fields are more difficult to plan. Acknowledgements. We are grateful to the medical, scientific, radiographic and secretarial staff without whom this work could not have been carried out. Dr Shiels received financial support from the Yorkshire Regional Health Authority and the West Riding Medical Research Trust.
449
REFERENCES
Best, J. J. K. & Blackledge, G. (1981). Do we need CIT. for the management of lymphomas? In Computed Axial Tomography in Oncology, ed. Husband, J. E. & Hobday, P. D., Chap. 8, pp. 59-68. Churchill Livingstone, Edinburgh. Blackledge, G., Mamtora, H., Crowther, D., Isherwood, I. & Best, J. J. K. (1981). The role of abdominal computed tomography in lymphomas following treatment. British Journal of Radiology, 54, 955-960. Crowther, D., Blackledge, G. & Best, J. J, K. (1979). The role of tomography of the abdomen in the diagnosis and staging of patients with lymphoma. Clinics in Haematology, 8, 567-591. Ellert, J. & Kreel, L. (1980). The role of computed tomography in the initial staging and subsequent management of the lymphomas. Journal of Computer Assisted Tomography, 4, 368-391. Jonsson, K., Karp, W., Landberg, J., Mortensson, W., Tennvall, J. & Tylen, U. (1983). Radiologic evaluation of subdiaphragmatic spread of Hodgkin's disease. Acta Radiologica Diagnosis, 24, 153-159. Kademian, M. J. & Wintaner, G. W. (1977). Accuracy of bipedal lymphography in Hodgkin's disease. American Journal of Roentgenology, 129, 1041-1042. Lee, J. K. I. & Balfe, D. M. (1982). Computed tomography evaluation of lymphoma patients. CRC Critical Reviews in Diagnostic Imaging, 18, 2-28. Rostock, R. A., Giorgreco, A., Wharam, M. D., Lenhard, R. E., Siegelman, S. S. & Order, S. E. (1982). C.T. modifications in the treatment of mediastinal Hodgkin's disease. Cancer, 49, 2267-2275. Rostock, R. A., Siegelman, S. S., Lenhard, R. E., Wharam, M. D. & Order, S. E. (1983). Thoracic C.T. scanning for mediastinal Hodgkin's disease: results and therapeutic implications. International Journal of Radiation Oncology, Biology, Physics, 9, 1451-1458. Sutcliffe, S. B., Timothy, A. R. & Lister, J. A. (1979). Staging in Hodgkin's disease. Clinics in Haematology, 8, 593-609.
000%9260/84/329447502.00
Priorities for Computed Tomography and Lymphography in the Staging and Initial Management of Hodgkin's Disease R. A. SHIELS, J. STONE, D. V. ASH, S. C. CARTWRIGHT, H. J. CLOSE, T. S. WORTHY and P. J. ROBINSON*
Regional Radiotherapy Centre, Cookridge Hospital and *Diagnostic Imaging Unit, St James's University Hospital, Leeds
Thirty-five patients with Hodgkin's disease were staged with the aid of chest radiographs, bipedal lylnphograms and computed tomography (CT) scans. Computed tomographic findings altered management in only two patients (6%) by indicating enlargement of their radiotherapy fields. After lymphography, five patients (14%) were changed from Stage II (clinical and CT staging) to Stage III, so altering their management. Because either technique may show more extensive disease, CT and lymphography are complementary. Computed tomography should be performed initially. If it reveals no abnormality in the lymphogram area, lymphography, too, should be undertaken. Inverted Y fields are easier to visualise and design from lymphograms than from CT sections.
Both treatment and prognosis of patients with Hodgkin's disease vary with the extent of disease at diagnosis; staging is, therefore, important. Several investigators have reported on the place of computed tomography (CT) in the staging.process (Crowther et al., 1979; Ellert and Kreel, 1980; Lee and Balfe, 1982; Jonsson et al., 1983). In particular, it has been suggested that CT scans render chest tomography and bipedal lymphography unnecessary (Blackledge et al., 1981). We have attempted to assess whether CT (1) influenced the choice of treatment method, (2) caused radiotherapy fields to be changed from those which would have been chosen using lymphography, liver and spleen scintigraphy and chest radiography and (3) improved the detection of infradiaphragmatic disease.
at 24ram intervals from the suprasternal notch to the symphysis pubis were obtained using a rotating fanbeam machine (Philips T300). Half an hour before scanning, patients were given 800 ml of 4% Gastrografin (Schering AG). Intravenous contrast media were not used routinely. Nodes were regarded as abnormal if their smallest diameter measured over 10 mm in the mediastinum, abdomen and pelvis, or over 6ram in the retrocrural space. Patients usually had bipedal lymphography before CT; chest radiographs, mediastinal and hilar tomography were also carried out. RESULTS Comparison of Chest Radiography and CT See Table 2. In two of 10 cases where chest radiography and CT were both positive, CT demonstrated more extensive disease but did not alter management. In one of these patients, CT showed erosion of the sternum which was not detected on chest radiography and, in the other, CT demonstrated additional intrapulmonary lesions. In two patients with normal chest radiographs, CT showed mediastinal disease. One of these was already assessed as Stage IV for other reasons and so received chemotherapy, but the other was changed from Stage I to Stage II and required wider mantle radiotherapy fields. In another patient, CT showed no abnormality but tomography and bone scintigraphy revealed rib lesions. Although this patient was changed from Stage III to Stage IV, management by chemotherapy was unchanged. Comparison of Lymphography and CT
PATIENTS Between May 1980 and June 1983, 35 new patients with Hodgkin's disease were staged with the aid of chest radiography, bipedal lymphography and CT scans of thorax, abdomen and pelvis. Nine patients underwent laparotomy and splenectomy. Staging and histology are shown in Table 1. Of the 35 patients, 13 received radiotherapy, 10 radiotherapy plus chemotherapy and 12 chemotherapy alone.
See Table 3. In the five of nine patients where lymphography and CT were both positive, the latter demonstrated more extensive disease. In four of these, CT showed enlarged nodes in areas not opacified during lymphography, including the mesentery (two cases), splenic hilum and porta hepatis. In the fifth, CT showed hepatic lesions although scintigraphy
Table 1 - Staging and histology METHODS Computed tomography scans using 12mm sections Address for correspondence: Dr P. J. Robinson, Diagnostic ImagingUnit, St James'sUniversityHospital,BeckettStreet, Leeds LS9 7TF.
Initial stage IA IIA IIIA IVA Total
4 9 6 5
Histology IB IIB IIIB IVB 35
0 2 6 3
Lymphocytepredominant Nodular sclerotic Mixed cellularity Lymphocytedepleted
5 17 11 2 35
448
CLINICAL RADIOLOGY Table 4 - Comparison of laparotomy findings and CT scans (nine patients)
Table 2 - Comparison of chest radiography and CT scans
CT scans above diaphragm
Chest radiography + ve
-- ve
+ve -ve
12 23
10 1
2 22
Totals
35
11
24
Table 3 - Comparison of lymphography and CT scans
CT scans below diaphragm
Spleen + Nodes + Liver + -
Laparotomy
CT
3 6
2 (1 false - v e by CT) 7 (1 false +ve by CT)
2 7
1 (false +ve by CT) 8 (2 false - v e by CT)
0 9
0 9
Lymphograrns +VC
--ve
+ve -ve
11" 24
9 8
2* 16
Totals
35
17
18
*Two cases of splenomegaly.
showed only slight hepatomegaly. In no case was lymphography normal and CT abnormal in an area covered by the lymphogram. In two patients the lymphogram showed more extensive nodal disease below the diaphragm than did CT but this did not alter staging or management. In eight others, the lymphogram was abnormal when CT below the diaphragm was normal. In five of these cases, the lymphograms changed the patient's stage from Ii to III because of the demonstration of subdiaphragmatic disease. As a result, management was altered from radiation alone to chemotherapy or to a combination of these two. Comparison of Laparotomy Findings and CT See Table 4: CT demonstrated splenomegaly in two patients, confirmed at laparotomy, with positive histology in one. In the other patient, the enlarged spleen was histologically free of disease. A third patient had a spleen of normal size with histological involvement. One patient with enlarged nodes on CT was free of nodal disease histologically and two patients with normal-sized nodes had positive histology (Table 4). Effect of CT on Treatment Policy Of four patients in whom CT showed more extensive disease than did chest radiography, one was changed from Stage I to Stage II and required an increase in the size of the mantle field. One was changed from Stage III to Stage IIIE without affecting management and in the others neither stage nor management was affected. Of five patients in whom CT showed more extensive disease than was shown by lymphography below the diaphragm, the radiotherapy fields in one were enlarged as a result of CT demonstration of involved retrocaecal nodes, although staging (II) was not affected. In the others, management by chemotherapy or chemotherapy plus radiotherapy was unchanged. One patient, found on CT to have an enlarged spleen, might have been over-treated if laparotomy had not been performed, since subsequent histology showed that the spleen was not involved.
DISCUSSION Previous studies (Rostock et al., 1982, 1983) have found that introducing CT data modified the treatment of Hodgkin's disease in 32-60% of patients. In our series, the management of only two (6%) out of 35 patients was so changed, in both cases by enlarging the radiotherapy portals. Most of the modifications recorded by Rostock et al. (1982, 1983) arose from the demonstration by CT of previously undetected extranodal involvement (e.g. chest wall or pericardial invasion by a mediastinal mass). The lesser impact of CT in our series is partly explained by our more frequent use of chemotherapy or combination therapy, but we have also recorded a lower incidence of extranodal disease in our patient population. Without histological evidence obtained at laparotomy it is impossible in cases of discrepant results to decide which test is correct. The risks of overtreatment are probably smaller than the losses involved in under-treatment so there is a tendency to regard all discrepant results as positive. Certainly, we accepted either an abnormal CT scan or an abnormal lymphogram as evidence of disease below the diaphragm and treated accordingly rather than subjecting all patients to laparotomy. In fact, none of our eight patients with normal CT scans and abnormal lymphograms underwent laparotomy. Ellert and Kreel (1980) reported 25 patients with Hodgkin's disease staged by lymphography and CT. Two of 10 with negative CT scans had abnormal lymphograms in this series. In a larger series, reported by Best and Blackledge (1981), 60 patients underwent lymphography and CT between 1976 and 1979. No patients with normal CT scans had abnormal lymphograms. However, three of the abnormal CT scans showed abnormalities which were within the lymphogram area but were not detected by lymphography. Ten CT scans showed abnormalities outside the lymphogram area. Thus, these authors suggested that, in staging Hodgkin's disease, CT could replace lymphography. None of these series compared lymphography with laparotomy and histology. With any investigative technique , however, there are bound to be some false interpretations. Published series on large numbers of patients with Hodgkin's disease give lymphography a false positive rate of about 10% and a false negative rate of about 3% (Kademian and Wintaner, 1977; Sutcliffe et al., 1979).
CT AND LYMPHOGRAPHY IN HODGKIN'S DISEASE
Computed tomography may be compared with laparotomy findings. Ellert and Kreel (1980) reported on 22 patients who underwent CT and laparotomy. There was one false positive and one false negative CT scan. Best and Blackledge (1981) reported on 105 patients who underwent CT and laparotomy. There were 42 positive laparotomies, 10 true positive CT scans, one false positive CT scan and 31 false negative CT scans (30 of which missed splenic disease). Our own series also includes both false positive and false negative CT results. If patients are not all to undergo laparotomy, then their staging investigations should be as thorough as possible in order not to miss disease. Lymphography and CT are complementary procedures. Patients need not all undergo lymphography but those with normal CT scans in the lymphogram area should do so unless their stage is already so advanced (III or IV) that the demonstration of additional disease could not alter management (Lee and Balfe, 1982) or they are too frail. The design of inverted Y radiotherapy fields and of shielding blocks is greatly facilitated by a lymphogram and intravenous urogram whereas, with CT scanning alone, the fields are more difficult to plan. Acknowledgements. We are grateful to the medical, scientific, radiographic and secretarial staff without whom this work could not have been carried out. Dr Shiels received financial support from the Yorkshire Regional Health Authority and the West Riding Medical Research Trust.
449
REFERENCES
Best, J. J. K. & Blackledge, G. (1981). Do we need CIT. for the management of lymphomas? In Computed Axial Tomography in Oncology, ed. Husband, J. E. & Hobday, P. D., Chap. 8, pp. 59-68. Churchill Livingstone, Edinburgh. Blackledge, G., Mamtora, H., Crowther, D., Isherwood, I. & Best, J. J. K. (1981). The role of abdominal computed tomography in lymphomas following treatment. British Journal of Radiology, 54, 955-960. Crowther, D., Blackledge, G. & Best, J. J, K. (1979). The role of tomography of the abdomen in the diagnosis and staging of patients with lymphoma. Clinics in Haematology, 8, 567-591. Ellert, J. & Kreel, L. (1980). The role of computed tomography in the initial staging and subsequent management of the lymphomas. Journal of Computer Assisted Tomography, 4, 368-391. Jonsson, K., Karp, W., Landberg, J., Mortensson, W., Tennvall, J. & Tylen, U. (1983). Radiologic evaluation of subdiaphragmatic spread of Hodgkin's disease. Acta Radiologica Diagnosis, 24, 153-159. Kademian, M. J. & Wintaner, G. W. (1977). Accuracy of bipedal lymphography in Hodgkin's disease. American Journal of Roentgenology, 129, 1041-1042. Lee, J. K. I. & Balfe, D. M. (1982). Computed tomography evaluation of lymphoma patients. CRC Critical Reviews in Diagnostic Imaging, 18, 2-28. Rostock, R. A., Giorgreco, A., Wharam, M. D., Lenhard, R. E., Siegelman, S. S. & Order, S. E. (1982). C.T. modifications in the treatment of mediastinal Hodgkin's disease. Cancer, 49, 2267-2275. Rostock, R. A., Siegelman, S. S., Lenhard, R. E., Wharam, M. D. & Order, S. E. (1983). Thoracic C.T. scanning for mediastinal Hodgkin's disease: results and therapeutic implications. International Journal of Radiation Oncology, Biology, Physics, 9, 1451-1458. Sutcliffe, S. B., Timothy, A. R. & Lister, J. A. (1979). Staging in Hodgkin's disease. Clinics in Haematology, 8, 593-609.