Radiographer screening for incidental pulmonary emboli on routine contrast-enhanced computerised tomography scans at a cancer centre.

AIM: To introduce and assess effectiveness of a radiographer-led screening programme for the detection of unsuspected pulmonary emboli on routine contrast-enhanced computed tomography (CT), and to evaluate radiographer response to this extended role. MATERIALS AND METHODS: A training programme was devised for all radiographic staff working in CT. The screening service was introduced and monthly quality assurance performed with cumulative analysis of the first 2 years. Clinical effectiveness before and after screening was evaluated by comparing the time interval between the scan and the start of a clinical consultation for anticoagulant prescription. A satisfaction survey was sent to all participating staff. RESULTS: Thirty-two radiographers completed the training. During the training period, the radiographer detection rate of incidental pulmonary emboli was 89%. Main, lobar, segmental, and subsegmental emboli were detected. The overall detection rate after full introduction of the programme was 92% for the first 2 years. The time interval between the scan and clinical consultation for anticoagulant prescription dropped from a mean of 1.5 days to a mean of 26 minutes and ensured that treatment was commenced at the same patient attendance. Eighty-four percent of staff completed the satisfaction survey and all were satisfied with the extended role. CONCLUSION: Radiographer screening for incidental pulmonary emboli was effective and accurate. It resulted in immediate communication with the responsible physician and commencement of anticoagulation therapy at the same hospital attendance, creating a "one-stop" service. Radiographer satisfaction with the extended role was high.

Enhancing fraction measured using dynamic contrast-enhanced MRI predicts disease-free survival in patients with carcinoma of the cervix.

BACKGROUND: There is a need for simple imaging parameters capable of predicting therapeutic outcome. METHODS: This retrospective study analysed 50 patients with locally advanced carcinoma of the cervix who underwent dynamic contrast-enhanced MRI before receiving potentially curative radiotherapy. The proportion of enhancing pixels (E(F)) in the whole-tumour volume post-contrast agent injection was calculated and assessed in relation to disease-free survival (DFS). RESULTS: Tumours with high E(F) had a significantly poorer probability of DFS than those with low E(F) (P=0.011). INTERPRETATION: E(F) is a simple imaging biomarker that should be studied further in a multi-centre setting.

The value of hyoscine butylbromide in pelvic MRI.

AIM: To evaluate the effect of hyoscine butylbromide (HBB) on image quality and lesion and organ visualization in pelvic magnetic resonance imaging (MRI) MATERIALS AND METHODS: A prospective, ethically approved study was undertaken of 47 patients attending for pelvic MRI at a cancer centre. T2-weighted transverse and sagittal sequences were performed before and after intravenous injection of 20 mg HBB. Three radiologists independently scored anonymized image series for overall image quality, visualization of pelvic lesions and visualization of individual pelvic organs. Statistical analysis was performed to assess improvements in radiologists' scores post-HBB administration. Radiologists also assessed pre-HBB administration T1-weighted images for degree of bowel peristalsis to determine whether this could predict improvement in post-HBB T2-weighted image scores. Side effects of HBB were recorded using a patient questionnaire. RESULTS: Radiologists' scores for image quality and lesion visualization were significantly higher on the post-HBB administration T2-weighted series (p<0.0005). Scores for the visualization of the bladder, rectum, pelvic bowel, prostate, and seminal vesicles (all p<0.0005), cervix (p=0.019) and vagina (p=0.0001) were also significantly higher post-HBB administration. Scores for the degree of peristalsis on T1-weighted images were not related to improvement in image quality or lesion visualization on T2-weighted images post-HBB administration. Side effects of HBB were mild and self-limiting. CONCLUSION: Intravenous HBB administration improves image quality and lesion visualization in oncological pelvic MRI and is recommended for routine use.

Magnetic resonance imaging of primary vaginal carcinoma.

AIMS: To describe the magnetic resonance imaging (MRI) features of vaginal carcinoma and to suggest a role for MRI in its management. MATERIALS AND METHODS: Twenty-five patients with primary vaginal carcinoma treated at our institution between 1996 and 2005 were included in the study. The MRI examinations were reviewed and tumour dimensions, signal characteristics and involvement of pelvic structures were documented, as were sites of enlarged lymph nodes and metastases. Details of patient treatment and outcome were obtained from the clinical notes. RESULTS: The median patient age was 54 years (range 31-86 years). Tumour maximum diameter ranged from 1.6-11.3 cm (mean 3.7 cm). Most tumours were of iso-intense signal to muscle on T1-weighted images and hyper-intense to muscle on T2-weighted images. Eighty-eight percent of patients had tumour extending beyond the vagina and 56% of patients had Figo stage III or above tumours. Sixteen patients were treated with radiotherapy (two with chemoradiotherapy), five with surgery and four with supportive care. Ten patients (40%) died of their disease during the study period. The MRI stage of the tumour correlated with survival. CONCLUSION: MRI identified over 95% of primary vaginal tumours in the present study, enabled radiological staging, which correlated with outcome, and provided information of use in treatment planning.

Magnetic resonance imaging of anal cancer.

AIM: The purpose of this study was to evaluate the magnetic resonance imaging (MRI) appearances of primary and recurrent anal carcinoma, and to demonstrate the commonest patterns of local and distant disease spread. METHODS: A retrospective review was performed of 27 cases of biopsy-proven anal carcinoma, where MRI was used for primary staging (9 patients) or suspected recurrence (18 patients). Two oncological radiologists reviewed the MR images, following a standardized approach. The size, extent and signal characteristics of the anal tumour were documented. Metastatic disease spread to lymph nodes, viscera and bone was recorded. In all, 7 patients with recurrent disease underwent surgery and subsequent histological correlation was performed. RESULTS: Primary and recurrent tumours were of high signal intensity relative to skeletal muscle on T2-weighted images (T2WI), and of low to intermediate signal intensity on T1-weighted images (T1WI). Lymph node metastases were of similar signal intensity to the anal cancer. Recurrent tumours were more locally advanced than primary tumours and extended into adjacent organs and the pelvic skeleton. Recurrent lymph node disease involved perirectal, presacral and internal iliac nodes more commonly than did primary lymph node disease. CONCLUSION: MRI can be useful in the primary staging of bulky tumours or of those with a long craniocaudal extent. MR has a role in the preoperative evaluation and surgical planning of cases of recurrent disease following radiotherapy.

Cross-sectional imaging in non-melanoma skin cancer of the head and neck.

AIM: To investigate in head and neck non-melanoma skin cancers (NMSCs) the accuracy of cross-sectional imaging for detection of local tumour extent, recurrent tumour and prediction of patient outcome. METHODS: This retrospective study included 33 NMSC patients (22 men, 11 women, median age 69 years) with 8 primary and 25 suspected recurrent tumours. The findings of magnetic resonance imaging (MRI) and computed tomography (CT) were compared with histopathology, and accuracy of MRI or CT in detecting local recurrence was determined. Extent of disease on imaging was compared with patient outcome assessed by clinical follow-up to a mean of 26.4 months. RESULTS: Lesions were identified in 29 patients, whose mean disease-free survival (DFS) was 25.5 months. In 4 of these cases, where imaging showed no invasion of deep structures, DFS was 56 months. In the other 25 cases DFS was 20.6 months, irrespective of treatment but varying with site of involvement. Of 19 patients treated with surgery, imaging of 16 showed deep invasion, which was confirmed at histology in 15 (93.7% accuracy), and 3 had superficial tumours on imaging all confirmed by histology (100% accuracy). Imaging accuracy for identifying recurrent tumour was 96% (24 of 25 patients). CONCLUSION: In NMSC, cross-sectional imaging accurately identifies tumour extent and local recurrence. The extent of disease and invasion of deeper structures predicts patient outcome.

Tositumomab and iodine I 131 tositumomab for recurrent indolent and transformed B-cell non-Hodgkin's lymphoma.

PURPOSE: An open-label phase II study was conducted at two centers to establish the efficacy and safety of tositumomab and iodine I 131 tositumomab at first or second recurrence of indolent or transformed indolent B-cell lymphoma. PATIENTS AND METHODS: A single dosimetric dose was followed at 7 to 14 days by the patient-specific administered radioactivity required to deliver a total body dose of 0.75 Gy (reduced to 0.65 Gy for patients with platelets counts of 100 to 149 x 10(9)/L). Forty of 41 patients received both infusions. RESULTS: Thirty-one of 41 patients (76%) responded, with 20 patients (49%) achieving either a complete (CR) or unconfirmed complete remission [CR(u)] and 11 patients (27%) achieving a partial remission. Response rates were similar in both indolent (76%) and transformed disease (71%). The overall median duration of remission was 1.3 years. The median duration of remission has not yet been reached for those patients who achieved a CR or CR(u). Eleven patients continue in CR or CR(u) between 2.6+ and 5.2+ years after therapy. Therapy was well tolerated; hematologic toxicity was the principal adverse event. Grade 3 or 4 anemia, neutropenia, and thrombocytopenia were observed in 5%, 45%, and 32% of patients, respectively. Secondary myelodysplasia has occurred in one patient. Four patients developed human antimouse antibodies after therapy. Five of 38 assessable patients have developed an elevated thyroid-stimulating hormone; treatment with thyroxine has been initiated in one patient. CONCLUSION: High overall and CR rates were observed after a single dose of tositumomab and iodine I 131 tositumomab in this patient group. Toxicity was modest and easily managed.

Staging of advanced cervical carcinoma using MRI-predictors of outcome after radical radiotherapy.

AIM: To assess the prognostic significance of imaging findings used in magnetic resonance imaging (MRI) staging of cervical carcinoma by correlation with survival after radiotherapy. MATERIALS AND METHODS: MRI examinations of 99 cervical carcinoma patients were reviewed. Tumour involvement of pelvic structures was assessed. Lymph node sites, short axis diameters and signal characteristics were recorded. MRI staging was compared with clinical [International Federation of Gynaecology and Obstetrics (FIGO)] staging. Univariate analysis was performed for MRI stage, clinical stage, nodal status and pelvic structure involvement against disease-specific (DSS) and disease-free survival (DFS). RESULTS: MRI staging correlated with DSS (p=0.006) and DFS (p=0.007) but clinical staging did not. Pelvic nodes > or = 10 mm and juxtaregional or distant nodes > or = 8 mm short axis were most strongly associated with survival (p=0.014, p=0.011 and p=0.001, respectively, for association with DSS). Tumour involvement of pelvic bowel loops, pelvic sidewall and bladder mucosa were significantly associated with poor DSS and DFS (p<0.05). Tumour dimensions and bladder muscle involvement alone were not associated with poor survival. CONCLUSION: MRI staging is a better predictor of survival than clinical staging in patients receiving radiotherapy for cervical carcinoma. MRI assessment of lymph node enlargement and tumour involvement of pelvic structures gives valuable prognostic information.

Symptoms of colorectal liver metastases: correlation with CT findings.

AIM: To correlate CT appearances of colorectal liver metastases (LM) with pattern and severity of symptoms. MATERIALS AND METHODS: One hundred and twenty patients with treated primary colorectal carcinoma were prospectively assessed by questionnaire for recent symptoms when attending for CT examination. Thorax, abdomen and pelvic CT scans were prospectively assessed for LM and extrahepatic disease (EHD). The number of LM, percentage liver replaced by LM and distribution of LM were recorded. RESULTS: Patients' ages ranged from 35 to 89 years (median 60) and 74/120 (62%) were male. Four subgroups were compared: group 1 - LM only (n = 30); 2 - EHD only (n = 22); 3 - both LM and EHD (n = 28); 4 - neither LM/EHD (n = 40). Anorexia was significantly worse in gp2 vs gp4 (P = 0.016) and lower abdominal pain (LAP) was significantly worse in gp2 vs gpl (P = 0.019). General pain was the worse symptom in all groups but notstatistically greater in any group. Patients with more than 10 LM had significantly worse anorexia (P = 0.002). general pain (P < 0.001) and LAP (P = 0.001). There was a trend (P > 0.05) towards worse symptoms with either volume of diseased liver or subcapsular LM. CONCLUSION: With increasing liver tumour burden there was an increase in symptomatology but extrahepatic abdominal metastatic tumour produced more symptoms than LM alone. Symptoms, particularly pain, therefore are not good predictors of hepatic metastatic disease.

Cerebral metastasis and other cerebral events in women with ovarian cancer.

AIM: To determine the incidence, imaging findings and prognostic significance of cerebral metastases and other cerebral events in women with ovarian cancer. METHOD: A 5-year retrospective review of all women with ovarian cancer who had cranial imaging was undertaken at two major gynaecological oncology centers. RESULTS: Of 1222 women under clinical review, 78 underwent cranial imaging and 13 (1.1%) had cerebral metastasis. Computed tomography (CT) was diagnostic of parenchymal disease in 12 and magnetic resonance imaging (MRI) showed leptomeningeal disease in two. The women were aged between 23 and 73 years and all had stage III or IV disease at presentation. Cerebral metastasis occurred at 6-60 months from initial diagnosis, with death occurring predominantly within 12 months, but with five survivors at 4-45 months. Of the remaining 65 women, 10 had cerebrovascular disease and three had unrelated lesions. CONCLUSION: Cerebral metastasis remains a rare event in women with ovarian cancer but may be an isolated late event associated with survival beyond a year after neurosurgery and chemotherapy. CT should be the first investigation as the incidence of cerebrovascular disease is similar to that of metastatic disease.

The assessment of irradiated bladder carcinoma using dynamic contrast-enhanced MR imaging.

AIM: To evaluate the role of dynamic contrast-enhanced magnetic resonance imaging (DCEMRI) in distinguishing residual or recurrent tumour from radiation change in patients with bladder carcinoma. MATERIALS AND METHODS: Forty patients with biopsy proven bladder carcinoma were imaged before and at 4 and 12 months after radiotherapy (XRT) using conventional and dynamic contrast-enhanced magnetic resonance imaging at 0.5 Tesla. An enhancement of >1.54 times above baseline at 80 s post-contrast injection proved a reliable indicator of tumour before radiotherapy and was therefore applied to the assessment of patients after XRT. Conventional MR images and dynamic enhancement profiles (DEPs) from the site of previous tumour were scored by three radiologists for the presence of tumour at 4 and 12 months after XRT. Findings were compared with cystoscopic biopsy. RESULTS: Dynamic contrast-enhanced magnetic resonance imaging had negative predictive values of 100% and 93% for tumour recurrence at 4 and 12 months, respectively. The positive predictive values, sensitivity and specificity were 48, 100 and 48% at 4 months and 50, 80 and +76% at 12 months post XRT, respectively. CONCLUSION: Dynamic contrast-enhanced magnetic resonance imaging may prove reliable in excluding the presence of persistent or recurrent tumour up to 12 months after XRT.

The effect of radiotherapy treatment changes on sites of relapse in stage I testicular seminoma.

AIM: To evaluate relapse patterns in stage I testicular seminoma related to changes in radiotherapy practice. METHOD: Four hundred and six patients with stage I testicular seminoma were treated with adjuvant radiotherapy following orchidectomy: 338 patients received para-aortic radiotherapy only and 68 patients with added risk factors had radiotherapy extended to include the pelvis. Computed tomograms of relapsed patients were reviewed and sites of relapse were documented with correlation to the radiotherapy field. RESULTS: Thirteen relapses were identified; 10 occurring in the para-aortic radiotherapy group (3.0% relapse rate) and three in the extended radiotherapy field group (4.4% relapse rate). Sites of relapse were; five pelvis, three mediastinum, one lung, one scapula, one scrotum, while one patient had multiple relapse sites including the pelvis and one had a tumour marker relapse with no site identified. All the pelvic relapses occurred in the para-aortic radiotherapy group. CONCLUSION: Pelvic relapse only occurred when radiotherapy had been confined to the para-aortic region. Since para-aortic radiotherapy achieves equivalent outcome to wider field radiotherapy with reduced toxicity, it is likely to become standard practice in stage I seminoma and pelvic relapses will therefore increase in frequency. It is therefore important to include pelvic imaging when relapse is suspected.

Accuracy of axillary MR imaging in treated breast cancer for distinguishing between recurrent tumour and treatment effects: does intravenous Gd-DTPA enhancement help in cases of diagnostic dilemma?

AIM: To evaluate the sensitivity and specificity of axillary magnetic resonance imaging (MRI) in symptomatic patients, who had previously been treated for breast cancer, compared with clinical outcome after a minimum of 1 year. METHODS: One hundred and five patients underwent axillary MRI examinations and were diagnosed as axillary tumour, metastatic tumour, treatment effect or normal. RESULTS: At MRI, 48 patients had axillary tumour, 51 had metastatic tumour (37 had both), 27 had treatment effect and 22 were normal. At outcome (median follow-up, 484 days), 54 patients were positive for axillary tumour, 59 for metastatic disease (40 had both), 21 had treatment effect alone and 18 were clear. Magnetic resonance imaging showed 89% sensitivity, 100% specificity and 94% accuracy for recurrent axillary tumour, and 85% sensitivity, 98% specificity and 90% accuracy for metastatic tumour. Soft tissue plaques were the commonest axillary disease pattern seen (37). Small volume soft tissue plaques gave the most diagnostic difficulty. Non-dynamic enhancement with intravenous Gadopentetate dimeglumine (Gd-DTPA) in a subset of 34 patients improved sensitivity for axillary tumour from 40 to 74%, and improved diagnostic confidence in 11 patients (32%). Magnetic resonance imaging had a positive management impact leading to treatment alteration in 45 patients, 43 of whom had recurrent axillary and/or metastatic tumour. CONCLUSIONS: Tumour plaques were the commonest pattern of recurrent axillary disease. Forty-eight percent of the patients had metastatic deposits identified by MRI. Magnetic resonance imaging had excellent specificity (100%) and good sensitivity (89%) for recurrent axillary tumour compared with outcome at 1 year, which was improved by non-dynamic administration of Gd-DTPA in 32% of the subset who received it.

The post-cystectomy pseudotumour sign: MRI appearances of a modified chronic pelvic haematoma due to retained haemostatic gauze.

The magnetic resonance imaging (MRI) findings of five men who had incidental similar pelvic masses identified after radical cystectomy are presented. In all patients the haemostatic agent Kaltostat had been used. In one patient, surgical resection of the mass was performed and histological evaluation showed a foreign body inflammatory reaction within a chronic haematoma. The differentiation between this lesion and other post-operative collections or tumours is discussed.

Tumour oxygenation levels correlate with dynamic contrast-enhanced magnetic resonance imaging parameters in carcinoma of the cervix.

BACKGROUND AND PURPOSE: The Eppendorf pO(2) histograph is the 'gold standard' method for measuring tumour oxygenation. The method is not suitable for widespread application because its use is limited to accessible tumours. A non-invasive imaging technique would be an attractive alternative. Therefore, the relationships between tumour oxygenation and dynamic contrast-enhanced magnetic resonance imaging (MRI) parameters were investigated. MATERIALS AND METHODS: The study comprised 30 patients with carcinoma of the cervix. Tumour oxygenation was measured pre-treatment as median pO(2) and the proportion of values less than 5 mmHg (HP5) using a pO(2) histograph. Repeat measurements were obtained for nine patients following 40-45 Gy external beam radiotherapy giving a total of 39 measurements. Dynamic contrast-enhanced MRI using gadolinium was performed prior to obtaining the oxygenation data. Time/signal intensity curves were generated to obtain two standard parameters: maximum enhancement over baseline (SI-I) and the rate of enhancement (SI-I/s). RESULTS: Using the 39 measurements, there was a significant correlation between SI-I and both median pO(2) (r=0.59; P<0.001) and HP5 (r=-0. 49; P=0.002). There was a weak, borderline significant correlation between SI-I/s and both median pO(2) (r=0.29; P=0.071) and HP5 (r=-0. 34; P=0.037). There was a significant relationship between tumour size and SI-I (r=0.54; P<0.001), but not SI-I/s. In 29 tumours, where data were available, there was no relationship between histological assessment of tumour angiogenesis (intra-tumour microvessel density; IMD) and either MRI parameter. CONCLUSIONS: Tumour oxygenation levels measured using a pO(2) histograph correlate with dynamic contrast-enhanced MRI parameters. Therefore, non-invasive dynamic MRI may be a method for measuring hypoxia in human tumours.

Relationship of MRI and clinical staging to outcome in invasive bladder cancer treated by radiotherapy.

AIM: To compare MRI and clinical staging of invasive bladder cancer prospectively and identify additional prognostic features on MRI before radiotherapy. METHODS AND MATERIALS: 143 patients with a pathological diagnosis of transitional cell carcinoma underwent MRI (1.0 T) of the abdomen and pelvis before radical radiotherapy. Tumour size, site, degree of infiltration, presence of adenopathy and hydronephrosis were assessed and an appropriate radiological stage assigned. Following radiotherapy all patients received regular cystoscopic follow-up. Date of first relapse and date of death were recorded. RESULTS: The median follow-up was 2.8 years for survivors. Those patients upstaged from T2a clinically to T3b on MRI had a significantly worse outcome (P = 0.0078). In univariate analysis a number of MRI features were significantly associated with adverse outcome: tumour size, circumferential tumour extent, and presence of hydronephrosis (all P < 0.05). After adjustment for clinical T stage and histological grade, all these MRI features and the MRI T stage were found to confer additional prognostic information in predicting early disease relapse and death (P < 0.05). CONCLUSION: This study demonstrates that MRI before radiotherapy provides valuable additional prognostic information compared to clinical staging.

Radiotherapy for carcinoma of the posterior pharyngeal wall.

Posterior pharyngeal carcinoma has an extremely poor prognosis regardless of the method of treatment. The purpose of this study was to assess the local control and survival in patients with carcinoma of the posterior pharyngeal wall treated with definitive radiotherapy and to determine prognostic factors which may be relevant to the current UICC staging classification. Between January 1991 and December 1995, 22 patients with a mean age of 60 years (range 44-82) received definitive radiotherapy, using a homogeneous technique, for carcinoma of the posterior pharyngeal wall. The median follow-up was 42 months (range 25-66). The overall 3-year survival and local control for the whole group was 50% and 73% respectively. Patients with early stage (T1 and T2) disease had a significantly better overall 3-year survival rate of 77% compared to 11% for patients with advanced stage (T3 and T4) disease (p=0.0010). Similarly, patients with early stage disease had a significantly improved 3-year local control rate compared to patients with more advanced stage disease (92% and 44% respectively, p=0.0080). Patients with node positive disease had an inferior survival rate of 29% compared to 60% for those with node negative disease though the difference did not reach statistical significance. In addition only one patient with initial node negative disease had isolated nodal relapse. There was no significant late morbidity. For patients with early stage disease we have obtained local control and survival rates comparable to other groups with a once daily, short fractionation radiotherapy scheme but with reduced morbidity. In late stage disease altered fractionation schemes should be considered in order to achieve better local control and survival. Isolated nodal relapse was not a significant problem in this cohort of patients. Outcome correlates with primary tumour size and this is reflected in the current UICC staging classification.

Magnetic resonance appearance of normal inguinal nodes.

AIM: To identify adult inguinal lymph node anatomical subgroups using magnetic resonance imaging (MRI), to derive a normal range for nodal number and size and to describe their morphology. MATERIALS AND METHODS: Eighty-three oncology patients with low stage pelvic tumours had inguinal lymph node assessment by MRI. Nodes were divided into proximal superficial (PS), distal superficial (DS) and deep inguinal (DI) subgroups, their number counted in two planes, and their transaxial short axis diameter recorded. Consistency of the largest node was recorded for each anatomical subgroup and two vertical distances measured, between the skin surface and the ipsilateral pubis, and between the skin surface and the deepest node. RESULTS: Transaxial plane maximum nodal number at the three sites was: PS 5, DS 8, DI 3; and in the coronal plane: PS 7, DS 5, DI 3. Nodal size ranges were: PS 3-10 mm (mean 4 mm), DS 3-15 mm (mean 6 mm) and DI3-10 mm (mean 6 mm). There was no correlation between nodal size or number and age or gender. Nodes were usually uniformly solid (PS 44%; DS 37%, DI 45%), or fatty (PS 39%; DS 33%; DI 25%). The range of distances between the skin and deepest lymph node was 2.5-16 cm depending on patient fatness. CONCLUSION: The mean number of nodes counted in the axial plane was six and in the coronal plane five. A maximum short axis diameter of 15 mm was recorded for inguinal lymph nodes.

Assessment and significance of mediastinal bulk in Hodgkin's disease: comparison between computed tomography and chest radiography.

PURPOSE: In Hodgkin's disease (HD), mediastinal bulk is currently defined from chest radiograph (CXR) measurements as a ratio of the maximum transverse mass diameter to the internal thoracic diameter at T5/6 level > or = 0.33. We evaluated how computed tomographic (CT) measurements of bulk correspond to those obtained from the CXR and correlated nodal mass long axis diameter with freedom from progression. METHODS: Ninety-five adult patients who had a CXR thoracic ratio of greater than 0.3 and a CT scan within 28 days of the CXR were included in the study, provided that both investigations were performed before the start of treatment. Measurements of the widest mediastinal diameter and internal thoracic diameter were made on both CXR and CT scan. The thoracic ratio (TR) was calculated for each modality and compared using paired t tests. The longest diameter of the largest individual nodal mass (LIM(CT)) was also measured from the CT and correlated with freedom from progression using Cox regression. RESULTS: There was excellent correlation between CT and CXR for measurement of TR, with TR(CT) greater than TR(CXR) (mean difference of 2%). A TR(CT) of 0. 35 was found to be equivalent to a TR(CXR) of 0.33. No single measurement of nodal size correlated with the current definition of bulk. However LIM(CT) greater than 10 cm did correlate with increased risk of progressive HD (P =.03), even after adjustment for other prognostic variables (chemotherapy regimen and Hasenclever Prognostic Index). CONCLUSION: Excellent correlation was observed between assessment of TR by CXR and CT scan. The longest diameter of the LIM(CT) greater than 10 cm was found to be associated with an increased risk of disease progression.