An Attempt to Characterize Malignant Lymphoma in Spleen, Liver and Lymph Nodes with Magnetic Resonance Imaging

An attempt was made to explore whether relaxation times and/or normalized image intensities obtained from magnetic resonance imaging (MRI) can separate malignant and non-malignant lymphomatous tissue. Spin-echo (SE) techniques with repetition times of 500 and 1500 ms and echo times of 35 and 70 ms were used for estimating T1 and T2. Estimation of T1 and T2 with such a low number of spin-echo sequences resulted in considerable variation in the data especially when T1 was long. Similar information was also extracted by normalizing the image intensities to the ‘proton density’ image (1500/35), and the spread of the data was then markedly reduced. Therefore, the method of normalizing was considered a more appropriate way of handling the image data when only a few sequences were available. No significant difference could be discerned in the MRI parameters between normal spleens and spleens infiltrated with malignant lymphoma, between normal livers and livers in patients with malignant lymphoma and between lymph nodes with low or high grade non-Hodgkin lymphoma. Lymphomatous tissue had similar MRI characteristics irrespectively of whether the cells were malignant or not, or located in spleens or in lymph nodes. The main biologic explanation for variation in data seems to be mostly the variable amounts of fibrosis, necrosis, oedema and/or iron content.

Reliability of splenic index to assess splenic involvement in pediatric Hodgkin's disease

The purpose of this study was to determine the reliability of the splenic index (SI) in children with Hodgkin's disease (HD). Seventeen patients who underwent staging laparotomy for HD were included in this study. Pretreatment computed tomography scans of these patients were reviewed retrospectively to determine the SI. The specificity, sensitivity, positive and negative predictive values, and accuracy of the SI were calculated. The sensitivity and specificity of the SI were 50% and 66%, respectively. The SI alone accurately identified or ruled out involvement with HD in 10 of 17 patients. Positive and negative predictive values of the SI were 57% and 60%, respectively. Even with the use of the SI, computed tomography alone remains unreliable to determine splenic involvement in children with HD. Additional imaging studies, especially fluorodeoxyglucose positron emission tomography, may improve the clinical staging of HD.

Diagnostic accuracy of ultrasound and computed tomography in the staging of Hodgkin's disease. Verification by laparotomy in 100 cases

BACKGROUND

A staging laparotomy still is considered the gold standard to detect occult abdominal involvement in Hodgkin's disease. Computed tomography and ultrasound are routinely available for diagnostic imaging. To the authors' knowledge, the exact contribution of ultrasound for the staging of Hodgkin's disease has not been reported in a large series of patients before this study.

METHODS

The diagnostic accuracy of abdominal ultrasound was compared with that of computed tomography and laparotomy in 100 patients with biopsy-proven Hodgkin's disease. Liver, spleen, paraaortic, and iliac lymph nodes were evaluated separately.

RESULTS

Seventeen patients had a higher disease stage after surgery (17%). Considering only patients without known abdominal disease (supradiaphragmatic involvement), 14/79 (18%) had a positive staging laparotomy. Ultrasound had superior sensitivity for detecting splenic involvement with Hodgkin's disease (sensitivity, 63% compared with 37% for computed tomography). The specificity of both methods for detecting splenic disease was identical (99% vs. 96%). Inhomogeneities of structure or small nodular infiltrates were detected preferentially by ultrasound. Hepatic involvement also was visualized better by ultrasound than by computed tomography. Lymph nodes at the splenic hilus were recognized by both methods with identical sensitivity (64% vs. 62%). Paraaortic and iliac lymph nodes were recognized with greater sensitivity by computed tomography than by ultrasound (sensitivity, 93% and 100% vs. 77% and 67%, respectively).

CONCLUSIONS

These results indicate that ultrasound and computed tomography each have their weaknesses and strengths and therefore should be combined, if possible. Ultrasound is the fastest and least invasive method and has particular accuracy for detecting splenic involvement, whereas computed tomography is more accurate in detecting involvement of paraaortic or iliac lymph node. If cost is important in the staging of Hodgkin's disease and if computed tomography is considered the standard, patients whose results are negative by computed tomography should be examined by ultrasound, focusing on splenic texture and size. Even in the era of combined modality treatment, surgical staging may be necessary to detect occult abdominal disease in a certain number of cases.

Comparison of initial lymphoma staging using computed tomography (CT) and magnetic resonance (MR) imaging

While MR is known to be superior to other imaging methods for detecting marrow involvement by lymphoma, MR is also capable of detecting abnormal lymph nodes. Our objective was to determine whether MR employing short TI inversion recovery (STIR) was comparable to CT in the initial staging of 23 patients with Hodgkin's disease (12 patients) and non-Hodgkin's lymphoma (11 patients). MR images of chest, abdomen, pelvis, and femoral marrow were obtained using the STIR and T1-weighted spin-echo (T1-SE) techniques, employing a protocol initially designed for marrow assessment. In all cases, CT-detected adenopathy was also found by MR. Four patients had marrow involvement by MR, undetected by CT. We conclude that MR and CT may be equivalent imaging modalities in the detection of nodal involvement, and that MR has an advantage in its ability to diagnose marrow involvement. Given the high frequency of focal marrow abnormalities detected by MR in patients with Hodgkin's disease and high-grade non-Hodgkin's lymphoma, MR may be the preferred staging modality for these patients.

Tumour imaging of indium-111 oxine-labelled autologous lymphocytes as a staging method in Hodgkin's disease

Following indium-111 oxine-labelled autologous lymphocyte infusion, 39 lymphocyte scintigraphies were performed in 35 patients with Hodgkin's disease (HD). Lymphocytes were obtained after leukapheresis and Lymphoprep gradient centrifugation and were further purified by an adherence step to eliminate monocytes. A median of 1.2 (0.2-3.7) X 10(9) cells were labelled with 6.7 (range 1.9-16.6) Mbq indium-111 oxine and reinfused to the patients. Gamma camera imaging was performed with a GE 400 AT. An accumulation of radioactivity was seen at 54 of 61 sites with enlarged lymph nodes. Increased radioactivity was also seen at 16 sites with no previous clinical evidence of tumour involvement. These data seem very promising and with further methodological improvement lymphoscintigraphy may prove to be an important complement in the staging procedure and follow-up of patients with HD.