Pathology and Medical Imaging MRI

Pathology and Medical Imaging

MRI Features of Larger Lesions Predict Malignancy in Cirrhotic Liver

The aim of the study by Jeong-Sik Yu et al. (2007) was to establish whether the presence of malignant liver cirrhosis could be predicted by studying the lesions present in the patient through MRI imaging. During the course of hepatocarcinogenesis, premalignant lesions can undergo fatty change preceding hepatocellular carcinoma, and the research studied these changes to establish how this may present in a patient with malignant rather than benign lesions; the predictive value of MRI for diagnosing malignancy was established.

Method total of 38 patients were selected to take part in the study. All patients were selected on the basis of having liver cirrhosis, and also focal lesions containing fat which were confirmed through chemical shift gradient-echo MRI. The patients were divided into two groups, with group a patients having four lesions or less, and group B. patients having more than 10. Positive predictive values (PPVs) for benignity and malignancy were based on lesion size, T1-weighted hypointensity, T2-weighted hyperintensity and arterial hypervascularity on the initial MRI.

Results

In group a patients, 47% of lesions became malignant, whereas in group B, all lesions were benign. It was found that there was a mean diameter of 18.8mm in malignant lesions, compared to 10.5 in benign lesions. The PPV of lesions over 15mm for malignancy was 85%, with 55% of immediately diagnosed hepatocellular carcinomas being entirely hypotense on the unenhanced in-phase T1-weighted images. The PPV for predicting malignancy in group a patients was 100% for T2-weighted hyperintensity and arterial hypervascularity.

Conclusion

The findings of the study indicate that the project was successful in the original aim of establishing a method for detecting malignancy through MR imaging. The results show that for patients who present with lesions over 15mm, with T1-weighted hypointensity on in-phase images, the findings are indicative of malignancy of the fat-containing nodules. The findings also suggest that the presence of many smaller lesions is indicative of benign disease.

PET/CT, Whole-body MRI Each Have Merits in Metastatic Breast Disease

Introduction

The study aimed to establish whether FDG PET/CT scanning would be more effective in detection of metastatic disease in patients with breast cancer than MR imaging. It has been noted that MRI scanning may miss malignant lymph node tumors which are not pathologically enlarged, but the study aimed to show that PET/CT would make spotting such tumors easier. PET/CT scanning should also make it clear when swelling in lymph nodes is due to inflammation as opposed to tumour (Tilke, 2007).

Method

30 patients were selected for the study based on having breast cancer and elevated tumor markers. Each patient underwent FDG-PET/CT scanning and MRI scanning. The imaging methods consisted of coronal T1-weighted SE/STIR sequences of the entire body, contrast-enhanced T1-weighted SE sequences of the brain, abdomen and pelvis, and whole body MRI scanning at 1.5- or 3-tesla. Two radiologists read the scan images and noted the presence of recurrent tumor, lymph node involvement and other distant metastases.

Results

From the whole-body MRI images, 195 malignant foci were detected by the two radiologists. In comparison, PET/CT images showed 188. Both methods revealed two recurrent tumors. It was found that in addition to the differences in the numbers of lesions identified in the two images, PET/CT imaging was superior for elucidating lymph node metastases and distant lung and skin metastases; whole-body MRI scans were superior for liver, bone and brain metastases. Overall sensitivity for whole-body and PET/CT was 96% and 93% respectively, while specificity was 87% and 88% respectively.

Conclusions

The study showed that both methods are relevant for detecting metastases in breast cancer patients, but also showed that the method chosen should be appropriate to the type of tumor which may be suspected based upon the laboratory and clinical findings. While the study found that both 1.5- and 3-tesla MRI scanning were feasible for the purpose of finding metastases, the team discovered that there is still much work to be done on refining the methods used for 3-tesla scanning. This could lead to further improvements in the results yielded by this method.

Breast gamma imaging spots DCIS better than mammo, MR

Introduction

Breast-specific gamma imaging (BSGI) is a functional study which is unaffected by dense tissue, implants or architectural distortion in the breast. The aim of the study was to assess the effectiveness of this method in detecting ductal carcinoma in situ (DCIS) in comparison to mammography and MRI methods (Pal, 2007).

Methods

22 women were selected for the study based on the presence of DCIS which had been confirmed through biopsy. The study did not take into account family history, genetic predisposition or breast density. BGSI was conducted both before biopsy, to assess indeterminate findings, and after biopsy, to determine foci and extent of disease. BGSI was conducted after injection of 25-30 mCi of technetium 99m-sestamibi, and results were measured by radio-tracer uptake on a five point scale, with 1 being normal, and 5 being abnormal, with marked focal tracer uptake. Mammography was performed in craniocaudal and mediolateral positions, and gadolinium enhanced MRI was performed with a 1.5-tesla scanner.