RSNA 2009 

Abstract Archives of the RSNA, 2009


SSE19-03

Potential Use of Normalized Subtraction of Serial Brain FDG PET Scans to Evaluate for Post-treatment Brain Changes

Scientific Papers

Presented on November 30, 2009
Presented as part of SSE19: Nuclear Medicine (Imaging the Brain: Structure and Function)

Participants

Hussein Rabie Farghaly MD, Abstract Co-Author: Nothing to Disclose
John Dombrowski, Abstract Co-Author: Nothing to Disclose
Anis Sami Nassif MD, Abstract Co-Author: Nothing to Disclose
Aaron S. Nelson MD, Abstract Co-Author: Owner, MIMvista Corp, Cleveland, OH Employee, MIMvista Corp, Cleveland, OH
Jon W Piper BEng, Abstract Co-Author: Employee, MIMvista Corp Stockholder, MIMvista Corp
Nghi Co Nguyen MD, PhD, Presenter: Nothing to Disclose

PURPOSE

Normalized subtraction (NS) can be used to identify differences in image volumes of the same modality. We aimed to explore whether NS may be helpful in characterization of post-treatment brain changes.  

METHOD AND MATERIALS

Nine cancer patients (pts) with history of brain metastases underwent at least two FDG PET/CT scans for staging/restaging that were done either before and/or after local treatment (Tx). Brain PET images of two different time points each were co-registered with each other and with contrast enhanced MRI images. NS was then applied for the two PET data sets using a 15% threshold level (MIMvista, Corp.) that in our experience yields the best trade-off between sensitivity and specificity.  

RESULTS

All 9 pts had radiotherapy (RT) to the brain: 7 had stereotactic radiosurgery, 2 had surgical resection and RT. A total of 29 PET scans were evaluated, 3 performed prior, 13 in early (1-6 mo) and 13 in late (7-20 mo) follow-up. A minimum of 2 PET scans were used in each pt, but 4 pts had 4 or more scans for review. When comparing pre-tx to post-tx scans, the FDG uptake in metastatic lesions decreased rapidly and showed no further decrease at 3 months of successful Tx. In one patient, however, there was FDG increase at previous metastatic site at 8 months after radiation which later correlated with worsening findings on MRI, suggestive of tumor recurrence. In patients with marked tumor associated vasogenic edema, an FDG increase within the edema region and adjacent cortex was observed for 6-10 months after completion of Tx. The corresponding FDG increase was shorter and last only for about 3 months if the vasogenic edema was less severe.   

CONCLUSION

Normalized subtraction of serial FDG PET scans can be a promising method to evaluate for post-treatment changes of the brain and tumor recurrence. It appears that lesions with marked associated vasogenic edema may result in delayed metabolic recovery of normal brain tissue compared to those with less vasogenic edema.

CLINICAL RELEVANCE/APPLICATION

 Normalized subtraction of serial FDG PET scans can be a promising method to evaluate for post-treatment brain changes

Cite This Abstract

Farghaly, H, Dombrowski, J, Nassif, A, Nelson, A, Piper, J, Nguyen, N, Potential Use of Normalized Subtraction of Serial Brain FDG PET Scans to Evaluate for Post-treatment Brain Changes.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8016360.html