MR spectroscopy may help avoid invasive procedures for recurrent brain lesions
A clinical decision model, which uses MR spectroscopy to help physicians differentiate between recurrent tumors and changes in the brain tissue due to radiation treatments, may help patients avoid invasive procedures and treatments, according to a study performed at the University of Michigan Health System in Ann Arbor, Mich.
The study, which appears in the February issue of the American Journal of Roentgenology, included 33 patients who had undergone radiation treatment for brain tumors and had MRI examinations that showed new lesions; either a recurrent tumor or radiation changes.
The patients then underwent MR spectroscopy and the ratios of three metabolites, choline (Cho), creatine (Cr) and n-acetylaspartate (NAA) were calculated. The researchers then created an ROC curve and prediction model, which determined the post-test probability of a patient having a recurrent tumor.
For the study, the final number of patients with recurrent tumors was 20 and those with post-radiation change were 13.
The authors found that an elevated Cho/NAA ratio correlated with evidence of tumor recurrence and allowed creation of a prediction rule to aid in lesion classification.
"MR spectroscopy looks inside each area of the brain and determines concentrations of metabolites that are there," said the study's lead author Ethan Smith, MD. "Choline, creatine and NAA are the three metabolites that are most commonly looked at. They are found inside the cells of the brain and the ratios between each change depending on if the cells are malignant or benign."
"With further work, this technique could save some patients from unnecessary brain biopsies," he noted.
The study, which appears in the February issue of the American Journal of Roentgenology, included 33 patients who had undergone radiation treatment for brain tumors and had MRI examinations that showed new lesions; either a recurrent tumor or radiation changes.
The patients then underwent MR spectroscopy and the ratios of three metabolites, choline (Cho), creatine (Cr) and n-acetylaspartate (NAA) were calculated. The researchers then created an ROC curve and prediction model, which determined the post-test probability of a patient having a recurrent tumor.
For the study, the final number of patients with recurrent tumors was 20 and those with post-radiation change were 13.
The authors found that an elevated Cho/NAA ratio correlated with evidence of tumor recurrence and allowed creation of a prediction rule to aid in lesion classification.
"MR spectroscopy looks inside each area of the brain and determines concentrations of metabolites that are there," said the study's lead author Ethan Smith, MD. "Choline, creatine and NAA are the three metabolites that are most commonly looked at. They are found inside the cells of the brain and the ratios between each change depending on if the cells are malignant or benign."
"With further work, this technique could save some patients from unnecessary brain biopsies," he noted.