Correlation Between Mucoceles and CSF Leak (PI: Valerie Jewells)

Purpose:  Determine if there is a correlation between mucoceles and CSF leak and if specific types of trauma increase the risk of CSF leak.

Participants: Patients who have had prior CSF leak surgery and/or have demonstrated mucoceles on brain MRI.

Procedures (methods): Retrospective evaluation of patients in a data base who have had CSF leak repair surgery by Dr. Brent Senior (private data base) as well as those patients with the key words mucocele and frontal, sphenoid and/or ethmoid as key words in their report from a brain MRI will be assessed through the medical record and imaging to ascertain if there is a true correlation between mucoceles and CSF leak and if specific types of trauma will increase the risk for CSF leak and mucoceles.

Evaluation of Brain Metastases on PET/CT vs MRI (PI: Valerie Jewells)

Purpose: Determine the utility of PET CT for visualization of brain metastases compared to MRI

Participants: Individuals who present with Melanoma, Breast and Lung Cancer who have had staging PET/CT and brain MRI with contrast.

Procedures (methods): Using the Carolina Data Warehouse and ICD codes we will retrospectively identify patients who have Melanoma, Breast and Lung Cancer who have had staging PET/CT and brain MRI with contrast within 30 days of each other.  After identification, the imaging exams will be reviewed by a resident and staff radiologists in nuclear medicine and neuroradioogy and a consensus will be reached with regard to the identification of metastatic foci to brain for PET/CT and MRI to determine accuracy rates for PET/CT compared to the gold standard MRI with contrast. The large numbers of patients will be prohibitive to HIPPA and signed consent requirements and therefore, these will not be obtained.

Automated Assessment of Leptomeningeal Collaterals on CT Angiograms (PI: Yueh Lee)

Purpose: The purpose of this study is to improve the platform’s integration with clinical workflows, address the clinical opportunities from computing CT perfusion measures (the evolving standard in stroke assessment), and conduct a study of the individual and combined utility of CTA and CT perfusion for stroke patient management.

Participants: One hundred (100) subjects presenting at the participating institutions between January 1, 2010 and September 5, 2018 will be included in this 2 part retrospective study.  The first group of 50 patients will be included to further refine the platform.  The second group of 50 patients will be included to evaluate the additive value of the output in patient care.

Procedures (methods):  We will retrospectively review both groups of individuals.  The first group will implement the CT perfusion algorithm and validate the measurements against actual patient studies against clinically available perfusion post-processing systems (Siemens and iSchemiaView) and follow-up final infarct volume as determined by CT or MR imaging.  The second group will evaluate the utility of the collateral score and core volumes as independent and combined predictors of mRs at 90 days. We will also examine the number of potential changes in treatment by clinicians after being offered the additional collateral status information.

Air Locules in the External Auditory Canal on Trauma Head CTs: Predictive Value for Detection of Nondisplaced Temporal Bone Fracture (PI: Carlos Zamora)

Purpose: To determine the positive predictive value of fluid with air locules in the external auditory canal on trauma head CTs for detection of nondisplaced temporal bone fractures.

Participants: Adult patients with head trauma undergoing screening head CTs in the Emergency Department.

Procedures (methods): Retrospective evaluation based on our imaging and EPIC databases. Search query will be performed through our PACS system for head CTs performed using our trauma protocol. Clinical data will include: age at the time of head CT and gender. Imaging evaluation will include the presence of fluid with air locules in the external auditory canal as a primary test variable and fluid within the middle ear cavity and mastoid air cells as secondary variables. The presence of nondisplaced temporal bone fractures will be assessed on screening trauma head CTs and confirmed on subsequent temporal bone imaging where available.

Cranial Nerve Cisternal Segment Enhancement in Multiple Sclerosis: Prevalence on 3.0 Tesla Volumetric T1 MRI and Clinical Implications (PI: Carlos Zamora)

Purpose: We propose a retrospective study that would evaluate the presence and clinical significance of cranial nerve cisternal segment enhancement using a large cohort of multiple sclerosis patients who received brain MRI at UNC Hospitals.

Participants: Patients with confirmed MS diagnosis by McDonald’s criteria, and at least one 3T brain MRI at UNC Hospitals over the study period January 1, 2013 to December 31, 2017.

Procedures (methods): This will be a retrospective study to review patients with multiple sclerosis treated at UNC.  We will characterize imaging characteristics. We will correlate our findings with several clinical measures, which previously have been reported sparingly and for trigeminal involvement alone. Lastly, to provide a better measure of the implications of enhancing lesions and to characterize their natural history, we will study patients at multiple time points, a strategy that has scarcely been used.

Evaluating the Predictive Value of Intra-spinal Incidental Findings on Lumbar Spine MRI for Malignancy (PI: Benjamin Huang)

Purpose: The purpose of this study is to complete a retrospective review of about 3000 patients who had lumbar spine MRI and evaluate their follow-up images for the presence or absences of incidental findings. We want to determine if these incidental findings have any predictive value. In this study we will compare the percentage of people with a spinal incidental finding who are later diagnosed with a malignancy vs the percentage of people without any incidental findings who are later diagnosed with a malignancy.

Participants: Approximately 3000 UNC patients who received a lumbar spine MRI between 7/1/14 and 6/30/15. Only including patients who were at least 18 years old at the time of imaging

Procedures (methods): This is a retrospective study. We will obtain a list of patients via a data request through the Carolina Data Warehouse. Then, we will review the included patients’ MRI images will be evaluated by both a board certified attending radiologist and either a radiology resident or fellow for incidental findings within the lumber spine. Data will be collected from each patient regarding the presence of incidental findings, and if any are found, they will be classified as clearly benign, clearly malignant, or indeterminate. The number and size of any lesions present will be noted for further subanalysis. If any patient has received multiple lumbar spine MRI studies, we will use the first scan on which an incidental finding is found as the baseline, and use any future studies as follow up.

Utility of Computed Tomography (CT) of the Sinuses in Bone Marrow Transplant Patients (PI: Benjamin Huang)

Purpose: This study aims to evaluate the radiologic features and clinical utility of Sinus CT ordered for assessment of sinusitis bone marrow transplant patients.

Participants: Retrospective review of nonconsecutive patients with history of bone marrow transplant, treated at UNC Hospitals within the past 12 years.

Procedures (methods): Patients will be identified from the UNC bone marrow transplant unit registry and/or the Carolina Data Warehouse for health. Patients with sinus CT ordered for sinusitis, fever, immunocompromise, and other clinical features will be assessed.  De-identified CT data will be evaluated with a grading scale, and multivariate analyses performed. Cost-benefit analysis will be performed.

Measuring Brain Iron Levels in Patients with Multiple Sclerosis (PI: Valerie Jewells)

Purpose: Currently, there is no imaging bio-marker for Multiple Sclerosis (MS). Reasons for the lack of a bio-marker include, but are not limited to; poor correlation between clinical findings and probability for disease progression, poor correlation between MRI T1/T2 lesion loads and disease progression. It is well known that iron is deposited in the brain in MS patients and that it is correlative with disease progression. Therefore, pursuit of a bio-marker for disease progression is the purpose of this study.

Participants: Adult clinically Isolated Syndrome (CIS) and Relapsing Remitting MS (RRMS) patients.

Procedures (methods):Adult CIS and RRMS patients will be selected through EHR as patients who have been treated through the MS clinic in neurology at UNC. We will then review the 3T MRIs that have been performed with a new method of susceptibility weighted imaging (SWI) that quantifies iron deposition in the brain. This quantification assessment includes cortical and deep grey matter areas currently under-assessed with conventional MRI as well as white matter lesional and peri-lesional deposition. Following this, we will correlate the iron deposition levels to T1/T2 lesion loads and whole brain atrophy analysis in the same MRI as well as the clinical information in the EHR including Expanded Disability Status Score (EDSS) routinely performed at office visits, relapse rate, and the multiple sclerosis severity score (MSSS).