Neil Thakur, MD

Abstract

Lumbar spine (LS) MRI overuse may be identified in administrative data, but these data may lack the detailed clinical information needed to correctly assess overuse. The aim of this study was to compare chart review with analysis of administrative data to determine the appropriateness of LS MRI.The sensitivity and specificity of the administrative method were determined, with inappropriateness regarded as the positive result, as if chart review determined the true state. Patients were the first 146 veterans who underwent LS MRI in the outpatient setting in fiscal year 2012 at the Veterans Affairs Palo Alto Health Care System. The InterQual criteria for chart review and the method of evaluating administrative data developed by CMS and endorsed by the National Quality Forum were used. Slight modifications were made to each measure to ensure completeness and comparability.Of the 146 scans reviewed, 23% were considered inappropriate by the administrative measure, whereas 59% were considered inappropriate by chart review. Compared with chart review, the administrative measure had specificity of 82% for identifying inappropriate scans and sensitivity of 27% for identifying appropriate scans.Compared with chart review, analysis of administrative data identified scans that were appropriate but underestimated inappropriate ordering. Contrary to expectations, chart review resulted in more scans being classified as inappropriate. The administrative method is economically feasible for identifying the overuse of LS MRI, but it underestimates the true extent of inappropriate ordering.

View details for DOI 10.1016/j.jacr.2016.04.013

View details for PubMedID 27344246

Abstract

The purpose of this study is to describe the findings and patterns of injury on magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) after whole-body hypothermia treatment for neonatal hypoxic ischemic encephalopathy.A retrospective review of consecutive term neonates treated with whole-body hypothermia was performed. Data recorded included demographics and MRI and MRS findings, and day of life (DOL) studies were performed. Injury patterns were classified on MRI as deep, cortical, mixed, or diffuse. The relative apparent diffusion coefficient (rADC) was plotted against DOL scanned and the presence of lactate was recorded.MRI was performed in 44 infants, 34 of whom also underwent MRS. MRI was abnormal in 32% of neonates, 29.5% of whom were imaged at DOL 4 to 8. rADC values were lowest in neonates scanned on DOL 4 and 5 and remained low up to DOL 8. The deep brain nuclei were involved in hypoxic ischemic encephalopathy in 93% of neonates with abnormal MRIs and lactate was identified on MRS in 18% of neonates between DOL 4 and 8.MRI performed after therapeutic cooling was abnormal in 29.5% of neonates scanned on DOL 4 to 8. Deep nuclear injury was identified in 93% of neonates. Lactate was present on MRS in 18% of neonates, and rADC values were most reduced on MRI between DOL 4 and 8.

View details for DOI 10.1097/SMJ.0b013e3182967d38

View details for Web of Science ID 000330323900004

View details for PubMedID 23736175

Abstract

Isolated borderline low conus medullaris is a frequent finding on screening lumbar sonography of unknown significance that often prompts further imaging and clinical follow-up.To determine the clinical outcome and utility of follow-up neuroimaging in infants with isolated borderline low conus on lumbar sonography.We reviewed 748 consecutive spinal sonograms identifying infants with conus terminating between L2-L3 disc space and mid-L3 level without other findings of tethered cord. We excluded infants with conditions associated with developmental delay and those who passed away, and compared the age of gross motor milestone achievement to normal ranges. Follow-up imaging was reviewed.Isolated borderline low conus was found in 90 of 748 infants (12%) on sonography. Seventy of those infants met inclusion criteria. Follow-up imaging in 11 children (10 MRI, 1 sonogram), showed change in conus position to "normal" level in 10, no change in 1, and no new findings within lumbar spine. Clinical follow-up was available in 50 of 70 (71%) children meeting inclusion criteria, with normal motor milestones met in all 50 children.Isolated borderline low conus is a common finding in infants who meet normal developmental milestones suggesting that follow-up evaluation has little utility and is likely unwarranted.

View details for DOI 10.1007/s00247-010-1889-y

View details for Web of Science ID 000288758300011

View details for PubMedID 21079942