High Rate of MRI Found in Study of Peripheral Neuropathy Diagnosis
But analysis did not examine whether imaging was appropriate or not.
ISSUE: APRIL 2012 by Rosemary Frei, MSc
With more than one dozen diagnostic tools, identifying peripheral neuropathy (PN) can be complex and costly. That is why investigators at the University of Michigan attempted to define diagnostic practice patterns in order to identify opportunities to improve efficiency of PN care.
The researchers identified 1,031 individuals diagnosed with PN between 1996 and 2007 through the Health and Retirement Study and the linked Medicare Standard Analytical Files (Ann Intern Med 2011;172:127-132). They included patients who were at least 65 years old in 1996, and created a matched comparison group. Among the subjects, 41.5% had diabetes and 44.4% of these had diabetic neuropathy; 80% of the nondiabetic individuals had idiopathic neuropathy.
The researchers focused on 15 relevant tests for PN to determine the number and patterns of tests used six months before and after the incident neuropathy diagnosis. After assessing 15 PN diagnosis tests, the investigators found that four were performed most often, on average, but testing patterns were highly variable, with more than 400 different patterns identified. The most common testing pattern was used only in 4.8% of patients.
About one-fourth—23.2%—had at least one magnetic resonance imaging (MRI) of the brain or spine, whereas a glucose tolerance test was rarely obtained (1%). Almost one-fifth (19.8%) of patients with neuropathy received electromyography. A complete blood cell count was ordered in 73.1% of patients, thyrotropin level in 55.2%, comprehensive metabolic panel in 53.2%, erythrocyte sedimentation rate in 28.7% and an antinuclear antibody test in 11.2%.
Additionally, a fasting glucose level test was ordered in 23.4% of patients with neuropathy. A hemoglobin A1c level was ordered in 43.2% of those with neuropathy and 17.1% of nondiabetic patients. Vitamin B12 levels were ordered in 32.6% of patients with neuropathy and in 40.6% of nondiabetic patients.
The large number of tests translated into high levels of Medicare expenses during the diagnostic period, at $14,362 per patient per year.
“High MRI use is probably for many reasons including physicians not being confident that someone has PN, the fact that no cause is identified for many cases with neuropathy, which pushes physicians to order more tests and because patients often prefer more testing, especially MRIs,” said Brian Callaghan, MD, assistant professor of neurology at University of Michigan Medical School, Ann Arbor.
The team concluded that more research is needed on the optimal approach for diagnosing PN. “First, I think we need to firmly establish what the best tests are for the evaluation of this condition. Next, we need to increase awareness among physicians including internists and neurologists that see this common condition,” said Dr. Callaghan.
Despite the high use of MRI, the evidence indicates two-hour oral glucose tolerance, fasting glucose, vitamin B12 levels and serum protein electrophoresis provide the highest true-positive rate and the greatest potential for guiding subsequent interventions (Neurology 2009;72:185-192).
Vera Bril, MD, suggested that the high use of MRI likely is linked to testing for comorbid conditions, rather than for PN itself. “You need to know why the physicians ordered the MRIs; it’s a leap to assume it’s inappropriate,” said Dr. Bril, professor of neurology at the University of Toronto and head of neurology at the University Health Network and Mount Sinai Hospital, Toronto, Canada.
“The patients in my clinic of this age often have spinal degenerative disease in addition to PN. I don’t do MRIs in my straightforward PN patients, but if they have spinal involvement. I’ll look further at it, including possibly with MRI.”
http://www.painmedicinenews.com/ViewArticle.aspx?d=Clinical%2bPain%2bMedicine&d_id=82&i=April+2012&i_id=826&a_id=20593
The researchers identified 1,031 individuals diagnosed with PN between 1996 and 2007 through the Health and Retirement Study and the linked Medicare Standard Analytical Files (Ann Intern Med 2011;172:127-132). They included patients who were at least 65 years old in 1996, and created a matched comparison group. Among the subjects, 41.5% had diabetes and 44.4% of these had diabetic neuropathy; 80% of the nondiabetic individuals had idiopathic neuropathy.
The researchers focused on 15 relevant tests for PN to determine the number and patterns of tests used six months before and after the incident neuropathy diagnosis. After assessing 15 PN diagnosis tests, the investigators found that four were performed most often, on average, but testing patterns were highly variable, with more than 400 different patterns identified. The most common testing pattern was used only in 4.8% of patients.
About one-fourth—23.2%—had at least one magnetic resonance imaging (MRI) of the brain or spine, whereas a glucose tolerance test was rarely obtained (1%). Almost one-fifth (19.8%) of patients with neuropathy received electromyography. A complete blood cell count was ordered in 73.1% of patients, thyrotropin level in 55.2%, comprehensive metabolic panel in 53.2%, erythrocyte sedimentation rate in 28.7% and an antinuclear antibody test in 11.2%.
Additionally, a fasting glucose level test was ordered in 23.4% of patients with neuropathy. A hemoglobin A1c level was ordered in 43.2% of those with neuropathy and 17.1% of nondiabetic patients. Vitamin B12 levels were ordered in 32.6% of patients with neuropathy and in 40.6% of nondiabetic patients.
The large number of tests translated into high levels of Medicare expenses during the diagnostic period, at $14,362 per patient per year.
“High MRI use is probably for many reasons including physicians not being confident that someone has PN, the fact that no cause is identified for many cases with neuropathy, which pushes physicians to order more tests and because patients often prefer more testing, especially MRIs,” said Brian Callaghan, MD, assistant professor of neurology at University of Michigan Medical School, Ann Arbor.
The team concluded that more research is needed on the optimal approach for diagnosing PN. “First, I think we need to firmly establish what the best tests are for the evaluation of this condition. Next, we need to increase awareness among physicians including internists and neurologists that see this common condition,” said Dr. Callaghan.
Despite the high use of MRI, the evidence indicates two-hour oral glucose tolerance, fasting glucose, vitamin B12 levels and serum protein electrophoresis provide the highest true-positive rate and the greatest potential for guiding subsequent interventions (Neurology 2009;72:185-192).
Vera Bril, MD, suggested that the high use of MRI likely is linked to testing for comorbid conditions, rather than for PN itself. “You need to know why the physicians ordered the MRIs; it’s a leap to assume it’s inappropriate,” said Dr. Bril, professor of neurology at the University of Toronto and head of neurology at the University Health Network and Mount Sinai Hospital, Toronto, Canada.
“The patients in my clinic of this age often have spinal degenerative disease in addition to PN. I don’t do MRIs in my straightforward PN patients, but if they have spinal involvement. I’ll look further at it, including possibly with MRI.”
http://www.painmedicinenews.com/ViewArticle.aspx?d=Clinical%2bPain%2bMedicine&d_id=82&i=April+2012&i_id=826&a_id=20593
No comments:
Post a Comment
All comments welcome but advertising your own service or product will unfortunately result in your comment not being published.