Patients with stroke often require extensive inpatient and outpatient care that is associated with high expenditures and very high morbidity and mortality rates. Neuroimaging is an important component of care in patients with acute ischemic stroke, especially for guiding the use of thrombolysis. Imaging is associated with high costs and is being used more often today. It is therefore important to have high-quality economic evaluations of medical imaging technology for patients with stroke. If done appropriately, this could be an important tool to generate accurate estimates of the impact of current neuroimaging technologies on patient outcomes and costs.
Kirsteen R. Burton, MBA, MSc, MD, from the University of Toronto Department of Medical Imaging and Institute of Health Policy, Management, and Evaluation in Canada, and colleagues assessed the quality of economic evaluations of imaging for patients with acute ischemic stroke using a systematic review of the available economic evaluations (Burton KR, et al. Stroke. 2014;45:807-814). The annual costs of stroke were recently calculated to be $2.7 billion (Canadian dollars).
The results of the study showed that economic evaluations of imaging modalities after acute ischemic stroke were generally of high methodological quality (with scores ranging from 7.19%-93.5%); however, all the analyses included in this study were missing important clinical components specific to radiology, including appropriate imaging comparators, as well as data on complications, recurrent stroke, and intracerebral hemorrhage.
The investigators performed a comprehensive search of relevant electronic databases using the following terms: stroke, cost, cost-benefit analysis, and imaging. Of the 568 potential studies identified, only 5 were appropriate for inclusion in this study. All 5 studies used noncontrast computed tomography (NCCT) as the base-case imaging modality to select patients for intravenous thrombolysis.
The 5 studies used different comparators, including 3 studies of computed tomography perfusion (CTP) and 2 studies of magnetic resonance imaging (MRI); however, none of the studies included all the possible comparators associated with NCCT.
Dr Burton and colleagues concluded that this is an important flaw and stated that studies should compare NCCT modality with each of the currently available imaging modalities, as well as with plausible combinations of neuroimaging modalities. Another flaw they identified was the omission of outcomes data with CTP, a relatively new technology.
“It is apparent that these models are hampered by these missing data, and that CTP-specific outcome data are needed to improve these models,” the investigators wrote.
They also pointed out that the studies that included computed tomography (CT) or MRI comparators did not account for the presence of contrast-induced nephropathy and nephrogenic systemic fibrosis. “Although uncommon, these are clinically important complications and should be included in any models that assess CT…or MRI,” Dr Burton and colleagues noted.
Another missing important type of data was MRI claustrophobia that is causing patients to bail out of MRI, which occurs in an estimated 1% to 30% of patients, and is a potentially important component related to the assessment of cost-effectiveness of MRI.
Of the 5 studies evaluated, 4 did not explicitly publish incremental cost-effectiveness ratios but provided enough data to calculate them. Even though patients diagnosed with acute ischemic stroke have approximately a 14% risk for recurrent stroke, only 1 of the 5 studies looked at the rate of stroke recurrence. In addition, data on intracerebral hemorrhage, another potentially significant cost driver, were missing in 2 of the studies.
“Future economic evaluations of stroke imaging should compare all reasonable neuroimaging modalities, incorporate imaging modality sensitivities/specificities, include CTP-specific outcome data, and should integrate the probabilities of hemorrhagic transformation and recurrent stroke after treatment,” the investigators concluded.