Patients with a transient ischemic attack (TIA) or minor stroke and suspected carotid artery stenosis (CAS) should undergo duplex ultrasound (US) and then CT angiography (CTA) after positive US results as the most effective and cost-effective imaging protocol, according to a study published in this month's Radiology. Immediate CTA and surgery for 50 to 99 percent stenosis is indicated for patients with a high-risk profile, a high CAS probability or those who can proceed to surgery without delay, added the authors.
CAS of 70 percent or more causes 10 to 30 percent of TIAs, and detection is critical to reducing the probability of a recurrent stroke, wrote Aletta T.R. Tholen of Erasmus University Medical Center in Rotterdam, the Netherlands, and colleagues. Although carotid endarterectomy reduces the risk of ipsilateral ischemic stroke, benefits from surgery decrease with delay between symptom onset and surgery, explained Tholen.
The researchers sought to assess the effectiveness and cost-effectiveness of state-of-the-art noninvasive imaging modalities including CTA alone, compared to duplex US and MR angiography using a decision model to evaluate and compare various diagnostic strategies. They also developed a Markov model with a one-year cycle length to extrapolate and evaluate long-term outcomes and subsequent treatment. Researchers paid particular attention to the time window between first symptoms and carotid endarterectomy and the cut-off value chosen as surgical indication (70 to 99 percent versus 50 to 99 percent stenosis).
Researchers included 194 male and 139 female patients who presented with TIA or minor stroke at Erasmus between November 2002 and January 2005 in the study. The average age was 62 for males and 61 for females, and age range was 19 to 90 years. All patients underwent duplex US and CTA of the carotid arteries. If either exam showed a stenosis of 50 to 99 percent the patient was referred for digital subtraction angiography.
Researchers reported the sensitivity and specificity of CTA at a cutoff point of 70 to 99 percent stenosis were 0.91 and 0.99, respectively, and 1.00 and 0.98, respectively, at a cut-off point of 50 to 99 percent stenosis.
Researchers assessed the diagnostic performances of US and contrast-enhanced MR angiography using eight meta-analysis that included 41 studies published between January 1987 and April 2004. The sensitivity and specificity of duplex US were 0.89 and 0.84, respectively, at a cutoff point of 70 to 99 percent stenosis and 0.84 and 0.83, respectively, at a cutoff point of 50 to 99 percent stenosis. MR angiography provided sensitivity and specificity of 0.94 and 0.93, respectively, at a cutoff point of 70-99 percent stenosis and 0.96 and 0.96 at a cutoff point of 50-99 percent stenosis.
Researchers also compared diagnostic strategies in terms of costs, effects (in quality-adjusted life-years [QALYs]), incremental cost-effectiveness ratios and net health benefits by using a willingness-to-pay of €50,000 ($66,018) per QALY.
“All strategies that used CT angiography or contrast-enhanced MR angiography, either as a solo strategy or in combination with an initial duplex US exam, demonstrated similar costs and effectiveness, presuming a two to four week delay in surgery,” wrote the authors. Assuming a €50,000 QALY, the duplex US/CTA combination strategy yielded the highest net health benefits in both men and women, continued the authors. Surgical timing affected net health benefits of treatment strategies.
They also found that the duplex US as a solo test was the least cost-effective strategy, which conflicts with earlier studies.
“The results demonstrated that the duplex US/CT angiography strategy with a cutoff of 70-99 percent stenosis is the most cost-effective strategy,” concluded the authors. An increase in the patient risk profile, higher disease probability or possibility of surgery without delay results in more lenient surgical criterion at 50-99 stenosis with solo CT as the preferred strategy.
“Implementing the results of this study should always be done with consideration of the individual patient and the local situation,” concluded the authors. Prior probability of CAS and the risk profile may affect the diagnostic strategy.