Purpose: To evaluate the diagnostic accuracy of different non-invasive cardiac imaging technologies in patients suspected of CAD and employ a novel technique of integrating multiple HTAs to inform policy decision-making.
Method: Evidence-based reviews of the literature from 2004 to 2009 were conducted to assess diagnostic accuracy using coronary angiography as the reference standard. Five diagnostic imaging technologies were identified: CMR, SPECT, CT Angiography, stress ECHO and stress contrast ECHO. Estimates of diagnostic accuracy were calculated to compare across the technologies and an expert panel was struck to assist in the contextualization of the evidence. Re-aggregation of results to a mega-HTA was based on the 4 OHTAC decision determinants: clinical effectiveness, value for money, social values and feasibility of implementation.
Result: Estimates of pooled sensitivity ranged from 0.98-0.84 and ranked highest to lowest as CT Angio, CMR, AC and traditional SPECT, gated SPECT and stress contrast ECHO and lastly, stress ECHO. For specificity, estimates ranged from 0.71-0.84 and ranked highest to lowest as stress ECHO, AC SPECT, CMR, stress contrast ECHO, CT Angio, gated and traditional SPECT. Sensitivity and specificity were equally weighted for policy-decision making, thus AUCs were calculated based on SROC curves and ranked as follows: CT Angio (0.94), cardiac MRI (0.93), AC SPECT (0.91; n=13), stress contrast ECHO (0.90), stress ECHO (0.89), gated SPECT (0.89) and lastly traditional SPECT (0.89).
Conclusion: For overall clinical benefit, CT Angio and CMR had the highest diagnostic accuracies. For implementation, stress ECHO (contrast) was deemed the most feasible. Social values were similar across technologies. Value for money is being assessed through an Ontario-based cost-effectiveness analysis. Mega-HTA results are currently being used to prioritize investments in cardiac imaging in Ontario.
See more of: The 32nd Annual Meeting of the Society for Medical Decision Making