Article type
Year
Abstract
ABSTRACT Background: Methods of detecting coronary artery disease (CAD) have become increasingly important in informing therapies for people with angina or following myocardial infarction (MI). Single photon emission computed tomography (SPECT) myocardial perfusion scintigraphy (MPS) allows comparison of tomographic images of the myocardium following stress and at rest to assess the presence, location and extent of CAD.
Objectives: To assess the effectiveness of SPECT MPS for the management of people with angina and following myocardial infarction (MI).
Methods: We sought prognostic studies involving SPECT, stress electrocardiography (sECG) and/or coronary angiography (CA) in people with suspected or known CAD. Outcomes included cardiac mortality, non-fatal MI and revascularisation. We searched the following databases: MEDLINE (1966 to October 2002), EMBASE (1980 to week 44 2002), Science Citation Index (1981 to December 2002), the Cochrane Central Register of Controlled Trials (The Cochrane Library issue 3, 2002), Health Management Information Consortium (1979 to 2002), Health Technology Assessment Database (October 2002), PREMEDLINE (5 November 2002) and BIOSIS (1985 to December 2002). Non-English language studies were noted but excluded from the review, as were studies with fewer than 100 participants. Two reviewers independently extracted data and assessed the quality of the included studies. Results: Twenty-one observational studies involving over 53,000 people reported the general prognostic value of SPECT. In multivariate analysis, SPECT variables provided independent and incremental prognostic information to clinical, sECG and angiographic variables in predicting future cardiac events. The variety of predicting variables and measures used to report them, however, prevented a straightforward comparison between the different studies. Three studies comparing a SPECT-CA strategy with direct CA reported a lower subsequent revascularisation rate for SPECT-CA (6-21%) than for direct CA (16-44%). Two of the comparative studies reported that a greater percentage of people referred to the direct CA strategy had normal angiograms compared with the SPECT-CA strategy (33% versus 18%; 43% versus 33%). Four observational studies involving over 2000 people reported the prognostic value of SPECT for people with previous MI. In multivariate analysis including clinical history, sECG, SPECT and angiographic variables, SPECT variables provided independent and incremental prognostic performance in predicting future cardiac events. Conclusions: SPECT provides independent and incremental prognostic information to that from sECG and/or CA that helps to risk-stratify people with CAD and influence the way in which their condition is managed. Increasing the use of strategies involving SPECT may reduce the number of CA tests required, as SPECT may be able to better identify those at lower-risk for whom CA might be avoided. All of the prognostic studies, however, were observational and susceptible to bias by unknown confounding factors. Acknowledgements: The Health Services Research Unit is supported by a core grant from the Chief Scientist Office of the Scottish Executive Health Department; it also receives grant support from the National Institute for Clinical Excellence (NICE). The work described formed the basis of a broader systematic review commissioned by the NHS R&D HTA Programme on behalf of NICE. The views expressed are those of the authors and not necessarily shared by the funders.
Objectives: To assess the effectiveness of SPECT MPS for the management of people with angina and following myocardial infarction (MI).
Methods: We sought prognostic studies involving SPECT, stress electrocardiography (sECG) and/or coronary angiography (CA) in people with suspected or known CAD. Outcomes included cardiac mortality, non-fatal MI and revascularisation. We searched the following databases: MEDLINE (1966 to October 2002), EMBASE (1980 to week 44 2002), Science Citation Index (1981 to December 2002), the Cochrane Central Register of Controlled Trials (The Cochrane Library issue 3, 2002), Health Management Information Consortium (1979 to 2002), Health Technology Assessment Database (October 2002), PREMEDLINE (5 November 2002) and BIOSIS (1985 to December 2002). Non-English language studies were noted but excluded from the review, as were studies with fewer than 100 participants. Two reviewers independently extracted data and assessed the quality of the included studies. Results: Twenty-one observational studies involving over 53,000 people reported the general prognostic value of SPECT. In multivariate analysis, SPECT variables provided independent and incremental prognostic information to clinical, sECG and angiographic variables in predicting future cardiac events. The variety of predicting variables and measures used to report them, however, prevented a straightforward comparison between the different studies. Three studies comparing a SPECT-CA strategy with direct CA reported a lower subsequent revascularisation rate for SPECT-CA (6-21%) than for direct CA (16-44%). Two of the comparative studies reported that a greater percentage of people referred to the direct CA strategy had normal angiograms compared with the SPECT-CA strategy (33% versus 18%; 43% versus 33%). Four observational studies involving over 2000 people reported the prognostic value of SPECT for people with previous MI. In multivariate analysis including clinical history, sECG, SPECT and angiographic variables, SPECT variables provided independent and incremental prognostic performance in predicting future cardiac events. Conclusions: SPECT provides independent and incremental prognostic information to that from sECG and/or CA that helps to risk-stratify people with CAD and influence the way in which their condition is managed. Increasing the use of strategies involving SPECT may reduce the number of CA tests required, as SPECT may be able to better identify those at lower-risk for whom CA might be avoided. All of the prognostic studies, however, were observational and susceptible to bias by unknown confounding factors. Acknowledgements: The Health Services Research Unit is supported by a core grant from the Chief Scientist Office of the Scottish Executive Health Department; it also receives grant support from the National Institute for Clinical Excellence (NICE). The work described formed the basis of a broader systematic review commissioned by the NHS R&D HTA Programme on behalf of NICE. The views expressed are those of the authors and not necessarily shared by the funders.