Meta-analysis: Accuracy of contrast-enhanced magnetic resonance angiography for assessing steno-occlusions in peripheral arterial disease.

Abstract

BACKGROUND: Contrast-enhanced magnetic resonance angiography (MRA) is a noninvasive, radiation-free imaging method for studying peripheral arterial disease (PAD) of the lower extremities. PURPOSE: To summarize evidence of prospective studies about how well MRA identifies or excludes arterial steno-occlusions (50% to 100% lumen reduction) in adults with PAD symptoms. DATA SOURCES: PubMed and 3 other databases were searched from 1998 to 2009 without language restrictions. STUDY SELECTION: Two independent reviewers selected 32 studies that compared MRA with intra-arterial digital subtraction angiography in PAD. Eligible studies were prospective and provided data to reconstruct 2 x 2 or 3 x 3 contingency tables (<50% stenosis vs. > or =50% stenosis or occlusion of arterial segments) in at least 10 patients with PAD symptoms. DATA EXTRACTION: Two reviewers independently assessed the study quality and extracted the study data, with disagreements resolved by consensus. DATA SYNTHESIS: The 32 included studies generally had high methodological quality. About 26% of the 1022 included patients had critical limb ischemia with pain at rest or tissue loss. Overall, the pooled sensitivity of MRA was 94.7% (95% CI, 92.1% to 96.4%) and the specificity was 95.6% (CI, 94.0% to 96.8%) for diagnosing segmental steno-occlusions. The pooled positive and negative likelihood ratios were 21.56 (CI, 15.70 to 29.69) and 0.056 (CI, 0.037 to 0.083), respectively. Magnetic resonance angiography correctly classified 95.3%, overstaged 3.1%, and understaged 1.6% of arterial segments. LIMITATION: Similar to most studies of computed tomographic angiography in PAD, the primary studies reported the diagnostic accuracy of MRA on a per-segment basis, not a per-patient basis. CONCLUSION: This meta-analysis of 32 prospective studies further increases the evidence that contrast-enhanced MRA has high accuracy for identifying or excluding clinically relevant arterial steno-occlusions in adults with PAD symptoms. PRIMARY FUNDING SOURCE: None.