Case 2: Contrast-enhanced magnetic resonance angiography
(MRA) for arterial stenosis disease
In this case, we use meta-analysis data from the study of Menke and
Larsen24 summarizing evidence about how well MRA
detects arterial stenosis. A total of 32 studies were included in the
analysis.
Figure 4 depicts the meta-analytic summary plots. It includes the
summary ROC curve, individual study estimate, and summary point estimate
of the “traditional” measures of performance of MRA in diagnosis of
arterial stenosis. As with figure 1, the interpretation of traditional
statistics in terms of test performance is difficult. For example, an
MRA test diagnosing arterial stenosis is 96% specific and 78%
sensitive. The combination of these values is difficult to interpret and
may lead to inappropriate assessment. In terms of MI however, the test
reduces diagnostic uncertainty by 49%, which indicates a clearly useful
test.
Figure 5, demonstrates the meta-analysis of MI, in which it is shown
that the information content of MRA in diagnosis of arterial stenosis is
0.53 (CI: 0.48, 0.57).
Figure 6 depicts the RMI for reported by each study as well as the pool
estimate, which is approximately 67% for pre-test probability of
disease 25%. That is, the MRA reduces uncertainty related to arterial
stenosis by 67%.