Page 24 - Cardiac testing for coronary artery disease in potential kidney transplant recipients
P. 24

Summary of results  (Continued)


                 characteristics (e.g. prevalence of chest pain) and test application (diagnostic test threshold, criteria for positive test, choice of stress
                 agent and stress protocol, and operator variability) likely contributed to heterogeneity, we were hindered from estimating their
                 contributions because of relatively sparse data, which resulted in low power
                 Partial verification, where not all patients who received screening tests also received coronary angiography, occurred in 5/25 compar-
                 isons. This may have affected estimates of sensitivity and specificity
                 Two different reference standard thresholds (≥ 70% stenosis or ≥ 50% stenosis) were used in the included studies, with most studies
                 only using one reference standard threshold or the other. An overall analysis pooling the results of all studies regardless of threshold
                 may introduce additional heterogeneity due to a threshold effect

                 Results

                 Test                           DSE              MPS
                 Number of studies [all studies]  13             9

                 Number of participants [all studies]  745       582

                 Pooled sensitivity (95% CI) [all studies]  0.79 (0.67 to 0.88)  0.74 (0.54 to 0.87)

                 Pooled specificity (95% CI) [all studies]  0.89 (0.81 to 0.94)  0.70 (0.51 to 0.84)

                 Number of studies [≥ 70% stenosis]  9           7

                 Number of participants [≥ 70% stenosis]  668    517

                 Pooled sensitivity (95% CI) [≥ 70% steno- 0.76 (0.60 to 0.87)  0.67 (0.48 to 0.82)
                 sis]

                 Pooled specificity (95% CI) [≥ 70% steno- 0.88 (0.78 to 0.94)  0.77 (0.61 to 0.88)
                 sis]

                 Number of false diagnoses of ≤ 70% coro- 24 (13 to 40)  33 (18 to 52)
                 nary artery stenosis in a standard popula- per 100  per 100
                 tion of 100 patients (false negative rate)

                 Number of false diagnoses of ≥ 70% coro- 12 (6 to 22)  23 (12 to 39)
                 nary artery stenosis in a standard popula- per 100  per 100
                 tion of 1000 patients (false positive rate)

                 Positive likelihood ratio [≥ 70% stenosis] 6.44 (3.03 to 13.70)  2.89 (1.39 to 5.99)
                 (95% CI)

                 Negative likelihood ratio [≥ 70% stenosis] 0.26 (0.13 to 0.50)  0.43 (0.23 to 0.80)
                 (95% CI)
                 Post test probability after positive screening 42% to 72%  24% to 54%
                 test result for a patient with low risk (10%
                 to 29% pre test probability) disease



                Cardiac testing for coronary artery disease in potential kidney transplant recipients (Review)  22
                Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
   19   20   21   22   23   24   25   26   27   28   29