Meta-analysis of the diagnostic accuracy of computed tomography angiography compared with invasive coronary angiography in preoperative cardiac surgery planning: a focus on valve surgery patients

Figures

• 

  Figure 1. PRISMA 2020 flow diagram illustrating study selection process (n=3123 records identified, 28 studies included). CABG, coronary artery bypass grafting; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses; TAVI, transcatheter aortic valve implantation; TAVR, transcatheter aortic valve replacement.

• 

  Figure 2. SROC curve for diagnostic performance of CCTA compared with ICA in preoperative CAD assessment (AUC=0.97). AUC, area under the curve; CAD, coronary artery disease; CCTA, coronary CT angiography; FPR, false-positive rate; ICA, invasive coronary angiography; ROC, receiver operating characteristic; SROC, summary receiver operating characteristic; TPR, true-positive rate.

• 

  Figure 3. Forest plot of study-level diagnostic performance for CCTA versus ICA. (A) Sensitivity estimates with 95% CIs demonstrate consistently high diagnostic accuracy across most studies. (B) Specificity estimates with 95% CIs show greater variability across studies, particularly in TAVI cohorts with extensive coronary calcification, reflecting a higher false-positive rate. CCTA, coronary CT angiography; ICA, invasive coronary angiography; TAVI, transcatheter aortic valve implantation.

• 

  Figure 4. (A) SROC curve—valve surgery subgroup (19 studies; 5132 patients). Pooled sensitivity of 92% (95% CI 89% to 94%) and specificity of 87% (95% CI 82% to 91%). (B) SROC curve for the TAVI/mixed subgroup (nine studies; 3393 patients). Pooled sensitivity was 89% (95% CI 84% to 92%) and specificity was 89% (95% CI 85% to 93%), with a DOR of 41.9 (95% CI 8.94 to 196.4), reflecting reduced discriminatory power compared with surgical cohorts. DOR, diagnostic OR; SROC, summary receiver operating characteristic; TAVI, transcatheter aortic valve implantation.

• 

  Figure 5. Box plot of sensitivity stratified by population type (valve surgery vs TAVI vs mixed). TAVI, transcatheter aortic valve implantation.

• 

  Figure 6. Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) risk of bias assessment across four domains (patient selection, index test, reference standard and flow/timing).