Risk stratification for coronary artery disease in multi-ethnic populations: Are there broader considerations for cost efficiency?

Table 2 Sensitivity and specificity non-invasive test

Test	Guideline indication	Sensitivity	Specificity	Stress modality	Advantage	Disadvantage
ECG	1st L-PTP	45-50	85-90	Physical	Simple and safe	Accuracy
	2nd I-PTP				Availability	ECG artifact
					Lower cost	False positives
Echo	1st L-PTP	80-85	80-88	Physical	Simple and safe	Suboptimal image quality e.g., resting wall motion defects, lung disease, respiratory artifact,
	2nd L-PTP			Pharma*	Availability	Image capture within 90 sec of peak HR
					Lower cost	Cost of contrast
					No radiation
					ECG independent
					Mobility independent*
					Ischemia: Quantify and localize; greater spatial resolution (subendocardial)
Myocardial perfusion scintigraphy (SPECT, PET)	1st L-PTP	73-92	63-87	Physical	Accurate quantification ischemic area	Cost
	2nd L-PTP	90%	75-87	Pharma*	Ischemia: Quantify and localize; greater spatial resolution (subendocardial)	Availability
					↑ accuracy with septal defects	Radiation and retesting
						Ischemia: ↓spatial resolution e.g. for subendocardial ischemia
						Pharma: CI, SE, ↓ sensitivity for multivessel disease
						↑ acquisition time
						Artifacts: Lung motion, breast tissue, diaphragm attenuation
MRI
Ischemia	1st I-PTP	79-88	81-91	Pharma*	Body habitus/lung window independent	Cost
Perfusion	2nd L-PTP	67-94	61-85		Accurate	Availability
					No radiation	Expertise
					Operator independence	↓ Gating: Rhythm and rate
					High spatial resolution
					Can perform absolute quantification of perfusion
CA Score	1st L-PTP	95-99	64-83	Direct visualization coronary artery	Availability	Radiation
CTCA	2nd I-PTP				Non-invasive	Cost
					Anatomical information	Ca score role
					FFR	No functional information
						Contrast
						↓ Gating: Rhythm and rate

Guideline referral for diagnostic evaluation is guided by PTP. All males start with a PTP of > 15% and thus warrant at least a Stress ECG. Females between 30 and 39 years have PTP 10% and clinical judgement may suffice. When PTP is > 85%, e.g., males > 60 years with typical symptoms, coronary imaging can be considered first line or complimentary based on clinical judgement. Females are not given a PTP > 85% and thus should at least receive a functional test. The choice of functional test is described in the body of the table. Non-imaging stress ECG has three information sources - PTP, subjective (test symptoms and hemodynamics) and solitary objective (ECG) marker of the ischemic cascade. Imaging stress testing adds baseline myocardial function to the PTP and additional objective components of the ischemic cascade improving accuracy. Greater advancement in addressing issues of access, cost and reimbursement, complementary (dual) modality testing, additional means to assess ischemic cascade components (e.g., tissue strain), image quality issues (e.g., contrast echo) could alter the test indication and its accuracy. PET - imaging has lower radiation, higher resolution and quantify blood flows (including microvasular), but cost and availability are issues. MRI identifies wall motion changes via dobutamine or myocardial perfusion by vasodilators. Exercise: treadmill; bicycle; right ventricular pacing. Pharmacological: dobutamine; adenosine; dypiridamole, ragadenoson. Imaging Sources: SPECT isotope: Technetium 99m, thallium 201. CA: Calcium; CI: Contraindication; CTCA: Computerized tomography coronary angiography; ECG: Electrocardiography; Echo: Echocardiography; FFR: Fractional flow reserve; H-PTP: High pretest probability; I-PTP: Intermediate pretest probability; L-PTP: Low pretest probability; PET: Positron emission tomography; PTP: Pretest probability; MRI: Myocardial resonance imaging; SE: Side-effects; SPECT: Single photon emission computerized tomography (Data partly synthesized from reference 38, 100, 101). *Physical exercise is the preferred stress modality. It provides higher physiological stress and workloads with greater opportunity for corroboration with patient’s symptoms. Although studies have shown similar accuracy between physical and pharmacological stress test, we feel there is insufficient evidence to draw similar long term conclusions on the diverse etiologies and severity of CAD emanating from more diverse patient populations.