Table 5.

Adjusted within-pair differences in plasma inflammatory biomarker levels based on CFR discordance status.

	Geometric Means of Inflammatory Markers (SE) (122 Twin Pairs)				Geometric Means of Inflammatory Markers (SE) (50 twin pairs)
	Twin with Lower CFR	Twin with Higher CFR	Within-Pair Differenc e (%)	P	Twin with CFR <2.5	Twin with CFR ≥2.5	Within-Pair Differenc e (%)	P
CRP (mg/L)	1.42 (1.10)	1.11 (1.10)	22.2	0.03	1.77 (1.17)	1.09 (1.17)	62.9	0.006
IL-6 (pg/dL)	2.05 (1.07)	1.65 (1.07)	19.4	0.02	2.63 (1.11)	1.74 (1.11)	51.2	0.003
WBC (1000/μL)	6.38 (1.02)	5.89 (1.02)	7.6	0.003	6.42 (1.04)	5.94 (1.04)	8.2	0.06
ICAM-1 (ng/mL)	322 (1.02)	301 (1.02)	6.5	0.02	332 (1.04)	302 (1.04)	9.7	0.04
VCAM-1 (ng/mL)	610 (1.03)	570 (1.03)	6.6	0.049	605 (1.05)	580 (1.05)	4.3	0.44

CFR: coronary flow reserve; CI: confidence interval; SE: standard error; CRP: C-reactive protein; IL-6: interleukin-6; WBC: white blood cell count; VCAM-1: vascular cell adhesion molecule-1; ICAM-1: intercellular adhesion molecule-1.

CFR is defined as the ratio of myocardial blood flow during adenosine to flow at baseline. All values in table are derived from mixed models with random intercept for pair.

All estimates adjusted for perfusion defect score, Framingham risk score (including age, current smoking, systolic blood pressure, LDL-cholesterol, HDL-cholesterol and diabetes), body mass index, and study year. Adjusting for additional covariables, including education, marital status, habitual physical activity, rate-pressure product during the adenosine infusion, previous history of coronary heart disease, and medications (statins, beta-blockers, aspirin), did not materially change the results.