Table 2. Linear models of the paired change in cardiovascular markers and inflammation biomarkers over 12 weeks of ART.

	Cardiovascular markers (change in regression outcome)
Inflammation biomarkers (change in log- transformed regression predictor)	Isoprostane-to-Creatinine ratio	Augmentation Index	Reactive Hyperemia Index	QTc	RMSSD	Triangular Index	Low frequency variations	High frequency variations	Power Ratio	Systolic blood pressure	Diastolic blood pressure	Heart rate
C-reactive protein	0.00, p=0.96	-2.22, p=0.54	-0.09, p=0.07	1.35, p=0.74	-5.76, p=0.37	-1.32, p=0.18	5.60, p=0.65	-22.80, p=0.18	0.05, p=0.81	1.62, p=0.60	-0.09, p=0.97	3.43, p=0.23
Tumor necrosis factor-α receptor 1	-0.08, p=0.52	-6.22, p=0.54	-0.26, p=0.03	9.62, p=0.31	-10.09, p=0.69	-2.13, p=0.41	12.78, p=0.70	-38.72, p=0.40	0.02, p=0.97	0.20, p=0.98	-3.98, p=0.38	14.10, p=0.06
Soluble CD163	0.02, p=0.86	-14.06, p=0.04	-0.11, p=0.17	2.71, p=0.68	-2.05, p=0.91	-1.75, p=0.34	28.46, p=0.16	1.86 p=0.95	0.33, p=0.23	5.19, p=0.30	2.34, p=0.46	9.44, p=0.06
Soluble CD14	-0.03, p=0.83	-15.37, p=0.25	-0.05, p=0.77	-4.08, p=0.73	11.25, p=0.76	0.41, p=0.91	9.91, p=0.79	-21.07, p=0.72	-0.03, p=0.96	-2.51, p=0.78	2.45, p=0.67	13.31, p=0.17

Models estimate the relationship between the change in log-transformed inflammation biomarker and change in cardiovascular marker using ordinary least squares regression, adjusted for baseline values. Bold values represent statistically significant findings.

Abbreviation: QTC, corrected QT interval; RMSSD, root mean square of the successive differences.

Reactive hyperemia index is natural log transformed