Table 3.

Cox proportional hazard model for the time to late PH.

Variable	Unadjusted		Adjusted
	HR (95% CI)	p-value	HR (95% CI)	p-value
Atrial septal defect (Yes vs. No)	2.37 (1.29, 4.36)	0.005*	2.44 (1.27, 4.68)	0.007*
Respiratory support at 28 days(yes vs. no)	0.68 (0.35, 1.29)	0.237	0.73 (0.36, 1.46)	0.366
Sepsis (yes vs. no)	1.76 (0.96, 3.22)	0.070	2.31 (1.19, 4.49)	0.013*
African American Maternal race (yes vs. no)	3.00 (1.41, 6.40)	0.005*	3.34 (1.39, 8.06)	0.007*
Birthweight (< 800 g vs. >800 g)	2.66 (1.36, 5.20)	0.004*	2.24 (1.09, 4.59)	0.028*
Intrauterine growth restriction (yes vs. no)	1.36 (0.54, 3.45)	0.541	1.63 (0.63, 4.23)	0.319

The hazard for the development of late pulmonary hypertension was determined using unadjusted Kaplan-Meier survival probabilities. In adjusted Cox proportional hazard models, the effect of atrial septal defect on the outcome of late PH was determined after controlling for variables that were significant in descriptive analyses (sepsis, African American maternal race, and birth weight) or in the literature (respiratory support, intrauterine growth restriction). Hazard ratios (HR) and 95% confidence intervals (CI) were determined. Statistical significance was defined as p-value < 0.05.