Skip to main content
. Author manuscript; available in PMC: 2011 Jan 1.
Published in final edited form as: Pediatr Res. 2010 Jan;67(1):77–82. doi: 10.1203/PDR.0b013e3181c0b16c

Table 4.

Median regressions for oxygen exposure in the prediction of elevated levels of cytokines, adjusting for covariates.

p-value§ Parameter estimate (95% CI)** Significant covariates*
MMP 9 – 1 day 0.80 19512 (−199921, 209885) GA
MMP 9 – 3 day 0.20 −117004 (−325486, 12894) GA, gender, RDS
MCP-1 – 3 day 0.003 471 (173890) GA
MIP-1β – 1 day 0.10 85 (−22, 217) EOS
MIP-1β – 3 day 0.47 −30 (−98, 56) Apgar5, EOS
RANTES – 3 day 0.18 −5836 (−19243, 2406) none
§

P-values from likelihood ratio tests.

**

The parameter estimate can be interpreted as the adjusted median difference in plasma levels between the two comparison groups (for example, 3 day MCP level for the oxygen exposed group is on average 471 pg/ml higher than that for the non-exposed group, after adjusting for covariates)

*

Covariates included: gestational age, gender, 5 minute Apgar < 5, any ANS (including infants who received ANS but did not have a complete course), severity of RDS (defined as > 2 doses of surfactant or any high frequency ventilation HFV vs. ≤ 2 doses of surfactant and no HFV) and early onset sepsis (EOS).