. Author manuscript; available in PMC: 2010 May 5.

Published in final edited form as: Crit Care Med. 2009 Aug;37(8):2387–2393. doi: 10.1097/CCM.0b013e3181a960d6

Table 3.

Longitudinal analysis of cytokine concentrations over time

Characteristics	Interleukin-6 Slope Estimate^a (95% Confidence Interval)	p Value	Tumor Necrosis Factor Slope Estimate^a (95% Confidence Interval)	p Value
Donor age, per 5 yrs	0.06 (−0.08, 0.20)	.44	−0.02 (−0.10, 0.06)	.64
Female gender	0.92 (−0.06, 1.91)	.067	0.56 (−0.25, 1.30)	.19
Time after brain death, hrs	−0.10 (−0.14, −0.06)	<.0001	−0.008 (−0.019, 0.003)	.14
Corticosteroid use	−0.25 (−1.57, 1.08)	.71	−0.37 (−1.05, 0.32)	.29
Preload responsiveness	1.69 (0.36, 3.03)	.013	0.78 (0.02, 1.53)	.044

Cytokine concentrations are expressed as natural logarithms in pg/mL and modeled as a linear function of donor age, gender, corticosteroid use, duration after brain death, and preload responsiveness over time. Log IL-6 levels over time are not significantly associated with donor age or corticosteroid use. However, female donors have a trend of increased log IL-6 concentrations compared with male donors over time (p = .067). On average, preload responsive donors have significantly higher log IL-6 concentrations compared with preload unresponsive donors (mean difference 1.7 log, p = .013). Whereas log IL-6 concentrations decreased over time for all donors (overall rate of decline 1/10 of a log per hour, p < .0001), the rate of decline was not different between preload responsive and unresponsive donors (data not shown). Similarly, log TNF concentrations were significantly higher for preload responsive donors compared with preload unresponsive donors (p = .044); however, there were no differences in TNF concentrations by age, gender, corticosteroid use, or duration after brain death.