Skip to main content
. 2023 May 31;15:454–467. doi: 10.1016/j.xjon.2023.04.022

Table 3.

Clinical trial primary and subset outcomes

Outcome n/N evaluated (%)
red blood cell group
Relative risk
(95% CI)
Absolute risk difference
(95% CI)
P value
Fresh Standard issue
Primary outcome: MODS
 New and/or progressive MODS§ 43/89 (48.3) 38/88 (43.2) 1.12 (0.81-1.54) 5.1 (−9.5 to 19.8) .49
 New MODS 43/89 (48.3) 37/88 (42.1)
 Progressive MODS 0/89 (0.0) 1/88 (1.1)
Subset outcomes: MODS
 Age (d)
 ≤28 15/15 (100.0) 13/13 (100.0)
 29-365 15/36 (41.7) 20/39 (51.3) 0.81 (0.50-1.33) −9.6 (−32.1 to 12.9) .51
 >365 13/38 (34.2) 5/36 (13.9) 2.46 (0.98-6.21) 20.3 (0.1-39.2)
 STAT mortality category
 1-3 23/65 (35.4) 21/69 (30.9) 1.15 (0.71-1.86) 4.5 (−11.5 to 20.5) .58
 4 or 5 20/24 (83.3) 17/20 (85.0) 0.98 (0.76-1.27) −1.7 (−23.3 to 20.0)
Sensitivity analysis on primary outcome: MODS
 Adjusted for baseline covariates 1.01 (0.90-1.12)∗∗ .89

MODS, Multiple organ dysfunction syndrome; STAT, Society of Thoracic Surgeons–European Association for Cardio-Thoracic Surgery.

In all comparisons, the fresh red blood cell group was used as the reference. Superiority was checked for the primary outcome and for all secondary outcomes analyzing patients according to their randomization groups. The principal analysis was performed using an unadjusted χ2 comparing the proportion of patients who acquire new or progressive MODS after randomization. The principal measure of effect is an unadjusted absolute risk difference with 95% CI. Dichotomous secondary outcomes were analyzed using risk differences and 95% CI followed by logistic regression procedures. Continuous outcomes were analyzed using independent t tests or Wilcoxon rank-sum tests depending on distribution of data.

Refers to number of patients with outcome/number of patients evaluated (proportion). n refers to number analyzed when it is less than the group total.

The primary outcome is described in the Methods section.

§

Defined by Proulx and colleagues.17

One patient who was randomized to the standard-issue group died in the operating room during cardiac surgery and had no data available for the primary outcome and some secondary outcomes. This patient was not included in the primary outcome but was included in mortality analyses.

Adjusted for weight at admission to ICU, kg (log transformed), P < .001, hemoglobin level before first transfusion, P = .071, cardiopulmonary bypass duration (log transformed), P = .002, circulatory arrest, P = .013, therapeutic hypothermia, P < .001 (P values are values from univariate analysis) and STAT score (as categorical).

∗∗

Odds ratio (95% CI) from multivariate logistic regression.