Initial Red Cell Infusion Rate and Escalation of Respiratory Support in Critically Ill Adults: A Target Trial Emulation
- 1Department of Emergency Medicine, West China Hospital, Sichuan University, Chengdu 610041, China
- 2Department of General Surgery, West China Tianfu Hospital, Sichuan University, Chengdu 610041, China
- 3West China School of Nursing, West China Hospital, Sichuan University, Chengdu 610041, China
Abstract
Background/Objectives: Current transfusion guidelines address when red cells should be given, not how fast. We estimated the association between the initial infusion rate of a first red cell transfusion and escalation of respiratory support in critically ill adults. Methods: We emulated a target trial in MIMIC-IV v3.1. Of 2790 eligible adults, 1787 received a slower (50–175 mL/h) and 1003 a faster (>175–350 mL/h) initial rate; the primary outcome was escalation to advanced respiratory support or death within 12 h. Overlap weighting estimated marginal risk differences in the overlap population, with 2000 bootstrap resamples refitting the propensity-score model. Sensitivity and continuous dose–response analyses were pre-specified. Results: The primary outcome was classifiable in 2671 patients, with 143 events. Weighted 12 h risks were 4.95% and 6.60%: risk difference 1.65 percentage points (95% CI −0.41 to 3.74), risk ratio 1.33 (0.93 to 1.92). Invasive mechanical ventilation or death, a nested key secondary outcome unadjusted for multiplicity, gave 2.55 percentage points (0.71 to 4.59); 28-day mortality gave −1.45 (−4.10 to 1.30). In the pre-specified dose–response analysis, standardised 12 h risk was lowest near 140 mL/h (4.27%) and rose to 8.83% (6.56 to 11.28) at 350 mL/h. Conclusions: The primary analysis did not distinguish the two initial-rate strategies, although the interval remained compatible with a clinically important increase in risk. The secondary and dose–response findings are hypothesis-generating and do not support changing current infusion-rate practice. These estimates support the design and sample-size planning of a prospective trial assigning infusion rate.
Related
- Early piperacillin exposure during continuous infusion in critically ill patients: the impact of renal and liver dysfunctions
- Cefepime exposure in prolonged or intermittent infusion in critically ill children
- Early midazolam exposure and short-term risk of acute exacerbation of COPD in critically ill patients: a target trial emulation using the MIMIC-IV database.
- Temporal patterns of respiratory rate exposure and mortality in critically ill patients: a retrospective multi-database cohort study
- Association of antifungal treatment duration with 90-day mortality in critically ill adults with candidemia: a multicenter target trial emulation in France
BibTeX
@article{zhao2026initial,
title = {{Initial Red Cell Infusion Rate and Escalation of Respiratory Support in Critically Ill Adults: A Target Trial Emulation}},
author = {Zhenkun Zhao and Xiaoye Chen and Zhongjing Pan and Ao Sha and Yadong Zhou and Xiaoli Chen and Tao Cheng and Lu Gan},
journal = {Journal of Clinical Medicine},
year = {2026},
month = oct,
doi = {10.3390/jcm15208147},
url = {https://www.mdpi.com/2077-0383/15/20/8147},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Initial Red Cell Infusion Rate and Escalation of Respiratory Support in Critically Ill Adults: A Target Trial Emulation
AU - Zhenkun Zhao
AU - Xiaoye Chen
AU - Zhongjing Pan
AU - Ao Sha
AU - Yadong Zhou
AU - Xiaoli Chen
AU - Tao Cheng
AU - Lu Gan
PY - 2026
DA - 2026/10/01
JO - Journal of Clinical Medicine
DO - 10.3390/jcm15208147
UR - https://www.mdpi.com/2077-0383/15/20/8147
AB - Background/Objectives: Current transfusion guidelines address when red cells should be given, not how fast. We estimated the association between the initial infusion rate of a first red cell transfusion and escalation of respiratory support in critically ill adults. Methods: We emulated a target trial in MIMIC-IV v3.1. Of 2790 eligible adults, 1787 received a slower (50–175 mL/h) and 1003 a faster (>175–350 mL/h) initial rate; the primary outcome was escalation to advanced respiratory support or death within 12 h. Overlap weighting estimated marginal risk differences in the overlap population, with 2000 bootstrap resamples refitting the propensity-score model. Sensitivity and continuous dose–response analyses were pre-specified. Results: The primary outcome was classifiable in 2671 patients, with 143 events. Weighted 12 h risks were 4.95% and 6.60%: risk difference 1.65 percentage points (95% CI −0.41 to 3.74), risk ratio 1.33 (0.93 to 1.92). Invasive mechanical ventilation or death, a nested key secondary outcome unadjusted for multiplicity, gave 2.55 percentage points (0.71 to 4.59); 28-day mortality gave −1.45 (−4.10 to 1.30). In the pre-specified dose–response analysis, standardised 12 h risk was lowest near 140 mL/h (4.27%) and rose to 8.83% (6.56 to 11.28) at 350 mL/h. Conclusions: The primary analysis did not distinguish the two initial-rate strategies, although the interval remained compatible with a clinically important increase in risk. The secondary and dose–response findings are hypothesis-generating and do not support changing current infusion-rate practice. These estimates support the design and sample-size planning of a prospective trial assigning infusion rate.
SN - 2077-0383
ER -