Lower oxygen delivery on cardiopulmonary bypass was linked to higher 30-day mortality

Design

Single-centre retrospective cohort

Population

4,358 adults having cardiac surgery with cardiopulmonary bypass, 2018–2024

Primary outcome

30-day mortality

Effect

aOR 1.16 (95% CI 1.09–1.23) per 10 mL/min/m² lower DO2i; adjusted mortality 1.69% vs 0.74% at 25th vs 75th centile

A single US tertiary centre calculated time-weighted indexed oxygen delivery (DO2i) on bypass for 4,358 adults having cardiac surgery between 2018 and 2024, from pump flow, haemoglobin, arterial saturation and body surface area.

Thirty-day mortality was 1.7% overall. Each 10 mL/min/m² fall in DO2i was associated with 16% higher odds of death (adjusted OR 1.16). Adjusted mortality was 1.69% at the 25th centile of DO2i (about 226 mL/min/m²) and 0.74% at the 75th centile (about 296). Lower DO2i was also associated with acute kidney injury and slightly longer ICU and hospital stays, but not with stroke.

This is observational: sicker patients may have lower haemoglobin or flows for reasons that also raise their risk. It supports watching oxygen delivery rather than flow or haematocrit alone, and it justifies trials of goal-directed perfusion, but it does not set a proven target. It was published on 28 September 2026.

• Track indexed oxygen delivery during bypass, not flow or haemoglobin in isolation.
• Discuss with perfusionists how haemoglobin, flow and saturation are balanced when DO2i runs low.
• Treat low DO2i as a risk marker for kidney injury after bypass.
• Do not transfuse to a DO2i target on this evidence alone; the link is observational.

Why it matters

It points to a single, measurable variable that combines three things perfusion teams already manage.

Don't overread it

An association in one centre does not show that raising oxygen delivery lowers mortality.

The statistics, in plain English

An adjusted odds ratio of 1.16 per 10-unit fall means each step down in oxygen delivery was linked to about 16% higher odds of death, after accounting for measured differences. With only 75 deaths, estimates are less precise than the large cohort suggests. Adjustment cannot account for everything that makes a patient sicker.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.