Home › Cardiac arrest › Higher CPR SpO2 linked to ROSC, not good outcome

Cardiac arrest

Higher CPR SpO2 linked to ROSC, not good outcome

Associations between pulsed oxygen saturation during out-of-hospital cardiopulmonary resuscitation, cardiopulmonary resuscitation quality and return of spontaneous circulation
Resuscitation · Added 25 September 2026

Study of 98 OHCAs asking whether pulse oximetry readings during CPR relate to outcomes — higher event-average SpO2 was associated with sustained ROSC (aOR 1.05, AUC 0.77, cutpoint ~71%) and survival to admission, but not neurological intact survival, and cycle-average SpO2 did not track compression quality metrics.

Study type

Cohort

Design

Retrospective observational cohort of OHCA resuscitations

Participants

98 index OHCAs (mean age 62.1 years, 32% female)

Main finding

Median event-average SpO2 was 55.0; higher event-average SpO2 was associated with sustained ROSC (aOR 1.05, 95% CI 1.02–1.08; AUC 0.77, cutpoint ~71) and survival to hospital admission (aOR 1.03, 95% CI 1.01–1.05) but not neurological intact survival (aOR 1.00, 95% CI 0.96–1.05). Cycle-average SpO2 was not associated with chest compression quality metrics.

Caveat

Very small sample (98 events) and low SpO2 values during CPR raise measurement reliability concerns

Bottom line

Hypothesis-generating

Learning goal

Weigh the value of pulse oximetry readings during cardiac arrest resuscitation as a marker of effective perfusion.

More

• Topic: Cardiac arrest
• Daily scan: 25 September 2026

Related papers on Cardiac arrest

• No evidence on screening arrest survivors' psychology

  Resuscitation · 26 Sep 2026 · Systematic review · Context only

  Systematic review of structured identification and assessment of psychological, emotional or cognitive problems in cardiac arrest survivors — of 1,605 records screened, none met eligibility, so there is no evidence base for which screening instruments, thresholds, timing or referral pathways should be used.

• Drone AED network modelling for Scotland

  Reliability Engineering & System Safety · 25 Sep 2026 · Modelling · Hypothesis-generating

  Bayesian optimisation of a drone-assisted AED delivery network using geographically referenced Scottish cardiac arrest data — station placement driven by OHCA survival probability and existing EMS coverage, with cost-effectiveness analysis suggesting drone AED delivery is cost-effective and could meaningfully improve response coverage in rural areas with long ambulance response times.

• rCAST score in young children after arrest

  Resuscitation Plus · 25 Sep 2026 · Cohort · Context only

  External validation of the rCAST post-cardiac arrest prognostic score in 75 infants and young children after targeted temperature management — good discrimination (AUC 0.947 for unfavourable neurologic outcome, 0.857 for mortality) with no favourable outcomes in the high-severity group, but 76% of low-score patients did well and the authors caution against using it for clinical decisions.

• Dispatcher CPR barriers delay bystander CPR, cut survival

  Resuscitation · 24 Sep 2026 · Cohort · Context only

  Retrospective cohort of 3,263 transcribed Swedish 112 calls linked to the national cardiac arrest registry quantified barriers to dispatcher-assisted CPR after the dispatcher recognised arrest — 43.7% of calls hit at least one barrier (commonest: inability to move the patient), bystander CPR reached 85% in barrier-free calls but stayed below 50% with three or more, and any barrier was associated with lower ROSC (adjusted RR 0.89) and 30-day survival (adjusted RR 0.57).