Supraglottic Airway vs. Intubation in OHCA: What a New Meta-Analysis Changes After JC-01

Thirteen studies, almost a quarter of a million patients, and a result that flips depending on whether you look at registries or randomized trials. What does it mean for an ambulance crew with a supraglottic airway in the bag?

The first part of our Journal Club (JC-01) covered the AIRWAYS-2 and PART trials and concluded that a supraglottic airway (SGA) is a fully valid option in out-of-hospital cardiac arrest (OHCA). In September 2026, Resuscitation Plus published a new systematic review and meta-analysis (Costa et al.) that pooled randomized and observational data. Its conclusions are more cautious and, on one point, contradictory. In part seven we look at what it really shows, where it is weak, and what applies to a Slovak paramedic in a non-physician crew (RZP), who is not allowed to intubate.

Clinical question (PICO)

Airway management during CPR is one of the few areas where the work of a paramedic and a physician overlap. The paramedic crew (RZP) inserts an SGA; the physician crew (RLP) or HEMS (LZS) may intubate. Whether the choice of device matters for survival therefore concerns both. Randomized trials so far leaned towards SGA, while large registries from Asia and Germany leaned towards intubation [1–3, 6].

Methods

The authors followed the Cochrane Handbook and PRISMA 2020 and pre-registered the protocol in PROSPERO (CRD420261337076). In January 2026 they searched PubMed, Embase and the Cochrane Library without language restriction. They included randomized and observational studies directly comparing SGA and TI in adults with OHCA. Where registries overlapped, the newer or larger dataset was kept [1].

Risk of bias was assessed with ROBINS-I (observational studies) and RoB 2 (randomized trials), certainty of evidence with GRADE. Results were pooled with a random-effects model and expressed as risk ratios (RR). A value below 1 favours intubation, above 1 favours SGA. The authors added subgroup analyses (device type, region, study design, timing of assessment), meta-regression, leave-one-out sensitivity analyses and an exploratory trial sequential analysis (TSA) [1].

Results

The main result is neutral. Survival to hospital discharge (9 studies) did not differ between SGA and intubation, RR 0.96 (95% CI 0.82–1.12). Heterogeneity was high (I² 80%) and the prediction interval of 0.58–1.58 allows for both benefit and harm in a future study [1].

For ROSC (12 studies), SGA was associated with a lower chance of return of circulation, RR 0.86 (0.77–0.96), especially for ROSC achieved in the field, RR 0.82 (0.71–0.95). This is where the key twist comes. In observational studies SGA did worse, RR 0.81 (0.73–0.89), but in both randomized trials it did better, RR 1.09 (1.03–1.16). When the authors pooled adjusted estimates from individual studies instead of raw counts, the direction for ROSC also reversed in favour of SGA, OR 1.24 (1.02–1.51) [1].

Favourable neurological outcome did not differ overall, RR 0.99 (0.88–1.11). Only in an exploratory subgroup of studies assessing neurology one month or later after the arrest did SGA come out borderline worse, RR 0.90 (0.81–0.99) [1].

Why the result flips

In a registry nobody randomizes which patient gets which device. Intubation is more often performed by an experienced physician, frequently later in the resuscitation or in a patient with a different prognosis. During a long resuscitation, a patient is also more likely to “make it” to an advanced airway even though the prognosis is already poor (resuscitation time bias). Randomization removes these biases. The authors themselves state that confounding by indication and resuscitation time may explain the gap between registries and RCTs, and the reversal in adjusted estimates supports that [1].

Critical appraisal

Device type. Importantly for Slovak practice, the meta-analysis cannot compare individual SGAs. The i-gel subgroup consisted of a single study, RR 1.10 (0.50–2.43), the laryngeal tube RR 1.00 (0.58–1.75); several large registries did not report the device at all [1]. The German HACORE registry, in patients with a shockable rhythm and ROSC, reported more anoxic brain injury with the laryngeal tube than with intubation (38% vs. 21%). It is, however, a small observational cohort (47 patients with LT) [6].

EMS system. The only European study in the survival analysis came from the physician-staffed German system, and SGA did worse there, RR 0.76 (0.65–0.89) [1]. That fits the logic that intubation helps where an experienced operator with high first-pass success does it. It does not automatically apply to a system where the advanced airway is placed by a paramedic with a supraglottic device.

Randomized trials. In AIRWAYS-2 (England, 9,296 patients, i-gel), good functional outcome at 30 days did not differ (6.4% vs. 6.8%), and initial ventilation was more often successful with SGA (87.4% vs. 79.0%) [2]. In PART (USA, 3,004 patients, laryngeal tube), initial airway success was 90.3% with LT versus 51.6% with intubation, and 72-hour survival 18.3% vs. 15.4% [3]. A meta-analysis of four RCTs by Forestell et al. found that SGA probably increases ROSC and secures the airway 2.5 minutes faster, with no effect on longer-term survival [5]. The Franco-Belgian CAAM trial showed that even bag-mask ventilation was not clearly equivalent to intubation: non-inferiority was not demonstrated and the mask group had more regurgitation and ventilation failure [4].

Comparison with Slovak practice and legislation

For an RZP crew, the question “SGA or intubation” in practice becomes “SGA now, or wait for the physician”. Intubation is a procedure for the RLP or HEMS physician. A Slovak Ministry of Health standard procedure specifically on airway management in prehospital OHCA could not be found on standardnepostupy.sk while preparing this article (as of 15 Oct 2026). Crews rely on the decree, their provider’s internal protocols and the ERC 2025 guidelines.

ERC 2025 recommend a stepwise approach starting with basic techniques according to the rescuer’s skills. Intubation should only be attempted by rescuers with a high success rate (by expert consensus more than 95% within two attempts), with a compression pause under 5 seconds and confirmation by continuous waveform capnography. Among supraglottic devices they prefer the i-gel over the laryngeal tube. Once an advanced airway is in place, ventilate at about 10 breaths per minute with continuous compressions. If an SGA leaks so much that ventilation is inadequate, return to 30:2 [7]. The i-gel-over-LT recommendation is a practical issue in Slovakia, as many ambulances carry a laryngeal tube. Changing equipment is the provider’s decision, not the crew’s.

Take-home for the field

The new meta-analysis gives no reason to change what RZP crews do. It does not show that SGA worsens survival. At the same time, it withdraws the overly strong conclusion of JC-01 that SGA is “proven better”. What matters is the quality of the procedure, not the device itself.

Practical implications for paramedics