Upper airway obstruction presents as a falling saturation

Why it matters

Airway obstruction in the recovering patient is the emergency the PACU exists for, and it does not always announce itself. In the Australian Incident Monitoring Study, laryngospasm was clinically obvious in 77% of reports; in the remainder it presented as airway obstruction in 14%, as regurgitation or vomiting in 5%, and as desaturation alone in 4% visvanathan-2005-laryngospasm. The companion analysis of natural-airway obstruction carries the same warning in a different setting: obstruction may present simply as desaturation visvanathan-2005-airway.

The consequence of getting this wrong is not the obstruction, which usually resolves. It is what obstruction does on the way to resolving: post-obstructive pulmonary oedema, aspiration and severe hypoxaemia are named by the AIMS authors as the morbidity of poorly managed laryngospasm visvanathan-2005-laryngospasm.

Mechanism

Laryngospasm is a reflex closure of the glottis, triggered by stimulation of the larynx during light planes of anaesthesia, most often at emergence. The tongue obstructs by falling back against the posterior pharynx when pharyngeal muscle tone is lost, which is the common mechanism of obstruction in the recovering patient who is not yet awake. Both produce the same downstream physiology: a patient making strong inspiratory efforts against a closed or narrowed upper airway generates a large negative intrathoracic pressure, which draws fluid across the pulmonary capillary into the alveolus. That is negative-pressure pulmonary oedema, and it can follow either mechanism. Meanwhile the obstructed patient desaturates, and the unprotected airway is exposed to whatever is in the pharynx.

Evidence

The AIMS laryngospasm analysis reviewed 189 reports among the first 4000 incidents reported to the registry visvanathan-2005-laryngospasm. Desaturation occurred in over 60% of cases, bradycardia in 6% (23% in patients under one year), pulmonary oedema in 4% and pulmonary aspiration in 3% visvanathan-2005-laryngospasm. The authors state that, had a structured approach been used, earlier recognition or better management may have occurred in 16% of cases visvanathan-2005-laryngospasm. The same paper cites the incidence from the large Scandinavian denominator study as 0.78% in more than 130,000 anaesthetics; that figure is quoted here from the AIMS paper because the primary record has not been read visvanathan-2005-laryngospasm olsson-1984.

The companion AIMS analysis of natural-airway obstruction supplies the observation that obstruction may present simply as desaturation, which is why a falling saturation in a recovering patient is an airway finding until the airway has been looked at visvanathan-2005-airway.

A 2025 series describes five cases of negative-pressure pulmonary oedema during recovery from general anaesthesia at one institution: four attributed to glossoptosis after extubation and one to laryngeal mask displacement, all with typical clinical and imaging features, and none with a poor outcome bmc-nppe-2025. It carries the point that in the recovery period the precipitant is not always laryngospasm.

What this does not show

Every source on this page is an incident registry or a case series visvanathan-2005-laryngospasm visvanathan-2005-airway bmc-nppe-2025. Registry data describe how events presented and what followed them among reported cases; they do not give a true incidence, because reporting is voluntary, and they do not test any management. The 16% figure is the authors’ judgement about what a structured approach might have changed, not a measured effect visvanathan-2005-laryngospasm.

One figure that circulates from this series, that intubation was required in 43% of cases, does not appear in the paper’s abstract or results and is not used here visvanathan-2005-laryngospasm. The bradycardia figure sometimes quoted at 11% is also wrong for the whole series; the paper gives 6%, with 23% in infants visvanathan-2005-laryngospasm.

The management ladder the director’s content list expects here is not in the held sources: [NUMBER NEEDED: laryngospasm management sequence, the Larson manoeuvre, and intramuscular versus intravenous succinylcholine dosing]. The five-case series describes supportive management of negative-pressure pulmonary oedema only in outline bmc-nppe-2025.

At the bedside

Treat a falling saturation in a recovering patient as an airway problem until you have looked at the airway. Both AIMS analyses say the same thing from different directions: obstruction can present as nothing but desaturation visvanathan-2005-laryngospasm visvanathan-2005-airway.

Distinguish the larynx from the tongue. In the recovery period the tongue is the likelier culprit, and the negative-pressure oedema that follows either looks the same bmc-nppe-2025. A jaw thrust and an oral or nasal airway address one; they do not address the other.

Use a structured approach and know it before you need it, because the registry authors judged that recognition or management could have been better in about one case in six visvanathan-2005-laryngospasm. [PRACTICE VARIES: the institutional laryngospasm algorithm and the dose and route of succinylcholine it specifies differ between hospitals; know yours.]

After any significant obstruction, watch for the oedema, the aspiration and the hypoxaemia that the registry names as its downstream harm, and do not discharge on the strength of the obstruction having resolved visvanathan-2005-laryngospasm.

Check yourself

Three items. Every option carries an explanation. Progress is not saved.