Pediatrics · Medical Literature
Effects of laryngeal mask airway removal under different anesthesia states on pediatric airway complications: a systematic review and meta-analysis.
Abstract
This meta-analysis was designed to compare airway complication rates between deep vs. awake laryngeal mask airway (LMA) removal in pediatric non-airway surgery, while discussing the impact of anesthetic agents, patient position and age. We systematically searched PubMed, Web of Science, and Cochrane Library to identify studies meeting the inclusion criteria. The primary outcome measured was the incidence of overall airway complications, which included laryngospasm, airway obstruction, desaturation, breath-holding, cough, and excessive secretions. Secondary outcomes included subgroup analyses stratified by patient positioning during LMA removal, inhalational anesthetic agents and age. Pooled analysis of 10 randomized controlled trials (RCTs) (n = 1,713) revealed no significant difference in overall airway complications (relative risk (RR) of 0.93, 95% confidence interval (CI) [0.65-1.32], p = 0.68) or specific outcomes (laryngospasm, desaturation, breath-holding, cough) between deep and awake LMA removal; however, deep removal was associated with a significantly higher airway obstruction risk and a lower incidence of excessive secretions. Subgroup analyses demonstrated that inhalational anesthetic agents type (sevoflurane/isoflurane) did not alter most complications but universally increased airway obstruction risk, whereas positioning during deep removal reduced secretions in both lateral/supine positions yet increased obstruction, with lateral positioning uniquely lowering desaturation risk. Notably, age did not modulate the airway obstruction risk profile: children under 6 years and over 6 years exhibited comparably elevated obstruction risks with deep removal. This meta-analysis suggests awake LMA removal is associated with lower airway obstruction risk than deep removal, which primarily reduces secretions. Given the higher clinical severity of obstruction, clinicians may prioritize its prevention. This finding appears consistent across age groups, though caution is advised for infants (<2 years) due to limited data. Lateral positioning may mitigate desaturation during deep removal. We suggest risk-stratified protocols, with lateral positioning and monitoring advised if deep removal is performed.
Clinical relevance
In pediatric patients undergoing non-airway surgery with a laryngeal mask airway (LMA), does removal during deep anesthesia compared with removal in the awake state result in a different overall rate of airway complications, and do anesthetic agent type, patient positioning, and age modify this effect? __SNAPSHOT__ {"studyDesign":"Systematic review and meta-analysis of randomized controlled trials (RCTs)","setting":"Pooled multi-institutional data; databases searched: PubMed, Web of Science, and Cochrane Library; individual trial settings not reported in abstract","sampleSize":"n = 1,713 across 10 RCTs (intervention/control allocation per individual trial not reported in abstract)","population":"Pediatric patients undergoing non-airway elective surgery under general anesthesia with LMA; specific age range, ASA status distribution, and comorbidity criteria not fully reported in available abstract","intervention":"LMA removal during deep anesthesia (deep removal)","comparator":"LMA removal in the awake state (awake removal)","primaryOutcome":"Incidence of overall airway complications, defined as the composite of laryngospasm, airway obstruction, oxygen desaturation, breath-holding, cough, and excessive secretions","followUp":"Perioperative/emergence period; exact follow-up duration not reported","fundingSource":"Not reported"}
Key findings
The pooled analysis of 10 RCTs (n = 1,713) demonstrated no statistically significant difference in overall airway complications between deep and awake LMA removal (RR 0.93, 95% CI 0.65–1.32, p = 0.68). Disaggregated analysis of individual complication types — laryngospasm, oxygen desaturation, breath-holding, and cough — also revealed no significant difference between groups. However, two specific outcomes diverged by removal technique: deep LMA removal was associated with a significantly higher risk of airway obstruction compared with awake removal, while awake removal was associated with a significantly higher incidence of excessive secretions compared with deep removal (specific RR, CI, and p-values for these subgroup outcomes not reported in the available abstract). Subgroup analyses by anesthetic agent (inhalational versus intravenous), patient positioning (lateral versus supine), and age were conducted, but the full quantitative results of these subgroup analyses were not available in the truncated abstract. NNT and NNH cannot be calculated due to absence of reported absolute event rates per group.
Practice implications
This meta-analysis provides moderate-certainty evidence that, for overall airway complications, the timing of LMA removal (deep vs. awake) does not confer a meaningful aggregate advantage in pediatric non-airway surgery. However, the finding that deep removal carries a significantly higher risk of airway obstruction is clinically important: upper airway obstruction in a deeply anesthetized child with attenuated laryngeal protective reflexes may require active airway maneuvers or instrumentation, and rare but serious events (including laryngospasm with obstruction) have been documented in registry-level data from deep removal practice. Conversely, the higher rate of excessive secretions with awake removal may reflect intact gag reflexes triggering hypersalivation and is generally a lower-acuity concern. The subgroup findings regarding anesthetic agent (inhalational versus propofol-based TIVA), patient position (lateral versus supine), and age are potentially clinically significant — prior literature suggests sevoflurane-maintained awake removal carries higher adverse event rates than propofol-based TIVA, and lateral positioning during deep removal substantially reduces airway obstruction risk — but the complete subgroup statistics from this review are not available in the truncated abstract and therefore cannot be fully adjudicated. The findings are most applicable to ASA I–II pediatric patients undergoing elective non-airway surgery; applicability to infants, neonates, children with reactive airway disease, obesity, or upper respiratory tract infections is uncertain. IMPORTANT LIMITATION: Only a truncated abstract was available for this review; full manuscript data including complete subgroup statistics, heterogeneity metrics (I²), risk-of-bias assessments for individual trials, and sensitivity analyses have not been reviewed, and conclusions should be tempered accordingly. __LIMITATIONS__ Only a truncated abstract was available; complete subgroup quantitative data, heterogeneity statistics (I²), and risk-of-bias assessments per trial could not be independently verified.,Only 10 RCTs with a total of 1,713 participants were pooled; individual trials are likely underpowered to detect rare serious events such as severe laryngospasm or aspiration.,Substantial heterogeneity is anticipated given the variability in anesthetic agents (sevoflurane, halothane, isoflurane, propofol TIVA), definitions of 'deep' and 'awake' states, patient positioning, age ranges, and surgical case mix across included trials.,Restriction to non-airway surgery limits generalizability; children at highest airway complication risk (e.g., tonsillectomy, reactive airway disease, URI) may not be adequately represented.,The composite primary outcome obscures clinically distinct events of differing severity; laryngospasm and airway obstruction carry substantially greater harm potential than cough or excessive secretions.,Blinding of outcome assessors is practically difficult in this context, introducing potential performance and detection bias.,The truncated abstract prevents assessment of publication bias, funnel plot analysis, or GRADE evidence certainty ratings.,Funding source is not reported, precluding assessment of sponsorship bias.,Age-stratified subgroup results are not available from the abstract, limiting conclusions about differential risk in infants versus older children.,Statistical significance of individual secondary outcomes (airway obstruction, excessive secretions) is reported directionally but without specific RR, CI, or p-values in the available abstract. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ In pediatric patients undergoing non-airway elective surgery, deep versus awake LMA removal produces equivalent overall airway complication rates (RR 0.93, 95% CI 0.65–1.32, p = 0.68), but deep removal is associated with significantly more airway obstruction while awake removal carries more excessive secretions. Clinicians should weigh the risk of unprotected airway obstruction in a deeply anesthetized child against the lower-acuity nuisance of secretions with awake removal, and individualize technique based on anesthetic agent used, patient positioning (lateral positioning appears protective with deep removal), and patient-specific risk factors; neither technique is universally superior.