Preethy J Mathew, Joseph L Mathew

The laryngeal mask airway (LMA) is a safe and effective modality to maintain the airway for general anaesthesia during surgical procedures. The LMA is removed at the end of surgery and anaesthesia, when the patient maintains an adequate respiratory rate and depth. This removal of the LMA can be done either when the patient is deep under anaesthesia (early removal) or only after the patient has regained consciousness (late removal). It is not clear which of these techniques is superior. The objective of this review was to compare the safety of LMA removal in the deep plane of anaesthesia (early removal) versus removal in the awake state (late removal) for participants undergoing general anaesthesia. We searched the Cochrane Central Register of Controlled Trials (CENTRAL; 2014, Issue 8); MEDLINE (1966 to August 2014); EMBASE (1980 to August 2014); LILACS (1982 to August 2014); CINAHL (WebSPIRS; 1984 to August 2014); and ISI Web of Science (1984 to August 2014). We searched for ongoing trials through various trial registration websites. In addition, we searched conference proceedings and reference lists of relevant articles. We included randomized controlled trials (RCTs) on adults and children undergoing elective general anaesthesia using the LMA, that compared early removal of the LMA (defined as removal of the LMA in the deep plane of anaesthesia) versus late removal of the LMA (defined as removal of the LMA after the patient is awake). Two authors independently assessed trial quality and extracted data. We contacted study authors for additional information. We used a random-effects model to generate forest plots from the data. We identified a total of 9188 citations and included 15 RCTs conducted on 2242 participants in this review. All trials used the LMA Classic in American Society of Anesthesiologists (ASA) physical status I or II for patients undergoing elective general anaesthesia. Children were enrolled in 11 trials and adults in five trials. None of the trials were of high methodological quality. Eight of the 15 studies had adequate generation of random sequence, whereas only one trial had adequate concealment of random sequence. Three trials had blinded the outcome assessor. Thus, the majority of the studies appeared to have a high risk of bias in the study design.Using the GRADE approach, we found low quality evidence that the risk of laryngospasm was similar with early removal of the LMA (3.3%) versus late removal (2.7%): risk ratio (RR) 1.23, 95% confidence interval (CI) 0.74 to 2.03; 11 trials, 1615 participants. The quality of evidence was very low that the risk of coughing was less after early removal (13.9%) than late removal (19.4%): RR 0.52, 95% CI 0.29 to 0.94; 11 trials, 1430 participants. The quality of evidence for the risk of desaturation was also very low; there was no difference between early removal (7.9%) and late removal (10.1%): RR 0.68, 95% CI 0.4 to 1.16; 13 trials, 2037 participants. We found low quality evidence that the risk of airway obstruction was higher with early removal (15.6%) compared to late removal of the LMA (4.6%): RR 2.69, 95% CI 1.32 to 5.5; eight trials, 1313 participants. This systematic review suggests that current best evidence comparing early versus late removal of the LMA in participants undergoing general anaesthesia does not demonstrate superiority of either intervention. However, the quality of evidence available is either low or very low. There is a paucity of well designed RCTs and a need for large scale RCTs to demonstrate whether early removal or late removal of the LMA is better after general anaesthesia.