Risk of pulmonary aspiration with laryngeal mask airway and tracheal tube

Anaesthesia 2009;64:1289-1294

Bernardini A, Natalini G

Abstract

Purpose            The purpose of this study was to test the hypothesis that pulmonary aspiration of gastric contents was more likely when the airway was managed with an LMA than an endotracheal tube when patients were mechanically ventilated.

Background            The Laryngeal Mask Airway (LMA) has a number of advantages over an endotracheal tube and it can be useful in the management of a difficult airway when ventilation through a facemask is difficult. Nevertheless, the LMA does not seal the trachea against entry of gastric contents as well as an endotracheal tube (ETT). Positive pressure ventilation may result in gastric inflation and increase the risk of regurgitation and aspiration during LMA use. There is, however, little evidence about the actual risk of pulmonary aspiration with an LMA vs. an ETT during mechanical ventilation. Criteria considered to contraindicate the use of an LMA are based primarily on opinion and not on evidence.

The overall incidence of pulmonary aspiration (all types of airway management) has been reported to be between 1 in 3,216 and 1 in 14,139 general anesthetics.

Methodology            This retrospective study examined data in a preexisting quality improvement database. Data had been collected over a 10+ year period. Records were included in the analysis if either an LMA or ETT was used during a general anesthetic with mechanical ventilation. The selection of an LMA for airway management was driven by locally accepted contraindications to LMA use including: patients who were not NPO, intestinal obstruction, pregnancy, procedures involving the airway, and prone position. An LMA was occasionally used when one or more of these contraindications were present in a difficult airway situation.

Default volume controlled ventilator settings were tidal volume 8 – 10 mL / kg with respiratory rate adjusted according to the end tidal CO2. Surgical procedures included major abdominal, urologic, gynecologic, retroperitoneal, and laparoscopic surgery. Pulmonary aspiration of gastric contents was defined as the presence of gastric contents or bilious fluid in tracheal aspirate; bilious fluid on the LMA or in the oropharynx; or postoperative dyspnea, hypoxia, or “auscultatory abnormalities.” When a case of aspiration was identified in the quality improvement database the original chart was reviewed to verify the data.

Result            Slightly over 1,000,000 cases were contained in the database. Of those, 65,712 met the criteria for inclusion; general anesthesia and mechanical ventilation with either an LMA or ETT. Of those, 2,517 were major abdominal surgery or laparoscopy performed with an LMA. The airway was managed with an LMA in 1.7% of cases in which a contraindication to LMA use was present.

Aspiration occurred in 10 cases. Of these, 4 occurred during an elective procedure; 2 with an LMA and 2 with an ETT. The other 6 occurred during a non-elective procedure; 1 with an LMA. Only 2 of these patients were admitted to intensive care due to aspiration related problems; 1 of the ICU admissions was an LMA patient and 1 was an ETT patient. The occurrence of pulmonary aspiration in patients whose airway was managed with an LMA and mechanical ventilation was no different than those whose airway was managed with an ETT (odds ratio 95% confidence interval 0.09-1.4; P=0.141). The power of the study (the probability of rejecting the hypothesis when it is false) was calculated after the fact to be 0.69.

The primary factor associated with pulmonary aspiration was emergency surgery, not the airway management device. The overall incidence of aspiration was 1 in 6,571 anesthetics.

Conclusion            The incidence of pulmonary aspiration was no greater when an LMA was used than when an ETT was used during general anesthesia with mechanical ventilation. Institutional contraindications to LMA use may have influenced the results.



Comment

The focus of this study was the risk of aspiration when mechanical ventilation was used with an LMA. Accepted contraindications to LMA use were observed. The study was not about the risk of using an LMA for any and every case. It was about the added risk (if there is any) of using mechanical ventilation with an LMA. I’m glad to see this question addressed. Many of my colleagues are hesitant to use a ventilator with an LMA, yet manually bagging with an LMA is fine. Both are positive pressure ventilation. If our concern is insufflating the stomach with air it can happen with either method of ventilation. In my view, it is all about technique. I sometimes use mechanical ventilation with an LMA but I don’t use the ventilator the same way I’d use it with an ETT. I take care, for example, to keep peak pressures below 20 cm H2O. This study indicated that there may not be any increased risk of aspiration when mechanical ventilation is used with an LMA (see limitations in the following paragraphs). While I’m not ready to use an LMA as widely as this group apparently did, I do believe that with proper technique mechanical ventilation can safely be used with an LMA.

While the study is retrospective, it is unlikely it could have been conducted any other way, and, as such, it is probably nearly the best we are going to get. As retrospective studies go, this was a good one. Most retrospective studies are not based upon a detailed quality improvement database. The authors also took the wise step of reviewing the charts of all patients identified by the QI database as having experienced pulmonary aspiration. And, it has the advantage of including a large number of patients, especially important for a relatively low frequency event such as aspiration. Still, few aspirations were detected and there may have been a true difference in the aspiration rate that was undetectable as a result. Based upon the detailed statistical tests they reported and, in my judgment, one error in analysis, I believe there is a reasonable chance that the actual incidence of pulmonary aspiration may have been greater with the LMA than with an ETT (see note following).

The investigators rightly point out that the QI database may not have detected all cases of aspiration because complications were reported to the database by the anesthesiologist caring for the patient. Self reporting tends to underestimate the incidence of a complication because the clinician unconsciously gives themself the benefit of the doubt and because reporting can get put off until later and forgotten.

While it has a couple limitations, this study is important and enlightening because it is generally well done and encompasses a very large group of patients. The institution where this study was conducted does a great service by taking the time and effort to maintain a detailed quality improvement (QI) database for over a decade. Without it, this study would not have been possible.

Michael Fiedler, PhD, CRNA

NOTE: The upper boundary of the adjusted odds ratio 95% CI of the risk of aspiration with an LMA was 5.62 times greater than an ETT. A one sided test was used to calculate the power of the study.

© Copyright 2009 Anesthesia Abstracts · Volume 3 Number 12, December 31, 2009