Do the choices of airway affect the post-anesthetic occurrence of nausea after knee arthroplasty? A comparison between endotracheal tubes and laryngeal mask airways.

OBJECTIVES The primary goal of this study was to assess the impact of airway devices on the incidence of nausea after knee arthroplasty and their interaction with the use of nitrous oxide. METHODS Charts were reviewed for 499 patients after knee arthroplasty under general anesthesia. Demographic data, type of airway device, nitrous oxide, sevoflurane, desflurane, isoflurane, fentanyl, metoclopramide, ondansetron, dexamethasone, rocuronium and neostigmine were analyzed. Fisher's exact test was used to compare the categorical factors and t-test was used for continuous variables. Sinclair scores were used for post-operative nausea and vomiting (PONV) risk stratification. Multivariate logistic regression model was constructed to identify the factors contributing to the frequency of PONV. RESULTS PONV was documented in 10.3% of patients. Nitrous oxide was associated with a higher frequency of PONV than those received air mixture (12.5% vs. 8.7%, P < 0.01). Prior to risk stratification, the frequency of PONV was 17% in the endotracheal tube (ETT) vs. 6.7% in the laryngeal mask airway (LMA) group (P < 0.01). Sinclair score was 0.51 +/- 0.17 for the ETT group and 0.74 +/- 0.12 for the LMA group (P < 0.001). After risk stratification and matching, the incidence of PONV was 15.8% with the use of ETT compared with 7.9% for LMA (P < 0.05). CONCLUSION The frequency of PONV was almost twice with ETT as with LMA. Longer duration of anesthesia, neuromuscular blockade and non-standardized antiemetic regimen may have contributed to the increase PONV in ETT group. Prospective randomized studies are necessary to further explore whether and to which extend airway devices influence the incidence of PONV.