Research Article: Is obstructive sleep apnea associated with difficult airway? Evidence from a systematic review and meta-analysis of prospective and retrospective cohort studies

Date Published: October 4, 2018

Publisher: Public Library of Science

Author(s): Mahesh Nagappa, David T. Wong, Crispiana Cozowicz, Satya Krishna Ramachandran, Stavros G. Memtsoudis, Frances Chung, Shahrad Taheri.


Difficult airway management and obstructive sleep apnea may contribute to increased risk of perioperative morbidity and mortality. The objective of this systematic review and meta-analysis (SRMA) is to evaluate the evidence of a difficult airway being associated with obstructive sleep apnea (OSA) patients undergoing surgery.

The standard databases were searched from 1946 to April 2017 to identify the eligible articles. The studies which included adult surgical patients with either suspected or diagnosed obstructive sleep apnea must report at least one difficult airway event [either difficult intubation (DI), difficult mask ventilation (DMV), failed supraglottic airway insertion or difficult surgical airway] in sleep apnea and non-sleep apnea patients were included.

Overall, DI was 3.46-fold higher in the sleep apnea vs non-sleep apnea patients (OSA vs. non-OSA: 13.5% vs 2.5%; OR 3.46; 95% CI: 2.32–5.16, p <0.00001). DMV was 3.39-fold higher in the sleep apnea vs non-sleep apnea patients (OSA vs. non-OSA: 4.4% vs 1.1%; OR 3.39; 95% CI: 2.74–4.18, p <0.00001). Combined DI and DMV was 4.12-fold higher in the OSA vs. non-OSA patients (OSA vs. non-OSA: 1.1% vs 0.3%; OR 4.12; 95% CI: 2.93–5.79, p <0.00001). There was no significant difference in the supraglottic airway failure rates in the sleep apnea vs non-sleep apnea patients (OR: 1.34; 95% CI: 0.70–2.59; p = 0.38). Meta-regression to adjust for various subgroups and baseline confounding factors did not impact the final inference of our results. This SRMA found that patients with obstructive sleep apnea had a three to four-fold higher risk of difficult intubation or mask ventilation or both, when compared to non-sleep apnea patients.

Partial Text

Obstructive sleep apnea (OSA) is characterized by intermittent episodes of either complete or partial upper airway obstruction resulting in desaturation and recurrent arousal episodes from sleep. In the general population, the prevalence is 9–25%[1] with a higher prevalence in the bariatric surgical population.[2] Despite the strong association between OSA and adverse perioperative complications,[3–5] the majority of OSA cases remain undiagnosed and untreated at the time of surgery,[6] The difficult airway in OSA patients is considered to be a main contributing factor to the higher rate of adverse respiratory events.[7]

Fig 1 summarizes our strategy for literature search. The full search strategy used is shown in the “S2 File”. We identified 4,806 citations and 25 studies were retrieved for complete review and data extraction. Nine studies were excluded (S3 File: Excluded studies). Sixteen studies with a total of 266,603 patients (32,052 OSA vs. 234,551 non-OSA) and a variety of surgical procedures: head and neck, thoracic, abdominal, vascular, genitourinary and orthopedic surgeries were incorporated into the meta-analysis.[12–27] Several studies reported on more than one difficult airway events. Of the sixteen studies, twelve studies provided data on DI (total 19,581: 1,775 OSA vs. 17,806 non-OSA),[12,13,15,18–23,25–27] six on DMV (total 71,489: 5,129 OSA vs. 67,759 non-OSA),[16,18,19,22,24,25] two on combined DI and DMV (total 191,049: 26,361 OSA vs. 164,688 non-OSA),[17,22] and two on failed supraglottic airway (total 15,832: 662 OSA vs. 15,170 non-OSA).[14,21] No study showed data on OSA and surgical airway. Eleven studies were prospective[14,17–20,22–27] and five retrospective in nature.[12,13,15,16,21] The patient characteristics of all included articles are described in Table 1. The summary of clinical characteristics between the OSA and non-OSA group were compared in Table 2.

To date, this is the first systematic review and meta-analysis comparing the difficult airway between the OSA and non-OSA patients undergoing surgery. We found that difficult intubation, mask ventilation and both difficult intubation & mask ventilation was 3.4, 3.4 and 4.1-fold higher in OSA patients compared to non-OSA patients respectively. There was no significant difference in LMA failure rates in OSA vs non-OSA patients. Meta-regression analysis adjusting for baseline confounding factors and subgroup analysis did not substantially change results. No data is available in the literature on the relationship between OSA and surgical airway.