Postoperative Length Of Hospital Stay And One-year Mortality After Laryngeal Mask Airway Or Endotracheal Intubation Of General Anesthesia For Geriatric Hip Fracture

Background:The impact of laryngeal mask airway(LMA)or endotracheal intubation(ETT)under general anesthesia combined with peripheral nerve block on outcomes of elderly patients with hip fractures after surgical treatment remains an unknown issue in the clinical literature.Objective: To explore the effect of LMA or ETT under general anesthesia combined with peripheral nerve block on the postoperative length of hospital stay(LOS)and 1-year mortality among geriatrics after hip fracture repair compared to endotracheal intubation(ETT)anesthesia.Design: Single-center,retrospective cohort studyPatients: Consecutive patients with age 65 or over with hip fracture admittied to our hospital between January 1,2015 and December 31,2015 were included.Exposure: General anesthesia with LMA or ETT on top of peripheral nerve blocks.Outcomes: The primary outcome was the postoperative length of hospital stay(LOS),while the secondary outcome was 1-year mortality after surgery.Statistical analysis: Both univariable and multivariable analyses were performed to compare the effect of LMA and ETT on outcomes.Propensity score-matched(PSM)was used to control the preoperative condition between the two groups.To account for missing data and those lost to follow-ups,we performed general estimating equation(GEE)and multiple imputation analysis.Subgroup analyses were conducted in patients with different gender,different fracture types,different age groups,and those with different ASA classification,to further investigate the association of outcomes with those two interventions.Results:1.Of 978 patients in the whole cohort,80.8% received LMA(n=790)and 19.2%received ETT(n=188).2.In total,the median LOS was 4 days(IQR: 3-4 days)and 51.1%(n=355)patients had prolonged postoperative LOS(defined as postoperative LOS>3 days).When considered all confounding factors,our conditional logistic regression analysis showed that there was no significant difference in prolonged postoperative LOS between LMA and ETT: 49.2%(n=389)in LMA group vs 59.0%(n=111)in ETT group [OR,1.539;95%CI,0.938-1.969;p=0.105].After PSM was completed,we had 183 patients in each group.When intraoperative and postoperative confounding factors were adjusted by multivariable logistic regression model,we found no significant difference in prolonged postoperative LOS by the two airway devices: ETT group 59.6%(n = 109)vs LMA group 55.2%(n =101)[OR,1.173;95% CI,0.733-1.875;p = 0.506].Selecting different percentiles(10%,50%,75% and 90%)as cutoff points when defining prolonged postoperative LOS,then performing further multivariable logistic regression analysis,the results were consistent with our previous findings,manifesting that the two types of airway management,LMA and ETT,had no obvious correlation with prolonged postoperative LOS.Supplementary analyses yield prolonged postoperative LOS findings to be similar to our main analyses except for patients with ASA status 1-2(OR,1.655;95%CI,1.059-2.856;p=0.027).3.Among the total population,279 patients lost follow-up in the telephone visits,and the rate was 28.5%.Totally,36 patients died,for a mortality rate of 5.2%.The result of our GEE revealed that the difference in 1-year mortality was still insignificant between the two groups(OR,1.456;95%CI,0.632-3.357;p=0.378).After we conducted multiple imputation and performed further logistic regression,the type of airway management was not associated with the postoperative 1-year mortality rate(OR,0.827;95%CI,0.321-2.127;p=0.689).After the exclusion of those who lost follow-ups,LMA and ETT group respectively had 543(77.7%)and 156(22.3%)patients.Our conditional logistic regression analysis result showed that there was no significant difference in 1-year mortality between the two groups:4.8%(n=26)in LMA group compared with 6.4%(n=10)in ETT group(OR,1.457;95% CI,0.579-3.667;p=0.424).After propensity score matching(153 matched patients)and adjusting the intraoperative and postoperative confounding factors,the overall likelihood of 1-year mortality remained insignificant: 5.2%(n=8)in LMA group vs6.5%(n=10)in ETT group(OR,0.923;95% CI,0.257-3.318;p=0.902).what's more,oursubgroup analysis indicated that ETT was associated with a higher odds of 1-year mortality relative to LMA only when patients aged 80 and over(OR,3.874;95% CI,1.083-13.859;p=0.037).Conclusions: The two different types of airway management had little impact on prolonged postoperative LOS and 1-year mortality in geriatric patients who underwent hip fracture surgery under general anesthesia combined with proper peripheral nerve block.