Background:Degenerative disorders of the lumbar spinedecrease the mobility and quality of life of elderly patients.Lumbar fusion surgery is the primary methodof treating degenerative lumbar spine disorders; however, the surgical stress response associated with major surgery has been linked to pathophysiological changes in the elderly,resulting in undesirable postoperative morbidity, complications, pain, fatigue, and extended convalescence. In the present study, we aimed to determine whether enhanced recovery after surgery (ERAS) significantly improvedsatisfaction and outcomes in elderly patients(>65 years old) with short-level lumbar fusion.Methods:A total of 192 patients were included, 96 in the ERAS group and 96 case-matched patients in the non-ERAS group.Data including demographic, comorbidity and surgical informationwere collected from electronic medical records. ERAS interventionswere categorisedas preoperative,intraoperative and postoperative. We also evaluated primary outcome,surgicalcomplication, length of stay (LOS), postoperative pain scores and 30-day readmissionrates.Results:There were no statistically significant intergroup differences in regards to demographics, comorbidities、American Society of Anaesthesiologists (ASA) grade、or the number of fusion levels. There were also no differences between mean surgery time of intraoperative blood loss between the ERAS and non-ERAS groups. In addition, the mean preoperative Japanese Orthopaedic Association (JOA) score、Visual Analogue Score (VAS) for the back and legs and Oswestry Disability Index (ODI) score were not significantly different between the two groups. Overall, ERAS pathway compliance was 92.1%. There were no significant differences inthe number of complicationsor themortality rates between the ERAS and non-ERAS groups. Furthermore, the mean postoperative JOA score、VAS for the back and legs,ODI and readmission ratesscore revealed no significant differencesbetween the groups at 30-day- follow-up point. However, we observed a statistically significant decrease in LOS in the ERAS group (12.30±3.03 of ERAS group versus 15.50±1.88 in non-ERAS group, p=0).Multivariable linear regression showed thatcomorbidities (p=0.023) and implementation of ERAS program (p=0.002) were correlated with prolonged LOS.Multivariable logistic regression showed that no characteristics were associated with complications.Conclusions:This report describes the first ERAS protocol used in elderly patients after short-level lumbar fusion surgery. Our ERAS program is safe and could help decreases LOS in elderly patients with short-level lumbar fusion.