134. Physical Therapy on Postoperative Day Zero Following Cervical Spine Surgery Decreases Length of Stay

BACKGROUND CONTEXT Postoperative length of stay (LOS) is a major factor in overall cost for inpatient anterior cervical discectomy and fusion (ACDF), and its duration often depends on discharge clearance from physical therapy (PT). PURPOSE The study goal was to compare postoperative LOS for inpatient ACDF patients who initiated formal PT on postop day (POD) 0 versus POD 1. STUDY DESIGN/SETTING Retrospective cohort study. PATIENT SAMPLE The reviewed patient sample consisted of 3,003 patients who underwent elective, inpatient 1-2 level ACDF by 13 spine surgeons at one institution from 2011-2019. OUTCOME MEASURES The primary outcome measure was postoperative LOS, as measured in hours, for patients who initiated formal PT on POD 0 versus POD 1. Methods A retrospective review was conducted of 3,003 patients who underwent elective, inpatient 1-2 level ACDF by 13 spine surgeons at one institution from 2011-2019. Patients were categorized by timing of postoperative PT, with POD 0 defined as within 24 hours after surgery. Patients were excluded if they had incomplete data, admitted under a trauma visit, admitted to the ICU between surgery and discharge for observation or complications, did not see PT before discharge, or had qualifying surgery at >2 levels. Demographics, perioperative variables, and postoperative variables were collected. Statistical methods for categorical bivariate analysis included Chi-square test of independence and Fisher's Exact test, if Chi-square assumptions were not met. For continuous outcomes, normality was tested using a Shapiro-Wilk test and a two-sample t-test or Wilcoxon Rank Sum test was used to determine differences across POD groups. Data were analyzed using R version 4.0.2 with two-sided alpha <0.05 considered significant. Results Of 3,003 patients, 374 met inclusion criteria. Formal PT was initiated on POD 0 for 240 patients, and POD 1 for 134 patients. There was no significant difference between POD 0 and POD 1 groups regarding: preoperative myelopathy (p=0.51), age (p=0.98), gender (p=0.16), BMI (p=0.31), use of an assistive gait device preoperatively (0.17), or discharge destination (eg, home) (p=0.08). The difference in LOS (9.9 hours) was large enough to be of statistical and clinical significance (p<0.0001) (median 46.8 hours for POD 0 PT [interquartile (IQ) range, 25.7-75.4] vs 72.5 hours for POD 1 PT [IQ range 46.8-101.3], (p<0.0001). The difference in distribution of postoperative nights spent in the hospital was also clinically and statistically significant, as 51.8% of POD 0 patients discharged in two nights or less compared to 26.1% of POD 1 patients (p<0.0001). Conclusions Early initiation of PT has been proposed to facilitate discharge following cervical spine surgery. Our study demonstrated a clinically and statistically significant difference in postoperative LOS when PT was initiated on POD 0 versus POD 1. Earlier discharge may potentially lower care costs and improve value, given our institution's direct daily cost of $724 per patient-day for inpatient floor status. With the role of LOS in quality metrics and reimbursement, it remains important for spine surgeons to optimize postoperative LOS. A multidisciplinary approach which includes physical therapists, along with early formal postoperative PT, may facilitate discharge following ACDF. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs. Postoperative length of stay (LOS) is a major factor in overall cost for inpatient anterior cervical discectomy and fusion (ACDF), and its duration often depends on discharge clearance from physical therapy (PT). The study goal was to compare postoperative LOS for inpatient ACDF patients who initiated formal PT on postop day (POD) 0 versus POD 1. Retrospective cohort study. The reviewed patient sample consisted of 3,003 patients who underwent elective, inpatient 1-2 level ACDF by 13 spine surgeons at one institution from 2011-2019. The primary outcome measure was postoperative LOS, as measured in hours, for patients who initiated formal PT on POD 0 versus POD 1. A retrospective review was conducted of 3,003 patients who underwent elective, inpatient 1-2 level ACDF by 13 spine surgeons at one institution from 2011-2019. Patients were categorized by timing of postoperative PT, with POD 0 defined as within 24 hours after surgery. Patients were excluded if they had incomplete data, admitted under a trauma visit, admitted to the ICU between surgery and discharge for observation or complications, did not see PT before discharge, or had qualifying surgery at >2 levels. Demographics, perioperative variables, and postoperative variables were collected. Statistical methods for categorical bivariate analysis included Chi-square test of independence and Fisher's Exact test, if Chi-square assumptions were not met. For continuous outcomes, normality was tested using a Shapiro-Wilk test and a two-sample t-test or Wilcoxon Rank Sum test was used to determine differences across POD groups. Data were analyzed using R version 4.0.2 with two-sided alpha <0.05 considered significant. Of 3,003 patients, 374 met inclusion criteria. Formal PT was initiated on POD 0 for 240 patients, and POD 1 for 134 patients. There was no significant difference between POD 0 and POD 1 groups regarding: preoperative myelopathy (p=0.51), age (p=0.98), gender (p=0.16), BMI (p=0.31), use of an assistive gait device preoperatively (0.17), or discharge destination (eg, home) (p=0.08). The difference in LOS (9.9 hours) was large enough to be of statistical and clinical significance (p<0.0001) (median 46.8 hours for POD 0 PT [interquartile (IQ) range, 25.7-75.4] vs 72.5 hours for POD 1 PT [IQ range 46.8-101.3], (p<0.0001). The difference in distribution of postoperative nights spent in the hospital was also clinically and statistically significant, as 51.8% of POD 0 patients discharged in two nights or less compared to 26.1% of POD 1 patients (p<0.0001). Early initiation of PT has been proposed to facilitate discharge following cervical spine surgery. Our study demonstrated a clinically and statistically significant difference in postoperative LOS when PT was initiated on POD 0 versus POD 1. Earlier discharge may potentially lower care costs and improve value, given our institution's direct daily cost of $724 per patient-day for inpatient floor status. With the role of LOS in quality metrics and reimbursement, it remains important for spine surgeons to optimize postoperative LOS. A multidisciplinary approach which includes physical therapists, along with early formal postoperative PT, may facilitate discharge following ACDF.