Adoption of Enhanced Surgical Recovery (ESR) Protocol for Lumbar Fusion Decreases In-Hospital Postoperative Opioid Consumption

Abstract

Study Design:

Retrospective observational cohort.

Objectives:

We sought to evaluate the impact of ESR on in-hospital and 90-day postoperative opioid consumption, length of stay, urinary catheter removal and postoperative ambulation after lumbar fusion for degenerative conditions.

Methods:

We evaluated patients undergoing lumbar fusion surgery at a single, multi-surgeon center in the transition period prior to (N = 174) and after (N = 116) adoption of ESR, comparing in-hospital and 90-day postoperative opioid consumption. Regression analysis was used to control for confounders. Secondary analysis was preformed to evaluate the association between ESR and length of stay, urinary catheter removal and ambulation after surgery.

Results:

Mean age study participants was 52.6 years with 62 (47%) females. Demographic characteristics were similar between the Pre-ESR and ESR groups. ESR patients had better 3-month pain scores, ambulated earlier, had urinary catheters removed earlier and decreased in-hospital opioid consumption compared to Pre-ESR patients. There was no difference in 90-day opioid consumption between the 2 groups. Regression analysis showed that ESR was strongly associated with in-hospital opioid consumption, accounting for 30% of the variability in Morphine Milligram Equivalents (MME). In-hospital opioid consumption was also associated with preoperative pain scores, number of surgical levels, and insurance type (private vs government). Pre-op pain sores were associated with 90-day opioid consumption. Secondary analysis showed that ESR was associated with a shorter length of stay and earlier ambulation.

Conclusions:

This study showed ESR has the potential to improve recovery after lumbar fusion for degenerative conditions with reduced in-hospital opioid consumption and improved postoperative pain scores.

Introduction

The current opioid epidemic in the United States has necessitated a shift in perioperative protocols towards multimodal pain control and reduced reliance on opioid analgesia. In the United States the rate of opioid overdose has almost tripled and opioid prescriptions have nearly quadrupled over the last 2 decades. ¹ It is well documented in the literature the morbidity and mortality implications of chronic opioid uses, abuse, or dependence also carry a significant direct and indirect annual healthcare costs estimated to exceed $78.5 billion. ¹ Studies have shown that opioid-naive patients who receive opioid analgesia within 7 days of their surgery are 44% more likely to develop long term opioid usage. ² Substantial opioid regimen is often needed for sufficient postoperative analgesia after lumbar fusion surgery due to the pain inherent to the procedure.^3,4 Thus, the immediate postoperative period after lumbar fusions presents a considerable opportunity to improve patient outcomes with appropriate minimization of opioid analgesia.

The Enhanced Surgical Recovery (ESR) protocols are evidence-based multidisciplinary perioperative protocols integrating preoperative optimization, patient engagement, and multimodal pain management to reduce opioid consumption and improve postoperative outcomes. ⁵ Similar protocols were initially described in colorectal surgery and are currently implemented most general surgery subspecialties. ⁶ All ESR adaptations maintain the following core components: 1) preoperative counseling, 2) nutritional and medical optimization, 3) standardized pain management and anesthesia, and 4) early mobilization. ⁷ ESR implementation in multiple surgical subspecialties including orthopaedic total joint surgery have seen decreases in narcotic consumption, complications, and hospital length of stay.^6,8,9 There is emerging evidence for ESR implementation in spine surgery. Emerging ESR protocols in spine surgery focus on multi-disciplinary preoperative optimization, multimodal pain control, early oral intake, and early ambulation.^10-12 However, few studies have investigated the implementation of ESR protocols in lumbar fusion surgery. The aim of this study was to evaluate the effect of ESR implementation on postoperative opioid consumption. We hypothesized that ESR protocol implementation would decrease post-surgical opioid use.

Methods

Pre-ESR Protocol

Prior to ESR implementation our institution’s perioperative patient management aligned with standard practices without a standardized protocol. Preoperative counseling included medical and cardiology clearance as indicated with comorbidity optimization and smoking cessation discussions occurred at the discretion of the surgeon and patients’ primary care provider. Intraoperatively, anesthesia protocols and intraoperative fluid administration were determined by the anesthesia team. Pain management often included a combination of long-acting opioids, opioid patient-controlled analgesia (PCA), and IV opioid analgesia for breakthrough pain. Postoperative diets were advanced as tolerated after flatus, and foley catheters were removed after patients began ambulating. Patients were generally discharged with long and short-acting opioid medications and muscle relaxers.

ESR Protocol

Our institution developed an ESR protocol adapted for patients undergoing spinal surgery. Prior to a patient’s preoperative appointment, their comorbidities and nutritional status were optimized together with the surgical team, patient, and primary care provider. All patients were counseled on smoking cessation, proper nutrition, and what to expect throughout the perioperative period including the postoperative pain regimen. Patients taking opioid pain medications preoperatively were seen by pain management physicians to optimize their opioid medication regimen.

Prior to surgery, patients were instructed to consume 2 carbohydrate rich drinks one at 10:00 PM the night before surgery and one 4 hours prior to surgery. Patients could consume clear liquids until 2 hours prior to surgery. Opioid naïve patients were administered the following preoperative medications by the preop holding nurse: 600 mg PO gabapentin, 1000 mg PO acetaminophen, 200 mg PO celecoxib, 750 mg PO methocarbamol, 15 mg PO extended-release morphine, 4 mg PO ondansetron, and scopolamine patch. The only difference in preoperative medications for opioid tolerant patients was the dose of extended-release oral morphine, calculated at 1.5 times total daily MME dose, which was often guided by the pain specialist preoperatively.

Intraoperatively, anesthesia placed transversus abdominis plane (TAP) blocks on all anterior approaches. The surgeon-initiated time-out included anticipated estimated blood loss (EBL), tranexamic acid (TXA), antibiotics, and local anesthetics if indicated. 0.25% bupivacaine with epinephrine was injected into local subcutaneous and intramuscular tissues on all cases without TAP blocks at the maximal dose as determined by the anesthesiologist. In all cases patient core body temperature was maintained above 35°C. 2 g IV magnesium bolus and on “open cases” with substantial possible, blood loss. 10 mg/kg TXA bolus were administered intraoperatively; TXA was not used on MIS or percutaneous cases with minimal expected blood loss. TXA bolus was given in posterior cases with an expected EBL greater than 200 cc. TXA was not administered if patients had a history of stroke, deep vein thrombosis (DVT), or another contraindication. Anesthesia with up to one half monitored anesthesia care (MAC) was utilized during somatosensory evoke potentials (SSEPs). Total intravenous anesthesia (TIVA) was used as an option to obtain better neuromonitoring data when needed. Intraoperative IV lidocaine and ketamine drips were implemented as anesthesia standard of care unless contraindicated. Throughout the operation anesthesia utilized non-invasive continuous hemodynamic monitoring with ClearSight EV-1000™ system (Edwards Lifesciences, Irvine CA, USA) to guide goal-directed fluid resuscitation with lactated ringers and maintain euvolemia. The system utilizes a non-invasive finger cuff to continuously monitor cardiac output (CO), stroke volume (SV), stroke volume variation (SVV), systemic vascular resistance (SVR), and mean arterial pressure (MAP).

Postoperatively multimodal analgesic protocol is initiated in post anesthesia recovery unit (PACU) including the following: 300 mg PO gabapentin, 1000 mg PO acetaminophen, 750 mg methocarbamol, and 20 mg famotidine. Patients were additionally prescribed 4 mg IV ondansetron as needed for nausea and PO oxycodone every 6 hours as needed for pain based on visual analog pain score (VAS) and preoperative opioid used. Opioid naïve patients were prescribed oxycodone 5 mg for 4-6 out of 10 pain, oxycodone 10 mg for 7-10 out of 10 pain, and 4 mg PO hydromorphone for breakthrough pain. Opioid tolerant patients were prescribed oxycodone 5 mg for 1-3 out of 10 pain, oxycodone 10 mg for 4-6 out of 10 pain, oxycodone 15 mg for 7-10 out of 10 pain, and 8 mg PO hydromorphone for breakthrough pain. IV opioid PCA was not included in the protocol. All protocol medications and doses were reviewed with the patient’s history and allergies and modified or held when appropriate based on the physician’s best judgement.

On the surgical floor patients continued their multimodal analgesic protocol with routine dosing of the multimodal medicine routine every 8 hours, incentive spirometry 10 times every hour, and early ambulation beginning with walking from the door of the room to their bed upon arrival to the floor. Patients’ vital signs were monitored every 4 hours, and ins and outs were tracked every shift. Foley catheters were discontinued by postoperative day (POD) one. Patients’ diets were advanced proactively as tolerated and patients were asked to ambulate 3 times on POD zero and consume meals out of bed in a chair beginning on day of surgery. Patients with poor intake were given protein shakes to supplement diet. All patients were prescribed prophylactic stool softeners.

Patient Selection

Consecutive patients undergoing lumbar fusion surgery for degenerative spine conditions between April 2017 and April 2019 were included. All patients undergoing fusions of 4 levels or less with a minimum of 90 day postoperative follow up were eligible for inclusion. Patients with following conditions: spinal deformity, pregnancy, malignancy, paralysis, or muscular dystrophy were not eligible for inclusion. All patients or their legal representatives signed written informed consent, included in the Notice of Privacy Practices, for study enrollment and publication prior to surgery.

Study Design

From a single, multi-surgeon (6 surgeons) center, we evaluated patients undergoing primary lumbar fusion surgery for degenerative conditions in the transition period prior to (Pre-ESR, N = 174) and after (ESR, N = 116) adoption of ESR. All lumbar fusion surgeries were performed as staged or same day anterior lumbar interbody fusion (ALIF) and posterior instrumented fusion (PIF). Standard demographic and surgical data was collected as well as preoperative VAS scores, length of hospital stay, and days to removal of the urinary catheter and ambulation. In-hospital and 90-day postoperative opioid consumption was also collected and converted to MMEs. ¹³

All statistical analysis was performed using IBM SPSS Statistics for Windows, Version 26.0 (Armonk, NY). Pre-ESR and ESR patients were compared using unpaired t-tests for continuous variables and Fishers Exact test for categorical variables with a P-value threshold set at 0.05 to be considered statistically significant. In addition, regression analysis was used to control for confounders including age, gender, pre-op VAS, number of surgical levels, surgical approach, staged vs same-day surgery, insurance type and pre-op opioid use to determine the impact of ESR on in-hospital and 90-day postoperative opioid consumption.

Results

The Pre-ESR and ESR cohorts were similar in terms of age, gender distribution, smoking status, and body mass index (Table 1). There was a greater proportion of patients in the ESR group who had private insurance (81%) compared to the Pre-ESR group (68%, P = 0.013). Preoperative pain scores were statistically significantly lower in the ESR group (5.47) compared to the Pre-ESR patients (6.04, P = 0.041).

Table 1.

Pre-ESR vs. ESR Cohort Comparison.

	Pre-ESR	ESR	P-value
N	174	116
Age, years, Mean (SD)	54.56 (15.31)	54.63 (13.05)	0.969
Female, N (%)	90 (52%)	58 (50%)	0.774
Smoker, N (%)	12 (7%)	15 (14%)	0.127
BMI, kg/m2, Mean (SD)	29.04 (5.66)	29.53 (5.40)	0.462
Insurance Type, N (%)			0.013
Private	118 (68%)	94 (81%)
Public	56 (32%)	22 (19%)
Pain Scores (0 to 10), Mean (SD)
Preoperative	6.04 (2.12)	5.47 (2.42)	0.041
Immediate Post-op	4.70 (2.51)	4.52 (2.29)	0.518
90-days Post-op	3.57 (2.51)	2.89 (2.21)	0.021
Same Day vs Staged, N (%)			0.013
Same-Day	46 (26%)	16 (14%)
Staged	128 (74%)	100 (86%)
Number of Surgical Levels, Mean (SD)
Anterior	1.39 (0.78)	1.47 (0.68)	0.303
Posterior	1.49 (0.87)	1.62 (0.72)	0.179
Total Estimated Blood Loss, mL, Mean (SD)	129.34 (183.98)	128.86 (193.50)	0.983
Operative time, min, Mean (SD)	224.87 (90.98)	267.09 (92.29)	0.000
Length of Stay, days, Mean (SD)	3.91 (1.68)	3.69 (1.44)	0.228
Postoperative day catheter removed, days, Mean (SD)	1.44 (0.98)	1.14 (0.60)	0.002
Postoperative day Ambulated, days, Mean (SD)	0.84 (0.57)	0.39 (0.54)	0.000
Total In-Hospital MME, Mean (SD)	781.26 (718.89)	374.43 (298.13)	0.000
90 Day Global MME, Mean (SD)	3764.26 (6021.24)	3869.40 (5007.72)	0.872

The number of surgical levels was similar between the 2 groups anteriorly and posteriorly. A greater proportion of patients in the ESR group had a staged procedure (86%) compared to the Pre-ESR group (74%, P = 0.013) and the ESR group had a longer operative time (267.09 mins) compared to the Pre-ESR group (224.87 mins, P = 0.000). Estimated blood loss was similar between the 2 groups.

Patients on ESR ambulated earlier (0.39 vs 0.84, P = 0.000) and had their urinary catheter removed earlier (1.14 vs 1.44, P = 0.002) compared to Pre-ESR patients. Although the in-hospital opioid consumption was more than 50% lower in the ESR group (374.43 MMEs) compared to the Pre-ESR group (781.25 MMEs, P = 0.000), the 90-day opioid consumption was similar between the 2 groups.

Multivariable regression analysis was performed to control for the differences between the 2 groups and showed that greater in-hospital opioid consumption was seen in patients in 1) the Pre-ESR group, 2) undergoing staged procedures, 3) on government insurance, and 4) with worse pre-op pain scores (Table 2). The only variable associated with 90-day opioid consumption was the preoperative pain score. Gender, age, BMI, and staged versus same day were not associated with either in-hospital or 90-day opioid consumption.

Table 2.

Summary of Multivariable Regression Analysis.

	Beta	Sig.
Dependent Variable: Total opioid consumption (MME) in-hospital	Beta	P-value
ESR	-0.359	0.000
Staged	0.208	0.000
Government insurance	-0.180	0.001
Preoperative Pain score	0.176	0.001
ESR	-0.359	0.000
Dependent Variable: Total 90-day opioid consumption
Preoperative Pain score	0.184	0.002

Discussion

This study was designed to evaluate the impact of ESR implementation on in-hospital and 90-day postoperative opioid consumption. Our practice developed and implemented an ESR protocol tailored to patients undergoing lumbar fusion surgery for degenerative conditions. Our study found a greater than 50% decreased in-hospital opioid consumption in ESR group (374.43 MMEs) compared to the Pre-ESR group (781.25 MMEs, P = 0.000). In the last 2 decades United States has seen an increase in the morbidity and mortality associated with opioid consumption and a concurrent rise in opioid prescriptions. ¹ Studies have shown the majority of patients who develop opioid use disorders report their first exposure to opioids occurs in the perioperative perioid.^1-4 These trends have driven efforts to reduce postoperative opioid consumption and reduce morbidity and mortality in our postoperative patients. Prior studies in several general surgery subspecialties have found that ESR implementation was associated with significant reductions opioid consumption during hospital stays and the days immediately following surgery^14-19; however few studies have evaluated its implementation in spine surgery. Grasu et al. reported preliminary results of ESR implementation in oncologic spine surgery at a single center. They found a trend towards reduced opioid consumption in Pre-Enhanced Recovery After Spine Surgery (ERSS) group compared to ERSS group. ¹⁰ Another study conducted by Brusko et al. demonstrated a statically significant reduction in postoperative in-hospital opioid consumption with ESR implementation for 1- to 3- level lumbar fusions with a small sample size (N = 97). ¹² The current study seeks to expand upon these finding with a larger sample size and analysis of opioid consumption during the 90-day postoperative. A pilot study by Ali et al. found a significant reduction in intravenous opioid consumption and a smaller proportion of patients using opioids at 1 month postoperatively after ESR implementation for elective spine and peripheral nerve surgeries. ²⁰ Two additional studies from earlier this year by Kerolus et al. and Adeyemo et al. found similar reduction in LOS, opioid consumption, and readmission rates with ESR protocol implementation.^21,22 Kerolus et al. found patients undergoing single level MIS TLIFs consumed an average of 252.74 MMEs during their admission before ERAS implementation compared to 455.91 MMEs prior to ERAS (P = 0.001). ²¹ Adeyemo et al. found similar reductions in opioid consumption during the immediate postoperative period after open thoracolumbar fusion for spinal deformity. ²² Our study suggests these reductions in immediate postoperative opioid consumption are also applicable in thoracolumbar fusion surgery for degenerative conditions. The current study found the ESR group had significantly decreased in-hospital opioid consumption, supporting the findings of the current literature. The current study found 90-day opioid consumption was similar in Pre-ESR and ESR groups. 90-day opioid consumption was estimated based upon total prescribed MMEs during this period. To the authors knowledge no previous studies have evaluated opioid consumption after hospital discharge in spine surgery. Two studies in colorectal surgery found patients were prescribed more MMEs at discharge after the initial implementation of ESR protocols compared to similar patients prior to implemetation.^23,24

The results of the current study also demonstrated reduced time to ambulation (0.39 days vs 0.84, P = 0.000) and urinary catheter duration (1.14 days vs 1.44, P = 0.002) in the ESR group compared to the pre-ESR group. Ali et al.’s pilot study found similar reductions in time to ambulation in spine surgery patients with ESR implemention. ²⁰ Decreased urinary catheter duration with ESR implementation was also consistent with the finding of similar studies implementing enhanced recovery protocols in spine surgery.^11,25,26 ESR has the potential to mitigate discharge barriers with early catheter removal and encouraging early ambulation and to decrease risk of catheter associated urinary tract infections (CAUTIs).

ESR and Pre-ESR patients were similar in terms of basic demographic, preoperative and surgical data. However, preoperative pain scores were lower in the ERS group 5.47 compared to 6.04 (P = 0.041). The minimum clinically important difference (MCID) determines the smallest change in patient reported outcomes. The MCID threshold for VAS back pain is 1.2. ²⁷ A 0.57 difference in VAS pain score may not be clinically relevant as such a small difference is likely unperceivable to the patient. A larger proportion of patients in the ESR group had a staged procedure (86%) compared to the Pre-ESR group (74%, P = 0.013). This may have accounted for the subsequent longer operative time in the ESR group (267.09 mins) compared to the Pre-ESR group (224.87 mins, P = 0.000). EBL and the number of surgical levels were similar between the 2 groups.

To our knowledge, the present study is the only one of its kind to examine consecutive perioperative cases during ESR implementation after thoracolumbar fusion for degenerative conditions. Additionally, we believe this study is the largest single center multi-surgeon analysis of ESR in spine surgery in the current literature. This study was limited its retrospective design and single multi-surgeon center study population. Our 90-day postoperative opioid consumption data was limited as it estimated actual consumption based on total prescriptions filled during the 90 days, and this clearly may not reflect the numbers of doses taken during this period. We plan to evaluate 90-day postoperative opioid consumption in ESR patients on a more granular level in future studies for this patient population.

Conclusions

This study demonstrated a greater than 50% decreased in-hospital opioid consumption, earlier ambulation, and reduced urinary catheter duration with ESR implementation in lumbar fusion patients. These findings highlight the potential of ESR to effectively improve outcomes in perioperative patient care. Our results indicate many of the beneficial outcomes of ESR implementation seen in other surgical fields are transferable to lumbar fusions. However continued evaluation and further adaption are needed leverage its full benefits in spine surgery. The authors are evaluating 90-day postoperative opioid consumption in ESR patients on a more granular level in an ongoing study.