1 Introduction

The opioid crisis is still ongoing and contributes to significantly increased health care cost and patient burden.^1–4 Since the identification of the opioid crisis, research into diminishing opioid consumption has burgeoned, and findings have directed the orthopedic community to utilize a multimodal pain regimen to manage pain and decrease the need for opioids.^5,6 The orthopedic surgery community, despite accounting for 2.5% of all physicians, are attributed with writing 7.7% of all opioid prescriptions. This places orthopedic surgeons in an important position to impact the opioid crisis.⁷

The multimodal approach to postoperative pain management improves patient outcomes through decreased length of hospital stays after total joint arthroplasty and shows adequate pain control with diminished reliance on opioid medication.^5,6 As a result, multimodal pain management regimens are standard of care following joint arthroplasty.^6,8 Current literature has explored the utility of NSAIDs, acetaminophen/paracetamol, gabapentinoids, transcutaneous electrical stimulation, cryotherapy, regional and local anesthesia, cognitive strategies, and opioids for pain control and recovery following orthopedic surgeries.⁷ Despite research in orthopedic trauma and spine surgery associating benefit with the use of skeletal muscle relaxants, the use of these medications including cyclobenzaprine in decreasing opioid consumption following joint arthroplasty procedures remains unexplored.^9–12

The intent of this study is to determine the most relevant elements of a multimodal protocol in diminishing opioid utilization and promoting physical recovery following joint arthroplasty with a particular focus on cyclobenzaprine's efficacy as it related to postoperative narcotic pain medication.

2 Materials and methods

2.1

2.1 Structure and data collection

This was a retrospective cohort study in which 471 patients undergoing 554 orthopedic surgeries by a single surgeon at a tertiary care center were evaluated. Institutional Review Board approval was obtained. The initial eligibility criteria included patients undergoing elective joint arthroplasty via total knee arthroplasty (TKA), unicondylar knee arthroplasty (UKA), patellar arthroplasty with prosthesis (PAwP), total hip arthroplasty (THA), and hip hemiarthroplasty (HAH) between January 1, 2019, and April 15, 2021. Patients with a minimum of 12 week postoperative follow up were included in the study and had to be a minimum of 18 years of age at the time of surgery. Subject inclusion characteristics are detailed in Fig. 1.

Close

Fig. 1 Study flow chart.

Export to PPT

Patient age, sex, procedure, and BMI were noted, and patient comorbidities were consolidated to calculate a Charleston Comorbidity Index (CCI) score. Type of anesthesia (general/spinal without regional or general/spinal with regional) was noted. Preoperative opioid use status was determined and stratified to delineate oxycodone, hydrocodone, fentanyl, codeine, morphine, and tramadol use. Furthermore, preoperative benzodiazepine use was noted and stratified to delineate alprazolam, clonazepam, lorazepam, and diazepam use. Other pr-operative controlled substance use was determined and stratified to delineate marijuana, cocaine, methamphetamine, and testosterone use. Patients with a substance use disorder listed in the electronic medical record were recorded.

The use of general, spinal, and regional anesthesia during surgery was noted. Additionally, hospital-administered perioperative and post-operative hydrocodone, oxycodone, hydromorphone, fentanyl, morphine, and tramadol were noted. These values were standardized and summed to calculate a perioperative and postoperative morphine milliequivalents (MME) measurement.¹² The preoperative characteristics listed previously will all be referred to as “pertinent covariates” in future sections.

Postoperative pain was measured via a visual analog scale (VAS) measurement on postoperative day zero and postoperative day one including both pain at rest and with activity. Baseline preoperative ambulatory capacity was obtained using an activity measure for post-acute case (AMPAC) score. Additional AMPAC scores were obtained on postoperative day zero and postoperative day one, and the difference of these scores were calculated.

Patients were retrospectively separated into cohorts based on postoperative prescription of cyclobenzaprine. The decision to include cyclobenzaprine into the postoperative pain control regimen was at the discretion of the surgical team and generally was utilized in younger patients and those that had issues with pain control immediately after surgery in the post-anesthesia recovery room. In the cohort receiving cyclobenzaprine postoperatively, the total amount was calculated as a sum during the recovery period. In all patients, postoperative opioid use was noted, summed, and delineated as morphine, oxycodone, hydromorphone, tramadol, and codeine. Additionally, the number of prescription opioid refills were recorded. To standardize postoperative opioid consumption measurements, a postoperative MME value was calculated.

Recovery period range of motion was measured at 6 weeks and 12 weeks for patients undergoing TKA or UKA, and all patients were reviewed to note any hospital readmissions and reoperations on the surgical site of interest. Length of follow up was calculated from the date of surgery to last follow up day regarding the surgical site of interest.

3 Results

Analysis of preoperative demographics, shown in Table 1, yielded no significant difference between cohorts in sex, BMI, preoperative opioid use, preoperative controlled substance use, and preoperative benzodiazepine use. The cyclobenzaprine cohort was found to be significantly younger with a mean age of 58.69 (SD 8.79) verses a mean age of 63.96 (SD 9.43) in the group that did not receive the medication (p-value <0.0001). The no-cyclobenzaprine cohort was found to have a statistically significantly higher Charleston Comorbidity Index with a mean CCI of 2.85 (SD 1.66) compared to the cyclobenzaprine cohort with a mean CCI of 2.33 (SD 1.71). Analysis of operative demographics demonstrated no statistically significant differences between perioperative morphine milliequivalents, procedure type, procedure laterality or anesthesia.

Recovery outcome analysis, shown in Table 2, indicated the unadjusted mean pain on visual analog scale (VAS) at rest on day of surgery was 1.11 units higher among patients who took cyclobenzaprine medication compared to those who had not (95% CI (0.60, 1.62), p < 0.0001). The adjusted mean pain on visual analog scale (VAS) at rest on day of surgery was 1.08 units higher among patients who took cyclobenzaprine medication compared to those who had not (95% CI (0.59, 1.56), p < 0.0001 after adjusting for age, sex, BMI, CCI, perioperative MME, laterality, procedure, anesthesia, pre-operative opioid use, preoperative other controlled substance uses and preoperative benzodiazepine use. The adjusted mean pain on visual analog scale (VAS) when active on day of surgery was also 1.25 units higher among patients who took cyclobenzaprine medication compared to those who had not (95% CI (0.78, 1.72), p < 0.0001) after adjusting for pertinent covariates outlined in the methods section.

Analysis of the activity measure for postacute care (AMPAC) scores showed statistically insignificant differences in movement scores on postoperative day zero and postoperative day one in adjusted and unadjusted models. The adjusted model controlled for pertinent covariates, outlined previously, and baseline AMPAC values. Additionally, both cohorts had statistically insignificant differences in mobility score improvements from postoperative day zero to postoperative day one shown in “change in AMPAC score” analysis. Range of motion (ROM) was seen to be statistically insignificantly different in both cohorts at 6 and 12 weeks. Both cohorts had statistically insignificantly different incidences of readmission in <90 days. Additionally, both cohorts had statistically insignificantly different average lengths of stay following surgery.

Upon analysis of postoperative morphine milliequivalents (MMEs) between cohorts, an unadjusted model indicated patients who took cyclobenzaprine had a comparatively 29% higher geometric mean of post-operative MMEs (unadjusted mean ratio: 1.29, 95% CI (1.07, 1.55), p = 0.007). In adjusted models, patients who took cyclobenzaprine had statistically insignificant differences in postoperative morphine milliequivalents (base model mean ratio: 1.06, 95% CI (0.87,1.29), p = 0.5599) (full model mean ratio: 1.09, 95% CI (0.91,1.30), p = 0.3608).

4 Discussion

The goal of this study was to determine cyclobenzaprine's efficacy in diminishing opioid utilization and promoting physical recovery following joint arthroplasty as an adjunct to current multimodal recovery protocols in patients undergoing hip or knee arthroplasty. This study provides evidence that cyclobenzaprine could be useful in patients with higher rated postoperative pain immediately following joint arthroplasty.

Analysis of baseline preoperative and operative patient demographics – other than age and CCI – failing to show statistically significant differences between cohorts indicates that the patient cohorts were comparable overall. Notably, patients were of similar biologic demographics and had no significant difference in key confounding variables. The statistically significant difference in age between cohorts may be reflected by judicious use of cyclobenzaprine in elderly patients to avoid its deleterious side effect profile that may promote sedation and falls in this cohort. While the difference is statistically significant, the 5.27-year difference in average cohort age may not be clinically significant. While the CCI's between both cohorts were statistically significantly different, the clinical significance of a 0.52 difference in CCI is likely insignificant in the setting of this study. The no-cyclobenzaprine cohort had a mean age of between 60 and 69 years old when compared to the cyclobenzaprine cohort with a mean age between 50 and 59 years old. These age clusters correspond to an average one-point increase in CCI among the no-cyclobenzaprine cohort due to age alone. This supports that the difference in mean age caused the observed statistical difference in cohort CCIs which is likely not clinically significant.

Analysis of multivariate results indicates an initial increased VAS pain measurement in patients that were prescribed cyclobenzaprine. This difference may be explained by preferential prescription of cyclobenzaprine as an adjunct to recovery in patients experiencing increased postoperative pain. Following the recovery period of both cohorts, the cyclobenzaprine and no-cyclobenzaprine groups had no difference in mobility after surgery, range of motion at 6 and 12 weeks, or readmissions in <90 days. These values indicate noninferior recovery in the cyclobenzaprine cohort compared to the current standard of care. Patients that took cyclobenzaprine were found to have an increased unadjusted MME value indicating increased opioid utilization in this cohort. When adjusted for initial pain scores, this difference failed to reach statistical significance. Overall, similar recovery metrics between groups and opioid utilization similar when adjusted for initial pain scores cyclobenzaprine's non-inferiority to standard multimodal pain management and suggest that cyclobenzaprine may have utilization in patients with higher rated postoperative pain and has a role in the multimodal approach to postoperative pain following joint arthroplasty.

Limitations of this retrospective study design include bias in cyclobenzaprine administration. The cohorts were similar; however, differed in several aspects that could be controlled for in a prospective study design. Additionally, all patients received the current evidence-based multimodal recovery protocol; however, amounts of components other than opioids and cyclobenzaprine were not studied. In theory, a difference in amounts of other multimodal pain control agents may account for some of the differences seen in this study.

In conclusion, cyclobenzaprine was able to serve as complimentary medication for postoperative pain in a multimodal anesthetic approach after joint arthroplasty in patients with a higher baseline pain. Patients with a 1-point higher VAS pain score immediately after joint arthroplasty were prescribed this skeletal muscle relaxant resulting in no greater narcotic requirement or limitation in mobility compared to patients with lower baseline pain scores after surgery. Despite being limited by the retrospective nature of the study, this work demonstrates the potential use of cyclobenzaprine as an adjunct medication for patients undergoing joint arthroplasty.

Ethical statement

This study was approved by the Biomedical Institutional Review Board of The Ohio State University.

Funding statement

The authors received no funding for this study.

Guardian/patient's consent

Consent was not required by our institution for this retrospective study.

CRediT authorship contribution statement

Nicholas F. Turner: Conceptualization, Methodology, Data curation, Writing – original draft. Nicholas J. Greco: Investigation, Supervision, Writing – review & editing.