記住我
The utilization of cannabis for medical applications dates back some 10,000 years.1 In 1970, cannabis use was markedly reduced after being classified as a Schedule I substance by the US federal government.2 Since then, the social, political, and legal nature of marijuana has undergone numerous changes, particularly in favor of both medical and recreational use. To date, all but four states in the Unites States have legalized some form of cannabis or cannabis-related products. These changes have led to a notable number of orthopaedic patients using marijuana in the perioperative surgical setting.
The current literature surrounding cannabis use in the orthopaedic surgical and nonsurgical patient populations is heterogeneous. The objectives of this study were to compare patients with a completely negative drug screen (control) against those positive for marijuana only (tetrahydrocannabinol [THC+]) and patients positive for marijuana with other controlled substance (multidrug) after routine preoperative urine drug screening (UDS). The primary outcome of interest was postoperative opioid consumption in morphine milligram equivalents (MMEs). The secondary outcomes were 90-day emergency department (ED) visit, readmission rates, and length of stay (LOS).
MethodsInstitutional review board approval was obtained before initiation of the study with a waiver of informed consent per institutional protocol. A retrospective study of all patients from a single orthopaedic spine surgeon's (D.P.) practice undergoing elective spine surgery between January 2013 and January 2019 was collected for this study. All participants had initially presented to an ambulatory clinic before being indicated for surgical spine intervention. As part of the primary surgeon's standard practice, all patients were required to undergo routine UDS. Before UDS collection, patients were informed that the results of their drug screen would be used for quality improvement and would not change their clinical treatment plan or prevent them from receiving surgery. Patients excluded from the study were those undergoing spine surgery through an ED-admission encounter or those without a UDS. Three cohorts were developed: a control group, a THC-only group, and a THC with opioid group based on the UDS.
Data were extracted from the electronic health records: Epic Hyperspace and AthenaHealth. A chart review was conducted for ED visits and hospital admissions within 90 days of surgery discharge (ED-90 and RA-90, respectively) for all patients. Total MMEs within 90 days of surgery discharge (MME-90) was calculated by reviewing filled prescriptions confirmed through SureScripts.
Patient demographics including age, sex, body mass index, race, smoking status, and marital status were collected. In addition, surgical levels, presence of fusion constructs, LOS, ED-90, RA-90, MME-90, Charlson Comorbidity Index conditions, and spine surgical invasiveness index were also obtained. The spine surgical invasive index is a validated scoring system that factors the number of vertebral levels decompressed, fused, or instrumented as well as the surgical approach.3
Statistical analyses were subsequently conducted using the Anaconda (version 5.3.0; Anaconda) distribution of Python (version 3.6.6; Python Software Foundation, www.python.org), in addition to the PyCharm Professional 2020.2 Integrated Developer's Environment (JetBrains, Prague, Czech Republic). The following libraries were used in this study: pandas, numpy, scikit-learn, and their respective library dependencies. Shapiro-Wilks test was conducted to evaluate normality of continuous variables. Owing to non-normally distributed data, continuous variable outcomes were evaluated with a Kruskal-Wallis test while categorical variables were assessed using χ2 tests. Multivariable logistic regression was conducted to evaluate what factors affect ED-90 and RA-90, as well as a separate a multivariable linear regression to evaluate what factors may affect MME-90. Covariates were selected based on a P-value <0.10. Logistic regression coefficients were converted and reported as odds ratios. A P-value of <0.05 was deemed statistically significant.
ResultsA total of 353 patients met inclusion criteria, with the control group containing 252 patients, the THC-only–positive group containing 54 patients, and the multidrug-positive group containing 47 patients (Table 1). The overall opiate-positive rate was 13.31%, and the total THC-positive rate was 28.61%. Control patients were older (60.46 ± 15.04 years) compared with multidrug-positive patients (58.53 ± 10.95 years) and THC-only users (51.90 ± 15.12 years). 34.78% of multidrug patients while 14.68% of control patients and 20.37% of THC-only patients were current smokers (P < 0.01). Control patients were more likely to be former smokers at 31.35%, compared with 15.22% of multidrug-positive patients and 16.67% of THC-only users formerly smoked (P < 0.01). Over half of THC-only users (62.96%) were never smokers, and 50.00% of multidrug users were never smokers (P < 0.01). Only the rate of solid tumors was statistically different, with respect to the Charlson Comorbidity Index (Table 2). THC-only users trended toward lower rates of spinal fusion procedures than control patients and THC-positive and opioid-positive patients but remained nonsignificant (P = 0.05). Similarly, the average spine surgical invasive index scores were lower for THC-only patients and control patients compared with multidrug-positive patients, but was again not statistically significant (Table 3; P = 0.11).
Table 1 - Patient Demographics for the Control, THC-Only, and THC-Opioid–Positive Groups Factor ControlBMI = body mass index, THC = tetrahydrocannabinol
THC = tetrahydrocannabinol
THC = tetrahydrocannabinol
No significant difference was observed when evaluating the rate of 90-day ED visits, rate of 90-day readmissions, time to 90-day ED visits, and time to 90-day readmissions (Table 4). Differences in MME-90 days, however, were notable. The control group used 424 ± 627.77 MMEs, THC-only users used 680.23 ± 880.45 MMEs, and multidrug users used 2,543.51 ± 2,678.6 MMEs within 90 days after surgery discharge (P < 0.0001).
Table 4 - Postoperative Length of Stay, 90-Day Emergency Department and Re-admission Rates, and 90-Day Postoperative Morphine Milligram Equivalent Factor ControlED = emergency department, MME = morphine milligram equivalent
Linear regression demonstrated a significant increase in postoperative opioid prescriptions for multidrug users by 2087.10 MMEs (confidence interval, 1,717.89 to 2,456.32; P < 0.0001) as well as increasing age by 8.78 MMEs per year (0.27; confidence interval, 17.30; P <0.05, Table 5). THC-only users did not demonstrate a statistically significant increase in opioid prescriptions when compared with control subjects, although there was an increased trend of 306.24 MMEs (P = 0.09). Logistic regression evaluating with 90-day ED visits and readmissions demonstrated no difference among control subjects, THC-only patients, and multidrug users (P > 0.05, Tables 6 and 7). Increasing age did demonstrate an increasing 1.03 odds ratio (confidence interval, 1.00 to 1.06, P <0 .05) for readmission.
Table 5 - Postoperative Opioid Prescriptions (Morphine Milligram Equivalent) Factor Coefficient (±2 Standard Error) P Age 8.78 (0.27 to 17.30) <0.05 Smoking status Never Ref Ref Former −295.63 (−671.943 to 80.684) 0.12 Active −217.43 (−547.72 to 112.87) 0.20 Fusion No Ref Ref Yes 32.23 (−224.32 to 288.766) 0.81 Substance utilization Control Ref Ref THC only 306.24 (−42.12 to 654.59) 0.09 Multidrug 2,087.10 (1,717.89 to 2,456.32) <0.0001THC = tetrahydrocannabinol
THC = tetrahydrocannabinol
THC = tetrahydrocannabinol
The opioid epidemic has driven the medical community and in particular orthopaedics to find a safer option for pain control. Several studies have examined cannabis use and outcomes in general orthopaedics, nonsurgical pain control, arthroplasty, trauma, sports medicine, and spine surgery.2,4-7 The cannabis plant has 500 distinct compounds and 144 different cannabinoids.1 Delta-9-THC and cannabidiol (CBD) are the two main chemically active cannabinoids. There is a proposed synergistic effect between the two.6 Both cannabinoids interact with the endocannabinoid system, which is a separate system from the opiate pathway.8 THC is primarily associated with marijuana's physical and psychotropic effects while CBD has a more calming and anti-inflammatory effect. THC has a higher affinity for cannabinoid-receptor type 1, which is more concentrated in the central and peripheral nervous systems.2 CBD has a higher affinity for cannabinoid-receptor type 2, found in higher concentrations in the bone, muscle, and immune system.2,9 A complete understanding of the endocannabinoid system is still being researched, and there is not a clear understanding of the risk-benefit profile, particularly in chronic cannabis use. Multiple studies have identified cross-tolerances between cannabis, barbiturates, prostaglandins, chlorpromazine, and opioids among others.7 Others have shown that the endocannabinoid system has an important role in bone healing and bone homeostasis and that THC may have an inhibitory effect on bone metabolism.10
The effect of cannabis on orthopaedic outcomes is difficult to study because of marijuana's heterogeneity in intake, strain, amount, and frequency of THC and CBD. The current published literature is retrospective and recorded marijuana use is often self-reported—likely underestimating the actual amount of cannabis use.7 In 2021, Ryan et al evaluated a cohort of 423,978 spine patients using the 2012 to 2015 Nationwide Inpatient Sample database and reported 0.56% diagnosed with cannabis use disorder.7 While pragmatic in its study design, cannabis use disorder refers to an International Classification of Disease-10 code (F12.10) referring to a formal disorder that results in clinically notable impairment of the patient.7 Because not all cannabis utilizers are substantially impaired with day-to-day function, cannabis use disorder likely underestimates actual usage for in the United States, which other studies have reported to range from 6% to 15% of the total population.7 In contrast to prior studies, our cohort's THC and opioid usage was confirmed with UDS as part of a routine preoperative assessment. Our rates further validate a higher utilization of marijuana in the United States.
D'Antonio et al published a retrospective cohort study which showed that preoperative cannabis use increased the likelihood of spine revision surgery for any indication after lumbar fusion. The cohorts examined in their study only included THC-positive and negative; however, there was no statistically significant difference in perioperative opioid (both preoperative and postoperative) use between the two. They showed that although not statistically significant, the 90-day all-cause readmission rate and 90-day readmission rate for surgical intervention were higher in the cannabis-positive group. Trends were noted for THC-only patients to present earlier to the ED at 12.20 days, versus 24.16 or 26.30 days for control subjects and multidrug users, respectively. Similar findings were also noted for average 90-day readmissions at 23.25 days for THC-only participants, versus 32.12 and 35.43 days for control subjects and multidrug users, respectively. These patients may benefit from earlier postoperative follow-up to prevent unnecessary ED visits.
Although not statistically significant, multidrug patients were noted for a trend toward increased ED visits and readmissions. These results are underpowered and interpretation is limited. Review of the data does demonstrate a markedly higher rate of solid tumors, as well as a trend toward higher rates of COPD at 17.02% compared with 6.75% and 7.41% of control subjects and THC-only patients, respectively. It is suspected these patients are afflicted by more serious medical conditions which require additional medical care. Unlike our study, D'Antonio reported no statistically significant difference in MME usages between the two groups. Of note, the patient's cannabis use was self-reported in their study.
In slight contrast to the abovementioned study, Jakoi et al published a retrospective cohort study examining marijuana's effects on outcomes in patients who specifically underwent a transforaminal lumbar interbody fusion.4 The authors found that Oswestry Disability Index scores were not different and that there were similar rates of revision surgery and confirmed fusion. This, however, was only confirmed by using radiograph and not CT. LOS was less in the marijuana-negative group. Similar to our results, the authors did find a higher rate of concomitant tobacco use in cannabis users compared with non-users. Our study, unfortunately, did not study the fusion rates of patients because our goal of this study was to aid in the immediate postoperative period, pain control, emergency room visit, and emergency room admission. The interaction of THC and smoking is varied and can influence the fusion rate in patients undergoing arthrodesis and should be further investigated in a longer time frame study.
Most importantly, the results of our study again bring into question the potential interaction of the endocannabinoid system, opioids, and overall pain management in postoperative spine patients. This is highlighted particularly when comparing the spine surgical invasive index and postoperative opioid consumption between the cohorts. Our data do support this because 46.53% of THC-positive patients were also opioid-positive, which is higher than previous publications of 8.9% to 31.8%.11 The clinical relevance of this interaction is difficult to assess because of the high variability in the currently published data.12,13 More importantly, our data indicate that there may be two subpopulations of marijuana utilizers: marijuana only and those who are multidrug users. Those who only use marijuana seem to have a similar or slightly higher requirement for postoperative opioid pain requirements. Conversely, those who use other drugs such as opiates require markedly greater opioids and may indicate a greater propensity for substance abuse or higher level of medical comorbidity.
A limitation to this study is incorporating both fusion and non-fusion patients. It is well recognized that patients undergoing fusion typically require more aggressive pain regimens. We, therefore, included the surgical invasiveness score, which demonstrated no notable difference among the groups. Furthermore, the presence of fusion was included in the multivariable linear and logistic regressions, which demonstrated no notable alteration in the outcomes of interest. Inclusion of all patients also improves the study's generalizability. This facilitates the surgeon in identifying patients at increased risk of greater postoperative opioid pain medication requirements, while avoiding stigmatization of others.
A second limitation of this study is its retrospective nature. Third, patient's THC usages were binary: THC-negative or positive. The frequency or amount of cannabis used, administration route, strain, or THC/CBD concentrations were not quantified. The indication for THC use was not deciphered as medicinally or recreational. These factors may have an underlying effect not appreciated in our data. This, however, is a common limitation seen in most studies involving cannabis. Patients may have also started using cannabis only after surgery and, therefore, would have been grouped into the THC-negative group, despite using cannabis in the perioperative period. Fourth, it is possible that some postoperative patients who returned to the ED or who were readmitted could have presented to an outside institution and, therefore, would not have been captured in our data collection. However, this number would likely be small as our institution's electronic medical record connects 22 hospitals and over 300 outpatient facilities. Fifth, this was a single-surgeon, single-institution study, which limits the generalizability of data and its potential application at a national level. In addition, our cohort's overall usage of preoperative THC and opioids is on par with the current literature, albeit slightly higher.11 In addition, this study was conducted in a state where marijuana is legal for both medical and recreational use; therefore, our patient population may not represent orthopaedic patients in states where there are more strict usage laws. Sixth, the reasons for return ED visits and readmissions were not compared. It is possible that patient's perioperative hospital visits were unrelated to their surgery. However, given the limited amount of data regarding cannabis use and spine surgery, we believe that our study still provides valuable data. Seventh, MMEs were calculated based on prescriptions and not pill counts and are, therefore, an estimate. Thus, although patients filled their prescription, this does not mean they used the entire prescription of opioids. This is a common limitation for these studies.
Overall, based on our single-surgeon study, patients who are THC-positive before having surgery have higher opioid consumption compared with THC-negative patients, regardless of preoperative narcotic use. Although not statically significant, there is also a trend toward less emergency room presentation in THC-positive patients. More importantly, the use of THC increases the risk of more MMEs needed in the postoperative spine period, especially if those patients were on opiates as well together. The results of this study allow orthopaedic spine surgeons to inform their patient's about potential risks of perioperative THC usages.
References 1. Johal H, Vannabouathong C, Chang Y, Zhu M, Bhandari M: Medical cannabis for orthopaedic patients with chronic musculoskeletal pain: Does evidence support its use? Ther Adv Musculoskelet Dis 2020;12:1759720X20937968. 2. Kleeman-Forsthuber LT, Dennis DA, Jennings JM: Medicinal cannabis in orthopaedic practice. J Am Acad Orthop Surg 2020;28:268-277. 3. Cizik A, Lee M, Martin B, et al.: Using the spine surgical invasiveness index to identify risk of surgical site infection: A multivariate analysis. J Bone Joint Surg Am 2012;94:335-342. 4. Jakoi AM, Kirchner GJ, Kerbel YE, Iorio JA, Khalsa AS: The effects of marijuana use on lumbar spinal fusion. Spine (Phila Pa 1976) 2020;45:629-634. 5. D'Antonio ND, Lambrechts MJ, Heard JC, et al.: The effect of preoperative marijuana use on surgical outcomes, patient-reported outcomes, and opioid consumption following lumbar fusion. Glob Spine J 2022; July 18 [Epub ahead of print]. 6. Shah RM, Saklecha A, Patel AA, Divi SN: Analyzing the impact of cannabinoids on the treatment of spinal disorders. Curr Rev Musculoskelet Med 2022;15:133-142. 7. Chiu RG, Patel S, Siddiqui N, Nunna RS, Mehta AI: Cannabis abuse and perioperative complications following inpatient spine surgery in theUnited states. Spine (Phila Pa 1976) 2021;46:734-743. 8. Price RL, Charlot KV, Frieler S, Dettori JR, Oskouian R, Chapman JR: The efficacy of cannabis in reducing back pain: A systematic review. Glob Spine J 2022;12:343-352. 9. Chin G, Etiz BAF, Nelson AM, Lim PK, Scolaro JA: Knowledge and opinion on cannabinoids among orthopaedic traumatologists. JAAOS: Glob Res Rev 2021;5:e21.00047. 10. O'Connor CM, Anoushiravani AA, Adams C, Young JR, Richardson K, Rosenbaum AJ: Cannabinoid use in musculoskeletal illness: A review of the current evidence. Curr Rev Musculoskelet Med 2020;13:379-384. 11. Runner R, Luu A, Nassif N, et al.: Use of tetrahydrocannabinol and cannabidiol products in the perioperative period around primary unilateral total hip and knee arthroplasty. J Arthroplasty 2020;35:S138-S143. 12. Moon A, LeRoy T, Yacoubian V, Gedman M, Aidlen J, Rogerson A: Cannabis use is associated with increased use of prescription opioids following posterior lumbar spinal fusion surgery. Global Spine J 2024;14:204-210. 13. Renslo B, Greis A, Liu CS, Radakrishnan A, Ilyas AM, Ilyas A: Medical cannabis use reduces opioid prescriptions in patients with osteoarthritis. Cureus 2022;14:e21564.
留言 (0)