Cannabis has been legalized for medical and recreational use in 30 states of the United States and in Canada. Some researchers believe that the cannabis plant and its derivatives, referred to as cannabinoids, can play a role in treating many conditions including chronic pain,1–4 although definitive data are lacking.5–8 It is difficult to accurately assess the prevalence and trends of cannabinoid use among patient populations because of biases associated with self-reporting, high variability in terms of route/form and amount of consumption, and even social stigmas.9,10
Patients with advanced arthritis have chronic pain and may turn to cannabinoids for symptomatic relief prior to proceeding with joint replacement surgery. Studies on cannabinoid use in orthopedic patients have looked mainly through the lens of substance use disorder. The reported prevalence of preoperative cannabis use disorder in joint replacement patients has been based on coded diagnoses, resulting in rates of less than 1%.11–13 Due to the inherent limitations of database research, however, it is possible that such studies underestimate the prevalence of use, and none can assess whether cannabinoids were actively being used around the time of surgery.
At the authors' hospital, all patients undergoing total joint arthroplasty (TJA) are routinely tested with a urine toxicology (UTox) screening during the 30-day preoperative window. The authors have used this information to postpone surgery on patients testing positive for methamphetamines and cocaine, substances that have been shown to increase the risk of perioperative complications.14,15 Since cannabinoids became legal for medical use in the authors' state in 1996, a positive screen has not excluded patients from surgery at their institution. The authors sought to use this unique UTox screening dataset to assess the recent prevalence and trends of cannabinoid use relative to opioid use and also to identify demographic characteristics of patients who screened positive for cannabinoids to detect an association between preoperative cannabinoid use and postoperative morbidity.
Materials and Methods
Institutional review board approval was obtained for this retrospective study. The authors reviewed all patients who underwent primary total hip arthroplasty (THA) or total knee arthroplasty (TKA) for end-stage arthritis at their hospital from calendar years 2012 to 2017. Patients were excluded if they underwent unicompartmental knee arthroplasty (UKA), conversion from UKA to TKA, THA for acute trauma, or if no UTox screening was obtained within the 30 days prior to surgery.
The primary outcome was the result of the preoperative UTox screening for opioids and cannabinoids. Tetrahydrocannabinol is the only cannabinoid included in the UTox screening at the authors' institution. Secondary outcomes were characteristics of the cohort gathered from the patients' electronic medical records, including patient age at surgery, sex, American Society of Anesthesiologists (ASA) Physical Status Classification System score, disposition, all readmissions within 90 days, and any reoperation of the same joint. Veterans Affairs Surgical Quality Improvement Program (VASQIP) data also were retrieved for operations performed at the authors' institution between 2012 and 2017. Thirty-day complications and 90-day mortality captured at any Veterans Affairs hospital were recorded for the subset of patients sampled by this database.
Patients who underwent more than one TJA during the study period were recorded as separate data points, one for each operation. Where applicable, data are described as mean±SD. Chi-square tests for categorical variables and t tests for continuous variables were performed using Microsoft Excel, with two-tailed significance defined as alpha being .05. Cochran–Armitage tests for trend (also known as chi-square tests for trend) were performed using the EpiTools Epidemio-logical Calculator (AusVet). Multivariable regression analyses were performed using The R Project for Statistical Computing ( http://www.r-project.org).
A total of 1778 operations performed on 1519 patients met inclusion criteria for this study (Table 1). Mean age at surgery was 65.9±8.4 years for 1678 (94.4%) men and 100 (5.6%) women. There were 1161 (65.3%) TKAs and 617 (34.7%) THAs. Mean ASA score was 2.79±0.43. A total of 1443 (81.2%) patients were discharged to home, 172 (9.7%) were discharged to a skilled nursing facility, and 162 (9.1%) were discharged to an acute rehabilitation unit. One patient died prior to discharge. A total of 54 (3%) patients were readmitted within 90 days of primary TJA, and 93 (5.2%) patients required at least 1 reoperation on the same joint at any time point.
Urinary Toxicology Screening Results
The UTox screening was performed within 30 days prior to surgery in 97% of primary TJA surgery (1778 of 1833). A total of 201 (11.3%) patients screened positive for cannabinoids, and 412 (23.17%) patients screened positive for opioids. Of these, 64 (3.6%) patients were UTox positive for both cannabinoids and opioids. The UTox screening for both substances was negative in 1229 (69.1%) operations. Patients who used cannabinoids were more likely to use opioids than those who screened negative for cannabinoids (31.8% and 22.1%, respectively; P=.002). Similarly, patients who used opioids were more likely to use cannabinoids than those who screened negative for opioids (15.6% and 9.3%, respectively; P<.001). A greater proportion of patients undergoing THA were positive for cannabinoids (13.6%), opioids (28.2%), and both substances (5.2%) than those undergoing TKA (10.1%, 20.5%, and 2.8%, respectively) (P=.025, P<.001, and P=.009, respectively).
Between calendar years 2012 and 2017, there was a statistically significant increase in cannabinoid use (9.4% to 14.9%, P=.049) and a statistically significant decrease in opioid use (23.8% to 16.7%, P=.040) for all operations (Figure 1). Furthermore, chi-square testing of proportion trends (Cochran–Armitage test for trend) demonstrated a significant increase in cannabinoid use throughout all years (P=.012) and a significant decrease in opioid use (P=.005).
Trends in cannabinoid (A) and opioid (B) use from 2012 to 2017. A significantly different increase in cannabinoid use (P=.049) and decrease in opioid use (P=.040) occurred comparing 2012 with 2017. There also was a statistically significant trend (Cochran–Armitage test for trend) regarding increased cannabinoid use (P=.012) and decreased opioid use (P=.005) throughout the years reviewed.
Age and Sex
Patients who screened positive for cannabinoids were significantly younger than those who screened negative (mean age, 62.3 and 66.4 years, respectively; P<.001) (Table 2). Furthermore, patients who screened positive for only cannabinoids also were significantly younger than patients who screened positive for only opioids (mean age, 62.6 and 64.7 years, respectively; P=.007). Finally, patients taking both cannabinoids and opioids were younger than those who did not use either (mean, 61.3 and 66.9 years, respectively; P<.001). Limited by the small number of female patients who underwent TJA, there was no significant difference in cannabinoid or opioid use between males and females (P=.160 and P=.840, respectively).
Urine Toxicology Screening Results by Cohort
Mean ASA score was statistically significantly lower in patients who used cannabinoids compared with those who did not (2.72 and 2.79, respectively; P=.024). Cannabinoid users also had a lower mean ASA score than opioid users (2.72 and 2.86, respectively; P<.001), excluding patients who screened positive for both substances.
A total of 175 of 201 cannabinoid-positive patients were discharged home after TJA (87.1%) compared with 1268 of 1577 patients who did not use cannabinoids (80.4%) and 333 of 412 patients who used opioids (80.8%) (Table 2). Multivariable regression analyses controlling for age, sex, surgery type, and ASA score revealed no statistically significant associations between cannabinoid or opioid use and postoperative disposition.
Readmissions and Reoperations
Readmissions and reoperations documented in the institution's electronic medical record were evaluated (Table 2). Multivariable regression analyses controlling for demographic features did not identify any significant differences in 90-day re-admissions between cannabinoid users (2 readmissions, 1.00%) and nonusers (52 readmissions, 3.30%), or between cannabinoid users and opioid users (10 readmissions, 2.43%). Similarly, no significant differences were found for reoperations of the same joint at any time point for cannabinoid users (8 reoperations, 3.98%) vs nonusers (85 reoperations, 5.39%) and cannabinoid users vs opioid users (17 re-operations, 4.13%).
Of the 1778 operations, 1187 (66.7%) were sampled by the VASQIP database (Table 2). Of these cases, the number of 30-day complications reported was 22 (Table 3), and no 90-day mortality events were reported. Given the numbers available, the authors could not identify any significant differences using multivariable regression regarding complications between patients who screened positive and negative for cannabinoids (2 [1.49%] and 20 [1.90%] complications, respectively) and between patients who screened positive for cannabinoids vs opioids (2 [1.49%] and 4 [1.57%] complications, respectively).
Veterans Affairs Surgical Quality Improvement Program Database Complication Details
The prevalence of cannabinoid use in this patient population, as identified by preoperative UTox screening in the month prior to surgery, rose from 9% to 15% between 2012 and 2017. This compares with a prevalence of 0.28% to 0.7% previously reported in the orthopedic literature regarding cannabis-use disorder.12,13 It is possible that the authors' patient population is unique; however, no prior study has quantified the prevalence of cannabinoid use among TJA patients through actual laboratory testing prior to surgery; this is an important distinguishing factor for the current study.
Prior studies on the use of cannabinoids among orthopedic patients have relied on searching diagnosis codes in large administrative databases. This can lead to three potential sources of error for estimating substance use among TJA patients and for assessing effect on outcomes. First, some large databases, such as Medicare, only include patients older than 65 years, a cohort of patients found by the current authors to have a lower prevalence of cannabinoid use. Second, codes in administrative databases usually are based on disordered substance use. Use, which is detected by the UTox screening, does not necessarily equate with disorder, which is implied in International Classification of Diseases, Ninth Revision, codes such as cannabis dependence (304.3) and non-dependent cannabis abuse (305.2). Patients diagnosed with a disorder are likely to be a subset of users with other associated health and psychosocial problems. Finally, coded diagnoses can remain in the medical record indefinitely and do not always reflect active use. On the basis of their urine toxicology screening data, the authors believe cannabinoid use among TJA patients has been vastly underreported in the literature.
Interestingly, between 2012 and 2017, the authors found an approximately 60% increase in cannabinoid use and a 30% decrease in opioid use. In 2017, the number of patients using cannabinoids was nearly the same as those using opioids. With data from this observational study, the authors are unable to conclude whether these trends are related. It may be that attention directed toward the problems of opioid misuse would have led to a decrease in opioid use during this time period, independent of any affect from cannabinoid use. Alternatively, it is possible that some patients and treating physicians have substituted cannabinoids for opioids in the treatment of chronic pain.16–18 Demographic trends have shown cannabinoid use has increased in adults older than 50 years, which may support this possibility.19,20 If cannabinoids are being used in this way, it remains to be determined whether or not this is desirable.
The authors found cannabinoid users were significantly younger than both nonusers and opioid users. Patients using cannabinoids also had a statistically lower ASA score (mean, 2.72) than nonusers (mean, 2.79) and opioid users (mean, 2.86), although this may not be clinically meaningful. Notably, cannabinoid users in this study were significantly more likely to be taking opioids compared with patients who did not use cannabinoids. A recent survey-based study concluded cannabis use actually increased the risk of developing opioid use disorder.21 Prospective studies are needed to determine whether cannabinoid use is actually predictive of future opioid use.
Finally, the authors were unable to show that patients using cannabinoids had different rates of postoperative complications, readmissions, or mortality compared with nonusers. This study was not specifically designed or powered to test these associations, but the authors used all of the data available for the time interval of this study. The authors had a 97% UTox screening rate, and during a 6-year period, the authors evaluated the 1187 patients who were sampled by VASQIP. With these numbers, the authors had 80% power to detect differences in VASQIP complications, reoperations, 3-month readmissions, and deaths if the differences between cannabis users and noncannabis users were greater than 3.5%, 4.3%, 2.3%, and 4.6%, respectively. This puts an upper limit on the largest differences in complication, reoperation, re-admission, and death rates that may exist between these 2 populations.
Interestingly, other studies have found cannabis use results in a higher complication rate,11–13,22 but it may be that these studies selected only the most severe cannabis users through disorder diagnosis codes. It remains unresolved whether simple use of cannabinoids, rather than disuse, or route of administration are independent risk factors for complications or mortality following TJA.
There were several limitations to this study. First, the results are from a single Veterans Affairs Medical Center in California, a state that legalized medical cannabis in 1996 (although cannabinoids still are not available through the federal health care system). Although cannabinoids are more available in California than in other states, the number of states legalizing cannabis for medical or recreational use is growing, and relevance to states beyond California is expanding.
Second, this study was performed in the Veterans Affairs Health Care System, which serves a population of patients that is overwhelmingly male and has a different profile of medical social comorbidities than the broader population of THA and TKA patients. Some literature indicates that men are more likely to report cannabis use.23 However, a recent nationwide survey actually found that compared with nonveterans, cannabis use was similar, if not lower, among veterans.24 Although the exact rates of cannabinoid use found in the current study may be different from the broader TJA population, the trends in cannabis use observed in this patient cohort reflect national trends in cannabis use that have been reported for patients older than 50 years.19
Third, given UTox data alone, the authors were unable to determine the route, frequency, and quantity of cannabinoid or opioid use. Furthermore, given that cannabinoids are not regulated by the Food and Drug Administration, it is not possible to identify the formulations used by patients nor draw meaningful conclusions about how product variability impacts patterns of use, side effects, and drug interactions. The authors also could not identify whether cannabinoid use was recreational, medical, or both, and whether the cannabinoid was obtained illicitly or legally.
Finally, the laboratory at the authors' institution only screens for one cannabinoid, tetrahydrocannabinol, which is the primary psychoactive compound in cannabis. Given that there are hundreds of natural and synthetic cannabinoids, it is possible that this study underestimated the number of patients using cannabinoids at the time of surgery.25
In a veteran population undergoing TJA, preoperative UTox screening indicated that the use of cannabinoids TJA is greater than previously reported in the literature, that the prevalence is growing, and that opioid use has concurrently decreased. With the numbers available, preoperative cannabinoid use could not be associated with either complications or early implant failure. With this rising trend of cannabinoid use among patients, further study is needed to understand the effects of cannabis and its derivatives on the outcomes of TJA.
- Nugent SM, Morasco BJ, O'Neil ME, et al. The effects of cannabis among adults with chronic pain and an overview of general harms: a systematic review. Ann Intern Med. 2017;167(5):319–331. doi:10.7326/M17-0155 [CrossRef]. PMID:28806817
- Campbell G, Hall WD, Peacock A, et al. Effect of cannabis use in people with chronic non-cancer pain prescribed opioids: findings from a 4-year prospective cohort study. Lancet Public Health. 2018;3(7):e341–e350. doi:10.1016/S2468-2667(18)30110-5 [CrossRef]. PMID:29976328
- Manzanares J, Julian M, Carrascosa A. Role of the cannabinoid system in pain control and therapeutic implications for the management of acute and chronic pain episodes. Curr Neuropharmacol. 2006;4(3):239–257. doi:10.2174/157015906778019527 [CrossRef]. PMID:18615144
- Russo EB. Cannabinoids in the management of difficult to treat pain. Ther Clin Risk Manag. 2008;4(1):245–259. doi:10.2147/TCRM.S1928 [CrossRef]. PMID:18728714
- Hill KP, Palastro MD, Johnson B, Ditre JW. Cannabis and pain: a clinical review. Cannabis Cannabinoid Res. 2017;2(1):96–104. doi:10.1089/can.2017.0017 [CrossRef]. PMID:28861509
- Ware MA, Wang T, Shapiro S, et al. Smoked cannabis for chronic neuropathic pain: a randomized controlled trial. CMAJ. 2010;182(14):E694–E701. doi:10.1503/cmaj.091414 [CrossRef]. PMID:20805210
- Deshpande A, Mailis-Gagnon A, Zoheiry N, Lakha SF. Efficacy and adverse effects of medical marijuana for chronic noncancer pain: systematic review of randomized controlled trials. Can Fam Physician. 2015;61(8):e372–e381. PMID:26505059
- Volkow ND, Baler RD, Compton WM, Weiss SRB. Adverse health effects of marijuana use. N Engl J Med. 2014;370(23):2219–2227. doi:10.1056/NEJMra1402309 [CrossRef]. PMID:24897085
- Gray KM, Watson NL, Christie DK. Challenges in quantifying marijuana use. Am J Addict. 2009;18(2):178–179. doi:10.1080/10550490902772579 [CrossRef]. PMID:19283572
- Prince MA, Conner BT, Pearson MR. Quantifying cannabis: a field study of marijuana quantity estimation. Psychol Addict Behav. 2018;32(4):426–433. doi:10.1037/adb0000370 [CrossRef]. PMID:29771542
- Best MJ, Buller LT, Klika AK, Barsoum WK. Outcomes following primary total hip or knee arthroplasty in substance misusers. J Arthroplasty. 2015;30(7):1137–1141. doi:10.1016/j.arth.2015.01.052 [CrossRef]. PMID:25765129
- Moon AS, Smith W, Mullen S, et al. Marijuana use and mortality following orthopedic surgical procedures. Subst Abuse. 2019;40(3):378–382. doi:10.1080/08897077.2018.1449054 [CrossRef]. PMID:29558287
- Law TY, Kurowicki J, Rosas S, et al. Cannabis use increases risk for revision after total knee arthroplasty. J Long Term Eff Med Implants. 2018;28(2):125–130. doi:10.1615/JLongTermEffMedIm-plants.2018027401 [CrossRef]. PMID:30317962
- Tominaga GT, Garcia G, Dzierba A, Wong J. Toll of methamphetamine on the trauma system. Arch Surg. 2004;139(8):844–847. doi:10.1001/archsurg.139.8.844 [CrossRef]. PMID:15302693
- Nicholson Roberts T, Thompson JP. Illegal substances in anaesthetic and intensive care practices. Contin Educ Anaesth Crit Care Pain. 2013;13(2):42–46. doi:10.1093/bjaceaccp/mks050 [CrossRef]
- Choo EK, Feldstein Ewing SW, Lovejoy TI. Opioids out, cannabis in: negotiating the unknowns in patient care for chronic pain. JAMA. 2016;316(17):1763–1764. doi:10.1001/jama.2016.13677 [CrossRef]. PMID:27802551
- Reiman A, Welty M, Solomon P. Cannabis as a substitute for opioid-based pain medication: patient self-report. Cannabis Cannabinoid Res. 2017;2(1):160–166. doi:10.1089/can.2017.0012 [CrossRef]. PMID:28861516
- Lucas P, Walsh Z. Medical cannabis access, use, and substitution for prescription opioids and other substances: a survey of authorized medical cannabis patients. Int J Drug Policy. 2017;42:30–35. doi:10.1016/j.drug-po.2017.01.011 [CrossRef]. PMID:28189912
- Han BH, Sherman S, Mauro PM, Martins SS, Rotenberg J, Palamar JJ. Demographic trends among older cannabis users in the United States, 2006-13. Addiction. 2017;112(3): 516–525. doi:10.1111/add.13670 [CrossRef]. PMID:27767235
- Lloyd SL, Striley CW. Marijuana use among adults 50 years or older in the 21st century. Gerontol Geriatr Med. 2018;4:2333721418781668. doi:10.1177/2333721418781668 [CrossRef]. PMID:29977980
- Olfson M, Wall MM, Liu S-M, Blanco C. Cannabis use and risk of prescription opioid use disorder in the United States. Am J Psychiatry. 2018;175(1):47–53. doi:10.1176/appi.ajp.2017.17040413 [CrossRef]. PMID:28946762
- Roche M, Law TY, Sodhi N, et al. Incidence of drug abuse in revision total knee arthroplasty population. J Knee Surg. 2018;31(10):928–933. doi:10.1055/s-0038-1669915 [CrossRef]. PMID:30193389
- Cuttler C, Mischley LK, Sexton M. Sex differences in cannabis use and effects: a cross-sectional survey of cannabis users. Cannabis Cannabinoid Res. 2016;1(1):166–175. doi:10.1089/can.2016.0010 [CrossRef]. PMID:28861492
- Davis AK, Lin LA, Ilgen MA, Bohnert KM. Recent cannabis use among veterans in the United States: results from a national sample. Addict Behav. 2018;76:223–228. doi:10.1016/j.addbeh.2017.08.010 [CrossRef]. PMID:28858694
- Lafaye G, Karila L, Blecha L, Benyamina A. Cannabis, cannabinoids, and health. Dialogues Clin Neurosci. 2017;19(3):309–316. PMID:29302228
| TKA||1161 (65.3%)|
| THA||617 (34.7%)|
| Male||1678 (94.4%)|
| Female||100 (5.6%)|
|Age, mean (SD), y||65.9 (8.4)|
|ASA physical status score, No.|
| 1||5 (0.3%)|
| 2||378 (21.3%)|
| 3||1380 (77.6%)|
| 4||15 (0.8%)|
|UTox result, No.|
| Cannabinoid positive||201 (11.3%)|
| Opioid positive||412 (23.2%)|
| Both positive||64 (3.6%)|
| Negative||1229 (69.1%)|
| Home||1443 (81.2%)|
| SNF||172 (9.7%)|
| ARU||162 (9.1%)|
| Deceased||1 (0.06%)|
Urine Toxicology Screening Results by Cohort
|All (N=1778)||Cannabinoid positive (n=201)||Cannabinoid negative (n=1577)||Opioid positive (n=412)|
|Age, mean, y||65.9||62.3a||66.4||64.2|
|ASA physical status score, mean||2.79||2.72a||2.80||2.86|
| Home||1443 (81.20%)||175 (87.10%)||1268 (80.40%)||333 (80.80%)|
| SNF||172 (9.67%)||12 (5.97%)||160 (10.10%)||49 (11.90%)|
| ARU||162 (9.11%)||14 (6.70%)||148 (9.38%)||30 (7.28%)|
| Deceased||1 (0.06%)||0||1 (0.06%)||0|
|Reoperationsb (at any time point), No.||93 (5.23%)||8 (3.98%)||85 (5.39%)||17 (4.13%)|
|Readmissionsb (within 90 days), No.||54 (3.04%)||2 (0.10%)||52 (3.30%)||10 (2.42%)|
|VASQIP complications (n=1187), No.||22 (1.85%)||2 (1.49%)||20 (1.90%)||4 (1.57%)|
Veterans Affairs Surgical Quality Improvement Program Database Complication Details
|Deep surgical site infectiona||4|
|Superficial surgical site infection||4|
|Deep venous thrombosis||3|
|Urinary tract infection||3|
|Sepsis (unrelated to surgical site)||2|
|Wound disruption (without infection)||1|