Buprenorphine has been used for opioid detoxification, addiction therapy, as well as acute and chronic pain management since 2002.1 It has unique properties that make it a distinct option in chronic pain and opioid-replacement therapy. It is a partial Mu agonist with high Mu receptor affinity, slow dissociation from the Mu receptor, and demonstrates kappa antagonist properties.2 Buprenorphine exhibits 25-50 times the potency and a partition coefficient 1,000 times that of morphine.3 In human and animal models, buprenorphine has been shown to produce full analgesic effect depending on the intensity of the stimulus.4 Intravenous buprenorphine (0.2 mg·kg−1 and 0.4 mg·kg−1) reduced experimental pain with a ceiling effect on ventilation.5 Dahan explained this by suggesting that buprenorphine acts as a partial agonist at the Mu opioid receptors involved in respiratory depression while simultaneously acting as a full agonist at opioid receptors in analgesic pathways. Other theories have suggested the partial agonist effect on respiratory depression due to a predominantly spinal mechanism of action.5

Suboxone® (Indivior Inc., Richmond, NJ, USA) is an abuse-deterrent sublingual formulation of buprenorphine and naloxone indicated for the treatment of opioid use disorder (OUD) and used off-label for management of chronic pain and opioid withdrawal. Buprenorphine’s wide safety profile and purported full agonist effects for analgesia have made it increasingly prescribed in patients with chronic pain and addiction, with wide-ranging reported success.3 Table 1 describes the different formulations of buprenorphine available in Canada.

Table 1 Buprenorphine: formulations available and pending distribution in Canada
Full size table

Budd6 demonstrated the analgesic efficacy of intravenous buprenorphine in 50 patients recovering from elective Cesarean delivery under general anesthetic. All patients were demonstrated to receive complete analgesia with 0.4-7 mg of buprenorphine. All of these factors make buprenorphine an important alternative in the management of pain among patients with complex pain and addiction history.

Indeed, one of the toughest challenges in pain medicine given the advent of the new Centers for Disease Control and Canadian Opioid Guidelines is the management of patients with opioid doses well beyond the suggested safe dose of 90 mg·day−1 of daily morphine equivalents. Suboxone has been used to rotate these patients off high-dose and misuse-prone opioids with a reduction in their pain symptoms as well as excellent success rates in the context of a behavioural support program (acceptance and commitment therapy).7 Additional formulations (Table 1) of buprenorphine indicated for the management of addiction are due to be released in Canada, substantiating the need for more research on this topic.

Several narrative reviews have summarized existing literature and include expert recommendations for perioperative management of patients taking buprenorphine.1,3,8,9,10,11,12 We conducted a systematic review to summarize the following features regarding the perioperative management of buprenorphine: 1) the indication for buprenorphine use; 2) the preparation and preoperative dose of buprenorphine; 3) whether buprenorphine therapy was continued perioperatively; 4) analgesic outcomes; 5) adverse events; and 6) success of deterrence against opioid use.

Methods

A literature search was performed (by M.E.) involving Medline (1946–June Week #1 2017), Medline In-Process/ePubs (June 13, 2017), Embase (1947–June 13, 2017), Cochrane Central (April 2017), Cochrane Database of Systematic Reviews (2005–June 9, 2017), PsycINFO (1806–June Week #1, 2017) (all via the Ovid search interface), Web of Science (1900–June 13, 2017) (Clarivate), Scopus (1960–June 2017) (Elsevier), CINAHL (1982–June 2017) (EbscoHost), and PubMed (1946–June 14, 2017, excluding Medline records) (NLM), to identify studies in which patients using buprenorphine and undergoing a surgical procedure were studied in the perioperative period. Searching was completed on June 14, 2017. Search terminology included blocks of terms for (surgery or perioperative or preoperative or postoperative terms) + (buprenorphine) + (drug use or chronic pain). The search strategies were not limited by study type. We supplemented the results with searches for book chapters, theses/dissertations, and ongoing clinical trials. Searches were limited to human studies. The retrieved citations were imported to Endnote™ (Clarivate Analytics) and checked for duplicates. The search strategy is further outlined in Appendix 1.

Eligibility criteria, study selection, and data extraction

We used the following inclusion criteria in our review:

Types of studies

All relevant studies, including randomized controlled trials (RCTs), observational studies including case control studies and cohort studies, as well as case series and case reports (CR). Conference proceedings were removed from the search strategy.

Participants

All human participants with no age restrictions, being administered buprenorphine prior to surgery for either chronic pain or addiction management.

Intervention

All reports involving the perioperative management of patients taking buprenorphine in the preoperative period to manage addiction or pain were included. Reports where patients were administered buprenorphine solely in the intraoperative or solely in the postoperative period were excluded. Any cases in which buprenorphine was continued beyond 12 hr postoperatively were considered to be in the buprenorphine “continued” category. Other information about descriptors of management was collected where possible, including: 1) rationale for preoperative buprenorphine use; 2) maintenance dose and preparation of buprenorphine; and 3) mode of continuation or discontinuation depending on management strategy suggested by the author.

Outcomes

The review did not aim to pool data, hence the outcomes are reported in Table 2 and important results are summarized. Relative effectiveness of either continuing buprenorphine or stopping buprenorphine in the perioperative period was reported as either proportion of patients with successful outcomes, or as mean scores with standard deviation. Successful outcomes included cases in which the authors did not highlight any complications (i.e., respiratory depression, apnea). Complications were noted to either be surgically related or due to other factors, including management of pain. Pain parameters and other patient reported outcomes were noted when available. Long-term follow-up, including information about opioid relapse and chronic pain information was noted when available.

Table 2 Results: summary of randomized-controlled trials, observational studies, and case reports where patients taking buprenorphine were managed in the perioperative setting
Full size table

Data analysis and interpretation of findings

Studies were grouped into RCTs and/or controlled studies (CS), observational comparator-controlled studies, and CR. In individual groups, the extracted data were organized in tabular form. Data extracted for the various pre-specified outcomes were collated, interpreted, and summarized into a narrative format. Standardized forms like the one shown in Appendix 1 were used to collect data. This was performed by two persons (A.G. and S.A.) in duplicate, and data were compared to ensure uniformity. Conflicts were assessed by the supervising expert (H.C.).

Results

The literature search yielded 3,630 results; after removal of duplicates, 2,632 articles were identified by two independent reviewers (A.G. and S.A.). After undergoing initial screening, and subsequent screening to remove previously unidentified duplicates and studies that did not meet our inclusion criteria, 18 papers were finally included, shown in the PRISMA flow diagram (Figure). The kappa agreement score between the investigators was 0.9. A majority, 72% (13/18), were CRs or case series (Table 2).13,14,15,16,17,18,19,20,21,22,23,24,25 Only one true case series was identified, with five of the CRs including single patients who underwent two surgeries at different time points. A total of 17 patients were included in the CRs and case series, five of whom underwent two procedures at different time points. The remaining five studies consisted of four observational studies and one CS that did not differentiate between buprenorphine and methadone in their controlled randomization (Table 2).26,27,28,29,30 Of the four observational studies, only two had sample sizes greater than 50, while the CS studied 19 Cesarean patients.30

Figure
figure 1

PRISMA search diagram: outline of the search strategy for final included studies

Full size image

Buprenorphine management

Of the 12 single surgery cases, six of the patients (50%) were continued on buprenorphine throughout the perioperative period. Of the five CRs with patients undergoing surgery at two different time points, three CRs (60%) described opposite management strategies (continued vs stopped). When buprenorphine was stopped, the stop date ranged from several months preoperatively to 12 hr after surgery. Gupta et al.26 and Meyer et al.28 described outcomes in patients who continued buprenorphine in the obstetrical patient population while Macintyre et al.27 and Hansen et al.29 described observational studies in which 50% of the study patients were continued on buprenorphine. The CS29 similarly compared opioid-replacement therapy in 37 study patients (three of which had two pregnancies) with 80 controls. The opioid-replacement therapy group included buprenorphine and methadone as one group (37 patients in the study group were randomized to buprenorphine or methadone), limiting conclusions regarding the causal relationship between buprenorphine and the main outcomes.

Buprenorphine indication

Preoperative buprenorphine use was characterized into OUD, chronic pain (CP), and postoperative analgesia (PA). In certain cases, there were multiple indications. The CS and observational studies quoted OUD as the indication for buprenorphine therapy in their respective cohorts with one observational study indicating both OUD and CP, and one that did not report the indication. Of the 17 patients described in the case series, buprenorphine was prescribed for OUD in five patients (29%), and CP alone in ten patients (59%). Buprenorphine was prescribed for both OUD and CP in one out of these 17 patients. One study had both OUD and PA as the indication.

Postoperative analgesic requirements

Controlled observational studies

One CS by Hoflich et al.30 compared parturients on opioid-replacement therapy with controls but did not stratify their results by buprenorphine or methadone use for this outcome. As a group, females on an opioid-replacement therapy using methadone (mean daily dose 63.89 mg) or buprenorphine (mean daily dose 15.33 mg) received significantly less opioid analgesics and significantly more non-steroidal anti-inflammatory drug therapy following Cesarean delivery compared with controls. Of note, this study also showed that the rates of Cesarean delivery were higher in methadone continued patients (68.4%) compared with buprenorphine continued patients (31.8%) although this result was not statistically significant.30 Pain scores and relapse rates were not reported.

Observational studies

Among the observational studies, Hansen et al.29 performed a prospective matched cohort study from a total joint arthroplasty database of patients receiving opioid-replacement therapy (patients receiving buprenorphine or methadone were pooled into one group and not differentiated in the analysis or results). When patients who were continuing opioid-replacement therapy (n = 17) were compared with matched counterparts who discontinued opioid-replacement therapy perioperatively (n = 34), one-year postoperative pain scores were not statistically different. Nevertheless, the study group had a mean increase in postoperative morphine use (793 mg·day−1) compared with the control group (109 mg·day−1). Relapse rates and route of delivery were not reported.

Meyer et al.28 compared parturients (44 vaginal and 19 Cesarean delivery) taking buprenorphine to controls in a case control study. This observational study demonstrated that females who continued buprenorphine perioperatively had similar pain and analgesic requirements during labour, but experienced more post-partum pain (pain scores 5.1 vs 3.3 in age matched controls not taking buprenorphine) and required 47% more analgesic following Cesarean delivery. Relapse rates and indication for buprenorphine were not reported.

Macintyre et al.27 studied buprenorphine/naloxone and methadone in a retrospective cohort study. They showed no difference in patient-controlled analgesia (PCA) morphine equivalents or pain scores when comparing perioperative continuation vs cessation of buprenorphine. There was no significant difference in days requiring acute pain service or PCA. They did show higher rates of sedation in buprenorphine continued patients, but this did not correlate with lower respiratory rates. Nevertheless, continuation of buprenorphine in this study was inconsistent with 14/22 (64%) continuing on day of surgery and 11/22 (50%) continuing on postoperative day (POD) 1. Relapse rates were not reported.

Case reports

Buprenorphine Continued Patients

Kornfeld et al.23 reported five cases, four of which were successfully managed on buprenorphine maintenance therapy (≤ 24 mg daily) in the perioperative setting with increased PA requirements and well controlled pain levels. Relapse rates were not reported.

Silva and Rubinstin15 report the case of a patient (24 mg buprenorphine sublingually [SL] daily) who underwent two total knee replacements (TKR) over a two-year period with well controlled pain. This same patient required higher doses of morphine (150 mg daily) to manage his postoperative pain when buprenorphine was discontinued for his second surgery. This patient relapsed into hydrocodone misuse at an unreported postoperative date after his second procedure.

Chern et al.16 and Huang et al.13 reported unsuccessful management of postoperative pain after a vaginal mesh removal and a Clagett window closure, with the former taking 24 mg buprenorphine SL daily and the latter taking 32 mg buprenorphine SL daily during a 41-day hospital stay. The same patient in Chern et al.16 discontinued buprenorphine in a follow-up procedure (addressed in the next section). Huang et al. only reported continued abstinence from opioids at a three-week follow-up and Chern et al. did not report this outcome.at any time period.

Book et al.14 reported the successful management of postoperative pain after removal of breast implants using supplemental buprenorphine alone in a patient continued on 24 mg SL daily.

Jones et al. (18 mg buprenorphine SL daily)18 describes a Cesarean delivery in which good pain control was achieved with daily morphine doses of 48 mg and 180 mg respectively. McCormick et al.21 described an emergent bilateral thigh and calf fasciotomy for compartment syndrome where a daily dose of 24 mg buprenorphine SL was continued until 12 hr postoperatively. Pain scores were reported as being well controlled. Book et al.,14 Jones et al.,18 and McCormick et al.21 did not report relapse rates.

Buprenorphine discontinued patients

Kornfeld et al.23 included two cases of five where 2 mg buprenorphine SL daily was discontinued and pain was well controlled. The first patient had two procedures with buprenorphine continued in the first and discontinued in the second. Pain was well controlled for both procedures. In the second patient, buprenorphine was restarted at POD 3 and their admission was uncomplicated and their pain was well controlled. The relapse rate was not reported.

Chern et al.16 reported the case of a 37-yr-old female (24 mg buprenorphine SL daily) who underwent two procedures for vaginal mesh removal and buprenorphine was discontinued for the second procedure, switched to hydromorphone five days before surgery. Intraoperatively, doses up to 1000 µg of fentanyl were required in the induction period when buprenorphine was discontinued. This patient reported poorly controlled pain after both procedures and noted the postoperative pain during the second case (when buprenorphine was discontinued) to be “unbearable”. This patient required an additional 100 µg of fentanyl with several adjunctive analgesics to manage postoperative pain in this instance.

Brummett et al. (16 mg buprenorphine SL daily)25 reported a case of posterior lumbar spinal fusion with buprenorphine discontinued on the day of surgery. The authors administered a dexmedetomidine infusion to improve their patient’s symptoms after failure to manage pain with a hydromorphone PCA in the intensive care unit. The relapse rate was not reported.

Israel et al.17 reported the case of a 37-yr-old female who presented for bilateral mastectomy and was transitioned from buprenorphine to a fentanyl patch three days preoperatively; this patient subsequently required up to 480 MEq·day−,1 adjuncts, and an APS consultation. Dose and relapse rates were not reported.

Marcucci et al.19 described the case of a patient with OUD who used illicitly obtained buprenorphine as part of a self-guided “crash detoxification program” three days before surgery to obliterate detectable cocaine levels in a urine test that was to be performed preoperatively. Buprenorphine was discontinued on the day of surgery and pain levels were poorly controlled. The dose was only reported as one tablet q4h and the relapse rate was not reported.

Rodgman and Pletsch22 reported the case of a 29-yr-old heart transplant patient (24 mg buprenorphine SL daily) who had their buprenorphine stopped in the preoperative setting; this patient required 90 MEq·day−1 and had high pain scores on re-induction of buprenorphine. The relapse rate was not reported.

Buprenorphine change or pain levels not reported

Hassamal et al.24 included a case of a 25-yr-old woman who underwent a tricuspid and aortic valve repair. She was taking 12 mg buprenorphine by an unreported route daily for CP. Buprenorphine was discontinued and restarted but the authors failed to report the timing of these changes, whether the discontinuation was preoperative or postoperative. Nevertheless, this was the only study to properly report the relapse rate with the patient remaining abstinent at a six-month follow-up.

Khelemsky et al.20 did not report pain levels in their case of a 44-yr-old woman who underwent two procedures. She was continued on 24 mg buprenorphine SL daily for an anterior cervical corpectomy but had her buprenorphine discontinued for an anterior cervical fusion four days later. The authors reported increased movement during surgery and higher anesthetic requirements when the buprenorphine was continued. Relapse rates were not reported. In a retrospective cohort study, Gupta26 compared vaginal and Cesarean deliveries. Daily peri-partum rescue analgesics in buprenorphine continued patients did not change with peri-partum neuraxial intervention. This study was underpowered to assess this outcome, lacking a control group and not reporting pain levels, rate of relapse, or indication of buprenorphine.

Discussion

The quality of evidence regarding perioperative management of patients on buprenorphine is weak. The number of studies is limited, and few directly evaluate the question of continuation versus discontinuation of buprenorphine. Among the studies that address this question, controls are scant with none being randomized. Of the observational studies (matched cohort, prospective cohort, retrospective cohort) that included patients on buprenorphine as part of their outcomes, four met our inclusion criteria and only two studied the effects of buprenorphine as a main outcome. The only CS combined patients taking buprenorphine and methadone into one group. This limits this study’s applicability to this systematic review to the level of an observation study as the lack of differentiation between methadone and buprenorphine makes the controlled randomization ineffectual. Thirteen CRs and a case series were identified. Many of these publications reported variables such as pain levels, buprenorphine dose, and perioperative buprenorphine continuation. Only three studies—all CRs—included relapse rates with one extending beyond the three-week time point.

There is insufficient evidence to support the decision to discontinue buprenorphine perioperatively. The main impetus for discontinuation, i.e., inadequate pain management, may be based on expert opinion and not on the existing evidence, as other authors have noted.19,20 The evidence does support the receptor occupancy theory described by Volpe et al.,31 which suggests near-maximal receptor occupancy at buprenorphine (SL) doses of 16 mg daily. Interestingly, in CRs with complete reporting (eight articles, 16 cases in total), every patient whose buprenorphine was discontinued and experienced poorly controlled pain was taking 16 mg SL daily or greater preoperatively.16,22,25 In fact, pain was successfully controlled in all but one of the patients taking 16 mg buprenorphine SL daily or greater who continued buprenorphine.13,14,15,16,18,23 Clinically correlated pharmacokinetic studies are required to confirm this cut-off, especially in the context of high inter-patient variability.

The existing evidence does suggest that the traditionally conservative approach of discontinuing and reducing opioids perioperatively may not be the most effective way to manage this complex patient population. Some evidence suggests that continuation with supplemental doses may offer the most effective analgesia while maintaining opioid-replacement therapy.14,15,23 Silva and Rubinstein directly address this theory with their observation that pain control was “easier to achieve” with greater functional recovery when buprenorphine was “continued throughout the perioperative period”.15

In addition to problematic pain management, discontinuation may hinder harm reduction with respect to addiction. Some expert opinions suggest improved treatment retention and lower misuse rates with discontinuation but do not acknowledge the greater risk of destabilizing a pre-existing CP condition or OUD when opioid-replacement therapy is stopped. According to the reviewed literature, there is no evidence to suggest that discontinuation of buprenorphine is the preferred method of preventing OUD relapse. Relapse rates are poorly defined in the reviewed literature, a surprising result given the importance of addiction management in this population. Also concerning is the lack of reporting the indication for buprenorphine use. The majority of reviewed studies report CP as the main indication vs OUD (ten vs five). This failure to report the indication for buprenorphine therapy in the existing literature may reflect the lack of awareness surrounding addiction therapy among perioperative physicians. If patient well-being beyond the operative room is to be factored into the decision-making process, current guidelines seem insufficient in addressing this matter.

Existing guidelines are largely driven by expert opinion with little reference to peer-reviewed primary evidence (Table 3). Potential weaknesses in the existing guidelines include the recommendation to transition patients to short-acting opioids before surgery.32 Evidence to the contrary shows lower relapse rates in the OUD patient population that are maintained on buprenorphine.9 Other guidelines disagree with this practice and do not recommend replacing buprenorphine with full mu agonists in the perioperative period.10 Lembke et al. most recently editorialized their support of perioperative buprenorphine continuation with evidence from CRs and series.33

Table 3 Summary of major existing guidelines for perioperative management of buprenorphine
Full size table

Other flaws in the existing guidelines include recommendations to prescribe full mu agonists at discharge for patients who had their buprenorphine discontinued.32 Some authors point to “opioid debt” i.e., (an insufficient opioid dose) as a potential complicating factor of this strategy.30 A CR by Rodgman and Pletsch indicate poor outcomes with this strategy22 with another by Khelemsky et al. reporting severe opioid withdrawal, apparently the most cited reason patients have OUD relapse while on opioid-replacement therapy.20

Overall, the current evidence to continue or discontinue buprenorphine perioperatively is limited. This gap in the literature represents an important area of research for those hoping to understand and appropriately manage the iatrogenic causes of the current opioid crisis. To better manage these patients, physicians caring for patients on buprenorphine in the perioperative setting need to incorporate harm reduction into their goals and decisions. In every case, connecting with outpatient primary care physicians and addiction specialists during the preoperative period is advised to ensure proper follow-up for these patients. During preoperative assessment, attention should be paid to each patient’s buprenorphine dose, indication, and risk for relapse (Table 4). There is a paucity of circumstances where the benefits of buprenorphine discontinuation (which could lead to relapse) outweigh the risks of continuation. Discontinuation is not benign and may impact relapse and result in poor acute pain management. The authors herein reinforce perioperative continuation of buprenorphine to be safe.

Table 4 Potential risk factors for opioid use disorder exacerbation in the perioperative setting
Full size table