“Opioid-free anesthesia may be feasible. Nevertheless, it appears neither logical nor beneficial to patients.”
Fashions come and fashions go. Changes in clothing, automobiles, and restaurants follow popular trends and are often periodic and cyclical. Ideally, medical change is driven not by fashion but instead by concepts of effectiveness and safety, and these concepts should improve and refine as better data become available. Trends hold true also in anesthesia practice, where for example intraoperative opioid selection has swung from long-duration to ultra-short duration and then at least partially back again.1 Recent years have witnessed a new fashion in anesthesiology: “opioid-free anesthesia.” For some, the opioid pendulum has swung clear past rational opioid use in balanced anesthesia to eliminating opioids intraoperatively and sometimes also postoperatively (opioid-free analgesia). Eradicating opioids from intraoperative and postoperative analgesic plans has been termed a “movement,”2 and like many “movements,” it has attracted passionate proponents and spirited debate.3,4 Nevertheless, clinical research and peer-reviewed evidence on the potential benefits and risks of opioid-free anesthesia have remained scant, needed, and called for.5–7
This issue of Anesthesiology features two articles on opioid-free anesthesia that bring some light to the heat of the debate. The original investigation by Beloeil et al.8 reports the results of a randomized clinical trial of balanced anesthesia with either remifentanil or dexmedetomidine (opioid-free). The review by Shanthanna et al.9 provides a narrative exposition of opioid-free versus opioid-sparing approaches in the perioperative period. These articles are timely and important.
The clinical trial8 was investigator-initiated, multicenter, randomized, prospective, parallel-group, and single-blind, conducted in 10 centers in France, with an independent data and safety monitoring board to oversee the conduct and review safety data. Patients undergoing major noncardiac surgery received intraoperative balanced anesthesia featuring opioids (remifentanil infusion and morphine) or dexmedetomidine infusion (opioid-free) and postoperative morphine. The primary outcome was a composite of opioid-related adverse events (hypoxemia, ileus, or cognitive dysfunction) in the 48 h postoperatively. Secondary outcomes included postoperative pain, opioid consumption, nausea and vomiting, and times to extubation and postanesthesia care unit discharge.
Frighteningly, the investigation was stopped prematurely because of safety concerns. Specifically, there were five cases of severe bradycardia in the dexmedetomidine group, including three cases of asystole. The composite primary endpoint (adverse events) occurred in significantly more dexmedetomidine (78%) than opioid (68%) patients (relative risk, 1.16), because of a greater incidence of hypoxemia but not more ileus or cognitive dysfunction. Times to extubation and postanesthesia care unit discharge were significantly longer in the dexmedetomidine than the opioid group. Although dexmedetomidine patients had lower rates of nausea and vomiting, postoperative morphine consumption was statistically but not clinically meaningfully less (median, 3 mg difference), and postoperative pain scores, intensive care unit admission, and length of stay did not differ between groups. The authors concluded that balanced opioid-free anesthesia with dexmedetomidine increased serious adverse events and they questioned the benefit of eliminating intraoperative opioids in favor of dexmedetomidine.
This investigation is important because it evaluated both the benefits and risks of an opioid-free anesthetic regimen in a carefully controlled and blinded randomized trial. Any benefits were negligible. It is rare and concerning when a clinical trial is terminated for safety. Moreover, the adverse events in this trial were not strictly related to dexmedetomidine dose, meaning that they are less predictable and that reducing the dose may not be a strategy for mitigation. This investigation exemplifies the critical need for carefully assessing the safety of multimodal regimens (and, in this case, opioid-free anesthesia), in addition to potential benefit, as has been called for previously in this journal10–12 and by others.13,14 Case in point, a four-drug intraoperative nonopioid regime added no benefit to standard anesthesia and did not appreciably change postoperative opioid use or pain scores despite the greatly increased complexity of the anesthetics.12 The trial reported by Beloeil et al.8 clearly demonstrates that we can do more harm than good by eliminating intraoperative opioids, depending on how we replace them.
The narrative review9 is comprehensive and concludes that although complete opioid sparing is possible in some contexts and procedures, there is no evidence that opioid-free strategies (1) have benefits above and beyond opioid-sparing strategies, (2) influence the risk or prevent persistent postoperative opioid use, or (3) prevent postoperative opioid overprescription. In fact, there is little reason to suppose any of these hoped-for benefits would accrue. The narrative remarks that opioid-free strategies, however noble in their cause, do not fully acknowledge the limitations and gaps in existing evidence and clinical practice considerations; do not allow analgesic titration based on patient needs; are unclear about optimal nonopioid components; have ill-defined roles in different surgical settings and perioperative phases; necessitate additional equipment, monitoring, resources, and cost; ignore concerns about safety and drug interactions; do not serve to decrease the risk of persistent opioid use; and, as a result, are unrealistic in routine clinical practice. Perhaps most importantly, the undue focus on opioid-free anesthesia may be distracting us from optimizing pain relief and minimizing long-term harms.
Looking Back
What are the antecedents and potential drivers of “opioid-free anesthesia”? These may include inappropriate overreaction to the opioid crisis, opioid sparing taken to an extreme, and misinterpretation of contemporary emphasis on enhanced recovery after surgery guidelines.
The Opioid Crisis
The current opioid crisis emanated from inappropriate prescribing, marketing, diversion, and misuse of oral prescription opioids for acute and chronic noncancer pain. The human and economic toll is staggering (400,000 U.S. deaths over the past two decades, one quarter of which may be suicides, and $2.5 trillion costs for just 2015 to 2018, including $700 billion in 2018 alone—3.4% of the U.S. gross domestic product),15–18 and the problem is not limited to the United States.19 Whether via the medical or lay press, no one is unaware of the opioid crisis, and opioid overdose has worsened with the COVID-19 pandemic.
The opioid crisis has multiple antecedents, and the roots are even deeper than popular narratives and current medical literature portray. Opioids have been used medicinally for millennia. However, more recent and successive waves of use have culminated in our current predicament. In the first wave (1960s), acute pain management with oral opioids was so common that propoxyphene was the second most commonly dispensed drug in the United States.20 A second wave (1990s) saw further increases in oral opioid use for chronic pain treatment despite unclear evidence of effectiveness.20 A third wave (2001 to 2010) saw further expanded oral opioid use, with aggressive marketing, liberalized prescribing, and proliferating managed care, regulatory agency, and practice efficiency guidelines that encouraged expanded opioid prescribing. Indeed, from 2006 to 2010, hydrocodone–acetaminophen was the most widely prescribed drug in the United States.21 Commensurate with increased oral opioid use during that time, the annual overdose deaths involving prescription oral opioids escalated by 182% (fig. 1).22,23 The fourth, most recent, and most devastating wave of opioid overdoses relates to escalating heroin and illicit fentanyl use (fig. 1), economic stress and employment deterioration,24 a shift of the abused opioid supply chain from medical to illicit sources, and a fluid substitution of illicit fentanyl for prescription opioids. Overall opioid prescribing has fortunately receded,25 yet overdose deaths have negligibly changed because of this more lethal wave of heroin and illicit fentanyl abuse.26,27
U.S. drug overdose deaths involving opioids 1999 to 2020 (November), by type of opioid. The data can be accessed at https://www.cdc.gov/nchs/nvss/vsrr/drug-overdose-data.htm.
U.S. drug overdose deaths involving opioids 1999 to 2020 (November), by type of opioid. The data can be accessed at https://www.cdc.gov/nchs/nvss/vsrr/drug-overdose-data.htm.
The point is this: Notably absent from this entire historical narrative is any mention of, association with, or causal attribution of the opioid crisis to intraoperative and immediate postoperative use of opioids to treat moderate-to-severe surgical pain. Similarly, it is a great and unfounded leap to say that there is an opioid crisis and therefore we must stop using opioids for surgery. However, the steady drumbeat of the opioid crisis can influence practitioner behavior, voluntarily or subconsciously. If the concept and practice of “opioid-free” anesthesia is a response to the opioid crisis, it is unnatural, unfounded, and perplexing.
Opioid Sparing versus Opioid Eradication
With the shift from balanced intraoperative anesthesia and postoperative analgesia with opioids as the only analgesic component to current concepts of balanced anesthesia and multimodal analgesia with opioids and other drugs, perhaps then just dropping opioids altogether was somehow seen as a logical or natural progression. After all, we have regional anesthesia, lidocaine, ketamine, and other drugs that can contribute to the analgesic components of anesthetics. Opioids do have untoward side effects, the most bothersome being nausea and vomiting and the most dangerous being respiratory depression. Multimodal analgesic therapy is based on the theory of combining drugs to achieve additive benefit without additive risk or synergistic benefit with only additive risk. Multimodal approaches can clearly reduce pain and opioid consumption in some circumstances. All is not pure benefit, however, because many adjuvant drugs can actually increase the very side effects that are unwanted or introduce new risks. For example, gabapentinoids, sedative-hypnotics, benzodiazepines, and “muscle relaxants” (e.g., baclofen and carisoprodol) can produce additive or even potentiate opioid-induced respiratory depression.28–30 This is not more benefit with less risk, but in fact more risk with less benefit.
Perhaps then the thinking was to change further, to ideally eliminate untoward opioid drug interactions by just eliminating opioids altogether. The problem is that opioids are the most efficacious analgesics available in many situations, and the remaining multimodal components alone are insufficiently effective against moderate-to-severe pain. No amount of wishing will turn acetaminophen into hydromorphone. Although opioid sparing (if it means avoiding excessive opioid use) is appropriate, opioid eradication, as identified by Shanthanna et al.,9 is not, based on evidence to date. Opioid sparing is not synonymous with opioid absence.
Enhanced Recovery after Surgery and Opioids
The objective of enhanced recovery after surgery protocols is to restore patients to their presurgical state of health as quickly as possible, as well as to reduce complications and standardize perioperative care to minimize variability and improve outcomes.13 Postoperative recovery is influenced primarily by minimally invasive (laparoscopic) surgical approaches and postoperative care (e.g., oral intake, early ambulation), and only one of the five key enhanced recovery protocol components originally associated with improved outcomes relates to anesthesia (epidurals in open colectomy).13,14 Nevertheless, enhanced recovery protocols have been increasingly laden by well intentioned clinicians with more and more elements, simply added under the assumption they will provide additional benefit. Unfortunately, these protocols and assumptions are often untested, unvalidated, and of unknown benefit and risk.14
Also problematic is that enhanced recovery protocols and consensus statements may overstate and/or misinterpret published literature to support assertions about opioids and the value of opioid eradication, and it is important to separate evidence from conventional wisdom and assumption. For example, a consensus review of optimal perioperative care in colorectal surgery from the Enhanced Recovery After Surgery Group31 claimed that the adverse effects of opioids can hamper recovery and prolong length of stay, citing a Cochrane review, yet that review did not address opioids. A set of clinical practice guidelines for enhanced recovery after colon and rectal surgery from the American Society of Colon and Rectal Surgeons and Society of American Gastrointestinal and Endoscopic Surgeons claimed that multiple prospective studies (citing four) demonstrated that minimizing opioids was associated with earlier return of bowel function and shorter length of stay, yet inspection of those four studies found that none evaluated the specific contribution of postoperative opioids in isolation, and none eliminated opioid use entirely.32 Less may be better, but it is not clear that none is best. Such overstatements notwithstanding, enhanced recovery protocols do not universally advocate for opioid eradication over rational use. Even the most recent enhanced recovery guidelines state that opioids cannot necessarily be avoided and that analgesia is best provided by opioid sparing, not by opioid eradication.33
Looking Forward
It is time to regain our balance in anesthesia. Opioid-free anesthesia may be feasible. Nevertheless, it appears neither logical nor beneficial to patients. Feasibility is not a therapeutic indication, nor is it patient-centric. In a hierarchical progression of values from feasible (can it be done?) to viable (does it have value?) to worthwhile (does it meet patient needs?) to desirable (do patients want it?) to optimal (does it best meet patient needs and optimize outcome?), feasibility is the lowest of value propositions.
Surgery hurts. Patients suffer from acute postoperative pain. More than 80% of patients report their postoperative pain is not adequately treated, a metric unchanged for decades.34–36 Depriving patients of analgesic opioids is unlikely to improve that metric. Additionally, poorly controlled acute postoperative pain is the greatest risk factor for chronic postoperative pain, which afflicts 10 to 60% of surgical patients.37,38 Moreover, acute postoperative pain is associated with complications, both generalized (postoperative cognitive decline, delirium, and sleep disturbances39 ) and surgical (surgical site and urinary tract infections,40 30-day infection rates,41 and higher readmission rates42 ). So too is chronic postsurgical pain associated with postoperative complications.43 Patients who experience postoperative pain have poorer recovery, less satisfaction, and more regret about having had surgery.44 Nonopioid approaches to treating acute postoperative pain such as local and regional anesthesia, which are hoped to prevent the transition to chronic postoperative pain, and nonopioid analgesics such as nonsteroidal anti-inflammatory drugs, N-methyl-d-aspartate–receptor antagonists, and gabapentinoids have, unfortunately, not been successful.38 The history of excessive and inappropriate outpatient prescribing of oral opioids for chronic pain should not motivate the withholding of opioids for acute perioperative surgical pain. As stated previously, “perioperative practitioners should avoid making surgical patients pay with unnecessary suffering for the opioid overprescribing sins of others.”7
The clinical trial report8 and the comprehensive review9 in this issue of Anesthesiology demonstrate that it is time for the opioid pendulum to swing back from the unjustified extreme of opioid eradication to a more balanced and rational approach.7 Although it is well known that excessive opioid use can have adverse consequences that can and should be mitigated, swinging completely to the absence of opioids, in the name of whatever well-intentioned motivation, can also have adverse consequences that can and should be mitigated. There is a Goldilocks zone for opioids in which patients are provided the optimal balance of analgesia with minimal or no side effects or risks. The problem is that one dose regimen does not fit all. Personalized medicine is difficult to protocolize.
The challenge is to devise regimens, optimize drug selection, and deliver care that provides optimal intraoperative and immediate postoperative analgesia and ideally also confers long-lasting benefit with improved recovery trajectories.45 For example, and the exact opposite of opioid eradication, the long-duration opioid methadone not only reduces intraoperative and immediate postoperative pain and opioid requirements but also reduces pain and opioid consumption long after patients have left the healthcare institution.46–49 An “opioid-sparing opioid” like methadone might confer advantage over opioid eradication.
Optimal postoperative analgesia will only be achieved by using the available tools in a manner tailored to the specific needs of individual patients. Our preoccupation with opioid-free anesthesia and analgesia distracts us from this challenge. What is clear is that our current analgesic pharmacologic armamentarium is inadequate to the task of providing adequate pain relief with minimal side effects, and it has not materially changed in decades. While academia and industry regroup after years of failures, hopefully new molecular, cellular, and systems-based approaches will help us to discover novel approaches to analgesia. In the meantime, we can work on using existing tools more effectively. We have not defined the optimal components of multimodal analgesic pathways, nor have we developed tailored approaches consistent with the goals of precision medicine. Rather than toss aside opioids, some of our most powerful tools, we should focus on defining the optimal use of these important drugs.
Research Support
Supported by grant No. R01 DA042985 from the National Institutes of Health, Bethesda, Maryland (to Dr. Kharasch) and grant No. I01 BX000881 from the Department of Veterans Affairs, Washington, D.C. (to Dr. Clark).
Competing Interests
Dr. Clark has a consulting agreement with Teikoku Pharma USA, San Jose, California. Dr. Kharasch declares no competing interests.
