Reducing Opioid Use in Patients Undergoing Cardiac Surgery — Preoperative, Intraoperative, and Critical Care Strategies

Abstract

Patients undergoing cardiothoracic surgery are exposed to opioids in the operating room and intensive care unit and after hospital discharge. Opportunities exist to reduce perioperative opioid use at all stages of care and include alternative oral and intravenous medications, novel intraoperative regional anesthetic techniques, and postoperative opioid-sparing sedative and analgesic strategies. In this review, currently used and investigational strategies to reduce the opioid burden for cardiothoracic surgical patients are explored.

PATIENTS UNDERGOING cardiac surgery are exposed to opioids throughout all phases of their care, from the operating room (OR) to hospital discharge. Opioid exposure may be particularly high in the intensive care unit (ICU), where opioids are used to provide immediate postoperative analgesia and sedation to patients requiring mechanical ventilation. Opioids are associated with a multitude of undesirable side effects, including ICU-associated delirium,^1–6 prolonged mechanical ventilation,^7–9 and delayed intestinal recovery after surgery.^10–12 Opioid use also is associated with the development of tolerance, hyperalgesia, and withdrawal,^6,13,14 significant concerns from a public health perspective as opioid abuse has become a global public health crisis, with increases in associated morbidity, mortality, and addiction worldwide.¹⁵ Opioid abuse is responsible for approximately 130 deaths each day in the United States,¹⁶ and anyone who takes opioids is at risk of developing an addiction.

Chronic pain may be broadly defined as persistent pain that lasts longer than 3 months.¹⁷ Chronic pain is common after cardiac surgery, affecting up to 56% of patients,¹⁸ and may affect up to 73% of patients in general who receive critical care.¹⁹ Persistent postoperative pain after cardiac surgery has been described as moderate to severe in half of patients, and even 2 years after cardiac surgery, up to 17% of patients still may experience pain.²⁰ The most common pain locations are the chest and leg, and pain is most commonly neuropathic.²⁰ Pain is significant enough that approximately 10% of patients develop persistent opioid use after cardiac surgery.²¹

A variety of pain scales and techniques have been used to assess pain after cardiac surgery,²² both in the immediate postoperative and chronic phases. This can be particularly challenging in patients who are sedated or require mechanical ventilation. Accurately assessing a patient’s pain is critically important because opioid analgesic dosing is closely related to a patient’s stated pain level or a provider’s perception of the patient’s pain if the patient is unable to participate in the assessment.

Duration of opioid exposure has been identified in postsurgical patients as one of the strongest predictors of developing opioid misuse,²³ and this has contributed to an interest in developing alternative strategies to minimize opioid use in surgical patients. The majority of cardiac surgery is highly invasive and pain-provoking, and many patients require prolonged periods of mechanical ventilation and continuous intravenous sedation and analgesia in the ICU. Although it would be impossible to eliminate opioids from the regular armamentarium of drugs used to treat this patient population, there are opportunities to reduce patients’ exposure to these medications, including during the preoperative, intraoperative, and critical care periods. The Enhanced Recovery After Cardiac Surgery Society recommends a multimodal, opioid-sparing analgesic regimen postoperatively (class I evidence),²⁴ but there are opportunities to reduce opioid use at all phases of a patient’s care. In this clinically oriented review, current practice and novel ways to reduce perioperative and critical care opioid exposure for patients undergoing cardiac surgery or receiving prolonged cardiopulmonary support in the ICU are considered.

Preoperative Interventions to Reduce Opioid Use

Opportunities to reduce opioid usage in patients undergoing cardiac surgery begin before the surgical incision is made and primarily are focused on preemptive analgesia. Two drug families—acetaminophen and anticonvulsants—have been studied regarding their potential to reduce opioid requirements when administered immediately before cardiac surgery (Fig 1). Intravenous acetaminophen (IVA) first was approved for use in the United States in 2010. Despite its widespread use, the mechanism of action is not entirely understood. IVA may work through multiple mechanisms, including inhibiting inflammatory marker production and blockade of pain pathways; through an inhibition of prostaglandin production; by affecting serotonin and spinal nociception; and through its effect on glutamate, substance P, and neurokinins.²⁵ IVA is associated with higher plasma and cerebrospinal fluid bioavailability compared with oral administration and has a faster onset of action and reduced production of the drug’s toxic metabolite, N-acetyl-p-benzoquinoneimine, potentially making IV a safer route of administration.²⁵ Peak hepatic concentration is predicted to be significantly reduced with IV administration,²⁶ which is desirable in the cardiac surgery patient population in whom hepatic dysfunction may occur in up to 10% of patients after cardiopulmonary bypass.²⁷ IVA should be avoided in patients with severe liver disease. IVA has minimal significant side effects, but perhaps the greatest limiting factor to IVA is financial, with a cost 400 times greater than oral administration.²⁸

Fig 1.

Non-opioid analgesics commonly used in perioperative cardiac surgery. COX, cyclooxygenase; NMDA, N-Methyl-D-aspartate; NSAID, nonsteroidal anti-inflammatory drug.

IVA has demonstrated mixed effects on opioid use and pain after cardiac surgery when started before surgery or intraoperatively, in some patients reducing opioid consumption postoperatively but not necessarily opioid-induced side effects,²⁹ and in other patients reducing pain scores but not total opioid requirements.³⁰ Jelacic et al. randomly assigned 70 patients undergoing a variety of procedures (coronary artery bypass grafting [CABG], valve repair and replacement, or aortic surgery) to six doses of IVA 1,000 mg, beginning before incision and finishing 24 hours after surgery.²⁹ IVA was associated with a 27% reduction in opioid use in the 24 hours postoperatively and increased patient satisfaction with pain control but no difference in opioid-related side effects.²⁹

A recent superiority trial compared IVA beginning in the OR and continued to 24 hours postoperatively with placebo in 147 patients having undergone median sternotomy and demonstrated that IVA was superior to placebo in terms of pain scores but not opioid requirements.³⁰ The study found no differences between IVA and placebo in other outcomes, including postoperative nausea or duration of mechanical ventilation. IVA’s ability to improve pain is a more consistent observation than its ability to decrease opioid usage.³⁰ Improved patient pain scores and satisfaction are not insignificant benefits of IVA and may be reason enough to incorporate the drug into a multimodal preoperative analgesic strategy. This is especially true given the favorable safety profile of IVA. The use of IVA in the pericardiac surgery period remains an area of active research.³¹

The anticonvulsants also have been studied for their opioid-sparing analgesic properties and potential to reduce the development of chronic pain after surgery. The incidence of chronic pain after cardiac surgery is approximately 11%³² and may adversely affect patients’ quality of life.³³ Gabapentin and pregabalin are the two most studied drugs, are both analogs of γ-aminobutyric acid,³⁴ and likely work through interaction with calcium channels in the central nervous system, leading to a decrease in neurotransmitter release.³⁵ Pregabalin is more potent than gabapentin and in general has fewer side effects. Initiating pregabalin before cardiac surgery has been shown to reduce opioid consumption immediately postoperatively and reduce chronic pain three months after surgery in elderly patients undergoing CABG or single valve repair or replacement.³⁶ In the same study, pregabalin treatment also was associated with less confusion on postoperative day one but an increased time to extubation.³⁶ Preoperative pregabalin also has been associated with a 60% reduction in postoperative tramadol requirement in patients undergoing off-pump CABG but no improvement in chronic pain up to three months postoperatively.³² The data are inconsistent with regard to pregabalin’s benefit in reducing chronic pain, with a recent study of 150 patients undergoing sternotomy for elective cardiac surgery being randomly assigned to preoperative and postoperative pregabalin with or without postoperative intravenous ketamine, or neither, showing a significant reduction in pain prevalence at three and six months after surgery in both groups when pregabalin was administered.³⁷ Postoperative pain scores at 24 hours after surgery also were reduced as were morphine requirements 24 hours after surgery in both the pregabalin and pregabalin plus ketamine groups.

Numerous studies of gabapentin suggest its efficacy in perioperative pain control in general.³⁸ Similar to pregabalin, study results in patients undergoing cardiac surgery are mixed. In a trial of 60 patients randomly assigned to preoperative gabapentin 1,200 mg, followed by 4 postoperative doses of 600 mg, Rapchuk et al. found no differences in postoperative pain, opioid use, or patient perceived quality of recovery.³⁸ Other studies have shown improved analgesia and reduced opioid consumption associated with gabapentin,^39–41 including a study by Menda et al. that demonstrated improved pain and opioid-related outcomes after a single preoperative dose of gabapentin 600 mg in patients undergoing CABG.³⁹

A meta-analysis of randomized controlled trials (RCTs) of gabapentin concluded that it did not reduce opioid use after cardiac surgery; and it may reduce pain scores but was associated with prolonged periods of mechanical ventilation.⁴¹ The same meta-analysis concluded that pregabalin, however, did reduce opioid consumption and pain scores and did not prolong mechanical ventilation. Specific clinical concerns of these drugs in the cardiac surgery population include their potential clearance by the cardiopulmonary bypass machine³⁸ and optimizing dosing in a patient population who experience frequent fluctuations in renal function.

Intraoperative Strategies to Reduce Opioid Use

Efforts to reduce intraoperative opioid use have been a pillar of the fast-track cardiac anesthesia movement, which, for the past decade, has focused on reducing time to extubation, length of stay, and hospital resources. Outcomes have been favorable enough that use of fast-track, or early recovery after surgery, techniques have been called a “global standard of care.”⁴² Fast-track cardiac anesthesia traditionally has aimed to facilitate extubation within six hours of patient arrival in the ICU, although currently an interest in ultrafast-track anesthesia strives for immediate extubation at the end of surgery.⁴³ A variety of strategies have been used or are actively being studied within fast-track protocols, including shorter-acting narcotics, such as remifentanil,⁴⁴ and lower doses of benzodiazepines⁴⁵ combined with preoperative medications, as previously discussed. Many hospitals have written their own protocols, including the authors of the present review at the University of Pennsylvania (Fig 2). More recently, regional anesthetics have been explored as a way to further reduce opioid and medication use in general. The use of regional anesthesia for intraoperative and/or postoperative analgesia after cardiac surgery is an exciting development in cardiovascular medicine.

Fig 2.

Fast-track cardiac surgery protocol used at the University of Pennsylvania, Department of Anesthesiology and Critical Care. CABG, coronary artery bypass grafting; IV, intravenous; OR, operating room; POD, postoperative day.

Regional Anesthesia

Regional anesthesia has tremendous potential to reduce opioid use after surgery, and the advent of liposomal bupivacaine (LipoB) has furthered interest in these techniques. Preoperative or intraoperative nerve blocks may be ultrasound-guided or placed under direct visualization. Ultrasound-guided peripheral nerve blocks offer an alternative to neuraxial techniques, including intrathecal, epidural, or paravertebral blockade. These techniques can provide profound somatic pain and sympathetic blockade; however, they largely have been replaced because peripheral blockade often is easier to perform and has a lower risk of spinal or lung injury, incompressible bleeding, hemodynamic instability, or local anesthetic (LA) systemic toxicity.⁴⁶

A promising evolving subset of peripheral nerve blockade, known as fascial plane blocks, relies on passive spread of LA placed between tissue planes to target peripheral nerves along their anatomic course. Moreover, depending on the particular nerve block’s location, some LA actually may spread proximally to provide sympathetic coverage in addition to peripheral somatic pain coverage. However, unlike direct peripheral nerve blocks, these “indirect” blocks may be less consistent or have high interindividual variation.⁴⁷

Blocks that are promising in cardiac surgery include erector spinae plane blocks (ESPB), serratus anterior plane blocks (SAPB), pectoralis nerve blocks (PEC I and II), intercostal nerve blocks, and various parasternal nerve blocks (Fig 3). They may be placed during any phase of perioperative care and can provide prolonged analgesia with either extended release formulations of LA or the placement of a continuous perineural catheter.

Fig 3.

Regional anesthesia techniques in cardiac surgery. PEC, pectoralis nerve block.

The ESPB involves injection of LA into a fascial plane between the thoracic transverse processes and the overlying erector spinae muscles. Its exact mechanism of analgesia is unclear, but a combination of intercostal blockade and spread to both the paravertebral and epidural spaces has been suggested.⁴⁶ As such, this spread theoretically will cover both somatic pain and sympathetic pathways. Several recent studies in patients, undergoing cardiac surgery via median sternotomy, have shown that bilateral ESPB improved pain scores, reduced opioid consumption and time to extubation, and improved other outcomes compared with control patients receiving intravenous (IV) medications or thoracic epidurals.^48–50

The SAPB involves injection of LA either superficial or deep to the serratus anterior muscle, typically near the midaxillary line. This facilitates blockade of the intercostobrachialis, long thoracic, thoracodorsal, and lateral cutaneous branches of the T3-to-T9 intercostal nerves.⁴⁶ Although some studies have shown improved outcomes for SAPB compared with continuous wound infiltrations or controls, other studies have found mixed results, especially compared with epidural and paravertebral blockade.^51–55 These findings partially may be a result of the limited duration of the block because perineural catheters can be dislodged easily at this site, and there is a lack of studies using extended-release LA.

The PEC I block involves injection of LA between the pectoralis major and minor muscles to anesthetize the medial and lateral pectoral nerves. The PEC II block involves both PEC I and a second injection between the pectoralis minor and serratus anterior muscles at the level of the fourth rib along the anterior axillary line. The PEC II block targets the long thoracic, thoracodorsal, and T2-to-T6 intercostal nerves.⁴⁶ One recent randomized controlled trial found that bilateral single-shot PEC II blocks improved pain scores and time to extubation for sternotomy patients compared with control patients.⁵⁶ In addition, case reports have described PEC II blocks as a rescue analgesic after mini-thoracotomy; however, to the authors’ knowledge, no RCTs exist.^57,58

There are several types of anterior chest wall blocks used for median sternotomy incisions that target the anterior cutaneous branch of the intercostal nerves, effectively offering a more distal and superficial target than the previous nerve blocks discussed. These nerves can be targeted as individual intercostal nerve blocks or as they ascend through fascial tissue planes in the parasternal region, either between the transversus thoracis and internal intercostal muscles (transversus thoracic plane block) or between the internal intercostal and pectoralis major muscles (pectointercostal fascial plane block or parasternal block).⁴⁷

To date, there are no large RCTs comparing the superiority of one particular block, but some smaller studies and case reports have shown promising results.^46,59–62 It is particularly important to discuss these blocks with the surgeon, given the possible risk of surgical site infection or even damage to the internal thoracic artery graft. Moreover, surgical dissection and graft preparation in this area may alter the fascial plane’s integrity and LA deposition, potentially limiting block efficacy.

These ultrasound-based chest wall techniques also can be supplemented or substituted by nerve blocks under direct visualization by the surgeons. For example, surgeons can target intercostal nerves directly as they course beneath their respective ribs or even consider performing intrapleural blockade. In this scenario, a surgeon injects LA or leaves a peripheral catheter between the parietal and visceral pleura, causing both somatic blockade of intercostal nerves and theoretical sympathetic blockade from proximal spread to the paravertebral chain, although this has been studied primarily in minimally invasive procedures.⁶³

Commonly used LAs for these regional blocks include ropivacaine and bupivacaine. However, these medications typically only have a clinical effect of approximately six-to-12 hours, depending on location of block and relative vascular uptake. Even with the addition of various adjuvants (eg, epinephrine, clonidine, dexamethasone), these single-shot blocks typically will not extend beyond 18-to-24 hours.⁶⁴ An alternative option to prolong duration is to place a perineural or fascial plane catheter for a continuous infusion of LA. However, this may not be feasible for certain fascial plane blocks that require multiple injection sites. In addition, it also is considered a more invasive procedure than single-shot injections and may pose additional bleeding and infection risks and potential interference in the surgical field.⁴⁷

For these reasons, there has been increasing interest in regional anesthesia in cardiac surgery as a result of the new availability of LipoB. LipoB is an extended-release formulation of bupivacaine that initially was approved by the US Food and Drug Administration in 2011 for wound infiltration after hemorrhoid and bunion surgeries.⁶⁵ Its approved uses have expanded to include breast, hernia, knee, and shoulder surgeries. LipoB uses DepoFoam liposomal technology, which permits a duration of action of approximately 72 hours.⁶⁵ Among cardiothoracic patients, LipoB has been studied for intercostal nerve blocks in patients undergoing thoracotomy and video-assisted thoracoscopic surgery (VATS), for which it has been shown to be a safe and effective alternative to epidural analgesia.⁶⁶ When compared with regular bupivacaine for intercostal nerve block in patients undergoing VATS, LipoB was associated with a significant reduction in postoperative opioid requirements, although this reduction was not necessarily present throughout the entire postoperative period.^67,68 However, a comparison between regular bupivacaine and LipoB in patients undergoing robotic-assisted thoracic surgery found no significant differences in postoperative analgesia.⁶⁹

At the University of Pennsylvania, LipoB has transformed the analgesic strategy for patients undergoing thoracic surgery. Epidural catheters have been replaced with intercostal LipoB injected intraoperatively by the surgeon for all patients, with the exception of lung transplantation and select decortication procedures when the surgeon anticipates difficulty with identifying the intercostal nerve because of pleural fibrosis. The use of LipoB as part of an enhanced recovery after surgery protocol has led to a 44% improvement in initial post-anesthesia care unit pain scores for patients undergoing VATS and a 74% improvement after thoracotomy. Postoperative inpatient oral opioid use decreased 56% after VATS and 11% after thoracotomy. A large reduction in discharge opioids also was observed after the introduction of LipoB (87% for VATS and 86% for thoracotomy).⁷⁰

There are very few studies of LipoB use in patients undergoing cardiac surgery. Subgroup analysis of a retrospective study, comparing LipoB with standard care (opioids ± thoracic epidural analgesia) in 58 patients who underwent sternotomy, demonstrated a reduction in postoperative pain scores but no reduction in opioid requirements.⁷¹ A recent randomized, prospective trial of LipoB versus placebo, when administered as a parasternal nerve block in patients undergoing coronary revascularization via median sternotomy, likewise failed to show a reduction in postoperative opioid requirements.⁷² Even without strong data to support it, there is growing interest in LipoB in cardiac surgical patients, and this represents an exciting area of research.

Neuraxial analgesia, in the form of intrathecal or epidural medications, has been used with varying effects in cardiac surgery to reduce opioid consumption. The use of preoperative intrathecal morphine to reduce postoperative opioid use after cardiac surgery is a class IIb recommendation within guidelines issued by the European Association of Cardio-Thoracic Surgery.⁷³ Similarly, the use of epidural analgesia is a class IIb recommendation within the same guidelines.⁷³

A number of studies have compared general anesthesia with and without thoracic epidural analgesia in cardiac surgery, and although no clear mortality improvement has been demonstrated, some positive outcomes have been observed, including a reduced length of mechanical ventilation.⁷⁴ The use of intrathecal morphine in cardiac surgery is an area of active research, with at least one active trial investigating its use in robotic cardiac surgery, registered on clinicaltrials.gov.⁷⁵

Critical Care Strategies to Reduce Opioid Use

Achieving adequate analgesia may be a challenge in the ICU after cardiac surgery, with significant clinical consequences. Untreated pain may present as agitation, which has been associated with increased postoperative delirium.⁷⁶ A vicious cycle of increased sedation and opioid analgesics may lead to delayed extubation, further delirium, and increased ICU length of stay. Patients have rated pain after cardiac surgery to be the most stressful experience in the ICU.⁷⁷ As discussed, this has led to development of enhanced recovery after cardiac surgery pathways that favor opioid-sparing analgesic techniques. In addition to continuation of patients’ preoperative nonopioid analgesics (pregabalin and gabapentin) and intraoperative opioid-sparing strategies (IVA and regional anesthetics), the critical care team has a variety of medications at its disposal, including nonsteroidal anti-inflammatory drugs (NSAIDs), dexmedetomidine (DEX), ketamine, and lidocaine (see Fig 1).

NSAIDs

NSAIDs decrease prostaglandin synthesis through cyclooxygenase (COX) enzyme inhibition, which produces their analgesic, anti-inflammatory, and antipyretic effects. The following two COX enzymes are inhibited: COX-1 and COX-2. COX-1 is systemic in the majority of tissues and causes the majority of side effects associated with NSAIDs (gastric irritation, platelet inhibition, renal injury).⁷⁸ COX-2 is increased during periods of inflammation, and its inhibition mediates the analgesic effects of NSAIDs.⁷⁸ NSAIDs have demonstrated benefit in decreasing postoperative opioid requirements, facilitating earlier extubation and improved mobility after cardiac surgery.⁷⁹ Despite these benefits, in 2005, the US Food and Drug Administration issued a black box warning for NSAID use in the immediate postoperative period after CABG surgery. This warning was in response to studies on the use of two COX-2—specific NSAIDs (parecoxib and valdecoxib), finding increased complications after CABG.⁸⁰

Despite the black box warning, Kulik et al. found that NSAID use continued in 32.5% of patients after CABG surgery,⁸⁰ with ketorolac and ibuprofen being the most commonly prescribed NSAIDs in this population, with respective rates of use of 89.2% and 12.9%. Ketorolac is a commonly used intravenous NSAID in the perioperative setting and may be used as part of a multimodal analgesic strategy. Rafiq et al. compared an NSAID-based multimodal regimen with a morphine-based one after cardiac surgery and demonstrated better analgesia and less nausea in the NSAID multimodal group.⁸¹ The multimodal treatment included ketorolac or ibuprofen, gabapentin, dexamethasone, and paracetamol, and the morphine group included paracetamol. That study demonstrated a trend toward a larger increase in serum creatinine levels in the multimodal group; however, this was not statistically significant.⁸¹ A meta-analysis in 2010 concluded that when NSAIDs are used in patient populations at low risk for renal dysfunction, there is no increased risk of renal injury.⁸²

In addition to analgesic effects, NSAIDs may provide the additional benefits of reducing atrial fibrillation and decreasing mortality after CABG surgery.^83–85 Engoren et al. used propensity-score matching to demonstrate decreased mortality after CABG, although the NSAID group was younger, predominantly male, had higher ejection fractions with shorter cardiopulmonary bypass and cross-clamp times, and received fewer blood products.⁸⁴ NSAIDs appear safe when used in the appropriate patient population with few comorbidities, specifically patients at low risk for bleeding and renal injury in uncomplicated cardiac surgeries.

DEX

DEX, best known for its sedative properties, is a potent intravenous alpha-2 agonist with a receptor- binding affinity seven times greater than clonidine. Alpha-2 agonists are believed to exert their analgesic effects by reducing the sympathetic outflow from the central nervous system; specifically, DEX has demonstrated opioid-sparing effects and decreased opioid-induced hyperalgesia.⁷⁸ DEX has a rapid onset and achieves peak effect within one hour of initiation. DEX undergoes hepatic metabolism; therefore, care must be taken in patients with severe liver disease because accumulation can occur. DEX may produce a dose-dependent bradycardia, which may have hemodynamic implications in the cardiac surgical population.

DEX has been shown to provide analgesia and allow for opioid reduction in surgical populations.^86,87 In 2003, Herr et al. demonstrated the opioid-sparing effects of DEX after CABG surgery; when compared with propofol for sedation in the ICU, propofol was associated with an increased morphine requirement compared with DEX.⁸⁷ DEX also was associated with a lower antiemetic requirement.⁸⁷ In addition, DEX has been associated with reductions in intraoperative fentanyl and postoperative morphine requirements in patients undergoing off-pump CABG.⁸⁸ Barletta et al. compared DEX with propofol for sedation as part of a fast-track cardiac surgery protocol. DEX was associated with a decreased opioid requirement, with 64% of patients receiving DEX not requiring opioids while sedated before extubation compared with 26% of patients receiving propofol sedation.⁸⁹

IV. Lidocaine

IV lidocaine, best known for its local anesthetic and anti-arrhythmic properties, is gaining traction as a postoperative analgesic because of its anti-inflammatory and anti-hyperalgesia properties. These effects occur through a dose-dependent inhibition of sodium, potassium, and calcium channels and G-coupled protein and n-methyl-d-aspartate receptors.⁷⁸ IV lidocaine has a rapid onset after bolus administration and can be maintained via continuous infusion. Lidocaine’s metabolism is primarily hepatic with renal excretion, and it is customary to follow IV levels while administering continuous infusion. IV lidocaine has been evaluated as an adjunct to analgesia for both acute postoperative and chronic pain and as a cardioprotective agent in the setting of myocardial ischemia and reperfusion injury.

IV lidocaine has demonstrated varied results when used for noncardiac surgical postoperative pain. In both laparoscopic and open abdominal procedures, IV lidocaine infusion reduced postoperative pain and facilitated earlier return of bowel function.⁹⁰ In open prostatectomy, thoracic surgery, and spine surgery, IV lidocaine infusion decreased postoperative pain and subsequently decreased total opioid consumption.⁹⁰ However, there are fewer data available for its use in cardiac surgery patients. Lee et al. found that intraoperative IV lidocaine reduced remifentanil requirements, in addition to decreasing myocardial injury as evidenced through cardiac enzyme release, in patients undergoing off-pump CABG surgery.⁹¹ Guinot et al. used boluses of lidocaine, ketamine, and dexamethasone, followed by a continuous intraoperative infusion of IV lidocaine, for an opioid-free anesthetic in patients undergoing cardiac surgery requiring cardiopulmonary bypass.⁹² They demonstrated that the opioid-free anesthetic group required significantly less total postoperative morphine (5 mg compared with 15 mg) to achieve similar pain scores and had shorter times to extubation and duration of ICU stay.⁹² IV lidocaine is a promising area for further research within cardiac critical care.

Ketamine

Ketamine, best known as a dissociative anesthetic, is a reversible n-methyl-d-aspartate antagonist; however, at subanesthetic doses, ketamine has demonstrated antihyperalgesic and antinociceptive effects. Ketamine primarily is excreted through the urine; therefore, care must be taken in patients with hepatic or renal dysfunction.⁷⁸ Ketamine may have considerable effects on the cardiovascular system—a sympathomimetic effect causing tachycardia and hypertension—and negative inotropy and increased myocardial oxygen demand; however, these findings are at anesthetic doses and not the subanesthetic doses of <0.5 mg/kg/h used for continuous infusions.⁷⁸

The data regarding ketamine’s effects on opioid requirements in the ICU have been mixed. Ketamine has been shown to reduce opioid requirements in patients admitted to surgical ICUs. Buchheit et al. found that low-dose ketamine infusions (up to 0.3 mg/kg/h) were beneficial at reducing both opioid and vasopressor requirements.⁹³ Pruskowski et al. reported that even though ketamine infusions reduced opioid requirements, there was an increase in the use of DEX or ziprasidone to maintain adequate sedation in a trauma ICU population.⁹⁴ A recent RCT of low-dose ketamine in mechanically ventilated patients in a general ICU found no reduction in opioid requirements but a reduction in the incidence of delirium.⁹⁵ A trial in cardiac surgery patients found that a 48-hour infusion of low-dose ketamine (1.25 μg/kg/min), compared with placebo, decreased total opioid requirement and found no difference in pain scores at rest or during deep breathing.⁹⁶ Ketamine may be a useful adjunct in patients with a tenuous respiratory status, with previous opioid tolerance, or at risk for opioid dependence; however, care must be taken given its associated risks and limited evidence for use in a cardiac surgical population.

Special Populations

Transcatheter Aortic Valve Replacement

Introduction of the transcatheter aortic valve replacement (TAVR) procedure has changed the landscape of cardiac surgery and has provided the opportunity for valve replacement in patients who otherwise would not have been candidates for open cardiac surgery. At its inception, the TAVR procedure was performed exclusively with the patient under general anesthesia; however, data over the past several years support use of monitored anesethesia care (MAC) for TAVR, despite 83% of TAVRs still being performed with the patient under general anesthesia as of 2015.⁹⁷ Importantly, the procedural success rate is similar at >97% regardless of anesthetic technique used.⁹⁷ Patients who received MAC saw a statistically significant shorter ICU length of stay and lower in-hospital all-cause mortality.^98,99 The risk of postoperative cognitive dysfunction is likely lower in patients for whom general anesthesia is avoided. In addition, Kiramijyan et al. noted that at their center, 79.2% of TAVRs performed with MAC also included intraoperative transesophageal echocardiography.⁹⁹

Opioid exposure is decreased in patients who undergo TAVR under MAC because patients do not require induction opioids to mitigate the hemodynamic effects of laryngoscopy. Successful strategies for sedation include propofol and DEX infusions. In a retrospective, observational study of a single center over 18 months, anesthesiologists providing MAC for TAVR used a fentanyl-dominant sedation strategy in only 4% of patients.¹⁰⁰ At the Hospital of the University of Pennsylvania, the MAC protocol for TAVR includes DEX and remifentanil infusions. Given the rapid metabolism of remifentanil and short context-sensitive halftime, no opioid given during the procedure remains postoperatively, and patients are ordered modest, as-needed doses of fentanyl or hydromorphone while in the post-anesthesia care unit for incisional or musculoskeletal pain.

Traditionally, femoral artery cutdown was considered a contraindication to performing TAVR with MAC. However, a multidisciplinary team at the University of Pittsburgh developed and investigated a protocol allowing for incision and cutdown using an ilioinguinal nerve block with 0.5% bupivacaine in addition to subcutaneous lidocaine at the area of incision.⁹⁷ This was in addition to IV acetaminophen, dexamethasone, and a propofol infusion titrated to a bispectral index of 75 to 85. That study found that the rate of conversion from MAC to general anesthesia intraoperatively also was lower than that of other reports—only 3.9% compared with 6% to 12% nation-wide.^97–99 Among the procedures that converted to general anesthesia, the most common reasons were related to the procedure itself, including valve or vascular issues, and not as a result of patient discomfort or oversedation.⁹⁷ As more cardiac surgical cases become catheter-based, including recent advances with transcatheter mitral valve replacement, additional research into optimizing an opioid-free anesthetic will improve patient care.

Extracorporeal Membrane Oxygenation

Extracorporeal membrane oxygenation (ECMO) and cardiopulmonary bypass may provide support for a patient’s failing cardiopulmonary circulation. There are similarities among all extracorporeal circuits, including exposure of blood to tubing and foreign membrane surfaces, which can result in altered pharmacodynamics and pharmacokinetics. Striking a balance between the constantly changing volume of distribution and drug sequestration of lipophilic drugs is extremely challenging with ECMO. Ex vivo studies have demonstrated that an ECMO circuit primed with whole blood and then administered fentanyl and midazolam, both lipophilic agents, had substantial drug loss at 24 hours.¹⁰¹

Commonly used analgesic regimens in the ICU include hydromorphone, fentanyl, and morphine, and these often are combined with sedatives including propofol, benzodiazepines, and DEX. Historically, patients on ECMO were administered large doses of opioids and sedation to alleviate the discomfort associated with large-bore cannulation and to reduce metabolic demands, oxygen consumption, and ventilator-patient dyssynchrony.

As more sophisticated approaches have developed, both for cannulation and the maintenance of ECMO circuitry, reduction of opioid use has followed. One of the key areas has been the move away from surgical cannulation to ultrasound-guided percutaneous cannulation for both venovenous and venoarterial (VA) ECMO. Percutaneous cannulation has been associated with fewer local infections, similar rates of limb ischemia, and improved 30-day survival.¹⁰² In addition, local anesthesia with lidocaine and low-dose anxiolytics can alleviate the pain and anxiety associated with percutaneous cannulation. Clinicians with advanced ultrasound experience can make this process quick and efficient. If awake cannulation is performed in the OR, inhalation mask ventilation also can be incorporated into a multimodality approach to avoid intubation and opioid administration.

The requirement for ongoing sedation and analgesia in a patient on ECMO depends on the indication for ECMO and underlying patient conditions. It is possible for patients to be awake, comfortable, interactive, and mobilizing during cannulation and maintenance of ECMO. Mori et al. described 19 VA ECMO patients managed with awake ECMO. No complications were experienced in this group, and multivariate analysis demonstrated a significant reduction in the risk of death in the awake VA ECMO group compared with the matched non-awake VA ECMO group.¹⁰³

A second cohort of patients who may be considered ideal candidates for non-sedated awake ECMO is the pre-lung transplantation patient population. In a study of 26 awake ECMO patients versus 34 mechanically ventilated patients, survival at six months after lung transplantation was 80% versus 50%, and patients in the awake ECMO group had shorter postoperative mechanical ventilation and shortened hospital stays.¹⁰⁴ Although the baseline physiologic characteristics of those who receive awake cannulation may have fewer risk factors and comorbidities than the mechanically ventilated group, there have been ademonstrated reductions in opioid and sedative use with awake pre-lung transplantation ECMO and increased mobilization and physical therapy, leading to better post-lung-transplantation outcomes. Some patients, for reasons varying from hemodynamic instability to underlying conditions, may require deep sedation while on ECMO. In these patients, alternative modalities can be used to limit exposure to opioids and sedatives. Reduction of opioid use is critical, with a recent study demonstrating that clinically significant iatrogenic withdrawal may occur in the short term after ECMO decannulation, and anxiety and depression occur in the long term.¹⁰⁵

The following three pharmacologic agents have been studied in an attempt to reduce opioid use in patients on ECMO: acetaminophen, ketamine, and DEX. An in vitro study of IVA concentration over six hours demonstrated a relatively constant concentration over time, irrespective of circuit age.¹⁰⁶ Acetaminophen has been recommended as long as concurrent liver injury is not present. A study of 26 patients receiving a ketamine infusion demonstrated decreased vasopressor use, and 35% of patients had a decrease in opioid use, while maintaining a Richmond Agitation and Sedation Scale score of −4 at 24 hours.¹⁰⁷ In a separate study, patients were administered DEX for weaning of venovenous ECMO at 0.7 ug/kg/h, and this led to a reduction in opioid infusion. Unfortunately, up to 25% of patients had adverse complications, including bradycardia.¹⁰⁸ With the increasing use of ECMO worldwide, there is enormous opportunity to study these drugs and improve strategies to reduce opioid use while safely managing patients on ECMO.

Conclusion

The importance of reducing opioid usage in perioperative medicine cannot be overstated. Traditional pharmacologic strategies may be supplemented by emerging techniques in regional anesthesia for patients undergoing thoracic and minimally invasive cardiac procedures, and opioid-minimizing sedation strategies increasingly are being used for patients requiring prolonged sedation while on ECMO. Additional research and development of novel sedatives and analgesics that effectively control intraoperative and postoperative pain have the potential to help not only patients on an individual level, but contribute to solving the global crisis of opioid dependence and abuse.