Chapter 67. Regional Analgesia in the Critically Ill

Intensive care specialists have become increasingly more interested in the prevention and treatment of physiologic and psychological stress in critically ill patients1–3 in order to prevent detrimental consequences ranging from systemic inflammatory response syndrome,4 to cardiac complications,5,6 to posttraumatic stress disorder.7–9 Studies have addressed the questions of an optimal sedation regimen and several evidence-based guidelines and strategies have been published.10–14 The analgesic component for sufficient stress relief, however, has not been addressed extensively, and few recommendations, primarily based on individual clinical practices, are currently available.15

In view of the side effects of opioids, especially respiratory depression, altered mental status, and reduced bowel function, regional analgesia utilizing neuraxial and peripheral nerve blocks offer significant advantages. The lack of a universally reliable pain assessment tool (“analgesiometer”) in the critically ill contributes to the dilemma of adequate analgesia. Many patients in the critical care unit are not able to communicate or use a conventional visual or numeric analog scale to quantify pain. Alternative assessment tools derived from pediatric16–18 or geriatric19 practice that rely on grimacing and other physiologic responses to painful stimuli might be useful, but have been inadequately studied in the intensive care unit (ICU). Changes in heart rate and blood pressure in response to nursing activities, dressing changes, or wound care can also serve as indirect measurements of pain,20 and sedation scores like the Ramsey or Riker and colleagues21,22 scale might be helpful, although not specifically designed for pain assessment.

The objective of this chapter is to describe the indications, limitations, and practical aspects of continuous regional analgesic techniques in the critically ill based on the available evidence, which at the moment is limited to case reports, cohort studies, expert opinion, and extrapolation from studies looking primarily at intraoperative use of regional anesthesia extending into the postoperative ICU stay. The evidence level of our recommendation is therefore mostly grade C and D according to the “Grades of Recommendation” published by the Oxford University Centre for Evidence Based Medicine (http://www.cebm.net/index.asp).

Epidural analgesia is probably the most commonly used regional analgesic technique in the ICU setting.23 Some indications, in which epidural analgesia may not improve mortality rates but facilitates management and improves patient comfort in the ICU, include chest trauma,24–27 thoracic28,29 and abdominal surgery,5,30,31 major vascular surgery,32,33 major orthopedic surgery,34 acute pancreatits,35 paralytic ileus,36–39 cardiac surgery,40,41 and intractable angina pain.42,43 Although high-risk patients seem to profit most from epidural analgesia,44,45 the current literature does not address the specific circumstances of the critically ill patient with multiple comorbidity and organ failure. For that reason, an individual approach is necessary when considering application of epidural analgesia in this population.46

In a survey of 216 general ICUs in England, Low47 found that 89% of the responding units used epidural analgesia, but only 32% had a written policy governing its use. Although 68% of the responding units would not place an epidural catheter in a patient with positive blood cultures, only 52% considered culture-negative sepsis, or systemic inflammatory response syndrome (SIRS), to be a contraindication. The majority of respondents did not list lack of consent or the need for anticoagulation after the catheter was placed as contraindications to the insertion an epidural catheter. Although the issues of consent, possible coagulopathy, and infection can be addressed rather easily in elective procedures, they become major problems in newly admitted patients, for example, those with multiple trauma or painful intraabdominal processes, especially acute pancreatitis. There is also controversy about the safety of placing epidural catheters in sedated patients,48,49 and confirmation of a good catheter position can be difficult in the critically ill patient if sensory level testing is not reliable.

Positioning the patient for the procedure can be a problem, depending on the underlying injury, the number and position of tubes and catheters, or external fixation devices. Table 67–1 summarizes indications, contraindications, and practical problems with the placement of epidural catheters.

The help of trained nursing staff is essential for good positioning and safe handling of tubes and catheters during the procedure. Maximum barrier precautions similar to the placement of central lines should be considered for placing epidural catheters in the critically ill as well. Tunneling the catheter should be considered to prevent dislocation and reduce possible catheter site infections.50 To confirm the correct position of the epidural catheter, electrical stimulation during placement or a postplacement radiograph with a small amount of nonneurotoxic contrast medium may be beneficial.51,52

Bolus injections of long-acting local anesthetics such as bupivacaine, ropivacaine, or levobupivacaine or the discontinuation of continuous infusion every morning will allow neurologic assessment when necessary. Monitoring of motor-evoked potentials (MEP) to the lower extremities and somatosensory-evoked potentials (SSEP) of the tibial nerve may serve as indicators when the neurologic examination is doubtful due to the patient's altered mental status. Although routinely used in the operating room for monitoring spinal cord integrity and for diagnosis and prognosis of spinal cord injury, the use of this technology in the ICU in the context of epidural analgesia has not been adequately assessed.

The most common side effects of epidural blocks are bradycardia and hypotension related to sympathetic block. Hemodynamic changes can be more pronounced with intermittent bolus dosing, in patients with hypovolemia, or those with reduced venous return secondary to high positive end-expiratory pressure (PEEP) ventilation. Based on data from lumbar punctures and meningitis from the beginning of the twentieth century,53 current sepsis and bacteremia are considered contraindications for intrathecal opioid applications and, by analogy, for placement of epidural catheters. However, many ICU patients, especially after trauma and major surgery, present with a clinical picture of SIRS. Fever and increased white blood cell counts alone, that is, in the absence of positive blood cultures, do not provide a reliable diagnosis of bacteremia. The combination of the serum markers C-reactive protein (CRP), procalcitonin, and interleukin-6, on the other hand, have been shown to indicate bacterial sepsis with a high degree of sensitivity and specificity54–57 and can guide the decision to place an epidural catheter. Regarding the patient's coagulation status, the current recommendations of the American Society of Regional Anesthesia (ASRA)58,59 should be followed. Adequate safety intervals during the administration of anticoagulative drugs are equally important for the placement and removal of epidural catheters.60,61 Although there is no compelling evidence of increased risks of epidural bleeding with developing coagulopathy or therapeutic anticoagulation while an epidural catheter is in place, the benefits of epidural analgesia should be weighed against this potential, highly detrimental complication.

At the time of this writing, there are no randomized, controlled trials or large prospective trials evaluating the use of peripheral nerve blocks for the upper extremity in critically ill patients. Nevertheless, severe trauma to the shoulder or arm is often part of the multiple injuries due to traffic or workplace accidents, often in combination with blunt chest trauma requiring mechanical ventilation. These injuries can contribute to severe pain, especially during positioning of the patient. If the orthopedic injury is part of complex trauma including brain injury in which the mental status of the patient is altered and opioid-based analgesic regimens might mask the neurologic situation, sufficient analgesia can be achieved for the shoulder or upper limb with either continuous interscalene,62–64 continuous cervical paravertebral,64–67 or infraclavicular68 approaches to the brachial plexus.

Particular concerns arise concerning the placement of regional blocks in ICU patients with impaired mental status due to neurologic injury or therapeutic sedation. Benumof reported a small series of serious complications, including spinal cord injury related to the interscalene approach, which might have been associated with sedation or general anesthesia.69 His case descriptions relate to the spinal cord injury in heavily sedated or anesthetized patients and not to injury of the peripheral nerves. Despite this, performance of blocks anatomically close to the centroneuraxis can indeed carry a higher risk of spinal cord needle or injection injury. In sedated critically ill patients a combination of ultrasound and nerve stimulation for the placement of interscalene catheters and a technique with a less medial needle direction70,71 should help to minimize the risk of complications. Perhaps most importantly, such blocks should be performed only by clinicians with adequate experience. The unavoidable blocking of the phrenic nerve and the loss of hemidiaphragmatic function72 should be considered while planning the intervention. Although phrenic nerve blockade has negligible effects in mechanically ventilated patients, it may impair weaning from mechanical ventilation in high-risk patients. Furthermore, the proximity of the insertion site of the interscalene catheter to a tracheostomy tube might increase the risk of infection, and careful, standardized monitoring of the puncture site is therefore needed. Positioning problems might limit the use of the cervical paravertebral approach, which provides good analgesia for shoulder, arm, and hand.

The continuous infraclavicular64,68,73–75 and axillary64,76–78 approaches provide good analgesia for most of the arm, elbow, and hand. Bolus injection of local anesthetic through the catheter should especially be considered in patients who need surgical anesthesia, for example, for painful dressing changes or debridements for burns or large soft tissue wounds in the affected area. The small risk of pneumothorax must be considered against the slightly higher success rate and easier catheter maintenance of the infraclavicular versus the axillary approach, but can be ignored if a chest tube is already in place. A more lateral infraclavicular approach79–81 might help to reduce the pneumothorax risk further.

Femoral nerve catheters are helpful in the management of acute pain from femoral neck fractures, in the period between the injury to shortly after surgical stabilization of the fracture.82,83 Skilled use of ultrasound84 might limit the unavoidable pain with nerve stimulation in this situation, which otherwise can be treated with small doses of intravenous remifentanil (0.3–0.5 mcg/kg) or ketamine (0.2–0.4 mg/kg). A fascia iliaca compartment block85,86 might be a technical alternative.

Continuous femoral catheters in combination with a sciatic block provide excellent pain relief for the whole leg and even surgical anesthesia for procedures like external fixation.87 Whether an anterior88 or posterior approach (midgluteal,89 subgluteal90 classical Labat approach with one or two injections91) to the sciatic nerve is chosen depends largely on the skills of the operator and the ability to adequately position the patient for the procedure. If a combination of catheter techniques is used, as is often needed for the lower extremity, the total daily dose of local anesthetic should be adjusted based on catheter location, admixtures like epinephrine, drug interactions, and disease states as summarized in a recent review by Rosenberg and coworkers.92

Bolus injection of long-lasting local anesthetics in combination with clonidine93 or buprenorphine94 may help to reduce the overall amount of local anesthetics needed and minimize the effects of local anesthetic toxicity, although research results about these adjuvants are equivocal at present.95,96

Celiac plexus blocks may provide excellent analgesia for pancreatitis and cancer-related upper abdominal pain, but technical difficulties in the critically ill (CT guidance, fluoroscopy, or transgastric ultrasound) and the need for repeated injections limit its value for acutely critically ill patients.

Intrapleural catheters for pain control after chest trauma are of limited value secondary to concurrent drainage from chest tubes. The risk of pneumothorax limits their benefits for the management of pain after conventional cholecystectomy compared with the epidural or paravertebral technique in ventilated patients.

Thoracic paravertebral catheters can be a valuable alternative to epidural catheters for the management of unilateral pain restricted to a few dermatomes (eg, rib fractures97 or zoster neuralgia).

Table 67–2 provides a summary of the most often utilized continuous peripheral catheters.

Single-injection nerve blocks, for example, intercostal blocks for the placement of chest tubes, scalp blocks98 for the placement of halo fixation, and sufficient local infiltration anesthesia for typical ICU procedures like placement of arterial and central venous catheters, lumbar punctures, or ventriculostomies are often forgotten, although they are easy and safe to perform. If EMLA cream is used for topical anesthesia it needs to be applied 30–45 min before the procedure to achieve an optimal effect.

Intrathecal morphine99–102 injections as a single shot and via spinal catheters (microcatheters are currently not approved in the United States but are available in Europe) can be an alternative to epidural catheters, especially if only short-term use after surgery is anticipated.

Local anesthetics have been shown to have several positive systemic effects (including analgesic, bronchodilatory, neuroprotective, anti-inflammatory, antiarrhythmic, and antithrombotic properties)103 when given or absorbed in adequate quantities (the exact dose–response relationships are widely unknown). They also have negative effects, such as, neurotoxicity (dose-dependent), myotoxicity,104,105 inhibition of wound healing, cardiotoxicity (dose-dependent), and central nervous excitation or depression (dose-dependent).106

To prevent local anesthetic systemic toxicity from accidental intravascular injection, a test dose of local anesthetic or saline with 1:200,000 epinephrine can be used with catheter placement, but the sensitivity of heart rate, blood pressure increase, and T wave changes103 might be altered in ICU patients, especially those treated with beta-blockade and α2-agonists or catecholamines. Careful aspiration to check for blood return should be performed before each bolus injection.

Most studies examining plasma levels of local anesthetics were not performed in critically ill patients. Scott and colleagues described the safe use of epidural ropivacaine 0.2% for 72 h with plasma levels far below the toxic threshold,107 and Gottschalk and associates observed safe plasma levels after 96 h in patients treated with thoracic epidural ropivacaine 0.375%, indicating no significant accumulation over time.108

In general, given the lack of cooperation and communication in many ICU patients, regional analgesia techniques using continuous catheters in the ICU require a higher level of vigilance than needed for regular ward patients. A close cooperation between the ICU team and the acute pain or anesthesia service of the hospital is required.

Critical care nursing personnel should be specifically trained in handling regional analgesia catheters and must be aware of the potential complications and their early warning signs. Because of the frequently large and confusing numbers of various infusion catheters in critically ill patients, the risk of drug errors and wrong administration of drugs through continuous regional analgesia catheters may be higher in these patients. Well-trained and highly qualified personnel are the best safeguard against these complications besides eye-catching labels, standardized care protocols, and perhaps specially designed connectors for those catheters.

Comprehensive diagnostic approaches, including magnetic resonance imaging or computed tomography, should be undertaken when there are clinical signs of possible bleeding complications (eg, suspected epidural or retroperitoneal hematoma). Structured observations of catheters for infectious complications and careful adherence to aseptic technique during placement and tunneling of the catheters, as well as the possible use of antibiotic-coated catheters in the future may reduce possible infectious complications. Catheters should not be removed routinely after certain time intervals, but only when clinical signs of infection appear.

If catheters get disconnected a study by Langevin109 suggests that when the fluid in the catheter is static the proximal 25 cm of the catheter can be immersed in a disinfectant, cut, and reconnected to a sterile connector. This technique is only feasible for catheters in which the fluid column can be observed. Stimulating catheters should never be cut because of the danger of unwinding the internal metal spiral wire, which conducts electrical current. No study has looked at the risk of reconnecting these catheters after thorough disinfection of the outer surface, which is likely a common practice in many institutions. Cuvillon and colleagues110 reported a high overall incidence of colonization (57%) of femoral catheters without septic complications. Therefore the decision to reconnect or remove the catheter must be based on a case-by-case basis and specific clinical circumstances. The overall risk of permanent neurologic damage (from direct trauma, bleeding, or serious infections) or death from regional anesthesia and analgesia seems to be low in the perioperative setting as shown by large surveys by Auroy and coworkers111,112 and Moen and associates.113 Although both studies certainly include critically ill patients, there are no specific subgroup data available.

In summary, regional analgesia, whether utilizing single-injection regional blocks or continuous neuraxial and peripheral catheters, can play a valuable role in a multimodal approach to pain management in the critically ill patient to achieve optimum patient comfort and to reduce physiologic and psychological stress. By avoiding high systemic doses of opioids, several complications like withdrawal syndrome, delirium, mental status changes, and gastrointestinal dysfunction can be reduced or minimized. Because of limited patient cooperation during placement and monitoring of continuous regional analgesia, indications for their use must be carefully chosen based on anatomy, clinical features of pain, coagulation status, and logistic circumstances. High-quality, trained nursing care and well-trained physicians are essential prerequisites when using these techniques safely in the critical care environment. These recommendations are based on small series, uncontrolled trials, or extrapolated from controlled trials in the perioperative setting (grade C and D recommendations); further research on the use of regional analgesia techniques in the critically ill is needed before definitive guidelines can be recommended.