Acute and Chronic Post-Thoracotomy Pain
David R. Lindsay
1. Without adequate analgesia, most patients would experience severe pain following thoracic surgery.
2. Epidural analgesia is widely practiced and has been shown to provide superior pain relief compared with systemic opioids.
3. Multimodal analgesic strategies improve overall outcomes including patient satisfaction.
4. Chronic post-thoracotomy pain (CPTP) is common and remains a challenging condition to treat. Further investigation into prevention of this syndrome is needed.
The patient is a 64-year-old man who underwent a left thoracotomy and extrapleural pneumonectomy for mesothelioma. A mid-thoracic epidural catheter was placed preoperatively and used to deliver 0.6mg of hydromorphone prior to incision. Intraoperatively, no medications were administered through the epidural catheter to avoid sympathectomy and hemodynamic instability.
Upon the patient’s arrival to the intensive care unit, an epidural infusion of bupivacaine 0.125% and hydromorphone 10 mcg/mL was initiated at 6 mL/h. The patient initially experienced 8/10 pain, requiring an epidural bolus of local anesthetic and an increase of the infusion rate. These adjustments resulted in reduction of his pain to 3/10. With improved analgesia, the patient was able to improve incentive spirometry performance, but he still continued to experience shoulder pain. He continued to do well with adequate pain control in the intensive care unit (pain score 3-5/10). His epidural was discontinued on postoperative day 3. He was discharged to home on postoperative day 5 with oral oxycodone as needed.
At his 2-month postoperative evaluation, the patient complained of significant chest wall pain localized to the thoracotomy incision. He described his pain as burning and aching.
The importance of postoperative pain management is well established.1 Postoperative pain following thoracic procedures causes a reversible restrictive pattern of ventilation with a decrease in vital capacity (VC) and functional residual capacity (FRC), impaired cough, rapid, shallow breathing, and often retention of secretions. These physiologic changes are particularly significant in thoracic surgery patients with preexisting pulmonary comorbidities, and may result in atelectasis, hypoxemia, and respiratory failure.2 Effective postoperative analgesia is of critical importance in these individuals. Nonetheless, effective treatment strategies for acute and chronic post-thoracotomy pain remain a significant challenge.3,4
The adverse effects of poor analgesia are not limited to the pulmonary system. Pain has been associated with increased myocardial oxygen demand, myocardial dysfunction, increased catecholamine release, poor glycemic control, deep vein thrombosis, and pulmonary embolism.5,6 These complications of inadequate pain control have been shown to lead to increased mortality and morbidity, prolonged length of hospitalization, and increased cost of patient care.7,8 In addition, several recent retrospective reviews suggest that a higher intensity of early (first week) postoperative pain is a risk factor for development of persistent pain.9,10
In this chapter, we will briefly review the mechanisms of thoracic pain. We will then discuss management strategies for acute postoperative pain and outline key concepts regarding chronic post-thoracotomy pain. The reader is referred to Chapter 6 for a more detailed discussion on the mechanisms of thoracic pain.
Nociceptive impulses from the thoracotomy incision, chest tube site, rib, muscle, and parietal pleural damage are transmitted along the intercostal nerves to the dorsal horn of the spinal cord. The autonomic nerves transmit noxious input from damaged visceral pleura. Pleural and bronchial manipulation may signal visceral pain through the afferent fibers of the vagus and phrenic nerves, which appear to be responsible for shoulder pain in some cases of acute and persistent pain.11,12
Tissue disruption triggers the release of inflammatory mediators such as prostaglandins, histamine, bradykinins, and potassium. These inflammatory mediators can directly activate nociceptors or enhance nociceptor activity. Furthermore, they cause a reduction in the pain threshold of the peripheral nerves. As a result, the severity of pain experienced by mechanical stimuli such as coughing or deep breathing may be intensified.13 This process is known as peripheral sensitization. Continued nociceptive stimulation will cause hyper-excitability of the nerves in the dorsal horn and other central pain centers, a process known as central sensitization.14 Central sensitization lowers the pain threshold of dorsal neurons and is associated with activation of NMDA receptors via substance P, calcitonin Gene Related Peptide (CGRP), and glutamate.15 Additionally, ongoing nociceptive stimulation alters neural function, resulting in neuroplasticity within the central nervous system. Despite the healing of tissue injury and the absence of inflammation, some patients continue to experience pain via central mechanisms (Figure 24-1).
Figure 24–1. Mechanisms of acute and chronic pain. (Modified from Pyati S, Gan TJ. Perioperative pain management. CNS Drugs. 2007;21(13):185-211, with permission from Adis. © Springer International Publishing AG 2007. All rights reserved.)
It is estimated that more than 70% patients undergoing thoracotomy experience moderate to severe pain,16 even when a minimally invasive surgical approach is employed. Appropriate treatment of pain is necessary not only to prevent discomfort but also to prevent other negative sequelae. Despite effective management of somatic pain, some patients also complain of ipsilateral shoulder discomfort, occasionally severe in nature. The etiology of this visceral component of post-thoracotomy pain is suggested to be from contributions of the phrenic and vagus nerves.17 The phrenic nerve supplies sensory branches to the diaphragmatic and mediastinal pleura and the pericardium. Although clinicians may have reservations about phrenic nerve infiltration and diminished pulmonary function, this does not appear to manifest clinically.17,18 When evaluating patients before surgery, it is important to determine whether they have preexisting pain in the area of the proposed surgery, to document the intensity of such pain, and to note any preoperative narcotic use. The presence of preexisting pain may indicate greater difficulty in management of postoperative pain, necessitating a more aggressive analgesic approach. Pain assessment is often performed using the visual analogue scale (0-10) or another more sophisticated pain assessment tool, such as the brief pain inventory or McGill pain questionnaire.
Because of the involvement of multiple pain generators from the chest wall, viscera, and the parietal, diaphragmatic and mediastinal pleura, post-thoracotomy pain can be difficult to treat with a single analgesic modality. An ideal analgesic combination should reduce the intensity of movement-related pain, decrease the surgical stress response, and reduce the duration of hospitalization. Several techniques are employed to manage acute post-thoracotomy pain. For the purpose of this discussion, it is most convenient to treat each modality separately and to bear in mind that a multimodal analgesic approach is the most effective management strategy to control moderate to severe pain19 (Figure 24-2). Multimodal strategies may include epidural analgesia, local anesthetic infiltration, opioids, nonsteroidal anti-inflammatory drugs (NSAIDs), and adjuvant medications (Table 24-1). Preoperative insertion of a thoracic epidural catheter is commonly performed to provide analgesia during the perioperative period. If an epidural catheter cannot be inserted, paravertebral blocks are the preferred analgesic technique. In a survey of post-thoracotomy pain management, approximately 80% respondents reported using epidural analgesia, with over 90% recommending a combination of a local anesthetic drug and opioids.20
Figure 24–2. Multimodal analgesia. Action of analgesics at various sites of the pain pathway. (Modified from Pyati S, Gan TJ. Perioperative pain management. CNS Drugs. 2007;21(13):185-211, with permission from Adis. © Springer International Publishing AG 2007. All rights reserved.)
Thoracic epidural catheterization is typically performed at mid-thoracic levels before induction of anesthesia. The paramedian approach to the epidural space is often the recommended technique because the obliquity of thoracic spinous processes makes the midline approach potentially difficult. Either a “loss of resistance” technique or “hanging drop” method may be used to identify the epidural space.
Most anesthesiologists are experienced with placement of lumbar epidural catheters using a “loss of resistance” technique. Although there are anatomic differences between the thoracic and lumbar spine, the fundamental technique of epidural space identification is consistent. Therefore, use of “loss of resistance” for placement of thoracic epidural catheters capitalizes on the anesthesiologist’s technical experience and knowledge.
“Hanging drop” technique entails placing a small amount of saline in the hub of the epidural needle after the stylet has been withdrawn. Upon entering the epidural space, negative pressure causes the meniscus of the fluid to disappear into the hub of the epidural needle. There is debate about whether the negative pressure created with the “hanging drop” technique is secondary to negative intrathoracic pressure, tenting of the dura with needle advancement, or both. The authors find the “hanging drop” technique performed in the sitting position particularly useful when working with trainees: withdrawal of fluid into the hub of the epidural needle provides an unambiguous visual end-point for the supervising physician.
Thoracic epidural local anesthetic administration has been shown to reduce lower extremity motor block as compared with lumbar placement.21 Additionally, given that the site of epidural needle/catheter placement determines the distribution pattern of neural blockade, there is a considerable theoretical advantage for thoracic catheter placement with a thoracic incision.22,23 Common adverse effects and complications of epidural analgesia include dural puncture, post-dural puncture headache, excessive motor blockade, and unsuccessful placement. Uncommon complications include epidural abscess, nerve injury, epidural hematoma, and paraplegia. Epidural analgesia is generally contraindicated in patients who are anticoagulated, septic, or have evidence of infection at the intended site of epidural placement. Guidelines for the management of epidural analgesia in the setting of anticoagulation are included in Table 24-2. Thoracic epidural catheter placement is often considered to be significantly riskier than insertion at the lumbar spine level due to concerns for possible spinal cord injury; evidence does not support this perceived increased risk.24,25
Table 24–2. Suggested Guidelines for Epidural Analgesia and Anticoagulation
Continuous infusion of local anesthetic and/or opioid into the epidural space is commonly used in the acute postoperative management of thoracic surgery patients. Compared with intravenous opioids, patients with epidural analgesia have superior pain control, less respiratory depression, and fewer pulmonary complications.26–28 In a comparative study of thoracic epidural analgesia versus systemic analgesia (PCA), there were well-demonstrated improvements in analgesia and quality of life in patients receiving epidurals.29 Investigations have not shown a dramatic difference in analgesia between the thoracic and lumbar route when opioid alone is used for epidural administration.30–32 Modest improvements in analgesic potency and clinical outcomes have been demonstrated with thoracic versus lumbar epidural administration of combined opioid and local anesthetic.27,28 The procedure-specific postoperative pain management (PROSPECT) working group recommends thoracic epidural or paravertebral blocks as first-line analgesic methods.33
A wide variety of opioids have been used in epidural analgesic regimens. The hydrophilic opioids morphine and hydromorphone have been shown to have similar efficacy when compared to postoperative analgesia.34 Hydromorphone has a favorable side effect profile with less respiratory depression, pruritus, and urinary retention when compared with morphine.35 Thoracic epidural administration of the hydrophobic opioid fentanyl was initially considered to have the advantage of segmental spinal analgesia, as this phenomenon is present during bolus administration. However, during continuous administration the analgesic effect of epidural fentanyl appears largely due to systemic absorption.36,37 The predominantly systemic distribution of epidural fentanyl infusion potentially makes this drug less desirable when other options are available. Nonetheless, patients with a history of pruritus, nausea, and other side effects with hydrophilic opioids such as morphine and hydromorphone may benefit from the choice of this lipophilic drug.38
Epidural administration of local anesthetic improves analgesia, but may produce sympathetic blockade, bradycardia, peripheral vasodilation, and hypotension. Sympathetic blockade is greater with thoracic epidural catheters than with lumbar epidural catheters,39 although it rarely requires removal of the local anesthetic from the infusion. The combination of epidural local anesthetic and opioid is noted to provide superior analgesia and reduction of side effects when compared with either drug class alone.40,41 Additionally, epidural local anesthetics have been shown to improve oxygenation and reduce pulmonary complications when compared with systemic analgesics (Table 24-3).26
Table 24–3. Drugs Used in Continuous Thoracic Epidural Analgesia
Patient controlled epidural analgesia (PCEA) has also been used in the delivery of thoracic epidural analgesia, and provides the ability to deliver effective analgesia with reduced opioid and local anesthetic doses.21,42 The advent of PCEA offers the promise of improved analgesia and patient satisfaction.43 PCEA has been shown to improve pain scores and cough as compared with continuous epidural analgesia and systemic analgesia.44 It allows patients to control pain by administering a bolus dose of local anesthetic and opioid mixture according to their individual need, tailoring the drug requirements to their activity level.45 A recent survey of PCEA with bupivacaine and hydromorphone demonstrated good analgesia without significant side effects in orthopedic patients.46 Although the literature regarding PCEA use for thoracic surgery is more limited, the side effect profile seems comparable to that of the lumbar space.47,48 The recommended dosing schedule for PCEA is included in Table 24-4.
Table 24–4. Recommended Dosing Schedule for PCEA
In patients with poorly controlled pain, the use of clonidine may be considered. Epidural clonidine (0.2-0.5 μg/kg/h) may improve symptom control, especially in patients with preexisting chronic pain states.49 Clonidine, an alpha-2 receptor agonist, has been found to provide analgesia, especially in neuropathic pain states.50 It appears to have analgesic activity in the periphery, the spinal cord dorsal horn, and the brainstem. Clonidine’s analgesic mechanism of action appears to predominantly involve spinal cholinergic activation and hyperpolarization of the primary afferent neuron.49,51 While it may be quite effective in potentiating epidural opioid analgesia, dosing may be limited by bradycardia, hypotension, and sedation.52
Ketamine has also been used in the epidural space to potentiate the analgesic effect of opioids53–55; however, its safety in the neuraxis needs to be established prior to recommending it for clinical use. Current scientific evidence supports the use of ketamine as an intravenous infusion for patients with preoperative opioid tolerance. A loading dose of 0.5 mg/kg on induction of anesthesia and a continuous infusion of 10 μg/kg/min during surgery and terminated at wound closure has been shown to significantly reduce morphine consumption at 24 and 48 hours as well at 6 weeks postoperatively in this population.56
Thoracic Paravertebral Blockade
Paravertebral blockade (PVB) is an alternative analgesic technique to thoracic epidural placement. Studies comparing surgically placed paravertebral catheters with thoracic epidural catheters have demonstrated similar analgesic benefit after thoracic surgery with a better side effect profile with paravertebral catheterization.57,58 Research involving percutaneously placed paravertebral catheters has been more limited to date, although there appears to be a similar analgesic result when this approach is compared with the surgically performed approach.59 The technique was first introduced in the early 1900s for analgesia following abdominal surgery; it was not until the 1970s that paravertebral blockade was commonly used for thoracotomies.60 It is well suited for treating post-thoracotomy pain because of the unilateral action of the paravertebral blocks. While single-shot injections performed at multiple levels may provide adequate analgesia, given the limited duration of analgesia obtained, catheter placement and continuous infusion may be preferable.
ANATOMY AND TECHNIQUE
The thoracic paravertebral space (PVS) is a wedge-shaped space that lies on each side of the vertebral column (Figure 24-3). Paravertebral blockade involves injection of local anesthetic in the vicinity of each spinal nerve root within the paravertebral space, resulting in unilateral anesthesia for thoracic surgery. The PVS connects medially with the epidural space via the intervertebral foramen and laterally with the intercostal space. The injectate may migrate to these spaces, but the magnitude of spread is unpredictable. On rare occasions, if the dural sleeve around the spinal nerve is inadvertently entered during the procedure, a subarachnoid injection may result. Injectate also frequently spreads both superiorly and inferiorly within the paravertebral space. While the spinal nerves course through the PVS as they exit the spinal canal, the sympathetic chain traverses the space anteriorly. The intercostal arteries, hemiazygos veins and lymphatics also pass through the PVS.
Figure 24–3. Transverse section of the thoracic spine depicting the boundaries, contents and structures surrounding the paravertebral space.
For thoracotomy, the levels blocked are usually between T4 and T9 when single-shot injection technique is used. Care must be exercised when counting the vertebrae before injection. Because of the steep angulations of the thoracic spinous processes, the tip of one spinous process corresponds to the transverse process and PVS of the vertebrae below (ie, T5 spinous process corresponds to T6 transverse process (Figure 24-4A and B). The thoracic transverse processes project laterally a mean distance of 3 cm from the midline. The needle should therefore be inserted 2½ cm lateral to the spinous process at each level. The mean depth from the skin to the PVS is 5.5 cm.
Figure 24–4. Landmarks for paravertebral block. A. Surface landmarks. The needle should be inserted 2½ cm lateral to the spinous process of each level to be blocked. B. Bony structure of the thoracic spine and ribcage. Note the steep angulations of the thoracic spinous processes. The tip of one spinous process corresponds to the transverse process and paravertebral space of the vertebra below. (Only part A reproduced with permission from Hadzic A: The New York School of Regional Anesthesia Textbook of Regional Anesthesia and Acute Pain Management. McGraw-Hill, Inc. 2007. Figure 43-6.)
For single-shot injection, the sites are labeled, the skin cleansed, and a 20-gauge Tuohy needle is advanced, seeking contact with the transverse process. The needle is then withdrawn and redirected caudally, advancing about 1 cm further than the distance to the transverse process until a tactile “pop” is noted as the needle passes through the costotransverse ligament into the PVS. Alternatively, a nerve stimulation technique (0.5-0.6 mA) is used to identify contraction of intercostal muscles as a result of spinal nerve stimulation. Three to four milliliters of local anesthetic (eg, 0.5% ropivacaine with 1:400,000 epinephrine) are administered after careful aspiration. The total volume of local anesthetic injected should be kept below the toxic dose when multiple injections are performed.
Continuous PVB can be instituted either by threading a catheter through an epidural needle after an initial bolus dose of local anesthetic or by placement of a catheter under direct vision by the surgeon.61 Various infusions have been used for paravertebral catheters; the reduced severity of sympathectomy often seen with the paravertebral technique may allow the use of more concentrated local anesthetics than are tolerable in the epidural space.58
The contraindications for PVB are similar to other regional anesthetic techniques and include local skin infection, coagulopathy, and hemodynamic instability.
ADVERSE EFFECTS OF PVB
The most common adverse event reported is technical failure. The published incidence of block failure ranges between 6% and 11%.62,63 Block failure consists of inability to provide adequate analgesia. Unintentional vascular puncture can occur in PVBs and should be recognized before injection of local anesthetic. Other common side effects of PVBs are mild hypotension, hematoma, or pain at the site of injection. Unintentional pleural puncture and consequent pneumothorax can occur with deep insertion of the needle. Therefore, it is necessary to watch for aspiration of air and/or cough during needle insertion. In a small proportion of patients, epidural spread of injectate can occur.