Introduction

Severe pain and chronic pain are frequent chief complaints among patients presenting for care. According to the Institute of Medicine, chronic pain affects over 100 million Americans and the socioeconomic costs of pain in the USA are estimated at over $500 billion per year. This means pain affects more patients than diabetes, cancer, and heart disease combined [1]. Beyond socioeconomic costs, the degree of morbidity and human suffering is immeasurable. Pain also has costs within our own clinics, particularly time, difficult and emotionally taxing patients, fear of bad outcomes, fear of lawsuits, and difficulty discerning drug-seeking patients from patients with real pain. Although pain is a common reason for healthcare visits, most health profession education programs provide minimal pain evaluation and management education, and unsurprisingly, many physicians feel unprepared to properly evaluate and treat these patients [2].

Within the medical community, a dogma that operating for pain will only cause more pain has led many pain specialists and primary care physicians to avoid referring patients with severe or chronic pain to surgeons [3, 4]. In many ways, surgeons share the guilt for perpetrating this dogma as many surgeons do not know how they can help or they do not want to take the time to thoroughly interview and examine these challenging patients. Consequently, many patients with pain caused by surgically manageable conditions are forced to either take chronic pain medications for life and deal with their side effects, or suffer. Recent literature indicates that surgical management of carefully diagnosed underlying pain generators results in pain improvement in 70–80% patients [5,6,7]. Regardless of surgery type, 10–30% of patients go on to have chronic pain and this pain is usually neuropathic [8, 9]. Understanding who is at risk for developing persistent pain, why, and how to prevent or treat it is an active area of research in many surgical specialties.

Therefore, the role of surgeons who treat peripheral nerve lesions and musculoskeletal complaints must be re-examined. We believe that the diagnosis of intractable, chronic, centralized pain should be a diagnosis of exclusion made only after ruling out treatable conditions of the peripheral nerves, bones, joints, muscles, and tendons. After all, patients with chronic pain frequently have or develop painful conditions amenable to treatment, like carpal tunnel, osteoarthritis, or peripheral neuropathy. The purpose of this paper is to highlight diagnosis and management of conditions that cause persistent, chronic, or severe pain and are amenable to surgical treatment. We also review the pathophysiology of pain and its implications on surgical management so that surgeons can be more active participants in a multidisciplinary pain team and non-surgeons understand what conditions might merit surgical referral. Many of the surgical techniques described in this paper are not new; however, growing understanding of the nature of pain, the frequency of neuropathic origins, and growing comfort with intraneural anatomy and dissection facilitated by the growing field of nerve transfers allows for better outcomes than in the past.

Pain Definitions and Biology

Pain is a physiologic response to a noxious stimulus intended to trigger protective behaviors and limit tissue damage [10]. The International Association for Study of Pain defines pain as an unpleasant sensory and emotional experience associated with actual or potential tissue damage. Perception of pain requires actualization of three stages: activation of nociceptive receptors, relay of the message to and through to the central nervous system, and pain response [11]. Unfortunately, like any finely tuned system, pain perception can become dysregulated. This process is known as central sensitization [12, 13]. The focus of this paper will be on identifying and stopping sources of nociceptive activation in the periphery, recognizing that modulation and perception of pain signals within the central nervous system also play an important role in pain experience and management. Methods to modify pain modulation (sympathetic blocks, neuroleptic medications, spinal cord, dorsal root ganglion, and peripheral nerve stimulation) and perception (relaxation, cognitive behavior therapy, psychotherapy, pharmacologic treatment of depression, anxiety, pain catastrophizing) will also briefly be discussed but only in the context of team-based care, multimodal therapy, pre-optimizing patients for surgery, and peri-operative pain management.

Peripheral nociceptive pain can result from injuries or inflammation of the musculoskeletal system or the peripheral nerves. Pain derived from injury or inflammation of the nerves is called neuropathic pain and often much more severe and distressing. Sustained input from the peripheral nervous system can lead to central sensitization, which results in amplification of pain signals and response within the central nervous system and manifests as more severe pain, often in response to non-painful stimuli and beyond the anatomic territory of injury. Physiologic and psychologic factors can predispose patients to central sensitization and heightened experience of pain. Therefore, it is important for all severe or chronic pain patients to have each of the three stages of pain be treated with the cooperation of specialists in the pharmacologic management of pain (pain medicine doctors), psychologic management of pain (psychiatrists, psychologists, and case managers), and management of musculoskeletal and nerve disorders (surgeons, physical medicine and rehabilitation doctors, physical and occupational therapists). This is ideally done within a team setting that facilitates close communication and coordination between various team members. At a minimum, the patient should be established with a pain management physician who will manage their pain medications and take over care when healing from surgery is complete. Surgeons should not serve as patient’s sole, or primary, pain management physicians.

The role of the surgeon within the pain team is to identify peripheral nociceptive and neuropathic sources of pain and eliminate or minimize them. These sources of pain fall into three categories: nerve compression, nerve injury (neuroma), and musculoskeletal injury/inflammation. Complex regional pain syndrome (CRPS) is a runaway pain syndrome with prominent physiologic changes frequently associated with nerve compression or neuroma that requires special attention. When seeing a patient with CRPS, nerve compression or neuroma should be suspected. A thorough discussion of sources of musculoskeletal pain is beyond the scope of this paper and not relevant to most readers of this journal. However, it is important to remember that musculoskeletal conditions are common and frequently co-exist with nerve injuries. Many of the same mechanisms that cause neuropathic pain such as swelling, inflammation, and muscle imbalance can also cause tendinitis, bursitis, or arthritis. It is typical for painful nerve injuries to result in postural modifications that then themselves generate secondary nerve compression. Consequently, the role of physical therapy to evaluate and treat posture-related extremity pain is critical and physicians that treat pain should not hesitate to refer patients with pain to physical and/or occupational therapy. Additionally, joint denervation is one treatment option in the algorithm for the management of arthritic pain [14,15,16,17,18,19].

Compression of a nerve can occur at any point along its trajectory from the spinal cord or dorsal root ganglion to its terminal end. There are more than 25 common points of nerve compression distal to the nerve roots, which can themselves be compressed as they exit the spine, in the upper and lower extremities and many more in the head, neck, and trunk (Table 1) [20,21,22,23,24]. Carpal tunnel is the most well-known and studied nerve compression syndrome. While anesthesia, dysesthesia, or pain are prominent symptoms of carpal tunnel, other compressions may present primarily with weakness or be subtler [25]. The physiology of nerve compression is the same at other points of compression [26, 27]. Changes from injury or compression at one point cause intraneural changes and edema along the nerve’s entire course, predisposing it to compression at other known points of compression. This is known as the double/multiple crush phenomenon [28,29,30].

Table 1 Common points of nerve compression in the upper and lower extremities resulting in pain
Full size table

Painful neuromas usually develop following trauma or surgery and affect 2–60% of patients with nerve injury [31,32,33,34]. Neuromas form after all nerve injuries. Why some patients develop painful neuromas and others do not is an active area of research and is incompletely understood. Neuromas can also have associated nerve compressions due to the double crush phenomenon. There are certain nerves that are commonly injured in different surgeries; physicians treating pain should be familiar with these and have a high index of suspicion for them if a patient presents with extreme pain that began shortly following an operation (Table 2).

Table 2 Sensory nerves at risk for painful neuroma after common surgeries
Full size table

CRPS is a characteristic runaway pain syndrome with prominent features of pain out of proportion to degree of tissue injury and physiologic changes that include edema, erythema, warmth, skin and nail atrophy, and abnormal sweating [6]. In the acute phase, these typically improve with sympathetic blockade although this response diminishes as the disorder becomes chronic. CRPS is divided into three categories: type I involves trauma but no identifiable nerve injury, type II has a neuropathic source, and type III has no apparent traumatic cause. Types I and III are rare [35]. Both neuroma and nerve compression can be causes of CRPS. CRPS is often caused by trauma, which can include surgery or simple immobilization. The role of the surgeon who typically is managing these patients at the time CRPS develops is to (1) diagnose it quickly; (2) refer the patient for sympathetic blocks, physical therapy, and pharmacologic pain management; and (3) correct any underlying musculoskeletal injury, nerve compression, or neuroma. Delaying treatment must be avoided as CRPS that persists beyond 1 year rarely resolves [36].

A diagnosis of CRPS is made according to the Refined Budapest Criteria (Table 3). Review of the diagnostic criteria reveals that CRPS has multiple underlying pathologies that are incompletely understood. These include changes in the sympathetic nervous system, neurogenic inflammation, autoimmunity, and changes within the central nervous system [37]. As such, multimodal team management is even more important for patients with CRPS. Notably, postural response to the pain of CRPS may cause or exacerbate nerve compression, including edema, and dystonia. This can further exacerbate and reinforce the pain response. For example, the characteristic protective position in which patients with CRPS guard their arm increase pressure on the nerves in the arm (Fig. 1).

Table 3 Refined International Association for the Study of Pain Budapest Diagnostic Criteria for complex regional pain syndrome [38]
Full size table
Fig. 1
figure 1

This patient demonstrates the typical guarded posture of complex regional pain syndrome (CRPS). Note that the shoulder is adducted and forward flexed causing muscle imbalance, shoulder pain, and traction on the brachial plexus at the middle scalene and axillary at the quadrangular space. The elbow is flexed compressing the ulnar nerve at the cubital tunnel. The forearm is pronated compressing the radial and median nerves at the supinator and pronator. The wrist is also flexed compressing the median nerve. Also note that the metacarpal–phalangeal joints are extended, not flexed in a clenched fist position. The authors thank Dr. Catherine Curtin, MD for allowing us to use this photograph.

Full size image

Examination of the Patient with Pain

The importance of careful and correct diagnosis when considering surgical intervention for pain cannot be overemphasized as the surgical management of nerve compression, neuroma, myofascial/muscle imbalance, or musculoskeletal pain are very different. In our experience, misdiagnosis and subsequent mismanagement can make a patient’s pain much worse. Patients with chronic pain often have a history of major trauma or multiple surgeries, or both. It is not uncommon for them to have multiple sources of pain. Therefore, a systematic approach to diagnosis is essential.

Standardized assessment tools that include a subjective assessment of patients’ pain level, quality, and location, as well as its impact on their function and psychological wellbeing, are very helpful. We like to use a form that we have customized which includes numerous visual analogue scales that are specific to each side, elaborates various time frames, includes an avatar for patients to mark the location of their pain, pain descriptors, questions about various factors that may intensify pain, and asks impact on their function (Fig. 2). Inspection of these forms often allows the physician to instantly identify the likely source of a patient’s pain. Neuropathic pain will frequently be drawn in a dermatomal distribution or along the course of a nerve, while musculoskeletal pain will be more localized. Numerous other validated measures are available for use (Table 4). We suggest pairing a simple assessment of pain, like the visual analogue scale with a more detailed pain and function questionnaire the DASH or PROMIS [39, 40]. Serial assessment with these forms facilitates tracking progress that is sometimes not apparent to the patients themselves.

Fig. 2
figure 2figure 2figure 2

Our preferred pain evaluation form includes multiple visual analogue scales of pain, depression, anxiety, and anger. It also includes an avatar for patients to mark their pain location(s) and numerous questions about relevant comorbidities

Full size image
Table 4 Common validated pain assessment scales
Full size table

History

A thorough history and timeline of patients’ pain is essential. Despite the subjective and individual experience of pain, different underlying pathologies will have typical histories and presentations (Table 5). It is important to take time to really understand the patients’ pain quality, trajectory, alleviating and aggravating factors, and specific distribution. Physicians should ask about when and how the pain started, the type of pain the patient is experiencing, how this has changed over time, and any event that caused that to change. Often, patients will present with a history of pain for which they had surgery, and that pain resolved but was replaced with much worse pain immediately after surgery. This should raise suspicion for iatrogenic nerve injury and subsequent neuroma. Symptoms that progress more gradually may indicate nerve compression or musculoskeletal pathology and are typically less painful than neurotmetic injuries (Fig. 3). Whenever available, previous imaging, electrodiagnostic studies, and operative reports must be thoroughly reviewed.

Table 5 Common sources of chronic pain and their characteristic presentation
Full size table
Fig. 3
figure 3

Diagrams of patients’ pain can help understand pain’s source. a Patients with neuroma will often describe an initial injury accompanied by pain with certain descriptive adjectives. They will then have a surgery and immediately develop much worse pain with new, different, often neuropathic-associated adjectives. Their pain level is often near 10 out of 10. b Patients with pain relating to compression neuropathy, muscle imbalance, or myofascial pain will describe a gradual worsening of their symptoms without a major change in pain adjectives. Giving patients colored pencils and asking them to draw their pain in this way, with different colors representing different pain, can be very informative

Full size image

Pain adjectives and associated descriptions can also help identify sources of pain. Electrical, burning, zapping, numbing, swelling and radiating, and stabbing are classically thought of as neuropathic symptoms while aching, pulling, smarting, deep pain is more often thought of as musculoskeletal. However, neuromas can also present as an aching, deep pain [35]. If patients describe multiple adjectives, we suggest asking them which ones are the most bothersome and working your way down the list from there to determine if the source of pain is likely neuropathic, or something else. In general, the more adjectives used, the more serious their pain response. Following surgery, the number of adjectives reported decreases [49]. Typically, pain from compression neuropathy is not as painful as that from neurotomy and surgery to treat a compression neuropathy is reliably successful.

Arthritic pain is usually worst in the morning and improves throughout the day, while neuropathic pain usually gets worse throughout the day. Allodynia is pain caused by a normally non-painful stimulus confined to a certain dermatomal or autonomous distribution. Dysesthesia is an unpleasant, abnormal sensation. Hyperpathia is an abnormally painful reaction to a stimulus that progresses slowly and beyond the distribution of any one dermatome. Allodynia, dysesthesia, or hyperpathia should raise suspicion of a neuropathic source of pain [6].

Patients’ complaints will rarely fit neatly into just one diagnosis. Cutaneous nerve distributions often overlap and plexus formation between nerves is common [50, 51]. One must bear in mind the distinct possibility of the double crush phenomenon with proximal nerve compression and distal nerve injury or vice versa [28,29,30]. It is also possible for patients with chronic pain to develop muscle imbalances, contractures, or arthritis due to altered and use movement patterns. Therefore, asking about activities, how a patient uses the affected extremity, and how the patient holds the extremity can be insightful. For example, we recently examined a patient after a bike accident who underwent numerous surgeries for a clavicle fracture, one of which dramatically increased his pain and resulted in radiating pain down his arm. Electromyography revealed a lateral cord injury. His neuropathic symptoms largely resolved; however, he continued to complain of a deep, dull ache in the shoulder and he was referred for “nerve pain.” History revealed that pressing on his anterior shoulder improved his pain. Rotator cuff evaluation was performed, and an anterior labral tear was diagnosed. Treatment of the labrum and rotator cuff improved most of his pain symptoms. Another example is a patient with a pan-plexus injury who recovered wrist and finger flexion with some degree of spasticity but no extension, resulting in him holding his wrist in constant flexion. He presented complaining of burning pain in his radial digits and requesting amputation. The distribution was not characteristic of root avulsions, thoracic outlet, nor proximal nerve injury. Nerve conduction studies revealed carpal tunnel syndrome that was easily managed with release of the transverse carpal ligament resulting in improvement of the patient’s pain.

Once we understand what we believe is the cause(s) of their pain, patient education about the anatomy of peripheral nerves, their dermatomes, and their function is imperative. This encourages patients to be active participants in their care and allows the patient to identify and confirm for you the specific nerve that is injured. We run through a differential diagnosis and what would be expected with each and let the patient to tell us if one sounds more like what they experience than the others. This discussion between patient and physician helps build trust, a strong physician–patient relationship, and can itself be therapeutic [52].

Although exceedingly rare, there are some factitious pain syndromes in which surgery must be avoided (Table 6). Histories in these patients will not clearly fit any combination of the previously mentioned pain sources and treatments that should provide some relief will often make these patients worse. Warning signs of factitious disorders include severe swelling, especially with a circumferential band mark proximal to the swelling, multiple ulcerations, a clenched fist, multiple wounds or scars, worsening or changing symptoms without apparent cause, lack of a clear diagnosis despite multiple previous tests, operations, or expert consultations, and a desire for further testing and treatment despite previous unsuccessful treatments [53]. Listening to your gut and paying attention to your feelings when interacting with these patients can help you identify these difficult, and sometimes hostile, patients [54, 55]. Remember that the pre-operative period is finite while the postoperative period is infinite. Having a psychologist or psychiatrist involved in a pain team can facilitate triaging these patients to their much-needed care.

Table 6 Factitious pain disorders to be avoided
Full size table

Beyond simple diagnosis of underlying sources of pain, it is incumbent on any physician treating pain to identify confounding factors that may intensify a patient’s experience of pain. These include depression, anxiety, sadness, stress, post-traumatic stress disorder, pain catastrophizing, drug-abuse, and a history of physical, sexual, or emotional abuse, especially in childhood [8, 56,57,58]. Failure to concurrently address these factors before an operation increases the chance of treatment failure. We ask most of these questions on our intake form (Fig. 2). Screening for pain catastrophizing can be accomplished using the pain catastrophizing scale [59].

Physical Examination

Objective findings of nerve or musculoskeletal injury that corroborate a patient’s history and symptoms are important before considering any operative intervention. Observation begins the moment the patient walks into your office and continues throughout the interaction. Note if the patient protects the injured extremity. If they complain of severe pain but use the extremity that should catch your attention and direct you away from surgery. Similarly, if they arrive with their extremity protected by a pillow or blanket that should indicate to you that they likely need other aspects of pain generation (sympathetic overactivity, pain catastrophizing, and other maladaptive behaviors) addressed before any consideration is given to surgery. Pay attention to their affect; lack of eye contact and hiding behind dark glasses and a hat are bad signs. Note what their spouse is doing. Ideally, they are interested and engaged but not speaking for, doting on, or otherwise enabling them. Conversely, if they are disengaged that is also a bad sign.

Physical examination of the patient with chronic pain requires empathy and buy-in from the patient. Before beginning, it is important to ask the patient if it is okay to touch them, their point of maximum pain, and if there are any areas that you should avoid. This information is very helpful in identifying trigger points of pain, the severity of pain, and narrowing the differential diagnosis. If a patient touches the dermatome of the suspected injured nerve, that suggests low-grade injury and need for decompression but not neuroma excision and repair or transposition. Apprehension or tears at the thought of being touched in this area, allodynia, and hypesthesia usually indicate higher grade nerve injury or central sensitization of pain that must be addressed. One should then begin with the most innocuous aspects of the exam like observation and sensory examination before progressing to potentially painful maneuvers like strength testing or provocative maneuvers. Areas of pain are saved for last.

Complete neuromuscular examination of the affected limb is attempted, and results are compared to the contralateral limb. First, observe for swelling, skin changes, warmth, sweating, and any scars. Sensation is tested with the ten test which compares patients’ perceived amount of sensation from 0 to 10 in the affected limb to the corresponding regions of contralateral normal limb before attempting 2-point discrimination [60]. Using a piece of ice in a similar fashion to assess abnormalities in cold perception is also used by some in detecting neuropathies [61]. If the patient will tolerate it, 2-point discrimination is then performed. Nerve distributions of the various cutaneous nerves or nerve root dermatomes are specifically tested and documented. Active and passive ranges of motion of the joints in the extremity are documented. Strength testing is then performed, specifically isolating muscles specific to the various nerves in the extremity proximal and distal to known points of compression. Provocative tests for points of nerve compression are then attempted (Phalen’s test for carpal tunnel, resisted pronation for pronator syndrome, etc.). The scratch-collapse can be very helpful in testing for proximal points of compression and determining which point among many is most significant in the case of multiple entrapment points along the same nerve [62,63,64,65,66,67,68]. Finally, the Tinel test is then performed [69, 70]. Start proximally to the area of suspected injury and progress distally along the nerve course. If you start at the point of maximum anticipated pain, you may provoke a significant pain response and end the exam. Inability to complete even innocuous portions of the exam indicates the need for aggressive medical and psychological management before any surgery is considered.

Once you have completed the physical examination, diagnostic blocks can be performed to test your suspected diagnosis and determine what effect decompression or neurectomy might have [67]. Blocks provide patients with an idea of the potential area of altered sensation with nerve resection but does not guarantee operative success [71]. Nerve blocks can usually be performed in a clinic, but occasionally require sedation or advanced imaging only available to pain specialists. Before any injection, objective measures such as VAS pain level, grip strength, pinch strength, and joint range of motion should be recorded. If a cutaneous neuroma is suspected, start with an injection of a small amount of local anesthetic very superficially within the scar. This should improve pain and function. Then, move to block the nerve(s) you suspect are injured in order of their suspected contribution to the patient’s pain. A proximal point along the nerve that is isolated from other nerves is chosen to allow assessment of that one nerve’s contribution. Wait an adequate time for the injection to have an effect and reassess the objective measures of pain and function before moving onto the next nerve, if more than one nerve is suspected to be involved. To prevent nerve injection injury, perform your injection proximal to the zone of previous injury and advise the patient to inform you of any symptom of nerve irritation. Ultrasound may also be used in experienced hands. Carefully note changes in the patients’ reported pain level, function, and behavior after each block. Failure to improve pain with nerve block should dissuade you from operating.

If at this point you are unsure of any aspect of your examination, bring the patient back for repeat examination. Additional imaging can also be ordered but is not necessary if you have a convincing story and physical examination. Nerve conduction studies are the gold standard diagnostic modality for nerve entrapment, and should be obtained in most patients, provided they can tolerate the exam [72]. However, electrical studies only measure myelinated axons and therefore can miss compression only affecting small pain fibers, reducing their sensitivity in patients in whom the predominant complaint is pain [73, 74]. Ultrasound has become a widely available tool for assessing peripheral nerves but should not be substituted for careful physical examination [75, 76]. It has the added benefit of being non-invasive and allowing the examiner to correlate the patient saying where the point of maximum pain is with the area of imaging [67]. In our experience, ultrasound is highly operator dependent and not particularly sensitive for nerve compression or neuroma [77]. MR neurography has also become an accepted method for evaluating compression or injury of the peripheral nerves and has the added advantage of giving high-resolution imaging of the surrounding anatomy and associated inflammation [78,79,80]. Unfortunately, in many centers, it is not readily available.

Surgical Management

Pre-Optimization

In our opinion, the best way to treat chronic pain is to prevent it from happening in the first place. This can be accomplished by judicious patient selection, recognizing and treating risk factors for chronic pain before, during, and after surgery, and recognizing patients who are not improving as expected early, so that adjuvant interventions can be implemented. About 20% of patients develop chronic pain after any surgery and pre-operative pain intensity and pain catastrophizing are two of the strongest risk factors for developing chronic pain [8, 81,82,83]. Management of these risk factors is again facilitated in a team setting. Study of the effect of pre-operative psychological intervention is a promising area of research [56].

In patients with CRPS, it is our practice to defer treatment until the disorder is at least partially controlled with the use of pre-operative sympathetic blocks and pain medications. Like the treatment of migraines, there are numerous medications that have been shown to help treat different facets of CRPS and selection of these agents is patient specific and directed by the patient’s pain physician. At our institution, most patients with chronic pain are on a tricyclic antidepressant and gabapentin or pregabalin prior to any surgery. Vitamin C has been shown in four randomized trials to reduce the risk of CRPS following trauma or surgery and we suggest that they take 500 mg or more daily [84].

Peri-Operative Care

The patient pain diagram is reviewed with the patient on the day prior to surgery to reconfirm the diagnosis, assess pre-operative pain management, and solidify the surgical plan. If operating for neuroma, we have the patient mark their point of maximum tenderness and any trigger points with a permanent marker. After undergoing the induction of general anesthesia, patients receive a pre-operative IV regional block (Bier) with lidocaine and dexmedetomidine [85,86,87,88]. If this is not possible because we are not operating on a limb amenable to tourniquet application, a pre-operative proximal nerve block or epidural is performed. In this case, the block should be done with the patient awake and advised to declare any nerve symptom during block placement. A catheter is left in place for post-operative analgesia. We never perform a nerve block in a sedated patient.

When operating for compression, especially in the setting of pain, you get one best chance. Minimally invasive, limited access incisions should be avoided. It is important to get wide exposure to optimize your ability to visualize and protect the nerve. If performing revision surgery, it is important to identify the nerve within normal anatomy proximal and distal before entering the scarred and distorted anatomy of the previously dissected field. All possible points of compression within the field should be visualized and released. This includes any new points of kinking or tightness if you are transposing the ulnar nerve [89]. Neurolysis of the nerve from surrounding scar or synovitis is performed. The nerve should lay loosely within a healthy wound bed, ideally protected deep within the surrounding muscles. The senior author’s preferred techniques for nerve decompression are available at https://surgicaleducation.wustl.edu, in her book “Nerve Surgery” [90], or in “Green’s Operative Hand Surgery” [27].

When operating for neuroma, it is rarely necessary to first expose the neuroma within the scarred, previously operated field. Rather, exposure of the nerve proximal is much easier and safer. The “tug test,” wherein you gently pull the nerve, will demonstrate if you have the correct nerve. In this case, you will see the trigger point or point of maximum pain marked by the patient retract. Prior to making any cut to the nerve, we crush the nerve as proximally as possible within the operative field, creating an axonotmetic injury (Sunderland II), to set the point front of regeneration far away from the cut nerve end [49]. Our clinical experience in patients with neuropathic pain suggests that neurotmetic injury (Sunderland IV, V, or Mackinnon VI) is much more painful than an axonotmetic injury. We believe that this proximal crush allows the brain to experience only one less severe insult (if any given the regional anesthesia) and minimizes or eliminates the brain’s perception of the distal cut or neuroma excision that is performed.

In our experience, it is always best to reconstruct a painful nerve injury if possible. We prefer to reconstruct mixed and motor nerves with autograft whenever possible [91]. We will also reconstruct critical sensory nerves with autograft. Non-critical sensory nerves with a small diameter and gap less than 3 cm are reconstructed with acellular allograft [92, 93]. When selecting autograft, we prefer to borrow from the injured nerve to avoid creating new areas of anesthesia and potentially painful new neuromas [94]. If this is not possible, we will use the medial antebrachial cutaneous nerve if already operating in the upper extremity, or the sural nerve if operating in the lower extremity or depending on patient preference. Whenever harvesting a nerve autograft, we prefer to leave enough length at the distal end of the nerve perform an end-to-side coaptation to a nearby non-painful cutaneous nerve. This allows spontaneous sprouting of sensory axons to return some sensation to the donor nerve dermatome and we believe this prevents potentially painful sprouting from pain sensitized adjacent dermatomes [50, 95].

If the nerve affected by neuroma is not critical and reconstruction is impossible or impractical, then our preference is to perform a proximal crush, followed by transposition of the cut nerve end deep within muscle [71]. In patients with CRPS, keratinocytes at the skin have been shown to have altered gene expression and to secrete pain-inducing chemokines [37]. Therefore, our preference is to move the nerve into a deep muscle with minimal excursion so that stimulation of the cut nerve end is minimized. In recent years, we have also been “capping” the nerve end with a 5-cm nerve allograft in the hopes of arresting axonal regeneration [92, 93]. Our group recently performed a meta-analysis of surgical neuroma management that found no difference in outcomes among the various ways of managing neuroma so long as the neuroma was excised. However, transposition or coverage with a flap to move the neuroma away from the skin was significantly better in patients who had 2 or more prior neuroma operations, or who had pain lasting longer than 2 years [5]. Our own clinical experience with neuroma crush and deep transposition shows improvement in most patients [7, 49, 96].

Recently, targeted muscle reinnervation (TMR) and regenerative peripheral nerve interface (RPNI) procedures have been reported to have positive outcomes for the treatment of neuroma pain [97]. These two techniques were developed to allow volitional control of muscles, or pieces of muscles, in amputated limbs by cut nerves that normally power the amputated body parts. The idea is that a nerve signal is converted into a muscle movement that can be recorded and used to direct a motor-powered prosthesis. Proponents of these techniques suggest that giving the nerve end an active target will improve neuroma outcomes. To date, published literature of the effectiveness of these techniques consists of only very small studies whose outcomes do not differ significantly from other published techniques [5, 98, 99]. However, we suspect that TMR, which involves an end-to-end coaptation of the cut nerve to a motor branch into muscle on the deep surface of that muscle, essentially transposing the cut nerve end under a muscle, will likely produce better results in amputees than traditional methods of nerve management in amputation stumps (traction neurectomy or suture ligation, both techniques used elsewhere to produce pain in rat pain models [100, 101]).

Post-Operative Care

At the time of incision closure, a catheter that infuses Marcaine is inserted in the operative bed to provide ongoing anesthesia for 2 days. A very lose, bulky splint is placed to protect and immobilize them. Patients are admitted for overnight observation, ensuring adequate pain control. They are placed on scheduled acetaminophen and ibuprofen and their home pain medications are also continued. On post-operative day 2, we see them back in the clinic, remove their pain catheter, and remove their dressing. They are sent immediately to physical therapy to commence gentle stretching, range of motion, and nerve gliding exercises [102]. Issues of pain exacerbation and narcotic seeking are uncommon in our carefully selected and pre-optimized surgical patients. In fact, over half of the patients chronically using opioid pain medications that we operate on for pain have stopped taking those medications 3 months after surgery [7]. We encourage patients to take time to relax, meditate, practice deep breathing, continue psychological therapy and psychiatric treatment, get adequate sleep, eat a healthy diet, exercise, perform their physical therapy exercises, be active, and socialize. We believe that giving them lots of self-care activities to perform provides, if nothing else, some illusion of control over their pain and allows them to become more active participants in their recovery.

Conclusions

In conclusion, chronic intractable pain should be a diagnosis of exclusion made only after treatable sources of neuropathic and musculoskeletal pain have been ruled out by surgeons and physicians experienced in treating those problems. This is best done in the setting of a pain management team that includes surgeons, pain management doctors, psychiatrists or psychologists, physical or occupational therapists, and case managers. With careful patient examination, selection, and pre-operative optimization, pain can be clinically significantly improved in most patients. Mastering the management of patients with pain improves your ability to treat, identify and treat your own patients who may be developing persistent pain. It can also be quite rewarding, as you will never have a patient as grateful as the one you rescue from devastating, life-altering pain.