Role of Rehabilitation Medicine and Physical Agents in the Treatment of Cancer-Associated Pain

Abstract

Purpose

To provide an overview of rehabilitation medicine– and physical modality–based approaches to cancer pain management, and to highlight the fact that these approaches are generally used in conjunction and that a majority are focused on minimizing pain during periods of mobility and the performance of activities of daily living.

Methods

We performed a nonsystematic literature review and provide a description of the current standard of care.

Results

Rehabilitative and physical modalities used to manage pain can be grouped into four categories: those that modulate nociception, stabilize or unload painful structures, influence physiological processes that indirectly influence nociception, or alleviate pain arising from the overloading of muscles and connective tissues that often occurs after surgery or with sarcopenia in late-stage cancer. Most modalities have been pragmatically refined over the years, and many have an evidence base, although few have been explicitly validated in the oncologic setting. With few exceptions, they are patient controlled and free of adverse effects.

Conclusion

Physical modalities and rehabilitation medicine offer a range of pain management approaches that may serve as beneficial adjuncts to the conventional systemic and interventional analgesic strategies used to control cancer-related pain. These approaches may be particularly beneficial to patients with movement-associated pain and those who are ambivalent regarding pharmacoanalgesia.

INTRODUCTION

Pain is a frequent aspect of cancer, for which treatment is often less effective and associated with more adverse effects than we would like. All pain is limiting, but bone pain is particularly problematic in that it is aggravated by movement and therefore has profound effects on an individual's mobility and independence.

Rehabilitation medicine, with its focus on optimizing function despite a patient's symptom burden or impairments, involves strategies that may be beneficial in pain of all types, but they are disproportionately targeted toward pain associated with movement. These approaches are often used in combination and, in almost all cases, serve as adjuncts to, rather than replacements for, conventional analgesic care. Most have been pragmatically refined over the years, and many have an evidence base. Although many have not been explicitly validated in the oncologic setting, common sense and extensive experience argue persuasively for their clinical effectiveness. In addition, with few exceptions, they are patient controlled and largely free of adverse effects.

Physical modalities used to manage pain can be grouped into four categories: those that modulate nociception, stabilize or unload painful structures, influence physiological processes that indirectly influence nociception, or alleviate pain arising from the overloading of muscles and connective tissues that often occurs after surgery or with sarcopenia in late-stage cancer. Table 1 lists the findings of systematic reviews since 2005 for the modalities discussed in this article, with ratings of evidence described according to GRADE¹ and/or Cochrane levels of evidence, depending on the approach used. The evidence for specific body parts is listed separately, because systematic reviews seldom conflate findings from different anatomic regions. It will be noted that many modalities are used in conjunction and that a majority are focused on minimizing pain during periods of mobility and the performance of activities of daily living.

Table 1.

SRs and MAs Since 2005 of Rehabilitation and Physical Modalities Used to Manage Pain, Described in Terms of Grade and Cochrane Level of Evidence

Condition	Grade	Cochrane	No. of SRs or MAs	Date of Most Recent Report	Evidence in Cancer	Authors
	Modulation of Nociception
Topical heat					No
LBP		SE	1	2006		French et al1
Shortwave diathermy					No
LBP	Very low		1	2007		Chou and Huffman2
Ultrasound					No
LBP	Very low	IE	3	2011		Chou and Huffman,2 Seco et al,3 Poitras and Brosseau4
Shoulder	Low		1	2010		Alexander et al5
Cold					No
LBP	Very low		1	2006		French et al1
Knee pain	Very low		1	2011		Lake and Wofford6
Transcutaneous nerve stimulation					Robb et al7*
LBP	Low	IE	3	2009		Poitras and Brosseau,4 Machado et al,8 Khadilkar et al9
Knee	Moderate	IE	2	2009		Rutjes et al10
Acute nociceptive pain	Moderate	IE	2	2014		Simpson et al,11 Walsh et al12
Neck		IE	1	2013		Kroeling et al13
Neuropathic pain	Moderate		3	2010		Jin et al,14 Mulvey et al,15 Dubinsky and Miyasaki16
Fracture		IE	1	2011		Abou-Setta et al17
Chronic	Moderate	IE	2	2008		Johnson and Martinson,18 Nnoaham and Kumbang19
Interferential current therapy
Musculoskeletal	Low		2	2010	No	Poitras and Brosseau,4 Fuentes et al20
	Stabilization and Unloading of Painful Structures
Compensatory strategies and adaptive devices
Hand arthritis	High		1	2010	No	Valdes and Marik21
Orthotics					Lee et al22*
Back		IE	2	2012		van Duijvenbode et al,23 Longo et al24
Hand arthritis and CTS	High	SE	2	2012		Valdes and Marik,21 Page et al25
Knee	Low to moderate	IE	4	2012		Swart et al,26 Raja and Dewan,27 Brouwer et al,28 Beaudreuil et al29
	Influence on Local Physiologic Processes Affecting Nociception
Laser and light therapy					Bensadoun and Nair,30 Bjordal et al,31 Clarkson et al32†
Arthritis	Moderate	SE	3	2011		Brosseau et al,33 Ye et al,34 Jamtvedt et al35
LBP		IE	3	2011		Chou and Huffman,2 van Middelkoop et al,36 Yousefi-Nooraie et al37
Neck	Moderate to high		> 7	2013		Graham et al,38 Leaver et al,39 Gross et al,40 Kadhim-Saleh et al,41 Chow and Barnsley,42 Chow et al,43 Hurwitz et al44
Orofacial	Low		2	2013		He et al,45 Petrucci et al46
Manual lymphatic drainage	Low		1	2009	No	Vairo et al47
	Reduction of Pain-Associated Muscle and Connective Tissue Pathology
Corticosteroid injections					No
Knee	Moderate to high	SE	3	2012		Cheng et al,48 Bellamy et al,49 Hepper et al50
LBP	Moderate to high	IE	3> 7	32013		Ammendolia et al,51 Quraishi,52 Benny and Azari,53 Roberts et al,54 Buenaventura et al,55 Staal et al56,57
Shoulder	Moderate to high			2009		Gaujoux-Viala et al,58 Arroll and Goodyear-Smith,59 Koester et al60
Trigger-point injections and dry needling					No
Neck		SE	1	2007		Peloso et al61
Nonspecific musculoskeletal	Low		2	2009		Scott et al,62 Tough et al63
Massage					Ernst,64 Wilkinson et al,65 Bardia et al66*
Shoulder	Low to moderate		5	2013		Koog et al,67 Kong et al,68 Ho et al,69 van den Dolder et al,70 Verhagen et al71
Neck	Low to moderate	IE	> 7	2013		Kong et al,68 Verhagen et al71 Brosseau et al,72 Patel et al,73 Bryans et al,74 Furlan et al,75 Vernon and Humphreys,76 Ezzo77
LBP	Low to moderate	IE	> 7	2013		Chou and Huffman,2 van Middelkoop et al,36 Brosseau et al,72 Furlan et al,75 Bronfort et al,78 Kumar et al,79 Furlan et al80,81
Therapeutic exercise					Carvalho et al82‡
Hip	Moderate to high	SE	4	2013		Gill and McBurney,83 Fransen et al,84,85 Zhang et al,86 Hernández-Molina et al,87 Bartels et al88
Knee	High		> 7	2014		Gill and McBurney,83 Juhl et al,89 Uthman et al,90 Tanaka et al,91 Wang et al,92 Gill SD 2013; Smith et al,93 Escalante et al94
Shoulder	Moderate to high		> 7	2012		Hanratty et al,95 Kromer et al,96 Littlewood et al,97 Brudvig et al,98 Marinko et al,99 Kuhn,100 Smidt et al101
Neck	Moderate to high	SE	4	2014		Leaver et al,39 Hurwitz et al,44 O'Riordan et al,102 Kay et al,103 Miller et al104
LBP	Moderate to high		6	2010		Chou and Huffman,2 van Middelkoop et al,36,105 Hayden et al106–108
Manipulation					No
Shoulder	Low to moderate		4	2011		Brantingham et al,109 Maund et al,110 Pribicevic et al,111 McHardy et al112
Spine (neck and low back)	Very low to moderate	IE to SE	> 7	2013		Miller et al,104 Rubinstein et al,113–116 Posadzki,117 Goertz et al,118 Smith et al,119 Huisman et al,120 Vincent et al,121 Gross et al122

Abbreviations: CTS, carpal tunnel syndrome; IE, insufficient evidence to recommend; LBP, low back pain; MA, meta-analysis; SE, sufficient evidence to recommend; SR, systematic review.

^*IE.

^†Moderate.

^‡High.

Modulation of Afferent Nociceptive Activity

Rehabilitation uses two approaches to modulate the input of nociceptive signals into the CNS. The first, the use of heat and cold, which is traditionally associated with rehabilitation medicine, is being de-emphasized but still has a roll. The second, epitomized by transcutaneous electronic nerve stimulation (TENS), uses benign afferent sensory input to modulate nociceptive activity, rather than attempting to block the input of painful stimuli. This latter approach, based on the gate theory of pain of Melzack and Wall,¹²⁴ has been challenged but never seriously debunked, even though the precise neural pathways and biochemical reactions continue to be studied.

Heat and Cold

Heat and cold have clear effects on a variety of physiologic processes. As such, their main use has been for the control of pain, typically focused on the musculoskeletal system, where there is support for benefit of heat and cold as adjuncts to exercise.^1,2,125,126 Metabolic and enzymatic processes are remarkably temperature sensitive; changes of only a few degrees Celsius are capable of altering nerve conduction, blood flow, and collagen extensibility.^127–130 Systemic temperature changes of 0.3 to 0.4°C are possible in the clinical situation. However, local effects, as might be expected, are far more pronounced.¹³¹ Ice massage, for example, can reduce knee intra-articular temperatures by as much as 6°C,¹³² and ultrasound and short-wave diathermy can increase deep-tissue temperatures by similar amounts.¹³² Although the heating agents differ, most gain their effects by inducing analgesia or hyperemia or reducing muscle tone. Research into the effectiveness of heat and cold in the clinical setting has been limited by the variety of conditions studied and the specifics of the techniques used. However, the evidenced-based consensus supports the longstanding belief of the clinician that although heat and cold in and of themselves can be beneficial, in almost all cases their benefits are larger when combined with a well-planned program of exercise and mobilization.^133–135

Electrical Stimulation

Electrical stimulation is used for a variety of indications that range from functional electric stimulation, where stimulation is used to assist a patient to move an impaired limb, to analgesia and even the healing of soft tissue injuries and fractures. However, this article is restricted to its analgesic applications. A number of electromagnetic approaches to pain control have been developed and continue to be studied in the field of rehabilitation. TENS is the best studied and most widely used of these agents, and as such, it is discussed here.

TENS was introduced in the early 1960s as a noninvasive means to provide the afferent sensory stimuli posited to block nociceptive signals.¹³⁶ A few successful trials led to its rapid acceptance. However, acceptance and use have not completely clarified its benefits or best means of application.

TENS units are typically small and programmable with ≥ one signal generator. Although an infinite number of stimulation parameters are possible, all choices tend to involve output currents on the order of 100 mA, pulse rates < 100 to 120 Hz, and pulse widths from 10 to a few hundred microseconds. A variety of waveform modulation patterns are used, with the goal of increasing its effectiveness and comfort.

Electrodes are often placed over the painful area; however, positioning over the superficial portions of afferent nerves and acupuncture points is often trialed as well. Two stimulation approaches are the most common. In the first (ie, low-intensity or conventional TENS), stimulation is set at approximately 40 to 80 Hz and is barely perceptible by the patient (ie, benign afferent signal of gate theory) and hence generally more comfortable. The second, which some consider a counterirritant approach, is in many ways the reverse; frequencies are relatively slow, (ie, 1 to 8 Hz), and the intensity is moderately uncomfortable and must be tolerated for 20 to 30 minutes.

Response is difficult to predict, and TENS studies range in quality from well-designed, prospective, randomized controlled trials to (particularly in its earlier days) small inadequately blinded trials. Even today, trials comparing TENS with active controls remain rare.

Many of the earlier studies focused on postoperative and early labor pain and found that TENS use resulted in benefits comparable to those of limited amounts of narcotics.^135–137 Subsequent research has yielded more mixed results, with recent evidence-based clinical guidelines and systematic reviews finding little evidence that TENS can lessen neck or back pain.^138–141 The situation may be somewhat more positive for knee osteoarthritis.

Cancer-related pain has thus far received only limited attention. For example, although theoretic arguments can be made that TENS is capable of improving movement- or weight-bearing–associated cancer pain, a recent Cochrane review, despite casting a wide net for acceptable studies, found that only three met its criteria for inclusion. Issues with design heterogeneity and quality were noted, and although treatment was well tolerated by participants, the authors were unable to conclude that the evidence was strong enough to support the use of TENS.¹⁴²

TENS relies on repetitive trains of stimuli, which raises concern that benefits may lessen with time as a result of habituation and tachyphylaxis. Although many have tried to avoid this issue by producing waveforms of varying shape, frequency, and packet size, newer approaches using rather intense stimuli with randomly varying waveforms may avoid this issue and become more effective than TENS, particularly for neuropathic pain.¹⁴³

TENS has few safety issues other than skin irritation and mild discomfort during use. Cardiac pacemakers seem relatively resistant to TENS signals, but reasonable concerns about real or apparent introduction of dysrhythmias or malfunction restrict its use in that setting. It also seems prudent to avoid treatment near the carotid sinus and epiglottis and on the abdomen or low back of pregnant women.

Why have TENS units continued to be used despite equivocal evidence of effectiveness? The reasons may help oncologists in deciding whether to consider TENS for their patients. First, TENS assessment is limited by the myriad of differing conditions and stimulation settings reported. Thus, although systematic reviews have found limited evidence of effectiveness, the old adage that “the absence of evidence of efficacy does not necessarily mean evidence of absence of efficacy” holds true, and they have been unable to state that the approach is ineffective. Second, adverse effects are minimal. Third, efficacy in both nociceptive and neuropathic pain syndromes—a mixture of which is experienced by most patients with cancer^144,145—is to some extent supported by reduced patient pain ratings. Fourth, a TENS trial can be incorporated into a course of physical therapy (PT) without significantly interfering with other potentially beneficial activities (eg, therapeutic exercise). Lastly, the prolonged use of TENS by a subgroup of patients suggests that it may benefit certain individuals. In all, reasonable candidates are patients whose localized pain is inadequately controlled by conventional treatments, who experience untenable medication adverse effects, or who prefer to try nonpharmacologic approaches.

Counterstimulation and Desensitization Techniques

There are a variety of other rehabilitation techniques designed to increase benign afferent sensory drive to attenuate pain intensity. Some, such as desensitization, have this as their sole objective. Others (eg, massage and compression garments) increase afferent sensory drive in conjunction with other treatment goals, such as the control of edema or decreasing muscle tone. Desensitization techniques in which tolerance for increasingly intense stimuli (initially benign and subsequently noxious) is systematically cultivated warrant particular mention, because they are a first-line rehabilitation medicine treatment for neuropathic and complex regional pain syndromes, including chemotherapy-induced peripheral neuropathy.¹⁴⁶

STABILIZATION AND UNLOADING STRATEGIES

Cancer is often associated with a decreased ability of the body to bear weight, move, or tolerate the forces placed on it by even routine activities. As a consequence, the limbs, spine, and muscles frequently become pain generators, particularly with movement. The most prevalent example is metastatically induced bone pain. However, even intact musculoskeletal elements can become secondarily painful as a consequence of the biomechanical changes induced by cancer and its treatment.¹⁴⁷

Four approaches are used to reduce the forces placed on painful bony or connective tissues: displacing painful forces onto external supports, improving the load-bearing capacity of intact anatomic elements, immobilizing painful joints, and reducing the physical effort required by an activity. It should be noted that although we present activities separately for the sake of clarity, their application in clinical treatment generally involves an integrated combination of approaches.

Assistive Devices for Mobility and Performance of Activities of Daily Living

Patients with cancer rank the maintenance of independent mobility as among their most important concerns, and loss of functionality is strongly associated with depression and desire for hastened death.^148,149 A wide range of devices are available to enhance the safety and autonomy of a patient's mobility. Each, even the single-point cane, has its own strengths, limitations, and fitting requirements. Although loads can be reduced by ≥ 30% with these devices,¹⁵⁰ there are times when mechanized or electrical alternatives may be required. For example, a Hoyer lift, scooter, or wheelchair may free a patient from pain during transfers and locomotion. The need for the guidance of a physical therapist and the importance of professionally supervised trials and fitting in all but the simplest situations cannot be overstated.

A diverse array of assistive devices can be used to protect inflamed or otherwise vulnerable structures from activities of daily living (ADL) –related forces. Figures 1A to 1D show a variety of possibilities. In contrast to assistive devices for mobility, those directed toward maintenance of ADL performance are generally designed to reduce the amount of reaching, bending, or twisting required to complete an activity. Often even a brief session with a knowledgeable occupational therapist and a few simple devices (eg, bath bench, built-up utensil) can preserve independence and reduce activity-related pain for a remarkable length of time.

Fig 1.

Assistive devices for performance of activities of daily living: (A) zipper pull, (B) dressing stick/sock aid, (C) weighted button aid, and (D) sock donner.

Compensatory Strategies

Compensatory strategies use the same principle as assistive devices, offloading the forces required to perform a painful activity. In fact, these strategies so often rely on assistive devices that it becomes almost contrived to separate them. Nonetheless, there are important distinctions, which, if appreciated, may help clinicians generate more effective and comprehensive PT and occupational therapy plans.

Our daily activities consist of orchestrated combinations of coordinated movements. Often, only a limited number of the movements required to execute an activity produce pain. By deconstructing painful activities to their constituent movements, physical and occupational therapists can isolate those that are painful and devise alternative, compensatory strategies to achieve a patient's goal. Simply helping patients recognize the movements that trigger pain frequently enables them to develop their own strategies. Activities that can be deconstructed in this manner vary widely, from getting into or out of a vehicle, to using the toilet, to putting on a bra. Engaging caregivers, using architectural supports and durable medical equipment, and altering the home environment can be combined to optimize patients' nonpainful functioning.

Therapeutic Exercise

Therapeutic exercise, often in conjunction with other modalities and designed to assist in obtaining specific goals, may be the most effective treatment in the rehabilitation armamentarium. For example, although there may be specific restrictions to the approach (eg, unstable spine), muscles are dynamic and often provide the most effective means of stabilization and immobilization. In fact, their use in this manner, often termed dynamic stabilization, has been a mainstay in sports medicine and the treatment of low back pain for years. As a result, therapeutic exercises aimed at enhancing the strength and stamina of the core musculature and muscles capable of splinting a painful body part can be remarkably effective adjuncts to conventional analgesia. Furthermore, consideration of the translation of these techniques into the realm of cancer rehabilitation seems warranted, given the promising results of unloading painful areas and mobilizing patients with metastatic disease.¹⁵¹

Therapeutic exercises should be chosen and implemented by a therapist with the skills and time to design a program that can effectively splint or constrain the movement of pain-generating structures. For the most part, muscles should be strengthened in a fixed position through isometric contractions that avoid pain-producing positional changes. Common examples of exercises used to stabilize pain-generating bony structures include isometric strengthening of the abdominal and hip abductor muscles to unload painful vertebrae and hip joints, respectively.¹³³

Orthotics

Orthotics are used to stabilize, unload, and protect compromised musculoskeletal structures. Orthotics that perform this function come in many forms, but typically, they either immobilize the entire affected body part or apply pressure at select points to restrict the motion of a specific joint or joints. Many are commercially available, whereas others may require custom construction. Those used with the goal of avoiding pain in cancer care are, in large part, directed at stabilizing the neck, trunk, and low back. However, orthotics for the extremities, such as molded-ankle foot orthoses and spica or hand and wrist splints, may benefit patients with distal limb pain.

Spinal braces warrant specific mention. Often their prescription is initiated and coordinated by an orthopedist or neurosurgeon when a patient's spine is deemed unstable. Molded body jackets, which require custom fabrication, are most frequently prescribed in these cases because of their superior immobilizing properties.¹⁵² Although body jackets unquestionably limit motion better than other spinal orthoses, they are costly, hot, uncomfortable, and poorly tolerated. In these cases, it may be necessary to consider a less ideal alternative, such as the semirigid braces that are often commercially available and are also capable of restricting spinal motion.¹⁵³

Less expensive, and more tolerable, alternatives include modular and prefabricated orthoses that encompass different segments of the spine. Thoracolumbosacral orthoses, such as the widely available CASH and Jewett braces (Figs 2A and 2B), are generally well tolerated. These three-point braces apply pressure at two points on the anterior upper and lower trunk and at a third posteriorly on the midback. A nice attribute of these off-the-shelf orthoses is that they can be tried to assess their benefits without the need for an expensive investment in either purchase or custom fabrication. It should be noted that although these braces do limit spinal flexion, they do little or nothing to constrain truncal extension. However, because the anterior vertebral column is the most common site of metastatic involvement,¹⁵⁴ anterior support is what is needed. However, if the posterior columns of the spine are unstable, surgical consultation and a molded body jacket should be considered.

Fig 2.

Prefabricated thoracolumbosacral orthoses to limit spinal flexion: (A) CASH and (B) Jewett braces.

Lumbosacral orthoses (LSOs) control spinal motion from roughly the epigastrium to the lower lumber area. LSOs are, for the most part, variations on the abdominal corset, with rigid struts and other support. The mechanisms of their benefits, whether from unloading the spine through compression of the abdominal contents, restriction of movement, or simply the warmth they produce, remain surprisingly unclear.¹⁵³ LSOs are generally well tolerated, can be easily tried with a physical therapist or orthotist without financial outlay, and are relatively inexpensive. A number of LSOs have a drawstring or one-handed cinching mechanism to facilitate easy donning, removal, and adjustment.

Cervical orthoses also offer a continuum of support, ranging from the fixed immobilization of a halo brace to the essentially tactile feedback and limited support of a soft cervical collar.¹⁵⁵ Between the two extremes are a range of prefabricated options, including the Miami J, Philadelphia, and sterno-occipital mandibular immobilizer braces. All provide variable amounts of movement restriction and comfort.¹⁵⁶ If pain control, rather than stabilization, is the impetus for considering a cervical orthotic, a trial of the differing options as well as a soft collar is warranted, because patient preferences and degree of benefit are difficult to predict.

Coordinating an orthotic trial is generally straightforward and within the practice scope of most physical and occupational therapists. Patients can also be sent directly to an orthotist, but this may require prior determination of which brace will be dispensed. Many medical centers have established alliances with orthotic suppliers. However, it should be remembered that (as is true for much durable medical equipment) payers may restrict provider choice, and coverage should be explored before initiating an assessment or fitting.

Positioning

The strategic use of pillows, bolsters, hospital beds, and adaptive equipment to support patients and constrain them to a pain-free range of motion is so common sense that it almost does not warrant mention. However, these effective approaches receive little attention in the formal guidance given to both professional and lay caregivers. For patients with severe weakness and motor deficits, stabilizing at-rest positions can not only reduce pain but also protect vulnerable skin over bony prominences. Like orthotics, pillows and bolsters can be used to reduce the forces placed on compromised muscles and tissue. For example, use of pillows and arm rests to support the arms of patients with head and neck disease with weakened cervical and shoulder stability after neck dissections can radically reduce pain and discomfort in the residual musculature. Figures 3A to 3D illustrate a range of widely available and inexpensive positioning aids.

Fig 3.

Positioning aids: (A) bed bolster, (B) armrest bolster, (C) contracture cushion, and (D) positioning wedge.

MODALITIES WITH PHYSIOLOGIC EFFECTS THAT INDIRECTLY INFLUENCE NOCICEPTION

Light and Laser Therapies

Low-power lasers were first used in the mid 1960s, and early reports of benefits, albeit often anecdotal, were enthusiastic and rapidly led to the application of lasers to a wide variety soft tissue injuries and conditions. With time, lasers have for the most part been replaced with a variety of nonlaser monochromatic and often infrared light sources. However, treatment parameters have remained relatively stable and typically involve the use of 30- to ≥ 150-mW devices and energy intensities of a few joules per square centimeter. As such, tissue temperatures are not elevated more than a few tenths of 1°C, and the theoretic support for effectiveness lies in the fact that irradiation at these wavelengths and intensities is known to alter a variety of cellular and metabolic processes. Light therapy was accepted by the US Food and Drug Administration via a 510K process for use primarily as an adjunct to the treatment of pain roughly 10 years ago.

As is true for many modalities, the clinical benefits of light and laser therapies remain difficult to quantitate. A large number of clinical trials have evaluated the efficaciousness of laser therapy over the years, with mixed results. Although many results are quite encouraging, systematic assessment has been limited, with recent evidence-based reviews on nonspecific low back pain and venous stasis ulcers unable to support benefit.^37,157 However, the situation seems somewhat better for patients with rheumatoid arthritis, shoulder pain,¹³⁴ or cancer receiving treatment for oral mucositis,^32,158 although even here, the need has been voiced for additional trials to ensure an adequate evaluation.

In summary, evidence for the benefits of light therapy remains mixed, although there is some level of support above that of an individual study. Treatment is associated with minimal adverse effects, and a recent review of the literature revealed not only acceptance for use in cancer treatment–related conditions, such as mucositis,^32,159 but also no evidence of an increase in the frequency of cancer recurrence or metastatic disease.¹⁶⁰

Manual Lymphatic Drainage

Manual lymphatic drainage, lymphatic massage, or Vodder-type massage is a highly specialized technique designed to reduce edema. Gentle and rhythmic movements are used to stimulate the contraction of lymphatic vessel smooth muscles, decongest tissues, and reduce inflammation through an enhanced removal of potentially inflammatory macromolecules.¹⁶¹ Massage is limited to finger or hand pressures of approximately 30 to 45 mmHg, with treatments initiated proximal to lymphostatic regions and gradually progressing distally.

The analgesic properties of manual lymphatic drainage have been well recognized through its extensive use in the management of lymphedema, with the consequence that it is increasingly being applied in pain syndromes in which lymph congestion is thought to play a role.^162,163 Treatment tends to be well tolerated, even among patients with moderate to severe allodynia. Benefits depend on the skill of the practitioner, and the recommendation of a knowledgeable person or use of a registry, such as the Lymphology Association of North America¹⁶⁴ and the Vodder School,¹⁶⁵ is highly recommended.

REHABILITATION APPROACHES TO MANAGING MUSCULOSKELETAL PAIN

Cancer-related musculoskeletal pain arises from four principal mechanisms: direct tumor invasion, maladaptive changes induced by cancer treatment or local tumor effects, exacerbation of pre-existing musculoskeletal pain, and hypertonicity and spasm related to any of the above. Although to our knowledge it never seems to have been subjected to epidemiologic study, the third mechanism (ie, exacerbation of pre-existing pain) seems surprisingly common across all cancer populations. Common approaches to the treatment of musculoskeletal pain are outlined here, with several often integrated in a unified treatment plan. Pain arising from tumor invasion is far more likely to definitively respond to treatment with antineoplastic therapies; however, even here, rehabilitation medicine approaches may be attempted once tumor control has been optimized.

Principles of Rest, Ice Compression, and Elevation

The roles of heat and cold were reviewed to some extent earlier in this article. However, the principles of rest, ice compression, and elevation (PRICE), although simple, continue to be widely used despite some concern about the effectiveness of cold in controlling pain and acute inflammation.¹⁶⁶ Musculoskeletal pain of abrupt onset or with a clear precipitant (eg, overuse or trauma) warrants a trial, with or without nonsteroidal anti-inflammatory drugs, of as many PRICE components as possible.

Deep-Heat Modalities

Although cold is generally administered topically, with ice packs or massage, heating modalities that target deeper tissues (eg, muscles and joints) are commonly used as adjuncts in PT practice. Ultrasound, in particular, continues to be a mainstay for enhancing tissue elasticity before range-of-motion activities and fibrous-release techniques. Systematic reviews and meta-analyses have established the efficacy of ultrasound in a range of benign pain states.¹⁶⁷

Injections

Injections share the common goal of delivering an analgesic or anti-inflammatory agent at high concentration to a localized pain generator to maximize therapeutic effect while minimizing systemic toxicity. An impressive array of pharmaceuticals are injected, but steroids, local anesthetics, and, more recently, botulinum toxin are the most common. Ultrasound guidance is increasingly used to optimize localization; however, the benefits of this approach, particularly with more mobile lipophilic injectates, have not been clearly established.¹⁶⁸ Inflamed tendons, bursae, and synovium are the most common targets.

When the long-term benefits of steroid injections have been scrutinized, they have seemed less effective than PT alone, although benefits are slower to accrue with PT.^169,170 To date, some of the most definitive work has been done in lateral epicondylitis or tennis elbow. There seems to be a rebound effect after the near-term anti-inflammatory benefits of the injection, with eventual worsening.¹⁷¹ This is relevant to patients with cancer, because their prognoses and involvement with cancer treatments vary radically. For a disease-free survivor with a good prognosis, PT in conjunction with or independent of an injection may offer greater collective benefit than an injection alone. In contrast, patients with far advanced cancer need near-term relief and may not survive to experience the more delayed and sustained benefits offered by PT. The role of botulinum toxin in treating mucloskeletal pain sources, such as myofascial pain, given its newer introduction, remains unclear but intriguing.¹⁷²

Injections are relatively safe and toxicity free. Patients should not receive > three intra-articular injections per year. Most physiatrists, orthopedists, and rheumatologists, as well as many primary care practitioners, perform these procedures. However, for more specialized injections (eg, involving smaller structures of the hand), the involvement of specialists who frequently target these joints is recommended.

Myofascial Release Techniques and Trigger Points

Myofascial pain is a syndrome that affects millions of people and is most commonly located in upper back musculature.^173,174 Its most salient findings include tenderness on palpation and the presence of taut bands of increased muscle tone and trigger points (ie, small areas of increased tenderness that when pressed generated stereotypical patterns of referred pain). Although the nature and cause of the syndrome remain controversial, its effects on patients can be large. Massage, exercise, and trigger-point or botulinum injections are the mainstays of treatment.

Massage, often accompanied with heat, muscle tension–release techniques, and relaxation exercises, is frequently employed. Like most approaches to myofascial pain, multiple treatment sessions are required for sustained benefit. Trigger-point injections and dry needling mechanically stimulate the discreet taut bands that are the hallmark of myofascial pain. Controversy persists as to the benefits of introducing an injectate (eg, local anesthetic, botulinum toxin,¹⁷² or steroid) versus simply penetrating taut bands with a needle. No direct comparisons offer an empiric basis to choose one over the other. However, because positive benefits have been reported for approaches using narrow-gauge (eg, > 30) needles that inflect less trauma and discomfort, these approaches may be preferred for initial needling trials.

As we have discussed, therapeutic exercise, alone or combined with other approaches, plays a critical role in normalizing the derangements that predispose patients to develop musculoskeletal pain in general. Overuse is the most common source of myofascial pain among patients with cancer and occurs when muscles, for whatever reason, must work harder or differently than they customarily do. It may not be possible to reverse the precipitating cancer-associated injuries that created the symptoms. However, the severity and chronicity of the resultant pain can generally be improved through therapeutic exercise. Evaluation by a physician or therapist familiar with cancer treatment–related changes, as well as comprehensive myofascial pain management, offers the best chance of developing an appropriate and individualized exercise program targeting all implicated muscle groups.

Massage and Body Work

Definitive research on massage as a pain-relieving modality has been hampered, similar to other modalities, by the heterogeneity of types and treatment schedules. As a stand-alone treatment, massage provides immediate or short-term pain relief for mechanical neck and low back pain.^73,79 Accupuncture-like, structural, and relaxation massages may offer greater or more sustained benefit, although the evidence base is limited.^80,175 Massage in isolation does not seem to yield sustained benefit and should be incorporated into an integrated program of exercise.³⁶ A similar limitation of benefit to the near term has characterized trials evaluating massage for cancer pain.¹⁷⁶

Therapeutic Exercise

In addition to its role in optimizing control of myofascial pain and enhancing the stabilization of painful body areas, it cannot be overemphasized that therapeutic exercise is the cornerstone of all rehabilitative approaches to pain arising from muscles, tendons, and ligaments. The structured application of specific demands to muscle and connective tissues reliably elicits desirable physiologic changes. Such demands may be resistive, aerobic, or tensile, depending on the desired alterations in muscle anatomy and physiology.

DISCUSSION

Physical modalities and rehabilitation medicine offer a range of pain management approaches that may serve as beneficial adjuncts to the conventional systemic and interventional analgesic strategies used to control cancer-related pain. These approaches may be particularly beneficial to patients with movement-associated pain and those who are ambivalent regarding pharmacoanalgesia.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author(s) indicated no potential conflicts of interest.

AUTHOR CONTRIBUTIONS

Conception and design: All authors

Collection and assembly of data: Jeffrey R. Basford

Data analysis and interpretation: Jeffrey R. Basford

Manuscript writing: All authors

Final approval of manuscript: All authors