The objectives of this chapter are to review the literature for articles about skilled therapy services applied to treat painful conditions; report primary outcomes of therapy participation (improved flexibility, strength, and endurance); and report secondary outcomes of therapy participation, including improved pain, disability, and quality of life. When the literature is informative, patient and provider attributes that guide patient selection for therapy are discussed. The terms “physical therapy for chronic pain” and “human,” were used to search the literature from January 1999 to January 2013. The search yielded 4940 articles, of which 200 were selected because they describe the application of licensed physical therapy¹ or the use of multidisciplinary programs² or functional restoration^3,4 as treatment for a painful disorder. That Cochrane reviews, systematic reviews, meta-analyses, randomized controlled trials (RCTs), and small series are included. Licensed physical therapy, as defined by the Centers for Medicare and Medicaid Services (CMS), is under the scope of PM&R and is a therapeutic activity funded by CMS and other payers in the United States. For articles not written in the United States, articles were selected that implied that a medical professional with an equivalent educational and licensing level of “licensed physical therapist” provided therapy services. Excluded were articles pertaining to exercise done at home or in a commercial gym, club, or spa, biofeedback, complementary and alternative medicine ([CAM]: homeopathy, acupuncture, hydrotherapy, massage), chiropractic manipulation, Tai Chi, Qi Gong, yoga, Pilates exercises, spas, balneotherapy (warm springs), mindfulness therapy, and cognitive-behavioral therapy (CBT) independent of a multidisciplinary program. Also excluded were articles that described exercise administered in nonmedical settings (commercial gyms or spas or home) and by trainers, instructors, certified trainers, “qualified lab personnel,” and kinesiologists. These exclusions were made because these activities, settings, or personnel do not fall under the scope of PM&R (e.g., CBT alone) or are activities that neither require medical prescription nor are reimbursed by medical insurance.⁵ (Fig. 96-1).

Discussions of rehabilitation applications for pain in the low back, knee, neck and pain caused by fibromyalgia and chronic regional pain syndrome, which were covered in the second edition chapter, are updated here. The current literature review permits a discussion of PM&R treatments for pain in the head and pelvic regions and for pain due to Achilles tendinopathy. The constructs of “back schools,” multidisciplinary programs, and functional restoration are also discussed.

The PM&R chapter in the second edition⁶ described the scope and philosophy of PM&R, explained the components of a detailed therapy script, and outlined basic prescriptions for various painful diagnoses. Simply put, PM&R focuses on function and performance and on restoring or compensating ability despite disease and impairment. The practice of PM&R spans the settings of home health, inpatient, outpatient, and community reentry. Our literature search confirms that most PM&R services that are accessed to treat pain are delivered in the outpatient setting.

The practice of PM&R has a long multidisciplinary tradition in which the physician works in concert with physical therapists (PTs), occupational therapists (OTs), and speech therapists, as well as social workers, nurses, pharmacists, psychologists, and recreational and vocational (or child life) specialists. Physical therapists help patients with gait or alternative mobility and trunk and lower limb strengthening, and OTs help patients with upper limb dexterity and strength, self-care, and activities of daily living. The literature search confirms that treatment of simple pain disorders may require only the physician and PT or OT; however, complex chronic pain disorders that cause severe disability and handicap may require treatment in a multidisciplinary program or a functional restoration program in which the services of a psychologist, nurse, or social worker are enlisted.

A physician writes a prescription to communicate the specifics of treatment to another allied health care provider, the medical record, and also the patient. There is no gold standard for what a therapy prescription should include, but Currie and Marberger⁷ suggest patient name and diagnosis, goals of treatment, discipline of the treating therapist (T, OT, speech therapist), precautions, and date of reevaluation. A physician familiar with options provided by a PT, OT, or speech therapist may elect specific types of exercise or treatments (Table 96-1). However, precautions, contraindications, or limitations to exercise are most important and should be discussed first.

A prudent physician pauses before prescribing a therapeutic moiety and, in the case of PM&R, considers any attribute of the patient that contraindicates participation in exercise or use of a thermal or electrical modality. It is second nature for a physician to ask about allergies prior to prescribing a medication. Exercise allergy is possible even when the antigen is unknown.⁸ Exercise mobilizes leukocytes and induces an initial inflammatory response; albeit rare, derangements of this inflammatory response, up to and including anaphylaxis, as well as exercise-induced asthma, and exercise-induced urticaria are possible.⁸ Medical and surgical conditions may limit, delay, or preclude exercising a patient. Writing about resistance exercises, Wai states that unstable medical or surgical conditions preclude participation.⁹ Wai cites the American Heart Association (AHA) article that recommends that persons with unstable ischemic, valvular, hypertensive, or arrhythmic conditions should not participate in resistance (strengthening) exercises¹⁰ (Table 96-2). Persons with these untreated or unstable diagnoses should not participate in aerobic,^11,12 aquatic, or flexibility exercises. Resources from the American College of Sports Medicine (ACSM) note other unstable medical conditions, including metabolic disorders (e.g., uncontrolled diabetes), hematologic disorders (e.g., acute deep venous thrombosis, severe anemia, coagulopathies, disorders of hemostasis), open wounds and certain skin eruptions, and unstable or untreated spinal or extremity fractures^11,12 (see Table 96-2). May notes that the exclusion criteria of studies may inform decisions about patients “felt to be unfit for physical therapy treatment,” for example, individuals with fracture, grade III or IV spondylolisthesis, malignancy, inflammatory arthropathies, pregnancy, significant neurologic loss, and prior spine surgery.¹³ In order to avoid burns, hot or cold modalities should never be used on insensate or vascularly compromised skin.⁶ In the United States, millions of people access PM&R services annually, but a complete history and physical examination are requisite to confirm diagnoses and conditions that limit, delay, or proscribe exercise participation. If there is a question regarding the patient’s ability to participate in an exercise program or how to adjust an exercise program so that a patient may participate, consultation with a PM&R physician may be in order.

A physician familiar with the spectrum of treatments provided by a PT, OT, or speech therapist may elect to specify certain types of exercise or treatments (see Table 96-1). The goal of the second edition chapter was to familiarize physicians with exercise options. The ACSM guidelines are excellent resources for information about the medical prescription and monitoring of exercise and applied exercise physiology.^11,12

Stretch and flexibility are initial steps in an exercise program and prevent injury to soft tissue.⁶ Slow sustained stretch held for 60 seconds is optimal.⁶ The activity of stretching may not distend the myotendonous unit as much as increase tolerance of the activity. Discussing hamstring stretch, Halbertsma demonstrates no change in the length or elasticity of hamstrings with applied stretch; rather increased tolerance of stretching activity occurs.¹⁴ The presence of tubes, wires (new spinal stimulator implantation), and lines may limit participation. Stretching has been shown to relieve pain at least briefly. Lewitt studied patients with myalgic pain and found that stretch applied immediately after an isometric contraction gave immediate pain relief in 94% of patients.¹⁵

Strength refers to the ability of a muscle group to apply force.¹⁶ There are three types of muscle contractions and three types of strength training: isometric, isotonic, and isokinetic. The first type, isometric (“same length”) contraction occurs when the muscle is hardened, or “set,” and the length of the muscle does not change.⁶ The body builder posing with his elbows flexed and biceps contracted is an example of isometric contraction. In contrast to static isometric contraction, isotonic contraction has motion and contracts a muscle through all or part of its normal range of motion (ROM) while lifting a constant amount of weight. Lifting a dumbbell as the elbow moves from full extension to full flexion is an example of an isotonic contraction of the biceps. There are two types of isotonic contractions: shortening, that is, concentric contraction, and lengthening that is, eccentric contraction. Concentric strengthening creates power, but eccentric strengthening increases tendon tensile strength or lengthens tendon or the myotendonous junction.¹⁷ The third type, isokinetic contraction, occurs when the muscle contracts against a fixed torque. Club machines with cams and fixed axes or a heavy door on a hydraulic governor provide examples if one pulls against them. Strength training improves the force production of a group of muscles by any of the following mechanisms: an increase in the number of motor units activated, an increase in the rate of activation, an increase in the synchronization of motor units firing, or the hypertrophy of muscle fibers.¹⁶

Strengthening occurs as long as exercise continues to the point of fatigue. The various formulas for strength training involve lifting some percentage of the “one-repetition-maximum” that a muscle group can move and doing repetitions to the point of fatigue. For example, if the heaviest dumbbell an individual can lift (safely) with the biceps is 20 pounds, the one repetition maximum, then a strengthening strategy might be lifting 10 pounds for 10 repetitions and then 15 pounds for 10 repetitions or less if fatigue occurs. When weights are handled through a set of repetitions, additional weights can be added. For strengthening to be accomplished, strengthening regimens of each target muscle group must be performed to fatigue about three to five times per week.⁶

Contraindications and precautions for strength training exist.⁶ Strengthening exercise is contraindicated in the presence of fracture; the orthopedic surgeon should prescribe permitted activity and denote prohibited activity for a fractured limb and the contiguous joints.⁶ If this is not specified, the primary physician or health care provider should ask the orthopedist to determine allowed ROM, weight-bearing, and exercises permitted. Strengthening exercise acutely increases blood pressure, and this vital sign needs to be monitored in patients with hypertension.⁶ Persons should have a “spotter” or partner when performing strengthening exercises, especially with free weights.⁶ Strengthening exercises are contraindicated in the presence of acute or unstable cardiopulmonary disease.⁶

The last exercise type is endurance, also known as aerobics, which involves the rapid repeated motion of large muscle groups under low load, often to accomplish locomotion (jogging, swimming, biking) for a prolonged period. Endurance is the time that a person can maintain either a static force or a power level involving a combination of concentric and eccentric muscle actions.¹⁸ Endurance is “the ability to continue a prescribed task in the desired manner.”¹⁹ Although strengthening exercise increases muscle force, endurance exercise increases aerobic capacity, or maximal oxygen uptake (VO₂). As a result of endurance training, the number and size of mitochondria in muscle increase, the activity of mitochondrial enzymes increases, and blood flow to muscles increases because of increased numbers of capillaries and improved efficiency of blood flow shunting.¹⁸ Adaptations in the heart and vasculature include increased stroke volume, expanded blood volume, decreased resting heart rate, and decreased resting systolic and diastolic blood pressure.¹⁸ To achieve an endurance effect, the patient needs to participate in 15 to 60 minutes of continuous aerobic activity three to five times per week at sufficient intensity to raise heart rate to 60 to 90% of maximum.⁶ A maximal exertion exercise treadmill test can determine maximal heart rate; however, an easy approximation of maximal heart rate for a given age is arrived at by subtracting the patient’s age from 220.

Howley writes about the classification of leisure and occupational activity in terms of aerobic and metabolic physiologic expenditures.²⁰ Patients often report that they get “plenty of exercise” in this manner. However the randomness of both the vector and intensity of work and leisure activity although it may expend calories, may not achieve strength, flexibility or cardiac endurance above the demand requirements the individual encounters and may expose the individual to risk and injury not inherent in isolated measured exercise.

To reiterate, this chapter is a review of the recent literature for articles pertaining to PM&R applications, that is, supervised exercise by a licensed therapist or multidisciplinary and functional restoration programs for the treatment of painful disorders, and to assess and report the effectiveness of these treatments. There are limitations for the review, and Mayer elegantly comments “that studies on therapeutic exercises often fail to provide details on the specific exercise techniques used and the exact exercise protocol that was prescribed or followed (e.g., dose, timing, intensity) and guidelines and systematic reviews frequently combine various forms of therapeutic exercise and ignore important differences among the types of exercise.”²¹ Exercise to relieve pain and applied at home or a commercial gym or the effectiveness of Eastern exercise and lifestyle traditions for the relief of pain is a different topic and not answered in this chapter.

In the second edition, a physical therapy prescription for neck pain that included corner stretches, cervical-thoracic stabilization exercises, and postural correction of head thrust position was described.⁶ The literature review for this edition supports that stretching for chronic neck pain is effective²² (Table 96-3). The literature finds that low-intensity postural neck exercise is not supported.^23,24 Deep cervical flexor (DCF) strengthening, which is believed to affect head and neck posture, does show correlations between improved electromyographic readings (a proxy for strength) decreased neck pain, and disability^25–29 (see Table 96-3). The review article by Ylinen reports that the effectiveness of long-term isotonic and isometric strengthening exercises of the neck and shoulders for chronic and recurring neck disorders was supported.²³ Pain and disability due to chronic neck pain are significantly improved not only by strengthening but also by endurance and coordination regimens³⁰ (see Table 96-3). The treatment of acute and chronic whiplash with exercise modalities continues to be problematic.^31–33 Only one study was found that evaluated the predictors of poor short-term and long-term outcomes for patients undergoing a rehabilitation program for chronic neck pain: Cecchi et al. concluded that poor outcome was predicted by pain-related medication intake in the short term and long term and by catastrophizing in the long-term.³⁴

A Cochrane review on applied exercise for mechanical neck disorders concludes that combinations of cervical and scapulothoracic stretching and strengthening for chronic neck pain improved function in the short and intermediate term.³⁵ However, chronic neck pain does not respond to upper extremity stretching and strengthening or a general exercise program.³⁵ Neck strengthening exercise in acute cervical radiculopathy relieves pain in the short term.³⁵ Many, but not all, studies analyzed in the Cochrane review describe the administration of an exercise program by a PT; however, some exercise regimens were self-administered or administered by a nonrehabilitation practitioner such as a chiropractor, which are applications outside the scope of PM&R.³⁵ The studies subsequently discussed are performed by a PT within a PM&R setting.

Cunha et al. found that stretching significantly reduces neck pain immediately and at 6 weeks.²²

Cunha et al. compared prolonged stretch (15 min) with manual therapy (pommage) versus conventional stretch after pommage, performed twice weekly for 6 weeks.²² Both were equally and significantly effective in reducing pain and improving the ROM and quality of life of women patients with chronic neck pain immediately after treatment (VAS 6.6 to 2, combination and 7.2 to 1.6 conventional stretch only, P > .000) and at a 6-week follow-up (VAS 3.2 combination, and 2.7 conventional stretch only)²² (see Table 96-3). The authors concluded that stretching exercises should be prescribed to chronic neck pain patients.²²

The literature review identified one active comparator trial²⁴ and one review article²³ regarding postural correction. The active comparator trial by Griffiths suggests that although posture exercises alone or combined with neck stabilization exercises (four sessions over 6 wk) may show a trend toward improving disability.²⁴ A significant finding in Griffiths’s study is that neck stabilization groups were less likely to be taking pain medication at 6-week follow-up (P < .02).²⁴ Ylinen concludes that the effectiveness of postural and proprioceptive low-intensity exercise regimens for chronic neck pain is not supported.²³

Jull et al.²⁶ discuss imbalance or impaired function of the deep cervical flexors (DCF) associated with chronic neck pain. Impaired function of the DCF may allow extension in the upper cervical segments and flexion in the lower segments and a head-forward posture. Rudolfsson et al. studied sagittal movement of persons with chronic neck pain and reported reduced extension in the upper cervical levels and reduced flexion for the lower levels.²⁹ Additional altered ratios between ROM for the upper and lower levels was observed with less contribution of motion of lower cervical levels to the total sagittal ROM compared with pain-free controls, but the findings were not explained by greater forward head posture.²⁹ These sagittal changes may relate to DCF weakness. Strengthening the DCF results in improved pain rating²⁵ and neck disability scores^25,26 and in improved muscle strength as indicated by electromyographic proxy.^25–28

In an RCT of low-load exercises of the DCF, C-CF versus higher load neck flexion exercises, Jull et al.²⁶ report improved electromyographic (EMG) amplitude DCF after C-CF training as well as decreased EMG amplitude of the superficial flexors (scalenes and sternocleidomastoid). The relative latency between the activation of the deltoid and the DCF during rapid arm movement in the C-CF group shortened compared to the strength group (P < .05).²⁶ Average pain intensity (C-CF decreased 1.7+/−2.0 and strength group decreased 1.0 +/− 3.3, both P < .001)) and neck disability index score significantly decreased in both exercise groups (C-CF −5.0 +/− 4.2 (P < .05) and strengthening group −3.5 +/− 2.3, P < .001) (see Table 96-3). Baseline pain on a scale of 10 was 4.5 +/− 1.0 in the C-CF and 4.2 +/− 1.0 in the strength group while baseline disability on a 50 point scale was 11.0 +/− 2.7 and 9.6 +/− 3.1, respectively.²⁶ O’Leary evaluated change in pain (visual analog scale [VAS]) and change in pain pressure threshold (PPT) after a single performance of C-CF versus neck flexion exercises.²⁵ He found an immediate small but significant improvement in pain and in PPT (change on 10 cm VAS 0.42 cm vs. 0.11 cm, P < .04) (PPT increase of 21% (P < .001)) in the C-CF group.²⁵ Falla et al. undertook a prospective strengthening program for the DCF 14 subjects with supervision from a PT once a week for 6 weeks with home repetition twice a day.²⁸ The activation of the DCF increased (P < .0001) most in patients with the lowest amplitude of DCF EMG at baseline.²⁸ A significant (P < .05) relationship existed between initial pain intensity, change in pain level with training, and change in EMG amplitude for the DCF during C-CF.²⁸ Pain decreased with the exercise regimen from 4.1 +/− 1.7 to 2.3 +/− 2.3 (P < .01) VAS 0-10) and neck disability index (possible 50) decreased from 10.2 +/− 2.7 to 5.5 +/− 4.4, P < .001) (see Table 96-3). The baseline characteristics of some test subjects demonstrated only moderate pain and disability related to chronic neck pain, but for persons with similar symptomatology C-CF exercises demonstrated efficacy for reducing pain and disability.

Another controlled trial by Falla et al.²⁷ used C-CF low load exercises as a comparator versus a neck flexion endurance-strength regimen for the cervical flexors to evaluate if superficial cervical flexor muscle fatigue could be improved in patients with chronic nonsevere neck pain. Maximal voluntary contraction of the anterior scalenes and sternocleidomastoid was significantly improved in the endurance-strength group (10.1 +/− 17.3 N, baseline 75.5 +/− 17.9 N) versus the C-CF group (1.8 +/− 10.6 N, baseline 78.2 +/− 19.1 N) (P < .05)²⁷ Average intensity of pain about 1 point on a 10 point scale in each group (baseline ~4) and neck disability index score decreased about 3 points in both intervention groups (baseline ~10).²⁷ Falla et al. note that these changes although small are significant. This study underscores the specificity of exercise for both targeted muscle (in this study, the superficial flexors) and type of exercise; that is, strength-endurance regimens improved strength-endurance.²⁷ Pain improves regardless of the exercise regimen, and the authors draw no conclusions about the interaction or directionality of improved pain and improved strength endurance.²⁷

O’Leary et al. compare mobility, endurance, and coordination exercises showed training specific benefits with between-group comparisons revealed significantly greater gains in endurance (P < .02) by the endurance group, and significantly greater gains in coordination (P < .01) by the coordination training group. However, all three groups had improvement in pain (P < .01) and disability (P < .01) (see Table 96-3).³⁰ The authors comment that subjects with mild to moderate symptoms are recruited so that the study regimens can be completed but they hypothesize that patients with more severe symptoms might benefit.

A RCT to determine the dose or intensity of exercise required to decrease chronic neck pain was accomplished by Nikander.³⁶ Patients in the control group received baseline strength measurements and several days of instruction in stretching with a recommendation to complete aerobics activities three times a week. Patients in the two active groups were trained by a PT for 12 days in strength or endurance exercises and continued the exercise program at home for 1 year.³⁶ Strength and endurance training (compared to the control group) decreased perceived neck pain and disability.³⁶ Declines in neck pain and disability correlated positively with the amount of specific training, and specifically neck, shoulder, and upper-extremity training for more than 8.75 metabolic equivalent of task (MET)-hours per week was an effective training dose for decrease of neck pain.³⁶ One MET-hour of training per week accounted for a 0.8-mm decrease of neck pain on a VAS (100 range VAS) and a 0.5-mm decrease on a disability index.³⁶ Improvements in pain and disability occurred in both the strength and endurance-training groups. The pain and disability baselines in this study indicate a more symptomatic group.

Ylinen et al. studied isometric strengthening versus dynamic endurance training trained over four sessions with a PT and then carried over at home for 12 months. In the strength training and aerobic training groups the greatest gains in neck strength, as well as decrease in neck pain and disability, were achieved during the first 2 months, and improvements continued up to 12 months.³⁷ The isometric strengthening group achieved the greatest strength gains at all follow-ups, and change in neck pain and disability indices correlated with the isometric neck strength.³⁷ The baseline VAS of the subjects was 58 (VAS range 100) and at 12 months, the strength group improved 69%, the endurance group improved 61%, and the control group improved only 28% (P < .001)³⁷ (see Table 96-3). Ylinen et al. crossed the control group who had only minor changes in pain and functional measures in this study over high-intensity strength training.³⁸ Significant decreases in neck pain and disability indices occurred, and maximal isometric neck strength increased in flexion and rotation and extension at the 2-year follow-up.³⁸

Other PM&R strategies applied to treat neck pain include grade exercise, a multidisciplinary rehabilitation program, and a multidisciplinary or functional restoration type program with exercise and fear avoidance training. Pool et al.³⁹ studied behaviorally graded progressive exercises (BGA) versus manual therapy each applied by PT for the treatment of subacute (4-12-wk duration) neck pain. Pain and function were improved about 90% of the time.³⁹ The statistically significant overall effect was found on the NDI in favor of the BGA treatment.³⁹ An active comparator trial of multimodal group rehabilitation practicing stability, strengthening, and proprioceptive exercises with an educational program, 1 hour a week for 6 weeks, was compared to control group treated as deemed appropriate by their physiotherapist.⁴⁰ Both groups significantly improved in both function and pain scores (4.6 +/− 2.3 and 4.5 +/− 2.2, P <0.01). There was no significant difference in improvements in disability or pain between groups.⁴⁰ No comparison based on program cost or patient time expenditure or convenience was made. Taimela et al.⁴¹ compared a functional restoration type program to an active comparator of home exercise with two educational sessions to an inactive control group of education alone for the treatment of chronic neck pain. The exercise regimen for chronic neck pain, administered by a physiotherapist, which included 24 sessions of cervicothoracic stabilization, relaxation training to reduce unnecessary muscle tension, behavioral support to reduce anxiety and fear, and seated wobble-board training to improve postural control, yielded a significant reduction in neck symptoms and improvement in general health and self-reported working ability at 3 and 12 months.⁴¹ A trend of improved psychological well-being was observed at 3 and 12 months but was not significant.⁴¹

Little regarding modalities and neck pain was identified in our literature search. A study group in Hong Kong found that TENS (150 us square pulses at 80 Hz) applied over acupressure points for 30 minutes was as effective as strengthening exercise (twice a week for 6 weeks) compared to infrared radiation (heat lamp for 20 min) to improve disability, isometric neck muscle strength, and pain at 6 weeks and at the 6-month follow-up.⁴²

Lastly, the literature suggests that the neck disorder whiplash, both in acute and chronic states, has limited PM&R treatment options. Regardless of active or inactive treatment of acute whiplash Kongsted et al. showed 50% of subjects with pain at 1 year.³¹ The randomized parallel group trial for acute whiplash showed no differences among three treatment groups lasting 6 weeks. The treatment arms included immobilization with a neck collar for 2 weeks and then mobilization, “act-as-usual,” or mobilization, with weekly evaluation and instruction by a PT in ROM and exercise completed at home.³¹ The progressive mobilization was based on principles of mechanical diagnosis and therapy (MDT), which are based on repetitive movements directed by pain response. At 1 year, about one-half of subjects reported considerable neck pain and disability and one-seventh had not returned to work.³¹ A RCT for chronic whiplash demonstrated that exercise and advice was significantly more effective than advice alone to reduce pain intensity and pain annoyance and improve function at 6 weeks; however, the effect did not persist at 12 months.³² In the exercise group, activities included aerobic exercise (e.g., a walking or cycling program), stretches, functional activities, activities to build speed, endurance and coordination, and trunk and limb strengthening exercises, all performed with the PT two to three sessions per week for 6 weeks.³² Greater treatment effect was observed with patients with higher levels of initial pain or disability.³² A review of RCT and non-RCT analyzing the effect of defined physical therapy protocols or multidisciplinary programs on whiplash associated disorder (WAD) support the use of interdisciplinary interventions and chiropractic manipulation but not strongly.³³ Regarding exercise the review concluded that more effective regimens were supervised, and started earlier but aggressive subacute programs like work hardening could be counterproductive.³³

In considering PM&R treatments for neck pain, whiplash may be a distinct entity because of its less favorable outcome after these interventions. However for patients with moderate chronic neck pain and related disability, stretch, strengthening, endurance training, and deep neck flexor strengthening supervised in a PM&R setting have demonstrated significant improvement.

In the second edition, a physical therapy prescription was outlined including instruction in cane use and joint protection, hamstring stretch and progressive strengthening of the quadriceps, and low-impact or pool aerobics, as well as the use of TENS or ice for pain.⁶ The interim literature for the third edition chapter yielded five review articles (two on exercise,^42,43 two on TENS,^44,45 and one on bracing⁴⁶) and four controlled trials (two on exercise,^47,48 one on exercise with bracing,⁴⁹ and one on diathermy⁵⁰) for the treatment of knee pain, generally due to osteoarthritis but also anterior knee pain.⁴⁹ Strengthening and aerobic types of exercise improve pain and function,⁴² walking pain and locomotor function,⁴³ and pain and strength⁴⁷ in osteoarthritis. The Jessep study compared cost of individual supervised therapy to cost of activity and education administered at a community center with simple equipment and found no differences at 12 months in clinical outcomes (WOMAC pain and function, HADS anxiety and depression, and aggregated physical performance); however, there was significant (50%) cost savings with the community center approach⁴⁸ (Table 96-4). Regarding patient selection, only the Brakke review article noted that patient traits that predict a response to physical therapy include “milder disease (i.e., unilateral OA, symptoms for less than 1 year, and a 40-m self-paced walking test faster than 25.9 seconds)” and those who rate their pain 6 or greater on the numerical pain rating scale.⁴² Fatalism and fear that activity can worsen arthritis may bode work against a positive physical therapy outcome.⁴² Overall the interim literature does not support the use of TENS in knee arthritis.^42,45 Bracing may be helpful, but compliance after 6 months declines.⁴⁶

Exercise continues to be a primary PM&R treatment for knee arthritis, and Brakke et al.⁴² conclude that “strength training, aquatic therapy, and balance and perturbation therapy were the most beneficial with respect to reducing pain and improving function.” Bennell and Hinman⁴³ reviewed application of exercise for primarily knee osteoarthritis across the spectrum of self-administered to therapist-supervised scenarios and reported that strengthening improves strength, pain, and physical function while aerobic exercise benefits pain, joint tenderness, functional status, and respiratory capacity. Bennell and Hinman also caution that the literature does not clarify the optimal exercise modality and dosage for osteoarthritis of the knee and the existing literature has not compared exercise regimens on the basis of exercise modality, intensity, duration, and/or frequency.⁴³ Specific to therapist administered exercise, Bennell and Hinman cite an 8-week program that led to significantly greater improvements in locomotor function and walking pain long-term at 12 months. Bennell and Hinman also report a smaller treatment effect with regimens fewer than 12 sessions but 12 or more supervised sessions having a moderate effect.⁴³ One study comparing types of exercise showed isokinetic exercise to improve strength and pain better than isometric exercise of the knee.⁴⁷ Hernandez-Rosa et al. compared 8 weeks of isometric to isokinetic strengthening applied every third day.⁴⁷ Isokinetic exercises had greater effectiveness for strength gains and pain relief (see Table 96-4), but ROM was similar.⁴⁷ The Jessep article highlights cost-efficiency in the delivery of skilled therapy services.⁴⁸ The community-based ESCAPE program starts with patient education and group discussions and instruction and supervision in the performance of an exercise set with the PT.⁴⁸ Over time the ESCAPE patients become independent with the carryover of the regimen. Patients in the ESCAPE paradigm were compared with patients receiving individualized therapy.⁴⁸ Although the authors use inconsistent terms to describe the number of visits, the authors suggest that the ESCAPE patients had about twice as many (8, presumed median) visits for strengthening exercise as the individualized therapy group (median 4) at about one-half the cost, £583 (currency expressed as pounds) per subject for the individualized group versus £320 for an ESCAPE subject.⁴⁸ Outcomes on the Western Ontario and McMaster Universities Osteoarthritis (WOMAC) pain and function indices, the Hospital Anxiety and Depression Scale (HADS), and the aggregated functional performance time (AFPT) for four tasks were similar between groups.⁴⁸ In sum, aerobic and strengthening (preferably isotonic or isokinetic) exercises improve pain and function related to symptomatic knee arthritis. Twelve or more visits may have a more substantial dose effect than fewer than 12 visits for this diagnosis. The setting in which supervised therapy is delivered did not affect outcome, and paradigms for economical therapy service delivery exist.

In addition to osteoarthritis other painful conditions of this joint exist. For refractory anterior knee pain, Schneider⁴⁹ described a study comparing 8 weeks of proprioceptive neuromuscular facilitation with 16 exercises to a training program using a special resistance-controlled knee splint for 15 minutes three times daily. Increased electromyographic activity in the vastus medialis muscle increased in both groups but a significant improvement in pain was observed only in those treated by knee splint.⁴⁹

A Cochrane review in 2000 reported that TENS and acupuncture-like TENS were more effective than placebo for the relief of pain and stiffness due to knee osteoarthritis.⁴⁴ However the updated review in 2009 of interim smaller studies of questionable quality did not confirm that TENS is effective for pain relief.⁴⁵ Brakke also concludes that electrical stimulation likely has little impact for knee osteoarthritis and also that evidence regarding manual therapy is equivocal.⁴²

A modality for knee osteoarthritis includes diathermy (433.92 MHz microwave), and Giombini et al. showed that the modality reduces pain and improves physical function in patients with moderate knee arthritis symptoms when applied 30 minutes three times a week for 4 weeks and compared to sham treatment.⁵⁰ The last review article of physical therapy interventions for knee pain due to unicompartmental arthritis reported that patellar taping reduces pain and knee unloader braces improve symptoms; however, about one-half of patients discontinue brace use within 6 months.⁴⁶

Achilles tendinopathy is one of the most common foot and ankle complaints related to sports and overuse injuries. A discussion of the exercise treatment of Achilles tendinopathy is included because the clinical complaint is common and the regimen is simple. In contrast to other painful diagnoses discussed in this chapter, Achilles tendinopathy and its treatment fall in the realm of sports medicine, in which a primary goal of intervention is the expedient return of the athlete to his or her sport, whether at the level of a professional athlete or as a tenacious weekend warrior. Unlike patients with chronic pain who may need supervision and education to overcome fear avoidance and kinesiophobia, athletes may require supervision and education to allow adequate healing and to improve techniques to prevent reinjury and safe return to play.

Eccentric training for chronic Achilles tendinopathy is described in the literature. After initial supervised instruction with a PT, the exercises can be performed independently. Petersen et al. demonstrate the technique that compared eccentric stretching to an ankle bracing device to both treatments.⁵¹ Eccentric training can be done with the patient standing on tiptoe, that is, the foot plantar flexed, on a step edge and slowly lowering the heel to achieve maximal dorsiflexion at the ankle.⁵¹ Petersen et al. reported that Air Heel brace was as effective as eccentric exercise for pain relief, but no synergistic effect with a combination of bracing and exercise was observed.⁵¹

Eccentric stretching has not been found to be harmful, and one study demonstrated improved microcirculatory tendon levels without any evidence of adverse effects in both midportion and insertional Achilles tendinopathy.⁵² A review of the literature on eccentric stretch for Achilles tendinopathy by Kingma et al. concluded that because of the methodologic limitations of the trials, the effects of eccentric overload training are inconclusive but promising.⁵³

Verrall et al. studied athletes with chronic Achilles tendinopathy and reported that after 6 weeks of stretching, pain significantly decreased (VNS 7.2 to 2.9, P <.01).⁵⁴ Patient satisfaction was rated as excellent for 80% of athletes, and average time to return to premorbid activity was 10 weeks.⁵⁴ However, long-term results may be limited. In a 5-year follow-up of Alfredson heel drop exercises, about 40% of patients were pain-free; however, about 50% of patients had elected other therapies.⁵⁵ Subjects were average age 50 years, and about 70% engaged in sports or recreational activities. Additional PM&R treatments for Achilles tendinopathy include supportive night splints, but no recent literature was located on this topic.

In the second edition, a physical therapy prescription for chronic regional pain syndrome (CRPS) included instruction in pacing activity and admonitions to withhold therapy if vasomotor sudomotor instability worsened.⁶ Passive and active ROM and stretch were recommended.⁶ Four review articles or treatment guidelines^56,59 summarize older PMR as well as medical strategies to treat CRPS and introduce newer therapy strategies.^58,59 A Canadian review summarized medications and injection therapies as well as physical therapeutics for CRPS.⁵⁶ Contrast baths and stress loading are older techniques and are discussed in a review by Li et al.⁵⁷ Stress loading has patients do “scrubbing” and “carrying.”⁵⁷ A recent review of therapy techniques used in the treatment of CRPS include graded exposure to activity, movements, and light touch; pain-adapted exercises and desensitization activities, exercise with stretching and active ROM, water therapy, stress loading, and mirror visual feedback.⁵⁸ British guidelines for the treatment of CRPS recommend referral to PT and integrated interdisciplinary treatment and the use of newer therapy paradigms such as graded motor imagery and mirror therapy.⁵⁹ Articles are reviewed for current treatment strategies with motor imagery programs or graded motor imaging, mirror therapies, pain exposure physical therapy (PEPT), and interdisciplinary programs for adults and children.

Motor imagery program (MIP) described by Moseley⁶⁰ is an exacting technique and uses 2 weeks each of recognition of hand laterality with pictures and then imagined and then mirrored movements.⁶⁰ This specific order of the component treatments needs to be maintained in order to improve pain and disability as was shown in later research with sequence changes that were unsuccessful.⁶¹ Moseley compared the MIP technique to a control group who received twice to three times weekly active and passive mobilization of the limb, systematic desensitization, and hydrotherapy in a department of physical therapy.⁶⁰ Compared to conventionally treated subjects, the MIP group reported significantly less pain and the neuropathic pain scale (NPS) (Table 96-5), finger circumference and response time to recognize the affected hand significantly improved (P < .01), and the beneficial effects of treatment were replicated when the conventionally treated controls were crossed over to the MIP group.⁶⁰ No improvement was noted in the control group while they were receiving conventional therapy. Twelve weeks after the study four patients in the initial MIP group and two patients in the crossover MIP group no longer fulfilled the criteria for CRPS.⁶⁰ Moseley concludes that response to MIP supports the hypothesis that cortical abnormalities are involved in the development of this disorder.⁶⁰ Because a specific sequence of activity is required, Moseley postulated that a sequential activation of cortical motor networks may occur.⁶¹ However, motor graded imagery (GMI) did not show improvement in pain outcomes in a prospective clinical audit of “real world” patients with CRPS treated in physical therapy clinics.⁶² Average pain intensities did not change from pre- to posttreatment, and 3 of 32 patients reported that pain had decreased by at least one-half.⁶² However, a secondary outcome of pain interference with activities of daily life was significantly improved at one center.⁶²

Moseley et al. studied mirror therapies to evaluate if tactile training resulted in improvement in tactile acuity (as determined by two-point discrimination[TPD]); when patients watched the reflected image of their unaffected limb in a mirror during training, they looked toward the stimulated (CRPS) body part and could see the skin of the opposite body part in the mirror.⁶³ TPD was 8 mm less 2 days after training compared with before training ([95% CI = 1.5-14.3 mm], P < .001).⁶³ Reduction in pain and change in TPD over the session were strongly related (R = .83, P < .001). However the effect is short-lived and there was no residual effect on pain at 2-day follow-up.⁶³

Van de Meent et al. explain pain exposure physical therapy (PEPT) as “treatment for patients with chronic regional pain syndrome type 1 (CRPS-1) that consists of a progressive-loading exercise program and management of pain-avoidance behavior without the use of specific CRPS-1 medication or analgesics.”⁶⁴ Ek et al.⁶⁵ studied an application of physical therapy of the affected CRPS limb that ignored or neglected pain and was directed at functional improvement only and normal use of the limb despite pain.⁶⁵ The authors emphasize that extensive explanation and disclosure is required with this methodology. Additionally, the subjects included were considered “end stage” and had failed multiple prior treatments and had symptoms of CRPS longer than 9 months. Activities included traction and translation of the stiff joints, assisted or active movement of the joint combined with passive stretching of contiguous muscles, and manual friction of tender points if needed. The patients underwent about four treatments over 3 months with encouragement to use the limb normally between sessions. It should be noted that the authors did not obtain institutional review board review prior to performing this study, rationalizing that the patients had already failed all available treatments. The function of the affected arm or leg improved in 95 patients and full functional recovery (defined outcome) was experienced in 49 (18 upper limb and 31 lower limb) (46%).⁶⁵ However, functional improvement did not imply a reduction in pain, and in 23 patients functional recovery but pain also increased.⁶⁵ Seventy-six patients had a significant reduction in pain.⁶⁵ At completion pain increased in 14 patients and did not change in 12. Four patients dropped out of the study related to pain increase.⁶⁵ No harm or injury from this therapy regimen is reported. Van de Meent et al.⁶⁴ studied the safety of PEPT in 20 patients and found that the physical signs of CRPS did not worsen (edema, color change, temperature, and joint mobility). The mean and mode for the number of sessions was five, and the maximum number allowed was six. Pain, upper and lower limb disability and function, and quality of life improved significantly at 12 months. Visual analog scale (57% decrease at 12 months from baseline, P < .001), pain intensity (48%), muscle strength (Newton) (improved 52% (upper limb) and 59% (lower limb) of the difference at baseline between affected and normal limb), arm/shoulder/hand disability (improved 36%, 71.7 +/− 16.2 at baseline 5.7 +/− 18.2, P < .001) (disability of arm, shoulder, and hand [DASH] measure), 10-meter walking speed (improved 29%), pain disability index (improved 60% 37.8 +/− 9.4 improved to 15.3 +/− 13.7 at 12 months (P < .001), kinesiophobia (18%) (Tampa Scale of Kinesiophobia), and the domains of perceived health change in the SF-36 survey (26.9% [corrected]) improved. Three patients initially showed increased vegetative signs but improved. Two patients had increased edema that resolved.⁶⁴ No injuries were reported.

The frequently cited Oerleman article is a RCT of PT versus OT versus a control group (social service interview) on 135 subjects with upper limb CRPS of less than 1 year’s duration.⁶⁶ Therapy regimens and frequencies were individualized for the patient, but the goals of PT were increasing pain control and optimizing coping and extinguishing the source of the ongoing pain and improving skills while the goals for OT were reducing symptoms of inflammation and/or protecting and supporting the hand in the most functional and comfortable position and normalizing sensibility and improving functional abilities of the hand and improving independence in activities of daily living.⁶⁶ Outcome measures were ROM and pain evaluation with VAS and the McGill pain questionnaire (MPQ). Results indicated that PT improved VAS pain scores somewhat faster than OT and significantly faster than the control group.⁶⁶ PT significantly improved scores on the McGill pain questionnaire (MPQ) compared with OT and the control group at 1 year.⁶⁶ Raw scores are not provided. Physical therapy or pursuit of the treatment goals assigned to that discipline in this study’s methodology (increasing pain control, optimizing coping, extinguishing the source of the ongoing pain, and improving skills) improved the symptomatology of upper limb CRPS more than OT or pursuit of its assigned treatment goals.

The Lee⁶⁷ pediatric CRPS article is often cited to indicate the frequency of therapies for CRPS.⁸ Twenty children with lower limb CRPS received physical therapy once a week for 6 weeks or three times a week for 6 weeks and both groups received six sessions of CBT.⁸ All patients were treated with transcutaneous electrical nerve stimulation (TENS), progressive weight-bearing, tactile desensitization, massage, and contrast baths.⁶⁷ All measures of pain and function improved significantly in both groups after treatment: VAS pain improved from 6.4 to 0.6; VAS effect improved from 5.4 to 0.6; allodynia (7-point Likert scale) improved from 5 to 7 (7 is anchored to no allodynia, 1 is anchored to extreme allodynia); stair climbing impairment score and gait impairment score also improved.⁶⁷ Sustained benefit was observed in most patients at long-term follow-up (average 66 weeks), but was not permanent as the authors note that recurrent episodes of CRPS were reported in one-half of subjects and about one-third eventually received sympathetic blockade.⁶⁷ The possibility of recurrence is noted in a review article by Bialocerkowski and Daly.⁶⁸ The review of the treatment of CRPS in children with PT or OT concluded that “low volume and poor to fair quality evidence which suggests that physiotherapy prescribed with other interventions may lead to short-term improvement in the signs and symptoms of CRPS-1 or functional ability in children with CRPS and relapse rate may be moderately high.”⁶⁸

Additional treatment strategies for children with CRPS include an interdisciplinary day hospital program.⁶⁹ The day hospital program described by Logan et al. demonstrated clinically and statistically significant improvements from admission to discharge in pain intensity (P <.001), functional disability (P <.001), subjective report of limb function (P <.001), timed running (P <.001), occupational performance (P <.001), medication use (P < .01), use of assistive devices (P <.001), and emotional functioning (anxiety, P < .001; depression, P < .01), and functional gains were maintained or further improved at follow-up (range 2-24 mo).⁶⁹ Treatment consisted of intensive daily physical, occupational, and psychological therapies 8 hours a day, 5 days a week for an average of 3 weeks.⁶⁹

Singh et al.⁷⁰ reported on a prospective study of 4 weeks of interdisciplinary management for adult patients with CRPS. The treatment consisted of 20 sessions of PT, 20 sessions of OT, 12 sessions of water therapy, 20 sessions of group psychotherapy, stellate ganglion blocks, and drug therapy that resulted in significantly improved upper limb function for weight tolerance and fine and gross motor skills and physical activity evaluated as maximum isometric force and endurance.⁷⁰ The MPQ was administered but does not appear to have been an outcome measure.⁷⁰ Results seem to be largely observational. After 2 years, 11 of 12 subjects could be contacted and 9 were employed, 8 had the same or less pain, 4 reported the spread of pain, and 5 reported using opioids.⁵

CRPS remains a difficult pain disorder to treat. Old treatment strategies like desensitization, contrast baths, and stress loading are still referred to in the literature. Newer treatment paradigms such as motor imagery program (MIP) which requires exact sequencing of therapy stages have had significant results in the laboratory but have not translated into real world practice. Mirror therapy alters sensory discrimination but has only short-lived effects on pain. It appears that PEPT significantly improves pain and function in patients with upper and lower limb CRPS and have not been shown to cause harm; however, it is indicated only for end stage patients who have failed available treatments. Children with CRPS show significant response to physical therapy given one or three times per week, but improvement may not be lasting and recurrence of CRPS happens one-half the time and one-third of patients opt for interventional treatment.

In the second edition, a physical therapy prescription for primary fibromyalgia syndrome (PFS) included generalized upper and lower limb flexibility and aerobic activity to a heart rate of over 70% predicted maximum.⁶ A review of exercise for fibromyalgia by Busch et al.⁷¹ catalogs the self-administered, community-based, and skilled-therapy exercise for fibromyalgia. Exercise regimens applied to treat PFS include stretch, stretching in a warm pool, Pilates, lifestyle exercise, Tai Chi, yoga, aquatic breathing, Nordic walking, vibration boards, aerobics, strengthening combinations of flexibility, aerobics, strengthening, aquatics, and multidisciplinary programs.⁷¹ Nijs recommends “primary care physical therapy” for patients with FMS that includes education, aerobic exercise, and strengthening exercise.⁷² Nijs recommends against passive treatments, activity management, and relaxation, citing less evidence supporting use as primary treatment.⁷² The search strategy used for this chapter yielded few examples^73–75 of physician-prescribed and physical therapy-supervised exercise for the treatment of fibromyalgia; a secondary literature search of references from the Busch article⁷¹ and others was also undertaken, but few additional examples of PT-supervised exercise for PFS were identified.

Although the descriptions of exercise to treat fibromyalgia in the literature are frequent, the specific trials of skilled PT within the prescriptive authority of a physician are not. It is not clear why skilled therapy does not seem to be used as often, or at least reported in the literature as often, with PFS patients compared to patients with other pain disorders. It is possible that the patients do not select skilled therapy services. The Valencia group commenting on their dropout rate during a stretching trial (25%) postulate that “sometimes, an fibromyalgia patient’s idea of how a rehabilitation program should be, or how it is going to affect their pathology, is often a wishful thinking; and their expectations are poorly satisfied.”⁷⁶ It is possible that the straightforward regimens described, for example walking 45 minutes three times a week or cycling for 60 minutes, do not require the skilled and more expensive attention of a PT. Busch et al.⁷¹ have observed that for patients with PFS, exercises with self-selected intensity appear well tolerated; by contrast, Busch et al. also note that PFS symptoms, like stiffness, may increase or new symptoms, like plantar pain, may emerge if the patient adheres to vigorous and even moderate-intensity exercise.⁷¹ They recommend “supervised programs encouraging PFS participants to perform short bouts of self-selected physical activity” as an initial exercise endeavor, but they then encourage progression to “self-efficacy,” “mastery,” and “modeling” of exercise for the ongoing management of PFS symptoms through instruction and practiced behavioral techniques.⁷¹ The recent review⁷¹ and Cochrane review⁷⁷ by Busch et al are recommended for the physician seeking a detailed discussion of exercise applications for PFS, and caveats for the exercise of PFS patients are provided. If the physician prescribes physical therapy for the treatment of fibromyalgia, aerobic, strengthening, and flexibility exercises may be specified; however, it is best to proceed slowly, to rest and repeat, and to advance weight, duration, or speed of exercise in a slower manner than might be expected for the patient’s age. Education about pacing and about carrying over exercise, and discrimination of old symptoms (tender points) from new ones (e.g., plantar fasciitis) or from postexertional myalgia, and about self-treatment with ice or heat should be reviewed. Patients who wish to pursue alternatives like Tai Chi, Nordic walking, yoga, and aquatic- or gym-based offerings do not require prescription and may not be reimbursed by medical insurance, but advice about pacing, proper athletic and foot wear, and exertional myalgia can be offered as counsel.

Valencia et al. evaluated stretching regimens to treat fibromyalgia.⁷⁶ Patients were randomized to a program of kinesiotherapy and active muscular stretching (self-administered) or to techniques of Meziere’s Global Myofascial Physiotherapy⁷⁶ (global stretching postures and manual myofascial and articular mobilization in relation to the patient’s respiratory dynamics).⁷⁸ Both treatment arms were twice weekly for 150 minutes per week for 12 weeks; with 20 total participants whose average duration of symptoms was 7 years, one-fourth dropped out (3 from control and 2 from Meziere). Both groups achieved a statistically significant reduction in the severity of the disease as measured by fibromyalgia impact questionnaire (FIQ) (Table 96-6) during treatment but then significantly worsened in the follow-up period.⁷⁶ Regarding tender points, while both groups showed improved counts, only the control group’s change was significant; both groups returned to baseline at follow-up (see Table 96-6).⁷⁶ Significant improvement in flexibility was demonstrated in both groups during treatment, and significant improvements remained in the Meziere group at follow-up (see Table 96-6).⁷⁶ Stretching in either form yielded short-term improvement for number of tender points and disease impact with a notable return to baseline in the intermediate term while short- and intermediate-term effects were noted for flexibility.