Poststroke Rehabilitation Care

Recent advances in acute stroke treatment and neurocritical care have led to higher rates of stroke survival, although persistent disability often still occurs. At 6 months poststroke, up to 53% of survivors remain dependent on others for at least 1 of their activities of daily living (ADL).¹ By 2030, an additional 3.4 million adults in the US will have had a stroke,² creating a pressing need for effective collaboration between neurologists and neurorehabilitation experts to optimize function and prevent poststroke complications. In 2016, a writing group of American Heart Association (AHA)/American Stroke Association (ASA) experts with training in physical medicine and rehabilitation (PM&R), neurology, physical therapy (PT), occupational therapy (OT), speech-language pathology (SLP), and nursing convened to review the evidence base and provide a synopsis of best clinical practices in adult stroke rehabilitation.³ Recommendations were ranked using standard levels of evidence and recommendation classes, ranging from Class I (useful and effective) to Class III (potentially harmful). This review incorporates the AHA/ASA guidelines and provides an overview of the organization and practice of poststroke rehabilitation care, including factors affecting recovery prognosis.

Organization of Poststroke Rehabilitation

The Stroke Rehabilitation Team

Stroke rehabilitation requires the coordinated efforts of a multidisciplinary team that includes physicians trained in rehabilitation medicine, nurses, physical therapists, occupational therapists, speech-language pathologists (SLPs), psychologists, nutritionists, social workers, and case managers. The patient and family are furthermore considered essential members of the rehabilitation team, and rehabilitation is organized in a patient-centered manner that considers the patient’s prior level of function and social roles, available level of caregiver support, and the degree of accessibility in the home and community environment to which the patient is discharging.

Rehabilitation Settings

Stroke rehabilitation can take place in many different settings, which are often described using a litany of acronyms that can confuse physicians, families, and patients alike. In the Figure, we try to disambiguate this terminology by depicting the relative levels of medical and rehabilitation services provided at each rehabilitation setting, which can be useful when questions arise from patients and families about discharge planning.

Inpatient Rehabilitation Facilities. The inpatient rehabilitation facility (IRF), also known as the acute rehabilitation hospital, provides hospital-level care for those expected to benefit from an intensive, 24-hour-a-day rehabilitation plan of care carried out under the direct supervision of a rehabilitation-trained physician. Medicare regulations also stipulate IRFs provide at least 3 hours/day of rehabilitation therapy (eg, PT, OT, or SLP) at least 5 days per week. IRFs are staffed by a multidisciplinary team of rehabilitation professionals with training in PM&R, OT, PT, SLP, and rehabilitation nursing. In some care settings, Neurologists with fellowship training in Rehabilitation may provide care alongside or in lieu of PM&R physicians. Social workers, psychologists, psychiatrists, and counselors also play essential roles.

Skilled Nursing Facilities. A skilled nursing facility (SNF) can also provide rehabilitation care to stroke survivors, but at a lower frequency and intensity compared with IRFs. The SNF setting may be appropriate for people recovering from stroke who are unable to tolerate more intensive therapy and whose medical needs can be met by skilled nursing services. Medicare requirements for SNFs only stipulate that a physician sees the patient at least once every 30 days for the first 90 days after admission, and at least once every 60 days thereafter.^a There is no requirement for direct supervision by a rehabilitation-trained physician, and although some SNFs contract with rehabilitation physicians as consultants to guide the plan of care, more often the physician role is staffed by internists or hospitalists.

The AHA/ASA guideline recommends “stroke survivors who qualify for and have access to IRF care receive treatment in an IRF in preference to a SNF” (Class I, Level B). Studies that compared outcomes have generally shown greater functional recovery and rates of return to community living after discharge to IRFs compared with SNFs or nursing homes. Discharges to SNFs also have higher rates of rehospitalization and poorer survival. It is important to temper interpretation of these results by considering the limitations inherent to these studies, including reliance on administrative data, observational designs, few study sites, and most notably, potential differences in patient case mixes between settings that could bias findings.

Recent studies try to account for several of these important potential confounders. For example, in a cohort with no significant differences in age, comorbidities, infarct volume, or recanalization rates, those discharged to SNF vs IRF had significantly worse outcomes.⁴ Another cohort study of 99,185 stroke survivors found that receiving rehabilitation services at an IRF vs SNF was associated with greater improvement in mobility and self-care after accounting for patient, clinical, and facility characteristics at admission.⁵ The differences may be explained by differing frequency and intensity of therapy services received, level of medical surveillance available, and degree of subspecialty expertise available to generate rehabilitation programs tailored to the patient’s specific needs and limitations.

Secondary Complications of Stroke

Spasticity

Spasticity is common after stroke with estimated prevalence of approximately 25%; incidence of spasticity after a first stroke is nearly 40%.⁶ The most well-known definition of spasticity describes it as “a motor disorder characterized by a velocity-dependent increase in tonic stretch reflexes (muscle tone) with exaggerated tendon jerks, resulting from hyperexcitability of the stretch reflex.”⁷ Clinically, patients may present with muscle stiffness, spasms, joint contractures, muscle fatigue, and abnormal posture, increasing the risk of complications (eg, pain, pressure ulcers, joint deformity, and poor hygiene) and contributing to disability and decreased quality of life.

Quantifying and consistently tracking the degree of spasticity using well-established metrics is essential to evaluate progression and response to treatment. The rate of change in joint angle is of critical importance because spasticity is velocity dependent by definition. Spasticity is commonly assessed using either of 2 well-known scales; the Modified Ashworth Scale (MAS; Table 1) or the Tardieu Scale. In clinical practice, the MAS is used most often in adults.³ In brief, when testing flexor muscles, spasticity is assessed by placing the joint crossed by that muscle in a maximally flexed position, then moving to a position of maximal extension over 1 second. The opposite applies when assessing extensor muscles, starting in the maximally extended position, then moving to maximal flexion. The assessor then notes the degree of tone elicited, and the range of motion in which it occurs.

Spasticity treatment options include pharmacologic and nonpharmacologic treatments, with surgical treatments or implantable pumps being reserved for more refractory cases. Common nonpharmacologic treatments include stretching, splinting, bracing, or serial casting to improve the range of motion. There is also evidence that neuromuscular electric stimulation (NMES) can improve range of motion and help decrease spasticity when combined with other interventions.⁸ Pharmacologic treatment options include systemic oral medications, injectable medications for neurolysis (eg, botulinumtoxin or phenol), and intrathecal therapy (eg, baclofen pumps). Common oral agents include baclofen, diazepam, clonazepam, tizanidine, and dantrolene (Table 2). The AHA/ASA guidelines note that oral antispasticity agents can be useful for generalized spastic dystonia (Class IIa, Level A), but also caution regarding dose-limiting sedation or other side effects.

Decision making about the choice of agent and route of delivery is largely driven by whether the spasticity being treated is diffuse/generalized or can be addressed by targeting a few key muscles. When a few muscles can be targeted for strategic injection to improve a specific functional task, focal injections may offer a good alternative or adjunct to systemic medications. The AHA/ASA guidelines recommend targeted injection with botulinum toxin A to reduce spasticity, improve range of motion, and improve dressing, hygiene, limb positioning, and spasticity that interferes with gait function (Class I, Level A). Phenol is less costly and can be used as an adjuvant to reduce the total amount of botulinum toxin needed for poststroke spasticity management without increased side effects.⁷ For severe spastic hypertonia unresponsive to other interventions, intrathecal baclofen therapy may be useful (Class IIb Level A). In clinical practice, a combination of systemic, injection, and implantable therapies is commonly utilized.

Key to effective spasticity treatment is comprehensive assessment and quantification of symptoms and identification of specific functional goals. For example, improving heel contact with the floor to stabilize gait by reducing ankle plantar flexor tone offers a well-defined goal based on specific muscle groups to be treated. Once an appropriate treatment goal is identified and baseline measures of spasticity are obtained, the treatment plan should be strategically planned to meet the individual’s needs, starting with the least invasive treatments and escalating based on symptoms and risk of adverse effects. Because spasticity can be exacerbated by noxious stimuli, careful evaluation to identify and treat secondary triggers of spasticity (eg, urinary tract infection [UTI], bowel or bladder distention, venous thromboses, skin lesions, infections, poorly fitting equipment, or constrictive clothing) should be prioritized.

Urinary Dysfunction

Managing bladder function is an essential aspect of stroke rehabilitation. Poststroke urinary dysfunction is common and negatively affects outcomes and quality of life. Of people hospitalized poststroke, 40% to 60% experience urinary incontinence, 25% have persistent incontinence upon hospital discharge, and 15% remain incontinent 1 year after stroke.⁹ Notably, not all urinary incontinence stems from a primary neuro-urologic problem. Functional incontinence is commonly caused by other primary issues (eg, mobility, communication, or cognition) despite a physiologically intact genitourinary system.¹⁰

Conceptually, bladder dysfunction can be divided into 2 broad categories as the inability to store urine (eg, bladder spasms causing urge incontinence) or the inability to empty urine (eg, detrusor weakness resulting in overfilling). The most commonly reported etiology of poststroke urinary dysfunction is detrusor hyperreflexia, presenting with urinary frequency and urgency that most often results in urge incontinence. Detrusor weakness and hyporeflexia may also occur, affecting 21% to 47% of stroke survivors within the first 72 hours poststroke with an initial loss of bladder tone that results in hesitancy, dribbling, and incomplete bladder emptying.^9,11 Urostasis associated with detrusor weakness can predispose individuals to a UTI. Conversely, UTI, which occurs in approximately 10% to 28% of stroke survivors, can lead to incomplete bladder emptying by contributing to an inflammatory outflow obstruction. UTI is associated with increased length of stay, poorer functional outcomes, and higher rates of institutionalization.^12,13 Risk factors for UTI include female sex, increasing age, higher modified Rankin Scale score, and urinary retention (ie, postvoid residual [PVR] volume >100 mL).¹⁴

Common strategies for poststroke management of urinary dysfunction in the AHA/ASA guidelines include: 1) behavioral strategies such as prompted/timed voiding (eg, every 4 hours); 2) assessing for urinary retention (eg, bladder scanning protocol); and 3) pelvic floor exercises. Guidelines further highlight the importance of obtaining a detailed history of any prestroke urologic issues, and assessing for urinary retention after an attempted void via bladder scanning or by recording the volumes obtained upon intermittent catheterization (Class 1, Level B). The guidelines do not endorse specific medications for poststroke urinary dysfunction, in part due to a lack of high-quality evidence. Studies on bladder management in the poststroke population are an important priority for future research.

Skin Breakdown

Often related to urinary incontinence and spasticity, skin breakdown is a common poststroke complication that should be proactively assessed. It is especially important to check for areas of skin breakdown associated with malpositioning of a spastic limb (eg, elbow flexion causing pressure on the olecranon) and around any splints/braces (eg, ankle-foot orthoses) during routine follow up. Counseling to check for blanchable versus nonblanchable erythema during dressing and bathing tasks is a good teaching point, and AHA/ASA guidelines recommend that patients, staff, and caregivers be specifically educated about preventing skin breakdown (Class I, Level C).

Shoulder Pain

Poststroke loss of arm function causes shoulder pain in up to 70% of survivors and arises from etiologies that include subluxation, impingement, neuropathic pain, adhesive capsulitis and spasticity. The AHA/ASA guidelines support patient and family education on positioning and range of motion (Class I, Level C) and suggest it is reasonable to consider the use of supportive devices and slings for shoulder subluxation (Class IIa, Level C), or NMES for shoulder pain (Class IIb, Level A). Proper positioning is important to reduce mechanical strain at the shoulder. Distally, this may include supports for the distal forearm (eg, an arm trough or Givmohr sling). Proximally, support with athletic taping to reposition the scapula or with NMES to engage the muscles of the shoulder girdle and proximal arm can be trialed.

Factors Affecting Prognosis for Functional Recovery

Recovery trajectories very depending on the specific neurological domain(s) affected (eg aphasia and paraplegia show differing responsiveness to treatment).^15-17 The prognosis for poststroke recovery also varies widely among individuals because each person’s recovery pattern uniquely reflects the combined influence of their lesion size and location, baseline health status, time to initial treatment, and response to medical treatment or rehabilitation, as well as other intrinsic and extrinsic factors. Models to predict recovery potential are an area of vigorous ongoing research. In the Early Prediction of Functional Outcome After Stroke (EPOS) study, when voluntary finger extension and shoulder abduction were present 48 hours poststroke, the probability of achieving hand dexterity at 6 months (measured by the Action Research Arm Test) was 98%. This simple bedside test offers a useful starting point for discussions about prognosis. Testing early shoulder abduction and finger extension is often referred to by the acronym SAFE, and is now an essential component in the more nuanced Predict Recovery Potential (PREP)¹⁸ and PREP2 algorithms.¹⁹ PREP2 has been shown to correctly predict upper extremity recovery at 90 days in up to 75% of stroke survivors and maintains comparable accuracy at 2 years poststroke.²⁰

Timing Initiation of Rehabilitation

The question of when to begin poststroke rehabilitation has long been a source of controversy. Concerns about the safety of intensive rehabilitation initiation during the acute poststroke phase have cast a long shadow since a trial showed negative effects of early intensive mobilization on 3-month outcomes after severe strokes.²¹ It is important to note, however, that the specific therapy under investigation involved upright mobilization. We argue that these results should not be interpreted as a general caution against early rehabilitation. Rather, these results remind us that when designing early mobility programs, it is important to consider the type and intensity of the exercise prescribed in the context of the individual’s stroke severity and etiology, giving particular consideration to the potential adverse effects of positional changes on blood pressure and cerebral perfusion. In cases where the risk of hypoperfusion with upright positioning is a significant concern, there are therapy options that can be delivered in the supine position, including some emerging wearable technologies to provide adaptive training in bed.²²

Another important recent study²³ on the timing of poststroke rehabilitation assessed whether individuals are optimally responsive to motor training during a specific time period after stroke. Participants were randomly assigned to receive standard therapy only or standard therapy plus a 20 hour bolus of additional motor training. This additional training bolus was delivered during either the acute (30 days), subacute (2-3 months) or chronic (≥6 months) stage poststroke. Relative to those who received standard therapy only, at 1 year participants who received additional motor training in the subacute period had significantly increased motor function. Those who had additional motor training in the acute period had a significant, although smaller, improvement. In contrast, those given intensive motor therapy in the chronic period had no significant improvement compared with those who received standard care.

The authors of the study concluded an additional bolus of intensive motor training was most effective within the first 2 to 3 months after stroke. The subsequent paraphrase in Neurology stating that intensive therapy “may provide significant motor improvements if initiated 60 to 90 days after acute ischemic stroke”²³ is, however, an oversimplification. The difference in treatment effects in the acute (<30 days) vs subacute groups (2-3 months) was 1.62 points, which is nominal because the mean clinically important difference is 5.7 points for this metric. The distinction is not trivial. Such paraphrased statements may be cited by insurers as justification to deny earlier intensive rehabilitation therapies, even though the timing and amount of therapy received as standard of care was neither controlled for nor considered as a covariate in this study. It is important for neurologists and rehabilitation physicians to recognize these facts so that they can advocate on behalf of their patients to obtain the services most appropriate to the individual’s degree of medical complexity and rehabilitation needs.

Conclusion

Stroke rehabilitation relies on the coordinated efforts of experts in multiple specialties using multidisciplinary approaches that are tailored to the individual patient’s deficits and goals of care. Although the optimal timing and dosing of therapy remain a topic of debate, there are clear benefits to rehabilitation after stroke based on an extensive literature reviewed in the AHA/ASA guidelines. Neurologists can play a central role in the stroke recovery process by understanding the basic principles of rehabilitation outlined here, which will better allow them to advise patients and caregivers on the natural history of stroke recovery, to ensure that therapy and medical services are provided in the appropriate location, and to screen for common complications that arise during the course of stroke recovery.