A model for multisystem evaluation treatment of individuals with Parkinson's disease.

A Model for Multisystem Evaluation Treatment of Individuals with Parkinson's Disease In this article, we describe a method for systematically making decisions regarding evaluation and treatment of the individual with Parkinson's disease. Parkinson's disease is a progressive degenerative disease of the central nervous system that leaves the patient increasingly immobile and dependent on others for assistance. The cardinal signs of Parkinson's disease are rigidity, bradykinsia, tremor, and postural instability. [1,2] In addition, there may be autonomic nervous system abnormalities including hypotension, excessive salivation, and sweating. [3] The clinician who treats the patient with Parkinson's disease is faced with decisions regarding which of the patient's problems can be corrected, which require compensatory strategies, and which are intractable to physical therapy intervention. The determination of which problems can and cannot be corrected is particularly difficult in Parkinson's disease becuse of the progressive, degenerative nature of the disorder.

The system we present for making decisions regarding evaluation and treatment of the patient with Parkinson's disease is based on a model that we recently developed for evaluation and treatment of individuals with neurologic dysfunction. [4] In this article, we apply that model to the patient with Parkinson's disease. We first demonstrate use of the model to interpret the underlying causes and symptoms of Parkinson's disease. We next demonstrate how the model can be used for decision making regarding physical therapy evaluation, monitoring, and treatment of the patient with Parkinson's disease. Finally, we demonstrate how the model can be used to focus research related to the efficacy of physical therapy intervention for the person with Parkinson's disease.

Definition of Impairments and

Disabilities of Patients with

Parkinson's Disease

The first step in using this model is to differentiate between the impairments and the disabilities of Parkinson's disease. (A glossary of all terms and definitions used in this article is provided in the Appendix.) Impairments refer to the abnormalities of anatomic, physiologic, or psychologic origin within specific organs or systems of the body. [5,6] Rigidity, bradykinesia, tremor, and postural instability are the cardinal signs of Parksinon's disease [1,2] and would all fit the definition of impairments of the CNS. Disability refers to functional restriction or inability to perform within the range of normal. [5,6] Functional disability can be subdivided into four domains: physical, mental, emotional, and social. [6] By these definitions, difficulty with bed mobility, transfers, and gait are all physical disabilities. The dementia, depression, social withdrawal, and isolation that accompany Parkinson's Disease [7] could be considered mental, emotional, and social disabilities. Individuals may have impairments from a disease without being disabled by the disease. [6] Early in the disease, for example, an individual may have impairments of rigidity and tremor but may be functioning within normal limits in the physical, mental, social, and emotional domains. The individual with Parkinson's disease is likely to develop disabilities in some or all of these domains as the disease progresses.

In summary, impairments relate to abnormalities within components of human function, whereas disabilities relate to abnormalities within the individual's total capacity to function. In evaluating a patient, it is useful to differentiate between the patient's impairments and disabilities. The patient's disabilities are the outcomes of the total impact of his or her impairments; the impairments are the underlying cause. Disabilities may be best corrected when the underlying impairments are identified and addressed. We will demonstrate this approach in the remainder of this article.

The second step to using this model is to try to differentiate between the origins of the various impairments that arise. First, do impairments occur that are the direct result of nervous system pathology? Second, do impairments occur in systems other than the CNS? Third, do impairments occur that are a composite effect of the CNS and non-CNS impairments? Figures 1 and 2 provide examples of impairments that are direct, indirect, and composite effects of CNS pathology.

The cardinal signs (impairments) of Parkinson's disease can be attributed directly to nervous system pathology. [2,8-10] We therefore consider these impairments to be direct effects of the lesion. In addition, other impairments such as autonomic nervous system dysfunction, dementia, and dyskinesia [3,7,10,11] can be attributed directly to the CNS pathology. Musculoskeletal [12,13] and cardiopulmonary [14] impairments are examples of impairments in systems other than the CNS, as illustrated in Figure 1. We therefore consider these impairments to be indirect effects of the lesion. Many of the impairments of the patient with Parkinson's disease can be considered composite effects of the disease resulting from a combination of nervous system impairments (direct effects) and non-nervous system impairments (indirect effects).

The differentiation between direct, indirect, and composite effects of the pathology of Parkinson's disease serves to emphasize that impairments occur both within the CNS and as sequelae in systems other than the CNS. It may be difficult or impossible for the physical therapist to correct those impairments that arise directly from nervous system pathology. It may be much more realistic for the physical therapist to prevent or reduce the effects of the lesion that lead to impairments in other systems.

Postural instability, or faulty balance, associated with Parkinson's disease is an illustration of an impairment that occurs because of composite effects of the disease (Fig. 2). The balance disorder has been considered a direct effect of the disease process [15] because of the brainstem and basal ganglia pathology. [8,10] In addition, however, faulty balance occurs as a consequence of other direct effects of the lesion including rigidity and hypokinesia. n10 Furthermore, as illustrated in Figure 2, musculoskeletal impairments, which are considered indirect effects of the lesion, could also contribute to faulty balance. Specifically, we propose that lack of trunk and pelvic mobility could provide a mechanical limitation to the patient's capacity to perform a balance response. The reader can experience the effects of a typical parkinsonian posture on his or her balance by standing with flexed knees, kyphotic trunk, and a forward head and then attempting to resist a backward shove by an associate. In addition, vestibular and neck afferent proprioceptive feedback are used in the balance response. [16] We therefore predict that lack of neck mobility would decrease the proprioceptive input and would contribute to the faulty balance.

In the next section of this article, we will demonstrate how a knowledge of the pathology of Parkinson's disease can lead to a prediction of the direct and indirect effects of the disorder. We then demonstrate that many of the impairments of Parkinson's disease may have multiple underlying causes. Some of the impairments that may be due to composite effects of the disease include faulty balance, bradykinesia, decreased arm swing in gait, drooling, difficulty swallowing, hypoventilation, and fatigue.

Pathology of Parkinson's

Disease and Its Direct Effects

It has long been known that individuals with Parkinson's disease have a degeneration of a group of pigmented neurons in the substantia nigra of the midbrain. [8] These neurons produce dopamine, which is normally transmitted to the basal ganglia. [8] A simplistic interpretation of Parkinson's disease is that loss of dopamine in the basal ganglia results in an imbalance between the dopamine and other neurotransmitters in the basal ganglia.

Some of the other neurotransmitters identified in this imbalance include acetylcholine (ACh), glutamate, asparatate, and gamma-aminobutyric acid (GABA). This imbalance is thought to be the cause of many of the signs and symptoms of Parkinson's disease. [8,10] Recently, however, it has been observed that there are different dopamine receptors in the basal ganglia, and these receptors may also undergo progressive pathological changes as a result of Parkinson's disease. [8,10] Furthermore, it is probable that dopamine is not the only neurotransmitter affected by Parkinson's disease. [8,10]

Much remains to be learned about the pathophysiology of Parkinson's disease. Nevertheless, many of the impairments of the disease can be explained by the neuroanatomic and neurophysiologic information that is currently available (Table). Both the brainstem and the basal ganglia have been identified as structures that contribute to mechanisms of postural responses. [10,15,16] Thus, faulty balance likely occurs, in part, as a direct effect of Parkinson's disease. The basal ganglia have been identified as one of the important structures in motor planning and programming. [17,18] Basal ganglia dysfunction has been implicated in the dyskinesia, such as resting tremors that occur with Parkinson's disease. [9,19] This unique form of rigidity, clinically described as "plastic" or "lead-pipe rigidity," is specifically associated with lesions of the basal ganglia [19] and may be linked with an alteration of muscle stiffness. [20] Deficits in the alerting mechanisms have been attributed to disturbances of projections between the basal ganglia and the cortex. [17,21] These deficits may be one of the factors that contribute to the difficulty in initiating movement and the hypokinesia that is typical of Parkinson's disease. [17,21] Brainstem dysfunction has been implicated in the impairments of autonomic nervous system functions that lead to hypotension, profuse sweating, and excessive salivation of the patient with Parkinson's disease. [2,3] Finally, cognitive dysfunction is considered a sign of Parkinson's disease. [7] It is unclear at this time whether the cognitive failure is due to cortical degeneration or to subcortically mediated defects of planning and attention. [22]

The presentation of Parkinson's disease can vary between patients and can even vary for a given patient during the course of the illness. [23,24] Current concepts of the pathophysiology of Parkinson's disease help to explain this variability. It is possible that the predominance of symptoms that an individual experiences may depend on which neurotransmitters are most affected by the disease. [10] This predominance of symptoms may account for the finding that some patients with Parkinson's disease are more impaired by rigidity and others by tremor. Drug therapy differs depending on the patient's symptoms. The chosen therapy reflects the difference in pathology involved. Early in the disease, when loss of dopamine-producing neurons occurs, the basal ganglia attempt to compensate for the decrease in available neurotransmitters. [10] This compensation is accomplished by a proliferation of receptors for dopamine in the basal ganglia neurons. [10] Later, as the disease progresses, there is progressive loss of dopamine receptors in addition to the loss of dopamine-producing cells. [10] Early in the disease, while there is a compensatory proliferation of dopamine receptors, dopaminergic agents such as carbidopa (Sinemet[R] *1) may be effective in replacing dopamine. Later, when the dopmaine receptors have also degenerated, such agents cannot be effective. [10,25]

Furthermore, other factors such as gastrointestinal absorption and blood-brain barrier transfer of medications (eg, Sinemet[R]) may account for the temporal variability of response to drug treatment. [24] A peak effect may occur as the drug is absorbed, and then the benefit decreases as the drug is metabolized. Thus, an individual's symptoms may wax and wane throughout the day in relation to time of medication. Early in the disease, the patient has a greater tolerance to fluctuations in drug concentration. Later in the disease, the "window" of tolerance changes and loss of effectiveness or toxic side effects can occur.

In summary, the neuroanatomical pathology of Parkinson's disease allows the physical therapist to predict some of the expected impairments associated with the disease. Pathophysiology helps to explain the variability among patients, within a given patient over time, and within a given patient during the course of the day. Parkinson's disease was originally considered to be a disease only affecting dopamine production in the substantia nigra. In recent years, it has become apparent that the disease is considerably more complex, involving mesolimbic, thalamic, cortical, and subcortical structures and affecting a variety of both neurotransmitters and receptors. [10]

Indirect Effects of Parkinson's Disease

Over time, the patient with Parkinson's disease develops a typical stooped-flexed posture that is characteristic of the disease. [12] This posture appears to result from the effects of rigidity, which especially affects the flexor musculature. In addition, the patient may become increasingly sedentary, and this sedentary life style may contribute to the patient's tendency to remain in postures of flexion. Eventually, muscle length changes may occur so that the individual loses the musculoskeletal flexibility required for extension. This loss of musculoskeletal flexibility has not been well documented in the literature, although a few studies are available related to loss of range of motion of specific muscles. [13] Nevertheless, clinical observation and experience would predict that chronic postural changes result in chronic length changes of muscle, an impairment of the musculoskeletal system that is indirectly caused by the neuroanatomic pathology. Unllike the impairments that arise directly from Parkinson's disease, there is little literature documenting the indirect musculoskeletal impairments. The literature, however, does contain commonly accepted clinical descriptions of these changes, and some experimental evidence does exist documenting the extent of musculoskeletal limitations. In this section, we summarize known and presumed musculoskeletal impairments of the patient with Parkinson's disease and discuss some of the predicted consequences of these impairments.

Musculoskeletal impairments appear to begin proximally, [12,26] affecting contractile and noncontractile tissue length and flexibility first of the trunk and pelvic girdle and then of the moral distal musculature. Impairments may frequently be observed first unilaterally and later bilaterally. [27] The muscle groups of the neck, thorax, and limbs become shortened and limit extension throughout these structures. [12,26,27] The antagonistic muscle groups become concomitantly lengthened. In addition, it is well known that rotation is difficult in the patient with Parkinson's disease. [12] This impairment suggests that the rotator muscles become contracted, or "tight," limiting trunk and limb rotation. The patient's movements appear increasingly stiff and are dominated by flexion and extension with limited side bending and rotation even when required. This predominance of straight-plane motions suggests that musculoskeletal flexibility for rotation is lost.

Musculoskeletal impairments may contribute to the patient's apparent bradykinesia. For example, these individuals have difficulty with motions such as rolling or turning, whether in sitting or standing. We suggest that these motions become slow, in part, because of the loss of capability for rotation. The reader can experience the impact of loss of trunk rotation and the resulting slowness of movement by attempting to roll from a supine or prone position without rotation (log roll).

Loss of trunk flexibility may occur anywhere throughout the vertebral column and may contribute to the patient's overall difficulty. [12] For example, loss of extension of the pelvis on the femur (loss of hip joint extension) may result from tightness of muscles such as iliacus muscle. This tightness may prevent posterior movement of the pelvis for a posterior pelvic tilt. Loss of extension of the lumbar and thoracic spine frequently occurs as well, as evidenced by the kyphotic posture of the patient with Parkinson's disease. Combined tightness of soft tissue, ligaments, and joint capsules may leave the patient unable to either posteriorly or anteriorly tilt the pelvis. This inability to tilt the pelvis would probably severely limit the individual's capability to use normal movement patterns in standing and in postural responses. Tightness of muscles and soft tissues of the lumbar and thoracic spine could contribute to loss of trunk rotation and loss of lateral flexion. Combined shortness of the rotator muscles, such as the abdominal muscles throughout and the short extensors in the lumbar region, may contribute to the loss of rotation and lateral flexion. We propose that these losses in mobility further compound the difficulties of the patient with Parkinson's disease, contributing to the consequent impairments of the disease. We suggest that ultimately musculoskeletal impairments contribute to the patient's total physical disability, restricting all activities of daily living.

The ability of a person to rise from a chair or weight shift while sitting appears to rely heavily on functional mobility of the pelvis fand lumbar spine. We propose that loss of these normal motions seriously limits normal functional movement, potentially interfering with all ADL.

We propose that lateral flexion of the lumbo-thoraco-cervical spine is required for the "trunk elongation" that occurs during righting reactions and that limitation in trunk lateral flexion would therefore limit righting reactions. Likewise, we assume that lack of thoracic extension, rotation, and lateral flexion interferes with the extension and trunk elongation components of the balance response, potentially contributing to falls, fractures, and hospitalization of individuals with Parkinson's disease. [28] Lack of thoracic mobility could also affect respiration by reducing vital capacity. Respiratory complications are the most common causes of death of the individual with Parkinson's disease. [27,29] The decreased chest expansion attributable to tight thoracic musculature presumably contributes to the respiratory disability of these patients, perhaps even contributing to terminal pneumonia. Alterations in muscle length of the cervicothoracic spine might contribute to poor closure of the mouth. Hence, these alterations may contribute, with the brainstem autonomic nervous system nuclei changes, [2,3] to impaired swallowing with drooling, which in turn contributes to the patient's social disability.

The stooped posture and shuffling gait of the parkinsonian are hallmarks of the disease. [12] For these individuals, step lengths are often short, there is a lack of dorsiflexion on heel-strike, counterrotation of the trunk is absent, and there is a loss of reciprocal arm swing. [30] The musculoskeletal limitations associated with Parkinson's disease might contribute to this characteristic gait. This influence is illustrated by an analysis of gait that could result with specific musculoskeletal limitations. For example, if an individual has soft tissue tightness combined with hip flexor muscle tightness while holding the pelvis in a posterior tilt relative to the spine, then knee flexion of the stance extremity might be required to allow the swinging foot to hit the ground. Hip flexor muscle tightness could prevent adequate hip extension during the late stance phase of gait, and knee flexor muscle tightness could limit knee extension in both the stance and swing phases of gait, contributing to a shortened step length.

An important determinant of gait, contributing to lengthening the swing extremity, is the lateral list (tilt) of the pelvis [31] accompanied by protraction of the pelvis, which is accomplished by rotation of the pelvis relative to the stance femur and to the lumbar spine. Tightness of hip rotator and lumbar lateral flexor musculature might preclude these motions from occurring. Together, these factors could result in a shortened step length. Gastrocnemius muscle tightness could limit dorsiflexion, preventing the normal heel-strike at initial contact. The shortened step length combined with a flatfoot at contact could then contribute to the shuffling character of the parkinsonian gait. Loss of trunk counterrotation may occur if gait is slow (as occurs in Parkinson's disease). Additionally, if these is a lack of available mobility required for trunk rotation, the counterrotation cannot occur. Reciprocal arm swing is designed to counterbalance trunk counterrotation. When gait is slow and when trunk rotation is reduced, arm swing would not be expected to occur.

In summary, the typical posture, functional movement, and gait of the patient with Parkinson's disease may occur in part because of musculoskel-etal constraints. The reader can experience the impact of these musculoskeleted limitations by assuming a typical parkingsonian posture and then trying to rise from a chair, walk, or turn 360 degrees.

Cardiopulmonary deconditioning provides another example of a non-nervous system impairment of the patient with Parkinson's disease. [13] This type of impairment is illustrated in Figure 1. Because of the rigidity and hypokinesia, the patient may become sedentary. This sedentary condition could eventually contribute to the observed cardiopulmonary deconditioning. This deconditioning represents an impairment in a system other than the nervous system. The deconditioning as described results indirectly from the nervous system pathology. Furthermore, the typical kyphotic posture (a musculoskeletal impairment) may because an additional sources of the Cardiopulmonary impairment (Fig. 1): Limited chest expansion could contribute to the decreased vital capacity associated with Parkinson's disease. Decreased vital capacity to can lead to impaired coughting can lead to impaired coughing, inability to clear secretions, and susceptibility to pneumonia. [14] The kyphotic posture could also contribute to decreased cardiac output. Decreased vital capacity, combined with decreased cardiac output, could in turn contribute to the individual's deconditioned state.

Musculoskeletal and cardiopulmonary impairments have been designated indirect effects of the pathology of Parkinson's disease because they represent impairments in a body system other than the CNS. Impairments that arise indirectly do not occur initially but develop over time. As postural deformity becomes fixed, and as cardiopulmonary deconditioning increases, motion is presumably slower, more laborious, and awkward and requires more energy and effort. These musculoskeletal impairments that arise appear to further compound the motor difficulties that occur directly as a result of the lesion.

Composite Effects of

Parkinson's Disease

Many of the impairments of the patient with Parkinson's disease that physical therapists treat represent a composite of direct and indirect effects of the disease process. Faulty balance, caused by impairments of both the CNS and the musculoskeletal system, serves as an example (Fig. 2). Our current knowledge of the disease process also suggests several additional potential causes of faulty balance such as impairment of alerting mechanisms and impairment of motor planning and programming.

Bradykinesia is another impairment that can arise because of composite underlying causes. Bradykinesia has been identified as a direct effect of Parkinson's disease. [9,18,19] As outlined in Figure 3, however, it is becoming increasingly evident that multiple potential causes of bradykinesia exist. [9,17,18] Bradykinesia can result because alerting mechanisms are impaired [9]; because motor programs are not well coordinated for efficient use [18]; or because the patient's motor-planning capability is faulty, precluding smooth integration of multiple motor plans. [17] In addition, musculoskeletal limitations that prevent rotatory movements also possibly contribute to bradykinesia. The reader can experience this limitation by attempting to roll without trunk rotation or to turn in stance without pelvifemoral (hip joint) rotation. Faulty balance may also contribute to bradykinesia; an individual may move more slowly for fear of falling. Finally, fatigue and decreased cardiopulmonary endurance might be contributing factors in bradykinesai. [13] As illustrated in Figure 3, some of these potential contributions to bradykinesia are direct effects of the lesion; others are not.

Drooling is another impairment of Parkinson's disease that has been attributed directly to the nervous system pathology but that might be due to composite effects of the lesion (Fig. 4). Drooling is considered to result from the increased salivation that occurs as an autonomic nervous system dysfunction. [3] We propose, however, that the typical forward head-neck posture and accompanying musculoskeletal limitations may make it difficult for the patient to swallow, therefore contributing to the drooling. Thus, drooling may be an impairment with composite causes. The reader can assume a posture of excessive neck flexion, as occurs in Parkinson's disease, and attempt to swallow water to experience the effect that musculoskeletal alignment might have on drooling. Faulty balance, bradykinesia, and drooling are only a few of the impairments of the patient with Parkinson's disease that may have multiple underlying causes.

So far we have considered the impairments and disabilities that arise from pathology. Additionally, disabilities may be a further source of impairment. For example, depression and withdrawal may contribute to the bradykinesia and to postural alterations. That is, composite impairments may be a consequence not only of direct and indirect effects of a lesion but also of the resulting disability.

Figure 5 provides a summary of the model we have developed as applied to Parkinson's disease. As shown in Figure 5, both the direct and indirect effects of the degenerative processes of Parkinson's disease may contribute to impairments that are composite effects. Ultimately, all impairments may contribute to the patient's disability. Furthermore, disability may feed back into the cycle, further contributing to impairments and to the resulting disability.

Evaluation, Interpretation,

and Treatment

The purpose of the model we have developed in to allow the clinician to make educated and systematic judgments about what to evaluate, how to interpret evaluation findings, and what different intervention strategies can and cannot do for the patient with Parkinson's disease. The potential direct, indirect, and composite effects of Parkinson's disease discussed in this article should be systematically evaluated. Findings should be interpreted and hypotheses made regarding which impairments are composite effects of direct and indirect impairments. The role of specific impairments in producing the patients' total disability should then be analyzed. We suggest specific guidelines for evaluating patients, setting goals, and designing treatment programs (see accompanying article by Schenkman et al in this issue). We will focus next on how knowledge of impairments leads us to a systematic interpretation of the relative contributions of physical therapy and pharmacologic management in remediating the patient's disability.

We propose that impairments of Parkinson's disease that are a direct result of CNS pathology may be intractable to physical therapy. For example, we would not expect physical therapy intervention to correct fundamental neurotransmitter defects in alerting mechanisms or in motor planning and programming mechanisms. Physical therapy, however, might be effective in teaching the patient compensatory mechanisms for these impairments and how to reduce potential sequelae. For example, the physical therapist may be able to teach the patient to temporarily reduce rigidity through techniques of self-relaxation and to utilize the remaining postural response mechanisms through conscious voluntary practice. We suggest that reducing the consequences of impairments that occur directly through the CNS damage is of greatest value if this task is coupled with reducing or preventing musculoskeletal impairments or with improving specific functional activities.

Pharmacologic intervention is directed toward restoration of a normal balance of neurotransmitters and, therefore, in contrast to physical therapy management, is presumed to be most effective in reducing the impairments that are direct effects of the lesion. (Pharmacologic intervention, however, would not be expected to reverse all chronic muscle-length changes.) In general, dopamine agonists such as L-dopa (levodopa) are most effective when rigidity is the predominant feature of Parkinson's disease; the anticholinergics are most effective when tremor is the predominant symptom. Anticholinergic drugs have not proven effective against hypokinesia. [32] There is no evidence in the literature regarding which agents, if any, are most effective in remediating the balance disorder of the patient with Parkinson's disease.

Pharmacologic intervention is not without drawbacks. Side effects have been a major problem with all pharmacologic agents used in treatment of patients with Parkinson's disease. These side effects include nausea, vomiting, headache, fatigue, dizziness, psychosis, and involuntary movements (dyskinesai). [10,33] One problem with levodopa is that in many patients it has a narrow therapeutic window. In a given individual, too much levodopa may induce drooling, involuntary movements, mental change, and sleep disturbances, whereas too little levodopa leaves the patient with stiffness, difficulty of movement, and tremor. A variety of unwanted effects including toxicity may occur with long-term use of drugs such as levodopa, [1,10,34] Some individuals experience a "wearing-off" phenomenon in which the signs of Parkinson's disease reoccur several hours after medication. [33] Some individuals experience an "on-off" phenomenon in which there is a sudden return of symptoms unrelated to the time of medication. [33,35] Other individuals develop involuntary movements, and for some the drug loses efficacy. There are also non-nervous system side effects (indirect impairments) that result from pharmacologic intervention. For example, bromocriptine can cause pulmonary fibrosis and respiratory decompensation. [14]

Because the use of pharmacologic agents presents its own set of problems for the patient with Parkinson's disease, it is advantageous to delay onset of drug therapy as long as possible. [34] We hypothesize that physical therapy management of impairments that occur indirectly may be beneficial in delaying the need for pharmacologic intervention.

Musculoskeletal impairments of patients with Parkinson's disease may be the most responsive to physical therapy intervention and the least responsive to pharmacologic management. We propose that appropriate physical therapy intervention can delay, reverse, or prevent loss of soft tissue length and flexibility, thereby reducing musculoskeletal contributions to consequent impairments of postural adjustment, bradykinesia, and cardiopulmonary conditioning. We further suggest that cardiovascular conditioning may reduce the cycle of worsening cardiac and pulmonary complications for these individuals and may contribute to their overall fitness. We hypothesize that physical therapy management can significantly improve the functional ability of the patient with Parkinson's disease, especially if it is initiated early in the disease before fixed musculoskeletal alternations occur. Increased physical capability and increased self-control over the disease process should also have an indirect impact on the patient's social and emotional disabilities.

An understanding of the relationship between pathology and its direct and indrect effects is helpful in identifying the types of approaches that may be most effective. For example, rigidity appears to occur in part because of disruption in the patient's ability to relax muscle tone, [9,19] and it appears to contribute to the stooped, kyphotic posture typical of patients with Parkinson's disease. [13] Physical therapists, thereofre, should teach these patients relaxation techniques that they can use to temporarily reduce rigidity in order to maintain ROM, lessen abnormal posturing, and produce long-term musculoskeletal change. It is unlikely that the physical therapy intervention can permanently reverse the rigidity, which is a direct result of the nervous system pathology. Loss of trunk mobility appears to occur early in the disease; thus, self-relaxation and ROM techniques should focus on restoration or maintenance of trunk mobility with special emphasis on rotational motions at each spinal region from the neck to the pelvis.

In summary, the focus of physical therapy intervention should be on the composite effects of disease and on the patient's resulting disability (eg, difficulty with transfers and gait, faulty balance). The model we have developed is intended to provide the clinician with guidenace regarding what impairments to evaluate and regarding interpretation of the relative contributions of direct and indirect effects of the disease in producing the patient's disability. Armed with a good working hypothesis regarding the causes of observed impairments, the clinician is in a good position to make sound clinical judgments regarding those impairments that can be corrected by physical therapy intervention and those that require pharmacologic management.

The model we have developed should demonstrate the need for integrated pharmacologic and physical therapy management of the patient with Parkinson's disease. Neither type of intervention alone can address all of the impairments of patients with Parkinson's disease. Neither type can prevent the inevitable course of the disease. Physical therapy management in conjunction with pharmacologic management, however, can provide the greatest possibility of maintaining functional ability as long as possible.

Implications for Research

Numerous hypotheses are inherent in the model that is outlined in this article. These hypotheses can be tested to support the model or to identify aspects of the model that are incorrect. For example, the model suggests that musculoskeletal limitations, an indirect effect of Parkinson's disease, contribute to bradykinesia, which has traditionally been considered a direct effect of the disease. One hypothesis that could be tested is that physical therapy intervention to increase trunk mobility (cervical through lumbopelvo-femoral) will decrease the time for the patient to carry out functional activities. This hypothesis tests the proposed relationship between musculoskeletal limitations and badykinesia. The hypothesis can be tested by an experiment in which measurements of trunk mobility and timing of functional activity are obtained during nonintervention and intervention periods (eg, baseline 1, treatment 1, treatment withdrawal, treatment 2).

The model also suggests a relationship between musculoskeletal limitations and automatic motor behavior such as posture and balance control. Posture and balance control can be measured using motion analysis techniques. Studies can be used to test the hypothesis that improvement in muscle length and flexibility will result in improvements in posture and balance control. A corollary to this hypothesis is that posture and balance are correlated; as posture worsens, so does balance. This hypothesis can be tested by measuring posture and balance of the same individual over time as the disease progresses. Alternatively, posture and balance can be measured for a large number of individuals with Parkinson's disease, and the strength of the correlation can be determined.

The model suggests that both direct and indirect effects of Parkinson's disease contribute to some of the typical impairments associated with the disease. A hypothesis that can test this interaction is that physical therapy intervention combined with pharmacologic therapy will be more effective in correcting impairments of balance than either intervention alone. This hypothesis can be tested using a two-group crossover experimental design. The first group receives the physical therapy intervention alone followed by the addition of an appropriate pharmacologic agent. The second group receives the pharmacologic intervention alone followed by the addition of the physical therapy intervention. This hypothesis predicts that the greatest capability for balance control will be achieved when both the drug and physical therapy interventions are combined. Subjects for this study would either receive drugs and physical therapy for the first time, or they would require an increase in drug dosage.

The hypotheses outlined above are only a few of the many that can be identified from the model and tested. Other hypotheses can be identified to test whether physical therapy intervention alters impairments that are a direct effect of the disease. Hypotheses can also be tested regarding the efficacy of physical therapy intervention early in the disease as compared with later, after musculoskeletal alterations become fixed. In summary, the model serves as a framework from which many important questions can be asked.

Summary

A model has been presented that differentiates between impairments that occur as direct effects of Parkinson's disease and those that occur as indirect effects of the disease. We have demonstrated that many of the impairments of the patient with Parkinson's disease may have contributions of both a CNS and a non-CNS nature. We suggest, therefore, that disabilities of the individual with Parkinson's disease occur both as a direct result of the degenerative process and as a result of sequelae indirectly causing impairment of other systems such as the musculoskeletal and cardiopulmonary systems. The model provides a system for identifying important factors to evaluate in the patient with Parkinson's disease and for interpreting evaluation findings with respect to the causes and effects of specific impairments. This model specifically suggests a direction for treatment of the patient with Parkinson's disease. The model is formulated in a manner that leads to development of specific hypotheses and to experimental tests of the hypotheses. Thus, the model has immediate clinical applicability and also can be used to guide research.

Acknowledgments

We acknowledge Victoria Rugo de Cartarya, PT, and Alan M Jette, PhD, PT, for useful discussions as we developed this manuscript. We also thank Steven J Rose, PhD, PT, FAPTA; Steven L Wolf, PhD, PT, FAPTA; and Rebecca Porter, MS, PT, for their assistance with an earlier draft of the manuscript.

(*1) Merck Sharp & Dohme, Div of Merck & Co, Inc, West Point, PA 19486.

References

[1] Ilson J, Bressman S, Fahn S: Current concepts in Parkinson's disease. Hosp Med 19:33-58, 1983

[2] Quinn NP, Husain FA: Parkinson's disease. Br Med J [Clin Res] 293:379-382, 1986

[3] Bannister RJ, Oppenheimer DR: Degenerative disorders of the nervous system associated with autonomic failure. Brain 95:457-474, 1972

[4] Schenkman M, Butler RB: A model for multisystem evaluation, interpretation, and treatment of individuals with neurologic dysfunction. Phys Ther 69:538-547, 1989

[5] Wood PHN: The language of disablement: A glossary relating to disease and its consequences. Int Rehabit Med 2:86-92, 1980

[6] Jette A: Functional disability and rehabilitation of the aged. Topics in Geriatric Rehabilitation 1(3):1-17, 1986

[7] Mayeux R, Stern Y, Rosenstein R, et al: An estimate of the prevalence of dementia in idiopathic Parkinson's disease. Arch Neurol 45:260-262, 1988

[8] McGeer PL, McGeer EG, Itagoak S, et al: Anatomy and pathology of the basal ganglia. Can J Neurol Sci 14(Suppl):363-371, 1987

[9] Hallett M: Physiology and pathophysiology of voluntary movement. In Tyler HR, Dawson DM (eds): Current Neurology. Boston, MA, Houghton Mifflin Co, 1979, vol 2, pp 351-376

[10] Lieberman A: Dopamine agonists: New Perspectives. Trends in Clinical Neurology 4(3):1-19, 1988

[11] Fahn S: The varied clinical expressions of dypstonia. In Jankovic J (ed): Neurological Clinics. Philadelphia, PA, W B Saunders Co, 1984

[12] Webster D: Critical analysis of the disability in Parkinson's disease. Modern Treatment 5:257-282, 1986

[13] Alba A, Trainor FS, Ritter W, et al: A clinical disability rating for Parkinson's patients. J Chron Dis 21:507-522, 1968

[14] McElvany NG, Wilcox PG, Chung A, et al: Pleuropulmonary disease during bromocriptine treatment of Parkingson's disease. Arch Intern Med 148:231-236, 1988

[15] Rothwell JC: Control of Voluntary Movement. Rockville, MD, Aspen Publishers Inc, 1987

[16] Baloh RW, Honrubia V: Clinical Neurophysiology of the Vestibular System. Philadelphia, PA, F A Davis Co, 1979

[17] Marsden CD: Then mysterious motor function of the basal ganglia: The Robert Wartenberg Lecture. Neurology 32:514-539, 1982

[18] Brooks VB: The Neural Basis of Motor Control. New York, NY, Oxford University Press Inc, 1986

[19] Cote L, Crutcher MD: Motor functions of the basal ganglia and disease of neurotransmitter metabolism. In Kandall ER, Schwartz JH (eds): Principles of Neural Science, ed 2. New York, NY, Elsevier Science Publishing Co Inc, 1985, pp 523-534

[20] Dietz V, Berger W, Horstmann GA: Posture in Parkinson's disease: Impairment of reflexes and programming. Ann Neurol 24:660-669, 1988

[21] Mesulam MM: Principles of Behavior Neurology. Philadelphia, PA, F A Davis Co, 1985

[22] Brown RG, Marsden CD: Internal versus external cues in the control of attention in Parkinson's disease. Brain 111:323-345, 1988

[23] Barbeau A, Roy M: Familial subsets in idiopathic Parkinson's disease. Can J Neurol Sci 11:144-150, 1984

[24] Nutt JG, Woodward WR, Hammersted JP, et al: The "on-off' phenomenon in Parkinson's disease: Relations to levodopa absorption and transport. N Engl J Med 310:483-488, 1984

[25] Agid Y, Javoy F, Glowinsky J: Hyperactivity of the remaining neurons after partial destruction of the nigrostriatal dopaminergic system in the rat. Nature 245:150-151, 1973

[26] Weinrich M, Koch K, Garcia F, et al: Axial versus distal motor impairment in Parkinson's disease. Neurology 38:540-545, 1988

[27] Hoehn M, Yahr M: Parkinsonism: Onset, progression, and mortality. Neurology 17:427-442, 1967

[28] Staeheli JW, Frassica FJ, Sim FH: Prosthetic replacement of the femoral head for fracture of the femoral neck in patients who have Parkinson's disease. J Bone Joint Surg [Am] 70:565-568, 1988

[29] Vinken WG, Gauthier SG, Dollfuss RE, et al: Involvement of upper-airway muscles in extrapyramidal disorders: A cause of airflow limitation. N Engl J Med 311-438-442, 1984

[30] Murray MP, Speci SB, Gardner GM, et al: Walking patterns of men with parkinsonism. Am J Phys Med 57:248-294, 1978

[31] Inman VT, Ralston HJ, Todd F: Human Waling. Baltimore, MD, Williams & Wilkins, 1981

[32] Burton K, Calne DB: Pharmacology of Parkinson's disease. Neurol Clin 2:461-472, 1984

[33] Feddman R, Lannon M: Parkinson's disease and related disturbances. In Feldman R (ed): Neurology: The Physician's Guide. New York, NY, Thieme Medical Publishers Inc, 1984, pp 147-162

[34] Melamed E: Initiation of levodopa therapy in parkinsonism patients should be delayed until the advanced stages of the disease. Arch Neurol 43:402-405, 1986

[35] Gauthier S, Gauthier L: Current status of levodopa therapy in idiopathic Parkinsonhs disease. Can J Neurol Sci 14(Suppl 3):452-454, 1987

M Schenkman, PhD, PT, is Assistant Professor, Graduate Program in Physical Therapy, MGH Institute of Health Professions, 15 river St, Boston, MA 02108-3402 (USA).

RB Butler, MD, is Chief of Medicine, Emerson Hospital, Old Road to Nine Acre Corner, Concord, MA 01742, and Assistant Professor, Department of Neurology, Boston University, Boston, MA.