The Relationship of Lower-Limb Muscle Force to Walking Ability in Patients With Amyotrophic Lateral Sclerosis.

Key Words: Gait; Muscle performance, lower extremity; Neuromuscular disorders, general.

The disablement models[1,2] suggest a relationship between impairments such as diminished muscle force and functional limitations such as difficulty walking. These relationships have been explored by researchers studying whether interventions directed at ameliorating impairments produce improvements in function or prevent functional limitation.[3-6] We believe that understanding the nature of such relationships is critical because many, if not most, treatments provided by physical therapists are based on the assumption that altering impairment leads to improved function.

Amyotrophic lateral sclerosis (ALS) is a motor disease that affects motoneurons in the spinal cord and the brain stem, resulting in progressive muscle weakness and loss of function. Although each patient's rate of disease progression is remarkably linear, muscle force is lost at varying rates, dependent on the type of onset (upper limb, lower limb, or bulbar), the muscle group under consideration, the sex of the patient, and the duration of disease.[7] The most common temporal pattern is initial limb involvement with subsequent bulbar symptoms.[7] Initially, distal muscles are more severely affected than proximal muscles.[7] Flexor muscles demonstrate greater weakness than extensor muscles throughout the course of disease.[7] Loss of muscle force in individuals with ALS has been shown to be related to loss of function.[8]

Many researchers[3-6,8-20] have explored the relationship between muscle force and walking ability among individuals with myriad characteristics and diseases. Changes in force production have not been shown to cause a specific change in function, but studies provide support for some type of relationship between force and function.[3,6,9,12,14-16,18,20] Several problems of interpretation arise, however, from the types of data and analyses used in these studies. We believe one problem results from the fact that impairments in force production have often been measured using units of force or work.[4,6,10-12,15,18] The force or work required for walking must vary as a function of an individual's body size. For example, the lower-extremity force that is adequate for walking in a petite 80-year-old woman may not be adequate to allow a large young man to walk. Additionally, the force generated by a relatively small, yet strong, upper-extremity muscle may be interpreted as weakness if measured in a large lower-extremity muscle. The studies that have addressed the problems associated with using units of force have normalized muscle force data by accounting for a subject's weight[4,16] or body mass index (BMI).[9] There is evidence, however, that sex differences remain after accounting for BMI.[9,20] Furthermore, muscle mass is lower in proportion to height in individuals over 60 years of age than it is in younger individuals,[21] and this difference will affect the use of BMI to normalize muscle force. Perry et al[18] described force as a percentage of normal, citing unpublished data. Absolute force values, however, appear to have been used in their analyses. Damiano and Abel[5] calculated force as a percentage of normal and used these calculations in their analyses. Their subjects, however, were children, and their sample consisted of only 16 subjects.

In most studies, walking ability has been quantified in units of energy expenditure,[5,16] speed,[3,6,9,11-14,17,18,20] or other gait variables.[5,6,17,18] Although certain community functions such as crossing the street during a standard light cycle may be affected by gait speed, the ability to walk in other than an urban, outdoor setting may not be affected by speed. For example, Perry et al[17] have shown that although gait speed increased for each higher functional level of walking in individuals with hemiplegia, those who were able to walk independently in the community had a much lower gait speed than did individuals without hemiplegia. In addition, although gait speed distinguished among categories of community walkers who needed assistance or not, the walking speeds of those individuals who were able to walk in the community with some assistance were not different from the walking speeds of those who were considered unlimited in household walking. Moreover, although gait variables such as stride length and speed affect gait efficiency, it seems possible that individuals with inefficient gait may be able to function in the community by reducing their energy demand or accepting short-term high energy demands not normally encountered by individuals without functional limitations. Using discriminant analysis, Perry et al[17] were not able to show that stride characteristics allowed differentiation of functional walking categories in patients with hemiplegia.

Because it may be difficult to interpret the clinical importance of, for example, a change of 0.016 m/s in gait speed that is predicted to result from a 1-kg change in force,[15] use of categories of walking ability may be more conceptually appealing than use of continuous scales of measurement. In some studies examining the relationship of force to walking, walking ability has been categorized by degree of independence or assistance required in the community or at home.[10,17] In one study,[10] however, the subjects were hospitalized patients, so the classifications were for very-low-level walking abilities, not including community walking. In 2 other studies in which walking ability was classified according to degree of independence, measurements of muscle force were not made.[17,22]

In recent studies using continuous measures of force and function, the authors have described the relationship as linear.[9,12,18] There is growing evidence, however, that the relationship is curvilinear.[13-15] Those studies in which a curvilinear relationship[13,14] has been described, as well as some studies in which categories of function have been examined,[20] support the concept of a threshold for muscle force for functional activities, above which notable improvements in force do not result in improvements in physical function. Identification of force thresholds for various functional activities has important clinical implications. Knowledge of the level of force below which a person might lose, or above which a person might gain, an important physical function could help focus the type and timing of interventions aimed at prevention, improvement, or palliation.

The purpose of our study was to examine the lower-extremity muscle force associated with 3 levels of walking ability in individuals with ALS. We wanted to ascertain the relationship of normalized muscle force (measured as a percentage of the predicted normal value) for dorsiflexion, knee extension and flexion, and hip extension and flexion to the ability to walk independently in the community, in the community with assistance, and in the home only.

Method

Subjects

Data from 118 patients were retrospectively analyzed for our study. Data were drawn from a data set containing information from 662 patients with ALS referred to the New England Medical Center Neuromuscular Research Unit from 1979 to 1995. Patients were referred to the laboratory to confirm the diagnosis of ALS, monitor clinical disease progression, or participate in clinical trials. All subjects signed informed consent forms. Subjects were scheduled for periodic visits and had a median of 5 visits (minimum = 1, maximum = 55). The mean age of the larger sample of subjects at their first visit was 58.38 years (SD = 12.21, range = 22.65-85.95); 58% were men and 42% were women.

Procedure

Muscle force measures. The force exerted during maximal voluntary isometric contractions for the muscles selected was measured (in kilograms) using an electronic strain gauge tensiometer.(*)[1] The apparatus consisted of an examination table surrounded by an aluminum orthopedic frame. Adjustable clamps with rings attached to the frame. Adjustable length straps were connected to the strain gauge, which, in turn, was attached to the rings clamped on the frame. The limb to be tested was connected to the other end of the strap. Force was transduced electronically by the amount of distortion within the strain gauge, then amplified and recorded. The protocol used in this study has been described as part of the Tufts Quantitative Neuromuscular Examination (TQNE).[23] Intrarater reliability (r = .97-.99) and interrater reliability (r = .92-.99) have been established for dorsiflexion, knee flexion, knee extension, hip flexion, and hip extension in patients with ALS using this protocol.[23]

Lower-extremity muscles were tested by 1 of 4 physical therapists trained in the protocol.[23-25] The protocols for testing muscle force were developed as part of the TQNE by a team that included 2 of the authors. Lower-extremity muscle forces were measured bilaterally. Test positions were ordered to allow for patient comfort and standardization. No warm-up was provided. The same test order of positions was used with each subject at each visit: dorsiflexion, knee flexion, knee extension, hip flexion, and hip extension. We used this test order in an effort to minimize fatigue by keeping the total test time at a minimum and avoiding having the subject change positions multiple times. For each movement, 2 maximal isometric contractions, held for 3 to 4 seconds, were performed 5 to 8 seconds apart. Verbal encouragement was provided, with the tester telling the subject to push as hard as possible against the strap secured to the limb segment. The test generating the higher force was used for data analysis.

To test dorsiflexion, each subject was positioned supine with a strap secured around the metatarsals while the tester stabilized the proximal calf. Knee flexion and extension were tested with the subject sitting with the hip and knee at 90 degrees and a strap placed proximal to the lateral malleolus. During knee flexion testing, the tester stabilized the subject over both shoulders. During knee extension testing, the tester stabilized the subject over the distal thigh. To test hip flexion and extension, the subject was positioned supine with the trunk supported on a wedge. To test hip flexion, the subject's thigh was supported on the table at about 20 degrees of flexion, and the strap was placed proximal to the knee; the knee was positioned at 90 degrees, with leg off the end of the table. The tester supported the opposite lower extremity with the hip and knee at 90 degrees. Hip extension was tested in a similar position, with the hip supported at 20 degrees of flexion with a strap proximal to the knee joint. The tester stabilized the subject over the anterior superior iliac spine.

The predicted normal maximal force (in kilograms) for each movement based on each subject's age, sex, height, and weight was derived using a regression equation based on data from a sample of nearly 500 men and women without known muscle disorders who were tested with the procedure we used.[25] For each subject, actual force data for each muscle group were then recorded as a percentage of predicted normal maximal force (%PMF).[24] An example of the calculation of %PMF is shown in Table 1. The use of %PMF, in our opinion, avoids the problems encountered in using a measure of force, and we believe it allows comparisons among muscle groups and among individuals of different sizes, ages, and sexes. Because the %PMF values for right and left lower extremities were highly correlated (r=.82-.93), the right lower-extremity values were used in all analyses. Additionally, an average of the %PMF for the 5 movements was calculated for each subject.

Table 1.

Sample Calculation of Percentage of Predicted Normal Maximal Isometric Force (%PMF)(a)

Predicted normal right knee extension for a 50-year-old
man who weighs 180 lb and is 72 in tall:

Regression equation: Predicted Normal Isometric Force =
  (Sex x 13.95) - (Age x 0.38) + (Weight x 0.14) +
  (Height x 0.08) + 24.60
  = (13.95) - (19) + (25.2) + (5.76) + 24.60
  = 50.51 kg

Subject's actual right knee extension force is 31 kg,
  %PMF = (31/50.51) x 100 = 61%

(a) 1 1b = 0.4536 kg, 1 in = 2.54 cm.

Walking ability classification. Hoffer et al[26] described functional ambulation categories for children with myelodysplasia, distinguishing between the ability to ambulate in the community and at home. We modified these criteria for our study. Each subject's walking ability was classified, through patient interview and observation, into 1 of 4 categories by 1 of the 4 physical therapists trained in the protocol. The categories were: unable to walk (the patient cannot ambulate even with assistance), walking only at home (the patient can ambulate in the home with or without assistance or assistive devices but cannot walk in the community), walking in the community with assistance (the patient can ambulate in the community but requires the assistance of a device or another person), or independent walking in the community (the patient can walk independently in the community without assistive devices). The validity of the scale has been supported by work by Perry et al.[17] The reliability, however, has not been tested. Because the functional status of patients with ALS declines at varying rates and the time between visits varied, subjects often remained in the same walking category for several visits.

Data Analysis

The data we selected for analysis were those from any 2 consecutive visits in which a subject had all muscle force data recorded and changed from one functional category to the next lower functional category. This selection of data resulted in 3 categories of subjects: (1) subjects who changed from being able to walk independently in the community to being able to walk in the community with assistance (n = 78), (2) subjects who changed from being able to walk in the community with assistance to being able to walk only at home (n = 63), and (3) subjects who changed from being able to walk only at home to being unable to walk (n = 25). Because subjects entered and left the unit at various stages of disease and function and because not all muscle groups were always tested, the subjects in each subset varied. All analyses were performed using SPSS, Version 7.5 for Windows.([dagger])

In order to determine whether the levels of lower-extremity muscle force are associated with levels of walking ability, descriptive statistics were calculated and box plots were constructed for the %PMF values for each muscle group and for the lower-extremity average at the 2 consecutive visits over which functional status declined by one category in each subset of subjects. Paired t tests were used to determine the differences in mean %PMF for each of the muscle groups and the lower-extremity average across the change in walking ability for each subset of subjects.

Results

Ages and sexes for each subset of subjects are shown in Table 2. For all subsets of subjects, the mean %PMF at each functional level was greatest at the hip and lowest at the ankle. In each subset of subjects, there was a decrease in %PMF in all muscle groups from the visit during which the walking function was higher to the visit during which the walking function was lower. Data are displayed in Table 3 and in Figures 1 through 3.

[Figures 1-3 ILLUSTRATION OMITTED]

Table 2.

Sample Demographics 

                                      Age (y)
                                      [bar]X    SD      Range

Independent community walking to      52.4      11.1    26.26-76.22
 assisted community walking (n=78)    52.6      11.1    26.35-76.38
Assisted community walking to         53.5      11.1    26.41-76.78
 walking only at home (n=63)          53.7      11.0    26.50-76.93
Walking only at home to               56.4       8.0    38.07-71.19
 unable to walk (n=25)                56.6       8.0    38.19-71.52

Full sample (N=662)                   58.38     12.21   22.65-85.95

                                     Sex (%)
                                     Male      Female

Independent community walking to     61        39
 assisted community walking (n=78)
Assisted community walking to        48        52
 walking only at home (n=63)
Walking only at home to              40        60
 unable to walk (n=25)
Full sample (N=662)                  58        42

Change From Walking Independently in the Community to Walking in the Community With Assistance

The mean lower-extremity average %PMF for the 5 movements was 54.01% (SD = 12.76%, range = 18.77-84.23) during the last visit when subjects were able to walk independently in the community. Measurements of %PMF ranged from 66.45% (SD = 21.59%, range = 30.20-136.21) for hip extension to 40.18% (SD = 22.30%, range = 0.00-92.90) tar dorsiflexion. During the next visit when subjects required assistance in the community, an average of 7 weeks had elapsed. The mean lower-extremity average %PMF was 50.19% (SD = 14.38%, range = 16.05-89.22), ranging from 62.08% (SD = 20.35%, range = 20.37-128.48) for hip extension to 36.76% (SD = 22.53%, range = 0.00-86.90) for dorsiflexion. The mean decline in %PMF of all muscle groups ranged from 2.76% to 4.37%.

Change From Walking in the Community With Assistance to Walking Only at Home

The mean lower-extremity average %PMF for the 5 movements was 37.52% (SD = 15.17%, range = 7.31-79.55) during the last visit that subjects were able to walk in the community with assistance. The %PMF ranged from 48.25% (SD = 21.40%, range = 12.73-118.46) for hip extension to 25.54% (SD = 19.70%, range = 0.00-69.87) for dorsiflexion. An average of 8 weeks elapsed before the next visit when subjects were able to walk only at home. The mean lower-extremity average %PMF for the 5 movements was 32.18% (SD = 13.83%, range = 3.54-75.86); %PMF ranged from 42.47% (SD = 22.08%, range = 3.76-113.00) for hip extension to 21.36% (SD = 18.53%, range = 0.00-69.87) for dorsiflexion. The mean decline in %PMF for all movements ranged from 4.18% to 6.53%.

Change From Walking Only at Home to Unable to Walk

The mean lower-extremity average %PMF for the 5 movements was 19.12% (SD = 9.08%, range = 3.54-38.93) during the visit when subjects were last able to walk at home. The mean %PMF ranged from 32.59% (SD = 17.23%, range = 3.76-62.39) for hip extension to 8.30% (SD = 10.60%, range = 0.00-33.34) for dorsiflexion. The mean lower-extremity average %PMF for the 5 movements was 13.70% (SD = 7.36%, range = 0.00-29.60) during the next visit, 10 weeks later, when patients were unable to walk. The mean %PMF ranged from 21.95% (SD = 15.99%, range = 0.00-65.95) for hip extension to 5.34% (SD = 7.65%, range = 0.00-25.93) for dorsiflexion. The mean decline in %PMF for all movements ranged from 2.95% to 10.64%.

Discussion

The decline in walking ability observed in our subjects appears to have been precipitated by relatively small changes in muscle force, suggesting the possibility of force thresholds below which walking cannot be maintained. The data suggest that, on average, the transition from independent walking to walking in the community with assistance occurs when the average lower-extremity %PMF drops below approximately 54%. The transition from walking in the community with assistance to being able to walk only at home occurs when the average lower-extremity %PMF drops below approximately 37%. With an average lower-extremity %PMF of less than approximately 19%, individuals are no longer able to walk in the home. Consistent with the pattern and progression of ALS described in the literature,[7] we found greater losses of muscle force in the distal musculature as compared with the proximal musculature at every functional walking category.

Use of serial data from individuals with ALS, with their inexorable, linear decline in force over time,[27,28] provides a unique opportunity for locating specific muscle force thresholds and describing a profile of lower-extremity muscle force at critical points in time vis-a-vis transitions in function. In a previous study using data from this data set, we analyzed the data from all individuals on their third visit to the clinic.[19] This crosssectional analysis demonstrated that increasing levels of %PMF in all muscle groups improved the chances that individuals would be able to walk in the community as opposed to being limited to the home. The approach to analysis, however, did not allow determination of the level of %PMF at which transitions in function might occur.

Previous studies have shown the relationship between force and walking ability in people without deficits in force-generating capacity[3,9,11] as well as in individuals with a variety of diagnoses.[5,10,11,16-19] In studies in which walking ability was classified in some manner, the evidence suggests that force generation of the lower-extremity muscles is related to higher functional levels of walking. Damiano and Abel[5] found that children with cerebral palsy who were independently walking in the community had greater force-generating capacity than those who were limited in their community walking. Bassey et al[9] found that individuals who walked with a walker had lower force-generating capacity than those who did not use a walker. Perry et al[17] found that voluntary control of knee muscle contraction in the upright position was related to being able to walk in the community as opposed to being able to walk only at home in patients with hemiplegia. Ferrucci et al[15] reported that older women who were unable to walk or walked with an aid had less force in the knee extensors and hip flexors than women who were able to walk with no aids.

Most studies that have used continuous scales to measure walking ability have shown, at best, moderate correlations between force and walking ability, depending on the muscle group and variable analyzed.[3,9-15,18,20] Other studies[4-6] have demonstrated the force-walking relationship by showing improvements in walking ability with programs designed to improve force. In some studies,[3,10,18] investigators have used multiple regression analyses to attempt to determine the factors or muscle groups most predictive of walking ability. This is a compelling endeavor; however, as other studies[3,13,14] demonstrate, force measurements within a limb are highly correlated. Multicollinearity, or correlation among independent variables in a regression analysis, reduces the ability to interpret the results of the analysis.

Because no other studies that have identified muscle force levels required for function have measured muscle force in terms of %PMF, it is somewhat difficult to compare our results. Another problem affecting comparisons is that summary scores for lower-extremity force have been used.[9,14] Additionally, force thresholds for different functions related to walking have been described. For example, Sonn et al[20] have suggested a threshold of 70 N [multiplied by] m for the knee extensors of women for overall independence in instrumental activities of daily living and a threshold of 120 N [multiplied by] m for men. Bassey et aP suggested a threshold of 1.2 W/kg of body mass for leg extensor force during unassisted walking. Buchner et al[14] described a potential threshold of 275 N [multiplied by] m for total lower-extremity force (sum of knee flexion, knee extension, dorsiflexion, and plantar-flexion groups), above which improvements would not result in increases in gait speed. Ferrucci et al[15] found that hip flexor force predicted walking speed only when it was below 15 kg. One problem noted in our analyses as well as those of other authors[15] is that despite the fact that muscle force differs among walking categories, force measurements show wide variability within categories of ability and overlap across categories (Figs. 1-3). This finding suggests the difficulty of using measures of central tendency to define thresholds.

Examination of the relationship between force and walking ability in individuals with ALS has some inherent limitations. Because ALS is a disease of both the central nervous system and motoneurons and involves bulbar functions as well, the ability to walk is potentially affected by factors such as reflex integrity, balance,[10] depression,[11] poor motor control,[17,22] and endurance.[3,16]

Tang et al[22] have shown that patterns of motor control in the lower limbs of individuals with incomplete spinal cord injury can help predict characteristics of ambulation such as use of assistive devices. They suggested that timing of muscle contractions and the ability to isolate muscle contractions may affect the ability to ambulate. Perry et al[17] demonstrated that, in patients with hemiplegia, the ability to isolate voluntary contractions of the muscle groups of the involved knee was associated with being able to walk in the community. Although the effect of central nervous system symptoms on muscle force and walking ability in people with ALS may be a factor, evidence suggests that motoneuron dysfunction is primarily responsible for the weakness.[29] Further more, even in people with increased reflex activity, greater force is associated with greater gait speed and cadence and gross motor function.[5] The relationship between force and walking in people with ALS may also depend on when the relationship is examined, as well as the way in which force is measured. For example, one study of the relationship of knee flexor torque at 180 [degrees]/s to walking speed in patients with ALS showed high correlations early in the disease process and moderate correlations later.[11]

An important limitation of this study is the correlational design. This design does not allow any conclusion regarding the effect of loss of force-generating capacity on decline in walking ability. Additionally, the analyses did not control for the myriad additional factors that might have an impact on walking ability in patients with ALS. The design also included many statistical analyses, leading to an increased possibility of erroneously demonstrating differences in muscle force across categories of walking ability. The reader may refer to Table 3 to evaluate the clinical importance of each of the reported differences in muscle force across categories of walking ability. Another limitation of our study was the lack of testing of the reliability of the classification of walking ability. Classification was accomplished, however, by 4 individuals who worked over a period of 5 to 9 years with patients with ALS in the study setting. An additional potential limitation was the standardization of testing order for muscle groups. Although the order was held constant over time, we cannot discount a possible effect of fatigue for the last muscle group. This fact may limit our ability to draw conclusions about the effect of weakness in specific muscles. The order of testing, however, was the same as that used to derive the norms for muscle force[25] and the same as that found in the literature on force testing in patients with ALS.[23,24,27]

Table 3.

Percentage of Predicted Muscle Force at Transitions in Walking Ability

                                  Dorsiflexion     Knee Flexion

Independent community walking
 [bar]X                           40.18            52.42
 SD                               22.30            18.39
 Range                             0.00-92.90      18.10-105.20
Assisted community walking
 [bar]X                           36.76            48.95
 SD                               22.53            19.39
 Range                             0.00-86.90      13.47-102.02
Difference
 [bar]X                            3.42             3.48
 CI(a)                             0.58-6.26        2.65-6.09
Assisted community walking
 [bar]X                           25.54            33.65
 SD                               19.70            19.39
 Range                             0.00-78.51       1.77-87.97
Walking only at home
 [bar]X                           21.36            27.12
 SD                               18.53            16.66
 Range                             0.00-69.87       1.18-79.18
Difference
 [bar]X                            4.18             6.53
 CI                                1.38-6.99        4.15-8.91
Walking only at home
 [bar]X                            8.30            11.31
 SD                               10.60             8.48
 Range                             0.00-33.34       0.00-27.81
Unable to walk
 [bar]X                            5.34             7.51
 SD                                7.65             6.21
 Range                             0.00-25.93       0.00-20.95
Difference
 [bar]X                            2.95             3.80
 CI                                0.76-5.14        1.60-6.00

                                  Knee
                                  Extension      Hip Flexion

Independent community walking
 [bar]X                           52.37          63.88
 SD                               19.79          17.01
 Range                             9.83-113.18   18.02-116.96
Assisted community walking
 [bar]X                           48.31          61.12
 SD                               45.11          18.92
 Range                             7.50-108.80   19.44-110.53
Difference
 [bar]X                            4.07           2.76
 CI(a)                             2.19-5.94      0.09-5.61
Assisted community walking
 [bar]X                           38.69          46.84
 SD                               21.37          20.91
 Range                             2.04-86.23     6.08-108.41
Walking only at home
 [bar]X                           33.56          42.32
 SD                               19.26          19.37
 Range                             3.07-71.98     6.29-96.38
Difference
 [bar]X                            5.13          4.52
 CI                                3.02-7.24     1.78-7.27
Walking only at home
 [bar]X                           20.86          26.18
 SD                               11.18          17.39
 Range                             2.57-40.38     0.00-85.01
Unable to walk
 [bar]X                           15.64          20.56
 SD                                9.74          13.92
 Range                             0.00-34.99     0.00-61.66
Difference
 [bar]X                            5.22          5.62
 CI                                3.12-7.32     1.77-9.47

                                                    Limb
                                  Hip Extension     Average

Independent community walking
 [bar]X                           66.45             54.01
 SD                               21.59             12.76
 Range                            30.20-136.21      18.77-84.23
Assisted community walking
 [bar]X                           62.08             50.19
 SD                               20.35             14.38
 Range                            20.37-128.48      16.05-89.22
Difference
 [bar]X                            4.37              3.82
 CI(a)                             1.99-6.75         2.45-5.19
Assisted community walking
 [bar]X                           48.25             37.52
 SD                               21.40             15.17
 Range                            12.73-118.46       7.31-79.55
Walking only at home
 [bar]X                           42.47             32.18
 SD                               22.08             13.83
 Range                             3.76-113.00       3.54-75.86
Difference
 [bar]X                            5.77              5.33
 CI                                2.79-8.75         3.61-7.06
Walking only at home
 [bar]X                           32.59             19.12
 SD                               17.23              9.08
 Range                             3.76-62.39        3.54-38.93
Unable to walk
 [bar]X                           21.95             13.70
 SD                               15.99              7.36
 Range                             0.00-65.95        0.00-29.60
Difference
 [bar]X                           10.64              5.42
 CI                                4.48-16.81        2.97-7.86

(a) CI = confidence interval.

Another potential limitation in interpreting the results of our study is that the categories of walking ability, by their nature, included more than the simple function of walking on a level surface. Walking in the community requires negotiating curbs, obstacles, doors, and possibly stairs. Walking at home requires negotiating obstacles such as furniture, thresholds, and carpets and may depend on the distances between essential areas of the home such as bed and bath. Therefore, the categorization of patients used in this study may actually measure both functional limitations and disability. Based on disablement models,[2] we might expect less of an association between muscle force and disability than between muscle force and simple walking. This conceptual problem in the classification scheme might explain the wide overlap of muscle force found across walking classifications.

Disablement models suggest that factors such as coping skills, motivation, and necessity, as well as functional capacity, play a role in determining function.[2] In addition, it is possible, given that several muscle groups in the lower-extremity function at more than one joint, that relative force in one muscle group may compensate tot weakness in another muscle group in performing an activity.[18] Severe neck or trunk weakness may also affect a person's ability to ambulate, despite adequate lower-extremity force. Future studies that examine force and function longitudinally and account for variables that affect both force and walking ability are needed to determine whether clear-cut force thresholds can be identified for meaningful physical abilities.

Based on the findings of previous studies and our own study, few definitive conclusions can be drawn with regard to muscle force thresholds for walking ability. In view of the possible intervening factors, it is possible that the difficulties in identifying thresholds cannot be easily overcome. We believe, however, that because our methods allowed us to define the relationship between force and walking in terms of conceptually understandable measures of function and easily derived and meaningful measurements of force, we have provided data that can be both useful in further conceptualizing research in this area and a practical starting point for clinicians interested in thinking about patients' prognoses and designing suitable interventions.

Conclusion

This study provides a profile of lower-extremity muscle force at critical transitions in walking ability in patients with ALS. For each of the lower-extremity movements tested, there was a decrease in %PMF from the subject's visit during which the walking function was at a higher level to the next visit during which the function was lower. The findings of our study provide support for a relationship between impairment and functional limitation and suggest possible thresholds for lower-extremity muscle force required for various functional levels of walking. Wide variations in muscle forces within and between categories of walking ability, however, suggest the difficulty of identifying force thresholds related to physical functions.

(*) Interface Inc, 7401 E Buttherus Dr, Scottsdale, AZ 85260.

([dagger]) SPSS Inc, 444 N Michigan Ave, Chicago, IL, 60611.

References

[1] Nagi SZ. Disability concepts revised: implications for prevention. In: Pope A, Tarlov A, eds. Disability in America: Toward a National Agenda for Prevention. Washington, DC: National Academy Press; 1991:193-203.

[2] Verbrugge LM, Jette AM. The disablement process. Soc Sci Med. 1994:38:1-14.

[3] Buchner DM, Cress ME, Esselman PC, et al. Factors associated with changes in gait speed ill older adults. J Gerontol A Biol Sci Med Sci. 1996;51:M297-M302.

[4] Chandler JM, Duncan PW, Kochersberger G, Studenski S. Is lower extremity strength gain associated with improvement in physical performance and disability in frail, community-dwelling elders? Arch Phys Meal Rehabil. 1998;79:24-30.

[5] Damiano DL, Abel MF. Functional outcomes of strength training in spastic cerebral palsy. Arch Phys Meal Rehabil. 1998;79:119-125.

[6] Lord SR, LLoyd DG, Nirui M, et al. The effect of exercise on gait patterns on (rider women: a randomized controlled trial. J Gerontol A Biol Sci Med Sci. 1996;51:M64-M70.

[7] Brooks BR. Natural history of ALS: symptoms, strength, pulmonary function, and disability. Neurology. 1996;47 (suppl 2):S71-S82.

[8] ALS CNTF Treatment Study Phase I-II Study Group. The ALS Functional Rating Scale: assessment of activities of daily living in patients with amyotrophic lateral sclerosis. Arch Neurol. 1996;53: 141-147.

[9] Bassey EJ, Fiatarone MA, O'Neill EF, et al. Leg extensor power and functional performance in very old men and women. Clin Sci (colch). 1992;82:321-327.

[10] Bohannon RW. Standing balance, lower extremity muscle strength, and walking performance of patients referred for physical therapy. Percept Mot Skills. 1995;80:379-385.

[11] Brooks BR, Sufit RL, Clough JA, et al. Isokinetic and functional evaluation of muscle strength over time in amyotrophic lateral sclerosis. In: Munsat TL, ed. Quantification of Neurologic Deficit. Boston, Mass: Butterworth; 1991:143-154.

[12] Brown M, Sinacore DR, Host HH. The relationship of strength to function in the older adult. J Gerontol A Biol Sci Med Sci. 1995;50: M55-M59.

[13] Buchner DM, deLateur BJ. The importance of skeletal muscle strength to physical function in older adults. Ann Behav Med. 1991;13: 95-98.

[14] Buchner DM, Larson EB, Wagner EH, et al. Evidence for a nonlinear relationship between leg strength and gait speed. Age Ageing. 1996;25:386-391.

[15] Ferrucci L, Guralnik JM, Buchner D, et al. Departures from linearity in the relationship between measures of muscular strength and physical performance of the lower extremities: the Women's Health and Aging Study. J Gerontol A Biol Sci Med Sci. 1997;52:M275-M285.

[16] Kramer JF, MacPhail HEA. Relationships among measures of walking efficiency, gross motor ability, and isokinetic strength in adolescents with cerebral palsy. Pediatric Physical Therapy. 1994;6:3-8.

[17] Perry J, Garrett M, Gronley JK, Mulroy SJ. Classification of walking handicap in the stroke population. Stroke. 1995;26:982-989.

[18] Perry J, Mulroy SJ, Renwick SE. The relationship of lower extremity strength and gait parameters in patients with post-polio syndrome. Arch Phys Med Rehabil. 1993;74:165-169.

[19] Slavin MD, Jette DU, Andres PL, Munsat TL. Lower extremity muscle force measures and functional ambulation in patients with amyotrophic lateral sclerosis. Arch Phys Med Rehabil. 1998;79:950-954

[20] Sonn U, Frandin K, Grimby G. Instrumental activities of daily living related to impairments and functional limitations in 70-year-olds and changes between 70 and 76 years of age. Scand J Rehabil Med. 1995;27: 119-128.

[21] Deurenberg P, van der Kooij KM, Evers P, Hulshof T. Assessment of body composition by bioelectrical impedance in a population aged greater than 60 years. Am J Clin Nutr. 1990;51:3-6.

[22] Tang SFT, Tuel SM, McKay WB, Dimitrijevic MR. Correlation of motor control in the supine position and assistive device used for ambulation in chronic incomplete spinal cord-injured persons. Am J Phys Med Rehabil. 1994;73:268-274.

[23] Andres PL, Hedlund W, Finison L, et al. Quantitative motor assessment in amyotrophic lateral sclerosis. Neurology. 1986;36:937-941.

[24] Andres PL, Skerry LM, Thornell B, et al. A comparison of three measures of disease progression in ALS. J Neurol Sci. 1996;139(suppl): 64-70.

[25] The National Isometric Muscle Strength Database Consortium. Muscular weakness assessment: use of normal isometric strength data. Arch Phys Med Rehabil. 1996;77:1251-1255.

[26] Hoffer MM, Feiwell E, Perry R, et al. Functional ambulation in patients with myelomeningocele. J Bone Joint Surg Am. 1973;55:137-148.

[27] Munsat TL,, Andres PL, Skerry LM. The use of quantitative techniques to define amyotrophic lateral sclerosis. In: Munsat TL, ed. Quantification of Neurologic Deficit. Boston, Mass: Butterworth; 1991: 129-142.

[28] Pradas J, Finison L, Andres PL, et al. The natural history of amyotrophic lateral sclerosis and the use of natural history controls in therapeutic trials. Neurology. 1993;43:751-755.

[29] Kent-Braun JA, Walker CH, Weiner MW, Miller RG. Functional significance of upper and lower motor neuron impairment in amyotrophic lateral sclerosis. Muscle Nerve. 1998;21:726-768.

DU Jette, DSc, PT, is Professor and Program Director, Graduate Program in Physical Therapy, Graduate School for Health Studies, Simmons College, Boston, Mass, and Research Program Manager, Department of Rehabilitation Services, Beth Israel Deaconess Medical Center, Boston, Mass. Address all correspondence to Dr Jette at the Graduate Program in Physical Therapy, Graduate School for Health Studies, 300 The Fenway, Simmons College, Boston, MA 02115 (USA) ([e-mail]).

MD Slavin, PhD, PT, is Associate Professor, Graduate Program in Physical Therapy, Graduate School for Health Studies, Simmons College.

PL Andres, PT, was Research Physical Therapist, Neuromuscular Research Unit, New England Medical Center, and Lecturer in Neurology, Tufts Medical School, Boston, Mass, at the time the data for this study were collected. Ms Andres is currently an independent consultant.

TL Munsat, MD, is Director, Neuromuscular Research Unit, Department of Neurology, New England Medical Center, and Professor of Neurology, Tufts Medical School.

Concept and research design were provided by Jette, Slavin, Andres, and Munsat; writing, by Jette and Slavin; data collection, by Andres; data analysis, by Jette; fund procurement, by Munsat; subjects, by Andres and Munsat; facilities and equipment, by Andres; institutional liaisons, by Munsat; and consultation (including review of manuscript prior to submission), by Slavin, Andres, and Munsat.

This study was approved by the Institutional Review Board of New England Medical Center.

This article was submitted October 14, 1998, and accepted March 22, 1999.