Archives of Physical Medicine and Rehabilitation
Volume 88, Issue 3 , Pages 361-367, March 2007

Recovery of Gait After Short-Stay Total Hip Arthroplasty

  • Inge van den Akker-Scheek, MSc

      Affiliations

    • Department of Orthopedics, University Medical Center Groningen, University of Groningen, The Netherlands
    • Corresponding Author InformationReprint requests to Inge van den Akker-Scheek, MSc, Dept of Orthopedics, University Medical Center Groningen, University of Groningen, PO Box 30001, 9700 RB Groningen, The Netherlands
    • ,
  • Martin Stevens, PhD

      Affiliations

    • Department of Orthopedics, University Medical Center Groningen, University of Groningen, The Netherlands
    • ,
  • Sjoerd K. Bulstra, PhD, MD

      Affiliations

    • Department of Orthopedics, University Medical Center Groningen, University of Groningen, The Netherlands
    • ,
  • Johan W. Groothoff, PhD

      Affiliations

    • Department of Health Sciences, University Medical Center Groningen, University of Groningen, The Netherlands
    • ,
  • Jim R. van Horn, PhD, MD

      Affiliations

    • Department of Orthopedics, University Medical Center Groningen, University of Groningen, The Netherlands
    • ,
  • Wiebren Zijlstra, PhD

      Affiliations

    • Center for Human Movement Sciences, University Medical Center Groningen, University of Groningen, The Netherlands.

Article Outline

Abstract 

van den Akker-Scheek I, Stevens M, Bulstra SK, Groothoff JW, van Horn JR, Zijlstra W. Recovery of gait after short-stay total hip arthroplasty.

Objective

To describe recovery of gait after total hip arthroplasty (THA) based on the assessment of spatiotemporal gait parameters determined with an ambulatory system.

Design

A 6-month inception cohort study.

Setting

Inpatient and outpatient setting in an academic hospital.

Participants

Sixty-three patients participating in a short-stay program for THA.

Intervention

Primary unilateral THA.

Main Outcome Measures

Walking speed, step length, step duration, and variability coefficient assessed at different walking speeds while performing an additional cognitive task and an endurance test. All measures were obtained preoperatively and 6 weeks and 6 months postoperatively.

Results

Patients improved significantly over time; however, extent and speed of recovery of gait parameters differed for each test part. The relation between walking speed and step length showed systematic improvement when analyzed over a range of speeds. At 6 months, the variability coefficient of the additional task test part was comparable with the preferred walking variability coefficient. The endurance test results could be predicted from the results of preferred walking.

Conclusions

Assessment of recovery of gait function requires more than only assessment of “normal” walking. Particularly, an analysis of walking at different speeds and walking while performing an additional cognitive task demonstrate different aspects of gait recovery after THA.

Key Words: Arthroplasty, replacement, hip, Gait, Rehabilitation

 

BESIDES SEVERAL WIDELY used self-report questionnaires to describe recovery after total hip arthroplasty (THA), objective measures can be obtained and are considered complementary.1, 2 In particular, measures of gait function are relevant to assess outcome. Because gait is highly important in everyday life, gait function is closely linked to overall functioning. For example, walking speed has been related to independent living and the ability to perform various activities of daily living, like safely crossing a traffic intersection and risk of falling.3, 4, 5 Hence, gait function is an important indicator of functional recovery. Gait function depends on the ability to maintain safe gait while navigating in a complex and changing environment, whereby gait has to be adjusted to different task demands.6 Thus, from a behavioral approach, important elements of gait function are the ability to walk at different speeds, combine gait with attention-demanding tasks, and walk longer distances.7 These aspects should be incorporated into a test battery assessing THA patients. Next to questionnaire research, predominantly limited gait analyses are performed (ie, only “normal” walking) on THA patients. With the development of ambulatory measuring devices that use body-fixed sensors, gait analysis is less time-consuming compared with the methods applied in laboratories by using camera systems, forceplates, treadmills, and electromyography.8, 9, 10, 11, 12, 13, 14, 15, 16, 17 It is therefore possible to measure different aspects of gait function within an acceptable amount of time and burden. Additionally, gait analysis can be performed outside a laboratory setting, which is a more representative environment of the patients’ everyday life.

The purpose of this study is to describe the spatiotemporal measures of gait recovery after THA, whereby the different aspects of gait function are measured with an ambulatory measuring device in an extensive test protocol, including walking at 3 speeds, walking with an additional cognitive task, and a walking endurance test. The hypothesis is that this test protocol will provide measures of gait recovery that cannot be detected when only preferred walking speed is assessed.

Back to Article Outline

Methods 

Participants 

From September 2002 to August 2004, patients were included in the study on admission to the Orthopedic Department of the University Medical Center Groningen for a unilateral primary THA while participating in the short-stay program. Because not all patients are capable of following a short-stay program, selection was done by using the following criteria: estimated surgery time less than 120 minutes, body mass lower than 110kg, expected hospital stay less than 6 days, absence of relevant cardiovascular comorbidity, no severe mobility disablement, no psychologic dysfunction, and no severe deformity of the spine. All patients were allowed to start walking with aids on the first postoperative day and were discharged on the fifth postoperative day, unless there were complications. No physical therapy was prescribed.

Within the short-stay program, 2 groups could be distinguished. One group received usual short-stay care, another received an additional support program called the Groningen Orthopedic Exit Strategy (GOES).18 The GOES was developed to respond to the insecurity and inactivity many patients show during the rehabilitation period, which as a consequence of the short-stay program is moved from the hospital setting to the home situation and aims to provide patients with extra support. The GOES treatment consists of showing patients a video with role models, providing extra information through newsletters and telephone consultation hours, and emphasizing partner participation. The hypothesis that this treatment would lead to more self-efficacy and social support and better pain coping, and eventually result in a better and quicker rehabilitation, was not supported, as measured by questionnaires.19 Because no effect on the psychosocial variables was found, it is assumed that no differences in gait recovery between these 2 patient groups will be present and the 2 groups can be evaluated as a whole.

Design and Data Collection 

All patients included in the study performed the gait analysis on admission to the hospital preoperatively and 6 weeks and 6 months postoperatively when visiting the outpatient clinic. The greatest incremental improvements in gait parameters have been shown to occur in the first 6 months after surgery.12, 13, 20 Additionally, a reference population consisting of relatives of the THA patients performed the gait analysis in the same way the patients did. The study was conducted in accordance with the guidelines stated in the Declaration of Helsinki and the regulations of the Medical Ethical Committee of the University Medical Center Groningen.

Measurements 

For gait analysis, we used the DynaPort system,a which consists of a data recorder (dimensions, 125×95×34mm; weight, 295g) that is attached to the lower back of the patient with a neoprene belt around the waist. The data recorder contains 3 uniaxial, piezoresistive accelerometers that measure acceleration in the frontal, sagittal, and transverse planes and a memory card on which data are stored. Three penlight batteries are attached to the belt. The position of the data recorder is adjustable so that the sensors can be positioned vertically.

Measurements were performed in a normal hospital corridor under supervision of the first author. Two cones were positioned 20m apart. Patients were instructed to walk 20m at their preferred speed, to walk as fast as possible (not running), and to walk at a slow speed. Additionally, patients were instructed to walk at their preferred speed while counting backward out loud from 50, each time subtracting 3 (dual task). Finally, patients were instructed to walk at their preferred speed for 6 minutes (walking endurance test). Because we wanted a test that is representative of an everyday life situation, we chose to instruct patients to walk at their preferred speed for 6 minutes rather than instruct them to cover as much ground as possible within 6 minutes, which is the instruction in the often-used 6-minute walk test (6MWT). Patients were allowed to quit if they experienced too much pain or became too fatigued. The amount of time patients could walk and the distance they covered were measured. During all test parts, patients were allowed to use crutches if they were used to walking with them. The total test duration was about 20 minutes.

After each measurement, the data were transferred from the memory card to a personal computer by using DynaScope software and displayed graphically. The beginning and end of each test part was selected and marked by hand. Analysis of accelerometer signals and extraction of gait parameters was performed by McRoberts BV.a Several movement features were obtained from the accelerometer signals. Of these features, walking speed, step length, step duration, and the standard deviation (SD) of the step durations were used in this study. The variability coefficient (variability coefficient = SD of step duration/mean step duration × 100%) was calculated to indicate individual variability in step duration.

The ambulatory method is used to determine spatiotemporal gait parameters by using lower-trunk accelerations. In previous studies, this method has been proven a valid means to determine gait parameters.17, 21, 22

Statistical Analysis 

Descriptive statistics (mean, SD) were used to describe the total study population and the results of the gait analysis. We examined walking speed, step length, step duration, and variability coefficient in an analysis of variance with time of measurement (pretest vs 6-wk follow-up vs 6-mo follow-up) as a within-subjects factor. If the sphericity assumption was not met, we applied the Huynh-Feldt correction. A significant main effect for measurement time was analyzed by single degree of freedom, repeated contrasts. Effect sizes were computed as partial η2 values.

The distance patients could walk at a preferred speed within 6 minutes was described and analyzed in a random coefficient analysis to determine if a significant improvement over time was seen. Again, the interaction term group by time was included as the covariate to test our assumption that the gait parameters of the patients of the GOES, and usual care groups were similar. Furthermore, to see whether patients showed fatigue and whether this diminishes over time, the mean values of the first and final three 20-m distances were calculated for all gait parameters at the 3 different measurement times. Consequently, patients who could not perform the walking endurance test at all or who could not walk at least six 20-m distances before quitting at 1 or more measurement times were excluded from this analysis.

We performed all statistical analyses by using the statistical package SPSS.b A P value lower than .05 was considered statistically significant.

Back to Article Outline

Results 

Demographic Characteristics 

A total of 80 THA patients participated in the short-stay program during the inclusion period of the study. Ten of them could not perform the second or third measurement or both and were therefore excluded: 2 patients had severe complications and were unable to walk, 3 patients had health problems unrelated to the arthroplasty at the time of the third measurement, 2 patients did not show up at the outpatient clinic for unknown reasons, 1 patient moved out of the country, and 2 patients underwent an arthroplasty of the contralateral hip before the final measurements took place. These 10 patients had a mean age of 62.2 years, and 70% were women. Body mass index was 29.2kg/m2. Six other patients had to be excluded because of technical problems and 1 patient because of the fact that not all test parts of 1 measurement were performed. The data of 63 patients were therefore available for analysis. One person from the reference group had to be excluded because of technical problems so the data of 19 persons were used in the analysis.

Table 1 presents baseline characteristics of the patients and the reference population. The patients’ mean age was 62.0 years, and most (68.3%) patients were women. There were no significant differences between the patients and the reference group on any variable. Mean hospital length of stay (LOS) was 7.0±3.3 days. The mean LOS was raised by 1 patient with an LOS of 26 days because of an infection and prolonged wound leakage. Not including this particular patient, the mean LOS was 6.7±2.3 days.

Table 1. Patient Characteristics at Baseline
CharacteristicsPatient GroupReference Group
N6319
Men20(31.7)8(42.1)
Women43(68.3)11(57.9)
Age (y)62.0±12.661.7±9.4
BMI (kg/m2)26.4±3.324.9±2.3
Height (m)1.72±0.101.71±0.11
Length of stay (d)7.0±3.3NA

NOTE. Values are n (%), mean ± SD, or as otherwise indicated.

Abbreviations: BMI, body mass index; NA, not applicable.

Gait Parameters 

Gait characteristics for the THA patients as well as for the reference population are displayed in table 2. All main effects for measurement time were significant with a P value lower or equal to .001, meaning that on all gait parameters and on all test parts, the patients improved significantly over time.

Table 2. Gait Parameters and Results of the General Linear Model Analysis
Patient Group (N=63)Preoperatively6 Weeks6 MonthsGLM RM (main time effect)GLM RC (preop to 6wk)GLM RC (6wk to 6mo)Reference Group (n=19)
Walking at slow speed
Speed (m/s)0.74±0.160.80±0.170.89±0.16F2,124=37.12F1,62=8.41F1,62=36.960.92±0.13
P<.001; partial η2=.37P=.005; partial η2=.12P<.001; partial η2=.37
Step length (m)0.51±0.080.55±0.080.58±0.09F1.83,113.52=38.03F1,62=25.66F1,62=15.520.62±0.05
P<.001; partial η2=.38P<.001; partial η2=.29P<.001; partial η2=.20
Step duration (s)0.71±0.100.71±0.090.65±0.07F1.88,116.42=16.17F1,62=0.11F1,62=28.710.68±0.07
P<.001; partial η2=.21P=.74; partial η2=.00P<.001; partial η2=.32
Variability coefficient (%)13.1±5.810.4±4.610.5±4.0F1.80,11.26=7.43F1,62=10.35F1,62=0.028.2±2.7
P=.001; partial η2=.11P=.002; partial η2=.14P=.90; partial η2=.00
Walking at preferred speed
Speed (m/s)0.93±0.190.95±0.171.14±0.18F1.80,111.48=56.16F1,62=0.99F1,62=105.741.32±0.15
P<.001; partial η2=.48P=.32; partial η2=.02P<.001; partial η2=.63
Step length (m)0.56±0.090.58±0.080.63±0.10F1.83,113.34=40.14F1,62=7.86F1,62=44.590.71±0.07
P<.001; partial η2=.39P=.007; partial η2=.11P<.001; partial η2=.42
Step duration (s)0.61±0.070.62±0.070.55±0.04F1.88,116.42=30.08F1,62=0.38F1,62=69.120.54±0.03
P<.001; partial η2=.33P=.54; partial η2=.01P<.001; partial η2=.53
Variability coefficient (%)12.0±5.610.5±3.69.6±3.1F1.85,114.48=7.55F1,62=6.27F1,62=2.698.1±3.2
P=.001; partial η2=.11P=.02; partial η2=.09P=.11; partial η2=.04
Walking at fast speed
Speed (m/s)1.14±0.261.13±0.211.37±0.25F1.77,109.97=57.11F1,62=0.01F1,62=141.431.73±0.17
P<.001; partial η2=.48P=.73; partial η2=.00P<.001; partial η2=.70
Step length (m)0.62±0.110.63±0.090.69±0.11F1.83,113.31=44.71F1,62=1.20F1,62=87.230.80±0.08
P<.001; partial η2=.42P=.28; partial η2=.02P<.001; partial η2=.59
Step duration (s)0.55±0.070.56±0.060.51±0.05F2,124=32.21F1,62=2.15F1,62=82.020.47±0.03
P<.001; partial η2=.34P=.15; partial η2=.03P<.001; partial η2=.57
Variability coefficient (%)10.7±4.810.9±4.09.0±3.3F2,124=6.72F1,62=0.21F1,62=14.8911.0±5.9
P=.002; partial η2=.10P=.65; partial η2=.00P<.001; partial η2=.19
Walking with additional task
Speed (m/s)0.79±0.210.85±0.210.99±0.21F1.86,115.10=40.62F1,62=6.65F1,62=52.371.09±0.16
P<.001; partial η2=.40P=.012; partial η2=.10P<.001; partial η2=.46
Step length (m)0.56±0.090.58±0.080.62±0.10F1.59,98.59=19.28F1,62=8.20F1,62=20.510.67±0.07
P<.001; partial η2=.24P=.006; partial η2=.12P<.001; partial η2=.25
Step duration (s)0.73±0.140.71±0.110.63±0.07F1.73,107.30=21.94F1,62=1.42F1,62=41.030.63±0.10
P<.001; partial η2= 0.26P=.24; partial η2=.02P<.001; partial η2=.40
Variability coefficient (%)14.0±8.711.7±5.910.0±4.0F1.75,108.37=7.85F1,62=4.40F1,62=4.937.4±3.8
P=.001; partial η2=.11P=.04; partial η2=.07P=.03; partial η2=.07

NOTE. Values are mean ± SD.

Abbreviations: GLM RC, general linear model repeated contrasts; GLM RM, general linear model repeated measures.

Between the preoperative and 6-week postoperative measurements, walking speed improved significantly only in the slow-walking and the dual-task test parts, whereas step length and the variability coefficient improved significantly in the slow-walking, preferred-walking, and dual-task test parts. Step duration did not improve significantly in any test part. Between the 6-week and 6-month postoperative measurements, significant improvement was seen in walking speed, step length, and step duration in all test parts. The variability coefficient improved significantly only in the fast walking and dual-task test parts. The relation between the parameters of step length and walking speed for the 3 measurement times is shown in figure 1.

  • View full-size image.
  • Fig 1. 

    Relation between speed and step length preoperatively and 6 weeks and 6 months postoperatively. Legend: a, walking with additional task test part; f, fast-walking test part; p, preferred-walking test part; s, slow-walking test part.

Between the different test parts preoperatively, most variability was seen in the dual-task test part (fig 2). At 6 weeks, all variability coefficients improved, except in the fast-walking test part. Still, the variability coefficient observed in the dual-task test part was the highest. At 6 months postoperatively, most variability was seen in the slow-walking test part.

  • View full-size image.
  • Fig 2. 

    The variability coefficient of the different test parts preoperatively and 6 weeks and 6 months postoperatively. Note: Bars represent SDs. Legend: closed squares represent the slow-, preferred-, and fast-walking speed test parts; open squares represent the walking with additional task test part.

Patients improved significantly in the distance they could walk within 6 minutes (F1.7,105.1=43.68, P<.001; partial η2=.41). Preoperatively, 294.0±103.8m were covered on average, 315.9±75.3m at 6 weeks, and 387.4±79.5m after 6 months. On average, the reference group walked 475.6±49.2m. Preoperatively as well as postoperatively, most patients were able to walk for 6 minutes. The mean values of the first and final three 20-m distances were calculated for all gait parameters at the 3 different measurement times (table 3). Six patients who could not walk at least six 20-m distances were excluded from this analysis. Patients improved significantly over time on the difference between the first and final mean values of the parameters walking speed, step length, and step duration (P<.05); at 6 months, the mean walking speed of the final three 20-m distances was even higher than the mean walking speed of the first 3. Differences between measurement times were very small though; this was expressed in low partial η2 values (.11, .06, .11, respectively). No significant improvement was seen on the difference between the mean first and final three 20-m distances of the variability coefficient.

Table 3. Gait Parameters of the First and Final Three 20m During the Walking Endurance Test
Patient Group (n=57)Preoperatively6 Weeks6 MonthsReference Group (n=19)
First 3Final 3First 3Final 3First 3Final 3First 3Final 3
Speed (m/s)1.00±0.210.97±0.211.00±0.200.99±0.211.20±0.201.21±0.211.55±0.231.53±0.19
Step length (m)0.60±0.090.60±0.090.61±0.090.62±0.100.66±0.100.67±0.100.78±0.090.78±0.08
Step duration (s)0.61±0.060.63±0.070.62±0.060.63±0.070.55±0.040.56±0.050.51±0.030.51±0.03
Variability coefficient (%)10.5±4.210.3±4.29.7±3.49.5±3.78.7±2.98.5±3.37.3±3.87.2±3.8

NOTE. Values are mean ± SD.

Of the 63 included patients, 35 received the GOES treatment. The GOES patients were comparable on all patient characteristics measured at baseline to the group of patients receiving usual short-stay care. No significant group by time effect was found on any of the gait parameters of the slow-, preferred-, and fast-walking or dual-task test parts; the patients of the GOES group did not improve significantly over time any differently than those of the usual short-stay group. The walking endurance test showed no difference between the 2 groups either. The 2 groups differed in the use of aids during walking at 6 weeks, but this did not result in a difference between groups on the gait parameters. It was therefore justified to perform the analyses with the overall group.

Back to Article Outline

Discussion 

The purpose of this study was to describe gait recovery after THA based on spatiotemporal gait parameters measured by an ambulatory measuring device while using a test protocol that includes walking at 3 speeds, walking with an additional cognitive task, and a walking endurance test. Our hypothesis was that this test protocol provides measures of gait recovery that cannot be detected when only walking at a preferred speed is assessed.

The results of the study showed that the test protocol provided added value over measurement of only preferred speed. Six weeks after surgery, the results of walking at a preferred speed indicate a very small increase in walking speed and step length and a small decrease in gait variability (variability coefficient). A larger increase in speed and step length is first seen after 6 months. In contrast, already after 6 weeks, the speed–step length relationship during walking at different speeds shows a small but consistent increase in speed and step length (see fig 1). In addition, the assessments of walking at a fast speed show a large speed and step length increase between 6 weeks and 6 months after surgery. In accordance with findings of others,23 our data of walking at different speeds show that gait variability is speed dependent before surgery (see fig 2). Six weeks after surgery, no decrease of gait variability with speed is seen; instead, gait variability increases slightly. However, after 6 months, an overall decrease of gait variability can be seen and again gait variability decreases with increasing speed. Some caution is needed when interpreting these data because most patients used walking aids 6 weeks after surgery, which might have influenced gait variables.24 All of these findings would not have been obtained if only gait at 1 speed had been analyzed, as is done in most of the previous research.1, 10, 12, 13, 14, 25, 26, 27 Kyriazis and Rigas28 did analyze gait parameters at 2 speeds; they compared preferred and fast speed of 25 patients preoperatively and 1 and 8 to 10 years after the THA with a conductive walkway system by using telemetry. They concluded that fast walking did not provide information that was not already shown in preferred gait. However, our results show that analyzing gait by assessing walking at 3 different speeds offers more insight in recovery of gait function after THA than assessments of preferred walking alone. As is also shown by our results on gait variability, an additional advantage of assessing gait at different speeds is that when other aspects of gait are studied (eg, muscle activities, angular displacements, trunk accelerations), the test protocol can take into account the speed dependency of these aspects.16, 29

Walking while performing an additional attention-demanding cognitive task is a means to measure whether patients are able to walk automatically.6 To our knowledge, this is the first study that uses this measure in THA patients. Analysis is based on the principle of dual-task interference, which is the worsening of performance of the main task (eg, walking) as a result of simultaneously performing an attention-demanding cognitive task (eg, counting backward). Automaticity of the main task is reflected by a low or absent dual-task interference effect. Measuring the interference effect over time gives an indication of the level of returning automaticity.6 The dual-task principle has been used in other orthopedic patient populations (eg, after limb-saving surgery of the lower limb).30 Nonautomaticity can have an impact on daily living; gait changes caused by performing an additional cognitive task while walking are associated with increased risk of falling among older adults.31

Our results indicate that the inclusion of a dual task is valuable as a means to determine (recovery of) gait automaticity. Particularly the variability coefficients showed improved gait during dual-task performance. Already after 6 weeks, gait variability is reduced. After 6 months, gait variability has further reduced and does no longer differ from the variability that can be expected based on the chosen gait speed (see fig 2). During dual-task conditions, gait speed, but not step length, is reduced in comparison to preferred walking. However, this reduction in gait speed is smaller after 6 months than preoperatively or after 6 weeks. Thus, all our results indicate a return of gait automaticity.

The results of the walking endurance test did not show the expected decrease in speed and step length and increase in gait variability when the first and final 20-m distances were compared. The results also did not show the expected improvement after surgery. Existing differences, although statistically significant, were very small and cannot be considered as clinically relevant. Whether or not a patient completed this test part, the results from the longer walking distance could already be predicted from the results of walking at preferred speed.

Results of the dual-task test part presented an unexpected interference that has not been reported in other similar studies: the association between walking speed and step length deviates from the association seen in the other test parts. The tendency to enunciate a number at each step resulted in a slow gait with relatively large step lengths, which can be because of the rhythmic character of additional task interfering with the walking rhythm.32 In gait research, there are different attention-demanding tasks that vary in complexity and suitability for different populations and age groups. Gait research on patients recovering from limb-saving surgery of the lower limb has used the auditory Stroop test.30 Further research is necessary to determine the applicability of this test in THA patients; some consider this task less appropriate for older adults who often have difficulties hearing.32 Moreover, it is advised that patients also perform the cognitive task as a single task (eg, while sitting down) so that performance on the task with and without walking can be compared too.31 In our study, we found that performing an additional cognitive task affects gait parameters. In theory, patients can also perform worse on the additional task and not on walking, despite the instructions to direct attention to the counting task while walking. This might be a sound “safety-first” strategy, but with this strategy changes in performance of the cognitive task remain unrecognized.

The walking endurance test appeared not to be a measure of endurance. Either the test has to be prolonged, which makes it difficult to incorporate into daily clinical practice, or the instruction should be given to walk as fast as possible, which is done in the protocol of the 6MWT. We considered this as not representative of independent functioning in daily life. However, comparison with previous studies performed by using the 6MWT is therefore impossible. We do have to keep in mind that the measurements were assessed in relatively healthy patients because only patients without other health problems could participate in the short-stay program. Additionally, 6 patients could not perform the test. These patients did complete the slow-, preferred-, and fast-walking test parts and performed worse on all parameters at all 3 measurement times than the included patients. A remarkable observation during the endurance test was that patients walked faster than during the walking at preferred-speed test part. This occurred despite the fact that participants were instructed to walk at their preferred speed in both test parts. The walking endurance test might be experienced as a competition. Thus, despite instructions, some participants may want to cover as many meters within the 6 minutes as possible. Another explanation is that walking the 20-m walkway several times back and forth is experienced differently than walking it once. It has been observed that gait speed on a long walkway (eg, 60m) is higher compared with short walkways.33

The administration of the measurements was easy and well accepted by all patients. None reported any discomfort caused by the portable measuring device so any possible influence thereof on gait can be neglected. The total duration of a measurement was approximately 20 minutes and can therefore be integrated into routine clinical practice.

Replacing the walking endurance test with a “walking on different ground surfaces” test part seems relevant and can be the object of future research.7, 16 Patients often complain about problems when walking on uneven pathways, especially after surgery. This can be the result of balance problems because it is known that balance during gait is altered by severe osteoarthritis, and some of these balance problems seem to remain after THA.34 Stability during walking should therefore be assessed while walking under circumstances that challenge the postural control system to obtain an indication of patients’ ability to respond to multiple, irregular perturbations.35 With the ambulant measuring device, tests can be performed outside the laboratory. Hence, gait can be studied under more complex, unpredictable, and real-life circumstances, and this may reveal essential information,7 as is also indicated by the results of our dual-task test part.

Back to Article Outline

Conclusions 

In this study, a test protocol consisting of walking at 3 different speeds while performing an additional cognitive task and an endurance test was used to obtain insight into recovery of gait function after total hip replacement. The hypothesis that the test protocol would provide measures of gait recovery that cannot be detected when only preferred walking speed is assessed was confirmed by a part of our results. First, the use of different speed instructions made it possible to analyze gait parameters, especially the link between walking speed and step length and gait variability over a range of speeds. Already after 6 weeks, the speed–step length relation during walking at different speeds shows a small but consistent increase in step length, whereas preferred walking speed gait parameters showed almost no improvement. Second, the additional cognitive task test part is a valuable addition as a measure of (return of) automaticity of walking. However, the walking endurance test revealed no additional information. By analyzing walking at a range of speeds and while performing an additional cognitive task, one can make statements about gait function of THA patients that cannot be made based on gait parameters of 1 walking speed.

Suppliers

Back to Article Outline

Acknowledgments 

We thank Sander Heikens and Rienk van der Slikke, from McRoberts BV, for their technical assistance.

Back to Article Outline

References 

  1. Berman AT, Quinn RH, Zarro VJ. Quantitative gait analysis in unilateral and bilateral total hip replacements. Arch Phys Med Rehabil. 1991;72:190–194
  2. Kennedy D, Stratford PW, Pagura SM, Walsh M, Woodhouse LJ. Comparison of gender and group differences in self-report and physical performance measures in total hip and knee arthroplasty candidates. J Arthroplasty. 2002;17:70–77
  3. Bohannon RW, Andrews AW, Thomas MW. Walking speed: reference values and correlates for older adults. J Orthop Sports Phys Ther. 1996;24:86–90
  4. Isobe Y, Okuno M, Otsuki T, Yamamoto K. Clinical study on arthroplasties for osteoarthritic hip by quantitative gait analysis (Comparison between total hip arthroplasty and bipolar endoprosthetic arthroplasty). Biomed Mater Eng. 1998;8:167–175
  5. Kerrigan DC, Lee LW, Collins JJ, Riley PO, Lipsitz LA. Reduced hip extension during walking: healthy elderly and fallers versus young adults. Arch Phys Med Rehabil. 2001;82:26–30
  6. Mulder T, Zijlstra W, Geurts A. Assessment of motor recovery and decline. Gait Posture. 2002;16:198–210
  7. Geurts A, Mulder T, Rijken R, Nienhuis B. From the analysis of movements to the analysis of skills—bridging the gap between laboratory and clinic. J Rehabil Sci. 1991;4:9–12
  8. Ajemian S, Thon D, Clare P, Kaul L, Zernicke RF, Loitz-Ramage B. Cane-assisted gait biomechanics and electromyography after total hip arthroplasty. Arch Phys Med Rehabil. 2004;85:1966–1971
  9. Heller MO, Bergmann G, Deuretzbacher G, et al. Musculo-skeletal loading conditions at the hip during walking and stair climbing. J Biomech. 2001;34:883–893
  10. Loizeau J, Allard P, Duhaime M, Landjerit B. Bilateral gait patterns in subjects fitted with a total hip prosthesis. Arch Phys Med Rehabil. 1995;76:552–557
  11. Long WT, Dorr LD, Healy B, Perry J. Functional recovery of noncemented total hip arthroplasty. Clin Orthop Relat Res. 1993;288:73–77
  12. Madsen MS, Ritter MA, Morris HH, et al. The effect of total hip arthroplasty surgical approach on gait. J Orthop Res. 2004;22:44–50
  13. Miki H, Sugano N, Hagio K, et al. Recovery of walking speed and symmetrical movement of the pelvis and lower extremity joints after unilateral THA. J Biomech. 2004;37:443–455
  14. Perron M, Malouin F, Moffet H, McFadyen BJ. Three-dimensional gait analysis in women with a total hip arthroplasty. Clin Biomech (Bristol, Avon). 2000;15:504–515
  15. Rosler J, Perka C. The effect of anatomical positional relationships on kinetic parameters after total hip replacement. Int Orthop. 2000;24:23–27
  16. Moe-Nilssen R. A new method for evaluating motor control in gait under real-life environmental conditions (Part 2: Gait analysis). Clin Biomech (Bristol, Avon). 1998;13:328–335
  17. Zijlstra W. Assessment of spatio-temporal parameters during unconstrained walking. Eur J Appl Physiol. 2004;92:39–44
  18. Stevens M, van den Akker-Scheek I, Spriensma A, Boss NA, Diercks RL, van Horn JR. The Groningen Orthopedic Exit Strategie (GOES): a home-based support program for total hip and knee arthroplasty patients after shortened hospital stay. Patient Educ Couns. 2004;54:95–99
  19. van den Akker-Scheek I, Zijlstra W, Groothoff JW, van Horn JR, Bulstra SK, Stevens M. Groningen orthopaedic exit strategy: validation of a support program after total hip or knee arthroplasty. Patient Educ Couns. 2007;65:171–179
  20. Wall JC, Ashburn A, Klenerman L. Gait analysis in the assessment of functional performance before and after total hip replacement. J Biomed Eng. 1981;3:121–127
  21. Zijlstra W, Hof AL. Assessment of spatio-temporal gait parameters from trunk accelerations during human walking. Gait Posture. 2003;18:1–10
  22. Aminian K, Rezakhanlou K, De Andres E, Fritsch C, Leyvraz PF, Robert P. Temporal feature estimation during walking using miniature accelerometers: an analysis of gait improvement after hip arthroplasty. Med Biol Eng Comput. 1999;37:686–691
  23. Bhatt T, Wening JD, Pai YC. Influence of gait speed on stability: recovery from anterior slips and compensatory stepping. Gait Posture. 2005;21:146–156
  24. Sonntag D, Uhlenbrock D, Bardeleben A, Kading M, Hesse S. Gait with and without forearm crutches in patients with total hip arthroplasty. Int J Rehabil Res. 2000;23:233–243
  25. Olsson E, Goldie I, Wykman A. Total hip replacement (A comparison between cemented (Charnley) and non-cemented (HP Garches) fixation by clinical assessment and objective gait analysis). Scand J Rehabil Med. 1986;18:107–116
  26. Mattsson E, Brostrom LA, Linnarsson D. Walking efficiency after cemented and noncemented total hip arthroplasty. Clin Orthop Relat Res. 1990;254:170–179
  27. Sicard-Rosenbaum L, Light KE, Behrman AL. Gait, lower extremity strength, and self-assessed mobility after hip arthroplasty. J Gerontol A Biol Sci Med Sci. 2002;57:M47–M51
  28. Kyriazis V, Rigas C. Temporal gait analysis of hip osteoarthritic patients operated with cementless hip replacement. Clin Biomech (Bristol, Avon). 2002;17:318–321
  29. den Otter AR, Geurts AC, Mulder T, Duysens J. Speed related changes in muscle activity from normal to very slow walking speeds. Gait Posture. 2004;19:270–278
  30. de Visser E, Veth RP, Schreuder HW, Duysens J, Mulder T. Reorganization of gait after limb-saving surgery of the lower limb. Am J Phys Med Rehabil. 2003;82:825–831
  31. Beauchet O, Dubost V, Gonthier R, Kressig RW. Dual-task-related gait changes in transitionally frail older adults: the type of the walking-associated cognitive task matters. Gerontology. 2005;51:48–52
  32. Beauchet O, Dubost V, Aminian K, Gonthier R, Kressig RW. Dual-task-related gait changes in the elderly: does the type of cognitive task matter?. J Mot Behav. 2005;37:259–264
  33. Oberg T, Karsznia A, Oberg K. Basic gait parameters: reference data for normal subjects, 10-79 years of age. J Rehabil Res Dev. 1993;30:210–223
  34. Majewski M, Bischoff-Ferrari HA, Gruneberg C, Dick W, Allum JH. Improvements in balance after total hip replacement. J Bone Joint Surg Br. 2005;87:1337–1343
  35. Menz HB, Lord SR, Fitzpatrick RC. Acceleration patterns of the head and pelvis when walking on level and irregular surfaces. Gait Posture. 2003;18:35–46
  • a McRoberts BV, Raamweg 43, 2596 HN, The Hague, The Netherlands.
  • b Version 12.0.1; SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.

 Supported by the University Medical Center Groningen, University of Groningen, The Netherlands.

 No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated.

PII: S0003-9993(06)01534-6

doi:10.1016/j.apmr.2006.11.026

Archives of Physical Medicine and Rehabilitation
Volume 88, Issue 3 , Pages 361-367, March 2007

Article Tools

* Image Usage & Resolution