Gait Analysis Parameters and Walking Activity Pre- and Postoperatively in Children With Cerebral Palsy : Pediatric Physical Therapy

Secondary Logo

Journal Logo

RESEARCH REPORTS

Gait Analysis Parameters and Walking Activity Pre- and Postoperatively in Children With Cerebral Palsy

Nicholson, Kristen PhD; Lennon, Nancy PT, MS; Church, Chris MPT; Miller, Freeman MD

Author Information
Pediatric Physical Therapy 30(3):p 203-207, July 2018. | DOI: 10.1097/PEP.0000000000000512
  • Free

INTRODUCTION

Cerebral palsy (CP) describes a group of disorders of the development of movement and posture that are attributed to a nonprogressive disturbance that occurs in the developing fetal or infant brain.1 Prevalence of CP worldwide is approximately 2.11 per 1000 live births,2 making CP the most common motor deficiency that causes disability in children.3 Although CP is a nonprogressive condition, progressive loss of physical function has been observed.4–6 To maintain or improve mobility, many youth with CP have single-event multilevel surgery (SEMLS), including muscle tendon lengthenings, tendon transfers, realignment osteotomies, and joint stabilization procedures.5 The goal of SEMLS is to correct deformities and reduce the strain of walking.7 Easing the burden of walking is thought to allow improved mobility function and quality of life.8

Surgery is a complex and resource-intensive event. Each surgery represents a significant investment for the patient, family, and health care system.7 Therefore, it is important for both surgeons and parents to understand the outcomes of SEMLS. In order to evaluate the effects of SEMLS, a variety of impairment level and capacity level gait measures have been used in clinical outcome studies.9 Changes in gait quality have been examined in a laboratory, revealing improvements at the impairment (kinematic and kinetic) and capacity levels (gait speed) following surgery.9,10 Changes in everyday quantity of walking activity are a measure of mobility performance and are meaningful as indicators of decline, improvement, or maintenance. Average daily strides have been positively associated with personal care, mobility, and recreation.11 While both technical and functional outcomes following SEMLS have been studied, understanding of the association between these outcomes is lacking or unknown.

The Gait Deviation Index (GDI) is a common technical parameter obtained through gait analysis. The GDI incorporates 3-dimensional rotation angles for the pelvis and the hip, sagittal plane angles for the knee and the ankle, and foot progression. Using these parameters, a score is assigned that represents the relative deviation from typical in the overall gait pattern. The developers of the GDI reported scores of 73.9 to 129.9, with a mean of 100 and a standard deviation of 10 for a group of children without known gait deviations.12 The GDI and the Gross Motor Function Measure (GMFM) are quantitative measures that describe the severity of gait abnormalities and altered motor ability in children with CP.13 The GDI and the GMFM have been found to be highly correlated, supporting the argument that the GDI is a useful measure for characterizing gait abnormalities in children with CP.14 In 2012, Maanum et al15 reported a low correlation between the GDI and the 6-minute walk test, suggesting that gait and functional endurance are different constructs. In 2015, Wilson et al16 also reported a low correlation between the GDI and the 6-minute walk test (0.47), but reported a moderate correlation between the GDI and the step count (0.58). Their research provides evidence that higher GDIs are associated with higher daily step totals and stated that the study results provide cautious support that postoperative improvements in joint kinematics may be reflected in increased walking activity. However, the correlation between changes in gait parameters, such as the GDI, and changes in walking activity postoperatively has not been studied. The purpose of this study was to explore the relationship between the GDI and walking activity before and after lower extremity surgery.

METHODS

This Institutional Review Board approved retrospective study examined preoperative and postoperative GDIs and habitual walking activity in youth with spastic CP who are ambulatory, Gross Motor Function Classification System (GMFCS) levels I to IV. Data from 93 youth with CP who had a preoperative and postoperative gait analysis with walking activity monitoring were initially reviewed. Data from 19 youth were excluded based on the StepWatch (SW) (Modus: Washington, District of Columbia) data inclusion criteria. Data from 74 youth with CP were included in the analysis. All youth were between the ages 4 and 19 years at their preoperative visit and classified as GMFCS levels I to IV (Table 1). The preoperative GDI was calculated using gait parameters collected during a clinical gait analysis 1 to 16 months prior to surgery. The postoperative GDI was calculated using gait parameters collected during a clinical gait analysis 10 to 26 months following surgery. All evaluations included kinematic assessment with an 8-camera motion analysis system (Motion Analysis, Santa Rosa, California), data reduction using OrthoTrack (Motion Analysis), and a complete physical examination. Kinematic assessments included a minimum of 15 left and 15 right gait cycles. All trials were completed barefoot. Youth who used an assistive device to walk (walker, crutches, hand-held assist, etc) used the same or a comparable device during the preoperative and postoperative gait analyses. A physical therapist used data collected during the kinematic assessment and the physical examination to assign the GMFCS level. Walking activity monitoring coincided with the preoperative and postoperative gait analysis visits.

TABLE 1 - Participant Characteristics
Characteristic Participant (n = 74)
Age at preoperative visit, mean (SD) 11 y 0 mo (4 y 0 mo)
Months before surgery of preoperative visit, mean (SD) 5 mo (4 mo)
Months after surgery of postoperative visit, mean (SD) 1 y 3 mo (4 mo)
Sex, n (%)
Male 31 (42)
Female 43 (58)
GMFCS, n (%)
Level I 7 (9)
Level II 64 (64)
Level III 24 (24)
Level IV 2 (3)
Time of postoperative visit, n (%)
≤13 mo 37 (50)
>13 mo 37 (50)
Abbreviations: GMFCS, Gross Motor Function Classification System; SD, standard deviation.

Walking activity was monitored using the SW. The SW is a validated wearable Food and Drug Administration class 2 device that has been used in research to document walking activity in children, youth, and adolescents with CP.6,17–19 The SW was calibrated using a standard protocol,20 sent home for 8 days, and returned by prepaid mail. The SW monitor measures only the steps of 1 leg. Thus, the outcome variables analyzed were stride counts and not step counts. Patient data collection episodes were included in the analysis if there were at least 3 days with 8 or more hours of wear from each collection period. When more than 3 days of data were available, days to include were chosen so the maximum number of days was included while keeping the ratio of weekdays to weekend days consistent from the preoperative visit to the postoperative visit.

Outcome variables analyzed included the right GDI, left GDI, average GDI, and mean total daily strides. Wilson et al16 chose to examine the GDI as the average of right and left GDIs because it includes parameters from multiple levels and from both limbs and may better relate to the total step count, which is influenced by both limbs. However, because we were examining the effects of SEMLS, we would expect that if a patient had unilateral surgery, the GDI on that side would improve more than the GDI on the nonoperated side, or the average GDI. Therefore, in addition to the average GDI, we examined the changes in the GDI that reflected the laterality of the SEMLS by reporting a Surgery GDI; the average GDI for bilateral procedures, the left GDI for left unilateral procedures, and the right GDI for right unilateral procedures.

Descriptive statistics were calculated for all variables. Variables were compared preoperatively and postoperatively using 2-tailed paired t tests. The association between the change in the GDI and the change in walking activity (represented by mean total daily strides) was examined with Pearson correlation coefficients. The strength of the correlation was assessed according to Evans21 (0.00-0.19: very weak, 0.20-0.39: weak, 0.40-0.59: moderate, 0.60-0.79: strong, and 0.80-1.0: very strong). In the presence of a moderate or strong correlation, the associated linear equation was used to examine the relationship between the GDI and walking activity.

Mobility, quantified with the Functional Mobility Scale scores, decrease in the first 6 months following SEMLS, improve to preoperative levels at 12 months, and further improve in the second year.5,22 To ensure that correlation between the GDI and the stride count was not affected by collecting stride counts at 1 year postoperative instead of 2 years postoperative, the change in stride counts for the participants who had 10- to 13-month follow-up (n = 37) was compared with the change in stride counts for the participants who had a follow-up greater than 13 months postoperatively (n = 37) with a 2-sample unequal variance t test.

RESULTS

The 74 participants had a total of 348 procedures, and the majority were soft-tissue procedures (Table 2). Procedures listed as “other” included foot surgery, knee capsulotomies, and patellar tendon plication and advancement. There were 10 patients with surgical procedures on the left side only, 15 patients who had surgical procedures on the right side only, and 49 patients who had bilateral surgery.

TABLE 2 - Participant Surgical Procedures
Surgical Procedure Number Performed (n = 348)
Hamstring lengthening 80
Gastrocnemius lengthening 68
Adductor lengthening 19
Achilles lengthening 22
Rectus transfer 30
Tibial osteotomy 20
Femoral osteotomy 26
Other 83
Total procedures 348
Average procedures per patient 4.7

The average GDI and the Surgery GDI significantly increased postoperatively (Table 3). The mean daily stride counts had no significant change postoperatively. To ensure that large stride standard deviations between GMFCS levels did not affect significance of stride change postoperatively, the preoperative and postoperative mean daily strides were further divided by GMFCS levels. The change in strides was not significantly different for any level. Preoperatively, the average GDI and mean daily stride counts were moderately correlated (r = 0.494). Postoperatively, the average GDI and mean daily stride counts were moderately correlated (r = 0.436).

TABLE 3 - Preoperative and Postoperative GDI and Mean Daily Strides
Preoperative Postoperative P Value
Average GDIa 69.5 (14.0) 74.4 (11.9) .001b
Surgery GDIc 68.1 (14.2) 75.4 (13.0) .001b
Mean daily strides 3296.1 (1923.0) 3366.3 (2133.8) .643
Mean daily strides GMFCS I 6460.4 (1356.4) 6199.1 (1997.7) .571
Mean daily strides GMFCS II 3603.5 (1510.3) 3765.9 (1816.7) .448
Mean daily strides GMFCS III 1522.2 (817.7) 1560.8 (812.4) .867
Mean daily strides GMFCS IV 961.7 (740.1) 393.6 (113.3) .382
Abbreviations: GDI, Gait Deviation Index; GMFCS, Gross Motor Function Classification System.
aAverage GDI: the average of the left and right GDI.
bP < .05.
cSurgery GDI: GDI reflecting the laterality of patient's surgery; average GDI for bilateral procedures, left GDI for left unilateral procedures, and right GDI for right unilateral procedures.

A significant weak correlation was present between the change in the average GDI and the change in mean daily strides (Table 4). A significant moderate correlation was present between the change in the Surgery GDI and the change in mean daily strides (r = 0.386, P = .001; Figure). When separated into GMFCS levels, moderate correlations were found between the change in the Surgery GDI and the change in mean daily strides, although only the GMFCS level II correlation was significant (Table 4). For participants who had a 10- to 13-month follow-up, there was a significant moderate correlation between the change in the Surgery GDI and the change in mean daily strides. Similarly, the data from participants who had a follow-up greater than 13 months had a significant moderate correlation. No significant difference was found between the 2 periods (P = .21).

F1
Fig.:
Change in Surgery GDI and change in mean daily strides from preoperative to postoperative levels. GDI indicates Gait Deviation Index.
TABLE 4 - Correlations Between GDI and Mean Stride Counts by Group: Laterality, GMFCS Level, Postsurgical Time
Pearson Correlation Coefficient P Value
Change in average GDI vs change in mean strides 0.282 .015a
Change in Surgery GDI vs change in mean strides
All 0.386 .001a
GMFCS I 0.430 .335
GMFCS II 0.374 .009a
GMFCS III 0.392 .108
Postoperatively ≤13 mo 0.386 .018a
Postoperatively >13 mo 0.391 .017a
Abbreviations: GDI, Gait Deviation Index; GMFCS, Gross Motor Function Classification System.
aP < .05.

Despite the moderate positive correlations between the Surgery GDI and stride counts, an increase in the GDI did not always correspond to an increase in walking activity. Seventy-four percent of the patients had an increase in the Surgery GDI following SEMLS. However, only 49% of the participants had increased walking activity following SEMLS. The associated linear equation (ΔStrides = 38.80 × ΔGDI − 208.53) supported that the Surgery GDI had to improve by more than 5 points before there was an increase in walking activity.

DISCUSSION

Prior to surgery, the mean GDI was moderately correlated with stride counts (0.49). These findings agree with those of Wilson et al,16 who found a correlation of 0.58 between the mean GDI and step counts. This moderate correlation, however, does not persist at surgical follow-up. Both the mean GDI and the GDI reflective of the laterality of the surgery increase significantly postoperatively. Stride counts however remain relatively unchanged at surgical follow-up.

The change in the mean GDI and the change in the mean daily stride count for the 74 participants were only weakly correlated. When only looking at the change in the GDI for the side(s) that received surgery, there was a moderate correlation between the change in the Surgery GDI and the change in stride counts. When separated into GMFCS levels, all GMFCS levels had moderate correlations, but GMFCS levels I and III had smaller samples sizes, possibly resulting in nonsignificant correlations. The majority of the participants were categorized as GMFCS level II. This group had a significant moderate correlation.

Postsurgery, the GDI reflective of the laterality of the surgery increased to the range that reflects typical movement patterns.12 Typical kinematics, though, do not necessarily correspond to motor ability. The average daily stride count remained well below stride counts reflective of typical levels in youth without disability (8381 strides).18 A number of researchers and clinicians with an interest in the activity levels of children with CP report similarly low levels of walking activity.6,17,23 For some children, the purpose of SEMLS may be to maintain walking ability. Therefore, no change in stride counts following surgery could be considered a success. However, with a significant improvement in the GDI and walking quality, an increase in walking quantity would be expected. These findings are helping to drive a shift in focus of rehabilitation programs from minimizing deficits to enhancing activity participation. Our results support the necessity of this shift, as we have shown that, while gait surgery leads to a significant improvement in gait quality, it has no effect on gait quantity 1 to 2 years after surgery.

In addition to physical impairments, factors such as motor coordination, physical fitness, social skills, and children's perception of their ability to participate all play a role in a child with CP's physical activity level. Surgery is aimed to reduce physical impairments such as altered muscle balance, and mechanical malalignment, but does little to affect motor abilities and social skills. While the ability to move in a typical pattern improves, walking activity remains low. Further studies need to explore other possible causes of low walking activity, such as motor coordination, social skills, and perceptions. Anecdotally, when children with preoperative GDIs in the 74 to 130 range12 are removed from the analysis, the average change in mean daily stride counts for the remainder of the group increases rather than remaining unchanged. This suggests that gait impairment drives physical activity to a certain extent, when movement patterns are in the typical range, other factors are contributing to low habitual physical activity and when movement patterns deviate further from typical, the potential for gains in walking activity following corrective gait surgery improves. Physical activity in children with CP is a complex topic that warrants further study and investigation.

As with any research, there are some limitations to this study. The data analyzed for this study were collected for the clinical purpose of monitoring each child's surgical recovery. Home-based assessment of the quantity of walking activity is less well controlled than a gait laboratory assessment of the quality of mobility. Step totals vary from day to day, and while we included enough days to theoretically represent habitual walking activity, there is no way to confirm that the recorded data represent average walking activity for an individual child. In addition, children are more active on school days than on nonschool days. While weekend days and weekdays were kept consistent from one visit to the next, the time of year (school year or summer) was not controlled. The limited sample size also imposed several limitations on the data analysis. There were not enough subjects to evaluate the effect of age, different surgical combinations, or GMFCS level.

Single-event multilevel surgery improves gait deviations in children with CP. Contrary to previously hypothesized, the improvement in gait pattern has limited correlation with postoperative change in walking activity at 1 to 2 years, suggesting that movement pattern recovers much more quickly than mobility performance. Our results demonstrate a need to pair surgical with additional intervention during recovery to affect any long-term improvements in walking activity and participation.

REFERENCES

1. Bax M, Goldstein M, Rosenbaum P, Leviton A, Paneth N. Proposed definition and classification of cerebral palsy. Dev Med Child Neurol. 2005;47:571–576. doi:10.1017/S001216220500112X.
2. Oskoui M, Coutinho F, Dykeman J, Jetté N, Pringsheim T. An update on the prevalence of cerebral palsy: a systematic review and meta-analysis. Dev Med Child Neurol. 2013;55(6):509–519. doi:10.1111/dmcn.12080.
3. van Eck M, Dallmeijer AJ, Beckerman H, Pam VDH, Voorman JM, Becher JG. Physical activity level and related factors in adolescents with cerebral palsy. Pediatr Exerc Sci. 2008;20(1):95–106. http://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=2009871482&site=ehost-live.
4. Opheim A, Jahnsen R, Olsson E, Stanghelle JK. Walking function, pain, and fatigue in adults with cerebral palsy: a 7-year follow-up study. Dev Med Child Neurol. 2009;51(5):381–388. doi:10.1111/j.1469-8749.2008.03250.x.
5. Harvey A, Rosenbaum P, Hanna S, Yousefi-Nooraie R, Graham K. Longitudinal changes in mobility following single-event multilevel surgery in ambulatory children with cerebral palsy. J Rehabil Med. 2012;44(2):137–143. doi:10.2340/16501977-0916.
6. Stevens SL, Holbrook EA, Fuller DK, Morgan DW. Influence of age on step activity patterns in children with cerebral palsy and typically developing children. Arch Phys Med Rehabil. 2010;91(12):1891–1896. doi:10.1016/j.apmr.2010.08.015.
7. Wilson NC, Chong J, Mackey AH, Stott NS. Reported outcomes of lower limb orthopaedic surgery in children and adolescents with cerebral palsy: a mapping review. Dev Med Child Neurol. 2014;56:808–814. doi:10.1111/dmcn.12431.
8. Steele KM, Rozumalski A, Schwartz MH. Muscle synergies and complexity of neuromuscular control during gait in cerebral palsy. Dev Med Child Neurol. 2015;57(12):1176–1182. doi:10.1111/dmcn.12826.
9. Rutz E, Donath S, Tirosh O, Graham HK, Baker R. Explaining the variability improvements in gait quality as a result of single event multi-level surgery in cerebral palsy. Gait Posture. 2013;38(3):455–460. doi:10.1016/j.gaitpost.2013.01.014.
10. Mcginley JL, Dobson F, Ganeshalingam R, Shore BJ, Rutz E, Graham HK. Single-event multilevel surgery for children with cerebral palsy: a systematic review. Dev Med Child Neurol. 2012;54(2):117–128. doi:10.1111/j.1469-8749.2011.04143.x.
11. Bjornson KF, Zhou C, Stevenson RD, Christakis D. Relationship of stride activity and participation in mobility-based life habits among children with cerebral palsy. Arch Phys Med Rehabil. 2014;95(2):360–368. doi:10.1016/j.apmr.2013.10.022.
12. Schwartz MH, Rozumalski A. The gait deviation index: a new comprehensive index of gait pathology. Gait Posture. 2008;28(3):351–357. doi:10.1016/j.gaitpost.2008.05.001.
13. Haumont T, Church C, Hager S, et al. Flexed-knee gait in children with cerebral palsy: a 10-year follow-up study. J Child Orthop. 2013;7(5):435–443. doi:10.1007/s11832-013-0505-8.
14. Malt MA, Aarli Å, Bogen B, Fevang JM. Correlation between the Gait Deviation Index and gross motor function (GMFCS level) in children with cerebral palsy [published online ahead of print May 13, 2016]. J Child Orthop. doi:10.1007/s11832-016-0738-4.
15. Maanum G, Jahnsen R, Stanghelle JK, Sandvik L, Larsen KL, Keller A. Face and construct validity of the Gait Deviation Index in adults with spastic cerebral palsy. J Rehabil Med. 2012;44(3):272–275. doi:10.2340/16501977-0930.
16. Wilson NC, Signal N, Naude Y, Taylor D, Stott NS. Gait Deviation Index correlates with daily step activity in children with cerebral palsy. Arch Phys Med Rehabil. 2015;96(10):1924–1927. doi:10.1016/j.apmr.2015.05.024.
17. Bjornson KF, Belza B, Kartin D, Logsdon R, Mclaughlin JF. Ambulatory physical activity performance in youth with cerebral palsy and youth who are developing typically. Phys Ther. 2007;87(3):248–257. doi:10.2522/ptj.20060157.
18. Bjornson KF, Zhou C, Stevenson R, Christakis D, Song K. Walking activity patterns in youth with cerebral palsy and youth developing typically. Disabil Rehabil. 2014;36(15):1279–1284. doi:10.3109/09638288.2013.845254.
19. van Wely L, Becher JG, Balemans ACJ, Dallmeijer AJ. Ambulatory activity of children with cerebral palsy: which characteristics are important? Dev Med Child Neurol. 2012;54:436–442. doi:10.1111/j.1469-8749.2012.04251.x.
20. Nicholson K, Weaver A, George A, Hulbert R, Church C, Lennon N. Developing a clinical protocol for habitual physical activity monitoring in youth with cerebral palsy. Pediatr Phys Ther. 2017;29(1):2–7. doi:10.1097/PEP.0000000000000320.
21. Evans JD. Straightforward Statistics for the Behavioral Sciences. Pacific Grove, CA: Brooks/Cole Publishing; 1996.
22. Lennon N, Hulbert R, Church C, Miller F. Surgical burden and recovery of walking performance in youth with cerebral palsy. Dev Med Child Neurol. 2015;57(S5):96–97. doi:10.1111/dmcn.12761.
23. Van Wely L, Dallmeijer AJ, Balemans ACJ, Zhou C, Becher JG, Bjornson KF. Walking activity of children with cerebral palsy and children developing typically: a comparison between the Netherlands and the United States. Disabil Rehabil. 2014;36(25):1–7. doi:10.3109/09638288.2014.892639.
Keywords:

cerebral palsy; Gait Deviation Index; StepWatch; walking activity

© 2018 Academy of Pediatric Physical Therapy of the American Physical Therapy Association