Factors Associated With Short-Term Recovery Following Single-Event Multilevel Surgery for Children With Cerebral Palsy : Pediatric Physical Therapy

Secondary Logo

Journal Logo


Factors Associated With Short-Term Recovery Following Single-Event Multilevel Surgery for Children With Cerebral Palsy

Lennon, Nancy PT, DPT, MS; Gerry, Grace BS; Biermann, Isabel BA; Beaman, Jason PT; Mamula, Nicole MOT, OTR/L; Gilmore, Abigail MS, OTR/L; Niiler, Tim PhD; Shrader, M. Wade MD; Owens, Laura L. MD

Author Information
Pediatric Physical Therapy 35(1):p 93-99, January 2023. | DOI: 10.1097/PEP.0000000000000976
  • Free



Cerebral palsy (CP) is a condition caused by perinatal insult to the developing brain resulting in a disorder of movement and posture.1 Cerebral palsy is classified by movement dysfunction, including spastic, dyskinetic, and ataxic, through topography including hemiplegia, diplegia, and quadriplegia, and based on gross motor function in sitting and standing using the Gross Motor Function Classification System (GMFCS).2,3 During the past 30 years, orthopedic and rehabilitation management of the secondary musculoskeletal problems common to spastic forms of CP has seen innovations in surgical methods, neurologic treatment, therapy approaches, and assistive technology.4 Interventions aimed at improving mobility outcomes for children and youth with CP who are able to walk are widely described in the clinical literature.

As children grow who have spastic forms of CP, musculotendinous contractures and bony deformities often develop. Such deformities can suppress gross motor development in the young child and diminish gross motor function during the teenage years.3,5 Hanna et al5 used data from serial evaluations of the Gross Motor Function Measure (GMFM) as evidence that children with CP functioning at GMFCS levels III, IV, and V are at high risk for decline in motor function during adolescence. This critical developmental time from early to late adolescence needs to be closely monitored by orthopedic and rehabilitation specialists such that interventions to preserve mobility can be introduced and evaluated for effectiveness.

Orthopedic surgery is commonly performed to correct malalignments with the intent of preventing or counteracting motor loss.6 According to Gorton et al,6 gait outcomes as measured by the Gillette Gait Index of 75 children with CP who underwent lower extremity surgery improved 1 year postsurgery compared with counterparts who did not undergo surgery. Such improvements in gait can contribute to improved performance of daily mobility tasks. Surgical procedures are reported to improve alignment and increase motion during walking4,7 but less frequently report elevated mobility function, participation, and life satisfaction.8

The concept of single-event multilevel surgery (SEMLS) was first introduced by Norlin and Tkaczuk in 1985.9 By definition, an SEMLS is performed when 2 or more bony or soft-tissue procedures are performed at 2 or more anatomical levels during 1 surgery.4 Following an orthopedic multilevel surgery, some expected areas of improvement after full recovery include postural alignment, range of motion, and motor function.10 Institutions with access to gait analysis use SEMLS as the standard treatment to fix bony and soft-tissue abnormalities in children with CP.11 Single-event multilevel surgery has gained popularity because of the single hospital admission with 1 intense recovery period of rehabilitation.12 Lamberts et al12 report consistent positive outcomes in gait kinematics after lower extremity SEMLS. More recently, the concept of SEMLS has evolved because many children need more than 1 round of surgical intervention as a result of the effects of growth and spasticity. Many surgeons and researchers use the term “multilevel surgery” to better describe that the surgical intervention addresses many or all of the impairments at a single point in time but also implies that additional correction may be needed in the future.13

Rehabilitation following surgery is often included for secondary musculoskeletal issues. Rehabilitation following SEMLS focuses on maximizing passive and active range of motion, improving functional strength, and minimizing the development of contractures.3 According to a systematic review of physical therapy following SEMLS, muscle strength is severely reduced at 6- and 12-month follow-up visits after multilevel surgery compared with presurgery.14 van Bommel and colleagues14 describe variations in therapy activities, treatment dosage, and duration of sessions due to the heterogeneity found among children with CP. In addition to factors specific to the child, surgical factors might also limit a child's participation in therapy activities postoperatively.

The purpose of this study was to examine factors that affect recovery of mobility function immediately following SEMLS in youth with CP. We hypothesized that child-related factors (gross motor function) and treatment-related factors (surgical burden, therapy dose) will impact postoperative recovery.



This was an Institutional Review Board–approved retrospective case-controlled cohort study. Data were collected from June 2017 to June 2019. Inclusion criteria were a diagnosis of CP or a CP-like condition (neuromotor condition with onset at birth, genetic etiology), a preoperative instrumented gait analysis visit, a lower extremity SEMLS, and an episode of rehabilitation at our center. Rehabilitation episodes included both physical therapy and occupational therapy sessions. The inclusion criteria were consistent with the standard of care for treating postoperative participants with CP. Exclusion criteria included no diagnosis of CP or a similar condition, no preoperative instrumented gait analysis visit, no lower extremity SEMLS, or on-site rehabilitation episode within 1 month of surgery. As this was a retrospective study, no recruitment of participants was required. Charts were reviewed to identify participants who met inclusion criteria.


Measurement Tools. The Gross Motor Function Measure dimension D (GMFM-D) was administered as part of a preoperative instrumented gait analysis visit. The GMFM-D is 1 of 5 dimensions of the GMFM-88, which is validated for children with CP. It consists of 13 items scored on the Likert scale from 0 to 3, with total scores ranging from 0 to 39.15 Gait speed, Gait Deviation Index (GDI), and GMFCS level were also recorded at each participant's preoperative gait analysis visit. The GDI is a numerical score that represents a child's degree of gait abnormalities based on kinematic measurements. A GDI score of 100 (±10) indicates the absence of gait pathology, while a score of less than 90 indicates the presence of gait pathology, with lower scores representing more significant gait deviations.16 The GMFCS is a classification system with 5 levels representing the gross motor function of children with CP.17

The Functional Independence Measure for children (WeeFIM) was given at the admission and discharge dates of each participant's inpatient rehabilitation episode. The WeeFIM instrument is an outcomes measurement tool that documents the severity of the disability for children under 3 domains that include self-care, mobility, and cognition, with the motor score consisting of the sum of the self-care and mobility scores.18,19 Each of the 18 items is scored on a scale of 1 to 7, with 1 indicating that the child needs total assistance and 7 indicating complete independence. The minimum score is 18 and the maximum score is 126.

Other Independent Variables. Information about participant demographics, number of osteotomies per surgery, surgical complications (wounds, wound infections, and hardware complications), and medical comorbidities (seizures, gastrostomy tube, and cardiopulmonary issues) were recorded with medical chart review and past department reports. Higher surgical burden, calculated on the basis of the number of osteotomies, predicts lower walking activity after SEMLS.20 Surgical complications and medical comorbidities were recorded as “Yes” if any of the 3 corresponding complications or comorbidities were present and “No” if none were present.

Therapy dose was reported as the sum of the number of physical therapy and occupational therapy sessions during the participant's rehabilitation episode. Rehabilitation episodes consisted of the inpatient rehabilitation program, Comprehensive Outpatient Rehabilitation Program (CORP), or a combination of both. The CORP is a day hospital program for participants who require the intensity of inpatient therapy services up to 5 days per week, but their medical needs do not necessitate an inpatient hospital admission. Length of stay was dependent on rehabilitation progression, child/family goals, and in some cases insurance coverage.

Dependent Variable. The change in Gross Motor Ability Estimator (GMAE) from admission to discharge of the rehabilitation episode was used as the dependent variable. The GMAE allows calculation of the GMFM-66 using an abbreviated item set, with total scores ranging from 0 to 100.21,22

Data Analysis

Regression analysis was used to determine how preoperative measurements, rehabilitation admission measurements, and other dependent variables impact change in GMAE. Objective measures included GMFCS level, preoperative GMFM-D, gait speed, GDI, and rehabilitation admission GMAE and WeeFIM (cognition and motor). Other dependent variables included number of osteotomies, age at surgery, surgical complications, medical comorbidities, and therapy dose.

Many of our collected variables had the potential to change GMAE. Relevant variables were first identified using correlational analysis; only those with at least a moderate correlation to change in GMAE (r > 0.3) were included in the regression model. Additional predictor variables were winnowed using the Akaike Information Criteria (AIC) to identify the smallest subset of variables, which explained the largest proportion of the variability in the change in GMAE. To estimate both the import and the stability of various coefficients, this model was then run 1,000 times with data sampled with replacement. The results of this bootstrapping technique were presented in both standard tabular form (showing means) and in graphical form (showing the full range of results with median values shown in box plots). Standard coefficients whose medians were farther from 0 (range of these is typically from −1 to 1 in this case), whose percent variability explained was larger, and whose median P value was smaller were considered good predictors.

Another regression analysis was run with AIC to see whether only presurgical variables were statistically significant in predicting change in GMAE post SEMLS. The variables used in the regression analysis were number of osteotomies, age at surgery, medical comorbidities, GMFCS level, WeeFIM cognition, preoperative GMFM-D, preoperative GDI mean, and preoperative gait speed. Variables were treated as either categorical (GMFCS, complications, comorbidities) or continuous (preoperative GMFM-D, gait speed, GDI, rehabilitation admission GMAE, WeeFIM cognition/motor, number of osteotomies, age at surgery, and therapy dose) in all regression analyses.


One hundred fifty children were reviewed with CP, a preoperative gait analysis, and orthopedic surgery. Eighty-one children were excluded because they did not have rehabilitation at our center. One child was removed because the inpatient rehabilitation admission date was too long after surgery skewing this child's baseline data. The 68 remaining children, classified via GMFCS levels (levels I [4%], II [35%], III [38%], and IV [22%]), were of an average age of 13.8 years. The cohort was 47% male and 53% female. The cohort was 68% White, 18% African American, 1% Asian, and 12% as either mixed or other (1% declined to answer) (Table 1). The average number of osteotomies per participant SEMLS was 2.6 (Table 2). The participants who had inpatient rehabilitation only or both inpatient and CORP received an average of 95.9 inpatient therapy sessions, and participants who participated in CORP only or both inpatient and CORP received an average of 77.2 CORP sessions (Table 2).

TABLE 1 - Participant Characteristics
Characteristic Participant (n = 68)
Type of rehabilitation, n (%)
Inpatient only 26 (38)
Comprehensive Outpatient Rehabilitation Program (CORP) only 4 (6)
Both inpatient and CORP 38 (56)
Sex, n (%)
Male 32 (47)
Female 36 (53)
Age at surgery, mean (SD), y 13.8 (3.8)
GMFCS, n (%)
Level I 3 (4)
Level II 24 (35)
Level III 26 (38)
Level IV 15 (22)
Race, %
White 68
African American 18
Asian 1
Other and 2 or more 11
Prefer not to answer 1
Ethnicity, %
Non-Hispanic or Latino 93
Some Hispanic, Latino, or Spanish origin 6
Prefer not to answer 1
Abbreviation: GMFCS, Gross Motor Function Classification System.

TABLE 2 - Preoperative, Surgical, and Postoperative Variables
Variable Participant (n = 68)
Number of osteotomies, mean (range) 2.6 (0-7)
Surgical procedures, n (%)
Soft-tissue procedures only 8 (12)
Bony procedures only 6 (9)
Both soft-tissue and bony procedures 54 (79)
Surgical complications, n (%)
Yes 19 (28)
No 49 (72)
Medical comorbidities, n (%)
Yes 38 (56)
No 30 (44)
Therapy dose, mean (SD) 138.0 (81.7)
Number of inpatient sessions 95.9 (47.2)
Number of CORP sessions 77.2 (61.4)
Days after surgery, mean (SD)
Inpatient rehabilitation admission 15.5 (18.7)
Inpatient rehabilitation discharge 50.9 (25.0)
CORP admission 53.5 (22.4)
CORP discharge 125.7 (67.4)
Measurement battery, mean (SD)
Preoperative GMFM-D 22.7 (10.1)
Preoperative gait speed, cm/s 63.7 (31.4)
Preoperative left-side GDI 62.5 (19.5)
Preoperative right-side GDI 62.6 (19.3)
Rehabilitation admission GMAE 37.5 (6.5)
Rehabilitation admission WeeFIM (cognition) 22.6 (9.8)
Rehabilitation admission WeeFIM (motor) 31.3 (11.9)
Abbreviations: CORP, Comprehensive Outpatient Rehabilitation Program; GDI, Gait Deviation Index; GMAE, Gross Motor Ability Estimator; GMFM-D, Gross Motor Function Measure dimension D; WeeFIM, Functional Independence Measure for children.

The AIC regression analysis, used to winnow the possible predictor variables, was significant (Preg < .001) and explained 63.7% of the variability with 9 variables although not all were statistically significant. These variables included the number of osteotomies, total therapy dose, surgical complications, medical comorbidities, baseline cognition, baseline WeeFIM motor, preoperative GMFM-D, admission GMAE, and preoperative gait speed, and are given by the following equation:

delta_GMAE = β0 + β1*Total Therapy Dose Sessions + β2*Number of Osteotomies + β3*Surgical Complications + β4*Medical Comorbidities + β5*Total Cognition + β6*Total Motor +β7*Preoperative GMFMD + β8*Admission GMAE + β9*Preoperative Gait Speed

where the β values correspond to the regression coefficients in Table 3. The bootstrapped regression results were quite similar to these results, with slightly less significance on average but explained more of the variability (73.8%) (Table 3). In the final model, 33 children were included.

TABLE 3 - Akaike Information Criteria and Bootstrapped Regressions
Variable, Mean (SD) AIC Regression Adjusted R 2 = 0.637, P reg < .001 Bootstrapped Regression Adjusted R 2 = 0.738
β P β P
Intercept 26.68 (8.48) .0045a 26.58 (8.50) .014a
Number of osteotomies −1.82 (0.66) .011a −1.86 (0.74) .025a
Total therapy sessions 0.046 (0.017) .011a 0.048 (0.023) .042a
Surgical complications −5.23 (2.64) .060 −4.83 (3.34) .15
Medical comorbidities −3.03 (2.51) .24 −2.94 (3.06) .26
Baseline total cognition −0.42 (0.19) .038a −0.39 (0.22) .094
Baseline total motor 0.23 (0.14) .12 0.20 (0.16) .21
Preoperative GMFM-D 0.22 (0.14) .14 0.22 (0.14) .17
Admission GMAE −0.54 (0.15) .0017a −0.55 (0.16) .0084a
Preoperative gait speed 0.064 (0.043) .14 0.066 (0.047) .17
Abbreviations: AIC, Akaike Information Criteria; GMFM-D, Gross Motor Function Measure dimension D.
aSignificant value, P < .05.

The most stable significant variables with larger absolute value standard regression coefficients were the number of osteotomies, the number of total therapy sessions, and the admission GMAE value (Table 3 and Figure 1a-d). Three variables were statistically significant at the P value of less than .05 level, including number of osteotomies, therapy dose, and admission GMAE (Table 3 and Figure 1d). With each added osteotomy, the change in GMAE decreased by 1.86 points. With each therapy session, the predicted change in GMAE increased by 0.048 points. With each negative point in the admission GMAE, the predicted change in GMAE increased by 0.55 points (Table 3 and Figure 1a). Altogether, 26.56% of the variability in change in GMAE was explained by number of osteotomies, therapy dose, and admission GMAE (Figure 1c). Surgical complications, medical comorbidities, WeeFIM cognition, WeeFIM motor, preoperative GMFM-D, and preoperative gait speed also contributed to the regression analysis, explaining 38.34% of the variability in change in GMAE, but were not statistically significant as individual variables (Table 3 and Figure 1c and 1d). Preoperative GDI, GMFCS level, and age at surgery did not contribute to predicted change in GMAE in the regression analysis.

Fig. 1.:
Results from bootstrapped regression analysis. (a) Mean regression coefficients with full ranges shown. Variables with larger ranges or that cross 0 should be considered less reliable as predictors. (b) Using standardized regression coefficients allows for comparison of coefficients on an equitable basis. In these boxplots, the median values are given by the horizontal lines. As before, coefficients whose bounds cross 0 are suspect as they cannot predict whether a variable has a positive or negative contribution to the outcome. (c) The most important variables in the model can also be looked at in terms of variability explained. (d) When these are combined with the variables having the most stable P values (and significance across the bootstrapping, it is possible to determine which variables are most likely to have impact on recovery: N Osteotomies, Total Therapy Dosage, and Adm GMAE. ADM GMAE indicates admission Gross Motor Ability Estimator; B Total Cognition, Functional Independence Measure for children—Cognition; B Total Motor, Functional Independence Measure for children—Motor; MedComorbid, medical comorbidities; N Osteotomies, number of osteotomies; PreOpGaitSpeed, preoperative gait speed; PreOpGMFMD, preoperative Gross Motor Function Measure dimension D; SurgComp, surgery complications.

Notably, the mean percent variabilities per variable slightly exceed that of the adjusted R2 in the first model (Figure 1c). This is because the variables are not completely uncorrelated; however, the relative import of these variables remains.

It is noteworthy that the final change in GMAE is dependent on both variables known prior to surgery and some variables known only after surgery. The WeeFIM motor and admission GMAE are evaluated postoperatively and at the start of therapy, surgical complications are revealed at or shortly after the time of surgery, and total therapy dose is known only at the end of the rehabilitation episode. Although our aforementioned model makes it possible to estimate the effects of these variables independently, running the same model while limiting variables to those known prior to surgery indicates the importance of postsurgical variables. The result of the regression based only on preoperative variables (adjusted R2 = 0.32, Preg = .007; Table 4) lends import to the idea that these postsurgical variables, in particular the number of therapy doses as well as the admission GMAE, were significant predictors of outcome. The elimination of these variables results in a reduction in variability explained by a factor of 2.

TABLE 4 - Regression With Preoperative Variables Only
Variable, Mean (SD) AIC RegressionAdjusted R 2 = 0.3193, P reg = .008
β P
Intercept 5.42 (7.36) .47
Number of osteotomies −1.31 (0.83) .13
Medical comorbidities 0.076 (3.25) .98
Baseline total cognition −0.18 (0.18) .33
Preoperative GMFM-D 0.26 (0.19) .18
Preoperative gait speed 0.10 (0.056) .085
Abbreviations: AIC, Akaike Information Criteria; GMFM-D, Gross Motor Function Measure dimension D.

Ultimately, the number of osteotomies, total therapy dose, and admission GMAE were the most reliably significant variables required to predict change in GMAE during the post-SEMLS rehabilitation period. The WeeFIM cognition score is also important but is not as reliable in predicting change as revealed in the bootstrap analysis.


The literature reporting outcomes after SEMLS in youth with CP provides consistent evidence of improved gait patterns after surgical correction, though less evidence of improved gross motor function and functional mobility outcomes.8,23 Postoperative physical and occupational therapy as part of a comprehensive rehabilitation program is the gold standard of specialized care for youth with CP. Despite its routine use in specialty centers, descriptions of motor recovery or the factors that influence motor recovery during the months following SEMLS are lacking. In clinical practice, recovery from SEMLS can vary widely among youth with CP, with some participants “bouncing back” relatively quickly and others experiencing drawn-out recoveries. This clinical retrospective study sought to examine gross motor outcomes during the post-SEMLS rehabilitation period.

One significant predictor of gross motor change during the rehabilitation phase was the magnitude or “burden” of orthopedic surgery on the child. For each osteotomy performed as part of an SEMLS event, the expected gain in GMAE was reduced by 1.8 points. We attributed this, at least in part, to the negative effects of orthopedic surgery on acute lower extremity strength and weight-bearing ability, as others have reported.14 A proportional effect of surgical burden on gross motor recovery with “more” surgery associated with less progress. Our related work revealed a similar proportional effect of surgical burden on recovery of walking activity as we found that higher surgical burden resulted in lower walking activity 1 year postoperatively in youth with CP.20

Outside of the surgical setting, pain during daily activities plays a role in the decline in gross motor function in youth with CP.24 Higher pain levels are experienced by people who undergo higher-burden surgical procedures.25 Based on this evidence, it can be inferred that pain maps play a role in the lower GMAE change scores seen in youth who underwent higher-burden surgical procedures. Recognizing and treating pain in the postoperative therapy setting might not only improve participant comfort, this might also improve gross motor progress during early SEMLS rehabilitation. To improve gross motor function after high-burden surgery, as well as outcomes such as walking activity, it is essential to identify and treat the induced weakness and pain associated with multiple osteotomies in the context of SEMLS.

Lower GMAE scores upon admission to the rehabilitation service were associated with higher change in GMAE at discharge. In addition, the multiple regression revealed that higher preoperative GMFM-D scores and gait speed were associated with greater improvement in GMAE. Taken together, these results indicate that youth with higher preoperative gross motor function followed by an acute decline immediately postoperatively may have greater potential for increases in GMAE at discharge from rehabilitation. This acute decline in gross motor function is an expected part of the recovery process, as Chang et al26 report, a decrease in GMFM-66 in children with CP 6 weeks post SEMLS with improvement in GMFM-66 noted at 6 months post SEMLS. Although factors in the study predict changes of only fractions of a point on the GMAE, this could represent valuable functional skills. The difference of 1 point on the GMAE (corresponding to 21 more therapy sessions and starting with a baseline GMAE 2 points lower) could represent new abilities such as the ability to get up from the floor without using hands or standing independently for 3 seconds.

Finally, a strong positive association exists between the number of physical therapy and occupational therapy sessions, or “therapy dose,” and change in gross motor function during the rehabilitation episode. Hanna and associates5 have described the detrimental effects of muscle weakness in youth with CP, especially for those at higher GMFCS levels. Our finding of greater gross motor function gains for youth who receive more therapy is consistent with reports from other authors who have described functional decline post SEMLS, followed by recovery of function during episodes of therapy.14,27 Formal physical and occupational therapy during the post-SEMLS rehabilitation phase appears to be critically important for achieving functional gross motor and daily living goals for youth with CP.12,27 Prospective studies with controlled therapy doses for specific orthopedic interventions are needed to provide specific recommendations related to ideal therapeutic protocols.


The retrospective nature of this study imposes several limitations. There was variability in therapy dose and timing post SEMLS. Although all rehabilitation episodes began in the first few weeks after surgery, discharge from rehabilitation ranged from 1 month to more than 6 months from surgery. We included both physical and occupational therapy sessions in calculating the therapy dose, and different interventions were used depending on participant goals. There was variability in the specific interventions administered as part of the rehabilitation episodes. Although this presents a limitation in standardization, it is consistent with individualized approaches needed, given the heterogeneity of youth with CP.28,29 Pain tolerance and motivation were not included as variables and these factors likely had some effect on motor progress post SEMLS.14 Finally, our clinical cohort included youth functioning at GMFCS level IV; these participants have very different goals (maintaining household standing function) than youth at GMFCS levels I to III (improving gross motor function), and this may have contributed to variability in results.5


Therapy dose, pre- and postoperative gross motor function, and surgical burden impact recovery of functional mobility in rehabilitation and should be considered in surgical and postoperative planning.

What this adds to the evidence:

Rehabilitation following SEMLS is common practice to address postoperative muscle weakness and is considered essential to successful functional outcomes in children with cerebral palsy, but there is a lack of evidence regarding care and its association with achieving functional mobility goals. Results of this clinical retrospective study add to the understanding of SEMLS recovery, providing evidence for the importance of postoperative rehabilitation dose, especially for youth with cerebral palsy who undergo high-burden surgical procedures.


1. Rosenbaum P, Paneth N, Leviton A, et al. A report: the definition and classification of cerebral palsy April 2006. Dev Med Child Neurol Suppl. 2007;109:8–14.
2. Rethlefsen SA, Ryan DD, Kay RM. Classification systems in cerebral palsy. Orthop Clin North Am. 2010;41:457–467. doi:10.1016/j.ocl.2010.06.005.
3. Tecklin JS. Pediatric Physical Therapy. Philadelphia, PA: Lippincott Williams & Wilkins: Wolters Kluwer; 2015.
4. 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;54:117–128. doi:10.1111/j.1469-8749.2011.04143.x.
5. Hanna SE, Rosenbaum PL, Bartlett DJ, et al. Stability and decline in gross motor function among children and youth with cerebral palsy aged 2 to 21 years. Dev Med Child Neurol. 2009;51:295–302. doi:10.1111/j.1469-8749.2008.03196.x.
6. Gorton GE III, Abel MF, Oeffinger DJ, et al. A prospective cohort study of the effects of lower extremity orthopaedic surgery on outcome measures in ambulatory children with cerebral palsy. J Pediatr Orthop. 2009;29(8):903–909. doi:10.1097/BPO.0b013e3181c11c0c.
7. Thomason P, Baker R, Dodd K, et al. Single-event multilevel surgery in children with spastic diplegia: a pilot randomized controlled trial. J Bone Joint Surg Am. 2011;93(5):451–460. doi:10.2106/jbjs.J.00410.
8. Dequeker G, Van Campenhout A, Feys H, Molenaers G. Evolution of self-care and functional mobility after single-event multilevel surgery in children and adolescents with spastic diplegic cerebral palsy. Dev Med Child Neurol. 2018;60(5):505–512. doi:10.1111/dmcn.13683.
9. Norlin R, Tkaczuk H. One-session surgery for correction of lower extremity deformities in children with cerebral palsy. J Pediatr Orthop. 1985;5(2):208–211.
10. Harryman SE. Lower-extremity surgery for children with cerebral palsy: physical therapy management. Phys Ther. 1992;72(1):16–24. doi:10.1093/ptj/72.1.16.
11. Thomason P, Graham HK. Rehabilitation of children with cerebral palsy after single-event multilevel surgery. In: Morris ME, Iansek R, eds. Rehabilitation in Movement Disorders. Cambridge: Cambridge University Press; 2013;203–216.
12. Lamberts RP, Burger M, du Toit J, Langerak NG. A systematic review of the effects of single-event multilevel surgery on gait parameters in children with spastic cerebral palsy. PLoS One. 2016;11(10):e0164686. doi:10.1371/journal.pone.0164686.
13. Dreher T, Thomason P, Švehlík M, et al. Long-term development of gait after multilevel surgery in children with cerebral palsy: a multicentre cohort study. Dev Med Child Neurol. 2018;60(1):88–93. doi:10.1111/dmcn.13618.
14. van Bommel EEH, Arts MME, Jongerius PH, Ratter J, Rameckers EAA. Physical therapy treatment in children with cerebral palsy after single-event multilevel surgery: a qualitative systematic review. A first step towards a clinical guideline for physical therapy after single-event multilevel surgery. Ther Adv Chronic Dis. 2019;10:2040622319854241. doi:10.1177/2040622319854241.
15. Lundkvist Josenby A, Jarnlo GB, Gummesson C, Nordmark E. Longitudinal construct validity of the GMFM-88 total score and goal total score and the GMFM-66 score in a 5-year follow-up study. Phys Ther. 2009;89(4):342–350. doi:10.2522/ptj.20080037.
16. 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.
17. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997;39(4):214–223. doi:10.1111/j.1469-8749.1997.tb07414.x.
18. Msall ME, DiGaudio K, Rogers BT, et al. The Functional Independence Measure for Children (WeeFIM). Conceptual basis and pilot use in children with developmental disabilities. Clin Pediatr (Phila). 1994;33(7):421–430. doi:10.1177/000992289403300708.
19. Ottenbacher KJ, Taylor ET, Msall ME, et al. The stability and equivalence reliability of the functional independence measure for children (WeeFIM). Dev Med Child Neurol. 1996;38(10):907–916. doi:10.1111/j.1469-8749.1996.tb15047.x.
20. Niiler TA, Nicholson K, Fischer L, Lennon N. Factors influencing post-surgical variability in StepWatch data in youth with cerebral palsy. Gait Posture. 2019;72:234–238. doi:10.1016/j.gaitpost.2019.06.017.
21. Beckers LW, Bastiaenen CH. Application of the Gross Motor Function Measure-66 (GMFM-66) in Dutch clinical practice: a survey study. BMC Pediatr. 2015;15:146. doi:10.1186/s12887-015-0459-8.
22. Russell DJ, Avery LM, Walter SD, et al. Development and validation of item sets to improve efficiency of administration of the 66-item Gross Motor Function Measure in children with cerebral palsy. Dev Med Child Neurol. 2010;52(2):e48–e54. doi:10.1111/j.1469-8749.2009.03481.x.
23. Edwards TA, Prescott RJ, Stebbins J, Wright J, Theologis T. What is the functional mobility and quality of life in patients with cerebral palsy following single-event multilevel surgery? J Child Orthop. 2020;14(2):139–144. doi:10.1302/1863-2548.14.190148.
24. Bartlett DJ, Hanna SE, Avery L, Stevenson RD, Galuppi B. Correlates of decline in gross motor capacity in adolescents with cerebral palsy in Gross Motor Function Classification System levels III to V: an exploratory study. Dev Med Child Neurol. 2010;52(7):e155–e160. doi:10.1111/j.1469-8749.2010.03632.x.
25. Guinet AL, Khouri N, Desailly E. Rehabilitation after single-event multilevel surgery for children and young adults with cerebral palsy: a systematic review. Am J Phys Med Rehabil. 2022;101(4):389–399. doi:10.1097/PHM.0000000000001864.
26. Chang CH, Chen YY, Yeh KK, Chen CL. Gross motor function change after multilevel soft tissue release in children with cerebral palsy. Biomed J. 2017;40(3):163–168. doi:10.1016/j.bj.2016.12.003.
27. Akerstedt A, Risto O, Odman P, Oberg B. Evaluation of single event multilevel surgery and rehabilitation in children and youth with cerebral palsy—a 2-year follow-up study. Disabil Rehabil. 2010;32(7):530–539. doi:10.3109/09638280903180171.
28. Collado-Garrido L, Parás-Bravo P, Calvo-Martín P, Santibáñez-Margüello M. Impact of resistance therapy on motor function in children with cerebral palsy: a systematic review and meta-analysis. Int J Environ Res Public Health. 2019;16(22):4513. doi:10.3390/ijerph16224513.
29. Hoffman RM, Corr BB, Stuberg WA, Arpin DJ, Kurz MJ. Changes in lower extremity strength may be related to the walking speed improvements in children with cerebral palsy after gait training. Res Dev Disabil. 2018;73:14–20. doi:10.1016/j.ridd.2017.12.005.

cerebral palsy; gait; gross motor function; multi-level surgery (MLS); orthopedic surgery; single-event multi-level surgery (SEMLS); therapy dose

© 2023 Academy of Pediatric Physical Therapy of the American Physical Therapy Association