1 81

2 82
3 83
4 84
5 85
PM R XXX (2017) 1-7
6 www.pmrjournal.org 86
7 87
8 Original Research 88
9 89
10 90
11
12
Long-Term Effects of Orthoses Use on the Changes of Foot and 91
92
13
14
Ankle Joint Motions of Children With Spastic Cerebral Palsy 93
94
15 95
16 96
17 Q8 Xue-Cheng Liu, MD, David Embrey, PhD, Channing Tassone, MD, Kim Zvara, MD, 97
18 98
19 Brenna Brandsma, DPT, Roger Lyon, MD, Karin Goodfriend, MD, 99
20 100
21
Sergey Tarima, PhD, John Thometz, MD 101
22 102
23 103
24 104
25 105
26 106
Abstract
27 107
28 108
29 Background: Orthoses commonly are prescribed to children with cerebral palsy (CP) to provide foot correction and to improve 109
30 ambulatory function. Immediate effects of ankle foot orthosis (AFOs) have been investigated, but long-term kinematic effects are 110
31 111
32 lacking clinical evidence. 112
33 Objective: To determine changes in 3-dimensional ankle and foot segment motion in pediatric patients with CP between initial 113
34 and follow-up visits (18-month average time differences) in both barefoot gait and gait with their AFO. We also investigated 114
35 115
36
intravisit changes between barefoot and AFO gait. 116
37 Design: A prospective cohort study. 117
38 Setting: Children’s Hospital of Wisconsin, Department of Orthopaedic Surgery, Medical College of Wisconsin. 118
39 Patients: A total of 23 children with CP, mean age 10.5 years (6.2-18.1 years) were clinically prescribed either a solid ankle foot 119
40 120
41 orthotic (SAFO), hinged ankle foot orthotic (HAFO), or supramalleolar orthotic. 121
42 Methods: Holes were cut in the study orthoses so that electromagnetic markers could be directly placed on the skin. A 6-foot 122
43 segment model was used. 123
44 124
45
Outcome Measurements: Kinematic and kinetic data was recorded for each patient’s initial and follow-up visit (18-month follow- 125
46 up average, 15-20 months range). 126
47 Results: For the SAFO group (gait with AFO), a significant decrease in dorsiflexion was found between the initial and third visit 127
48 128
(P ¼ .008). Furthermore, the SAFO group (barefoot gait) had an increased eversion at the midfoot for most of the gait cycle
49 129
50 (P < .008). Sagittal forefoot range of motion was reduced for all 3 groups between the barefoot and AFO groups. 130
51 Conclusion: The use of AFOs long term either maintained or improved foot deformities or dysfunction. 131
52 Level of Evidence: To be determined. 132
53 133
54 134
55 135
56 136
57 137
58 138
59 Introduction spastic CP have foot or ankle deformities [3]. Two of the 139
60 most common foot deformities in children with CP are 140
61 141
62 Children with cerebral palsy (CP) usually have equinovarus and pes valgus [4,5]. 142
63 impaired ambulation and reduced balance compared Orthoses are believed to help correct abnormal gait 143
64 144
65
with children who are typically developing, and orthoses for patients with CP by allowing for a maintenance of a 145
66 often are prescribed to improve their ambulatory neutral or slight dorsiflexion and a reduction of ankle 146
67 function. In addition, patients with CP have an altered plantarflexion during swing phase (compared with 147
68 148
69 pattern of lower limb and trunk muscle activation, barefoot); these reduce the risk of the foot making 149
70 particularly with a greater muscle activation rate [1]. contact with the floor during swing phase and prevents 150
71 151
72 This may lead to a repetitive loss of balance and foot drop during swing [5,6]. Other gait deviations can 152
73 abnormal foot position. Later, it can lead to a soft-tissue be improved by the use of specific types of foot ortho- 153
74 154
75
contracture and osseous deformity [2] and ultimately ses. For example, the supramalleolar orthosis (SMO) is 155
76 foot deformities. An estimated 93% of children with prescribed primarily to control hindfoot and midfoot 156
77 157
78 158
79 1934-1482/$ - see front matter ª 2017 by the American Academy of Physical Medicine and Rehabilitation 159
80 http://dx.doi.org/10.1016/j.pmrj.2017.08.438 160

FLA 5.5.0 DTD
PMRJ1973_proof
26 September 2017
5:10 pm
ce

 2 Long-Term Effects of Orthotics in Children With CP Q1

161 varus/valgus and is thought to have a limited influence shoe, they do not measure underlying foot motion. We 241
162 242
163 in the sagittal plane control during stance phase [5,7,8]. hypothesized that a long-term (18 month) application of 243
164 Hinged ankle foot orthoses (HAFOs) are prescribed to HAFOs, SAFOs, and SMOs would demonstrate improve- 244
165 245
166
reduce foot equinus during gait to improve ankle dor- ments in ankle joint and foot segment motion, although 246
167 siflexion in mid-stance and to prevent foot drop during the SAFO improvements may be less evident. Particu- 247
168 the swing phase [6]. larly, improvements are most evident by reducing dor- 248
169 249
170 However, there are concerns that in some cases the siflexion at the forefoot during the third rocker 250
171 HAFO allows excessive dorsiflexion and may promote (terminal stance phase, or 40%-60% of gait cycle) [17]. 251
172 252
173
overstretching of the gastrocnemius [7]. Solid ankle foot 253
174 orthoses (SAFOs) often are prescribed to help provide Methods 254
175 foot and ankle stability. SAFOs also control excessive 255
176 256
177 ankle plantarflexion while preventing excessive Study Recruitment 257
178 dorsiflexion. 258
179 259
180 Although most SAFOs allow some ankle dorsiflexion, For the initial visit, 23 patients were recruited. 260
181 some have proposed that HAFOs provide a more natural Before the final visit, 2 patients discontinued the study. 261
182 262
183
kinematic and kinetic function [6,9]. Our previous study The institutional review board approved this study, and 263
184 investigated the immediate effects of orthoses and all families signed informed consent before their 264
185 confirmed that SAFO constrained forefoot range of mo- participation. 265
186 266
187 tion (ROM) and did not control excessive plantar flexion 267
188 [10]. Contrary to previous reports, the SMO showed no AFO Assessment 268
189 269
190
effects in the coronal plane. However, our study 270
191 demonstrated immediate effects of an orthosis inter- The mean time between the initial and final visit was 271
192 vention [10]. 18 months (15-20 months). All participants used clini- 272
193 273
194 There are few studies that have investigated the cally prescribed HAFOs, SAFOs, or SMOs and were 274
195 effect of orthoses on patients with CP over an extended divided into groups according to the orthoses pre- 275
196 276
197
period of time [11]. One study followed 12 children with scribed. During the first visit, 7 participants were diag- 277
198 CP over a 2-year period and discontinued their AFO use nosed as hemiplegic and the other 16 were diagnosed 278
199 Q3 for 4 weeks [12]. The investigators found that during with diplegia; 2 patients with diplegia dropped out of 279
200 280
201 these time periods, ROM and overall gait deteriorated. the study by the final visit. There were 9 participants in 281
202 Furthermore, a retrospective study examined 2 gait the SMO group, with 3 using SMOs bilaterally and 6 282
203 283
204 analysis results (average 18 months apart) of 18 patients unilaterally. Initially, 5 participants were in the SAFO 284
205 with CP [13]. Although AFOs were not analyzed as a group, with all participants using them bilaterally; 1 285
206 286
factor affecting gait, 16 of the subjects in this study subject dropped out before the final visit. Initially, 10
207 287
208 used or had used AFOs. Findings showed subjects with participants were in the HAFO group, with 6 using them 288
209 CP had worsened spatiotemporal parameters and bilaterally and 4 unilaterally; 1 bilateral HAFO user 289
210 290
211 decreased joint motion at the latter gait analysis [13]. dropped out before the final visit. A participant who 291
212 In addition to orthoses affecting a patient’s gait, wore a right HAFO and a left SMO was assigned to both 292
213 293
214
evidence shows that shoes by themselves can affect gait the HAFO and SMO groups. During the initial and final 294
215 in patients with CP. A study comparing barefoot gait, visit, the average age was 8.6 years (range 4.5-16.6 295
216 shod gait, and shod gait with orthoses found that the years) and 10.5 years (range 6.2-18.1), respectively. All Q4 296
217 297
218 use of shoes significantly alters gait in children with participants had their function classified using the Gross 298
219 spastic hemiplegia. This study found that the signifi- Motor Function Classification System (GMFCS) by a 299
220 300
221
cance in gait kinematic and kinetic changes when or- physical therapist. At the final visit, 10 participants 301
222 thoses were worn varied depending on whether the were level 1, 5 participants were level 2, 4 participants 302
223 subjects were barefoot or shod [14]. Studies in healthy were level 3, and 2 participants were level 4. A partic- 303
224 304
225 children also found that shoes could have an impact on ipant with level 1 GMFCS and a participant with level 3 305
226 gait parameters [15,16]. GMFCS from the initial visits discontinued the study. 306
227 307
228 For this study, we investigated the long-term effects During the initial visit, a pedorthotist constructed the 308
229 of HAFOs, SAFOs, or SMOs on kinematic function clinically prescribed orthosis (with holes for data 309
230 310
231
between the initial patient visit and final visit (18 collection) for each participant, which was used at the 311
232 months). Measurements were collected by the use of initial and final visits. The dorsiflexion of the ankle in 312
233 direct foot markers directly on the skin and specially the HAFOs and SAFOs are set up at plantigrade (0 of 313
234 314
235 constructed orthoses. In most studies that examine the dorsiflexion) or 1-5 of dorsiflexion, depending on the 315
236 gait of children with CP with orthoses, investigators individual cases. Orthoses without holes were con- 316
237 317
238
attach markers to the outside of the orthoses and/or the structed (using the same molds) for everyday wear by 318
239 shoe to track movement. Although markers placed in the subjects in the study. Slip-resistant material was 319
240 this manner may measure movement of the orthosis or used to modify the study orthosis and to provide 320

FLA 5.5.0 DTD
PMRJ1973_proof
26 September 2017
5:10 pm
ce

 X.-C. Liu et al. / PM R XXX (2017) 1-7 3

321 traction. Three patients who had surgery between the Table 1 401
322 Significant differences in foot segment motion between initial and 402
323 initial and final visits were seen at least 6 months after 403
final visit for the barefoot and SAFO conditions
324 their surgery. 404
325 P 405
326 Condition Joint Plane Phase Initial Visit Final Visit Value 406
327 Motion Capture 407
328 Barefoot Midfoot C LR e3.8  8.2 e16.0  17.9 .008 408
329 MSt e2.4  8.1 e14.4  19.7 .008 409
330 An electromagnetic motion tracking system 410
ISw 1.1  8.3 e11.4  16.1 .008
331 composed of StarTrak hardware (Polhemus, Colchester, 411
332
MSw 1.9  5.5 e9.7  15.3 .008 412
333
VT) and 6D Research software (Skill Technologies, TSw 0.5  4.8 e11.1  14.5 .008 413
334 Phoenix, AZ) used square sensors (w1.5  1.5 cm) to SAFO Forefoot S MSt 4.4  5.5 e1.4  6.6 .008 414
335 measure kinematic data. It used a 6-segment foot TSt 8.4  7.8 0.8  8.4 .008 415
336 416

337 marker configuration model, which was used previously Mean  SD, P < .01, difference >5 . 417
338 to measure changes in foot motion in pediatric subjects In the sagittal plane, the positive values denote dorsiflexion. In the 418
339 coronal plane, a positive value denotes inversion. The transverse 419
340 with and without CP [10,18-20]. The external sensors 420
plane, the positive values denote internal rotation.
341 were placed on the foot to track the motion of the tibia, SAFO ¼ solid ankle foot orthosis; C ¼ coronal; LR ¼ loading response;
421
342 422
343
calcaneus, cuboid, navicular, first metatarsal, and MSt ¼ mid-stance; ISw ¼ initial swing; MSw ¼ mid-swing; TSw ¼ ter- 423
344 hallux. These landmarks were positioned by palpating minal swing; S ¼ sagittal; Tst ¼ terminal stance. 424
345 anatomic landmarks. Exact sensor locations could be 425
346 426
347 referred to our previous AFO study [10,19]. Holes were 427
examined the soft-tissue movement on the foot and
348 cut in the orthoses by a pedorthotist during the first visit 428
349 found that the mean difference measured between skin 429
350
to allow sensor placement on the anatomic landmarks of 430
mounted markers and markers implanted in the bones of
351 the foot and expanded if necessary during the final visit. 431
352
the foot only exceeded 5 during 3.5% of the gait cycle 432
Before data collection, the participants stood in their
353 [22]. To account for system measurement error and skin 433
354 neutral standing positions and the sensor coordinates 434
movement error, all changes in mean angle of rotation
355 were calibrated with respect to the participants’ 435
356 or ROM <5 were disregarded. The 6-segment foot 436
357
neutral position. To help ensure a consistent neutral 437
model used in this study was found to have moderate to
358 standing posture between the barefoot trials and the 438
359 high interrater and intrarater reliability in the sagittal 439
orthoses trials, a custom-built positioning device, made
360 and coronal planes (intraclass correlation coefficient Q6 440
361 of polyvinyl chloride (ie, PVC) piping, was used in 441
362
0.36-0.94) and low reliability in the transverse plane 442
conjunction with a foot tracing to measure the neutral
363 (intraclass correlation coefficient < .30) [16]. 443
364 standing posture of a participant. Sensor coordinates 444
365 were recalibrated if a sensor fell off during gait. Before 445
366 446
367
each orthosis trial, the participant was positioned in the Statistical Analysis 447
368 barefoot neutral position with the position device and 448
369 foot tracing, and the sensor coordinates were then 449
370 For comparisons between the first and final visits, the 450
371 recalibrated for the orthosis trial. Wilcoxon signed rank test for paired data was used for 451
372 For both visits, subjects walked at a self-selected the difference between the kinematic data collected
452
373 453
374
pace. At least 2 datasets were gathered for both the for patients barefoot and in orthoses. The presence of 454
375 barefoot and the orthoses walking trials. During the highly skewed distributions caused us to choose 455
376 initial visit, 7 participants used an assistive device or 456
377 nonparametric tests. The tests were applied separately 457
378 needed parental support, but only 6 participants on 7 intervals of the gait cycle (4 intervals during 458
379 needed support for the final visit. Motion between 459
380 460
381
several foot segments were recorded, including the 461
382 tibiaecalcaneus, calcaneusecuboid, navicularefirst 462
383 metatarsal, and first metatarsalehallux. The motion 463
384 464
385 data were divided into the stance and swing phase, 465
386 which were further subdivided. The stance phase was 466
387 467
388 divided into 4 intervals [17]: loading response (LR: 468
389 0%-17% stance), mid-stance (MSt: 17%-50% stance), ter- 469
390 470
391
minal stance (TSt: 50%-83% stance), and pre-swing (PSw: 471
392 83%-100% stance). The swing phase was divided into 472
393 3 intervals [17]: initial swing (0%-33% swing), mid-swing 473
394 474
395 (33%-68% swing), and terminal swing (68%-100% swing). 475
396 The mean angle of rotation and ROM of joint for all 476
397 477
398
conditions were calculated. Measurement errors of up Figure 1. Significant differences in mid-foot segment motion in the 478

399 Q5 to 2 have been found for the SkillTech system when coronal plane for the SAFO barefoot gait between first and final visits. 479
400 recording angles between sensors [21]. Another study SAFO ¼ solid ankle foot orthosis. 480

FLA 5.5.0 DTD
PMRJ1973_proof
26 September 2017
5:10 pm
ce

 4 Long-Term Effects of Orthotics in Children With CP

481 Barefoot Versus Orthoses at Final Visit 561

482 562
483 563
484 Kinematic differences were observed during the final 564
485 565
486
visit in the SMO group between the subjects’ barefoot 566
487 gait and gait with an SMO (Table 2). At the ankle joint, 567
488 the SMO increased eversion during terminal swing by 568
489 569
490 5.3 (P < .002). At the forefoot, the SMO decreased 570
491 dorsiflexion for most of the gait cycle (P < .002) 571
492 572
493
(Figure 3). Differences in ROM were observed in all 3 573
494 groups during the final visit between the barefoot and 574
495 orthoses groups (Table 2). The HAFO demonstrated 575
496 576
497 decreased forefoot sagittal ROM during PSw by 7.8 (P < 577
498 Figure 2. Significant differences in mid-foot segment motion in the 578
.001). The SAFO decreased sagittal ROM at the forefoot
499 sagittal plane for the SAFO between first and final visits. SAFO ¼ solid 579
500 ankle foot orthosis. by 7.6 (P < .008). The SMO decreased sagittal ROM at 580
501 the forefoot during LR and TSt (P < .001). 581
502 582
503 583
504 Discussion 584
505
stance, 3 intervals during swing), 4 joints, and 3 plains. 585
506 A P value < .01 for clinically relevant changes (>5 ) was 586
507 considered to be both statistically and clinically signif- During the initial visit of our study, the use of the HAFO 587
508 produced an increase in ankle dorsiflexion and a decrease 588
509 icant. This approach compensated for excessive false- 589
510 discovery rate associated with multiple comparisons in forefoot sagittal motion [20]. Romkes et al [23] also 590
511 found an immediate improvement after HAFO use; there 591
512
and minimized the chance of reporting clinically irrel- 592
513 evant but statistically significant findings. was a reduced ankle plantarflexion during the swing phase 593
514 (e26.7 BF and 0.7 HAFO) and toe off (e17.1 BF and 2.7 594
515 HAFO) and increased step and stride length. However, our 595
516 596
517 Results final visit observed no difference between barefoot and 597
518 HAFO motion at the ankle and foot segments. 598
519 599
520 Initial Versus Final Visit When compared with our previous study, the sagittal 600
521 forefoot dorsiflexion decreased throughout the entire 601
522 602
523
When we compared the kinematics between the first barefoot trial but not enough to show clinical signifi- 603
524 visit and the final visit, kinematic differences were cance [20]. However, in our current study, there was no 604
525 observed for all 3 groups. However, significant differ- significant difference between the HAFO and barefoot 605
526 606
527 ences only emerged in the SAFO group (Table 1). After trials. This means that the HAFO may have modified or 607
528 18 months, there was an increased eversion at the promoted the sagittal ankle and forefoot motion 608
529 609
530
midfoot for most of the gait cycle (P < .008) (Figure 1). during the 18 months of the study, enough where the 610
531 When participants walked with the SAFO, there was group’s barefoot gait matched the gait they had while 611
532 decreased dorsiflexion at the forefoot during MSt and using the HAFO. Furthermore, no clinically significant 612
533 613
534 TSt (P < .008) (Figure 2). No significant ROM changes changes for the ankle or midfoot groups in any plane 614
535 were observed between the first and final visit for any indicates that the HAFO is able to maintain good foot 615
536 616
537
orthosis group. position (Table 3). 617
538 618
539 619
540 Table 2 620
541 Significant differences between barefoot and orthosis walking for the gait intervals at the final visit 621
542 622
543 Orthosis Variable Joint Plane Phase Barefoot Orthosis P Value 623
544 624
HAFO ROM Forefoot S PSw 13.5  13.9 5.7  4.2 .001
545 625
546 SAFO ROM Forefoot S TSt 11.6  6.3 4.0  2.2 .008 626
547 SMO Mean value Ankle C TSw 7.3  4.5 2.0  5.4 .002 627
548 Forefoot S LR 6.2  4.1 0.4  4.5 .001 628
549 TSt 7.5  6.4 1.6  4.4 .002 629
550 PSw 17.9  8.0 4.5  6.4 .0005 630
551 631
ISw 9.4  5.5 e2.1  7.0 .0005
552 632
553 MSw 6.5  3.8 e3.5  7.2 .001 633
554 TSw 11.3  4.0 0.2  6.4 .0005 634
555 ROM Forefoot S LR 9.9  3.7 4.1  2.6 .001 635
556 TSt 12.3  3.5 5.5  2.1 .001 636
557 637
558 Mean  SD, P < .01, difference >5 . 638
559 HAFO ¼ hinged ankle foot orthosis; ROM ¼ range of motion; S ¼ sagittal; PSw ¼ pre-swing; SAFO ¼ solid ankle foot orthosis; TSt ¼ terminal 639
560 stance; C ¼ coronal; TSw ¼ terminal swing; SMO ¼ supramalleolar orthotic; LR ¼ loading response; ISw ¼ initial swing; MSw ¼ mid-swing. 640

FLA 5.5.0 DTD
PMRJ1973_proof
26 September 2017
5:10 pm
ce

 X.-C. Liu et al. / PM R XXX (2017) 1-7 5

641 In both visits of our study, the SAFO participants’ ROM 721
642 722
643 and mean motions between the AFO and barefoot trials 723
644 revealed no significant differences other than a reduced 724
645 725
646
ROM of the forefoot during TSt. In addition, there was 726
647 an increase in eversion at the midfoot between the first 727
648 and final visits for barefoot gait for most of the gait 728
649 729
650 cycle (Figure 2). This confirms that SAFOs may not be 730
651 the best treatment option for controlling midfoot varus/ 731
652 732
653
valgus deformities, especially after long-term use. 733
654 Figure 3. Sagittal forefoot motion during AFO gait and during barefoot SAFOs particularly are known for providing maintenance 734
655 gait for the final visit. Visit 1 changes could be referred to our initial in the sagittal plane. Yet, there is a decrease in forefoot 735
656 736
visit study [10]. AFO ¼ ankle foot orthosis; SMO ¼ supramalleolar dorsiflexion motion during the third rocker (terminal
657 737
658 orthotic. 738
stance phase, or 40%-60% of gait cycle) between the
659 739
660 initial and final visits [17]. This may indicate that the 740
661 For the HAFO groups, other studies have also been group also had its sagittal motion pattern worsen over 741
662 able to observe modifications after an extended period 742
663
the 18 months. The participant’s limited ROM during 743
664 of time. Radtka et al [6] found an increased ankle dor- growth may have had an effect. Our previous study 744
665 siflexion during TSt (e1.3 BF to 16.1 HAFO), LR (7.2 confirms that SAFO constrained forefoot ROM and did 745
666 746
667 BF to 13.3 HAFO), and MSt (0.7 BF to 11.7 HAFO) after not control excessive plantar flexion [10]. For this study, 747
668 1 month of orthosis use. Dalvand et al [24] found a sig- we confirmed that an SAFO puts a stronger restriction on 748
669 749
670
nificant improvement of gross motor function after 3 the forefoot ROM compared with an SMO or HAFO, and, 750
671 months of HAFO use in patients with spastic diplegic CP. when combined with its long-term use, may provide 751
672 After 18 months, it appears that the HAFO is not only negative consequences for midfoot function. However, 752
673 753
674 able to provide improvements in sagittal ankle motion the changes between the 2 visits were comparatively 754
675 but also slightly modify and maintain sagittal forefoot small, indicating that there is some prevention of 755
676 756
677
motion. However, because spastic CP is a neurologic further deformities or dysfunction rather than a 757
678 disorder, the HAFO may be able to prevent further correction in foot alignment. In addition, the SAFO was 758
679 spasticity of the muscles that may cause foot and ankle able to maintain ankle alignment for the coronal and 759
680 760
681 joint deformities, and continued HAFO use may be sagittal planes long term, indicating it may be able to 761
682 required to maintain these improvements. prevent deformities further in the ankle (Table 3). 762
683 763
684 764
685 765
686 Table 3 766
687 Summary of intravisit and long-term kinematic and ROM changes from AFO, HAFO, SAFO, and SMO use 767
688 768
689 AFO Initial Visit changes Between Final Visit Changes Between AFO Final Visit Long-Term Change in Barefoot or 769
690 Prescription AFO and Barefoot Trials [10] and Barefoot Trials AFO Trials 770
691 771
HAFO  Increases dorsiflexion at the ankle joint  No significant changes in ankle or  Modfied ankle joint in the sagittal plane
692 772
693 during PSw, ISw, LR, and TSt midfoot and maintained ankle joint in the 773
694  No change in midfoot  Reduced sagittal forefoot ROM sagittal plane 774
695  Decreased forefoot dorsiflexion at the during TSt  Maintains midfoot in the sagittal and 775
696 end of St and Sw coronal planes 776
697  Decreased ROM at forefoot during TSt  Maintains coronal forefoot motion 777
698 778
699
and ISw  Long-term use of HAFO modified sagittal 779
700 forefoot motion by decreasing 780
701 dorsiflexion for most of gait cycle 781
702 SAFO  No significant changes in ankle or  No significant changes in ankle or  Maintains ankle in the sagittal and coronal 782
703 midfoot midfoot planes 783
704 784
 Reduced sagittal forefoot ROM during  Reduced sagittal forefoot ROM  Midfoot varus/valgus allignment worsened
705 785
706 TSt during TSt for gait cycle 786
707  Reduced sagittal forefoot motion during 787
708 MSt and TSt 788
709  Maintained forefoot in coronal plane 789
710 SMO  No siginificant changes in ankle  Increased ankle eversion during  Maintains ankle, forefoot, and midfoot 790
711 791
 Reduced plantarflexion at the midfoot TSw for the sagittal and coronal planes
712 792
713 during ISw  No siginificant changes in midfoot  No significant changes in ROM 793
714  Reduced forefoot dorsiflexion for Sw  Reduced forefoot dorsiflexion for 794
715  Reduced sagittal forefoot ROM at LR swing phase, LR, and TSt 795
716 and TSt  Reduced sagittal forefoot ROM at 796
717 TSt and LR 797
718 798
719 ROM ¼ range of motion; AFO ¼ ankle foot orthosis; HAFO ¼ hinged ankle foot orthosis; SAFO ¼ solid ankle foot orthosis; SMO ¼ supramalleolar 799
720 orthotic; PSw ¼ pre-swing; ISw ¼ initial swing; LR ¼ loading response; TSt ¼ terminal stance. 800

FLA 5.5.0 DTD
PMRJ1973_proof
26 September 2017
5:10 pm
ce

 6 Long-Term Effects of Orthotics in Children With CP

801 Only the SMO group showed changes in mean values fit. If the orthosis was too small in terms of foot width, 881
802 882
803 between barefoot and this orthosis group for the final we fabricated a new one. If the orthosis was 1-2 cm 883
804 visit, indicating that the SMO continues to change foot short of foot length, we added the orthotic length in the 884
805 885
806
alignment with immediate use. Long term, this study forefoot. Otherwise, a new orthosis was acquired. These 886
807 also confirms that SMOs are able to perform their pri- changes were made identically to both the study and 887
808 mary function of controlling hind foot and midfoot prescribed orthoses. Based on our observations, we did 888
809 889
810 varus/valgus deformities. However, contrary to other not notice any remarkable differences in stiffness be- 890
811 studies, we found that the use of an SMO during the first tween the study and prescribed orthosis, although we 891
812 892
813
and final visits decreased dorsiflexion at the forefoot for did not perform a biomechanical analysis. Furthermore, 893
814 most of the gait cycle [5,7,8]. During the final visit, the at the end of the 18 months, the prescribed orthoses 894
815 SMO also decreased forefoot dorsiflexion at the first may not have suited the patients’ needs due to growth, 895
816 896
817 rocker (LR, or 0%-10% of gait cycle), which was not seen increases or decreases in spasticity, or changes in 897
818 during the initial visit [17]. Immediate use of an SMO weight. Finally, the compliance of participants in the 898
819 899
820 limits sagittal forefoot motion overall and ROM during use of their daily use of orthoses was not monitored 900
821 stance phase. There were no significant differences carefully. 901
822 902
823
between the first and final visits in motion and ROM 903
824 between visits, indicating that the SMO was able to Conclusion 904
825 prevent foot deformity in the sagittal plane as well 905
826 906
827 (Table 3). However, a decrease in sagittal forefoot Overall, there was some improvement or maintained 907
828 dorsiflexion at the first rocker also could be a concern in status in kinematics found for 3 orthoses groups (HAFO, 908
829 909
830
supporting the foot for an improved heel contact. In SMO, or SAFO). HAFOs appear to alter the forefoot motion 910
831 addition, a decrease in forefoot dorsiflexion and sagittal and reduce dorsiflexion overall but not enough to show 911
832 ROM during PSw and TSt could be a concern for power 912
833 clinical significance. SAFOs were able to provide the 913
834 production. status quo of the foot segments and ankle joint move- 914
835 The intravisit effect of the orthoses on ROM was 915
836
ment between the first and final visits, but not as well as 916
837
consistent during first and final visits. The use of the the other prescriptions for the midfoot coronal plane. 917
838 HAFO, SMO, or SAFO reduced the sagittal ROM at the The SMO group is the only group that continues to show 918
839 forefoot compared with barefoot walking for the third 919
840 intravisit changes, particularly in the sagittal forefoot 920
841 rockers. This could have an effect on power generation motion along with a decrease in ankle eversion. Future 921
842 and overall gait efficiency. Sagittal ankle ROM and 922
843
studies may investigate the effects of varying stiffness of 923
844 maximum plantarflexion during PSw has been shown to orthosis material and how that is able to preserve power 924
845 be a contributing factor for power production [25,26]. and improve kinematic and kinetic function. 925
846 926
Furthermore, energy expenditure index scores have
847 927
848 been found to be correlated strongly with ankle ROM 928
849
Acknowledgments 929
[25]. Our findings show a reduction in the ROM for all 3
850 930
851 conditions, with the SAFO group having the lowest ROM. 931
We thank J. AL-Ramahi, BS, and C. Marquez-
852 Maximum power generation also has been shown to be 932
853 Barrientos, MS, Center for Motion Analysis, Depart- 933
854
reduced for all 3 orthoses groups compared with bare- 934
ment of Orthopaedic Surgery, for data collection and
855 foot walking, with SAFO being the most reduced as well 935
856 manuscript preparation. 936
[26]. In an attempt to improve power generation, a few
857 937
858 individuals have proposed prescribing a carbon com- 938
859 posite, spring like AFO [27]. The carbon composite References 939
860 940
861
design was able to improve maximal power generation 941
at the ankle and decrease net energy cost even with a 1. Prosser LA, Lee SC, Barbe MF, VanSant AF, Lauer RT. Trunk and hip
862 942
863 muscle activity in early walkers with and without cerebral palsyda 943
reduced ankle ROM [27]. A reduction of sagittal forefoot frequency analysis. J Electromyogr Kinesiol 2010;20:851-859.
864 944
865 ROM after immediate orthoses use or a change in the 2. Kedem P, Scher DM. Foot deformities in children with cerebral 945
866 AFO stiffness could reduce power production or energy palsy. Curr Opin Pediatr 2015;27:67-74. 946
867 3. Connell PA, D’Souza L, Dudeney S, Stephens M. Foot deformities in 947
868 efficiency for all 3 orthoses conditions. However, there 948
869 are limited studies that investigated midfoot and fore- children with cerebral palsy. J Pediatr Orthop 1998;18:743-747. 949
870 4. Davids JR, Rowan F, Davis RB. Indications for orthoses to improve 950
871
foot effects on power generation or energy expenditure, gait in children with cerebral palsy. J Am Acad Orthop Surg 2007; 951
872 which could be a promising future study. 15:178-188. 952
873 5. Davids JR. The foot and ankle in cerebral palsy. Orthop Clin North 953
874 954
Study Limitations Am 2010;41:579-593.
875 955
876 6. Radtka SA, Skinner SR, Johanson ME. A comparison of gait with 956
877 solid and hinged ankle-foot orthoses in children with spastic 957
878
A number of limitations exist in this study. Slight diplegic cerebral palsy. Gait Posture 2005;21:303-310. 958
879 adjustments were made for some participants after the 7. Novacheck TF, Kroll GJ, Gent G, et al. Orthoses. In: Gage J, 959
880 initial fabrication or study orthoses to allow for better Schwartz M, Koop S, Novacheck TF, eds. The Identification and 960

FLA 5.5.0 DTD
PMRJ1973_proof
26 September 2017
5:10 pm
ce

 X.-C. Liu et al. / PM R XXX (2017) 1-7 7

961 Treatment of Gait Problems in Cerebral Palsy. London: Mac Keith 18. Liu XC, Embrey D, Tassone C, et al. Does the Anatomical Foot and 1046
962 Press; 2009, 327-348. Ankle Motion Really Change Following Orthosis Intervention in 1047
963 1048
964 8. Carlson WE, Vaughan CL, Damiano DL, Abel MF. Orthosis manage- Children with Spastic CP? Miami, FL: Gait and Clinical Movement 1049
965 ment of gait in spastic diplegia. Am J Phys Med Rehabil 1997;76: Analysis Society; 2010. 1050
966 219-225. 19. Liu XC, Thometz J, Lyon R, et al. Six-segmental foot model using 1051
967 9. Rethlefsen S, Kay R, Dennis S, Forstein M, Tolo V. The effects of ETS. 20th ISB/ 29th ASB, Cleveland, OH, July 31 to August 5, 2005. 1052
968 1053
969 fixed and articulated ankle-foot orthoses on gait patterns in sub- 20. Marquez-Barrientos C, Liu XC, Lyon R, Tassone C, Thometz J, 1054
970 jects with cerebral palsy. J Pediatr Orthop 1999;19:470-474. Tarima S. Correlation between anatomic foot and ankle movement 1055
971 10. Liu XC, Embrey D, Tassone C, et al. Foot and ankle joint move- measured with MRI and with a motion analysis system. Gait Posture 1056
972 ments inside orthoses for children with spastic CP. J Orthop Res 2012;36:389-393. 1057
973 1058
974 2014;32:531-536. 21. Kobayashi K, Gransberg L, Knutsson E, Nolen P. A new system for 1059
975 11. Morris C, Bowers R, Ross K, Stevens P, Phillips D. Orthosis man- three-dimensional gait recording using electromagnetic tracking. 1060
976 agement of cerebral palsy: Recommendations from a consensus Gait Posture 1997;6:63-75. 1061
977 conference. NeuroRehabilitation 2011;28:37-46. 22. Nester C, Jones RK, Liu A, et al. Foot kinematics during walking 1062
978 1063
979 12. Hainsworth F, Harrison MJ, Sheldon TA, Roussounis SH. A pre- measured using bone and surface mounted markers. J Biomech 1064
980 liminary evaluation of ankle orthoses in the management of chil- 2007;40:3412-3423. 1065
981 dren with cerebral palsy. Dev Med Child Neurol 1997;39:243-247. 23. Romkes J, Brunner R. Comparison of a dynamic and a hinged ankle- 1066
982 13. Johnson DC, Damiano DL, Abel MF. The evolution of gait in foot orthosis by gait analysis in patients with hemiplegic cerebral 1067
983 1068
984 childhood and adolescent cerebral palsy. J Pediatr Orthop 1997; palsy. Gait Posture 2002;15:18-24. 1069
985 17:392-396. 24. Dalvand H, Dehghan L, Feizi A, et al. The impacts of hinged and 1070
986 14. Desloover K, Molenaers G, Van Gestel L, et al. How can push-off solid ankle-foot orthoses on standing and walking in children with 1071
987 be preserved during use of an ankle foot orthosis in children spastic diplegia. Iran J Child Neurol 2013;7:12-19. Q7 1072
988 1073
989 with hemiplegia? A prospective controlled study. Gait Posture 25. Ballaz L, Plamondon S, Lemay M. Ankle range of motion is key to 1074
990 2006;24:142-151. gait efficiency in adolescents with cerebral palsy. Clin Biomech 1075
991 15. Oeffinger D, Brauch B, Cranfill S, et al. Comparison of gait with and (Bristol, Avon) 2010;25:944-948. 1076
992 without shoes in children. Gait Posture 1999;9:95-100. 26. Buckon CE, Thomas SS, Jakobson-Huston S, Sussman M, Aiona M. 1077
993 1078
994 16. Wolf S, Simon J, Patikas D, Schuster W, Armbrust P, Döderlein L. Comparison of three ankle-foot orthosis configurations for children 1079
995 Foot motion in children shoes: A comparison of barefoot walking with spastic hemiplegia. Dev Med Child Neurol 2001;43:371-378. 1080
996 with shod walking in conventional and flexible shoes. Gait Posture 27. Kerkum YL, Buizer AI, van den Noort JC, Becher JG, Harlaar J, 1081
997 2008;27:51-59. Brehm MA. The effects of varying ankle foot orthosis stiffness on 1082
998 1083
999 17. Perry J, Burnfield J. Gait Analysis: Normal and Pathological Func- gait in children with spastic cerebral palsy who walk with exces- 1084
1000 tion. Thorofare, NJ: Slack Incorporated; 1992. sive knee flexion. PLoS One 2015;10:e0142878. 1085
1001 1086
1002 1087
1003 1088
1004 1089
1005
Disclosure 1090
1006 1091
1007 1092
X.-C.L. Center for Motion Analysis at Orthopedic Surgery, Medical College of R.L. Center for Motion Analysis at Orthopedic Surgery, Medical College of
1008 1093
1009 Wisconsin, Milwaukee; and Department of Orthopaedic Surgery, Children’s Wisconsin, Milwaukee; and Department of Orthopaedic Surgery, Children’s 1094
1010 Hospital of Wisconsin, Milwaukee, WI. Address correspondence to: X.-C.L., Hospital of Wisconsin, Milwaukee, WI 1095
1011 Pediatric Orthopedics, Suite C360, 9000 W Wisconsin Avenue, PO Box 1997, Disclosure: nothing to disclose 1096
1012 Milwaukee, WI 53201; e-mail: xcliu@mcw.edu 1097
1013 Q2 Disclosure: nothing to disclose K.G. Center for Motion Analysis at Orthopedic Surgery, Medical College of 1098
1014 Wisconsin, Milwaukee; and Physical Medicine and Rehabilitation, Children’s 1099
1015 1100
D.E. Children’s Therapy Unit, MulitCare Good Samaritan Hospital, Puyallup, WA Hospital of Wisconsin, Milwaukee, WI
1016 1101
1017 Disclosure: nothing to disclose Disclosure: nothing to disclose 1102
1018 1103
1019 C.T. Center for Motion Analysis at Orthopedic Surgery, Medical College of S.T. Division of Biostatistics, Medical College of Wisconsin, Milwaukee, WI 1104
1020 Wisconsin, Milwaukee; and Department of Orthopaedic Surgery, Children’s Disclosure: nothing to disclose 1105
1021 Hospital of Wisconsin, Milwaukee, WI 1106
1022 Disclosure: nothing to disclose 1107
J.T. Center for Motion Analysis at Orthopedic Surgery, Medical College of
1023 1108
1024 Wisconsin, Milwaukee; and Department of Orthopaedic Surgery, Children’s 1109
1025 K.Z. Center for Motion Analysis at Orthopedic Surgery, Medical College of Hospital of Wisconsin, Milwaukee, WI 1110
1026 Wisconsin, Milwaukee; and Physical Medicine and Rehabilitation, Children’s Disclosure: nothing to disclose 1111
1027 Hospital of Wisconsin, Milwaukee, WI 1112
1028 The National Institute on Disability and Rehabilitation Research (H133G 1113
Disclosure: nothing to disclose
1029 G060155) funded this study. 1114
1030 1115
B.B. Children’s Therapy Unit, MulitCare Good Samaritan Hospital, Puyallup, WA Submitted for publication January 11, 2017; accepted August 19, 2017.
1031 1116
1032 Disclosure: nothing to disclose 1117
1033 1118
1034 1119
1035 1120
1036 1121
1037 1122
1038 1123
1039 1124
1040 1125
1041 1126
1042 1127
1043 1128
1044 1129
1045 1130

FLA 5.5.0 DTD
PMRJ1973_proof
26 September 2017
5:10 pm
ce