ORIGINAL ARTICLE

Knee Pain and Patellofemoral Symptoms in Patients With Cerebral Palsy Susan A. Rethlefsen, PT, DPT,* Danny T. Nguyen, BA,* Tishya A. L. Wren, PhD,*w Matthew D. Milewski, MD,z and Robert M. Kay, MD*w

Background: Knee pain in cerebral palsy (CP) is associated with increased patellofemoral forces present when walking with flexed knees. In typically developing children, knee pain and patellofemoral dysfunction are associated with obesity, genu valgum, femoral anteversion, and external tibial torsion. These problems are also common in CP, and may contribute to knee problems in this population. The purposes of this study were to define the prevalence of knee pain and patellofemoral dysfunction in children with CP, and to identify physical and gait characteristics (using 3-dimensional gait analysis data) that predispose them to such problems. Methods: Retrospective review of 121 children with CP, Gross Motor Function Classification System level I to IV, who underwent computerized gait analysis testing. Demographics, range of motion, body mass index and hip, knee, and ankle kinematics were compared between subjects with and without knee pain. Results: Twenty-five of 121 subjects (21%) reported knee pain at the time of testing. Three of 121 subjects (2%) had a history of patellar subluxation/dislocation. Age and sex were significantly related to presence of knee pain. The likelihood of knee pain was almost 5 times higher in females (odds ratio = 4.9, [95% confidence interval, 1.8-13.3], P = 0.002), with a prevalence of 40% (17/42) in females versus 10% (8/79) in males. The likelihood of knee pain increased with age by approximately 13% per year (odds ratio = 1.13, [95% confidence interval, 1.00-1.28], P = 0.058). Malignant malalignment syndrome showed a potential relationship to more severe knee pain (P = 0.05), which warrants further investigation. Body mass index, pes valgus, and degree of stance knee flexion showed no statistically significant relationships to knee pain (P > 0.16). Conclusions: The prevalence of knee pain in ambulatory patients with CP is approximately 21%. Patellar subluxation (2%) and dislocation are rare in these patients. Knee pain is not always related to crouch, femoral anteversion, external tibial torsion, From the *Children’s Orthopaedic Center, Children’s Hospital; wDepartment of Orthopaedic Surgery, Keck School of Medicine of USC, Los Angeles, CA; and zElite Sports Medicine, Connecticut Children’s Medical Center, Farmington, CT. Study was conducted at the Children’s Hospital Los Angeles, Los Angeles, California. The authors declare no conflicts of interest. Reprints: Susan A. Rethlefsen, PT, DPT, Children’s Orthopaedic Center, Children’s Hospital, 4650 Sunset Blvd., M/S 69, Los Angeles, CA 90027. E-mail: [email protected]. Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.

J Pediatr Orthop



genu valgum, or pes valgus. Knee pain in these patients is more prevalent in females, and increases with increasing age. Level of Evidence: Level III—case-control study. Key Words: patellofemoral dysfunction, knee pain, gait, cerebral palsy (J Pediatr Orthop 2015;35:519–522)

P

ain in children with cerebral palsy (CP) is a problem that negatively impacts quality of life.1 Bodily pain in these patients persists into adulthood and increases in prevalence with age.1–3 Prevalence of knee pain in adolescents with CP has been reported at 26% in ambulatory patients at all functional ability levels.4 Although anterior knee pain in patients with CP has been well documented, its causes have not been determined conclusively.5,6 Anterior knee pain is often associated with excessive knee flexion during the stance phase of gait, due to increased patellofemoral joint forces present when walking with a knee flexed posture.7 Senaran et al5 studied a group of patients with intractable anterior knee pain, and found that all had flexed knee gait and positive radiographic abnormalities including patella alta, inferior pole fracture, and patellar subluxation/dislocation. Surgical and nonsurgical treatment of knee flexion significantly reduced anterior knee pain. A study by Rodda et al8 showed that anterior knee pain in patients with CP resolves after surgical correction of flexed knee gait, despite persistence of patella alta. Using cine phase contrast magnetic resonance imaging, Sheehan and colleagues found abnormal patellar kinematics in subjects with CP who have knee pain (more superior, posterior, extension, and valgus orientation) as compared with their pain-free peers when performing non–weight-bearing active, cyclical knee flexion/extension motions. Their data suggest that anterior knee pain in these patients is likely due to abnormal forces acting between the patella and femur due to abnormal kinematics.9 Various factors contribute to anterior knee pain and patellofemoral dysfunction in the nondisabled population. These include obesity, genu valgum (quadriceps angle), femoral anteversion, and external tibial torsion.10–13 As these problems are often present in patients with CP, they may contribute to anterior knee pain and patellofemoral dysfunction in this population as well.

Volume 35, Number 5, July/August 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

www.pedorthopaedics.com |

519

J Pediatr Orthop

Rethlefsen et al

The purpose of this study was to define the prevalence of knee pain and patellofemoral dysfunction in children with CP, and to identify physical or gait characteristics that predispose them to such problems.

METHODS Institutional review board approval was obtained before initiation of the study. Retrospective review of medical records and gait studies was conducted for a consecutive sample of 121 children with CP, Gross Motor Function Classification System (GMFCS) level I to IV, who were seen at the Motion Analysis Laboratory between October 2011 and August 2013. Of the 121 patients, 97 exhibited bilateral involvement, and 24 had unilateral involvement. The distribution of GMFCS levels was 35 patients in GMFCS I, 47 GMFCS II, 31 GMFCS III, and 8 GMFCS IV. All subjects had undergone gait analysis testing at our laboratory. Testing included a history (including pain assessment), a complete clinical examination, and collection of 3-dimensional kinematic (joint motion) data. Passive joint range of motion (ROM) was measured by an experienced physical therapist using a goniometer, according to standardized procedures. Malignant malalignment syndrome14 was considered present if there was excessive internal hip rotation with excessive external foot progression (> 1 SD more than mean for able-bodied peers). Pes valgus was considered present if there was hindfoot and/or forefoot valgus/pronation or midfoot break in stance or swing, as identified visually by the examining physical therapist. Three-dimensional kinematic data were acquired using an 8-camera VICON 3-dimensional motion analysis system (Vicon Motion Systems, Los Angeles, CA). A set of 15 to 19 passive retroreflective markers was attached over specific bony landmarks of the pelvis and lower extremities and the participants made several passes down a 15-m path with markers in place. Data from multiple trials were processed and averaged together. Kinematic variables potentially related to patellofemoral pain and/or dysfunction were exported for analysis. These included maximum knee extension in stance, maximum knee flexion in stance, average knee flexion/extension in stance, average knee varus/valgus in stance, average hip rotation in stance, and average foot progression in stance.



Volume 35, Number 5, July/August 2015

Data were compared statistically between subjects with knee pain (Pain group) and those with no knee pain (NoPain group). One side per subject was included in the statistical analysis (either the left side or the painful side for bilaterally involved patients, and the involved side for those with unilateral symptoms; none of the unilaterally involved patients reported pain on the contralateral side). Data were compared between groups using Fisher exact tests for categorical variables and t tests for continuous variables. Variables showing statistical significance were then included in a multivariate logistic regression analysis. The significance level was set at P < 0.05.

RESULTS Twenty-five of 121 subjects (21%) reported experiencing knee pain at the time of the gait analysis. Severity of pain was rated on a 0 to 10 scale by the 20 subjects who were able to communicate such information. Severity averaged 5 ± 2 (range, 2 to 7; median = 5; mode = 6). Subjects reported having knee pain an average of 3 ± 7 times per week (range, 0.02 to 46 times per week). Subjects reported their pain interfered with their ability to sleep [4/25 (16%)] and to go up and down stairs [14/25 (56%)]. Although only 2/25 subjects (8%) reported having increased knee pain when sitting with the knees bent, 7/25 (28%) reported that their pain was relieved by straightening the knees. Six of 25 (24%) reported experiencing painful kneecap movements at some point in time. Three of 121 subjects (2%) had a history of patellar subluxation/dislocation (recurrent dislocations in 2 subjects) and 1 subject went on to have medial patellofemoral ligament reconstruction surgery. Comparisons between the Pain and NoPain groups are detailed in Tables 1, 3, and 4. The Pain group was significantly older and had a higher proportion of females than the NoPain group (Table 1). Body mass index percentile was similar between groups, indicating that obesity was not related to knee pain. GMFCS level also was not related to knee pain. The frequency of pain reported by patients at each GMFCS level is detailed in Table 2. Femoral anteversion (as measured by the trochanteric prominence angle test and passive hip internal and external rotation ROM), external tibial torsion (as measured by the thigh-foot angle), and presence of pes

TABLE 1. Summary of Demographic Characteristics and Results of Univariate Analysis Demographics

No Knee Pain (N = 96)

Age (y) [mean ± SD (range)] Sex [N (%)] Male Female Body mass index percentile [mean ± SD (range)] GMFCS [N (%)] I II III IV

10.5 ± 3.7 (4-19) 71 (74) 25 (26) 54 ± 35 (0-99.8) 27 37 24 8

Knee Pain (N = 25)

P

13 ± 4.1 (5-20)

0.004 < 0.001

8 (32) 17 (68) 59 ± 30 (7-96)

(28) (38) (25) (8)

0.51 0.60

8 (32) 10 (40) 7 (28) 0

Continuous variables are presented as mean ± SD (range). Categorical variables are presented as N (%).

520 | www.pedorthopaedics.com

Copyright

r

2014 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

J Pediatr Orthop



Volume 35, Number 5, July/August 2015

TABLE 2. Frequency of Report of Knee Pain by GMFCS Level GMFCS Levels

Frequency of Knee Pain

I II III IV

23% 21% 23% 0%

(8/35) (10/47) (7/31) (0/8)

valgus were not related to presence of knee pain (Table 3). Malignant malalignment syndrome was more common in the Pain group compared with the NoPain group, but this difference was not statistically significant. None of the kinematic variables (minimum, maximum and average knee flexion in stance, knee valgus, hip rotation, and foot progression) were different between groups (Table 4). Data analysis was repeated dividing the Pain group into subgroups with mild pain (ratings of 1 to 5/10) and severe pain (6 to 10/10). Age (P = 0.001) and sex (Pr0.0001) remained statistically significantly related to knee pain regardless of its severity. Presence of malignant malalignment syndrome showed a nearly statistically significant association with knee pain severity (P = 0.05). Age and sex were entered into the multivariate logistic regression analysis. Age and sex were both related to the presence of knee pain independent of each other. The likelihood of having knee pain was almost 5 times higher in females than in males (odds ratio = 4.9, [95% confidence interval, 1.8-13.3], P = 0.002), with a prevalence of 40% (17/42) in females compared with 10% (8/79) in males. The likelihood of having knee pain also increased with age by approximately 13% per year (odds ratio = 1.13, [95% confidence interval, 1.00-1.28], P = 0.058).

DISCUSSION The prevalence of knee pain in the current study was 21%, which is similar to previous reports.4 The prevalence of patellar subluxation and dislocation was very low in our cohort (2%), and only 1 subject required surgical intervention for recurrent patellar dislocation. Interestingly, none of these 3 patients exhibited significant stance knee flexion during gait. The 2 who did not require surgery had malignant malalignment syndrome, pes valgus, external tibial

Knee Pain and Patellofemoral Symptoms in CP

torsion, and femoral anteversion, but the subject who went on to have medial patellofemoral ligament reconstruction had none of these predisposing factors. Only age and sex were related to knee pain in our patients, with females and older children having a higher prevalence. These findings are consistent with other reports of higher rates and severity of pain in girls, and with increasing age.3,4 Our results indicate that girls with neurological disabilities are almost 5 times more likely to have knee pain than boys. It has been reported that in young active healthy females the prevalence of anterior knee pain is probably between 12% and 15% with females being twice as likely as males to develop patellofemoral pain syndrome.15,16 Thus, young females with or without neurological disabilities seem to be at added risk for development of anterior knee pain or patellofemoral symptoms. Our results also indicate that these patients are 1.13 times more likely to have knee pain with each additional year of life. It was surprising that body mass index or obesity was not correlated with increased knee pain but this echoes findings in healthy adolescent females where the incidence of patellofemoral pain was not correlated with relative body mass composition.17 Our data contradict previous reports of patellofemoral dysfunction in subjects with severe stance knee flexion during gait.5,8 We found no difference in either hamstring or knee extension passive ROM or dynamic knee position in the stance phase of gait between subjects with and without knee pain. Follow-up analysis revealed that the subjects who walked with the greatest stance knee flexion were those functioning at GMFCS level IV (average knee flexion in stance 54 vs. 19, 22, and 30 degrees at levels I, II, and III, respectively, P < 0.0001). None of these patients reported knee pain. However, cognitive and/or communicative limitations may have impacted accuracy of reports of pain in these subjects. In addition, patients at GMFCS level IV do not ambulate as much, and do not participate in sports and other activities that place demands on the knee and patellofemoral joints as their more functionally independent peers. Both these factors may explain the lack of knee pain reported by the subjects with the greatest stance knee flexion in the current study. Knee pain was equally prevalent among patients functioning at levels I to III.

TABLE 3. Summary of Static Measures and Results of Univariate Analysis Range of Motion Hip internal rotation (deg.) Hip external rotation (deg.) Femoral anteversion Popliteal angle (deg.) Knee extension (deg.) Thigh-foot angle (deg.) Malignant malalignment syndrome Pes valgus

No Knee Pain (N = 96)

Knee Pain (N = 25)

P

65 ± 15 (30-90) 39 ± 13 (5-65) 36 ± 17 ( 20 to 70) 50 ± 13 (10-85) 5 ± 10 (35-degree hyperextension to 20-degree flexion) 7 external ± 14 (40-degree external to 35-degree internal) 13/96 (14%)

68 ± 10 (45-85) 37 ± 11 (5-55) 36 ± 15 (10-70) 50 ± 12 (30-70)  2 ± 9 (20-degree hyperextension to 15-degree flexion) 6 external ± 12 (30-degree external to 20-degree internal) 7/25 (28%)

0.33 0.44 0.95 0.96 0.12

34/96 (35%)

5/25 (20%)

0.16

0.81 0.12

Continuous variables are presented as mean ± SD (range). Categorical variables are presented as N (%).

Copyright

r

2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

www.pedorthopaedics.com |

521

J Pediatr Orthop

Rethlefsen et al



Volume 35, Number 5, July/August 2015

TABLE 4. Summary of Dynamic Measures Obtained From 3-dimensional Gait Analysis and Results of Univariate Analysis Kinematic Variables Max knee extension in stance (deg.) Max knee flexion in stance (deg.) Average knee flexion/extension in stance (deg.) Average knee varus/valgus in stance (deg.) Average hip rotation in stance (deg.) Average foot progression in stance (deg.)

No Knee Pain (N = 96)

Knee Pain (N = 25)

17 ± 20 (16-degree hyperextension to 95-degree flexion) 45 ± 14 (21-108) 26 ± 18 (4-degree hyperextension to 102-degree flexion)  0.2 valgus ± 3.0 (12-degree valgus to 9-degree varus) 5 internal ± 13 (23-degree external to 38-degree internal) 2 external ± 21 (64-degree external to 59-degree internal)

11 ± 19 (20-degree hyperextension to 55-degree flexion) 41 ± 14 (17-68) 21 ± 18 (12-degree hyperextension to 60-degree flexion) 0.7 varus ± 2.4 (3-degree valgus to 6-degree varus) 9 internal ± 11 (12-degree external to 35-degree internal) 4 external ± 17 (34-degree external to 30-degree internal)

P 0.16 0.20 0.18 0.24 0.15 0.76

Values are presented as mean ± SD (range).

We were unable to determine whether factors associated with patellofemoral pain in able-bodied children (femoral anteversion and external tibial torsion, pes valgus, knee valgus) were causative of knee pain in patients with disabilities. The prevalence of malignant malalignment syndrome was twice as high in the subjects with knee pain as in the pain-free group [particularly patients with severe pain (rated 6 to 10/10)], but the difference did not reach the level of statistical significance. Nevertheless, the data suggest further study of a potential relationship between malignant malalignment syndrome and knee pain may be warranted. As with all retrospective research, the current study was limited by several factors. The sample included only patients being evaluated for possible surgical intervention to improve their gait, and may not be representative of the general population. We did not record the cognitive level of patients, and in some cases pain questions were answered by a parent. Even when answered by the subject him or herself, subjects with cognitive limitations may not accurately identify the location or intensity of their pain. Not all of the subjects received their medical care at this facility. Therefore, we were not able to record surgeries that were performed at other facilities after gait analysis had been done. Finally, lack of available knee radiographs for subjects with and without knee pain limited our ability to ascertain relationships between radiographic abnormalities and knee pain in our subjects. Reports of relationships between radiographic findings and knee pain in patients with CP are mixed.5,8,18 This remains an area for future study.

CONCLUSIONS The prevalence of knee pain in ambulatory patients with CP is approximately 21%. Patellar subluxation (2%) and dislocation are rare in these patients. Knee pain is not always related to flexed knee gait, femoral anteversion, external tibial torsion, genu valgum, or pes valgus. Knee pain is significantly more prevalent in females, and increases with increasing age. Further study is needed to examine radiographic differences between children with disabilities with and without knee pain.

522 | www.pedorthopaedics.com

REFERENCES 1. Dickinson HO, Parkinson KN, Ravens-Sieberer U, et al. Self-reported quality of life of 8-12-year-old children with cerebral palsy: a crosssectional European study. Lancet. 2007;369:2171–2178. 2. Andersson C, Mattsson E. Adults with cerebral palsy: a survey describing problems, needs, and resources, with special emphasis on locomotion. Dev Med Child Neurol. 2001;43:76–82. 3. Jahnsen R, Villien L, Aamodt G, et al. Musculoskeletal pain in adults with cerebral palsy compared with the general population. J Rehabil Med. 2004;36:78–84. 4. Doralp S, Bartlett DJ. The prevalence, distribution, and effect of pain among adolescents with cerebral palsy. Pediatr Phys Ther. 2010;22:26–33. 5. Senaran H, Holden C, Dabney KW, et al. Anterior knee pain in children with cerebral palsy. J Pediatr Orthop. 2007;27:12–16. 6. Rosenthal RK, Levine DB. Fragmentation of the distal pole of the patella in spastic cerebral palsy. J Bone Joint Surg Am. 1977;59: 934–939. 7. Steele KM, Demers MS, Schwartz MH, et al. Compressive tibiofemoral force during crouch gait. Gait Posture. 2012;35:556–560. 8. Rodda JM, Graham HK, Nattrass GR, et al. Correction of severe crouch gait in patients with spastic diplegia with use of multilevel orthopaedic surgery. J Bone Joint Surg Am. 2006;88:2653–2664. 9. Sheehan FT, Babushkina A, Alter KE. Kinematic determinants of anterior knee pain in cerebral palsy: a case-control study. Arch Phys Med Rehabil. 2012;93:1431–1440. 10. Post WR, Teitge R, Amis A. Patellofemoral malalignment: looking beyond the viewbox. Clin Sports Med. 2002;21:521–546, x. 11. Post WR. Clinical evaluation of patients with patellofemoral disorders. Arthroscopy. 1999;15:841–851. 12. Lee TQ, Anzel SH, Bennett KA, et al. The influence of fixed rotational deformities of the femur on the patellofemoral contact pressures in human cadaver knees. Clin Orthop Relat Res. 1994;302: 69–74. 13. Eckhoff DG, Montgomery WK, Kilcoyne RF, et al. Femoral morphometry and anterior knee pain. Clin Orthop Relat Res. 1994; 302:64–68. 14. Gage JR, Schwartz MH. Pathological gait and lever arm dysfunction. In: Gage JR, ed. The Treatment of Gait Problems in Cerebral Palsy. London: Mac Keith Press; 2004:198–200. 15. Roush JR, Curtis Bay R. Prevalence of anterior knee pain in 18-35 year-old females. Int J Sports Phys Ther. 2012;7:396–401. 16. Boling M, Padua D, Marshall S, et al. Gender differences in the incidence and prevalence of patellofemoral pain syndrome. Scand J Med Sci Sports. 2010;20:725–730. 17. Barber Foss KD, Hornsby M, Edwards NM, et al. Is body composition associated with an increased risk of developing anterior knee pain in adolescent female athletes? Phys Sports Med. 2012;40:13–19. 18. Topoleski TA, Kurtz CA, Grogan DP. Radiographic abnormalities and clinical symptoms associated with patella alta in ambulatory children with cerebral palsy. J Pediatr Orthop. 2000;20:636–639.

Copyright

r

2014 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

Knee Pain and Patellofemoral Symptoms in Patients With Cerebral Palsy.

Knee pain in cerebral palsy (CP) is associated with increased patellofemoral forces present when walking with flexed knees. In typically developing ch...
100KB Sizes 0 Downloads 7 Views