RELATIONSHIP BETWEEN FUNCTION AND PRESENCE AND TYPE OF DEFORMITY IN PATIENTS WITH CEREBRAL PALSY

Objective: To evaluate the gross motor function (GMFCS) with respect to the prevalence and type of scoliosis in patients with cerebral palsy (CP). Methods: This was an analytical, cross-sectional study. We evaluated medical records and imaging studies of 100 patients randomly assigned to a specialist rehabilitation center for the care of such patients. The patients were classified according the gross motor function (GMFCS) and those with deformities were classified as per the kind of scoliosis through the classification of Lonstein and Akbarnia). A correlation was made among the presence of deformity, the variables of the type of deformity and motor function by GMFCS. Results: Of the 100 patients evaluated, 69 had scoliosis. The mean age of patients with scoliosis was higher than that of patients without deformity (12.63 and 10.46 years). Thirty-nine (57%) patients had spastic tetraparesis and 32 (46%) spastic diparesis. The most frequent curve pattern was the thoracolumbar and the average angular value of the main curve was 27 degrees. There was a positive correlation between the presence of scoliosis and GMFCS level V. There was also a positive correlation between the Lonstein Group II and GMFCS V. Conclusion: There is a positive correlation between the presence of scoliosis and greater involvement of gross motor function (GMFCS V). In patients with deformities, there is also a positive correlation between the Group II of Lonstein and GMFCS V.


INTRODUCTION
The patient with cerebral palsy (CP) has a high risk of presenting scoliosis. 1Its prevalence varies from 15 to 80%, depending on the definition of scoliosis used, the age, and the severity of the neurological involvement. 1,2The degree of neurological involvement is usually classified according to the type of lesion (spasticity, dyskinesia, and ataxia) and the location (hemiplegia, diplegia, or tetraplegia). 1,3][7] The type of scoliosis found in patients with CP also varies according the presentation of the disease.The most used classification is that of Lonstein and Akbarnia, with Group I being more prevalent in ambulatory patients and Group II in nonambulatory patients. 8There is no study in the literature that considers both the type of deformity and the GMFCS motor function.
The objective of this study is to evaluate motor function in relation to the prevalence and the type of deformity according to the Lonstein classification.

METHOD
This is a cross-sectional analytical study.Following approval by the Institutional Review Board as number 52871115.1.0000.0023, the medical records and exams of 100 patients (Table 1) who were in outpatient follow-up in a rehabilitation center specializing in the care of this type of patient (CRER -Centro de Reabilitação e Readaptação Dr. Henrique Santillo -Goiânia -GO) were evaluated.The patients were chosen randomly from all the patients with a primary diagnosis of CP in the electronic medical records system.All the patients were classified by level of locomotion and motor function using the GMFCS 4 in consultations with the physiatrist and the neurologist.(Table 2) All the patients had full spine radiographs in anteroposterior and lateral views that were taken as scoliosis screening exams during routine evaluation by the physiatric team.The radiographs were taken with ambulatory patients standing and with nonambulatory patients sitting.The radiographic evaluation assessed the presence or absence of scoliosis (deformity in the coronal plane greater than 10 degrees by the Cobb method), the presence of pelvic obliquity (PO), the type of curve, and the presence or absence of trunk imbalance in the sagittal plane.The trunk was considered in balance when a plumb line from C7 fell on the upper plateau of S1, anteriorly decompensated when it fell in front, and posteriorly decompensated when it fell behind. 9he patients with scoliosis were also classified during the evaluation according to the Lonstein and Akbarnia classification system 8 as shown in Figure 1.
A logistic regression test was performed to evaluate the correlation between the presence or absence of scoliosis and the different variables and the chi square test was used to assess the relationship between the Lonstein classification and the GMFCS level (SPSS for Windows, version 15.0).The significance level was set at 5% and values of p-value less than 0.05 were considered to be statistically significant.

RESULTS
Table 1 displays the demographic data of the patients.Of the 100 patients evaluated, 61 were male and 39 were female.The average age was 11.96 (SD 4.68) years, ranging from 4 to 36.The prevalence of scoliosis was 69%.Of the 69 patients with deformity, 43 (62.3%) were male and their average age was 12.63 (SD 5.12), statistically higher than the group without deformity (p<0.05).
In terms of motor function, most of patients with scoliosis were classified as levels IV and V according to the GMFCS, with a statistically significant correlation with type V (p=0.04,OR 1.54).(Table 3) In the radiographic evaluation, the most frequent types of curves were thoracolumbar (24 out of 69 cases) and lumbar (18 out of 69 cases).(Figure 2) The average angle of the principal curve was 27 degrees, ranging from 10 to 77 degrees.PO was present in 75% (52 out of 69) of the patients with scoliosis, with an average angular value of 8.4 degrees (minimum of 3 and maximum of 21).In the sagittal plane, 30 (43%) patients presented anterior trunk imbalance, 18 (26.1%)presented posterior trunk imbalance, and 21 (30.4%) were in balance.
As for the type of deformity, 40 patients were in group 1A, 13 in group IB, two in group IIA, and 14 in group IIB.Table 4 shows that there was a statistical correlation between group II and GMFCS motor function level V. Fifteen of the 16 patients with group II deformities presented GMFCS IV or V. (Table 4) Table 2. Classification of Gross Motor Function (GMFCS).

Level I
Walks without restrictions, with limitations for running and jumping.
Level II Walks with assistance from small appliances and/or crutches, with slight community ambulation limitations.

Level III
Walks with the assistance of walker and/or crutches, with community ambulation difficulties.

Level IV
Walks with the assistance of walker but with limitations and requires a wheelchair for community ambulation.

Level V
Severely limited mobility, even with appliances and adaptations, with wheelchair adaptations required.

DISCUSSION
1][12] Most studies correlate the presence of scoliosis with the degree of neurological involvement 2,[5][6][7] and few correlate the deformity with any type of functional classification such as the GMFCS. 1 In addition, no study correlates the type of deformity to motor function.
Perrson-Bunke et al. 1 evaluated a population in southern Sweden and observed an incidence of scoliosis of 192 cases among 666 children with CP.This screening was performed based on a physical examination of the patients, which could justify the much lower incidence than that found in our population.Another factor is that our case series consists of patients who are undergoing outpatient follow-up in an institution specializing in the treatment of this type of pathology, while the study in question is based on a register of the population.Radiographs were performed in 76 of these patients, in which predominantly thoracolumbar curves were observed, most of them between 21 and 40 degrees, similar to those found in our population.This same study 1 showed that there is a positive correlation between the level of involvement according to the GMFCS and the presence of deformity that is more significant than the type and topography of the CP involvement.In fact, the higher prevalence of scoliosis among the patients with spastic tetraparesis or dyskinesia is justified by the high prevalence of patients with GMFCS types IV and V in these CP subtypes.These findings are in agreement with those found in our case series.However, this study did not take the presence of PO or the type of deformity into account and we know that they are important factors in the reduction of function in patients with CP. 13 Lonstein and Akbarnia 8 described the curve patterns affecting patients with CP and mental retardation in 1983.In their original study, they observed a higher prevalence of group I deformities (balanced double curves) among patients with mental retardation and of group II deformities (imbalanced lumbar or thoracolumbar curves) among nonambulatory patients and those with PO.Our study was the first to correlate the type of curve with motor function and, just as with the presence or absence of deformity, there was a positive correlation between group II deformities and patients classified as GMFCS V.
Loeters et al. 14 conducted a literature review on the risk factors for the progression of scoliosis in patients with CP.Based on 10 studies, they failed to demonstrate any specific risk factor with a high level of evidence.They only suggested a possible relationship to the severity of the pathology.They concluded their review, stating that it was very important to conduct new studies based on some functional scale, such as the GMFCS.

CONCLUSION
The CP patients at GMFCS IV and V had a higher risk of presenting scoliosis.In addition, these patients had a higher incidence of Lonstein group II deformities with PO.
All the authors declare that there are no potential conflicts of interest regarding this article.

Figure 1 .
Figure 1.Lonstein and Akbarnia Classification.Group I is characterized by a compensated trunk (IA double curve and IB thoracic) and Group by a decompensated trunk with pelvic obliquity (IIA fractionated curve above the sacrum and IIB with the sacrum making up part of the principal curve).

Figure 2 .
Figure2.Example of a patient with diparetic spastic CP, with lumbar scoliosis (apex L3) on the right and PO with elevation of the left hemipelvis.This is an example of a type IIA curve since there is a fractionated lumbosacral curve with the pelvis not being part of the curve.This patient underwent surgical treatment with correction of the curve and arthrodesis from T4 to the ilium.

Table 4 .
Correlation between the type of deformity and the GMFCS motor function.

Table 1 .
Demographic data, presence or absence of scoliosis, degree of neurological involvement, and function according to the GMFCS.

Table 3 .
Relationship between gross motor function and the presence or absence of deformity.