Femoral Head Coverage Assessment in Healthy Children Younger than 6 Years

Introduction Developmental dysplasia of the hip (DDH) is one of the commonest hip joint pathologies in children; to treat it properly, hip surgeons should know the normal femoral head (FH) coverage by the acetabulum. This paper aims to assess the femoral head coverage in healthy children younger than 6 years. Methods 270 hip joint CT scans were selected, and digital pelvic models were created according to these scans. FH coverage by the five acetabular regions was assessed according to patient's age and sex. Results Normal reference values of FH coverage by different acetabular regions were obtained. It was found that the growth process of different acetabular regions occurs nonlinearly with the periods of acceleration. Anterior and superior-anterior acetabular regions grow more intensively in boys up to 3 years old and between 4 and 5 years old both in boys and girls; superior-posterior, posterior-superior, and posterior-inferior acetabular regions grow more intensively in boys and girls up to 3 years old and between 4 and 5 years old (p ≤ 0.005). The following sex differences in FH coverage by the acetabulum were found: more superior-anterior FH coverage was found in boys and posterior FH coverage in girls (p ≤ 0.005).


Introduction
Developmental dysplasia of the hip (DDH) is one of the commonest hip joint pathologies in pediatric orthopedist practice [1]; without treatment, it leads to early hip arthritis [2]. In younger patients, treatment is usually nonsurgical, but in older patients (who began to walk), surgical treatment is preferable [3]. Among all surgical options, the best results were obtained after pelvic osteotomies [4][5][6]. In patients under 6 years of age (when the most active acetabular development occurs [7,8]), three pelvic osteotomies are widely used-Salter, Pemberton, and Dega osteotomies.
ese osteotomies may improve both anterior-superior or posterior-superior femoral head (FH) coverage [9,10]. To apply these pelvic osteotomies more effectively, it is necessary to understand the normal acetabulum morphology and its maturation process in healthy children of 1-6 years old.
Previously, acetabulum morphology was evaluated by other authors, but in adults [11][12][13][14][15][16][17][18][19][20][21][22], the youngest age in these studies was 8 years. [16]. ere are no studies dedicated to evaluating acetabulum morphology in children under 6 years. As we know this study is the first one, the reliable way to assess acetabular morphology is to assess the FH coverage by the acetabulum as described by others [12,13,16]. e goal of this study is to assess the femoral head coverage by the acetabulum in healthy children younger aged 1-6 years. joint pathologies; all of them underwent abdominal and pelvic CT scanning to detect neoplasms metastases. ese pelvic CT scans with a slice thickness <1.5 mm (for better visualization of pelvic bones) were selected for further investigation. e scanning was performed on "Siemens SOMATOM Definition AS, USA." Totally, 270 CT scans of 135 patients were selected for further work: 156 hip joint CT scans of 78 male patients and 114 hip joint CT scans of 57 female patients.
Selected CT scans were transported into Mimics 20.0 software (Materialize Inc., Leuven, Belgium), and automatic segmentation of pelvic bones was performed. Digital pelvic models were exported from Mimics software in STL format. Furthermore, these models were transported into custommade software for the FH coverage assessment. FH coverage was assessed by fitting the virtual sphere into the acetabulum with the least-squares method. en, the contact between the virtual sphere and acetabular borders was highlighted (with a contact spot). After that, the reference line was drawn through the centers of both spheres. Finally, intersection angles from the sphere's center between the reference line and acetabular edges were measured circumferentially. e measurement step was 1°, so each virtual sphere had 360 contact points with the acetabular edge (both with pelvic bones and triradiate cartilage limbs).
us, 360 angles were measured. is data was represented both as numerical values and graphically displayed in the chart. e numerical values were exported to Microsoft Corporation (2007) and Microsoft Excel for further analysis. e pelvic spatial alignment (as described by others [12,13]) was not performed. is was not necessary because pelvic bones in children are separated with triradial cartilage limbs that allow identifying each pelvic bone in the acetabulum without any reference points. e process of FH coverage assessment is shown in Figure 1.
e FH coverage was assessed by the pubis, ilium, and ischium bones (anterior, superior, and posterior acetabular coverage, respectively). For a more detailed assessment, the regions of superior and posterior acetabulum coverage were further divided into the superior-anterior, superior-posterior, posterior-superior, and posterior-inferior regions. us, FH coverage by the 5 acetabular regions (anterior, superior-anterior, superior-posterior, posterior-superior, and posterior-inferior) was evaluated. FH coverage by different acetabular regions is presented in Figure 2.
Before the assessment of the FH coverage in selected children, we had the following learning curve. e FH coverage in other 70 children (140 hip joints) of both genders 1-6 years old without hip joint pathologies was assessed.
is practice taught us how to overcome with more shallow acetabulum in younger children while fitting in it the virtual  sphere. Also, after this training, we have decided to select only CT scans with slice thickness <1.5 mm for better pelvic bone visualization for further FH coverage assessment.

Statistical
Analysis. Data were normalized as follows: 360 numerical values (intersection angles values) were entered for each child in Microsoft Corporation (2007), Microsoft Excel. en, all children and the abovementioned numerical values were sorted according to their sex (male and female children separately) and age (from 1 to 6 years old). Furthermore, these numerical values were divided according to the FH coverage by an anterior acetabular region (covered by pubis bone), superior region (ilium bone), and posterior region (ischium bone). e numerical values of superior and posterior FH coverage were divided into two parts, which reflected the superior-anterior, superior-posterior, posterior-superior, and posterior-inferior FH coverage. e arithmetic means in each of the five FH coverage regions were calculated according to the child's sex and age. ese arithmetic mean values presented themselves as the raw data for further statistical analysis.
To determine normal FH coverage reference values, arithmetic mean values of the FH coverage by different acetabular regions were normally distributed according to children's age and sex. e mean, maximal, minimum values and standard deviation were calculated for each group.
To determine the development intensity in acetabular regions, the mean values of the FH coverage by the same acetabular region in children of the adjacent age and the same sex were compared. e null hypothesis (Ho) was the absence of intensive development of certain acetabular regions in a certain age period. An alternative hypothesis (Ha) was the presence of such intensive development. e hypothesis was tested with Student's t-test. e level of significance (α) was set at 5%. At a "p" value <0.05, the development of a certain acetabular region in a certain age period was considered intensive.
Sex differences in the FH coverage by the same acetabular region in children of the same age were assessed similarly. e null hypothesis (Ho) was the absence of sex difference in the FH coverage by the same acetabular region in male and female children of the same age. An alternative hypothesis (Ha) was the presence of such a difference. e hypothesis was tested with Student's t-test. e level of significance (α) was set at 5%. At a "p" value <0.05, the sex difference in the FH coverage by the same acetabular region in male and female children of the same age was considered significant.
To assess the reproducibility of our measurements the intrarater correlation was calculated by the same rater (Suvorov Vasyl). FH coverage by different acetabular regions was assessed according to the method described above (in the "Methods" section). 4 months after the initial assessment, an intrarater correlation was calculated using intraclass correlation (ICC). e reproducibility was considered poor with an ICC value <0.50; moderate-with 0.50 ≤ ICC < 0.75; good-with 0.75 ≤ ICC < 0.90; and excellent-with ICC ≥0.90.

Results
Normal reference values of FH coverage by different acetabular regions in children according to their age and sex are shown in Table 1.  According to the growth of different acetabular regions it was revealed that the anterior region grows more intensively in male children between 1-2 and 4-5 years (p ≤ 0.05); in female children, it grows rapidly between 1 and 2 years, then there is retardation up to 3 years of age (p ≤ 0.05). e superior-anterior acetabular region grows intensively in male children between 2-3 and 4-5 years and in female children between 4 and 5 years (p ≤ 0.05). e superiorposterior acetabular region grows intensively in male children up to 3 y.o. and between 4 and 5 years, in female children up to 2 y.o. and between 4 and 5 years (p ≤ 0.05). e posterior-superior acetabular region grows intensively in male and female children up to 3 y.o. and between 4 and 5 years (p ≤ 0.05). e posterior-inferior acetabular region grows intensively in male and female children for up to 3 y.o.
(p ≤ 0.05). e comparison of FH coverage by different acetabular regions in male and female children according to their age is shown in Table 2.
Regarding the sex differences in FH coverage in male and female children, it was found that superior-anterior FH coverage is more in male children at 3 y.o. (p ≤ 0.05). Contrarily, it was revealed that the posterior-superior FH coverage is more in girls at 2 and 6 y.o. (p ≤ 0.05); the posterior-inferior FH coverage is more in girls at 2, 4, and 5 y.o. (p ≤ 0.05). Sex differences in FH coverage in children according to their age are shown in Table 3.
According to the reproducibility of our measurements, the moderate-to-good intrarater agreement was seen in all cases. e table of intraclass correlation according to the patient's age and FH coverage region is presented in Table 4.

Discussion
is study aimed to investigate the femoral head coverage by the five acetabular regions in healthy children of 1-6 years old. In this study, FH coverage was assessed from the standpoint of pelvic osteotomies application for DDH treatment, as this pathology is the commonest hip joint pathology in children 1-6 years old, which requires pelvic osteotomies. It is important for the hip surgeon to know the normal reference values of the FH coverage, as this allows for differentiating pathological conditions from the normal extremes and planning hip joint reconstructive surgeries. However, if needed, the results of this study are also applicable for other pathologies that involve hip joint in children of 1-6 years old (cerebral palsy, multiple epiphyseal dysplasia, and secondary acetabular dysplasia in Legg-Calve-Perthes disease). e method of FH coverage assessment, described in this article, allows hip surgeons to appreciate acetabular morphology indirectly and to compare it with reference values.
ese reference values will be useful when determining the type of acetabular deficiency and for better preoperative planning; however, pelvic CT scans are needed for this.
Analyzing the acetabulum maturation process, we have found out that the most intensive acetabular growth occurs during the first 5 years of life, which is consistent with the results of Li, L. Y., and Novais [7,8,17]. is justifies the early application of reconstructive pelvic osteotomies for DDH treatment. Also, this may explain better results after pelvic osteotomies in younger patients compared to older ones [23][24][25][26][27][28].
Regarding sex differences in the FH coverage, we have revealed that posterior FH coverage is more in female children and anterior FH coverage is more in male children. Our findings are in concordance with the results of other authors about gender differences in hip joint morphology in adults [12,13,16].
is may raise the question if these differences arise from childhood. e drawbacks of this work are as follows: (1) only the boney part of the acetabulum was included in this study as CT scans were evaluated; this does not allow for fully appreciating FH coverage in children due to a large amount of chondral tissue in the acetabulum (that is not seen on CT scans). us, MRI studies are preferable to CT scans for the assessment of FH coverage in children. However, CT scans instead of MRI are widely used in children for different abdominal and pelvic cavities pathologies, so they were chosen as research objects in this study. (2) e method of FH coverage assessment described in this article is sophisticated. It may be inconvenient for routine application as pelvic CT scans and reference data of the FH coverage by different acetabular regions are needed for it. Moreover, the results and reference values described in this article may be different in other countries (because of possible differences in pelvic morphology). (3) e validation of our technique was not performed.

Conclusion
After the assessment of the FH coverage by different acetabular regions, normal reference values of this coverage were obtained. ese reference values may be used by hip surgeons during preoperative planning of pelvic osteotomies for DDH treatment (or for other hip joint pathologies) in patients of 1-6 years old. Evaluating the maturation process of different acetabular regions, it was found that it occurs nonlinearly with periods of acceleration (up to 3 y.o. and between 4 and 5 y.o.); this also should be taken into account during preoperative planning (to prevent overcorrection in those regions that still growth). Assessing sex differences in   Advances in Orthopedics 5 the FH coverage, more intensive superior-anterior coverage in male children and posterior coverage in female ones were found. is information should be considered by the hip surgeon during routine preoperative planning without CT scans (to prevent overcorrecting FH coverage in those regions that better grow).

Data Availability
Pelvic CT scans, digital pelvic models, and software for femoral head coverage used to support the findings of this study are available from the corresponding author upon request. Other data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.