Radiographic analysis of the proximal femoral anatomy in the Croatian population

Frequency of total hip arthroplasty (THA), as effective treatment for end stage of hip osteoarthritis, has been increased all over the world [1]. The goal of THA is to achieve painless and functional hip for long period. Many factors influence the longevity of THA, such as implant design, material type, body weight, surgical technique, and hip anatomy reconstruction [2]. Multiple authors in their studies described the effects of incorrectly reconstructed hip anatomy with THA, resulting in patients’ dissatisfaction, leg length inequality, limping, pain, increased material wear, and loosening of hip prosthesis [3], and in many of this cases, revision of total hip replacement is required. Accurate reconstruction of hip anatomy grossly depends on implant design [4]. In order to select an implant in correspondence with the native hip, the proximal femoral morphology has been in focus of many studies in the past years. Studies have shown significant differences in the anatomy of the proximal femur between races, ethnic groups, and genders but also between geographic regions of the same population [5]. Therefore, these studies established the need for developing ethnic- [6] and gender-specific implants [7]. Reviewing the literature, we did not find too many data on the hip anatomy of the southeast Europe population. The subject of this study is to analyze proximal femoral geometry by radiographic evaluation in the Croatian population, which geographically belongs to the mentioned part of Europe.

We conducted a retrospective study analyzing the antero-posterior (AP) and axial radiographies of the hip, obtained within the last four years from the database of the Clinic for Orthopaedic Surgery Lovran. This study was approved by the ethical committee. The number of studied patients was 300, 168 women and 132 men. Average age of analyzed patients was 64.28 ± 13.17 (women 62.80 ± 14.72, men 66.16 ± 10.76). We excluded from this study patients with hip disorders, previous hip surgery, hip fracture history, and any infectious lesion, and OA changes grade IV according to Kellgren-Lawrence. AP view was obtained by standard pelvis positioning protocol for hip arthroplasty with the beam of the X-ray directed toward the midline above the symphysis pubis and with both lower extremities in 15° of internal rotation. The axial view of the hip was taken with the patient in the supine position, the image receptor placed superior to the iliac crest and angled approximately 20–45° to match the angle of the femoral neck, the central ray angled to be perpendicular to the long axis of the femoral neck, the centering point 13 cm distal to the neck of the femur, and patient’s unaffected hip flexed and abducted. Radiological measurements of proximal femoral geometric parameters were performed using the Agfa IMPAX Orthopaedic Tools program. The proximal femoral geometric parameters assessed were as follows: femoral head diameter, neck-shaft angle, angle of femoral neck anteversion, and lateral femoral offset (Figs. 1, 2, 3, and 4).

The results of radiographic measurements of proximal femoral geometric parameters in Croatian population are presented in Table 1. We compared proximal femoral geometry mean values of both genders. Results are presented in the Table 2. Analyzing the data revealed that mean values of female femoral head diameter and lateral femoral offset were significantly smaller than the male values of same parameters. There was no statistical significance of mean values in other femoral geometric parameters between genders.

Multiple studies analyzed the proximal femoral morphology using different specimens and methods of measuring [10, 11]. We decided to analyze conventional radiographies since they are used as a standard in pre-operative planning for total hip arthroplasty. Proximal femoral anatomy became very important, because reconstruction of the native individual values was recognized as a prerequisite factor for the success in total hip arthroplasty [12]. Hip anatomy is a subject to a high individual variability [13]. Gender is one of the parameters associated with anatomical hip variability [14]. In our group of patients, we observed statistically significant differences between genders in FHD and LFO parameters, while FNA and FNSA were of similar values in both genders. Many studies observed differences in femoral geometry between races and ethnic group [5, 6]. We compared values from this study with values from similar studies which described proximal femoral anatomy of different races and ethnic groups, but we also compared our results with the results of authors who analyzed Caucasian proximal femoral geometry [15‐19]. We detected differences in proximal femoral geometry between Croatian population and other ethnic groups. The comparison of the mean values of proximal femoral geometry between Croatian population and various ethnic groups is shown in Table 3. Comparing our results of proximal femoral geometry with the Asian population from Korea [20] and China [21], we observed that Croatian population has significantly smaller FNSA and FNA but higher LFO. Comparing the results of the Croatian population with the results of Indian studies, proximal femoral geometry has shown similarities in FNSA with the results of Rawal et al. [6] but smaller values in comparison with the study of Minakshi et al. [22]. The results for FHD are smaller and for LFO higher in our study compared with both of these studies. FNA is higher in the Croatian population in comparison with the findings by Rawal et al. [6]. Comparing our results with results from varius studies of the Caucasian population, we also detected some differences in proximal femoral geometry. The FHD in our study was 38.84 ± 5.32 mm, while the median value of the femoral head diameter in the study of Rubin et al. [16] was 43.4 ± 2.6 mm. Unnanuntana et al. [19] analyzed proximal femoral morphology in American Caucasians, and the diameter of the femoral head in his study was 52.09 ± 4.4 mm, significantly larger than in the Croatian population. With regard to FNSA, varying ranges have been described as reference ranges. Boese et al. reported the value ranging from 98 to 160° in the healthy population [13]. Normal range of the FNSA is generally considered between 120 and 140° [23] with a global mean of 126.4° [24]. Values < 120° are classified as coxa vara and > 140° as coxa valga [25]. FNSA together with femoral neck length directly affects the LFO. The reconstruction of LFO largely depends on femoral stem design. Offset reduction of more than 15% or more than 5 mm in comparison with native value reduces the abductor moment arm influencing the gate pattern [26]. The FNSA in our study was 125.34 ± 4.26°, and this is significantly lower in comparison with that of the Turkish [15] 129.71 ± 4.4°, French 129.2 ± 7.8° [17], and Norwegian population [18] 127.7 ± 7.6° but higher than in the Swiss population [16] 122.9 ± 7.6. The mean value for LFO in the Croatian population is 51.22 ± 8.44 mm. This was the highest value in comparison with the values reported in all analyzed studies regardless of race or ethnicity, and the difference was statistically significant. Another parameter of proximal femoral anatomy that should be reconstructed during hip arthroplasty is FNA. Error in adjusting the version of the femoral component of endoprosthesis will modify the lever arms, foot position, and the gait pattern and is recognized as a risk factor for hip dislocation [27] and can decrease periprosthetic bone density [28]. The literature revealed a discrepancy between native femoral neck anteversion and version of the femoral component of endoprosthesis, ranging in excessive anteversion to retroversion, especially in cementless prostheses [29]. In most studies, the degree of version of the femoral component was significantly increased compared to the degree of native femoral neck anteversion [30]. Previous studies have shown that femoral anteversion of Asians is generally larger than that of Caucasians where the mean value is about 10° [12]. FNA in the Croatian population is 16.53 ± 1.97°, between Asian and Caucasian values.

Our results support the observations from similar studies on high diversity in the morphology of the proximal femur, not only between racial and ethnic groups but also depending on the geographic regions of the same population. Compared with other ethnic groups, our study showed specificity of the Croatian population in most parameters of proximal femoral anatomy. We hope that our results will improve understanding of proximal femur morphology and may help to choose implant in correspondence with the anatomy of the hip for the majority of our population.