In TKA, precise shape matching between the prosthesis and the distal femoral resected surface is a determining factor for a long-term successful outcome. The femoral component fitting is important, especially in the anterior-posterior direction and the medial-lateral direction. Mismatch of the prosthesis in the two directions will have a different impact on clinical outcomes and even affect the survival rate of the prosthesis. Oversizing of the femoral component in the anterior-posterior direction alters the flexion gap, leading to patellofemoral tightness or overstuffing, which can cause unexplained pain or stiffness postoperatively [15]. However, undersizing of the femoral component may require over-resection of the distal femur, which leads to the elevation of the joint line and thus causes laxity in flexion [16]. Furthermore, Downsized femoral component may increase the risk of anterior cortical notching in the use of a posterior referencing system [17]. In the medial-lateral direction, both overhang and underhang (inadequate coverage of the distal end of the femur) of the femoral component were frequently encountered. Bonnin et al have reported that ML overhang is observed in 66% of femurs (84% in females). The overhang can cause soft-tissue irritation, residual pain, poor knee flexion, and poor functional outcome [5, 18]. Whereas components underhang may expose increased contact stress to more cancellous bones, which results in early subsidence and loosening of the prosthesis [19]. Current prosthesis designs and surgical techniques should be able to cope with morphologic variances by making surgical adjustments to fit a proper component on the resected surface, avoiding overhang and soft tissue impingement caused by bigger prostheses or instability due to smaller prostheses.
Lee et al [25] investigated CT scan measurements of knee morphologic data and found that CT scan measurements are comparable to intraoperative measurements. In this study, we choose CT data for the anthropometric measurement because its advantage in obtaining the anatomic landmarks and measurement on the 3D model of the femur can reduce the interobserver error generated from different view compared with the intraoperative measurement or the cadaver measurement. Moreover, the data of CT is more convenient to be obtained. Regarding all the dimensions of the distal femur measured in this study, including ML, AP, MAP, LAP, those of females were significantly smaller than those of males. The ARs of females were smaller than that of males (P = 0.003). In addition, the linear fitting for the AP and ML dimensions of the females lies below those of males, showing that females generally have a smaller ML dimension than males for a given AP dimension, i. e. that the femurs of females are “narrower” comparing to males. (Fig. 2). In our study, the MAP and LAP on the femoral surface were also measured, and it was found that the MAP was larger than the LAP, similar to other studies [17, 20]. The anthropometric differences between females and males shown in our study were consistent with the results reported in other previous studies as the list in Table 3. Conley et al [21] identify three notable anatomic differences in females: a less prominent anterior condyle, an increased Q angle, and a reduced aspect ratio. Since the above differences exist and females account for almost two-thirds of TKA [22], the need for gender-specific knee prosthesis was proposed. But the necessity has been an issue of debate. Some studies have suggested that gender-specific prostheses do have an advantage on better radiographic fit than the standard unisex prosthesis[23, 24]. However, many studies and even systematic reviews do not give evidences to support that gender-specific prosthesis brings more favorable clinical outcomes [25–28].
Table 3
Summary of the morphometry of the distal femoral resected surface reported by different authors
Authors | Population | Journal | Published time | Measured method | ML (mm) | AP (mm) | ML/AP(AR) | MAP (mm) | LAP (mm) |
Our study | Chinese | | 2022 | CT | 66.36 ± 4.61(C) | 58.39 ± 3.81(C) | 1.14 ± 0.07(C) | 49.81 ± 3.35(C) | 45.97 ± 4.62(C) |
71.47 ± 3.84(M) | 61.87 ± 3.38(M) | 1.16 ± 0.06(M) | 50.57 ± 3.89(M) | 46.83 ± 4.21(M) |
64.78 ± 3.57 (F) | 57.31 ± 3.26(F) | 1.13 ± 0.07(F) | 49.57 ± 3.14(F) | 45.70 ± 4.71 (F) |
Ho WP et al.[15] | Chinese | The KNEE | 2006 | intraoperative | 70.2 ± 5.4 | 63.7 ± 5.1 | 1.09 ± 0.06 | | |
Cheng FB et al.[7] | Chinese | The KNEE | 2009 | CT | 71.0 ± 3.0(C) | 64.1 ± 2.7(C) | 111.1 ± 2.7%(C) | 51.3 ± 3.3(C) | 50.7 ± 4.0(C) |
74.4 ± 2.9(M) | 66.6 ± 2.4(M) | 111.7 ± 3.3%(M) | 52.6 ± 2.4(M) | 51.8 ± 3.7(M) |
66.8 ± 3.1(F) | 61.0 ± 2.7(F) | 109.6 ± 3.6%(F) | 49.8 ± 3.2(F) | 49.3 ± 4.1(F) |
Yang B et al.[14] | Chinese | PLOS ONE | 2014 | CT | 79.0 ± 5.0(M) | 66.8 ± 4.0(M) | 1.18 ± 0.06(M) | | |
71.2 ± 4.3(F) | 61.3 ± 3.3(F) | 1.16 ± 0.05(F) | | |
Ha CW et al.[8] | Korean | The Journal of Bone and Joint Surgery | 2012 | intraoperative | 74.8 (M) | 66.3 (M) | 1.12 (M) | 67.4 (M) | 66.3 (M) |
68.2 (F) | 60.8 (F) | 1.12 (F) | 61.4 (F) | 60.8 (F) |
Lim HC et al.[16] | Korean | The KNEE | 2013 | MRI | 78.6 ± 5.1(C) | 58.7 ± 3.81(C) | 1.25(C) | 59.6 ± 4.75(C) | 58.7 ± 3.81(C) |
81.5 ± 5.7(M) | 59.0 ± 4.01(M) | 1.19(M) | 62.7 ± 4.10(M) | 59.0 ± 4.01(M) |
76.7 ± 3.71(F) | 58.4 ± 3.10(F) | 1.30(F) | 56.8 ± 3.31(F) | 58.4 ± 3.10(F) |
Bellemans J et al.[17] | European | Clinical orthopaedics and related research | 2010 | CT | | | 1.31 ± 0.06(M) | | |
| | 1.29 ± 0.06(F) | | |
Loures FB et al.[18] | Brazilian | PLOS ONE | 2019 | intraoperative | 79.9 ± 5.7(M) | 68.4 ± 5.6(M) | 1.16 ± 0.1(M) | 68.6 ± 5.7(M) | 68.4 ± 5.6(M) |
70.2 ± 5.4(F) | 61.6 ± 5.2(F) | 1.15 ± 0.1(F) | 60.7 ± 5.2(F) | 61.6 ± 5.2(F) |
Serhat Mutlu et al.[19] | Turkish | Journal of Back and Musculoskeletal Rehabilitation | 2021 | CT | 72.1 ± 2.8(C) | 66.1 ± 2.1(C) | 1.09(C) | 53.7 ± 2.8(C) | 53.0 ± 3.7(C) |
75.8 ± 2.8(M) | 68.6 ± 2.6(M) | 1.10(M) | 54.6 ± 2.7(M) | 53.8 ± 3.6(M) |
67.9 ± 2.7(F) | 63.2 ± 2.2(F) | 1.08(F) | 52.2 ± 3.1(F) | 51.7 ± 4.0(F) |
Hafez MA et al.[20] | Arabian | The Journal of Arthroplasty | 2016 | CT | 72.04 ± 6.6(C) | 68.1 ± 7.75(C) | 1.06 ± 0.14(C) | 51.82 ± 6.06(C) | 49.45 ± 6.24(C) |
78.48 ± 7.05(M) | 72.88 ± 4.57(M) | 1.08 ± 0.08(M) | 55.16 ± 4.55(M) | 54.91 ± 5.57(M) |
70.5 ± 5.48(F) | 67.26 ± 7.39(F) | 1.06 ± 0.15(F) | 51.02 ± 6.12(F) | 48.14 ± 5.68(F) |
ML, mediolateral dimension; AP, anteroposterior dimension; MAP, medial anteroposterior dimension; LAP, lateral anteroposterior dimension; AR, femoral aspect ratio; M, Male; F, Female; C, Combined. |
To date, many recent anthropometric studies have demonstrated knee anthropometric differences in Caucasians and other ethnicities [6, 9, 10, 29–31]. Among them, the difference between the Asian race and Caucasian race has been studied the most. And it is shown most clearly that all the measured distal femoral dimensions of the Asians and as well as the ARs of Asians [7, 8, 14, 32, 33] were generally smaller than those of the Caucasians. The smaller ARs of Asian knees means the distal femur seemed to be “narrower” than that of the Caucasian knees. The morphometric data of the distal femoral resected surface in our study was compared with that of previous investigations obtained from other ethnics in Table 3. Our present study revealed that the average ARs of the distal femoral resected surface was 1.14 ± 0.07, which was consistent with the data measured in other studies of Chinese knees [7, 14, 32], and similar to those of knees of other Asian ethnics, like Korean, but smaller than those of Caucasian knees [10]. Bellemans J et al.[34] reported that the average ARs of the distal femoral resected surface of European was 1.31 ± 0.06 for males, and 1.29 ± 0.06 for females. Hafez MA et al. reported that the size of Arab knees was generally smaller than Caucasian’s and larger than Asian’s [19]. However, the comparison of the differences of the diameters and AR values between different ethnics could not be authentically realized due to differences in methods of measurement and imaging technology.
In 2019, the total number of TKAs performed in China was 374,833, of which 54.8% adopted imported prostheses [35]. The currently imported prostheses used in China are designed based on the anthropometric data of Caucasian knees and do not accommodate the Chinese knees. Cheng FB et al. [7] compared the morphometric data of Chinese knee with those of five prostheses currently used in China: Scorpio and Duracon (Stryker Howmedica Osteonics, Allendale, NJ); PFC sigma (DePuy-Johnson and Johnson, Warsaw, IN); Nexgen (Zimmer, Warsaw, IN) and DC-Dynamic (Plus-Fosun ortho Co., Beijing, China), four of which were imported prostheses. The results suggest that the imported prostheses which are suitable for Caucasian patients may be larger in ML dimension, and this situation was more evident in females.
The above situation not only exists in China, relevant studies in South Korea, Japan, Thailand, and other countries also have reached similar conclusions. The current design of prosthesis according to Caucasian knees does not cater to femoral anthropometric differences among races, which will bring the risks of overhang in the femoral component [6, 8, 10, 30, 31]. And thus, the idea of the racial-specific prosthesis was proposed. However, more studies are needed to determine whether the long-term clinical outcome of racial-specific prostheses will be as controversial as gender-specific ones.
In addition to the differences among gender and race, a wide inter-individual variation was observed in the present study ranging from 1.00 to 1.34, which deserves more attention and is worthy of our exploration on how to select the proper component size for coping with such a large variety of distal femoral resected surface. In this study we distributed ARs according to the quartiles and defined the morphometric features of the distal femoral resected surfaces into three morphotypes: “narrow” (AR values from 1.00 to1.09, smallest, accounting for 25%), “moderate” (AR values from 1.09 to1.18, immediate, accounting for 50%), and “dumpty” (AR values from 1.18 to1.34, largest, accounting for 25%) (Fig. 3.). As the number of current component sizes available to surgeons was limited, there were two situations we may encounter during the operation: When the components with a small AR value, the resected femoral surface is “narrow”; when the components with a larger AR value, the femoral resected surface is “dumpty”. After the distal femoral resection, we choose the femoral component size by referring to AP dimension or ML dimension on the resected surface. In the “narrow” surface, referring to the AP dimension can lead to an overhang in the ML direction; referring to the ML dimension can cause excess osteotomy of the posterior condyle or notch in the anterior cortex. While in the “dumpty” surface, referring to the AP dimension can lead to underhang in ML direction; referring to ML dimension can cause overstuffing. However, we only defined the morphotypes of the distal femoral resected surface according to the calculated quartiles of ARs in this study, but didn’t investigate component position on the postoperative radiography and the clinical outcomes of TKA knees with different morphometric features. In the further study, we will focus on the proposal of a classification system of the distal femoral resected surface to instruct us how to choose the proper femoral component size.
Our study has some limitations. Firstly, all participants enrolled in the present study received the treatment limited to a single orthopedic center in southwest China, which may not sufficiently represent the whole Chinese population. However, research regionalization is the common problem in the current study of femoral morphology. Secondly, all the measurements and analysis in this study were carried out by the same person, which may lead to information bias. Thirdly, we did not combine the postoperative radiography and the clinical outcomes with the variable AR values. This part of the research requires careful design and more data on clinical outcomes, we will continue this part of the research in the future.