The most important finding of the present study was that femoral anterior bowing was an influential factor for implant positioning in TKA with conventional femur osteotomy. Many authors have concluded that correct alignment of the prosthesis was correlated with clinical success in TKA[11]. Moreover, improper positioning of components during surgery, may lead to early loosening of the implant due to impingement between the cam and the post in posterior stabilized systems[25]. Extensive study has been performed on ideal coronal alignment of the femoral component in total knee prostheses[3]. In contrast, there are limited studies addressed sagittal alignment of femoral component. In general, there is no consensus on sagittal mechanical axis of the alignment of femur and femoral component[4].
Lasam et al.[23] found that the proportion of knees with cFSB of was 42.2 % in the TKA group in women, and Yasushi Akamatsu et al.[18] reported that the proportion of knees with lateral FSB of angulation of > 5°was 37.8%, which was similar to our finding in this study. It is controversial whether age, BMI, gender was the factor associated with sFSB or cFSB. In this study, we found that FSB was not associated with age, BMI or gender. Although some researchers thought sFSB was associted with BMI, and anterior FSB and age were correlated in women[18, 26], there was no clear conclusion. Yehyawi TM et al.[8] demonstrated that large variances of sagittal femoral bowing and the taller and heavier patients had less distal bowing, and men had greater proximal and less distal bowing than women. Egol et al.[27] stated that there was no correlation between age and anterior FSB. Walensky[28] shown that American-Indians exhibited greater anterior curvature than Caucasians and African-Americans, and the femora of Eskimos was more closely related to American-Indians. Tang et al.[29] studied the alignment of femur in sagittal plane in Chinese population, they demonstrated that the distal one-third of the femur was not just curved antero-posteriorly, but it was more bowed than the middle and upper segments. The finding of Kim JM et al.[12] study was that femoral anterior bowing was an influential factor for implant positioning in conventional TKAs. If the femoral component is too flexed, the impingement of the femoral cam on the anterior aspect of the polyethylene post can cause accelerated wear of the post. Therefore, it should draw more surgeons’ attention to sFSB, especially in the East-Asians. Tang et al.[29] recommend that the patients with obvious femoral bowing at the distal femur as seen on the preoperative long film should be used with caution when performing intramedullary guide in TKA. Similar results were found by Bao et al.[19].
In this study, we found that there was a significant correlation between sFBA and cFBA, and little literature discussed the correlation between sFBA and cFBA. We conjectured the relationship between sFBA and cFBA maybe regulated by femur growth and development. However, the patients with sFSB usually presented non-cFSB. According to our data, we observed that cFBA was not correlated with sFBA in the patients with sFSB, and cFBA in the patients with sFBA showed no significant difference when comparing with non-sFBA. Our data suggested that there was no correlation between coronal femoral shaft bowing and sagittal femoral shaft bowing in the OA patients with undergoing TKA.
The implant alignment is an issue of high importance in TKA. Young-Hoo Kim et al.[30] studied 3018 patients who underwent total knee arthroplasty, they thought that when total knee components in the position of:femoral coronal alignment, 2–8°valgus; femoral sagittal alignment, 0–3°; tibial coronal alignment, 90°; tibial sagittal alignment, 0–7°; femoral external rotational alignment, 2–5°; tibial external rotational alignment, 2–5°; and overall anatomical knee alignment at an angle of 3-7.5°valgus, the survival rate of the prosthesis could improve. Whether coronal femoral shaft bowing or sagittal femoral shaft bowing, the femoral component alignments in the coronal plane or sagittal plane were affected[7, 23]. The view was further confirmed in our study. Comparision between β angle in the sFSB patients and the non-sFSB patients was significant difference, also, there was significant difference between mTFA of the patients with cFSB and the patients with non-cFSB. As mentioned in many previous publications[20, 23, 31, 32, 33], the presence of coronal femoral shaft bowing may lead to varus orientation of the femoral component during the implantation with using intramedullary guiding system. Similarly, we found that the cFBA was correlated with mTFA. Coronal variations in femoral shape may result in a distal cut which was not perpendicular to the femoral mechanical axis. Accurate distal femoral resection is challenging because it was difficult to determine the mechanical axis during surgery. The standard practice to determine distal cutting angle referenced off the femoral intramedullary guide was usually to measure the angle between the anatomical and mechanical femoral axis on preoperative radiographs[12]. The position of TKA implant significantly affects the outcome of joint replacement in the coronal plane [34, 35, 36]. Although femoral component alignment has been thoroughly studied, the importance of femoral
sagittal bowing in TKA has not been widely studied[11]. No clear safety margin has been documented in the sagittal plane. Sagittal femoral shaft bowing should be considered in TKA because the axis of the distal femur was more flexed than the sagittal femoral mechanical axis when sagittal femoral bowing angle increased[37]. Kazemi SM et al.[4] studied 25 patients who underwent TKA using cruciate retaining knee prosthesis, they found that there was a significant positive correlation between femoral component flexion in relation to mechanical axis and amount of sagittal femoral bowing, amount of flexion in relation to mechanical axis and DACL was 8.4 ± 2.9 degrees (range: 4 to 14.3 degrees) and 1.7 ± 0.9 degrees (range: 0 to 3 degrees). In our study, we found that the sFBA was correlated with β angle(CC = 0.543, p < 0.01) and cFBA was correlated with β angle(CC = 0.241, p = 0.035). On the other hand, femoral component alignment in sagittal plane may be affected by the coronal femoral bowing angle. As known, alignment of the femoral component in sagittal plane is determined by some factors including the entry point of intramedullary guide, depth of guide insertion, sagittal femoral shaft bowing, implant design[4]. Furthermore, sagittal femoral position had a significant influence on patellar kinematics[5] and increased femoral component flexion decreased the flexion gap and altered condylar lift-off and tibiofemoral kinematics[38, 39]. Sebastien Lustig et al.[40] demonstrated the angle between distal femoral cut in the sagittal plane and the mechanical axis more than 3.5°was an independent risk factor for clinically detectable flexion contracture. Consequently, Hiroyuki N et al.[1] advocated that upsizing or downsizing of femoral component could occur if the femoral osteotomy was performed in at least 3°extension or flexion. Usually, distal femoral cutting error in the sagittal plane which lead to hyperflexion of femur component position was due to using intramedullary alignment guide in the femur with sagittal femoral shaft bowing, as shown in Fig. 2. Computer navigation may achieve more accurate alignment[21, 41], the previous studies showed that the use of navigation improves alignment. The overall incidence of mechanical tibiofemoral angle outliers was lower in the navigation group (15.4% versus 24.9%)[23]. However, there was controversial, Chen et al.[7] found that navigated TKAs resulted in a higher risk of hyperextension of the femoral components and Jae Han Ko et al.[7] found that the femoral implant position was more extended in navigated TKAs than in conventional TKAs. Xu et al.[24] invented an extramedullary device which was easy and convenient and this instrument could help the surgeons perform TKAs with achieving better alignment in both coronal and sagittal planes.
Several limitations in this study must be acknowledged. First, the main problem was insufficient sample size during the study period about how to determine and measure sagittal alignment of the components to provide the statistical power. Second, we used only one conventional instrumentation system (Genesis II), and different knee prosthesis may produce different results. Third, this was a radiological study and we did not assess any relation between alignment and function.
In summary, we first assessed the correlation between sFBA and cFBA, in the study, we found that the sFBA was correlated with cFBA. Using conventional intramedullary alignment guide, sFSB will affect the femoral component alignment in the sagittal plane and cFSB will affect the femoral component alignment in the coronal plane. It was unclear whether there was any correlation between age, BMI or gender and sFSB or cFSB. The patients with sFSB usually presented non-cFSB.