Total knee arthroplasty (TKA) is one of the most recent and successful orthopaedic procedures for knee osteoarthritis (OA). However, several conditions, including proper coronal alignment of the implant, must be met for a successful TKA. In this regard, the TKA revision rate has reportedly increased in outlier groups, in which coronal alignment deviated from the mechanical axis by > ± 3° with regard to implant placement in TKA [17, 30]. Therefore, surgeons have developed advanced surgical techniques, such as navigation surgery [4, 6, 10, 27] and patient-specific instrumentation [28, 29, 33], to allow placement of implants within ± 3° of the mechanical alignment. However, the need for strict coronal alignment to ensure long-term results is questionable because medium- to long-term TKA revision rates in neutrally aligned patients were not different from those of patients with ± 3° deviations from the mechanical alignment [1, 33]. Nevertheless, residual postoperative pain, knee function, and patient satisfaction are remaining concerns in TKA [12], and coronal alignment may affect clinical outcomes.
Bellemans et al. [8] conceptualised constitutional varus, stating that 32% of men and 17% of women experience varus alignment after skeletal maturity. In such cases, returning to neutral alignment after TKA may have adverse effects and induce overcorrection. Moreover, Vanlommel et al. preoperatively evaluated the data of 143 patients with varus deformity using radiography and classified them into three groups (i.e., the neutral, mild [3°-6°], and moderate [6°-9°] varus groups) according to alignment immediately after TKA. Among these, the mild varus group had the most desirable outcomes [34].
While mechanical alignment was the primary goal of TKA, new concepts have recently emerged, including the goal of kinematic alignment for the constitutional varus knee [16]. However, such concepts are still debatable [2]. Nevertheless, surgeons should attempt to place the implant within 3° of the target alignment to achieve a stable outcome following TKA. Alignment accuracy is evaluated via postoperative imaging, and a deviation of > 3° from the target angle is often reported as an outlier. Additionally, measurement error > 1° from the actual alignment may create a major problem during evaluation.
Although standing long-leg radiography (LLR) is the standard evaluation method for coronal alignment after TKA, lower extremity rotation and limited knee extension at the time of imaging reportedly affected measurements [5, 13, 18, 21, 23, 24, 35]. In this regard, Mizu-uchi et al. [27] found a significant difference between three-dimensional computed tomography (3DCT)- and LLR-measured angles for coronal alignment after TKA. Moreover, the postoperative coronal alignment measurements assessed through navigation and LLR in TKA were reportedly not correlated, with LLR exhibiting low validity [19, 35]. However, Boonen et al. [11] concluded that coronal alignment the assessment by LLR was sufficiently valid as compared to that by 3DCT. Nevertheless, the validity of LLR measurements remains controversial. Therefore, this study aimed to compare the reliability and validity of the coronal alignment angle measurements after TKA using LLR and 3DCT. LLR reproducibility was evaluated by comparing the lower extremity alignment measurements of LLR taken on the fifth postoperative day (5-day LLR) with those taken 6 months postoperatively (6-month LLR). We hypothesised that LLR was less consistent and had poor reproducibility at different postoperative imaging time points compared to 3DCT. Therefore, LLR might not be a valid evaluation method for the coronal alignment after TKA.