While other sub-specialties of orthopedics have incorporated stress radiographs into common practice [13,14,15], adult reconstruction has infrequently utilized this diagnostic modality. Flexion instability remains one of the most common causes of revision for TKA patients, and may be a significant contributor to patient dissatisfaction; however, diagnostic evaluation and indications for surgical intervention remain highly subjective and surgeon dependent [1, 4,5,6, 16]. As a result, outcomes for revision TKA for flexion instability have been inconsistent in the literature [7, 8]. The current study found that a commercially available product did not adequately reproduce tibial translation when compared to manual stress. However, radiographic translation demonstrated good to excellent inter- and intra-observer reliability, indicating that stress radiographs may be reproducibly interpreted.

Current radiographic assessment of TKAs for flexion instability includes evaluation for excessive posterior slope and reduced posterior condylar offset on lateral X-rays as well as joint line elevation on AP radiographs [4, 5]. Unfortunately, no objective cutoff values or quantitative measures for these parameters have been identified. The existing literature indicates that A–P translation on physical examination > 10 mm should be considered marked translation, and there is suggestion that correction to < 5 mm translation is more appropriate for improved clinical outcomes [3, 4, 6]. However, A–P translation based on physical examination is inherently subjective, and no current literature to our knowledge utilizes objective quantitative measures to identify these patients. Importantly, prior authors have highlighted the subjective nature of knee laxity testing in TKA, with poor inter-rater reliability using clinical testing alone, discouraging its use in isolation [17].

It has previously been noted by Stambough et al. [4] that stress radiographs could be important in the diagnostic evaluation of flexion instability. However, no prior study has demonstrated that these radiographs can be reliably interpreted or correlated to outcomes after revision surgery. Prior to determining if stress radiographs correlate to patient outcomes, it is first prudent to determine if these radiographs can be reliably and reproducibly interpreted across providers. The current study is an important first step in creating objective criteria to identify flexion instability.

Seon et al. [10] utilized the Telos device for anterior as well as posterior drawer stress radiographs, and showed an association between postoperative range of motion and total anterior–posterior laxity. However, the measurement itself was not validated, and it was not utilized to determine patient satisfaction, but rather early range of motion. While these authors did find tibial translation with the commercially available device, they compared anterior to posterior drawer rather than anterior drawer to rest, which was performed in the current study. Other authors have utilized the stress device in a native knee for posterior drawer testing, and found high reproducibility for quantification of posterior instability [11]. However, it is unclear whether or not similar results can be seen in patients after TKA. The current study was unable to demonstrate successful tibial translation in patients after TKA.

There are several limitations to this study worth noting. First, flexion instability is a dynamic process which involves both anterior and posterior tibial translation throughout functional motion. The current study evaluated only anterior tibial translation at 90° of flexion in a static position. This was performed for two reasons: first, it is important to validate the radiographic measurements rather than correlate to patient outcome at this stage, and second, dynamic evaluation of translation with functional X-rays or 3-D imaging is unlikely to be widely utilized outside of tertiary academic centers. Additionally, this study was limited in that a single provider performed the stress radiographs, and it remains unclear if stress radiographs performed by different providers would have reproducible findings. Lastly, this study did not attempt to identify patients with flexion instability for inclusion. This was done intentionally, as the purpose of the current study was only to attempt to validate the radiographic measurement.

The current study was unable to demonstrate added value to a commercially available stress device for evaluation of flexion instability. Tibial translation with a stress device was significantly less than manual stress performed by a provider on physical examination, which is likely due to method of force application, rather than amount of force applied. However, the inter- and intra-observer reliability of radiographic measurements of stress radiographs for flexion instability was high. This is important to consider as future studies evaluating flexion instability can utilize manually performed stress radiographs, and tibial translation can be reproducibly measured. With this, adult reconstruction may be able to utilize stress radiographs to quantitatively identify patients with flexion instability, and correlate changes to their outcomes following revision surgery.