From My Perspective… The Quadriceps Malalignment Question: No Evidence to Externally Rotate the Femoral Component in Kinematic Alignment

Should the femoral component be externally rotated during kinematically aligned total knee arthroplasty (KA TKA) when a preoperative CT scan of the distal one-third of the femur and knee shows a large quadriceps tendon axial angle (QTAx) or a laterally positioned proximal quadriceps tendon? A recent hypothesis, derived from mechanical alignment (MA) TKA with a near 0° prosthetic trochlear groove (PTG), argues that it should. This Perspective examines whether that recommendation is supported by evidence generated in KA TKA with a 6° valgus PTG.

The clinical claim under review is specific: that these two quadriceps tendon measurements identify a subset of patients — described as having “quadriceps tendon malalignment” (QTM) — in whom the femoral component must be externally rotated to avoid patellofemoral complications. The evidence relevant to that claim — including a body of KA TKA outcomes that predates the QTM measurements — does not support it.

What the KA TKA record already demonstrated, before the QTM hypothesis

KA TKA has been performed for nearly twenty years with femoral components designed for mechanical alignment, featuring a typical 6° valgus PTG. The femoral component is not externally rotated in KA — the technique aligns it flush with the pre-arthritic or native posterior joint line, which averages 3 to 5° less external rotation than MA (Dossett et al. 2014).

A 2017 series of 3,212 KA TKAs performed by a single surgeon with four MA-designed implants (Vanguard CR, Triathlon CR, Sigma CR, Persona CR — all 6° valgus PTG) reported 13 cases of patellofemoral instability, a 0.4% incidence (Nedopil et al. 2017). Case-control analysis found no association with preoperative deformity, sex, age, BMI, or tibial rotation; the only distinguishing variable was increased femoral component flexion (11° vs 6°, p = 0.0012), which limits the proximal reach of the trochlea and hampers early patellar engagement (Brar et al. 2016). Correcting the flexion prospectively eliminated further cases in the next 715 KA TKAs (Nedopil et al. 2017). The cause of patellofemoral instability in KA TKA is therefore sagittal, not rotational. A 2020 analysis of the Australian and New Zealand joint registries confirmed the same at population scale: in 20,512 Triathlon CR TKAs (2011–2013), the 416 performed with KA using patient-specific instrumentation (PSI) had a 7-year cumulative revision rate of 3.1% — statistically identical to the 3.0% for the 20,096 performed with computer-assisted or conventional MA instrumentation (P = .89), with only one revision for patellar instability in the KA-PSI group (Klasan et al. 2020).

This record puts the QTM prediction to a direct test: if the QTM measurements were as predictive as the hypothesis suggests, KA TKA — performed without femoral component external rotation in thousands of patients — would show a higher rate of patellar subluxation and dislocation than MA TKA, where the femoral component is routinely externally rotated by 3° to 5°. That prediction is not observed. Rates in KA TKA are comparable to MA TKA — not higher.

QT differences in PF OA reflect the arthritis, not evidence of malalignment

Consider the HKA angles measured in the varus and valgus limbs of a windswept knee: are those values evidence of preexisting malalignment, or a consequence of the arthritic process? The same question applies to the 2023 anatomic paper’s quadriceps tendon axial angle (QTAx). Twenty-five knees with severe lateral facet PF OA had a greater mean QTAx than 25 matched controls (17° ± 11° vs 3° ± 7°), against an underlying cadaveric range of 42° internally rotated to 44° externally rotated in 110 knees (Talbot et al. 2023). A cross-sectional comparison of arthritic to native values cannot answer the cause-and-effect question.

For the windswept case, the answer is on the record: the coronal deformity is arthritically acquired, not preexisting. Windswept knees are nearly twenty times more common in OA than in healthy populations (8% vs 0.4%) and are linked to meniscectomy, rheumatoid arthritis, and osteotomy (Moore et al. 2024). Arthritis caused the deformity: KA TKA restores paired varus and valgus knees to matching post-operative alignment, meaning both had the same pre-arthritic bone geometry before OA drove them apart (Howell et al. 2021). By the same logic, the QTAx values are not preexisting malalignment but a consequence of patellofemoral OA. Lateral facet PF OA translates the patella laterally and rotates it externally; because the quadriceps tendon inserts at the superior pole, it undergoes the same lateral translation and external rotation. The QTAx values measured in the arthritic cohort reflect that process, not preexisting deformity. Using arthritic values to define “malalignment” has no scientific basis.

Why the QTM analysis in MA TKA does not apply to KA TKA

The QTM analysis used to justify femoral component external rotation is based on MA TKA, not KA TKA. It comes from a 2025 series of 388 MA TKAs performed with the SAIPH MRK (MatOrtho) — an MA-designed femoral component featuring a near 0° prosthetic trochlear groove (Talbot et al. 2025). MA TKA with this implant compromises patellofemoral tracking in ways that KA TKA does not: it increases the Q-angle in over 70% of constitutionally varus knees (Bellemans et al. 2012; Lozano et al. 2019), it prevents the posterior rollback and internal tibial rotation that would typically medialize the tibial tubercle, promoting patellar tracking during flexion, and its near-vertical PTG does not match the native groove distribution in approximately 94% of patients, calculated from the reported distribution of native trochlear angles (mean 1.6 degrees valgus, SD 6.6 degrees) (Rosa et al. 2023). A lateralized quadriceps tendon becomes significant in this context because the implant kinematics and unforgiving trochlear anatomy cannot accommodate it. Nonetheless, the QTM-versus-non-QTM KOOS-12 difference in that series was 7 points — below the minimal clinically important difference of 11.1 — which the paper describes as “clinically insignificant” (Talbot et al. 2025).

The two direct tests of the QTM hypothesis in KA TKA

Two independent series in 367 KA TKAs, all performed with a 6° valgus PTG and no femoral component external rotation, tested the QTM hypothesis directly.

In 2024, a 313-knee series of KA TKAs with a 6° valgus PTG (GMK Sphere, Medacta) and patellar resurfacing found no correlation between post-operative quadriceps tendon location and 2-year Forgotten Joint Score or Oxford Knee Score; no patient reported patellofemoral instability or required reoperation for it (Razick et al. 2024). The post-operative quadriceps-tendon-to-trochlear-groove distance (11 ± 8 mm) is comparable to the 9 +/- 8 mm measured in 110 cadaveric knees (Talbot et al. 2023), refuting the concern that immediate post-operative scans distort the landmark.

In 2026, a 54-knee series with the same implant but without patellar resurfacing reported negligible correlations between the QTAx and every patient-reported outcome (FJS, OKS, pain VAS, and satisfaction) (Akagawa et al. 2026); the high-QTAx subgroup trended slightly better. Two patients (3.7%) required secondary patellar resurfacing — within the published MA TKA range — and one had a QTAx of 7°, well below the 14° “malalignment” threshold.

Together, the two series show no association between the QTM measurements and outcomes in KA TKA — with or without patellar resurfacing. Neither series was designed to test the QTM findings in MA TKA; both show those findings do not apply to KA TKA.

A KA-optimized 20° valgus PTG further reduces the utility of the QTM measurements and any need for femoral component external rotation

External rotation of the femoral component has been proposed as a fix for the valgus phenotypes (CPAK 3 and CPAK 6) and the apex-distal, neutral phenotype (CPAK 2), where the quadriceps line of pull is lateral to the 6° prosthetic trochlear groove (Sappey-Marinier et al. 2022). The evidence-based response is not to externally rotate the femoral component; it is to redesign the prosthetic groove to accommodate every CPAK phenotype. The two main features of the KA-optimized femoral component are the 20° valgus PTG, which opens the lateral trochlear ridge to accommodate the quadriceps line of pull, and the flattened medial ridge, which addresses anterior over-stuffing. Together, they align the prosthetic trochlea with the native quadriceps line of pull rather than constraining the patella into an unnatural track. In a 437-knee KA series, switching from a 6° valgus PTG (GMK Sphere) to a 20° valgus PTG (SpheriKA) raised the median Forgotten Joint Score by 6 points overall and cut the proportion of patients with a poor FJS (< 40) from 24% to 8% (Howell et al. 2026). The largest gains were in exactly the phenotypes where the quadriceps vector is lateral to the 6° PTG — CPAK 6 (median FJS 97 vs 54, n = 2), CPAK 3 (+20), and CPAK 2 (+7) — with no measurable cost in any other CPAK type (Howell et al. 2026, 2023, 2025). By contrast, deviating the posterior medial condyle anteriorly by 1 to 2.5 mm — the effect of externally rotating the femoral component — lowers the FJS by 34 points (Rak et al. 2023).

No clinical series of KA TKA — with either an MA-designed femoral component or a KA-optimized implant — has been published in support of the QTM hypothesis. The recommendation is therefore a critique of, and prescription for, an operation with no published clinical data.

About this Perspective

A letter to the editor challenging the 2024 Razick KA TKA study of the QTM measurements was forwarded to the study authors, who submitted a formal response in March 2026. The response was not published. The subsequent discussion moved to social media, which lacks the editorial process, methods section, and peer review the underlying question deserves. This Perspective is that response, placed in a venue where the evidence can be laid out in full.

Disclosures relevant to this Perspective

I developed the kinematic alignment technique and run the educational platform kaoptimizedtka.com. I receive consulting fees and royalties from Medacta International, which manufactures the GMK Sphere and the SpheriKA implants discussed in this editorial. That is a relevant disclosure and the reader should weigh it.

The relevant papers carry the following disclosures. The 2023 anatomic paper reports no conflicts (Talbot et al. 2023). The 2025 clinical TKA series that anchors the external-rotation prescription uses the SAIPH MRK (MatOrtho) implant (Talbot et al. 2025). Industry funding does not invalidate any of this work; it is context readers may weigh.

Four things a KA surgeon should take from this

First, the long history of KA TKA performed without femoral component external rotation — more than 3,200 knees in one single-surgeon series and 416 in a national registry — produced a 0.4% patellofemoral instability rate, comparable to MA TKA performed with routine external rotation (Nedopil et al. 2017; Klasan et al. 2020). The instability that did occur was caused by femoral component flexion, not by rotation.

Second, the clinical case for external rotation comes from an MA TKA series with an MA-designed implant and a 7-point KOOS-12 effect described in that paper as clinically insignificant (Talbot et al. 2025). The data do not transfer to KA TKA with a KA-optimized implant.

Third, in two independent KA TKA series totaling 367 knees performed without external rotation using an MA-designed femoral component, the location of the quadriceps tendon showed no association with 1- and 2-year patient-reported outcomes or with patellofemoral instability (Razick et al. 2024; Akagawa et al. 2026).

Fourth, when the quadriceps line of pull sits lateral to a 6° PTG, the evidence-based solution is a KA-optimized femoral component with a 20° valgus PTG, not femoral component external rotation. Switching the implant raises the FJS and cuts the proportion of poor scores by two-thirds (Howell et al. 2026); rotating the femoral component lowers the FJS by 34 points (Rak et al. 2023).

The bottom line

No published evidence supports external rotation of the femoral component in KA TKA. With a KA-optimized 20° valgus PTG available, no argument for it remains.

Declaration of Generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the author used Claude (Anthropic, San Francisco, CA, USA) to improve the language, readability, and clarity of the manuscript. After using this tool, the author reviewed and edited the content as needed and takes full responsibility for the publication’s content.

Conflicts of interest

The author is the developer of the kinematic alignment technique for total knee arthroplasty and runs the educational platform www.kaoptimizedtka.com. He receives consulting fees and royalties from Medacta International, which manufactures the GMK Sphere and SpheriKA implants discussed in this editorial. He receives no financial support from any other implant or robotic surgical platform manufacturer.

Funding

None.