A Propensity-Matched Analysis of Anatomic Risk Factors for Periprosthetic Patellar Fractures After Total Knee Arthroplasty

Braden V Saba; Farouk Khury; Chloe Fong; David Novikov; Daniel Sherwood; Joshua C Rozell

doi:10.1016/j.arth.2026.03.079

A Propensity-Matched Analysis of Anatomic Risk Factors for Periprosthetic Patellar Fractures After Total Knee Arthroplasty

J Arthroplasty. 2026 Mar 31:S0883-5403(26)00303-7. doi: 10.1016/j.arth.2026.03.079. Online ahead of print.

Authors

Braden V Saba¹, Farouk Khury², Chloe Fong¹, David Novikov¹, Daniel Sherwood¹, Joshua C Rozell¹

Affiliations

¹ Division of Adult Reconstruction, Department of Orthopedic Surgery, NYU Langone Health, New York, New York.
² Division of Adult Reconstruction, Department of Orthopedic Surgery, NYU Langone Health, New York, New York; Division of Orthopedic Surgery, The Ruth and Bruce Rappaport Faculty of Medicine, Haifa, Israel.

PMID: 41933602
DOI: 10.1016/j.arth.2026.03.079

Abstract

Background: Periprosthetic patellar fracture (PPPF) after total knee arthroplasty (TKA) is a rare complication, but can dramatically affect patient function and implant survival. This study sought to better identify radiographic and anatomical risk factors for PPPF compared to a propensity-matched cohort.

Methods: We retrospectively queried 22,092 TKAs from January 2011 to December 2024 with patellar resurfacing at a single, urban, academic institution, yielding 44 (0.2%) verified cases of PPPF after TKA. Using propensity score matching on the basis of age, sex, body mass index, race, and Charlson comorbidity index, 44 control TKA patients who had patellar resurfacing without fracture were identified and analyzed using the same methods.

Results: The mean time to PPPF was two years after TKA (range, 10 days to 10 years), and 46% were atraumatic. A decreased native lateral patellar tilt (20.0 versus 22.6°, P = 0.039) and thinner native patellar thickness (22.6 versus 24.0 mm, P = 0.018) were associated with increased PPPF risk. Lateralization of the patella during resurfacing also increased risk (P = 0.011), as well as increased patellar component size (P = 0.034). On the receiver operating characteristic analysis, thinner native patella thickness was the most predictive of fracture risk (area under the curve = 0.621); however, this result was underpowered, thus an optimal cutoff value could not be meaningfully established.

Conclusions: The PPPFs are rare complications following TKA and may occur with or without trauma. Native patellar thickness was inversely correlated with fracture risk, and lateral positioning during resurfacing and increased component size were associated with increased risk. Patients who had PPPFs also demonstrated lesser lateral patellar tilt compared to nonfractured controls. Further investigation with a larger cohort may enable more precise risk factor stratification.

Keywords: patellar fracture; periprosthetic patellar fracture; radiographic; risk factors; total knee arthroplasty.