2.3. Statistical analysis

SPSS (SPSS, Inc, Chicago, IL) was used for the statistical analysis. The clinical data were analyzed using the means and standard deviations. The level of statistical significance was set at P < .05. Independent 2-sided paired t-tests were performed to determine the difference in the HSS knee score, WOMAC score, and the ROM. For categorical variables, chi-square analysis was used to compare the difference. The Kaplan–Meier method was used for the survivorship analysis. For the patients who could not complete the clinical evaluation during the latest follow-up, we presumed that the implants worked well in situ if no problems with the TKA were described at the last visit. The survivorship between different factors was evaluated in a log rank test. The risk factor of patellar related complications was analyzed with Cox proportional hazards model. Hazard ratio (HR) with 95% confidence interval (CI) was used to evaluate the relative risk between the different groups. A 2-tailed post hoc power analysis was conducted to detect a 10% difference in clinical outcome assuming an α error of 5% by using the software G∗Power (version 3.1.9.2, Franz Faul, University of Kiel, Kiel, Germany).