PREPATELLAR FAT THICKNESS RATIO AS AN INDICATOR OF INFECTION RISK FOLLOWING TOTAL KNEE ARTHROPLASTY

Purpose: Investigation of the correlation between obesity, as defined by BMI, and risk of infection following TKA has been met with mixed results. Several studies show an increased risk of infection in the obese, while others show an acceptable success rate. It is our belief that a specific study of the fat distribution surrounding the knee itself versus the general BMI measurement will yield a more precise corollary to the risk of infection. We plan to assess the fat distribution over the knee quantitatively by creating a "prepatellar fat thickness ratio", which would compare the soft tissue thickness anterior to the patella with the actual thickness of the patella. Methods: An alpha level of 0.05 was used to determine statistical significance. Descriptive statistics including means, standard deviations, frequencies and confidence intervals were computed for the variables of interest in the study. Chi-square tests were used to determine if a patient's development of an infection was associated with their gender, smoking or diabetes status. Linear regression were used to determine if the prepatellar ratio was significantly affected by a patient's BMI or age. Logistic regression and odds-ratios were computed to examine the effect that BMI, smoking, diabetes, gender or the prepatellar ratio had on developing an infection post total knee arthroplasty. Results: There were no significant associations detected between developing an infection and the patient's gender, age, BMI, smoking or diabetes status. Using logistic regression, it was determined that for every 0.925mm increase in the ratio between the thickness of the patella and the thickness of the prepatellar soft tissue, a patient's risk of developing an infection post-surgery increased approximately 2.523 times. Conclusions: Many authors believe obesity plays a significant part in increasing infection risks, and used BMI to verify this risk; however, the prepatellar fat thickness ratio proved to be the most accurate predictor of infection in our study. BMI is a not a composition specific measurement, and has met mixed results when used to predict infection risks in some studies. In contrast, the prepatellar ratio is both composition and site specific for TKA. Further investigation is warranted, but it is our belief that this measurement holds promise as a pre-operative indicator of infection risk