Correlation between lumbar multifidus fat infiltration and lumbar postoperative infection: a retrospective case–control study

BACKGROUND: The aim of this study was to investigate the correlation between lumbar multifidus fat infiltration and lumbar postoperative surgical site infection (SSI). Several clinical studies have found that spine postoperative SSI is associated with age, diabetes, obesity, and multilevel surgery. However, few studies have focused on the correlation between lumbar multifidus fat infiltration and SSI. METHOD: A retrospective review was performed on patients who underwent posterior lumbar interbody fusion (PLIF) between 2011 and 2016 at our hospital. The patients were divided into SSI and non-SSI groups. Data of risk factors [age, diabetes, obesity, body mass index (BMI), number of levels, and surgery duration] and indicators of body mass distribution (subcutaneous fat thickness and multifidus fat infiltration) were collected. The degree of multifidus fat infiltration was analyzed on magnetic resonance images using Image J. RESULTS: Univariate analysis indicated that lumbar spine postoperative SSI was associated with urinary tract infection, subcutaneous fat thickness, lumbar multifidus muscle (LMM) fat infiltration, multilevel surgery (≥2 levels), surgery duration, drainage duration, and number of drainage tubes. In addition, multiple logistic regression analysis revealed that spine SSI development was associated with sex (male), age (> 60 years), subcutaneous fat thickness, LMM fat infiltration, and drainage duration. Receiver operating characteristic curve analysis indicated that the risk of SSI development was higher when the percentage of LMM fat infiltration exceeded 29.29%. Furthermore, Pearson’s correlation analysis demonstrated that LMM fat infiltration was correlated with age but not with BMI. CONCLUSION: Indicators of body mass distribution may better predict SSI risk than BMI following PLIF. Lumbar Multifidus fat infiltration is a novel spine-specific risk factor for SSI development.