Are lower preoperative serum sodium levels associated with postoperative surgical site infection? Results from a propensity matched case-control study

Abstract

Background: We previously reported a statistical trend toward a harmful association between lower preoperative serum sodium levels and surgical site infection (SSI) in South African (SA) laparotomy patients. Serum sodium tests are widely available and could serve as a cost-effective method for preoperatively identifying patients at risk for SSI who might benefit from additional preventative strategies. We sought to investigate the possible association between lower serum sodium levels and SSI further, in a larger sample of SA patients undergoing various surgical procedures. Objective: To determine if lower preoperative serum sodium levels are associated with SSI in SA surgical patients. Method: This was a propensity matched case-control study involving data from 729 surgical patients who attended a quaternary SA hospital between 01 January 2012 and 31 July 2016. Cases were defined as patients who developed SSI. Controls were defined as patients who did not develop SSI. Multivariate logistic regression was used to investigate the association between preoperative serum sodium levels (in mmol/L) and SSI. Results: Lower preoperative serum sodium levels were associated with a higher risk of SSI (odds ratio per 1.0 mmol/L decrease in serum sodium: 1.051, 95% confidence interval: 1.007–1.097; p = 0.026). Conclusion: Although we report a statistically significant association between lower preoperative serum sodium levels and a higher risk of SSI, the magnitude of this effect size (odds ratio) is minimal and clinically insignificant. Preoperative serum sodium levels are unlikely to be useful for SSI risk stratification in our setting.