Wound infection rates are generally higher in patients undergoing surgery for spinal metastasis. Risk factors of wound infection in these patients are poorly understood.
Purpose: To identify demographic and clinical variables that may be associated with patients experiencing a higher wound infection rate.
Study design: Retrospective study with prospectively collected data of spinal metastasis patients operated consecutively at a University Teaching Hospital, adult spine division which is a tertiary referral centre for complex spinal surgery.
Patient sample: Ninety-eight patients were all surgically treated, consecutively from January 2009 to September 2011. Three patients had to be excluded due to inadequate data.
Outcome measures: Physiological measures, with presence or absence of microbiologically proven infection.
Methods: Various demographic and clinical data were recorded, including age, serum albumin level, blood total lymphocyte count, corticosteroid intake, Malnutrition Universal Screening Tool (MUST) score, neurological disability, skin closure material used, levels of surgery and administration of peri-operative corticosteroids. No funding was received from any sources for this study and as far as we are aware, there are no potential conflict of interest-associated biases in this study.
Results: Higher probabilities of infection were associated with low albumin level, seven or more levels of surgery, use of delayed/non-absorbable skin closure material and presence of neurological disability. Of these factors, levels of surgery were found to be statistically significant at the 5 % significance level.
Conclusion: Risk of infection is high (17.9 %) in patients undergoing surgery for spinal metastasis. Seven or more vertebral levels of surgery increase the risk of infection significantly (p < 0.05). Low albumin level and presence of neurological disability appear to show a trend towards increased risk of infection. Use of absorbable skin closure material, age, low lymphocyte count, peri-operative administration of corticosteroids and MUST score do not appear to influence the risk of infection.