Use of vancomycin and gentamicin-impregnated calcium sulfate beads for the prevention of surgical site infection in adult spinal deformity

BACKGROUND CONTEXT Surgical site infections (SSI) occur at higher rates in adult spinal deformity surgery and are a significant driver of morbidity and increased health system cost. Use of intrawound vancomycin powder has demonstrated equivocal results, prompting interest in delivery mechanisms such as antibiotic-laden calcium sulfate beads which offer improved elution profiles and have demonstrated success in extremity surgery. PURPOSE Given preliminary reports of superiority of vancomycin-impregnated calcium sulfate beads relative to vancomycin powder but persistence of gram-negative infections, the current study sought to investigate the use of vancomycin and gentamicin calcium sulfate beads for combined gram-positive and gram-negative coverage. We hypothesized that combination therapy beads would result in fewer gram-negative infections with a consistent safety profile. STUDY DESIGN/SETTING This was a retrospective cohort study at a tertiary academic referral center on adult patients undergoing thoracolumbar instrumented fusion for adult spinal deformity. PATIENT SAMPLE Patients either had vancomycin-impregnated calcium sulfate beads implanted into the surgical site prior to closure (vancomycin bead group [VB]) or combination vancomycin and gentamicin calcium sulfate beads used in a similar fashion (“Vanc/Gent bead group”[VGB]). These groups were compared with a historical control using intrawound vancomycin (“Vanc powder group”[VP]). OUTCOME MEASURES The primary outcome was deep surgical site infection in accordance with Centers for Disease Control (CDC) criteria. Nephrotoxicity was assessed using Acute Kidney Injury Network (AKIN) criteria. METHODS Bivariate analyses were performed using Wilcoxon rank sum testing, chi-squared, or Fisher's exact testing. Logistic regression modeling yielded odds ratios with adjustment for significant values on bivariate testing. RESULTS One-hundred sixty patients met inclusion criteria: 66 in the VB group, 58 in the VGB group, and 36 in the VP group as a historical control. The surgical site infection rate was 1.5% (1/66) in the VB and 5.2% (3/58) in the VGB, compared with 11.1% (4/36) in the VP group, with no significant difference between groups. The sole SSI in the VB group was in a 73-year-old diabetic male who developed a postoperative deep SSI with Klebsiella species on postoperative day 20. All three infections in the VGB group were polymicrobial, whereas pathogens in the VP group varied amongst gram positive and negative monomicrobial infections. The overall rate of AKI was 8.8% (14/160), however there was no significant difference in AKI rates between groups. CONCLUSIONS SSI rates comparing vancomycin-impregnated calcium sulfate beads with vancomycin and gentamicin-impregnated beads were not statistically different, however the low rates in both bead groups compared with powder use (1.5% and 5.2% compared with 11.1% in the powder group) are encouraging. Notably, all infections seen with combination vanc/gent therapy were polymicrobial, and future studies may investigate the clinical consequences of polymicrobial infections and the associated need for broad spectrum antibiosis. Further study is also critical to confirm the safety profile of combined therapy given the high overall rate of AKI in this cohort. FDA Device/Drug Status Stimulan Calcium Sulfate Beads (Approved for this indication) Surgical site infections (SSI) occur at higher rates in adult spinal deformity surgery and are a significant driver of morbidity and increased health system cost. Use of intrawound vancomycin powder has demonstrated equivocal results, prompting interest in delivery mechanisms such as antibiotic-laden calcium sulfate beads which offer improved elution profiles and have demonstrated success in extremity surgery. Given preliminary reports of superiority of vancomycin-impregnated calcium sulfate beads relative to vancomycin powder but persistence of gram-negative infections, the current study sought to investigate the use of vancomycin and gentamicin calcium sulfate beads for combined gram-positive and gram-negative coverage. We hypothesized that combination therapy beads would result in fewer gram-negative infections with a consistent safety profile. This was a retrospective cohort study at a tertiary academic referral center on adult patients undergoing thoracolumbar instrumented fusion for adult spinal deformity. Patients either had vancomycin-impregnated calcium sulfate beads implanted into the surgical site prior to closure (vancomycin bead group [VB]) or combination vancomycin and gentamicin calcium sulfate beads used in a similar fashion (“Vanc/Gent bead group”[VGB]). These groups were compared with a historical control using intrawound vancomycin (“Vanc powder group”[VP]). The primary outcome was deep surgical site infection in accordance with Centers for Disease Control (CDC) criteria. Nephrotoxicity was assessed using Acute Kidney Injury Network (AKIN) criteria. Bivariate analyses were performed using Wilcoxon rank sum testing, chi-squared, or Fisher's exact testing. Logistic regression modeling yielded odds ratios with adjustment for significant values on bivariate testing. One-hundred sixty patients met inclusion criteria: 66 in the VB group, 58 in the VGB group, and 36 in the VP group as a historical control. The surgical site infection rate was 1.5% (1/66) in the VB and 5.2% (3/58) in the VGB, compared with 11.1% (4/36) in the VP group, with no significant difference between groups. The sole SSI in the VB group was in a 73-year-old diabetic male who developed a postoperative deep SSI with Klebsiella species on postoperative day 20. All three infections in the VGB group were polymicrobial, whereas pathogens in the VP group varied amongst gram positive and negative monomicrobial infections. The overall rate of AKI was 8.8% (14/160), however there was no significant difference in AKI rates between groups. SSI rates comparing vancomycin-impregnated calcium sulfate beads with vancomycin and gentamicin-impregnated beads were not statistically different, however the low rates in both bead groups compared with powder use (1.5% and 5.2% compared with 11.1% in the powder group) are encouraging. Notably, all infections seen with combination vanc/gent therapy were polymicrobial, and future studies may investigate the clinical consequences of polymicrobial infections and the associated need for broad spectrum antibiosis. Further study is also critical to confirm the safety profile of combined therapy given the high overall rate of AKI in this cohort.