Advertisement

Does intrawound vancomycin powder reduce surgical site infection after posterior instrumented spinal surgery? A propensity score-matched analysis

      Abstract

      Background Context

      Recent reports suggested that placing vancomycin powder into surgical wounds before closure can prevent surgical site infections (SSIs) in spinal surgery.

      Purpose

      The present study aimed to evaluate if intrawound vancomycin powder could prevent SSIs after spinal surgery with posterior instrumentation.

      Study Design

      This is a multicenter retrospective cohort study using propensity score matching.

      Patient Sample

      We reviewed all spinal surgeries performed with posterior instrumentation from July 2012 to December 2014 at 11 institutions among patients aged ≥15 years.

      Outcome Measures

      The incidence of SSIs was compared between patients who received intrawound vancomycin powder (vancomycin group) and those who did not (control group).

      Methods

      Demographic and operative data and microbiological findings of SSI cases were analyzed. After a preliminary whole-cohort analysis, we performed one-to-one propensity score matching to adjust for the differences between the two groups and then compared the incidence of SSIs between the matched groups. No funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript.

      Results

      A total of 2,859 patients were included in the study. In the vancomycin and control groups (n=694 and n=2165, respectively), 12 (1.73%) and 21 (0.97%) patients developed SSIs, respectively, but the difference was not statistically significant (p=.10, chi-square test). During the propensity score-matched analysis, 507 pairs were analyzed. No significant change in the rate of SSIs was seen between the vancomycin and control groups (8 SSIs [1.58%] vs. 9 SSIs [1.78%], respectively; p=.81, chi-square test). Microbiological analysis revealed that 5 of 12 (42%) and 11 of 21 (52%) SSIs in the vancomycin and control groups, respectively, were caused by Staphylococcus (p=.72, Fisher exact test).

      Conclusions

      Intrawound application of vancomycin powder was not associated with a significant decrease in the incidence of SSIs after posterior instrumented spinal surgeries in a propensity score-matched analysis. However, the rate of infections caused by Staphylococcus species was lower in the vancomycin group.

      Keywords

      Introduction

      Surgical site infections (SSIs) are one of the most serious complications of surgery and a common concern among surgical specialties. Surgical site infections are associated with high morbidity, mortality, and health-care costs [
      • de Lissovoy G.
      • Fraeman K.
      • Hutchins V.
      • Stoner J.
      • Li J.
      • Puckett T.
      Surgical site infection: incidence and impact on hospital utilization and treatment costs.
      ,
      • Kirkland K.B.
      • Briggs J.P.
      • Trivette S.L.
      • Angotti J.M.
      • Kalobwe E.
      • Muhlbauer M.S.
      • et al.
      The impact of surgical-site infections in the 1990s: attributable mortality, excess length of hospitalization, and extra costs.
      ]. Despite the ubiquity of prophylactic antibiotics and aseptic techniques, SSIs occur at a rate of 0.6%–12% in patients undergoing instrumented spinal surgery [
      • Imajo Y.
      • Taguchi T.
      • Yone K.
      • Okawa A.
      • Otani K.
      • Ogata T.
      • et al.
      Japanese 2011 nationwide survey on complications from spine surgery.
      ,
      • Cooper K.
      • Glenn C.A.
      • Martin M.
      • Stoner J.
      • Li J.
      • Puckett T.
      Risk factors for surgical site infection after instrumented fixation in spine trauma.
      ,
      • Ehlers A.P.
      • Khor S.
      • Shonnard N.
      • Oskouian Jr, R.J.
      • Sethi R.K.
      • Cizik A.M.
      • et al.
      Intra-wound antibiotics and infection in spine fusion surgery: a report from Washington State's SCOAP-CERTAIN collaborative.
      ,
      • Pull ter Gunne A.F.
      • van Laarhoven C.J.
      • Cohen D.B.
      Surgical site infection after osteotomy of the adult spine: does type of osteotomy matter?.
      ,
      • Swank S.M.
      • Cohen D.S.
      • Brown J.C.
      Spine fusion in cerebral palsy with L-rod segmental spinal instrumentation. A comparison of single and two-stage combined approach with Zielke instrumentation.
      ,
      • Koutsoumbelis S.
      • Hughes A.P.
      • Girardi F.P.
      • et al.
      Risk factors for postoperative infection following posterior lumbar instrumented arthrodesis.
      ].
      Staphylococcus aureus and Staphylococcus epidermidis are the most common organisms associated with SSIs [
      • Massie J.B.
      • Heller J.G.
      • Abitbol J.J.
      • McPherson D.
      • Garfin S.R.
      Postoperative posterior spinal wound infections.
      ,
      • Gerometta A.
      • Rodriguez Olaverri J.C.
      • Bitan F.
      Infections in spinal instrumentation.
      ]; therefore, the use of intrawound vancomycin is considered to reduce the frequency of SSIs by killing these gram-positive cocci. However, the results of published studies have been inconsistent [
      • Ehlers A.P.
      • Khor S.
      • Shonnard N.
      • Oskouian Jr, R.J.
      • Sethi R.K.
      • Cizik A.M.
      • et al.
      Intra-wound antibiotics and infection in spine fusion surgery: a report from Washington State's SCOAP-CERTAIN collaborative.
      ,
      • Sweet F.A.
      • Roh M.
      • Sliva C.
      Intrawound application of vancomycin for prophylaxis in instrumented thoracolumbar fusions: efficacy, drug levels, and patient outcomes.
      ,
      • O'Neill K.R.
      • Smith J.G.
      • Abtahi A.M.
      • Archer K.R.
      • Spengler D.M.
      • McGirt M.J.
      • et al.
      Reduced surgical site infections in patients undergoing posterior spinal stabilization of traumatic injuries using vancomycin powder.
      ,
      • Caroom C.
      • Tullar J.M.
      • Benton Jr, E.G.
      • Jones J.R.
      • Chaput C.D.
      Intrawound vancomycin powder reduces surgical site infections in posterior cervical fusion.
      ,
      • Strom R.G.
      • Pacione D.
      • Kalhorn S.P.
      • Frempong-Boadu A.K.
      Decreased risk of wound infection after posterior cervical fusion with routine local application of vancomycin powder.
      ,
      • Theologis A.A.
      • Demirkiran G.
      • Callahan M.
      • Pekmezci M.
      • Ames C.
      • Deviren V.
      Local intrawound vancomycin powder decreases the risk of surgical site infections in complex adult deformity reconstruction: a cost analysis.
      ]. Moreover, most published data have been based on observational before–after analyses at one center.
      The purpose of the present study was to evaluate if local vancomycin powder could prevent SSIs after spinal surgery with posterior instrumentation in a large-scale multicenter study. We used propensity score matching to adjust for patient characteristics and operative data. We also analyzed which infectious organisms caused the SSIs to assess whether the distribution of infectious organisms changed through the use of intrawound vancomycin powder.

      Materials and methods

      Patient selection

      We conducted a surveillance study of SSIs after spinal surgery from July 2012 to December 2014; all spinal surgeries performed at 11 hospitals were registered. We included patients aged ≥15 years who underwent surgeries with posterior instrumentation. We did not differentiate according to surgical site (cervical, thoracic, or lumbar spine), and included combination surgeries using the anterior and posterior approaches. Patients who underwent surgeries to treat infective pathologies (including extradural abscess and pyogenic discitis), those who underwent surgeries only for rod elongation, and those who did not provide consent to participate in the study were excluded. Each patient had undergone follow-up for a minimum of 1 year. The institutional review boards of the participating hospitals approved the present surveillance study, and informed consent was obtained from each patient.

      Standard preoperative and postoperative systemic prophylactic antibiotic regimen

      Patients received standard systemic antibiotic prophylaxis consisting of intravenous (IV) cefazolin before surgical incision and additional intraoperative IV cefazolin if needed [
      • Mangram A.J.
      • Horan T.C.
      • Pearson M.L.
      • Silver L.C.
      • Jarvis W.R.
      Guideline for prevention of surgical site infection, 1999.
      ,
      • Bratzler D.W.
      • Dellinger E.P.
      • Olsen K.M.
      • Perl T.M.
      • Auwaerter P.G.
      • Bolon M.K.
      • et al.
      Clinical practice guidelines for antimicrobial prophylaxis in surgery.
      ]. Postoperative IV cefazolin was also used for at least 24 hours after surgery. If the patient was allergic to beta-lactams, clindamycin was administered instead. Other infrequently used intravenous antibiotics included fosfomycin, minomycin, piperacillin, and vancomycin.

      Treatment and control cohorts

      We grouped the patients into two cohorts. The vancomycin group consisted of patients who received 1 or 2 g of intraoperative vancomycin powder, spread throughout the local wound immediately before closure. The powder was placed directly on the implants and muscles. The control group consisted of patients who did not receive intrawound vancomycin powder. No cases lacked documentation of affirmative or negative usage of intrawound vancomycin. The decision to use vancomycin and the amount were based on the surgeon's preference.

      Diagnostic evaluation

      Surgical site infections were identified using the 1999 definitions of the Centers for Disease Control and Prevention [
      • Mangram A.J.
      • Horan T.C.
      • Pearson M.L.
      • Silver L.C.
      • Jarvis W.R.
      Guideline for prevention of surgical site infection, 1999.
      ]. We did not distinguish between the extent of the SSIs (superficial, deep, or organ or space) in the present study. All patients participating in the study underwent standard laboratory tests on admission to the hospital, including peripheral white blood cell counts, C-reactive protein levels, and a complete urine analysis. Abnormal C-reactive protein levels were based on the cut-off value of each hospital (0.3–0.9 mg/dL). Bacterial identification and susceptibility testing for SSI cases were performed according to the 1999 guidelines of the Centers for Disease Control and Prevention [
      • Mangram A.J.
      • Horan T.C.
      • Pearson M.L.
      • Silver L.C.
      • Jarvis W.R.
      Guideline for prevention of surgical site infection, 1999.
      ].

      Clinical parameters

      Pre- and intraoperative data were collected for each patient. The documented risk factors for SSI included body mass index, American Society of Anesthesiologists physical status classification, diagnosis of diabetes mellitus, hemodialysis, current smoking habits, chronic steroid or other immunosuppressant use, and previous surgery at the same site. Pertinent operative details, including estimated blood loss, operating time, and emergency surgery, as well as the use of autologous bone grafting, endoscopy, microscopy, fluoroscopy, and biological clean rooms, were also collected.

      Statistical analysis

      During our preliminary analysis, we compared backgrounds and SSI rates of the vancomycin and control groups. The null hypothesis was that no difference would be found in the postoperative SSI rate between patients who were treated with intrawound vancomycin powder and those who were not. Because of marked differences in the observed characteristics between the two groups, we established a one-to-one matching model using propensity scores to make the two groups directly comparable. In this model, each patient in the vancomycin group was closely matched with a patient from the control group, according to the likelihood that they would receive intrawound vancomycin powder. To estimate the propensity score, we fitted a logistic regression model for the use of local vancomycin powder as a function of 18 variables, including patient characteristics and operative data (Table 1). The C-statistic for evaluating the goodness of fit was calculated. Each patient of the vancomycin group was matched with a patient from the control group with the closest estimated propensity on the logit scale within a specified range (≤0.25 of the pooled standard deviation of estimated logits) to reduce differences between treatment groups by at least 90%. If two or more patients in the control group met this criterion, then we randomly selected one patient for matching.
      Table 1Variables used to fit a logistic regression model to calculate propensity scores
      Patient characteristics
       Gender, age, BMI, ASA PS, diagnosis of DM, previous surgery of the same site
       Smoking habit, current steroid use, hemodialysis, current immunosuppressant use
      Operative details
       EBL, operative time, emergency surgery
       Use of autologous bone grafting, endoscope, microscope, fluoroscope, BCR
      BMI, body mass index; ASA PS, American Society of Anesthesiologists physical status classification; DM, diabetes mellitus; EBL, estimated blood loss; BCR, biological clean room.
      During the descriptive analyses of unmatched and matched cohorts, we compared patient characteristics, operative data, and postoperative SSI rates. We used the chi-square test to compare nominal scales, except for data including an expected cell frequency ≤5; in these cases, the Fisher exact test was performed. The Mann-Whitney U test was used to analyze ordinal scales. The threshold for significance was set at p<.05. All statistical analyses were conducted using JMP, version 13 (SAS Institute Japan, Tokyo, Japan).

      Results

      Analysis of the whole cohort

      Table 2 shows the characteristics of the initial study cohort of 2,859 patients before they were matched according to their propensity scores to receive intraoperative vancomycin powder. In the sample of unmatched patients, 694 patients (24.3%) received intrawound vancomycin powder (vancomycin group) and 2,165 patients (75.7%) did not (control group).
      Table 2Clinical and surgical data of unmatched patients
      n=2,859Vancomycin: 694 patientsControl: 2,165 patientsp
      Characteristics
       Gender
      Male, no. (%)326 (47.0%)999 (46.3%)
      Female, no. (%)368 (53.0%)1,161 (53.8%).74
       Age (mean), y68.565.0<.0001
       American Society of Anesthesiologists physical status classification<.0001
      1107 (15.9%)370 (14.1%)
      2452 (67.3%)1,384 (52.7%)
      3110 (16.4%)200 (7.6%)
      43 (0.5%)0
       BMI (mean)23.223.5.07
       Diabetes mellitus146 (21.1%)225 (10.4%)<.0001
       Previous surgery of the same site271 (39.0%)346 (16.0%)<.0001
       Smoking habit40 (5.8%)213 (9.9%).001
       Current steroid use46 (6.6%)98 (4.5%).03
       Hemodialysis43 (6.2%)52 (2.4%)<.0001
       Immunosuppressant use29 (4.2%)48 (2.2%).006
      Operative details
       EBL (mean), mL759458<.0001
       Operative time (mean), min243210<.0001
       Autologous bone grafting124 (18.1%)623 (28.9%)<.0001
       Endoscopy26 (3.8%)508 (23.5%)<.0001
       Microscopy9 (1.4%)9 (0.4%).007
       Fluoroscopy88 (12.7%)907 (41.9%)<.0001
       Biological clean room176 (25.4%)733 (33.9%)<.0001
       Emergency surgery53 (7.7%)98 (4.5%).001
      Outcomes
       SSI12 (1.73%)21 (0.97%).10
      BMI, body mass index; EBL, estimated blood loss; SSI, surgical site infection.
      Patient characteristics and operative data were significantly different between the two groups. Patients who received intrawound vancomycin powder were significantly older, had more comorbidities such as diabetes mellitus or chronic kidney disease, were more likely to have had previous surgeries at the same site, and had longer surgeries with greater blood loss.
      During the preliminary analysis of the unmatched cohorts, the vancomycin group had a higher SSI rate (n=12; 1.73%) than the control group (n=21; 0.97%; p=.10); however, this difference was not significant.

      Analysis of the matched cohorts

      Using one-to-one propensity score matching, 507 pairs of patients in the control and vancomycin groups were selected. Table 3 shows the characteristics of the matched cohorts. The C-statistic for the goodness of fit was 0.800 in the propensity score model. After matching, no statistically significant differences in patient characteristics and operative data were noted between the two groups.
      Table 3Clinical and surgical data of propensity score-matched patients
      n=1,014Vancomycin: 507 patientsControl: 507 patientsp
      Characteristics
       Gender
      Male, no. (%)234 (46.2%)233 (46.0%)
      Female, no. (%)273 (53.9%)274 (54.0%).95
       Age (mean), y67.567.1.99
       American Association of Anesthesiologists physical status classification.73
      180 (15.8%)81 (16.0%)
      2348 (68.6%)338 (66.7%)
      379 (15.6%)88 (17.4%)
      4
       BMI (mean)23.123.3.22
      Diabetes mellitus92 (18.2%)96 (18.9%).75
       Previous surgery of the same site150 (29.6%)149 (29.4%).95
       Smoking habit26 (5.1%)29 (5.7%).68
       Current steroid use33 (6.5%)35 (6.9%).80
       Hemodialysis27 (5.3%)26 (5.1%).89
       Immunosuppressant use25 (4.9%)25 (4.9%)1.00
      Operative details
       EBL (mean), mL678652.24
       Operative time (mean), min244239.23
       Autologous bone grafting84 (16.6%)74 (14.6%).39
       Endoscope11 (2.2%)9 (1.8%).65
       Microscope7 (1.4%)4 (0.8%).36
       Fluoroscope66 (13.0%)59 (11.6%).50
       Biological clean room148 (29.2%)136 (26.8%).40
       Emergency surgery46 (9.1%)47 (9.3%).91
      Outcomes
       SSI8 (1.58%)9 (1.78%).81
      BMI, body mass index; EBL, estimated blood loss; SSI, surgical site infection.
      In the matched cohorts, eight (1.58%) and nine (1.78%) patients in the vancomycin and control groups developed SSIs, respectively. Although the incidence of SSIs was lower in the vancomycin group than in the control group, the difference was not significant (p=.81, Fig. 1).
      Fig. 1
      Fig. 1Incidence of postoperative surgical site infections in patients who received and who did not receive intrawound vancomycin powder in the propensity score-matched cohorts.

      Microbiology

      A total of 26 pathogens were detected in 33 SSI cases. Detailed descriptions of all SSIs are shown in Table 4, Table 5 for the vancomycin group and control group, respectively.
      Table 4Patient characteristics and microbiology reports of infected cases in the vancomycin group
      PtAgeGenderBMIDMOther risk factorsOperative time (min)EBL (mL)Primary causative organism
      173F23.3Previous surgery155240MSSA
      285F22.1204300MRSA
      371F16.4Immunosuppressant and steroid use239450MRCNS
      460F21.3Hemodialysis, previous surgery234415MRCNS
      575F21.1Hemodialysis, previous surgery215300MRCNS
      668F21.9Previous surgery2121,300P. aeruginosa
      771M26.1Previous surgery201680P. aeruginosa
      874M28.7Immunosuppressant and steroid use181350P. aeruginosa
      963M22.8Previous surgery144440Enterococcus faecalis
      1061F30.8225300Anaerobic gram-negative bacilli
      1172F20.1Steroid use, previous surgery234200Unknown
      1276F20.319950Unknown
      Pt, patient; BMI, body mass index; DM, diabetes mellitus; F, female; M, male; MSSA, methicillin-susceptible Staphylococcus aureus; MRSA, methicillin-resistant Staphylococcus aureus; MRCNS, methicillin-resistant coagulase-negative staphylococci; P. aeruginosa, Pseudomonas aeruginosa.
      Table 5Patient characteristics and microbiology reports of infected cases in the control group
      PtAgeGenderBMIDMOther risk factorsOperative time (min)EBL (mL)Primary causative organism
      1378F23.5156230MSSA
      1478F24.2Previous surgery143430MSSA
      1575F17.8230400MSSA
      1672M23.3Smoking4751,430MSCNS
      1715M14.8290735MSCNS
      1838M43.8449850MRSA
      1982M25.8Previous surgery302300MRSA
      2020F30.43051765MRSA
      2173M24.4174120MRCNS
      2261M20.7Previous surgery343256MRCNS
      2372F20.4Immunosuppressant and steroid use640750MRCNS
      2476M24.6+5471,300P. aeruginosa
      2554M27.2+Hemodialysis, previous surgery311490P. aeruginosa
      2681F24.7162100P. aeruginosa
      2731F24.0Previous surgery7891,520Enterobacter cloacae
      2865F26.9+187305Finegoldia magna
      2958M21.2188400Gram-positive bacilli
      3047M24.2Smoking1671,979Unknown
      3181M26.7203950Unknown
      3278F15.3204110Unknown
      3365M31.2226300Unknown
      Pt, patient; BMI, body mass index; DM, diabetes mellitus; F, female; M, male; MSSA, methicillin-susceptible Staphylococcus aureus; MSCSN, methicillin-susceptible coagulase-negative staphylococci; MRSA, methicillin-resistant Staphylococcus aureus; MRCNS, methicillin-resistant coagulase-negative staphylococci; P. aeruginosa, Pseudomonas aeruginosa.
      The most common organisms cultured in both groups were Staphylococcus species, including S. aureus (both methicillin-susceptible S. aureus and methicillin-resistant S. aureus [MRSA]) and coagulase-negative staphylococci. The rate of Staphylococcus species in SSI cases was 42% (n=5) in the vancomycin group and 52% (n=11) in the control group (Fig. 2). These ratios were not significantly different between the two groups (p=.72, Fisher exact test). There were three MRSA infections in the control group and one MRSA infection in the vancomycin group.
      Fig. 2
      Fig. 2Causative organisms in infected cases. MSSA, methicillin-susceptible Staphylococcus aureus; MRSA, methicillin-resistant Staphylococcus aureus; Staph., Staphylococcus species; GNRs, gram-negative rods.
      The rate of gram-negative rods, mainly Pseudomonas aeruginosa, in SSI cases was higher in the vancomycin group (n=4; 33%) than in the control group (n=4; 19%). However, the rates were not significantly different between the two groups (p=.42, Fisher exact test).

      Discussion

      The use of intrawound vancomycin powder has rapidly spread among spinal surgeons since Sweet et al. [
      • Sweet F.A.
      • Roh M.
      • Sliva C.
      Intrawound application of vancomycin for prophylaxis in instrumented thoracolumbar fusions: efficacy, drug levels, and patient outcomes.
      ] first reported its potential ability to decrease the incidence of SSIs after spinal surgeries. Ehlers et al. [
      • Ehlers A.P.
      • Khor S.
      • Shonnard N.
      • Oskouian Jr, R.J.
      • Sethi R.K.
      • Cizik A.M.
      • et al.
      Intra-wound antibiotics and infection in spine fusion surgery: a report from Washington State's SCOAP-CERTAIN collaborative.
      ] reported that intrawound antibiotics were used in 55% of spinal fusion surgeries performed at 20 Washington State hospitals, despite the paucity of robust data supporting its use.
      This multicenter retrospective cohort study aimed to assess if local intrawound vancomycin powder could prevent SSIs after posterior instrumented spinal surgeries using propensity score matching. Our data revealed that intrawound vancomycin powder did not result in a significant decrease in SSIs. These results are not consistent with those of most of the existing studies, which have reported decreased rates of postoperative SSIs with the use of intrawound vancomycin [
      • Sweet F.A.
      • Roh M.
      • Sliva C.
      Intrawound application of vancomycin for prophylaxis in instrumented thoracolumbar fusions: efficacy, drug levels, and patient outcomes.
      ,
      • O'Neill K.R.
      • Smith J.G.
      • Abtahi A.M.
      • Archer K.R.
      • Spengler D.M.
      • McGirt M.J.
      • et al.
      Reduced surgical site infections in patients undergoing posterior spinal stabilization of traumatic injuries using vancomycin powder.
      ,
      • Caroom C.
      • Tullar J.M.
      • Benton Jr, E.G.
      • Jones J.R.
      • Chaput C.D.
      Intrawound vancomycin powder reduces surgical site infections in posterior cervical fusion.
      ,
      • Strom R.G.
      • Pacione D.
      • Kalhorn S.P.
      • Frempong-Boadu A.K.
      Decreased risk of wound infection after posterior cervical fusion with routine local application of vancomycin powder.
      ,
      • Theologis A.A.
      • Demirkiran G.
      • Callahan M.
      • Pekmezci M.
      • Ames C.
      • Deviren V.
      Local intrawound vancomycin powder decreases the risk of surgical site infections in complex adult deformity reconstruction: a cost analysis.
      ]. Three meta-analyses concluded that intrawound vancomycin powder was associated with a decreased risk of SSIs after spinal surgeries [
      • Khan N.R.
      • Thompson C.J.
      • DeCuypere M.
      • et al.
      A meta-analysis of spinal surgical site infection and vancomycin powder.
      ,
      • Chiang H.Y.
      • Herwaldt L.A.
      • Blevins A.E.
      • Cho E.
      • Schweizer M.L.
      Effectiveness of local vancomycin powder to decrease surgical site infections: a meta-analysis.
      ,
      • Bakhsheshian J.
      • Dahdaleh N.S.
      • Lam S.K.
      • Savage J.W.
      • Smith Z.A.
      The use of vancomycin powder in modern spine surgery: systematic review and meta-analysis of the clinical evidence.
      ]. In contrast, two randomized controlled trials (RCTs) reported no significant changes in the rate of SSIs. Tubaki et al. [
      • Tubaki V.R.
      • Rajasekaran S.
      • Shetty A.P.
      Effects of using intravenous antibiotic only versus local intrawound vancomycin antibiotic powder application in addition to intravenous antibiotics on postoperative infection in spine surgery in 907 patients.
      ] compared open spinal surgeries at any level with and without the use of intrawound vancomycin powder. In their RCT involving 907 patients, they found no statistically significant decrease in the rate of postoperative SSIs (vancomycin group 1.61% vs. control group 1.68%). Mirzashahi et al. [
      • Mirzashahi B.
      • Chehrassan M.
      • Mortazavi S.M.J.
      Intrawound application of vancomycin changes the responsible germ in elective spine surgery without significant effect on the rate of infection: a randomized prospective study.
      ] conducted another RCT that demonstrated a relative increase in SSI rates with the use of intrawound vancomycin (vancomycin group 5.2% vs. control group 2.7%; p=.2), although their method of applying vancomycin was not identical to that of previous studies. They applied vancomycin powder after closure of the fascia, whereas other studies reported that it was applied in the subfascial layer [
      • Sweet F.A.
      • Roh M.
      • Sliva C.
      Intrawound application of vancomycin for prophylaxis in instrumented thoracolumbar fusions: efficacy, drug levels, and patient outcomes.
      ,
      • O'Neill K.R.
      • Smith J.G.
      • Abtahi A.M.
      • Archer K.R.
      • Spengler D.M.
      • McGirt M.J.
      • et al.
      Reduced surgical site infections in patients undergoing posterior spinal stabilization of traumatic injuries using vancomycin powder.
      ,
      • Caroom C.
      • Tullar J.M.
      • Benton Jr, E.G.
      • Jones J.R.
      • Chaput C.D.
      Intrawound vancomycin powder reduces surgical site infections in posterior cervical fusion.
      ,
      • Strom R.G.
      • Pacione D.
      • Kalhorn S.P.
      • Frempong-Boadu A.K.
      Decreased risk of wound infection after posterior cervical fusion with routine local application of vancomycin powder.
      ,
      • Theologis A.A.
      • Demirkiran G.
      • Callahan M.
      • Pekmezci M.
      • Ames C.
      • Deviren V.
      Local intrawound vancomycin powder decreases the risk of surgical site infections in complex adult deformity reconstruction: a cost analysis.
      ,
      • Emohare O.
      • Ledonio C.G.
      • Hill B.W.
      • Davis R.A.
      • Polly Jr, D.W.
      • Kang M.M.
      Cost savings analysis of intrawound vancomycin powder in posterior spinal surgery.
      ,
      • Hey H.W.
      • Thiam D.W.
      • Koh Z.S.
      • Thambiah J.S.
      • Kumar N.
      • Lau L.L.
      • et al.
      Is intraoperative local vancomycin powder the answer to surgical site infections in spine surgery?.
      ,
      • Chotai S.
      • Wright P.W.
      • Hale A.T.
      • Jones W.A.
      • McGirt M.J.
      • Patt J.C.
      • et al.
      Does intrawound vancomycin application during spine surgery create vancomycin-resistant organism?.
      ].
      Similarly, some observational studies reported no significant decrease in the rate of SSIs with the use of intrawound vancomycin. Martin et al. [
      • Martin J.R.
      • Adogwa O.
      • Brown C.R.
      • Bagley C.A.
      • Richardson W.J.
      • Lad S.P.
      • et al.
      Experience with intrawound vancomycin powder for spinal deformity surgery.
      ] reported no significant decrease in the rate of deep SSIs (vancomycin group 5.1% vs. control group 5.3%; p=.936) in their observational study including 306 patients. Gaviola et al. [
      • Gaviola M.L.
      • McMillian W.D.
      • Ames S.E.
      • Endicott J.A.
      • Alston W.K.
      A retrospective study on the protective effects of topical vancomycin in patients undergoing multilevel spinal fusion.
      ] also reported similar results (vancomycin group 5.2% vs. control group 11.0%; p=.08) in their observational study involving 326 patients.
      However, these studies may lack statistical power owing to the relatively small number of patients. The estimated sample size needed for the RCT was approximately 4,600 participants. This number was based on the assumption that the SSI rate of the control cohort is 2.0% for posterior instrumented spinal surgeries [
      • Imajo Y.
      • Taguchi T.
      • Yone K.
      • Okawa A.
      • Otani K.
      • Ogata T.
      • et al.
      Japanese 2011 nationwide survey on complications from spine surgery.
      ], the rate of Staphylococcus species in SSI cases is 70% [
      • Massie J.B.
      • Heller J.G.
      • Abitbol J.J.
      • McPherson D.
      • Garfin S.R.
      Postoperative posterior spinal wound infections.
      ,
      • Gerometta A.
      • Rodriguez Olaverri J.C.
      • Bitan F.
      Infections in spinal instrumentation.
      ], and that two-thirds of staphylococcal infections are prevented by local application of vancomycin powder; consequently, the SSI rate of the vancomycin cohort was estimated to be 1.0%. In this estimation, we also assumed that the probability of falsely rejecting a true null hypothesis (α) was 0.05, and that the probability of failing to reject a false null hypothesis (1 − β) was 0.80.
      Our preliminary analysis of the whole cohort showed a relative increase in the SSI rate in the vancomycin group (1.73% in the vancomycin group vs. 0.97% in the control group; p=.10). This was likely due to patients in the vancomycin group being at significantly higher risk of SSIs than patients in the control group, as shown in Table 2. This difference could have resulted from most surgeons' preference of using intrawound vancomycin for high-risk populations.
      To adjust for differences in patient characteristics and operative data between the two groups, we calculated propensity scores and established one-to-one matched cohorts. Patient parameters and surgical data were not significantly different between the two matched groups (p>.05 for each variable, Table 3). During the matched analysis, the vancomycin group showed a lower incidence of SSIs than the control group, but the difference was not significant (1.58% in the vancomycin group vs. 1.78% in the control group; p=.81). The sample size of our study might have been too small to detect whether vancomycin powder could prevent SSIs, although the estimation of the required sample size for propensity score matching is not considered the same as that for RCTs.
      The decrease in the rate of infections by Staphylococcus species compared with all other infectious organisms in SSI cases in the vancomycin group in the present study (Fig. 2) might indicate the ability of vancomycin to prevent Staphylococcus infections. Conversely, the rate of infection caused by gram-negative rods increased in the vancomycin group. Other authors also reported comparable results for SSI-causative organisms [
      • Chotai S.
      • Wright P.W.
      • Hale A.T.
      • Jones W.A.
      • McGirt M.J.
      • Patt J.C.
      • et al.
      Does intrawound vancomycin application during spine surgery create vancomycin-resistant organism?.
      ,
      • Martin J.R.
      • Adogwa O.
      • Brown C.R.
      • Bagley C.A.
      • Richardson W.J.
      • Lad S.P.
      • et al.
      Experience with intrawound vancomycin powder for spinal deformity surgery.
      ]. The small increase in infections caused by gram-negative rods could have compensated for the decrease in Staphylococcus infections in the vancomycin group, thereby resulting in the absence of a significant change in the SSI rate caused by all pathogens through the use of intrawound vancomycin powder.
      The present study was based on data collected prospectively in the surveillance study of SSIs. Therefore, it contains information regarding several potential risk factors for spinal SSIs, including current smoking habits and chronic steroid or other immunosuppressant use, as well as body mass index, American Society of Anesthesiologists physical status, diagnosis of diabetes mellitus, and hemodialysis. In the propensity score-matched analysis, all these factors were adjusted to ensure an even distribution in both groups. Nonetheless, there was no significant decrease in the SSI rate in the vancomycin group.
      The present study had several limitations. First, we did not record the amount of applied vancomycin. Although the most commonly used dose of intrawound vancomycin is 1 g [
      • Caroom C.
      • Tullar J.M.
      • Benton Jr, E.G.
      • Jones J.R.
      • Chaput C.D.
      Intrawound vancomycin powder reduces surgical site infections in posterior cervical fusion.
      ,
      • Strom R.G.
      • Pacione D.
      • Kalhorn S.P.
      • Frempong-Boadu A.K.
      Decreased risk of wound infection after posterior cervical fusion with routine local application of vancomycin powder.
      ,
      • Theologis A.A.
      • Demirkiran G.
      • Callahan M.
      • Pekmezci M.
      • Ames C.
      • Deviren V.
      Local intrawound vancomycin powder decreases the risk of surgical site infections in complex adult deformity reconstruction: a cost analysis.
      ], some authors reported that they changed the dose according to the length of the incision [
      • Chotai S.
      • Wright P.W.
      • Hale A.T.
      • Jones W.A.
      • McGirt M.J.
      • Patt J.C.
      • et al.
      Does intrawound vancomycin application during spine surgery create vancomycin-resistant organism?.
      ]. The appropriate dose of vancomycin has not yet been determined. Second, we did not record some factors that might be associated with SSIs, such as the diagnosis of coronary heart disease, chronic obstructive pulmonary disease, or osteoporosis [
      • Koutsoumbelis S.
      • Hughes A.P.
      • Girardi F.P.
      • et al.
      Risk factors for postoperative infection following posterior lumbar instrumented arthrodesis.
      ], and the surgeons' experience [
      • Olsen M.A.
      • Nepple J.J.
      • Riew K.D.
      • Lenke L.G.
      • Bridwell K.H.
      • Mayfield J.
      • et al.
      Risk factors for surgical site infection following orthopaedic spinal operations.
      ]. As these unrecorded factors cannot be adjusted via propensity score matching, they could have affected the SSI rates. This is a limitation attributable to the retrospective nature of the present study. Finally, the study did not include any patient-reported outcome scores, which might have been beneficial to the overall assessment of patients. A randomized, prospective study that includes patient-reported outcomes is needed.
      Despite these limitations, our study indicates that the use of intrawound vancomycin powder does not reduce the incidence of postoperative SSIs in posterior instrumented spinal surgeries. Our data also suggest that the distribution of causative organisms might change with the use of intrawound vancomycin powder.

      Conclusions

      The intrawound application of vancomycin powder was not associated with a significant decrease in the rate of SSIs after posterior instrumented spinal surgery. The rate of infections by Staphylococcus species was lower in the vancomycin group than in the control group, but the difference was not significant. The small increase in infections caused by gram-negative rods could have compensated for the decrease in Staphylococcus infections in the vancomycin group, leading to the non-significant change in the overall SSI rate between the two groups.

      References

        • de Lissovoy G.
        • Fraeman K.
        • Hutchins V.
        • Stoner J.
        • Li J.
        • Puckett T.
        Surgical site infection: incidence and impact on hospital utilization and treatment costs.
        Am J Infect Control. 2009; 37: 387-397
        • Kirkland K.B.
        • Briggs J.P.
        • Trivette S.L.
        • Angotti J.M.
        • Kalobwe E.
        • Muhlbauer M.S.
        • et al.
        The impact of surgical-site infections in the 1990s: attributable mortality, excess length of hospitalization, and extra costs.
        Infect Control Hosp Epidemiol. 1999; 20: 725-730
        • Imajo Y.
        • Taguchi T.
        • Yone K.
        • Okawa A.
        • Otani K.
        • Ogata T.
        • et al.
        Japanese 2011 nationwide survey on complications from spine surgery.
        J Orthop Sci. 2015; 20: 38-54
        • Cooper K.
        • Glenn C.A.
        • Martin M.
        • Stoner J.
        • Li J.
        • Puckett T.
        Risk factors for surgical site infection after instrumented fixation in spine trauma.
        J Clin Neurosci. 2016; 23: 123-127
        • Ehlers A.P.
        • Khor S.
        • Shonnard N.
        • Oskouian Jr, R.J.
        • Sethi R.K.
        • Cizik A.M.
        • et al.
        Intra-wound antibiotics and infection in spine fusion surgery: a report from Washington State's SCOAP-CERTAIN collaborative.
        Surg Infect (Larchmt). 2016; 17: 179-186
        • Pull ter Gunne A.F.
        • van Laarhoven C.J.
        • Cohen D.B.
        Surgical site infection after osteotomy of the adult spine: does type of osteotomy matter?.
        Spine J. 2010; 10: 410-416
        • Swank S.M.
        • Cohen D.S.
        • Brown J.C.
        Spine fusion in cerebral palsy with L-rod segmental spinal instrumentation. A comparison of single and two-stage combined approach with Zielke instrumentation.
        Spine. 1989; 14: 750-759
        • Koutsoumbelis S.
        • Hughes A.P.
        • Girardi F.P.
        • et al.
        Risk factors for postoperative infection following posterior lumbar instrumented arthrodesis.
        J Bone Joint Surg Am. 2011; 93: 1627-1633
        • Massie J.B.
        • Heller J.G.
        • Abitbol J.J.
        • McPherson D.
        • Garfin S.R.
        Postoperative posterior spinal wound infections.
        Clin Orthop Relat Res. 1992; 284: 99-108
        • Gerometta A.
        • Rodriguez Olaverri J.C.
        • Bitan F.
        Infections in spinal instrumentation.
        Int Orthop. 2012; 36: 457-464
        • Sweet F.A.
        • Roh M.
        • Sliva C.
        Intrawound application of vancomycin for prophylaxis in instrumented thoracolumbar fusions: efficacy, drug levels, and patient outcomes.
        Spine. 2011; 36: 2084-2088
        • O'Neill K.R.
        • Smith J.G.
        • Abtahi A.M.
        • Archer K.R.
        • Spengler D.M.
        • McGirt M.J.
        • et al.
        Reduced surgical site infections in patients undergoing posterior spinal stabilization of traumatic injuries using vancomycin powder.
        Spine J. 2011; 11: 641-646
        • Caroom C.
        • Tullar J.M.
        • Benton Jr, E.G.
        • Jones J.R.
        • Chaput C.D.
        Intrawound vancomycin powder reduces surgical site infections in posterior cervical fusion.
        Spine. 2013; 38: 1183-1187
        • Strom R.G.
        • Pacione D.
        • Kalhorn S.P.
        • Frempong-Boadu A.K.
        Decreased risk of wound infection after posterior cervical fusion with routine local application of vancomycin powder.
        Spine. 2013; 38: 991-994
        • Theologis A.A.
        • Demirkiran G.
        • Callahan M.
        • Pekmezci M.
        • Ames C.
        • Deviren V.
        Local intrawound vancomycin powder decreases the risk of surgical site infections in complex adult deformity reconstruction: a cost analysis.
        Spine. 2014; 39: 1875-1880
        • Mangram A.J.
        • Horan T.C.
        • Pearson M.L.
        • Silver L.C.
        • Jarvis W.R.
        Guideline for prevention of surgical site infection, 1999.
        Infect Control Hosp Epidemiol. 1999; 20: 247-280
        • Bratzler D.W.
        • Dellinger E.P.
        • Olsen K.M.
        • Perl T.M.
        • Auwaerter P.G.
        • Bolon M.K.
        • et al.
        Clinical practice guidelines for antimicrobial prophylaxis in surgery.
        Surg Infect (Larchmt). 2013; 14: 73-156
        • Khan N.R.
        • Thompson C.J.
        • DeCuypere M.
        • et al.
        A meta-analysis of spinal surgical site infection and vancomycin powder.
        J Neurosurg Spine. 2014; 21: 974-983
        • Chiang H.Y.
        • Herwaldt L.A.
        • Blevins A.E.
        • Cho E.
        • Schweizer M.L.
        Effectiveness of local vancomycin powder to decrease surgical site infections: a meta-analysis.
        Spine J. 2014; 14: 397-407
        • Bakhsheshian J.
        • Dahdaleh N.S.
        • Lam S.K.
        • Savage J.W.
        • Smith Z.A.
        The use of vancomycin powder in modern spine surgery: systematic review and meta-analysis of the clinical evidence.
        World Neurosurg. 2015; 83: 816-823
        • Tubaki V.R.
        • Rajasekaran S.
        • Shetty A.P.
        Effects of using intravenous antibiotic only versus local intrawound vancomycin antibiotic powder application in addition to intravenous antibiotics on postoperative infection in spine surgery in 907 patients.
        Spine. 2013; 38: 2149-2155
        • Mirzashahi B.
        • Chehrassan M.
        • Mortazavi S.M.J.
        Intrawound application of vancomycin changes the responsible germ in elective spine surgery without significant effect on the rate of infection: a randomized prospective study.
        Musculoskelet Surg. 2018; 102: 35-39
        • Emohare O.
        • Ledonio C.G.
        • Hill B.W.
        • Davis R.A.
        • Polly Jr, D.W.
        • Kang M.M.
        Cost savings analysis of intrawound vancomycin powder in posterior spinal surgery.
        Spine J. 2014; 14: 2710-2715
        • Hey H.W.
        • Thiam D.W.
        • Koh Z.S.
        • Thambiah J.S.
        • Kumar N.
        • Lau L.L.
        • et al.
        Is intraoperative local vancomycin powder the answer to surgical site infections in spine surgery?.
        Spine. 2017; 42: 267-274
        • Chotai S.
        • Wright P.W.
        • Hale A.T.
        • Jones W.A.
        • McGirt M.J.
        • Patt J.C.
        • et al.
        Does intrawound vancomycin application during spine surgery create vancomycin-resistant organism?.
        Neurosurgery. 2017; 80: 746-753
        • Martin J.R.
        • Adogwa O.
        • Brown C.R.
        • Bagley C.A.
        • Richardson W.J.
        • Lad S.P.
        • et al.
        Experience with intrawound vancomycin powder for spinal deformity surgery.
        Spine. 2014; 39: 177-184
        • Gaviola M.L.
        • McMillian W.D.
        • Ames S.E.
        • Endicott J.A.
        • Alston W.K.
        A retrospective study on the protective effects of topical vancomycin in patients undergoing multilevel spinal fusion.
        Pharmacotherapy. 2016; 36: 19-25
        • Olsen M.A.
        • Nepple J.J.
        • Riew K.D.
        • Lenke L.G.
        • Bridwell K.H.
        • Mayfield J.
        • et al.
        Risk factors for surgical site infection following orthopaedic spinal operations.
        J Bone Joint Surg Am. 2008; 90: 62-69