Abstract
Study Design
Retrospective cohort analysis.
Objectives
Surgical site infection (SSI) is a major complication following spine surgery. This is especially true with posterior cervical spinal surgery, contributing to increased hospital stay, socioeconomic burden, morbidity, and mortality. This study aimed to investigate risk factors for cervical SSI, identify organisms responsible, and characterize antibiotic treatment regimens to inform surgeons.
Methods
A retrospective review of patients who underwent posterior cervical decompression and fusion (PCDF) was performed at a single academic quaternary care center. Outcome variables included SSI, culture speciation, and antibiotic regimens. Associations between demographic, clinical, and surgical factors with SSI were analyzed using univariate statistics and multiple logistic regression. A sub-analysis was conducted for gram-negative and gram-positive pathogens. P<0.05 was considered significant.
Results
In 590 patients undergoing PCDF, 47 (8.0%) SSIs were identified, and average time to operative revision was 31.4 days. Coagulase-negative Staphylococcus (20, 42.6%), Staphylococcus aureus (17 [MRSA: 6, MSSA: 11], 36.2%), and Cutibacterium species (10, 21.3%) were most common. Eighteen (38.3%) SSIs were polymicrobial. Increased BMI was associated with SSI on univariate (p=0.018) and multivariate regression (p=0.021). Operative time was associated with increased risk of gram-negative infection (p=0.038).
Conclusion
SSI following PCDF remains a major complication with an incidence of 8%. Elevated BMI was an independent risk factor, underscoring importance of patient optimization. Longer operative time was associated with gram-negative infection in univariate analysis, however, this did not remain significant following multivariable adjustment. These findings highlight the need for awareness of resistant pathogens in tailoring empiric antibiotic regimens.
Keywords: PCDF, SSI, spine, complications, antibiotics
Introduction
Surgical site infection (SSI) is a significant concern following spine surgeries, contributing to extended hospital stays, increased costs, and morbidity.1-3 It remains a leading cause of unplanned 30- and 90-day readmission.4-6 Published work has shown a difference in SSI rates and considerations depending on the operative region, instrumentation, and surgical approach. 7 Prior studies have reported SSIs in thoracolumbar spine surgery with granularity, investigating associated risk factors, outcomes, and microbiologic profiles. 8 However, the literature on SSI after cervical spinal procedures remains sparse. Posterior cervical fusion procedures have significantly higher rates of SSI as opposed to an anterior approach, thought to originate from extensive dissection of the paraspinal muscles and complications associated with tissue re-approximation, as well as risk factors inherent to posterior cervical wounds. 9 Prior studies reveal a total complication rate following posterior cervical decompression and fusion (PCDF) to be between 8.6% to 49.1%, and an estimated incidence of SSI of 2.9% to 10.3%. 10
In most cases, SSI requires return to the operating room for exploration of the surgical wound, debridement and complex closure, and long-term antibiotic treatment. The regimen surrounding SSIs typically involves initial empiric antibiotics and intraoperative cultures to inform a targeted treatment course. Posterior cervical wounds pose unique challenges and are prone to microbiomes that are different from thoracolumbar wounds. 11 Thus, further investigation of these infections may better inform empirical treatment for patients presenting with potential SSI following posterior cervical decompression and fusion. This study aimed to characterize SSI following PCDF at a single, quaternary care institution and characterize the species and resistance profiles of the isolated organisms.
Methods
A retrospective review of patients who underwent PCDF between 2012 and 2022 was performed at a single academic quaternary care center. This study adhered to STROBE reporting guidelines for cohort studies. Institutional Review Board approval (STUDY 230730202) was given with a waiver of informed consent given the retrospective design. The electronic medical record was used to extract demographic, clinical, surgical, and outcome variables. SSI was defined as any case requiring wound washout and/or debridement following posterior cervical fusion, in which a pathogenic organism was isolated from wound cultures. Cases in which the treating surgical and infectious disease teams determined that culture growth represented contamination rather than true infection (i.e., treatment for SSI was not recommended) were excluded. For each SSI, the time to return to the operating room, pathogenic organism, antibiotic type and duration, and resistance matrices were recorded. Antibiotic treatment courses were recorded following culturing, speciation, and sensitivity testing. The drugs used in the primary treatment following microbiological testing were recorded, along with their route of administration: intravenous (IV) and oral (PO). The utilization of suppressive antibiotics following the primary course was also noted, along with the length of suppression being long (≥6 months) or short (<6 months).
Statistical Analysis
Continuous variables were compared using independent t-tests, discrete variables using Mann-Whitney U-tests, and categorical variables using Chi-square or Fisher’s exact tests where appropriate. Logistic regression was used for univariate and multivariate analyses of SSI risk factors. Variables included in multivariate modeling were BMI, age, diabetes, smoking status, indication, and operative time. Separate subgroup analyses were performed for gram-positive versus gram-negative infections. A further analysis of BMI was performed, wherein BMI was split into clinical categories. For all analyses, p-values, along with odds ratios (OR) and 95% confidence intervals (CI) were calculated where applicable. All statistical analyses were performed in R Version 4.3.3 (R Foundation for Statistical Computing, Vienna, Austria) with p < 0.05 considered to be significant.
Results
Clinical and hospitalization records for 590 patients who underwent posterior cervical fusions were evaluated, and 47 infections (8.0%) were identified. Average follow-up was 2.90 ± 2.71 years. The average age was 62.38 ± 12.48 years, and the majority were male (354, 60.1%). The most common indication for surgery was degenerative disease (294, 50.5%). The average operative time was 211.98 ± 104.76 minutes, the average length of stay was 8.27 ± 13.33 days, and the average time to surgical debridement was 31.4 ± 12.75 days (Figure 1).
Figure 1.
Distribution of the time in days between index surgery and surgical debridement procedure for the patients with surgical site infection following posterior cervical decompression and fusion
Between the groups, continuous BMI (p=0.009) and categorical BMI clinical classifications (p=0.002) were significantly different. Full demographics and surgical characteristics are shown in Table 1.
Table 1.
Demographic and Surgical Characteristics of the Study Cohort. Groups are Stratified by Surgical Site Infection (SSI) Postoperatively to Posterior Cervical Decompression and Fusion (PCDF). P-Values are the Result of T-test and Mann Whitney U-Test for Continuous Variables, and Chi Square and Fisher Exact Test for Categorical Variables, Where Appropriate
| Variable | All (n= 590) | No SSI (n=543) | SSI (n=47) | p-value |
|---|---|---|---|---|
| Age, mean ± SD | 62.38 ± 12.48 | 62.53 ± 12.59 | 60.63 ± 11.13 | 0.276 |
| Operative Time (Minutes), mean ± SD | 211.98 ± 104.76 | 210.75 ± 106 | 225.96 ± 89.33 | 0.276 |
| Length of Stay (Days), mean ± SD | 8.27 ± 13.33 | 8.36 ± 13.81 | 7.21 ± 5.23 | 0.234 |
| BMI (kg/m 2 ), mean ± SD | 29.74 ± 7.31 | 29.51 ± 7.35 | 32.29 ± 6.42 | 0.009 |
| BMI category, no. (%) | | | | 0.002 |
| Underweight (BMI <18.5) | 12 (2.3%) | 12 (2.5%) | 0 (0%) | |
| Healthy Weight (BMI 18.5-24.9) | 135 (25.8%) | 128 (26.6%) | 7 (16.3%) | |
| Overweight (BMI 25-29.9) | 156 (29.7%) | 148 (30.8%) | 8 (18.6%) | |
| Obesity Class 1 (BMI 30-34.9) | 112 (21.4%) | 100 (20.8%) | 12 (27.9%) | |
| Obesity Class 2+ (BMI ≥ 35) | 109 (20.8%) | 93 (19.3%) | 16 (37.2%) | |
| Sex, no. (%) | | | | 0.939 |
| Female | 235 (39.9%) | 216 (39.9%) | 19 (40.4%) | |
| Male | 354 (60.1%) | 326 (60.1%) | 28 (59.6%) | |
| Osteoporosis, no. (%) | | | | 0.262 |
| Yes | 27 (4.7%) | 23 (4.4%) | 4 (8.7%) | |
| No | 544 (95.3%) | 502 (95.6%) | 42 (91.3%) | |
| Smoking Status, no. (%) | | | | 0.922 |
| Current | 157 (27.2%) | 144 (27.1%) | 13 (27.7%) | |
| Former | 187 (31.7%) | 173 (32.6%) | 14 (29.8%) | |
| Non-smoker | 234 (40.5%) | 214 (40.3%) | 20 (42.6%) | |
| Heart Disease, no. (%) | | | | 0.703 |
| Yes | 138 (23.4%) | 126 (23.6%) | 12 (26.1%) | |
| No | 442 (76.2%) | 408 (76.4%) | 34 (73.9%) | |
| Diabetes, no. (%) | | | | 0.495 |
| Yes | 147 (25.4%) | 137 (25.8%) | 10 (21.3%) | |
| No | 431 (74.6%) | 394 (74.2%) | 37 (78.7%) | |
| Preoperative Steroids, no. (%) | | | | 0.787 |
| Yes | 48 (8.3%) | 45 (8.5%) | 3 (6.4%) | |
| No | 529 (91.7%) | 485 (91.5%) | 44 (93.6%) | |
| Preoperative Rheumatoid Drugs, no. (%) | | | | 0.192 |
| Yes | 10 (1.7%) | 8 (1.5%) | 2 (4.3%) | |
| No | 567 (98.3%) | 522 (98.5%) | 45 (95.7%) | |
| Indication, no. (%) | | | | 0.173 |
| Degenerative | 294 (50.5%) | 263 (49.2%) | 31 (66%) | |
| Revision | 103 (17.7%) | 97 (18.1%) | 6 (12.8%) | |
| Trauma | 130 (22.3%) | 124 (23.2%) | 6 (12.8%) | |
| Deformity | 55 (9.5%) | 51 (9.5%) | 4 (8.5%) | |
| Prior Cervical Surgery, no. (%) | | | | 0.595 |
| Yes | 200 (35%) | 183 (34.7%) | 17 (38.6%) | |
| No | 372 (65%) | 345 (65.3%) | 27 (61.4%) | |
| Number of Levels, median (IQR) | 8 (6-8) | 8 (6-8) | 8 (6.5-8) | 0.417 |
Values are based on available data. Totals may not sum to the overall sample size because of missing data.
Analysis of culture data revealed Coagulase-negative Staphylococcus (20), Staphylococcus aureus (17), and Cutibacterium species (10) being the most common organisms (Figure 2). Eighteen SSIs included multiple pathogens. Patients with SSI had an average antibiotic course of 32.6 ± 17.5 days. The frequency of antibiotics prescribed in the primary regimen used to treat these infections following pathogen culture/determination is shown in Figure 3, with some patients contributing more than one antibiotic to the prescription totals. The antibiotic regimens used to treat SSI were IV antibiotics only (12, 25.5%), primary IV with long suppressive oral antibiotics (11, 23.4%), primary IV and oral antibiotics followed by long oral suppression (7, 14.9%), oral antibiotics only (7, 14.9%), primary IV followed by short oral suppression (6, 12.8%), primary IV and oral antibiotics followed by short oral suppression (3, 6.4%), and primary oral antibiotics followed by long oral suppression (1, 2.1%) (Figure 4). Ten (21.3%) of the patients were taken to the operating room multiple times for debridement and washout. Sensitivity matrices were recorded for 35 of the positive cultures, with 20 (57%) exhibiting resistance to at least one antibiotic. The most common resistances were oxacillin (11) and erythromycin (6) (Figure 5).
Figure 2.
Pathogenic organisms cultured from surgical site infection following posterior cervical fusion
Figure 3.
Antibiotics prescribed following surgical site infection from posterior cervical fusion
Figure 4.
Antibiotic regimens prescribed following surgical site infection following posterior cervical fusion. Suppressive oral antibiotics ≥ 6 months were considered long suppression and oral suppression regimens < 6 months were considered short suppression. IV: Intravenous, PO: Oral
Figure 5.
Antibiotic-resistant strains of surgical site infection from sensitivity matrices. The horizontal axis represents organisms cultured, the vertical axis contains the antibiotics tested in the sensitivity matrices, and the numbers in the boxes represent the number of patients wherein surgical site cultures revealed a specific organism resistant to that antibiotic
In univariate regression, BMI was positively associated with risk of infection (OR 1.046, 95% CI 1.007-1.086, p=0.018). In multiple logistic regression, BMI was positively associated with risk of infection (OR 1.049, 95% CI 1.006-1.092, p=0.021). The full results of univariate and multiple regression are shown in Table 2.
Table 2.
Factors Affecting Incidence of Surgical Site Infection
| Variable | Odds ratio (95% confidence interval) | P-value |
|---|---|---|
| Univariate Regression | ||
| Heart Disease | 1.143 (0.522-2.345) | 0.703 |
| Diabetes | 0.778 (0.336-1.649) | 0.495 |
| Female | 1.024 (0.526-1.955) | 0.939 |
| Smoker (Former or Recent) | 0.912 (0.479-1.762) | 0.763 |
| Preoperative Steroids | 0.735 (0.141-2.443) | 0.787 |
| Preoperative Rheumatoid Drugs | 2.892 (0.291-15.100) | 0.192 |
| Osteoporosis | 2.075 (0.499-6.484) | 0.262 |
| Prior Cervical Surgery | 1.187 (0.590-2.326) | 0.595 |
| Age | 0.988 (0.966-1.012) | 0.321 |
| Length of Stay (Index Surgery) | 0.983 (0.925-1.012) | 0.516 |
| Body Mass Index | 1.046 (1.007-1.086) | 0.018 |
| Operative Time (minutes) | 1.002 (0.999-1.005) | 0.315 |
| Number of Levels | 1.037 (0.877-1.182) | 0.626 |
| Multiple Regression | ||
| Body Mass Index | 1.049 (1.006-1.092) | 0.021 |
| Age | 0.992 (0.965-1.022) | 0.580 |
| Diabetes | 0.600 (0.258-1.272) | 0.206 |
| Non-smoker | 1# | |
| Smoker - Former | 0.779 (0.352-1.668) | 0.525 |
| Smoker - Current | 0.913 (0.390-2.074) | 0.829 |
| Indication - Degenerative | 1# | |
| Indication - Revision | 0.402 (0.116-1.074) | 0.100 |
| Indication - Trauma | 0.402 (0.116-1.073) | 0.100 |
| Indication - Deformity | 0.672 (0.190-1.847) | 0.482 |
| Operative Time (minutes) | 1.002 (0.998-1.005) | 0.380 |
| BMI Category Analysis (Reference: Healthy Weight, BMI 18.5–24.9) | ||
| Underweight (BMI <18.5) | 0 (0-45971758) | 0.984 |
| Overweight (BMI 25-29.9) | 0.988 (0.346-2.891) | 0.983 |
| Obesity Class 1 (BMI 30-34.9) | 2.194 (0.851-6.087) | 0.112 |
| Obesity Class 2+ (BMI ≥35) | 3.146 (1.288-8.467) | 0.015 |
1#: indicates reference group.
In the subgroup analysis of gram-negative and gram-positive organisms, operative time was positively associated with odds of gram-negative bacterial infection in univariate regression (gram-negative: 260.4 ± 119.3 minutes, gram-positive: 209.8 ± 67.6 minutes, p=0.038, OR 1.005, 95% CI 1.000-1.009), but this value did not reach significance in multiple regression (OR 1.004, 95% CI 0.999-1.009, p=0.060) (Table 3).
Table 3.
Association of Clinical, Demographic, and Surgical Factors With Gram-Negative Infection in SSI Cohort
| Variable | Odds ratio (95% confidence interval) | P-value |
|---|---|---|
| Univariate Regression | ||
| Heart Disease | 1.623 (0.427-5.321) | 0.366 |
| Diabetes | 1.068 (0.244-3.676) | 1.000 |
| Female | 0.540 (0.124-1.852) | 0.289 |
| Smoker | 1.021 (0.319-3.536) | 0.969 |
| Preoperative Steroids | 1.724 (0.183-7.980) | 0.359 |
| Preoperative Rheumatoid Drugs | 4.366 (0.094-35.608) | 0.233 |
| Osteoporosis | 0 (0-6.293) | 1 |
| Prior Cervical Surgery | 0.500 (0.089-1.924) | 0.398 |
| Age | 0.989 (0.954-1.031) | 0.589 |
| Length of Stay (Index Surgery) | 1.000 (0.921-1.020) | 0.993 |
| Body Mass Index | 1.056 (0.992-1.115) | 0.065 |
| Operative Time (minutes) | 1.005 (1.000-1.009) | 0.038 |
| Number of Levels | 1.099 (0.860-1.284) | 0.308 |
| Multiple Regression | ||
| Body Mass Index | 1.056 (0.988-1.122) | 0.087 |
| Age | 1.003 (0.958-1.055) | 0.918 |
| Diabetes | 0.822 (0.217-2.561) | 0.750 |
| Non-smoker | 1# | |
| Smoker - Former | 0.893 (0.221-3.254) | 0.865 |
| Smoker - Current | 1.315 (0.327-5.094) | 0.689 |
| Indication - Degenerative | 1# | |
| Indication - Revision | 0.275 (0.015-1.486) | 0.224 |
| Indication - Trauma | 0.605 (0.090-2.438) | 0.530 |
| Indication - Deformity | 0.444 (0.024-2.441) | 0.447 |
| Operative Time (minutes) | 1.004 (0.999-1.009) | 0.060 |
1#: indicates reference group.
In further analysis of BMI clinical classifications, individuals in Obesity class 2+ (BMI ≥35 kg/m2) had significantly higher odds of an SSI following PCDF as compared to the healthy BMI group (OR 3.146, 95% CI 1.288-8.467, p=0.015) (Table 2).
Discussion
In the present study, a rate of infection of 8.0% was reported after PCDF, which fell within the previously reported SSI incidence range of 2.9-10.3%,1,2,4-6,10,12 and may reflect the complex surgical population treated at this quaternary referral center, including revision surgery, deformity, and trauma cases.7,10 Altogether, SSI may contribute to significant increases in hospital costs and patient morbidity.2,7 The majority of institutions across the U.S. follow prophylactic measures such as chlorhexidine showers before surgery, antibiotics before the incision, extensive irrigation, and vancomycin powder administration to the wound.13,14 In spite of these strategies, there remains little consensus, however, on baseline risk factors and perioperative variables that could predict or treat SSI.
The majority of patients undergoing elective PCDF at this institution receive preoperative evaluation, education, and optimization by a comprehensive preoperative team. While they may be optimized from the perspective of various medical conditions, patient BMI remained a driver of SSI, which was in line with prior smaller reports.10,12,15 In this study, BMI ≥35 in the sub-classification of class 2 or class 3 obesity was significantly associated with SSI. Furthermore, the highest SSI rate was among the BMI ≥ 35 kg/m2 group, wherein the SSI rate was 14.7%. In manifestations of severe cervical disease, particularly with myelopathy, patients can have limited mobilization and become functionally dependent. In these instances, adherence to BMI as a criterion and recommendation of weight loss before moving forward with surgery can be a challenge. Extensive informed discussion with the patient is needed regarding risks and benefits of the operation. Otherwise, with nonurgent cervical spinal pathology and BMI over 35, they should be enrolled in intensive weight loss programs preoperatively. The off-label use of GLP-1 receptor agonist within the past decade, while effective for weight loss, should be explored further in this patient population.
Similarly, Donnally et al and Mehta et al found that subcutaneous fat thickness and lamina-to-skin distance predispose patients to SSI.16,17 More obese patients require longer incision, more extensive tissue dissection and blood loss, longer operation, and lastly, more challenging closure. Subcutaneous fat also predisposes patients to tissue necrosis and infection. 10 Outside of targeting excessive body fat/BMI for mitigating the risk of SSI, numerous strategies emerge for preoperative optimization, including antiseptic protocols. For example, Deng et al reported in a thoracolumbar spine surgery cohort of 2252 patients that their institutional Chlorhexidine gluconate (CHG) shower protocol preceding surgery was associated with lower rates of SSI (p=0.001), and offers a standardized preoperative skin-decolonization guideline for patient preparation prior to spine procedures. 8
Risks to wound healing in the posterior neck include soft tissue creases, wound pressure while recumbent, tension with range of motion, and additional bacterial colonization from hair and sweat. The most common bacteria were coagulase-negative Staphylococcus, Staphylococcus aureus, and Cutibacterium species. Coagulase-negative Staphylococcus is commonly identified in nosocomial infections due to its colonization on skin. It is also a common culture contaminant, providing a false-positive in wound swabs. 18 Thus, the principles of antibiotic stewardship suggest that the presence of coagulase-negative Staphylococcus on culture be correlated with the severity of clinical presentation and correlation with deep cultures when prescribing potentially extensive antibiotic courses.
Nevertheless, the relative abundance of Staphylococcus aureus in this identified SSI cohort is consistent with its predominance in SSI surveys. Patel et al found Staphylococcus aureus infections to contribute to 49.3% of pooled SSIs across North America, Europe, and Asia. 19 This commonality underscores the need to maintain coverage for both methicillin-sensitive Staphylococcus aureus (MSSA) and methicillin-resistant Staphylococcus aureus (MRSA) in empiric regimens with agents such as vancomycin, which was included in primary antibiotic treatments 20 times in this cohort.
Cutibacterium species are commonly understood as another bio-film producing, skin commensal organism implicated in SSI following implanted hardware, that has increasingly been recognized in spine surgery SSI and a target for prophylactic therapy. 20 Interestingly, this surgical site colonization may depend on implant material, as Garcia et al found Cutibacterium acnes to have higher adherence to polyether ether ketone (PEEK) than metal implants such as titanium, cobalt chromium, or other mixed alloys. 21 Of note, the treatment for Cutibacterium acnes infections generally includes β-lactams, which were well represented in the treatment courses.
Infections caused by gram-negative pathogens are often associated with increased severity and pose significant therapeutic challenges. 22 Although longer operative time was associated with gram-negative infection on univariate analysis, this association did not remain statistically significant following multivariable adjustment. Nevertheless, the observed trend may warrant further investigation in larger cohorts. Gram-negative bacteria are known to be prevalent in hospitals, and the increased time in the operating room with an open wound increases the odds that sufficient seeding will occur to develop an SSI. 23 Previous studies have also found operative time to be significantly associated with odds of gram-negative infection, with Alexiou et al finding that for every increase in operative time by one minute, odds of multi-drug-resistant gram-negative bacterial infection increased by 0.7%. 24 In cases with longer expected operative duration, implementation of additional protocols to avoid contamination should be considered, which could include minimizing operative staff change, increasing glove changes, and irrigating the wound with increased frequency.
Other strategies in infection prevention stand out as viable options to mitigate infection risk as well. Intraoperative prevention strategies such as vancomycin powder, institutional SSI prevention bundles, and enhanced surgeon adherence to perioperative infection prevention protocols have all been shown to reduce postoperative infection rates and improve surgical outcomes.25-28
Postoperative Management and Antibiotics
In the immediate postoperative period, SSI prophylaxis and wound care remain important; however, evidence for strategies such as prolonged antimicrobial prophylaxis, wound drainage systems, and frequent dressing changes remains limited. 29 Bains et al found that maintaining the original operative dressing for the first five postoperative days was associated with lower SSI rates compared to routine early dressing changes. 30 These findings further emphasize the importance of preoperative and intraoperative infection prevention strategies.
In collaboration with infectious disease colleagues and antibiotic stewardship guidelines, this study found that vancomycin was a common agent used, due to a high frequency of Staphylococcus species including MRSA. The use of broader-spectrum antibiotics such as cefepime or meropenem indicates a subset of gram-negative infections and more resistant organisms in SSI. The predominance of β-lactams in treatment aligns with prior findings that prophylactic β-lactam administration is associated with lower SSI rates compared to non-β-lactam regimens. 31 These findings, combined with the observed resistance patterns, highlight the need for empiric antibiotic strategies that not only include coverage for MRSA and other gram-positive coverage but also consider early coverage for resistant or polymicrobial infections in select high-risk patients. Importantly, once culture results are available, regimens should be rapidly de-escalated to minimize selective pressure and antimicrobial resistance.
The utilization of suppressive oral antibiotics following completion of primary coverage was well documented in this cohort. Over half of the patients were placed on an antibiotic regimen with suppression, and over 40% were prescribed suppressive antibiotics for over six months. Especially in the setting of retained hardware, long-term suppression is commonly employed to mitigate risk of persistent or recurrent infection when complete source control is not feasible. 32 While the principles of antibiotic stewardship suggest minimizing unnecessary antibiotic exposure, the potential benefits of long-term suppression often outweigh the risks. The mean course of primary antibiotics was 32.6 ± 17.5 days, reflective of the need for prolonged therapy in the setting of instrumented spine surgery for 4 to 6 weeks. 32 This variability may reflect physician preferences in treatment; however, further work should establish and employ empiric protocols for PCDF treatment.
Sixteen different antibiotics were used in primary antibiotic treatment, and many regimens included both IV and oral antibiotics in the primary treatment course (10, 21.3%). The heterogeneity likely reflects organism-specific sensitivities, polymicrobial infections, outpatient logistics and feasibility, evolving infectious disease guidelines, and patient-specific renal function, which all must be considered for prescribing antibacterial regimens following organism culturing. Importantly, 21.3% of patients with SSI required multiple operative washouts and debridements, highlighting the complexity of cervical infections, which often prohibit wound closure and incur substantial surgical burden.
The most common antibiotic resistance was to oxacillin, which is not surprising since the majority of S. epidermidis isolates are resistant to this antibiotic, suggesting that empiric coverage with anti-staphylococcal penicillins may be suboptimal in this population. Instead, β-lactams should be considered, including the use of first generation cefazolin or fourth generation cefepime. The high prevalence of oxacillin resistance in this cohort reinforces the continued need for empiric MRSA-active agents, such as vancomycin or daptomycin, in initial postoperative coverage. Additionally, standard susceptibility profiles may miss low-level oxacillin resistance, leading to suboptimal antibiotic selection. 33 Importantly, borderline oxacillin-resistant Staphylococcus aureus (BORSA) may present similarly to methicillin-resistant Staphylococcus aureus (MRSA) and is often clinically more severe than methicillin-susceptible Staphylococcus aureus (MSSA). 34 Once resistance profiles are known, prompt de-escalation remains critical to reduce selective pressure and limit further resistance emergence as increased antibiotic course duration is associated with acquisition of drug resistance.
Limitations
Constraints of the present study include its retrospective nature. Additionally, SSI was defined as cases requiring operative debridement with positive intraoperative cultures. This definition prioritized diagnostic specificity but may have underestimated the true incidence of postoperative infection by excluding superficial infections or infections successfully treated without operative intervention. Accordingly, the reported infection rate should be interpreted as representing surgically managed, culture-confirmed SSI rather than all postoperative wound infections. This study also examined the antibiotic profile for patients with SSI who have undergone PCDF at one major medical center. Antibiotic strains and resistance profiles can vary geographically and across institutions, which has culminated in individualized antimicrobial stewardship guidelines in respective centers. This limits the external validity of resistance profiles on a per strain basis, as well as antibiotic regimens, as institutional data likely guided similar empiric recommendations across the cohort. Nevertheless, this study identified trends in SSI and highlights an issue consistent across all hospitals and hardware-associated procedures. 12 Additionally, this study did not track whether current smokers truthfully followed preoperative guidelines on smoking cessation, or whether patients resumed smoking following surgery, which is an important criteria in clinical practice. Smoking is a recognized high-risk behavior due to impaired wound healing and immune system dysregulation, and both former and current smoking have been independently associated with increased SSI risk per the ACS NSQIP general surgical database, for example. 15 Furthermore, although the multivariable model adjusted for several clinically relevant variables, additional factors including estimated blood loss, nutritional status, wound-related characteristics, and number of instrumented levels were not included and may represent residual confounders. Future prospective studies incorporating these variables may provide a more comprehensive assessment of independent SSI risk factors.
Conclusion
SSI following posterior cervical decompression and fusion remains one of the most common complications with an incidence of 8% in this cohort. Elevated BMI was identified as an independent risk factor, underscoring the importance of patient optimization prior to surgery. Additionally, longer operative time was associated with increased risk of gram-negative infection in univariate analysis, potentially indicating exposure risk to more severe microbiome during longer operations; however, this did not remain significant upon multivariable adjustment. Coagulase-negative Staphylococcus and Staphylococcus aureus were the most isolated organisms. These findings highlight the need for prolonged wound monitoring after surgery and the need for tailored antibiotic regimens once an infection is identified.
Footnotes
Funding: The authors received no financial support for the research, authorship, and/or publication of this article.
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
ORCID iDs
Benjamin J. Carnovale https://orcid.org/0000-0002-3870-8907
Hansen Deng https://orcid.org/0000-0002-9486-8390
Nitin Agarwal https://orcid.org/0000-0003-0256-3897
References
- 1.de Lissovoy G, Fraeman K, Hutchins V, Murphy D, Song D, Vaughn BB. Surgical site infection: incidence and impact on hospital utilization and treatment costs. Am J Infect Control. 2009;37(5):387-397. [DOI] [PubMed] [Google Scholar]
- 2.Kuhns BD, Lubelski D, Alvin MD, et al. Cost and quality of life outcome analysis of postoperative infections after subaxial dorsal cervical fusions. J Neurosurg Spine. 2015;22(4):381-386. [DOI] [PubMed] [Google Scholar]
- 3.Parker SL, Shau DN, Mendenhall SK, McGirt MJ. Factors influencing 2-year health care costs in patients undergoing revision lumbar fusion procedures. J Neurosurg Spine. 2012;16(4):323-328. [DOI] [PubMed] [Google Scholar]
- 4.Bernatz JT, Anderson PA. Thirty-day readmission rates in spine surgery: systematic review and meta-analysis. Neurosurg Focus. 2015;39(4):E7. [DOI] [PubMed] [Google Scholar]
- 5.McCormack RA, Hunter T, Ramos N, Michels R, Hutzler L, Bosco JA. An analysis of causes of readmission after spine surgery. Spine (Phila Pa 1976). 2012;37(14):1260-1266. [DOI] [PubMed] [Google Scholar]
- 6.Patel V, Metz A, Schultz L, et al. Rates and reasons for reoperation within 30 and 90 days following cervical spine surgery: a retrospective cohort analysis of the Michigan Spine Surgery Improvement Collaborative (MSSIC) registry. Spine J. 2023;23(1):116-123. [DOI] [PubMed] [Google Scholar]
- 7.Zhou J, Wang R, Huo X, Xiong W, Kang L, Xue Y. Incidence of Surgical Site Infection After Spine Surgery: A Systematic Review and Meta-analysis. Spine (Phila Pa 1976). 2020;45(3):208-216. [DOI] [PubMed] [Google Scholar]
- 8.Deng H, Chan AK, Ammanuel S, et al. Risk factors for deep surgical site infection following thoracolumbar spinal surgery. J Neurosurg Spine. 2020;32(2):292-301. [PubMed] [Google Scholar]
- 9.Aleem IS, Tan LA, Nassr A, Riew KD. Infection prevention in cervical spine surgery. J Spine Surg. 2020;6(1):334-339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Badiee RK, Mayer R, Pennicooke B, Chou D, Mummaneni PV, Tan LA. Complications following posterior cervical decompression and fusion: a review of incidence, risk factors, and prevention strategies. J Spine Surg. 2020;6(1):323-333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Long DR, Bryson-Cahn C, Waalkes A, et al. Contribution of the patient microbiome to surgical site infection and antibiotic prophylaxis failure in spine surgery. Sci Transl Med. 2024;16(742):eadk8222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Pahys JM, Pahys JR, Cho SK, et al. Methods to decrease postoperative infections following posterior cervical spine surgery. J Bone Joint Surg Am. 2013;95(6):549-554. [DOI] [PubMed] [Google Scholar]
- 13.Emohare O, Ledonio CG, Hill BW, Davis RA, Polly DW, Jr., Kang MM. Cost savings analysis of intrawound vancomycin powder in posterior spinal surgery. Spine J. 2014;14(11):2710-2715. [DOI] [PubMed] [Google Scholar]
- 14.Godil SS, Parker SL, O'Neill KR, Devin CJ, McGirt MJ. Comparative effectiveness and cost-benefit analysis of local application of vancomycin powder in posterior spinal fusion for spine trauma: clinical article. J Neurosurg Spine. 2013;19(3):331-335. [DOI] [PubMed] [Google Scholar]
- 15.Sebastian A, Huddleston P, 3rd, Kakar S, Habermann E, Wagie A, Nassr A. Risk factors for surgical site infection after posterior cervical spine surgery: an analysis of 5,441 patients from the ACS NSQIP 2005-2012. Spine J. 2016;16(4):504-509. [DOI] [PubMed] [Google Scholar]
- 16.Donnally CJ, Henstenburg JM, Pezzulo JD, et al. Increased Surgical Site Subcutaneous Fat Thickness Is Associated with Infection after Posterior Cervical Fusion. Surg Infect (Larchmt). 2022;23(4):364-371. [DOI] [PubMed] [Google Scholar]
- 17.Mehta AI, Babu R, Karikari IO, et al. Young Investigator Award winner: The distribution of body mass as a significant risk factor for lumbar spinal fusion postoperative infections. Spine (Phila Pa 1976). 2012;37(19):1652-1656. [DOI] [PubMed] [Google Scholar]
- 18.Kleinschmidt S, Huygens F, Faoagali J, Rathnayake IU, Hafner LM. Staphylococcus epidermidis as a cause of bacteremia. Future Microbiol. 2015;10(11):1859-1879. [DOI] [PubMed] [Google Scholar]
- 19.Patel H, Khoury H, Girgenti D, Welner S, Yu H. Burden of Surgical Site Infections Associated with Select Spine Operations and Involvement of Staphylococcus aureus. Surg Infect (Larchmt). 2017;18(4):461-473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Baroudi M, Daher M, Parks RD, et al. Cutibacterium acnes in spine surgery: pathophysiology, diagnosis, and treatment. Spine J. 2024;24(9):1545-1552. [DOI] [PubMed] [Google Scholar]
- 21.Garcia D, Mayfield CK, Leong J, et al. Early adherence and biofilm formation of Cutibacterium acnes (formerly Propionibacterium acnes) on spinal implant materials. Spine J. 2020;20(6):981-987. [DOI] [PubMed] [Google Scholar]
- 22.Breijyeh Z, Jubeh B, Karaman R. Resistance of Gram-Negative Bacteria to Current Antibacterial Agents and Approaches to Resolve It. Molecules. 2020;25(6):1340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Rezaei AR, Zienkiewicz D, Rezaei AR. Surgical site infections: a comprehensive review. J Trauma Inj. 2025;38(2):71-81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Alexiou VG, Michalopoulos A, Makris GC, Peppas G, Samonis G, Falagas ME. Multi-drug-resistant gram-negative bacterial infection in surgical patients hospitalized in the ICU: a cohort study. Eur J Clin Microbiol Infect Dis. 2012;31(4):557-566. [DOI] [PubMed] [Google Scholar]
- 25.Bakhsheshian J, Dahdaleh NS, Lam SK, Savage JW, Smith ZA. The use of vancomycin powder in modern spine surgery: systematic review and meta-analysis of the clinical evidence. World Neurosurg. 2015;83(5):816-823. [DOI] [PubMed] [Google Scholar]
- 26.Toi M, Maruo K, Arizumi F, et al. Prevention of Surgical Site Infection after Spine Operation with Care Bundle. Surg Infect (Larchmt). 2025;26(8):584-590. [DOI] [PubMed] [Google Scholar]
- 27.Agarwal N, Agarwal P, Querry A, et al. Implementation of an infection prevention bundle and increased physician awareness improves surgical outcomes and reduces costs associated with spine surgery. J Neurosurg Spine. 2018;29(1):108-114. [DOI] [PubMed] [Google Scholar]
- 28.Agarwal P, Gaba F, Agarwal N, Crall V, Mazurkiewicz A, Friedlander RM. Physician Awareness Combined With Perioperative Infection Prevention Bundles Results in Durable Neurosurgical Infection Control and Cost Savings. Neurosurgery. 2025;97(4):780-785. [DOI] [PubMed] [Google Scholar]
- 29.Tan T, Lee H, Huang MS, et al. Prophylactic postoperative measures to minimize surgical site infections in spine surgery: systematic review and evidence summary. Spine J. 2020;20(3):435-447. [DOI] [PubMed] [Google Scholar]
- 30.Bains RS, Kardile M, Mitsunaga LK, Bains S, Singh N, Idler C. Postoperative Spine Dressing Changes Are Unnecessary. Spine Deform. 2017;5(6):396-400. [DOI] [PubMed] [Google Scholar]
- 31.Agarwal P, Kumar RP, Oleksiuk LM, et al. Non-β-Lactam Antibiotic Use, β-Lactam Allergy, and Surgical Site Infections. JAMA Surg. 2025;160(11):1260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Palmowski Y, Bürger J, Kienzle A, Trampuz A. Antibiotic treatment of postoperative spinal implant infections. J Spine Surg. 2020;6(4):785-792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Modak S, Mane P, Patil S. A Comprehensive Phenotypic Characterization of Biofilm-Producing Coagulase-Negative Staphylococci: Elucidating the Complexities of Antimicrobial Resistance and Susceptibility. Cureus. 2025;17(2):e79039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hryniewicz MM, Garbacz K. Borderline oxacillin-resistant Staphylococcus aureus (BORSA) - a more common problem than expected? J Med Microbiol. 2017;66(10):1367-1373. [DOI] [PubMed] [Google Scholar]