Impact of a Multidisciplinary Approach on Surgical Site Infections (SSIs) in Pediatric Posterior Spine Fusions (PSF): Insights From a Ten-Year Cohort Study (2012-2022) Authors: Tyler Sullivan, Kelea Damp, Carson Keeter, Mark Erickson, Sumeet Garg Published: 2026 ABSTRACT Background:Surgical site infections (SSIs) after pediatric posterior spine fusion (PSF) remain a notable concern. Our institution launched the Target Zero (TZ) initiative in 2007 to reduce SSIs through standardized prevention protocols. This study evaluates SSI rates from 2012 to 2022, following the initiative's early years (2006 to 2012).Methods:After IRB approval, we conducted a retrospective review of 1476 PSF patients aged 10 to 18 from July 2012 to December 2022. Patients were stratified into high-risk (HR) and low-risk (LR) groups based on comorbidities. Infection rates and associated factors were analyzed.Results:Of 1476 patients, 495 were HR and 981 LR. SSI rates were significantly higher in HR patients than in LR patients (6.87% vs 0.92%, P < 0.001). The number of levels fused and estimated blood loss did not differ significantly between cohorts with and without infection. HR patients who developed an SSI had longer procedures (302.4 vs 252.5 minutes; P < 0.03). Time of procedure was not significantly different in LR patients with or without infection. Compliance with TZ antibiotic protocols was high (100% LR, 98% HR). Compared with the previous cohort (2008 to 2012), SSI rates in LR patients declined from 2.7% (2008 to 2012) to 0.92% (2012 to 2022), while rates in HR patients remained stable (6.87% vs 6.5%). Protocol compliance improved from 85% to 98%.Conclusion:SSI rates in LR patients have declined over the past decade, reflecting sustained effectiveness of TZ protocols. Rates in HR patients remained unchanged, highlighting the need for targeted strategies for high-risk populations. These findings are consistent with known risk factors of SSIs in pediatric spine fusion surgery. Surgical site infection (SSI) associated with spine fusion is both costly and inconvenient for patients and families. Many variables should be considered when attempting to decrease SSIs, including appropriate prophylactic antibiotic regimens, preincision scrub, number of people in the OR suite, procedure duration, blood loss, and high-risk comorbidities. Patients with adolescent idiopathic scoliosis (AIS) who undergo posterior spine fusion (PSF) are at a 0.31-1.25% risk of developing overall wound infections, compared with nonidiopathic patients, who have reported risks of 4 to 20% and a 2.82% risk of deep infections.1 As a result of complications, the hospital length of stay for both idiopathic and nonidiopathic patients may increase by as much as 29 days.2 The average extended hospitalization charges due to complications are $154,537, with a range of $26,977 to $961,722.2,3 In July 2007, a multidisciplinary spine infection task force, designated “Target Zero” (TZ), was formed, consisting of representatives from departments such as orthopaedic surgery, infectious disease, epidemiology, pharmacy, anesthesia, quality improvement, and the operating room. A formal antibiotic protocol was written, including preoperative, intraoperative, and postoperative initiatives. Ballard et al4 compared infection rates in both low-risk (LR) and high-risk (HR) pediatric patients undergoing spinal surgery at a tertiary pediatric hospital before and after implementation of a spine fusion infection task force. Patients were organized into two cohorts: low-risk (LR) and high-risk (HR). In 2013, the department opted to revise the definitions of both LR and HR used in the previous and current study. Patients were considered LR (standard prophylaxis) if they had idiopathic scoliosis with no underlying comorbidities. Patients were considered HR (expanded prophylaxis) if they had conditions or comorbidities such as cardiac anomalies, congenital scoliosis, cerebral palsy/brain injuries, Down syndrome, kyphosis, Marfan syndrome, muscle disease, neurofibromatosis, spina bifida, spondylolisthesis, spondylolysis, syndromic, and tumors (Table 1). Table 1Spine Surgery Standard and Expanded Infection Risk Prophylactic Antibiotic GuidelinesInfection RiskPreoperative (Before Incision)Intraoperative Re-dosingPostoperativeSTANDARD:• Idiopathic scoliosis with no underlying comorbidities and BMI less than the 95th percentile for any surgical procedure (e.g., PSF or VBT)• Trauma: typically, single-system trauma—surgeon discretion for standard vs. expanded prophylaxisThe patient is < 13 years, premenarchal, or MRSA−cefazolinFrom start of preincision dose: cefazolin every 4 hours+vancomycin every 8 hours(if given before incision)ALSOFor every 50% blood volume loss, give ½ dosesb: cefazolin+vancomycin(if given before incision)The patient is < 13 years, premenarchal, or MRSA−:cefazolinevery 8 hours x 3 dosesThe patient is ≥ 13 years, postmenarchal, MRSA+or<13 years with unknown MRSA status:Cefazolin+vancomycinThe patient is ≥ 13 years, postmenarchal, MRSA+Or<13 years with unknown MRSA status:cefazolinevery 8 hours x 3 doses+Vancomycinevery 8 hours x 3 doses• Anterior cervical fusions• All cervical spine fusion cases not specified for expanded prophylaxisEXPANDED:• All other patients with scoliosis who are not idiopathic or with underlying conditions that increase the risk of infection, e.g., neuromuscular, syndromic, kyphosis, spondylolisthesis, and scoliosis related to other underlying conditions/processes• BMI > 95th percentile• History of spine infection• Revision surgery (noncervical)• Anterior spine fusions with intra-abdominal, retroperitoneal, and transthoracic approaches• Trauma: Typically, multisystem trauma, including any injury to the abdomen—surgeon discretion for standard vs. expanded prophylaxisThe patient is < 13 years, premenarchal, or MRSA−:cefazolin+ceftriaxoneFrom start of preincision dose:cefazolin every 4 hours+vancomycin every 8 hours(if given before incision)ALSOFor every 50% blood volume loss, give ½ dosesb:Cefazolin+vancomycin (if given before incision)+ceftriaxoneThe patient is < 13 years, premenarchal, or MRSA: cefazolin every 8 hours x 3 dosesThe patient is ≥13 years, postmenarchal, MRSA +Or<13 years with unknown MRSA status:Acefazolin+vancomycin+ceftriaxoneThe patient is ≥ 13 years, postmenarchal, MRSA+Or<13 years with unknown MRSA status:cefazolinevery 8 hours x 3 doses+Vancomycinevery 8 hours x 3 doses• Posterior cervical cases with one of the following: instrumentation of > 6 levels, history of myelomeningocele or multisystem trauma (within 7 days) If patient is currently on antibiotics, consult Epidemiology MD for recommended antibiotic prophylaxis Bolded text denotes guidelines for each phase of care (pre-op, intra-op, and post-op). Spine Surgery Standard and Expanded Infection Risk Prophylactic Antibiotic Guidelines Bolded text denotes guidelines for each phase of care (pre-op, intra-op, and post-op). A total of 394 patients who underwent spine surgery from April 2006 to September 2008 were retrospectively reviewed to identify patients who developed an SSI after surgery. HR and LR patients were compared before TZ implementation and after TZ implementation, and the study found that TZ implementation helped prevent 1 infection in every 16 patients in the HR group and 1 in 30 overall. In addition, infection rates reduced from 12.9% to 6.5% in the HR group and from 4.9% to 2.7% in the LR group.4 In 2018, Vandenberg et al examined a cohort of 511 patients who underwent primary spine procedures at the same hospital from 2008 to 2012 and found that 85% of patients were compliant with the antibiotic protocol and that strict adherence prevented 1 SSI within 90 days of surgery for every 26 patients.5 The objective of this study was to compare infection rates between low-risk and high-risk pediatric patients undergoing index posterior spine fusion surgery at a tertiary pediatric hospital over a 10-year period (2012 to 2022) and to evaluate improvements relative to the previous study (2008 to 2012). The infection protocol for spine surgery at our hospital undergoes ongoing updates; these changes are described in the Methods section. This study investigates a previously unreported cohort of patients who underwent surgery after our last reported cohort.5 We also aimed to investigate the risk factors for developing SSI, including length of stay, estimated blood loss (EBL), 90-day postoperative complications, and compliance with an antibiotic protocol. We hypothesized that compliance with the antibiotic protocol would reduce SSI rates after PSF in both HR and LR patients. Methods Patient Population After obtaining approval from the Institutional Review Board (IRB), a retrospective chart review was conducted to collect demographic and clinical data from 1476 patients who underwent index posterior spine fusion surgery (revisions, removals, and exchanges were excluded) at a single, tertiary pediatric hospital between July 1, 2012, and December 31, 2022. Patients were followed for a minimum of 90 days postoperatively. To stay consistent with the previous publication analyzing the 2008-2012 data set, the same definition was applied to both the LR and HR patient cohorts. Patients were organized into LR and HR cohorts, and the following data were collected: demographics (BMI, race/ethnicity, sex, age at surgery), diagnosis, risk category (HR or LR), levels fused/length of fusion, procedure time, EBL (based on recorded intraoperative cell salvage volume), posterior column osteotomies with PSF procedures, compliance with the hospital's antibiotic protocol, surgical site infections or complications within 90 days postoperatively, and unplanned return to the operating room. Patients who developed an SSI met the Centers for Disease Control and Prevention—National Healthcare Safety Network (CDC-NHSN) criteria for SSI occurring within 30 to 90 days of the procedure.6 Antibiotic Protocol The TZ team developed an antibiotic protocol with hopes to reduce the risk of infections after pediatric spine surgery (Table 1). Three criteria measured antibiotic bundle compliance: timely preincision antibiotic administration, proper intraoperative re-dosing, and antibiotic discontinuation after surgery. Postoperative antibiotic use was discontinued, with no surgeon extending antibiotics beyond 24 hours after surgery. Alternative antibiotics were administered to patients with documented allergies to any of the primary agents. Staff were provided with a standardized list of approved alternatives to ensure consistency and protocol control (Table 2). To be considered compliant, all three criteria were required to be completed. Table 2Alternative Antibiotics for AllergiesAllergyAlternativeCefazolinVancomycinCeftriaxoneLevofloxacinIf true allergy to vancomycin (not infusion reaction)Cefazolin + clindamycinIf allergic to cefazolin + true allergy to vancomycinClindamycin Alternative Antibiotics for Allergies Evolution of Protocol Changes Iterative changes have been made to the antibiotic protocol since its implementation in 2007. In 2010, vancomycin IV was initiated for patients ≥13 years or after menarche. In 2011, the length of stay goal for LR patients was ≤6 days and ≤7 days for HR patients. In 2012, the length of stay goal changed to ≤5 days for LR patients; ceftriaxone was added to the expanded prophylaxis protocol for HR patients, and vancomycin powder was implemented. In 2013, the length of stay goal changed to ≤4 days for LR patients. In 2015, providers began using a topical skin adhesive combined with a self-adhering mesh system (DERMABOND PRINEO Skin Closure System, Ethicon, Inc., a Johnson & Johnson company) for postoperative incision in LR patients. In 2017, noninvasive skin wound closure systems were implemented (Zip Surgical Skin Closure, Stryker Corporation) for HR patients. In 2018, surgeons started using a closed-incision negative-pressure therapy (ciNPT) device to promote wound healing and reduce infection risk (Prevena Incision Management System, 3M Health Care) among HR patients and those with BMI > 35%. In 2019, providers began administering cefazolin to all patients, even if they were receiving vancomycin IV. All postoperative antibiotics were discontinued by all surgeons. In 2022, one surgeon began using the single-use negative-pressure wound therapy (sNPWT) system for all patients, specifically the PICO sNPWT (Smith + Nephew). By 2024, all spine surgeons had adopted PICO sNPWT (Smith + Nephew) for sNPWT in all surgical patients. Protocol changes were standardized across the study period through department-wide implementation and staff training to minimize variation and potential bias. However, incremental modifications to the protocol throughout the years likely influenced SSI rates among both LR and HR patients. Statistical Analyses To avoid overfitting a multivariate model, univariate logistic regression models were used to evaluate the relationship between the odds of SSI development and various predictors such as sex, age, study year, risk level, levels fused, procedure time, and estimated blood loss. Statistically significant associations will be interpreted as odds ratios (ORs). For improved interpretability, time-based variables (length of stay, anesthesia time, and procedure time) were log2-transformed so that ORs reflect the change in the odds of SSI development for each doubling of the predictor value. All other variables remain untransformed. Summaries are also given for each variable, along with demographic variables, including means and standard deviation (SD), median and interquartile range (IQR), and range. Counts of missing data are also reported. For continuous variables, P-values were obtained using independent t-tests, while for categorical variables, P-values were derived from chi-square tests with Monte Carlo simulations based on 2000 replications. Statistical significance is set at α = 0.05. Results A total of 1476 patients (495 HR and 981 LR) who underwent index PSF were included in this 10-year cohort study and were followed for at least 90 days postoperatively. 69% of patients were female, and 31% were male. The average age of the patients was 14 years (range: 3-25). A total of 43 of 1476 patients developed an SSI from 2012 to 2022 (2.9% rate). No significant differences in the incidence of SSI were observed by sex, BMI, age, or study years (Table 3). Table 3SSI Rate Comparison by Patient CharacteristicsCharacteristicNo Infection, N = 1,4331Infection, N = 431 P 2 Sex0.6 F994 (69%)28 (65%) M439 (31%)15 (35%)Age (years)>0.9 Mean (SD)14.18 (2.58)14.14 (2.50) Median (IQR)14.00 (13.00-16.00)14.00 (12.12-16.00) Range3.00, 25.008.00, 18.01BMI0.3 Mean (SD)51.41 (1149.60)19.70 (5.66) Median (IQR)19.45 (17.31-22.38)17.72 (16.15-21.02) Range0.78-43,080.0012.33-38.26Missing3011= n (%)2= Pearson's Chi-squared test; Welch Two Sample t-test SSI Rate Comparison by Patient Characteristics = n (%) = Pearson's Chi-squared test; Welch Two Sample t-test Rate of SSI Among High-Risk (HR) vs Low-Risk (LR) Patients The HR group accounted for 79% of infections while the LR group made up 21% (P < 0.001, Figure 1 and Table 4). We compared infection rates with our previously reported cohort from 2008 to 2012.5 Since the implementation of TZ, the SSI rate in the LR group decreased from 2.7% (2008-2012) to 0.9% (2012-2022) and has remained relatively stable in the HR group from 6.5% (2008-2012) to 6.9% (2012-2022). LR patients exhibited a 0% infection rate from 2016 to 2022, but the infection rate among HR patients remains elevated (Figure 2). On average, HR patients are 7.9 times more likely to have an SSI than LR patients (OR = 7.925, 95% CI, 3.934 to 17.71). Chart demonstrating the total count of SSIs among HR vs LR patients who underwent posterior spine fusions over a 10-year period (2012 to 2022). Table 4Univariate Logistic Regression Table for SSI and Risk LevelsCharacteristicORSEStatistic95% CI P Risk level Low———— High7.9250.3795.463.934, 17.71<0.001CI = confidence interval; OR = odds ratio; SE = standard error. Univariate Logistic Regression Table for SSI and Risk Levels CI = confidence interval; OR = odds ratio; SE = standard error. Plot demonstrating the percent SSI incidence among HR and LR patients from 2012 to 2022. This graph shows the percentage of SSI incidence in high-risk and low-risk patient populations over a 10-year period (2012 to 2022). For LR patients, the incidence of SSIs has remained at 0% since 2016. By contrast, the incidence of SSIs in HR patients has remained relatively stable, with no significant upward or downward trend over the same period. Data points represent annual percentages of patients who developed an SSI within each group. Levels Fused The SSI rate was not significantly different based on the number of levels fused in either the LR or HR cohort (Table 5). Table 5Patient and Surgical Characteristics by Risk Group and Infection StatusCharacteristicLow RiskHigh RiskNo InfectionN = 972InfectionN = 9 P a No InfectionN = 461InfectionN = 34 P a Levels fused0.80.3 Mean (SD)10.77 (2.58)10.56 (3.00)13.74 (4.18)14.68 (4.99) Median (Q1, Q3)11.00 (10.00-13.00)11.00 (10.00-12.00)14.00 (12.00-17.00)17.00 (13.00-18.00) Min, max0.00-19.004.00-15.000.00-28.000.00-19.00Procedure time (minutes)0.80.030 Mean (SD)204.37 (99.60)199.67 (49.76)252.48 (118.68)302.44 (124.97) Median (Q1, Q3)179.00 (148.00-236.00)195.00 (193.00-219.00)224.00 (169.00-307.00)257.00 (209.00-401.00) Min, max15.00-898.00114.00-292.0016.00-902.00163.00-604.00Estimated blood loss (mL)0.70.2 Mean (SD)612.10 (490.61)552.78 (513.63)893.10 (673.06)1019.53 (552.26) Median (Q1, Q3)475.00 (300.00-750.00)350.00 (250.00-700.00)725.00 (425.00-1200.00)950.00 (600.00-1200.00) Min, max5.00-6000.0050.00-1750.0010.00-4600.00160.00-2500.00 Missing7010Length of stay (days)0.80.4 Mean (SD)3.21 (1.60)3.33 (1.22)6.31 (8.31)6.94 (4.03) Median (Q1, Q3)3.00 (2.00-4.00)3.00 (3.00-4.00)5.00 (3.00-7.00)6.00 (4.00-8.00) Min, max1.00-28.002.00-6.001.00-103.002.00-21.00aWelch two-sample t-test. Patient and Surgical Characteristics by Risk Group and Infection Status Welch two-sample t-test. Estimated Blood Loss (EBL) The SSI rate was not significantly different based on EBL in either the LR or HR cohort (Table 5). Procedure Time A significant difference was observed between procedure duration and the incidence of SSI in the HR population (P = 0.030). HR patients who developed an SSI experienced longer procedures, lasting 302.4 minutes, compared with 252.5 minutes for those without an SSI. No significant difference in SSI rates was observed based on procedure length in LR patients (P = 0.80). In the cohort, the mean duration was 199.7 minutes for patients who developed an SSI and 204.4 minutes for those who did not (Table 5). Length of Stay Patients who developed an SSI had a significantly longer average length of stay (6.2 days) than patients without an SSI (4.2 days). A statistically significant positive relationship between log2(length of stay) and the odds of SSI was observed. Furthermore, there was a significant difference in length of stay between HR and LR patients. HR patients spent significantly longer in the hospital (6.4 days) than LR patients (M = 3.2 days, P < 0.001). Unplanned Return to the Operating Room (UPROR) Findings demonstrate a significant difference in SSI-related UPROR and non–SSI-related UPROR; 3.3% of patients returned to the OR for non–SSI-related postoperative complications while 1.5% returned for SSI-related complications (P < 0.001). 95% of patients did not have UPROR. TZ Antibiotic Compliance A significant difference was observed in the odds of SSI between the patients who were considered “compliant” with the TZ antibiotic protocol and the patients who were not compliant. Noncompliance was defined as failure to meet one or more of the three antibiotic compliance criteria. Patients who were not compliant with the TZ protocol were 33.9 times more likely to develop an SSI after index PSF than patients who were compliant (P = 0.013, 95% CI, 1.326 to 867.2). Discussion The purpose of this study was to evaluate the rate of SSI among pediatric orthopaedic spine patients who underwent index PSF procedures over a 10-year period and to determine whether compliance to the TZ antibiotic protocol contributed to a reduction in infection rates. Furthermore, we aimed to investigate any postoperative complications and unplanned hospital readmissions related to incidence of SSI after the PSF procedures. Our hypothesis is that SSI rates would be lower among LR patients after surgery. In addition, it was confirmed that compliance with an antibiotic protocol will markedly reduce the likelihood of developing SSIs after PSF. Risk Factors of SSI Incidence Previous literature has suggested multiple risk factors for developing SSI after spine surgery in both idiopathic and nonidiopathic patients. One study found that procedure type had a notable effect on SSI incidence; specifically, nonidiopathic children and adolescents undergoing fusion procedures were at an increased risk of SSI when compared with patients undergoing growing rod procedures.7,8 Our study focused on posterior spine fusion procedures and excluded all other procedures to understand the outcomes of PSF on SSI incidence. In addition, comorbidities such as diabetes, hypertension, cardiac anomalies, incontinence, ambulatory status, and osteoporosis, among others, have been associated with the risk of SSI.9,10 Other risk factors including BMI, levels fused, procedure duration, number of blood transfusions, and compliance with an institution's antibiotic protocol have also been identified.11-13 Our findings align with these studies—demonstrating a notable relationship between incidence of SSI and levels fused, procedure and anesthesia duration, compliance with the antibiotic protocol, risk level (HR vs LR), and EBL. However, BMI did not markedly affect the rate of SSI in our cohort. TZ Antibiotic Protocol Compliance The results of this study demonstrate a significant reduction in SSI among the LR cohort and a stable rate of SSI among the HR cohort, compared with our earlier studies from 2008 to 2012, after implementation of the TZ protocol, confirming the significant clinical importance of adhering to antibiotic protocols.4,5 The problem of HR populations experiencing SSIs after index PSF procedures persists, and additional investigations are necessary to determine modifiable risk factors in this population that could reduce risk of SSI. Limitations Our retrospective study had several limitations. First, we were limited to the data already present in the patients' medical chart, resulting in missing information on the counts in earlier years. EBL was determined using intraoperative cell salvage volume, which is recognized as an imprecise measurement. Microbiological data for SSIs were not collected; therefore, it is unclear whether the reported infections were due to organisms covered by our antibiotic prophylaxis. Next, we did not collect Cobb angles or blood transfusion data to analyze the effect of curve severity and the number of blood transfusions on incidence of SSI, which are known risk factors in SSI rates. Additional investigations are necessary to determine the underlying relationship between the rate of SSI and blood transfusions. The changes in infection prevention practices described in our Methods section from 2012 to 2022 may have affected the reported infection rates. Finally, owing to the 2013 change in the institutional definitions of LR and HR, infection rates for the HR population group may have been underreported in the previous study. Importantly, both previous and current studies use the same definitions for each cohort. Clinical Implications Moving forward, it is important to know the risk factors that can lead to the development of an SSI. HR patients generally will have longer fusions, longer OR times, and increased EBL compared with LR patients. These specific surgical factors are all associated with an increased risk of SSI. Optimizing surgical techniques to reduce procedure time may help lower the incidence of infection. In addition, minimizing blood loss and the need for blood transfusions may further reduce the risk of infection. Previous studies have indicated that having a dedicated surgical spine team for each case can reduce total OR time by 50 minutes.14 By ensuring that team members have specialized training and shadowing experience in spine surgeries, OR efficiency is markedly improved. Moreover, controlling factors such as the opening and closing of OR doors, administering appropriate intraoperative medications such as tranexamic acid, and implementing preoperative planning—such as a multidisciplinary blood conservation program and transfusion awareness initiatives—markedly reduce OR time and blood loss in patients undergoing PSF for AIS. It is essential to continue educating both patients and their families about the potential risks and the critical role of preoperative and postoperative care in reducing the risk of SSI.15,16 In addition, continuing to educate nursing staff, when necessary, on optimal postoperative care and administration of antibiotics preoperatively, intraoperatively, and postoperatively will help mitigate the risk of SSI. Most importantly, institutions should continue monitoring the compliance with their antibiotic protocols to minimize UPROR and length of stay and ultimately reduce the likelihood of infection. The challenges associated with HR patients—such as longer fusion surgeries, longer OR times, and increased EBL—pose notable hurdles in reducing the risk of SSIs. These factors not only complicate the surgical process but also elevate the likelihood of complications. However, optimizing these HR patients through appropriate preoperative, intraoperative, and postoperative care is crucial for improving outcomes. Evidence suggests that reducing OR time and minimizing blood loss, along with reducing the need for blood transfusions, can markedly decrease the risk of infections. Implementing the antibiotic protocol for all patients undergoing PSF has decreased the overall SSI rate. While these strategies have been effective in LR patients, they remain equally, if not more, critical for high-risk patients. By focusing on enhancing OR efficiency and implementing effective blood management protocols, we can better manage the specific challenges presented by HR patients, leading to improved surgical outcomes and a reduction in the occurrence of SSIs. Conclusion These results are consistent with previous literature on the risk factors that influence the likelihood of SSIs after pediatric spine fusion surgery. HR patients with complex diagnoses, longer procedure times, and increased EBL are strongly associated with higher SSI rates and ultimately longer hospital stays after surgery. The TZ antibiotic protocol has been effective in reducing SSI in LR patients, but SSI in HR patients remains a problem requiring additional targeted interventions. Enhanced protocols and continued investigation are necessary to improve surgical outcomes in the HR population.