A Clinical Risk Model for Surgical Site Infection Following Pediatric Spine Deformity Surgery

Utilizing a comprehensive machine learning approach to identify patients at high risk for extended length of stay following spinal deformity surgery in pediatric patients with early onset scoliosis

PURPOSE

Early onset scoliosis (EOS) patient diversity makes outcome prediction challenging. Machine learning offers an innovative approach to analyze patient data and predict results, including LOS in pediatric spinal deformity surgery.

METHODS

Children under 10 with EOS were chosen from the American College of Surgeon's NSQIP database. Extended LOS, defined as over 5 days, was predicted using feature selection and machine learning in Python. The best model, determined by the area under the curve (AUC), was optimized and used to create a risk calculator for prolonged LOS.

RESULTS

The study included 1587 patients, mostly young (average age: 6.94 ± 2.58 years), with 33.1% experiencing prolonged LOS (n = 526). Most patients were female (59.2%, n = 940), with an average BMI of 17.0 ± 8.7. Factors influencing LOS were operative time, age, BMI, ASA class, levels operated on, etiology, nutritional support, pulmonary and neurologic comorbidities. The gradient boosting model performed best with a test accuracy of 0.723, AUC of 0.630, and a Brier score of 0.189, leading to a patient-specific risk calculator for prolonged LOS.

CONCLUSIONS

Machine learning algorithms accurately predict extended LOS across a national patient cohort and characterize key preoperative drivers of increased LOS after PSIF in pediatric patients with EOS.

What Is the Risk of Irrigation and Debridement Following Foot and Ankle Surgery? Development of a Risk Severity Scoring System

BACKGROUND

Surgical site infection (SSI) after foot and ankle surgery has serious negative effects on patient health and quality of life. While previous studies have looked at predisposing factors for SSI, to our knowledge, no study has proposed a risk severity score.

QUESTIONS/PURPOSES

Can a risk severity score, based on patient demographic characteristics and surgical variables, be developed for preoperative use in patients undergoing foot and ankle surgery that will calculate the risk of an irrigation and debridement (I&D) procedure within 90 days of surgery utilizing data from previous surgeries?

METHODS

A retrospective chart study was performed on patients undergoing foot and ankle surgery. Data on demographic characteristics including age, sex, and BMI were recorded. Data on patient factors including diabetes and smoking history were also recorded. Surgical details including length of surgery, procedure type, surgeon, antibiotic delivery time, antibiotic type, and antibiotic dose were analyzed. Of 2979 procedures, 1% (36) of I&Ds were performed within 90 days. The mean age at surgery was 49 ± 17 years, and 57% (1702) of patients were female. The mean BMI was 28 ± 6 kg/m2. The primary outcome was I&D within 90 days postoperatively. Descriptive statistics of differences in patient characteristics between those who underwent I&D and those who did not were examined using chi-square tests and t-tests (p < 0.05 was taken as significant). Significant variables from a simple regression analysis were included in a multiple logistic regression model with a forward stepwise procedure for variable selection. We required all data in the model to be categorical; thus, continuous variables such as time were dichotomized. We factored odds ratios determined by multiple regression for significant variables into the final risk severity score, and an easy-to-use tool based on this risk severity score was created in Excel (Microsoft).

RESULTS

Current tobacco use, diabetes, and longer operative times were the only factors associated with I&D within 90 days postoperatively. A risk severity score was developed using current tobacco use, diabetes, and length of surgery greater than 60 minutes as factors. A patient with a severity score of 0 (no risk factors) had a 0.6% chance of I&D within 90 days, while a severity score of 1 indicated a 1.1% chance, a score of 2 indicated a 2.1% chance, a score of 3 indicated a 4.0% chance, and a score of 4 (all risk factors) indicated a 7.5% chance of I&D within 90 days. A spreadsheet that can be used at the point of care was created based on these findings.

CONCLUSION

Our risk severity score may help inform preoperative patient guidance and operative planning. Calculating the score in the office setting during preoperative visits can also improve communication between physician and patient. Future research should focus on validation of this risk severity score at multiple institutions.

LEVEL OF EVIDENCE

Level III, prognostic study.

Predicting early return to the operating room in early-onset scoliosis patients using machine learning techniques

PURPOSE

Surgical treatment of early-onset scoliosis (EOS) is associated with high rates of complications, often requiring unplanned return to the operating room (UPROR). The aim of this study was to create and validate a machine learning model to predict which EOS patients will go on to require an UPROR during their treatment course.

METHODS

A retrospective review was performed of all surgical EOS patients with at least 2 years follow-up. Patients were stratified based on whether they had experienced an UPROR. Ten machine learning algorithms were trained using tenfold cross-validation on an independent training set of patients. Model performance was evaluated on a separate testing set via their area under the receiver operating characteristic curve (AUC). Relative feature importance was calculated for the top-performing model.

RESULTS

257 patients were included in the study. 146 patients experienced at least one UPROR (57%). Five factors were identified as significant and included in model training: age at initial surgery, EOS etiology, initial construct type, and weight and height at initial surgery. The Gaussian naïve Bayes model demonstrated the best performance on the testing set (AUC: 0.79). Significant protective factors against experiencing an UPROR were weight at initial surgery, idiopathic etiology, initial definitive fusion construct, and height at initial surgery.

CONCLUSIONS

The Gaussian naïve Bayes machine learning algorithm demonstrated the best performance for predicting UPROR in EOS patients. Heavier, taller, idiopathic patients with initial definitive fusion constructs experienced UPROR less frequently. This model can be used to better quantify risk, optimize patient factors, and choose surgical constructs.

LEVEL OF EVIDENCE

Prognostic: III.

The effect of antibiotic-impregnated calcium sulfate beads and Medical Optimization Clinic attendance on the acute surgical site infection rate in high-risk pediatric neuromuscular and syndromic scoliosis patients

BACKGROUND

Neuromuscular and syndromic (NMS) scoliosis patients are at higher risk of acute surgical site infections (SSIs). Despite following POSNA's endorsed consensus-based guidelines for SSI prevention, our institutional rates of acute SSI have varied dramatically. This variability drove simultaneous strategies to lower SSI rates: the creation of a preoperative Medical Optimization Clinic (MOC) and use of antibiotic-impregnated (Abx-I) calcium sulfate beads.

METHODS

Patients undergoing index PSF at a single institution between 2016 and 2022 were retrospectively reviewed. Patients with ≥ 2 risk factors were included: (1) BMI < 18.5 or > 25; (2) incontinence; (3) instrumentation to pelvis; (4) non-verbal; (5) GMFCS IV/V. SSI was defined as deep infection within 90 days. We compared patients who attended MOC and received Abx-I (MOC + Abx-I) to those receiving neither intervention (control) nor a single intervention.

RESULTS

282 patients were included. The overall infection rate was 4.26%. Higher GMFCS (p = 0.0147), non-verbal status (p = 0.0048), and longer fusions (p = 0.0298) were independently associated with infection rate. Despite the MOC + Abx-I group having larger Cobb angles (88° ± 26°), higher GMFCS levels (4.5 ± 0.9), ASA class (3 ± 0.4), and more frequent instrumentation to the pelvis (85%), they had the lowest infection rate (2.13%) when compared to the control (4.2%) or single intervention groups (5.7%, 4.6%) (p = 0.9).

CONCLUSION

The study examined the modern infection rate of NMS patients following the implementation of two interventions: MOC and Abx-I. Despite having higher risk factors (curves (88°), GMFCS level (4.5), ASA class (3), higher % instrumentation to the pelvis (85%)), the patients treated with both interventions demonstrated the lowest infection rate (2.13%).

Risk stratification for early postoperative infection in Pediatric spinal deformity correction: development and validation of the Pediatric scoliosis infection risk score (PSIR score)

BACKGROUND CONTEXT

Postoperative infection after spinal deformity correction in pediatric patients is associated with significant costs. Identifying risk factors associated with postoperative infection would help surgeons identify high-risk patients that may require interventions to minimize infection risk.

PURPOSE

To investigate risk factors associated with 30-day postoperative infection in pediatric patients who have received posterior arthrodesis for spinal deformity correction.

STUDY DESIGN/SETTING

Retrospective review of prospectively collected data.

PATIENT SAMPLE

The National Surgical Quality Improvement Program Pediatric database for years 2016-2021 was used for this study. Patients were included if they received posterior arthrodesis for scoliosis or kyphosis correction (CPT 22,800, 22,802, 22,804). Anterior only approaches were excluded.

OUTCOME MEASURES

TThe outcome of interest was 30-day postoperative infection.

METHODS

Patient demographics and outcomes were analyzed using descriptive statistics. Multivariable logistic regression analysis using likelihood ratio backward selection method was used to identify significant risk factors for 30-day infection to create the Pediatric Scoliosis Infection Risk Score (PSIR Score). ROC curve analysis, predicted probabilities, and Hosmer Lemeshow goodness-of-fit test were done to assess the scoring system on a validation cohort.

RESULTS

A total of 31,742 patients were included in the study. The mean age was 13.8 years and 68.7% were female. The 30-day infection rate was 2.2%. Reoperation rate in patients who had a post-operative infection was 59.4%. Patients who had post-operative infection had a higher likelihood of non-home discharge (X2 = 124.8, p < 0.001). In our multivariable regression analysis, high BMI (OR = 1.01, p < 0.001), presence of open wound (OR = 3.18, p < 0.001), presence of ostomy (OR = 1.51, p < 0.001), neuromuscular etiology (OR = 1.56, p = 0.009), previous operation (OR = 1.74, p < 0.001), increasing ASA class (OR = 1.43, p < 0.001), increasing operation time in hours (OR = 1.11, p < 0.001), and use of only minimally invasive techniques (OR = 4.26, p < 0.001) were associated with increased risk of 30-day post-operative infection. Idiopathic etiology (OR = 0.53, p < 0.001) and intraoperative topical antibiotic use (B = 0.71, p = 0.003) were associated with reduced risk of 30-day postoperative infection. The area under the curve was 0.780 and 0.740 for the derivation cohort and validation cohort, respectively.

CONCLUSIONS

To our knowledge, this is the largest study of risk factors for infection in pediatric spinal deformity surgery. We found 5 patient factors (BMI, ASA, osteotomy, etiology, and previous surgery, and 3 surgeon-controlled factors (surgical time, antibiotics, MIS) associated with risk. The Pediatric Scoliosis Infection Risk Score (PSIR) Score can be applied for risk stratification and to investigate implementation of novel protocols to reduce infection rates in high-risk patients.

Time-Dependent Bacterial Contamination of a Surgical Suction Tip

Background: No in vitro surgical study has evaluated the time-dependent contamination of surgical suction tips compared with controls. Our purpose was to determine the difference in suction tip bacterial contamination rates between suction-positive and suction-negative tips. Materials and Methods: A matched-pair analysis of the contamination of surgical suction tips over a six-hour period was performed in two clean operating rooms. One suction tip was connected to standard wall suction (suction-positive group), with a matched control tip not connected to wall suction (suction-negative group). At time zero and then at hourly intervals for six hours, the distal 3 cm of suction tips were removed, placed in nutrient broth for 48 hours, then plate cultured. One hundred tips were collected for each time interval. Results: Eighty-two of 700 (11.7%) suction tips had bacterial contamination. Sixty-three (18.0%) of 350 suction-positive tips were contaminated, with 19 (5.4%) of the 350 suction-negative tips contaminated (χ2 = 26.7, p < 0.001). Suction tip contamination was time-dependent with the first significant difference between groups occurring after two hours of continuous suction (χ2 = 4.0, p = 0.04). Contamination rate in the suction-positive group increased significantly after one hour compared with time-zero controls (χ2 = 7.1, p = 0.008). There was no significant difference in frequency of positive cultures over time in the suction-negative group compared with time-zero controls. Conclusions: This is the first controlled laboratory study suggesting a time-dependent increase in positive suction tip cultures. From our data, operating room staff should have an awareness that suction tips represent a potential source of bacterial concentration. We recommend that when not in use, suction tip valves be closed if this feature is available, that hosing be manipulated to cease suction when not needed, that suckers be disconnected from tubing, or that suckers be exchanged at frequent intervals. Doing so may reduce bacterial contamination on the suction tip.

Can we automate spine fusion surgical site infection data capture?

PURPOSE

Spine fusion surgical site infection (SSI) rate is reported to national quality databases and used as a benchmark for orthopedic departments and hospital systems. However, accurate data require resource-heavy administrative review and even this has shown to vary. We aimed to create a passive electronic medical record (EMR) algorithm to automatically capture spine fusion SSI and compared its accuracy against the administrative chart review and self-reported morbidity and mortality (M&M) rates.

METHODS

We retrospectively reviewed a single institution's spine fusion records for 7 years for all 90-day post-operative SSIs. We used Centers for Disease Control and Prevention (CDC) SSI definition coupled with intention to treat as an infection by orthopedics/infectious disease service as the gold standard. We compared our gold standard to administrative hand-checked SSI data, anonymously reported departmental M&M, and a passive EMR algorithm (ICD-9 or -10 post-operative SSI diagnosis code entered, or all four of: positive culture, antibiotic prescription between 3-90 days post-op, re-operation/re-admission, and a qualifying diagnosis).

RESULTS

Nine hundred and fourteen spine fusions were included, with a 2.8% SSI rate (0.9% superficial and 2.0% deep). Passive EMR algorithm was the most sensitive at 89% (vs 76% administrative review, 73% M&M); all were highly specific at 99-100%. M&M was 100% positively predictive, administrative review 95%, and EMR 79%.

CONCLUSION

Our passive EMR algorithm was more sensitive to pediatric spine fusion 90-day SSI than self-reported M&M and hand-checked administrative chart review. Although EMR may over-report, it can be used by others to narrow the initial sample for review, reduce resource burden involved with administrative spine SSI review, and provide a quality check for M&M self-reporting.

LEVEL OF EVIDENCE

III.

Strategies reducing risk of surgical-site infection following pediatric spinal deformity surgery

BACKGROUND

Identifying beneficial preventive strategies for surgical-site infection (SSI) in individual patients with different clinical and surgical characteristics is challenging. The purpose of this study was to investigate the association between preventive strategies and patient risk of SSI taking into consideration baseline risks and estimating the reduction of SSI probability in individual patients attributed to these strategies.

METHODS

Pediatric patients who underwent primary, revision, or final fusion for their spinal deformity at 7 institutions between 2004 and 2018 were included. Preventive strategies included the use of topical vancomycin, bone graft, povidone-iodine (PI) irrigations, multilayered closure, impermeable dressing, enrollment in quality improvement (QI) programs, and adherence to antibiotic prophylaxis. The CDC definition of SSI as occurring within 90 days postoperatively was used. Multiple regression modeling was performed following multiple imputation and multicollinearity testing to investigate the effect of preventive strategies on SSI in individual patients adjusted for patient and surgical characteristics.

RESULTS

Univariable regressions demonstrated that enrollment in QI programs and PI irrigation were significantly associated, and topical vancomycin, multilayered closure, and correct intraoperative dosing of antibiotics trended toward association with reduction of SSI. In the final prediction model using multiple regression, enrollment in QI programs remained significant and PI irrigation had an effect in decreasing risks of SSI by average of 49% and 18%, respectively, at the individual patient level.

CONCLUSION

Considering baseline patient characteristics and predetermined surgical and hospital factors, enrollment in QI programs and PI irrigation reduce the risk of SSI in individual patients. Multidisciplinary efforts should be made to implement these practices to increase patient safety.

LEVEL OF EVIDENCE

Prognostic level III study.

Best Practice Guidelines for Surgical Site Infection in High-risk Pediatric Spine Surgery: Definition, Prevention, Diagnosis, and Treatment

BACKGROUND

Prior "best practice guidelines" (BPG) have identified strategies to reduce the risk of acute deep surgical site infection (SSI), but there still exists large variability in practice. Further, there is still no consensus on which patients are "high risk" for SSI and how SSI should be diagnosed or treated in pediatric spine surgery. We sought to develop an updated, consensus-based BPG informed by available literature and expert opinion on defining high-SSI risk in pediatric spine surgery and on prevention, diagnosis, and treatment of SSI in this high-risk population.

MATERIALS AND METHODS

After a systematic review of the literature, an expert panel of 21 pediatric spine surgeons was selected from the Harms Study Group based on extensive experience in the field of pediatric spine surgery. Using the Delphi process and iterative survey rounds, the expert panel was surveyed for current practices, presented with the systematic review, given the opportunity to voice opinions through a live discussion session and asked to vote regarding preferences privately. Two survey rounds were conducted electronically, after which a live conference was held to present and discuss results. A final electronic survey was then conducted for final voting. Agreement ≥70% was considered consensus. Items near consensus were revised if feasible to achieve consensus in subsequent surveys.

RESULTS

Consensus was reached for 17 items for defining high-SSI risk, 17 items for preventing, 6 for diagnosing, and 9 for treating SSI in this high-risk population. After final voting, all 21 experts agreed to the publication and implementation of these items in their practice.

CONCLUSIONS

We present a set of updated consensus-based BPGs for defining high-risk and preventing, diagnosing, and treating SSI in high-risk pediatric spine surgery. We believe that this BPG can limit variability in practice and decrease the incidence of SSI in pediatric spine surgery.

LEVEL OF EVIDENCE

Not applicable.

Surgical Site Infection after Bone Tumor Surgery: Risk Factors and New Preventive Techniques

The management of malignant bone tumors requires multidisciplinary interventions including chemotherapy, radiation therapy, and surgical tumor resection and reconstruction. Surgical site infection (SSI) is a serious complication in the treatment of malignant bone tumors. Compared to other orthopedic surgeries, the surgical treatment of malignant bone tumors is associated with higher rates of SSIs. In patients with SSIs, additional surgeries, long-term administrations of antibiotics, extended hospital stays, and the postponement of scheduled adjuvant treatments are required. Therefore, SSI may adversely affect functional and oncological outcomes. To improve surgical outcomes in patients with malignant bone tumors, preoperative risk assessments for SSIs, new preventive techniques against SSIs, and the optimal use of prophylactic antibiotics are often required. Previous reports have demonstrated that age, tumor site (pelvis and tibia), extended operative time, implant use, body mass index, leukocytopenia, and reconstruction procedures are associated with an increased risk for SSIs. Furthermore, prophylactic techniques, such as silver and iodine coatings on implants, have been developed and proven to be efficacious and safe in clinical studies. In this review, predictive factors of SSIs and new prophylactic techniques are discussed.

Pediatric surgical site infections in 287 hospitals in the United States, 2015-2018

Among 287 US hospitals reporting data between 2015 and 2018, annual pediatric surgical site infection (SSI) rates ranged from 0% for gallbladder to 10.4% for colon surgeries. Colon, spinal fusion, and small-bowel SSI rates did not decrease with greater surgical volumes in contrast to appendix and ventricular-shunt SSI rates.