Construct and Validate a Predictive Model for Surgical Site Infection after Posterior Lumbar Interbody Fusion Based on Machine Learning Algorithm

Risk factors for deep surgical site infection following open posterior lumbar fusion: A retrospective case-control study

This study aimed to identify risk factors for deep surgical site infection (SSI) following open posterior lumbar fusion (OPLF). We retrospectively analyzed the clinical data of patients who underwent OPLF between January 2014 and December 2022. Patients were divided into SSI and non-SSI groups according to whether deep SSI occurred following OPLF. Patient's sex, age, body mass index (BMI), history of diabetes mellitus and smoking, American Society of Anesthesiologists score, surgical segment, surgical time, preoperative albumin level, local use of vancomycin, and cerebrospinal fluid (CSF) leakage were compared between the 2 groups. Univariate and multivariate logistic regression analyses were used to identify risk factors for postoperative deep SSI. The deep SSI rate was 5.0% (63/1256). Among them, age (P < .001), BMI (P = .008), surgical segment (P < .001), surgical time (P < .001), prevalence of diabetes mellitus (P = .036), and CSF leakage (P < .001) were significantly higher in the SSI group, whereas the preoperative albumin level (P < .001) and proportion of local use of vancomycin (P = .046) were significantly lower in the SSI group than those in the non-SSI group. Multivariate analysis indicated that higher age (P = .046, odds ratio [OR]: 1.036, 95% confidence interval [CI]: 1.001-1.073), BMI (P = .038, OR: 1.113, 95% CI: 1.006-1.232), lower preoperative albumin level (P = .041, OR: 0.880, 95% CI: 0.778-0.995), higher surgical segment (P = .004, OR: 2.241, 95% CI: 1.297n3.871), and CSF leakage (P = .046, OR: 2.372, 95% CI: 1.015-5.545) were risk factors, and the local use of vancomycin (P < .001, OR: 0.093, 95% CI: 0.036-0.245) was the protective factor for deep SSI following OPLF. We identified 5 risk factors (older age and BMI, lower preoperative albumin level, higher surgical segment, and CSF leakage) and 1 protective factor (local use of vancomycin powder) for deep SSI following OPLF. To address these risk and protective factors, comprehensive evaluations and recommendations should be provided to patients to reduce SSI rates.

Revolutionizing spinal interventions: a systematic review of artificial intelligence technology applications in contemporary surgery

Leveraging its ability to handle large and complex datasets, artificial intelligence can uncover subtle patterns and correlations that human observation may overlook. This is particularly valuable for understanding the intricate dynamics of spinal surgery and its multifaceted impacts on patient prognosis. This review aims to delineate the role of artificial intelligence in spinal surgery. A search of the PubMed database from 1992 to 2023 was conducted using relevant English publications related to the application of artificial intelligence in spinal surgery. The search strategy involved a combination of the following keywords: "Artificial neural network," "deep learning," "artificial intelligence," "spinal," "musculoskeletal," "lumbar," "vertebra," "disc," "cervical," "cord," "stenosis," "procedure," "operation," "surgery," "preoperative," "postoperative," and "operative." A total of 1,182 articles were retrieved. After a careful evaluation of abstracts, 90 articles were found to meet the inclusion criteria for this review. Our review highlights various applications of artificial neural networks in spinal disease management, including (1) assessing surgical indications, (2) assisting in surgical procedures, (3) preoperatively predicting surgical outcomes, and (4) estimating the occurrence of various surgical complications and adverse events. By utilizing these technologies, surgical outcomes can be improved, ultimately enhancing the quality of life for patients.

Digital wound monitoring with artificial intelligence to prioritise surgical wounds in cardiac surgery patients for priority or standard review: protocol for a randomised feasibility trial (WISDOM)

INTRODUCTION

Digital surgical wound monitoring for patients at home is becoming an increasingly common method of wound follow-up. This regular monitoring improves patient outcomes by detecting wound complications early and enabling treatment to start before complications worsen. However, reviewing the digital data creates a new and additional workload for staff. The aim of this study is to assess a surgical wound monitoring platform that uses artificial intelligence to assist clinicians to review patients' wound images by prioritising concerning images for urgent review. This will manage staff time more effectively.

METHODS AND ANALYSIS

This is a feasibility study for a new artificial intelligence module with 120 cardiac surgery patients at two centres serving a range of patient ethnicities and urban, rural and coastal locations. Each patient will be randomly allocated using a 1:1 ratio with mixed block sizes to receive the platform with the new detection and prioritising module (for up to 30 days after surgery) plus standard postoperative wound care or standard postoperative wound care only. Assessment is through surveys, interviews, phone calls and platform review at 30 days and through medical notes review and patient phone calls at 60 days. Outcomes will assess safety, acceptability, feasibility and health economic endpoints. The decision to proceed to a definitive trial will be based on prespecified progression criteria.

ETHICS AND DISSEMINATION

Permission to conduct the study was granted by the North of Scotland Research Ethics Committee 1 (24/NS0005) and the MHRA (CI/2024/0004/GB). The results of this Wound Imaging Software Digital platfOrM (WISDOM) study will be reported in peer-reviewed open-access journals and shared with participants and stakeholders.

TRIAL REGISTRATION NUMBERS

ISRCTN16900119 and NCT06475703.

Prediction models for risk assessment of surgical site infection after spinal surgery: A systematic review

Background

Spinal surgeries are a common procedure, but there is significant risk of adverse events following these operations. While the rate of adverse events ranges from 8% to 18%, surgical site infections (SSIs) alone occur in between 1% and 4% of spinal surgeries.

Methods

We completed a systematic review addressing factors that contribute to surgical site infection after spinal surgery. From the included studies, we separated the articles into groups based on whether they propose a clinical predictive tool or model. We then compared the prediction variables, model development, model validation, and model performance.

Results

About 47 articles were included in this study: 10 proposed a model and 5 validated a model. The models were developed from 7,720 participants in total and 210 participants with SSI. Only one of the proposed models was externally validated by an independent group. The other 4 validation papers examined the performance of the ACS NSQIP surgical risk calculator.

Conclusions

While some preoperative risk models have been validated, and even successfully implemented clinically, the significance of postoperative SSIs and the unique susceptibility of spine surgery patients merits the development of a spine-specific preoperative risk model. Additionally, comprehensive and stratified risk modeling for SSI would be of invaluable clinical utility and greatly improve the field of spine surgery.

Limitations in Evaluating Machine Learning Models for Imbalanced Binary Outcome Classification in Spine Surgery: A Systematic Review

Clinical prediction models for spine surgery applications are on the rise, with an increasing reliance on machine learning (ML) and deep learning (DL). Many of the predicted outcomes are uncommon; therefore, to ensure the models' effectiveness in clinical practice it is crucial to properly evaluate them. This systematic review aims to identify and evaluate current research-based ML and DL models applied for spine surgery, specifically those predicting binary outcomes with a focus on their evaluation metrics. Overall, 60 papers were included, and the findings were reported according to the PRISMA guidelines. A total of 13 papers focused on lengths of stay (LOS), 12 on readmissions, 12 on non-home discharge, 6 on mortality, and 5 on reoperations. The target outcomes exhibited data imbalances ranging from 0.44% to 42.4%. A total of 59 papers reported the model's area under the receiver operating characteristic (AUROC), 28 mentioned accuracies, 33 provided sensitivity, 29 discussed specificity, 28 addressed positive predictive value (PPV), 24 included the negative predictive value (NPV), 25 indicated the Brier score with 10 providing a null model Brier, and 8 detailed the F1 score. Additionally, data visualization varied among the included papers. This review discusses the use of appropriate evaluation schemes in ML and identifies several common errors and potential bias sources in the literature. Embracing these recommendations as the field advances may facilitate the integration of reliable and effective ML models in clinical settings.

Is Prolonged Use of Antibiotic Prophylaxis and Postoperative Antithrombotic and Antispasmodic Treatments Necessary After Digit Replantation or Revascularization?

BACKGROUND

Ensuring the patency of repaired vessels is pivotal in improving the success rate of digit replantation. There is no consensus on how to best approach postoperative treatment for digit replantation. The influence of postoperative treatment on the risk of failure of revascularization or replantation remains unclear.

QUESTIONS/PURPOSES

(1) Is there an increased risk of postoperative infection with early discontinuation of antibiotic prophylaxis? (2) How are anxiety and depression affected by a treatment protocol consisting of prolonged antibiotic prophylaxis and administration of antithrombotic and antispasmodic drugs and by the failure of a revascularization or replantation procedure? (3) Are there differences in the risk of revascularization or replantation failure based on the number of anastomosed arteries and veins? (4) What factors are associated with failure of revascularization or replantation?

METHODS

This retrospective study was conducted between July 1, 2018, and March 31, 2022. Initially, 1045 patients were identified. One hundred two patients chose revision of amputation. In all, 556 were excluded because of contraindications. We included all patients in whom the anatomic structures of the amputated part of the digit were well preserved, and those with an ischemia time for the amputated part that did not exceed 6 hours. Patients in good health without any other serious associated injuries or systemic diseases and those without a history of smoking were eligible for inclusion. The patients underwent procedures that were performed or supervised by one of four study surgeons. Patients were treated with antibiotic prophylaxis (1 week); patients treated with antithrombotic and antispasmodic drugs were categorized into the prolonged antibiotic prophylaxis group. The remaining patients treated with antibiotic prophylaxis for less than 48 hours and no antithrombotic and no antispasmodic drugs were categorized into the nonprolonged antibiotic prophylaxis group. Postoperative follow-up was for a minimum of 1 month. Based on the inclusion criteria, 387 participants with 465 digits were selected for an analysis of postoperative infection. Twenty-five participants with a postoperative infection (six digits) and other complications (19 digits) were excluded from the next stage of the study, in which we assessed factors associated with the risk of failure of revascularization or replantation. A total of 362 participants with 440 digits were examined, including the postoperative survival rate, variation in Hospital Anxiety and Depression Scale scores, the association between the survival rate and Hospital Anxiety and Depression Scale scores, and the survival rate based on the number of anastomosed vessels. Postoperative infection was defined as swelling, erythema, pain, purulent discharge, or a positive bacterial culture result. Patients were followed for 1 month. The differences in anxiety and depression scores between the two treatment groups and the differences in anxiety and depression scores based on failure of revascularization or replantation were determined. The difference in the risk of revascularization or replantation failure based on the number of anastomosed arteries and veins was assessed. Except for statistically significant variables (injury type and procedure), we thought that the number of arteries, number of veins, Tamai level, treatment protocol, and surgeons would be important. A multivariable logistic regression analysis was used to perform an adjusted analysis of risk factors such as postoperative protocol, injury type, procedure, number of arteries, number of veins, Tamai level, and surgeon.

RESULTS

Postoperative infection did not appear to increase without prolonged use of antibiotic prophylaxis beyond 48 hours (1% [3 of 327] versus 2% [3 of 138]; OR 2.4 [95% confidence interval (CI) 0.5 to 12.0]; p = 0.37). Intervention with antithrombotic and antispasmodic therapy increased the Hospital Anxiety and Depression Scale scores for anxiety (11.2 ± 3.0 versus 6.7 ± 2.9, mean difference 4.5 [95% CI 4.0 to 5.2]; p < 0.01) and depression (7.9 ± 3.2 versus 5.2 ± 2.7, mean difference 2.7 [95% CI 2.1 to 3.4]; p < 0.01). In the analysis based on the failure of revascularization or replantation, the Hospital Anxiety and Depression Scale scores for anxiety (11.4 ± 4.4 versus 9.7 ± 3.5, mean difference 1.7 [95% CI 0.6 to 2.8]; p < 0.01) and depression (8.5 ± 4.6 versus 7.0 ± 3.1, mean difference 1.5 [95% CI 0.5 to 2.5]; p < 0.01) were higher in the failed revascularization or replantation group than in the successful revascularization or replantation group. There was no increase in the artery-related risk of failure (one versus two anastomosed arteries: 91% versus 89%, OR 1.3 [95% CI 0.6 to 2.6]; p = 0.53). For patients with anastomosed veins, a similar outcome was observed for the two vein-related risk of failure (two versus one anastomosed vein: 90% versus 89%, OR 1.0 [95% CI 0.2 to 3.8]; p = 0.95) and three vein-related risk of failure (three versus one vein anastomosed: 96% versus 89%, OR 0.4 [95% CI 0.1 to 2.4]; p = 0.29). Factors associated with failure of revascularization or replantation included the mechanism of injury (crush: OR 4.2 [95% CI 1.6 to 11.2]; p < 0.01, avulsion: OR 10.2 [95% CI 3.4 to 30.7]; p < 0.01). Revascularization had a lower risk of failure than replantation (OR 0.4 [95% CI 0.2 to 1.0]; p = 0.04). Treatment with a protocol of prolonged antibiotics, antithrombotics, and antispasmodics was not associated with a lower risk of failure (OR 1.2 [95% CI 0.6 to 2.3]; p = 0.63).

CONCLUSION

With proper wound debridement and patency of repaired vessels, prolonged use of antibiotic prophylaxis and regular antithrombotic and antispasmodic treatment may not be necessary for successful digit replantation. However, it may be associated with higher Hospital Anxiety and Depression Scale scores. Postoperative mental status is associated with digit survival. Well-repaired vessels, instead of the number of anastomosed vessels, could be critical to survival and decrease the influence of risk factors. Further research on consensus guidelines that compare postoperative treatment and the surgeon's level of expertise after digit replantation should be conducted at multiple institutions.

LEVEL OF EVIDENCE

Level III, therapeutic study.