Non-operative management of spinal metastases: A prognostic model for failure

Developmental and Validation of Machine Learning Model for Prediction Complication after Cervical Spine Metastases Surgery

STUDY DESIGN

This is a retrospective cohort study utilizing machine learning to predict postoperative complications in cervical spine metastases surgery.

OBJECTIVES

The main objective is to develop a machine learning model that accurately predicts complications following cervical spine metastases surgery.

SUMMARY OF BACKGROUND DATA

Cervical spine metastases surgery can enhance quality of life but carries a risk of complications influenced by various factors. Existing scoring systems may not include all predictive factors. Machine learning offers the potential for a more accurate predictive model by analyzing a broader range of variables.

METHODS

Data from January 2012 to December 2020 were retrospectively collected from medical databases. Predictive models were developed using Gradient Boosting, Logistic Regression, and Decision Tree Classifier algorithms. Variables included patient demographics, disease characteristics, and laboratory investigations. SMOTE was used to balance the dataset, and the models were assessed using AUC, F1-score, precision, recall, and SHAP values.

RESULTS

The study included 72 patients, with a 29.17% postoperative complication rate. The Gradient Boosting model had the best performance with an AUC of 0.94, indicating excellent predictive capability. Albumin level, platelet count, and tumor histology were identified as top predictors of complications.

CONCLUSIONS

The Gradient Boosting machine learning model showed superior performance in predicting postoperative complications in cervical spine metastases surgery. With continuous data updating and model training, machine learning can become a vital tool in clinical decision-making, potentially improving patient outcomes.

LEVEL OF EVIDENCE

Level III.

Development and internal validation of machine-learning models for predicting survival in patients who underwent surgery for spinal metastases

STUDY DESIGN

A retrospective study.

PURPOSE

This study aimed to develop machine-learning algorithms for predicting survival in patients who underwent surgery for spinal metastasis.

OVERVIEW OF LITERATURE

This study develops machine-learning models to predict postoperative survival in spinal metastasis patients, filling the gaps of traditional prognostic systems. Utilizing data from 389 patients, the study highlights XGBoost and CatBoost algorithms̓ effectiveness for 90, 180, and 365-day survival predictions, with preoperative serum albumin as a key predictor. These models offer a promising approach for enhancing clinical decision-making and personalized patient care.

METHODS

A registry of patients who underwent surgery (instrumentation, decompression, or fusion) for spinal metastases between 2004 and 2018 was used. The outcome measure was survival at postoperative days 90, 180, and 365. Preoperative variables were used to develop machine-learning algorithms to predict survival chance in each period. The performance of the algorithms was measured using the area under the receiver operating characteristic curve (AUC).

RESULTS

A total of 389 patients were identified, with 90-, 180-, and 365-day mortality rates of 18%, 41%, and 45% postoperatively, respectively. The XGBoost algorithm showed the best performance for predicting 180-day and 365-day survival (AUCs of 0.744 and 0.693, respectively). The CatBoost algorithm demonstrated the best performance for predicting 90-day survival (AUC of 0.758). Serum albumin had the highest positive correlation with survival after surgery.

CONCLUSIONS

These machine-learning algorithms showed promising results in predicting survival in patients who underwent spinal palliative surgery for spinal metastasis, which may assist surgeons in choosing appropriate treatment and increasing awareness of mortality-related factors before surgery.

C-reactive Protein-to-albumin Ratio in Spinal Epidural Abscess: Association with Post-treatment Complications

INTRODUCTION

Spinal epidural abscess (SEA) is a complex medical condition with high morbidity and healthcare costs. Clinical presentation and laboratory data may have prognostic value in forecasting morbidity and mortality. C-reactive protein-to-albumin ratio (CAR) demonstrates promise for the prediction of adverse events in multiple orthopaedic and nonorthopaedic surgical conditions. We investigated the relationship between CAR and outcomes after treatment of SEA.

METHODS

We retrospectively evaluated adult patients treated within a single healthcare system for a diagnosis of SEA (2005 to 2017). Laboratory and clinical data included age at diagnosis, sex, race, body mass index, smoking status, history of intravenous drug use, Charlson Comorbidity Index, and CAR. The primary outcome was the occurrence of any complication; mortality and readmissions were considered secondarily. We used logistic regression to determine the association between baseline CAR and outcomes, adjusting for confounders.

RESULTS

We included 362 patients with a 90-day mortality rate of 13.3% and a 90-day complication rate of 47.8%. A reduced complication rate was observed in the lowest decile of CAR values compared with the remaining 90% of patients, a threshold value of 2.5 (27.0% versus 50.2%; odds ratio [OR] 2.66, 95% confidence interval [CI] 1.22 to 5.81). CAR values in the highest two deciles experienced significantly increased odds of complications compared with the lowest decile (80th: OR 3.44; 95% CI 1.25 to 9.42; 90th: OR 3.28; 95% CI 1.19 to 9.04).

DISCUSSION

We found elevated CAR to be associated with an increased likelihood of major morbidity in SEA. We suggest using a CAR value of 2.5 as a threshold for enhanced surveillance and recognizing patients with values above 73.7 as being at exceptional risk of morbidity.

LEVEL OF EVIDENCE

Level III observational cohort study.

Epidemiologic Study of Operative Treatment for Spinal Metastasis in Thailand : A Review of National Healthcare Data from 2005 to 2014

Objective

To study the factors relating to operative treatment for spinal metastasis in Thailand during 2005-2014 and to determine the hospital costs, mortality rate, and incidence of perioperative complication.

Methods

Inpatient reimbursement data from 2005 to 2014 was reviewed from three national healthcare organizations, including the National Health Security Office, the Social Security Office, and the Comptroller General's Department. The search criteria were secondary malignant neoplasm of bone and bone marrow patients (International Classification of Diseases 10th revision, Thai modification codes [ICD 10-TM], C79.5 and C79.8) who underwent spinal surgical treatment (ICD 9th revision, clinical modification procedure with extension codes [ICD 9-CM], 03.0, 03.4, 03.09, and 81.0) during 2005-2014. Epidemiology, comorbidity, and perioperative complication were analyzed.

Results

During the study period, the number of spinal metastasis patients who underwent operative treatment was significantly increased from 0.30 to 0.59 per 100000 (p<0.001). More males (56.14%) underwent surgical treatment for spinal metastasis than females. The most common age group was 45-64 (55.1%). The most common primary tumor sites were the unknown origin, lung, breast, prostate, and hepatocellular/bile duct. Interestingly, the proportion of hepatocellular/bile duct, breast, and lung cancer was significantly increased (p<0.001). The number of patients who had comorbidity or in-hospital complication significantly increased over time (p<0.01); however, the in-hospital mortality rate decreased.

Conclusion

During the last decade, operative treatment for spinal metastasis increased in Thailand. The overall in-hospital complication rate increased; however, the in-hospital mortality rate decreased.

Surgical Stabilization for Patients with Mechanical Back Pain Secondary to Metastatic Spinal Disease is Associated with Improved Objective Mobility Metrics: Preliminary Analysis in a Cohort of 26 Patients

OBJECTIVES

To investigate the effect of surgical stabilization for patients with metastatic spinal disease on objective mobility metrics.

METHODS

A retrospective chart review identified patients who had mechanical back pain from metastatic spinal disease and underwent spinal stabilization during 2017. Mobility metrics, the Activity Measure for Post-Acute Care (AM-PAC) inpatient mobility short form (IMSF) and the Johns Hopkins Highest Level of Mobility (JH-HLM), were reviewed.

RESULTS

A total of 26 patients were included in the analysis with median hospital stay of 8 days. Preoperative JH-HLM scores were available for 17 patients with a mean score of 5.4, increasing to mean score of 6.6 at last follow-up (P = 0.036). Preoperative AM-PAC IMSF scores were available for 14 patients with a mean score of 19.4, decreasing slightly to a mean score of 18.7 at last follow-up (P = 0.367). Last follow-up with mobility metrics occurred a median of 6.5 days postoperatively (range: 3-66 days). Multivariable analysis showed that American Spinal Injury Association and Karnofsky Performance Status scores were significantly associated with both JH-HLM and AM-PAC mobility scores at last follow-up. A higher JH-HLM or AM-PAC score was significantly associated with direct home discharge and a higher AM-PAC score was associated with shorter hospital stay.

CONCLUSIONS

Surgical stabilization for patients with mechanical back pain secondary to metastatic spinal disease might lead to an objective improvement in JH-HLM score. JH-HLM and AM-PAC scores may be correlated with length of hospital stay and discharge disposition. Future studies are encouraged to further characterize the role of these mobility metrics in the management plan of these patients.