Letter to "Preoperative prediction of early mortality after surgery for spinal metastases".

Updated external validation of the SORG machine learning algorithms for prediction of ninety-day and one-year mortality after surgery for spinal metastasis

Author's reply to "Preoperative prediction of early mortality after surgery for spinal metastases".

The objective of this study was to provide a convenient preoperative prediction of the risk of early postoperative mortality. This retrospective study included patients who underwent surgery for spinal metastasis at our hospital between 2009 and 2021. Preoperative blood test data of all patients were collected, and the survival time was calculated by dividing the blood data. A multivariate analysis was conducted using a Cox proportional hazards model to identify prognostic factors. The study population included 83 patients (average: 64.5years), 22 of whom died within 3 months. The most common lesion was the thoracic spine, and incomplete paralysis was observed in 57 patients. The surgical methods included posterior implant fixation (n=17), posterior decompression (n=31), and posterior decompression with fixation (n=35). In the univariate analysis, the presence of abnormal values was significantly associated with postoperative survival in six preoperative blood collection items (hemoglobin, C-reactive protein, albumin, white blood cell, gamma-glutamyl transpeptidase, and lactate dehydrogenase). In a multivariate analysis, four test items (hemoglobin, C-reactive protein, white blood cell, and lactate dehydrogenase) were identified as independent prognostic factors.Comparing cases with ≥3 abnormal values among the above four items (high-risk group; n=23) and those with ≤2 (low-risk group; n=60), there was a significant difference in survival time. In addition, it was possible to predict cases of early death within 3months after surgery with 73% sensitivity and 89% specificity. The study showed that four preoperative blood test abnormalities (hemoglobin, C-reactive protein white blood cell, and lactate dehydrogenase) indicated the possibility of early death within 3months after surgery.

Preoperative prognostication of short-term postoperative mortality in patients with spinal metastatic disease can improve shared decision making around end-of-life care. To (1) develop machine learning algorithms for prediction of short-term mortality and (2) deploy these models in an open access web application. The American College of Surgeons, National Surgical Quality Improvement Program was used to identify patients that underwent operative intervention for metastatic disease. Four machine learning algorithms were developed, and the algorithm with the best performance across discrimination, calibration, and overall performance was integrated into an open access web application. The 30-d mortality for the 1790 patients undergoing surgery for spinal metastatic disease was 8.49%. Preoperative factors used for prognostication were albumin, functional status, white blood cell count, hematocrit, alkaline phosphatase, spinal location (cervical, thoracic, lumbosacral), and severity of comorbid systemic disease (American Society of Anesthesiologist Class). In this population, machine learning algorithms developed to predict 30-d mortality performed well on discrimination (c-statistic), calibration (assessed by calibration slope and intercept), Brier score, and decision analysis. An open access web application was developed for the best performing model and this web application can be found here: https://sorg-apps.shinyapps.io/spinemets/. Machine learning algorithms are promising for prediction of postoperative outcomes in spinal oncology and these algorithms can be integrated into clinically useful decision tools. As the volume of data in oncology continues to grow, creation of learning systems and deployment of these systems as accessible tools may significantly enhance prognostication and management.

Objective To develop a predictive model of early postoperative morbidity and mortality with the purpose of assisting in the selection of the candidates for spinal metastasis surgery.Methods A retrospective analysis of consecutive patients operated for metastatic spinal disease. The possible prognostic preoperative characteristics were gender, age, comorbidities, tumor growth rate, and leukocyte and lymphocyte count in the peripheral blood. The postoperative outcomes were 30-day mortality, 90-day mortality and presence of complications. A predictive model was developed based on factors independently associated with these three outcomes. The final model was then tested for the tendency to predict adverse events, discrimination capacity and calibration.Results A total of 205 patients were surgically treated between 2002 and 2015. The rates of the 30-day mortality, 90-day mortality and presence of complications were of 17%, 42% and 31% respectively. The factors independently associated with these three outcomes, which constituted the predictive model, were presence of comorbidities, no slow-growing primary tumor, and lymphocyte count below 1,000 cells/µL. Exposure to none, one, two or three factors was the criterion for the definition of the following categories of the predictive model: low, moderate, high and extreme risk respectively. Comparing the risk categories, there was a progressive increase in the occurrence of outcomes, following a linear trend. The discrimination capacity was of 72%, 73% and 70% for 30-day mortality, 90-day mortality and complications respectively. No lack of calibration occurred.Conclusion The predictive model estimates morbidity and mortality after spinal metastasis surgery and hierarchizes risks as low, moderate, high and extreme.

Scoring systems for metastatic spine disease focus on predicting long- to medium-term mortality or a combination of perioperative morbidity and mortality. However, accurate prediction of perioperative mortality alone may be the most important factor when considering surgical intervention. We aimed to develop and evaluate a new tool, the H2-FAILS score, to predict 30-day mortality after surgery for metastatic spine disease. Using the National Surgical Quality Improvement Program database, we identified 1195 adults who underwent surgery for metastatic spine disease from 2010 to 2018. Incidence of 30-day mortality was 8.7% (n = 104). Independent predictors of 30-day mortality were used to derive the H2-FAILS score. H2-FAILS is an acronym for: Heart failure (2 points), Functional dependence, Albumin deficiency, International normalized ratio elevation, Leukocytosis, and Smoking (1 point each). Discrimination was assessed using area under the receiver operating characteristic curve (AUC). The H2-FAILS score was compared with the American Society of Anesthesiologists Physical Status Classification (ASA Class), the 5-item modified Frailty Index (mFI-5), and the New England Spinal Metastasis Score (NESMS). Internal validation was performed using bootstrapping. Alpha = 0.05. Predicted 30-day mortality was 1.8% for an H2-FAILS score of 0 and 78% for a score of 6. AUC of the H2-FAILS was 0.77 (95% confidence interval: 0.72-0.81), which was higher than the mFI-5 (AUC 0.58, p < 0.001), ASA Class (AUC 0.63, p < 0.001), and NESMS (AUC 0.70, p = 0.004). Internal validation showed an optimism-corrected AUC of 0.76. The H2-FAILS score accurately predicts 30-day mortality after surgery for spinal metastasis. Prognostic level III.

Commentary: Development of Machine Learning Algorithms for Prediction of 30-Day Mortality After Surgery for Spinal Metastasis.

61. Development and external validation of predictive algorithms for six-week mortality in spinal metastasis using 4304 patients from five institutions

Development and external validation of predictive algorithms for six-week mortality in spinal metastasis using 4,304 patients from five institutions

Retrospective study. This study aimed to evaluate the ability of the mortality and adverse events prediction following metastatic spinal surgery of MRI-based cross-sectional psoas muscle area (PMA). A retrospective chart review, 120 patients who had undergone metastatic spinal surgery were included. The cross-sectional area identified the PMA under MR-imaging at the L3 or L4 pedicle level, which was classified into 3 tertiles. We used univariate and multivariate cox proportional hazard regression to assess whether PMA was associated with 30-day, 90-day, 1-year, and overall mortality. The small psoas tertile group populations had a higher mortality rate than the large psoas tertile group. PMA in T1 and T2 had a probability of a higher 90-d mortality rate than PMA in T3 (T1 VS T3: P = .29 and T2 VS T3: P = .12). The median survival time was 7months, 9months, and 10months in PMA T1, T2, and T3, respectively. PMA in tertile 2 had a significantly higher mortality rate of 38% compared to PMA in tertile 3 (HR 1.38, 95% CI .83-2.32, P = .02). Considering PMA as a continuous variable, every 1mm2 increment of PMA resulted in the increase survivorship of 1% (HR .99 with 95% CI .99-1). The MRI-based cross-sectional PMA tends to predict the 90-d mortality rate and overall mortality rate in spinal metastasis patients who underwent spinal surgery. The PMA should be considered one of the prognostic factors in the treatment of metastatic spinal patients.

ObjectivePatients with spinal metastasis (SM) are at advanced stages of systemic cancer disease. Surgical therapy for SM is a common treatment modality enabling histopathological diagnosis and the prevention of severe neurological deficits. However, surgery for SM in this vulnerable patient cohort may require prolonged postoperative intensive care treatment, which could adversely affect the anticipated benefit of the surgery. We therefore assessed postoperative prolonged mechanical ventilation (PMV) as an indicator for intensive care treatment with regard to potential correlations with early postoperative mortality and overall survival (OS).MethodsBetween 2015 and 2019, 198 patients were surgically treated for SM at the author´s neurosurgical department. PMV was defined as postoperative mechanical ventilation of more than 24 hours. A multivariate analysis was performed to identify pre- and perioperative collectable predictors for 30 days mortality.ResultsTwenty out of 198 patients (10%) with SM suffered from postoperative PMV. Patients with PMV exhibited a median OS rate of 1 month compared to 12 months for patients without PMV (p < 0.0001). The 30 days mortality was 70% and after one year 100%. The multivariate analysis identified “PMV > 24 hrs” (p < 0.001, OR 0.3, 95% CI 0.02-0.4) as the only significant and independent predictor for 30 days mortality (Nagelkerke’s R2 0.38).ConclusionsOur data indicate postoperative PMV to significantly correlate to high early postoperative mortality rates as well as to poor OS in patients with surgically treated SM. These findings might encourage the initiation of further multicenter studies to comprehensively investigate PMV as a so far underestimated negative prognostic factor in the course of surgical treatment for SM.

SummaryThe AHFS90 was developed for the prediction of early mortality in patients ≥ 90 years undergoing hip fracture surgery. The AHFS90 has a good accuracy and in most risk categories a good calibration. In our study population, the AHFS90 yielded a maximum prediction of early mortality of 64.5%.PurposeIdentifying hip fracture patients with a high risk of early mortality after surgery could help make treatment decisions and information about the prognosis. This study aims to develop and validate a risk score for predicting early mortality in patients ≥ 90 years undergoing hip fracture surgery (AHFS90).MethodsPatients ≥ 90 years, surgically treated for a hip fracture, were included. A selection of possible predictors for mortality was made. Missing data were subjected to multiple imputations using chained equations. Logistic regression was performed to develop the AHFS90, which was internally and externally validated. Calibration was assessed using a calibration plot and comparing observed and predicted risks.ResultsOne hundred and two of the 922 patients (11.1%) died ≤ 30 days following hip fracture surgery. The AHFS90 includes age, gender, dementia, living in a nursing home, ASA score, and hemoglobin level as predictors for early mortality. The AHFS90 had good accuracy (area under the curve 0.72 for geographic cross validation). Predicted risks correspond with observed risks of early mortality in four risk categories. In two risk categories, the AHFS90 overestimates the risk. In one risk category, no mortality was observed; therefore, no analysis was possible. The AHFS90 had a maximal prediction of early mortality of 64.5% in this study population.ConclusionThe AHFS90 accurately predicts early mortality after hip fracture surgery in patients ≥ 90 years of age. Predicted risks correspond to observed risks in most risk categories. In our study population, the AHFS90 yielded a maximum prediction of early mortality of 64.5%.

Redo valve surgery is associated with increased risk of mortality that may be underestimated by current risk scores. In this study, we hypothesized that additional echocardiographic assessment of left ventricular diastolic and right ventricular systolic function would have independent prognostic value in the prediction of early postoperative mortality in patients undergoing redo valve surgery. We prospectively evaluated 145 patients who underwent redo mitral or aortic valve surgery at our center. All patients underwent comprehensive preoperative echocardiography. The primary outcome was all-cause mortality at 30days. The 30-day mortality rate was 11.7%. Independent of EuroSCORE II both preoperative left ventricular diastolic dysfunction and right ventricular systolic dysfunction were a significant multivariable predictors of 30-day mortality (HR 5.47; 95% CI 1.12-26.74, P=0.036 and HR 4.09; 95% CI 1.11-15.07, P=0.035, respectively) in addition to EuroSCORE II. Diastolic dysfunction remained significant when added to other clinically significant variables. The assessment of both parameters increased the discriminatory power of EuroSCORE II for prediction of early mortality and the combination identified a group at very high risk of mortality. Comprehensive preoperative echocardiography including assessment of left ventricular diastolic and right ventricular systolic function has independent prognostic value over and above EuroSCORE II in the prediction of early postoperative mortality in patients undergoing redo valve surgery. The results of preoperative echocardiography should be taken into account during the selection and perioperative management of patients undergoing redo valvular surgery.

This study examines the efficacy of the predicting power for hospital mortality and functional outcome of three different scoring systems for head injury in a neurosurgical intensive care unit (NICU). On the day of admission, data were collected from each patient to compute the Acute Physiology, Age, and Chronic Health Evaluation (APACHE) II and III, and Glasgow Coma Scale (GCS) scores. Hospital mortality was defined as the deaths of patients before discharge from hospital. Early mortality was defined as death before the 14th day after admission. Late mortality was defined as death after the 15th day from admission. Functional outcome was evaluated by Index of Independence in Activities of Daily Living (Index of ADL). An 8-bed NICU in a 1270-bed medical center in Taichung Veterans General Hospital. Two hundred non-selected patients with acute head injury were included in our study in a consecutive period of 2 years. Patients less than 14 years old were not included. None. Sensitivity, specificity and correct prediction outcome were measured by the chi-square method in three scoring systems. The Youden index was also obtained. The best cut-off point in each scoring system was determined by the Youden index. The difference in Youden index was calculated by Z score. A difference was also considered if the probability value was less than 0.05. The area under Receiver Operating Characteristic (ROC) curve was computed. Then the area under ROC of each scoring system was compared by Z score. There was statistical significance if p was less than 0.05. For prediction of hospital mortality, the best cut-off points are 55 for APACHE III, 17 for APACHE II and 5 for GCS. The correct prediction outcome is 82.4% in APACHE III, 78.4% in APACHE II and 81.9% in the GCS. The Youden index has best cut-off points at 0.68 for APACHE III 0.59 for APACHE II, and 0.56 for GCS. The area under Receiver Operating Characteristic (ROC) curve is 0.90 in the APACHE III, 0.84 in the APACHE II and 0.86 in the GVS. There are no statistical differences among APACHE III and II, and GCS in terms of correct prediction outcome, Youden Index and the area under the ROC curve. Other physiological variables excluding GCS in APACHE III and II (AP III-GCS, AP II-GCS) have less statistical value in the determination of mortality for acute head injury. For the prediction of late mortality, APACHE III and II yield significantly better results in the area under the ROC curve, correct prediction and Youden index than those of GCS. Other physiological variables (AP III-GCS and AP II-GCS) play an important role in the prediction of late mortality in APACHE scores. For prediction of the functional outcome of surviving patients with acute head injury, the APACHE III yields the best results of correct prediction outcome, Youden index and the area under the ROC curve. The APACHE III and II may not replace the role of GCS in cases of acute head injury for hospital or early mortality assessment. But for prediction of the late mortality, the APACHE III and II have better accuracy than GCS. Other physiological variables excluding GCS in the APACHE system play a crucial contribution for late mortality. GCS is simple, less time-consuming and economical for patients with acute head injury for the prediction of hospital and early mortality. The APACHE III provides better prediction for severe morbidity than GCS and APACHE II. Therefore, the APACHE III provides a good assessment not only for hospital and late mortality, but also for functional outcome.

BackgroundTraumatic brain injury (TBI) is a common cause of mortality and disability worldwide. Choosing an appropriate diagnostic tool is critical in early stage for appropriate decision about primary diagnosis, medical care and prognosis.ObjectivesThis study aimed to compare the Glasgow coma scale (GCS), full outline of unresponsiveness (FOUR) and acute physiology and chronic health evaluation (APACHE II) with respect to prediction of the mortality rate of patients with TBI admitted to intensive care unit.Patients and MethodsThis diagnostic study was conducted on 80 patients with TBI in educational hospitals. The scores of APACHE II, GCS and FOUR were recorded during the first 24 hours of admission of patients. In this study, early mortality means the patient death before 14 days and delayed mortality means the patient death 15 days after admitting to hospital. The collected data were analyzed using descriptive and inductive statistics.ResultsThe results showed that the mean age of the patients was 33.80 ± 12.60. From a total of 80 patients with TBI, 16 (20%) were females and 64 (80%) males. The mortality rate was 15 (18.7%). The results showed no significant difference among three tools. In prediction of early mortality, the areas under the curve (AUCs) were 0.92 (CI = 0.95. 0.81 - 0.97), 0.90 (CI = 0.95. 0.74 - 0.94), and 0.96 (CI = 0.95. 0.87 - 0.9) for FOUR, APACHE II and GCS, respectively. In delayed mortality, the AUCs were 0.89 (CI = 0.95. 0.81-0.94), 0.94 (CI = 0.95. 0.74 - 0.97) and 0.90 (CI = 0.95. 0.87 - 0.95) for FOUR, APACHE II and GCS, respectively.ConclusionsConsidering that GCS is easy to use and the FOUR can diagnose a locking syndrome along same values of subscales. These two subscales are superior to APACHI II in prediction of early mortality. Conversation APACHE II is more punctual in the prediction of delayed mortality.