scholarly journals Death After Discharge: Prognostic Model of 1-year Mortality in Traumatic Brain Injury Patients Undergoing Decompressive Craniectomy

2020 ◽  
Author(s):  
Wenxing Cui ◽  
Shunnan Ge ◽  
Yingwu Shi ◽  
Xun Wu ◽  
Jianing Luo ◽  
...  
Keyword(s):  
Traumatic Brain Injury ◽  
Logistic Regression ◽  
Brain Injury ◽  
Random Forest ◽  
Decompressive Craniectomy ◽  
Prognostic Model ◽  
Mortality Prediction ◽  
Random Tree ◽  
Prognostic Models ◽  
Biochemical Tests

Abstract OBJECTIVE: Despite advances in decompressive craniectomy (DC) for the treatment of traumatic brain injury (TBI), a high risk of poor long-term prognosis exists in these patients. The aim of this study is to predict 1-year mortality in TBI patients undergoing DC using the logistic regression and random tree models.METHODS: This was a retrospective analysis of TBI patients undergoing DC from January 1, 2015 to April 25, 2019. Patient demographic characteristics, biochemical tests and intraoperative factors were collected. 1-year mortality prognostic models were developed using multivariate logistic regression and random tree algorithms. Overall accuracy, sensitivity, specificity and area under the receiver operating characteristic curves (AUC) were used to evaluate model performance.RESULTS: Of the 230 patients, 70 (30.4%) died within 1 year. Older age (OR, 1.066; 95% CI, 1.045-1.087; P < 0.001), higher Glasgow coma score (GCS) (OR, 0.737; 95% CI, 0.660-0.824; P < 0.001), higher d-dimer (OR, 1.005; 95% CI, 1.001-1.009; P = 0.015), coagulopathy (OR, 2.965; 95% CI, 1.808-4.864; P < 0.001), hypotension (OR, 3.862; 95% CI, 2.176-6.855; P < 0.001) and completely effaced basal cisterns (OR, 3.766; 95% CI, 2.255-6.290; P < 0.001) were independent predictors of 1-year mortality. Random forest demonstrated better performance for 1-year mortality prediction, which achieved overall accuracy of 0.810, sensitivity of 0.833, specificity of 0.800, and AUC of 0.830 on the testing data, compared to logistic regression model.CONCLUSIONS: Random forest model showed relatively good predictive performance for 1-year mortality in TBI patients undergoing DC. Further external test is required to verify our prognostic model.

scholarly journals Death after discharge: prognostic model of 1-year mortality in traumatic brain injury patients undergoing decompressive craniectomy

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Wenxing Cui ◽  
Shunnan Ge ◽  
Yingwu Shi ◽  
Xun Wu ◽  
Jianing Luo ◽  
...  
Keyword(s):  
Traumatic Brain Injury ◽  
Logistic Regression ◽  
Brain Injury ◽  
Random Forest ◽  
Decompressive Craniectomy ◽  
Prognostic Model ◽  
Mortality Prediction ◽  
Random Tree ◽  
Prognostic Models ◽  
Biochemical Tests

Abstract Background Despite advances in decompressive craniectomy (DC) for the treatment of traumatic brain injury (TBI), these patients are at risk of having a poor long-term prognosis. The aim of this study was to predict 1-year mortality in TBI patients undergoing DC using logistic regression and random tree models. Methods This was a retrospective analysis of TBI patients undergoing DC from January 1, 2015, to April 25, 2019. Patient demographic characteristics, biochemical tests, and intraoperative factors were collected. One-year mortality prognostic models were developed using multivariate logistic regression and random tree algorithms. The overall accuracy, sensitivity, specificity, and area under the receiver operating characteristic curves (AUCs) were used to evaluate model performance. Results Of the 230 patients, 70 (30.4%) died within 1 year. Older age (OR, 1.066; 95% CI, 1.045–1.087; P < 0.001), higher Glasgow Coma Score (GCS) (OR, 0.737; 95% CI, 0.660–0.824; P < 0.001), higher d-dimer (OR, 1.005; 95% CI, 1.001–1.009; P = 0.015), coagulopathy (OR, 2.965; 95% CI, 1.808–4.864; P < 0.001), hypotension (OR, 3.862; 95% CI, 2.176–6.855; P < 0.001), and completely effaced basal cisterns (OR, 3.766; 95% CI, 2.255–6.290; P < 0.001) were independent predictors of 1-year mortality. Random forest demonstrated better performance for 1-year mortality prediction, which achieved an overall accuracy of 0.810, sensitivity of 0.833, specificity of 0.800, and AUC of 0.830 on the testing data compared to the logistic regression model. Conclusions The random forest model showed relatively good predictive performance for 1-year mortality in TBI patients undergoing DC. Further external tests are required to verify our prognostic model.

scholarly journals Lactate Albumin Ratio is Associated with Mortality in Patients with Moderate to Severe Traumatic Brain Injury

2020 ◽  
Author(s):  
Ruoran Wang ◽  
Min He ◽  
Xiaofeng Ou ◽  
Xiaoqi Xie ◽  
Yan Kang
Keyword(s):  
Risk Factors ◽  
Traumatic Brain Injury ◽  
Logistic Regression ◽  
Brain Injury ◽  
Hospital Mortality ◽  
Prognostic Model ◽  
Multivariate Logistic Regression ◽  
Albumin Ratio ◽  
Severe Tbi ◽  
Auc Value

Abstract Background: Traumatic brain injury (TBI) is a serious public health issue all over the world. This study was designed to evaluate the prognostic value of lactate to albumin ratio (LAR) on moderate to severe traumatic brain injury.Methods: Clinical data of 273 moderate to severe TBI patients hospitalized in West China Hospital between May 2015 and January 2018 were collected. Multivariate logistic regression analyses were used to explore risk factors and construct prognostic model of in-hospital mortality in this cohort. Nomogram was drawn to visualize the prognostic model. Receiver operating characteristic (ROC) curve and calibration curve were respectively drawn to evaluate discriminative ability and stability of this model.Results: Non-survivors had higher LAR than survivors (1.0870 vs 0.5286, p<0.001). Results of multivariate logistic regression analysis showed that GCS (OR=0.818, p=0.008), blood glucose (OR=1.232, p<0.001), LAR (OR=1.883, p=0.012), and red blood cell distribution (RDW)-SD (OR=1.179, p=0.004) were independent risk factors of in-hospital mortality in included patients. These four factors were utilized to construct prognostic model. The area under the ROC curve (AUC) value of single lactate and LAR were 0.733 (95%Cl; 0.673-0.794) and 0.780 (95%Cl; 0.725-0.835), respectively. The AUC value of the prognostic model was 0.868 (95%Cl; 0.826-0.909), which was higher than that of LAR (Z=2.5143, p<0.05).Conclusions: LAR is a readily available prognostic marker of moderate to severe TBI patients. Prognostic model incorporating LAR is beneficial for clinicians to evaluate possible progression and make treatment decisions in these patients.

scholarly journals Factors associated with shunt-dependent hydrocephalus after decompressive craniectomy for traumatic brain injury

2018 ◽  
Vol 128 (5) ◽  
pp. 1547-1552 ◽  
Author(s):  
Aditya Vedantam ◽  
Jose-Miguel Yamal ◽  
Hyunsoo Hwang ◽  
Claudia S. Robertson ◽  
Shankar P. Gopinath
Keyword(s):  
Traumatic Brain Injury ◽  
Logistic Regression ◽  
Brain Injury ◽  
Decompressive Craniectomy ◽  
Controlled Trial ◽  
Statistical Significance ◽  
Ventriculoperitoneal Shunting ◽  
Younger Age ◽  
Frontal Horn ◽  
Severe Tbi

OBJECTIVEPosttraumatic hydrocephalus (PTH) affects 11.9%–36% of patients undergoing decompressive craniectomy (DC) and is an important cause of morbidity after traumatic brain injury (TBI). Early diagnosis and treatment of PTH can prevent further neurological compromise in patients who are recovering from TBI. There is limited data on predictors of shunting for PTH after DC for TBI.METHODSProspectively collected data from the erythropoietin severe TBI randomized controlled trial were studied. Demographic, clinical, and imaging data were analyzed for enrolled patients who underwent a DC. All head CT scans during admission were reviewed and assessed for PTH by the Gudeman criteria or the modified Frontal Horn Index ≥ 33%. The presence of subdural hygromas was categorized as unilateral/bilateral hemispheric or interhemispheric. Using L1-regularized logistic regression to select variables, a multiple logistic regression model was created with ventriculoperitoneal shunting as the binary outcome. Statistical significance was set at p < 0.05.RESULTSA total of 60 patients who underwent DC were studied. Fifteen patients (25%) underwent placement of a ventriculoperitoneal shunt for PTH. The majority of patients underwent unilateral decompressive hemicraniectomy (n = 46, 77%). Seven patients (12%) underwent bifrontal DC. Unilateral and bilateral hemispheric hygromas were noted in 31 (52%) and 7 (11%) patients, respectively. Interhemispheric hygromas were observed in 19 patients (32%). The mean duration from injury to first CT scan showing hemispheric subdural hygroma and interhemispheric hygroma was 7.9 ± 6.5 days and 14.9 ± 11.7 days, respectively. The median duration from injury to shunt placement was 43.7 days. Multivariate analysis showed that the presence of interhemispheric hygroma (OR 63.6, p = 0.001) and younger age (OR 0.78, p = 0.009) were significantly associated with the need for a shunt after DC.CONCLUSIONSThe presence of interhemispheric subdural hygromas and younger age were associated with shunt-dependent hydrocephalus after DC in patients with severe TBI.

Prognostic significance of abnormal hematological parameters in severe traumatic brain injury requiring decompressive craniectomy

2020 ◽  
Vol 132 (2) ◽  
pp. 545-551 ◽  
Author(s):  
Jade-Marie Corbett ◽  
Kwok M. Ho ◽  
Stephen Honeybul
Keyword(s):  
Traumatic Brain Injury ◽  
Brain Injury ◽  
Decompressive Craniectomy ◽  
Prognostic Model ◽  
Prognostic Significance ◽  
Unfavorable Outcome ◽  
Severe Tbi ◽  
Hematological Abnormalities ◽  
The Impact ◽  
Predicted Risk

OBJECTIVEHematological abnormalities after severe traumatic brain injury (TBI) are common, and are associated with a poor outcome. Whether these abnormalities offer additional prognostic significance over and beyond validated TBI prognostic models is uncertain.METHODSThis retrospective cohort study compared the ability of admission hematological abnormalities to that of the IMPACT (International Mission for Prognosis and Analysis of Clinical Trials) prognostic model to predict 18-month neurological outcome of 388 patients who required a decompressive craniectomy after severe TBI, between 2004 and 2016, in Western Australia. Area under the receiver operating characteristic (AUROC) curve was used to assess predictors’ ability to discriminate between patients with and without an unfavorable outcome of death, vegetative state, or severe disability.RESULTSOf the 388 patients included in the study, 151 (38.9%) had an unfavorable outcome at 18 months after decompressive craniectomy for severe TBI. Abnormalities in admission hemoglobin (AUROC 0.594, p = 0.002), plasma glucose (AUROC 0.592, p = 0.002), fibrinogen (AUROC 0.563, p = 0.036), international normalized ratio (INR; AUROC 0.645, p = 0.001), activated partial thromboplastin time (AUROC 0.564, p = 0.033), and disseminated intravascular coagulation score (AUROC 0.623, p = 0.001) were all associated with a higher risk of unfavorable outcome at 18 months after severe TBI. As a marker of inflammation, neutrophil to lymphocyte ratio was not significantly associated with the risk of unfavorable outcome (AUROC 0.500, p = 0.998). However, none of these parameters, in addition to the platelet count, were significantly associated with an unfavorable outcome after adjusting for the IMPACT predicted risk (odds ratio [OR] per 10% increment in risk 2.473, 95% confidence interval [CI] 2.061–2.967; p = 0.001). After excluding 8 patients (2.1%) who were treated with warfarin prior to the injury, there was a suggestion that INR was associated with some additional prognostic significance (OR 3.183, 95% CI 0.856–11.833; p = 0.084) after adjusting for the IMPACT predicted risk.CONCLUSIONSIn isolation, INR was the best hematological prognostic parameter in severe TBI requiring decompressive craniectomy, especially when patients treated with warfarin were excluded. However, the prognostic significance of admission hematological abnormalities was mostly captured by the IMPACT prognostic model, such that they did not offer any additional prognostic information beyond the IMPACT predicted risk. These results suggest that new prognostic factors for TBI should be evaluated in conjunction with predicted risks of a comprehensive prognostic model that has been validated, such as the IMPACT prognostic model.

scholarly journals Sequential changes in Rotterdam CT scores related to outcomes for patients with traumatic brain injury who undergo decompressive craniectomy

2016 ◽  
Vol 124 (6) ◽  
pp. 1640-1645 ◽  
Author(s):  
Kenji Fujimoto ◽  
Masaki Miura ◽  
Tadahiro Otsuka ◽  
Jun-ichi Kuratsu
Keyword(s):  
Traumatic Brain Injury ◽  
Logistic Regression ◽  
Brain Injury ◽  
Decompressive Craniectomy ◽  
Roc Curve ◽  
Area Under The Curve ◽  
Prognostic Indicator ◽  
Unfavorable Outcome ◽  
Multivariable Logistic Regression Analysis ◽  
The Relationship

OBJECT Rotterdam CT scoring is a CT classification system for grouping patients with traumatic brain injury (TBI) based on multiple CT characteristics. This retrospective study aimed to determine the relationship between initial or preoperative Rotterdam CT scores and TBI prognosis after decompressive craniectomy (DC). METHODS The authors retrospectively reviewed the medical records of all consecutive patients who underwent DC for nonpenetrating TBI in 2 hospitals from January 2006 through December 2013. Univariate and multivariate logistic regression and receiver operating characteristic (ROC) curve analyses were used to determine the relationship between initial or preoperative Rotterdam CT scores and mortality at 30 days or Glasgow Outcome Scale (GOS) scores at least 3 months after the time of injury. Unfavorable outcomes were GOS Scores 1–3 and favorable outcomes were GOS Scores 4 and 5. RESULTS A total of 48 cases involving patients who underwent DC for TBI were included in this study. Univariate analyses showed that initial Rotterdam CT scores were significantly associated with mortality and both initial and preoperative Rotterdam CT scores were significantly associated with unfavorable outcomes. Multivariable logistic regression analysis adjusted for established predictors of TBI outcomes showed that initial Rotterdam CT scores were significantly associated with mortality (OR 4.98, 95% CI 1.40–17.78, p = 0.01) and unfavorable outcomes (OR 3.66, 95% CI 1.29–10.39, p = 0.02) and preoperative Rotterdam CT scores were significantly associated with unfavorable outcomes (OR 15.29, 95% CI 2.50–93.53, p = 0.003). ROC curve analyses showed cutoff values for the initial Rotterdam CT score of 5.5 (area under the curve [AUC] 0.74, 95% CI 0.59–0.90, p = 0.009, sensitivity 50.0%, and specificity 88.2%) for mortality and 4.5 (AUC 0.71, 95% CI 0.56–0.86, p = 0.02, sensitivity 62.5%, and specificity 75.0%) for an unfavorable outcome and a cutoff value for the preoperative Rotterdam CT score of 4.5 (AUC 0.81, 95% CI 0.69–0.94, p < 0.001, sensitivity 90.6%, and specificity 56.2%) for an unfavorable outcome. CONCLUSIONS Assessment of changes in Rotterdam CT scores over time may serve as a prognostic indicator in TBI and can help determine which patients require DC.

scholarly journals Effect of primary decompressive craniectomy on outcomes in severe traumatic brain injury with mass lesions and the independent predictors for operation decision

2020 ◽  
Author(s):  
Chen Yang ◽  
Jia-Rui Zhang ◽  
Gang Zhu ◽  
Hao Guo ◽  
Fei Gao ◽  
...  
Keyword(s):  
Traumatic Brain Injury ◽  
Logistic Regression ◽  
Brain Injury ◽  
Decompressive Craniectomy ◽  
Pupillary Light Reflex ◽  
Mass Lesion ◽  
Light Reflex ◽  
Pupillary Light ◽  
Mass Lesions ◽  
Basal Cisterns

Abstract Background: Although operative indications for traumatic brain injury (TBI) have been evaluated, neurosurgeons often face a dilemma of whether or not to remove the bone flap after mass lesion evacuation, and a useful predictive scoring model for which patients should be decompressive craniectomy (DC) has yet to be developed. The aim of this study was firstly to compare the outcomes of craniotomy and DC, and secondly to determine independent predictors and develop a multivariate logistic regression equation to determine whom should perform primary DC in TBI patients with mass lesions.Methods: A total of nine different variables were evaluated. All 245 patients with severe TBI in this study were retrospectively evaluated between June 2015 and May 2019 and all underwent decompressive craniectomy (DC) or craniotomy for mass lesion removal. The 6-month mortality and Extended Glasgow Outcome Scale (GOSE) were compared between DC and craniotomy. By using univariate, multiple logistic regression and prognostic regression scoring equations it was possible to draw Receiver Operating Characteristic curves (ROC) to predict the decision for DC.Results: The overall 6-month mortality in the entire cohort was 11.43% (28/245). DC patients had a lower mean preoperative Glasgow Coma Scale (GCS) (p = 0.01); more patients with GCS of 6 (p=0.007);more unresponsive pupillary light reflex (p< 0.001); more closed basal cisterns (p< 0.001); and more patients with diffuse injury (p=0.025) than craniotomy patients. Given the greater severity, patients undergoing primary DC had higher 6-month mortality than the remainder of the cohort. However, in the surviving patients, the favorable GOSE rate was similar in two groups. We found that pupillary light reflex and basal cisterns were independent predictors for DC decision. Using ROC curve to predict the probability of DC, the sensitivity was 81.6% and the specificity was 84.9%.Conclusion: Our preliminary findings showed that the primary DC may benefit subgroups of sTBI with mass lesions, and unresponsive pre-op pupil reaction, and closed basal cistern to predict the DC decision were useful. These sensitive variables can be used as a referential guideline in our daily practice to decide to perform or avoid primary DC.

Prognostic significance of blood-brain barrier disruption in patients with severe nonpenetrating traumatic brain injury requiring decompressive craniectomy

2014 ◽  
Vol 121 (3) ◽  
pp. 674-679 ◽  
Author(s):  
Kwok M. Ho ◽  
Stephen Honeybul ◽  
Cheng B. Yip ◽  
Benjamin I. Silbert
Keyword(s):  
Traumatic Brain Injury ◽  
Brain Injury ◽  
Blood Brain Barrier ◽  
Decompressive Craniectomy ◽  
Prognostic Value ◽  
Prognostic Model ◽  
Long Term Outcomes ◽  
Prognostic Information ◽  
Severe Tbi

Object The authors assessed the risk factors and outcomes associated with blood-brain barrier (BBB) disruption in patients with severe, nonpenetrating, traumatic brain injury (TBI) requiring decompressive craniectomy. Methods At 2 major neurotrauma centers in Western Australia, a retrospective cohort study was conducted among 97 adult neurotrauma patients who required an external ventricular drain (EVD) and decompressive craniectomy during 2004–2012. Glasgow Outcome Scale scores were used to assess neurological outcomes. Logistic regression was used to identify factors associated with BBB disruption, defined by a ratio of total CSF protein concentrations to total plasma protein concentration > 0.007 in the earliest CSF specimen collected after TBI. Results Of the 252 patients who required decompressive craniectomy, 97 (39%) required an EVD to control intracranial pressure, and biochemical evidence of BBB disruption was observed in 43 (44%). Presence of disruption was associated with more severe TBI (median predicted risk for unfavorable outcome 75% vs 63%, respectively; p = 0.001) and with worse outcomes at 6, 12, and 18 months than was absence of BBB disruption (72% vs 37% unfavorable outcomes, respectively; p = 0.015). The only risk factor significantly associated with increased risk for BBB disruption was presence of nonevacuated intracerebral hematoma (> 1 cm diameter) (OR 3.03, 95% CI 1.23–7.50; p = 0.016). Although BBB disruption was associated with more severe TBI and worse long-term outcomes, when combined with the prognostic information contained in the Corticosteroid Randomization after Significant Head Injury (CRASH) prognostic model, it did not seem to add significant prognostic value (area under the receiver operating characteristic curve 0.855 vs 0.864, respectively; p = 0.453). Conclusions Biochemical evidence of BBB disruption after severe nonpenetrating TBI was common, especially among patients with large intracerebral hematomas. Disruption of the BBB was associated with more severe TBI and worse long-term outcomes, but when combined with the prognostic information contained in the CRASH prognostic model, this information did not add significant prognostic value.

scholarly journals Intracranial Densitometry-Augmented Machine Learning Enhances the Prognostic Value of Brain CT in Pediatric Patients With Traumatic Brain Injury: A Retrospective Pilot Study

2021 ◽  
Vol 9 ◽  
Author(s):  
Young-Tak Kim ◽  
Hakseung Kim ◽  
Choel-Hui Lee ◽  
Byung C. Yoon ◽  
Jung Bin Kim ◽  
...  
Keyword(s):  
Machine Learning ◽  
Traumatic Brain Injury ◽  
Brain Injury ◽  
Length Of Stay ◽  
Hospital Mortality ◽  
Outcome Measures ◽  
Prognostic Model ◽  
Pediatric Patients ◽  
Prognostic Models ◽  
Brain Ct

Background: The inter- and intrarater variability of conventional computed tomography (CT) classification systems for evaluating the extent of ischemic-edematous insult following traumatic brain injury (TBI) may hinder the robustness of TBI prognostic models.Objective: This study aimed to employ fully automated quantitative densitometric CT parameters and a cutting-edge machine learning algorithm to construct a robust prognostic model for pediatric TBI.Methods: Fifty-eight pediatric patients with TBI who underwent brain CT were retrospectively analyzed. Intracranial densitometric information was derived from the supratentorial region as a distribution representing the proportion of Hounsfield units. Furthermore, a machine learning-based prognostic model based on gradient boosting (i.e., CatBoost) was constructed with leave-one-out cross-validation. At discharge, the outcome was assessed dichotomously with the Glasgow Outcome Scale (favorability: 1–3 vs. 4–5). In-hospital mortality, length of stay (&gt;1 week), and need for surgery were further evaluated as alternative TBI outcome measures.Results: Densitometric parameters indicating reduced brain density due to subtle global ischemic changes were significantly different among the TBI outcome groups, except for need for surgery. The skewed intracranial densitometry of the unfavorable outcome became more distinguishable in the follow-up CT within 48 h. The prognostic model augmented by intracranial densitometric information achieved adequate AUCs for various outcome measures [favorability = 0.83 (95% CI: 0.72–0.94), in-hospital mortality = 0.91 (95% CI: 0.82–1.00), length of stay = 0.83 (95% CI: 0.72–0.94), and need for surgery = 0.71 (95% CI: 0.56–0.86)], and this model showed enhanced performance compared to the conventional CRASH-CT model.Conclusion: Densitometric parameters indicative of global ischemic changes during the acute phase of TBI are predictive of a worse outcome in pediatric patients. The robustness and predictive capacity of conventional TBI prognostic models might be significantly enhanced by incorporating densitometric parameters and machine learning techniques.

Sign in / Sign up

Export Citation Format

Share Document