Classification of Level of Consciousness in a Neurological ICU Using Physiological Data.

BACKGROUND: Impaired consciousness is common in intensive care unit (ICU) patients, and an individual's degree of consciousness is crucial to determining their care and prognosis. However, there are no methods that continuously monitor consciousness and alert clinicians to changes. We investigated the use of physiological signals collected in the ICU to classify levels of consciousness in critically ill patients. METHODS: We studied 61 patients with subarachnoid hemorrhage (SAH) and 178 patients with intracerebral hemorrhage (ICH) from the neurological ICU at Columbia University Medical Center in a retrospective observational study of prospectively collected data. The level of consciousness was determined on the basis of neurological examination and mapped to comatose, vegetative state or unresponsive wakefulness syndrome (VS/UWS), minimally conscious minus state (MCS-), and command following. For each physiological signal, we extracted time-series features and performed classification using extreme gradient boosting on multiple clinically relevant tasks across subsets of physiological signals. We applied this approach independently on both SAH and ICH patient groups for three sets of variables: (1) a minimal set common to most hospital patients (e.g., heart rate), (2) variables available in most ICUs (e.g., body temperature), and (3) an extended set recorded mainly in neurological ICUs (absent for the ICH patient group; e.g., brain temperature). RESULTS: On the commonly performed classification task of VS/UWS versus MCS-, we achieved an area under the receiver operating characteristic curve (AUROC) in the SAH patient group of 0.72 (sensitivity 82%, specificity 57%; 95% confidence interval [CI] 0.63-0.81) using the extended set, 0.69 (sensitivity 83%, specificity 51%; 95% CI 0.59-0.78) on the variable set available in most ICUs, and 0.69 (sensitivity 56%, specificity 78%; 95% CI 0.60-0.78) on the minimal set. In the ICH patient group, AUROC was 0.64 (sensitivity 56%, specificity 65%; 95% CI 0.55-0.74) using the minimal set and 0.61 (sensitivity 50%, specificity 80%; 95% CI 0.51-0.71) using the variables available in most ICUs. CONCLUSIONS: We find that physiological signals can be used to classify states of consciousness for patients in the ICU. Building on this with intraday assessments and increasing sensitivity and specificity may enable alarm systems that alert physicians to changes in consciousness and frequent monitoring of consciousness throughout the day, both of which may improve patient care and outcomes.

Cost-Effective Machine Learning Based Clinical Pre-Test Probability Strategy for DVT Diagnosis in Neurological Intensive Care Unit.

In order to overcome the shortage of the current costly DVT diagnosis and reduce the waste of valuable healthcare resources, we proposed a new diagnostic approach based on machine learning pre-test prediction models using EHRs. We examined the sociodemographic and clinical factors in the prediction of DVT with 518 NICU admitted patients, including 189 patients who eventually developed DVT. We used cross-validation on the training data to determine the optimal parameters, and finally, the applied ROC analysis is adopted to evaluate the predictive strength of each model. Two models (GLM and SVM) with the strongest ROC were selected for DVT prediction, based on which, we optimized the current intervention and diagnostic process of DVT and examined the performance of the proposed approach through simulations. The use of machine learning based pre-test prediction models can simplify and improve the intervention and diagnostic process of patients in NICU with suspected DVT, and reduce the valuable healthcare resource occupation/usage and medical costs.

Thrombocytopenia and platelet course on hospital mortality in neurological intensive care unit: a retrospective observational study from large database.

BACKGROUND: Thrombocytopenia (TP) has been shown to be an independent predictor of mortality in the intensive care unit (ICU) patients. Studies are lacking in the neurological ICU (NICU) population. The aim was to evaluate the incidence of TP in NICU and the relationship between TP and outcomes. METHODS: We conducted a retrospective multicenter study of prospectively collected data of all patients admitted to the NICU between 2014 and 2015 from a large database (eICU Collaborative Research Database). The main exposure was TP at admission and TP developed during NICU stay. Multivariable logistic regression and Cox proportional hazard models were used to evaluate the relationship of TP at admission and platelet course with hospital mortality. The primary outcome was hospital mortality. RESULTS: 7450 patients in NICU from 17 hospitals were included. Hospital mortality was 9%. TP at admission was present in 20% of patients, TP developed during NICU stay was present in 13.2% of patients. TP at admission was not associated with hospital mortality after adjusting for confounders (OR 1.14 [95% CI 0.92-1.41, p = 0.237]). Hospital mortality of continuous TP during NICU stay was 15% while hospital mortality of recovery from TP at admission was 6% (p < 0.001). Patients with TP developed during NICU stay had higher odds ratio for hospital mortality (OR 1.65 [95% CI 1.3-2.09, p < 0.001]). CONCLUSIONS: Thrombocytopenia is common in NICU and patients who have thrombocytopenia not resolving have increased mortality. Patients' recovery from thrombocytopenia may predict a good prognosis.

Demographic and clinical predictors of multiple intracranial aneurysms in patients with subarachnoid hemorrhage.

OBJECTIVE Pathophysiological differences that underlie the development and subsequent growth of multiple aneurysms may exist. In this study, the authors assessed the factors associated with the occurrence of multiple aneurysms in patients presenting with aneurysmal subarachnoid hemorrhage (SAH). METHODS Consecutive patients presenting with aneurysmal SAH between 1996 and 2012 were prospectively enrolled in the Subarachnoid Hemorrhage Outcome Project. Patients harboring 1, 2, or 3 or more aneurysms were stratified into groups, and the clinical and radiological characteristics of each group were compared using multivariate logistic regression. RESULTS Of 1277 patients with ruptured intracranial aneurysms, 890 had 1 aneurysm, 267 had 2 aneurysms, and 120 had 3 or more aneurysms. On multinomial regression using the single-aneurysm cohort as base case, risk factors for patients presenting with 2 aneurysms were female sex (relative risk ratio [RRR] 1.80, p < 0.001), higher body mass index (BMI) (RRR 1.02, p = 0.003), more years of smoking (RRR = 1.01, p = 0.004), and black race (RRR 1.83, p = 0.001). The risk factors for patients presenting with 3 or more aneurysms were female sex (RRR 3.10, p < 0.001), higher BMI (RRR 1.03, p < 0.001), aneurysm in the posterior circulation (RRR 2.59, p < 0.001), and black race (RRR 2.15, p = 0.001). Female sex, longer smoking history, aneurysms in the posterior circulation, BMI, and black race were independently associated with the development of multiple aneurysms in our adjusted multivariate multinomial model. CONCLUSIONS Significant demographic and clinical differences are found between patients presenting with single and multiple aneurysms in the setting of aneurysmal SAH. These predictors of multiple aneurysms likely reflect a predisposition toward inflammation and endothelial injury.

Association of Extubation Failure and Functional Outcomes in Patients with Acute Neurologic Illness.

BACKGROUND: An association between extubation failure and neurologic and functional outcomes in patients with primary neurologic illness has not been investigated rigorously. We plan to conduct a retrospective chart review to study this association. METHODS: A total of 949 unique patients intubated and ventilated for at least 48 h in Neuro ICU (NICU) were obtained. Extubation failure was defined as need for reintubation within 48 h of initial extubation. Independent and dependent association between extubation failure and clinical parameters was assessed. RESULTS: The patients had a median age [interquartile range (IQR)] of 58.5 (23.0) years. 60.5% were male and 81.9% were Caucasian. Extubation failure occurred in 108 (12.8%) patients. There was no difference in age, APACHE 3 score, FOUR score, or GCS score of patients at ICU admission between those who experienced extubation failure and those who did not. Extubation failure was associated with longer NICU and hospital LOS [median (IQR); 13.7 (11.3) vs. 9.1(8.2) days, P < 0.01 and 24.5 (20.0) vs. 16.8 (16.7) days, P < 0.01]. Patients with extubation failure had worse functional outcomes at 6 months as measured by the modified Rankin score [MRS; median (IQR), 5.0 (2.0) vs. 4.0 (3.0), P < 0.01]. After adjusting for confounders, extubation failure was associated with longer hospital and ICU LOS and worse functional outcomes. CONCLUSIONS: In patients with acute neurological illness, extubation failure is associated with longer ICU and hospital stays but does not impact hospital mortality. Patients with extubation failure may experience a worsening of their functional status over time.

Clinical value of CereTom mobile CT scanner for patients in neurological ICU / 医疗卫生装备

Objective To explore the clinical value of cranial CT for the patients in neurological ICU by analyzing the application of mobile CT scanner CereTom in some hospital.Methods Retrospective analysis was carried out for the patients being hospitalized and undergoing cranial CT examination in some hospital from March 2012 to August 2014.Results Totally 261 patients and 325 times of examination were involved in, and two ones failed in the examination, with the success rate of 99.23%. There were 218 patients (83.52%) had the examination completed in one time and 43 ones (16.48%) in several times. It's proved that bedside CT could be applied clinically with high successful rate. The mean time of bedside CT examination was (18.3±3.8)min, significantly less that then of common examination.Conclusion Mobile CT may decrease moving-related risk of the patient and the time, manpower consumed for examination, and thus is worth popularizing clinically.