Outcome prediction in disorders of consciousness: the role of coma recovery scale revised.

BACKGROUND: To evaluate the utility of the revised coma remission scale (CRS-r), together with other clinical variables, in predicting emergence from disorders of consciousness (DoC) during intensive rehabilitation care. METHODS: Data were retrospectively extracted from the medical records of patients enrolled in a specialized intensive rehabilitation unit. 123 patients in a vegetative state (VS) and 57 in a minimally conscious state (MCS) were included and followed for a period of 8 weeks. Demographical and clinical factors were used as outcome measures. Univariate and multivariate Cox regression models were employed for examining potential predictors for clinical outcome along the time. RESULTS: VS and MCS groups were matched for demographical and clinical variables (i.e., age, aetiology, tracheostomy and route of feeding). Within 2 months after admission in intensive neurorehabilitation unit, 3.9% were dead, 35.5% had a full recovery of consciousness and 66.7% remained in VS or MCS. Multivariate analysis demonstrated that the best predictor of functional improvement was the CRS-r scores. In particular, patients with values greater than 12 at admission were those with a favourable likelihood of emergence from DoC. CONCLUSIONS: Our study highlights the role of the CRS-r scores for predicting a short-term favorable outcome.

Predicting Outcome of Traumatic Brain Injury: Is Machine Learning the Best Way?

One of the main challenges in traumatic brain injury (TBI) patients is to achieve an early and definite prognosis. Despite the recent development of algorithms based on artificial intelligence for the identification of these prognostic factors relevant for clinical practice, the literature lacks a rigorous comparison among classical regression and machine learning (ML) models. This study aims at providing this comparison on a sample of TBI patients evaluated at baseline (T0), after 3 months from the event (T1), and at discharge (T2). A Classical Linear Regression Model (LM) was compared with independent performances of Support Vector Machine (SVM), k-Nearest Neighbors (k-NN), Naïve Bayes (NB) and Decision Tree (DT) algorithms, together with an ensemble ML approach. The accuracy was similar among LM and ML algorithms on the analyzed sample when two classes of outcome (Positive vs. Negative) approach was used, whereas the NB algorithm showed the worst performance. This study highlights the utility of comparing traditional regression modeling to ML, particularly when using a small number of reliable predictor variables after TBI. The dataset of clinical data used to train ML algorithms will be publicly available to other researchers for future comparisons.

Dysautonomia and Clinical Outcome in Vegetative State.

A dramatic disorder tentatively attributed to diencephalic-hypothalamic damage or dysfunction, dysautonomia, affects recovery from brain injury. Its incidence, correlation with etiology, and relevance as a predictor of outcome were retrospectively surveyed in 333 patients in vegetative state (VS) for more than 2 weeks at admission. Outcome was assessed according to the Glasgow Outcome Scale. Data were treated statistically by multi-variate analyses. Dysautonomia occurred in 26.1% of patients, with greater incidence among post-traumatic (31.9%) than non-traumatic (15.8%) patients. Outcome was worse among non-traumatic than post-traumatic patients irrespective of dysautonomia, and worst among non-traumatic patients with dysautonomia. Dysautonomia proved common among patients in VS (with incidence depending on etiology and age) and influenced the patients' outcome through mechanisms still to be defined, but conceivably mediated by diencephalic-hypothalamic unbalance.

External Validation and Calibration of the DecaPreT Prediction Model for Decannulation in Patients with Acquired Brain Injury.

We propose a new set of clinical variables for a more accurate early prediction of safe decannulation in patients with severe acquired brain injury (ABI), during a post-acute rehabilitation course. Starting from the already validated DecaPreT scale, we tested the accuracy of new logistic regression models where the coefficients of the original predictors were reestimated. Patients with tracheostomy were retrospectively selected from the database of the neurorehabilitation unit at the S. Anna Institute of Crotone, Italy. New potential predictors of decannulation were screened from variables collected on admission during clinical examination, including (a) age at injury, (b) coma recovery scale-revised (CRS-r) scores, and c) length of ICU period. Of 273 patients with ABI (mean age 53.01 years; 34% female; median DecaPreT = 0.61), 61.5% were safely decannulated before discharge. In the validation phase, the linear logistic prediction model, created with the new multivariable predictors, obtained an area under the receiver operating characteristics curve of 0.901. Our model improves the reliability of simple clinical variables detected at the admission of the post-acute phase in predicting decannulation of ABI patients, thus helping clinicians to plan better rehabilitation.

The Impact of Medical Complications in Predicting the Rehabilitation Outcome of Patients With Disorders of Consciousness After Severe Traumatic Brain Injury.

In this study, we sought to assess the predictors of outcome in patients with disorders of consciousness (DOC) after severe traumatic brain injury (TBI) during neurorehabilitation stay. In total, 96 patients with DOC (vegetative state, minimally conscious state, or emergence from minimally conscious state) were enrolled (69 males; mean age 43.6 ± 20.8 years) and the improvement of the degree of disability, as assessed by the Disability Rating Scale, was considered the main outcome measure. To define the best predictor, a series of demographical and clinical factors were modeled using a twofold approach: (1) logistic regression to evaluate a possible causal effect among variables; and (2) machine learning algorithms (ML), to define the best predictive model. Univariate analysis demonstrated that disability in DOC patients statistically decreased at the discharge with respect to admission. Genitourinary was the most frequent medical complication (MC) emerging during the neurorehabilitation period. The logistic model revealed that the total amount of MCs is a risk factor for lack of functional improvement. ML discloses that the most important prognostic factors are the respiratory and hepatic complications together with the presence of the upper gastrointestinal comorbidities. Our study provides new evidence on the most adverse short-term factors predicting a functional recovery in DOC patients after severe TBI. The occurrence of medical complications during neurorehabilitation stay should be considered to avoid poor outcomes.

Paroxysmal Sympathetic Hyperactivity Rate in Vegetative or Minimally Conscious State after Acquired Brain Injury Evaluated by Paroxysmal Sympathetic Hyperactivity Assessment Measure.

The rate of paroxysmal sympathetic hyperactivity (PSH) was retrospectively assessed using the Paroxysmal Sympathetic Hyperactivity-Assessment Measure (PSH-AM) scale in patients with disorders of consciousness attributed to traumatic and non-traumatic acquired brain injury during the rehabilitation phase. These results were compared with previous studies carried out in the same clinical scenario, in order to verify the prevalence of PSH signs from 1998 to 2014. The entire sample consisted of 140 patients in vegetative state/unresponsive wakefulness syndrome or minimally conscious state admitted to a neurorehabilitation subacute unit from June 2010 to December 2014. PSH-AM revealed the presence of PSH in 16% of traumatic and 12% of non-traumatic younger patients. In the non-traumatic group, the rate was higher in patients with anoxia-hypoxia (37.5%) etiology than those with vascular brain injury (6.7%). A comparison with previous studies revealed a reduction in the number of PSH cases in traumatic patients. This study provides evidence that PSH-AM can be used prospectively to detect the rate of PSH and stratify severity of signs. Further longitudinal analysis is warranted to confirm the prevalence of PSH signs in non-traumatic brain injured patients.

The Arg allele in position 192 of PON1 is associated with carotid atherosclerosis in subjects with elevated HDLs.

Human serum paraoxonase (PON1) is an HDL-associated enzyme involved in the protection of lipoproteins from oxidation. A polymorphism at position 192 (Gln/Arg) influences its activity in a substrate-dependent manner. The aim of the present study was to evaluate, in vivo, the contribution of the PON1-192 polymorphism to the protective effect of HDLs. Three hundred and forty seven subjects in the upper and lower decile of sex-specific HDL cholesterol distribution were selected from participants in a cardiovascular disease prevention study. PON1 genotypes were determined by PCR amplification and restriction analysis. Blood pressure, height, weight, smoking and alcohol habits, as well as the presence of familial or personal history of CHD were recorded. Plasma lipids and blood glucose were measured by routine enzymatic methods. As a measure of antiatherogenic HDL effect, carotid atherosclerosis was assessed by ultrasonography. Allele and genotype frequencies did not differ significantly between low- and high-HDL groups. Similarly, no significant difference was observed among genotypes in all variables studied. Subjects with Gln/Arg or Arg/Arg had more carotid abnormalities than Gln/Gln with an adjusted odds ratio (OR) of 3.27 (95% CI, 1.61-6.64, P=0.001) for abnormal carotid score. Stepwise logistic regression analysis showed that in the whole population age and presence of low-HDL were the only independent predictors for abnormal carotid score. In low-HDL, age was the only independent predictor entered into the model (OR 1.09/year, P<0.0001); in high-HDL, age entered first (OR, 1.07/year, P=0.001), followed by the presence of Gln/Arg or Arg/Arg (OR, 2.94, 95% CI, 1.47-5.91, P=0.002). In conclusion, in subjects with low levels of HDL cholesterol, carotid atherosclerosis is not related to PON1-192 polymorphism. On the other hand, in subjects with elevated concentration of HDL cholesterol, the presence of carotid atherosclerosis is significantly associated with the arginine variant in position 192 of the PON1 gene.

Decreasing incidence of paroxysmal sympathetic hyperactivity syndrome in the vegetative state.

OBJECTIVE: To update knowledge of the incidence of paroxysmal sympathetic hyperactivity (PSH, also referred to as dysautonomia), an emergency condition tentatively attributed to sympathetic paroxysms or diencephalic-hypothalamic disarrangement associated with severe diffuse brain axonal damage or hypoxia. This condition is reportedly common in the vegetative state, threatens survival and affects outcome. METHODS: The results of a retrospective study on 333 subjects in a vegetative state admitted to a dedicated unit in 1998-2005 are compared with a survey on patients admitted to the same unit in 2006-2010. RESULTS AND COMMENT: In the 1998-2005 period, the incidence of PSH was 32% and 16% in post-traumatic and non-traumatic patients, respectively. It decreased to 18% and 7% in the 2006-2010 period. The PSH duration and the time spent in emergency units before admission and in the dedicated unit for the vegetative state after admission also decreased significantly. Incidence was greater among post-traumatic- patients; its effect on outcome does not appear to have changed.

Neurorehabilitation for severe disorder of consciousness: the S. Anna - RAN operational model.

The operational model and strategies designed for use in the S. Anna - Research in Advanced Neurorehabilitation Institute for the care and neurorehabilitation of subjects in the vegetative or minimally conscious states are described here. A total of 722 patients were admitted, cared for and discharged from the institute in the period 1998-2009. Application of the model approach has progressively shortened the time of hospitalization and rehabilitation and reduced costs.