Data-driven prediction of decannulation probability and timing in patients with severe acquired brain injury

Implications of the Consciousness State on Decannulation in Patients With a Prolonged Disorder of Consciousness

OBJECTIVE

To prospectively investigate the evolution of the consciousness state and the cannula-weaning progression in patients with prolonged disorders of consciousness.

DESIGN

Nonconcurrent cohort study.

SETTING

A rehabilitation unit.

PARTICIPANTS

Adult patients (N=144) with prolonged disorders of consciousness after a severe acquired brain injury admitted between June 2020 and September 2022.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Consciousness state was assessed by repeated Coma Recovery Scale-Revised (CRS-R) questionnaire administration at admission and weekly afterward. The dates of the first improvement of consciousness state and the achievement of decannulation were recorded. Decannulation followed an internal protocol of multiprofessional rehabilitation.

RESULTS

One hundred forty-four patients were included: age, 69 years; 64 (44.4%) with hemorrhagic etiology; time post onset, 40 days, CRS-R score at admission, 9, median length of stay, 90 days. Seventy-three (50.7%) patients were decannulated. They showed a significantly higher CRS-R (P<.001) and states of consciousness (P<.001) at admission, at the first improvement of the consciousness state (P=.003), and at discharge (P<.001); a lower severity in the Cumulative Illness Rating Scale at admission (P=.01); and a lower rate of pulmonary infections with recurrence (P=.021), compared with nondecannulated patients. Almost all decannulated patients (97.3%) improved their consciousness before decannulation. Consciousness states at decannulation were as follows: unresponsive wakefulness syndrome, 0 (0%); minimally conscious state (MCS) minus, 4 (5.5%); MCS plus, 7 (9.6%); and emergence from MCS, 62 (84.9%). Kaplan-Meier analysis showed a significant divergence between the curves with a higher probability of decannulation in patients who improved consciousness (P<.001).

CONCLUSIONS

This study showed that the presence of signs of consciousness, even subtle, is a necessary condition for decannulation, suggesting that consciousness may influence some of the components implied in the decannulation process.

Predictive Factors for Successful Decannulation in Patients with Tracheostomies and Brain Injuries: A Systematic Review

Neurological patients frequently have disorders of consciousness, swallowing disorders, or neurological states that are incompatible with extubation. Therefore, they frequently require tracheostomies during their stay in an intensive care unit. After the acute phase, tracheostomy weaning and decannulation are generally expected to promote rehabilitation. However, few reliable predictive factors (PFs) for decannulation have been identified in this patient population. We sought to identify PFs that may be used during tracheostomy weaning and decannulation in patients with brain injuries. We conducted a systematic review of the literature regarding potential PFs for decannulation; searches were performed on 16 March 2021 and 1 June 2022. The following databases were searched: MEDLINE, EMBASE, CINAHL, Scopus, Web of Science, PEDro, OPENGREY, OPENSIGLE, Science Direct, CLINICAL TRIALS and CENTRAL. We searched for all article types, except systematic reviews, meta-analyses, abstracts, and position articles. Retrieved articles were published in English or French, with no date restriction. In total, 1433 articles were identified; 26 of these were eligible for inclusion in the review. PFs for successful decannulation in patients with acquired brain injuries (ABIs) included high neurological status, traumatic brain injuries rather than stroke or anoxic brain lesions, younger age, effective swallowing, an effective cough, and the absence of pulmonary infections. Secondary PFs included early tracheostomy, supratentorial lesions, the absence of critical illness polyneuropathy/myopathy, and the absence of tracheal lesions. To our knowledge, this is the first systematic review to identify PFs for decannulation in patients with ABIs. These PFs may be used by clinicians during tracheostomy weaning.

Rehabilitative intervention for successful decannulation in adult patients with acquired brain injury and tracheostomy: a systematic review

PURPOSE

Tracheostomy and dysphagia are independently associated with increased complications and poorer functional outcome after acquired brain injury (ABI). The aim of this study was to identify and evaluate rehabilitation to restore functional swallowing ability and respiratory capacity during tracheal tube weaning.

MATERIALS AND METHODS

The review was conducted according to PRISMA guidelines. Any study design with adult patients with ABI and tracheostomy was eligible. The primary outcome was decannulation.

RESULTS

A total of 2647 records were identified and eight papers included. Four studies investigated pharyngeal electrical stimulation (PES), two explored Facial Oral Tract Therapy (F.O.T.T.), one respiratory physiotherapy (RPT), and one study investigated external subglottic air flow (ESAF). Two RCTs found a significant difference between intervention and control on successful decannulation and readiness for decannulation in favour of PES. Time from rehabilitation admission and tracheostomy to decannulation was significantly reduced after implementing an F.O.T.T.-based protocol.

CONCLUSION

Four interventions were identified, PES, F.O.T.T., RPT, and ESAF, all aimed at increasing oropharyngeal sensory input through stimulation. Due to heterogeneity of interventions, designs and outcome measures, effect could not be estimated. This review highlights the limited research on rehabilitative interventions and thus the limited evidence to guide clinical rehabilitation.

Predictors of long-term decannulation in patients with disorders of consciousness

Purpose

The predictors of tracheostomy decannulation in patients with disorders of consciousness (DOC) are not comprehensively understood, making prognosis difficult. The primary objective of this study was to identify predictors of tracheostomy decannulation in patients with disorders of consciousness (DOC). The secondary aim was to evaluate the feasibility and safety of the modified Evans blue dye test (MEBDT) in tracheostomized DOC patients.

Methods

This retrospective study included all patients with disorders of consciousness (DOC) who underwent tracheostomy and were admitted between January 2016 and September 2022. Age, sex, etiology, initial Glasgow coma scale (GCS), initial Coma Recovery Scale-Revised (CRS-R), diagnosis of unresponsive wakefulness syndrome (UWS) or minimal consciousness state (MCS), MEBDT, initial modified Rankin scale (mRS), and initial Functional Oral Intake Scale (FOIS) were collected upon study enrollment. The relationship between clinical characteristics and cannulation status was investigated through a Cox regression model.

Results

A total of 141 patients were included in the study. The average age of these patients was 52.5 ± 16.7 years, with 42 (29.8%) being women. During the study period, 86 subjects (61%) underwent successful decannulation. Univariate analysis revealed that decannulated patients exhibited a significantly better conscious state compared to those without decannulation (CRS-R: p < 0.001; GCS: p = 0.023; MCS vs. UWS: p < 0.001). Additionally, a negative modified Evans blue dye test (MEBDT) result was significantly associated with tracheostomy decannulation (p < 0.001). In the multivariate analysis, successful decannulation was associated with a higher level of consciousness (MCS vs. UWS, p < 0.001, HR = 6.694) and a negative MEBDT result (negative vs. positive, p = 0.006, HR = 1.873). The Kaplan-Meier analysis further demonstrated that MEBDT-negative patients and those in the MCS category had a higher probability of decannulation at 12 months (p < 0.001).

Conclusion

The findings of this study indicate that a negative MEBDT result and a higher level of consciousness can serve as predictive factors for successful tracheostomy decannulation in DOC patients.

EEG-based methods for recovery prognosis of patients with disorders of consciousness: A systematic review

OBJECTIVE

Disorders of consciousness (DoC) are acquired conditions of severely altered consciousness. Electroencephalography (EEG)-derived biomarkers have been studied as clinical predictors of consciousness recovery. Therefore, this study aimed to systematically review the methods, features, and models used to derive prognostic EEG markers in patients with DoC in a rehabilitation setting.

METHODS

We conducted a systematic literature search of EEG-based strategies for consciousness recovery prognosis in five electronic databases.

RESULTS

The search resulted in 2964 papers. After screening, 15 studies were included in the review. Our analyses revealed that simpler experimental settings and similar filtering cut-off frequencies are preferred. The results of studies were categorised by extracting qualitative and quantitative features. The quantitative features were further classified into evoked/event-related potentials, spectral measures, entropy measures, and graph-theory measures. Despite the variety of methods, features from all categories, including qualitative ones, exhibited significant correlations with DoC prognosis. Moreover, no agreement was found on the optimal set of EEG-based features for the multivariate prognosis of patients with DoC, which limits the computational methods applied for outcome prediction and correlation analysis to classical ones. Nevertheless, alpha power, reactivity, and higher complexity metrics were often found to be predictive of consciousness recovery.

CONCLUSIONS

This study's findings confirm the essential role of qualitative EEG and suggest an important role for quantitative EEG. Their joint use could compensate for their reciprocal limitations.

SIGNIFICANCE

This study emphasises the need for further efforts toward guidelines on standardised EEG analysis pipeline, given the already proven role of EEG markers in the recovery prognosis of patients with DoC.

Cross-validation of predictive models for functional recovery after post-stroke rehabilitation

Background

Rehabilitation treatments and services are essential for the recovery of post-stroke patients’ functions; however, the increasing number of available therapies and the lack of consensus among outcome measures compromises the possibility to determine an appropriate level of evidence. Machine learning techniques for prognostic applications offer accurate and interpretable predictions, supporting the clinical decision for personalised treatment. The aim of this study is to develop and cross-validate predictive models for the functional prognosis of patients, highlighting the contributions of each predictor.

Methods

A dataset of 278 post-stroke patients was used for the prediction of the class transition, obtained from the modified Barthel Index. Four classification algorithms were cross-validated and compared. On the best performing model on the validation set, an analysis of predictors contribution was conducted.

Results

The Random Forest obtained the best overall results on the accuracy (76.2%), balanced accuracy (74.3%), sensitivity (0.80), and specificity (0.68). The combination of all the classification results on the test set, by weighted voting, reached 80.2% accuracy. The predictors analysis applied on the Support Vector Machine, showed that a good trunk control and communication level, and the absence of bedsores retain the major contribution in the prediction of a good functional outcome.

Conclusions

Despite a more comprehensive assessment of the patients is needed, this work paves the way for the implementation of solutions for clinical decision support in the rehabilitation of post-stroke patients. Indeed, offering good prognostic accuracies for class transition and patient-wise view of the predictors contributions, it might help in a personalised optimisation of the patients’ rehabilitation path.

Predicting outcome of patients with prolonged disorders of consciousness using machine learning models based on medical complexity

Patients with severe acquired brain injury and prolonged disorders of consciousness (pDoC) are characterized by high clinical complexity and high risk to develop medical complications. The present multi-center longitudinal study aimed at investigating the impact of medical complications on the prediction of clinical outcome by means of machine learning models. Patients with pDoC were consecutively enrolled at admission in 23 intensive neurorehabilitation units (IRU) and followed-up at 6 months from onset via the Glasgow Outcome Scale-Extended (GOSE). Demographic and clinical data at study entry and medical complications developed within 3 months from admission were collected. Machine learning models were developed, targeting neurological outcomes at 6 months from brain injury using data collected at admission. Then, after concatenating predictions of such models to the medical complications collected within 3 months, a cascade model was developed. One hundred seventy six patients with pDoC (M: 123, median age 60.2 years) were included in the analysis. At admission, the best performing solution (k-Nearest Neighbors regression, KNN) resulted in a median validation error of 0.59 points [IQR 0.14] and a classification accuracy of dichotomized GOS-E of 88.6%. Coherently, at 3 months, the best model resulted in a median validation error of 0.49 points [IQR 0.11] and a classification accuracy of 92.6%. Interpreting the admission KNN showed how the negative effect of older age is strengthened when patients' communication levels are high and ameliorated when no communication is present. The model trained at 3 months showed appropriate adaptation of the admission prediction according to the severity of the developed medical complexity in the first 3 months. In this work, we developed and cross-validated an interpretable decision support tool capable of distinguishing patients which will reach sufficient independence levels at 6 months (GOS-E > 4). Furthermore, we provide an updated prediction at 3 months, keeping in consideration the rehabilitative path and the risen medical complexity.

Cross-Validation of Machine Learning Models for the Functional Outcome Prediction after Post-Stroke Robot-Assisted Rehabilitation

The state of the art is still lacking an extensive analysis of which clinical characteristics are leading to better outcomes after robot-assisted rehabilitation on post-stroke patients. Prognostic machine learning-based models could promote the identification of predictive factors and be exploited as Clinical Decision Support Systems (CDSS). For this reason, the aim of this work was to set the first steps toward the development of a CDSS, by the development of machine learning models for the functional outcome prediction of post-stroke patients after upper-limb robotic rehabilitation. Four different regression algorithms were trained and cross-validated using a nested 5×10-fold cross-validation. The performances of each model on the test set were provided through the Median Average Error (MAE) and interquartile range. Additionally, interpretability analyses were performed, to evaluate the contribution of the features to the prediction. The results on the two best performing models showed a MAE of 13.6 [13.4] and 13.3 [14.8] on the Modified Barthel Index score (MBI). The interpretability analyses highlighted the Fugl-Meyer Assessment, MBI, and age as the most relevant features for the prediction of the outcome. This work showed promising results in terms of outcome prognosis after robot-assisted treatment. Further research should be planned for the development, validation and translation into clinical practice of CDSS in rehabilitation. Clinical relevance- This work establishes the premises for the development of data-driven tools able to support the clinical decision for the selection and optimisation of the robotic rehabilitation treatment.

Consciousness-Domain Index: a data-driven clustering-based consciousness labeling

Assessing consciousness results in one of the most complex neurological diagnosis. Even more complex and uncertain is prognosticating on consciousness recovery. Currently, consciousness is assessed by using a six-items scale, the Coma Recovery Scale-Revised. Namely, scores on the sub-items can individually assign or not a specific level of consciousness to a patient. In our work, by using solely the six sub-items of the CRS-R, we implemented a clustering algorithm labeling patients with the Consciousness-Domains Index (CDI) starting from a dataset of 190 patients with a Disorder of Consciousness (DoC). Then, the CDI is compared with the clinical state at admission and at six months via univariate analysis. The number of clusters best dividing the groups resulted equal to two and the most influencing sub-items resulted the visual and motor one. The CDI closely resembles the clinical state at admission (CSA) (Cohen's k=0.85). On the other hand, when comparing CDI and CSA, a net improvement was found in the prognostic power of the neurological outcome at six months, targeted as presence/absence of a DoC ( ). Data-driven techniques pave the way for automated and model-based search of prognostic factors, together with the use of such prognostic factors in multivariate prognostic models. Future works will address the external validation of the CDI, together with the inclusion of the CDI in a multivariate supervised model, in order to assess the true potential of such novel index. Clinical Relevance- A completely data-driven index was derived from a clustering of CRS-R sub-items. It correlates with the neurological outcome at six months better than the state of consciousness at admission.

Merging Clinical and EEG Biomarkers in an Elastic-Net Regression for Disorder of Consciousness Prognosis Prediction

Patients with Disorder of Consciousness (DoC) entering Intensive Rehabilitation Units after a severe Acquired Brain Injury have a highly variable evolution of the state of consciousness which is a complex aspect to predict. Besides clinical factors, electroencephalography has clearly shown its potential into the identification of prognostic biomarkers of consciousness recovery. In this retrospective study, with a dataset of 271 patients with DoC, we proposed three different Elastic-Net regressors trained on different datasets to predict the Coma Recovery Scale-Revised value at discharge based on data collected at admission. One dataset was completely EEG-based, one solely clinical data-based and the last was composed by the union of the two. Each model was optimized, validated and tested with a robust nested cross-validation pipeline. The best models resulted in a median absolute test error of 4.54 [IQR = 4.56], 3.39 [IQR = 4.36], 3.16 [IQR = 4.13] for respectively the EEG, clinical and hybrid model. Furthermore, the hybrid model for what concerns overcoming an unresponsive wakefulness state and exiting a DoC results in an AUC of 0.91 and 0.88 respectively. Small but useful improvements are added by the EEG dataset to the clinical model for what concerns overcoming an unresponsive wakefulness state. Data-driven techniques and namely, machine learning models are hereby shown to be capable of supporting the complex decision-making process the practitioners must face.

Predicting SARS-CoV-2 infection duration at hospital admission:a deep learning solution

COVID-19 cases are increasing around the globe with almost 5 million of deaths. We propose here a deep learning model capable of predicting the duration of the infection by means of information available at hospital admission. A total of 222 patients were enrolled in our observational study. Anagraphical and anamnestic data, COVID-19 signs and symptoms, COVID-19 therapy, hematochemical test results, and prior therapies administered to patients are used as predictors. A set of 55 features, all of which can be taken in the first hours of the patient's hospitalization, was considered. Different solutions were compared achieving the best performance with a sequential convolutional neural network-based model merged in an ensemble with two different meta-learners linked in cascade. We obtained a median absolute error of 2.7 days (IQR = 3.0) in predicting the duration of the infection; the error was equally distributed in the infection duration range. This tool could preemptively give an outlook of the COVID-19 patients' expected path and the associated hospitalization effort. The proposed solution could be viable in tackling the huge burden and the logistics complexity of hospitals or rehabilitation centers during the pandemic waves. With data taken ad admission, entering a PCA-based feature selection, a k-fold cross-validated CNN-based model was implemented. After external texting, a median absolute error of 2.7 days [IQR = 3 days].