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

Lucca, Lucia Francesca; Lofaro, Danilo; Leto, Elio; Ursino, Maria; Rogano, Stefania; Pileggi, Antonio; Vulcano, Serafino; Conforti, Domenico; Tonin, Paolo; Cerasa, Antonio

doi:10.3389/fnhum.2020.570544

Cited by 13 publications

(12 citation statements)

References 41 publications

(65 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of the importance of a dynamically-evolving prognostication 33 , we developed a cascade ML model that, beside training and validating a model at admission, we added data collected at 3 months onto the prediction of the admission model. Differently from previous studies in literature 16 , 32 , in the present work data at admission and at 3 months were used in two separate models, connected by the prediction of the first model. In this way, it is possible to investigate the effect of MCs (risen in the first 3 months of stay in rehabilitative units) on the prediction done at admission.…”

Section: Discussionmentioning

confidence: 99%

“…We observed that younger age, higher CRS-R total and sub-scores, absence of an enteral feeding device and lower DRS total score were significant predictors of a favorable outcome. Several studies demonstrated that age is one of the main predictors of the functional improvement in patients with pDoC 15 , 32 , 54 , probably in relation with more premorbid medical illnesses 55 and lower age-related brain plasticity 56 in the elderly. In addition, we found a significant interaction between age and the level of consciousness at admission.…”

Section: Discussionmentioning

confidence: 99%

“…To our knowledge, previous solutions adopted data recorded at early stage after admission, disregarding the occurrence of MCs within the rehabilitative path. Given the high relevance of MCs 15 , 18 , 32 , we are convinced that intaking data collected during the hospital stay would foster the improvement of prognostic predictions. Up to our knowledge, the information on complications has always been merged together with clinical and instrumental variables, without evaluating whether complications could improve the prediction made at admission and/or the mutual influence between arisen complications and admission prognosis.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

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

Liuzzi

Magliacano

Bellis

et al. 2022

Sci Rep

View full text Add to dashboard Cite

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.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

Liuzzi

Magliacano

Bellis

et al. 2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…It was difficult to demonstrate their causal impact on the level of consciousness and recovery. However, general presumptions, which are usually negative, may be made based on the existing evidence [ 3 , 5 , 6 ]. A standardized tool, the Comorbidities Coma Scale (CoCoS), may be helpful in practice and future studies [ 11 ].…”

Section: Discussionmentioning

confidence: 99%

“…Currently, there does not exist any well-established epidemiological data for medical and neurological complications and comorbidities in the DoC population. Several articles have attempted to investigate those conditions [ 3 , 4 , 5 ] but focused on the correlation with outcome prediction [ 4 , 5 , 6 ]. There remains concern that the information pertaining to clinical evaluation and management of these complications and comorbidities may be underrepresented.…”

Section: Introductionmentioning

confidence: 99%

Many Faces of the Hidden Souls: Medical and Neurological Complications and Comorbidities in Disorders of Consciousness

et al. 2021

View full text Add to dashboard Cite

Early and goal-directed management of complications and comorbidities is imperative to facilitate neurorecovery and to optimize outcomes of disorders of consciousness (DoC). This is the first large retrospective cohort study on the primary medical and neurological complications and comorbidities in persons with DoC. A total of 146 patients admitted to a specialized inpatient DoC rehabilitation program from 1 January 2014 to 31 October 2018 were included. The incidences of those conditions since their initial brain injuries were reviewed per documentation. They were categorized into reversible causes of DoC, confounders and mimics, and other medical/neurological conditions. The common complications and comorbidities included pneumonia (73.3%), pain (75.3%), pressure ulcers (70.5%), oral and limb apraxia (67.1%), urinary tract infection (69.2%), and 4-limb spasticity (52.7%). Reversible causes of DoC occurred very commonly. Conditions that may confound the diagnosis of DoC occurred at surprisingly high rates. Conditions that may be a source of pain occurred not infrequently. Among those that may diminish or confound the level of consciousness, 4.8 ± 2.0 conditions were identified per patient. In conclusion, high rates of various complications and comorbidities occurred in persons with DoC. Correcting reversible causes, identifying confounders and mimics, and managing general consequences need to be seriously considered in clinical practice.

show abstract

Artificial Intelligence Applications for Traumatic Brain Injury Research and Clinical Management

Fabrizio

Termine

2023

Contemporary Clinical Neuroscience

View full text Add to dashboard Cite

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

Cited by 13 publications

References 41 publications

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

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

Many Faces of the Hidden Souls: Medical and Neurological Complications and Comorbidities in Disorders of Consciousness

Artificial Intelligence Applications for Traumatic Brain Injury Research and Clinical Management

Contact Info

Product

Resources

About