Predictors of Improvement in Disorders of Consciousness

Pauline Anderson

July 20, 2020

Younger age, shorter time since injury, higher level of responsiveness and consciousness, and EEG reactivity to eye opening are all factors that can predict 6-month improvement among patients in a vegetative or minimally conscious state from a severe brain injury, new research suggests.

In addition, information on such factors can be gathered at the bedside relatively easily and without sophisticated technical tools — and doing so is not time-consuming or expensive, lead author Anna Estraneo, MD, a neurologist at Nola General Hospital, Naples, Italy, told Medscape Medical News.

Dr Anna Estraneo

Determining factors that predict better outcome in patients with a "prolonged" disorder of consciousness (DoC), defined as lasting more than 28 days, is important for improving function and quality of life, she said.

"This study shows that by applying standardized multimodal assessment protocols, clinicians can identify people with DoC most suitable for tailored rehabilitation programs and therapeutic interventions," added Estraneo, who is also a scientific consultant at the Neurorehabilitation Unit, Don Carlo Gnocchi Foundation, Florence, Italy.

The findings were published online July 13 in Neurology.

Three Etiologies

Although previous studies have looked at the predictive value of individual factors, such as scores on the Coma Recovery Scale-Revised (CRS-R), this is the first to closely examine a large number of predictors in patients from multiple centers.

The study enrolled 147 patients (70% men; mean age, 49.4 years; mean time post injury, 59.6 days) from 12 centers in North America, Europe, and Asia.

Of these participants, 71 were diagnosed with vegetative state/unresponsive wakefulness syndrome (VS/UWS), and 76 with a minimally conscious state (MCS).

Those with MCS had a significantly higher mean total CRS-R score (12.4 vs 4.6) compared with patients with VS/UWS, as well as a higher nociception coma scale-revised (NCS-R) score (3.7 vs 2.1) and lower mean score on the disability rating scale (DRS) (22.3 vs 25.4) (all comparisons, P < .001).

There were no significant differences in age, gender, etiology, or time post-injury between the two groups.

All patients were classified into three etiologies: 55 had traumatic brain injury related to a severe concussion, 56 had a vascular injury related to an ischemic or hemorrhagic stroke, and 36 had an anoxic injury due mostly to carbon monoxide poisoning or cardiorespiratory arrest.

The traumatic group was significantly younger (mean age, 33.8 years) than the anoxic (mean age, 56) and vascular groups (mean age, 60).

Blinded neurophysiologists recorded and analyzed multimodel neurophysiological data including standard EEG, somatosensory-evoked potential, and event-related potentials within 15 days of study entry.

Researchers classified EEG background activity into five severity categories: normal, mildly abnormal, moderately abnormal, diffuse slowing, and low voltage. The MCS and VS/UWS groups differed significantly in baseline predominant EEG background activity.

EEG Reactivity

To analyze EEG reactivity to external stimuli, investigators randomly administered five kinds of stimulation during EEG recordings:

  • eye opening and eye closing

  • tactile, which included wiping cotton wool on forearm

  • noxious, which included pressing a pencil on fingernail beds on each hand

  • acoustic, which included hand clapping

  • intermittent light, which used a strobe lamp

Complete information on the study's primary outcome, which was diagnosis at 6 months post-injury, was available for 143 patients. Among these patients, about half (49.6%) were classified as not improved.

Among those who did improve, 27.8% progressed from VS/UWS to MCS; and 9.7% who were in a VS/UWS state and 62.5% in MCS emerged from MCS.

Univariate analysis showed that patients who improved were younger than those who did not improve (44.4 vs 55.4 years) and had a shorter time post-injury (51.8 vs 67.0 days). Better 6-month prognosis was also associated with being male, having an MCS diagnosis, and having a traumatic etiology.

Improved patients also had a higher CRS-R total score (10.1 vs 6.8) and lower DRS total score (23.1 vs 24.6) than those who did not improve. However, these groups did not differ in NCS-R total score.

Several patterns of EEG background activity significantly differed between patients who improved and those who did not. EEG reactivity to eye opening and closing (corrected P = .001) and to acoustic stimuli (corrected P = .004) was significantly more frequent in patients who improved, whereas reactivity to tactile, nociceptive, and intermittent photo stimulation was not.

A regression analysis that included data on 72 patients showed that younger age, shorter time post-injury, higher CRS-R total score, and presence of EEG reactivity to eye opening were all significantly associated with a better outcome.

On the other hand, gender, etiology, clinical diagnosis, DRS total score, EEG background activity, and reactivity to acoustic stimuli were not linked to a better outcome.

Estraneo noted that although patients in MCS usually show better clinical outcome than patients in VS, this is not always the case.

Identifying reliable prognostic markers would allow clinicians to identify patients with higher likelihood of clinical recovery, she said. "This in turn may help to make appropriate decisions concerning treatment and care," she added.

Hopefully, this study "can stimulate further research and drive international consensus on prognostic procedures for the clinical care of individuals with prolonged DoC," said Estraneo.

"Advances The Field"

Commenting on the findings for Medscape Medical News, Christine Blume, PhD, Centre for Chronobiology, University of Basel, Switzerland, who has done previous research on cognition and consciousness, said the study was "interesting, important, and certainly advances the field."

"It's a high-quality study, has very experienced researchers, and the protocol is well designed," said Blume, who was not involved with the current study.

However, she's not convinced these new results will make much of a difference to clinicians in their practice.

She pointed out that certain factors that were predictive in the univariate analysis were no longer significant in the regression analysis that included all predictors together in one model.

"Etiology, diagnosis, and EEG background activity are the three predictors that I'm surprised were not significant predictors any more in the regression model," Blume said.

A "constraint" of the statistical analysis was the small number of included patients relative to the number of predictors assessed, she added. "The more predictors you have, the more patients you need."

However, she also recognized how challenging it is to acquire high-quality data on patients with DoC, including the need for repeated assessments of responsiveness and consciousness.

The study was supported by grants from the European Union's Horizon 2020 research and innovation program, the National Institute of Neurological Disorders and Stroke, the NIH Director's Office, the James S. McDonnell Foundation, the Tiny Blue Dot Foundation, the European Union's Horizon 2020 Framework Programme for Research and Innovation, the Luminous project, the Belgian National Funds for Scientific Research, and the Italian Ministry of Health. The investigators and Blume have disclosed no relevant financial relationships.

Neurology. Published online July 13, 2020. Abstract

For more Medscape Neurology news, join us on Facebook and Twitter