Long-lasting coma

Abstract

In this report, we describe the case of a patient who has remained in a comatose state for more than one year after a traumatic and hypoxic brain injury. This state, which we refer to as long-lasting coma (LLC), may be a disorder of consciousness with significantly different features from those of conventional coma, the vegetative state, or brain death. On the basis of clinical, neurophysiological and neuroimaging data, we hypothesize that a multilevel involvement of the ascending reticular activating system is required in LLC. This description may be useful for the identification of other patients suffering from this severe disorder of consciousness, which raises important ethical issues.

Figure 1

Brain MRI performed 8 days after the acute event. Axial T2-weighted fluid-attenuated inversion recovery (FLAIR-T2w) images of the medulla oblongata (a), pons (b), midbrain (c), basal ganglia (d), corona radiata (e) and centrum semiovale (f). Axial spin echo T1-weighted images of the diencephalon (g) and vertex (h). Axial apparent diffusion coefficient (ADC) map from a diffusion-weighted imaging sequence acquired with b = 1000s/mm² (i). Predominant cerebral hypoperfusion signs: bilateral hyperintense edematous white matter tracts and gyri with compressed sulci in FLAIR-T2w image (arrow heads: b, c, e, f); hyperintense edematous globus pallidi in FLAIR-T2w image (asterisks: d) and deep white matter infarct within the watershed zone between anterior and middle cerebral artery territories showing restriction of diffusion in ADC maps (asterisks: i). The following signs of head trauma were evident: 1) a small hemorrhagic cerebral contusion involving the right superior temporal gyrus (long filled arrow in d) with evidence of T1 shortening caused by extracellular methemoglobin (late subacute hemorrhage; long filled arrow in g); 2) diffuse axonal injuries in the posterior arm of the left internal capsule (late subacute hemorrhagic lesions; short filled arrows in d and g) and at the vertex (h); 3) two small non-hemorrhagic shearing lesions in the right thalamus (empty arrow in d); 4) bilateral subdural frontotemporal hygromas that were difficult to recognize in FLAIR-T2w images but clear in the ADC map; 5) edema of splanchnocranium soft tissues with bleeding in the paranasal sinuses as an indirect sign of multiple fractures (a, b, c); and 6) extracranial subgaleal hematomas that were evident bilaterally (d, e, f, g, h, i).

Figure 2

EEG results. A. EEG revealing low amplitude (<20 μV) background activity at 3–4 Hz (delta and theta bands). B. Epileptiform discharges with spikes and spike-waves associated with bilateral shoulder myoclonus. Intermittent epileptiform discharges and associated myoclonus were responsive to intravenous administration of benzodiazepines, without changes in the level of consciousness, suggesting that the epileptiform activity was not affecting the patient’s consciousness. At the time of the EEG recording, the patient was medicated with 2000 mg of levetiracetam per day. Both A and B are from an EEG recorded approximately nine months after the brain injury. Spontaneous electrical brain activity was recorded from 20 electrodes (O₁, O₂, O_z, P₃, P₄, P₅, T₅, T₆, C₃, C₄, C_z, T₃, T₄, F₃, F₄, F_z, F₇, F₈, Fp₁, Fp₂) placed in accordance with the International 10–20 System (band-pass, 0.5–70 Hz; sampling rate, 200 Hz). G2 is an arbitrary name for the common ear-linked reference. The impedance of the recording electrodes was monitored during data acquisition and was always below 5 kΩ. The last three channels (in red) show the horizontal and vertical electro-oculogram and the electrocardiogram. These channels have been added to reveal any ocular or electrocardiographic artifacts. Both pages in the figure encompass a recording period of 15 seconds.

Figure 3

Brain glucose metabolism. Visual and statistical analysis of FDG PET data revealed widespread and severe cerebral hypometabolism, except in the anterior regions of both frontal lobes.