Recovery of multiply injured ascending reticular activating systems in a stroke patient

Consciousness is controlled by activation of the ascending reticular activating system (ARAS). The ARAS consists mainly of the lower and upper parts between the thalamus and cerebral cortex (Edlow et al., 2012; Yeo et al., 2013; Jang et al., 2014). Because the ARAS is composed of several neuronal circuits connecting the brainstem to the cortex. These neuronal connections begin from the reticular formation (RF) of the brainstem and the intralaminar nucleus of thalamus to the cerebral cortex (Gosseroes et al., 2011). In addition, the ARAS system also includes several brainstem nuclei (such as dorsal raphe, locus coeruleus, pedunculopontine nucleus, median raphe and parabrachial nucleus), non-specific thalamic nuclei, hypothalamus, and basal forebrain (Fuller et al., 2011). Development of diffusion tensor tractography (DTT), which is derived from diffusion tensor imaging (DTI), has enabled three-dimensional reconstruction and estimation of the ARAS of the live human brain (Yeo et al., 2013; Jang et al., 2014). Many studies using DTT have demonstrated injury of the ARAS in various brain pathologies including subarachnoid hemorrhage, cerebral infarction, traumatic brain injury, intracerebral hemorrhage, and hypoxic ischemic brain injury (Jang et al., 2015a, b; Jang and Kim, 2015; Jang and Lee, 2015a; Jang and Seo, 2015), while only a few studies have reported on the recovery of an injured ARAS in patients with brain injury (Jang et al., 2015c, 2016; Jang and Lee, 2015b). In this study, we report on a stroke patient who showed recovery of a multiply injured ARAS using DTT. A 57-year-old male patient with a spontaneous intraventricular hemorrhage (IVH) and left basal ganglia hemorrhage (ICH) underwent bilateral frontal extraventricular drainage (EVD) for IVH (Figure 1A). One month from symptom onset, he was transferred to the department of rehabilitation. He exhibited impaired alertness, with a Glasgow Coma Scale (GCS) score of 9 (eye opening: 4, best verbal response: 1, and best motor response: 4) and a Coma Recovery Scale-Revised (CRS-R) score of 5 (auditory function: 0, visual function: 0, motor function: 3, verbal function: 0, communication: 0, and arousal: 2) (Teasdale and Jennett, 1974; Giacino et al., 2004). He showed quadriplegia of motor function (shoulder abductor: 0/0, elbow flexor: 0/0, finger extensor: 0/0, hip flexor: 0/0, knee extensor: 0/0 and ankle dorsiflexor: 0/0). Brain MR images taken 1 month after onset showed multiple leukomalactic lesions in both frontal lobes, left subcortical white matter and basal ganglia (Figure 1A). He underwent comprehensive rehabilitative therapy, which included neurotropic drugs (pramipexole 3 mg, ropinirole 3 mg, amantadine 300 mg and levodopa 750 mg), physical therapy, occupational therapy until 7 months after symptom onset. At 7 months after onset, his GCS score had recovered to 15 (eye opening: 4, best verbal response: 5, and best motor response: 6) with a GRS-R score of 22 (auditory function: 4, visual function: 5, motor function: 6, verbal function: 3, communication: 1, arousal: 3). He presented some recovery of the left hemiplegia after 7 months (shoulder abductor: 0/4, elbow flexor: 0/4, finger extensor: 0/3, hip flexor: 0/3, knee extensor: 0/3 and ankle dorsiflexor: 0/2). However, we could not perform Mini-Mental State Exam (MMSE) evaluation at 1 and 7 months due to poor awareness and cognition. The patient’s wife provided signed, informed consent and our institutional review board approved the study protocol. DTI data were acquired twice (1 and 7 months after onset) using a 6-channel head coil on a 1.5 T Philips Gyroscan Intera (Philips, Best, The Netherlands) with single-shot echo-planar imaging. For each of the 32 non-collinear diffusion sensitizing gradients. Imaging parameters were as follows: acquisition matrix = 96 × 96; reconstructed matrix = 192 × 192; field of view = 240 × 240 mm; repetition time = 10,726 ms; echo time = 76 ms; b = 1,000 s/mm; number of excitations = 1; and a slice thickness of 2.5 mm with no gap. Analysis of diffusion-weighted imaging data was performed using the Oxford Centre for Functional Magnetic Resonance Imaging of the Brain (FMRIB) Software Library (FSL; www.fmrib.ox.ac.uk/fsl). Fiber tracking was performed using a probabilistic tractography method based on a multifiber model, and applied in the current study utilizing tractography routines implemented in FMRIB Diffusion (Behrens et al., 2007). For the lower dorsal ARAS, the seed region of interest (ROI) was given at the RF of the pons and the target ROI was given at the thalamic intralaminar nucleus (ILN) (Yeo et al., 2013). For the lower ventral ARAS, the seed ROI was given at the pontine RF and the target ROI was given at the hypothalamus (Jang and Kwon, 2015). Finally, for the neural connectivity of the upper ARAS, the seed ROI was given at the ILN (Jang et al., 2014). Out of 5,000 samples generated from the seed voxel, results for contact were visualized at a threshold for the lower (dorsal and