Covert Subclinical Neurocognitive Sequelae During the Rehabilitation Course of Severe Coronavirus Disease 2019

Abstract

Apart from respiratory symptoms, encephalopathy and a range of central nervous system complications have been described in coronavirus disease 2019. However, there is a lack of published literature on the rehabilitative course and functional outcomes of severe coronavirus disease 2019 with encephalopathy. In addition, the presence of subclinical neurocognitive sequelae during postacute rehabilitation has not been described and may be underrecognized by rehabilitation providers. We report the rehabilitative course of a middle-aged male patient with severe coronavirus disease 2019 who required intensive care and mechanical ventilation. During postacute inpatient rehabilitation for severe intensive care unit–related weakness, an abnormal cognitive screen prompted brain magnetic resonance imaging, which revealed destructive leukoencephalopathy. Subsequently, detailed psychometric evaluation revealed significant impairments in the domains of processing speed and executive function. After 40 days of intensive inpatient rehabilitation, he was discharged home with independent function. This report highlights the need for an increased awareness of covert subclinical neurocognitive sequelae, the role of comprehensive rehabilitation, and value of routine cognitive screening therein and describes the neurocognitive features in severe COVID-19.

A neurological spectrum in association with COVID-19 (coronavirus disease 2019) is increasingly being described.¹ Altered mental status and unspecified encephalopathy without distinct magnetic resonance imaging (MRI) or cerebrospinal fluid abnormalities seem to be common neurological manifestations associated with severe COVID-19.²

To date, the rehabilitation characteristics for severe COVID-19 patients with neurological sequelae are not well described. This is a case report of the acute intensive care unit (ICU) and rehabilitation course and outcome of a patient with severe COVID-19 who presented with reversible encephalopathy in the ICU and received inpatient rehabilitation for severe ICU-acquired weakness. He was subsequently found to have severe and diffuse cognitive impairment associated with white matter leukoencephalopathy on MRI during rehabilitation.

The patient gave written consent for publication of this case report. This study conforms to all CARE guidelines and reports the required information accordingly (see Supplemental Checklist, Supplemental Digital Content 1, https://links.lww.com/PHM/B164).

CASE REPORT

Initial Presentation and ICU Course

A 39-yr-old, right-handed male technician with a university degree presented to an acute hospital with a 5-day history of hemoptysis, fever, and dyspnea. He had no previous medical history or previous cognitive disorders. On admission, he was alert and fully oriented with a Glasgow Coma Scale (GCS) score of 15 with no focal neurological deficits. Severe acute respiratory syndrome coronavirus 2 infection was confirmed by reverse transcriptase polymerase chain reaction assay on a nasopharyngeal swab. In view of hemoptysis and rapidly worsening type 1 respiratory failure, he was intubated and placed on mechanical ventilation.

At day 14 of ICU stay, despite weaning sedation, he displayed fluctuating consciousness with predominantly depressed mental status, raising the possibility of encephalopathy.^3,4 Although neurological examination showed bilateral brisk reactive pupils of 3 mm, he had a GCS of 3 T with no grimace or motor response to noxious stimuli in limbs. A noncontrasted brain computed tomography performed on day 16 of ICU stay did not report any abnormalities (Fig. 1A). Brain MRI, electroencephalogram, and cerebrospinal fluid studies were not performed then because of severe medical instability. His encephalopathy gradually improved, and he attained a GCS score of 11 T on day 62 of hospitalization. He was intubated for a total of 67 days, with resolution of encephalopathy on day 69.

FIGURE 1: Computed tomography and MRI imaging of brain. A, Noncontrast computed tomography on day 17 showing normal findings. B, T1-weighted MRI performed on day 114 shows multiple well-defined rounded or tubular shaped focal areas of low-signal intensity in the deep cerebral white matter bilaterally (arrows). C and D, On fluid-attenuated inversion recovery images, these discrete white matter lesions show low-signal intensity.

Subacute Phase and Rehabilitation Course

He was weaned down to room air and was transferred to a recovery ward after successful extubation. He was alert and fully oriented, with a GCS score of 11 T. However, generalized muscle wasting and normal muscle stretch reflexes, with motor power of 2/5 in all 4 limbs secondary to deconditioning, were noted.

He was later transferred to an inpatient rehabilitation unit at day 103 after admission as he remained totally dependent in all aspects of mobility despite receiving ward-based physiotherapy. His tracheostomy was successfully decannulated before transfer. The Functional Independence Measure (FIM), comprising motor and cognitive subset FIM scores, was the main disability measure for the patient on admission and discharge from inpatient rehabilitation.⁵ Other functional measures including 6-min walk test,⁶ 30-sec sit-to-stand test,⁷ 10-meter walk test,⁸ and 2-min step test⁹ were recorded as well.

On admission to rehabilitation, he had a GCS of 15, and neurological examination of his cranial nerves, visual fields, and sensory system were again normal.

Administration of the Hospital Anxiety and Depression Scale were within normal limits (0/7 for both anxiety and depression scores), and no behavioral disorders were found. On admission to inpatient rehabilitation, his screening Montreal Cognitive Assessment score was 17/30 and revealed cognitive impairments in multiple domains: visuospatial function (1/5), naming (3/3), attention (4/6), language (1/3) abstraction (0/2), memory (3/5), and orientation (5/6).

In view of his grossly abnormal Montreal Cognitive Assessment, and newly emerging evidence of abnormal brain MRI findings in severe COVID-19 patients,¹⁰ MRI neuroimaging was performed on day 114 of hospitalization and this revealed multiple sharply marginated focal abnormalities limited to the deep cerebral white matter bilaterally (Fig. 1B) with low-signal intensity on T1-weighted MR images, high signal on T2-weighted images and decreased signal on fluid-attenuated inversion recovery images (Figs. 1C, D). They did not show contrast enhancement, restricted diffusion, or susceptibility abnormalities on a gradient-recalled echo MR image or susceptibility weighted images. Time-of-flight MR angiography did not reveal stenosis or occlusion in the Circle of Willis, and only five widely distributed tiny foci of susceptibility, interpreted as microhemorrhages, were detected on susceptibility weighted images.

The striking findings of white matter lesions on MRI were unexpected, and this prompted detailed neuropsychological evaluation, which was performed at day 124 of hospitalization by a clinical psychologist. These findings are summarized in Table 1. This revealed moderately severe cognitive impairment, which could possibly be explained by MRI abnormalities within the bilateral supratentorial deep cerebral white matter. For example, significant impairments were found especially in the domains of processing speed, working memory, visuospatial processing, and executive function, which correlated with the involved deep frontal white matter. Impairments in language processing and naming could be related to dominant hemisphere (left) frontal lobe lesions or memory retrieval. Notably, the motor and sensory tracts in the M1 and supplementary motor cortex were spared in our patient, as there were no demonstrable lesions in these structures on brain computed tomography or MRI. Moreover, focal neurological deficits such as limb and gait apraxia were absent, and sensory testing was normal.

TABLE 1 - Neuropsychological profile at 4 mos post illness

Domains	Tests	Status
Estimated premorbid level of functioning	Average, University level degree
Attention	WAIS-IV Digit Span Forward	Average
Working memory	WAIS-IV Digit Span Backward	Borderline
	WAIS-IV Digit Span Sequencing	Borderline
Processing speed	WAIS-IV Symbol Search	Extremely low
	WAIS-IV Coding	Extremely low
	CTT 1	Mildly to moderately impaired
Visuospatial	Simple copy	Intact
	Clock Drawing Test	Poor spatial arrangement of numbers
	RBANS Figure Copy	Average
	Rey CFT Copy	Mildly to moderately impaired
Visuoperceptual	RBANS Line Orientation	Low average
Language	RBANS Picture Naming	Moderately impaired
Memory
Verbal memory (immediate)	RBANS List Learning	Average
	RBANS Story Memory	Above average
Verbal memory (delayed)	RBANS List Recall	Average
	RBANS List Recognition	NA
	RBANS Story Recall	Superior
Visual memory (immediate)	Rey CFT Immediate Recall	Above average
Visual memory (delayed)	RBANS Figure Recall	Low average
Executive functioning
Planning and organization	Rey CFT	Used a piecemeal and haphazard approach
Strategy generation	BADS Key Search	Average
Mental flexibility	CTT 2	Severely impaired
	Color Form Sort	Able to sort initially but required demonstration to sort into different groups
Verbal fluency	RBANS Semantic Fluency	Mildly impaired

BADS, Behavioral Assessment of the Dysexecutive Syndrome; CTT, Color Trails Test; NA, not available; RBANS, Repeatable Battery for the Assessment of Neuropsychological Status; Rey CFT, Rey Complex Figure Test; WAIS-IV, Wechsler Adult Intelligence Scale Fourth Edition.

Because of his cognitive deficits, several graded rehabilitation strategies were used focusing on functional skills and compensatory strategy training. Environmental compensations included providing him with a daily routine using a modifiable visual chart of his daily schedule detailing meal, therapy, and recreational timings. Eventually, he was able to plan his own schedule independently upon discharge. He was also successfully taught to self-monitor for signs of physical or cognitive fatigue and take a rest break when required. Task-specific training was performed for ward-based topographical orientation and activities of daily living, and he was able to ambulate and perform ward-based basic activities of daily living independently by day 11 of rehabilitation. Apart from functional training, he was also engaged in virtual reality games using a Nintendo Wii platform (eg, table tennis, boxing) to improve visual learning, perceptual skills, processing speed, and cardiovascular fitness. As his recovery progressed, repetitive task-specific training was directed at facilitating independence in instrumental activities of daily living such as meal preparation and laundry. He was taught strategies for task simplification, such as proper organization of his workspace, and setting up of equipment before tasks (eg, preparation of cleaning kit for tracheostomy wound care). Given impairments in language, he was taught to rephrase his statements using simple sentence structures when communicating. The patient was also educated on his neuropsychological findings by a clinical psychologist who implemented compensatory strategies such as allowing additional time to process information, refraining from multitasking, requesting for short and simple instructions, providing step-by-step instructions in a written form, and rehearsal when learning new skills. He did not require any psychopharmacological agents.

Weekly rehabilitation team conferences were held to review goal setting, functional progress, and discharge planning. From an initial admission cognitive FIM score of 23/35, he made clinically significant improvements and was discharged with a cognitive FIM score of 33/35 after 40 days of inpatient rehabilitation.

To treat his severe ICU-acquired weakness and functional dependency, strength and endurance tasks were instituted together with functional mobility training by his physiotherapist.¹¹ Despite cognitive impairments, the patient was able to participate fully in rehabilitation. The patient had impaired exercise tolerance on admission, with an initial 6-min walk test of 90 m (terminated at the fourth minute of testing), a motor FIM score of 54/91, 10-meter walk test of 42 secs, 30-sec sit-to-stand test of 6 reps, and a 2-min step test of 45 reps. The general principles of cardiopulmonary exercise prescription were used, aiming for a moderate exercise intensity based on heart rate reserve.¹² He had continuous heart rate, pulse oximetry, and Borg Rating of Perceived Exertion monitoring, with 5 mins of warm-up and cool-down sessions. Intensity and duration of therapy sessions were incrementally increased according to the patient’s tolerance.

After his 40-day rehabilitation stay, he made significant overall gains in functional status with a discharge 6-min walk test of 485 m, a motor FIM score of 91/91, 10-meter walk test of 8 secs, 30-sec sit-to-stand test of 11 reps, and a 2-min step test of 58 reps.

He was reassessed at 3 mos after discharge by the multidisciplinary rehabilitation team, where he demonstrated further functional improvement. Assessment measures revealed a 6-min walk test of 474 m, 10-meter walk test of 5.5 secs, and a 30-sec sit-to-stand test of 17 reps (2-min step test not performed). He had also created his own home routine by then, was able to prepare his meals independently, was able to take public transport independently for his medical appointments, and did not express significant impairments in communication with his friends. He is currently on follow-up with the physiatrist, with plans for a repeat neuropsychological and occupational assessment in the following months to facilitate an eventual return-to-work program.

DISCUSSION

This case report illustrates the diverse rehabilitation challenges of severe COVID-19 that survivors encounter in the continuum of prolonged hospitalization involving ICU, post-ICU, and rehabilitation settings. In patients with severe infection who recover slowly after a prolonged ICU course, it is important to consider occult and subtle brain involvement despite normal serial neurological examinations, and further investigations (brain MRI, cerebrospinal fluid studies, neuropsychological assessments) may be warranted. However, it can be challenging to perform such investigations during a pandemic of an unknown, emerging pathogen in severe COVID-19 patients with multiorgan failure.

Brain MRI was useful during the rehabilitation phase to explain the patient’s severe and diffuse cognitive impairment. A lowered threshold for neuroimaging was appropriate as central nervous system complications for COVID-19 including stroke, encephalitis, acute disseminated encephalomyelitis, and hypoxia have been described in the literature. The exact pathomechanisms and onset for the MRI pattern of destructive leukoencephalopathy and neurocognitive findings remain speculative because of the lack of early brain MRI and cerebrospinal fluid or tissue studies to detect the cytopathic presence of the virus. Early descriptions of central nervous system involvement in the pandemic hypothesize pathomechanisms such as infarction, dysimmunity, encephalitis, and hypoxia,^1,13 although research into treatment, prognosis, and rehabilitation is ongoing.^14,15

In this patient, an overall good postrehabilitation outcome was achieved despite destructive leukoencephalopathy and moderately severe cognitive impairment. The positive functional outcome was likely contributed by his young age, good premorbid cognitive reserve, relatively preserved attention, visual and verbal memory domains, and an interdisciplinary rehabilitative impairment-based approach. Exercise-based rehabilitation has been advocated for the management of physical and functional disability after critical illness,¹⁶ and a closely supervised exercise program based on cardiopulmonary principles for this patient was well tolerated within a goal-orientated inpatient rehabilitation program.

Neuropsychological assessment was helpful in characterizing and explaining the neurocognitive effects of COVID-19. In general, cognitive deficits have not been well reported in COVID-19. An observational case series has been described consisting of 58 patients with COVID-19, of which 14 patients displayed an overt, and not subclinical, dysexecutive syndrome with obvious inattention, disorientation, or poorly organized movements in response to commands.¹⁷ There also seems to be an overlap in neurocognitive deficits in terms of global cognition and executive function with severely ill patients despite full physical recovery, although the pathomechanisms in both groups remain poorly defined.^18,19 Neuropsychological sequelae including impaired concentration and memory have been reported in severe acute respiratory syndrome and Middle East respiratory syndrome patients when evaluated at 6 wks to 39 mos.²⁰ Similarly, in a case series of nine patients with Nipah virus encephalitis,²¹ impaired attention and verbal memory were disproportionately reported. These findings were in contrast to our patient, for whom verbal memory and attention were relatively preserved. Future long-term follow-up studies, including longitudinal neuropsychological testing and MRI, are needed to determine the burden and reversibility of damage, and outpatient multidisciplinary rehabilitation follow-up is often warranted.²²

In conclusion, this case report highlights that in addition to targeting motor or pulmonary impairments,^22,23 a heightened awareness of the covert and subtle manifestations of neurocognitive deficits after severe COVID-19 is needed as part of a holistic rehabilitation approach. Neuroimaging, early cognitive screening and neuropsychological evaluation may be necessary to further characterize neurocognitive deficits and direct cognitive strategies in severe COVID-19, to lessen their potential impact on community and vocational reintegration.