INTRODUCTION
In March 2020, the World Health Organization declared COVID-19 a pandemic disease
with unknown consequences[1]. As of mid-May 2020, there had been almost five million reported infections, and
more than 300,000 deaths worldwide caused by SARS-CoV-2[2]. Although COVID-19 was initially described as a mainly respiratory disease, accumulating
evidence suggests neuropsychiatric complications play an important role in the disease[3],[4]. In a Chinese retrospective cohort, impaired consciousness was observed in 7.5%
of COVID-19 patients, and in 14.8% of severe cases[5]. Neurological signs are almost ubiquitous in very severe COVID-19 patients, with
agitation or positive CAM-ICU findings noted in over 65% of cases[3]. COVID-19 may also be complicated by encephalopathy, headache, and stroke.
Pre-existing neurological disease is associated with increased disease severity and
poorer outcome in COVID-19[6],[7],[8]. Likewise, comorbid SARS-CoV-2 infection can lead to higher rates of disability
in acute neurological patients[9]. There is also an emerging concern on the potential negative impact of COVID-19
in the management of chronic neurological diseases such as Parkinson’s disease[10] and epilepsy[11].
These issues underscore the importance of inpatient neurological evaluation and treatment
during the pandemic. Neurology societies have highlighted the importance of reorganizing
neurology consultation services for in- and outpatient clinics[12],[13],[14],[15]. Although a few retrospective case series have described the occurrence of specific
neurological symptoms and signs in COVID-19 patients[3],[4],[5], we were not able to find systematic reports of the real-life experience of consulting
neurologists, other than the anecdotal editorial by Bersano and Pantoni[12].
The objective of this study was to report the main reasons for requesting neurological
consultations by internists and intensivists in a COVID-19 dedicated hospital and
the frequency of individual neurological diagnoses in a real-life situation over a
60-day period.
METHODS
Study design and participants
In March 2020, the Instituto Central do Hospital das Clínicas da Faculdade de Medicina
da Universidade de São Paulo, Brazil (ICHC-HCFMUSP) was designated a COVID-dedicated
tertiary referral center, with 900 beds (including three hundred Intensive Care Units
- ICU - beds). Since its reorganization, a team of seven neurologists and eight neurology
residents was assigned to provide on-demand consultations exclusively for COVID-19
patients.
Patients with confirmed or suspected COVID-19 requiring hospitalization underwent
SARS-CoV-2 RNA detection with throat swab real-time reverse transcription-polymerase
chain reaction (RT-PCR). In high clinical suspicion PCR-negative cases, a repeat RT-PCR
test was performed.
This retrospective study enrolled exclusively PCR-positive cases evaluated by the
neurology inpatient consultation team between March 23rd and May 23rd, 2020. This is a real-life situation study: neurological consultations were requested
at the discretion of each patient’s attending physician.
Patient clinical evaluation included clinical examination, routine laboratory testing
(blood cell count, biochemical analysis, liver and renal function tests, C-reactive
protein, D-dimer, lactate dehydrogenase, creatine phosphokinase) and chest computed
tomography (CT). Patients were divided into two groups based on the severity of the
respiratory symptoms (severe and non-severe). Severe respiratory symptoms were defined
as respiratory insufficiency requiring mechanical ventilation.
For each case, the reason for requesting the neurological consultation and the final
neurological diagnosis were recovered from the specialist consult electronic form.
Neurological evaluation consisted of neurological examination and additional ancillary
tests (brain CT, Magnetic Resonance Imaging - MRI, cerebrospinal fluid - CSF - analysis),
requested at the neurologist’s discretion.
Difficult diagnosis cases were presented in a consensus meeting with the Neurology
Department staff at Hospital das Clínicas. In cases with two or more neurological diagnoses, the final diagnosis was that of
the most severe condition, or of the more plausible cause for the condition (for example:
in an encephalopathy due to a confirmed stroke, the final diagnosis was stroke). Neurological
diagnoses were finally classified into four main groups: encephalopathy, cerebrovascular
disease, epilepsy, and neuromuscular disease. A fifth group (named “others”) included
diagnoses not fitting in the four other groups (i.e., headache, vertigo), including management or follow-up of pre-existing neurological
conditions.
Statistical analysis
All analyses were performed with the Statistical Package for Social Sciences software,
version 21.0 (SPSS, IBM Statistics, Chicago, IL, USA). Categorical variables were
expressed as absolute and relative frequencies and compared using Pearson’s χ2 univariate analysis. All continuous variables were expressed as mean and standard
deviations and compared with Student’s t-test or Mann-Whitney test according to normality distribution. All tests were two-tailed.
Statistical significance was accepted at p<0.05.
Standard protocol approvals
The institutional review board approved the investigation protocol, and waived the
informed consent, since this was a retrospective study, and no patient had undergone
an experimental intervention.
RESULTS
During the studied period, 1,208 patients with confirmed COVID-19 were admitted to
the hospital and 89 neurological consultation requests were placed (7.4%). The number
of neurological consultations rose from 0.3 consults/day in the first week, to 3.6/day
in the last week, representing a twelve-fold increase ([Figure 1]).
Figure 1 The solid line displays the number of patients who underwent initial neurological
evaluation between March 23rd and May 23rd, 2020 at Instituto Central do Hospital das Clínicas da Faculdade de Medicina da Universidade
de São Paulo, Brazil (ICHC-HCFMUSP). The dotted red line indicates the 7-day moving
average. For any given day (D0), this number corresponds to the mean of all 7 values
ranging from D-3 to D+3, and annotated on D0, and is therefore available from March
25th through May 21st.
Approximately half of the patients had severe respiratory conditions, requiring orotracheal
intubation and mechanical ventilation ([Table 1]). Mean age of the 89 cases was 57.4+/-15.9 years, with no difference between groups;
the majority of patients (61.8%) were men. Hypertension and diabetes mellitus were the most common comorbidities. The most prevalent typical symptoms of COVID-19
were fever, cough, and dyspnea. Anosmia and dysgeusia were rarely reported by patients.
Thirty-nine patients (43.8%) had a previous neurological diagnosis ([Table 1]).
Table 1
Clinical features of patients with COVID-19 evaluated by the Neurology team between
March 23rd and May 23rd, 2020 at Instituto Central do Hospital das Clínicas da Faculdade de Medicina da Universidade
de São Paulo, Brazil (ICHC-HCFMUSP).
|
Characteristics
|
Nonsevere respiratory condition
|
Severe respiratory condition
|
Total
|
p-value
|
|
Number of patients
|
44
|
45
|
89
|
|
|
Age, mean (SD), y
|
57.7 (16.2)
|
56.9 (15.8)
|
57.4 (15.9)
|
0.811
|
|
Gender
|
|
Female
|
15 (34.1%)
|
19 (42.2%)
|
34 (38.2%)
|
0.515
|
|
COVID-19 symptoms
|
|
Fever
|
24 (60%)
|
33 (76.7%)
|
57 (68.7%)
|
0.155
|
|
Cough
|
30 (75%)
|
29 (67.4%)
|
59 (71.1%)
|
0.478
|
|
Dyspnea
|
22 (56.4%)
|
37 (84.1%)
|
59 (71.1%)
|
0.008
|
|
Myalgia
|
5 (12.8%)
|
13 (29.5%)
|
18 (21.7%)
|
0.108
|
|
Fatigue
|
15 (38.5%)
|
14(31.8%)
|
29 (34.9%)
|
0.645
|
|
Diarrhea
|
5 (12.8%)
|
3 (6.8%)
|
8 (9.6%)
|
0.465
|
|
Anosmia
|
3 (8.1%)
|
5 (11.4%)
|
8 (9.9%)
|
0.570
|
|
Dysgeusia
|
0
|
3 (9.4%)
|
3 (4.8%)
|
0.135
|
|
Comorbidities
|
|
Hypertension
|
23 (59.0%)
|
28 (65.1%)
|
51 (62.2%)
|
0.651
|
|
Diabetes
|
12 (30.8%)
|
19 (44.2%)
|
31 (37.8%)
|
0.257
|
|
Heart disease
|
8 (20.5%)
|
9 (20.9%)
|
17 (20.7%)
|
0.590
|
|
Malignancy
|
4 (10.3%)
|
4 (9.1%)
|
8 (9.6%)
|
1.000
|
|
Chronic renal disease
|
11 (28.2%)
|
6 (14.0%)
|
17 (20.7%)
|
0.172
|
|
Lung disease
|
5 (12.8%)
|
5 (11.6%)
|
10 (12.2%)
|
1.000
|
|
Smoking
|
8 (18.2%)
|
10 (22.2%)
|
18 (20.2%)
|
0.376
|
|
Previous neurological disease
|
|
Cerebrovascular disease
|
7 (16.7%)
|
9 (20.9%)
|
16 (18.8%)
|
0.782
|
|
Epilepsy
|
8 (18.2%)
|
3 (6.7%)
|
11 (12.4%)
|
0.253
|
|
Dementia
|
4 (9.5%)
|
4 (9.1%)
|
8 (9.3%)
|
1.000
|
|
Outcome
|
|
Death
|
2 (5.0%)
|
12 (27.3%)
|
14 (16.7%)
|
0.008
|
The main reasons for neurological consultation were: altered level of consciousness,
muscle weakness, and psychomotor agitation. In eleven cases, the neurologist was called
to assist in the management and follow-up of patients with pre-existing neurological
diseases without acute neurological manifestations ([Table 2]). In some cases, two or more symptoms were reported by the clinician on neurological
consultation requests. Main ancillary exams performed were brain CT (in 71 patients)
and lumbar puncture (in 27 patients), while brain MRI was performed in only 15 patients
(20.0%).
Table 2
Reason for Neurological Consultation in patients with COVID-19 between March 23rd and May 23rd, 2020 at Instituto Central do Hospital das Clínicas da Faculdade de Medicina da Universidade
de São Paulo, Brazil (ICHC-HCFMUSP).
|
Symptom (reported by the internist)
|
Nonsevere respiratory condition
|
Severe respiratory condition
|
Total
|
p-value
|
|
Altered level of consciousness
|
17 (38.6%)
|
18 (40.0%)
|
35 (39.3%)
|
0.534
|
|
Psychomotor agitation
|
8 (18.2%)
|
4 (8.9%)
|
12 (13.5%)
|
0.166
|
|
Slow awakening from sedation
|
0
|
5 (11.1%)
|
5 (5.6%)
|
0.029
|
|
Focal neurological signs
|
2 (4.5%)
|
1 (2.2%)
|
3 (3.4%)
|
0.491
|
|
Muscle weakness
|
6 (13.6%)
|
10 (22.2%)
|
16 (18.0%)
|
0.219
|
|
Vertigo
|
1 (2.3%)
|
1 (2.2%)
|
2 (2.2%)
|
0.747
|
|
Headache
|
3 (6.8%)
|
0
|
3 (3.4%)
|
0.117
|
|
Seizure and transient changes in consciousness
|
3 (6.8%)
|
4 (8.9%)
|
7 (7.9%)
|
0.513
|
|
Seizure in epileptic patient
|
4 (9.1%)
|
1 (2.2%)
|
5 (5.6%)
|
0.173
|
|
Movement disorder*
|
2 (4.5%)
|
4 (8.9%)
|
6 (6.7%)
|
0.349
|
|
Management or follow-up of pre-existing neurological disease
|
6 (13.6%)
|
5 (11.1%)
|
11 (12.4%)
|
0.484
|
*Myoclonus, nonspecific psychomotor agitation.
The most prevalent diagnoses were delirium/encephalopathy, stroke, previous neurological
diseases, seizures, neuromuscular disorders, other acute brain lesions, as well as
other mild nonspecific symptoms ([Table 3] and [Figure 2]).
Table 3
Final neurological diagnoses of COVID-19 patients evaluated by the Neurology team
between March 23rd and May 23rd, 2020 at Instituto Central do Hospital das Clínicas da Faculdade de Medicina da Universidade
de São Paulo, Brazil (ICHC-HCFMUSP).
|
Neurological diagnosis
|
Nonsevere respiratory condition
|
Severe respiratory condition
|
Total
|
|
Diffuse encephalopathy
|
20 (45.5%)
|
20 (44.4%)
|
40 (44.4%)
|
|
Stroke
|
5 (11.4%)
|
6 (13.3%)
|
11 (12.3%)
|
|
Cerebral venous thrombosis
|
1 (2.3%)
|
1 (2.2%)
|
2 (2.2%)
|
|
Intracranial hemorrhage
|
1 (2.3%)
|
1 (2.2%)
|
2 (2.2%)
|
|
Acute non-vascular structural encephalic lesions*
|
1 (2.3%)
|
2 (4.4%)
|
3 (3.4%)
|
|
Seizure in epileptic patient
|
2 (4.5%)
|
2 (4.4%)
|
4 (4.5%)
|
|
Acute symptomatic or provoked seizure
|
2 (4.5%)
|
2 (4.4%)
|
4 (4.5%)
|
|
Peripheral neuropathy
|
0
|
3 (6.7%)
|
3 (3.4%)
|
|
Rhabdomyolysis
|
2 (4.5%)
|
0
|
2 (2.2%)
|
|
Pre-existing neurological disease
|
4 (9.1%)
|
4 (8.9%)
|
8 (9.0%)
|
|
Others**
|
6 (13.6%)
|
4 (8.9%)
|
10 (11.2%)
|
*Wernicke encephalopathy and central pontine myelinolysis (01 case), acute disseminated
encephalomyelitis (01 case), and Posterior Reversible Encephalopathy Syndrome (01
case).
**Headache, vertigo, syncope, somatoform disorder, primary psychosis.
Figure 2 Graphic distribution of neurological diagnoses by disease group in the period between
March 23rd and May 23rd, 2020 at Instituto Central do Hospital das Clínicas da Faculdade de Medicina da Universidade
de São Paulo, Brazil (ICHC-HCFMUSP). Group “Others” included “management or follow-up
of previous neurological disease”, headache, vertigo, syncope, somatoform disorder,
and primary psychosis.
Associations between some reasons for neurological consultation requested by the internist
and the final diagnoses established by the neurologist were found: altered level of
consciousness correlated with encephalopathy, muscle weakness with cerebrovascular
disease, seizure and transient changes in consciousness with epilepsy ([Table 4]).
Table 4
Association between reason for neurological consultation and final diagnosis by disease
group.
|
Reason for Neurological consultation
|
Final diagnosis by disease group
|
p-value
|
|
Encephalopathy
|
Seizure and Epilepsy
|
Cerebrovascular disease
|
Neuromuscular disorder
|
Others
|
|
Altered level of consciousness
|
28 (80.0%)
|
1 (2.9%)
|
4 (11.4%)
|
1 (2.9%)
|
1 (2.9%)
|
<0.001
|
|
Psychomotor agitation
|
7 (58.3%)
|
1 (8.3%)
|
2 (16.7%)
|
1 (8.3%)
|
1 (8.3%)
|
0.925
|
|
Slow awakening from sedation
|
4 (80.0%)
|
0
|
1 (20.0%)
|
0
|
0
|
0.562
|
|
Focal neurological signal
|
1 (33.3%)
|
0
|
2 (13.3%)
|
0
|
0
|
0.215
|
|
Muscle weakness
|
3 (18.8%)
|
0
|
7 (43.8%)
|
5 (31.3%)
|
1 (6.3%)
|
<0.001
|
|
Vertigo
|
0
|
0
|
1 (50.0%)
|
0
|
1 (50.0%)
|
0.403
|
|
Headache
|
0
|
0
|
0
|
0
|
3 (100%)
|
0.004
|
|
Seizure and transient changes in consciousness
|
4 (57.1%)
|
3 (42.9%)
|
0
|
0
|
0
|
0.011
|
|
Seizure in epileptic patient
|
0
|
4 (80.0%)
|
0
|
0
|
1 (20.0%)
|
<0.001
|
|
Movement Disorder
|
4 (66.7%)
|
0
|
1 (16.7%)
|
0
|
1 (16.7%)
|
0.808
|
|
Management or follow-up of pre-existing neurological disease
|
3 (27.3%)
|
0
|
1 (9.1%)
|
2 (18.2%)
|
5 (45.5%)
|
0.001
|
DISCUSSION
In this study, 89 neurological consults were analyzed, requested among 1,208 COVID-19
patients admitted to a single University-based tertiary referral hospital dedicated
to severe COVID-19 patient care (7.4% of total cases). In this setting, patients were
evaluated by neurologists with the same training; more difficult cases were discussed
in a weekly consensus meeting with the neurology department staff.
More than half of these complications (65.1%) were severe or potentially severe, consisting
of encephalopathy, acute cerebrovascular disease, and non-vascular acute encephalic
lesions, some of which can cause serious neurological sequelae. These severe complications
were seen both in patients who required mechanical ventilation and in those with less
severe degrees of the disease.
In the early stage of the COVID-19 pandemic, Mao et al. reported the occurrence of
neurological manifestations in 214 patients admitted to three COVID-19 dedicated hospitals
in Wuhan, China[5]. In a retrospective analysis of medical records, the authors found that 78 (36.4%)
of the patients had central or peripheral nervous system involvement. Manifestations
ranged from mild, such as anosmia and headache, to stroke and encephalopathy. Although
the reported frequency in the Wuhan study was high, severe central nervous system
(CNS) diagnoses (acute cerebrovascular diseases and impaired consciousness) were reported
in 28.2% of these 78 patients.
Another study applying a similar methodology to that used in Wuhan[5] found neurological manifestations in 512 (57.4%) of 841 patients in two COVID-19
hospitals in Spain[4]. Severe manifestations of CNS involvement, using the same disease categorization
used in the present study, were described in 36.5% of the cases[4].
In an observational case series, Helms et al. recorded neurological manifestations
in 58 consecutive patients admitted to the hospital due to COVID-19-related acute
respiratory distress syndrome, in Strasbourg, France. All patients were treated in
ICUs. Agitation was noted in 69% of the patients when neuromuscular blockade was discontinued.
Diffuse corticospinal tract signs were observed in 67% of patients. The severity of
the neurological manifestation in these cases with acute respiratory failure was even
higher than in our sample[3].
The lower frequency and the greater severity of neurological involvement that we found
were most likely due to case ascertainment, similarly to the real-life situation,
in which neurologists are called to evaluate patients who present signs or symptoms
that concern internists or intensivists. Therefore, a bias toward potentially more
serious and, therefore, less frequent manifestations was present in our study.
Our data also revealed that neurological manifestations are frequent both in cases
that required mechanical ventilation, as well as in less severe ones. The fact that
more serious neurological conditions, such as stroke and encephalopathy, were equally
frequent in more and in less severe patient groups was unexpected.
Neurological manifestations in COVID-19 can be due to indirect nervous system involvement
resulting from systemic diseases, such as hypoxia, uremia, coagulopathy, and critical
care neuropathy, due to the invasion of the CNS by SARS-CoV-2, or caused by other
mechanisms such as inflammatory and immune-mediated (including post-infectious) reactions.
Factors that support an indirect (systemic) mechanism for neurological complications
of SARS-COV2 infection, such as encephalopathy, are the absence of SARS-CoV2 genetic
material in the CSF, and the lack of signs of brain involvement on neuroimaging (CT
and MRI) in the vast majority of cases in which these tests were obtained[16]. More recently, postmortem pathological examination of brain specimens obtained
from 18 patients who died from COVID-19 failed to demonstrate inflammatory changes
indicative of encephalitis or changes related to a direct viral effect and disclosed
only hypoxic changes[17].
On the other hand, case reports and small case series provided evidence of direct
CNS involvement in COVID-19 infection. Moriguchi et al., in Japan, reported a meningoencephalitis
case with medial temporal lobe lesions on MRI, in which SARS-CoV-2 was detected in
the CSF by RT-PCR[18]. In the previously cited study by Helms et al., leptomeningeal enhancement was noted
on MRI in 8/13 patients, and bilateral frontotemporal hypoperfusion was noted in all
11 patients who underwent perfusion MRI[3]. In another report, MRI showed a subtle hyperintensity in the olfactory bulbs and
cortical hyperintensity in the right gyrus rectus, consistent with local invasion
and direct brain parenchymal lesion by SARS-CoV-2 in a case with anosmia[19]. Most remarkable was the documented finding of the presence of SARS-CoV-2 in neurons
and endothelial cells in the frontal lobe of a case of COVID-19 on electronic microscopy,
which was confirmed by detection of SARS-CoV-2 genetic material in the brain with
RT-PCR[20]. In that case, PCR for SARS-CoV-2 was negative in the CSF.
SARS-CoV-2 encephalopathy may be also caused by acute necrotizing encephalopathy[21],[22] and encephalitis related to an exaggerated inflammatory reaction in the CNS, with
high cytokine and other inflammatory mediators levels in the CSF[23],[24], probably related to a hyperinflammatory systemic reaction against SARS-CoV-2, with
a breakdown of the blood-brain barrier. Alternatively, encephalitis may be caused
by a CNS-specific inflammatory response[23],[24]. Acute disseminated encephalomyelitis (ADEM) has been reported in SARS-CoV-2 patients,
indicating a probable virally triggered immune-mediated mechanism leading to CNS inflammatory
lesions[25],[26].
In our study, the occurrence of neurological manifestations in cases with less severe
COVID-19 systemic disease, may indicate that neurological symptoms, at least in part
of our cases, occurred regardless of systemic involvement, and that neurological symptoms
are probably due to heterogeneous causes. Further studies are needed to elucidate
the precise mechanism underlying the neurological manifestations of SARS-CoV-2[27], which will contribute to the development of more rational therapeutic interventions
and, possibly, reduce the severity of the impact of SARS-CoV-2 in the nervous system.
The limitations of our study are those of a retrospective study. The history of COVID-19
pre-admission symptoms and neurological symptoms, for instance, was frequently obtained
from relatives of critically ill or confused patients and could be prone to recall
bias. Some potentially relevant clinical and demographical information, such as the
prevalence of obesity or angiotensin-converting enzyme inhibitors, were not available.
Our patient population originated from neurology consultation requested by internists,
which corresponds to 7.4% of the total COVID-19 population in our hospital. This makes
our study inadequate to evaluate the total prevalence of specific signs and diagnoses
in the general COVID-19 population. Perception of the need for a neurological evaluation,
especially in severely ill patients, may have varied among different medical teams.
Of note, the daily number of requests for neurological consultations increased along
the study period, probably reflecting an increased awareness of the frequency and
severity of neurological manifestations of SARS-CoV-2 infection, as well as of the
importance of neurology specialized care for this patient population. Like all studies
carried out in COVID-19 dedicated hospitals, there were restrictions on the availability
of MRI, CT, electroencephalography, CSF, electroneuromyography, and neuropathological
examinations. Also, the purpose of this work was not to evaluate the details of each
case to establish the pathophysiology of stroke, encephalopathy, seizures, and other
neurological conditions, which we intend to do in forthcoming manuscripts. Along the
same lines, neuroimaging and CSF findings are also being prepared for future communications.
The main conclusions of this study are:
-
Severe neurological complications that require neurologist interventions are relatively
common in COVID-19 patients.
-
These complications are equally frequent in patients with severe respiratory failure,
who require respiratory assistance, and less severely ill patients, without significant
respiratory symptoms.
-
Neurologists should be part of the medical team of COVID-19 in dedicated hospitals,
and internists must be trained to recognize signs of neurological complications and
to appropriately request neurology consultations in such cases.
Future studies are needed to further understand the pathophysiology of nervous system
manifestations in SARS-CoV-2 patients, so that more effective therapeutic interventions
can be implemented to prevent and minimize severe neurological complications.
