Neurological manifestations and etiological risk factors in patients hospitalized with COVID-19 in Turkey

The medical records of patients who were hospitalized for COVID-19 and who sought neurological examination between March 2020 and March 2021 at our university hospital, which is the reference pandemic hospital in our region of Turkey for COVID-19 patients, were retrospectively reviewed. Approval of the Trabzon Kanuni Training and Research Hospital management and Local Ethics Committee for Non-Pharmaceutical Clinical Research (approval No. 2021/72, 2021 April 20) was obtained for the study, which was performed following the principles of the contemporary revision of the Declaration of Helsinki (64th WMA General Assembly 2013) and ICH Good Clinical Practice (GCP) Guidelines. Informed consent was specifically waived by the ethics committee because of the retrospective nature of the study. We used STROBE and RECORD statement guidelines for reporting the study [15, 16].

The patients had been diagnosed with COVID-19 by positive SARS-CoV-2 reverse-transcriptase polymerase chain reaction (RT-PCR) test result on a combined sample taken from throat and nasopharynx or negative SARS-CoV-2 RT-PCR (first, second, or third sample) along with characteristic findings on thorax computed tomography (CT) (characteristic lesions including diffuse ground-glass lesion bilaterally and/or consolidation zones). Exclusion criteria included the following: patients with suspected SARS-CoV-2 infection who were excluded from the diagnosis of COVID-19 by the results of the examination upon consultation request; patients who had no acute neurological event or no worsening of a current neurological condition; patients who were consulted in the emergency department without hospitalization; and patients who had undiagnosed pre-existing neurological disease before SARS-CoV-2 infection, and patients with incomplete data in their medical records.

Patients were then divided into 2 groups; patients with an acute neurological event (Group 1) and patients with progression of a current neurological condition (Group 2) depending on the neurological disease status.

Demographic data of patients, comorbidities, hospitalization duration in the ward and/or intensive care unit (ICU), discharge status (live–dead), and laboratory levels were recorded. The Modified Charlson Comorbidity Index (MCCI) was calculated.

Neurological consultation was requested for 238 of 4,295 patients who were hospitalized due to a diagnosis of COVID-19 between March 2020 and March 2021. After exclusions, we included data from 150 patients (94 male, 63%) in the present study (Figure 1). The mean age ± SD of the patients was 68.6 ± 16.0 years (range 21–97 years). The SARS-CoV-2 RT-PCR test result was positive in 140 (93%) patients and negative in 10 (7%) patients. Abnormal neurological findings were detected at a mean of 7.8 ± 9.7 days following RT-PCR or radiological test positivity. The average hospitalization duration of all patients was 15.2 ± 15.5 days. There were 110 patients in Group 1 and 40 patients in Group 2. There was no difference between these groups in terms of age, sex, or hospitalization in ICU (Table 1). The most frequent neurological events in Group 1 were ischemic cerebrovascular events (CVEs), seizure, and encephalopathy, while in Group 2, the most frequent neurological events were deterioration in consciousness, epileptic seizure, and Parkinson disease (Tables 2 and 3). While the mortality rate was 36.3% in Group 1, it was 25% in Group 2, but we found no significant difference overall (P = 0.19). Comorbid diseases are shown in Figure 2. Demographic and clinical data of the patients in Group 1 and Group 2 are summarized in Tables 2 and 3. There were 5 patients with undefined transient neurological findings such as visual hallucination (n = 2), anisocoria (n = 1), neuropathic pain (n = 1), and gait disturbance (n = 1).

Figure 1

Figure 2

For patients in ICU (n = 72), the average duration of stay was 14.9 ± 15.6 days. The MCCI of all patients was 4.69 ± 2.27. Fifty (a third) of the 150 patients died. The logistic regression model for possible factors that may affect mortality found significance with advanced age (P = 0.005), increase in MCCI score (P < 0.001), and prolonged duration of hospitalization (P = 0.04); however, sex (P = 0.28) and time of abnormal neurological finding detection after COVID-19 diagnosis (P = 0.49) were not significantly related to mortality.

Comparison of laboratory test results between dead and alive patients found in deceased patients significantly higher levels of creatinine (1.8 ± 1.6 mg/dL versus 1.0 ± 1.3 mg/dL, P = 0.005), lactate dehydrogenase (LDH) (578.0 ± 457.6 IU/L versus 310.2 ± 125.9 IU/L, P = 0.001), ferritin (1,000.8 ± 771.3 μg/L versus 572.2 ± 551.9 μg/L, P = 0.003), procalcitonin (4.8 ± 13.4 versus 0.5 ± 2.5, P = 0.03), C-reactive protein (CRP) (153.1 ± 88.1 mg/L versus 82.9 ± 67.1 mg/L, P < 0.001), white blood cells (WBC) (12,456.2 ± 8,075.3 cells/μL versus 8,467.3 ± 4,228.5 cells/μL, P = 0.002), neutrophils (12.1 ± 10.4 cells/μL versus 6.7 ± 4.2 cells/μL, P = 0.001), International normalized ratio (INR) (1.19 ± 0.2 versus 1.09 ± 0.26, P = 0.04), and dimerized plasmin fragment D (D-dimer) (2,788.2 ± 4,749.1 μg/mL versus 1,212.8 ± 2,300.9 μg/mL, P = 0.03). Platelet cell count (205,750 ± 888,855 platelets/μL versus 255,187 ± 145,525 platelets/μL, P = 0.01) was significantly lower in deceased patients. We found no significant differences for other laboratory test results (Table 4).

In the present study, we evaluated acute neurological complications and neurological disease progression in patients hospitalized with COVID-19. The 3 most common comorbid diseases accompanying neurological events were hypertension, coronary artery disease–heart failure, and diabetes mellitus; and several things could explain this association including increased prothrombotic risk, inflammation, or poor baseline cognition [5, 6]. Advanced age, increase in MCCI score, and prolonged duration of hospitalization were found to be associated with mortality.

We found an overall mortality rate of 33% for patients with COVID-19 who developed neurological complications. The mortality rates for Group 1 and Group 2 were 36.3% and 25%, respectively. In deceased patients, acute phase reactants (LDH, ferritin, procalcitonin, CRP, neutrophil), INR, and D-dimer were found to be significantly higher, and platelet cell count was found to be significantly lower, than in those who survived. Increased levels of acute-phase reactants, INR, D-dimer, and decreased platelet cell count may solely be due to COVID-19 infection, or the results of anticoagulant drug use, comorbid diseases, or secondary infections. Previous studies demonstrated that blood levels of D-dimer are higher in patients with neurological symptoms than patients without neurological symptoms [6, 17]. Platelet cell count and D-dimer levels were higher in patients with COVID-19 related pneumonia than in those with non-COVID-19 related pneumonia [18]. In accordance with the literature, D-dimer levels were also found higher in deceased patients, by contrast with the lower platelet cell count found in the present study.

Acute neurological findings in COVID-19 patients involve the CNS, PNS, and musculoskeletal system [6]. The most common neurological finding has been reported to be headache [17]. Although the incidence of headache varied between 11% and 14% in patients with COVID-19 [17], we found 8% of our patients had headache in the present study. This relatively lower incidence of headache in our study might be associated with the organization of medical treatment by the physician without consultation in hospitalized patients. We found that headache itself had no association with mortality.

Another important complication of CNS involvement is acute CVE (ischemic, hematoma, subarachnoid bleeding). In the present study, the incidence of an acute CVE in patients with COVID-19 was 23.3%. Incidence of CVE in COVID-19 patients has been reported as between 1% and 3% [6, 19, 20]. The increasing tendency toward coagulation downregulates natural anticoagulant mechanisms through inflammatory mediators; the reduction of the mortality rate by anticoagulant treatment supports the hypothesis of a deteriorated coagulation system [5, 21, 22].

Another complication detected in our study in CNS involvement was seizure. Generalized seizures have been reported in COVID-19 patients without any history of epilepsy [23]. The possible cause is triggering of neuronal hyperexcitability through receptor activation that depends upon the release of inflammatory cytokines, tumor necrosis factor α, and granulocyte colony-stimulating factor [23]. We detected a seizure rate of 16.6% in the present study. The incidence of seizures during COVID-19 has been reported as between 0.5% and 12.5% [24,25,26]. The higher rate of seizure in our present study might be related to the higher frequency of acute CVEs, as well as hypoxia and electrolyte imbalance. Detection of status epilepticus in 2 cases suggests that the virus may cause hyperexcitability during disease progress more than is currently considered. Because antiviral drugs such as lopinavir and ribavirin may trigger epilepsy, investigation of drugs used in case of epileptic seizure becomes important [27].

In the present study, the incidence of optic neuropathy among PNS complications was 1.3%. This incidence is consistent with reports of the incidence of visual loss as 1.4% [6]. Although many underlying causes may exist, inflammation and ischemia are considered possible causes for the development of optic neuropathy. Guillain Barré Syndrome (GBS) is also a PNS involvement together with optic neuropathy [28]. GBS was considered as the preliminary diagnosis in one patient; however, the diagnosis could not be confirmed as an electromyography (EMG) test could not be performed because of rapid deterioration of vital signs of the patient.

Mao et al. [6] identified skeletal muscle injury in the presence of myalgia and higher serum creatine kinase levels (>200 U/L) in musculoskeletal system injury during SARS-CoV-2 infection. Although the underlying mechanism of muscle injury was not clarified, increased proinflammatory cytokine-induced immune-mediated reaction was considered as the direct cause. These authors reported a rate of 19.3% for musculoskeletal system injury in patients with COVID-19. We observed a 2.6% rate of musculoskeletal system injury in the present study. The musculoskeletal system injury was found as a risk factor for mortality; however, the death of a 21-year-old postpartum patient without any recognized risk factor in the present study suggests that not all risk factors were revealed.

A major limitation of the current study is its retrospective design with a relatively low patient number. Common and mild symptoms during SARS-CoV-2 infection such as headache, dizziness, and loss of smell and taste could not be assessed. Furthermore, a neurological consultation may not have been requested because such symptoms may have been considered as a part of the natural progression of the disease. In the present study, the true incidence of symptoms and signs could not be revealed, because only patients hospitalized with COVID-19 and receiving treatment were evaluated. Neurological complications associated with COVID-19 such as transverse myelitis and sinus thrombosis were undetected in the present cross-sectional study, which might be a result of patients not being able to undergo further examinations because of their unstable condition.

COVID-19 may present with a range of neurological symptoms. Neurological findings can either develop acutely or as a progression of neurological disease. An ischemic CVE was the most common acute neurological event, while deterioration in consciousness was observed in those with the progression of neurological disease. Although the overall mortality in our Turkish patients with the disease was low, the rate was higher in patients who develop neurological complications, and in patients with advanced age, high MCCI score, and prolonged hospitalization duration as consistent with other findings [29]. Differences in neurological complication rates compared with the literature may be associated with regional and genetic differences. Further multicentered prospective studies with longer monitoring periods are needed to determine the underlying pathogenesis of neurological disease in Turkish populations and the neurological findings that appear due to COVID-19.