Complex Interplay of COVID-19 ARDS with Guillain-Barré Syndrome and Cerebral Infarction: A Case Study

Introduction

Coronavirus disease (COVID-19) can cause various complications. Neurological manifestations of COVID-19 have been reported in up to 36.4% of cases [1]. COVID-19 has been reported to be associated with various neurological symptoms, including dizziness, headache, confusion, myalgia, ageusia, and anosmia [1]. Several neurologic illnesses, such as encephalitis/encephalopathy, meningitis, and stroke, have reportedly occurred along with COVID-19 [1]. We report a case of severe COVID-19 acute respiratory distress syndrome (ARDS) complicated by neurological manifestations, including Guillain-Barré syndrome (GBS) and cerebral infarction, as well as pulmonary embolism (PE).

Case Report

A 55-year-old Japanese man with a history of ulcerative colitis (UC), hyperlipidemia, and hypertension was admitted to our hospital with dyspnea after 11 days of fever. He tested positive for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA on reverse transcriptase polymerase chain reaction (RT-PCR) and was diagnosed with COVID-19. The patient’s C-reactive protein level was 14.14 mg/dL. He was treated with ciclesonide, favipiravir, and methylprednisolone for COVID-19. His respiratory status worsened, and he developed ARDS, requiring intubation on hospital day (HD) 3. He received midazolam, propofol, and fentanyl for analog-sedation. On HD 8, he was able to grip with his hands and elevate his arms. Then, his respiratory condition worsened significantly. On HD 12, he developed a right pneumothorax, and a chest drain was inserted. He experienced anaphylactic shock, presumably due to xylocaine, during the insertion of a chest drain. Despite the insertion of a chest drain, oxygenation deteriorated. Therefore, a muscle relaxant, rocuronium, was added to his regimen to reduce breathing effort. Simultaneously, the dosage of propofol was increased, and on HD 14, his level of consciousness was GCS 1T1. On HD 16, contrast computed tomography revealed PE, and by HD 18, there was an added deterioration in respiratory status, presumably due to this PE, necessitating the initiation of veno-venous extracorporeal membrane oxygenation (V-V ECMO) in conjunction with ARDS. Rocuronium was temporarily resumed during the procedure. On HD 20, he exhibited progression of metabolic acidosis, and continuous hemodiafiltration (CHDF) was initiated. His respiratory status subsequently improved, and on HD 22, tracheostomy was performed. On HD 23, V-V ECMO was successfully discontinued. The doses of sedative agents were gradually decreased, and sedatives were then discontinued on HD 24 (Figure 1). On HD 44, he was able to follow commands with visual pursuit and nodding but was unable to move his extremities. The patient regained consciousness and was able to communicate by sticking out his tongue. He was noted to be completely quadriplegic. Magnetic resonance imaging (MRI) showed bilateral scattered cerebral infarctions (small high-intensity areas in the left and right precentral gyri, right superior frontal gyrus, and right occipital lobe) on HD 76 (Figure 2). We confirmed the absence of intracardiac thrombi on echocardiography and conducted a bubble study, which also confirmed the absence of a patent foramen ovale. The heparin therapy for PE, which had been discontinued on HD 64 due to anemia, was resumed. However, this did not explain the patient’s quadriplegia. Neurological examination revealed amyotrophy of all limbs, total paralysis of the extremities and the bulbar and facial muscles, and external ophthalmoplegia. Deep-tendon reflexes were absent in all limbs. Babinski reflex was absent. These findings were symmetrical. Cranial nerve involvement, specifically bilateral facial nerve weakness, was noted. A nerve conduction study showed delayed peripheral nerve conduction velocity, disappearance of the F wave, and decreased amplitude in the left median nerve. These findings were indicative of severe axonal neuropathy. Cerebrospinal fluid showed albuminocytologic dissociation, no white blood cells (0/mm3), and elevated protein levels (124 mg/dL). Blood serum testing showed strong positivity for IgG antibodies to the ganglioside GD1a. Antibodies to the gangliosides GQ1b and GM1 were not detected. These findings were consistent with the diagnosis of GBS. He received intravenous immunoglobulin (IVIG) (0.4 g/kg) for 5 days, starting on HD 89. His neurological symptoms slightly improved after therapy, and he was able to move both shoulders and his right hand (manual muscle testing, 2/5). On HD 95, the patient developed interstitial nephritis, presumably due to mesalazine.

Discussion

In this case, it was difficult to determine the onset of GBS because the neurological findings were difficult to evaluate in an isolated intensive care setting. Early detection of neurological complications is imperative to providing immediate intervention and achieving a better prognosis, but it is difficult to detect the initial symptoms of COVID-19-associated GBS. The initial symptoms included flaccid paralysis and facial diplegia, occurring 5–10 days after the onset of acute respiratory symptoms [2]. The initial symptoms of neurological complications in severe COVID-19 are challenging to diagnose because the patients are isolated, making neurological evaluation difficult, especially for intubated patients.

It was also difficult to diagnose the neurological complications of COVID-19, including GBS and stroke, because MRI and neurophysiological tests, including electroencephalography and nerve conduction studies, are difficult to perform in an intensive care setting. Moreover, safe nursing and adequate infection control need to be practiced.

Furthermore, it was difficult to differentiate GBS from other neurological diseases. Our patient was receiving V-V ECMO and was sedated with several drugs, including a muscle relaxant. General weakness was challenging to differentiate from ICU-acquired weakness (ICU-AW) and other neurological diseases. Muscle injury and myalgia are commonly reported neurological findings in COVID-19 (19.2%, 95% CI 15.4–23.2%) [3]. The weakness attributed to cerebral infarction on MRI was considered in the differential diagnosis. However, both pyramidal signs, including absent tendon reflexes and absent Babinski reflex, were noted, leading us to suspect a peripheral rather than central origin. Critical illness polyneuropathy (CIP) was one of the essential differential diagnoses in this patient. However, in CIP, peripheral facial paralysis is uncommon, and antiganglioside antibodies are not detected [4]. To differentiate GBS from other neurological diseases, antiganglioside antibody testing was useful in confirming the diagnosis of GBS, especially for an isolated intensive care setting because antiganglioside antibodies can be collected without moving, and are strongly associated with certain forms of GBS [5].

In our case, COVID-19 was complicated by GBS, cerebral infarction, and PE. These complications can be associated with cytokines. SARS-CoV-2 can induce an excessive immune reaction with increases production of cytokines, such as IL-6, by activated leukocytes. These stimulate the inflammatory cascade, leading to extensive tissue damage [6]. In this case, the patient developed respiratory failure, requiring V-V ECMO. In this state, a hyperinflammatory cytokine profile, often termed a “cytokine storm,” can be fatal. Cytokine storm is one of the clinical complications of SARS-CoV-2 infection [6]. In patients with GBS with mechanical ventilation, our patient required V-V ECMO (Table 1) [7–17]. In this case, the patient`s respiratory condition deteriorated to the point that V-V ECMO was required due to pulmonary embolism complicating ARDS caused by COVID-19. It is conceivable that this PE, similar to GBS and cerebral infarction, could have originated from a cytokine storm induced by COVID-19.

Cytokine storm was also associated with increased IL-6 levels, resulting in hyperviscosity and increased risk of stroke [6]. Hypercoagulability was also a possible cause of PE. In our case, the patient had COVID-19 complicated by stroke and PE. This case emphasized that multiple coagulation complications can occur sequentially in COVID-19.

Conclusions

Neurological complications of severe COVID-19 ARDS are challenging to diagnose because patients are deeply sedated, and neurological symptoms need to be distinguished from ICU-AW. Neurologic examinations in COVID-19 patients are also limited by isolation guidelines. Early detection and treatment of neurological complications are essential to achieve a good prognosis. This case emphasizes that neurological complications are not uncommon and are prone to delays in detection. When evaluating patients with COVID-19, neurological complications, including GBS and stroke, should be strongly suspected.