HIGH-FLOW NASAL CANNULA COMBINED WITH PRONE POSITIONING AS AN INTUBATION ALTERNATIVE IN COVID-19: A CASE REPORT AND REVIEW OF THE LITERATURE

SUMMARY

The use of high-flow nasal cannula (HFNC) in COVID-19 patients is a controversial topic due to the benefits and risks which may occur in patients and healthcare workers. The goal of this treatment modality is potential avoidance of invasive mechanical ventilation, but generation of aerosol and increased healthcare professional infection risk must be considered. We present a case of a SARS-CoV-2-positive 71-year-old male with acute hypoxemic respiratory failure, who was successfully treated with HFNC combined with prone positioning. Furthermore, we discuss recent literature concerning potential issues of HFNC treatment in COVID-19 patients.

Introduction

The use of high-flow nasal cannula (HFNC) to provide oxygen supplementation and improve the ventilation-perfusion ratio is a well-established treatment modality in patients with hypoxemic respiratory failure (HRF) (1). However, its use in COVID-19 patients is discouraged due to the potential aerosol spread and increased infection risk (2). We present a patient with borderline moderate to severe HRF in whom invasive mechanical ventilation was avoided by combined use of HFNC and prone positioning.

Case Report

In April 2020, a 71-year-old male patient with arterial hypertension and mild obesity (body mass index 27.5 kg/m²) was admitted to the Intensive Care Unit (ICU) in primary respiratory intensive care center specialized for COVID-19 patients (PRIC). Eight days prior to the onset of symptoms, he was in contact with a friend who was SARS-CoV-2 positive. On day 7 after symptom onset and positive SARS-CoV-2 real-time polymerase chain reaction test, he was admitted to PRIC. Symptoms and signs included fatigue, muscle pain, fever (up to 39 ∞C), non-productive cough, and anosmia. Two days prior to hospital admission, he developed mild dyspnea and occasional chest pressure. Physical examination at admission showed mild respiratory distress with the following vital signs: axillary temperature 37.2 °C, heart rate 86/min, blood pressure 170/90 mm Hg, respiratory rate 20/min, peripheral oxygen saturation (SpO₂) 84% on room air in upright position and 90% in lateral decubital or prone position. Arterial blood gas analysis showed moderate hypoxemia and mild hypocapnia (PaO₂ 6.5 kPa, PaCO₂ 4.4 kPa) with normal base deficit and lactate levels. Chest x-ray showed discrete inhomogeneous opacities in the right lower lobe (Fig. 1). Immediately after admission and blood sampling, the patient received 4 L/min of O₂ through facemask, which resulted in fast relief of dyspnea with improvement of SpO₂ to 98% and PaO₂ to 10.8 kPa. According to the hospital protocol, chloroquine phosphate (total of 5 days) and azithromycin (total of 3 days) were prescribed. QTc interval and liver enzymes were regularly checked given the possible side effects of the previously mentioned drugs. After day 4 of ICU stay, the patient showed fatigue during mild physical exertion (for example, going to the bathroom) with SpO₂ 82%. Repeated chest x-ray showed worsening of bilateral infiltrates and small pleural effusion on the left side. From that moment, the patient was encouraged to lie in a prone position for several hours through day and night, which led to relief of symptoms and improvement of SpO₂.

Fig. 1.

Chest x-ray at admission.

Nevertheless, on day 6 of ICU stay, his state worsened further with aggravation of hypoxemia (with 6 L/min of O₂ through facemask SpO₂ fell to 82%, PaO₂ to 6.2 kPa, and PaO₂/FiO₂ 105 mm Hg). Disparity between the clinical condition (primarily in the severity of dyspnea) and oxygen blood levels was present, hence the patient was not in severe respiratory distress nor did he complain of shortness of breath to the extent it would be expected. Also, further progressive changes in radiographic findings were recorded. Figure 2 demonstrates extensive bilateral infiltrates affecting the entire lung with predominance in the lower and medial lung fields. Computed tomography (CT) showed subpleural perihilar diffuse alveolar consolidations, ground-glass opacities, and intra- and interlobular septal thickening as common CT features in COVID-19.

Fig. 2.

Chest x-ray on day 6 of Intensive Care Unit stay.

Considering worsening of hypoxemia and chest radiograph, HFNC (Hi-Flow Star, Dräger, Lübeck, Germany) was initiated, with FiO₂ 1.0, flow 35 L/min and temperature of 37 ∞C as initial settings, which were subsequently adjusted to the aimed SpO₂ >93%. When necessary, maximum HFNC flow was utilized (50 L/min). Additionally, surgical mask was used over the nose and mouth to reduce the risk of aerosol spread. The patient reported alleviated breathing approximately 30 minutes after starting HFNC, accompanied with gradual rise of SpO₂ up to 92% and PaO₂ to 8.7 kPa, which were even more pronounced in prone position (SpO₂ up to 94%). Due to respiratory function impairment and worsening of gas exchange, intravenous methylprednisolone was applied (1 mg/kg for 3 days, tapered to 0.5 mg/kg for the next 4 days). Laboratory findings reflected elevated systemic inflammation markers (white blood cell count, serum C-reactive protein, procalcitonin, ferritin and D-dimer). Lung ultrasound was regularly performed to reduce ionizing radiation exposure, and a decrease of B-line density was recorded over time. The patient was hemodynamically stable at all times without significant signs of other organ damage. Restrictive fluid administration regimen and daily respiratory therapy were implemented.

High-flow nasal cannula was applied for a total of 15 days, and the patient was gradually weaned to facemask and subsequently to room air. The basic principle we used in weaning HFNC therapy was firstly gradual reduction of FiO₂ (approximately to 0.3-0.35) and if the respiratory rate and SpO₂ were satisfactory, gradual reduction of flow was performed (3). In total, supplemental oxygen therapy was needed for a total of 40 days. At 22 days of admission, nasopharyngeal swabs were confirmed negative for SARS-CoV-2. The patient was discharged home 46 days after admission fully recovered with complete regression of pulmonary infiltrates, normal parameters of gas exchange and other laboratory findings.

Discussion

Endotracheal intubation and mechanical ventilation are still the gold standard in the treatment of acute HRF but it is associated with a high complication rate and increased mortality in COVID-19 patients (4). There is a hypothesis that certain patients with COVID-19 have presented a new phenotype that differs from clinical features previously seen in hypoxemic patients. In these patients, respiratory mechanics are preserved, and hypoxemia may be attributed to pulmonary thromboembolism and loss of hypoxic pulmonary vasoconstriction (5, 6). It may be reversed by a combination of high FiO₂, positive airway pressure and reduced work of breathing created by HFNC coupled with decrease of intrapulmonary shunting and increase of lung compliance provided by prone positioning (7, 8). Simultaneous use of these two treatment modalities enabled avoidance of invasive mechanical ventilation in this patient. As shown in a study from China, in COVID-19 patients with PaO₂/FIO₂ <300 mm Hg combining prone position with HFNC significantly improved ventilation-perfusion mismatch and proved to be a good strategy in avoiding intubation and reducing the medical staff workload (9). It must also be noted that, although recently published data suggest that stress ulcer prophylaxis might increase the risk of bacterial superinfection (10), in this patient it was not the case and the collected sputum specimens showed no pathogen growth.

As COVID-19 pandemic erupted in early 2020, a main concern was limiting the spread of this highly virulent virus. Because of that, the use of HFNC was observed with skepticism and was initially discouraged in these patients (2, 11). However, both experimental and clinical published data showed HFNC to be safe to use when all the precautionary measures are employed (12) (confirmed by the fact that none of the healthcare personnel that treated this patient were infected with SARS-CoV-2) and may greatly reduce the intubation and complications rate, as well as ICU stay (13).