Early Hemoperfusion for Cytokine Removal May Contribute to Prevention of Intubation in Patients Infected with COVID-19

As with previous viral outbreaks, a percentage of patients with COVID-19 require intensive care and complex management [1, 2]. Current strategies for coronavirus disease 2019 (COVID-19) include extracorporeal membrane oxygenation (ECMO) in the most severe cases of acute respiratory distress syndrome (ARDS) [3], as well as high-flow nasal oxygen (HFNO) and mechanical ventilation for intubated patients [4]. Although these measures are effective in most cases, several COVID-19 patients may present a fatal outcome. Reports from around the world differ due to the variable intensive care unit (ICU) policy for intubation, mechanical ventilation, and application of ECMO. Yang et al. [5] reported in their study in China that nearly 94% of patients with COVID-19 admitted to ICUs and undergoing mechanical ventilation die. Furthermore, besides such a low chance of recovery hospitals may also experience a shortage of ventilator devices, a problem that some countries are still facing to some extent. Given the high mortality rate of patients requiring mechanical ventilation, all efforts should be made to prevent disease progression and need for intubation. We hypothesize that early application of hemoperfusion (HP), either isolated or in combination with continuous renal replacement therapy (CRRT) or hemodialysis (HD), may contribute to preventing the incidence and progression of ARDS, acute kidney injury (AKI), liver failure, and septic shock by removing cytokines, chemokines, interleukins, and other inflammatory mediators. This approach should allow a degree of immunomodulation, ultimately preventing or slowing down the progression to multiple organ failure [6] and the need for invasive ventilation.

A 54-year-old male with cough and dyspnea was referred to the emergency department 2 days ago. The patient received antiviral (chloroquine, coltra, and siltamivir), antibacterial (meropenem and vancomycin), and anticoagulant (heparin 10 U/kg/h) drugs. Due to the history of hypertension and diabetes mellitus, the patient also received antihypertensive and diabetes medications.

The patient treated in the present report was diagnosed with COVID-19 for shortness of breath, a decrease in O₂ saturation (80% with the reservoir bag mask), and transferred to the ICU. The patient had oliguria and a high level of BUN and creatinine before ICU admission (creatinine 2.7 and BUN 70). Our most important clinical parameters for patient monitoring were as follows: the clinical condition of the patient, respiratory rate, respiratory distress, O₂Sat, FiO₂, PaO₂, Pf ratio, IL6 and chest X-ray. Figure 1 shows the patient’s chest X-ray at ICU admission. The patient was heavily dependent on oxygen O₂ saturation, and it was rapidly decreasing to <70% upon mask removal. HP (HA280 cartridge, Jafron Biomedical Co., Zhuhai, China) in combination with CRRT was initiated. Reasons for starting CRRT include kidney function: before the onset of HP-CRRT, the patient has oliguria (urine <400 cm³ in 24 h). After HP-CRRT, the patient’s urine output reached 1,100 cm³ in 24 h. The patient had acute renal failure, and the patient had a history of diabetes and high blood pressure. Other reasons for using CRRT with HP were as follows: continuous veno-venous hemofiltration (CVVH) mode with a convection mechanism has the ability to remove middle-molecule toxins and helps to remove cytokines in the interval between HP sessions. Numerous articles have been shown to increase the performance of HP and CRRT when combined. In addition to kidney damage, septic shock and multiple organs failure can be mentioned as possible complications of patients with corona disease. CRRT prevents septic shock and multiple organ failure by removing toxins and middle-molecule cytokines. CRRT mode used was as follows: CVVH predilution and postdilution every 2 h, blood flow: 200–250 mL/min, substitution flow: 25 cm³/kg/h, UF rate: 20 cm³/h, heparin: 10 U/kg/h, patient weight: 70 kg. The HP cartridge was added to the CRRT circuit simultaneously with the start of CRRT, and HP and CRRT were started simultaneously. After 6 h (saturation time), the HP cartridge was removed from the CRRT circuit, and CRRT was continued. After 20 h, the second HP cartridge was added to the CRRT circuit, and it was used for 6 h and then removed. The fluid balance was maintained neutral. The O₂ saturation of the patient gradually increased after the first hour of HP/CRRT, reaching 95 and 100% after several hours. Figure 2 shows the patient’s chest X-ray after the second HP cartridge application. After 24 h, the patient’s clinical condition was significantly improved. CRRT was then stopped. After about 10 h from discontinuation, O₂ saturation gradually declined and after 20 h reached about 90%, and CRRT/HP was started again. Two HP sessions were performed within the following 24 h in conjunction with CRRT, and the O₂ saturation climbed above 95% again. CRRT continued for another 12 h, and the O₂ saturation was still above 95% using a mask with a reservoir bag. The patient’s creatinine and BUN levels also decreased significantly (creatinine 0.7 and BUN 30). The patient’s dependence on oxygen decreased, and he could remove the mask for a few minutes. The patient was monitored for loss of O₂ saturation, but O₂ saturation had stabilized, and there was no decrease in the following 24 h. After 5 days, the patient was finally discharged from the ICU. Figure 3 shows the patient’s chest X-ray at that time. Figure 4 shows the trend of the patient’s hemoglobin O₂ saturation during the treatment period from admission to ICU until discharge. Table 1 shows the comparison of inflammatory parameters and biochemical parameters before and after HP-CRRT.

ARDS is the most common cause of intubation in patients with COVID-19 and admission in the ICUs. Subsequently, septic shock, elevated liver enzymes and renal markers, acute hepatic and renal failure, and multiple organ failure occurred and resulted in death for the patient. Cytokine storm is addressed as one of the contributing factors to ARDS. Applying HP/CRRT with a mechanism of adsorption appears to capture and harvest cytokines from the blood, prevents them from lying on the wall of the alveoli and pulmonary arteries, and ultimately prevents the incidence of ARDS and/or its progress [7-10]. Thus, we consider that there is a rationale for early application of HP/CRRT before the patient’s clinical condition becomes so severe to require invasive mechanical ventilation. This is especially true in the absence of pharmacological remedies for the COVID-19 infection. Indications and duration of HP/CRRT should respond to specific criteria in order to be consistent in the application of this rescue treatment. High level of inflammatory markers and cytokines, the severe tendency to hypoxia, clinical signs of hemodynamic instability, and need for vasopressor support may represent a trigger for early application of HP/CRRT. Further studies are needed to confirm this hypothesis, but recent data are promising. In conclusion, with clinical experiences we have shown that the application of CRRT/HP in the early stages of ARDS, when the O₂ saturation of the patient’s blood with a reservoir oxygen mask is <90%, prevented the progression of the disease to moderate-to-severe ARDS, stabilized the O₂ saturation, and gradually led to increased O₂ saturation, prevention of intubation, improved clinical conditions, reduced dependence on oxygen, discharge from ICUs, and ultimately discharge from the hospital.

We would like to thank physicians and nurses of Qom Kamkar Hospital for their support.

The authors have no ethical conflicts to disclose. The study was conducted in accordance with policies and procedures approved by the local institution review board. The patient, described in this case report, has given his written consent to publish data and images.

CR consulted or advised in the last 3 years for ASAHI, Astute, Baxter, Biomerieux, B. Braun, Cytosorbents, ESTOR, FMC, GE, Jafron, Medtronic, and Toray. The other authors have no conflicts of interest.

The authors did not receive any funding.

A.E.V. and M.H.P. performed HP/CRRT for the patient and were involved in data collection and writing and editing of the manuscript. C.R. was involved in writing and editing of the manuscript and gave final approval for publishing. M.G. performed literature search and was involved in data collection and writing and editing of the manuscript. M.M. and H.R. were involved in writing and editing of the manuscript.