Our results showed that three sessions of extracorporeal hemoadsorption could improve the peripheral capillary oxygen saturation in six of the ten critically ill patients with COVID-19 disease. Mean SpO2 showed a significant improvement after the intervention.
Different aspects have been listed as the potential mechanisms of organ damage and disease severity in COVID-19 patients. One of the most important mechanisms is cytokine release syndrome (also known as cytokine storm) (17-19); and IL-6 has been represented as the most important causative cytokine in cytokine storm (10, 19). The COVID-19 disease progresses rapidly when the cytokine storm occurs and immune responses increase (17). Extracorporeal blood purification has been proposed as one of the treatment approaches to remove these inflammatory cytokines and could potentially be beneficial in patients with severe corona virus disease (7, 9-11). It seems that the improvement of peripheral capillary oxygen saturation during the blood purification is related to cytokine clearance rather than reduction in volume load; because during the CRRT procedure we adjusted the fluid outflow from the patient as the same amount of the fluid input. Cytokine removal following extracorporeal therapies could prevent cytokine-induced organ damages (10); and patients who underwent these therapeutic approaches in early phase of cytokine storm could have better clinical outcomes (11).
Interleukin-6 has been presented as a potential marker of disease severity in coronavirus infected patients. The upper limit point of serum IL-6 level in COVID-19 patients who had no severe pneumonia was reported 24.3 pg/mL, and the increased expression of IL-6 in serum is expected to predict the severity of the COVID-19 pneumonia and a poor prognosis of patients (20). C-reactive protein (CRP) is a biomarker which may increase at early stages of coronavirus disease, and higher value of this marker can be associated with more severe pulmonary lesions in these patients (21). In our research CRP showed a significant reduction after the intervention; and serum IL-6 decreased, although this reduction was not statistically significant. In addition to, in this study, the patients who expired showed a lower plasma lymphocytes count. Some previous studies demonstrated that COVID-19 patients with severe disease might have lower lymphocyte count compared to the mild ones (22, 23). As we mentioned above, extracorporeal blood purification treatment can effectively remove IL-6, IL-8, IL-1β, TNF-α and so on (13), however, due to the expenditures associated with measuring various inflammatory cytokines as well as the unavailability of some laboratory kits in initial and peak periods of COVID-19 epidemic in Iran, and considering the interleukin 6 as one of the most important inflammatory cytokines, only this cytokine has been measured in this research.
In this study half of the six patients who had a chronic underlying disorder improved after the intervention. In a research in China in which 1590 laboratory-confirmed hospitalized patients with COVID-19 were evaluated about the comorbidities, hypertension and diabetes mellitus have been reported as the most prevalent comorbidities; and nearly 8% of the individuals had two or more underlying disorders (22). A systematic review and meta-analysis reported hypertension, cardiovascular diseases, diabetes mellitus, smoking, chronic obstructive pulmonary disease, malignancy, and chronic kidney disease as the most prevalent underlying diseases among hospitalized COVID-19 patients (24). These risk factors might compromise and deteriorate the patients' clinical outcome. Based on our results, it seems that hemoperfusion can be beneficial in management of fluid overload, metabolic disorder, and cardiovascular dysfunction, besides to reduction of inflammatory mediators; as was mentioned in previous studies (25).
Cartridges which are used in hemoperfusion process are divided in selective and non-selective types. The Jafron resin hemoperfusion cartridges are classified as non-selective group. These cartridges are different based on the pore size distribution which determines their cutoff points for adsorption of different materials, and makes them applicable for different clinical outputs; for example HA-130 was used for improvement of uremic symptoms in chronic hemodialysis and HA-330 was effective on modulation of severe inflammatory processes (12). We used HA-280 and HA-230 cartridges because we did not have access to other types like cytosorb, Jafron HA-380, and HA-330 in our country. There is a limited scientific evidence about the application of HA-280 and HA-230 cartridges in clinical settings (26, 27). Previous study in which the efficacy of HA-330 resin-directed hemoperfusion has been assessed in acute respiratory distress syndrome, some considerable treatment outcomes including improved oxygenation, reduction in lung edema and histopathological signs of acute respiratory distress syndrome, and reduced circulating and alveolar cytokine levels have been resulted; and the authors concluded that this cartridge could beneficially influence the course of acute respiratory distress syndrome by attenuating systemic and pulmonary inflammatory cytokines (12). As the cartridges we used in the process of CRRT can act as non-selective ones to absorb inflammatory cytokines such as IL-6, by conducting hemoperfusion with a mode of CVVH, more cytokines are expected to be absorbed compared to CVVH only.
Randomized trial data about the effectiveness of hemoperfusion in COVID-19 patients is lacking, however, evidence shows that this therapeutic approach is tolerable to most patients if conducted with the assistance of nephrology specialists, in order to minimize risks of infection and bleeding (11).
The strength points of this study were well-defined condition of the patients and team-working of a multidisciplinary group of experts to conduct the intervention. The most important limitations of this research were absence of a control group, small sample size of the study population, and not to present fluid balance measures in patients. Some evidence recommends PaO2/FIO2 (P/F), rather than oxygen saturation, as the best marker of oxygenation in patients with acute respiratory distress syndrome (28); since this study was performed during the first weeks of the onset of COVID-19 epidemic, when a large number of patients were hospitalized in the state hospital, this variable was not measured.
In addition to, it is recommended to compare the impact of hemoperfusion plus CRRT on the removal of inflammatory cytokines with the effect of hemoperfusion alone in future studies. Large prospective multicenter trials in carefully selected patients are needed to definitely evaluate the efficacy of hemoadsorption in COVID-19 patients.