Critically Ill Patients with COVID-19 Pneumonia Requiring Continuous Renal Replacement Therapy with oXiris^® Membrane in a Third-Level Hospital in Northeast Mexico Free

The coronavirus disease 2019 (COVID-19) pandemic represented a global public health problem with devastating consequences that have challenged conventional medical treatments. It has tested all the knowledge and experience of every health personnel; in nephrology, this is not an exception. The development of acute kidney injury (AKI) in previously healthy patients can range from 0.5 to 29% in the intensive care unit (ICU), and approximately 5–10% of ICU patients with AKI from any etiology require RRT. The mortality of these patients ranges from 30 to 70%. The main indication for continuous renal replacement therapy (CRRT) in patients with AKI KDIGO 3 who require renal replacement therapy is hemodynamic instability [1]. CRRT presents different modalities for its application and allows the use of different membranes and cartridges depending on the patient requirements. The pathophysiological understanding of each cartridge has generated new strategies for controlling inflammatory cytokines. For these reasons, CRRT has earned a place in the ICU during the SARS-CoV-2 pandemic [1].

COVID-19 manifests in its most severe form as respiratory distress syndrome, septic shock, metabolic acidemia, hypercoagulability, and multiple organ failure. This represents a high mortality rate [1]. Unlike the MERS virus, where the severe disease was less than 10%, patients with COVID-19 had a severe disease in 13.8%, of which 4.7% became critically ill [2], defined as those with respiratory failure requiring mechanical ventilation, shock, or any other organ failure that requires ICU care [3]. In severe SARS-CoV-2, the main cause of death is septic shock induced by a cytokines storm such as TNF-alpha, IL-1 beta, IL-2, IL-6, IFN-alpha, and MCP-1 [4]. Because more than 67% of patients in the ICU with COVID-19 have multiple organ failure, providing adequate organ support and blocking or eliminating cytokine storm is an essential part of treatment [5]. In this scenario, a decrease in mortality has been reported in patients undergoing CRRT using the oXiris membrane because it helps control the cytokine storm.

Blood purification removes inflammatory factors and blocks the cytokine storm, thereby reducing inflammatory damage [3]. oXiris is a modified AN69 membrane designed by Baxter International and was approved by the US Food and Drug Administration (FDA) in April 2020 under emergency use authorization to treat patients with COVID-19. The positive charge on its surface gives it the ability to adsorb endotoxins and cytokines [6]. The presence of three different layers in this device gives it three important properties: renal replacement therapy, elimination of cytokines and endotoxins, and antithrombogenic properties [7]. In addition to the adsorption mechanism, this membrane can remove solutes through convection and diffusion [8].

Multiple studies have proven the efficacy of the modified AN69 membrane for the management of patients with sepsis due to COVID-19. Lobo et al. [9] reported an improvement in respiratory failure and encephalopathy after hemodiafiltration using oXiris. There was a reduction in IL-6, and serum creatinine, indicating adequate control of inflammation and metabolic clearance [9]. Ma et al. [10] reported a significant decrease in inflammatory markers in 3 patients with COVID-19 treated with plasma exchange and modified AN69 membrane in a hospital in Wuhan, China. While Zang et al. [11] in an observation study reported that the clinical effectiveness of AN69-oXiris remains unclear compared with AN69-ST filters on cytokine levels and clinical improvement in septic patients. In addition, the patients who used AN69-oXiris had a more remarkable improvement in hemodynamic status and lower cytokine levels; they showed no differences in clinical outcomes. A prothrombotic state is seen in severe COVID-19, this requires incremental doses of anticoagulation both to maintain circuit patency and to control the patient’s thrombophilia [11].

The following study was conducted to describe our experience with the modified AN69 membrane in patients with COVID-19 in the ICU of a third-level hospital in northeast Mexico. Because the need for data for its use in developing countries is limited, this study is considered of great importance.

An observational, retrospective, and analytical study was conducted with 13 patients. We included adult patients (aged >18 years) who were admitted to the ICU with a diagnosis of COVID-19 by real-time PCR test and AKI who required CRRT with modified AN69 membrane between January 2020 and August 2021. All patients who met the inclusion criteria with confirmed COVID-19 and elevated inflammatory markers, such as PCR, were provided with a modified AN69 membrane. All patients were previously treated with antimicrobials and corticosteroids based on their bacterial superinfection, ICU stay, and immunosuppression condition before starting CRRT. For the exclusion criteria, patients who did not meet all the previous data mention, patients who did not receive AN69 membrane, those who did not have a complete medical file, or more than 20% of data missing, or patients with AKI KDIGO 3 that require CRRT that did not have COVID-19 positive test were excluded. The primary outcomes were renal recovery and survival, and the secondary outcomes were the decrease in vasopressors and hemodynamic changes in parameters.

From the population included, they had at least 15 days hospitalized and at least another 15 days in the ICU. Sociodemographic data were obtained, and the total hospital and ICU stay, duration and specifications of CRRT, and its outcomes were analyzed. Continuous venovenous hemodiafiltration was performed using Prismaflex, Baxter. Blood flow rates were maintained between 100 and 200 mL/min, while prescribed dialysis doses were in the range of 25–35 mL/kg/h. We use heparin for anticoagulation in all patients. By the time CRRT was required, the patients had multiple organ dysfunction, with a Sequential Organ Failure Assessment (SOFA) >2 points. CPR (mg/dL), norepinephrine (μg/kg/min), and creatinine (mg/dL) were measured at the initiation and at 24 and 48 h post-CRRT.

Patient data were deidentified for statistical analysis. Frequencies and percentages were used to describe categorical variables. The normality of the numerical variables was assessed using the Shapiro-Wilk test. Variables with normal distribution are described using the mean and SD. Variables with nonnormal distribution were described using the median and the 25th and 75th percentiles (p25–p75). Categorical variables were compared using Pearson’s χ² test. For quantitative variables, groups were compared using the T test. Data were analyzed using SPSS version 25 (IBM, New York, NY, USA). A p value less than or equal to 0.05 was taken as the cut-off point for statistical significance.

Thirteen patients were included (Fig. 1), of whom 10 were men (77%). The mean age was 59.4 ± 12.9 years, and all patients were in mechanical ventilation at the time of CRRT. The most frequent comorbidities were arterial hypertension (53.8%) and type 2 diabetes mellitus (38.4%) (Table 1). The mean hospital stay days were 60.3 ± 44.9 and 45.8 ± 30 days in the ICU. The mean duration of CRRT was 8 days (3–11) as described and shown in Table 1, for both groups (survival with renal recovery or intermittent RRT and non-survival). The filters were changed either after clotting or 72 h of use, and the filter thrombosis mean rate was 55 h. The main indications for CRRT were anuria (61.5%), fluid overload (23%), and uremia (15.4%). None of the patients at the time of AKI were treated with Remdesivir or Baricitinib. Seven patients (53%) were treated with diuretics, all of them with furosemide, whereas 15% (2 of the 7 patients) were also treated with spironolactone. All patients received corticosteroids. All the patients were receiving antibiotic therapy based on what we described before, while 46% of them were receiving more than one antibiotic, all of which were individualized and based on antibiograms and bacterial superinfections. Of the total population, 3 (23%) recovered kidney function and survived, 2 (15.5%) survived and were discharged with intermittent hemodialysis, and 8 (61.5%) did not survive (Fig. 2). The initial PCR was mean 140 mg/L, whereas the mean PCR after 24 h of CRRT was 127 mg/L (Table 2). In the first 24 h after the initiation of CRRT with modified AN69 membrane, the vasopressor requirements (noradrenaline) decreased 12%, in addition to the decrease in creatinine and other inflammatory biomarkers, as shown in Table 2. The primary outcome was observed in 23% of the patients.

Among the possible mechanisms for developing AKI in COVID-19, the most important is the direct damage caused by the systemic inflammatory response. In its most severe expression, the cytokine storm, hypoxia, and crosstalk between organs, cardiorenal, hepatorenal, pneumorenal, and neurorenal, among others, either individually or in combination, lead to multiple organ failure. In this scenario, ECMO occurs, and in a more novel way, through the use of filters and cartridges with specialized characteristics that may participate in a septic patient or with acute respiratory distress syndrome. The need for RRT in patients with sepsis and AKI tends to occur more frequently in the setting of hemodynamic instability and fluid overload; therefore, if available, CRRT would be the first option of treatment [9].

In an in vitro study, the removal of sepsis-associated cytokines and mediators from oXiris, CytoSorb^®, and Toraymyxin^® resulted in a mean adsorption clearance of approximately 20 mL/min in the first 30 min with Toraymyxin versus 8 mL/min with oXiris (p < 0.05) and 1 mL/min with CytoSorb. At 120 min, there were no differences in endotoxin clearance between Toraymyxin and oXiris. The clearance rate of pro-inflammatory cytokines and other mediators was higher with the modified AN69 membrane and CytoSorb than with Toraymyxin. oXiris was the only device that showed endotoxin and cytokine removal through an ionic adsorption mechanism. Prior studies have shown IL-6 reduction rates of up to 70–93% after 2 h of extracorporeal treatment of septic patients with oXiris and an important reduction in vasopressor use [12].

The presentation of AKI in the evolution of the severity of this pathology is not unexpected since it has been shown that the risk factors for COVID-19 are the same as those for AKI: high blood pressure, with a frequency of 15–33%; diabetes mellitus from 7.4 to 20%; cardiovascular disease 40%; obesity; advanced age, any other known chronic disease; immunosuppression; or chronic inflammation [9]. A prospective, multicenter, observational study was conducted on data collected from the oXirisNet Registry in Italy of patients with COVID-19 in the ICU who received treatment with this membrane for immunomodulation and/or support during AKI. Data from 37 patients with a mean age of 59.5 ± 9.5 were analyzed. The median baseline IL-6 level was 1,230 pg/mL, which decreased during the first 72 h of treatment and was more significant during the first 24 h. Premature coagulation was reported in 18.9% of patients [13]. Cecci et al. [14] reported a decrease in the use of vasopressors as well as a decrease in the Sequential Organ Failure Assessment score with the use of modified AN69 membrane in patients with multiple organ failure [14]. Lumblergul and Srisawat [15] reported in 2021 a decrease in the use of vasopressor (norepinephrine) of 45.9%, whereas an increase in MAP by 6.1% with the use of an oXiris haemofilter into the CRRT circuit of Prismaflex by Baxter in 35 patients enrolled in the study with septic shock.

Zhang et al. [1] analyzed 5 patients with a mean age of 70.2 ± 19.6 years in a hospital in China with COVID-19. They underwent continuous venovenous hemodiafiltration using an oXiris membrane. The average treatment time was 172.8 ± 82.1 h. After treatment, a statistically significant decrease in IL-1 beta, IL-6, IL-10, and C-reactive protein levels was observed [1].

In our study, despite mortality, in patients who used the modified AN69 membrane (Fig. 3), there was an important decrease in inflammatory biomarkers as well as a decrease in vasopressor requirement. Renal function recovery was observed in 23% of the patients.

The limitations of this study were that some biomarkers had more than 20% missing data during the ICU stay and before the use of CRRT in all the enrolled patients (IL-6, procalcitonin), so they were eliminated, and the number of patients that met the inclusion criteria. We also recognize another limitation in measuring severity scores as SOFA after the use of the modified AN69 membrane.

The use of the oXiris membrane in patients with severe COVID-19 resulted in hemodynamic improvement, with 23% of patients achieving complete renal recovery as well as a decrease of the requirement of vasopressor in the first 24 h of use, accompanied by a decrease in inflammatory vasopressor requirement and biomarkers. As for mortality exceeding 50%, this has been shown in previous studies, and we consider that this outcome is dependent on multiple comorbidities and clinical situations. Therefore, its application should continue to be investigated. The literature on the benefit of CRRT with the modified AN69 membrane is based on pathophysiological data and clinical trials that show evidence of clinical improvement and better laboratory parameters in patients with COVID-19. However, in Mexico, studies in this regard are limited, and our research provides valuable information that may be used in the future.

We thank the Hospital Christus Muguerza Alta Especialidad and the biostatistics department for their assistance with all clinical data files.

This study was conducted in accordance with the Declaration of Helsinki of the World Medical Association This study protocol was reviewed and the need for written and informed consent was waived by “Comité de Investigación de Salud del Grupo CHRISTUS MUGUERZA” of the Christus Muguerza Hospital Alta Especialidad. The registration was approved in August 2020. Approval No. CMHAE-053-2020-Cl. Written informed consent from participants was not required in accordance with local/national guidelines.

The authors have no conflicts of interest to declare.

This manuscript has no funding sources.

Data are not publicly available due to ethical reasons. Further inquiries can be directed to the corresponding author.