Extracorporeal Carbon Dioxide Removal Using a Renal Replacement Therapy Platform to Enhance Lung-Protective Ventilation in Hypercapnic Patients With Coronavirus Disease 2019-Associated Acute Respiratory Distress Syndrome

Coronavirus disease 2019 (COVID-19)-associated acute respiratory distress syndrome (ARDS) is associated with high mortality. Lung-protective ventilation is the current standard of care in patients with ARDS, but it might lead to hypercapnia, which is independently associated with worse outcomes. Extracorporeal carbon dioxide removal (ECCO2R) has been proposed as an adjuvant therapy to avoid progression of clinical severity and limit further ventilator-induced lung injury, but its use in COVID-19 has not been described yet. Acute kidney injury requiring renal replacement therapy (RRT) is common among critically ill COVID-19 patients. In centers with available dialysis, low-flow ECCO2R (<500 mL/min) using RRT platforms could be carried out by dialysis specialists and might be an option to efficiently allocate resources during the COVID-19 pandemic for patients with hypercapnia as the main indication. Here, we report the feasibility, safety, and efficacy of ECCO2R using an RRT platform to provide either standalone ECCO2R or ECCO2R combined with RRT in four hypercapnic patients with moderate ARDS. A randomized clinical trial is required to assess the overall benefit and harm. Clinical Trial Registration: ClinicalTrials.gov. Unique identifier: NCT04351906.


INTRODUCTION
The percentages of coronavirus disease 2019 (COVID- 19) patients diagnosed with acute respiratory distress syndrome (ARDS) range between 20 and 67% of hospitalized patients (1,2) and 100% of mechanically ventilated patients (3) and are associated with high mortality (2). Lung-protective ventilation is the current standard of care for ARDS (4), which limits ventilator-induced lung injury but may lead to elevated FIGURE 1 | Schematic representation of the ECCO 2 R circuit either as standalone therapy (A) or in-line after the hemofilter for combined ECCO 2 R and RRT (B). ECCO 2 R in conjunction with RRT may be performed with regional citrate anticoagulation (1) or systemic heparinization (2). Ca, calcium chloride solution; Ci, trisodium citrate solution; Pump 1, blood line; Pump 2, effluent dialysate line; Pump 3, dialysate line; P1, in-flow pressure sensor; P2, pre-filter pressure sensor; P3, out-flow pressure sensor; P4, effluent dialysate pressure sensor; RRT, renal replacement therapy. carbon dioxide (CO 2 ) levels and respiratory acidosis, which are independently associated with worse outcomes in the setting of ARDS (5,6). In these patients, extracorporeal CO 2 removal (ECCO 2 R) may help to avoid the progression of clinical severity (5). Acute kidney injury (AKI) is common among critically ill COVID-19 patients, with ∼20% requiring renal replacement therapy (RRT) (7). Recent studies have proposed the integration of ECCO 2 R into continuous RRT (CRRT) platforms to provide combined CO 2 removal and renal support using low blood-flow levels (<500 mL/min) (5,8). Of note, only one study described the use of CRRT platform-driven ECCO 2 R without hemofilter to provide standalone ECCO 2 R in patients with mild to moderate ARDS. However, that trial used an ECCO 2 R membrane with a significantly lower surface area (0.32 m 2 as opposed to 1.35 m 2 in the current study), limiting the rate of maximal CO 2 removal (9). In centers with available dialysis, low-flow ECCO 2 R using CRRT platforms might be an option to efficiently allocate resources for patients with hypercapnia as the main indication. The use of ECCO 2 R has not been described so far in COVID-19associated ARDS.

MATERIALS AND METHODS
We report results of a single-center study evaluating the feasibility and safety of ECCO 2 R in combination with a CRRT platform as a standalone therapy or combined with CRRT for ARDS patients with refractory hypercapnia (arterial partial pressure of CO 2 [PaCO 2 ] > 55 mmHg) secondary to confirmed COVID-19 to effectively decrease CO 2 levels and enhance lung-protective ventilation.

Study Design and Participants
COVID-19 was diagnosed according to the World Health Organization (WHO) guidance (10). All patients were nursed in an isolation intensive care unit (ICU) with other patients suffering from COVID-19. The study was prospectively registered at http://clinicaltrials.gov (Identifier: NCT04351906). Patients were sedated with fentanyl, midazolam, and propofol. Other medications, including antibiotics, fluids, catecholamines, and transfusions, were left to the discretion of the attending physician.

Participants
In-patients ≥18 years of age with confirmed COVID-19 admitted to the University Hospital Giessen and Marburg, Giessen Medical Center, were enrolled in the feasibility study. Inclusion criteria were mild-to-moderate ARDS according to the Berlin definition (11), 100 mmHg < partial alveolar oxygen pressure/fraction of inspired oxygen (PaO 2 /FiO 2 ) <300 mmHg with positive end-expiratory pressure >5 cmH 2 O on mechanical ventilation expected to last >24 h; hypercapnia >55 mmHg with or without metabolic acidosis (pH < 7.3); bilateral opacities on chest imaging; with or without AKI requiring dialysis. Exclusion criteria were age <18 years, pregnancy, patients with decompensated heart failure or acute coronary syndrome, respiratory acidosis with persistent partial pressure of blood carbon dioxide (PaCO 2 ) levels >80 mmHg, acute brain injury, severe liver insufficiency (Child-Pugh scores > 7) or fulminant hepatic failure, decision to limit therapeutic interventions, catheter access to a femoral vein or jugular vein impossible, and pneumothorax.
Extracorporeal Carbon Dioxide Removal Operational Characteristic ECCO 2 R was provided using a polymethylpentene, hollow fiber, gas-exchanger membrane (multiECCO 2 R; Eurosets, Medolla, Italy), a labeled and certified European device to be used in conjunction with multiFiltrate CRRT platforms (Fresenius Medical Care, Bad Homburg, Germany) for combined respiratory and renal support. The manufacturer determined the multiECCO 2 R membrane's maximum duration to be 72 h. A 13.5-Fr dual lumen hemodialysis catheter (Niagara, Bard Access, Heidelberg, Germany) was percutaneously inserted under in the femoral vein. Sweep gas flow was set at a gas/blood flow ratio of 15:1. Data were collected before starting ECCO 2 R (baseline) and 1, 4, 24, and 48 h after initiation of ECCO 2 R. A bloodline warmer (Barkey S-line) and a thermal pad (both from Barkey, Leopoldshöhe, Germany) wrapped around the multiECCO 2 R, as well as a warming blanket, were used to avoid undercooling of the patient. Figure 1 depicts a schematic representation of the ECCO 2 R setup used in this study, either as standalone therapy ( Figure 1A) or in conjunction with RRT ( Figure 1B). The technical terminology of the extracorporeal circuit was based on a nomenclature developed for RRT (12). For standalone ECCO 2 R, the multiFiltrate was set in hemoperfusion mode. ECCO 2 R was commenced at a blood flow of 400 mL/min. Systemic heparinization was started after catheter insertion aiming for an activated partial thromboplastin time of 60-80 s.
For ECCO 2 R + CRRT, the multiFiltrate was set in continuous venovenous hemodialysis (CVVHD) mode, and the multiECCO 2 R was inserted in series after the hemofilter (Ultraflux AV 1000S, Fresenius Medical Care, Bad Homburg, Germany). ECCO 2 R + CRRT was commenced at a blood flow of 200 mL/min. CVVHD was delivered with an effluent dose of 25 mL/kg/h and regional citrate anticoagulation aiming a post-filter ionized calcium concentration of ∼0.25-0.35 mMol/L.

Definitions
Lung-protective ventilation strategies were the standard of care for invasive mechanical ventilation (4). Treatment strategies for COVID-19-associated ARDS were based on the WHO interim guidance (10), which were in line with our institutional standard of care for other forms of ARDS. Of note, at the time of patient recruitment, the WHO guidance on corticosteroids to treat patients with severe and critical COVID-19 was not available (13). Therefore, we did not routinely use corticosteroids for this patient population. Severe adverse events were defined as recently described (14). The feasibility of ECCO 2 R was assessed using Bowen et al.'s (15) feasibility framework. The use of RRT was at the discretion of the attending physician rather than by predefined biochemical or clinical criteria. However, RRT was initiated emergently when lifethreatening changes in fluid, electrolyte, and acid-base balance occurred (16). The Institute of Medical Virology (Justus Liebig University Giessen, Germany) processed nasopharyngeal swabs and bronchoalveolar lavage fluid specimens, and severe acute respiratory syndrome coronavirus 2 infection was confirmed by real-time PCR according to the previously described protocols (17).  Table 2). Patient 4 developed combined respiratory and metabolic acidosis secondary to hypercapnia and AKI, and ECCO 2 R + CRRT was commenced with a blood-flow rate of 200 mL/min, leading to a decrease of PaCO 2 from 58.7 to 46.5 mmHg within 1 h while pH and bicarbonate levels progressively increased. CRRT ultrafiltration (100 mL/h) was started at 38 h post-ECCO 2 R initiation due to oliguria. Tidal volume, plateau and driving pressure, as well as respiratory rate could be reduced during the second day of ECCO 2 R (from median 6.  Figure 2B). No patient-or ECCO 2 R/CRRT-related adverse events occurred. Downtime ranged from 2 to 8% of the total treatment time owing due to the turning of patients into the prone position. In all four patients, the ECCO 2 R treatment could be terminated after a median of 5.5 (4.5-7.5) days due to a sustained improvement in hypercapnia. In patient 4, however, CRRT was continued for another 4 days due to oliguria. Furthermore, patient 2 developed AKI stage 3, necessitating CRRT 6 days after the termination of ECCO 2 R as a sequel to septic shock.

DISCUSSION
Our data indicate that low-flow ECCO 2 R using CRRT platforms might be safe and feasible to provide either standalone ECCO 2 R or ECCO 2 R combined with CRRT. This minimally invasive approach leads to efficient CO 2 removal in the setting of moderate ARDS. No patient-or ECCO2R/CRRT-related adverse events occurred. Importantly, these data also implicate that CRRT, continuous renal replacement therapy; ECCO 2 R, extracorporeal carbon dioxide removal; FiO 2 , fraction of inspired oxygen; HCO 3 , bicarbonate; PaCO 2 , arterial partial pressure of carbon dioxide; PBW, predicted body weight; PCO 2 , venous partial pressure of carbon dioxide; PaO 2 , arterial partial pressure of oxygen; RR, respiratory rate; V E , minute volume; V T , tidal volume.
every ICU with available dialysis may apply RRT platformdriven ECCO 2 R to limit ventilator-induced lung injury or rescue uncontrollable respiratory acidosis even in situations where "standard" ECCO 2 R consoles are not available. To the best of our knowledge, this is the first description of ECCO 2 R in COVID-19. Although these data may provide the rationale for randomized clinical trials, the following limitations need to be acknowledged. Given the invasive nature of an ECCO 2 R therapy, future randomized trials are required to assess the overall benefit and harm before widespread implementation can be recommended. Also, eligibility criteria should be further examined, particularly in those without an indication for CRRT. Furthermore, if the ECCO 2 R is intended to be continuous, sustaining a blood flow of 400 mL/min with a temporary catheter may be challenging, particularly in patients with COVID-19 who are obese or require prone positioning. COVID-19 induces a hypercoagulable state in many patients, which may result in premature extracorporeal circuit failure (18). No studies are available to date to aid in the selection of anticoagulation strategy, in particular when introducing an extracorporeal circulation. Thus, close monitoring of the extracorporeal circuit performance is advisable to ensure maximal circuit patency, as the initial anticoagulation strategy may not be effective in all patients, and a stepwise escalation and/or alternative plans (e.g., combination of different anticoagulation strategies) may be required. However, if using CRRT, we suggest CVVHD or continuous venovenous hemodiafiltration to decrease filtration fraction and reduce the risk of circuit clotting (19). In addition, COVID-19associated ARDS may follow uncontrolled host immune response to the virus with the release of various immune mediators, especially cytokines, damage-associated molecular patterns, and pathogen-associated molecular patterns (20,21). Extracorporeal blood purification techniques (e.g., hemoperfusion; RRT with surface-modified AN69, polymethylmethacrylate, or high-cut off membranes) have been proposed as adjuvant therapy for critically ill patients with COVID-19 to restore immune homeostasis through the removal of these circulating mediators (7). As many healthcare agencies have authorized emergency use of various extracorporeal blood purification techniques, these treatments might be indicated as sequential extracorporeal therapies in special cases in which immuno-dysregulation is evident, inflammatory parameters or cytokines are elevated, and other supportive therapies are failing or insufficient. Nonetheless, careful patient selection is required if these are to be used, as the benefits and adverse effects in COVID-19 patients have not been formally studied. Finally, additional costs associated with the use of ECCO 2 R in conjunction with RRT platforms in COVID-19-associated ARDS may be offset by a potential cost reduction through the elimination of daily rental costs for standalone ECCO 2 R consoles, the recruitment of dialysis professionals in centers with available dialysis to operate ECCO 2 R, and a shorter length of ICU and hospital stay. However, large, multicenter randomized clinical trials are required to support the cost-benefit ratio of ECCO 2 R in conjunction with RRT platforms.
In conclusion, our data indicate that low-flow ECCO 2 R using CRRT platforms might be safe and feasible to provide either standalone ECCO 2 R or ECCO 2 R combined with CRRT. A multicenter randomized trial is warranted to assess the effects of CRRT platform-driven ECCO 2 R on clinical outcomes of patients with ARDS secondary to COVID-19 or other pathogenic factors.

DATA AVAILABILITY STATEMENT
The datasets used and/or analyzed during the study are available from the corresponding author upon reasonable request.

ETHICS STATEMENT
The study protocol was approved by the Ethics Committee of the Medical Faculty of the Justus Liebig University Giessen (AZ 63/20) and complied with the Declaration of Helsinki. Legally authorized representatives of the patients provided written informed consent. FH-S and IV: manuscript drafting and had full access to all study data and had final responsibility for submitting for publication. FH-S, JW, and IV: figure illustration. H-WB, H-DW, WS, and IV: study supervision. All authors shared the study design, data collection, data analyses, data interpretation, as well as preparation, review, and approval of the manuscript.