Elevated oxygen demand in a case of COVID-19 with severe ARDS: a point for optimal oxygenation therapy including ECMO management

Background: Coronavirus disease 2019 (COVID-19) caused by SARS-CoV-2 has become a global pandemic, and those developing critically ill conditions have been reported to have mortality in the range of 39% to 61%. Due to the lack of denitive treatments, mechanical ventilation and supportive oxygenation therapy are key management strategies for the survival of patients with acute respiratory distress syndrome (ARDS). Optimizing oxygenation therapy is mandatory to treat patients with severe respiratory failure, to suciently compensate for the oxygen (O 2 ) demand. We experienced a case of severe ARDS due to COVID-19 successfully treated with extracorporeal membrane oxygenation (ECMO) after increasing oxygen delivery according to O 2 consumption measurement by indirect calorimetry Case Presentation: A 29-year-old obese but otherwise healthy man was hospitalized for treatment of COVID-19 pneumonia presenting with a 4-day history of persisting cough, high fever, and dyspnea. Mechanical ventilation, nitric oxide inhalation, and prone positioning were initiated in the ICU against severe respiratory dysfunction. Indirect calorimetry on the 3 rd and 6 th ICU days revealed persistent elevation of oxygen consumption (VO 2 ) of 380 mL/min. Veno-venous ECMO was initiated on the 7 th ICU day after further deterioration of respiratory failure. Periodic events of SpO 2 decline due to effortful breathing was not resolved by neuromuscular blockade in attempt to reduce O 2 consumption. Increasing the ECMO ow induced hemolysis and hyperkalemia despite the use of large bore cannulas and ECMO circuit free of clots and defects. The hemoglobin management level was elevated from 10 g/dL to 13 g/dL to increase blood oxygen capacity, enabling the reduction of ECMO ow while attenuating respiratory effort and maintaining SpO 2 . Lung protective ventilation strategy and prone positioning were continued for successful weaning from ECMO on the 16 th ICU day, and the ventilator on the 18 th ICU day. Conclusion: The present case of severe ARDS due to COVID-19 was successfully treated with ECMO. Enhancing oxygen delivery was crucial to compensate for the elevated O 2 demand. Measuring O 2 consumption by indirect calorimetry can elucidate the oxygen demand for optimizing

Mechanical ventilation, nitric oxide inhalation, and prone positioning were initiated in the ICU against severe respiratory dysfunction. Indirect calorimetry on the 3 rd and 6 th ICU days revealed persistent elevation of oxygen consumption (VO 2 ) of 380 mL/min. Veno-venous ECMO was initiated on the 7 th ICU day after further deterioration of respiratory failure. Periodic events of SpO 2 decline due to effortful breathing was not resolved by neuromuscular blockade in attempt to reduce O 2 consumption. Increasing the ECMO ow induced hemolysis and hyperkalemia despite the use of large bore cannulas and ECMO circuit free of clots and defects. The hemoglobin management level was elevated from 10 g/dL to 13 g/dL to increase blood oxygen capacity, enabling the reduction of ECMO ow while attenuating respiratory effort and maintaining SpO 2 . Lung protective ventilation strategy and prone positioning were continued for successful weaning from ECMO on the 16 th ICU day, and the ventilator on the 18 th ICU day.
Conclusion: The present case of severe ARDS due to COVID-19 was successfully treated with ECMO. Enhancing oxygen delivery was crucial to compensate for the elevated O 2 demand. Measuring O 2 consumption by indirect calorimetry can elucidate the oxygen demand for optimizing the oxygenation therapy for successful management and survival of critically ill COVID-19 patients.

Background
Coronavirus disease 2019 (COVID-19) caused by SARS-CoV-2 has become a global pandemic (1-3), and those developing critically ill conditions have been reported to have mortality in the range of 39% to 61% (4)(5)(6). Due to the lack of de nitive treatments prove effective for COVID-19, mechanical ventilation and supportive oxygenation therapy are key management strategies for the survival of patients with acute respiratory distress syndrome (ARDS) (7).
The goal of oxygenation therapy for ARDS is to provide su cient oxygen (O 2 ) to meet the demand of the whole-body metabolism. Hypoxia can occur when oxygen delivery insu ciently compensates for the O 2 demand, either as a result of reduced oxygen uptake or the excessive elevation of oxygen consumption.
For progressive respiratory failure despite optimal conventional management such as prone positioning and nitric oxide (NO) administration, extracorporeal membrane oxygenation (ECMO) support is indicated (10,11). Moreover, measures to increase the e cacy of O 2 supplementation become critical when O 2 demand exceeds delivery, and reduction of O 2 consumption is not feasible. However, methods to determine O 2 demand of COVID-19 patients have not been well described to date.
Here, we encountered a case of severe ARDS due to COVID-19 successfully treated with ECMO by optimizing oxygen delivery against persistent elevation of O 2 demand revealed by indirect calorimetry.

Case Presentation
A 29-year old male admitted to a clinic after 4 days of persisting cough, high fever, and dyspnea, without gastrointestinal symptoms or loss of taste and smell. He was obese with body mass index (BMI) of 31.9 kg/m 2 (height 168 cm, weight 90.0 kg), but otherwise presented no underlying pathologies, did not travel abroad and worked at home for at least 14 days prior to the onset of the symptoms. He was diagnosed positive for SARS-CoV-2 polymerase chain reaction test the day after his admission and referred to our hospital for treatment of COVID-19 pneumonia. His brother, a nursing home employee, living in the same household was later diagnosed SARS-CoV-2 positive.
He was conscious and communicated verbally upon admission but presented high fever and tachypnea of 30/minute, with persisting cough requiring 6 L/min of O 2 to maintain SpO 2 95%. Blood tests revealed non-bacterial in ammatory signs, with white blood cell count of 5,900/mm 3 with slightly reduced lymphocyte count of 985/ mm 3 , CRP 5.29 mg/dL, procalcitonin 0.14 ng/mL, and interleukin-6 86.8 pg/mL [ Table 1]. He also presented bilateral ground-glass opacities and patchy consolidation on the lung computed tomography (CT) scan, typical ndings of COVID-19 pneumonia [ Figure 1a, 2a]. Favipiravir and azithromycin were started on the day of hospitalization, but SpO 2 fell below 92% under 15L/min O 2 reservoir mask on day 3, when he was intubated and admitted to the intensive care unit (ICU).
Mechanical ventilation was initiated, and nitric oxide was inhaled at 20 ppm for oxygenation support.
Chest X ray on the 3 rd ICU day presented severe progression of bilateral in ltrate, prompting prone positioning for lung recruitment. Indirect calorimetry [E-COVX, General Electric, USA] was conducted on the 3 rd ICU day for the evaluation of energy and O 2 consumption [ Table 2]. The resting energy expenditure Events of excessive respiratory effort were observed after the initiation of VV-ECMO, with a simultaneous SpO 2 decline below 80%. As the patient was already deeply sedated with midazolam and fentanyl to achieve Richmond agitation-sedation scale of -5, neuromuscular blocking with rocuronium was added to attenuate the excessive respiratory effort, thus to avoid lung damage and to reduce O 2 consumption.
However, unresolved SpO 2 decline with gradual increase of lactate from 1.2 to 1.8mmol/l under stable systemic circulation necessitated increasing the ECMO blood ow to 4.5 L/min. However, the increased ow induced hemolysis and hyperkalemia, without obvious signs of sucking down or clot formation in the ECMO circuit and arti cial lung. We strategically elevated the hemoglobin management level from 10 g/dL to 13 g/dL to enhance the blood oxygen capacity and thus the O 2 delivery, enabling the reduction of the ECMO blood ow to resolve hemolysis.
Chest X-ray ndings presented gradual recovery from bilateral in ltration, enabling reduction of ECMO ow to 2.6L/min for su cient oxygenation by the 12 th ICU day. He was weaned off from ECMO on the 16 th ICU day after 120 minutes of a successful sweep gas cessation trial. He was successfully weaned off from the ventilator on 18 th ICU day and was transferred back to the ward on the 23 rd ICU day. After 23 days of rehabilitation in the ward, he was discharged home on his 46 th day of hospitalization [ Figure 1b, 2c].

Discussion And Conclusions
We experienced a case of young COVID-19 patient suffering from severe ARDS. Oxygen delivery by VV-ECMO was optimized to compensate for the increased O 2 demand as observed in indirect calorimetry, for successful management and survival.
While young COVID-19 patients rarely develop critically ill conditions, (5, 12) obesity (BMI≧30 kg/m 2 ) has been observed in 46% of the critically ill COVID-19 patients (5). The current case had BMI of 31.9 kg/m 2 and presented bilateral ground-glass opacities and patchy consolidation on the CT scan, typical of acute respiratory distress syndrome in COVID-19 with reduced lung compliance (13). Although VV-ECMO is indicated, mortality for COVID-19 patients with severe respiratory failure has been reported to be as high as 50% (7). Appropriate management of ECMO is essential for the safety and e cacy of the treatment, thus survival of the patients (11).
While oxygenation failure can only be evaluated passively by the amount of O 2 supplementation required to maintain adequate blood oxygen levels, O 2 consumption can be measured by indirect calorimetry, a method to measure VO 2 and carbon dioxide production (VCO 2 ) by respiratory gas analysis to calculate energy expenditure (14). Oxygen demand in critically ill COVID-19 patients has not been well described, with only one study reporting a median REE of 4044 kcal/day in seven critically ill patients without ECMO, corresponding to VO 2 of 585 mL/min (18), more than twice the average (270mL/min) reported in critically ill patients (15). The elevated demand for oxygen is likely to be sustained during the course of respiratory failure, leading to the need for VV-ECMO and additional measures to increase the e cacy of O 2 delivery (17,19). As indirect calorimetry is not readily available for patients under ECMO (19,20), it is recommended to conduct calorimetry soon after initiation of mechanical ventilation to optimize O 2 delivery according to the O 2 demand.
Indirect calorimetry was conducted in the present case on the 3 rd and 6 th ICU days, before the initiation of VV-ECMO. VO 2 reached 380ml/min while being deeply sedated at RASS -5. The REE of 2583kcal/d corresponds to 1.6 times the energy requirements calculated by the Harris-Benedict equation using ideal body weight (1563kcal/d), comparable to stress factor for calculating energy demands in 50 % total body surface area burns. The elevated VO 2 persisted until the 6 th ICU day, re ecting the sustained O 2 demand.
Increasing O 2 delivery with VV-ECMO was required to maintain systemic oxygenation, under normal temperature management with ECMO and efforts to reduce O 2 consumption by administering sedatives and neuromuscular blockers to achieve RASS -5. Since increasing the ECMO ow induced hemolysis despite the use of large bore catheters (25Fr for drainage and 19Fr for out ow) and no obvious signs of suck down or clot formation in the circuit and the arti cial lung, blood hemoglobin level was elevated to increase the blood oxygen capacity and thus the O 2 delivery (17).
In conclusion, we experienced a COVID-19 patient with severe ARDS, successfully managed by VV-ECMO.    Figure 1 Lung CT Scan on admission and before discharge. a) Bilateral ground glass opacities and patchy consolidations typical of COVID-19 pneumonia was observed on admission. b) The initial ndings have Page 10/10 resolved with slight remains of ground glass appearance by the day before discharge.

Figure 2
Chest x-ray progression during the hospital stay. a) Minor shadows observed in the lower lobes on hospital admission. b) Severe bilateral in ltration progressed with the deterioration of respiratory function, requiring VV-ECMO. c) Although slight bilateral shadows remain, respiratory function su ciently recovered, no longer requiring oxygen supplementation at the time of hospital discharge.