ECMO or Removing CO2

The enthusiasm risen by first successful Extra Corporeal Membrane Oxygenation (ECMO) at the beginning of the seventies by Hill and al. led to the first large randomized trial launched in 1974 to compare veno-arterial (VA) ECMO versus conventional therapy in adult acute respiratory distress syndrome (ARDS) patients. This trial, in 1975, although not completed, already showed discouraging negative results. Dr. Kolobow, at National Institute of Health (NIH), was studying a new membrane lung with greater surface exchange and thinner membrane to optimize the CO2 removal, later called CDML . The underlying hope was that in COPD patients an intermittent CO2 dialysis could potentially improve the clinical scenario. As a new fellow of Dr. Kolobow at NIH, between 1975 and 1977, I was responsible of testing the performances of the CDML, in awake sheep, by measuring the CO2 input and output across the membrane lung, as well as the CO2 removed as gas from the expiration port of the membrane lung. As I was curious to see the respiratory response of the awake spontaneously breathing sheep during CO2 removal, through a closed respiratory circuit I measured the minute oxygen consumption and the CO2 exhaled from the animal. It was immediately evident the strong relationship between the CO2 removed by the artificial lung and the CO2 exhaled by the sheep. Being its metabolic CO2 production near constant, it became immediately evident that the CO2 exhaled by the sheep was decreased proportionally to the CO2 removed by the membrane lung, up to the complete apnea, when CO2 metabolicaly produced was completely cleared by the artificial lung. The oxygenation was provided by diffusion through the natural lung. The first set of experiments was published on Anesthesiology and the title of the paper focused on the capability of the membrane lung to control the spontaneous breathing. The idea of CO2 dialysis in chronic lungers was abandoned and the extracorporeal CO2 removal approach was extensively studied in experimental animals with the aim of applying it in ARDS patients to provide complete or partial lung rest. Therefore, between 1976 and 1980 a series of physiological studies explored the potential of CO2 removal, its physiology and the relationship between artificial and natural lungs. The best set we identified was the one providing complete CO2 clearance associated to 2-3 breath/minute to maintain lung volumes, while oxygenation was primarily performed by 200-300 mL/min at 100% oxygen insufflated into the trachea. The technique was called low frequency positive pressure ventilation with extra corporeal CO2 removal (LFPPV-ECCO2R). The first patients were treated in Milan, and the first successful extracorporeal CO2 removal, presented in a poster session during an Intensive Care meeting in Paris in 1980. Curiously, this first successful CO2 removal, presented as poster, had as a neighbor poster the first report of Lackman on inverted ratio ventilation (both posters were completely neglected). The first experiences with LFPPV-ECCO2-R on 3 patients were published on Lancet in 1980. and 6 years later we reported on JAMA the results obtained in a group of 43 patients. We found that more than 70% of the patients improved lung function and 21 patients eventually survived without major technical accidents in more than 8000 hours of perfusion. Therefore we concluded that this technique could be a reliable alternative to conventional treatments. These results led to the investigations into the technological development of extracorporeal support devices. In 1984 we reported a strict association between the need of LFPPV-ECCO2R and total static lung compliance in a group of 36 ARDS patients meeting mortality rate criteria (90%) as defined in the Zapol ECMO trial. Total static lung compliance (TSLC) was the only predictive value of success or failure of the management of severe ARDS patients unresponsive to conventional treatment. We found that patients with TSLC lower than 25 ml/cmH2O did not tolerate PC-IRV or CPAP, patients with TSLC higher than 30 cmH2O were successfully treated with CPAP while the other patients (TSLC comprised between 25 and 30 cmH2O) had to be treated with PC-IRV for more than 48 h, or were then placed on LFPPV-ECCO2R if PaCO2 rose prohibitively. The results of the study became clear after the quantitative CT scan analysis was introduced in the evaluation of respiratory failure. It was shown, in fact, that the TSLC is strictly related to the size of the ventilatable ARDS lung, which, at TSLC around 25 cmH2O has the size of the normal lung of 2-3-year child ("baby lung"). Therefore we found that the ARDS lung is not stiff but just small. In the nineties ECMO was mainly diffused as the treatment for neonates affected by respiratory failure unresponsive to conventional treatment as reported by the ELSO registry. The results of the second randomized clinical trial on extracorporeal support were published in 1994 by Morris et al. The authors compared the effects of pressure-controlled inverse ratio ventilation followed by LFPPV-ECCO2R to positive pressure ventilation in 40 ARDS patients (21 ECCO2R patients and 19 mechanically ventilated). The study was stopped for futility and survival rates were not significantly different in the 2 groups (33%vs 42% in the control group, p=0.8), despite mortality was impressively improved sing the seventies. The study rose a lot of criticism for little experience with the technique in humans, the high pressure (PEEP and Peak) ventilation and the elevate number of blood loss complications. The research in the field stopped until the new century when another prospective randomized trial on the efficacy and economic assessment of ECMO versus conventional mechanical ventilation was conducted in the United Kingdom between 2001 and 2007 (CESAR trial). The results were published in 2009 on Lancet. The treatment arm of the study was treated at Glenfeld Hospital, a single high volume center capable of treating patients with ECMO. The control group was treated at the hospital of admission or at the nearest one participating to the study. The primary endpoint of the study, the survival at 6 months free of disabilities, was 63% in the ECMO-referred patients (75% of them actually received ECMO) vs 47% in control group. The study was criticized for the randomization of the patients and for the lack of information on the ventilation settings in the control group, however the most important result is that the treatment of patients affected by respiratory failure unresponsive to conventional treatment in an high volume center with ECMO capabilities can significantly improve survival. The H1N1 flu pandemics of 2009 caused an impressive increase of the number of patients characterized by acute pneumonia with severe hypoxemia that were considered not safely ventilatable even with safe mechanical ventilation criteria. The experience of australian and New Zealand investigators led to renewed interest for extracorporeal support and hundreds of ARDS patients worldwide received ECMO. The authors reported that the proper rescue therapy for life-threatening hypoxemia was high flow VV bypass and the overall mortality rate was 21%. After this report, and also due to political support, several countries in Europe, United States, South America, Canada, and Asia faced the pandemic using ECMO as buy time maneuver waiting the resolution of the underlying pathology. Obviously the use of ECMO without a scientific background was criticized as the only evidence for ECMO application was the presence of sever life-threatening hypoxemia in patients untreatable with conventional mechanical ventilation. In Italy the Italian Health Authorities set up a national referral network (ECMOnet) of 14 selected intensive care units able to provide ECMO to face the H1N1 flu pandemic. Two clinical experts coordinated the communication between the authorities and the net and organized the operations. A call center service was set up to grant the communication between hospitals and the referral centers and a series of training courses were performed. A list of recommended national clinical criteria for early patient centralization and for ECMO eligibility was written up. Between August 2009 and March 2010, 153 patients were admitted to the 14 centers with suspected H1N1. Sixty patients were treated with ECMO; among them 49 patients had ARDS caused by H1N1, while 11 patients had ARDS because of other causes. Overall survival at hospital discharge was 41/60 (68.3%), while survival for confirmed H1N1 was 35/49 (72%) versus 6/11 (54%) for non confirmed H1N1. One patient died of cerebral hemorrhage, 16 patients had hemorrhagic complications and 10 of them had major bleeding events but none of them stopped the treatment. For what the ventilatory treatment concerned the setting was left to the referral center. In several centers in Italy ventilatory support was characterized by very low tidal volume and respiratory rate limited to 7-8 bpm with high mean airway pressure due to high PEEP. In Milan patients are initially treated with high PEEP (above 15 cmH2O) and low frequency ventilation. In 2011 a study published on JAMA by Noah et al. compared the hospital mortality of patients affected by H1N1-related ARDS treated with ECMO in 1 of the 4 adult ECMO centers in the United Kingdom during the pandemic with matched patients who were not referred for ECMO from the Swine Flu Triage study. The hospital mortality rate was significantly lower in ECMO-referred patients compared to non-ECMO-referred patients. This study further reinforced the result that new generation devices and the promotion of support from experienced centers seems relevant for a successful ECMO treatment and to reduce hospital mortality. At the time we are writing the "Extracorporeal Membrane Oxygenation for Severe Acute Respiratory Distress Syndrome (EOLIA)" (NCT01470703) is currently recruiting participants. The goal of the study is to evaluate the impact on morbidity and mortality of VV ECMO instituted early after the diagnosis of ARDS not evolving favorably after 3-6 hours of optimal treatment. In conclusion it appears that , as Dr. Bartlett wrote to me, "the pigs did for ECMO more than whatever randomized trial". However, as the people start to use this technique they realized how powerful it is, and this explains the reason why, despite a lack of conventional "evidence" so many centers do apply this technique. To story continues...