Show Summary Details
Page of

Treating respiratory failure with extracorporeal support in the ICU 

Treating respiratory failure with extracorporeal support in the ICU
Treating respiratory failure with extracorporeal support in the ICU

Giacomo Bellani

and Antonio Pesenti

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2020. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 03 December 2020

Key points

  • Veno-venous extracorporeal life support is a solid tool in rescuing patients with severe hypoxaemia.

  • While a low-flow bypass can remove comparatively high amounts of CO2 from blood, oxygenation is limited by venous haemoglobin saturation and therefore requires high extracorporeal blood flows.

  • As shown by several data registries, a shorter duration of mechanical ventilation before initiation of extracorporeal support is associated with a decreased mortality rate.

  • The recent improvements in efficacy and safety of extracorporeal support technology might lead to an earlier application of this technique to remove CO2, decreasing the need for ventilation and, hence, decreasing ventilator-induced lung injury.

  • Extracorporeal support still bears severe risks of complications. Centralizing patients to selected specialized centres with a high case volume is likely to improve the outcome of the treatment.

Historical background

Extracorporeal respiratory support technology was developed in the 1940s as an essential tool in cardiac surgery. Its successful use in an adult patient with acute respiratory distress syndrome (ARDS) was first described in 1971 by J. D. Hill. Following this first report, one multicentre randomized trial was conducted, but no effect on mortality was observed in patients treated with extracorporeal membrane oxygenation (ECMO) [1]‌. The trial was doomed by a very high incidence of severe bleeding and by the use of immature technology. The result of this and other studies led to an almost complete stop of ECMO in adults, confining its application to newborns [2]. The reasons leading to the failure of the adult early trial might probably be identified in the continuous lung damage caused by an injurious setting of mechanical ventilation (MV) in spite of extracorporeal support. Subsequently, Kolobow and Gattinoni suggested the concept that by removing CO2 through an artificial lung, ventilation could be reduced virtually down to zero [3], allowing to put the lung at rest minimizing the negative effect of ventilation.

Present status of ECMO

The concept of ECMO is apparently very simple [4,5]—blood is diverted from the patient to an artificial lung for gas exchange (oxygenation and CO2 removal) and then returned into the patient’s circulation once arterialized. The bypass may be veno-arterial, veno-venous, or arteriovenous (Table 105.1). While a low-flow bypass can remove comparatively high amounts of CO2 from the blood, oxygenation is limited by venous haemoglobin saturation and, therefore, requires high extracorporeal blood flows, in the range of 3–5 L/min. Several technical improvements led to a profound change in the safety and applicability of ECMO in recent years, and even allowed to transfer patients undergoing ECMO [6]‌. Surgical was largely replaced by percutaneous cannulation [7], with a greater bedside applicability and less bleeding complications. Moreover, double lumen cannulas [8], specifically designed to minimize recirculation, allow for single vessel access for both drainage and reinfusion. The introduction of heparin-coated surface for tubings and artificial lungs drastically reduced the need for systemic anticoagulation. Roller pumps were largely abandoned in favour of centrifugal pumps. Until the late 1990s, the flat sheet silicon rubber membrane artificial lungs were the standard of care. These oxygenators required a large priming volume, and had a very high flow resistance. A revolution in oxygenators came from the introduction of hollow-fibre membrane oxygenators whose initial problems with plasma leakage are now largely solved since polymethylpentene membranes came into use. These oxygenators have a much lower priming volume and resistance to blood flow (tens, rather than hundreds of mmHg/L/min). The very low resistance of these circuits permitted the design a pumpless system, which takes advantage of the arteriovenous pressure gradient to pump blood through the oxygenator; this circuit type is not very efficient in oxygenating blood, due to the relatively low blood flow and the high haemoglobin saturation of the blood entering the oxygenator [9]. However, the efficient CO2 removal allows a decrease in traditional ventilation needs.

Table 105.1 Extracorporeal respiratory support


Blood flow

Bypass type

Main suggested indications

Extracorporeal membrane oxygenation (ECMO)*

3–5 L/min

  • Veno-venous

  • Veno-arterial

Rescue of hypoxia in ARDS

Extracorporeal CO2 removal

0.5–2.5 L/min

  • Veno-venous

  • Arteriovenous

  • Prevention of VILI in ARDS

  • COPD

  • Bronchopleural fistulas with air leaks

  • Bridge to transplant

* Brings CO2 removal as a fringe benefit.

ECMO as a rescue procedure for the severely hypoxic patient

ECMO has been mainly proposed as salvage-therapy for the most severe ARDS patients. The substitution for the lung’s gas exchange function reduces the ventilatory requirement and warrants viable levels of oxygenation.

An important contribution in this field came from the Extracorporeal Life Support Organization (ELSO) [10], which collected in 1986 from data of patients treated with ECMO in 130 centres, mainly located in the United States. In 2009 the results from 1473 patients were published, constituting the largest published dataset on this topic. Notwithstanding the severe hypoxaemia (median PaO2/FiO2 was 57 mmHg) and high peak inspiratory pressure (median 40 mmHg) before ECMO institution, the survival rate was 50%. Interestingly, one of the factors associated with a lower mortality was a short duration of mechanical ventilation prior to ECMO. This underlines the importance of an early initiation of ECMO in the most severe patients, rather than a last desperate rescue after failure of ‘traditional’ therapies. Another important case series was published by the group of the Karolinska hospital [11]. The authors reported data from 16 patients with a very high lung injury score (average 3.5): the survival rate was 76%. Patients were managed with minimal sedation, with the use of pressure support ventilation, and accepting arterial saturation as low as 70%.

Until recently, however, no randomized trial had proven a clear survival in patients with ARDS treated with ECMO. In 2009, the publication of the results of the CESAR trial provided the first formal evidence in favour of ECMO application in adults [12]. This study was conducted in the UK and followed a particular design, modelled on the template of a previous ECMO study in neonates. Adult patients (18–65 years) with severe, but potentially reversible respiratory failure were enrolled. Severe respiratory failure was defined as a Murray score ≥3 or uncompensated hypercapnia with a pH <7.20. Patients randomized to ECMO were transferred to the ECMO centre in Leicester; control patients continued the conventional treatment according to the best available clinical practice. In most of the patients assigned to the treatment group, veno-venous ECMO via percutaneous cannulation was used. The system was designed to provide full substitution of pulmonary gas exchange with high blood flows (> 5 L/min), and high gas exchange surfaces. The average duration of bypass was 9 days. During ECMO the ventilator settings were gradually reduced to allow lung rest, limiting the peak inspiratory pressure to 30 cmH2O and respiratory rate to 10 breaths/min. In 5 years, 180 patients were enrolled onto the study. Survival at 6 months or the absence of severe disability was achieved in 63% of the ECMO patients, comparing very favourably with 47% of the control group. This accounted for one life without severe disability saved every six patients treated. In spite of the peculiarity of the study design, this is the first, long-expected, positive randomized clinical trial on adult ECMO application. These results are achieved through lung protection by the total or almost total substitution of the lung gas exchange function.

An important impulse to the early use of ECMO in ARDS came from the experience with the recent outbreak of H1N1 influenza. The pandemic was characterized by a high incidence of severe respiratory complications. Different series of patients with H1N1 infection requiring ICU admission have been described in Mexico, Canada, Australia, and United States. All patients required MV with high PEEP and frequent use of rescue therapies. Davies et al. have recently described a series of 68 patients with H1N1-associated ARDS treated with ECMO in 15 Australian ICUs—the median duration of extracorporeal support was 10 days and the overall mortality was 21% [13].

To cope with the H1N1 influenza epidemic outbreak in Italy, the Italian Ministry of Health organized a multicentre network for the transferring of more severe patients to specialized centres equipped to apply ECMO treatment. Eight of the 14 ECMO centres were also able to start ECMO in outside centres and transport the patient to the referral hospital, while on ECMO. Thus, if patient transportation was considered risky, patient was connected to ECMO at bedside and then transferred while on ECMO to the tertiary centre. Between August 2009 and March 2010, 60 patients with suspected H1N1 affected by severe ARDS were treated with ECMO, H1N1 diagnosis being confirmed later in 49. Twenty-eight patients were transported while on ECMO. Overall ICU discharge survival was 68%. No major complications were reported during the transportation of patients in ECMO. In keeping with the data of the ELSO registry, non-survivors had a longer duration of mechanical ventilation when compared with survivors, in addition to higher SAPS II and SOFA scores [14].

A smaller case series from Marseille, France had been previously published [15], where nine patients with H1N1 were treated with ECMO; six of these had been connected to ECMO in a referral hospital by a mobile team.

A similar system was set up in the UK, with similar results. The British ECMO group recently published a cohort study in which ECMO-referred patients, defined as all patients with H1N1-related ARDS, who were transferred to one of the four adult ECMO centres in the UK, were matched with similar non-ECMO-treated H1N1 ARDS patients using data from a concurrent, longitudinal cohort study. Detailed demographic, physiological, and comorbidity data were used in three different matching techniques. The hospital mortality rate for ECMO-referred patients was almost one half of that for non–ECMO-referred patients in all the three matching used [16].

Future perspectives in extracorporeal respiratory support: extracorporeal gas exchange to avoid injury from mechanical ventilation

Ventilator-induced lung injury (VILI) has been known as a potential mechanism of lung injury, which can further aggravate ARDS. The benefits of reduction of tidal volume and plateau airway pressure reduction are well known, but the need for CO2 removal, even if hypercapnia is tolerated to a certain extent, represent a limiting factor to a further ultra-preventive lung strategy. In this respect, CO2 removal by ECMO could dramatically decrease the ventilatory needs. As stated, in order to effectively remove a large fraction of total CO2 production (without the need for an effective oxygenation), a lower blood flow is required, further reducing the size of the cannulas, the need for anticoagulation and the invasiveness of the technique. Terragni et al. showed that extracorporeal CO2 removal by a low flow device allows the reduction of tidal volume from 6 to 4 mL/kg, with a reduction of circulating inflammatory cytokines and lung hyperinflation. However, more data are necessary in order to understand whether extracorporeal CO2 removal is able to improve the outcome of ARDS patients [17].

Pushing this concept to its furthest end, ECMO could be conceived as an alternative to MV and endotracheal intubation, although some preliminary reports indicate that COPD patients failing non-invasive ventilation might successfully be managed with [low flow] ECMO and avoid intubation [18].


The fire of ECMO has been smouldering under the ashes for many years, but improved technologies, deeper knowledge of the risk of VILI and flu pandemics gave a new burst to its flame. Clinicians should use ECMO with caution, in order to avoid unnecessary risks for their patients. The role of ECMO in severe hypoxia is hardly debatable, but, in such cases, an early application of the technique and a centralization of cases to a few specialized experienced centres (taking advantage of ECMO provided by mobile teams also for a safe patient transport) are mandatory to success. We will hopefully learn, in the next years, if low flow CO2 removal will become a lighthouse for safer, less invasive ventilation, eventually finding its place, in some patients.


1. Zapol WM, Snider MT, Hill JD, et al. (1979). Extracorporeal membrane oxygenation in severe acute respiratory failure. A randomized prospective study. Journal of the American Medical Association, 242(20), 2193–6.Find this resource:

2. Bartlett RH, Roloff DW, Custer JR, Younger JG, and Hirschl RB. (2000). Extracorporeal life support: the University of Michigan experience. Journal of the American Medical Association, 283(7), 904–8.Find this resource:

3. Gattinoni L, Agostoni A, Pesenti A, et al. (1980). Treatment of acute respiratory failure with low-frequency positive-pressure ventilation and extracorporeal removal of CO2. Lancet, 2(8189), 292–4.Find this resource:

4. Pesenti A, Zanella A, and Patroniti N. (2009). Extracorporeal gas exchange. Current Opinions in Critical Care, 15(1), 52–8.Find this resource:

5. Brodie D and Bacchetta M. (2011). Extracorporeal membrane oxygenation for ARDS in adults [Review]. New England Journal of Medicine, 365(20), 1905–14.Find this resource:

6. Isgro S, Patroniti N, Bombino M, et al. (2011). Extracorporeal membrane oxygenation for interhospital transfer of severe acute respiratory distress syndrome patients: 5-year experience. International Journal of Artificial Organs, 34(11), 1052–60.Find this resource:

7. Pesenti A, Rossi GP, Pelosi P, Brazzi L, and Gattinoni L. (1990). Percutaneous extracorporeal CO2 removal in a patient with bullous emphysema with recurrent bilateral pneumothoraces and respiratory failure. Anesthesiology, 72(3), 571–3.Find this resource:

8. Pesenti A, Kolobow T, Riboni AE, et al. (1982). Single vein cannulation for extracorporeal respiratory support. Life Support Systems, 1(Suppl.), 165–7.Find this resource:

9. Kopp R, Bensberg R, Wardeh M, Rossaint R, Kuhlen R, and Henzler D. (2012). Pumpless arterio-venous extracorporeal lung assist compared with veno-venous extracorporeal membrane oxygenation during experimental lung injury. British Journal of Anaesthesia, 108(5), 745–53.Find this resource:

10. Brogan TV, Thiagarajan RR, Rycus PT, Bartlett RH, and Bratton SL. (2009). Extracorporeal membrane oxygenation in adults with severe respiratory failure: a multi-center database. Intensive Care Medicine, 35(12), 2105–14.Find this resource:

11. Linden V, Palmer K, Reinhard J, et al. (2000). High survival in adult patients with acute respiratory distress syndrome treated by extracorporeal membrane oxygenation, minimal sedation, and pressure supported ventilation. Intensive Care Medicine, 26(11), 1630–7.Find this resource:

12. Peek GJ, Mugford M, Tiruvoipati R, et al. (2009). Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet, 374(9698), 1351–63.Find this resource:

13. Davies A, Jones D, Bailey M, et al. (2009). Extracorporeal Membrane Oxygenation for 2009 Influenza A(H1N1) Acute Respiratory Distress Syndrome. Journal of the American Medical Association, 302(17), 1888–95.Find this resource:

14. Patroniti N, Bellani G, Saccavino E, et al. (2012). Respiratory pattern during neurally adjusted ventilatory assist in acute respiratory failure patients. Intensive Care Medicine, 38(2), 230–9.Find this resource:

15. Roch A, Lepaul-Ercole R, Grisoli D, et al. (2010). Extracorporeal membrane oxygenation for severe influenza A (H1N1) acute respiratory distress syndrome: a prospective observational comparative study. Intensive Care Medicine, 36(11), 1899–905.Find this resource:

16. Noah MA, Peek GJ, Finney SJ, et al. (2011). Referral to an extracorporeal membrane oxygenation center and mortality among patients with severe 2009 influenza A(H1N1). Journal of the American Medical Association, 306(15), 1659–68.Find this resource:

17. Terragni PP, Del Sorbo L, Mascia L, et al. (2009). Tidal volume lower than 6 mL/kg enhances lung protection: role of extracorporeal carbon dioxide removal. Anesthesiology, 111(4), 826–35.Find this resource:

18. Kluge S, Braune SA, Engel M, et al. (2012). Avoiding invasive mechanical ventilation by extracorporeal carbon dioxide removal in patients failing noninvasive ventilation. Intensive Care Medicine, 38(10), 1632–9.Find this resource: