It’s Time To Reconsider Adult ECMO

Extracorporeal membrane oxygenation (ECMO) is a technology that provides long-term cardiopulmonary bypass support for patients with the most severe but potentially reversible forms of respiratory and/or cardiovascular derangements. This technology has a long history in modern medicine and has become the standard of care for many childhood disorders. Early investigations of ECMO in the adult population demonstrated that it was an expensive, invasive modality that did not improve long-term outcome (Morris et al. Am J Respir Crit Care Med. 1994;149[2Pt1]:295; Zapol et al. JAMA. 1979;242[20]:2193). However, more recent data suggest that ECMO has a place in the treatment of the adult patient with severe respiratory failure (Peek et al. Lancet. 2009;374[9698]:1351; Davies et al. JAMA. 2009;302[17]:1888). Advances in ECMO technology suggest that the time is right for a renaissance in adult ECMO use.

The technology behind the ECMO circuit is simple in concept. The essential components include a pump, oxygenator, tubing, and vascular access (Fig 1). Each of the circuit components has undergone decades of refinement such that a modern ECMO circuit looks little like the systems of the past. Older circuits required frequent maintenance to the tubing to prevent malfunction, and the tubing was subject to occasional rupture or other catastrophic failure due to the roller pumps that were used. Constant attendance of the device by a trained technician was, therefore, mandatory. These problems limited ECMO use to a few centers and increased operational costs.

Recent advances and mass manufacturing of extracorporeal technologies have offered the opportunity to provide ECMO in a simplified and cost-effective manner. Such improvements include low resistance and highly reliable centrifugal pumps with minimal risk of failure. The older silicone-based oxygenators that suffered from regular plasma leak have been replaced by efficient, low-resistance polymethylpentene devices. The new generation of oxygenators operate at lower pressures than the older units, resulting in less hemolysis and lower circuit pressures. Importantly, the new oxygenators can support a patient for weeks without failure. Bedside monitoring of anticoagulation using the activated clotting time point-of-care testing devices has been simplified with a shift to central lab-based activated partial thromboplastin times.

Vascular access has also undergone considerable enhancements. ECMO circuitry initially involved accessing the venous and arterial circulation through open surgical exposure of venous and arterial vessels. Blood was extracted from the venous circulation, run through an oxygenator, and reinfused into a major artery, similar to cardiopulmonary bypass (Fig 1). Today, percutaneous techniques are commonly used for vascular access. The recent FDA approval of double-lumen adult venous cannulae (Fig 2) has simplified vascular access by allowing clinicians to use only a single venovenous (VV) cannula for support. These cannulae are designed to drain venous blood from the superior and inferior vena cava and then infuse highly oxygenated blood directly into the right atrium via a specially designed inflow port (Fig 1). This technique offers respiratory support with a reduced risk of ischemia and other complications associated with arterial-based access. Another advantage of the VV devices is their similarity to other large multilumen central venous catheters used in the ICU. The ECMO VV cannulae can be inserted by nonsurgical intensivists with additional training. VV ECMO is now the preferred method of support for adult patients with acute respiratory failure.

All of these advances have made ECMO considerably simpler and safer. Although the technology still has a considerable level of complexity, it no longer requires a dedicated team to provide continuous support at the patient’s bedside. The modern ECMO system is simple enough that the nurse, with additional training on surveillance of the circuit and basic troubleshooting, can attend to both the patient and the ECMO circuit. Of course, this does not eliminate the need for the availability of knowledgeable and well-trained ECMO technicians. Such individuals are necessary to coordinate safe ECMO services, build and maintain ECMO systems, initiate and terminate ECMO support, and provide a higher level of ECMO troubleshooting than an ICU nurse could or should be expected to do.

Resistance to using ECMO for respiratory failure may linger from earlier and, arguably, no longer applicable studies. In 1979, 90 extraordinarily ill patients with severe ARDS were enrolled in an NIH-sponsored, randomized, prospective trial of VA ECMO vs standard mechanical ventilation (Zapol et al. JAMA. 1979; 242[20]:2193). Forty-eight patients were managed using ECMO, and 42 received conventional ventilator support. The survival difference was not significant between the groups (9.5% vs 8.3%). A subsequent trial published by Morris and colleagues (Am J Respir Crit Care Med. 1994;149[2 Pt1]:295) of 40 patients with severe ARDS, 21 of whom were randomized to ECMO for CO₂ removal, showed no difference in survival between groups.

While it is important to recognize that the early trials of ECMO did not show a survival benefit, it is also important to acknowledge how standard management of ARDS and the critically ill has evolved since these trials. Low tidal volume strategies, goal-directed sepsis management, DVT prophylaxis, and central line placement are all far different today than they were even 10 years ago. Expected mortality for a variety of critical illnesses was far higher in 1979 than expected today. These transitions in care of the patient with respiratory failure have made it exceptionally difficult to understand how ECMO may have actually impacted outcome during this period. Hence, more recent data may help illustrate the potential impact of the technology.

The first randomized trial data supporting the use of ECMO in the adult population were published in 2009 by Peak and colleagues (Lancet. 2009; 374[9698]:1351). The Conventional Ventilation or ECMO for Severe Adult Respiratory Failure (CESAR) trial, through a minimization strategy, enrolled 180 patients with severe ARDS (Murray score > 3 or pH < 7.20). They were randomized to either care at a tertiary care center or transfer and management at a single ECMO center. Of the 90 patients randomized for transfer to the ECMO center, 5 died prior to or during transfer, and 16 improved with conventional management. Sixty-eight patients received ECMO support, and overall survival or severe disability at 6 months was 63% for the ECMO group vs 47% for the conventional management group. Appropriate criticisms of this study have focused on the lack of standardized management in the controls. Also, statistical significance was lost if patients from the ECMO center cohort who did not receive ECMO were removed. Many have reflected on CESAR as a trial of transfer of patients with severe ARDS to a comprehensive center with ECMO support rather than a trial of the ECMO technology itself.

However, the recent ECMO experiences with ARDS in the United Kingdom and severe respiratory failure due to influenza A(H1N1) have led to a renewed interest in using ECMO in adult respiratory failure. A report on ECMO and survival rates outside of the clinical trial environment was provided in JAMA (Davies et al. JAMA. 2009;302[17]:1888). In this study, conducted in Australia and New Zealand, 68 patients suffering from influenza A(H1N1) with ARDS and treated with ECMO were evaluated. The patient population in the ECMO cohort was exceptionally ill, with a mean Pao₂ to Fio₂ ratio (P/F ratio) of 56 on an average of 18 cm H₂O of positive end-expiratory pressure prior to initiation of ECMO therapy. Follow-up communication reported that 17 patients died in hospital (Freebairn et al. JAMA. 2010;303[10]: 941; author reply 942). Overall survival to hospital discharge among patients treated with ECMO was 75%—higher than reported with other series, which may reflect a population with fewer comorbidities and a single pathologic condition (Bartlett et al. JAMA. 2000;283[7]:904).

ECMO today is far simpler and safer, and less technologic prowess and manpower are required at the bedside. In select populations, ECMO may provide a survival advantage in adults compared with traditional management. However, any institution interested in providing ECMO support must understand the high level of training and commitment required of all practitioners involved in the care of such patients. While ECMO is not considered the standard of care, recent advances suggest it is worthwhile to consider this approach when proven ventilation techniques are not adequate.

Dr James M. Blum; and
Dr Andrew L. Rosenberg
Division of Critical Care
Department of Anesthesiology
University of Michigan Medical Center
Ann Arbor, MI