Scavenging Gas Outlet Ports with COVID-19 ECMO and new FDA Recommendations for ECMO

In a previous article, I raised a question about the possibilities of viruses being transmitted from the gas outlet ports of membrane oxygenators, especially during long term support like ECMO/ECLS. From the responses, it seemed clear that perfusionists and ECMO specialists were concerned about scavenging this outlet gas during ECMO support COVID-19 patients. The use of scavenging from gas outlet ports during extracorporeal circulation has been used for many years in cardiac surgery (in many programs) to eliminate the release of volatile anaesthetic gas into the operating room. The use of vacuum on the gas outlet port of a membrane oxygenator is a safe practice, if done using vacuum <10 mmHg and tubing with an open leur connector, open wye connector or a cut in the vacuum tubing. It is also advisable to turn the vacuum source off when there is no gas flow through the membrane oxygenator.

This update will first repeat the article from February 20, 2020 and provide new information about a virus that is known to be transmitted by droplets, respiratory secretions, touch, fecal matter and blood. I will also touch on the new recommendations from the USFDA regarding the use of membrane oxygenators for ECMO and >6 hours of use.

The topic of invasive mechanical circulatory support for viral respiratory failure has been extensively reviewed over the past decade. However, one area that has never been covered in these reviews is the potential for transmission of pandemic viruses like H1N1, SARS, MERS, and COVID-19, during the use of Extracorporeal Membrane Oxygenation (ECMO), often called Extracorporeal Life Support (ECLS).

These invasive forms of extreme life support are often required when the patient’s native lungs fail to provide O2 transport and CO2 removal due to respiratory failure. ECMO/ECLS therapies temporarily take over gas transport for the lungs by diverting the patient’s blood through these extracorporeal devices and can maintain lung functions (and cardiac functions if necessary) for several hours to many weeks.

The question about their use with airborne/droplet transmissible viral diseases is this;

Can these viruses be transmitted through the membrane oxygenators polypropylene hollow fibers and enter the hospital environment through the membranes gas outlet? Historically, membrane oxygenators were originally designed to maintain lung and cardiac functions during cardiac surgery, trauma and organ system failure. ECMO use was always an option for respiratory support with transmissible pulmonary diseases, but prior to 2009-2010 it was not always a popular, acceptable option. After the H1N1 pandemic, ECMO/ECLS suddenly became an acceptable form of therapy in bacterial and viral pneumonia patients that advanced into severe respiratory failure (ARDS).

According to the Extracorporeal Life Support Organization (ELSO), the use of ECMO to support adult respiratory failures went from 3,280 patients in 2012 (64% survival), up to 19,482 cases in 2019 (69% survival). That is an increase of 494% over 7 years. From personal experience, I know that ECMO/ECLS is widely used in China to support respiratory failure with pandemic viral patients. Statistics from the World Health Organization (WHO) for COVID-19, demonstrated that » 5% of the affected patients in China developed respiratory failure. This potentially (or theoretically) means that there may have been 3,531 new patients supported on ECMO/ECLS in China since late November 2019.

To my knowledge, with four decades of experience in this specialty, there has never been any investigation into the presence of viral pathogens in the exhaust gases of membrane oxygenators, even though plasma leakage is still a possibility in membrane oxygenators (even polymethylpentene membranes) during long term use.

To better understand the design of this membrane technology, I include the following excerpts from my book “Biocompatible Materials and ExtraCorporeal Circulation” … available through the LivaNova Campus Educational Platform; (https://campus.livanova.com/public/cardiopulmonary-bypass.php)

— All hollow fiber membranes used for extracorporeal circulation are hydrophobic by design to allow for the passage of gas through the fibers and into the blood, while inhibiting the passage of plasma water into the gas phase. Hollow fiber membranes are initially porous to plasma until they are exposed to their first pass of plasma proteins. Therefore, there is a direct blood to gas contact during bypass when the hollow fiber micropores are eventually coated with the patient’s blood plasma proteins, after which gas exchange takes place through the micropores by direct contact.

The resistance of microporous hollow fibers to blood plasma leakage is dependent on it permeance. In the context of hollow fibers, permeance is the allowance of movement of either blood plasma or gas. When plasma infiltrates the hollow fibers, it markedly diminishes the membranes permeance, which results in a reduction of gas exchange due to the stagnant plasma within the fiber pores. Over time, a decrease in the membranes permeance due to the increase in plasma wetting can degrade an oxygenator’s performance, which is one of the contributing factors to the international standard organization (ISO) limiting ALL hollow fiber membrane oxygenator use to < 6 hours (uncoated, coated and polymethylpentene).

Some in vitro reports indicate that plasma wetting of the hollow fibers may occur between 4-12 hours, while others give a range of 8-16 hours. It is believed that plasma wetting may result from phospholipids, lipoproteins or proteins in blood that adsorb onto the internal lumen walls, rendering this surface to become more hydrophilic and thereafter allowing either partial or complete plasma infiltration of the fiber.

One of the ways to create plasma resistant fibers and prevent plasma wetting is to coat existing polypropylene hollow fibers with a thin (< 1.0µ thick) non-porous polymer layer, which are thereafter called skinned asymmetric and composite symmetric hollow fibers. The thin non-porous polymer creates a layer over the fiber surface (called a skin), which thereby renders it resistant to blood plasma infiltration.  The challenge to prevent plasma wetting with hollow fibers, is to reduce/eliminate the permeability of the fibers internal walls to plasma, but at the same time maintain the permeance of the outer walls to gas exchange. The most well-known of these skinned asymmetric fibers used for long term extracorporeal circulatory support is polymethylpentene (PMP). PMP hollow fibers (Membrana GmbH, Germany) have an inner diameter of 200 µ and a wall thickness of 90 ±10 µ. These skinned PMP fibers have an asymmetric pore structure that allows gas exchange to take place by diffusion without any direct contact between blood and gas across the skinned micropores, thereby preventing plasma leakage even with blood contact for prolonged periods of time. –

Although most membrane oxygenators used today for long term support during ECMO/ECLS use PMP technology, there is no evidence-based literature to suggest that viruses cannot permeate these hollow fiber materials. After all, most viruses range in size between 0.02 microns and 0.2 microns, and viruses in the Coronaviridae family (SARS, etc) range in size between 0.08 and 0.15 microns. Due to the size of airborne/droplet viruses, most hospital filtration masks are relatively ineffective at removing them. Only the N95 mask seems capable of removing viruses in an effective manner and is the reason these are in such high demand for health care workers.

However, the question still remains, should we be venting the membrane oxygenators exhaust gas when using ECMO/ECLS to support patients with transmissible viral respiratory conditions? In addition, should we avoid using ECMO/ECLS membranes that do not use PMP technology (regardless of surface coating) when supporting these types of viral organ failures.

Another concern with ECMO/ECLS devices in pandemic respiratory support cases is the manner in which these circuits are disposed of. For example, should these blood contact devices, used to support respiratory failure in COVID-19 patients, be disposed of in the same manner as other membranes used during routine cardiac surgical procedures or should they be handled differently?

As previously mentioned, during the pandemic of H1N1 (2009-2010), the use of ECMO/ECLS support became recognized as an extremely important form of therapy for respiratory failure in these patients. After that outbreak, the growth of ECMO/ECLS skyrocketed and became recognized as a good bridge to recovery during respiratory failure with viral pandemics.

However, the impact (containment) of air borne viral diseases on extracorporeal membrane technologies now needs to be investigated, to assure practitioners of their own safety when using ECMO/ECLS for long/short term respiratory failure.

Following are some of the more interesting comments to my article, that may shed some light on this topic.

Jessica Lei wrote;

Great considerations, thank you for the article. Is it possible to retrofit a Pall filter, such as Pall BB50T Breathing Circuit Filter (since their viral filtration efficiency is much greater) to the gas outlet of the ECMO circuit? To create some sort of an extension so it will fit?

Reply;

I guess it’s possible to retrofit any filter to an outlet by using reducing/increasing connectors and different tubing sizes. As we used to say “you are not a real perfusionist until you can get 1/4″ tubing onto a 3/8″ connector” For study purposes and periods of an hour or two, the filter may be ideal to collect data on viruses or bacteria, and not put the patients at risk. Otherwise, long term use of this filter could lead to increasing resistance and pressure build up inside the hollow fibers. As you know, this is a dangerous situation for sudden gas embolism. The filters do remove some viruses, but even in a ventilator circuit they get wet, increase resistance to flow and therefore have to be changed daily?? Not quite sure of the actual size particles they remove but would be interested in finding out.

You could scavenge the membranes gas exhaust port to the atmosphere by using hospital vac, like we do in the OR. Just make sure to cut notch or place a wye connector in vacuum tubing to prevent membrane pressure buildup.

Jessica Lei wrote;

“In a recent ELSO blog, the findings of a new paper out of China were discussed. The author of the blog stated that in this Chinese letter to editor (Wang W et al: JAMA March 11, 2020; doi:10.1001/jama.2020.3786), the COVID-19 virus was detected in only <1% of patient’s blood. The blog author stated that there is no viral load in the blood and PMP fibers are molecularly tight (only molecular water can pass through the fiber, ultimately becoming condensation on the fiber interior as a result of the cold gas). The blog author also stated there is no virus in the water droplets exiting a PMP oxygenator, and therefore recommends still following precautions recommended by American College of Surgeons (ACS): having viral filters on all exit points for gas on the anesthesia machine and appropriate N95 masks for staff along with protective eyewear to safeguard in the event of loss of airway. What are your thoughts?”

Reply …

“Fortunately, or unfortunately, we are part of a medical community that requires evidenced based literature/studies to provide or change the care to our patients. We are all entitled to our opinions based on theoretical technology and anecdotal evidence. Vacuum on the gas outlet ports of extracorporeal membranes has been used by many programs for decades to eliminate anaesthetic gases in the OR, without incident. As long as the vacuum is low (<10 mmHg) and there is a connector with an open leur port, wye connector or a cut in the exhaust tubing. Used in this manner it is impossible to create excessive negative pressure on the membrane bundle. As the author states, COVID-19 transmission through blood contact/transfusion may seem to be a very low possibility in the one study they reviewed.

However, even the FDA has indicated caution in this assumption, and state that the detection of SARS-CoV-2 has only occurred in severely ill patients. They also state that patients who were in contact with this virus should refrain from blood donations for a period of 28 days, after full recovery.

The patients we place on ECMO are severely ill patients, while the patients referenced in this ELSO post and cited Chinese paper were positive COVID – 19 patients, but were not critically ill ECMO patients. Over the past few days I have been speaking with some researchers in Europe who are keen on investigating any relationship between COVID-19 blood and the gas exhaust of ECMO membranes. Their problem to date, is that they have had a small amount of COVID-19 patients in their ICU, but none of them needed ECMO. I believe we should not dismiss any potential relationship based on an educated guess. We should wait until evidence-based research confirms/dismisses any relationships. Until that day, perfusionists need to use their best judgement and knowledge to determine any course of action (if any) in relation to their ECMO patients and COVID-19 (or for that matter, any transmissible virus).”

Christos Calaritis posted a document release from Maquet regarding their official statement on their membranes and the COVID-19 viral outbreak. I believe this may have been circulated to most Perfusion programs.

Reply;

“Thanks for posting this document from Maquet Christos … For recent evidence that plasma leakage occurs through PMP membranes in COVID-19 ECMO patients, see the presentation of Dr. K. Liu, from Tokyo, Japan. His presentation was given through a recent ELSO webinar (March 30, 2020) on Japan’s experience with ECMO and COVID-19 patients. You can access the webinar through one of my recent posted articles (posted 3 days ago), which was a follow up to my posted concerns from back in February, or you can access through the ELSO web site.”

Christos replied;

“The membrane used in Japan seems to be a hollow fiber type with plasma leakage at some point, the concerns I am sure raised by clinicians around the world is whether Polymethylpentene oxygenators such as the Quadrox D, Cardiohelp, Peds Quadrox, would spew out droplets of Coronavirus out of the exhaust port of the Membrane and whether we should scavenge this.”

Reply;

“The concerns you mentioned Christos are the exact reason I brought this topic up in the February post. As you know, we have been using membranes for long term support with viral outbreaks in much greater frequency since the H1N1 outbreak (2009-2010). But we have never examined the possibility of these viruses being present in fluids coming out the exhaust ports of these membranes (insensible fluid loss or plasma leakage). I agree, plasma leakage is less likely in PMP membranes, but insensible fluid losses are common in membrane oxygenators (uncoated, coated or PMP fibers). As you know, it is routine in most (neo/ped) centers to measure the insensible fluid loss from membranes so that it can be included in the patients input/output measurements (can be as much as 40-70 ml/day/lpm gas). This is the way we maintain the patient’s fluid balance over 24 hours.

To my knowledge, there has now been three published reports of plasma leakage using three PMP oxygenators during ECMO/ECLS. There were several theories about the failures, but the actual cause of plasma leakage in these three polymethylpentene oxygenators was not determined;

Pius L et al: Perfusion 2009, 24(1):51-2 — using Medos Hilite 7000 LT

Gill MC et al: Perfusion 2015, 30(7):600-3 — using Medos Hilite 2400 LT

Liu K et al: ELSO webinar March 30, 2020 — membrane unreported.

Since the Liu report about COVID-19 presence in plasma leakage fluid, and we have no evidence of the content of insensible fluid during ECMO, it would be prudent to isolate, scavenge or analyze all fluids during ECMO/ECLS support.”

The nature of viruses is that collectively, virus particles invade the respiratory mucosa first and then infect other cells, triggering a series of immune responses and the production of cytokine storm in the body, which may be associated with the critical condition of COVID-19 patients. The presence of this virus has been found in respiratory droplets, secretions, blood and fecal matter.

In another recent open access paper released from China (Wei Zhang et al: Molecular and serological investigation of 2019-nCoV infected patients: implication of multiple shedding routes – Emerging Microbes & Infections 2020, VOL. 9 https://doi.org/10.1080/22221751.2020.1729071), the authors found the molecular presence of COVID-19 in blood and plasma, with 6 out of 15 recovering patients (40%).

The authors conclusions were; “We detected the virus in oral swabs, anal swabs and blood, thus infected patients can potentially shed this pathogen through respiratory, fecal, oral or body fluid routes. In addition, we successfully applied serology test a large population and showed this could greatly improve detection rate. We provide a cautionary warning that 2019-nCoV may be transmitted through multiple routes, including respiratory, fecal and blood.”

To conclude … Until evidence based research determines that pandemic type viruses cannot be transmitted through plasma leakage or insensible fluid losses in a membranes gas outlet port, it would probably be in the care givers best interest to investigate and isolate these fluids during ECMO/ECLS support through the use of gentle vacuum suction.

As previously stated, The ISO guidelines are accepted as the global standard of operation for blood oxygenators, and therefore limits ALL membrane oxygenators to a maximum period of use to 6 hours or less (uncoated, coated and PMP hollow fiber membranes).

To reassure clinicians that oxygenators used for long term support could maintain O2/CO2 exchange over prolonged periods of time, some manufacturers tested their devices for gas exchange and called the process Validation. To be clear, Validation testing is different than the requirements used for ISO standards.

Most countries accepted the Validation process, but until this month the USFDA did not recognize Validation periods and stated the decision to use membrane oxygenators for >6 hours were up to the physician. As of April 2020, the USFDA released their ‘nonbinding recommendations’ for membrane oxygenators used ECMO therapy (fda.gov/media/136734/download). Due to the potential of COVID-19 causing increased numbers of severe respiratory distress syndromes that may be unresponsive to conventional ventilation, the FDA recommended certain types of ECMO membranes be allowed for use longer than the maximum 6-hour period.

However, the policy is only for a limited period of time, quote:

“This policy appears to be intended to remain in effect only for the duration of the public health emergency related to COVID-19 declared by the Department of Health and Human Services (HHS), including any renewals made by the HHS Secretary in accordance with section 319(a)(2) of the Public Health Service (PHS) Act.”