Chatter or chattering of the extracorporeal circuit access lines (aka fluttering, swinging, surging, or rattling) is a common phenomenon which can be observed (at any time) during ECMO run: the inflow line shakes randomly... what's happening??
Let's start from the beginning...
A core parameter of extracorporeal life support, the Blood Flow (BF), is generated and controlled by a pump; centrifugal pumps have become the pump of choice for ECMO. How do they work?
In centrifugal pumps a spinning rotor generates a sub-atmospheric pressure at the center of the pump-head, actively drawing blood through an upper inlet, directly connected to the inflow (access or drainage) cannula and line. The centrifugal force imparts kinetic energy to the blood, forced to the edge of the pump head and thrown out through the side outlet with a positive pressure (picture 1). PINLET or PVEN, the pump inlet pressure, is the negative pressure in the blood phase before the pump (also pre-pump or drainage pressure).
picture 1. Because of the sub-atmospheric pressure and the centrifugal force generated by a spinning rotor, blood enters the inlet at the top of the pump head, is forced to the edge and exits through a side outlet with a positive pressure, progressively decreasing due to encountered resistances (extracorporeal circuit/patient).
Usually, on ECMO support, the pump speed (in number of revolutions per minute - RPM parameter) is set, while the BF (in liters per minute - lpm) is the derived parameter (the inverse could be rarely true). When the centrifugal pump is running at high speed (= high RPM setting), in order to achieve higher BF, a considerably negative pressure could be generated in the pre-pump portion of the extracorporeal circuit. A "normal" PINLET during adult ECLS is above -100 mmHg, not below -40/-80 mmHg. However, as centrifugal pumps are non-occlusive and preload dependent/afterload sensitive, the speed is not the sole determinant of the flow: a decreased preload results in a further declining PINLET.
An inflow pressure lower than -100 mmHg may suggests impaired venous drainage, the so-called "access insufficiency", expression of a mismatch between the venous return to the access cannula and the degree of suction pressure at the set pump speed (too low return, excessive demands or a combination of both). There are many signs of access insufficiency:
- modest, palpable vibration of the access cannula and tubing;
- anomalous, major decline in PINLET, which may instantly (and fleetingly) become really low;
- a visible shacking, dancing movement of the extracorporeal drainage tubing, the most apparent and familiar sign, known as chattering (as seen in video 1). If particularly intense, chattering is referred to as kicking or kick back (as seen in video 2); sometimes chattering and kicking are simply use as synonyms. This appears to occur because the compliant vascular (Vena Cava) and eventually cardiac structures (right atria) intermittently collapse down against itself or around the drainage cannula, due to the excessive negative pressure, inducing a partial or complete occlusion of the access cannula inlet ports, transiently interrupting blood inflow. This is colloquially referred to as "suck down": the vein is sucked on to the access cannula side-holes; in seconds, ongoing venous return fills up the vein again, cannula sets free from the vessel wall, and side ports let the blood flow, in a cyclic pattern. As a result BF instantly falls, becoming erratic, pulsing.
- eventually, a swinging, back and forth movement of the drainage cannula.
video 1. Chattering inflow line on ECMO support.
video 2. Kicking inflow line on ECMO support.
Some actual ECMO circuits incorporate pressure sensors/transducers for continuous monitoring of drainage pressure and low pressures/low flow alarms (as seen in picture 2 and video 3 link https://youtu.be/uJnpt_Qhwlg), easing identification of access issues, and eventually servo regulation to stop flow if PINLET exceeds a set threshold, as safety feature (hidden risk: BF interruption = no extracorporeal support!). The insertion of a small collapsible chamber (bladder) in the venous line before the centrifugal pump, acting as a minimized reservoir and providing compliance to the circuit, has been proposed to reduce large negative PINLET spikes preventig continuing suction if the access line is occluded for more than a few seconds, but is an accessory component adding complexity to the extracorporeal circuit, and is seldom used with centrifugal pumps.
picture 2. Episode of access insufficiency on ECMO run with continuous monitoring of extracorporeal support parameters: look at the lowering inlet pressure (red circle) and fluctuating blood flow (blue asterisk) at stable pump speed (RPM setting, red asterisk).
video 3. continuous monitoring of extracorporeal support parameter during two episodes of access insufficiency with kicking line on ECMO; look at the lowering inlet pressure and fluctuating blood flow at stable pump speed (Cardiohelp system).
Complications of access insufficiency
Chattering could damage cardiovascular structures (if you have any doubt, check this video to understand what happens to IVC! originally published on ED ECMO blog, with permission). Moreover, the sudden interruptions in BF, due to the impairment of venous drainage despite a stable RPM setting, can have many other deleterious effects.
As BF becomes unstable, the degree of extracorporeal support fluctuates: even a transient decrease of the output of the pump lowers support and may not be tolerated in patients highly dependent on ECMO: hemodynamic instability and/or impairment of oxygenation can occur, which may temporary require other strategies to ensure adequate gas exchange/circulatory support while approaching the access issue. An unstable, dropping BF is prodromal for intra-circuit thrombosis due to flow turbolence/blood stagnation.
Another potential problem when the inlet line is occluded is cavitation, the occurrence of voids in the blood phase that result from a rapid decrease in fluid pressure. If the rotor of the centrifugal pump keeps spinning at high speed, ejecting blood out from the pump head but there is no preload to refill the voided space, as drainage line is occluded, instantaneously a vacuum forms within the pumphead. Each sudden interruption in BF may determine cavitation; the subsequent risk is degassing (or out-gassing), the release of blood-dissolved gases, with gaseous microemboli (GME), blood trauma and the resulting hemolysis. The evidence of GME has been observed with negative pressures of -200/250mmHg and onwards. Hemolysis is indicated by increased plasma free-haemoglobin level (fHb) above 0.50 g/dL (normal value <0.10 g/dL), decreased hemoglobin, increased bilirubin/lactate dehydrogenase [exclude other potential sources of hemolysis].
In order to avoid such complications, access insufficiency need to be promptly recognized, managed and fixed through early interventions. When chattering occurs the centrifugal pump speed should be markedly reduced (lowering RPM) to disengage the vessel wall from the drainage cannula, resolving chattering, and slowly increased to obtain a stable flow, while attempting to determine and correct the underlying cause/causes, implementing a solution.
Why chattering occurs?
One of the most common cause of access insufficiency is inadequate preload due to hypovolemia. If chattering is secondary to depletion of intravascular volume, as suggested by hemodynamic/echodynamic monitoring and by patient history (restrictive fluid therapy of aggressive diuresis? common to avoid overload/edema improving respiratory function in ARDS pts on veno-venous ECMO support), consider replacement: usually chattering resolves or is greatly improved with an intravenous fluid loading or transfusion. Consider ongoing bleeding as a potential cause too and relative hypovolemia (e.g. medication induced vasodilation).
However, this phenomenon can also occur independent of volume status, related to other several potential causes, including circuit/cannula related factors (picture 3):
• obstruction within the drainage side of extracorporeal circuit: check the access cannula and line to make sure no kinking/clamping or (less common) thrombosis is present. If a component of the ECMO circuit is obstructed by clots/thrombi and weaning is not an option, the affected portion needs to be replaced. Venous thrombosis may involve the cannulated vessel too, obstructing systemic venous return and cannula inlet ports → try to assess vessel/cannula patency with bedside echography.
• poorly positioned drainage cannula or displacement during the ECMO run → evaluate cannula location through ultrasound (transthoracic and/or transesophageal) or XR looking for eventual migration of the catheter's distal tip (e.g. into the hepatic veins). If misplacement is confirmed, achieving proper placement will result in rapid improvement in extracorporeal blood flow and resolution of chattering.
• positional flowing through drainage cannula: was the patient recently mobilized for therapeutic strategies (proning/supining/turning), physical therapy, nursing/diagnostic/interventional procedures, transportations,...? → consider optimizing patient position, carefully monitoring effects of repositioning on extracorporeal circulation parameters (things may worsen!).
• improper cannula size (too small): in order to optimize venous drainage, inflow cannula have a larger diameter than outflow ones, are usually longer and marked by the presence of multiple side holes (inlet ports for the blood inflow) over a large proportion of their length. Nevertheless, flow requests may be too high for the cannula diameter → consider replacement of the access cannula with a bigger one. If upsizing is not appropriate for vessel dimension, insertion of un additional drainage cannula may be needed if problems persists (VV-A or VV-V configurations, the two drainage lines jointed pre-pump by an Y connector).
and patient related factors:
• changes (increasing) in intra-thoracic pressure or intra-abdominal pressure related to coughing, straining, fighting the ventilator → if indicated, implement sedation and eventually administer neuromuscular blocking agents.
• any pathological process resulting in mechanical obstruction which might impair venous return: cardiac tamponade (↑ cardiac pressures), especially in postcardiotomy patients, tension pneumothorax, air trapping in severe asthma, abdominal compartment syndrome (or intra-abdominal hypertension-IAH) → decompress/treat primary cause.
• impairment of native cardiac function during veno-venous support → consider another extracorporeal support configuration if refractory to medical therapy.
• severe aortic regurgitation on veno-arterial ECMO.
picture 3. Troubleshooting in the event of chattering: systematic approach involve a prompt evaluation of the extracorporeal circuit, starting form the tip of the cannula, associated to a comprehensive and accurate monitoring of the patient, to resolve any issue before major related complications may occur. For the detailed explanation refer to the text.
A last note...
On veno-arterial ECLS, evidence of chattering of the extracorporeal outflow (arterial) line could be a sign of transmitted native cardiac function (synchronous with ventricular systole and becoming stronger as the heart function recovers) or, if an intra-aortic balloon pump is on site, rhythmic movement could be the result of the transmitted IABP inflation/deflation.
Lequier L, Horton SB, McMullan DM, Bartlett RH. Extracorporeal Membrane Oxygenation Circuitry. Pediatr Crit Care Med. 2013 Jun; 14(5 0 1): S7–12. open access link
Sidebotham D. Troubleshooting Adult ECMO. J Extra Corpor Technol. 2011 Mar; 43(1): P27–P32.
Pediatr Crit Care Med. Author manuscript; available in PMC 2014 Jun 1. open access link
Tulman DB, Stawicki SPA, Whitson BA, Gupta SC, Tripathi RS et al. Veno-venous ECMO: a synopsis of nine key potential challenges, considerations, and controversies. BMC Anesthesiol. 2014; 14: 65. open access link
Simons AP, Ganushchak YM, Teerenstra S, Bergmans DC, Maessen JG, Weerwind PW. Hypovolemia in extracorporeal life support can lead to arterial gaseous microemboli. Artif Organs. 2013 Mar;37(3):276-82. link
Fiza B, Tang M, Maile M. Management of cardiopulmonary assist devices in critically ill patients using point-of-care transthoracic echocardiography: a case series. Crit Ultrasound J. 2017; 9: 24. open access link
Douflé G, Roscoe A, Filio Billia F, Fan E. Echocardiography for adult patients supported with extracorporeal membrane oxygenation. Crit Care. 2015; 19: 326. open access link
Mossadegh C. Monitoring the ECMO in Nursing Care and ECMO. Mossadegh C, Combes A Ed, Springer 2017 link
Borrelli U, Costa C. Materials: Cannulas, Pumps, Oxygenators and Hockings L, Vuylsteke A. Troubleshooting Common and Less Common Problems in ECMO - Extracorporeal Life Support in Adults. Sangalli F, Patroniti N, Pesenti A. Ed, Springer 2014 link
and some other #FOAMed #FOAMcc resources
Complications and Troubleshooting ED ECMO blog open access link
Chatter causes IVC Trauma! ED ECMO blog open access link
An ECMO Earthquake? Intensive blog open access link
ECMO Troubleshooting: 3 Problems You NEED to Recognize Early! Maryland CC Project blog open access link
ECMO Pump Chatter Resus Review blog open access link
Published on August 01, 2018