Paul Mazurek, Survival Flight Clinical Educator
A male in his late 40’s presents to a local emergency department (ED) following a syncopal episode and subsequent fall at home. His history is significant for dilated cardiomyopathy with a low ejection fraction (EF) and congestive heart failure (CHF). He had a HeartMate 3 Ventricular Assist Device (VAD) surgically implanted into his left ventricle a year prior.
During his time in the referring ED, the patient has shown clinical signs of hypoperfusion with low flow alarms to the VAD. Laboratory results are significant for hypokalemia, an elevated Blood Urea Nitrogen (BUN), serum creatinine and Brain Natriuretic Peptide (BNP). There is also a mild leukocytosis with a left shift. The patient has 2, 20-ga peripheral IV’s, both in his left arm, has received over 4 liters of crystalloid and broad-spectrum antibiotic coverage. He is currently awake and alert, is maintaining his own airway and protective reflexes (i.e., positive cough), and lung sounds are significant for crackles in all fields. He has 3+ pitting edema to both lower extremities.
The consensus by both the transport team and referring staff is that the patient’s syncope and fall are related to the low flow alarm on the VAD. But why was the patient experiencing a low flow alarm? Initial thought was septic shock and fluid volume deficit. A natural assumption especially since the centrifugal flow properties of the pump are quite reliant upon preload (providing adequate “suction head” to the pump inlet). Given the fact that 4500 ml of crystalloid failed to improve this created even more confusion and angst for the referring staff. This is where an understanding of the device and how it incorporates itself into the patient’s intrinsic cardiac physiology is important.
Initial scrutiny of the device controller reveals a pump flow of 2.1 lpm at 5850 rpm. The cardiac rhythm upon transport team arrival showed…
Upon discovery of this rhythm (interpreted as a “polymorphic” ventricular tachycardia), the transport team made immediate preparations for cardioversion. This was at the initial reluctance of the referring physician who saw an awake and alert patient with a functioning (albeit sub-optimally) VAD. A quick parasternal long axis, parasternal short axis and apical 4-chamber view of the heart via ultrasound revealed little to no native cardiac contractility and a rapid ventricular rate. 150 mg of amiodarone was requested to be administered over 10-15 minutes and the patient was prepared for cardioversion (all the while awake, alert and understandably anxious).
The patient was sedated with modest doses of midazolam and ketamine, and received an “unsynchronized” shock at 200J (results below)
At this time, pump flow rose slightly to 2.9 lpm. The bradycardia was immediately treated with a “push-dose” pressor of epinephrine (20 mcg) with a subsequent heart-rate increase to between 80 and 100 bpm and increase in pump flow to above 3.0 lpm. A dopamine infusion was initiated to manage the bradycardia.
A venous blood gas with electrolytes revealed a metabolic acidosis, hypokalemia and hypocalcemia. The patient received appropriate electrolyte replacement as well as an infusion of sodium bicarbonate during the transport period. Subsequent blood gasses by point of care testing during transport showed improvement in pH without resolution in acidosis. Blood pressure was never obtainable during transport but the patient showed clinical signs of improved perfusion. Upon arrival to the referral center, the patient had device flows of 5.1 lpm with a pump speed of 5500 rpm.
A somewhat common misunderstanding that clinicians experience (particularly those that do not see these devices with any type of regularity) is that the ventricular assist device (VAD) can provide ALL of the required cardiac output for the patient. This case is a prime example of how volume resuscitation is not always the answer for device low flow alarms. While improving flows from the centrifugal device is important, the assistance of the “positive displacement” pump provided by the right and left ventricles cannot be underappreciated.
Blood from the right ventricle is needed by the inlet of the device in order to move it forward. Likewise, ANY native LV function can certainly improve these blood-flow characteristics. When this is removed from the equation (as in this instance) the patient will typically experience hypoperfusion and clinical signs of reduced cardiac output.
The final thought is that “Right Ventricular Resuscitation” should be in the mind of any clinician troubleshooting VAD issues. It starts with determining and / or establishing appropriate pump operation and quickly moving to address potential stroke volume issues such as preload, afterload and cardiac contractility. The video short below outlines this. While the LVAD (left-sided device) plays a significant role in prolonging life and often serves as a “bridge to transplant,” it still requires blood from the right ventricle in order to appropriately function.
