Ventricular Assist Devices

A more than moderate TR should be fixed, in particular when caused by structural problems of the TV.

Most LVAD patients suffer from venous pulmonary hypertension, i.e. as a consequence of LV failure. In these patients, pulmonary vasodilators are either uneffective or even dangerous. The best treatment is to treat LV failure, hence LVAD implantation! The only indication for pulmonary vasodilators is a situation in which the patient suffers additionally from another cause of pulmonary hypertension.

I am very restrictive concerning the use of milrinone as I am afraid of systemic vasodilation which can precipitate RV failure.

Only few data available on this. Some centers sucessfully use levosimendan in these patients. However, also levosimendan (comparable to milrinone) can cause severe systemic vasodilation which may jeopardize RV function. Sometimes, a trial with iNO is beneficial.

Gut feeling: norepinephrine 0.2-0.3 µg/kg/mun, epinephrine 0.1-0.2 µg/kg/min

All iuotropes have side effcets! In LVAD patients, I still prefer inopressors over inodilators.

Yes, sometimes, for preoperative optimization.

Inhaled nitric oxide or inhaled prostacyclins are of equal efficiency.

Yes, OLV is possible in Lvad patients. One have to be carful of the increased afterlaod due to possibe hypoxemia. Invasive arterial pressure monitoring is recommended

The Lvad is preload dependent. Potitioning that my reduce venous retourn can lead to instability and should be perfiremd carfully. Volume adjustemnet may be needed. Peritoneal insufflation sgould be made in steps. Spacial attention needs to be given to the drive line during positioning.

Many thanks for this interesting question. Cerebral oxygenation depends on cerebral oxygenation: flow and MAP. Therefore a pronounced decrease in MAP during cardiac vasoplegia syndrome, might cause a decrease on cerebral oxygenation.

The reported higher incidence of vasoplegia after pulsatile left ventricular assist device placement comes from and old publication from 1997. At that time, there were no continuous flow LVADs on the market, and thus included external or internal pulsatile LVADs. The definition of vasoplegia has an impact on the incidence. Argenziano et al used a MAP < 70 mm Hg despite Norepinpehrine administration > 8 µg/min and a CI > 2.5 l/min/m2 in the first 5 min after weaniong from cardiopulmonary bypass. It is obvious that this incidence seems quite high as a MAP of 70 mm Hg is a high cutt-off value, and the first 5 min after weaning from cardiopulmonary bypass is a quite unstable hemodynamic period due to adjusting preload, giving coagulation factors, such as protamine and blood products, and manipulation of the heart. Thus a high reported incidence on not really realistic grounds.

Thyroid disease is an independent risk factor of vasoplegia with an Odds Ratio of 2.7 (95%CI 1.0-7.0, p = 0.04). Hypothyreoidie means high TSH levels with low T4 and free T4. It was reported that hypothyreoidie is related to endothelial dysfunction (Taddei, J Clin Endocrinol Metab 2003; 88: 3731-7) with a decrease in NO release, with a decrease in cGMP and decrease in myosin dephosphorylation, and thus more netto vasoconstriction. On the other hand, higher thyroxine levels are associated with a decreased SVR (Klein et al, Thyroid hormone and the cardiovascular system, N Engl J Med 2001; 344: 501-509). Thus hypothyreoidie finally vasoconstriction, hyperthyreoidie finally vasodilation. In the publication of Patarroyo in JHLT reporting vasoplegia in HTx patients, thyroid disease (in this publication hypothyroidism) is an independent risk factor of vasoplegia. However, as the impact seems to have a less vasodilating effect, it seems controversial. Preoperative hypothyroidisms seems to have an impact on cardiac output and tissue oxygen delivery (decreased) and probably an increase in preoperative pro-and antiinflammatory biomarkers with subsequent impact on postoperative vasoplegia. So yet not clearly to be explained.

Yes, we gave did.
To summarise, there are four areas of controversy around external compressions: a) safety, b) efficacy, c) timing and d) delivery.
Resolution of these issues has been hampered by an absence of high-quality evidence and insufficient methodological detail to draw firm conclusions from published case reports.
a) Safety
This is the greatest concern because of the perceived risk of cannula dislodgement/anastomotic rupture and features in warnings issued by device manufacturers. It is probably reasonable to assume that this risk is greatest in the peri-operative period and there is limited evidence the risk reduces after healing from surgery. Our initiative was targeted at ambulance clinicians. LVAD recipients are generally discharged into the community around four weeks after surgery by which time one would expect wound healing to be well established. We concluded that risks of mechanical trauma to the LVAD/circulation interface as a result of chest compressions had probably been previously overestimated, particularly after the peri-operative period.
b) Efficacy
The most powerful resuscitative tool available is to restart an LVAD which a) was functioning well prior to stopping, b) has recently stopped. This is because the LVAD blood path is non-occlusive and there is retrograde blood flow from aorta to left ventricle during ventricular diastole if the LVAD has stopped which worsens the systemic perfusion already limited by the dysfunctional heart. It is reasonable to assume that chest compressions in a patient with an LVAD which has stopped would therefore be less effective than in a patient without an LVAD.

c) Timing
Sometimes chest compressions were administered prior to restarting the LVAD though it is unclear whether this conferred more benefit than attempting to restart the LVAD directly. As expected, short periods of chest compression were associated with better outcomes.

d) Delivery
The options are normal chest compressions (normal depth and frequency), light chest compressions or abdominal.
As chest compressions in a patient with a dysfunctional LVAD are likely to be less effective (in terms of restoration of systemic perfusion) than in a patient without an LVAD, we concluded that “light” chest compressions should not be recommended in the community. However, in the peri-operative setting, the risk of cannula dislodgement/anastomotic trauma is likely to be higher.

Abdominal chest compressions are unlikely to be particularly effective in patient with a dysfunctional LVAD for the reasons described in b). Moreover, there may be an elevated risk of right sided failure as a result of their implementation in an LVAD recipient. There is a report of a patient who received abdominal compressions peri-operatively during LVAD exchange for suspected pump thrombosis (Rottenberg et al. J Cardiothorac Surg 2011;6:91). Abdominal compressions were implemented due to intra-operative cardiac arrest and refractory hypotension (presumably due to partial/complete occlusion of the LVAD flow path with thrombus) prior to pump exchange and were chosen to allow ECMO to be commenced. Chest compressions were not selected because the surgeon did not want to compress the chest due to the perceived risk of cannula dislodgement/anastomotic rupture. It is unclear whether the LVAD was functioning at least partially during the episode. The patient expired post-operatively.

We concluded that chest compressions should be implemented as “a last resort” if attempts to restart the LVAD have failed or as a temporising measure until definitive treatment can be implemented.

Thank you for this important question. Yes, I believe there is a role for an algorithm for anaesthetists based on those I presented. Whilst there are many generalisable principles which pertain to both the intra-/perioperative and the outpatient settings it is important to remember that for the anaesthetist there are important distinctions, such as the possibility of concomitant RVAD, patient under general anaesthetic, etc. Critical considerations intra-operatively are not only a robust understanding of MCS but clear demarcation of responsibilities and good communication between the team which also involves the VAD co-ordinator/nurse and perfusionist. A key focus in the period after the cession of cardiopulmonary bypass is the preservation of right ventricular function as detailed in Professor Rex's presentation. It is important for the anaesthetist to be able to differentiate between LVAD related and medical issues which can have counterintuitive effects, e.g. a medical problem such as hypovolaemia can cause an LVAD low flow alarm. Conversely, an LVAD related problem can manifest as a medical problem.

This is a very interesting question and certainly preparedness and aupport of high risk PCI has been recently included in algorhythms. I feel that the pendulum is shifting towards Impella use compared to ECMO.

The study of Muslem looked at >240 patients with predominantly Haertmate 2 and Heartware impantatyion and assessed AKI by the KDIGO criteria. AKI criteria were metin 70% of patients,of whom nearly 50 % were AKI Satge I, about 10% in Stage II and 15 % in Stage III. These had evere influence on outcomes.

I think there is much knowledge on thyroid disease and heart failure but less in relation to LVAD. A recent study from Chicago showed that 44% of patients scheduled for LVAd had abnormal thyroid function.Despite lack of guidelines in this area, most centres aim to achieve an euthyroid state before device therapy or transplantation.

I have sought some advise for this question from Dr Bowles. We know that these patients can sleep on their side so we also believe that a short term procedure in the lateral position is permissible but extreme care should be exercised for positioning the drive line and monitoring. I would probably use TEE to ensure chamber geometry is preserved and suction phenomena are prevented and continuous blood flow to the device is maintained. Monitoring flow and pulsatility is of essence. OLV must be careful to maintain PVR as hypoventilation or high PEEP, hypoxia and hypercarbia could all destabilize RV function.