Q&A with Dr. Steven M. Schwartz

The CCAS editorial board, Drs. Latham and Nasr, interviewed Dr. Steven Schwartz, first author of Pharmacological Manipulation of Peripheral Vascular Resistance in Single Ventricle Patients (Stages I, II, and III of Palliation); Current Vascular Pharmacology, 2016, 14, 58-62.Dr. Schwartz is a professor of paediatrics, a senior associate scientist, Norine Rose Chair in Cardiovascular Sciences, and Head of the Division of Cardiac Critical Care Medicine at the University of Toronto and The Hospital for Sick Children.

---

Summary
Drs. Schwartz, Alejandro Floh, and Peter Laussen recently published a review of the current concepts and understanding of systemic vascular resistance (SVR) and its interplay with pulmonary vascular resistance (PVR) in children with single ventricle palliation. The effect of pharmacologic vascular manipulation differs at each stage of single ventricle palliation pathway, given the unique anatomic considerations of each stage. After Stage I palliation, there is evidence to suggest that acute postoperative afterload reduction to a certain extent may improve cardiac output and oxygen delivery (DO2) and as such improves outcomes. The benefit of SVR manipulation, however, is unclear after the Glenn or Fontan procedures.

Question 1

Editorial board
Around the turn of the century, work by people such as yourself1, Tweddel2, Hoffman3, Migliavacca4 and others drew attention to the benefit of afterload reduction after Stage I palliation and other congenital heart surgeries. What historical trends have you seen in this regard since then?

Dr. Schwartz
The papers to which you refer really got people thinking about manipulating systemic vascular resistance (SVR) rather than pulmonary vascular resistance (PVR). When I was in training, we thought that everything was about the PVR, and the most important thing was to focus on the ventilator. The real advance in knowledge that came about from the papers you cite1-3 was that ventilator manipulation wasn’t as effective as people thought and that vasodilation was a much more useful way to go about adjusting oxygen delivery. When the initial Tweddel paper2 came out, it helped make the case that lower blood pressure (i.e. lower afterload) could be associated with better markers of oxygen delivery. The immediate thought was that it showed that vasodilation redistributed an essentially fixed cardiac output between Qp and Qs in a way that favored the Qs. The Migliavacca paper4 and subsequent work from their group was perhaps less noticed but showed quite nicely that afterload reduction may work by improving overall cardiac output, rather than redistributing a fixed amount of flow. That this would be the most effective way to improve oxygen delivery makes sense, especially in light of what we can see from an even older but nonetheless important paper by Barnea and colleagues (Circulation. 1998;98:1407-1413). This paper has several figures that represent oxygen delivery as a function of cardiac output and other variables including the Qp:Qs ratio and oxygen saturation, each of which show the relatively large increase in oxygen delivery that accompanies increased in total cardiac output compared to alteration of other variables. This makes intuitive sense when one considers that if total cardiac output were unlimited, excess Qp would not be a limitation to oxygen delivery.

The understanding generated by this work led to a lot of enthusiasm for using vasodilators in the postoperative period following the Norwood operation. Later work, particularly that from the Pediatric Heart Network (PHN) sponsored Single Ventricle Reconstruction trial (N Engl J Med 2010; 362:1980-1992), has been unable to clearly establish that any one particular approach to management is better than any other approach. Pasquali and colleagues (J Thorac Cardiovasc Surg 2012;144:915-21) looked at variability in practice following the Norwood operations in the PHN study and found that eight of 15 centers used alpha-adrenergic blockade at least some of the time, and that within those eight centers, use varied from 2% to 100% of patients. This particular study did not look at relationship of practices to outcomes, but cumulatively the data from this trial has not supported any one specific approach. Furthermore, challenges with obtaining alpha-adrenergic blocking drugs has perhaps made their use somewhat less common. At this point, I think that it is reasonable to say that understanding and applying the principles discussed in our paper in Current Vascular Pharmacology is more important than applying any specific algorithm (see below).

Question 2

Editorial board
You state that after the Norwood procedure, afterload reduction will typically increase DO2, and in some cases, it may increase the systemic blood pressure upon favorable restoration of ventricular end-systolic elastance (Ees). What is your current strategy to delineate the degree of afterload reduction that optimizes DO2? And how do you determine optimal DO2?

Dr. Schwartz
My experience is that post-operative Norwood management is very center specific, likely with good reason. Surgical and bypass technique often differs, albeit subtly, between centers and even surgeons and may present the anesthesiologist and intensive care physician with different problems to manage. The key is not so much applying a specific cocktail of drugs as it is understanding that maximizing total cardiac output is the best way to improve oxygen delivery, and that when output is limited due to problems with ventricular function or atrioventricular valve insufficiency, for example, that then distribution of that cardiac output becomes even more important.

When SVR is high, perfusion/oxygen delivery is poor and blood pressure is high, lowering blood pressure via vasodilation is an obvious approach. When blood pressure is already low and cardiac output is poor, the only way that vasodilation can help the situation is if it allows a failing ventricle to move to a more linear point on the Ees curve, or even better, move to a steeper Ees curve, representing an actual increase in contractility. Since this is far from a certainty, it would be best to pursue vasodilation under these circumstances with ECMO backup ready.

As far as delineating how much afterload reduction to use, I presume you mean how do we assess the patient for a positive effect. There are challenges in terms of knowing how to balance blood pressure and vasoactive agents. Clearly, a patient needs perfusion to occur at an adequate pressure for an optimal outcome, so there’s a limit to how much one can use vasodilating agents. In the ideal case, as delineated by George Hoffman (J Thorac Cardiovasc Surg 2004;127:738-45) is that when vasomotor tone is appropriately low, blood pressure becomes a function of cardiac output more than it does a reflection of SVR. In other words, higher blood pressure is good when it occurs at a low SVR but not when it occurs at a high SVR. Since we don’t have a specific way to measure SVR, we’re left with using a mix of indicators. Clinically, we look at the overall hemodynamics, such as blood pressure, CVP, heart rate, SaO2, urine output and near-infrared spectroscopy (NIRS), as well as acid-base status, lactic acid levels, mixed-venous oxygen saturation (SmvO2) and the ventricular and atrioventricular valve function on echocardiography to determine if our support is adequate or if we need to adjust support of either contractility and/or vasomotor tone.

Question 3

Editorial board
Recent data by Mills et al5 demonstrated that a milrinone vasodilator protocol during Stage I palliation decreased postoperative cardiac arrest and improved other postoperative markers, which corroborates with your summary. What bolus and dosing regimen do you advocate? Postoperatively, what is your regimen in titrating and eventually discontinuing afterload reduction? Do you advocate the routine use of milrinone for the hybrid procedure as well?

Dr. Schwartz
Our approach, like that of many others, is to load with milrinone on bypass and come off on milrinone at 0.5 mcg/kg/min as well as epinephrine between 0.02 and 0.05 mcg/kg/min. For our particular situation, that seems to help us keep vasomotor tone at a reasonable level and aid contractility and thus cardiac output. If the function and anatomy are good, we can titrate the blood pressure up or down as needed with appropriate agents, keeping an eye on the arterial oxygen saturation, as this can sometimes be somewhat dependent on blood pressure, and on markers of oxygen delivery (NIRS and SmvO2). If there are problems with ventricular function, then we look at potential causes, which could be high SVR, but could also be hypotension or hypoxemia as well as residual anatomic lesions. We can titrate our drugs while watching the function, and markers of cardiac output to guide us. In the end, we aim for a mean arterial pressure between 40-50 mmHg early in the OR and 45-55 mmHg in the ICU, NIRS/SmvO2 > 45% and SaO2 of about 70-85%. If we can accomplish all of these things, we find that we can generally manage the patient medically. For some patients, this means using alpha-adrenergic blocking agents; for others, it means using norepinephrine or vasopressin. In the rare instance where we can’t accomplish these goals, early mechanical support can be necessary. We have found that the hemodynamics of the hybrid procedure are quite similar to that of a standard Norwood procedure in terms of the relationship between SVR and cardiac output (Circulation. 2007 Sep 11;116(11 Suppl):I179-87), which supports using the same overall strategy.

Question 4

Editorial board
Given the benefits of an afterload reduction strategy to optimize DO2 with Stage I palliation, do you think this strategy is similarly beneficial to other neonatal populations undergoing major cardiac surgery?

Dr. Schwartz
In general, the same principles can be applied to any patient with compromised ventricular function and high SVR. That said, many patients have good left or systemic ventricular function and therefore significant reserve in terms of the degree of pressure load that can be handled without adverse effects. Neonates undergoing the arterial switch operation, for example, may benefit from more focus on left ventricular afterload due to the relatively sudden imposition of a pressure load on the ventricle as a consequence of the surgery. On the other hand, a neonate who undergoes repair of tetralogy of Fallot is more likely to have issues with right ventricular function and may benefit from increasing afterload on a normally functioning left ventricle to take advantage of the effects of such an approach on ventricular-ventricular interactions (perhaps a conversation for another day). While the PRIMACORP study (Circulation. 2003 Feb 25;107(7):996-1002), in which low and high dose milrinone was compared with placebo for preventing low cardiac output syndrome after cardiopulmonary bypass for correction of congenital heart disease, was not directed at neonates, it’s worth noting that over 70% of subjects in the placebo group did not have low cardiac output syndrome, so aggressive afterload reduction without regard to the individual situation isn’t necessary as a rule.

Question 5

Editorial board
You state that after Stage II and especially Stage III palliation, the child is unlikely to need special consideration in terms of manipulation of systemic afterload. Do you think this is all due to the progressive anatomic change from in-parallel to in-series cardiopulmonary anatomy, or might age or other factors contribute?

Dr. Schwartz
Low cardiac output syndrome seems to be much less common after stage II and III procedures. I expect that the reasons for this include the change from a parallel to a series circulation with a significantly reduced ventricular workload, at least at the time of conversion from stage I to stage II, shorter or even non-existent aortic cross-clamp times, and patient selection. Pre-operative assessment tends to eliminate those with severely compromised ventricular function from the single ventricle pathway and turn them towards a transplantation approach if possible. Certainly, blood pressure management is still important, as hypertension is common after the stage II operation, but not so much as a way to ensure oxygen delivery. When oxygen delivery is compromised after a stage II operation, it is more often the result of ongoing and severe cyanosis rather than issues with cardiac output likely to be resolved by manipulation of afterload stress. In the event that one undertakes a higher risk cavo-pumonary operation in a patient with poor ventricular function, or when intraoperative events result in poor ventricular function, similar considerations to those for the Norwood operation would apply.

Question 6

Editorial board
What do you see as important next steps in research regarding SVR and PVR in postoperative single ventricle patients?

Dr. Schwartz
As a field, we’ve spent a lot of time and energy examining drug combinations and basic physiology in this population, and as a result we’ve seen surgical mortality rates drop dramatically at many institutions. We’ve also seen that one year (and longer-term) transplant-free survival remain significant problems, and that even for those who do survive longer-term, they often contend with a number of non-cardiac morbidity issues. With that in mind, what we’re really lacking are the answers to two very important questions. The first is how does our ICU care impact brain, kidney and other organ function over the long-term, and second, what can we do as a field to improve heart function beyond the stage I hospitalization? The second question is unlikely to be addressed by focusing on intensive care hemodynamics and is more likely to be addressed through novel therapies focused on cardiac regeneration or mechanical support.

The question about organ preservation though, is ripe for research. In the past, we’ve been limited in our ability to study hemodynamics because there is only so much data that a researcher can record. Recently, the ability to collect and store continuous physiologic data (Pediatr Crit Care Med. 2017 Dec 4. doi: 10.1097/PCC.0000000000001395. [Epub ahead of print]) has made it more realistic to fully characterize the intensive care course of these complex patients and to examine that course for periods of potentially inadequate oxygen delivery. This opens the door to performing studies of complex hemodynamic states and their association with various outcomes of interest. Hopefully this will take us beyond the relatively simple, immediate and somewhat arbitrary targets I identified in my answer to question 3 to a place where our care is more evidence-based and likely to impact important, but more distant events.

Citations

1. Taeed R, Schwartz SM, Pearl JM, et al. Unrecognized pulmonary venous desaturation early after norwood palliation confounds qp:qs assessment and compromises oxygen delivery. Circulation 2001; 103:2699-704. 

2. Tweddell JS et al. Phenoxybenzamine improves systemic oxygen delivery after the Norwood procedure. Ann Thorac Surg 1999; 67: 161-7.
3. Hoffman TM et al. Efficacy and safety of milrinone in preventing low cardiac output syndrome in infants and children after corrective surgery for congenital heart disease. Circulation. 2003 Feb 25;107(7):996-1002.
4. Migliavacca F et al. Modeling of the Nor- wood circulation: effects of shunt size, vascular resistances, and heart rate. Am J Physiol Heart Circ Physiol 2001; 280: H2076-86.
5. Mills KI et al. Phosphodiesterase Inhibitor-Based Vasodilation Improves Oxygen Delivery and Clinical Outcomes Following Stage 1 Palliation. J Am Heart Assoc. 2016; 5(11). DOI: 10.1161/JAHA.116.003554