Right Ventricular Assist Devices in Right Ventricular Infarction: Do They Augment Right Ventricular Function Sufficiently to Improve Prognosis?

Nearly 50% of all patients with acute inferior myocardial infarction (MI) have involvement of the right ventricle; however only 25% of these patients develop clinically evident hemodynamic manifestations.¹ Right ventricular infarction (RV MI) occurs when there is an occlusion of the right coronary artery proximal to the acute marginal branches. There is an increased incidence and perhaps worse prognosis in patients without a history of pre-infarction angina, which may result from a lack of adequately formed collateral vessels and/or preconditioning. Prognosis is worse in the elderly, or when associated with hypotension, ST elevation >1 mm in V4R, prior MI or RV hypertrophy.^2,3 With a large infarction, the RV becomes a non-functional conduit from the systemic veins to the pulmonary circulation. When inferior myocardial infarction is complicated by RV infarction, the in-hospital mortality is 31 percent, as compared with 6 percent for patients with inferior myocardial infarction without RV involvement.⁴ Elderly patients are at particularly high risk. RV MI has an in-hospital mortality of 47 percent compared to 10 percent mortality in STEMI in the absence of RV involvement.⁵ Revascularization with thrombolysis or percutaneous coronary intervention (PCI) increases the speed of RV functional recovery and improves in-hospital outcome but there is still a high mortality rate, similar to that of cardiogenic shock of LV origin.⁶

In this issue of the journal, Weiss et al⁷ present an interesting case of RV MI treated with multivessel PCI and a right ventricular assist device (RVAD). They suggest that an RVAD may be a valuable adjunct in the treatment of RVMI in the future. Since the cost of the equipment, insertion, and management is substantial, and the prognosis of RVMI is not good, the question raised is whether this may be a valuable expenditure of resources or merely another example of the triumph of hope over reason.

Pathophysiology of Right Ventricular Infarction

A review of the hemodynamics and pathophysiology of RV MI provides a starting point to consider whether, and in what cases, an RVAD might be considered and have potential positive effect on outcomes.

Coronary Blood Flow. Coronary blood flow occurs in both systole and diastole in the right ventricle. The main blood supply is the right coronary artery (serves the lateral wall, the posterior wall and posterior interventricular septum by the post descending artery). The RV anterior (free) wall is supplied by the conus artery and the marginal branches. The septum is supplied by septal branches of the LAD; often there is a large RV septal branch from the proximal or mid RCA. RV MI can occur without inferior wall MI (non-dominant RCA, acute marginal artery occlusion). A persistently occluded proximal RCA is the main culprit. The more proximal the right coronary artery occlusion, the larger the size of the infarction. The incidence of RVMI with inferior MI is reduced with patent infarct-related arteries, compared with persistent occlusion (14 vs. 48%).⁸ RV infarction noted at necropsy usually involves the posterior septum and posterior wall; the RV free wall is spared. This discrepancy arises from a high degree of collateralization from three sources: the left coronary artery, thebesian veins and diffusion of oxygen directly from the ventricular cavity. If occlusion occurs before the RV marginal branches, and collateral blood flow from the left anterior descending coronary artery is absent, then the size of infarction generally is greater.

Susceptibility of the RV to Ischemia. The RV is a thin walled chamber shaped like a pocket, wrapped around the LV, sharing the interventricular septum and pericardium. The RV is anatomically and physiologically designed to serve the low pressure pulmonary circulation. Despite the same cardiac output as the LV, the RV muscle mass is only 15% of LV mass and RV stroke work is 25% of LV function. The same cardiac output is generated because pulmonary resistance is 10% of systemic vascular resistance. Consequently the RV has lower oxygen demand than the LV. Its ability to extract oxygen can increase during hemodynamic stress. All of these factors make the RV less susceptible to infarction than the left ventricle

RV function improves in the majority of patients, including those who are not reperfused. In the majority of survivors, the clinical (and echocardiographic/radionuclide) manifestations of RV dysfunction return to normal. This observation suggests that the RV dysfunction is usually due to ischemic myocardium that remains viable. The RV may also be protected to a greater degree by ischemic preconditioning. The presence of pre-infarction angina within 72 hours of the MI is associated with a reduction in the incidence of RV infarction, hypotension and shock. However, RV hypertrophy increases the susceptibility of the RV to ischemia and increases its incidence in IWMI and its severity.

Hemodynamics of RV infarction. With proximal occlusion of the RCA there is compromise in RV free wall perfusion resulting in diminished contractility and depressed global RV performance. RCA occlusion proximal to the RV branches reduces RV free wall perfusion, resulting in RV (free wall) dysfunction, which diminishes transpulmonary delivery of LV preload, leading to decreased cardiac output despite intact LV contractility. The RV dilates to maintain stroke volume but results in global RV systolic dysfunction resulting in elevated intrapericardial pressure which, together with increased RV diastolic pressure, shifts the interventricular septum toward the volume deprived left ventricle, further limiting LV filling.⁹

The RV waveform demonstrates a sluggish upstroke, depressed and broadened systolic peak, delayed relaxation phase and markedly diminished RV stroke work. The immediate consequence is diminished transpulmonary delivery of LV preload, leading to decreased cardiac output despite relatively preserved LV function.

There is substantial diastolic dysfunction as well. In early diastole the RV is noncompliant and dilated, leading to further impairment of RV inflow. As filling continues the noncompliant RV manifests a steep pressure-volume curve, leading to a pattern of rapid diastolic pressure elevation, observed as a dip- and plateau pattern, and in the right atrium, as an increased Y descent. The ratio of RAP: PCWP is typically > 0.8 (< 0.6 normally). When the mean right atrial pressure is within 1–5 mmHg of pulmonary arterial wedge pressure, this finding has a sensitivity of 73% and specificity of 100%¹⁰ for RV MI.

These hemodynamic alterations impair LV filling, not only indirectly by decreased transpulmonary blood flow but directly through ventricular interdependence secondary to increased pericardial pressure and shifting of the interventricular septum. The elevated RV pressures shift the interventricular septum to the left, further decreasing the LV chamber size and volume. Abrupt RV dilatation within the non-compliant pericardium acutely increases intrapericardial pressure, decreasing both RV and LV compliance and further exaggerating diastolic ventricular interactions. As both ventricles fill they compete for space within the crowded pericardium, and the effects of pericardial constraint contribute to the pattern of equalized diastolic pressures and RV dip and plateau characteristics of RVI.

Consequently, contractility of the dilated noncompliant RV is highly dependent on diastolic pressure (preload dependent). RV contractility is highly dependent on diastolic pressure (preload dependent). When contractility is impaired and diastolic dysfunction develops in RVMI, the RV diastolic pressure increases substantially and systolic pressure decreases. Ventricular interdependence with high RV diastolic pressure causes bowing of septum into LV; alteration in geometry decreases LV stroke volume. Increased RV afterload may result, and RV output decreases dramatically. The LV is also noncompliant and under filled, also raising the afterload, but the pulmonary resistance is still much lower that the systemic circulation. Therefore, conditions and or treatment that reduce ventricular preload tend to be detrimental, whereas measures that optimize cardiac filling tend to be beneficial.

The lower afterload and myocardial oxygen demand of the RV [as compared with the left] results in lower oxygen extraction at rest and explains its relative resistance to irreversible ischemic damage. Recovery of RV function is partially due to decreased oxygen demand of the thin-walled RV and to increased oxygen extraction.^11,12 Actually, relatively little RV myocardium is necrotic or irreversibly damaged; the label RV “infarct” is therefore largely a misnomer: the acutely ischemic RV is predominantly viable. The severity of the hemodynamic abnormalities associated with RV infarction is related to the extent of RV ischemia and consequent RV dysfunction as well as to the restraining effect of the pericardium, LV function, and ventricular interdependence.

Normally, the RV ejects blood into a highly compliant, low resistance circulation; compared to the LV, a much lower proportion of RV stroke work goes to pressure generation and is directed towards generating flow. When RV free wall contraction is depressed, RV systolic pressure is generated mainly by contraction of the LV septum, due to mechanical displacement of the septum into the RV cavity. Augmented right atrial contraction compensates partially to optimize cardiac output. Loss of this booster pump function, due to atrioventricular dyssynchrony or ischemic depression of RA contractility resulting from very proximal RCA occlusions, further impairs RV performance and is associated with more severe hemodynamic compromise and higher mortality. With RV contractility decreased and afterload increased, the only driving force remaining is elevated right atrial pressure. Under these circumstances, the RV serves as a poorly functioning conduit between the right atrium and the pulmonary artery.

Potential Role of RVADs. Mechanical assist devices that solely support the left ventricle could potentially worsen the hemodynamic changes in RVMI due to volume loading of a stiff and dilated RV, as described above. Further, if a balloon counterpulsation device were inserted into the PA, similar to an intra-aortic balloon pump, it could be predicted to have minimal beneficial effect, as RV afterload is not the limiting factor and the other advantages when placed in the systemic circulation of increased coronary and cerebral perfusion would not apply. Right ventricular assist devices (RVAD) are being increasingly used for hemodynamic support after RVMI. In 2006, Atiemo and colleagues reported the first percutaneous RVAD implantation in the setting of RV failure after AMI.¹³ There are now a few published reports of percutaneous RVAD implanted for RV MI.^14–16

The TandemHeart™ (Cardiac Assist, Pittsburgh, Pennsylvania) uses a 21 Fr inflow cannula positioned in the RA with a second 21 Fr outflow cannula inserted into the main pulmonary artery. As a centrifugal pump that generates continuous flow with a minimal low amplitude pulsatile component, the TandemHeart approximates native RV function. The TandemHeart maximally provides centrifugal flow up to 5.0L/min from the RA to main PA, bypassing a poorly functioning RV, substituting for its contractile function. In most cases the RVAD is set at the maximum speed (7,500 rpm), which delivers approximately 3.4 L/min of cardiac output. Weaning protocols for VAD are varied and are yet to be standardized. Most studies reviewed utilized serial echocardiography both during RVAD use to monitor for complications and during turndown to monitor hemodynamic parameters.¹⁵

There are little data published on the hemodynamic effects of RVAD on patients with RV MI. One study¹³ tracked hemodynamic changes on a single patient implanted with a RVAD 4 days post RVMI. Hemodynamics were measured 2 hours after insertion of the RVAD with an immediate increase in MAP from a baseline of 57mmHg to 69 mmHg on day 1 to 81 mmHg on the day of device removal. Right atrial pressures were reduced from a baseline of 15 mmHg to 5 mmHg; PCWP was reduced from 17 mmHg to 9mmHg. By echocardiography there was improvement in RV wall motion as well as improvement in LV function, as implied by improved cardiac output. In the current study6, the authors add further to our understanding of the hemodynamic effects of the RVAD. The device was not placed until 24 hours after PCI, and the RA pressure continued to be elevated along with the wedge pressure. Within 24 hours, the CVP decreased from 14 to 7 and the cardiac output improved to a point where pressors could be discontinued.

There are several potential disadvantages of this device, including technical difficulties in placement,¹⁴ need for anticoagulation due to risk of clot formation in the outflow cannula and possible pulmonary embolism,¹⁵ cannula migration either anterograde presenting as hypoxic respiratory failure or hemothorax; or retrograde migration into the RV resulting in decreased CO due to tricuspid regurgitation or ventricular arrhythmias. These complications can be prevented by meticulous attention to cannula placement and echocardiographic documentation of anatomical positioning. Use of the TandemHeart as an RVAD is limited by its sensitivity to loading conditions. Due to its close resemblance to native RV functioning it requires optimal LV functioning for adequate RV support. In patients with baseline pulmonary hypertension right atrial to main pulmonary artery bypass may worsen pulmonary pressures by increased flow, and lead to higher risk of lung injury.¹⁶

Conclusion

We hypothesize, based on the sparse data available in the published literature, including that of Weiss et al⁷, that RVADs improve prognosis in RV MI primarily by improving forward flow into the pulmonary artery and the left side of the heart for a sufficient period of time for the RV to recover. Increased survival may be observed due to stabilization and normalization of cardiac output, allowing RV recovery to occur, depending on the acuity of the clinical syndrome on presentation, previous damage to the LV, and residual coronary artery disease. Consequently, the RA pressure decreases and native RV systolic function improves as part of the natural history of the pathophysiology of RV MI rather than from any specific therapeutic hemodynamic effect of the device. Survival may be possible if the RVAD is placed prior to end organ damage and if coronary blood flow is optimized by revascularization. Further in depth study will determine whether this hypothesis is correct.

References

1. Showkata H, Andassad M. Right ventricular infarction—Diagnosis and treatment. Clinical Cardiology 2000;23:473–482.
2. Zehender M, Kasper W, Kauder E, et al. Right ventricular infarction as an independent predictor of prognosis after acute inferior myocardial infarction. N Engl J Med 1993;328:981–988.
3. Goldstein JA, Barzilai B, Rosamond TL, et al. Determinants of hemodynamic compromise with severe right ventricularinfarction. Circulation 1990;82:359–368
4. Bowers, TR, O'Neill, WW, Grines, C, et al. Effect of reperfusion on biventricular function and survival after right ventricular infarction. N Engl J Med 1998;338:933.
5. Bueno H, López-Palop R, Bermejo JD, López-Sendón JL, Delcán JL. In-hospital outcome of elderly patients with acute inferior myocardial infarction and right ventricular involvement. Circulation 1997;96:436–441.
6. Jacobs AK, Leopold JA, Bates E, et al. Cardiogenic shock caused by right ventricular infarction. A report from the SHOCK Registry. J Am Coll Cardiol 2003;41:1273–1279.
7. Weiss S, Jolly N, Shah AP. Multivessel intervention and placement of a percutaneous right ventricular assist device in a patient with acute myocardial infarction complicated by cardiac arrest. J Invasive Cardiol 2011;23:248–251.  
8. Berger PB, Ruocco NA Jr, Ryan TJ, et al. Frequency and significance of right ventricular dysfunction during inferior wall left ventricular myocardial infarction treated with thrombolytic therapy (Results from the Thrombolysis In Myocardial Infarction [TIMI] II trial). Am J Cardiol 1993;71:1148–1152.
9. Goldstein JA. Right versus left ventricular shock. A tale of two ventricles. J Am Coll Cardiol 2003;41;1280–1282
10. Lopez-Sendon J, Coma-Canella I, Gamallo C. Sensitivity and specificity of hemodynamic criteria in the diagnosis of acute right ventricular infarction. Circulation 1981;64:515–525.
11. Williams JF. Right ventricular infarction. Clin Cardiol 1990;13:309.
12. Cohn JN. Right ventricular infarction revisited. Am J Cardiol 1979;43:666–668. 
13. Atiemo AD, Conte JV, Heldman IW. Resuscitation and recovery from acute right ventricular failure using a percutaneous right ventricular assist device. Catheterization and Cardiovascular Interventions 2006;68:78–82 .
14. Prutkin JM, Strote JA, Stout KK. Percutaneous right ventricular assist device as support for cardiogenic shock due to right ventricular infarction. J Invasive Cardiol 2008;20;E215–E216.
15. Marquez TT, D'Cunha J, John R, Liao K, Joyce L. Mechanical support for acute right ventricular failure: Evolving surgical paradigms. J Thorac Cardiovasc Surg 2009;137:e39–e40
16. Kiernan MS, Krishnamurthy B, Kapur NK. Percutaneous right ventricular assist via the internal jugular vein in cardiogenic shock complicating an acute inferior myocardial infarction. J Invasive Cardiol 2010;22:E23–E26.

____________________________________

From Advocate Illinois Masonic Medical Center, Chicago, Illinois.
The authors report no conflicts of interest regarding the content herein.
Address for correspondence: Lloyd W. Klein, MD, Professional Office Building, 3000 North Halsted, Suite #625, Chicago, IL 60614. Email: lloydklein@comcast.net