CASE REPORT
Left Main Coronary Artery Occlusion and the Collateral-Dependent Heart Benjamin N. Morris, MD,* Adair Q. Locke, MD,* Kristopher M. Foote, MD,†‡ and Roger L. Royster, MD*
T
HE PRESENCE OF collateral circulation in areas of myocardium with insufficient perfusion was noted as early as 1969 when Richard Lower described anastomoses between coronary arteries.1 Understanding the nature of collateral vessels and being aware of their specific pharmacologic responses are essential to protecting the myocardium supplied by these vessels. The authors present a case of chronic complete left main coronary artery occlusion in which the entire left circulation of the heart was supplied by collateral vessels from the right coronary circulation and from noncoronary collateral circulation. This case documents a rare finding and provides for discussion of important considerations regarding anesthetic management of patients with critical coronary disease and coronary collateral vessels. CASE DESCRIPTION
A 64-year-old man was referred to the authors’ hospital with new-onset symptoms of dyspnea and orthopnea, with a history of diabetes, hypertension, and peripheral vascular disease. Physical examination was remarkable for a II/VI systolic murmur heard best over the left lower sternal border and apex, basilar rales, and absence of lower extremity edema. A chest radiograph showed some mild interstitial fluid and borderline cardiomegaly. An electrocardiogram revealed a left bundle-branch block with Twave inversion across the lateral leads but no Q waves. Transthoracic echocardiogram showed global left ventricular hypokinesis with an ejection fraction (EF) of 20%, moderate-to-severe mitral regurgitation, and mild-to-moderate tricuspid regurgitation. The right ventricle appeared normal and there was mild pulmonary hypertension estimated by tricuspid flow velocities. Dobutamine stress magnetic resonance imaging demonstrated functional and metabolic responses indicating hibernating, viable myocardium without evidence of significant scarring. Coronary angiography prior to surgery revealed complete occlusion of the distal left main coronary artery (Fig 1), which appeared chronic in nature due to the well-developed collateral circulation. The left coronary circulation was visible primarily due to intercoronary collateral filling from the right coronary circulation (Figs 2–4). All major vessels of the left coronary circulation appeared suitable for coronary artery bypass grafting. The patient was placed on a heparin infusion and the cardiac surgery team was consulted, with coronary artery surgery and mitral valve repair/replacement scheduled for the following day. The chief senior cardiothoracic surgeon contacted the cardiac anesthesia team, voicing his concern about the case. A preoperative intraaortic balloon pump was discussed, but was not inserted due to the patient’s significant bilateral iliac vasoocclusive disease extending into the distal abdominal aorta. The importance of maintaining coronary collateral flow was discussed.
The anesthesia plan was to maintain both systolic and diastolic pressure at or above baseline with phenylephrine prior to, during, and after anesthetic induction. A nitroglycerin (NTG) infusion would be added to maintain collateral flow following anesthetic induction and central venous catheter insertion and to potentially reduce the amount of mitral regurgitation as guided by transesophageal echocardiography (TEE). The heparin infusion would be continued until bolus heparin was administered for full anticoagulation prior to cardiopulmonary bypass (CPB). Vasopressin was to be contraindicated prior to coronary revascularization. Since noncoronary collateral flow was likely, mean pressure on CPB was planned to be 50 mmHg to 60 mmHg to reduce washout of cardioplegia. On the morning of surgery, a left radial arterial catheter was placed prior to the induction of anesthesia during mild sedation with intravenous midazolam, 2 mg, while vital signs were monitored. There were no concerns about the patient’s airway. An intravenous induction was accomplished with fentanyl, 10 µ/kg, followed immediately with pancuronium, 0.1 mg/kg, and etomidate, 0.1 mg/kg. Intermittent boluses of phenylephrine were administered to maintain systemic resistance and coronary perfusion pressure (CPP) at baseline. Maintenance anesthesia consisted of isoflurane, 0.2% to 0.8%, and a fentanyl infusion at 100 µ/h to 150 µ/h. After induction, a TEE probe and a pulmonary artery catheter were placed for monitoring. Cardiac index was initially 1.8 L/min/m2, SvO2 was 83%, and pulmonary artery pressures were 40/22 mmHg, while blood pressure was 110/80 mmHg. Nitroglycerin was started at 20 µ/min with phenylephrine infusion to maintain CPP. Cardiac index increased to 2.1 to 2.2 L/min/m2, while pulmonary diastolic pressure decreased to 16 mmHg to 18 mmHg and SvO2 increased to 87%. The TEE revealed markedly reduced left ventricular function, with an EF of 20% and moderate-tosevere mitral regurgitation with restrictive-appearing leaflets.
From the *Department of Anesthesiology, Wake Forest School of Medicine, Winston-Salem, NC; †Department of Internal Medicine, New Hanover Regional Medical Center, Wilmington, NC; and ‡Department of Anesthesiology, University of Utah School of Medicine, Salt Lake City, UT. Address reprint requests to Benjamin N. Morris, MD, Department of Anesthesiology, Wake Forest School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1009. E-mail: bmorris@ wakehealth.edu © 2015 Elsevier Inc. All rights reserved. 1053-0770/2602-0033$36.00/0 http://dx.doi.org/10.1053/j.jvca.2015.04.003 Key words: coronary occlusion, coronary circulation collaterals, coronary artery disease
Journal of Cardiothoracic and Vascular Anesthesia, Vol ], No ] (Month), 2015: pp ]]]–]]]
1
2
Fig 1. Left anterior oblique cranial image from coronary angiograms revealing complete occlusion of the distal left main coronary artery.
MORRIS ET AL
Fig 3. Left anterior oblique image showing collateral flow predominantly from the posterior lateral branch filling retrograde into the left coronary circulation. Arrows illustrate large grade 3-collaterals.
A large left-to-right intra-atrial septal defect was discovered. Diastolic blood pressure was maintained at 80 mmHg until time to prepare for aortic cannulation. The heparin infusion was discontinued after 300 units/kg of heparin were administered. Intraoperatively, the mitral valve was exposed via a superior septal incision, and a large atrial septal defect was observed. There was no calcification of the mitral leaflets or of the annulus, but the leaflets and subvalvular apparatus were scarred and tethered. The valve was not amenable to repair. The
annulus sized to a 31-mm St. Jude Medical Epic Supra Stented Porcine Bioprosthesis (St. Jude Medical, Inc., St. Paul, MN). Four-vessel coronary artery bypass grafting was performed with vein grafts to the posterior descending, first obtuse marginal, and first diagonal arteries with an internal mammary artery graft to the left anterior descending. It was noted during distal anastomoses that continuous suction was required to
Fig 2. Left anterior oblique cranial image of enlarged right coronary artery with irregularities and the posterior lateral and posterior descending arteries. Arrows denote the critical lesions of the posterior descending artery.
Fig 4. Right anterior oblique cranial image showing collateral filling through the septal perforators to the left anterior descending (Arrow A) and a large diagonal branch (Arrow B) and the circumflex artery (Arrow C).
3
CORONARY OCCLUSION AND DEPENDENT HEART
visualize the anastomotic site despite intermittent retrograde cardioplegia that was indicative of significant noncoronary collateral flow. The proximal anastomoses were performed with a single cross-clamp technique. The patient separated uneventfully from CPB on inotropic support (milrinone and epinephrine) and was transported to the intensive care unit in stable condition. The patient had a stormy postoperative course that included respiratory failure and tracheostomy. Postoperative transthoracic echocardiography revealed the EF had improved to 30% to 35%, with prosthetic valve function normal. The patient eventually was discharged from the hospital with improvement in heart failure symptoms and no anginal symptoms.
DISCUSSION
A patient surviving a total left main coronary artery occlusion is extremely rare. This case was the first total left main occlusion discovered at angiography at the authors’ institution. Zimmern et al reported total occlusion of the left main coronary artery in only 12 (0.06%) of 20,197 patients before cardiac surgery.2 Collateral blood flow is absolutely necessary to survive such an occlusion of the coronary circulation. By definition, coronary collateral vessels provide anastomotic arterial connections without passing through a capillary bed, either between coronary arteries (intercoronary collaterals), between branches of the same artery (intracoronary collaterals), or from outside the native coronary circulation (noncoronary collaterals) (Table 1). These vessels are recruited to supply the ischemic myocardium and develop in the presence of occlusive coronary artery disease.3 Coronary collaterals develop through 2 primary mechanisms: Arteriogenesis and angiogenesis. Angiogenesis is the development of new capillaries from pre-existing arteries via stimulation through hypoxia-inducible factor 1-α (HIF1-α). While present in humans, this mechanism does not appear to be the primary mechanism of collateral formation, at least in the setting of single coronary occlusion where arteriogenesis appears to predominate.4 In arteriogenesis, coronary collateral vessels arise from preformed arterioles and are visible at angiography within hours5,6 to a few weeks after occlusion (Table 2).7 Maturation takes time, however, and collaterals take on the appearance of normal coronary arteries within 12 months.8 Since coronary collateral vessel patency can be critical in maintaining blood flow to jeopardized regions of myocardium, understanding the physiology and pharmacologic responses of the coronary collateral circulation is essential in Table 1. Sources of Coronary Collateral Flow Intracoronary collateral vessels: Collateral vessels that develop around areas of coronary occlusion within the same coronary artery. Intercoronary collateral vessels: Collateral vessels that develop between a coronary artery with normal or reduced flow to a vessel that is totally occluded. Noncoronary collateral vessels: Collateral vessels that develop from arteries outside of the coronary circulation to occluded vessels within the coronary circulation. These collaterals have been described from bronchial arteries, pericardial arteries, pulmonary veins, and internal mammary arteries.
Table 2. Grading of Coronary Collateral Vessels at Angiography Grade 0 Grade 1 Grade 2 Grade 3
No visible coronary collaterals Visible collateral vessels without filling of the occluded epicardial coronary artery Visible collateral vessels that partially fill an occluded epicardial coronary artery Visible collateral vessels that totally fill an occluded epicardial coronary artery
managing patients with coronary occlusion during cardiac anesthesia and surgery. Within normal coronaries, α-adrenergic stimulation causes mild vasoconstriction. However, overall myocardial blood flow is dependent on perfusion pressure, as well as coronary vascular resistance, and myocardial blood flow response to an α-agonist like phenylephrine either is unchanged or increases.9 Collateral vessels, however, do not constrict to α1-stimulation and are very pressure-dependent in maintaining collateral flow. Histologic studies have demonstrated the absence of α1-receptors within mature collaterals.10 Interestingly, studies have shown that prazosin infusions increase blood flow more to collateral-dependent regions versus noncollateral areas during exercise.11 β-Adrenergic antagonists traditionally have been considered cardioprotective; however, some data suggest that β-antagonists may decrease collateral blood flow. Billinger et al showed that Doppler coronary collateral flow velocity was reduced after metoprolol infusion during coronary angiography.12 This is in agreement with a similar study using the nonselective drug propranolol in which coronary collateral flow was reduced as well.13 Work by Traverse et al found that propranolol increased the resistance across the collateral vascular bed with resultant decreased flow.14 Increased resistance is likely due to inhibition of β-mediated vasodilation rather than unmasked α-adrenergic effects, as α-adrenergic receptors do not play a prominent functional role in collateral vessels.15 This effect is probably most relevant when external pacing is used to prevent rate reduction in the heart after β-blockade, thereby preventing a decrease in myocardial oxygen assumption. Despite the acute pharmacologic effects, it should be noted that chronic β-blockade is a predictor for increased collateral formation in patients with coronary artery disease. Some have proposed this phenomenon as a possible etiology for the improved benefits of long-term β-blockade use.16 Vasopressin causes vasoconstriction via the V1 receptor on vascular smooth muscle. Mature collateral vessels are hyperresponsive to vasopressin, producing a vasoconstriction greater than that of the normal coronary vessels. Even vasopressin levels encountered during periods of physiologic stress (major surgery, hemorrhage, bypass, etc) may result in vasoconstriction capable of diminishing flow to collateral-dependent myocardium and may result in ischemia in those tissues.17 Additionally, enhancement of the response of collateraldependant vessels to vasopressin has been seen as well.18 Vasodilation of coronary collateral vessels in response to NTG administration is a well-known effect of this drug.19 Collateral vessels not already maximally dilated via nitric oxide mechanisms displayed an enhanced vasodilatory response to
4
MORRIS ET AL
NTG, and systolic contractile function improves in the region of myocardium supplied.20 Given the lack of collateral vasoconstriction with α-agonists, the combination of phenylephrine and NTG appears to be synergistic.20 Even in the coronary circulation without collateralization, the combination of phenylephrine and NTG increases coronary flow to ischemic myocardium by maintaining CPP with phenylephrine while vasodilating coronary vessels into the ischemic areas with NTG.9,21 Heparin frequently is administered to patients with critical coronary lesions not only to prevent thrombosis and occlusion of stenotic vessels, but to cause vasodilation of coronary collateral vessels.22 Thus, NTG, heparin, and endogenous adenosine are the 3 primary vasodilators of the coronary collateral circulation. Noncoronary collaterals or extracardiac collateral vessels develop during critical stenoses when coronary collateral flow is inadequate. Noncoronary collaterals have been reported to occur in 20% or more of patients with normal hearts.23 Varying anastomotic connections between the coronary circulation and internal mammary arteries, bronchial arteries, pericardial arteries, and pulmonary veins have been described.24 Flow from noncoronary collateral sources can be seen during coronary revascularization when a coronary artery is opened for grafting, and continous suctioning of blood flowing in the artery is required when the aortic is cross-clamped and no cardioplegia is being administered, as was seen in the authors’ patient. Noncoronary collateral flow is also pressure-dependent, and, thus, maintaining a high mean pressure pre-CPB is of critical importance. However, if mean pressure on CPB is allowed to be too high, myocardial protection could be affected negatively as cardioplegia could be washed out of the coronary circulation more rapidly since noncoronary collateral flow is not affected by aortic crossclamping.25 Collateral-supplied myocardium also can be placed at risk by disruption of the collateral flow during the surgical procedure. In a study lookng to determine the flow characteristics of extracardiac collaterals, the extracollateral flow was shown to strongly correlate with the laterality of the lesion—left anterior descending extracardiac collaterals arising from the left internal mammary artery (LIMA) and right coronary artery extracardiac collaterals arising from the right internal mammary artery. Given that the authors’ patient had an occluded left main coronary, the common practice of using the LIMA as a left anterior descending graft theoretically places patients at risk for ischemia during the prepump phase when the LIMA is taken down.26 Another unique intraoperative consideration is related to the concept of coronary steal as it applies to coronary collaterals. As coronary collaterals develop to supply ischemic areas, the coronary arteries from which the collaterals develop are prone to occlusion themselves. If this obstruction is distal to the takeoff of the collateral, then flow will increase preferentially down the collateral. If the postcollateral obstruction is relieved, then flow will decrease down the collateral a left anterior descending
Table 3. Unique Coronary Collateral Characteristics 1) Anatomic a) Derived from pre-existing arterioles (arteriogenesis) or de b) c)
novo (angiogenesis) Flow more pressure driven due to less autoregulation May be extracardiac in origin
2) Pharmacologic a) Alpha-agonists—Cause vasoconstriction in normal coronary
b)
c)
arteries; however, do not appear to have any effect on collateral arteries Beta-agonists—Mild vasoconstriction during acute infusion, but decrease in oxygen supply likely disproportionate to decrease in overall oxygen consumption; chronic use appears to be associated with greater quantity of collaterals Vasopressin agonist—Exaggerated response compared to normal coronary flow
3) Physiologic a) Disruption of IMA can disrupt collateral flow to the b) c)
ipsilateral side Collateral flow may impair cardioplegia activity by premature washout Restoring flow to obstructed coronary lesions potentially can induce coronary collateral steal
Abbreviation: IMA, internal mammary artery.
a steal phenomenon due to the change in pressure gradients, analogous to the pharmocologic steal phenomenon seen with coronary vasodilators such as adenosine and persantine.4,27 Conceivably, this would manifest most dramatically when totally occluded coronaries are recanalized; severe (not totally occluded) lesions that are revascularized through percutaneous techniques are statistically more likely seen clinically. It is also theoretically possible that during off-pump coronary artery bypass graft procedures, the use of intracoronary shunts temporarily could induce the steal phenomenon. Intraaortic balloon pump counterpulsation theoretically should enhance coronary collateral flow with balloon inflation during diastole. However, there are very few studies in humans documenting this effect. External counterpulsation during diastole has demonstrated improvement in coronary collateral flow in patients with chronic coronary artery disease.28 The risks of an intra-aortic balloon pump should be considered versus any benefits, as was done in the patient discussed. In summary, patients with coronary occlusions often develop coronary collateral circulation that provides critical flow to regions of ischemic but still viable myocardium, resulting in reduced mortality.29 Awareness of the physiology of the collateral and noncoronary collateral circulations and the response to various vasoactive drugs can guide patient management and avoid potentially dangerous problems intraoperatively in patients with critical coronary stenoses, such as a total left main coronary occlusion (Table 3).
REFERENCES 1. Schaper W: The physiology of the collateral circulation in the normal and hypoxic myocardium. Ergeb Physiol 63:102-145, 1971
2. Zimmern SH, Rogers WJ, Bream PR, et al: Total occlusion of the left main coronary artery: The Coronary Artery Surgery Study (CASS) experience. Am J Cardiol 49:2003-2010, 1982
CORONARY OCCLUSION AND DEPENDENT HEART
3. Koerselman J, van der Graaf Y, de Jaegere PP, et al: Coronary collaterals: An important and underexposed aspect of coronary artery disease. Circulation 107:2507-2511, 2003 4. Zimarino M, D’Andreamatteo M, Waksman R, et al: The dynamics of the coronary collateral circulation. Nat Rev Cardiol 11: 191-197, 2014 5. Rentrop KP, Cohen M, Blanke H, et al: Changes in collateral channel filling immediately after controlled coronary artery occlusion by an angioplasty balloon in human subjects. J Am Coll Cardiol 5: 587-592, 1985 6. Cohen M, Sherman W, Rentrop KP, et al: Determinants of collateral filling observed during sudden controlled coronary artery occlusion in human subjects. J Am Coll Cardiol 13:297-303, 1989 7. Seiler C: The human coronary collateral circulation. Eur J Clin Invest 40:465-476, 2010 8. Wechsler AS: Development of coronary collateral circulation. Annu Rev Med 28:341-348, 1977 9. Miller RR, Awan NA, DeMaria AN, et al: Importance of maintaining systemic blood pressure during nitroglycerin administration for reducing ischemic injury in patients with coronary disease. Effects on coronary blood flow, myocardial energetics and left ventricular function. Am J Cardiol 40:504-508, 1977 10. Harrison DG, Chilian WM, Marcus ML: Absence of functioning adrenergic receptors in mature canine coronary collaterals. Circ Res 59: 133-142, 1986 11. Bache RJ, Dai XZ, Herzog CA, et al: Effects of nonselective and selective alpha 1-Adrenergic blockade on coronary blood flow during exercise. Circ Res 61:II36-II41, 1987 12. Billinger M, Raeber L, Seiler C, et al: Coronary collateral perfusion in patients with coronary artery disease: Effect of metoprolol. Eur Heart J 25:565-570, 2004 13. Kyriakides ZS, Kolettis T, Antoniadis A, et al: Beta-adrenergic blockade decreases coronary collateral blood flow in patients with coronary artery disease. Cardiovasc Drugs Ther 12:551-559, 1998 14. Traverse JH, Altman JD, Kinn J, et al: Effect of beta-adrenergic receptor blockade on blood flow to collateral-dependent myocardium during exercise. Circulation 91:1560-1567, 1995 15. Harrison DG, Chilian WM, Marcus ML: Absence of functioning alpha-adrenergic receptors in mature canine coronary collaterals. Circ Res 59:133-142, 1986
5
16. van der Hoeven NW, Teunissen PF, Werner GS, et al: Clinical parameters associated with collateral development in patients with chronic total coronary occlusion. Heart 99:1100-1105, 2013 17. Peters KG, Marcus ML, Harrison DG: Vasopressin and the mature coronary collateral circulation. Circulation 79:1324-1331, 1989 18. Sellke FW, Quillen JE, Brooks LA, et al: Endothelial modulation of the coronary vasculature in vessels perfused via mature collaterals. Circulation 81:1938-1947, 1990 19. Klassen CL, Traverse JH, Bache RJ: Nitroglycerin dilates coronary collateral vessels during exercise after blockade of endogenous NO production. Am J Physiol 277:H918-H923, 1999 20. Bache RJ: Effect of nitroglycerin and arterial hypertension on myocardial blood flow following acute coronary artery occlusion in the dog. Circulation 57:557-562, 1978 21. Borer JS, Redwood DR, Levitt B, et al: Reduction in myocardial ischemia with nitroglycerin or nitroglycerin plus phenylephrine administered during acute myocardial infarction. N Engl J Med 293: 1008-1012, 1975 22. Fujita M, Sasayama S, Asanoi H, et al: Improvement of treadmill capacity and collateral circulation as a result of exercise with heparin pretreatment in patients with effort angina. Circulation 77:1022-1029, 1988 23. Björk L: Angiographic demonstration of extracardial anastomoses to the coronary arteries. Radiology 87:274-277, 1966 24. Loukas M, Hanna M, Chen J, et al: Extracardiac coronary arterial anastomoses. Clin Anat 24:137-142, 2011 25. Olinger GN, Bonchek LI, Geiss DM: Noncoronary collateral distribution in coronary artery disease. Ann Thorac Surg 32: 554-557, 1981 26. Stoller M, de Marchi SF, Seiler C: Function of natural internal mammary-to-coronary artery bypasses and its effect on myocardial ischemia. Circulation 129:2645-2652, 2014 27. Farhad H, Murthy VL: Pharmacologic manipulation of coronary vascular physiology for the evaluation of coronary artery disease. Pharmacol Ther 140:121-132, 2013 28. Gloekler S, Meier P, de Marchi SF, et al: Coronary collateral growth by external counterpulsation: A randomised controlled trial. Heart 96:202-207, 2010 29. Meier P, Hemingway H, Lansky AJ, et al: The impact of the coronary collateral circulation on mortality: A meta-analysis. Eur Heart J 33:614-621, 2012
Left Main Coronary Artery Occlusion and the Collateral-Dependent Heart Benjamin N. Morris, MD,* Adair Q. Locke, MD,* Kristopher M. Foote, MD,†‡ and Roger L. Royster, MD*
T
HE PRESENCE OF collateral circulation in areas of myocardium with insufficient perfusion was noted as early as 1969 when Richard Lower described anastomoses between coronary arteries.1 Understanding the nature of collateral vessels and being aware of their specific pharmacologic responses are essential to protecting the myocardium supplied by these vessels. The authors present a case of chronic complete left main coronary artery occlusion in which the entire left circulation of the heart was supplied by collateral vessels from the right coronary circulation and from noncoronary collateral circulation. This case documents a rare finding and provides for discussion of important considerations regarding anesthetic management of patients with critical coronary disease and coronary collateral vessels. CASE DESCRIPTION
A 64-year-old man was referred to the authors’ hospital with new-onset symptoms of dyspnea and orthopnea, with a history of diabetes, hypertension, and peripheral vascular disease. Physical examination was remarkable for a II/VI systolic murmur heard best over the left lower sternal border and apex, basilar rales, and absence of lower extremity edema. A chest radiograph showed some mild interstitial fluid and borderline cardiomegaly. An electrocardiogram revealed a left bundle-branch block with Twave inversion across the lateral leads but no Q waves. Transthoracic echocardiogram showed global left ventricular hypokinesis with an ejection fraction (EF) of 20%, moderate-to-severe mitral regurgitation, and mild-to-moderate tricuspid regurgitation. The right ventricle appeared normal and there was mild pulmonary hypertension estimated by tricuspid flow velocities. Dobutamine stress magnetic resonance imaging demonstrated functional and metabolic responses indicating hibernating, viable myocardium without evidence of significant scarring. Coronary angiography prior to surgery revealed complete occlusion of the distal left main coronary artery (Fig 1), which appeared chronic in nature due to the well-developed collateral circulation. The left coronary circulation was visible primarily due to intercoronary collateral filling from the right coronary circulation (Figs 2–4). All major vessels of the left coronary circulation appeared suitable for coronary artery bypass grafting. The patient was placed on a heparin infusion and the cardiac surgery team was consulted, with coronary artery surgery and mitral valve repair/replacement scheduled for the following day. The chief senior cardiothoracic surgeon contacted the cardiac anesthesia team, voicing his concern about the case. A preoperative intraaortic balloon pump was discussed, but was not inserted due to the patient’s significant bilateral iliac vasoocclusive disease extending into the distal abdominal aorta. The importance of maintaining coronary collateral flow was discussed.
The anesthesia plan was to maintain both systolic and diastolic pressure at or above baseline with phenylephrine prior to, during, and after anesthetic induction. A nitroglycerin (NTG) infusion would be added to maintain collateral flow following anesthetic induction and central venous catheter insertion and to potentially reduce the amount of mitral regurgitation as guided by transesophageal echocardiography (TEE). The heparin infusion would be continued until bolus heparin was administered for full anticoagulation prior to cardiopulmonary bypass (CPB). Vasopressin was to be contraindicated prior to coronary revascularization. Since noncoronary collateral flow was likely, mean pressure on CPB was planned to be 50 mmHg to 60 mmHg to reduce washout of cardioplegia. On the morning of surgery, a left radial arterial catheter was placed prior to the induction of anesthesia during mild sedation with intravenous midazolam, 2 mg, while vital signs were monitored. There were no concerns about the patient’s airway. An intravenous induction was accomplished with fentanyl, 10 µ/kg, followed immediately with pancuronium, 0.1 mg/kg, and etomidate, 0.1 mg/kg. Intermittent boluses of phenylephrine were administered to maintain systemic resistance and coronary perfusion pressure (CPP) at baseline. Maintenance anesthesia consisted of isoflurane, 0.2% to 0.8%, and a fentanyl infusion at 100 µ/h to 150 µ/h. After induction, a TEE probe and a pulmonary artery catheter were placed for monitoring. Cardiac index was initially 1.8 L/min/m2, SvO2 was 83%, and pulmonary artery pressures were 40/22 mmHg, while blood pressure was 110/80 mmHg. Nitroglycerin was started at 20 µ/min with phenylephrine infusion to maintain CPP. Cardiac index increased to 2.1 to 2.2 L/min/m2, while pulmonary diastolic pressure decreased to 16 mmHg to 18 mmHg and SvO2 increased to 87%. The TEE revealed markedly reduced left ventricular function, with an EF of 20% and moderate-tosevere mitral regurgitation with restrictive-appearing leaflets.
From the *Department of Anesthesiology, Wake Forest School of Medicine, Winston-Salem, NC; †Department of Internal Medicine, New Hanover Regional Medical Center, Wilmington, NC; and ‡Department of Anesthesiology, University of Utah School of Medicine, Salt Lake City, UT. Address reprint requests to Benjamin N. Morris, MD, Department of Anesthesiology, Wake Forest School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1009. E-mail: bmorris@ wakehealth.edu © 2015 Elsevier Inc. All rights reserved. 1053-0770/2602-0033$36.00/0 http://dx.doi.org/10.1053/j.jvca.2015.04.003 Key words: coronary occlusion, coronary circulation collaterals, coronary artery disease
Journal of Cardiothoracic and Vascular Anesthesia, Vol ], No ] (Month), 2015: pp ]]]–]]]
1
2
Fig 1. Left anterior oblique cranial image from coronary angiograms revealing complete occlusion of the distal left main coronary artery.
MORRIS ET AL
Fig 3. Left anterior oblique image showing collateral flow predominantly from the posterior lateral branch filling retrograde into the left coronary circulation. Arrows illustrate large grade 3-collaterals.
A large left-to-right intra-atrial septal defect was discovered. Diastolic blood pressure was maintained at 80 mmHg until time to prepare for aortic cannulation. The heparin infusion was discontinued after 300 units/kg of heparin were administered. Intraoperatively, the mitral valve was exposed via a superior septal incision, and a large atrial septal defect was observed. There was no calcification of the mitral leaflets or of the annulus, but the leaflets and subvalvular apparatus were scarred and tethered. The valve was not amenable to repair. The
annulus sized to a 31-mm St. Jude Medical Epic Supra Stented Porcine Bioprosthesis (St. Jude Medical, Inc., St. Paul, MN). Four-vessel coronary artery bypass grafting was performed with vein grafts to the posterior descending, first obtuse marginal, and first diagonal arteries with an internal mammary artery graft to the left anterior descending. It was noted during distal anastomoses that continuous suction was required to
Fig 2. Left anterior oblique cranial image of enlarged right coronary artery with irregularities and the posterior lateral and posterior descending arteries. Arrows denote the critical lesions of the posterior descending artery.
Fig 4. Right anterior oblique cranial image showing collateral filling through the septal perforators to the left anterior descending (Arrow A) and a large diagonal branch (Arrow B) and the circumflex artery (Arrow C).
3
CORONARY OCCLUSION AND DEPENDENT HEART
visualize the anastomotic site despite intermittent retrograde cardioplegia that was indicative of significant noncoronary collateral flow. The proximal anastomoses were performed with a single cross-clamp technique. The patient separated uneventfully from CPB on inotropic support (milrinone and epinephrine) and was transported to the intensive care unit in stable condition. The patient had a stormy postoperative course that included respiratory failure and tracheostomy. Postoperative transthoracic echocardiography revealed the EF had improved to 30% to 35%, with prosthetic valve function normal. The patient eventually was discharged from the hospital with improvement in heart failure symptoms and no anginal symptoms.
DISCUSSION
A patient surviving a total left main coronary artery occlusion is extremely rare. This case was the first total left main occlusion discovered at angiography at the authors’ institution. Zimmern et al reported total occlusion of the left main coronary artery in only 12 (0.06%) of 20,197 patients before cardiac surgery.2 Collateral blood flow is absolutely necessary to survive such an occlusion of the coronary circulation. By definition, coronary collateral vessels provide anastomotic arterial connections without passing through a capillary bed, either between coronary arteries (intercoronary collaterals), between branches of the same artery (intracoronary collaterals), or from outside the native coronary circulation (noncoronary collaterals) (Table 1). These vessels are recruited to supply the ischemic myocardium and develop in the presence of occlusive coronary artery disease.3 Coronary collaterals develop through 2 primary mechanisms: Arteriogenesis and angiogenesis. Angiogenesis is the development of new capillaries from pre-existing arteries via stimulation through hypoxia-inducible factor 1-α (HIF1-α). While present in humans, this mechanism does not appear to be the primary mechanism of collateral formation, at least in the setting of single coronary occlusion where arteriogenesis appears to predominate.4 In arteriogenesis, coronary collateral vessels arise from preformed arterioles and are visible at angiography within hours5,6 to a few weeks after occlusion (Table 2).7 Maturation takes time, however, and collaterals take on the appearance of normal coronary arteries within 12 months.8 Since coronary collateral vessel patency can be critical in maintaining blood flow to jeopardized regions of myocardium, understanding the physiology and pharmacologic responses of the coronary collateral circulation is essential in Table 1. Sources of Coronary Collateral Flow Intracoronary collateral vessels: Collateral vessels that develop around areas of coronary occlusion within the same coronary artery. Intercoronary collateral vessels: Collateral vessels that develop between a coronary artery with normal or reduced flow to a vessel that is totally occluded. Noncoronary collateral vessels: Collateral vessels that develop from arteries outside of the coronary circulation to occluded vessels within the coronary circulation. These collaterals have been described from bronchial arteries, pericardial arteries, pulmonary veins, and internal mammary arteries.
Table 2. Grading of Coronary Collateral Vessels at Angiography Grade 0 Grade 1 Grade 2 Grade 3
No visible coronary collaterals Visible collateral vessels without filling of the occluded epicardial coronary artery Visible collateral vessels that partially fill an occluded epicardial coronary artery Visible collateral vessels that totally fill an occluded epicardial coronary artery
managing patients with coronary occlusion during cardiac anesthesia and surgery. Within normal coronaries, α-adrenergic stimulation causes mild vasoconstriction. However, overall myocardial blood flow is dependent on perfusion pressure, as well as coronary vascular resistance, and myocardial blood flow response to an α-agonist like phenylephrine either is unchanged or increases.9 Collateral vessels, however, do not constrict to α1-stimulation and are very pressure-dependent in maintaining collateral flow. Histologic studies have demonstrated the absence of α1-receptors within mature collaterals.10 Interestingly, studies have shown that prazosin infusions increase blood flow more to collateral-dependent regions versus noncollateral areas during exercise.11 β-Adrenergic antagonists traditionally have been considered cardioprotective; however, some data suggest that β-antagonists may decrease collateral blood flow. Billinger et al showed that Doppler coronary collateral flow velocity was reduced after metoprolol infusion during coronary angiography.12 This is in agreement with a similar study using the nonselective drug propranolol in which coronary collateral flow was reduced as well.13 Work by Traverse et al found that propranolol increased the resistance across the collateral vascular bed with resultant decreased flow.14 Increased resistance is likely due to inhibition of β-mediated vasodilation rather than unmasked α-adrenergic effects, as α-adrenergic receptors do not play a prominent functional role in collateral vessels.15 This effect is probably most relevant when external pacing is used to prevent rate reduction in the heart after β-blockade, thereby preventing a decrease in myocardial oxygen assumption. Despite the acute pharmacologic effects, it should be noted that chronic β-blockade is a predictor for increased collateral formation in patients with coronary artery disease. Some have proposed this phenomenon as a possible etiology for the improved benefits of long-term β-blockade use.16 Vasopressin causes vasoconstriction via the V1 receptor on vascular smooth muscle. Mature collateral vessels are hyperresponsive to vasopressin, producing a vasoconstriction greater than that of the normal coronary vessels. Even vasopressin levels encountered during periods of physiologic stress (major surgery, hemorrhage, bypass, etc) may result in vasoconstriction capable of diminishing flow to collateral-dependent myocardium and may result in ischemia in those tissues.17 Additionally, enhancement of the response of collateraldependant vessels to vasopressin has been seen as well.18 Vasodilation of coronary collateral vessels in response to NTG administration is a well-known effect of this drug.19 Collateral vessels not already maximally dilated via nitric oxide mechanisms displayed an enhanced vasodilatory response to
4
MORRIS ET AL
NTG, and systolic contractile function improves in the region of myocardium supplied.20 Given the lack of collateral vasoconstriction with α-agonists, the combination of phenylephrine and NTG appears to be synergistic.20 Even in the coronary circulation without collateralization, the combination of phenylephrine and NTG increases coronary flow to ischemic myocardium by maintaining CPP with phenylephrine while vasodilating coronary vessels into the ischemic areas with NTG.9,21 Heparin frequently is administered to patients with critical coronary lesions not only to prevent thrombosis and occlusion of stenotic vessels, but to cause vasodilation of coronary collateral vessels.22 Thus, NTG, heparin, and endogenous adenosine are the 3 primary vasodilators of the coronary collateral circulation. Noncoronary collaterals or extracardiac collateral vessels develop during critical stenoses when coronary collateral flow is inadequate. Noncoronary collaterals have been reported to occur in 20% or more of patients with normal hearts.23 Varying anastomotic connections between the coronary circulation and internal mammary arteries, bronchial arteries, pericardial arteries, and pulmonary veins have been described.24 Flow from noncoronary collateral sources can be seen during coronary revascularization when a coronary artery is opened for grafting, and continous suctioning of blood flowing in the artery is required when the aortic is cross-clamped and no cardioplegia is being administered, as was seen in the authors’ patient. Noncoronary collateral flow is also pressure-dependent, and, thus, maintaining a high mean pressure pre-CPB is of critical importance. However, if mean pressure on CPB is allowed to be too high, myocardial protection could be affected negatively as cardioplegia could be washed out of the coronary circulation more rapidly since noncoronary collateral flow is not affected by aortic crossclamping.25 Collateral-supplied myocardium also can be placed at risk by disruption of the collateral flow during the surgical procedure. In a study lookng to determine the flow characteristics of extracardiac collaterals, the extracollateral flow was shown to strongly correlate with the laterality of the lesion—left anterior descending extracardiac collaterals arising from the left internal mammary artery (LIMA) and right coronary artery extracardiac collaterals arising from the right internal mammary artery. Given that the authors’ patient had an occluded left main coronary, the common practice of using the LIMA as a left anterior descending graft theoretically places patients at risk for ischemia during the prepump phase when the LIMA is taken down.26 Another unique intraoperative consideration is related to the concept of coronary steal as it applies to coronary collaterals. As coronary collaterals develop to supply ischemic areas, the coronary arteries from which the collaterals develop are prone to occlusion themselves. If this obstruction is distal to the takeoff of the collateral, then flow will increase preferentially down the collateral. If the postcollateral obstruction is relieved, then flow will decrease down the collateral a left anterior descending
Table 3. Unique Coronary Collateral Characteristics 1) Anatomic a) Derived from pre-existing arterioles (arteriogenesis) or de b) c)
novo (angiogenesis) Flow more pressure driven due to less autoregulation May be extracardiac in origin
2) Pharmacologic a) Alpha-agonists—Cause vasoconstriction in normal coronary
b)
c)
arteries; however, do not appear to have any effect on collateral arteries Beta-agonists—Mild vasoconstriction during acute infusion, but decrease in oxygen supply likely disproportionate to decrease in overall oxygen consumption; chronic use appears to be associated with greater quantity of collaterals Vasopressin agonist—Exaggerated response compared to normal coronary flow
3) Physiologic a) Disruption of IMA can disrupt collateral flow to the b) c)
ipsilateral side Collateral flow may impair cardioplegia activity by premature washout Restoring flow to obstructed coronary lesions potentially can induce coronary collateral steal
Abbreviation: IMA, internal mammary artery.
a steal phenomenon due to the change in pressure gradients, analogous to the pharmocologic steal phenomenon seen with coronary vasodilators such as adenosine and persantine.4,27 Conceivably, this would manifest most dramatically when totally occluded coronaries are recanalized; severe (not totally occluded) lesions that are revascularized through percutaneous techniques are statistically more likely seen clinically. It is also theoretically possible that during off-pump coronary artery bypass graft procedures, the use of intracoronary shunts temporarily could induce the steal phenomenon. Intraaortic balloon pump counterpulsation theoretically should enhance coronary collateral flow with balloon inflation during diastole. However, there are very few studies in humans documenting this effect. External counterpulsation during diastole has demonstrated improvement in coronary collateral flow in patients with chronic coronary artery disease.28 The risks of an intra-aortic balloon pump should be considered versus any benefits, as was done in the patient discussed. In summary, patients with coronary occlusions often develop coronary collateral circulation that provides critical flow to regions of ischemic but still viable myocardium, resulting in reduced mortality.29 Awareness of the physiology of the collateral and noncoronary collateral circulations and the response to various vasoactive drugs can guide patient management and avoid potentially dangerous problems intraoperatively in patients with critical coronary stenoses, such as a total left main coronary occlusion (Table 3).
REFERENCES 1. Schaper W: The physiology of the collateral circulation in the normal and hypoxic myocardium. Ergeb Physiol 63:102-145, 1971
2. Zimmern SH, Rogers WJ, Bream PR, et al: Total occlusion of the left main coronary artery: The Coronary Artery Surgery Study (CASS) experience. Am J Cardiol 49:2003-2010, 1982
CORONARY OCCLUSION AND DEPENDENT HEART
3. Koerselman J, van der Graaf Y, de Jaegere PP, et al: Coronary collaterals: An important and underexposed aspect of coronary artery disease. Circulation 107:2507-2511, 2003 4. Zimarino M, D’Andreamatteo M, Waksman R, et al: The dynamics of the coronary collateral circulation. Nat Rev Cardiol 11: 191-197, 2014 5. Rentrop KP, Cohen M, Blanke H, et al: Changes in collateral channel filling immediately after controlled coronary artery occlusion by an angioplasty balloon in human subjects. J Am Coll Cardiol 5: 587-592, 1985 6. Cohen M, Sherman W, Rentrop KP, et al: Determinants of collateral filling observed during sudden controlled coronary artery occlusion in human subjects. J Am Coll Cardiol 13:297-303, 1989 7. Seiler C: The human coronary collateral circulation. Eur J Clin Invest 40:465-476, 2010 8. Wechsler AS: Development of coronary collateral circulation. Annu Rev Med 28:341-348, 1977 9. Miller RR, Awan NA, DeMaria AN, et al: Importance of maintaining systemic blood pressure during nitroglycerin administration for reducing ischemic injury in patients with coronary disease. Effects on coronary blood flow, myocardial energetics and left ventricular function. Am J Cardiol 40:504-508, 1977 10. Harrison DG, Chilian WM, Marcus ML: Absence of functioning adrenergic receptors in mature canine coronary collaterals. Circ Res 59: 133-142, 1986 11. Bache RJ, Dai XZ, Herzog CA, et al: Effects of nonselective and selective alpha 1-Adrenergic blockade on coronary blood flow during exercise. Circ Res 61:II36-II41, 1987 12. Billinger M, Raeber L, Seiler C, et al: Coronary collateral perfusion in patients with coronary artery disease: Effect of metoprolol. Eur Heart J 25:565-570, 2004 13. Kyriakides ZS, Kolettis T, Antoniadis A, et al: Beta-adrenergic blockade decreases coronary collateral blood flow in patients with coronary artery disease. Cardiovasc Drugs Ther 12:551-559, 1998 14. Traverse JH, Altman JD, Kinn J, et al: Effect of beta-adrenergic receptor blockade on blood flow to collateral-dependent myocardium during exercise. Circulation 91:1560-1567, 1995 15. Harrison DG, Chilian WM, Marcus ML: Absence of functioning alpha-adrenergic receptors in mature canine coronary collaterals. Circ Res 59:133-142, 1986
5
16. van der Hoeven NW, Teunissen PF, Werner GS, et al: Clinical parameters associated with collateral development in patients with chronic total coronary occlusion. Heart 99:1100-1105, 2013 17. Peters KG, Marcus ML, Harrison DG: Vasopressin and the mature coronary collateral circulation. Circulation 79:1324-1331, 1989 18. Sellke FW, Quillen JE, Brooks LA, et al: Endothelial modulation of the coronary vasculature in vessels perfused via mature collaterals. Circulation 81:1938-1947, 1990 19. Klassen CL, Traverse JH, Bache RJ: Nitroglycerin dilates coronary collateral vessels during exercise after blockade of endogenous NO production. Am J Physiol 277:H918-H923, 1999 20. Bache RJ: Effect of nitroglycerin and arterial hypertension on myocardial blood flow following acute coronary artery occlusion in the dog. Circulation 57:557-562, 1978 21. Borer JS, Redwood DR, Levitt B, et al: Reduction in myocardial ischemia with nitroglycerin or nitroglycerin plus phenylephrine administered during acute myocardial infarction. N Engl J Med 293: 1008-1012, 1975 22. Fujita M, Sasayama S, Asanoi H, et al: Improvement of treadmill capacity and collateral circulation as a result of exercise with heparin pretreatment in patients with effort angina. Circulation 77:1022-1029, 1988 23. Björk L: Angiographic demonstration of extracardial anastomoses to the coronary arteries. Radiology 87:274-277, 1966 24. Loukas M, Hanna M, Chen J, et al: Extracardiac coronary arterial anastomoses. Clin Anat 24:137-142, 2011 25. Olinger GN, Bonchek LI, Geiss DM: Noncoronary collateral distribution in coronary artery disease. Ann Thorac Surg 32: 554-557, 1981 26. Stoller M, de Marchi SF, Seiler C: Function of natural internal mammary-to-coronary artery bypasses and its effect on myocardial ischemia. Circulation 129:2645-2652, 2014 27. Farhad H, Murthy VL: Pharmacologic manipulation of coronary vascular physiology for the evaluation of coronary artery disease. Pharmacol Ther 140:121-132, 2013 28. Gloekler S, Meier P, de Marchi SF, et al: Coronary collateral growth by external counterpulsation: A randomised controlled trial. Heart 96:202-207, 2010 29. Meier P, Hemingway H, Lansky AJ, et al: The impact of the coronary collateral circulation on mortality: A meta-analysis. Eur Heart J 33:614-621, 2012
