Ann Thorac Surg 2001;72:S2220-S2226
© 2001 The Society of Thoracic Surgeons
a Department of Surgery, VA Boston Healthcare System, Brigham and Womens Hospital, Harvard Medical School, Boston, Massachusetts, USA
* Address reprint requests to Dr Crittenden, Surgical Service (112), VA Boston Healthcare System, 1400 VFW Parkway, West Roxbury, MA 02132, USA
Presented at Monitoring and Improving Patient Safety During and Following Cardiac Surgery, San Diego, CA, May 5, 2001.
Numerous clinical studies have corroborated the ability of intraoperative sampling of coronary sinus blood to measure changes in myocardial metabolism induced by ischemia and reperfusion. Among other changes, cardiac arrest induces a period of obligate myocardial lactate production that persists for an indeterminate amount of time after reperfusion. Coronary sinus lactate assays have been established as a standard method to compare various myocardial protection strategies. Current methodology requires detailed sample processing, precluding real-time feedback in the operating room. Newer devices hold promise in allowing the online assessment of myocardial metabolism; however, these methods await precise validation.
The resurrection of chemical cardioplegia by Gay and Ebert  in 1973 provided the means for cardiac surgeons to safely induce an elective arrest. Nearly 30 years later, the evolution of myocardial protection techniques has yielded distinct benefits for the vast majority of patients undergoing cardiac surgery, including an expanding population of high-risk candidates who require complex procedures . Despite extensive clinical and laboratory studies investigating myocardial preservation, the current methodology does not invariably prevent myocardial stunning or frank myocardial necrosis. Recognition of this failure to uniformly avert intraoperative myocardial injury has led to the development of a wide array of cardioprotective strategies , which are often divergent in nature (Table 1). Each approach has a loyal constituency who avidly proclaim primacy based on studies that have explored relevant histologic, hemodynamic, or metabolic variables to assess the adequacy of the method. By necessity, the histologic changes induced by ischemia and reperfusion during cardiac surgery can be evaluated only after an operation. Although the postischemic recovery of left ventricular function can be assessed intraoperatively, the routine clinical use of load-independent estimates of myocardial contractility is rare. The usual load-dependent measures of ventricular function are less precise because preload, afterload, and heart rate cannot be held constant . Evaluation of myocardial metabolism during cardiac surgery allows the investigator to quantify the degree of physiologic impairment at various stages of cardiac operations (eg, on initiation of bypass, after aortic cross-clamping, during early and late reperfusion). Clinically applicable methods for metabolic analysis include the following: myocardial tissue assays; serial collection of myocardial specific biomarkers; online measurement of intramyocardial gas tensions or pH; or direct cannulation of the coronary sinus metabolites to measure coronary blood flow and to analyze substrate use .
Bing and colleagues  first achieved cannulation of the coronary sinus in humans in 1947. This achievement was followed by detailed analyses of myocardial substrate metabolism obtained by measuring the arteriovenous extraction of carbohydrates, fat, ketones, and amino acids in vivo [12, 13]. These provocative studies led to the study of myocardial metabolism in a variety of populations: patients during exercise; patients with myocardial ischemia; patients with known coronary artery disease who were asymptomatic; patients with diabetes; and patients with cardiomyopathy and congestive heart failure. In 1968, Giannelli and colleagues  were the first to describe a technique for direct cannulation of the coronary sinus intraoperatively for the purpose of examining postoperative myocardial metabolism. A second, contemporaneous study performed by these investigators analyzed the postoperative left ventricular function and myocardial metabolism after mitral replacement or repair . The relationships between myocardial oxygen consumption, coronary blood flow, myocardial oxygen extraction, and lactate extraction were examined before and after an infusion of isoproterenol. Myocardial oxygen consumption1 (MVO2) was elevated immediately after surgery, and was associated with a concordant rise in coronary blood flow. On postoperative day 1, the MVO2 and coronary blood flow normalized; lactate extraction2 persisted and there was a reduction in the myocardial oxygen extraction ratio3 (O2ER). Isoproterenol improved the cardiac index, mean aortic pressure, and coronary blood flow. Myocardial oxygen consumption increased, leading to a fall in the magnitude of lactate extraction in some patients, and lactate production in others. Perioperative changes in lactate metabolism in coronary bypass patients were first reported by Carlson and associates  in 1972. At rest, all patients had normal lactate extraction preoperatively. However, with atrial pacing, most patients demonstrated lactate production. For those patients who were successfully revascularized, atrial pacing did not induce lactate production.
Several early studies analyzed the efficacy of myocardial protection by measuring myocardial oxygen and lactate extraction during cardiopulmonary bypass . Myocardial preservation was achieved with coronary perfusion, elective fibrillatory arrest, or intermittent anoxic arrest. There were several common themes. First, oxygen extraction increased with continuous coronary perfusion and elective fibrillatory arrest, whereas it decreased with intermittent anoxic arrest. Second, although most patients had normal lactate extraction before cardiopulmonary bypass, once they were placed on bypass, all of the patients developed lactate production regardless of technique. Finally, the bypass-induced lactate production would diminish with time after restoration of normal coronary flow. This usually coincided with cessation of cardiopulmonary bypass but could persist for several hours after that.
Evaluation of myocardial preservation
The ability to perform coronary sinus blood sampling provided researchers with the tools to evaluate the ability of potassium-based cardioplegia solutions to prevent irreversible degradation of metabolic processes. In a series of articles, various investigators  were able to characterize myocardial lactate and oxygen extraction throughout the duration of bypass, with the heart arrested, and during reperfusion. Cardioplegia-induced arrest consistently generated lactate production that was exacerbated once the clamp was removed ("washout"), and persisted for an indeterminate time thereafter, depending on the duration of ischemia and on the cardioplegia delivery protocol. Despite relative maintenance of myocardial oxygen consumption during bypass, oxygen extraction invariably fell to less than baseline values after removal of the aortic cross-clamp. The arteriovenous difference in oxygen content would return to baseline within the first 24 hours.
The revival of blood as a vehicle for potassium cardioplegia made comparisons between sanguinous and asanguinous solutions inevitable. Several investigators [21, 30, 31] found that blood cardioplegia reduced the magnitude of lactate production during the ischemic interval. Two of these studies [21, 30] but not the third  found that upon reperfusion, hearts arrested with blood cardioplegia extracted more oxygen and were able to extract lactate earlier than hearts arrested with an asanguinous solution. Despite demonstrating metabolic recovery that was equivalent  or superior [21, 30] to crystalloid cardioplegia, there were no consequential differences in broad measures of clinical outcome (ie, survival and inotrope use) in either study. Teoh and colleagues  addressed the "reperfusion lag" in metabolic recovery by delivering the terminal warm blood cardioplegia just before cross-clamp removal. The characteristic postischemic delay in the conversion of myocardial metabolism from net lactate production to net lactate extraction was virtually eliminated.
Contemporary myocardial protective strategies
Newer delivery strategies such as: continuous warm retrograde , simultaneous antegrade/retrograde cardioplegia [37, 38], and tepid cardioplegia [39, 40] have been evaluated in single-institution studies. Menasché and colleagues , using a technique for delivering warm retrograde cardioplegia, were able to maintain myocardial oxygen extraction at near-normal levels and to virtually eliminate lactate production both during and after aortic cross-clamping. Hayashida and colleagues have readdressed the optimal temperature for cardioplegia delivery. Based on the belief that interrupting normothermic cardioplegia may lead to warm ischemia, the concept of tepid cardioplegia was introduced. In a prospective trial, patients were randomized to receive cold, warm, or tepid blood cardioplegia to determine the optimal delivery temperature . As compared with warm or cold delivery temperatures, tepid cardioplegia yielded less lactate production and acid release immediately after cross-clamp removal. Early postoperative ventricular function was best in those patients who received tepid cardioplegia.
Free fatty acids (FFA) serve as the primary fuel for myocardial substrate metabolism in the fasting state, with glucose providing only 15% to 20% of energy production. In the postprandial state, carbohydrate assumes a larger percentage, rising to nearly 30% . There is evidence that after ischemia and reperfusion, myocardial substrate metabolism does not return to baseline for several hours . Stress and operative trauma stimulate gluconeogenesis and liberate FFA by means of systemic lipolysis . In experimental preparations, the presence of elevated plasma levels of FFA exacerbates a postischemic defect in myocardial substrate use, which leads to excess fatty acid oxidation. This preference for fatty acid interferes with glucose oxidation and may impede recovery of contractile function [46, 47].
In a series of patients undergoing coronary bypass, Teoh and colleagues  examined myocardial fatty acid oxidation using palmitate labeled with carbon-14 (14C). They found that during reperfusion, FFA were extracted by the heart without a concomitant rise in radiolabeled CO2 in the coronary sinus effluent. Additionally, the investigators found that lactate was consumed in preference to the elevated plasma levels of glucose and free fatty acids normally found during reperfusion. They concluded that myocardial fatty acid oxidation was impaired due to an alteration in substrate preferences. In a second, related study, Teoh and colleagues  hypothesized that lactate, rather than glucose and fatty acids, may be the preferred substrate in the postischemic reperfused heart. Patients undergoing myocardial revascularization were randomized to receive either an infusion of Ringers lactate solution or an equal volume of normal saline perioperatively. Both groups also received infusions of 14C-radiolabeled lactate. During reperfusion, patients in the high lactate group paradoxically demonstrated both lactate production (by means of chemical analysis) and lactate extraction (by scintillation counter). Patients receiving Ringers lactate solution had better postoperative systolic function and spilled less creatine kinaseMB (CK-MB) than those patients who received saline.
Svensson and colleagues  examined arteriovenous FFA, amino acid, lactate, and glucose extraction in coronary bypass patients randomized to receive either an insulin drip postoperatively or standard postoperative care without a continuous insulin infusion. These investigators hypothesized that impaired myocardial metabolism after aortic cross clamp removal may be due to insulin resistance. Despite characteristically high plasma levels of FFA during reperfusion, they were unable to demonstrate active myocardial uptake of FFA in either the experimental or control group. In the insulin-treated group, plasma levels of FFA decreased. Transmyocardial uptake of glucose, alanine, lactate, and pyruvate was enhanced with insulin infusion. The authors postulated that insulin increased the activity of pyruvate dehydrogenase, leading to improved myocardial uptake of lactate and pyruvate and an earlier return to aerobic metabolism. Following a similar theme, Rao and colleagues  examined the effect of insulin-enhanced cardioplegia on myocardial lactate metabolism, oxygen extraction and myocardial pyruvate dehydrogenase. Patients undergoing elective myocardial revascularization were randomized to receive blood cardioplegia solution either with or without insulin using a tepid, antegrade/retrograde delivery protocol. Both groups of patients produced lactate while the heart was arrested; however, in patients receiving insulin-enhanced cardioplegia, lactate was immediately extracted once the aortic cross-clamp was removed. Insulin-enhanced cardioplegia did not affect myocardial pyruvate dehydrogenase activity. Substrate-enhanced cardioplegia has been advocated as another possible method to restore impaired fatty acid or carbohydrate energy production . Substrate-enhanced cardioplegia solution was used in an attempt to improve postischemic lactate metabolism. There was less reperfusion lactate production with glutamate ; however, -ketoglutarate  did not make a difference. Vanhannen and associates  examined the effect of an intravenous infusion of aspartate after coronary bypass procedures. The aspartate infusion did increase myocardial uptake of aspartate and decreased myocardial uptake of glutamate. There was no effect on fatty acid metabolism or myocardial lactate kinetics.
Continuous online measurement of coronary sinus substrates
All of the previously mentioned metabolic studies had blood samples withdrawn from the coronary sinus at various intervals throughout the bypass period; however, none of the data generated were analyzed concurrently. Lactate assays and measurement of oxygen content require meticulous processing and are time intensive. Customization of oximetry catheters has allowed continuous measurement of coronary sinus oxygen saturation [53, 54]. Before operation, a flexible oximetry catheter can be inserted into the coronary sinus by way of the left subclavian vein. Coronary sinus O2 saturation (ScsO2) is displayed continuously on a monitor in the operating room. If, during the ischemic interval, the level decreases to less than 60%, then cardioplegia concentration or flow can be adjusted to raise O2 delivery. Recently the technique was modified to include a Doppler flow wire that was inserted into the coronary sinus through the previously mentioned flexible oximetry catheter. In this manner, coronary sinus blood flow and coronary sinus oxygen saturation could be followed simultaneously during ischemia and reperfusion .
Another real-time method involves rapid determination of plasma lactate levels. A coronary sinus catheter is placed in the usual manner. Once its position is confirmed, sequential samples of arterial and coronary sinus blood are drawn every 2 minutes and lactate determinations made using a lactate oxidase electrode . The lactate assay requires 40 seconds, with results displayed on a console. The direction of lactate extraction (ie, net negative = production; net positive = extraction) measured continuously in this fashion yields data similar to those described by previous investigators who obtained serial samples of coronary sinus blood. During cardiac arrest there is net negative lactate production. Net positive lactate extraction usually recovered early during reperfusion. This device was used to evaluate online lactate metabolism in a series of patients undergoing simple and complex cardiac procedures. The investigators found that patients who had an early "cross-over point"4 versus a later one required less inotropic support and had a larger increase in their postoperative ejection fraction than those patients having the later cross-over point .
Table 2 provides a partial list of constituents that have been measured in blood aspirated from the coronary sinus during cardiac operations. The length of this list certainly reflects the ease of inserting a coronary sinus catheter.
Limitations of coronary sinus measurements
A detailed description of cardiac venous anatomy is beyond the scope of this article; however, there are several features that may affect the composition of blood found in the coronary sinus.
In an autopsy study of more than 300 hearts that was designed to examine the variation in coronary sinus drainage, von Lüdinghausen  described five distinct patterns of venous drainage. More than half (51%) of the hearts studied demonstrated a venous drainage pattern in which only the great cardiac vein (ie, the anterior interventricular vein) and the middle cardiac vein (ie, the posterior interventricular vein) drained into the coronary sinus. In only 21% of the specimens examined did all of the major epicardial veins empty into the coronary sinus. The presence of these anatomic variations in concert with the loss of venous blood by way of the Thebesian system suggests that coronary sinus blood is an admixture of venous blood from many, but not all, regions of the heart. As such, coronary sinus blood sampling may not be fully representative of global myocardial metabolism.
The technique for transatrial placement of a coronary sinus cannula for retrograde cardioplegia has been described elsewhere . In practice, cannulation of the coronary sinus for metabolic monitoring is similar; however, as in placement for cardioplegia delivery, there is no consensus for where the tip of the cannula should lie. Early studies of human myocardial metabolism suggested that sample homogeneity was obtained when the catheter tip was in the mid to anterior portion of the coronary sinus . Alternative methods for coronary sinus cannulation include preoperative transvenous placement under fluoroscopic guidance and direct insertion through a right atrial incision or by puncture of the coronary sinus.
Lactate release from the heart can occur despite a normal supply of oxygen . It is incorrect to assume that in every circumstance, the heart is either extracting lactate or producing it . Another limitation of coronary sinus measurements is the assumption that under conditions of constant ischemia, lactate production is also constant. In an experimental preparation, Apstein and colleagues  demonstrated that lactate production decreased with time, thus casting doubt on the ability of coronary sinus lactate levels to reflect the magnitude of the ischemic insult, particularly during prolonged periods of aortic clamping. The absence of native coronary blood flow during ischemia may yield ambiguous information about the extent of lactate production. Sidi and Davis  analyzed lactate extraction and lactate flux5 before and during left anterior descending coronary artery (LAD) occlusion. Before LAD occlusion, lactate extraction and lactate flux were closely correlated (r = 0.76); however, during LAD occlusion, lactate extraction and lactate flux had only a weak correlation (r = 0.32). Thus, the presence of "no-flow" ischemia led to a significant overestimate of lactate production in this experimental model. Cardioplegia is infused during cardiac surgery to ameliorate the effects of global ischemia; therefore, it may be speculated that in the clinical setting, this flow-dependent limitation might be rectified. In a series of coronary bypass patients randomized to receive either blood or crystalloid cardioplegia, Fremes and colleagues  inserted a double thermistor-tipped catheter into the coronary sinus to measure blood flow and also to sample blood for metabolites. This instrumentation allowed these investigators to measure myocardial lactate extraction and then to calculate myocardial lactate flux at multiple time points perioperatively. At each interval, the values for lactate extraction and lactate flux responded comparably to ischemia and reperfusion. The only significant difference was that the ischemia-induced fall in lactate extraction (net lactate production) was more pronounced than the decline in lactate flux. Although calculation of lactate flux allows a more precise estimation of the ischemic burden than does calculation of lactate extraction, in the clinical setting of cardiac surgery, their correlation appears to be maintained. A critical flow-independent event in the reperfusion period is the conversion from net lactate production to lactate extraction, signaling the return of aerobic metabolism.
The intricate analysis required for coronary sinus blood sampling currently limit its use for real-time metabolic management intraoperatively. Except for the studies cited in the section of this paper entitled Continuous Online Measurement of Coronary Sinus Substrates, every other reference cited previously that measured coronary sinus metabolites during cardiac surgery required multiple serial blood sampling and meticulous handling of laboratory determinations followed by detailed, time-intensive postoperative analysis. All of the important information derived from these studies was ascertained after the fact.
Unfortunately, the investigators proffering real-time data acquisition techniques have not yet validated their methodology with other well-accepted means of evaluating myocardial metabolism. Until these devices have been corroborated in trials designed to evaluate clinically relevant intraoperative and postoperative end-points, coronary sinus measurement of myocardial metabolites will remain a tool for surgical investigators who are pursuing hypothesis-driven research protocols.
1 MVO2 = coronary blood flow x (arterial O2 content - coronary sinus O2 content).
2 Lactate extraction = arterial[lactate] - coronary sinus[lactate]/arterial[lactate] x 100, typically expressed as a percentage.
3 O2ER = arterial O2 content - coronary sinus O2 content/arterial O2 content.
4 A time during reperfusion where the net negative lactate extraction (production) crosses zero to become net positive lactate extraction. This flow-independent event signals the restoration of aerobic metabolism.
5 Lactate flux = (arterial[lactate] - coronary sinus[lactate]) x coronary blood flow.
This article has been cited by other articles:
M. A. Schroeder, K. Clarke, S. Neubauer, and D. J. Tyler
Hyperpolarized Magnetic Resonance: A Novel Technique for the In Vivo Assessment of Cardiovascular Disease
Circulation, October 4, 2011; 124(14): 1580 - 1594.
[Full Text] [PDF]
R. Jaumdally, C. Varma, R. J. Macfadyen, and G. Y.H. Lip
Coronary sinus blood sampling: an insight into local cardiac pathophysiology and treatment?
Eur. Heart J., April 2, 2007; 28(8): 929 - 940.
[Abstract] [Full Text] [PDF]
T. A. Vassiliades Jr, J. L. Nielsen, and J. L. Lonquist
Coronary perfusion methods during off-pump coronary artery bypass: results of a randomized clinical trial
Ann. Thorac. Surg., October 1, 2002; 74(4): S1383 - S1389.
[Abstract] [Full Text] [PDF]
|HOME||HELP||FEEDBACK||SUBSCRIPTIONS||ARCHIVE||SEARCH||TABLE OF CONTENTS|
|ANN THORAC SURG||ASIAN CARDIOVASC THORAC ANN||EUR J CARDIOTHORAC SURG|
|J THORAC CARDIOVASC SURG||ICVTS||ALL CTSNet JOURNALS|