Year : 2007 | Volume
: 51 | Issue : 4 | Page : 303-
Perioperative cardiac failure
Yatin Mehta1, Sangeeta Khanna2,
1 MD, DNB, FRCA, FAMS, Senior Consultant Anaesthesiologist, Platinum Lounge, Indraprastha Medical Corporation Limited, Sarita Vihar, New Delhi 110076, India
2 MD, Platinum Lounge, Indraprastha Medical Corporation Limited, Sarita Vihar, New Delhi 110076, India
Senior Consultant Anaesthesiologist, Platinum Lounge, Indraprastha Medical Corporation Limited, Sarita Vihar, New Delhi 110076
|How to cite this article:|
Mehta Y, Khanna S. Perioperative cardiac failure.Indian J Anaesth 2007;51:303-303
|How to cite this URL:|
Mehta Y, Khanna S. Perioperative cardiac failure. Indian J Anaesth [serial online] 2007 [cited 2019 Aug 20 ];51:303-303
Available from: http://www.ijaweb.org/text.asp?2007/51/4/303/61159
Acute cardiac failure is defined as the pathophysiologic state in which abnormal cardiac function is responsible for failure of the heart to pump sufficient blood for the need of the peripheral tissues  . This condition may arise in the perioperative period in patients with poor left ventricular function, following incomplete revascularization, poor preservation of myocardial function during cardiopulmonary bypass (CPB) and in patients with high risk for cardiac surgery  . All these patients may have difficulty in being weaned off the pump.
The factor that contributes the most to this is the pathology that existed immediately prior to operation. Even when reconstructive surgery is successful, it is unusual to see an immediate improvement in contractile function (without the administration of inotropes). Thus adequate cardiac reserve must be present to withstand the demands of heart surgery. Preoperative cardiac abnormalities are not limited to systolic function but may involve diastolic, valvular, electrophysiologic, and vascular function. Optimal perioperative care is aimed at compensating for this impaired preoperative function.
The operation itself may cause left ventricular dysfunction and low cardiac output and cardiac failure. Contributors to cardiac dysfunction include inadequate protection of the myocardium during periods of aortic cross-clamping, myocardial and pulmonary edema, acute left ventricular distension or other trauma, uncorrected valvular lesions, arrhythmias, acute cardiac tamponade, graft thrombosis, inadequate revascularization and reduced coronary blood flow. The investigation as to whether the operation has been adequate to correct preexisting pathology begins as soon as the aortic cross-clamp is removed. Search for graft occlusion, valvular incompetence, cardiac compression, or intracardiac shunting continues throughout the postoperative period. Useful tools include assessment of wall motion abnormalities by direct visualization or echocardiography (usually transesophageal),the electrocardiogram, ANP levels, cardiac enzymes, blood gases obtained from the pulmonary and systemic vasculature, and right and left atrial pressure measurements. Discovery of an inadequate operation usually demands immediate reoperation.
Predictors of difficult weaning from CPB are:
Prolonged aortic cross clamp time>60 minutesProlonged bypass timePoor myocardial preservation during CPBLeft ventricular ejection fraction (LVEF) Pre bypass delta PCO 2 >6 mmHg (veno-arterial and regional difference in the partial pressure of CO 2 )Pre operative diastolic dysfunction
Mild degree of acute cardiac failure post CPB can easily be reversed by careful control of peripheral circulation. If ventricular function is depressed following a cardiac operation, treatment with vasodilators and volume loading may not be sufficient to ensure adequate circulation. Ventricular contractility should be augmented, usually with inotropic agents. Inotropic agents may be divided into catecholamines and noncatecholamines. The former include natural or synthetic adrenergic agents that stimulate alpha and beta receptors in the heart, lungs, and peripheral vasculature. Noncatecholamines include calcium, digoxin, phosphodiesterase inhibitors, calcium channel sensitizers, glucose insulin potassium, and thyroid hormone.
Patients likely to benefit from catecholamine support are those with low cardiac output (CI -1 .m -2 ), with optimized heart rate, rhythm, ventricular preload, and after load, and without evidence of acute cardiac tamponade. Dopamine and dobutamine enhance heart rate and cardiac output equally, but dobutamine produces greater reductions in left ventricular preload and afterload  . Dobutamine augments myocardial coronary blood flow more than dopamine  . As with all adrenergic agents, the haemodynamic effects of these drugs depend on dosage. When dopamine is administered at less than 8 µg.kg -1 .min-1, beta and dopaminergic receptor stimulation predominates and enhances cardiac output and renal blood flow  . At doses greater than 10 µg.kg -1 .min -1 , alpha vasoconstrictor effects predominate, and tachycardia mayensue. The underlying pathophysiology and age of the patient may alter the expected response.
The effect of epinephrine on both alpha- and beta adrenergic receptors makes it a useful agent following cardiac surgery. The effects of epinephrine vary by dosage. At low doses ( -1 .min -1 ), epinephrine stimulates peripheral beta 2 receptors and causes vasodilation. Higher doses cause increasing cardiac effects, and the highest doses can cause vasoconstriction via peripheral alpha receptor stimulation. The response of an individual patient's peripheral and pulmonary haemodynamics is somewhat unpredictable, especially after cardiopulmonary bypass. Steen et al  investigated the effects of epinephrine after cardiac surgery and discovered a consistent increase in cardiac output but variable changes in mean arterial pressure. Stephenson et al  demonstrated similar increases in cardiac output with epinephrine infusions after heart surgery and showed that hypertension and tachycardia occurred when higher doses were used.
Norepinephrine, another naturally occurring catecholamine, is used after heart surgery when blood pressure is low. In addition to pronounced effects on peripheral alpha receptors, norepinephrine is a potent beta 1 agonist and therefore increases myocardial inotropy. The increased blood pressure that these two effects provide must be balanced against increased myocardial oxygen consumption and reduced renal, mesenteric, and peripheral perfusion that may ensue, especially at higher doses.
Isoproterenol stimulates beta 1 - and beta 2 -adrenergic receptors but has little alpha action. Isoproterenol increases heart rate and contractility and decreases systemic vascular resistance. Isoproterenol is potentially useful when reactive pulmonary hypertension and right-sided heart failure contribute to postoperative low cardiac output, as may occur following mitral valve surgery or cardiac transplantation. Its nonselective beta-adrenergic stimulation, which may cause tachyarrhythmias and systemic vasodilation, limits its utility in other situations.
Calcium, in its ionized form, is critical for excitation contraction coupling in cardiac muscle  . Low calcium ion concentrations depress ventricular function and peripheral resistance and contribute to hypotension and low cardiac output. In addition, adequate calcium is necessary for the action of many cardiovascular drugs, including catecholamines. Drop and Scheidegger  demonstrated that a calcium bolus injection is associated with increased myocardial contractility, an effect that is directly related to the initial calcium level. Shapira et al  also showed that a bolus of calcium can cause transient haemodynamic improvement in patients after cardiopulmonary bypass but that a continuous infusion of calcium does not sustain the beneficial effect. Because of these and similar observations, many surgeons administer a bolus of calcium (500 to 1000 mg) immediately before weaning from cardiopulmonary bypass when serum ionized calcium levels are generally low.
Inamrinone (initially named amrinone) is a noncatecholamine bypyridine derivative that inhibits phosphodiesterase to slow the hydrolysis of adenosine 3', 5'cyclic monophosphate (cAMP)  . Because its congener, milrinone, causes less thrombocytopenia, milrinone has largely replaced inamrinone in clinical use  . Milrinone potentiates action of catecholamines by inhibiting the breakdown of cAMP. This maybe particularly important in patients who were in CHF prior to operation, as β1-adrenergic receptors can be down-regulated in this population. Therefore, when ventricular dysfunction occurs despite standard catecholamine therapy, milrinone can provide additional and effective inotropic support  .
Milrinone has been shown to increase CI and improve haemodynamicsin a variety of cardiac surgical and CHF patients. Feneck  studied 99 adult patients after elective cardiac operation who had a low CO (CI -1 .m -2 ) with a pulmonary artery occlusion pressure of 8 mm Hg or higher. In this study, patients received a loading dose of milrinone (50 µg.kg -1 over a 10-min period) followed by a continuous infusion of one of three dosages-0.375, 0.5, or 0.75 µg.kg -1 .min -1 (low-, middle, and high-dose groups, respectively)-administered for a minimum of 12 hours. Patients were sequentially allocated to each dosage group. Haemodynamic measurements were made prior to therapy and up to 12 hours after the start of milrinone therapy. Milrinone treatment resulted in a rapid, well-sustained, and highly significant increase in CI in all three dosage groups and a similar reduction in pulmonary artery occlusion pressure in all groups. Significant reductions were also observed in systemic vascular resistance and pulmonary vascular resistance (PVR), although changes in the latter were less predictable and more dose dependent. Further analysis revealed that low CI(1.59 L.min -1 .m -2 ), high resting PVR (>200 dynes.s.cm -5 ), and low mean arterial pressure (64 mm Hg) prior to treatment were predictors of a good therapeutic response to milrinone.
Another placebo-controlled, double-blind study  demonstrated the benefits of milrinone in facilitating weaning of high-risk patients from CPB. Patients were randomized to receive either intravenous milrinone (50µg.kg -1 loading dose over 20 minutes followed by 0.5 µg.kg -1 .min -1 infusion) or placebo 15 minutes before withdrawal from CPB. Of the 30 patients who completed the study, bypass support was withdrawn successfully in all 15 patients randomized to receive milrinone but in only 5 of the 15 patients randomized to receive placebo. The remaining 10 patients in the placebo group who were unable to be separated from CPB had milrinone administered in an open-label phase. After receiving milrinone treatment, these remaining patients were successfully with drawn from CPB.
Once acute problems have subsided, some patients in sinus rhythm require chronic augmentation of contractile function. The use of digital is  in this setting has long been argued, but there is evidence that it increases contractility. It produces a dose related increase in myocardial contractility in both normal and failing heart. In the failing heart it improves contractility and cardiac output with decrease in SVR and venous tone. It is also useful to limit AV conduction when increase in heart rate is detrimental. However it has been found to be more useful in chronic heart failure. Newer agents such as enoximone, another type III phosphodiesterase inhibitor, and vesnarinone, a quinalone with immunomodulatory properties, may be available for treatment of chronic heart failure, perhaps used in combination with beta-blocking agents  .Their role in the immediate postoperative period is not defined.
Levosimendan  is a new drug which is a calcium channel sensitizer. It acts by
Increasing myofilament activity by binding to cardiac troponin-C in a calcium dependent mannerImproves diastolic relaxationProduces vasodilatationDoes not increase myocardial oxygen demandHas selective PDE-III inhibitory action.Follath et al  compared treatment of levosimendan 24mg.kg -1 followed by infusion and dobutamine 5μg.kg1 .min -1 infusion for 24 hours in 203 patients with heart failure. They observed a 35% increase in CI and 25% decrease in PCWP at 24 hours; but this was achieved in 28 percent patients in the levosimendan group compared to 15% patients in the dobutamine group. Levosimendan 8 or 24 mg.kg- 1 improved cardiac performance without increasing myocardial oxygen consumption or changing myocardial substrate utilization in 23 patients undergoing elective CABG. Del-Razo et al  concluded that levosimendan has a better profile of safety than its predecessors (inamrinone and milrinone) and improves haemodynamic parameters without increase in myocardial consumption and is a promising drug in this group of patients. Unlike agents that act through adrenergic pathways, levosimendan does not cause diastolic calcium overload, which can impair myocardial relaxation. Furthermore, it does not increase myocardial oxygen requirements, and may improve myocardial perfusion as a result of vasodilatation. Longterm-benefits could also result from levosimendan use, as the presence of a pharmacologically active metabolite with a long elimination half-life (75-80 h) could lead to persistent haemodynamic effects 
Glucose - Insulin-Potassium (GIK)
GIK has again been revived and received attention in patients with MI and cardiac surgery. Berger et al  observed that GIK;
Suppresses inflammatory cytokinesIncreases endothelial nitric oxide NOIncreases anti inflammatory cytokinesHas cardioprotective actionAccelerates recovery of ischaemic myocardium in patients undergoing cardiac surgeryGradinac et al  observed an increase in CI and decreased need for inotropic support in patients with acute cardiac failure following CPB when glucose 50%, insulin 80u/L and potassium 100meq/L were infused compared to sodium chloride. Szabo et al  observed decreased mortality and shorter ICU stay in patients with poor LV function who were infused high dose GIK in insulin 1U.kg-1 post operatively.
Several studies , have demonstrated that cardiopulmonary bypass and hypothermic cardiac arrest result in low serum levels of thyroid hormone. The pattern of thyroid hormonedepletion after cardiopulmonary bypass is similar to that seen in other acute nonthyroid illnesses and the euthyroid sick syndrome. Traditionally, triiodothyronine replacement has not been administered to this group of patients. Several studies have challenged this view by demonstrating improved recovery of ischaemic myocardiumin animals  and improved haemodynamics with a lower incidence of atrial fibrillation in a small number of patients after cardiopulmonary bypass  . Thyroid supplementation to unstable organ donors may improve outcomes in cardiac transplantation. Triiodothyronine (T3) augments cardiac output via nucleus-mediated mechanisms and direct stimulation of calcium-ATPase in the sarcolemma and sarcoplasmic reticulum. Enhancement of calcium transport reduces the cytoplasmic calcium concentration, which aids in myocardial relaxation. This action improves myocardial compliance and diastolic function in postischaemic stunned myocardium. Triiodothyronine also decreases cardiac work postoperatively by acutely decreasing systemic vascular resistance.
A number of studies  , however, including a trial of over 200 patients, have failed to demonstrate any significant haemodynamic benefit from administering triiodothyronine to patients after cardiopulmonary bypass. This may be related tointramyocardial T3 levels. Although serum levels of thyroid hormone are decreased for up to one week after cardiopulmonary bypass the myocyte may not be depleted of thyroid hormone. Concerns regarding enhanced myocardial oxygen demand and conflicting data about atrial arrhythmias have further limited widespread use of this hormone. Further investigation will help elucidate the multiple cardiovascular effects of triiodothyronine and possibly define the subset of patients who will benefit from this therapy.
So a whole range of inotropes inodilators and vasodilators are available to treat patients with acute cardiac failure. The choice of drugs depends on the patients underlying disease, severity of the low CO state, preoperative drug therapy and response to the individual inotropic and vasodilator drugs. Combination inodilator therapy with either one drug or a separate inotrope and vasodilator offers advantages over inotropic/vasodilator therapy alone as:
Lower PCWP can be obtainedMetabolic demand is reducedSide effects are decreasedGreater augmentation of CO is possibleBP and vital organ function is better maintainedSome combinations that can be used are shown in the [Table 1]:
Mechanical circulatory support
Pharmacologic inotropic support is the first line of therapy for the patient who fails to separate from cardiopulmonary bypass or who experiences pump failure in the early postoperative period. The decision to add mechanical support (IABP or a ventricular assist device) to chemical support is frequently based on the presence of low cardiac output (CI -1 .m -2 ) despite maximal inotropic support. A study showed that patients requiring three high-dose inotropes at the time of weaning from cardiopulmonary bypass had a mortality of 80%  . This series showed greatly improved hospital discharge rates for patients who underwent ventricular assist device (VAD) insertion based on a defined formula (cardiogenic shock despite administration of two high-dose inotropes) that emphasized early insertion. It is clear that high-dose pharmacologic inotropic support alone may stabilize a patient's haemodynamics and permit transfer to the intensive care unit, but that a significant proportion of these patients will ultimately succumb to multisystem organ failure.
Intra-aortic balloon counterpulsation
Intra-aortic balloon pumping (IABP) was first performed clinically by Kantrowitz et al  in 1968. IABP uses the principle of diastolic counter pulsation, in which the balloon inflates in synchrony with the cardiac cycle. Benefits of this technique include augmentation of diastolic coronary perfusion pressure, reduced systolic after load, and increased cardiac output with an improvement in the myocardial oxygen supply/demand ratio. Major contraindications to IABP use include severe atherosclerotic disease of the aorta oriliofemoral arteries, descending aorticdissection, and aortic insufficiency. If indicated, insertion of the IABP can be accomplished bedside in the ICU. IABPs are used perioperatively in 8% to 12% of cardiac surgical operations. There are significant practice pattern variations in regards to timing of insertion, with the percentage of IABPs inserted preoperatively ranging from 20% to 70%  . Complications of femoral IABP placement include lower extremity ischaemia and thrombocytopenia 
Ventricular assist devices
The intra-aortic balloon pump is an attractive form of circulatory support for the patient undergoing cardiac surgery because of its ease of insertion and removal. However, the IABP only yields a modest increase in cardiac output and does not displace a significant volume. Failure of the IABP to improve haemodynamic performance of the failing heart should prompt consideration of rapid institution of mechanical circulatory support. An ideal circulatory support device would be rapidly and easily implanted and explanted, permit uni-or biventricular assist, have minimal anticoagulation requirements, provide maximal LV unloading, reliably provide intermediate length support (7-14 days),permit ambulation, have a low infection rate, and be easy to convert to a long-term device. Such a device does not exist. Currently available options include the centrifugal pump, extracorporeal life support (ECLS), the Abiomed BVS-5000, and the Thoratec system.
Pennington et al  observed that in fact biventricular assist devices were more useful than LVAD alone in patients of biventricular failure post CPB hence a confirmed diagnosis of isolated left ventricular failure post CPB, was necessary before weaning from CPB.
One of the causes of acute cardiac perioperative failure is persistent mitral regurgitation (MR) after revascularization i.e correction of ischaemia. One of the newer techniques for correction of this is the insertion of Copasys device (Myocor) which is put through the mitral annulus (TEE guided) to reduce the size of the dilated annulus and hence reduce the MR. We have found excellent results with this device  .
Biventricular pacing  with resynchronization of the intra and interventricular conduction has also been suggested for weaning from CPB with acute cardiac failure. Biventricular stimulation leads to normal synchronized contraction of the ventricles resulting in improved cardiac output and stable haemodynamics postoperatively.
ANP (Atrial natriuretic peptide)
Plasma levels of ANP(pg/ml; radioimmunoassay) as a parameter for post ischaemic dysfunction and levels of troponin T (TnT) (ng/ml; ELISA test) as a parameter for postischaemic cellular damage were determined in 15 patients with coronary artery disease (CAD) (mean age: 58 + 6.1 years; 13 m, 2 w; with no history of myocardial infarction and no signs for congestive heart failure) prior to, during and after extracorporeal circulation (ECC).Ten days postop, the ANP level was with 262 + 33 pg/ml still increased threefold in comparison to the preoperative level The significantly increased ANP level up to the 10th day postoperatively indicate a very sensitive prolonged, post ischaemic dysfunction, which is not compensated 10 days postoperatively Perioperative levels of atrial natriuretic peptide and troponin-T in patients with uncomplicated coronary artery disease, undergoing CABG were measured  . Threefold increase of ANP 10 days postoperative and return of TnT levels to normal under consideration of datas of echo show, that ANP is suitable to indicate the mean term, functional, myocardial reduction. Increased ANP levels, atrial dilatation and dysfunction are important signs of cardiac functional reduction after cardiac surgery  .
SvO 2 is of prognostic value and due to its specificity it seems particularly useful for telling which patients are unlikely to develop cardio respiratory problems. Thus, this simple method for haemodynamic monitoring could contribute to cost containment as it seems that we can safely reserve Swan Ganz catheters for high-risk patients 
Transesophageal echocardiography (TEE)
TEE has been the recent and most important advances in cardiovascular monitoring. It gives a true picture of the loading condition of the left ventricle and also demonstrates new segmental wall motion abnormalities (SWMA) which may warrant addition/ revision of a graft. Ischaemic MR, position of IABC, LVAD can be detected on TEE  .
Pulmonary artery thermodilution catheter (PAC)
The Swan Ganz catheter is a very important monitoring tool. All haemodynamic data can be calculated using the PAC  . It is also useful in guiding fluid therapy and vasoactive therapy  .
These patients may require prolonged postoperative ventilation and may be kept on positive pressure ventilation (CPAP or BIPAP) post extubation. Diuretics and anti arrhythmics (Amiodarone) may be continued post operatively to maintain sinus rhythm and stable haemodynamics.
With the advances in technology a whole new spectrum of drugs and devices have become available. What and when to use requires good clinical acumen, understanding of cardiac physiology and expertise. The cost and duration of hospital stay are also important determining factors. In depth study of the technology and drug should be carried out before clinical use.
|1||ZeltsmanD,Acker MA. Surgical management of heart failure: an overview. Annu Med Rev 2002;53:383-91.|
|2||Kaplan JA, Guffin AV. Treatment of perioperative left ventricular failure. In Kaplan JA,(ed) Cardiac Anesthesia, 3rd edition. Philadelphia, WBSaunders Company 1993; 1058-94.|
|3||Di Sesa VJ, Brown E, Mudge GH Jr, et al. Hemodynamic comparison of dopamine and dobutamine in the postoperative volume-loaded, pressure-loaded, and normal ventricle. J Thorac Cardiovasc Surg 1982; 83:256.|
|4||Fowler MB, Alderman EL, Oesterle SN, et al. Dobutamine and dopamine after cardiac surgery: greater augmentation of myocardial blood flow with dobutamine. Circulation 1984; 70: I103.|
|5||Goldberg LI, Rajfer SI. Dopamine receptors: applications in clinical cardiology. Circulation 1985; 72:245.|
|6||Steen PA, Tinker JH, Pluth JR, et al. Efficacy of dopamine, dobutamine, and epinephrine during emergence from cardiopulmonary bypass in man. Circulation 1978; 57:378.|
|7||Stephenson LW, Blackstone EH, Kouchoukos NT. Dopamine vs epinephrine in patients following cardiac surgery: randomized study. Surg Forum 1976; 27:272.|
|8||Drop LJ. Ionized calcium, the heart, and hemodynamic function. Anesth Analg 1985; 64:432.|
|9||Drop LJ, Scheidegger D. Plasma ionized calcium concentration: important determinant of the hemodynamic response to calcium infusion. J Thorac Cardiovasc Surg 1980; 79:425.|
|10||Shapira N, Schaff HV, White RD, Pluth JR. Hemodynamic effects of calcium chloride injection following cardiopulmonary bypass: response to bolus injection and continuous infusion. Ann Thorac Surg 1984; 37:133.|
|11||Alousi AA, Johnson DC. Pharmacology of the bipyridines: amrinone and milrinone. Circulation 1986; 73:III 10-24.|
|12||Kikura M, Lee MK, Safon RA, et al. The effects of milrinone on platelets in patients undergoing cardiac surgery. Anesth Analg 1995; 81:44.|
|13||Lobato EB, Florete O Jr, Bingham HL.Asingle dose of milrinone facilitates separation from cardiopulmonary bypass in patients with pre-existing left ventricular dysfunction. Br JAnaesth 1998; 81:782.|
|14||Feneck RO. Intravenous milrinone following cardiac surgery: 1.Effects of bolus infusion followed by variable dose maintenance infusion. The European Milrinone Multicentre Trial Group.J Cardiothorac Vasc Anesth 1992; 6:554-62.|
|15||Levy JH, Bailey JM, Deeb M. Intravenous milrinone in cardiac surgery. Review Ann Thorac Surg 2002;73:325-330.|
|16||Cook LS, Toal KW, Elkins RC. Cardiovascular time course in patients with postoperative myocardial dysfunction requiring catecholamine administration. Curr Surg 1987; 44:124.|
|17||Gottlieb SS. New approaches to managing congestive heart failure. Curr Opin Cardiol 1995; 10:282.|
|18||Lilleberg J, Nieminen MS , Akkila J,et al. Effects of a new calcium sensitizer, levosimendan, on haemodynamic coronary blood flow and myocardial substrate utilization early after coronary artery bypass grafting. Eur Heart J 1998;19:660-68.|
|19||Follath F,Cleland JG, Just H, et al. Steering Committee and Investigators of the Levosimendan Infusion versus Dobutamine LIDO study. Efficacy and safety of intravenous levosimendan compared with dobutamine in severe low-output heart failure(the LIDO study)a randomized double blind trial. Lancet 2002;360:196-202.|
|20||Del-Razo OE, Carballal-Sanjuro JC, Campos Lara MG, et al. Levosimendan: a new option in the pharmacologic management of cardiac insufficiency. Gac Med Max 2003.139;87-89.|
|21||Yoshiyuki Tokuda * , Peter W. Grant, Hugh D. Wolfenden, et al. Levosimendan for patients with impaired left ventricular function undergoing cardiac surgery. Interact CardioVasc Thorac Surg 2006;5:322-326.|
|22||Berger MM, Mustafa I. Metabolic and nutritional support in acute cardiac failure. Curr Opin Clin Nutr Metab Care 2003; 6:195-201.|
|23||Gradinac S, Coleman GM, Toegtmeyer H, et al. Improved cardiac function with glucose-insulin-potassium after aortocoronary bypass grafting. Ann Thorac Surg1989;48;484-89.|
|24||Szabo Z, Hokanson E,Maros T, et al. High dose glucose-insulin potassium after cardiac surgery: a retrospective analysis of clinical safety issues. Masui 2003;52;420-23.|
|25||Holland FW, Brown PS Jr, Weintraub BD, Clark RE. Cardiopulmonary bypass and thyroid function: a "euthyroid sick syndrome."Ann Thorac Surg 1991; 52:46.|
|26||Murzi B, Iervasi G, Masini S, et al. Thyroid hormones homeostasis in pediatric patients during and after cardiopulmonary bypass. Ann Thorac Surg 1995; 59:481.|
|27||Dyke CM, Ding M, Abd-Elfattah AS, et al. Effects of triiodothyronine supplementation after myocardial ischemia. Ann Thorac Surg 1993; 56:215.|
|28||Novitzky D, Fontanet H, Snyder M, et al. Impact of triiodothyronine on the survival of high-risk patients undergoing open heart surgery. Cardiology 1996; 87:509.|