myocardial protection. during Prolonged Aortic Cross-Clamping wih Cold Blood Potassium Cardioplegia

Transcription

1 Mvocardial Protection wih Cold Blood Potassium Cardioplegia during Prolonged Aortic Cross-Clamping Frank P. Catinella, M.D., Joseph N. Cunningham, Jr., M.D., Peter X. Adams, M.D., Steven L. Snively, M.D., Ronald I. Gross, M.D., and Frank C. Spencer, M.D. ABSTRACT The efficacy of cold blood potassium cardioplegia during periods of aortic cross-clamping (greater than 100 minutes) was assessed in 127 patients undergoing a variety of open-heart surgical procedures at New York University Medical Center from January, 1978, to April, Ischemic intervals ranged from 100 to 267 minutes (mean, 128 minutes). Cardiac-related deaths occurred in only 3 patients (2.4'/0), and overall operative mortality was 8.7% (11 patients). The rate of perioperative was 10%. Fourteen patients (11%) required vasopressor support or balloon counterpulsation after cardiopulmonary bypass despite the lengthy crossclamp intervals. Multivariate analysis revealed no significant relationship between the length of cross-clamp time and operative mortality (p = 0.291, incidence of perioperative (p = 0.54), or the occurrence of low-output syndrome postoperatively (p = 0.68). These findings suggest that cold blood potassium cardioplegia provides adequate myocardial protection when periods of arrest as long as 3 to 4 hours are required for complex cardiac surgical procedures. Most cardiac surgical procedures can now be accomplished with minimal operative morbidity and mortality. Intraoperative myocardial and states of postoperative low cardiac output, common complications in the past, are seen far less frequently now. To a large degree, these changes can be attributed to the evolution of more effective methods of intraoperative myocardial protection. Clinical and experimental data presented by From the Department of Cardiovascular Surgery, New York University School of Medicine, New York, NY. Accepted for publication Apr 27, Address reprint requests to Dr. Cunningham, Department of Surgery, New York University Medical Center, Suite 6D, 530 First Ave, New York, NY Follette and associates [l] prompted us in 1978 to adopt cold blood potassium cardioplegia as our primary method of intraoperative myocardial protection. Preliminary results were quite encouraging and revealed that proper application of the cold potassium arrest technique preserved myocardial adenosine triphosphate (All'), myocardial ultrastructure, and postoperative myocardial function [2]. Other studies also have indicated the safety and reliability of this technique; therefore, we currently employ this method of arrest for all open-heart surgical procedures performed at this institution [31. This clinical study was undertaken to examine the efficacy of cold blood potassium arrest during periods of aortic cross-clamping (longer than 100 minutes) at the time of open-heart operation. We specifically wished to define the relationship of the duration of arrest by this method to the occurrence of perioperative complications such as myocardial, lowoutput syndrome, and cardiac or noncardiac causes of death. Methods and Materials Method of Arrest Following institution of cardiopulmonary bypass, systemic blood is cooled to 25 C and the aorta is cross-clamped. Cardiac arrest is achieved by immediate infusion of 1 to 1.5 liters of the cold blood arrest solution (10" to 15 C; potassium 30 meqll, ph 8, hematocrit 28%, 350 mosm) into the Aortic root or directly into the coronary ostia (Fig 1). Topical iced lavage (4 C Plasmalyte) is used in addition to lower myocardial temperature initially and to maintain profound myocardial hypothermia (less than 20 C) during the period of arrest. Reinjection of the arrest solution is carried out every by The Society of Thoracic Surgeons

2 229 Catinella et al: Myocardial Protection with Cold Blood Potassium Cardioplegia Reservoir Heat Exchanger Table 1. Procedures Performed on 127 Patients with Aortic Cross-Clamp Times Greater than 100 Minutes Pressure \ Fig 1. Apparatus for delivery of cold blood arrest solution. (EKG = electrocardiographic.) minutes (500 ml) or if electromechanical activity returns. Myocardial temperature is constantly monitored by an indwelling thermistor probe. Care is taken to record temperature in both the right and left ventricles (especially in patients with coronary disease) to ensure uniform cooling. Reinjection is carried out if temperature rises above 20 C. Injectate pressure is monitored, and the arrest solution is infused at a rate to maintain an aortic root pressure of 60 to 80 mm Hg in order to avoid hypoperfusion or hyperperfusion. After the clamps are removed, systemic flow rate and pressure are lowered for 2 to 3 minutes to avoid overperfusion injury of the flaccid, arrested heart. Serum potassium levels are monitored hourly during bypass, and treatment is instituted if needed. Material A total of 127 patients operated on between January, 1978, and April, 1979, were identified as having aortic cross-clamp times greater than 100 minutes. They ranged from 31 to 83 years old (mean, 58 years); 66% were men (84 patients) and 34% women (43 patients). Surgical procedures performed in this group of highrisk patients are depicted in Table 1. Fiftyseven patients (45%) were in New York Heart Association Functional Class I11 and 61 patients (48%), in New York Heart Association Functional Class IV [4]. Table 2 further outlines other preoperative indicators of high risk among this group of patients. Duration of aortic occlusion ranged from 100 Patients Procedure No. Percentage AVR MVR 5 4 CABG AVR + CABG MVR + CABG 4 3 Multiple valve replacement Multiple valve replacement CABG Other AVR = aortic valve replacement; MVR = mitral valve replacement; CABG = coronary artery bypass grafting. Table 2. Additional Preoperative Indicators of High Risk in 127 Patients Absolute Finding Frequency Percentage Bacterial endocarditis Previous myocardial History of congestive heart failure Previous cardiac operation Disease of > 3 vessels Abnormal ventriculogram LVEDP > 25 mm Hg LVEDP = left ventricular end-diastolic pressure. to 267 minutes (mean, 128 minutes). Figure 2 depicts the distribution of aortic cross-clamp time for these 127 patients. Half of the patients had aortic occlusion times greater than 2 hours, while 4 required cross-clamp times greater than 3 hours. As can be seen from Figure 2, aortic cross-clamp times for this group of patients were skewed toward occlusion times less than 120 minutes. For this reason, statistical analysis of a nonparametric or distribution-free nature (chisquare analysis) was employed to evaluate the relationship of length of aortic occlusion with certain other variables recorded in this study. Since postoperative serum glutamic-oxaloacetic transaminase (SGOT) and creatine phos-

3 230 The Annals of Thoracic Surgery Vol 33 No 3 March 1982 * TIME [Minutes] Fig 2. Cold blood potassium cardioplegia; frequency of distribution of cross-clamp times great than 100 minutes. phokinase (CPK) levels have shown to correlate well with electrocardiographic evidence of, we employed our previously published criteria in the designation of perioperative or postoperative myocardial [5]. Patients who exhibited the definite appearance of new Q waves in the postoperative period were considered to have sustained a myocardial. In addition, patients exhibiting serum values for SGOT and CPK of more than 200 and more than 2,000 IU, respectively, were considered to have sustained a myocardial. These criteria have proved to be more effective indicators of perioperative injury in our institution than CPK-MB enzyme studies. Results Operative mortality from all causes was 8.7% (11 patients); but only 3 patients died of cardiac-related causes (2.4%). Thirteen patients (10.2%) experienced perioperative myocardial as determined by the previously defined electrocardiographic or enzymatic criteria. These 13 patients all exhibited the appearance of new Q waves postoperatively and, in addition, had either elevations of SGOT or CPK isoenzymes of greater than 200 or 2,000 IU, respectively. No other patient in the series exhibited these electrocardiographic or enzymatic changes. Nine patients (7.1%) required pressor Table 3. Analysis of Operative Mortality following Prolonged Periods of Aortic Cross-Clamp Time (>ZOO Minutes) in 127 Consecutive Patients Cause of Death CARDIAC-RELATED DEATHS (N = 3) Myocardial Myocardial Myocardial Cross-Clamp Time (min) NONCARDIAC-RELATED DEATHS (N = 8) Aortic dissection Aortic dissection Renal failure Ruptured abdominal aneurysm Respiratory failure SBE Pulmonary hypertension Neurological SBE = subacute bacterial endocarditis support following discontinuation of cardiopulmonary bypass. Twelve patients required intraaortic balloon counterpulsation intraoperatively and 2 additional patients (total, 14; 11%) required balloon insertion during the first 24 hours after operation. Table 3 presents an overall analysis of the observed early and late mortalities. All 3 patients experiencing cardiac death sustained mas-

4 231 Catinella et al: Myocardial Protection with Cold Blood Potassium Cardioplegia NUMBER OF PATIENTS Cardiac-related deaths All deaths 2L CROSSCLAMP TIME [minutes] Fig 3. Relationship of aortic cross-clamp time to mortality in 127 patients with prolonged periods of aortic occlusion. Number of patients is based on total number of deaths, i.e., 100% = 11 patients at risk at 100 minutes (3 cardiac-related and 8 noncardiac-related deaths) For cardiac-related deaths, p = 0.54; for all deaths, p.= sive intraoperative myocardial s. The leading causes of noncardiac-related deaths followed no specific pattern and included aortic dissection, pulmonary or renal failure, endocarditis, or neurological problems. Figure 3 depicts the relationship of crossclamp time to mortality. No relationship was found between duration of aortic occlusion and cardiac-related deaths (p = 0.54). Length of cross-clamp time seemed to correlate better with total operative mortality, but the relationship was still insignificant (p = 0.29). No significant relationship could be shown to exist between length of cross-clamp time and the incidence of perioperative myocardial (p = 0.54) or need for postoperative inotropic support (p = 0.68). Comment The safe upper limit of cardioplegic arrest during open-heart operation has not yet been conclusively established. Earlier reports by Roe and associates [6] described excellent clinical results when single-dose crystalloid potassium arrest was combined with profound myocardial hypothermia for periods of up to 2 hours. Tyers [7], Conti [8], and their colleagues reported similar results when reinjection cardioplegic arrest techniques were used. These excellent clinical results were parallel to those previously reported from our institution using the technique of multidose crystalloid potassium arrest in combination with topical myocardial cooling 191. The search for a more ideal arrest solution continues on an international level. Several different approaches are popular, and each appears to have its own theoretical advantage. In 1980, Christlieb and Clark [lo] reported preliminary results favoring the use of intracellular-like calcium-bearing crystalloid solutions instead of extracellular-like cardioplegia solutions. Bleese and colleagues [ll] advocate use of an isoosmolar, low sodium (50 mm), procainecontaining cardioplegic mixture with increased onconicity (addition of hydroxyethyl starch) to achieve adequate arrest and protection clinically for patients requiring extended crossclamp times. In a recent report of 167 patients requiring cross-clamp times greater than 120 minutes, they had an operative mortality of 7.2%, which is similar to that described in this survey. Only two of their 12 deaths described resulted from perioperative [ll]. Braimbridge and Cankovic-Darracott [12] demonstrated the efficacy of cold arrest with the St. Thomas solution. During the past five years, they have noted a lowered mortality and have shown improved left ventricular preservation by use of needle biopsy techniques to assess cellular chemical changes. Bretschneider and his colleagues [13] suggested that up to 600 minutes of arrest at 15 C can be achieved safely in the dog heart by proper administration of their histidine-buffered cardioplegic solution. Follette and co-workers [l] suggested the use of oxygenated pump blood as a vehicle for cardioplegic arrest in We adopted their technique of multidose sanguineous potassium arrest using alkaline blood (ph 8) cardioplegia. Our preliminary and subsequent observations, on both a laboratory and clinical level, have been quite encouraging and reinforce our belief that cold blood potassium arrest provides a very effective means of myocardial protection during prolonged periods of aortic cross-clamping. Barner and co-workers [14] have compared intermittent cold crystalloid arrest with cold potassium cardioplegia in two consecutive series of patients undergoing coronary artery bypass grafting. Although the mean aortic cross-

5 232 The Annals of Thoracic Surgery Vol 33 No 3 March 1982 clamp times were double in the cold blood group, the incidence of perioperative decreased from 14.3% in the initial series to 5.6% in the cold blood arrest group [14]. Engelman and associates [15] also described the efficacy of blood cardioplegia in myocardial preservation by demonstrating oxygen utilization and uptake by the myocardium when cold blood was used as a vehicle for potassium arrest. Their study suggested that blood cardioplegia was significantly more effective in providing a medium for myocardial oxygen utilization than either oxygenated or unoxygenated crystalloid cardioplegic solutions [15]. These latter observations parallel similar findings from our institution, which have been documented repeatedly on a clinical level.* These indications of oxygen delivery to the myocardium are impressive in view of the wellunderstood effect of temperature on the oxyhemoglobin dissociation curve [161. We have reported previously [2] results in detailed biochemical, ultrastructural, and hemodynamic studies on patients undergoing open-heart operation during which a cold blood arrest technique was utilized. The findings clearly indicate that with proper application of the cold blood arrest technique, there was adequate preservation of high-energy phosphate stores, avoidance of irreversible ultrastructure changes, and absence of postoperative low-output syndrome even after prolonged periods of arrest. This same study and another [17] also demonstrated that improper application of the cold blood arrest technique with lack of attention to details, such as monitoring of myocardial temperature, frequency and pressure of reinjection, and electrocardiographic monitoring, usually results in failure to achieve adequate myocardial preservation. Clinical results obtained at our institution are similar to those demonstrated in the canine heart model. We have observed no change in high-energy phosphate level, left ventricular function, or compliance following reperfusion after 180 minutes of uninterrupted cold blood potassium arrest in the dog heart model Il81. Critics of use of blood cardioplegia have *Catinella FP et al: Unpublished data, suggested that the use of cold blood might result in the presence of aggregates or rouleau formation that would produce sludging in the coronary microcirculation. Laks and colleagues [191 investigated this potential problem and found no such formation of aggregates or rouleau, and consistently reported negative results of indirect Coomb s test. Our current methodology involves cooling the cardioplegic perfusate no lower than 5 C to prevent this potential problem. Cooling below this level might well result in lysis of red cells as the freezing point is approached. A second possible problem associated with usage of blood potassium cardioplegia is related to the non-newtonian characteristics of the solution. We have noted clinically that following addition of potassium to the cold blood solution, uniform concentrations of potassium can only be obtained if the solution is constantly recirculated. By use of sequential sampling techniques during delivery, we have noticed on occasions wide variations of potassium concentrations (ranging from 10 to 90 meqll) when such recirculation is not carried out. Because of the possible injurious effects of such high potassium concentrations on the myocardial vasculature, we advocate the use of this system to provide constant mixing of the arrest solution [2]. Although the information needed by most cardiac surgeons today would be best provided by control randomized series comparing the various existing formulas for myocardial protection, no such study has been reported on either a clinical or laboratory basis. Furthermore, reports of consecutive series utilizing cold crystalloid arrest versus cold blood potassium cardioplegia are sparse. The results of this present report, nevertheless, confirm our premise that proper application of the cold blood arrest technique provides adequate myocardial protection during prolonged periods of aortic cross-clamping. While we have not found the need clinically to extend any period of cross-clamping beyond 4.5 hours, we are confident that this can be safely achieved if necessary for correction of complex surgical problems. It is important to reemphasize the lack of any relationship shown in this study between the length of cross-clamp time and operative mortality, incidence of perioperative

6 233 Catinella et al: Myocardial Protection with Cold Blood Potassium Cardioplegia, and occurrence of low-output syndrome postoperatively. This observation is even more important in view of the fact that 50% of the patients had cross-clamp times greater than 120 minutes, and no cardiacrelated deaths or perioperative infarcts occurred in those patients arrested for periods greater than 4 hours. It is advantageous that arrest solutions deliver oxygen and substrate during periods of aortic cross-clamping. Substrate delivery can be easily accomplished by adding the essential components for maintenance of cellular energy stores to any arrest vehicle. Oxygenation, on the other hand, can be accomplished only by use of a solution such as blood or stroma-free hemoglobin. Stroma-free hemoglobin is impractical and bubbling of oxygen into asanguineous arrest solutions under high tension may be dangerous because of release of air bubbles from the solution into the coronary arteries. For this reason, it seems not only safer but simpler to use cold blood from the heart-lung machine to accomplish the dual goals of oxygenation and substrate replenishment during arrest. It is important to buffer the arrest solution to allow for anaerobic metabolism and equally necessary to continuously pay strict attention to all of the intricate details of the arrest technique. With proper application, the safe limit of arrest with cold blood potassium cardioplegia will almost certainly always exceed the time necessary to correct the existing cardiac defect. References 1. Follette DM, Mulder DG, Maloney JV Jr, Buckberg GD: Advantages of blood cardioplegia over continuous coronary perfusion or intermittent ischemia: experimental and clinical study. J Thorac Cardiovasc Surg 76: , Cunningham JN Jr, Adams PX, Knopp EA, et al: Preservation of ATP, ultrastructure and ventricular function after aortic crossclamping and reperfusion: clinical use of blood potassium cardioplegia. J Thorac Cardiovasc Surg 78: , Bamer HB, Laks H, Codd JE, et al: Cold blood as the vehicle for potassium cardioplegia. Ann Thorac Surg 28: , Criteria Committee of the New York Heart Association: Nomenclature and Criteria for Diagnosis of Diseases of the Heart and Great Vessels. Seventh edition. Boston, Little, Brown, 1973, p Rose MR, Glassman E, Isom OW, Spencer FC: Electrocardiographic and serum enzyme changes of myocardial after coronary artery bypass surgery. Am J Cardiol33: , Roe BB, Hutchinson JC, Fishman NH, et al: Myocardial protection with cold, ischemic, potassium-induced cardioplegia. J Thorac Cardiovasc Surg 73: , Tyers GFO, Manley NJ, Williams EH, et al: Preliminary clinical experience with isotonic, hypothermic potassium arrest. J Thorac Cardiovasc Surg 74: , Conti VR, Bertranou EG, Blackstone EH, et al: Cold cardioplegia versus hypothermia for myocardial protection: randomized clinical study. J Thorac Cardiovasc Surg 76: , Adams PX, Cunningham JN Jr, Trehan NK, et al: Clinical experience using potassium-induced cardioplegia with hypothermia in aortic valve replacement. J Thorac Cardiovasc Surg 75: , Christlieb IY, Clark RE: Enhancement of myocardial protection by intracellular-like cardioplegic solution. Arch Surg 115: , Bleese N, Doring V, Rodewald G: Myocardial protection by intermittent coronary cardioplegic reperfusion. In: Cardioplegia, the First Quarter Century. Proceedings of the Royal Society, London, June 9-10, 1980, pp Braimbridge MV, Cankovic-Darracott S: Five year experience with cardioplegia assessed by ventricular biopsy. In: Cardioplegia, the First Quarter Century. Proceedings of the Royal Society, London, June 9-10,1980, pp Bretschneider HJ, Gebhard MM, Preube CJ: The Bretschneider approach-intracellular formula tions. In: Cardioplegia, the First Quarter Century. Proceedings of the Royal Society, London, June 9-10, 1980, pp Barner HB, Kaiser GC, Codd JE, et al: Clinical experience with cold blood as the vehicle for hypothermic potassium cardioplegia. Ann Thorac Surg 29: , 1980 Engelman RM, Rousou JH, Dobbs W, et al: The superiority of blood cardioplegia in myocardial preservation. Circulation 62:Suppl , 1979 Harken AH: The surgical significance of the oxyhemoglobin dissociation curve. Surg Gynecol Obstet 144: , 1977 Takamoto S, Levine FH, LaRaia PJ, et al: Comparison of single-dose and multiple-dose crystalloid and blood potassium cardioplegia during prolonged hypothermic aortic occlusion. J Thorac Cardiovasc Surg 79:19-26, 1980 Catinella FP, Cunningham JN Jr, Paone G, et al: Blood potassium cardioplegia preserves All' and ventricular function for three hours of aortic crossclamping. Curr Surg (in press) Laks H, Bamer HB, Kaiser G: Cold blood cardioplegia. J Thorac Cardiovasc Surg 77: ,1979