Safety of prolonged aortic clamping with blood cardioplegia. II. Glutamate enrichment in energy-depleted hearts

Cold blood cardioplegic techniques are limited in their ability to protect energy-depleted hearts during aortic clamping. This study extends our previous observations on the benefits of amino acid (L-glutamate) enrichment of blood cardioplegic solutions as well as the added metabolic benefits of normothermic induction of cardioplegic arrest to increase the rate of repair of energy-depleted hearts. The results demonstrate that L-glutamate enrichment of blood cardioplegic solutions significantly improves metabolic recovery (greater oxygen consumption, better anaerobic metabolism) and ventricular performance. Normothermic induction of glutamate-enriched cardioplegia allowed complete recovery of myocardial metabolism and function compared to cold blood cardioplegic technique and may be used as a form of "active resuscitation" of energy-depleted hearts.     

Safety of prolonged aortic clamping with blood cardioplegia. I. Glutamate enrichment in normal hearts

This study compares the protection provided by prolonged (4 hours) aortic clamping with glutamate-enriched potassium blood cardioplegia (n = 8) to (1) prolonged (4 hours) aortic clamping with multidose potassium blood cardioplegia without glutamate (n = 4), (2) 4 hours of continuous perfusion of the beating empty heart (n = 7), and (3) 15 minutes of normothermic ischemia (n = 10). According to measurements of myocardial oxygen uptake, left ventricular compliance, left ventricular contractility, and stroke work performance, no statistical difference could be detected between those hearts receiving blood cardioplegia either with or without glutamate enrichment. In both of these groups, myocardial protection was excellent, as demonstrated by the following: postischemic myocardial oxygen uptake 43% (p less than 0.05) above control, 95% +/- 6% recovery of the left ventricular compliance, a 97% +/- 5% return of the left ventricular contractility, and a 91% +/- 6% recovery of stroke work index. Contrary to the excellent recovery of those hearts receiving blood cardioplegic protection, those hearts undergoing 4 hours of continuous perfusion showed a 45% +/- 16% (p less than 0.05) loss of left ventricular compliance and a 72% +/- 8% (p less than 0.05) recovery of stroke work index; those hearts experiencing 15 minutes of normothermic ischemia showed a 74% +/- 6% (p less than 0.05) return of left ventricular compliance, a 30% +/- 5% (p less than 0.05) decrease in contractility, and a 56% +/- 5% (p less than 0.05) recovery of postischemic left ventricular stroke work.     

Studies of myocardial protection in the immature heart. V. Safety of prolonged aortic clamping with hypocalcemic glutamate/aspartate blood cardioplegia

This study tests the hypothesis that multidose, hypocalcemic aspartate/glutamate-enriched blood cardioplegia provides safe and effective protection during prolonged aortic clamping of immature hearts. Of 17 puppies (6 to 8 weeks of age, 3 to 5 kg) placed on vented cardiopulmonary bypass, five were subjected to 60 minutes of 37 degrees C global ischemia without cardioplegic protection and seven underwent 120 minutes of aortic clamping with 4 degrees C multidose aspartate/glutamate-enriched blood cardioplegia ([Ca++] = 0.2 mmol/L), preceded and followed by 37 degrees C blood cardioplegic induction and reperfusion. Five puppies underwent blood cardioplegic perfusion for 10 minutes without intervening ischemia to assess the effect of the cardioplegic solution and the delivery techniques. Left ventricular performance was assessed 30 minutes after bypass was discontinued (Starling function curves). Hearts were studied for high-energy phosphates and tissue amino acids. One hour of normothermic ischemia resulted in profound functional depression, with peak stroke work index only 43% of control (0.7 +/- 0.1 versus 1.7 +/- 0.2 gm x m/kg, p less than 0.05). There was 70% depletion of adenosine triphosphate (7.6 +/- 1 versus control 20.3 +/- 1 mumol/gm dry weight, p less than 0.05) and 75% glutamate loss (6.6 +/- 1 versus control 26.4 +/- 3 mumol/gm, p less than 0.05). In contrast, after 2 hours of aortic clamping with multidose blood cardioplegia preceded and followed by 37 degrees C blood cardioplegia, there was complete recovery of left ventricular function (peak stroke work index 1.6 +/- 0.2 gm x m/kg) and maintenance of adenosine triphosphates, glutamate, and aspartate levels at or above control levels adenosine triphosphate 18 +/- 2 mumol/gm, aspartate 21 +/- 1 versus control 2 mumol/gm, and glutamate 25.4 +/- 2 mumol/gm). Puppy hearts receiving blood cardioplegic perfusion without ischemia had complete recovery of control stroke work index. We conclude that methods of myocardial protection used in adults, with amino acid-enriched, reduced-calcium blood cardioplegia, can be applied safely to the neonatal heart and allow for complete functional and metabolic recovery after prolonged aortic clamping.     

Avoidance of reperfusion injury after cardioplegia.

Myocardial damage incurred by ischaemia appears during and seems to be accelerated by reperfusion, which restores recoverable cells and disrupts badly damaged ones. Vicious cycles of oedema, calcium accumulation, acidosis, oxygen toxicity, fibrillation and air and platelet emboli contribute to the reperfusion injury. The philosophy of cool low-pressure reperfusion gradually restoring temperature and pressure to normal is here contrasted experimentally with that of immediate normothermic, normotensive perfusion after 90 minutes of ischaemic cool, cardioplegic arrest. The preparation was a canine heart which was treated according to the usual clinical protocol except that one group was reperfused at normal temperature and pressure, and the other group started reperfusion cool and at a low pressure and over the next 10 minutes pressure and temperature were restored to normal. Isovolumic ventricular function studies were done before ischaemia and after recovery and showed statistically significant differences between the groups in favour of the immediate restoration of normal temperature and pressure of perfusion. Contractile velocity and systolic pressure showed very highly significant (p = less than 0.005) differences, wall stress significant (p = less than 0.025) and compliance not significant differences between the groups. Reperfusion with optimal conditions may prevent "vicious cycle" changes in ischaemically damaged but recoverable myocardium. We have shown that a step in this direction is reperfusion with blood at normal temperature and pressure rather than initially at lowered temperature and pressure.     

Amelioration of reperfusion injury following hypothermic, ischemic cardioplegia in isolated, infarcted rat hearts

The left coronary artery was ligated and myocardial infarction developed in 28 rats. Three weeks later, the hearts were excised and mounted in an apparatus for perfusion of non-working isolated hearts (Langendorff). Hypothermic (15 degrees C), ischemic cardioplegia was induced for either 2 or 3 1/2 h followed by reperfusion for 45 min. Half of the hearts were reperfused with an initially gradual rise in temperature and pressure of the perfusion fluid, whereas the other half was reperfused directly with the perfusate at 37 degrees C and 100 cm H2O pressure. The hearts were examined by transmission electron microscopy and randomized for stereological analysis based on point counting on electron micrographs. Cardioplegia of 2 h duration was tolerated better than cardioplegia for 3 1/2 h (interstitial edema; P = 0.03, fraction of altered mitochondria; P = 0.001). Particularly in the hearts undergoing the longest cardioplegia, myocardial injury was less severe following a gentle reperfusion as compared with those exposed to the clinically common abrupt technique (fraction of mitochondria in the myocyte; P = 0.03, fraction of altered mitochondria; P = 0.008). In the interstitium, the luminal area of capillaries was significantly increased and the endothelial swelling less pronounced in the groups undergoing the gentle reperfusion technique, (luminal/endothelial fraction; P = 0.01). The study shows that previously infarcted hearts are susceptible to ischemic damage even after 2 h of regular hypothermic, ischemic cardioplegia and that a gentle reperfusion technique significantly ameliorates reperfusion injury.     

Improved energy preservation following gentle reperfusion after hypothermic, ischemic cardioplegia in infarcted rat hearts

The influence of temperature and pressure during early reperfusion after 2 h of hypothermic, cardioplegic ischemia was investigated. Adenosine triphosphate (ATP) and creatinephosphate (CP) were measured after 45-min reperfusion. The experiments were carried out in normal and previously infarcted rat hearts (the left coronary artery having been ligated 3 weeks carlier). Four groups, each containing six hearts, were studied. Group 1 consisted of normal hearts reperfused with an abrupt rise in temperature and pressure, group 2 of normal hearts exposed to slowly rising temperature and pressure, and group 3 and 4 of previously infarcted hearts. Reperfusion procedures in groups 3 and 4 were the same as in group 1 and 2, respectively. The study showed that previously infarcted hearts have a lowered tolerance to ischemia and that the reperfusion technique may influence the preservation of myocardial energetics, although this influence was not statistically significant in normal hearts following only 2 h of ischemia. The gently reperfused infarcted hearts had energy stores equal to the normal hearts after 2 h of ischemia and 45 min of reperfusion, whereas the infarcted hearts reperfused in a rougher mode had significantly lowered values (P<0.05 for ATP and P<0.01 for CP).     

Protection of the hypertrophied pig myocardium. A comparison of crystalloid, blood, and Fluosol-DA cardioplegia during prolonged aortic clamping

The myocardial protective effects of crystalloid, blood, and Fluosol-DA-20% cardioplegia were compared by subjecting hypertrophied pig hearts to 3 hours of hypothermic (10 degrees to 15 degrees C), hyperkalemic (20 mEq/L) cardioplegic arrest and 1 hour of normothermic reperfusion. Left ventricular hypertrophy was created in piglets by banding of the ascending aorta, with increase of the left ventricular weight-body weight ratio from 3.01 +/- 0.2 gm/kg (control adult pigs) to 5.50 +/- 0.2 gm/kg (p less than 0.001). An in vivo isolated heart preparation was established in 39 grown banded pigs, which were divided into three groups to receive aerated crystalloid (oxygen tension 141 +/- 4 mm Hg), oxygenated blood (oxygen tension 584 +/- 41 mm Hg), or oxygenated Fluosol-DA-20% (oxygen tension 586 +/- 25 mm Hg) cardioplegic solutions. The use of crystalloid cardioplegia was associated with the following: a low cardioplegia-coronary sinus oxygen content difference (0.6 +/- 0.1 vol%), progressive depletion of myocardial creatine phosphate and adenosine triphosphate during cardioplegic arrest, minimal recovery of developed pressure (16% +/- 8%) and its first derivative (12% +/- 7%), and marked structural deterioration during reperfusion. Enhanced oxygen uptake during cardioplegic infusions was observed with blood cardioplegia (5.0 +/- 0.3 vol%), along with excellent preservation of high-energy phosphate stores and significantly improved postischemic left ventricular performance (developed pressure, 54% +/- 4%; first derivative of left ventricular pressure, 50% +/- 5%). The best results were obtained with Fluosol-DA-20% cardioplegia. This produced a high cardioplegia-coronary sinus oxygen content difference (5.8 +/- 0.1 vol%), effectively sustained myocardial creatine phosphate and adenosine triphosphate concentrations during the extended interval of arrest, and ensured the greatest hemodynamic recovery (developed pressure, 81% +/- 6%, first derivative of left ventricular pressure, 80% +/- 10%) and the least adverse morphologic alterations during reperfusion. It is concluded that oxygenated Fluosol-DA-20% cardioplegia is superior to oxygenated blood and especially aerated crystalloid cardioplegia in protecting the hypertrophied pig myocardium during prolonged aortic clamping.     

Adverse effects of low-pressure reperfusion after hypothermic cardioplegia in normal and hypertrophic hearts.

The aim of this study was to determine the effect of low-pressure and high-pressure reperfusion, with and without ventricular fibrillation, on the recovery of hypertrophic and normal hearts after hypothermic cardioplegia. Fourteen hearts rendered hypertrophic by valvular aortic stenosis and 18 normal canine hearts were subjected to 1 hour of cardioplegic arrest at 28 degrees C during cardiopulmonary bypass. Each heart was then reperfused at a coronary pressure of either 40 mm Hg (low) or 80 mm Hg (high), initially in the empty beating state and then during ventricular fibrillation. Low-pressure reperfusion produced left ventricular subendocardial ischemia in hypertrophic and in normal hearts, shown by marked depression of subendocardial blood flow, myocardial pH, and myocardial oxygen consumption. In hypertrophic hearts the ischemia was more severe and resulted in a persistent depression of left ventricular function and myocardial oxygen consumption even when coronary pressure was returned to normal levels. High-pressure reperfusion was associated with rapid and complete recovery of myocardial metabolism and function in hypertrophic and in normal hearts. During low-pressure reperfusion, ventricular fibrillation exacerbated ischemia in hypertrophic and in normal hearts. During high-pressure reperfusion, a short period of ventricular fibrillation produced no adverse effects either in hypertrophic or in normal hearts. We conclude that low-pressure reperfusion produces subendocardial ischemia in normal and in hypertrophic hearts even in the empty beating state; in hypertrophic hearts it also impairs recovery of myocardial metabolism and function. The adverse effects of low-pressure reperfusion are exacerbated by ventricular fibrillation.     

Myocardial injury in hypertrophic hearts of patients undergoing aortic valve surgery using cold or warm blood cardioplegia.

OBJECTIVES: Myocardial protection techniques during cardiac surgery have been largely investigated in the clinical setting of coronary revascularisation. Few studies have been carried out on patients with left ventricular hypertrophy where the choice of delivery, and temperature of cardioplegia remain controversial. This study investigates metabolic changes and myocardial injury in hypertrophic hearts of patients undergoing aortic valve surgery using antegrade cold or warm blood cardioplegia. METHODS: Thirty-five patients were prospectively randomised to intermittent antegrade cold or warm blood cardioplegia. Left ventricular biopsies were collected at 5min following institution of cardiopulmonary bypass, 30min after cross-clamping the aorta and 20min after cross-clamp removal, and used to determine metabolic changes during surgery. Metabolites (adenine nucleotides, amino acids and lactate) were measured using high pressure liquid chromatography and enzymatic techniques. Postoperative myocardial troponin I release was used as a marker of myocardial injury. RESULTS: Ischaemic arrest was associated with significant increase in lactate and alanine/glutamate ratio only in the warm blood group. During reperfusion, alanine/glutamate ratio was higher than preischaemic levels in both groups, but the extent of the increase was considerably greater in the warm blood group. Troponin I release was markedly (P<0.05, Mean+/-SD) lower at 1, 24 and 48h postoperatively in the cold compared to the warm blood group (0.51+/-0.37, 0.37+/-0.22 and 0.27+/-0.19 vs. 0.75+/-0.42, 0.73+/-0.51 and 0.54+/-0.38ng/ml for cold vs. warm group, respectively). CONCLUSIONS: Cold blood cardioplegia is associated with less ischaemic stress and myocardial injury compared to warm blood cardioplegia in patients with aortic stenosis undergoing valve replacement surgery. Both cardioplegic techniques, however, confer sub-optimal myocardial protection.     

Long-term survival rates after prolonged aortic cross-clamping with St. Thomas' Hospital cardioplegia

Prolonged aortic cross-clamping (in excess of 120 min) was necessary in 154 cardiac surgical patients. St. Thomas' Hospital cardioplegia was used for myocardial preservation. Quantitative polarization microscopy enabling quantitative birefringence measurements to assess the change in birefringence of the muscle fibres in response to the addition of buffer containing ATP and calcium (i.e. myocardial contractility) was used to detect whether there had been any deterioration in right or left ventricular myocardium during the bypass period. 30 day survival was 90%, long-term (60 months) survival was 80%. In single valve replacements, patients with aortic valvular replacement had 100% survival up to 92 months, whereas patients with mitral valvular replacement had survival rates of 83% after 12 months and 27% after 60 months. Survival rates after 60 months were 89% for coronary artery bypass grafting, 80% for multiple valve replacements, and 74% for combined valvular and coronary artery bypass grafting surgery. Quantitative birefringence assessment of function showed that in the surviving patients 5% had functional deterioration during bypass whereas in the non-surviving patients 70% had functional deterioration. It may be concluded that after cardiac surgery necessitating prolonged aortic cross-clamping--once the initial operative problems are overcome--reasonable long-term results can be obtained by using St. Thomas' Hospital cardioplegia.     

Myocardial protection during prolonged aortic cross-clamping. Comparison of blood and crystalloid cardioplegia

We compared the ability of blood and crystalloid cardioplegia to protect the myocardium during prolonged arrest. Twelve dogs underwent 180 minutes of continuous arrest. Group I (six dogs) received 750 ml of blood cardioplegic solution (potassium chloride 30 mEq/L) initially and every 30 minutes. Group II (six dogs) received an identical amount of crystalloid cardioplegic solution (potassium chloride 30 mEq, methylprednisolone 1 gm, and 50% dextrose in water 16 ml/L of electrolyte solution). Temperature was 10 degrees C and pH 8.0 in both groups. Studies of myocardial biochemistry, physiology, and ultrastructure were completed before arrest and 30 minutes after normothermic reperfusion. Biopsy specimens for determination of adenosine triphosphate were obtained before, during, and after the arrest interval. Regional myocardial blood flow, total coronary blood flow, and myocardial oxygen consumption were statistically unchanged in Group I (p greater than 0.05). Total coronary blood flow rose 196% +/- 49% in Group II (p less than 0.005), and left ventricular endocardial/epicardial flow ratio fell significantly in this group from 1.51 +/- 0.18 to 0.8 +/- 0.09, p less than 0.01 (mean +/- standard error of the mean. The rise in myocardial oxygen consumption was not significant in this group (34% +/- 36%, p greater than 0.05). Ventricular function and compliance were statistically unchanged in both groups. In Group II, adenosine triphosphate fell 18% +/- 3.4% (p less than 0.005) after 30 minutes of reperfusion; it was unchanged in Group I. Ultrastructural appearance in both groups correlated with these changes. We conclude that blood cardioplegia offers several distinct advantages over crystalloid cardioplegia during prolonged arrest.     

Induction of cardioplegia with blood and crystalloid potassium solutions during prolonged aortic cross-clamping

Recent controversy concerns the proper vehicle for delivery of potassium cardioplegia. In the present study, adult dogs supported by cardiopulmonary bypass were subjected to 2 hours of multidose, hypothermic potassium cardioplegic arrest with 30 minutes of reperfusion with either autologous blood or crystalloid solution as the cardioplegic vehicle. Preservation of myocardial high-energy nucleotide stores was assessed by serial left ventricular biopsies assayed for adenosine triphosphate (ATP) and creatine phosphate. Preischemic and postischemic ventricular function was assessed by the use of an isovolumic intraventricular balloon. ATP stores were equally maintained at preischemic levels after ischemia and reperfusion by both autologous blood and crystalloid solution. Although creatine phosphate stores significantly declined (P less than 0.01, both groups) after 2 hours of arrest, reperfusion allowed equal restoration of preischemic levels. Maximum first derivative of left ventricular pressure and measured velocity were not depressed by either mode of protection. Similarly, myocardial compliance, as assessed by length-tension curves, showed no change following either autologous blood or crystalloid solution. The data show equal and significant myocardial protection by multidose, hypothermic potassium cardioplegia when both delivery vehicles were used.     

Effects of initial reperfusion temperature and pressure after prolonged cardioplegic ischemic arrest. A metabolic and functional study in rat hearts

The effects of temperature and pressure during early cardiac reperfusion after 3.5 hours of hypothermic, cardioplegic ischemia were investigated in isolated Langendorff-perfused rat hearts. The hearts were randomized in two groups and subjected to different techniques of reperfusion. The group I hearts were exposed to rapidly rising perfusion pressure and temperature, and in group II slowly rising pressure and temperature were employed. After 60 min of reperfusion, left ventricular developed pressure, coronary flow and tissue content of high-energy phosphates were evaluated. Left ventricular pressure and coronary flow were significantly better preserved in group II. Recovery of adenosine triphosphate and creatine phosphate was significantly lower in group I (5.27 +/- 0.38 and 8.72 +/- 0.62 mumol x g dry weight-1) than in group II (9.31 +/- 0.41 and 14.97 +/- 0.62). The study thus demonstrated that functional recovery, restoration of coronary flow and normalization of high-energy phosphate stores after long periods of hypothermic cardioplegic ischemia can be considerably influenced by the employed reperfusion technique.     

Reversal of ischemic damage with secondary blood cardioplegia.

After severe ischemic injury, it is usually necessary to prolong bypass to enhance recovery. This study tests the hypothesis that the best reversal of ischemic damage is achieved by briefly rearresting the postischemic heart with a continuous infusion of an oxygenated cardioplegic solution (secondary blood cardioplegia) during the period when bypass must be prolonged. Twenty dogs underwent 45 minutes of normothermic ischemic arrest. Fifteen minutes after unclamping, no heart could support the systemic circulation. In all dogs, oxygen demands were lowered by extending bypass for 30 minutes. In 10 of these dogs, demands were further lowered by rearresting the heart for 5 minutes with a continuous infusion of a 37 degrees C blood cardioplegic solution (K+28 mEq/L; pH 7.6; Ca++ 1 mEq/L) at a pressure of 50 mm Hg. Hearts treated with secondary blood cardioplegia showed greater recovery in the rate of contraction (-dP/dt 75% versus 62%, p less than 0.05) and relaxation (-dP/dt 76% versus 58%, p less than 0.05), better recovery of compliance (85% versus 51%, p less than 0.05), a higher stroke work index (0.72 versus 0.50 gm-m/Kg, p less than 0.05), and more ability to augment oxygen uptake (85% versus 45%, p less than 0.05) to meet the demands of the working heart than hearts treated by prolonging bypass alone. We conclude that rearresting the heart with a brief, continuous infusion of a blood cardioplegic solution results in more complete reversal of ischemic damage than possible by prolongation of a bypass alone. We believe that the increased recovery with secondary cardioplegia results from diversion of delivered oxygen toward reparative processes rather than its being expended needlessly on electromechanical work during the time when bypass must be prolonged.     

Myocardial protection with blood cardioplegia in ischemically injured hearts: reduction of reoxygenation injury with allopurinol

Myocellular injury mediated by oxygen radicals potentially limits myocardial protection in ischemically damaged hearts. This damage may be greater with oxygen-carrying blood cardioplegic solutions. A major mechanism of oxygen radical production is the conversion of hypoxanthine to uric acid by xanthine oxidase. In 16 anesthetized dogs, we studied whether adding allopurinol, a xanthine oxidase inhibitor, to blood cardioplegia would improve recovery of left ventricular (LV) performance and oxygen consumption. Millar transducer-tipped catheters and minor axis ultrasonic crystals were placed to assess LV performance by the slope of the end-systolic pressure-minor axis diameter relationships (Emax). Following total vented bypass, the hearts underwent 30 minutes of normothermic ischemia and then hypothermic blood cardioplegia with 1 mM allopurinol (N = 8) or without allopurinol (N = 8). Postischemic LV performance was significantly better with allopurinol than without (49.5 +/- 8.0 versus 17.4 +/- 4.1% of preischemic Emax; p less than 0.004). Postischemic LV oxygen consumption in the beating working state, calculated from LV blood flow (15 microm microspheres) and oxygen extraction, was comparable to preischemic values with and without allopurinol (10.2 +/- 1.2 versus 8.6 +/- 1.2 ml O2/100 gm/min). We conclude that allopurinol enhancement of blood cardioplegia increases myocardial protection in severely ischemic ventricles.     

Whole blood cardioplegia (minicardioplegia) reduces myocardial edema after ischemic injury and cardiopulmonary bypass

While blood:crystalloid cardioplegia is the clinical standard for patients undergoing cardiopulmonary bypass (CPB), it has been postulated that whole blood minicardioplegia may benefit the severely injured heart by reducing cardioplegic volume, thereby reducing myocardial edema. To test this hypothesis, we compared the cardioprotection of a popular 4:1 blood:crystalloid cardioplegia to whole blood minicardioplegia (WB) in a porcine model of acute myocardial ischemia. Yorkshire pigs (n = 20) were placed on atriofemoral bypass and subjected to 30 minutes of global normothermic ischemia. Animals were randomized to receive either 4:1 cold cardioplegia (n = 10) or WB cold cardioplegia (n = 10) delivered antegrade continuously for 90 minutes. Baseline (BL) echocardiographic determination of left ventricular mass (LVM) was compared within groups for cardiac edema (%) measured by histologic morphometrics. All (100%) animals receiving WB were successfully weaned off CPB, whereas only 40% of animals receiving 4:1 were successfully weaned off CPB. Cardiac edema percentage (p < .004) and LVM (p < .05) were significantly decreased in the WB group compared with 4:1. WB cardioplegia increases the number of hearts successfully weaned from CPB and decreases cardiac edema in our porcine model of acute myocardial ischemia. This finding implies whole blood cardioplegia may be more protective in a select group of patients undergoing extended CPB time by decreasing myocardial edema.     

Acute blood leukocyte reduction after liver reperfusion: a marker of ischemic injury

INTRODUCTION: Reperfusion injury occurs after ischemic storage of the liver. The release of free radicals from endothelial cells leads to increased adherence of polymorphonuclear neutrophils to endothelium and further release of proteases and free radicals that alter the microcirculation and produce graft dysfunction. Acute blood leukocyte reduction after reperfusion may be an expression of this sequestration and activation of neutrophils within hepatic sinusoids. This study sought to evaluate whether reduction in white blood cells occurring immediately after reperfusion was a marker of poor graft preservation and postoperative dysfunction. METHODS: The leukocyte count was evaluated at the end of anhepatic phase and at 5 minutes after reperfusion among 65 patients undergoing liver transplantation. Group A included patients with a leukocyte reduction between the two phases greater than 50%; group B, patients with less than 50%. Hepatic enzymes, blood lactate (60 and 120 minutes after graft reperfusion) and factor V and VII and bilirubin levels (daily for 15 days after transplantation) were compared between groups to assess graft injury and postoperative dysfunction. RESULTS: Alanine aminotransferase levels were significantly higher among group A than group B at both 60 and 120 minutes after graft reperfusion. No differences were observed in lactate, and factor V and VII levels. Total bilirubin was significantly higher among group A than group B patients at 10 and 15 days postoperative. CONCLUSIONS: The acute blood leukocyte reduction after reperfusion, probably due to sequestration and activation into hepatic sinusoids, seemed to be an early intraoperative marker for poor graft preservation and function.