Volumen
8 Número 3 Mayo - Junio 1996
Artículo de Revisión
| Jean-François Baron, MD |
|
|
La hemodilución intencional (HD) fue introducida ala práctica quirúrgica debido a que la sangre obtenida de donadores se asocia con riesgos principalmente infecciosos. De cualquier forma, estudios experimentales sugieren que las adaptaciones cardiovasculares inducidas por hemodilución pueden ser deletéreas o limitantes en pacientes con enfermedad arterial coronaria (CAD). Esta revisión contempla varios aspectos concernientes a las respuestas adaptativas cardiovascular o coronaria normales. Asimismo, se describen aspectos relacionados con los efectos de la HD en coronarias enfermas tanto en modelos experimentales como en la práctica clínica. Basados en los estudios tanto experimentales como clínicos, se considera que HD normovolémica puede ser una técnica de ahorro hemático bien tolerada durante cirugía en pacientes portadores de CAD. De cualquier forma, algunos puntos deben ser resaltados: 1) La HD normovolémica no debe ser realizada en pacientes con infarto miocárdico reciente, angina inestable o falla ventricular izquierda; 2) siempre debe contarse con monitoreo cardíaco adecuado para el paciente con CAD, y 3) no debe excederse un hematocrito superior a 30%.
|
Intentional hemodilution (HD) has been introduced into surgery because transfusion donor blood is associated with significant risks. However, it has been suggested by experimental studies that cardiovascular adaptations involved by hemodilution may be either deleterious or limited in patients with coronary artery disease (CAD). This review contemplates several aspects concerning normal cardiovascular and coronary adaptations to normovolemic HD. Furthermore, several aspects concerning hemodilution effects on diseased coronaries in experimental models are also described. Based both on experimental and clinical experiences, it is considered that normovolemic HD may be a well-suited method for blood management during surgery. However, some limitations must be reemphasized: 1) normovolemic HD must not be performed either in patients with recent myocardial infarction, disabling angina or impaired left ventricular function; 2) adequate cardiac monitoring must always be part of the technique for a patient with CAD, and 3) an hematocrit greater that 30% must be respected.
Palabras clave: coronarias, infarto, fisiología cardíaca, hemodilución.
Key words: coronary disease, infarction, cardiac physiology hemodilution.
Preoperative intentional hemodilution is induced by an isovolemic exchange of whole blood with colloid or crystalloids solutions to gain autologous blood while maintaining normovolemia. Intentional hemodilution has been introduced into surgery because transfusion donor blood is associated with significant risks. The appearance of AIDS and its possible transmission by transfusion despite as election of donors, is a very sensitive problem for patients and physicians. However, it is probably not the major problem since the risk hepatitis transmission is quantitatively more important. In addition, the evidence that transfusions of homologous blood can induce immunosuppression and thereby impair the host resistance of surgical patients is a cause of new concern. These problems with homologous transfusion have promoted all techniques of autotransfusion. Among these, intentional hemodilution is the one which has probably known the greater development since it is the simplest and least expansive.
The pathophysiology of normovolemic hemodilution has been extensively studied. It has been established that a reduction in hematocrit and, as a result, in arterial oxygen content, is not deleterious since compensating mechanisms are involved to maintain systemic oxygen transport. However, it has been suggested by experimental studies that the cardiovascular adaptations involved by hemodilution may be either deleterious or limited in patients with coronary artery disease. Nevertheless, some clinical studies have been conducted in patients with coronary artery disease and have brought some conflicting results. It remains from all these experimental and clinical studies that rather than to speak about hemodilution as a contraindication in patients with coronary artery disease, it may be probably more appropriate to call up its safe limits in this setting.
Cardiovascular Adaptations to Normovolemic Hemodilution
The decrease in hematocrit results in an improvement of rheologic properties of blood. In normovolemic conditions, the enhanced blood fluidity induces a large increase in cardiac output as a result of an enhanced venous return. The stroke volume increases, while at the same time, the emptying of the left ventricle is facilitated due to a reduced afterload. It is generally admitted that the increase in cardiac output is not due to a change in heart rate as long as normovolemia is preserved. Heart rate increases as hypovolemia occurs, due to either an insufficient infusion of the plasma substitute or to its rapid extravasation. However, when considering control groups, clinical studies found a trend to a higher heart rate in patients with hemodilution. In addition, an experimental study, comparing cardiac adaptations to normovolemic hemodilution in dogs with chronic cardiac denervation to those occurring in normal dogs, found a lower increase in cardiac output in former group than in the second. In this study, this result was due to a decrease in the relative contribution of an increased heart rate to the increase in cardiac output. These results suggest an involvement of the autonomic nervous system by hemodilution through an activation of chemoreceptors. The same mechanisms might explain the increase in myocardial contractility during normovolemic hemodilution.
Despite the decreased oxygen carrying capacity, the system oxygen transport is not compromised since cardiac output increases. Based on theoretical considerations, Hint first predicted that systemic oxygen transport may be at least maintained until hematocrit reaches 20%. These results were fully corroborated by many experimental and clinical studies. Global oxygen consumption is maintained during normovolemic hemodilution for a large range of hematocrits (from 60 to 20%). This is associated with a decrease in venous oxygen content. Continuous and simultaneous measurements of oxygen pressure distribution in various organs revealed that the local PO2 in the liver, pancreas, intestine, kidney and skeletal muscle increased slightly as a result of normovolemic hemodilution. The shift of the tissue PO2 distribution profiles towards higher P02 values while the hematocrit was lowered was interpreted as a reflection of a more homogeneous flow distribution of capillary flows.
Coronary Circulation Adaptations to Normovolemic Hemodilution
The increased cardiac output is distributed to the vital organs in approximately the same fraction during normovolemic hemodilution as at normal hematocrit with the exception of the coronary circulation. Coronary blood flow increases proportionately more than other local blood flows.
This finding indicates that the increase in coronary blood flow is not only due to the decrease in viscosity but that it also reflects a coronary vasodilation. In addition, if considering the coronary perfusion pressure as the difference between the mean arterial pressure and the right atrial pressure, hemodilution may induce a decrease in the coronary perfusion pressure since the mean arterial pressure generally slightly decreases while the right arial pressure increases.
Myocardial oxygen consumption is maintained for a large range of hematocrits (from 20 to 60%) as well as for the total body. However, no change in coronary sinus oxygen saturation is observed during hemodilution even at a very low hematocrit while a decrease in. mixed venous oxygen saturation is generally observed with hematocrits below 25%. This particular adaptation of the myocardium to hemodilution is imposed by the nearly maximal oxygen extraction in basal conditions. These experimental findings have been confirmed in the human coronary circulation when compared to other tissues, a strict maintenance of myocardial oxygen supply is necessary during hemodilution.
Furthermore, the endocardial/epicardial distribution of coronary blood flow may also be modified by hemodilution. Brazier et al. demonstrated that oxygen delivery to both subepicardium and subendocardium is adequately maintained over a wide range of hemoglobin levels (until 5g/100 ml) in normal dogs with patent coronary arteries. This was achieved by a proportional increase in coronary blood flow to both areas. With a hemoglobin concentration lower than 5g/100 ml, a further increase in both total coronary and subendocardial blood flow was observed. However, there was a significant reduction in the proportion of that flow delivered to the subendocardium, leading to a decrease in endocardial/ epicardial flow ratio. These changes were associated with myocardial ischemia detected with an intracavitary electrocardiogram. These results demonstrate that during hemodilution the subendocardium is more vulnerable to ischemia than other layers because it must receive most of all of its flow during diastole. Thus, for the subendocardium to receive the same amount of flow as other layers, it must have a lower vascular resistance. Since the capacity of subendocardial vessels to vasodilate is more limited, maximum coronary dilation will occur first in this region. In addition Geha showed that the maximal coronary blood flow is no modified during Normovolemic hemodilution. Asaresult the coronary flow reserve, or the ratio of peak increase in flow after 10 second occlusion to preocclusion flow is decrease by 50% at an hematocrit of 30%. This decrease is only due to the increase in the resting flow with hemodilution. Thus, at half normal hematocrit, coronary reserve is severely compromised, indicating cardiac vulnerability, especially if a rise in myocardial oxygen requirements occurs.
Hemodilution and Experimental Coronary Stenosis or Occlusion
Experimental studies were conducted with models of acute myocardial ischemia to determine whether normovolemic hemodilution is deleterious or not. A study in open-chest dogs revealed that a decrease in arterial oxygen content obtained during hemodilution does not increase myocardial ischemia evaluated by summating S-T elevations from epicardial electrocardiographic mapping.A same decrease in arterial oxygen content obtained by hypoxia at a normal hematocrit increases the severity of ischemia in the same experiments. Similar studies performed in isolated, isovolemic hearts under controlled hemodynamic conditions confirmed that hemodilution does not increase myocardial ischemia after coronary occlusion. These last experiments suggested that an increased collateral flow during normovolemic hemodilution is the mechanism by which a decrease in arterial oxygen content is not further deleterious when obtained by hemodilution rather than by hypoxia. This increase in collateral flow is probably directly related to the decreased viscosity at the microcirculatory level. These experiments are in agreement with those of Stucker et al, who compared the myocardial performances of isolated working hearts reperfussed after a prolongated global myocardial ischemia according to the hematocrit of the perfussate. They found that after a global ischemia coronary blood flow increased in the hemodiluted group in order to maintain the same myocardial oxygen transport than in the non hemodiluted group. In addition, performance evaluated by the aortic blood flow of this isolated working heart was better in the group with hemodilution after global ischemia. These results contrasted with the similarity in both groups of the myocardial oxygen consumption.
Few studies investigated the effects of hemodilution with experimental coronay artery stenosis. Hagi et al. demonstrated that after a partial occlusion of a coronary artery, the increase in coronary blood flow induced by hemodilution was extremely limited, while it was adapted in another normal coronary artery. As a result, oxygen transport was reduced in the region irrigated by the narrowed artery. However, these experimental preparations are probably not appropriate to simulate coronary adaptations in the setting of coronary artery disease. In addition, coronary adaptations probably depends on the degree of stenosis, as well as on the existence or not of a collateral circulation.
Hemodilution an d Coronary Artery Disease
Based on these experimental date, acute normovolemic hemodilution has been considered to be contraindicated inpatients with coronary artery disease. Nevertheless, such a technique is commonly used during cardiac surgery including coronary bypass procedures. During vascular surgery, hemodilution is also commonly used despite the high incidence of coronary artery disease in such patients. Although postoperative myocardial infarction is a frequent outcome after these types of surgery, normovolemic hemodilution has not been identified as a determinant or as a contributing factor for this complication. In fact, to our knowledge, no randomized clinical study has been conducted to determine whether hemodilution may influence the prevalence of this complication.
One approach to investigate the effects of normovolemic hemodilution on coronary dynamics and on myocardial metabolism in patients with coronary artery disease is to study myocardial thallium uptake. Such a study conducted by Laxenaire et al. did not reveale any further impairment in basal thallium uptake after a normovolemic hemodilution decreasing hematocrit to 30%. Furthermore, basal thallium uptake was improved in three patients. In addition, when comparing dipyridamole-thallium scans before and after normovolemic hemodilution, no change in thallium uptake was observed in 7 of the 10 patients included in this study, an improvement was observed in 2 patients and an increase in thallium. defects was observed in only 1 patient. This last patient had a disabling angina with chest pain at rest. These results indicated that despite the decrease in coronary vascular resistances induced by hemodilution resulting from both coronary vasodilation and decrease in blood viscosity, no coronary steal phenomenon was observed in these patients since no significant change in basal thallium uptake was observed. On the contrary, the improvement in basal thallium uptake in 3 patients suggested that the collateral flow may be improved and/or that the flow downstream to the coronary artery stenosis may be better distributed. However, the increase in dipyridamole-thallium scans defects in one patient with disabling angina suggested an increased dipyridamole-induced coronary steal and fixed a limit to the tolerance of hemodilution in patients with coronary artery disease. Furthermore, this study pointed out that left ventricular function at rest was not altered by hemodilution in these patients with coronary artery disease. However, one should wonder to what extent these results might be extrapolated to anesthetic situations in which both global and myocardial oxygen consumption may dramatically vary, especially during recovery from general anesthesia. In this situation, a myocardial oxygen imbalance may impair left ventricular function, which in turn may limit system oxygen transport.
We conducted a randomized study in anesthetized patients with coronary artery disease to investigate the effects of normovolemia hemodilution on left ventricular function, systemic oxygen transport and global oxygen consumption during the perioperative period. On preoperative myocardial thallium gammatomography, all patients included had at least one defect on initial scans with a redistribution on delayed scans. Their left ventricular ejection fraction determined by gated radionuclide angiography was greater than 0.50. After induction of general anesthesia, patients were randomly assigned to a hemodiluted group or to non-hemodiluted group. In the hemodiluted group, blood was withdrawn and simultaneously replaced by gelatins to achieve a final hematocrit of about 25%. In this group, hematocrit was maintained to its initial value in the nonhemodiluted group. We noted that during the preoperative period the cardiac index did not significantly increase in the hemodiluted group and was not different from the values of the control group. As a result, systemic oxygen transport was significantly lower after hemodilution than in the other group. Nevertheless, global oxygen consumption was not different between the two groups, this was achieved by a significant decrease in mixed venous oxygen content. The lack of increased cardiac index in these patients with coronary artery disease and normal left ventricular function may not be explained by an insufficient volume replacement since the same amount of colloids was infused. In addition, no significant changes in pulmonary capillary wedge pressure and in left ventricular end diastolic area monitored by transesophageal echocardiography, were observed. An impairment in left ventricular function due to myocardial ischemia was also unlikely since no significant chage in left ventricular ejection fraction area was noted. An increased venous compliance due to fentanyl benzodiazepine anesthesia may not be excluded and would have interfered with the effects of hemodilution on venous return. During the postoperative period, after rewarning, global oxygen consumption significantly increased to the same extent in both groups. However, cardiac index was not significantly higher in the hemodiluted group. This situation does not appear to be deleterious since there is a trend to a lower lactate concentration in the hemodiluted group. In addition, myocardial ischemic episodes retrospectively evidenced by the analysis of holter tapes were not more frequent in the hemodiluted group than in the control group. This study illustrates that normal hemodynamic adaptations to hemodilution may not occur during the intra and postoperative period in patients with coronary artery desease, even with a normal preoperative left ventricular function. Several mechanisms may lead to these results: interference with general anesthesia, impairement in left ventricular function, myocardial ischemia, decreased autonomic nervous system tone by general anesthesia. Some of these mechanisms are not deleterious as illustrated in this study. Therefore, a cardiac monitoring included ECG monitoring with ST-T segment analysis, and an hemodynamic monitoring to evaluate preload pressures, could be recommended in patients scheduled for major surgery.
|