Abstract
Background
The safety of homologous blood transfusion has become a major con-cern for patients and physicians. Current transfusion practice is highly variable and may be associated with inappropriate blood use. Inotropic agents have been almost routinely administered perioperatively to patients undergoing cardiac surgery to overcome low cardiac output due to cardiopulmonary bypass (CPB) and cardioplegia-induced cardiac ischemic arrest. In this study, we evaluated the feasibility of blood-less and non-inotropic open-heart surgery.
Methods
Perioperative clinical data were retrospectively collected from two groups of patients undergoing open-heart surgery by one surgeon in the same sea-son. Twenty consecutive patients underwent a bloodless approach and received isoflurane-based closed-circuit general anesthesia and 20 consecutive patients (comparison group) underwent fentanyl-based anesthesia. A cell-saver was used for all patients to collect the CPB circuit blood for retransfusion. In the comparison group, conventional criteria were applied for blood transfusion and inotropic support and the goal was to keep hemoglobin > 10 g/dL and cardiac index > 2.2 L/min/m2. In the bloodless group, new criteria for blood transfusion and inotropic support were used and included (1) low cardiac output syndrome, (2) impaired hemody-namic status and mixed venous oxygen saturation, (3) inadequate urine output, (4) metabolic acidosis, (5) ischemic signs on electrocardiography, and (6) patient’s autonomy after being informed of and discussing the benefits and risks of blood transfusion.
Results
In both groups, there was no inhospital mortality and all patients were discharged in a stable condition. Eighteen of 20 (90%) patients did not receive blood transfusion, while inotropic support was not provided in 17 of 20 (85%) patients in the bloodless group; in contrast, blood transfusion and inotropic support were required for all patients in the comparison group (both: p< 0.01). All patients in the bloodless group, except one with severe chronic obstructive pulmonary disease (1-second forced expiratory volume of 0.9 L), accomplished earlier extubation (mean ± standard deviation, 1.2 ± 1.1 hours) and shorter intensive care unit stay (3.1 ± 2.1 days), as compared with patients in the comparison group (19.5 ± 2.5 hours and 5.1 ± 1.7 days, respectively; both: p< 0.01). Systemic vascular resistance was significantly lower in the bloodless group.
Conclusion
In conclusion, bloodless and non-inotropic cardiac surgery is feasible with the aid of a cell-saver and closed-circuit anesthesia in combination with new practice guidelines.
Keywords
anesthesia; closed-circuit; blood loss; surgicalblood transfusion; cardiac surgical procedures; coronary artery disease;
1. Introduction
Transfusion of blood components during cardiac surgery has markedly reduced in recent years because of improvements in surgical, anesthetic and bloodconservation techniques. Nevertheless the incidence of transfusion remains significant and, in a recent report, was performed in 29.4% of patients.1 Blood transfusion is necessary in some clinical situations to increase oxygen delivery. On the other hand, trans fusion also increases the risk of transmission of blood-borne disease, acute or delayed hemolytic reactions, transfusion-related acute lung injury,2 and disturbed immunity.
First, it is necessary to define rational criteria for initiating transfusion that augment the benefits and minimize the risks associated with transfusion. However, the cutoff value of hemoglobin to initiate blood transfusion (transfusion trigger) varies widely between institutions. Second, in the current aging society, most patients who undergo open-heart surgery have more comorbidities than were previously encountered, which increase the severity and range of diseases. Therefore, inotropic support during the perioperative period becomes more important to stabilize the hemodynamic status of the patient. Third, closed-circuit anesthesia (CCA) has been proposed to provide better hemodynamic stability.3 It also has advantages such as low waste, low expense and being environmentally friendly.
In this study, we evaluated the feasibility of bloodless and non-inotropic cardiac surgery using CCA and a cell-saver. We also present new clinical recommendations.
2. Methods
We retrospectively collected data from two groups of patients undergoing open-heart surgery by one surgeon in the same season. Twenty consecutive cardiac surgical patients who received CCA were included in the bloodless group after a pilot study had confirmed the safety of CCA and new cardiac surgery intensive care unit (CSICU) criteria for blood transfusion and inotropic support had been introduced in our institution (Table 1). The comparison group included 20 consecutive patients who received fentanyl-based anesthesia. Transfusion of blood components and other factors, and inotropic support, were performed perioperatively to maintain hemoglobin (Hb) > 10 g/L and the cardiac index > 2.2 L/ min/m2 .
CCA was induced with fentanyl (100 μg) and thiopental (4−5 mg/kg). Intubation was facilitated with pancuronium priming (0.015 mg/kg) and succinylcholine (1.25 mg/kg). Two percent isoflurane in high oxygen flow (3 L/min) was initially given to each patient for 10 minutes to ensure that isoflurane filled the functional residual capacity of both lungs and the breathing circuit. After isoflurane wash-in, the oxygen flow was reduced to less than 250 mL/ min and the isoflurane vaporizer setting was adjusted to 3−5% to maintain the inspired isoflurane at 2%. In the fentanyl-based anesthesia group, anesthesia was induced with 20−40 μg/kg of fentanyl and 0.05 mg/kg of midazolam. Pancuronium (0.1− 0.15 mg/kg) was used for endotracheal intubation. Anesthesia was maintained by midazolam at 0.1 mg/ kg/hr and fentanyl at 5−20 μg/kg/hr. An inotropic agent was administered if the patient’s mean blood pressure dropped below 65 mmHg or decreased by > 20% of the pre-induction value.
In both groups, cardiopulmonary bypass (CPB) was performed while maintaining the circuit hematocrit at 20−25%, mixed venous oxygen saturation in the normal range, and urine output in good condition. A cell-saver was used to collect the CPB circuit blood for retransfusion at the end of bypass. Mortality and morbidity, Hb concentrations, platelet counts, use of inotropic agents, perioperative cardiac indices, systemic vascular resistance (SVR), CPB duration, extubation time, and length of CSICU stay were recorded.
This study was approved and individual patient consent was waived by the Ethics Committee because it was a retrospective study.
2.1. Statistical analysis
The statistical analysis was performed by SPSS software (SPSS Inc., Chicago, IL, USA). Data are presented as mean ± standard deviation or as n and proportions. Unpaired Student’s t test was used to determine between-group differences for continuous variables and the χ2 test was used to compare the rates of bloodless procedures and non-inotropic procedures. The differences of extubation time and CSICU stay were compared with the Kaplan-Meier log-rank method. A p value < 0.05 was considered statistically significant.
3. Results
The demographic data are shown in Table 2. The two groups were comparable, although there were more patients with coronary artery disease in the comparison group. In the bloodless group, the patients’ mean age was 62.7 ± 12.3 years (range: 48− 82 years). Of the 20 patients, 13 were diagnosed with coronary artery disease while the other seven had other diseases. The following operations were successfully performed: coronary artery bypass grafting (n= 13) with a mean number of grafts of 3.8 (range: 2−5), aortic valve replacement (n= 1), mitral valve and aortic valve replacements in combination with tricuspid valve annuloplasty (n= 4), aortic valve replacement in combination with ventricular septal defect repair and reconstruction of right ventricular outflow tract (n= 1), and orthotopic heart transplantation (n= 1). Postoperative urine output was good in all patients. None of the patients developed low cardiac output syndrome or ST−T changes on electrocardiography. The postoperative course was uneventful in all patients, except one patient who developed delayed prolonged activated clotting time (437 seconds) with significant mediastinal tube bloody discharge and another who showed sustained uterine myoma bleeding.
There was no inhospital mortality in either group. Bloodless procedures were achieved in 90% of patients (n= 18/20) in the bloodless group as compared with none in the comparison group (p< 0.01) (Table 3). Only two patients needed packed red blood cells (RBCs) for CPB priming, which was done without transfusion of other blood components. Non-inotropic procedures were achieved in 85% of patients (n= 17/20) in the bloodless group, although four patients had an ejection fraction of < 30%. In contrast, all 20 patients in the comparison group required inotropic support (p< 0.01) (Table 3). In the bloodless group, three patients needed perioperative inotropic support: one patient with rheumatic valvular heart disease developed a sustained episode of cardiorespiratory arrest, which was followed by successful cardiopulmonary cerebral resuscitation; one patient with coronary artery disease had a huge left ventricular aneurysm and left ventricular ejection fraction of 20%; and one patient who underwent heart transplantation was routinely treated postoperatively with isoproterenol infusion.
The Hb levels remained stable postoperatively in the bloodless group, although the Hb level was significantly lower than that in the comparison group (p< 0.01) (Figure 1). Platelet counts did not differ significantly between the two groups and began to recover from the first postoperative day without transfusion (Figure 2). The postoperative hemodynamic condition was stable in all patients, and SVR was significantly lower in the bloodless group than in the comparison group (p< 0.01) (Table 3). The mean extubation time was 1.2 ± 1.07 hours after arrival at the CSICU in the bloodless group, except for one patient with severe chronic obstructive pulmonary disease (1-second forced expiratory volume of 0.9 L), who needed ventilator support for 15 hours. In contrast, extubation time was significantly longer in the comparison group (> 16 hours; p< 0.01) (Table 3). The mean CSICU stay was also longer in the comparison group than in the bloodless group (5.1 ± 1.7 days vs. 3.1 ± 2.1 days, respectively; p< 0.01) (Table 3). On arrival at the CSICU, the patients’ feet in the bloodless group were warm and the capillary refill time was around 1.5 seconds, which was quicker than in the comparison group (approximately 3 seconds), and consistent with the lower SVR in the bloodless group.
Download full-size image
Download full-size image
4. Discussion
The major finding of this study is that, by using CCA and applying the criteria summarized in Table 1, 90% of the patients in the bloodless group did not require blood transfusion, and 85% of the patients did not require inotropic agents, although 20% of the patients had very poor cardiac function (left ventricular ejection fraction, < 30%). Extubation time and length of CSICU stay were also significantly shorter in the bloodless group than in the comparison group.
RBC transfusion is intended to increase the oxygen-carrying capacity and subsequent oxygen delivery to the tissues. The ultimate goal of RBC transfusion is to prevent cardiac and cerebral ischemia, particularly in patients with limited cardiac reserve. However, the optimal time to perform RBC transfusion varies greatly between physicians and institutions. Indeed, what is the most appropriate time to commence transfusion and what should be considered as the “transfusion trigger”? Animal studies and anecdotal case reports suggest that the risk of mortality increases significantly in otherwise stable patients once Hb reaches 3.5−4.0 g/dL, a level at which anemia leads to significant lactate production, and an oxygen extraction ratio > 50%. In a model of coronary stenosis, this was found to occur at 6.0−7.5 g/dL. However, the true lower limit of tolerance to anemia is still unknown.4 In our patients, the Hb levels were significantly lower in the bloodless group (Table 2). These patients did not receive blood transfusion based on the new recommendations, but the clinical outcomes were comparable with or even better than those of the comparison group (Table 3). Therefore, the widely acknowledged transfusion trigger of Hb < 10 g/dL is no longer appropriate. Greenburg’s definition of anemia as a change in the dynamics of oxygen delivery and consumption departs from the traditional emphasis on the Hb level and provides a basis for a physiological approach to treatment.5 The use of an oxygen extraction ratio as a transfusion trigger could reduce the rates of allogeneic blood transfusion.6 Patients should be transfused according to the clinical conditions. Karkouti et al developed criteria to predict the need for blood transfusion, and included preoperative Hb, weight, age and sex,1 and not just the Hb level. Clinically oriented criteria for blood transfusion are more rational than solely relying on a single numerical value (i.e. Hb level). The patient’s autonomy should also be considered in current medical practice, particularly because there is no consensus for treatment.
Transfusion thresholds are relatively lower in patients undergoing cardiac surgery than in patients without severe cardiac disease.7 In other words, complete surgical correction (e.g. complete coronary revascularization in the present study) plays a key role in determining a more appropriate trigger to commence blood transfusion; inotropic support should be emphasized in this process.
Inappropriate blood transfusion is associated with increased resource utilization and contributes to longer intubation time and ICU stay, in addition to greater postoperative morbidity and 30-day mortality rates.8 Furthermore, a dose-response relationship was reported between the number of units of whole blood or packed RBCs received and the prevalence of infection, which was believed to be related to transfusion-related immunosupression.9,10
The primary adaptations to anemia included increased release of oxygen from Hb and enhanced cardiac output. The molecular dynamics of Hb result in more efficient oxygen delivery in anemia. Small reductions in the venous oxygen tension result in a dramatic increase in oxygen release. The rightward shift of the dissociation curve in anemia begins at an Hb level of 9 g/dL and is prominent at levels below 6.5 g/dL, primarily the result of increased synthesis of 2,3-diphosphoglycerate, which takes 12−36 hours to complete. The level of 2,3-diphosphoglycerate is low in stored blood and transfusion may only augment oxygen delivery several hours later. The immediate effects of transfusion are detrimental to the myocardium because of impaired myocardial oxygen metabolism and potential volume overloading.
To reduce perioperative blood product use, several blood conservation techniques during openheart surgery have been proposed, including preoperative administration of human erythropoietin, the use of a crystalloid primer for the CPB equipment,11 the recycle of blood removed from the surgical field by cardiotomy suction,12 hemodilution during CPB, retransfusion of all contents of the oxygenator at the end of CPB, routine administration of antifibrinolytic drugs, supplemental mechanical devices (i.e. cell-saver), autotransfusion of shed mediastinal chest tube drainage, modified ultrafiltration, and autologous blood donation programs. However, the cost-effectiveness of the indiscriminate application of these expensive strategies for patients undergoing isolated coronary artery bypass grafting procedures has been questioned.12−14
Blood conservation strategies cannot rely on singlecomponent therapy, which will not eliminate homologous blood transfusion.15 In contrast, multimodal approaches will be necessary to provide better outcomes without increasing the risk of adverse events. The cell-saver and multicomponent bloodconserving strategy, including meticulous surgical techniques, recycling of blood removed from the surgical field by cardiotomy suction, hemodilution during CPB, and retransfusion of all contents in the oxygenator at the end of CPB, were applied to all of our patients in both groups.
Overall, CCA better preserves peripheral tissue perfusion of patients during surgery and circulatory changes can be minimized by CCA.3 CCA with isoflurane was shown to decrease the response of vascular smooth muscle to sympathetic impulses.16 Similar findings were noted in the bloodless group in our study. We believe that it is important to explore the additional effects of CCA on cardiac surgery, including its effects on myocardial protection, blood coagulation and hemostasis.
This study demonstrated that blood does not need to be transfused and inotropic agents are not necessary for patients undergoing cardiac surgery or in the CSICU. Table 1 summarizes the factors that should be considered by the surgeon to determine whether bloodless and non-inotropic cardiac surgery is feasible and safe in a specific patient. CCA is more physiological and avoids inappropriate sympathetic activation, therefore providing better peripheral perfusion. CCA also enhances earlier recovery of consciousness and earlier extubation. In addition, CCA is associated with less SVR and potentially greater cardiac output after the operation. Higher cardiac output requires lower blood Hb concentrations to deliver similar amounts of oxygen to tissues, which leads to a lower transfusion trigger point and reduces demand for inotropic support. These not only allow for earlier postoperative communication between the CSICU staff and the patients, easier postoperative CSICU care, less patient discomfort, and less workload for the medical staff, but also mean that bloodless and non-inotropic cardiac surgery is feasible and safe. The positive surgical outcomes in the patients in this study were based on the excellent cooperation between the cardiac surgeons and the cardiac anesthetic team, favoring complete surgical correction, and the use of advanced physiological CCA. Of note, hospital costs could be reduced by the reduced blood transfusion, inotropic support and length of CSICU stay.
There are several limitations to this study. First, this was a retrospective study conducted to demonstrate the differences between two common clinical scenarios (i.e. CCA in combination with new guidelines versus fentanyl-based anesthesia and conventional guidelines). However, the study was not designed to establish superiority of a specific approach. The mechanism by which CCA reduces the need for blood transfusion and inotropic support is still uncertain, and further research is needed to elucidate the relationships between these factors. Clearly, a larger, prospective randomized controlled trial is needed to study the effect of different anesthesia techniques (i.e. CCA vs. fentanylbased anesthesia) on the trigger for blood transfusion and inotropic support, and on other clinical outcomes. However, the ethics of conducting such studies should be considered when a pilot study has already confirmed the benefits of CCA and new clinical guidelines for blood transfusion and inotropic support. Second, the sample size was too limited to reflect the beneficial effects of bloodless and non-inotropic cardiac surgery on mortality, morbidity and hospital cost. Third, the bloodless group included patients without coronary artery disease. This raises the concern about the comparability of the two groups; however, the patients included in both groups were treated consecutively by the same surgeon in the same season to reduce selection bias.Never theless, from a clinical point of view, coronary or non-coronary disease should not influence the decision to use or not use inotropic support and/ or blood transfusion. Of note, more of the patients in the bloodless group had worse heart function than in the comparison group, which is likely to have resulted in greater perioperative blood transfusion and inotropic support. Finally, this study only presented perioperative data to demonstrate the feasibility of bloodless and non-inotropic cardiac surgery; we did not assess the long-term outcomes in either group, which warrants further study.
We conclude that bloodless and non-inotropic cardiac surgery is feasible with the aid of a cellsaver and CCA in combination with new clinical guidelines.
Acknowledgments
We wish to thank Professor Chung-Yuan Lin, Department of Anesthesia and Critical Care, University of Chicago, Chicago, U.S.A. for his dedicated visiting professorship in teaching and demonstrating the closed-circuit anesthesia.