| Research Article |
Open Access |
|
| Blood-Sparing Heart Surgery in Critically Anaemic Patients Refusing Red Blood Cell Transfusions |
| Luca Weltert |
| Cardiac Surgery Department, European Hospital, Rome, Italy |
| *Corresponding author: |
Luca Weltert
Cardiac Surgery Department
European
Hospital, Rome, Italy
Tel: +393478880617
Fax: +390665975215
E-mail:
lweltert@gmail.com |
|
| |
| Received October 02, 2011; Accepted February 13, 2012; Published February
19, 2012 |
| |
| Citation: Weltert L (2012) Blood-Sparing Heart Surgery in Critically Anaemic
Patients Refusing Red Blood Cell Transfusions. J Blood Disord Transfus S1:009.
doi:10.4172/2155-9864.S1-009 |
| |
| Copyright: © 2012 Weltert L. This is an open-access article distributed under the
terms of the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original author and
source are credited. |
| |
| Abstract |
| |
| Objective: The need for allogenic blood, plasma and platelets is an unresolved dilemma in contemporary
surgical practice. There is increasing evidence that transfusions worsen the prognosis and should be avoided
as far as possible. Patients refusing transfusion for personal reasons but requiring urgent surgery represent a
challenge to maximize blood-sparing strategies. |
| |
| Methods: As the referral hospital for heart surgery in Jehova’s Witnesses in Central Italy, the European
Hospital of Rome has developed a multimodal strategy for blood-sparing heart surgery. This strategy has been
in use since January 2006 and from early 2010 the protocol has consisted of four points: high-dose preoperative
erythropoietin administration; screening and optimization of platelet aggregation and serum coagulation;
intraoperative and early postoperative reinfusion of blood loss; and strict control of fluid administration. |
| |
| Results: Since January 2006, 202 patients who refused any kind of transfusion have been operated on at
our hospital, which represents around 4% of all people undergoing cardiac surgery at our hospital. All types of
heart surgery were undertaken, with a prevalence of Coronary Artery Bypass Grafting (CABG), ascending aorta
replacement and aortic valve replacement. Not a single unit of blood was transfused to these patients, whereas
in a comparison group consisting of 4700 patients with an average of 1.1 units of blood per patient were used.
The expected mortality as calculated by EURO score was 8.1%. The overall observed mortality was 3.1%, with
mortality directly related to anaemia accounting for 1.6%. |
| |
| Conclusion: A multimodal approach to the problem of anaemia, involving optimization of preoperative
haemoglobin level as far as possible in an emergency setting, optimization of coagulation and aggregation,
reinfusion of lost blood intra- and perioperatively and minimizing haemodilution, allows heart surgery to be
performed with a minimally increased risk in this subset of patients, casting new light on reducing the use of
allogenic transfusion in the general population. |
| |
| Introduction |
| |
| The use of blood products in cardiac surgery is still a difficult
problem in clinical practice. On the one hand their availability depends
on the rate of donations, making supply inconsistent and resulting in
frequent and unpredictable delays in the planning of interventions;
and on the other, it is now clear that transfusion can adversely affect
the patient’s prognosis and outcome in both the medium and the long
term [1,2]. In the specific context of cardiac surgery transfusion is also
an independent risk factor for both the onset of acute kidney injury
and the onset of atrial fibrillation [3]. These last two occurrences alone
represent over 70% of postoperative morbidity and are responsible for
stretches of in-hospital stay-time by over 30% [4]. The need to avoid
transfusion during major surgery is therefore a critical problem. In
everyday clinical practice, however, transfusion represents a rapid and
convenient short-term solution to life-threatening problems such as
under perfusion of vital organs, myocardial ischaemia and renal failure
[5,6]. |
| |
| In addition, established practice firmly supports the concept that
the availability and use of blood products, where clinically feasible, is
essential for responsible medical practice. In an era when medicine
is driven to avoid malpractice suits there is often a tendency to overtransfuse
patients. However, if a patient refuses transfusion – usually
for religious reasons – this cannot be ignored and so alternatives must
be found [7]. To address this situation, the European Hospital in Rome,
as a national centre of excellence for cardiac surgery and referral centre
in Italy for cardiac surgery in Jehovah’s Witnesses, has developed a
strong commitment to minimize the use of transfusions. Since January
2006 many different clinical trials have been completed, all aimed at the verification of such aids as are truly effective in reducing transfusions.
The results of these studies have been incorporated into a protocol
to provide a smooth, reliable decision-making aid in the struggle
against perioperative anaemia. The first element of this strategy is the
administration of erythropoietin [8,9]. Unfortunately, the urgency of
many of the conditions treated in cardiac surgery means that often
advantage cannot be taken of the protocols established in the 1990s to
increase the mass of circulating red blood cells (RBC), which require a
delay to surgery longer than the typical heart surgery setting of three or
four days. A new protocol was therefore developed for administering
high doses of erythropoietin in the immediate preoperative period.
The administration of 50 000 IU of erythropoietin reduced the relative
risk of transfusion to 0.43, saving approximately 0.44 blood-units
per patient. During this first study we observed no adverse events
of any sort, in either the short or the medium term; blood viscosity
and density caused no concern, as perioperative bleeding invariably reduced the haemoglobin content and the density of the blood [10].
Encouraged by this first study we began a second study, which is still
ongoing, using a single administration of 80,000 IU erythropoietin
with the expectation of achieving saturation of the receptor. The
preliminary results seem even better. The second element of the
strategy is assessment and correction of the patient’s aggregation and
coagulation status [11]. In addition to a history of drug administration
and withdrawal from coumadin, heparin, aspirin and clopidogrel, we
routinely evaluate antithrombin and INR and, when in doubt, perform
thromboelastography and aggregometry. All suboptimal results
undergo complete pharmacological correction. The third element
of this strategy is the recovery of intra- and postoperative blood
loss. The Cell Saver System and in particular the Cardiopat system, a
centrifugal pump which concentrates red blood cells extracted from
the haemorrhagic fluid, are used for salvage and recovery in the first 6
hours after the operation [12-14]. |
| |
| A randomized study currently in press has shown that this method
leads to a saving of almost 0.5 units of blood per patient (1.37 units
transfused in controls vs 0.94 in the Cardiopat group). The system is
versatile and can be used for any operation, but is still difficult for nurses
to manage and inadequate to handle heavy intraoperative bleeding.
Active collaboration with the developers will lead to the creation of a
second-generation product that bridges earlier defects. |
| |
| The fourth element of the strategy is a restrictive policy on fluid
dilution. A simple checklist designed to check the rate of infusion
of all the venous accesses can significantly reduce intraoperative
haemodilution. The simultaneous use of these four strategies is our gold
standard, to which it is crucial to add respect for transfusion triggers in
patients who agree. For example, lowering the transfusion trigger from
8.5 g/dl to 8.0 g/dl in the absence of poor organ perfusion can further
reduce the use of transfusion in the absence of additional risks. |
| |
| Materials and Methods |
| |
| Although in the general population the notion of transfusion is
totally acceptable, for Jehovah’s Witnesses it is not. |
| |
| Italian Law has established that the refusal to be transfused, when
expressed by a patient with self-awareness and mental clarity and
collected by a medical officer at the beginning of the inpatient episode
cannot be transgressed, even when immediate survival depends on
transfusion. This legal provision makes Jehovah’s Witnesses extremely
difficult to treat in cardiac surgery. In many cases interventions are
impossible to postpone and the need to avoid anaemia is paramount
in order to minimize organ ischaemia, primarily in the heart. Our
particular attention to blood transfusion has made our hospital the
referral centre for Jehovah’s Witnesses in Italy. In this subgroup we
apply the above-mentioned protocol with meticulous care, knowing
that in case of failure of the strategy transfusion cannot be used. Since January 2006, 202 patients have categorically refused transfusion
despite consenting to the intervention and its risks. This subset
represents approximately 4% of the total population undergoing
heart surgery at the European Hospital. The type of operation ranged
from myocardial revascularization, mitral plasty and aortic valve
replacement to replacement of the ascending aorta, as shown in Table 1.
Only one patient needed reoperation after previous surgery. We chose
to compare this population to a cohort of 4700 patients operated on
in the same timeframe. We therefore conducted a retrospective cohort
study, focused on mortality and adverse events. All patients’ risk factors
for ischaemic heart disease (family history, the presence of diabetes,
hypercholesterolemia, smoking, obesity, hypertension) as well as factors
included in the EUROScore analysis (age, gender, chronic pulmonary
obstructive disease, the presence of extracardiac arteriopathy,
neurological dysfunction, previous cardiac surgery, serum creatinine,
active endocarditis, critical preoperative state, unstable angina,
left ventricular dysfunction, recent myocardial infarct, pulmonary
hypertension, emergency conditions, post-infarct septal rupture)
and biometric parameters (height, weight and BSA) were collected
and stored in the database before running a custom randomization
application working in Windows XP. All deaths were recorded and
categorized according to the principal initial cause and adverse events
and postoperative findings were all recorded in the central database. After discharge patients returned for a 30-day postoperative check and
the follow-up file was then closed and reported for analysis. |
| |
|
Table 1: Distribution by type of operation in the subpopulation analysed. |
|
| |
| Statistical Analysis |
| |
| All data were processed by IBM SPSS version 17, including Analysis
of Variance (ANOVA) and risk analysis. |
| |
| The first step was to ascertain whether the two populations were
comparable in terms of baseline characteristics. ANOVA of the
dependent variables ‘Accepting transfusion’ or ‘Refusing transfusion’
was performed (ANOVA and UniANOVA as needed, CI 95%, p = 0.05)
on all preoperative data, namely general cardiovascular risk parameters, EUROScore risk parameters and biometric parameters. The second step
was to verify whether the two groups differed in terms of adverse events
in general and mortality. Every comparison was tested with several
statistical techniques (Student’s t-test, Yates’ continuity-corrected chi-
squared test), as appropriate. |
| |
| Results |
| |
| ANOVA confirmed the comparability of the two cohorts, as all
variables analysed proved to have no statistically significant difference
between the two groups, including primary disease characteristics. The
baseline demographic and clinical characteristics of each group are
summarized in Table 2. Table 3 shows the adverse events analysis, with
differences in terms of 30-day mortality, in-hospital complications and
mid-term morbidity. Briefly, the two populations did not differ for the
parameters considered, except in terms of atrial fibrillation and deep
vein thrombosis, which seemed to occur more frequently in the control
group (29.2% vs 33.4%, p 0.04; 1.7% vs 2.4%, p=0.03, respectively) and
of transfusions, which did not occur at all in the transfusion-refusing
population; on the other hand, the control group required 1.1 units
of blood per patient (p<0.001). Overall mortality was subdivided
according the recorded initial cause, as shown in Table 4.Within
the overall mortality in the no-transfusion subgroup we identified,
albeit with a certain degree of arbitrariness, those patients in whom
inability to correct the anaemia was the primary cause of a potentially
preventable death. To distinguish these conditions we used blood lactate
under conditions of sufficient flow measured by the thermodilution
Swan–Ganz. In these cases, the presence of anaemia < 7.0 g/dl was
considered an inducer of major organ distress. This corresponds to
about 1.6% of the subgroup, i.e. almost half the deaths in this category.
The comparison group had no occurrences in this category, as would
be expected and the inconsistency of absolute numbers (just 3 cases) in
the remaining categories clearly makes statistical inference unreliable.
Overall mortality was also analyzed with regard to the type of operation
(Table 5), although extreme fragmentation of the reports owing to the
increasing number of categories again makes any kind of statistical
inference unreliable. The expected mortality in the transfusion-refusing
group, calculated according to the EUROScore parameters, was 8.1%
and in the comparison population was 9.2%. The observed mortality
rate in the transfusion-refusing group was 3.1%. For comparison, in
the general population mortality in the same period amounted to 2.7%.
The difference was not statistically significant (p= 0. 10). Results are
shown in Figure 1. |
| |
|
Table 2: Baseline characteristics. |
|
| |
|
Table 3: Postoperative occurrences. |
|
|
| |
|
Table 4: Causes of death in the two populations. |
|
| |
|
Table 5: Causes of death in relation to type of operation. |
|
| |
|
Figure 1: Observed overall mortality, no-transfusion group vs global population, compared to predicted mortality as calculated by EUROSCore. P 0.11. |
|
| |
| Discussion |
| |
| The focus of this study was to assess whether the impossibility of
transfusing red blood cells in the setting of major surgery determines
an increase in risk of death and postoperative complications despite
a modern and tested approach using blood-saving techniques.
Blood-conservation techniques are of paramount importance in
cardiac surgery, as postoperative bleeding is common and allogenic
RBC transfusion carries the risks of clerical errors, immunological
reactions and transmission of blood borne pathogens, both familiar
and unrecognized. Moreover, allogenic blood products are quite
expensive in their production process and represent a limited resource worldwide. Nowadays, despite all efforts, a proportion of patients
undergoing cardiac surgery require allogenic RBC transfusion. The
design of this study was simple and adequately powered. All patients
presenting for heart surgery at our institution were eligible for the study
and no exclusion criteria were applied. All variables analysed showed
no statistically significant differences between the two groups. The key
findings were that observed postoperative occurrences did not differ
substantially and no difference could be established in terms of mortality.
Despite a consistent number of studies regarding each separate aspect
of blood conservation, in both general heart surgery patients and those
refusing transfusion, we could find no relevant publications with regard to an integrated strategy such as ours and this led us to disclose our
method. The ability to recruit a relatively large number of heart surgery
patients within a short period also yields interesting findings, as this
is intrinsically the best way to control for confounding factors due to
changes in approaches, drugs and techniques over time. The study lacks
the power of prospective and randomized studies, but randomization
with regard to this topic could easily be illegal and hence impossible to
realize. Another possible weakness is the lack of blindness: despite the
strict observance of internal protocols, the awareness of not being able
to transfuse can lead individuals to multiply their efforts at all times,
thereby biasing the ‘equal strategy’ assumption. |
| |
| Conclusion |
| |
| This observational study shows that it is possible to perform cardiac
surgery without allogenic transfusion, with a mortality rate that is
acceptable and comparable to that of the general population. The results
obtained in the specific subpopulation of those who refused blood
transfusions might in the future be of benefit to the entire population,
avoiding the detrimental effect of transfusion and bringing cardiac
surgery closer to the goal of not requiring blood. |
| |
|
| References |
| |
- Koch CG, Khandwala F, Li L, Estafanous FG, Loop FD, et al. (2006) Persistent effect of red cell transfusion on health-related quality of life after cardiac surgery. Ann Thorac Surg 82: 13-20.
- Koch CG, Li L, Duncan AI, Mihaljevic T, Loop FD, et al. (2006) Transfusion in coronary artery bypass grafting is associated with reduced long-term survival. Ann Thorac Surg 81: 1650-1657.
- Koch CG, Li L, Van Wagoner DR, Duncan AI, Gillinov AM, et al. (2006) Red cell transfusion is associated with an increased risk for postoperative atrial fibrillation. Ann Thorac Surg 82: 1747-1756.
- Saving and improving lives-National Blood Service. Annual report 2004
- Goodnough LT, Brecher ME, Kanter MH, AuBuchon JP (1999) Transfusion medicine. First of two parts--blood transfusion. N Engl J Med 340: 438-447.
- Karkouti K, Wijeysundera DN, Yau TM, Beattie WS, Abdelnaem E, et al. (2004) The independent association of massive blood loss with mortality in cardiac surgery. Transfusion 44: 1453-1462.
- Price S, Pepper JR, Jaggar SI (2005) Recombinant human erythropoietin use in a critically ill Jehovah's Witness after cardiac surgery. Anesth Analg 101: 325-327.
- Alghamdi AA, Albanna MJ, Guru V, Brister SJ (2006) Does the Use of Erythropoietin Reduce the Risk of Exposure to Allogeneic Blood Transfusion in Cardiac Surgery? A Systematic Review and Meta-Analysis. J Card Surg 21: 320-326
- Monk TG (2004) Preoperative recombinant human erythropoietin in anaemic surgical patients. Crit Care 8: 45-48.
- Weltert L, D'Alessandro S, Nardella S, Girola F, Bellisario A, et al. (2010) Preoperative very short-term, high-dose erythropoietin administration diminishes blood transfusion rate in off-pump coronary artery bypass: a randomized blind controlled study. J Thorac Cardiovasc Surg 139: 621-626.
- Diprose P, Herbertson MJ, O'Shaughnessy D, Deakin CD, Gill RS (2005) Reducing allogeneic transfusion in cardiac surgery: a randomized double-blind placebo-controlled trial of antifibrinolytic therapies used in addition to intra-operative CS. Br J Anaesth 94: 271-278.
- Murphy GJ, Allen SM, Unsworth-White J, Lewis CT, Dalrymple-Hay MJ (2004) Safety and Efficacy of Perioperative CS and Autotransfusion After Coronary Artery Bypass Grafting: A Randomized Trial. Ann Thorac Surg 77: 1553-1559.
- Carless PA, Henry DA, Moxey AJ, O'Connell D, Brown T, et al. (2006) Cell salvage for minimizing perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 17: CD001888.
- Klein AA, Nashef SA, Sharples L, Bottrill F, Dyer M, et al. (2008) A randomized controlled trial of cell salvage in routine cardiac surgery. Anesth Analg 107: 1487-1495.
|
| |
| |
|
|
|
This article |
DOWNLOAD |
|
CONTRIBUTE |
|
SHARE |
|
EXPLORE |
|
 |
 |
| |
|
| |
| |
| |
|
Untitled Document
|
|
|
|
|
