| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Click here to read this article as a CME activity
Department of Cardiothoracic–Vascular Anesthesia and Intensive Care, IRCCS Policlinico S. Donato, Milan, Italy
Accepted for publication October 2, 2007.
* Address correspondence to Dr Ranucci, Department of Anesthesia and Intensive Care, IRCCS Policlinico S. Donato, Via Morandi 30, San Donato–Milanese, Milan, 20097, Italy (Email: cardioanestesia{at}virgilio.it).
| Cardiothoracic anesthesiology:
The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
|
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentReferences |
|---|
Methods: A retrospective study on 4,546 consecutive patients who underwent coronary surgery was performed. The outcome (hospital mortality and length of stay in the intensive care unit) was evaluated according to body surface area, gender, and the presence of allogeneic blood transfusions.
Results: Female gender is not a risk factor for hospital mortality or prolonged intensive care unit stay. Small body surface area in men and large body surface area in women are associated with a prolonged intensive care unit stay. Transfusions are independent risk factors for both mortality and prolonged intensive care unit stay. Fresh-frozen plasma and platelet transfusion carry a higher mortality risk (odds ratio, 12) than transfusions of packed red blood cells (odds ratio, 5).
Conclusions: Female gender and small body surface area are associated with severe intraoperative hemodilution, and this may trigger blood transfusions, which are true determinants of adverse outcomes. A large body surface area in women is frequently associated with obesity (68%) and may prolong the intensive care unit stay, whereas it is not a risk factor in men. Conversely, a small body surface area is accompanied by a prolonged intensive care unit stay in men but not in women.
Female gender and small body size (body surface area, body mass index, or body weight) are considered independent risk factors for adverse outcomes in coronary surgery [1–12]. Several investigators have addressed the role of female gender [1, 3, 6, 9, 10], small body size [5, 7, 8, 11], or both [2] on outcomes in coronary surgery.
Female gender was found to be an independent predictor for early mortality [1, 2, 4, 6, 10]; one article reported increased early mortality but only for low- or medium-risk patients [3]. A higher morbidity rate was reported in women; this includes composite severe morbidity [1, 4], low cardiac output syndrome [2], perioperative myocardial infarction, need for inotropic agents, and prolonged mechanical ventilation time [9]. Interestingly, all studies addressing late mortality found no gender-related differences or even better results in female patients [1, 4, 10, 12].
Small body size was independently associated with early mortality [5, 7] as well as different kinds of morbidity, including perioperative myocardial infarction, stroke, length of hospital stay [5]; acute renal failure and prolonged hospital stay [7]; reoperation and lung dysfunction [11]; and other composite indices [8]. One study failed to detect body size–related increases in morbidity [2].
Female gender is generally associated with a smaller body size, but the respective role of the two factors is still unclear, as demonstrated by the fact that the most commonly used morbidity and mortality risk scores may include either gender [13–18], body size [18], or both [13, 15, 16, 18]. Moreover, some studies demonstrated that female gender and a small body size are associated with an increased risk of receiving allogeneic blood product transfusions [5, 7, 11, 19], and one recent study suggested that transfusions, rather than female gender, were responsible for mortality and morbidity (namely infections) after coronary operations [19].
The interaction among these three potential risk factors (female gender, small body size, and transfusions) was not completely addressed with adequate statistical tests in all the previous studies, therefore maintaining open the debate about their respective role in determining morbidity and mortality after coronary operations. We hypothesize that female gender and small body size may simply trigger transfusions, which could be the real determinant of adverse outcomes.
The aim of the present study is to provide a comprehensive analysis of the respective role of female gender, small body size, and transfusions as potential independent predictors of mortality or prolonged intensive care unit (ICU) stay after coronary operations.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Patients
A series of 4,546 consecutive patients, who underwent isolated coronary surgery with cardiopulmonary bypass (CPB) at our institution in the period December 1, 2001, through March 31, 2007, was retrospectively analyzed using a local computerized database. Sixty-three patients who underwent off-pump coronary surgery were excluded from study. All patients provided written consent to the scientific treatment of their data, and the local ethical committee waived the need for approval.
Intraoperative Anesthetic Management, Monitoring, and Cardiopulmonary Bypass
Anesthesia was induced and maintained with intravenous agents. All patients received tranexamic acid intravenously (15 mg/kg before CPB and 15 mg/kg after protamine administration). Aprotinin was never used (it is not commercially available in Italy). Cardiopulmonary bypass was conducted using either closed or open circuits, standard or phosphorylcholine-coated hollow-fiber oxygenators, and roller or centrifugal pumps according to availability. Regardless of the circuit used, the priming volume was always minimized to 800 to 1,000 mL.
Data Collection and Definitions
The following variables were collected for a preoperative profile: demographics (including body surface area [BSA] and body mass index), left ventricular ejection fraction, serum creatinine value (mg/dL), unstable angina, recent myocardial infarction, hematocrit (%, HCT), extracardiac arteriopathy, diabetes on medication, chronic obstructive pulmonary disease, neurologic dysfunction, previous cardiac surgery, chronic congestive heart failure, pulmonary hypertension, active endocarditis, and critical preoperative state.
The following variables were collected for perioperative data: urgent operation, emergency operation, CPB duration (minutes), number of distal anastomoses, internal mammary artery use, lowest HCT on CPB, lowest temperature on CPB, and need for allogeneic blood products during the operation or the ICU stay.
The following variables were collected as outcome variables: peak serum creatinine value (mg/dL), acute renal failure (peak serum creatinine value twice the baseline value), low cardiac output (need for inotropic agents for more than 24 hours), need for intraaortic balloon pump, myocardial infarction (new Q waves plus enzymatic criteria), severe lung dysfunction (more than 48 hours of mechanical ventilation as a result of poor arterial blood gas analyses), stroke, sepsis, severe morbidity (one of the following: low cardiac output, myocardial infarction, intraaortic balloon pump, stroke, acute renal failure, or sepsis), ICU stay (days), and hospital mortality (within 30 days from the operation).
Every risk condition was established according to EuroSCORE definitions [17].
Statistical Analysis
The outcome variables considered were hospital mortality and morbidity, defined as the length of ICU stay in surviving patients.
The first statistical step was to separately assess the association between gender, BSA, transfusions, and the outcome variables. The second step was identifying the preoperative and intraoperative variables associated with outcome variables using univariate analysis (relative risk with 95% confidence interval or unpaired Student’s t test when appropriate). Subsequently, outcome variables that were significantly associated (p < 0.05) with preoperative and intraoperative variables were examined with multivariable analysis (stepwise forward multivariate linear regression for continuous variables and stepwise forward multivariate logistic regression for binary variables). To avoid overfitting the model, the number of independent variables entered into each model was limited to 1 for every 10 events. When appropriate, a Bonferroni correction of the multivariable analyses was applied. Tolerance and inflation statistics were applied to assess intercorrelation between independent variables.
Other statistical procedures included the locally weighted scatterplot smoothing (LOWESS) [20] for investigating nonlinear association between continuous variables.
The data in the tables are reported as mean and standard deviation of the mean or as a count and percentage. A probability value of less than 0.05 was considered significant for all the statistical tests. Statistical calculations were performed using a computerized statistical program (SPSS 11.0, Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
|
Hemodilution during CPB was more pronounced in women than in men (lowest HCT on CPB, 23.7% ± 3.3% versus 27.6% ± 3.5%; p = 0.001), and BSA is significantly associated with hemodilution during CPB. These associations, if considered within a multivariable model of linear regression, remain significant (p = 0.001; Fig 1). The degree of hemodilution during CPB is associated (p = 0.001) with the risk of transfusing PRCs, but not FFP and platelets, after correction for the other confounding factors in a multivariable logistic regression model (Fig 2).
|
|
|
|
The same graphic relationship was created for men and women being transfused or not transfused (Fig 4). Data from patients who were not transfused confirmed the previous behavior; conversely, patients being transfused have the longer ICU stay in correspondence with larger BSA values, regardless of gender.
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
On the basis of the above information, we cannot confirm that female gender is per se a risk factor in coronary operations, as hypothesized by other authors [1–11]. We believe that this results from the inclusion of the lowest HCT on CPB within the possible confounding factors. This factor has been associated with morbidity and mortality in many previous studies [21–24].
Both female gender and a small BSA are associated with a higher degree of hemodilution during CPB; this may directly deteriorate the outcome, and even lead to transfusions as a result of a low hemoglobin level during and after the operation. Transfusions are a well-recognized factor responsible for adverse outcomes [25–27], and therefore this chain of events may justify the link between female gender or small BSA and adverse outcomes found by other authors. The deleterious effects of transfusions seem more pronounced for FFP and platelets. It is presently still debated whether platelet transfusions are associated with increased mortality and morbidity after cardiac surgery [28]. In our setting, FFP and platelet transfusions are generally used to treat postoperative bleeding, and surgical revision attributable to bleeding is always accompanied by FFP and often by platelet transfusion. Therefore, it is not surprising that these blood products, being a marker of postoperative bleeding, carry a higher mortality risk than simple PRC transfusions.
The possible role of BSA in prolonging postoperative ICU stay should be carefully addressed. Body surface area is a difficult variable from the statistical point of view: it appears that both small and large values may be associated with a prolonged ICU stay. Moreover, the behavior is different from men to women: the first have the worst outcome if small-sized, the second if large-sized. Analyzing the obesity rate on the basis of the body mass index, we could demonstrate that there is a highly different distribution of fat and muscle between large-sized men and women. Large-sized women are much more likely to be obese than are men, and this may justify their longer ICU stay: the greater amount of fat and the consequent relatively lower circulating red blood cell mass predispose them to hemodilution, transfusions, respiratory problems, and superficial and deep mediastinal infections. On the other side, small-sized men may suffer from a higher degree of hemodilution during CPB with the already mentioned relative complications.
A further matter of debate is introduced by the changes induced by transfusions. Patients who do not receive transfusions and women who are transfused correspond to the above-described gender-dependent behavior. Conversely, men being transfused have the longest ICU stay when large-sized. Our interpretation is that in this case the transfusions are probably triggered not by a low hemoglobin level (which is rare in this subset of patients), but by the presence of adverse events suggesting a higher target value of hemoglobin (eg, low output state, acute renal failure, stroke) or by a serious postoperative bleeding with or without the need for a surgical revision.
It is always difficult to address the impact of allogeneic blood products on the postoperative outcome of cardiac surgery patients. In the presence of severe complications, and especially when one or more organs are at risk for ischemia, the hemoglobin-based transfusion trigger is increased [27]. At our institution, the usual hemoglobin value for considering PRC transfusion is 8 g/dL, but it can be increased to 9 or even 10 g/dL in the presence of severe postoperative complications (low cardiac output, myocardial infarction, stroke, acute renal failure, sepsis). Therefore, the link between transfusions and adverse outcomes can be coincidental rather than causative.
The link between gender, body size, and transfusions, as well as their respective roles in changing the outcomes, has not been widely addressed in previous studies. However, Guru and coworkers [10] confirmed that female gender, if adjusted for BSA, is not an independent risk factor for morbidity after coronary operations. Koch and coworkers [9] approached the problem with a propensity-score matching, and once the gender-related groups were adequately matched, no differences in outcome were observed. Potapov and coworkers [8] again confirmed that a small body size is a risk factor independent of gender. Finally, two interesting articles discussed issues with transfusions: Schwann and coworkers [5] found that patients with a low body mass index were more hemodiluted during the operation and received more transfusions during and after the operation. They hypothesize that this could be one of the mechanisms underlying the higher morbidity rate, namely at the level of lung function and infections. Recently, Rogers and coworkers [19] confirmed that women were more prone to receive allogeneic blood products and this could explain their higher rate of infections.
The main limitation of our study is the retrospective nature of the data collection; however, it is difficult to hypothesize a prospective trial based on gender, BSA, and transfusions as independent variables, and the majority of the studies in this field are retrospective as well. Another limitation is that we are missing data about the quality of the coronary vessels in women and small BSA patients. This factor has been considered a reason for the higher morbidity and mortality rate in some studies [2, 3], together with other possible operation-related factors like the number of anastomoses and rate of internal mammary artery use (which were not significantly different between groups in our study).
The main strength of this study is that we applied a multivariable analysis including the intraoperative degree of hemodilution (which was missing in all the previous studies) in the model, and that we introduced the concept that BSA may behave differently, as a risk factor, in women versus men. This approach enabled us to conclude that a small BSA or female gender may be considered a morbidity or mortality risk factor for coronary operations basically because of the greater hemodilution during the operation and the associated higher transfusion rate. Finally, this study generates the hypothesis that applying adequate measures for containing the degree of hemodilution in women and small BSA patients could result in improvements in their postoperative outcome. This hypothesis requires adequate prospective trial testing for confirmation.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  C. G. Koch Invited Commentary Ann. Thorac. Surg., February 1, 2008; 85(2): 487 - 487. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
