|
|
||||||||
Ann Thorac Surg 2006;81:880-885
© 2006 The Society of Thoracic Surgeons
a Department of Anesthesiology and Intensive Care Medicine, Campus Charité Mitte, Charite-University Medicine Berlin, Berlin, Germany
b Department of Biometry, Campus Charité Mitte, Charite-University Medicine Berlin, Berlin, Germany
c Department of Cardiac Surgery, Campus Charité Mitte, Charite-University Medicine Berlin, Berlin, Germany
d Department of Anesthesia, New England Medical Center, Boston, Massachusetts
Accepted for publication September 30, 2005.
* Address correspondence to Dr Vargas Hein, Department of Anesthesiology and Intensive Care Medicine, Campus Charité Mitte, Charité-University Medicine Berlin, Schumannstrasse 20/21, D-10117 Berlin, Germany (Email: ortrud.vargas{at}charite.de).
| Abstract |
|---|
|
|
|---|
METHODS: Records from 2,683 cardiac surgery patients were retrospectively evaluated. Univariate and multivariate analyses for risk factors were performed for an ICU stay greater than 3 days. Thereafter, 2,563 patients were enrolled in a follow-up study for an observational time of 3 years after surgery.
RESULTS: Mortality was dependent on renal, respiratory, and heart failure, as well as age, elevated APACHE II scores, and reexploration. Long-term survival analyses demonstrated a significantly lower survival in patients with longer ICU stay. However, the 6-month to 3-year long-term survival was comparable with survival in patients without prolonged ICU stay.
CONCLUSIONS: Because of the increasing acuity of patients needing cardiac surgery, it is important to identify those at risk for a prolonged ICU course. It is therefore of paramount interest to implement measures throughout their entire hospital stay that would maximize organ function to improve survival and resource utilization.
| Introduction |
|---|
|
|
|---|
Hospital or 30-day mortality does not resemble patients' long-term prognosis [7]. In those patients needing a prolonged ICU stay, long-term survival has proven to be better for cardiac versus general surgical patients because of the mostly curative nature of heart surgery [1820]. The knowledge of long-term survival of these severely ill patients in comparison with patients without a prolonged ICU stay could assist decision-making about further therapeutic strategies and options [7].
The first aim of this study was to identify perioperative risk factors for the end point of ICU stay greater than 3 days in a consecutive cardiac surgery patient population. The second aim of this study was to compare long-term survival in these patients with and without a prolonged ICU stay greater than 3 days.
| Patients and Methods |
|---|
|
|
|---|
Study Population and Data Collection
The study population consisted of all cardiac surgery patients admitted postoperatively to the cardiothoracic ICU at the hospital Charité-University Medicine Berlin, Germany, for a period of 2 years from August 1, 2001, to August 31, 2003. All patients were treated intraoperatively and postoperatively according to the standard operating procedures of the Department of Anesthesiology and Intensive Care of this single facility [21]. Clinical and demographic data were collected into a computerized database by the ICU medical staff on admission and daily on every patient. Patient data were retrospectively evaluated for the study. Exclusion criteria included ICU stay less than 1 day (n = 14), age younger than 18 years (n = 118), patients readmitted to the ICU after their primary procedure owing to complications (n = 51), patients with chronic renal failure requiring dialysis (n = 35), acute liver failure patients (n = 5), patients with a primary indication for a ventricular assist device (VAD; n = 3), and patients with missing entry data (n = 344). Initially, 3,253 postoperative patients were evaluated, and records from 2,683 patients were finally involved in the study.
Definitions for Study End Point and Risk Factors
Prolonged ICU stay was defined as an ICU stay greater than 3 days. The preoperative, intraoperative, and postoperative variables assessed as possible risk factors for the study end point were based on previous published analyses, clinical judgment, and completeness of data collection [24, 11, 19, 2227]. The risk factors were defined as follows: respiratory failure, the necessity for mechanical ventilation greater than 48 hours postoperatively; acute renal failure, postoperative acute renal failure with the necessity for continuous venovenous hemofiltration (ARF-D); heart failure, the intraoperative or postoperative need for an intraaortic balloon pump (IABP); and emergency surgery, nonelective surgery. Further risk factors are defined in Table 2.
|
Statistics
A descriptive data analysis was performed for the patients in the study end point regarding risk factors described as median and 25th quartile to 75th quartile. The continuous variables were tested on normal distribution with the Lilliefors test, showing no normal distribution for all risk factors (data not shown). Univariate analyses were initially performed to identify risk factors associated with the study end point using the Mann-Whitney U test for the continuous variables and the
2 test for the categorical variables. Risk factors with a p less than 0.05 on univariate analyses were entered in the multivariate analyses to identify independent risk factors. The multivariate logistic stepwise regression was performed using a backward likelihood ratio technique for the two study end points. The continuous risk factors were entered as continuous variables. The odds ratio, 95% confidence intervals, and p values were calculated for each risk factor. The predictive performance of the risk model was assessed with the determination of the area under the receiver-operating characteristic curve. An area under the receiver-operating characteristic curve of 100% would equate to a predictive power of 100%. An arbitrary distribution of the predicted variables in both groups would produce a diagonal in the diagram, resulting in an area under the receiver-operating characteristic curve of 50%. The Hosmer-Lemeshow goodness-of-fit test was performed to evaluate how well the model was calibrated. A small
2 value and a p greater than 0.05 would show an acceptable adaptation.
Long-term survival was assessed with the use of Kaplan-Meier estimates. Data on patients who were withdrawn from follow-up were censored for analyses. Survival curves were performed for the patients with an ICU stay greater than 3 and 3 days or less. These curves were statistically compared using the log-rank test. Significance was assessed at the p less than 0.05 level.
Data were stored and analyzed by use of the software SPSS version 11 (SPSS Inc, Chicago, IL).
| Results |
|---|
|
|
|---|
|
|
0.001; Fig 1). In Figure 2, the 6-month to 3-year long-term survival is noted to be significantly decreased in the patients with an ICU stay greater than 3 days (survival, 90%) in comparison with the group of patients with an ICU stay 3 days or less (survival, 95%; p
0.001).
|
|
| Comment |
|---|
|
|
|---|
Independent risk factors for an ICU stay greater than 3 days were related to perioperative risk factors such as age, catecholamine therapy, IABP, ARF-D, respiratory failure, and reexploration. The risk factors with the highest adjustable odds ratio for the study end point ICU stay greater than 3 days were respiratory failure and ARF-D. Postoperative renal insufficiency was found in one study as an independent risk factor for a prolonged ICU stay greater than 24 hours [25]. Ryan and associates [9] found ARF-D as an independent risk factor for hospital mortality in patients with an ICU stay greater than 14 days. The incidence for respiratory failure (44%) and for ARF-D (32%) was significantly higher in patients with prolonged ICU stay in comparison with the patients with an ICU stay 3 days or less. These two organ failures are reported to be mainly related with an increased ICU stay.
The factors that lead to these organ failures may not be related directly with preoperative risk factors or surgery, but with ICU-associated problems such as ventilator-associated pneumonia [7, 9]. However, a protracted postoperative inflammatory response as a result of high catecholamine levels after cardiac surgery could also lead to organ failures [1012]. Cardiac failure, as highlighted by the need for an IABP and catecholamine therapy, were independent risk factors for an ICU stay greater than 3 days. The necessity for an IABP was identified as an independent risk factor for a prolonged ICU stay greater than 48 hours and 3 days or greater in two other studies [6, 28]. Turner and colleagues [25] also found inotropic therapy at the time of admission to the ICU to be an independent risk factor for a prolonged ICU stay greater than 24 hours. Cardiac failure seems to be relevant in the first postoperative days after surgery and has been described as postcardiotomy cardiogenic shock syndrome [26]. When patients recover promptly from this state and benefit from the curative approach of surgery, they do not have a protracted ICU stay [7]. Therefore, measures to reduce the risk of organ failure (such as the early insertion of an IABP) should be taken as early as possible to reduce the risk of a prolonged ICU stay and its associated high mortality [29, 30]. The use of phosphodiesterase inhibitors was applied but not entered in the database.
The necessity for a VAD showed a risk reduction. At first, this might seem illogical. However, a relatively short ICU stay in VAD patients is explainable owing to the early and high mortality rate of VAD patients. All the patients (n = 5) with a VAD and an ICU stay 3 days or less died in comparison with 17 (65%) of 26 patients with a VAD and an ICU stay greater than 3 days.
An increased APACHE II score was an independent risk factor for an ICU stay greater than 3 days. Studies investigating the APACHE II score for the end point ICU stay have not been published. The APACHE II score has been found as an independent risk factor for one or multiple organ failures, as well as for mortality [11, 31]. However, the APACHE II score as a predictive tool in cardiac surgery has been questioned because the scores may be to low owing to the fact that intensive care measures can normalize pathologic states, eg, hemodynamic variables [5, 16, 18].
Increased age and a reexploration were found to be independent risk factors for a prolonged ICU stay greater than 3 days. This combination of risk factors has consistently elevated mortality in cardiac surgery patients [22, 25, 32]. Undergoing a reexploration has also been found to be an independent risk factor in three studies with the study end points of ICU stay 3 days or greater and prolonged ventilation, and for the composite end point 30-day-mortality with at least one complication [4, 8, 23, 33]. Older patients (>75 years) have a longer hospital stay (51%, >14 days) with a prolonged ICU stay (3 to 7 days) and higher mortality rates [9, 24, 34, 35]. Three studies have found age to be an independent risk factor for the study end points ICU stay greater than 48 hours, 3 days or greater, and 6 days or greater [6, 28, 32]. A poorer outcome in older people was possibly related to higher preoperative comorbidity rates and severity of illness [24, 34, 36]. However, with advancing operative and anesthetic technology, morbidity and mortality rates in older patients have been decreasing in the last years [24, 34, 35]. The optimization of preoperative comorbidities should be a priority in this group of patients.
Long-term survival was significantly decreased in patients with a prolonged ICU stay in our study. One other study by Isgro and coworkers [37] compared 5-year long-term survival in cardiac surgery patients with an ICU stay greater than 5 days and less than 5 days, showing a significantly decreased survival rate of 43% in the patients with a prolonged ICU stay in comparison with the survival rate of 78% in the patients with an ICU stay less than 5 days. Holmes and colleagues [14] studied 1-year long-term survival in patients with an ICU stay greater than 48 days, showing a survival rate of 60%.
The course of the survival curve for the patients with a prolonged ICU stay showed a marked decline in survival in the first 6 months of follow-up. This decline reproduces to a high extent the high ICU and hospital mortality in patients with prolonged ICU stay mainly because of organ failures. This could be shown in our study. Therefore measures to reduce the risk of organ failure with its high mortality should be taken throughout the entire hospital course, ideally at an early stage of illness or as a preventive measure preoperatively. Measures such as the prophylactic use of an IABP in those patients with reduced ejection fraction or the early initiation of renal replacement therapy in patients with evidence of preoperative renal dysfunction have shown outcome improvements; however, the study sizes were small [30, 38]. The difference in survival decline after the first 6 months to the end of the observation time of 3 years between patients with an ICU stay greater than 3 days (7%) and patients with an ICU stay 3 days or less (4%) was numerically comparable, although significant. The same observation of a leveling in decline of survival between patients with complicated versus an uncomplicated ICU stay was made by Isgro and associates [37]. The reduced decline in death rate after 6 months has also been described in other studies after cardiac surgery with different study end points (respiratory failure and the necessity for an IABP) and has been explained with the direct effects of surgery persisting for 6 months or more [16, 20, 39]. It has been noted that high-risk patients who survive the postoperative period after curative cardiac surgery have a better long-term outcome than medically treated patients [1820]. The long-term survival results of this study show that cardiac surgery patients with a prolonged ICU stay have a comparable chance for long-term survival as patients who have uncomplicated courses if they overcome the first postoperative months. This is an important factor that should be taken into consideration when decisions have to be made about continuing therapy in critically ill postoperative cardiac surgery patients.
Data obtained in a retrospective observational study such as this can be influenced by differences in daily clinical practice. In an effort to minimize this problem, the patients in this study were uniformly treated according to standard operating procedures [21] germane to this institution. The data were obtained in a single cardiac surgery center. This fact may lead to one-sided results attributable to the influence of specific procedures and a unique patient population. This limitation was balanced with the inclusion of a large number of patients to ensure single-center reliable results. The long-term survival analyses did not include the quality of life of the surviving patients. Even though this is an important variable, our primary approach was to investigate long-term survival in a high number of cardiac surgery patients.
In conclusion, prolongation of ICU stay was mainly related to organ failures. Long-term survival analyses demonstrated a significant lower survival of patients with a longer ICU stay. However, the 6-month to 3-year long-term survival was comparable to survival in the patients without prolonged ICU stay. Measures to identify and reduce the risk of organ failure should be taken as early as possible to reduce the possibility of a prolonged ICU stay with the associated increase in morbidity and mortality.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. Dimopoulos, G.-P. Diller, E. Koltsida, A. Pijuan-Domenech, S. A. Papadopoulou, S. V. Babu-Narayan, T. V. Salukhe, M. F. Piepoli, P. A. Poole-Wilson, N. Best, et al. Prevalence, Predictors, and Prognostic Value of Renal Dysfunction in Adults With Congenital Heart Disease Circulation, May 6, 2008; 117(18): 2320 - 2328. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Gaudino, F. Girola, M. Piscitelli, L. Martinelli, A. Anselmi, C. Della Vella, R. Schiavello, and G. Possati Long-term survival and quality of life of patients with prolonged postoperative intensive care unit stay: Unmasking an apparent success J. Thorac. Cardiovasc. Surg., August 1, 2007; 134(2): 465 - 469. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. G. Healy, A. E. Wood, A. O'Neill, J. F. McCarthy, J. M. Fitzpatrick, and R. W. Watson Can preoperative modelling of individual neutrophil adhesion responses predict renal morbidity? Eur. J. Cardiothorac. Surg., June 1, 2007; 31(6): 1088 - 1093. [Abstract] [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 |