| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2008;7:389-395. doi:10.1510/icvts.2007.165795 © 2008 European Association of Cardio-Thoracic Surgery
EuroSCORE directed intraaortic balloon pump placement in high-risk patients undergoing cardiac surgery – retrospective analysis of 267 patients
a Martin-Luther-Universität Halle-Wittenberg, Department of Cardiothoracic Surgery, Ernst-Grube-Strasse 40, D-06097 Halle, Germany Received 24 August 2007; received in revised form 7 January 2008; accepted 8 January 2008
*Corresponding author. Tel.: +49 941 944 9846; fax: +49 941 9449802.
Objectives: Intraaortic ballon pump replacement (IABP) is the most widely used circulatory assist device today and is utilized in a wide range of serious cardiovascular conditions. We examined the effects on mortality of pre-, intra-, or postoperative IABP support in patients undergoing cardiac surgery compared to high-risk patients without IABP support. Methods: Between June 2001 and April 2004, 267 patients either received preoperative IABP support (n=62), an intra- or postoperative IABP (n=113) or had no IABP (n=92). Perioperative mortality was calculated with the EuroSCORE. Results: Patients with preoperative IABP and without IABP support had a lower ejection fraction [37 (29; 50) % and (39 (30; 53)) % vs. (50 (39; 65)) %, P=0.0001], more frequent unstable angina (38/62 and 53/92 vs. 37/113, P=0.0004) and recent myocardial infarctions (33/62 and 51/92 vs. 26/113, P=0.0001). The number of emergency procedures was also significantly higher (36/62 and 65/92 vs. 27/113, P  0.01). Patients with intra-, or postoperative IABP support and patients without IABP support had a longer ICU-stay [7.5 (5; 17.75)) and (7 (5; 15.5)) days vs. (6 (3; 10) days, P=0.023, P=0.015]. The overall hospital stay of patients without IABP [18.5 (14; 29) days] and intra-/postoperative IABP support [19, (14; 28) days] were significantly longer (P=0.007) compared to patients with preoperative support [14 (11.5; 20.5) days]. Whereas we found a trend towards reduced mortality in high-risk non-emergency patients with preoperative support, emergency patients and patients receiving intra- and postoperative support had significantly higher mortality rates than predicted by the EuroSCORE. Both emergency and non-emergency patients without IABP insertion had a significantly higher actual mortality than predicted (29.5% vs. 13.7%, P=0.03 and 38.1% vs. 26.3%, P<0.0001). The overall actual mortality between patients with preoperative IABP insertion and patients without preoperative IABP did not significantly differ (14/62 vs. 29/92, P=0.27). The EuroSCORE proved to be a valid predictor for perioperative mortality among high-risk non-emergency and emergency patients with preoperative IABP support at lower score sums, but failed at higher score sums (>8) and among patients with intra- and postoperative IABP insertion. Conclusion: Preoperative IABP support is indicated in high-risk non-emergency patients. The benefit of preoperative IABP insertion in emergency patients and intra- and postoperative IABP support still remains controversial and needs to be elucidated in further prospective, randomized studies.
Key Words: Intraaortic balloon pump; EuroSCORE; Mortality
The intraaortic balloon pump (IABP), introduced into clinical practice in the late 1960s, has become the most widely used circulatory assist device today [1, 2]. Today, this treatment modality is routinely used in a wide range of serious cardiovascular conditions, ranging from hemodynamic stabilization in patients suffering from complications of acute myocardial infarction (AMI) or cardiogenic shock, to high-risk patients undergoing coronary artery bypass grafting (CABG) or angioplasty [3–6]. The hemodynamic effects of an IABP depend on the pre-existing cardiovascular conditions: Besides a reduced afterload, decreased left ventricular wall tension and decreased myocardial oxygen consumption during systole, the diastolic antegrade coronary blood flow usually increases dependent on the grade of stenosed coronary arteries. However, a multicenter observational study by Baskett et al. showed in 2005 a lack of evidence of effectiveness in patients with preoperative IABP implantation [7], especially for ‘prophylactic’ implantation in hemodynamically stable patients. There is still an ongoing debate for whom preoperative IABP support might be useful and recommendable. Several score systems, e.g. the Parsonnet-Score [8] have been developed during the recent decades to predict the early mortality after adult cardiac surgery. In Europe, the EuroSCORE project represents the most recent development. Because of its simple, objective and up-to-date system for risk stratification, the EuroSCORE proved to be the most widely used score system in Europe [9]. Although the EuroSCORE was developed for the prediction of early mortality after cardiac surgery, many surgeons use it as an indicator whether or not to implant an IABP preoperatively [10]. The EuroSCORE item ‘critical preoperative state’ includes, besides several other criteria, the preoperative implantation of an IABP. Our study addressed two issues. First, we examined and compared the outcome of high-risk patients with and without preoperative IABP use with patients receiving intraoperative or postoperative IABP. Second, we wanted to know if the EuroSCORE might help to identify those patients for whom preoperative IABP use is recommendable.
2.1. Patients Participants were recruited from the Department of Cardiothoracic Surgery and data for 267 consecutive patients undergoing open heart surgery with (n=175) or without (n=92) an IABP from April 2001 to June 2004 were prospectively collected. Patients were grouped primarily into preoperative, intra- or postoperative IABP or without any IABP use. The groups were further subdivided into emergency and non-emergency, but high-risk cases. High-risk was defined as severely impaired left ventricular function as reported on ventriculography or echography (ejection fraction <40%) and/or unstable angina with critical coronary anatomy (left main stem stenosis >70% and right coronary stenosis >70%). An emergent patient was operated within 12 h after referral to avoid death from cardiogenic shock. EuroSCORE (additive and logistic) based on patient and procedural risk was calculated for all patients, using the current online-version (www.euroscore.org). The predicted mortality was derived from the total score sum and the calculated probability of perioperative death (logistic model). Patients with an additive score sum  6 were considered at high risk, whereas a score sum between 3–5 placed patients at medium risk. Preoperative IABP insertion was not included in the calculation of EuroSCORE. Mortality was defined as death in the hospital. 2.3. Intraaortic balloon pumps The IABP catheter used was an 8F 40 ml sheathed profile IABP catheter (Datascope, Oakland, NJ) connected to a Datascope computerized portable console. Percutaneous femoral artery insertion was employed in all patients. Preoperative IABP insertion was either performed two hours preoperatively at the Intensive Care Unit (ICU) under radiological control or in the catheterization laboratory. Postoperative IABP insertion was performed at the ICU under radiological control.Every patient in this study received preoperatively a Swan-Ganz-Catheter for hemodynamic monitoring. Data were collected in an Excel-worksheet and then transferred into SPSS (version 11.0.1). Categorical data were analyzed with Fisher's exact test (two sided). Continuous data were first tested for normality and then analyzed either with parametric tests, e.g. Student's t-test or analysis of variance, or if no normal distribution could be assumed non-parametric tests were used, e.g. Mann–Whitney U-test. Observed and expected frequencies were analyzed with the 2-test. Data are shown either as mean±S.D. or median plus interquartile range (if no normal distribution could be assumed). A P-value <0.05 was considered significant.
3.1. Preoperative data among the study population Table 1 summarizes the preoperative data among the study participants. Data did not significantly differ between patients with preoperative and without preoperative IABP insertion. However, patients with and without preoperative IABP use presented more frequently with NYHA III–IV, had more often unstable angina, a recent myocardial infarction and a lower ejection fraction in contrast to patients with intra-/postoperative IABP insertion.
3.2. Indications for IABP insertion Almost 50% (n=85) of all IABP insertions were performed intraoperatively in order to facilitate weaning from cardiopulmonary bypass. Twenty-eight patients (16%) received the IABP postoperatively because of a low cardiac output syndrome, whereas 62 patients required preoperative IABP support due to the following reasons: poor ejection fraction (n=21), acute myocardial infarction with hemodynamic instability (n=22) and support and stabilization in the catheter lab (n=19). 3.3. Operations and surgical technique Patients with preoperative IABP insertion underwent 51 coronary bypass graftings [with extracorporeal circulation (n=38), off-pump (n=3), beating heart (n=10)], two valve replacements and nine combinations of valve procedures and CABG. Patients with intra- or postoperative IABP implantation underwent 81 coronary bypass graftings [with extracorporeal circulation (n=66), off-pump (n=8), beating heart (n=7)], 6 valve replacements, 23 combinations of valve procedures and CABG as well as two other cardiac procedures, e.g. resection of a left ventricular aneurysm. Patients without IABP insertion had similar diseases compared to patients with preoperative IABP use. Cardioplegic arrest was achieved by antegrade intermittent warm blood cardioplegia (1:1 ratio of blood to cardioplegia).Whereas the number of grafts and the frequency of IMA use did not differ between the three groups, the bypass- and aortic clamping time was significantly higher in patients with intra-/postoperative IABP insertion (Table 2). More than 50% of procedures in the preoperative IABP group were emergency operations.
3.4. Outcome after operation Patients with intra- or postoperative IABP insertion and patients without preoperative IABP insertion had more frequently a low cardiac output syndrome. Furthermore, their stay at the ICU and their overall hospital stay were significantly longer compared with patients with preoperative IABP insertion (Table 3). The duration of the IABP support was similar in both groups.
There is a high number of patients with transient renal failure and prolonged ventilation (defined as ventilation time >24 h) among patients with intra- and postoperative IABP support. 3.5. Complications after IABP insertion The balloon pump had to be removed in five patients because of limb ischemia. There was one patient with a balloon leakage and two patients with a sonographically demonstrable vessel thrombus after balloon removal. In summary, there were 8/175 (4.75%) complications after IABP insertion, but not IABP related morbidity. The mean additive EuroSCORE in patients with preoperative IABP insertion was 9.94±3.65 reflecting a mean predicted mortality of 21±16.99%, whereas the mean additive EuroSCORE in patients without preoperative IABP was calculated as 9.65±3.51, which was equivalent to a mean predicted mortality of 20±14.4%. The EuroSCORE in patients with intra- and postoperative IABP insertion was 7.08±3.67 corresponding to a mean predicted mortality of 11.97±15.09%. The mean additive EuroSCORE and the predicted mortality between the two groups (preoperative IABP or no IABP, respectively, versus intra-/postoperative IABP) were significantly different (P=0.001). Whereas the actual mortality did not differ from the predicted mortality among patients with preoperative IABP insertion, the actual mortality among patients with intra- and postoperative IABP was significantly higher than predicted (27.5% vs. 11.97%, P<0.0001, Fig. 1). The actual overall mortality among patients without IABP support proved to be statistically higher than predicted (20% vs. 31.5%, P=0.01).
In a subgroup analysis, we also compared risk-adjusted mortality between emergency versus non-emergency patients with preoperative IABP insertion (Fig. 2a). The mean additive EuroSCORE between both subgroups differed significantly (11.06±3.37 vs. 8.6±3.52, P=0.001). However, the predicted versus the actual mortality did not differ significantly in both groups. Nonetheless, we observed a trend towards reduced mortality among non-emergency patients (15.2% to 7.69%, P=n.s.).
Fig. 2b displays the calculated EuroSCORE and predicted risk-adjusted mortality between emergency versus non-emergency patients with intra-/postoperative IABP insertion. The mean additive EuroSCORE differed significantly between emergency and non-emergency patients (4.8±2.8 vs. 9.2±3.72, P<0.001). Whereas the predicted and the actual mortality showed a significant difference among emergency patients (17.1% vs. 40.7%, P<0.01), the mortalities also differed between non-emergency patients (6.9% vs. 23.2% P<0.01). Fig. 2c summarises the predicted and actual mortalities among patients without IABP support. Emergency patients had a higher mean additive EuroSCORE compared to non-emergency patients (11.3±3.7 vs. 8.0±3.0, P<0.0001). The predicted versus actual mortality among emergency patients did not statistically differ, but non-emergency patients had a higher actual mortality than predicted by EuroSCORE (29.5% vs. 13.7%, P=0.03). 3.7. EuroSCORE as a predictor of perioperative mortality Fig. 3 shows the additive EuroSCORE plotted against observed – predicted mortality % (from logistic calculation) among our study population with preoperative IABP-implantation. If there is to be a good match between observed and predicted mortality – an indicator that the EuroSCORE is predicting mortality accurately – the majority of points will cluster around 0 on the Y-axis. The overall appearance from Fig. 3 is that logistic EuroSCORE does quite accurately predict mortality at lower score sums and failed at higher score sums (8) since the values cluster in a wide range around the y-axis. However, since the number of patients in some score groups was quite small, the overall conclusion from Fig. 3 should be weighed carefully (Table 4).
The IABP has long been established as a valuable mechanical support for temporary ventricular assistance in the treatment of the failing heart [11]. Besides significant afterload reduction and a more favorable myocardial blood supply through augmentation of diastolic pressures, the IABP also leads to a redistribution of coronary blood flow toward ischemic areas of the myocardium [12, 13]. Several studies have shown that the preoperative IABP insertion reduced perioperative myocardial ischemia and thus improved outcome in high-risk patients undergoing coronary artery bypass grafting with extracorporeal circulation [14]. The major findings of our study are as follows: First, the mean EuroSCORE predicted mortality and the actual mortality rates do not statistically differ among patients with preoperative IABP insertion. Second, patients without preoperative IABP insertion have a higher actual mortality than predicted. Third, the actual mortality among patients with intra- and postoperative IABP use also proved to be significantly higher than predicted by the EuroSCORE. However, subdividing the patients into emergency and non-emergency cases in all three groups gives further insight for whom an IABP insertion might be useful and provides a survival advantage. Although we found no statistical significance among patients with prophylactic preoperative IABP insertion in high-risk non-emergency patients, our data are in line with several recently published reports [14–16], which provided contradictory results. Even though there was a trend towards reduced mortality among these patients, a prospective study with a larger sample size with better power will be required to clarify this issue. We also examined the mortality among high-risk patients without any IABP support. Schmid et al. showed that 19% of high-risk patients scheduled for surgery without preoperative IABP required intra- or postoperative IABP insertion due to low cardiac output [17]. Another study demonstrated the benefit in terms of reduced mortality of early preoperative IABP support in high-risk coronary patients [18]. Our data also support this concept because patients without any IABP support had a significant higher actual mortality than predicted. Thus, especially non-emergency patients may benefit from an early preoperative IABP placement. As recently published by Baskett et al., there is still insufficient evidence of effectiveness of prophylactic IABP use in elective, but high-risk patients undergoing CABG [7, 19]. Two studies showed no survival advantage for prophylactic IABP insertion in hemodynamically stable high-risk patients [17, 20]. Irrespective of pre- or intra/postoperative IABP implantation, emergency patients in our study population had a similar additive EuroSCORE sum and their actual mortality rates were higher than predicted, but not statistically different. Our data also suggest that emergency patients (high-risk patients requiring an operation within 12 h) do not sufficiently benefit from an intraoperative or postoperative IABP support because the in-hospital mortality could not be reduced. This finding might be explained by several factors. We think that besides advanced myocardial damage, systemic stress factors, such as a systemic inflammatory response, counteract the potential benefit of an IABP support. Whereas an IABP might facilitate the intraoperative weaning from cardiopulmonary bypass, long-term outcome and in-hospital mortality seem not to be reduced. Our data do not fully answer the question when and for whom an IABP support may be helpful. Whereas we could not show a statistically significant reduction of perioperative mortality among high-risk non-emergency patients most likely because this subgroup was too small to reach significance, emergency patients with preoperative IABP insertion had a higher actual mortality than predicted by the logistic EuroSCORE. On the other hand, patients without preoperative IABP placement had the highest overall actual mortality and non-emergency patients performed significantly worse than predicted. This finding advocates an early preoperative IABP insertion to reduce mortality, at least to the predicted value. The EuroSCORE has been the best established and validated risk model for contemporary practice [21]. The logistic EuroSCORE was shown to be superior to the standard additive EuroSCORE in predicting mortality in high-risk cardiac surgical patients [22] and could also be applied to North American cardiac surgery [23]. However, our data also show that the EuroSCORE cannot be applied to patients requiring intra- or postoperative IABP support and their actual mortality rates have been grossly underestimated. Since the EuroSCORE was developed for preoperative assessment of perioperative mortality, intra- and postoperative aggravating factors such as low cardiac output or acute renal failure cannot be considered for the calculation. We conclude from our data that preoperative IABP support in high-risk non-emergency patients should be initiated as soon as possible because these patients seem to have a reduced EuroSCORE-predicted mortality, whereas high-risk emergency patients do not benefit from preoperative IABP support. Intra- and/or postoperative IABP support should be used where appropriate, but EuroSCORE-based mortality rates do not reflect the actual rates and thus should be valued carefully. A further prospective, randomized controlled trial with more study participants is needed to determine for whom IABP support is truly beneficial.
Related Articles
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
Top
Abstract
1. Introduction
