Ann Thorac Surg 2006;81:2084-2087
© 2006 The Society of Thoracic Surgeons
Original article: Cardiovascular
Limitations of EuroSCORE for Measurement of Risk-Stratified Mortality in Aortic Arch Surgery Using Selective Cerebral Perfusion: Is Advanced Age No Longer a Risk?
Kaoru Matsuura, MD,
Hitoshi Ogino, MD
*
,
Hitoshi Matsuda, MD,
Kenji Minatoya, MD,
Hiroaki Sasaki, MD,
Toshikatsu Yagihara, MD,
Soichiro Kitamura, MD
Department of Cardiovascular Surgery, National Cardiovascular Center, Suita, Osaka, Japan
Accepted for publication January 4, 2006.
* Address correspondence to Dr Ogino, Department of Cardiovascular Surgery, National Cardiovascular Center, 5-7-1 Fujishirodai, Suita, Osaka, 565-8565 Japan (Email: hogino{at}hsp.ncvc.go.jp).
 |
Abstract
|
|---|
BACKGROUND: The European system for cardiac operative risk evaluation (EuroSCORE) is a risk stratification tool for perioperative mortality of cardiothoracic surgery that was developed in Europe and validated in North America in more than 500,000 patients. The operative mortality of aortic arch surgery has been improved by various novel operative techniques and adjuncts, whereas the number of such procedures for elderly patients has recently been increasing. The aim of this study was to examine the usefulness of the EuroSCORE, and our modification of it regarding age, in predicting mortality after aortic arch repair performed with selective cerebral perfusion.
METHODS: We reviewed 358 consecutive patients with a mean age of 69 ± 10 years undergoing aortic arch repair with selective cerebral perfusion between January 1993 and February 2004. Observed in-hospital mortality was compared with predicted mortality as determined by both additive and logistic EuroSCOREs. We also evaluated a version of the EuroSCORE modified for age, which was obtained by subtracting the contribution of age from the original EuroSCORE. Score validities were assessed by calculating the areas under receiver operating characteristic curves.
RESULTS: Overall hospital mortality was 6.2% compared with 7.7% (additive EuroSCORE) and 11.8% (logistic EuroSCORE). Area under the receiver operating characteristic curve was 0.58 for the additive EuroSCORE and 0.58 for the logistic EuroSCORE as well. The overall age-unrelated EuroSCOREs were 5.1% (additive) and 5.2% (logistic), respectively, and areas under the receiver operating characteristic curve were 0.70 for additive and 0.69 for logistic.
CONCLUSIONS: The original additive and logistic EuroSCOREs overpredicted mortality in this patient group, whereas the age-unrelated EuroSCORE was better in predicting mortality.
|
Introduction
|
|---|
The number of aortic surgeries has recently increased, and the prevalence of aortic disease is higher in elderly patients than in the younger population [1, 2]. Because of the improvements of outcome with the development of operative techniques, adjuncts, and postoperative care, operative indications have been expanded for elderly patients [3–8]. Furthermore, the invasiveness of surgery and surgical mortality for elderly patients has been improved enough that advanced age is no longer a criterion for exclusion from aortic surgery. It is therefore important to determine whether advanced age is still a risk for mortality after aortic surgery.
Homogeneity of subject population is important in studies of thoracic aortic surgery, because this type of surgery features many variations in its adjuncts or procedures. In this study, we limited subjects to patients who underwent aortic arch surgery using selective cerebral perfusion (SCP) to eliminate other surgical subfactors affecting outcome. We have aggressively used SCP for patients undergoing surgery requiring circulatory arrest [9], and we now consider aortic arch repair using SCP a standard procedure.
Various risk stratification systems have been developed in the last decade, the validity of which has been widely accepted by clinicians throughout the world [10]. The European system for cardiac operative risk evaluation (EuroSCORE) (online calculator available at www.euroscore.org) is a risk stratification tool for perioperative mortality of cardiothoracic surgery that was designed and developed in Europe and validated in North America [11, 12]. In regard to age, every 1 point was added per 5 years for age older than 60 years.
One of the purposes of this study was to examine the usefulness of the EuroSCORE in predicting mortality after aortic arch repair using SCP. We also designed a modified EuroSCORE obtained by subtracting the contribution of age from the original EuroSCORE, and we examined the usefulness of it as well for the prediction of mortality.
|
Patients and Methods
|
|---|
Institutional approval for this study was obtained and each patient participating in the study gave informed consent to serve as a subject. In our institution, 358 consecutive patients with atherosclerotic aneurysm (n = 239) or aortic dissection (n = 119) underwent aortic arch repair using SCP between January 1993 and February 2004. All patients undergoing elective, urgent, or emergency aortic arch repair were included. Eighty-four patients underwent surgery on an emergency basis because of rupture or acute type A aortic dissection. Baseline demographic and clinical data were available for all patients. Mean age was 69 ± 10 years. Baseline characteristics and preoperative variables are shown in Table 1.
Operation
All operations were performed through a median sternotomy under standard anesthesia. Cardiopulmonary bypass was established with arterial cannulation and bi-caval drainage. Myocardial protection was achieved using intermittent antegrade and retrograde cold-blood cardioplegia. Selective cerebral perfusion was used in all patients for intraoperative brain protection. The patients were cooled to a core body temperature of 18°C to 28°C. All of the arch vessels were anastomosed individually with quadrifurcated grafts.
Both additive and logistic EuroSCOREs were calculated using the online calculator. Patients were divided into four groups by additive EuroSCORE: (1) low-risk group (3–5), (2) medium-risk group (6–8), (3) high-risk group (9–11), and (4) very high-risk group (>11). Observed in-hospital mortality was compared with predicted mortality as determined by both additive and logistic EuroSCOREs in each group. A modified version of the EuroSCORE for age (modified EuroSCORE) was obtained by subtracting the contribution of age from the original EuroSCOREs, both additive and logistic.
Statistical Analysis
All values are mean percentage ± standard deviation. Score validity was assessed by calculating the area under the receiver operating characteristic (ROC) curve, which was a plot of true positive rate against false positive rate for the different possible cut points for both the original EuroSCORE and the modified EuroSCORE. Areas under the ROC curve were also calculated to determine the accuracy of prediction. All analyses were performed using SAS statistical software, version 8.02 (SAS Institute Inc, Cary, NC).
|
Results
|
|---|
Overall observed hospital mortality was 6.2%. Overall predicted mortalities calculated by original and modified EuroSCOREs and area under the ROC curve are shown in Table 2. The predicted mortalities determined using additive and logistic EuroSCOREs were 7.7% and 11.8%, respectively. The ROC curves for the original additive EuroSCORE are shown in Figure 1. Area under the ROC curve was 0.58 for the original additive EuroSCORE and 0.58 for the original logistic EuroSCORE as well. The overall modified EuroSCOREs were 5.1% (additive) and 5.2% (logistic), respectively. The ROC curves for the modified additive EuroSCORE are shown in Figure 2. Area under the ROC curve was 0.70 for the modified additive EuroSCORE and 0.69 for the modified logistic EuroSCORE.
View this table:
[in this window]
[in a new window]
|
Table 2. Overall Predicted Mortalites Calculated by Original and Age-Unrelated EuroSCORE and Areas Under the Receiver Operating Characteristic Curve
|
|
Observed and predicted mortality stratified by risk band are shown in Table 3. The number of patients was classified by risk band: 40 = low-risk band, 212 = medium-risk band, 95 = high-risk band, and 11 = very high-risk band. Observed mortality and predicted mortalities were 2.5%, 4.6% (additive EuroSCORE), and 4.4% (logistic EuroSCORE) for the low-risk band (additive EuroSCORE < 5); 6.1%, 7.2% (additive EuroSCORE), and 9.1% (logistic EuroSCORE) for the medium-risk band (6–8); 7.4%, 9.7% (additive EuroSCORE), and 18.6% (logistic EuroSCORE) for the high-risk band (9–11); and 9.1%, 12.3% (additive EuroSCORE), and 32.8% (logistic EuroSCORE) for the very high-risk band (> 12).
The mean age at operation of the patients (n = 22) who died early after operation was 67.3 ± 10.1 years, whereas that of the patients who survived (n = 336) was 69.4 ± 10.0 years (p = 0.34).
|
Comment
|
|---|
Various risk stratification systems have been developed in the past decade, the validity of which has been widely accepted by clinicians throughout the world [10]. The EuroSCORE is a risk stratification tool for perioperative mortality of cardiothoracic surgery, which was established using a typical group of patients who underwent cardiac surgery [11, 12]. In the design of the original EuroSCORE, the major pathology of patients was coronary artery disease, and only a few patients underwent surgery of the ascending aorta. Although many previous reports have described the validity of risk stratification with the EuroSCORE for cardiac surgery, few have referred to aortic surgery [13–19]. Kawachi and colleagues [19] reported that the EuroSCORE is reliable not only for patients undergoing cardiac surgery, but also for those undergoing thoracic aortic surgery. However, thoracic aortic surgery includes various strategies and adjuncts which affect surgical outcome. Aortic arch surgery has been refined and standardized with antegrade selective cerebral perfusion and is now a commonly performed type of thoracic aortic surgery [4, 7]. In this study, we limited subjects to patients who underwent aortic arch surgery using SCP to eliminate other factors affecting outcome. Although one of the aims of this study was to examine the usefulness of the EuroSCORE in predicting mortality in this patient group, the areas under the ROC curve for the additive and logistic scores were low, reflecting the low sensitivity and specificity of the EuroSCORE in this group.
Another aim of this study was to improve the predictive value of the EuroSCORE for this group. Barmettler and colleagues [18] reported that a modified version of the EuroSCORE obtained by taking into account aortic dissection (6 points) and preoperative malperfusion (12 points) significantly improved the preoperative value of the EuroSCORE in patients undergoing thoracic aortic surgery. However, the surgical mortality in our group was lower than that predicted with the original EuroSCORE. Thus it does not seem to be useful to include additional factors in the original EuroSCORE when modifying it, and it is best to omit factors having little effect on the outcome.
As surgical outcomes have improved, surgical indications for elderly patients have increased in importance [1, 2]. Significant progress during the past decade, with significant lowering of mortality, has contributed to widening of indications for aortic surgery in elderly patients. In 1999, Okita and colleagues [1] reported that the early-term and long-term results of surgery for thoracic aortic aneurysm in patients older than 70 years of age were less satisfactory than those for patients younger than 70 years of age. However, other authors have reported benefits of surgery in such elderly patients, with a functional improvement comparable with that obtained in younger patients. Hagl and colleagues [2] reported in 2001 that aortic surgery utilizing hypothermic circulatory arrest in octogenarians can be performed with acceptable risk of mortality. Although age is treated as a significant risk factor for mortality in the EuroSCORE, surgical outcome has been greatly improved even for elderly patients. This is why the EuroSCORE may overestimate mortality. Collart and colleagues [14] evaluated the usefulness of the EuroSCORE in octogenarians who underwent valvular surgery, and concluded that the logistic EuroSCORE overestimates mortality in this high-risk group of patients. The question thus arises whether age is no longer a risk factor for aortic arch surgery in the current era of highly developed surgery. Based on statistical calculations, we modified the EuroSCORE by subtracting the contribution of age from the original EuroSCORE. With this modification, we could increase the area under the ROC curve from 0.58 to 0.70 in the additive model and from 0.58 to 0.69 in the logistic model. This modified version was thus better in predicting mortality than the original additive and logistic EuroSCOREs, and provided reasonable measurements of risk-stratified mortality.
Limitations
We examined a small number of patients in only a single institution in retrospective fashion. In addition, the operations were performed by more than one surgeon.
The original additive and logistic EuroSCOREs overpredicted mortality in this patient group, whereas the modified EuroSCORE, which was obtained by subtracting the contribution of age from the original EuroSCORE was better in predicting mortality.
|
References
|
|---|
- Okita Y, Ando M, Minatoya K, et al. Early and long-term results of surgery for aneurysms of the thoracic aorta in septuagenarians and octogenarians Eur J Cardiothorac Surg 1999;16:317-323.[Abstract/Free Full Text]
- Hagl C, Galla JD, Spielvogel D, et al. Is aortic surgery using hypothermic circulatory arrest in octogenarians justifiable? Eur J Cardiothorac Surg 2001;19:417-422.[Abstract/Free Full Text]
- Galloway AC, Colvin SB, LaMendola CL, et al. Ten-year operative experience with 165 aneurysms of the ascending aorta and aortic arch Circulation 1980;80(Suppl):249-256.
- Kazui T, Washiyama N, Muhammad BAH, Terada H, Yamashita K, Takinami M. Improved results of atherosclerotic arch aneurysm operations with a refined technique J Thorac Cardiovasc Surg 2001;121:491-499.[Abstract/Free Full Text]
- Ueda Y, Okita Y, Aomi S, Koyanagi H, Takamoto S. Retrograde cerebral perfusion for aortic arch surgeryanalysis of risk factors. Ann Thorac Surg 1999;67:1879-1882.[Abstract/Free Full Text]
- Spielvogel D, Strauch JT, Minanov OP, Lansman SL, Griepp RB. Aortic arch replacement using a trifurcated graft and selective cerebral antegrade perfusion Ann Thorac Surg 2002;74:S1810-S1814.[Abstract/Free Full Text]
- Strauch JT, Spielvogel D, Lauten A, et al. Technical advances in total artic arch replacement Ann Thorac Surg 2004;77:581-590.[Abstract/Free Full Text]
- Okita Y, Ando M, Minatoya K, Kitamura S, Takamoto S, Nakajima N. Predictive factors for mortality and cerebral complications in arteriosclerotic aneurysm of the aortic arch Ann Thorac Surg 1999;67:72-78.[Abstract/Free Full Text]
- Okita Y, Minatoya K, Tagusari O, Ando M, Nagatsuka K, Kitamura S. Prospective comparative study of brain protection in total aortic arch replacementdeep hypothermic circulatory arrest with retrograde cerebral perfusion or selective antegrade cerebral perfusion. Ann Thorac Surg 2001;72:72-79.[Abstract/Free Full Text]
- Geissler HJ, Holzl P, Marohl S. Risk statitification in heart surgerycomparison of six score systems. Eur J Cardiothorac Surg 2000;17:400-406.[Abstract/Free Full Text]
- Roques F, Nashef SAM, Michel P, et al. Risk factors and outcome in European cardiac surgeryanalysis of the EuroSCORE multinational database of 19030 patients. Eur J Cardiothorac Surg 1999;15:816-823.[Abstract/Free Full Text]
- Nashef SAM, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R, EuroSCORE study group European system for cardiac operative risk evaluation (EuroSCORE) Eur J Cardiothorac Surg 1999;16:9-13.[Abstract/Free Full Text]
- Karthik S, Srinivasan AK, Grayson AD, et al. Limitations of additive EuroSCORE for measuring risk stratified mortality in combined coronary and valve surgery Eur J Cardiothorac Surg 2004;26:318-322.[Abstract/Free Full Text]
- Collart F, Feier H, Kervaul F, et al. Valvular surgery in octogenariansoperative risk factors, evaluation of Euroscore and long term results. Eur J Cardiothorac Surg 2005;27:276-280.[Abstract/Free Full Text]
- Gogbashian A, Sedrakyan A, Treasure T. EuroSCOREa systematic review of international performance. Eur J Cardiothorac Surg 2004;25:695-700.[Abstract/Free Full Text]
- Toumpoulis IK, Anagnostopoulos CE, Swistel DG, DeRose Jr JJ. Does EuroSCORE predict length of stay and specific postoperative complications after cardiac surgery? Eur J Cardiothorac Surg 2005;27:128-133.[Abstract/Free Full Text]
- Zingone B, Pappalardo A, Dreas L. Logistic versus additive EuroSCOREa comparative assessment of the two models in an independent population sample. Eur J Cardiothorac Surg 2004;26:1134-1140.[Abstract/Free Full Text]
- Barmettler H, Immer FF, Berdat PA, Eckstein FS, Kipfer B, Carrel TP. Risk-stratification in thoracic aortic surgeryshould the EuroSCORE be modified?. Eur J Cardiothorac Surg 2004;25:691-694.[Abstract/Free Full Text]
- Kawachi Y, Nakashima Y, Toshima Y, Arinaga K, Kawano H. Risk stratification analysis of operative mortality in heart and thoracic aorta surgerycomparison between Parsonnet and EuroSCORE additive model. Eur J Cardiothorac Surg 2001;20:961-966.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
N. Motomura, H. Miyata, H. Tsukihara, S. Takamoto, and from the Japan Cardiovascular Surgery Database Org
Risk Model of Thoracic Aortic Surgery in 4707 Cases From a Nationwide Single-Race Population Through a Web-Based Data Entry System: The First Report of 30-Day and 30-Day Operative Outcome Risk Models for Thoracic Aortic Surgery
Circulation,
September 30, 2008;
118(14_suppl_1):
S153 - S159.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
L. D. Berrizbeitia
Invited commentary
Ann. Thorac. Surg.,
June 1, 2006;
81(6):
2088 - 2088.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
