| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
© 2004 European Association of Cardio-Thoracic Surgery
Mid-term survival after cardiac surgery in elderly patients: analysis of predictors for increased mortality
a Department of Cardiothoracic Surgery, The Cardiothoracic Centre Liverpool, Thomas Drive, Liverpool L14 3PE, UK
* Corresponding author. Tel.: +44-151-293-2336; fax: +44-151-288-2371 Received August 18, 2003; received in revised form December 22, 2003; accepted January 8, 2004
This study examines and quantifies the potential risk factors for increased mid-term mortality in elderly patients (  75 years old) undergoing cardiac surgery. We undertook a retrospective analysis of 840 consecutive elderly patients who underwent cardiac surgery (CABG and/or Valve) between April 1997 and March 2002. Deaths occurring as a function of time were described using the product limit methodology of Kaplan and Meier. Cox proportional hazards analysis was used to identify preoperative risk factors for mortality with hazard ratios (HR). One hundred and sixty-two (19.3%) deaths occurred during the study period, with a total follow-up of 1866 patient years (mean 2.2 years, SD 1.5 years). Observed freedom from death in the elderly patients at 5 years was 71.7%, compared to 70.9% for the age- and sex-matched general population  Multivariate analysis for independent predictors of increased mortality found that renal dysfunction (HR 3.2;  ), valves(s) surgery (HR 1.8; ), cerebrovascular disease (HR 1.8;  ), and catastrophic state (HR 2.2;  ) were the major risk factors. We have identified and quantified several risk factors, which need to be considered when assessing patients for cardiac surgery.
Key Words: Elderly; Cardiac surgery; Risk factors; Renal dysfunction
The proportion of patients aged 75 or over undergoing cardiac surgery has steadily increased year on year in the United Kingdom [1]. Several reasons have contributed to this. Firstly, with improved living standards, people are living longer, with the predicted increase in numbers of people aged between 75 and 84 in United Kingdom of around 40% from 1991 to 2031 [2]. Secondly, there is a rising prevalence of cardiovascular disease in the elderly [3]. Finally, there has been an increase in the number of elderly patients undergoing angiography that may represent a reduced reluctance to aggressively investigate the elderly. <!?tpor=1>Although there is evidence suggesting improved quality of life for elderly patients [4,5], these patients are still at high-risk of early mortality, morbidity, and increased hospital costs [6,7]. We therefore aimed to identify and quantify risk factors for increased mortality in elderly patients undergoing cardiac surgery.
2.1. Patient population and data We performed a retrospective study on a total of 840 consecutive patients aged 75 years or older undergoing cardiac surgery between 1st April 1997 and 31st March 2002 at the Cardiothoracic Centre Liverpool. We took a cut-off of 75 years of age for our study population as these patients score more than 10 points according to the Parsonnet risk stratification tool and therefore can be regarded as high-risk cases [1]. Data was collected prospectively during the patient's admission on the variables listed in Table 1. Definitions and data collection methods are available from www.nwheartaudit.nhs.uk. Urgency of operation was classified in accord with the Society of Cardiothoracic Surgeons of Great Britain and Ireland definitions [1]. Emergency or salvage cases that required preoperative IABP, ventilation or inotrope support, or were in cardiogenic shock prior to cardiac surgery were classified as in catastrophic state. Postoperative morbidity data was also collected for stroke, myocardial infarction, renal failure and deep sternal wound infections. Definitions for these complications have been previously published [8].
2.2. Patient follow-up Patient records were linked to the National Strategic Tracing Service, which records all-cause mortality in the United Kingdom. 2.3. Statistical methods Continuous variables are shown as median with 25th and 75th percentiles and categorical variables are shown as a percentage with 95% confidence intervals (CI). Comparisons were made with Wilcoxon rank sum tests and tests as appropriate. Deaths occurring as a function of time were described using the product limit methodology of Kaplan and Meier [9]. Expected survival for members of the general population matched for age and sex were derived from United Kingdom Life Tables using parametric modelling of the expected survival distribution [10]. Cox proportional hazards analysis was used to identify preoperative risk factors for mortality with hazard ratios (HR) and to calculate adjusted Kaplan–Meier survival curves [11]. To assess whether elderly patients receiving isolated CABG between April 2000 and March 2002 would benefit from non-surgical techniques, we compared this group with elderly patients receiving percutaneous coronary interventions (PCI) from the same time period at our institution. Survival curves for both CABG and PCI were compared after risk adjusting for variables identified as risk factors for 1-year mortality in this cohort, using Cox proportional hazards analysis [11]. In all cases, a P value <0.05 was considered significant. All statistical analysis was performed retrospectively with SAS for Windows version 8.2.
The number of elderly patients undergoing cardiac surgery at our institution has increased from 9.5% in the financial year 1997/1998 to 15.5% in 2001/2002 (P value <0.001). A comparison group of elderly patients undergoing PCI revealed that the proportion of elderly patients receiving non-surgical interventions is also increasing (6.9% in 2000/2001 to 9.4% in 2001/2002;  ). The median age at operation in this study cohort was 77.5 years (25th and 75th percentiles: 76.2–79.6 years); the maximum age was 92.6 years. Patient characteristics are shown in Table 1. Prior to hospital discharge, the percentage of elderly patients who had a postoperative stroke, myocardial infarction, renal failure requiring dialysis support, or a deep sternal wound infection were 3.2, 3.1, 3.8, and 1.2%, respectively. One hundred and sixty-two (19.3%) deaths occurred during the study period, with a total follow-up of 1866 patient years (mean 2.2 years, SD 1.5 years). Observed freedom from death in the elderly patients compared with that expected for the United Kingdom general population matched for age and sex is shown in Fig. 1.
The univariate predictors of mortality are shown in Table 1. A multivariate analysis using the Cox proportional hazards regression model was used to identify independent preoperative risk factors for increased mortality. These independent predictors with their respective risk ratios and 95% CI are shown in Table 2. Renal dysfunction, valves(s) surgery with or without CABG, cerebrovascular disease, and catastrophic state were all predictors for increased mortality. Fig. 2 shows the adjusted survival rates for elderly patients with or without renal dysfunction.
We went on to assess whether elderly patients receiving isolated CABG would benefit from a less invasive procedure such as PCI. We found that elderly patients undergoing CABG  or PCI  over a 2-year period at our institution had no significant difference in survival curves, after taking into account risk factors for mortality which included renal dysfunction, cerebrovascular disease, and emergency status (adjusted HR 1.8; 95% CI 0.7–3.7). Adjusted survival at 12 months for CABG was 92.4% compared to 95.6% for PCI 
Cardiac surgery has become a more common occurrence in the elderly population. During the 5-year period of this study (1997–2002), the proportion of elderly patients undergoing cardiac surgery increased by 63% at our institution. Similar increases have been seen across the United Kingdom [1]. Our results also highlight that the percentage of elderly cardiac patients receiving non-surgical interventions, through PCI, is also increasing. There has been extensive work done regarding the safety and efficacy of cardiac surgery in elderly patients. Peterson and colleagues, in a study of 24,461 CABG patients aged 80 years or older, concluded that elderly patients faced high surgical risks and expensive hospital costs [6]. A further report involving eight databases (NCN, STS, CCP, Veterans Administration, Emory, MidAmerica Heart, Northern New England, and New York State) with 161,776 patients aged 75 years or older concluded that although older age was associated with increasing risk of morbidity and mortality following CABG, these risks were modest and improving over time [7]. Other reports have also shown that surgery can be performed in the elderly population with increased but acceptable mortality and morbidity and these patients do benefit in terms of better quality of life [4,5]. Akins and co-workers found that 87% of patients questioned believed that having the operation after attaining age 80 years had been a good choice, and interestingly 87% said that their quality of life was as good or better than prior to surgery [5]. The survival of elderly patients following cardiac surgery is also comparable to the general population when matched for age and sex [4–6]. Although there tends to be a survival disadvantage in the first year after surgery, the 3-year mortality rate in elderly patients is nearly equal to that of the age- and sex-matched general population in our study (20.6 versus 19%). A multivariate Cox proportional hazards analysis revealed that preoperative renal dysfunction, cerebrovascular disease, valve surgery, and patients in a catastrophic state were independent predictors of increased mortality in elderly patients. Peterson and colleagues identified 10 independent predictors of 3-year mortality after CABG in octogenarians [6]. These included increasing age, female sex, non-white, acute myocardial infarction, congestive cardiac failure, cerebrovascular disease, peripheral vascular disease, respiratory disease, diabetes, and chronic renal disease. Preoperative renal dysfunction has long been recognised as a predictor of immediate adverse outcomes following cardiac surgery [12]. Further work by the Northern New England Cardiovascular Disease Study Group has shown that renal failure requiring dialysis support remains a highly significant predictor of decreased long-term survival in CABG patients [13]. We have shown that non-dialysis renal dysfunction has a mid-term impact on survival (only one patient in our study had preoperative dialysis support). Elderly patients with non-dialysis renal dysfunction had a 60% chance of death during a 5-year follow-up period compared to 25% without any history of renal dysfunction. This poorer prognosis could be due to the renal disease progressing to end stage renal failure or due to curtailment of benefits of surgery because of graft or prosthesis failure. Both Peterson [6] and Akins [5] identified renal insufficiency as a predictor of late death. Therefore, patients with renal impairment need careful evaluation regarding their mid- to long-term prospects. Also, strategies that offer better renal protection during surgery need to be carefully applied during intervention. Patients who have a history of cerebrovascular accidents are known to have increased risks for further strokes. Increasing age has also been shown to be a risk factor for stroke [14]. Not surprisingly, therefore, elderly patients with preoperative history of cerebrovascular events are at increased risk of peri-operative stroke and poor mid-term survival. This study shows that elderly patients with cerebrovascular disease have nearly a two-fold increased risk of mortality. Peterson and colleagues [6] showed a similar relationship in octogenarians with cerebrovascular disease having a 1.7 times increased risk. Akins and co-workers [5] went one step further by identifying postoperative strokes as a predictor of late death. Valve surgery and preoperative catastrophic state are known variables dictating poor immediate outcome following cardiac surgery. We have now shown that this impact is also carried on to the longevity of these patients. Patients with valve disease have been shown to suffer more postoperative complications than patients with isolated coronary disease [15]. This poorer outcome could be linked to immediate complications like stroke or renal failures that these patients sustained during the initial procedure and that this may have a mid- to long-term implication. A preoperative catastrophic state would also contribute to an increased rate of postoperative complications in this group of patients with implications on decreasing survival. Although we know that patients with preoperative renal impairment, history of cerebrovascular disease, undergoing valve surgery and in a catastrophic state do have poor mid-term survival we have not actively elucidated the mechanisms of this poor outcome including cause of death. This by itself could form a prospective study and help us to formulate management protocols for these high-risk patients. This will aid us in counselling patients when they attend for preoperative clinics and select candidates for more ardent follow-up in a postoperative cardiac surgical or cardiology clinic. The implication of our study is not to judge if surgery could be offered to these patients but to identify the patients at risk and quantify the effect. Our data suggests that after adjusting for risk factors (renal dysfunction, cerebrovascular disease and emergency intervention), elderly cardiac patients who receive PCI instead of CABG have a similar 1-year survival curve. Our data represents a recent and consecutive series of elderly patients undergoing cardiac surgery and therefore, represents contemporary surgical practice. The obvious limitation of this study is its retrospective and non-randomised nature. It is also limited by the fact that the cause of death of these patients is not known although it is reasonable to assume that it is more likely to be cardiac disease.
In conclusion, patients aged over 75 years who undergo cardiac surgery generally do have a mid-term mortality rate comparable to (a) the age- and sex-matched general population, and (b) matched patients receiving PCI. Patients with preoperative renal impairment have a poor prognosis in terms of mid-term survival. Other risk factors include history of cerebrovascular disease, preoperative catastrophic state and valve surgery. Defining specific risk factors will allow a more accurate assessment of preoperative risk and optimise the informed consent process. Strict follow-up measures may aid in improving outcomes in this high-risk elderly population.
We would like to acknowledge the co-operation given to us by all the Consultant Cardiac Surgeons and Cardiologists at the Cardiothoracic Centre Liverpool. We would also like to thank our Audit Officers, who ensure completeness of data collected in our registries. doi:10.1016/j.icvts.2004.01.002
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
