| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2005;4:184-188. doi:10.1510/icvts.2004.102707 © 2005 European Association of Cardio-Thoracic Surgery
Left internal mammary artery use in patients with poor left ventricular ejection fraction: a propensity-matched analysis of mid-term survival
1 Departments of Cardiothoracic Surgery, The Cardiothoracic Centre Liverpool, Thomas Drive, Liverpool, L14 3PE, UK Received 16 November 2004; received in revised form 1 February 2005; accepted 1 March 2005
*Corresponding author. Tel.: +44 151 293 2336; fax: +44 151 288 2371.
We aimed to determine whether the use of left internal mammary artery (LIMA) to the left anterior descending (LAD) artery during coronary artery bypass grafting (CABG) confers an improved survival benefit to patients with an impaired preoperative left ventricular ejection fraction (LVEF). Between April 1997 and March 2004, 7198 consecutive patients underwent first time CABG to the LAD. There were 627 patients who had an LVEF <30% and of these, 548 patients (87.4%) received a LIMA graft, while 79 patients (12.6%) did not. A propensity-matched analysis was performed to provide matched cohorts for analysis of deaths occurring over time, which were described using Kaplan–Meier techniques. Propensity-matching produced two cohorts of 77 patients with or without the use of LIMA. Patient characteristics were reasonably matched between the groups. Forty-six (29.9%) deaths occurred in the propensity-matched groups. Freedom from death in patients with LIMA used at 4-years was 77.1%, compared with 60.7% for the patients with no LIMA used (P=0.026). The use of the LIMA as a bypass conduit is not contraindicated in patients with a poor preoperative LVEF. The usage of LIMA markedly improves survival.
Key Words: LIMA; LAD; Poor ejection fraction; Coronary artery bypass surgery; Mid-term survival
Ischaemic heart disease is the commonest cause of heart failure in the developed world. Patients with poor left ventricular function caused by ischaemic heart disease are known to have a poor prognosis when treated medically. Several studies have shown that these patients have improved outcomes when undergoing coronary artery bypass grafting (CABG) compared to optimal medical therapy [1] or percutaneous coronary interventions [2]. CABG can be performed relatively safely in patients with a poor left ventricular ejection fraction (LVEF), with good medium- to long-term survival, improved left ventricular function, and improved quality of life [3,4]. The use of the left internal mammary artery (LIMA) has been shown to be safe as a conduit in a high proportion of patients with poor LVEF [5–7]. However, recent work by the North West Quality Improvement Programme in Cardiac Interventions has highlighted that patients with poor LVEF are still less likely to receive the LIMA compared to other subgroups undergoing CABG (Quality Improvement Programme in Cardiac Surgery. Left Internal Mammary Artery Use in Patients Receiving First Time Coronary Artery Bypass Grafting. Quality Improvement Note: May 2002; available from http://www.nwheartaudit.nhs.uk/LIMAimprovementnote1.pdf). The LIMA has proven survival benefits [8], therefore, we aimed to quantify the effect of the LIMA on mid-term survival in patients with poor LVEF undergoing first time CABG to the left anterior descending (LAD) artery at our institution.
2.1. Patient population and data We performed a retrospective study on a total of 7198 consecutive patients undergoing first time CABG to the LAD between 1st April 1997 and 31st March 2004 at the Cardiothoracic Centre-Liverpool. The circumflex or right coronary artery were also grafted in these patients when appropriate using either vein or other arterial conduits. Of the 7198 patients in the study, 6571 (91.3%) patients had a LVEF of 30% or more. These patients were removed from the study, leaving 627 (8.7%) patients for analysis who had a poor LVEF (<30%), as defined by the Society of Cardiothoracic Surgeons of Great Britain and Ireland [9]. Data were collected prospectively during the patient's admission as part of routine clinical practice on the variables listed in Tables 1, 2 and 3. Definitions and data collection methods are available from www.nwheartaudit.nhs.uk.
In-hospital mortality was defined as death within the same hospital admission regardless of cause. All patients transferred from the base hospital to another hospital were followed up to confirm their status at discharge. Re-exploration for bleeding was defined as bleeding that required surgical re-operation after initial departure from the operating theatre. 2.2. Patient follow-up Patient records were linked to the National Strategic Tracing Service (NSTS), which records all deaths in the United Kingdom. To establish current vital status, patients were matched to the NSTS based on patient name, National Health Service number, date of birth, gender, and postcode. The link to the NSTS was performed after LIMA and non-LIMA patients were matched for patient characteristics (see Statistical methods)2.3. Statistical methods A propensity score [10] of three-digits was used to match non-LIMA patients with patients whom had the LIMA used. Logistic regression was used to develop the propensity score which was constructed from core characteristics. These characteristics included age, sex, extent of disease, left main stem stenosis, priority, peripheral vascular disease, diabetes, renal dysfunction, respiratory disease, and body mass index as suggested by the American College of Cardiology/American Heart Association practice guidelines [11]. The C statistic for this model was 0.83. Continuous variables are shown as median with 25th and 75th percentiles and categorical variables are shown as a percentage. Comparisons were made with Wilcoxon rank sum tests and Chi-square tests as appropriate. Deaths occurring as a function of time were described using the product limit methodology of Kaplan and Meier. A Cox proportional hazards regression model was constructed to identify independent risk factors for mid-term mortality, using the forward stepwise technique. In all cases a P value <0.05 was considered significant. All statistical analysis was performed with SAS for Windows Version 8.2.
Of the 627 patients with a poor LVEF, 548 (87.4%) received the LIMA, while 79 (12.6%) patients did not. After performing the propensity-matching, we were able to successfully match 77 of the 79 patients who did not receive the LIMA. Table 1 shows patient and disease characteristics based on whether the LIMA was used in the propensity-matched groups. All characteristics were reasonably matched. The breakdown of the type of grafts used (arterial or vein) on the coronary artery vessels treated are shown in Table 2. Of the LIMA patients, 9 underwent CABG without cardiopulmonary bypass compared to 5 patients in the non-LIMA group (11.7% versus 6.5%; P=0.26). In-hospital outcomes are shown in Table 3. Forty-six (29.9%) deaths occurred in the propensity-matched patients during the study with a total follow-up period of 521.9 patient-years (mean follow-up of 3.4 years). The relative risk of mid-term mortality for LIMA patients was 0.5 (95% CI 0.3 to 0.9; P=0.026). Freedom from death in patients with LIMA used at 1, 2, 3, and 4 years was 86.9%, 85.1%, 80.9% and 77.1%, respectively, compared with 80.5%, 74.9%, 68.9% and 60.7% for patients with no LIMA used (Fig. 1).
Independent risk factors, adjusted for the propensity score, for mid-term mortality are shown in Table 4. Cerebrovascular disease, female sex, and avoidance of the LIMA were all identified as risk factors for mid-term mortality in this group of patients with poor LVEF.
With an increasingly aged population it follows that the prevalence of heart failure will increase. The Society of Thoracic Surgeons National Database confirms that the proportion of patients with a poor LVEF undergoing CABG has been increasing. The average ejection fraction has fallen from 62% in 1980 to 49% in 2000 [9]. In the United Kingdom the trend is different, the Society of Cardiothoracic Surgeons of Great Britain and Ireland report that the proportion of patients with a LVEF <30% has remained static [9], although this could be related to varying methods of estimating ejection fraction. The benefits of surgical revascularisation in patients with a poor left ventricle and hibernating myocardium are well documented and include: improved quality of life [5,6]; improved left ventricular contractility [2,4]; improved LVEF [3–6]; improved New York Heart Association (NYHA) class [5]; as well as good medium- [3,5,6] and long-term survival [1,3,5,6]. Medical management of heart failure caused by ischaemic heart disease has a poor prognosis. Transplantation probably offers the best clinical results [12] but is not a realistic treatment option for most patients because of the lack of availability. Studies since the mid-1980s have shown a survival advantage in patients with severe left ventricular dysfunction and predominantly angina who have undergone CABG. In these reports, 5–7 year survival for the surgical revascularisation group is 63–69% and for the medically treated groups 34–43% [13]. However, the accuracy of these data can be questioned, the reports are subject to bias by inclusion in the surgical group of patient with more severe angina, more extensive but operable CAD and less congestive cardiac failure, while patients with more severe congestive cardiac failure tended to be treated medically. These reports are also retrospective and non-randomised [13]. Despite the potentially large benefit to patient and community there is some reticence to operate on patients with ischaemic heart failure. Miller and colleagues [14] attribute the low rate of surgical revascularisation in patients with heart failure and a large reversible myocardial perfusion defect (revasculaisation rate of 13%) at their institute due to the absence of generalisability of the existing literature as to the optimal means of treating patients with heart failure and myocardial ischaemia. Percutaneous coronary intervention may be an alternative. Toda and colleagues studied 117 consecutive patients with severe left ventricular dysfunction who underwent either CABG or percutaneous coronary interventions. They concluded that patients receiving CABG had more complete revascularisation, improved left ventricular function, fewer cardiac events, and fewer target revascularisations compared to percutaneous coronary interventions. The mid-term survival was similar [2]. Whilst it is clear that patients with heart failure stand to benefit the most from revascularisation, they also pose a greater risk. Low ejection fraction and worse functional state independently increase the risk of CABG. Also, poor ejection fraction is a risk factor for adverse outcomes following CABG in most risk prediction tools [11], with the European System for Cardiac Operative Risk Evaluation Score allocating 3 points to patients who have an ejection fraction <30% [15]. The operative mortality is high and ranges from 4.7 to 15% in a series conducted in patients with ejection fraction lower than 20, 25, or 30% [3]. The use of the LIMA has been shown to have a statistically significant impact on improving survival in patients receiving CABG both in the short- and long-term [8]. Consequently, the use of LIMA has increased dramatically over the last two decades [9]. However, the North West Quality Improvement Programme in Cardiac Interventions has highlighted that patients with a poor ejection fraction are still less likely to receive the LIMA compared to other subgroups undergoing CABG (Quality Improvement Programme in Cardiac Surgery. Left Internal Mammary Artery Use in Patients Receiving First Time Coronary Artery Bypass Grafting. Quality Improvement Note: May 2002; available from//www.nwheartaudit.nhs.uk/LIMAimprovementnote1.pdf). This maybe partially explained by the belief that the use of the LIMA graft adds complexity to the CABG operation and consequently the risks associated with these patients. Elefteraides and colleagues have concluded, in two separate studies of patients undergoing CABG with an ejection fraction of less than 30%, that LIMA can safely be used as a conduit [5,6]. Interestingly, Jegaden and co-workers found that single internal mammary artery with sequential vein grafting was more beneficial than multiple arterial grafting in patients with severe left ventricular dysfunction (defined as less than 40%) [7]. In this recent series of CABG patients, we have shown that patients with a poor LVEF who receive the LIMA have a markedly improved survival over a 4-year follow-up period. After successfully propensity-matching (3-digit match) 77 of the 79 non-LIMA patients, the 4-year post-CABG survival rate was 77.1% with LIMA and 60.7% without (P=0.026). Use of the LIMA was also still shown to be beneficial in a Cox proportional hazards analysis looking at risk factors for mid-term mortality. Although patients who received a LIMA were noted to have a greater amount of blood loss in the intensive care unit, knowledge that the LIMA graft has a clinically significant survival benefit in patients with an impaired left ventricular, will no doubt help increase its use in this particular high-risk group. Our study has a number of limitations. Firstly, it is an observational report and therefore could be confounded by selection bias. We have attempted to minimise the risk of bias by performing a propensity-matched analysis, and ignoring the ‘mixed-fruit’ [10]. However, we may not have taken into account variables that may affect the mid-term survival for our patients when propensity-matching, but for this to have a dramatic effect on our conclusions, the variables used in the propensity-matching would have to be uncorrelated with the variables not matched for. It is also important to note that even though no significant differences existed between the propensity-matched patient characteristics, certain characteristics such as New York Heart Association class and renal dysfunction were three times as common in the non-LIMA group. Therefore, although as individual characteristics these failed to reach statistical significance, the combination of these differences could be important. To tackle this potential problem we included the propensity score of each patient in the Cox proportional hazards regression model to adjust for any underlying bias. Another limitation is that we only looked at mid-term survival and did not take into account other outcomes such as the incidence of recurrent angina, graft patency and quality of life. These outcomes will be of interest as our experience grows. In conclusion, we have quantified the mid-term survival differences in patients with poor left ventricular ejection fraction undergoing coronary artery bypass surgery with and without the use of the left internal mammary artery as a bypass conduit. The LIMA is not contraindicated in patients with a poor LVEF; indeed LIMA markedly improves survival and should be the conduit of choice for bypass surgery in this high-risk group.
We would like to acknowledge the co-operation given to us by all the Consultant Cardiac Surgeons at the Cardiothoracic Centre-Liverpool: Mr JAC Chalmers, Mr WC Dihmis, Mr BM Fabri, Miss EM Griffiths, Mr N Mediratta, Mr RD Page, Mr DM Pullan, Mr A Rashid, and Mr WI Weir. We would also like to thank Janet Deane, who maintains the quality and ensures completeness of data collected in our Cardiac Surgery Registry.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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