Left Internal Mammary Artery to the Left Anterior Descending Artery: Effect on Morbidity and Mortality and Reasons for Nonusage

Left Internal Mammary Artery to the Left Anterior Descending Artery: Effect on Morbidity and Mortality and Reasons for Nonusage Shishir Karthik, FRCS, Arun K. Srinivasan, FRCS, Antony D. Grayson, BS, Mark Jackson, PhD, and Neeraj K. Mediratta, FRCS Departments of Cardiothoracic Surgery and Research and Development, The Cardiothoracic Centre-Liverpool, Liverpool, United Kingdom Background. The left internal mammary artery to the left anterior descending artery is recognized as the gold standard for revascularization. Several studies have shown the benefits of the left internal mammary artery. However, a substantial portion of patients undergoing coronary artery bypass grafting does not receive this conduit. We sought to identify reasons for nonusage of the left internal mammary artery to the left anterior descending artery, while evaluating short- and mediumterm outcomes. Methods. Between April 1997 and September 2001 a total of 4406 consecutive patients underwent coronary artery bypass grafting with revascularization to the left anterior descending artery. All data were collected prospectively except reasons for nonusage of the left internal mammary artery. Logistic regression and Cox proportional hazards analyses were used to adjust in-hospital and medium-term outcomes, respectively. Selection bias was controlled by constructing a propensity score. A case note review was carried out to determine reasons for nonusage. Results. A total of 4047 patients (91.8%) received the left internal mammary artery to the left anterior descending artery leaving 359 patients (8.2%) who did not. We found no difference in hospital mortality or morbidity, however, patients receiving the left internal mammary artery had a survival advantage at 4 years. Reasons for not using the left internal mammary artery were damage to the conduit (n 44), poor flow (n 43), poor lung function (n 45), unstable symptoms (n 24), vascular problems (n 12), and 19 patients with other reasons (previous radiotherapy, chest wall deformity, and obese patients). Case notes had no reason stated for 146 patients and reasons for 26 patients were unobtainable. Conclusions. Left internal mammary artery usage is not associated with any increase in hospital mortality and morbidity, but these patients have better medium-term survival. Injury on harvesting, poor flow, and poor lung function were the most common reasons for not using this conduit. (Ann Thorac Surg 2004;78:142 8) 2004 by The Society of Thoracic Surgeons The left internal mammary artery (LIMA) has been used as a conduit for coronary artery revascularization since 1968 [1]. However, its use has become regular and frequent in the last two decades as its long-term benefits have become apparent [2 4]. Patients who receive LIMA experience better long-term survival, have fewer late myocardial infarctions, and sustain a greater cumulative event-free survival. Several studies have also shown an improved long-term patency. In the past, some doubts have been raised about the in-hospital outcomes and morbidity in patients who received LIMA compared with those who have received vein grafts. However, there have been several reports illustrating that the early outcomes are as good as, if not better, in patients who received LIMA [5 7]. In view of the short- and long-term benefits revealed in various studies [2 7], it would be appropriate to consider Accepted for publication Feb 6, 2004. Address reprint requests to Dr Karthik, Yorkshire Heart Centre, Leeds General Infirmary, Great George St, Leeds LS17 8AP, UK; e-mail: suchkats@yahoo.com. LIMA to be the gold standard conduit of choice for surgical revascularization of the left anterior descending artery (LAD) [8]. Whereas the worldwide trend in the last two decades has been a steady increase in the usage of LIMA, not much is documented about the reasons for not using LIMA. Hence, we have designed a study to look at the experiences at our center with regard to LIMA usage. Although we looked at the patient characteristics and early and midterm outcomes of those having LIMA versus those in whom it was not used, the main focus of this project was to look into the case notes of those patients in whom the LAD was grafted, but LIMA was not used, to determine reasons for nonusage of LIMA. Material and Methods Patient Population and Data A total of 4587 consecutive patients underwent first-time isolated coronary artery bypass graft (CABG) surgery between April 1, 1997 and September 30, 2001 at the Cardiothoracic Centre-Liverpool. One-hundred eighty- 2004 by The Society of Thoracic Surgeons 0003-4975/04/$30.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2004.02.025

Ann Thorac Surg KARTHIK ET AL 2004;78:142 8 USE OF THE LIMA TO LAD Abbreviations and Acronyms BMI body mass index CABG coronary artery bypass grafting CCS Canadian Cardiac Society CI confidence intervals COAD chronic obstructive airways disease IABP intra aortic balloon pump ICU intensive care unit LAD left anterior descending artery LIMA left internal mammary artery NSTS National Strategic Tracing Service NYHA New York Heart Association OR odds ratio SWI sternal wound infection one (3.9%) patients did not receive revascularization to the LAD and were excluded from this analysis. This left 4406 patients, who received revascularization to the LAD, as our study group. Data collection methods and definitions have been published previously [9]. Data collected prospectively included variables listed in Table 1 (preoperative) and Table 2 (in-hospital outcomes). A retrospective case note review was carried out to identify reasons for not using LIMA to LAD. 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. Postoperative stroke was defined as a new focal neurologic deficit and/or comatose states occurring postoperatively that persisted for more than 24 hours after its onset. We excluded confused states, transient cerebral events, and intellectual impairment from our study to avoid any subjective bias. Postoperative myocardial infarction was defined as a new Q-wave postoperatively in two or more contiguous leads on an electrocardiogram or notable rise in postoperative creatine-kinase with hemodynamic and echocardiographic signs of myocardial infarction. Postoperative bleeding was defined as bleeding that required surgical reexploration after initial departure from the operating theater. Renal failure was defined as patients with a postoperative creatinine level greater than 200 mol/l or patients requiring dialysis. Criteria for diagnosing superficial and deep sternal wound infections (SWI) were in accordance with the guidelines published by the Centers for Disease Control and Prevention [10]. Patient Follow-Up Patient records were linked to the National Strategic Tracing Service (NSTS), which records all causes of mortality in the United Kingdom to establish current vital statistics (as of July 31, 2002). Patients were matched to the NSTS based on patient name, National Health Service number, date of birth, gender, and postcode. 143 Statistical Analysis Categorical variables are shown as a percentage with 95% confidence intervals (CI). Comparisons were made with 2 tests as appropriate. Standard statistical tests were used to calculate odds ratios (OR) and 95% CI. Parsonnet scores were calculated, modified to a regional standard [9] to assess differences in patient case mix, and are shown as a median with 25th and 75th percentiles. Logistic regression was used to examine the impact of LIMA usage on in-hospital mortality and morbidity, while adjusting for differences in patient characteristics (treatment selection bias) [11]. Deaths occurring as a function of time were described actuarially using the product-limit methodology of Kaplan Meier [12]. To control for treatment selection bias, we used Cox proportional hazards analysis to calculate adjusted relative risk and to risk adjust the Kaplan Meier survival curve [13]. Treatment selection bias was controlled for by constructing a propensity score [14]. The propensity score was the probability that a patient would receive LIMA and was constructed from the variables listed in Table 1 (C statistic 0.75). Once the propensity score is constructed for each patient, there are three ways of using the score for comparisons: matching, stratification, and multivariable adjustment. Because of the small sample size available to us for this study, we have decided to use multivariable adjustment as matching would have reduced the study size even further and stratification can be difficult to interpret. The propensity score is then included along with the comparison variable (LIMA vs no LIMA) in multivariable analyses of outcome producing adjusted OR as shown in Table 2. The propensity score adjusts for the treatment selection bias, which is evident in Table 1, between the two groups [14]. Inall cases a p value less than 0.05 was considered significant. All statistical analysis was performed retrospectively with SAS for Windows Version 8.2 (SAS Institute, Inc., Chicago, IL). Results Left Internal Mammary Artery Use Of the 4406 patients who underwent revascularization of the LAD, 4047 (91.8%) received LIMA. The yearly usage of LIMA has increased from 90.9% in the financial year 1997/1998 to 95.7% in 2000/2001 (p value for trend 0.001). Table 1 lists study characteristics and the rate of LIMA use by patient subgroup. Clinical factors associated with low LIMA use included older age, females, smaller body mass index (BMI), increased severity of angina and dyspnea, poor left ventricular ejection fraction, emergency surgery, on-pump surgery, and patients with comorbid disease (history of diabetes, congestive cardiac failure, respiratory disease, cerebrovascular disease, and renal dysfunction). The median Parsonnet score for the LIMA group was 2.1 (25th and 75th percentiles: 1.0 4.2) compared with 4.2 (25th and 75th percentiles: 2.1 7.3) in the non-lima group (p 0.001). CARDIOVASCULAR

144 KARTHIK ET AL Ann Thorac Surg USE OF THE LIMA TO LAD 2004;78:142 8 Table 1. Patient Characteristics and Rates of LIMA Use % Patients % LIMA Use p Value Age at operation (years) 60 31.8 96.1 60 69 43.5 92.1 70 24.7 86.0 0.001 a Gender Male 80.1 93.3 Female 19.9 85.9 0.001 Body mass index (kg/m 2 ) 25 23.9 89.1 25 30 49.2 92.9 30 35 21.8 92.8 35 5.1 90.3 0.001 a Angina CCS class IV No 68.7 93.6 Yes 31.3 88.1 0.001 NYHA class IV No 93.8 92.4 Yes 6.2 83.2 0.001 Respiratory disease No 70.2 93.2 Yes 29.8 88.7 0.001 Current smoker No 84.9 91.8 Yes 15.1 91.9 0.987 Diabetes No 83.2 92.2 Yes 16.8 90.0 0.045 Peripheral vascular disease No 87.5 92.1 Yes 12.5 90.0 0.089 Cerebrovascular disease No 91.8 92.6 Yes 8.2 83.4 0.001 Renal dysfunction No 97.9 92.1 Yes 2.1 83.0 0.002 Congestive cardiac failure No 92.1 92.6 Yes 7.9 83.1 0.001 Ejection fraction 30% No 90.8 92.5 Yes 9.2 85.4 0.001 Three vessel disease No 15.8 93.5 Yes 84.2 91.5 0.079 Left main stenosis 50% No 81.5 91.9 Yes 18.5 91.6 0.731 Emergency procedure No 97.8 92.5 Yes 2.2 61.5 0.001 Off-pump surgery No 87.7 90.9 Yes 12.3 98.7 0.001 a 2 test for trend. CCS Canadian Cardiac Society; LIMA left internal mammary artery; NYHA New York Heart Association. Table 2. Crude and Adjusted Odds Ratio for In-Hospital Outcomes (LIMA Versus No LIMA) Crude OR with 95% CI Adjusted a OR with 95% CI Mortality 0.45 (0.26 0.77) 1.21 (0.62 2.39) p 0.003 p 0.576 Stroke 0.59 (0.33 1.07) 1.42 (0.69 2.90) p 0.079 p 0.338 Myocardial infarction 0.88 (0.53 1.47) 1.32 (0.75 2.33) p 0.631 p 0.344 Reop for bleeding 1.54 (0.62 3.81) 1.79 (0.68 4.68) p 0.349 p 0.237 Renal failure 0.37 (0.22 0.61) 0.77 (0.42 1.40) p 0.001 p 0.394 Deep SWI 0.59 (0.21 1.71) 0.77 (0.24 2.49) p 0.331 p 0.661 Superficial SWI 0.52 (0.29 0.90) 0.56 (0.31 1.04) p 0.018 p 0.067 Inotrope support 0.66 (0.53 0.82) 1.06 (0.83 1.35) p 0.001 p 0.670 IABP support 0.53 (0.31 0.93) 1.21 (0.62 2.36) p 0.025 p 0.581 Mechanical ventilation 0.66 (0.44 0.99) 1.37 (0.85 2.21) 24 hrs p 0.045 p 0.197 ICU stay 3 days 0.61 (0.44 0.85) 1.12 (0.77 1.63) p 0.003 p 0.558 Postoperative stay 14 0.55 (0.39 0.76) 1.19 (0.81 1.77) days p 0.001 p 0.366 a Adjusted for the propensity score. CI confidence intervals; IABP intraaortic balloon pump; ICU intensive care unit; LIMA left internal mammary artery; OR odds ratio; SWI sternal wound infection. Reasons For Nonusage of LIMA Of the 4406 patients in our study, 5 patients received alternate arterial grafts to the LAD. In all cases, the radial artery was the conduit used. The reasons for not using the LIMA were documented in the patients medical records in 187 cases (52.1%). These included 44 patient cases of damage/injury to LIMA on harvesting, 43 patient cases with poor flow in LIMA, 45 patient cases with poor lung function, 24 patient cases with unstable symptoms, 12 patient cases with vascular complications, and 19 patient cases with other reasons (previous radiotherapy to the chest wall, chest wall deformity, and obese patients with a short sternum). No apparent reason was stated for nonusage of LIMA in 146 patient cases (40.7%), whereas 26 patient case notes were unobtainable (7.2%). In-Hospital Outcomes Crude and adjusted OR for in-hospital outcomes (LIMA vs no LIMA) are shown in Table 2. In-hospital outcomes were adjusted for the propensity score which included emergency surgery, age at operation, off-pump surgery, sex, ejection fraction, cerebrovascular disease, respiratory disease, unstable angina, congestive cardiac failure, and body mass index. Hospital mortality after CABG was 2.0% (95% CI 1.6

Ann Thorac Surg KARTHIK ET AL 2004;78:142 8 USE OF THE LIMA TO LAD Table 3. Crude In-Hospital Outcomes Based on LIMA Use LIMA Used (n 4047) LIMA not Used (n 359) p Value 145 CARDIOVASCULAR Mortality 2.0 (1.6 2.5) 4.5 (2.6 7.3) 0.003 Stroke 2.2 (1.7 2.7) 3.6 (2.0 6.3) 0.079 Myocardial infarction 4.2 (3.6 4.9) 4.7 (2.9 7.6) 0.631 Reop for bleeding 2.1 (1.7 2.6) 1.4 (0.5 3.4) 0.349 Renal failure 2.1 (1.7 2.6) 5.6 (3.5 8.6) 0.001 Deep SWI 0.7 (0.4 1.0) 1.1 (0.4 3.0) 0.331 Superficial SWI 2.2 (1.7 2.7) 4.2 (2.4 6.9) 0.018 Inotrope support 31.4 (29.9 32.8) 40.9 (35.8 46.2) 0.001 IABP support 2.3 (1.8 2.8) 4.2 (2.4 6.9) 0.025 Ventilation 24 hrs 5.5 (4.8 6.3) 8.1 (5.6 11.5) 0.045 ICU stay 3 days 8.6 (7.8 9.4) 13.4 (10.1 17.4) 0.003 Postoperative stay 14 days 7.9 (7.1 8.7) 13.4 (10.1 17.4) 0.001 ICU intensive care unit; SWI sternal wound infec- IABP intraaortic balloon pump; LIMA left internal mammary artery; tion. 2.5) in patients receiving LIMA (n 4047) and 4.5% (95% CI 2.6 7.3) in patients not receiving LIMA (n 359) (p 0.003; Table 3). After multivariate adjustment was made with the propensity score, patients receiving LIMA compared with patients not receiving LIMA were no longer associated with a lower incidence of in-hospital mortality (Table 4). By univariate analyses there was no demonstrable association between LIMA use and postoperative stroke, myocardial infarction, reexploration for bleeding, and Table 4. In-Hospital Outcomes Based on LIMA Use Adjusted for the Propensity Score LIMA Used (n 4047) LIMA not Used (n 359) p Value Mortality 2.2 (1.8 2.8) 2.0 (1.2 3.2) 0.576 Stroke 2.4 (1.9 2.9) 1.9 (1.0 3.3) 0.338 Myocardial infarction 4.3 (3.7 5.0) 3.5 (2.1 5.6) 0.344 Reop for bleeding 2.2 (1.7 2.7) 1.3 (0.5 3.2) 0.237 Renal failure 2.3 (1.9 2.9) 2.8 (1.7 4.3) 0.394 Deep SWI 0.7 (0.5 1.0) 0.8 (0.3 2.3) 0.661 Superficial SWI 2.3 (1.9 2.9) 3.6 (2.2 5.7) 0.067 Inotrope support 32.2 (30.7 33.7) 31.4 (27.5 35.4) 0.670 IABP support 2.4 (2.0 3.0) 2.2 (1.3 3.6) 0.581 Ventilation 24 hrs 5.8 (5.1 6.7) 4.8 (3.3 6.8) 0.197 ICU stay 3 days 9.1 (8.3 10.0) 8.4 (6.5 10.5) 0.558 Postoperative stay 14 days 8.4 (7.6 9.4) 7.6 (5.7 9.8) 0.366 ICU intensive care unit; SWI sternal wound infec- IABP intraaortic balloon pump; LIMA left internal mammary artery; tion. Fig 1. Observed survival after coronary artery bypass surgery. deep SWI. LIMA patients, however, had lower rates of renal failure, superficial SWI, need for inotrope and intraaortic balloon pump (IABP) support, and a shorter duration of mechanical ventilation, intensive care unit (ICU) length of stay, and total postoperative length of hospital stay (Table 3). After multivariate adjustment with the propensity score, there was no association between LIMA use and postoperative stroke, myocardial infarction, reexploration for bleeding, renal failure, deep and superficial SWI, need for inotrope and IABP support, duration of mechanical ventilation, ICU length of stay and postoperative length of stay (Table 4). Follow-Up Mortality Three-hundred thirty-nine deaths (7.4% [95% CI 6.7 8.2]) occurred during the study period, with a mean follow-up of 2.9 years (standard deviation 1.4 years). The crude relative risk of long-term mortality for LIMA patients was 0.47 (p 0.001; Fig 1). Freedom from death in LIMA patients at 30 days, 1 year, 2 years, 3 years, and 4 years was 98.1%, 96.3%, 94.8%, 93.4%, and 92.0%, respectively, compared with 95.2%, 91.1%, 88.9%, 86.9%, and 83.4% for the non-lima patients. After multivariate adjustment for the propensity score, the adjusted relative risk of long-term mortality for LIMA patients was 0.66 (p 0.005). The adjusted Kaplan Meier survival curves are shown in Figure 2. The adjusted freedom from death in the LIMA patients at 30 days, 1 year, 2 years, 3 years, and 4 years was 98.1%, 96.2%, 94.8%, 93.5%, and 92.0%, respectively, compared with 97.1%, 94.3%, 92.3%, 90.3%, and 88.2% for the non-lima patients. Comment Use of LIMA during surgical revascularization can vary between different hospitals and different surgeons. Surgeon specific rates of LIMA use have been shown to be as low as 61% and as high as 97% across the four cardiothoracic centers in the northwest of England (see www.nw-

146 KARTHIK ET AL Ann Thorac Surg USE OF THE LIMA TO LAD 2004;78:142 8 Fig 2. Adjusted survival after coronary artery bypass surgery adjusted for the propensity score. heartaudit.nhs.uk). These variations can be explained partly by variations in case mix and also by differences in surgical practices. This study aimed to quantify reasons for nonusage of LIMA and to examine patient characteristics and outcomes. An analysis of our database revealed several variations in patient characteristics in the LIMA and the non-lima groups. Women were less likely to receive LIMA compared with men. This was illustrated in several previous studies [15 17]. Elderly patients were also less likely to receive LIMA to LAD compared with younger patients. The longer patency of LIMA and its attendant benefits are probably less likely to benefit older patients. Obese patients (BMI 35 kg/m 2 ) were less likely to receive LIMA than nonobese individuals. However, patients with a BMI less than 25 kg/m 2 were also less likely to receive LIMA to the LAD. A previous study has shown a similar relationship with body surface area [16]. Although the reason for this is unclear, a possible reason could be the smaller sizes of LIMA in this subgroup. The rates for LIMA usage were 61.5% in patient cases requiring emergency revascularization compared with 92.5% in the nonemergent group. The case note review revealed this to be a fairly common reason for nonusage of LIMA (0.54% of the total study group). The lower rate of LIMA usage in this group of patients is likely to be due to hemodynamic instability or a greater bleeding risk as these patients are often on antithrombotic and anticoagulant therapy. The perceived prognostic benefits of LIMA may not be as obvious in this group. Upon further analysis of the preoperative variables by multivariate logistic regression, we found elderly patients ( 70years), patients with respiratory diseases, females, diabetics, patients with ejection fraction (EF) less than 30%, and patients undergoing emergency CABG were less likely to receive LIMA to LAD. In the last two decades, LIMA usage has improved worldwide. Several patients, who would have been denied LIMA in the past because of the perceived early risks associated with LIMA usage, are now receiving it more often. There have been several observational studies in the last 10 years demonstrating not only the long-term benefits of LIMA usage, but also illustrating the protective effect of LIMA in the early postoperative period [5, 7]. In our study, there was a considerable difference in the crude mortality rates between the two groups with the patients receiving LIMA doing better. However, after adjusting for propensity score, this favorable effect was no longer evident. This also seems to be the case with the various measures of postoperative morbidity. Although the overall trend is toward a better outcome in the LIMA group, none of the differences were statistically significant after adjusting for the various factors enumerated in Table 1. The reason for this becomes apparent on analyzing the modified Parsonnet scores (0.51 of the calculated Parsonnet score) [9]. Although the median score of the LIMA group was 2.1, that of the non-lima group was 4.2. This indicates that the patients not receiving LIMA were higher risk patients. Hence, on accounting for these differences, there is no marked advantage with regard to in-hospital outcomes. What is quite apparent is that there is certainly no disadvantage in terms of early outcomes in the LIMA group. Several studies have shown some early benefits in patients with LIMA usage; however, no sound mechanism responsible for this advantage has been proposed. Among the reasons advanced for this early benefit are the early higher patency rates of LIMA. This however is debatable. Although Grondin and associates [18] have shown a 1-month occlusion rate of 10.9% for saphenous vein grafts versus 2.6% for IMA grafts, Sethi and associates [19] did not find any difference in the 7 10-day or 1-year patency rates of these two conduits. Dabal and associates [20] have recently shown that LIMA use is associated with better short-term outcomes in all patients including lower mortality rates, lower incidence of ventricular arrhythmias, shorter duration of ventilation and transfusion requirements, and shorter length of stay. They were able to demonstrate these benefits in certain high-risk groups such as women, diabetics, and elderly patients. Another possible reason for the improved operative survival in the LIMA group, especially when the other proximal anastomosis is performed with a side-biting clamp, is the perfusion of this vital territory by LIMA as soon as the cross-clamp is removed [5]. However, this remains speculative and is essentially unproven. Although the early benefits arising out of LIMA usage may be debatable, there have been several studies in the literature showing the long-term survival benefits of LIMA [3, 4]. In our study, we have been able to show a definite survival advantage for patients in whom LIMA was used and this advantage seems to be increasing with the passage of time. At 4 years, the adjusted freedom from death for the LIMA patients was 92%, whereas for patients without LIMA usage it was 88.2%. This is in accordance with most of the studies published. To identify practices with regard to LIMA usage, we carried out a case note review of the patients who did not

Ann Thorac Surg KARTHIK ET AL 2004;78:142 8 USE OF THE LIMA TO LAD have LIMA used. In 1% of the study population, the cause was injury to LIMA while harvesting. In another 1%, the LIMA was discarded after harvesting as it was deemed to have poor flow. In another 1% of the patient cases, LIMA was not harvested either because of poor lung function, chronic obstructive airways disease (COAD), or emphysematous lungs. Patients who were too unstable before surgery and those who needed inotropic agents/iabp on induction of anesthesia or had cardiac arrest before sternotomy were the other major group of patients not receiving LIMA (0.54%). Arteriopaths and patients with extensive peripheral vascular diseases and arch vessel diseases were another subset of patients who did not receive LIMA to LAD. Among the other reasons for not using the LIMA were previous radiotherapy to the chest wall, chest wall deformity, and obese patients with a short sternum. We acknowledge that often reasons mentioned in patient case notes represent the judgment of an individual surgeon to a given clinical situation. This study has definitely made us aware of a variety of reasons that may prevent LIMA usage in specific cases. One of the positive outcomes of this study has been a change in our data collection so that we can now collect reasons for not using the LIMA in individual cases. We anticipate that this will provide us with valuable information in the future which might help in regulating our practice further. A recent study of the Society of Thoracic Surgeons database reveals a LIMA usage of 81% across all coronary revascularization procedures in the United States [21]. We believe that a better awareness of reasons for nonusage of LIMA will probably lead to a gradual increase in this figure. We accept that LIMA usage will never be 100%; our aim during CABG should be to reach as close to that figure as possible with valid reasons for not using it in selected cases. As results of CABG are improving, surgeons are taking up more challenging patient cases and operating on high-risk patients. At the same time, the results of individual surgeons and institutes are coming under greater scrutiny. As most units now conduct regular audits, it is inevitable that LIMA usage will be perceived as a measure of quality. There are some limitations which may affect the conclusions drawn from our study. These include the retrospective observational design of the study. We have used propensity score adjustment to account for this, however this is no substitute for a properly designed randomized controlled trial. On the other hand, retrospective comparisons with propensity score adjustment are more versatile and may be more widely acceptable than randomized control trials [14]. A further limitation is that we have only assessed all-cause mortality rates, which means that cardiac related deaths are potentially overpredicted; however, we do not expect this to considerably affect our conclusions as the chances of a noncardiac related death are equal in both groups. This study also does not assess other long-term outcomes for these patients (eg, graft patency, reintervention rates, and quality of life). Such outcomes will be of interest as our experience grows. A final limitation of the study is the fact that we were only able to obtain a reason for nonusage in 52.1% of patient cases where the LIMA was not used. A more complete picture will become available to us in the next year or two with prospectively collected reasons for nonusage of LIMA. In conclusion, there has been a steady increase in LIMA to LAD grafting at our center during the study period. LIMA was not associated with in-hospital mortality and morbidity; however, there is a notable mortality advantage in LIMA patients over a period of 4 years after surgery. This study looks at reasons for nonusage of LIMA. The common reasons for not using the LIMA are damage during harvesting, poor flow in the LIMA after harvesting, and patients with poor lung functions/coad each of these being seen in at least 1% of the patient population. This study has also helped us identify patient groups in whom LIMA to LAD usage could be increased further. These groups include females, the obese, patients older than 70 years, and those undergoing emergency revascularization. We believe that maximizing LIMA to LAD usage is important, as it is quite likely to become a quality improvement marker for coronary artery bypass surgery. We would like to acknowledge the cooperation given to us by all of the Consultant Cardiac Surgeons at the Cardiothoracic Centre-Liverpool: John A. C. Chalmers, Walid C. Dihmis, Brian M. Fabri, Elaine M. Griffiths, Neeraj K. Mediratta, Richard D. Page, D. Mark Pullan, Abbas Rashid, and W. Ian Weir. We would also like to thank Janet Deane, who is responsible for maintaining the quality and ensuring the completeness of the data collected in our Cardiac Surgery Registry. References 147 1. Green GE, Strertzer SH, Reppert EH. Coronary arterial bypass grafts. Ann Thorac Surg 1968;5:443 8. 2. Lytle BW, Loop FD, Cosgrove FD, et al. Long-term (5 12 years) serial studies of internal mammary artery, and saphenous vein coronary bypass grafts. J Thorac Cardiovasc Surg 1985;89:248 58. 3. Loop FD, Lytle BW, Cosgrove DM, et al. Influence of internal mammary artery graft on 10-year survival and other cardiac events. N Engl J Med 1986;314:1 6. 4. Cameron AA, Green GE, Brogno DA, et al. Internal thoracic artery grafts: 20-year clinical follow-up. J Am Coll Cardiol 1995;25:188 92. 5. Grover FL, Johnson RR, Marshall G, et al. Impact of mammary grafts on coronary bypass operative mortality and morbidity. Ann Thorac Surg 1994;57:559 69. 6. Edwards FH, Clark RE, Schwartz M. Impact of internal mammary artery conduits on operative mortality in coronary revascularization. Ann Thorac Surg 1994;57:27 32. 7. Leavitt BJ, O Connor GT, Olmstead EM, et al. Use of internal mammary artery graft and in-hospital mortality and other adverse outcomes associated with coronary artery bypass surgery. Circulation 2001;103:507 12. 8. Cohn LH. Use of Internal Mammary Artery graft and inhospital mortality and other adverse outcomes associated with coronary artery bypass surgery. Circulation 2001;103: 483 4. 9. Wynne-Jones K, Jackson M, Grotte G, et al. On behalf of the north west regional cardiac surgery audit steering group. CARDIOVASCULAR

148 KARTHIK ET AL Ann Thorac Surg USE OF THE LIMA TO LAD 2004;78:142 8 Limitations of the Parsonnet score for measuring risk stratified mortality in the north west of England. Heart 2000;84: 71 8. 10. Mangram AJ, Horan TC, Pearson ML, et al. The Hospital Infection Control Practices Advisory Committee. Guideline for prevention of surgical site infection, 1999. Infect Control Hosp Epidemiol 1999;20:247 78. 11. Hosmer D, Lemeshow S. Applied logistic regression. New York: Wiley, 1989. 12. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:547 81. 13. Cox DR. Regression models and life-tables. J R Stat Soc 1972;34:187 220. 14. Blackstone E. Comparing apples and oranges. J Thorac Cardiovasc Surg 2002;123:8 15. 15. Cameron A, Davis KB, Green GE, et al. Clinical implications of internal mammary bypass grafts: the coronary artery surgery study experience. Circulation 1988;77:815 9. 16. Leavitt BJ, Olmstead EM, Plume SK, et al. Use of the Internal Mammary Artery Graft in Northern New England. Circulation 1997;96(Suppl II):32 7. 17. Jones JW, Oschner JL, Mills NL, et al. The internal mammary bypass graft: a superior second coronary artery. J Thorac Cardiovasc Surg 1978;75:625 31. 18. Grondin CM, Campeau L, Lesperance J, et al. Comparison of late changes in internal mammary artery and saphenous vein grafts in two consecutive series of patients 10 years after operation. Circulation 1984;70:I-208 12. 19. Sethi GK, Copeland JG, Moritz T, et al. Comparison of postoperative complications between saphenous vein and IMA grafts to left anterior descending coronary. Ann Thorac Surg 1991;51:733 8. 20. Dabal RJ, Gross JR, Maynard C, et al. The effect of left internal mammary artery utilization on short-term outcomes after coronary revascularization. Ann Thorac Surg 2003;76: 464 70. 21. Cleveland JC Jr, Shroyer ALW, Chen AY, et al. Off-pump coronary artery bypass grafting decreases risk-adjusted mortality, and morbidity. Ann Thorac Surg 2001;72:1282 9. Notice From the American Board of Thoracic Surgery The 2004 Part I (written) examination will be held on Monday, December 6, 2004. It is planned that the examination will be given at multiple sites throughout the United States using an electronic format. The closing date for registration is August 1, 2004. Those wishing to be considered for examination must apply online at www.abts.org. To be admissible to the Part II (oral) examination, a candidate must have successfully completed the Part I (written) examination. A candidate applying for admission to the certifying examination must fulfill all the requirements of the Board in force at the time the application is received. Please address all communications to the American Board of Thoracic Surgery, 633 N St. Clair St, Suite 2320, Chicago, IL 60611; telephone: (312) 202-5900; fax: (312) 202-5960; e-mail: info@abts.org. 2004 by The Society of Thoracic Surgeons Ann Thorac Surg 2004;78:148 0003-4975/04/$30.00 Published by Elsevier Inc