Late Outcome After Coronary Artery Bypass Graft Surgery in Patients <40 Years Old
Background Randomized trials confirm the long-term efficacy of coronary artery bypass graft surgery (CABG), although there are no randomized data in patients <40 years old. Because these patients have been reported to have an early recurrence of symptoms, the long-term postoperative outcome was examined.
Methods and Results The long-term outcome of patients (n=221) <40 years old undergoing CABG at Green Lane Hospital, New Zealand, from 1970 to 1992 was determined. The 30-day mortality rate was 1.8% for initial and 9.5% for redo CABG. The median times to angina or myocardial infarction (recurrent ischemic event), further intervention, and death were 6.0, 9.6, and 14.2 years, respectively. Factors associated with increased late mortality on univariate analysis included end-systolic volume (ESV) ≥80 mL (P=.004; 10-year mortality 19% versus 39% ESV ≥80 mL), no internal mammary conduit (P=.01), no lipid-modifying therapy (P=.005), and no postoperative aspirin use (P=.0002); the latter was also associated with increased recurrent ischemic events (P=.04) or increased reintervention (P=.02). On stepwise logistic regression analysis, factors associated with increased late mortality were increasing ESV (P=.004), no internal mammary artery conduit (P=.009), diabetes (P=.04), and no postoperative aspirin (P=.02); the latter was also associated with increased recurrent ischemic events (P=.02). Hypercholesterolemia (≥6.5 mmol/L) was present in 65% of patients at presentation and 45% at follow-up.
Conclusions To attempt to prevent recurrent ischemia or late death, patients <40 years old who require CABG should receive internal mammary conduits, aspirin, lipid-modifying therapy, therapy to inhibit ventricular dilatation, and strict diabetes management.
The surgical management of patients with severe coronary artery disease improves long-term survival, which has recently been confirmed in an overview of randomized trials of CABG.1 However, subgroup analysis according to clinical risk, which includes age, suggested that some low-risk patients may not have a significant mortality benefit from surgery. For example, when the subgroup of patients <47 years old was compared with older age groups in the European trial, there was no mortality benefit for surgery at 10 years.2 The reasons for this age-dependent difference in the relative mortality benefit of CABG are not immediately apparent but may reflect both a lower mortality in medically treated younger patients2 and surgical crossover. Also, a higher prevalence of factors associated with early graft occlusion, such as an elevated total and/or low HDL cholesterol and persistent cigarette smoking,3 4 5 has been reported in younger patients.6 7 8 9 10 The possible role of other potential factors that may be associated with an unfavorable long-term outcome, such as diabetes and hypertension, is equivocal.6 7 8 9 10 Since several studies have shown that in the year after CABG there is considerable inattention to correction of adverse risk factors by patients and/or physicians,11 12 13 suboptimal treatment of these may impact the long-term outcome after coronary surgery in younger patients. Antiplatelet therapy confers a 30% benefit in terms of graft patency at 1 year,14 15 but whether this is translated into a beneficial effect on long-term outcome in young patients after CABG is unknown.
In this study, we examined the entire surgical experience at Green Lane Hospital from 1970 to 1992 in patients who were <40 years old at the time of CABG to determine the attention paid to risk factors such as dyslipidemia, smoking status, and the use of antiplatelet therapy in this group of post-CABG patients. The impact of variables known to influence long-term outcome, such as ventricular function and the use of mammary conduits, was also assessed.
The long-term outcome was evaluated in all patients who underwent a first CABG operation at age <40 years at Green Lane Hospital from 1970 through 1992. Evaluation was performed in 1993. Patients with reoperation before age 40 years were included only once for analysis. Follow-up information was obtained by clinical and case note review as well as detailed questionnaires to physicians and patients similar to those previously described.16
The preoperative variables included sex, race, smoking history (current or stopped ≤1 year, stopped 1 to 10 years, stopped >10 years, or never smoked), serum lipids (and whether action had been taken to rectify any abnormalities), family history of premature coronary artery disease in first-degree relatives (women ≤60 and men ≤50 years old, respectively), hypertension (on therapy and/or blood pressure ≥150/95 mm Hg), diabetes (on insulin or fasting glucose ≥6.5 mmol/L), class of angina (Canadian Cardiovascular Society), previous Q-wave myocardial infarction, left ventricular end-diastolic pressure, left ventricular ESV, presence of left main coronary stenosis of ≥50% (reduction in lumen diameter), and myocardial score.17
The operative data included the number of distal grafts, the use of IMA conduits, bypass and cross-clamp times, and perioperative development of new Q waves and/or elevation of aspartate aminotransferase (≥100 U/mL).18
The long-term follow-up data included vital status, recurrence of angina or further infarction, the necessity for further cardiac intervention (CABG, angioplasty, or cardiac transplantation), follow-up lipid levels, use of lipid-modifying therapy, aspirin therapy, and postoperative smoking history (defined as admitting smoking regularly for at least 1 month in the postoperative period).
Actuarial analyses (Kaplan-Meier) were applied to assess unfavorable surgical outcomes, ie, recurrence of angina/myocardial infarction, further intervention, and death. Statistically significant differences between data sets were determined by two-tailed Student’s t tests and univariate analysis. Multivariate analysis using stepwise logistic regression and Cox proportional-hazards models examined the factors that were associated with an adverse outcome.
There were 239 CABG operations performed on patients <40 years old at Green Lane Hospital during the years 1970 through 1992, and patients were followed up for 8.3±5.1 years. Eleven patients had two or more operations at age <40 years, and 7 patients had CABG only as a secondary procedure at the time of aortic root replacement for dissection or aortic valve replacement. Therefore, the data presented here are from the 221 patients who had initial CABG as the primary procedure at age <40 years.
Preoperative Patient and Catheterization Data
The main preoperative patient characteristics are summarized in Table 1⇓. The majority of patients (148 of 218, 68%) underwent CABG for severe angina (Canadian Cardiovascular Society class III or higher). There was a high incidence of hypercholesterolemia (104 of 159 [65%] with total cholesterol ≥6.5 mmol/L [250 mg/dL]), with a mean cholesterol level of 7.21±1.65 mmol/L; current or recent smoking (61%); and a family history of premature coronary heart disease (39%); whereas the incidences of hypertension and diabetes were 17% and 5%, respectively.
The preoperative angiographic data are summarized in Table 2⇓. Twenty-six patients (12%) had ≥50% left main coronary stenosis. The mean preoperative ejection fraction on left ventriculography was 57.9±13.1%, and ESV was 79.3±47.4 mL. Sixty-four percent of patients (128/201) had a myocardial score of ≥10, ie, the approximate equivalent of triple-vessel disease17 on coronary arteriography, and 12% had ≥50% left main stenosis.
The 221 patients received a total of 518 saphenous vein graft distal anastomoses. Forty-four patients had single IMA conduits (no patients had bilateral mammary conduits). There were 4 deaths (1.8%) within 30 days of surgery. By a relatively conservative assessment of either new Q waves and/or a peak aspartate aminotransferase ≥100 IU/mL, 21 patients (9.5%) had perioperative infarcts.
Ten patients were lost to follow-up after the perioperative period. The median times until death, intervention, or a further ischemic event (either recurrent angina or myocardial infarction) were 170, 136, and 72 months, respectively (Fig 1⇓). Survival at 10 years after surgery was 74%. There were 58 further interventions (28%). Forty-two patients (20%) had further CABG, with a 30-day mortality at reoperation of 9.5%. In addition, 12 patients (5.6%) had postoperative coronary angioplasty, and 5 patients (2.4%) underwent cardiac transplantation. If the cohort was equally divided into pre–January 1, 1981 (n=107) and post–January 1, 1981 cohorts (n=104), there were no differences in the incidences of postoperative recurrent ischemia (P=.67) or late mortality (P=.98) (Fig 2⇓). After 1984, coronary angioplasty and the use of mammary conduits, aspirin, and lipid-modifying therapy became more prevalent. The cohort was therefore also divided into pre–January 1, 1985 (n=165; 11 CABGs per year) and post–January 1, 1985 (n=46; 5.8 CABGs per year); there were no differences in either late mortality (P=1.00) or recurrent ischemic events (P=.43).
On univariate analysis, several factors were associated with an increased actuarial mortality during long-term follow-up (Fig 3⇓). These included an increased ESV ≥80 mL (P=.004; the 10-year mortality was 39% versus 19% for ESV <80 mL), the lack of use of an IMA conduit (P=.01), the lack of postoperative aspirin therapy (P=.0002), and the lack of lipid-modifying therapy (P=.005). Of these, only the postoperative use of aspirin was associated with a significant reduction in further ischemic events (P=.04) or further intervention (P=.02).
Several potentially adverse factors were included in a multivariate model to predict determinants of long-term mortality (Table 3⇓). Stepwise logistic regression analysis showed that a smaller ESV (continuous variable, P=.004), the use of IMA conduits (P=.009), postoperative aspirin (P=.02), and the absence of diabetes (P=.04) were all associated with a better long-term mortality outcome. Multivariate analysis using the Cox proportional-hazards model was also performed. Increasing ESV (P=.01), the lack of aspirin postoperatively (P=.01), diabetes (P=.05), and the lack of lipid-modifying therapy (P=.03) were significant adverse factors (the lack of an IMA conduit, P=.11).
While postoperative smoking (28% of patients) showed a trend toward being an adverse factor affecting increased late mortality, hypertension did not (Table 3⇑). There were 83 patients (45%) who did not receive any postoperative aspirin. Of the 101 patients (55%) who received aspirin postoperatively, only 49 were treated preoperatively and 8 within 1 month of surgery, so the beneficial effect of aspirin may well be underestimated.
Lipid data were not always recorded, particularly in the early years of this cohort. However, on univariate analysis, those patients without baseline lipid data had the same late mortality (P=.34). The use of postoperative lipid-modifying therapy rather than cholesterol was included in the logistic regression model, but this factor did not reach statistical significance (P=.11). Inclusion of baseline cholesterol, which reduced the total number of observations, did not affect the significance of the other factors. At review, 45% of patients (59/132) remained hypercholesterolemic (>6.5 mmol/L) (Table 4⇓). There were no statistical differences in the other baseline characteristics of those patients who did and did not have follow-up lipid data, nor was there any difference between those who had complete lipid analysis versus those who had analyses of only total cholesterol and triglycerides.
This study has examined the actuarial outcome of patients <40 years old who underwent CABG at Green Lane Hospital between 1970 and 1992. During most of this era, the usual bypass conduit was the saphenous vein graft. IMA grafts were not routinely used at Green Lane Hospital before the mid-1980s, which coincided with an approximate halving of the annual surgical rate for younger patients due to the use of angioplasty as a revascularization procedure.
The 10-year postoperative mortality of 26% reported here was the same as that reported from Johns Hopkins19 for similar patients in the same era. Patients in the Johns Hopkins series had slightly more left main stenosis (15% versus 12%) and slightly better ventricular function than those reported here, with 89% having an ejection fraction ≥45% compared with 87% with an ejection fraction >40%. The number of grafts per patient was also similar (2.7 versus 2.5). However, the risk profiles of patients were somewhat different, with more patients having hypertension (35% versus 17%) and diabetes (17% versus 5%) and currently smoking (74% versus 61%) in the Johns Hopkins series; hypercholesterolemia (≥250 mg/dL, or ≥6.5 mmol/L) was similar to that reported here (64% versus 65%). In contrast to the Johns Hopkins experience, we found no increase in the percentage of women or diabetics in our post–January 1, 1981 cohort.
In a recent report of a smaller series from France, 86 patients <40 years old with less extensive disease requiring 1.9 grafts per patient had IMA grafts to the left anterior descending artery in all cases and an excellent long-term outcome, with only 7% mortality at 10 years,20 but these patients had fewer distal anastomoses compared with our series. In the European Coronary Surgery Study,2 patients <47 old years had a 10-year mortality of ≈20% following initial surgical or medical treatment, although these patients had generally less severe disease than those in our series. It is not clear whether this represents a better outcome with medical treatment or, alternatively, whether it may represent expeditious crossover to surgery. The Cleveland Clinic series6 of patients <36 years old also had slightly lower 10-year mortality, although patients in that series received more IMA conduits and had less marked dyslipoproteinemia, and fewer patients had left main stem stenosis (2.8%).
Although a medical cohort is not included in this series, there is an age-matched control group from the New Zealand population. Compared with age-matched control subjects in previous series from our institution, which reported long-term post-CABG outcome in all patients (mean age, 54 years) undergoing surgery in 1976 and 197716 and those ≥70 years old (mean age, 71 years) undergoing surgery in 1981 through 1985,21 patients in the present series did not have as good a relative late outcome. This may be due in part to an increased incidence of occult ischemic heart disease in older individuals in the general population. A recent report from Oregon of their entire surgical experience also showed that the youngest patients (<45 years old) had a poorer long-term mortality outcome than those in some older age groups.22
Although there have been improvements in myocardial protection over the past two decades, it is unlikely that perioperative technique was a major contributor to this relatively unsatisfactory late mortality outcome, since the operative mortality rate (death at <30 days) was 1.8%. There is clearly a significant risk at reoperation, with 9.5% 30-day mortality. When total bypass time and clamp time were included in the multivariate analysis, these variables were not significant predictors of an adverse late outcome.
As previously reported from this institution,23 24 we have shown that increased mortality is positively associated with poorer ventricular function but, in particular, increasing ESV. The 10-year postoperative mortality more than doubled (39% versus 19%) in those patients with a preoperative ESV of ≥80 mL. A recent report from our institution examining increased preoperative ESVs24 found that ESV >130 mL was most highly associated with increased mortality. This index of ventricular function was the strongest single predictor of a poor long-term mortality outcome on multivariate analysis.
The data show that in this group of young patients after CABG, there is a high incidence of dyslipidemia that has not always been optimally treated, with 45% of patients having cholesterol ≥6.5 mmol/L (250 mg/dL) at the time of review. Others have reported similar results in unselected patients after CABG.11 12 13 HMG-CoA reductase inhibitors can achieve a 30% to 40% reduction in total and LDL cholesterol25 26 and also increase HDL cholesterol. Lipid-modifying therapy can cause coronary artery disease regression, so that patients with coronary artery disease, especially those with saphenous vein grafts,3 25 should be treated aggressively. More than 60% of the patients in this series met eligibility criteria for enrollment in the Scandinavian Simvastatin Survival Study,27 which showed 30% and 42% reductions in total and coronary heart disease deaths, respectively.27 One would anticipate that the use of HMG-CoA reductase inhibitors in patients similar to those described here, especially for patients with cholesterol ≥5.5 mmol/L, may achieve similar benefits.
In this series, a large number of patients did not receive antiplatelet therapy. The significant beneficial effect of postoperative aspirin therapy on total mortality, the prevention of recurrent ischemic events and reintervention, emphasizes the importance of this therapy. These data suggest that the previously documented effects of aspirin on 1-year graft patency14 15 correlate with enhanced long-term survival. Some of the patients not receiving aspirin had undergone CABG before the general use of postoperative aspirin at this institution in the early 1980s, while others were no longer patients of this institution and/or had let their aspirin therapy lapse.
In this group of patients, as in other studies,6 7 8 9 10 high incidences of preoperative cigarette smoking and a family history of premature coronary heart disease have been documented. In this cohort, unfortunately, 28% of patients smoked regularly at some time after surgery. Since smoking has been reported to cause late saphenous vein graft thrombosis,4 difficulties in documenting this habit may have contributed to its showing only a trend toward being a significant adverse factor (P=.12). Hypertension was not associated with an adverse outcome (P=.95). Although there was a much smaller percentage of diabetics (5%) in this series than in the Johns Hopkins series19 (17%), this factor was associated with increased late mortality (P=.04). However, because diabetes and smoking are potentially modifiable risk factors, they should receive vigorous attention.
Our study does not address whether CABG is a better treatment than medical therapy. A trial randomizing patients without severely limiting angina or life-threatening coronary anatomy to aggressive lipid-modifying therapy either alone or in conjunction with CABG may be appropriate but, given that young patients are likely to be more insistent in requesting revascularization for symptoms, seems improbable.
To achieve a better long-term outcome in patients who have CABG at <40 years old, lipid-modifying therapy, regular postoperative aspirin, optimal treatment of diabetes, the use of IMA conduits, and the preservation of left ventricular function will need to be assiduously addressed.
Selected Abbreviations and Acronyms
|CABG||=||coronary artery bypass graft surgery|
|HMG-CoA||=||3-hydroxy-3-methylglutaryl coenzyme A|
|IMA||=||internal mammary artery|
This work was supported in part by the Green Lane Hospital Research and Educational Fund, Auckland, New Zealand. Dr Harvey White’s helpful advice is gratefully acknowledged.
Reprint requests to Dr John French, Cardiology Department, Green Lane Hospital, Epsom, Auckland 1003, New Zealand.
- Copyright © 1995 by American Heart Association
Yusuf S, Zucker D, Peduzzi P, Fisher LD, Takaro T, Kennedy JW, Davis K, Killip T, Passamani E, Norris R, Morris C, Mathur V, Varnauskas E, Chalmers TC. Effect of coronary artery bypass graft surgery on survival: overview of 10-year results from randomised trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration. Lancet. 1994;344:563-570.
Solymoss BC, Nadeau P, Millette D, Campeau L. Late thrombosis of saphenous vein coronary bypass grafts related to risk factors. Circulation. 1988;78(suppl I):I-140-I-143.
Campeau L, Enjalbert M, Lespérance J, Bourassa MG, Kwiterovich PJ, Wacholder S, Sniderman A. The relation of risk factors to the development of atherosclerosis in saphenous-vein bypass grafts and the progression of disease in the native circulation: a study 10 years after aortocoronary bypass surgery. N Engl J Med. 1984;311:1329-1332.
Cohen MV, Byrne M-J, Levine B, Gutowski T, Adelson R. Low rate of treatment of hypercholesterolemia by cardiologists in patients with suspected and proven coronary artery disease. Circulation. 1991;83:1294-1304.
Henderson WG, Goldman S, Copeland JG, Moritz TE, Hasker LA. Antiplatelet or anticoagulant therapy after coronary artery bypass surgery: a meta analysis of clinical trials. Ann Intern Med. 1989;111:743-750.
Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy, II: maintenance of vascular graft or arterial patency by antiplatelet therapy. Br Med J. 1994;308:159-168.
Merry AF, Ramage MC, Whitlock RML, Laycock GJA, Smith W, Stenhouse D, Wild CJ. First time coronary artery bypass grafting: the anaesthetist as a risk factor. Br J Anaesth. 1992;68:6-12.
Zehr KJ, Lee PC, Poston RS, Gillinov M, Greene PS, Cameron DE. Two decades of coronary artery bypass graft surgery in young adults. Circulation. 1994;90(part 2):II-133-II-139.
Buffet P, Colasante B, Bischoff N, Juillière Y, Danchin N, Feldmann L, Selton-Suty C, Amrein D, Mathieu P, Cherrier F. Fifteen-year follow-up study of coronary surgery with left internal mammary artery bypass grafting to the left anterior descending artery in patients younger than 40 years. Eur Heart J. 1993;14(suppl):171.
White HD, Norris RM, Brown MA, Brandt PWT, Whitlock RML. Left ventricular end-systolic volume as the major determinant of survival after recovery from myocardial infarction. Circulation. 1987;76:44-51.
Hamer AW, Takayama M, Abraham KA, Roche AHG, Kerr AR, Williams BF, Ramage MC, White HD. End-systolic volume and long-term survival after coronary artery bypass graft surgery in patients with impaired left ventricular function. Circulation. 1994;90:2899-2904.