Coronary Revascularization (Surgical or Percutaneous) Decreases Mortality After the First Year in Diabetic Subjects but not in Nondiabetic Subjects With Multivessel Disease
An Analysis From the Medicine, Angioplasty, or Surgery Study (MASS II)
Background— It is currently unknown whether revascularization procedures are associated with an improvement in mortality among diabetic subjects, as compared with a more conservative medical treatment.
Methods and Results— In MASS II, a total of 611 patients with stable multivessel coronary disease were randomly assigned to medical treatment, surgery, or angioplasty. From these, 190 patients had diabetes (medical, 75 patients; angioplasty, 56 patients; surgery, 59 patients) and comprised the present study population. Mortality rates were analyzed for the entire 5 years of follow-up. Separate analyzes were also performed for mortality at 2 time intervals: during the first year and after the first year of follow-up. We calculated the probability of death conditional on surviving to the start of the interval analyzed. The cumulative 5-year mortality as well as the mortality during the first year of follow-up was not significantly different among treatment groups, both for diabetic and for nondiabetic subjects. Also, during years 2 to 5, the mortality of the 3 treatment groups was not different for nondiabetic subjects. Among diabetic subjects, however, patients randomized to angioplasty or surgery had a significantly lower mortality between years 2 and 5 than those allocated to medical treatment (P=0.039).
Conclusion— Surgery, angioplasty, and medical treatment appear to be associated with similar mortality rates for non-diabetic subjects. For diabetic subjects, however, coronary revascularization (percutaneous or surgical) significantly decreased the risk of death after the first year and up to 5 years, compared with medical treatment alone.
Previous studies have shown that the presence of diabetes significantly decreases the adjusted long-term survival even after coronary revascularization either with percutaneous coronary intervention or with bypass surgery.1–4 When comparing both invasive treatments for patients with multivessel disease, investigators from the Bypass Angioplasty Revascularization Investigation (BARI) study demonstrated that diabetic subjects treated with balloon angioplasty were at increased risk for death in comparison with patients undergoing surgical revascularization.5 However, the effect of diabetes on the outcomes after invasive treatment still remain controversial. In BARI, the difference in long-term mortality between the study groups was mostly explained by the larger degree of completeness of revascularization in the surgical arm, compared with patients randomized to balloon angioplasty.6 Moreover, no difference in mortality was seen between coronary surgery and angioplasty in more recent trials using coronary stenting instead of plain balloon dilatation.7
Currently, scarce information is available on the outcomes of multivessel patients maintained on medical therapy. Medical treatment has improved in the past years with the widespread use of aspirin, beta-blockers, statins, and angiotensin-converting enzyme inhibitors. Most of the recent randomized trials on coronary artery disease did not include a conservative treatment arm. The Medicine, Angioplasty, or Surgery Study (MASS II) has previously compared the outcomes of patients with stable multivessel coronary disease randomized to surgery, angioplasty, or medical treatment, with no evidence of 1-year mortality difference for the overall population.8 However, to the best of our knowledge, no published trial to date have compared medical treatment with revascularization in diabetic patients followed-up for longer periods. It is still debatable whether a strategy of early revascularization compared with a more conservative approach reduces mortality in patients with type 2 diabetes presenting with mild and stable symptoms.
Therefore, in the present report, we aimed to analyze the 5-year mortality rates in patients with and without diabetes according to 3 therapeutic options, medical, surgical, and angioplasty in the MASS II population.
Study Design and Patient Population
The study design of MASS II has been previously published elsewhere.8 Briefly, the MASS II is a randomized, prospective, single-center study that compared 3 therapeutic strategies, medical, surgery, and angioplasty in patients with multivessel and stable coronary artery disease. A total of 611 patients were randomized between May 1995 and March 2000. The present report is a post-hoc analysis using the overall MASS II population, which was retrospectively divided into 2 groups according to the presence of diabetes at baseline. Baseline, procedural, and follow-up data were stored in a dedicated electronic database specifically built for the MASS II study.
Diabetes was defined as the presence of serum glucose concentration ≥140 mg/dL on at least 2 separate occasions. We also considered as diabetic subjects patients with previous diagnosis using specific medication, oral drugs, or insulin. These criteria were the most accepted definition for diabetes at the beginning of the study and were maintained in the present report.9 From the total population, 190 patients were diabetic subjects (75 in medical, 56 in angioplasty, and 59 in surgical groups) and 421 patients were nondiabetic subjects (128 in medical, 149 in angioplasty, and 144 in surgical groups).
This protocol was approved by the hospital ethics committee and is in accordance with the Declaration of Helsinki. Written, informed consent was obtained from every patient.
Device choice was left at the discretion of the operator and included stents, excimer laser, directional atherectomy, rotablator, and balloon angioplasty. The interventional cardiologist was encouraged to treat all arteries that were likely to contribute to ischemia and/or had lesion with ≥70% diameter stenosis. Angioplasty was performed according to a standard protocol that included administration of aspirin before the procedure. Dilatation of a stenotic vessel was considered successful if the residual stenosis of the lumen diameter was <50%. Patients treated with coronary stents were maintained on ticlopidine 250 mg twice daily for 1 month in addition to lifelong aspirin.
For patients assigned to the surgery, the cardiac surgeon was encouraged to intervene in all feasible stenosed arteries as an attempt to accomplish complete revascularization. Use of internal mammary conduits was strongly advised for all cases. Coronary bypass was executed using standard surgical techniques, under hypothermic arrest, with the use of blood cardioplegy.
The medication recommendations were not different for patients randomized to surgery, angioplasty, or medical therapy alone. All patients were treated with lifelong aspirin. Nitrates, angiotensin-converting enzyme inhibitors, and β-blocking agents were administered with the objective to keep the patient free of angina. Statins were prescribed with low-fat diet with the objective to keep low-density lipoprotein cholesterol <100 mg/dL. Anti-hypertensive drugs were used according to the physician’s judgment. For diabetes treatment, sulfonylurea, insulin, and metformin were used with the main objective to keep fasting glucose levels <130 mg/dL.
The study primary end point was to compare the impact of the 3 treatment strategies, medical, angioplasty, or surgery, in the mortality in diabetic and nondiabetic patients at 5 years. To better-evaluate the clinical outcomes, we analyzed separately the mortality trends, in 2 intervals: the first year of follow-up and for the next 4 years (years 2 to 5). The purpose of this analysis was to exclude deaths that were related to the invasive procedure, surgical or percutaneous, and to evaluate the potential protective effect of invasive strategies compared with the conservative arm in the long-term follow-up.
Categorical variables were compared using Fisher exact test and continuous variables were compared using Student t test. Event rates were estimated with the Kaplan-Meier method and compared using the log-rank test. Hazard rates were calculated from the life table analyses as the death rate during a time interval, conditional on survival to the end of the preceding time interval. P<0.05 was considered to be significant. All statistical analyses were performed with the statistical package SPSS 13.0; SPSS Inc, Chicago, Ill).
The authors had full access to the data and take full responsibility for their integrity. All authors have read and agree to the manuscript as written.
Baseline and Procedural Characteristics
Baseline characteristics were similar among patients randomized to medical treatment, angioplasty, or surgery within the diabetic and nondiabetic subgroups, except by a higher prevalence of previous myocardial infarction in nondiabetic subjects treated with angioplasty (compared with nondiabetic subjects in the other treatment groups) (Table 1). When pooled over treatment groups, diabetic subjects had lower prevalence of males, higher prevalence of hypertension, higher ejection fraction, and higher fasting serum glucose (Table 1). Triglycerides tended to be higher and low-density lipoprotein cholesterol tended to be lower in diabetic subjects compared with nondiabetic subjects (Table 1).
The rates of revascularization of the three coronary territories at the index procedure (right coronary artery, left circumflex, and left anterior descending) were significantly lower in patients treated with angioplasty than in surgical patients, both among diabetic subjects and among nondiabetic subjects (Table 1). Overall, stents were used in >80% of patients randomized to percutaneous intervention and left internal mammary artery (IMA) or right IMA grafts were used in 95% of surgical patients (Table 1).
Patients were followed-up for an average period of 1702±452 days (median, 1840 days). The Figure shows the 5-year mortality curves for patients randomized to medical treatment, angioplasty, or surgery in the diabetic (Figure, A) and in the nondiabetic (Figure, B) subpopulations. The cumulative 5-year mortality was not significantly different among treatment groups within diabetic subjects (P=0.25 by log-rank) and within nondiabetic subjects (P=0.63 by log rank) (Figure).
To better-evaluate the clinical outcomes along the study, the mortality trends were separately analyzed for the first year of follow-up and for the next 4 years thereafter (years 2 to 5). Both time intervals had their respective hazard rates calculated, which is an estimate of the probability of death conditional on surviving to the start of the interval (Table 2). During the first year, the mortality was not statistically different among the treatment groups, both for nondiabetic subjects and for diabetic subjects (P=0.2 and P=0.5 by log-rank, respectively). Similarly, during the second period of the follow-up (years 2 to 5), the mortality of the 3 treatment groups was not different for nondiabetic subjects, when analyzing patients who have survived the first year of follow-up (P=0.5 by log rank). Conversely, among diabetic subjects, patients randomized to medical treatment (who had survived the first year) had a significant increase in mortality between years 2 and 5 (P=0.039), compared with patients treated with percutaneous intervention or surgery (with no difference in mortality between 2 and 5 years for angioplasty versus surgery; P=0.7).
Cardiovascular medication usage at 5 years is shown in Table 3. Overall, diabetic subjects and nondiabetic subjects had similar prescription, except by a higher aspirin use among diabetic subjects. In total, 74% of patients were using statins at 5 years. Anti-ischemic drugs were significantly more frequent for patients randomized to medical treatment.
The main findings of the present study are that the initial therapeutic approach, medical treatment, angioplasty, or surgery, did not modify the mortality trend during 5 years for nondiabetic subjects with stable multivessel coronary disease. Also, the treatment modality did not influence the outcomes during the first year in diabetic subjects. However, from the first year and afterward, diabetic subjects undergoing treatment with invasive strategies (angioplasty or surgery) had significantly improved mortality rates in comparison with patients randomized to a more conservative medical strategy.
In the BARI study, diabetic subjects with multivessel coronary disease treated with balloon angioplasty had worse long-term survival than those undergoing surgery. There was no difference in long-term survival in the nondiabetic patients.5 Interestingly, the long-term benefit of bypass surgery for diabetic subjects could be explained by the increased rates of complete revascularization in the surgical arm.6 Based on these findings, it is reasonable to expect that patients with diabetes and multivessel disease treated with either angioplasty or surgery would have similar outcomes if both strategies could be performed with a comparable completeness of revascularization. Accordingly, more recent randomized trials with coronary stenting (when more aggressive and effective percutaneous approaches were used in comparison with early trials with balloon angioplasty) did not show a major difference in mortality between angioplasty and surgery for diabetic subjects with multivessel disease.7 Conversely, a large observational study have shown a survival benefit for patients treated with surgery, compared with angioplasty, for multivessel disease.10 Our findings strongly support the concept that the level of revascularization is inversely proportional to the risk of death during the follow-up, because diabetic subjects maintained on medical treatment presented a worse prognosis than patients treated invasively. It is interesting to observe that the impact of revascularization procedures for diabetic subjects was only evident after the first year of follow-up. It could be hypothesized that, during the first year, the protective effect of invasive treatment was overcome by the incidence of procedure-related deaths. After the first year, however, the benefit of revascularization for diabetic subjects was translated into a significant reduction in mortality as compared with patients randomized to a noninvasive treatment. Curiously, the beneficial effect of revascularization on diabetic subjects occurred in a population with a significant higher proportion of females, when compared with nondiabetic subjects, a factor that has been previously reported to increase the risk of long-term complications.11
Modern medical therapy is superior to that used in early studies, not only for coronary artery disease but also for diabetes mellitus. In MASS II, aspirin was prescribed for >90% of the patients, beta-blockers for >70%, and statins for >70%. It is important to notice that a suboptimum medication scheme did not seem explain the worse outcome observed for diabetic subjects allocated to medical treatment, which reinforces the beneficial effect of an invasive strategy for this population even in the context of optimized drug therapy. Nevertheless, the treatment of diabetes has markedly changed in the past years, with 2 different approaches (or their association) currently being undergoing intense investigation; the use of insulin or drugs that promote the increase of insulin secretion (insulin-providing approach) or the use of drugs that reduce insulin resistance (insulin-sensitizing approach).6 Many unanswered questions about the impact of the treatment of diabetes for patients with stable coronary artery disease will probably be answered by the ongoing BARI 2D trial.12
The present study has several limitations. It is a retrospective analysis of MASS II trial, which was not specifically designed to compare the effects of the 3 strategies in the diabetic population. Moreover, it was not powered to detect subtle differences in mortality, especially when considering the relatively low risk of the included patients, who had normal left ventricular function and triple vessel disease in only &50% of the cases. Also, the low number of insulin-dependent diabetic subjects in the study arms precluded a more detailed analysis on the outcomes of these subgroups. It should be noted that the population included in the present study was possibly at a higher risk for complications than a population selected according to a diagnostic criteria using lower levels of fasting glucose. Nevertheless, a clear reduction in the risk of death could be seen after the first year in diabetic subjects treated invasively. New therapeutic approaches for diabetes, as well as a change toward more rigid diagnostic criteria of the disease have been advocated in the past few years. Future studies should investigate whether the present findings will be reproduced in the context of the new diagnostic definitions of diabetes and more modern treatment schemes.
Surgery, angioplasty, and medical treatment did not influence the risk of death for nondiabetic subjects. For diabetic subjects, however, coronary revascularization (percutaneous or surgical) was associated with a protective effect compared with medical treatment alone, significantly decreasing the risk of death after the first year and up to 5 years.
Presented at the American Heart Association Scientific Sessions, Dallas, Tex, November 13–16, 2005.
Kip KE, Faxon DP, Detre KM, Yeh W, Kelsey SF, Currier JW. Coronary angioplasty in diabetic patients. The National Heart, Lung, and Blood Institute Percutaneous Transluminal Coronary Angioplasty Registry. Circulation. 1996; 94: 1818–1825.
Halon DA, Merdler A, Flugelman MY, Rennert HS, Weisz G, Shahla J, Lewis BS. Late-onset heart failure as a mechanism for adverse long-term outcome in diabetic patients undergoing revascularization (a 13-year report from the Lady Davis Carmel Medical Center Registry). Am J Cardiol. 2000; 85: 1420–1426.
Weintraub WS, Stein B, Kosinski A, Douglas JS Jr, Ghazzal ZM, Jones EL, Morris DC, Guyton RA, Craver JM, King SB III. Outcome of coronary bypass surgery versus coronary angioplasty in diabetic patients with multivessel coronary artery disease. J Am Coll Cardiol. 1998; 31: 10–19.
Barsness GW, Peterson ED, Ohman EM, Nelson CL, DeLong ER, Reves JG, Smith PK, Anderson RD, Jones RH, Mark DB, Califf RM. Relationship between diabetes mellitus and long-term survival after coronary bypass and angioplasty. Circulation. 1997; 96: 2551–2556.
Abizaid A, Costa MA, Centemero M, Abizaid AS, Legrand VM, Limet RV, Schuler G, Mohr FW, Lindeboom W, Sousa AG, Sousa JE, van Hout B, Hugenholtz PG, Unger F, Serruys PW; Arterial Revascularization Therapy Study Group. Clinical and economic impact of diabetes mellitus on percutaneous and surgical treatment of multivessel coronary disease patients: insights from the Arterial Revascularization Therapy Study (ARTS) trial. Circulation. 2001; 104: 533–538.
Hueb W, Soares PR, Gersh BJ, Cesar LA, Luz PL, Puig LB, Martinez EM, Oliveira SA, Ramires JA. The medicine, angioplasty, or surgery study (MASS-II): a randomized, controlled clinical trial of three therapeutic strategies for multivessel coronary artery disease: one-year results. J Am Coll Cardiol. 2004; 43: 1743–1751.
American Diabetes Association. Clinical practice recommendations 1997. Diabetes Care. 1997; 20 (Suppl 1): S1–S70.
Jacobs AK. Coronary revascularization in women in 2003: sex revisited. Circulation. 2003; 107: 375–377.
Sobel BE, Frye R, Detre KM. Burgeoning dilemmas in the management of diabetes and cardiovascular disease: rationale for the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) Trial. Circulation. 2003; 107: 636–642(Erratum in: Circulation. 2003;108:500.)