Abstract
Background
Patients with diabetes mellitus (DM) are at higher risk for adverse outcomes following percutaneous coronary intervention (PCI).
Methods
To determine whether outcomes have improved over time, we analyzed data from 2838 consecutive patients with medically-treated DM, including 1066 patients (37.6%) treated with insulin, in the National Heart Lung and Blood Institute Dynamic Registry undergoing PCI registered in Waves 1(1997–98), 2 (1999), 3 (2001–02), 4 (2004) and 5 (2006). We compared baseline demographics and 1-year outcomes in the overall cohort, and in analyses stratified by recruitment wave and insulin use.
Results
Crude mortality rates among those treated with insulin by chronological Wave were 9.5%, 12.5%, 8.9%, 11.6%, and 6.6% (p-valuetrend=0.33); and respectively, among patients treated by oral agents:9.7%, 6.5%, 4. 1%, 5.4%, and 4.7%, (p-valuetrend=0.006). The adjusted hazard ratios of death, myocardial infarction (MI), and overall major adverse cardiovascular events (death, MI, revascularization) in insulin treated patients with DM in waves 2–5 as compared to wave 1 were either higher or the same. In contrast, the similar adjusted hazard ratios for oral agent treated patients with DM were either similar or lower.
Conclusions
Significant improvements over time in adverse events by 1-year were detected in patients with DM treated with oral agents. In insulin-treated diabetic patients, despite lower rates of repeat revascularization over time, death and MI following PCI have not significantly improved. These findings underscore the need for continued efforts at optimizing outcomes among patients with DM undergoing PCI, especially those requiring insulin treatment.
Background
The number of patients with diabetes mellitus (DM) who undergo invasive cardiac evaluation has been increasing with the rising number of American s diagnosed with the disorder. The total prevalence of diabetes in the United States is projected to more than double (from 5.6% to 12.0%) from 2005 to 2050.1 Patients with DM are at higher risk of both short and long-term adverse events following percutaneous coronary intervention (PCI).2–5 Specifically, patients with DM are at higher risk for restenosis, target vessel revascularization, late myocardial infarction (MI), and death. Furthermore, the subset of patients with DM treated with insulin are a particularly high-risk group.6–8 An evaluation of temporal trend sin outcomes of patients with DM undergoing PCI has not been report ed. Also lacking are data assessing the association between treatment regimens for DM and clinical outcomes over time. We evaluated the outcome of patients with DM who underwent PCI in multiple sequential waves of the National Heart Lung and Blood Institute (NHLBI) Dynamic Registry to evaluate the temporal outcome of such patients.
Methods
Study Population
The specific methodologies and characteristics of the NHLBI Dynamic Registry have been reported previously.9 In brief, data were collected on approximately 2,000 consecutive patients undergoing PCI during five recruitment ‘waves’ across 27 clinical centers and incorporates an enriched sample of women and minorities(Wave 1: July 1997–February 1998;Wave 2: February–June 1999; Wave 3: October 2001–March 2002; Wave 4: February–May 2004; Wave 5: February 2006–August 2006). Patients were contacted via telephone interview at one year by trained nurse coordinators to assess vital status, symptoms, coronary events or cardiac-related hospitalizations, and medication status. Informed consent was obtained for all patients and the study protocol was approved by Institutional Review Boards at the respective clinical sites and at the University of Pittsburgh data coordinating center. The present analyses comprise data from patients with medically-treated DM undergoing PCI in Waves 1–5.
Definitions
Major in-hospital complications (death from any cause, myocardial infarction [MI] or coronary artery bypass graft [CABG])were recorded. MI was defined as evidence of two or more of the following: (1) typical chest pain > 20 min duration not relieved by nitroglycerin, (2) serial electrocardiogram recordings showing changes from baseline or serially in ST-T and/or Q-waves in ≥ 2 contiguous leads, (3) serum enzyme elevation of CK-MB > 5% [total CK >2X normal, LDH subtype 1 > LDH subtype 2, or troponin > 0.2 μg/ml], or (4) new wall motion abnormalities. A composite outcome of major adverse coronary event (MACE) rate was defined as a composite of death, MI and repeat revascularization. Successful lesion dilation was defined as an absolute 20% reduction in lesion severity with a final stenosis ≤ 50%. Angiographic success was defined as successful treatment of all lesions.
Statistical Analysis
A stratified analysis was performed because (1) interaction terms are often difficult to interpret by clinical personnel and (2) direct interpretation of the point estimates could be made for each of our diabetes groups. Trends in baseline clinical, angiographic, and procedural characteristics between the waves were compared by Mantel-Haenszel test for trend for discrete variables and the Jonckheere-Terpstra test for continuous variables. Clinical event rates at 1-year were calculated via the Kaplan Meier approach and compared using the log-rank test for trend. Patients who did not experience the outcome of interest were censored at the last known date of contact or at one year if contact extended beyond one year. The association between recruitment wave and 1-year safety(death, MI, death/MI and death/MI/CABG) and efficacy (CABG, repeat PCI) events were evaluated using Cox proportional hazards regression modeling. The reference group for all models was Wave 1. Important covariate variables considered for all models were initially screened for univariable association with each outcome of interest where p<0.20. For the final model the recruitment wave variable was forced to stay in the model and the remaining covariates (pentry<0.15, pstay<0.05) identified from the initial screening were assessed using backward stepwise process. The proportionality assumption was assessed for all Cox proportional-hazards models graphically and the hazard ratios (HR) represent the average risk during the year after the index procedure. The complete list of variables that were included in the regression models are in the appendix.
This study was supported by grant number HL-33292 from the National Heart, Lung, and Blood Institute of the National Institutes of Health. The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper and its final contents.
Results
We compared baseline demographics and 1-year outcomes in the overall cohort, and in analyses stratified by a) recruitment wave and b) insulin use. The study cohort comprised 2838 patients with medically-treated DM from Waves 1 (insulin treated n=215, oral agent n=338), 2 (insulin n=193, oral agent n=312), 3 (insulin n=194, oral agent n=333), 4 (insulin n=229, oral agent n=382), and 5 (insulin n=235, oral agent n=407). Baseline characteristics stratified by Wave of enrollment and diabetic treatment are presented in Table 1. In the overall cohort and both insulin-and oral agent-treated patients with DM, there was a lower prevalence of women and of a prior MI in more recent cohorts while the prevalence of prior percutaneous procedures and renal disease increased over time (p valuetrend<0.01 for each). The procedural characteristics of the overall cohort and diabetic patients stratified by treatment with insulin or oral agents are in Table 2. In the overall cohort and in the oral agent-treated diabetic patients, there was an increase in the percentage of three vessel disease in later cohorts. In the overall cohort and both treatment groups, the lesion length was significantly longer in more recent cohorts. The frequency of attempted lesions containing thrombus decreased with time in the overall cohort as well as in oral-agent treated patients, but this observation was not noted in the insulin-treated patients with diabetes.
Table 1.
Demographic Features of Diabetic Patients by Enrollment Wave and Medical Therapy
| Characteristic | Wave 1 | Wave 2 | Wave 3 | Wave 4 | Wave 5 | P-value trend | |
|---|---|---|---|---|---|---|---|
| Number of patients | |||||||
| Total | 553 | 505 | 527 | 611 | 642 | ||
| Insulin | 215 | 193 | 194 | 229 | 235 | ||
| Oral Agent | 338 | 312 | 333 | 382 | 407 | ||
| Mean age | |||||||
| Total | 63.6 | 63.1 | 65.2 | 64.3 | 63.6 | 0.53 | |
| Insulin | 63.9 | 62.2 | 64.2 | 63.5 | 62.5 | 0.65 | |
| Oral Agent | 63.5 | 63.7 | 65.7 | 64.8 | 64.2 | 0.26 | |
| Female (%) | |||||||
| Total | 45.9 | 48.1 | 43.5 | 40.9 | 37.5 | 0.0002 | |
| Insulin | 56.3 | 53.9 | 49.0 | 47.6 | 45.1 | 0.007 | |
| Oral Agent | 39.3 | 44.6 | 40.2 | 36.9 | 33.2 | 0.01 | |
| Hypertension (%) | |||||||
| Total | 73.1 | 78.4 | 86.3 | 87.6 | 88.5 | <0.0001 | |
| Insulin | 78.5 | 79.2 | 88.7 | 88.9 | 88.7 | <0.0001 | |
| Oral Agent | 69.6 | 77.9 | 84.9 | 86.8 | 88.4 | <0.0001 | |
| Congestive Heart Failure (%) | |||||||
| Total | 18.4 | 15.8 | 22.9 | 13.1 | 15.9 | 0.11 | |
| Insulin | 23.1 | 21.6 | 26.8 | 18.9 | 22.4 | 0.64 | |
| Oral Agent | 15.3 | 12.2 | 20.7 | 9.7 | 12.2 | 0.11 | |
| Prior PCI* (%) | |||||||
| Total | 33.2 | 36.3 | 41.3 | 37.9 | 43.9 | <0.0001 | |
| Insulin | 36.0 | 38.3 | 42.3 | 34.9 | 46.9 | 0.006 | |
| Oral Agent | 31.5 | 35.0 | 40.8 | 39.8 | 42.3 | 0.0003 | |
| Prior CABG† (%) | |||||||
| Total | 21.4 | 22.6 | 25.6 | 23.8 | 25.6 | 0.21 | |
| Insulin | 27.9 | 29.5 | 29.9 | 28.5 | 33.6 | 0.42 | |
| Oral Agent | 17.2 | 18.2 | 23.1 | 44.1 | 21.0 | 0.28 | |
| Prior Myocardial Infarction(%) | |||||||
| Total | 39.8 | 39.6 | 30.7 | 29.1 | 27.8 | <0.0001 | |
| Insulin | 40.9 | 47.4 | 31.2 | 28.1 | 31.5 | 0.0003 | |
| Oral Agent | 39.1 | 34.9 | 30.4 | 29.7 | 25.8 | <0.0001 | |
| Renal disease (%) | |||||||
| Total | 7.3 | 8.7 | 14.3 | 17.0 | 17.6 | <0.0001 | |
| Insulin | 11.3 | 12.6 | 22.2 | 27.3 | 29.3 | <0.0001 | |
| Oral Agent | 4.8 | 6.4 | 9.6 | 10.8 | 10.8 | 0.0006 | |
| Revascularization Indication (%) | |||||||
| Asymptomatic CAD‡ | |||||||
| Total | 1.8 | 5.0 | 9.9 | 9.2 | 14.8 | <0.0001 | |
| Insulin | 2.3 | 6.2 | 7.2 | 8.7 | 14.5 | <0.0001 | |
| Oral Agent | 1.5 | 4.2 | 11.4 | 9.4 | 15.0 | <0.0001 | |
| Stable Angina | |||||||
| Total | 23.0 | 19.8 | 20.1 | 24.7 | 17.0 | 0.15 | |
| Insulin | 20.1 | 19.2 | 19.6 | 21.8 | 15.3 | 0.39 | |
| Oral Agent | 24.9 | 20.2 | 20.4 | 26.4 | 18.0 | 0.24 | |
| Unstable Angina | |||||||
| Total | 49.6 | 49.5 | 45.2 | 37.6 | 37.3 | <0.0001 | |
| Insulin | 53.3 | 52.8 | 42.3 | 39.3 | 37.4 | <0.0001 | |
| Oral Agent | 47.3 | 47.4 | 46.8 | 36.6 | 37.2 | 0.0001 | |
| Acute Myocardial Infraction | |||||||
| Total | 19.4 | 22.4 | 21.8 | 22.6 | 24.3 | 0.06 | |
| Insulin | 19.6 | 18.1 | 28.9 | 23.6 | 23.8 | 0.14 | |
| Oral Agent | 19.2 | 25.0 | 17.7 | 22.0 | 24.6 | 0.21 | |
Percutaneous Coronary Intervention
Coronary Artery Bypass Graft Surgery
Coronary Artery Disease
Table 2.
Procedural Characteristics of Diabetic Patients by Enrollment Wave and Medical Therapy
| Wave 1 | Wave 2 | Wave 3 | Wave 4 | Wave 5 | P-value trend | ||
|---|---|---|---|---|---|---|---|
| Patient Level (n) | |||||||
| Total | 553 | 505 | 527 | 611 | 642 | ||
| Insulin | 215 | 193 | 194 | 229 | 235 | ||
| Oral Agent | 338 | 312 | 333 | 382 | 407 | ||
| Mean Ejection Fraction | |||||||
| Total | 52.4 | 51.6 | 49.4 | 50.8 | 51.2 | 0.56 | |
| Insulin | 50.5 | 48.3 | 48.4 | 51.8 | 49.3 | 0.45 | |
| Oral Agent | 53.7 | 53.8 | 50.0 | 50.2 | 52.2 | 0.14 | |
| Vessel Disease (%) | |||||||
| Total | 0.006 | ||||||
| Single | 34.0 | 33.3 | 29.8 | 27.7 | 28.7 | ||
| Double | 32.4 | 35.2 | 27.7 | 34.4 | 32.1 | ||
| Triple | 33.5 | 30.9 | 42.1 | 37.3 | 38.6 | ||
| Insulin | 0.23 | ||||||
| Single | 32.1 | 28.5 | 25.3 | 27.1 | 25.5 | ||
| Double | 28.8 | 38.3 | 25.3 | 34.9 | 32.8 | ||
| Triple | 39.2 | 32.6 | 49.0 | 37.6 | 41.3 | ||
| Oral Agent | 0.009 | ||||||
| Single | 35.2 | 36.2 | 32.4 | 28.0 | 30.5 | ||
| Double | 34.6 | 33.3 | 29.1 | 34.0 | 31.7 | ||
| Triple | 29.9 | 29.8 | 38.1 | 37.2 | 37.1 | ||
| Mean Significant Lesions | |||||||
| Total | 3.4 | 3.2 | 3.6 | 3.4 | 3.4 | 0.05 | |
| Insulin | 3.5 | 3.4 | 3.9 | 3.7 | 3.8 | 0.16 | |
| Oral Agent | 3.3 | 3.0 | 3.5 | 3.3 | 3.3 | 0.11 | |
| # Lesions Attempted | |||||||
| Total | 0.02 | ||||||
| 1 | 63.9 | 70.4 | 64.9 | 72.5 | 69.3 | ||
| 2 | 27.2 | 22.5 | 25.4 | 21.8 | 22.9 | ||
| >2 | 8.8 | 7.2 | 9.7 | 5.7 | 7.8 | ||
| Insulin | 0.72 | ||||||
| 1 | 65.6 | 70.3 | 64.9 | 71.2 | 68.9 | ||
| 2 | 30.2 | 21.9 | 24.2 | 24.5 | 23.0 | ||
| >2 | 7.9 | 3.7 | 1.5 | 1.3 | 1.7 | ||
| Oral Agent | 0.008 | ||||||
| 1 | 62.9 | 70.4 | 64.9 | 73.3 | 69.5 | ||
| 2 | 25.2 | 22.8 | 26.1 | 20.2 | 22.9 | ||
| >2 | 11.9 | 6.7 | 9.0 | 6.5 | 2.0 | ||
| GP IIb/IIIa inhibitor* (%) | |||||||
| Total | 23.9 | 30.1 | 51.4 | 33.1 | 31.2 | 0.01 | |
| Insulin | 20.5 | 30.6 | 48.2 | 38.4 | 29.8 | 0.01 | |
| Oral Agent | 26.0 | 29.8 | 53.3 | 29.8 | 31.9 | 0.23 | |
| Lesion Level (n) | P-value trend | ||||||
| Total | 823 | 705 | 778 | 823 | 903 | ||
| Insulin | 300 | 274 | 286 | 309 | 331 | ||
| Oral Agent | 523 | 431 | 492 | 514 | 572 | ||
| Mean Lesion Length | |||||||
| Total | 12.7 | 13.6 | 13.6 | 16.0 | 17.2 | <0.0001 | |
| Insulin | 12.9 | 13.2 | 13.9 | 15.7 | 17.6 | <0.0001 | |
| Oral Agent | 12.6 | 13.8 | 13.5 | 16.1 | 17.1 | <0.0001 | |
| Evidence of thrombus (%) | |||||||
| Total | 16.7 | 13.5 | 10.3 | 10.0 | 11.9 | 0.0005 | |
| Insulin | 16.5 | 10.7 | 15.4 | 10.6 | 13.9 | 0.40 | |
| Oral Agent | 16.8 | 15.4 | 7.6 | 9.7 | 10.8 | 0.0001 | |
| Calcified Lesion (%) | |||||||
| Total | 28.3 | 25.8 | 20.5 | 23.7 | 34.9 | 0.01 | |
| Insulin | 27.9 | 31.2 | 22.4 | 29.3 | 33.2 | 0.25 | |
| Oral Agent | 28.6 | 22.4 | 19.4 | 20.4 | 35.9 | 0.02 | |
| Stent Use (%) | |||||||
| Total | 55.5 | 68.8 | 74.3 | 89.8 | 91.7 | <0.0001 | |
| Insulin | 52.3 | 74.5 | 71.7 | 88.7 | 91.2 | <0.0001 | |
| Oral Agent | 57.4 | 65.2 | 75.8 | 90.5 | 92.0 | <0.0001 | |
| Drug-Eluting Stent Use (%) | |||||||
| Total | 0.0 | 0.0 | 0.0 | 67.2 | 83.6 | <0.0001 | |
| Insulin | 0.0 | 0.0 | 0.0 | 68.3 | 83.4 | <0.0001 | |
| Oral Agent | 0.0 | 0.0 | 0.0 | 66.5 | 83.7 | <0.0001 | |
| Angiographic Success (%) | |||||||
| Total | 94.5 | 95.7 | 96.8 | 96.0 | 96.6 | 0.05 | |
| Insulin | 93.3 | 96.0 | 96.2 | 96.1 | 96.1 | 0.13 | |
| Oral Agent | 95.2 | 95.6 | 97.2 | 95.9 | 96.6 | 0.27 | |
Glycoprotein IIb/IIIa inhibitor
The in-hospital and overall cumulative 1-year event rates in patients with medically treated DM are presented in Table 3. The overall cohort of medically-treated diabetic patients had a significant reduction in in-hospital mortality over time. Diabetic patients treated with oral agents had a significant reduction in in-hospital mortality and in the combined endpoint of death/any MI/any CABG in later cohorts. Insulin-treated patients experienced no significant change in any of the in-hospital adverse event rates over time. The percentage of patients with medically-treated DM in the overall cohort and in the subgroups treated with oral agents and insulin who were discharged on angiotensin converting enzyme inhibitors, beta blockers and statins increased significantly over time (p-valuetrend for each <0.001).
Table 3.
In-Hospital and 1-YearCumulative Event Rates of Diabetic Patients by Enrollment Wave and Medical Therapy
| In-Hospital Cumulative Events | Wave 1 | Wave 2 | Wave 3 | Wave 4 | Wave 5 | P-value trend | |
|---|---|---|---|---|---|---|---|
| Number of patients | |||||||
| Total | 553 | 505 | 527 | 611 | 642 | ||
| Insulin | 215 | 193 | 194 | 229 | 235 | ||
| Oral Agent | 338 | 312 | 333 | 382 | 407 | ||
| Death | |||||||
| Total | 3.1 | 1.4 | 1.1 | 2.3 | 0.8 | 0.03 | |
| Insulin | 1.4 | 2.6 | 1.5 | 3.9 | 0.9 | 0.95 | |
| Oral Agent | 4.1 | 0.6 | 0.9 | 1.3 | 0.7 | 0.003 | |
| Myocardial infarction | |||||||
| Total | 1.6 | 1.4 | 1.5 | 2.8 | 1.1 | 0.90 | |
| Insulin | 0.5 | 1.0 | 2.1 | 3.1 | 0.9 | 0.31 | |
| Oral Agent | 2.4 | 1.6 | 1.2 | 2.6 | 1.2 | 0.55 | |
| CABG | |||||||
| Total | 0.5 | 1.4 | 0.2 | 1.0 | 0.3 | 0.43 | |
| Insulin | 0.5 | 0.5 | 0.5 | 1.3 | 0.9 | 0.40 | |
| Oral Agent | 0.6 | 1.9 | 0.0 | 0.8 | 0.0 | 0.08 | |
| Death/Any MI/Any CABG | |||||||
| Total | 4.9 | 3.8 | 2.7 | 5.6 | 1.9 | 0.08 | |
| Insulin | 2.3 | 4.1 | 3.6 | 7.4 | 2.1 | 0.51 | |
| Oral Agent | 6.5 | 3.5 | 2.1 | 4.5 | 1.7 | 0.005 | |
| 1-Year Cumulative Events | |||||||
| Death | |||||||
| Total | 9.6 | 8.8 | 5.9 | 7.7 | 5.2 | 0.005 | |
| Insulin | 9.5 | 12.5 | 8.9 | 11.6 | 6.6 | 0.33 | |
| Oral Agent | 9.7 | 6.5 | 4.1 | 5.4 | 4.7 | 0.006 | |
| Myocardial Infarction | |||||||
| Total | 5.8 | 5.3 | 7.7 | 7.1 | 5.8 | 0.7 | |
| Insulin | 4.6 | 5.9 | 11.5 | 9.6 | 9.5 | 0.05 | |
| Oral Agent | 6.6 | 5.0 | 5.5 | 5.6 | 3.9 | 0.19 | |
| CABG | |||||||
| Total | 8.0 | 8.7 | 5.3 | 3.6 | 3.2 | 0.0001 | |
| Insulin | 7.1 | 6.2 | 3.9 | 3.3 | 2.7 | 0.02 | |
| Oral Agent | 8.6 | 10.2 | 6.1 | 3.8 | 3.6 | 0.0002 | |
| Repeat PCI after discharge | |||||||
| Total | 19.3 | 16.3 | 16.3 | 12.2 | 12.4 | 0.0002 | |
| Insulin | 21.6 | 18.2 | 16.1 | 14.2 | 13.8 | 0.02 | |
| Oral Agent | 17.8 | 15.2 | 16.5 | 11.0 | 13.0 | 0.02 | |
| CABG/Repeat PCI after discharge | |||||||
| Total | 25.2 | 22.3 | 19.7 | 14.7 | 15.0 | 0.0001 | |
| Insulin | 27.5 | 23.8 | 18.0 | 16.3 | 15.0 | 0.0006 | |
| Oral Agent | 23.6 | 21.6 | 20.6 | 13.7 | 16.3 | 0.0005 | |
| Death/MI/CABG/Repeat PCI | |||||||
| Total | 33.7 | 31.0 | 27.9 | 23.1 | 21.5 | 0.0001 | |
| Insulin | 34.6 | 35.5 | 32.0 | 27.8 | 23.5 | 0.005 | |
| Oral Agent | 33.1 | 28.2 | 25.5 | 20.3 | 21.9 | 0.00001 | |
The overall cumulative 1-year mortality rate in patients with medically-treated DM was 7.3%, including 9.7% in insulin-treated patients and 5.9% among those treated with oral agents. At one year, the overall cohort of diabetic patients had a significant reduction in mortality over time with no change in MI rates. (Table 3) While there was no change in cumulative 1-year mortality rates over time in insulin-treated patients, oral-agent treated patients died less frequently in more recent recruitment waves. The cumulative MI rates by wave and diabetes treatment demonstrated a significantly higher 1-year rate of MI in the insulin-treated diabetic patients in later waves, which was not seen in the oral-agent treated patients. There was a significant reduction in the cumulative 1-year need for repeat PCI after discharge in the overall cohort and the treatment subgroups. There was a significant reduction in the combined 1-year event rate by recruitment wave of repeat PCI after discharge and CABG in the overall cohort, the insulin-treated patients, and the oral agent-treated patients. The cumulative 1-year MACE rates were significantly lower in later recruitment waves in the overall cohort, in patients treated with insulin, and in oral agent treated patients.
The adjusted proportional hazard models for 1-year adverse events in diabetic patients stratified by treatment with insulin and oral agents are presented in Figures 1 and 2, respectively. After adjustment for clinical and angiographic predictors of risk, diabetic patients treated with insulin demonstrated a significant increase in mortality in waves 2 and 4 compared with wave 1. There was also a significant increase in the hazard ratio of MI in waves 3 and 4 compared with wave 1, but a decrease in the hazard ratio of repeat PCI/CABG in wave 3, 4, and 5 compared with wave 1. This resulted in no significant difference in the insulin-treated patients for the hazard ratio of MACE by enrollment wave. Diabetic patients treated with oral agents had no significant difference in the adjusted hazard ratio of mortality by wave, a significant decrease in myocardial infarction in wave 5 compared with wave 1, and a significantly reduced adjusted hazard ratio of revascularization with PCI or CABG in waves 4 and 5 compared with wave 1. This resulted in a significant reduction in the combined adjusted hazard ratio of MACE in Waves 4 and 5 compared to Wave 1.
Figure 1.
Adjusted 1-year adverse events In Insulin-Treated Diabetic Patients
Figure 2.
Adjusted 1-year adverse events In Oral Agent-Treated Diabetic Patients
Discussion
In this evaluation of temporal outcomes of diabetic patients in a contemporary PCI registry, we found that the adjusted risk of major adverse cardiac events among diabetic patients treated with oral agents decreased over time. However, among insulin-treated diabetic patients, the adjusted risk for major adverse cardiac events remained steady, despite a reduction in repeat revascularization with PCI and CABG. Additionally, the risk of MI grew in these patients despite advancements in medical therapy.
There are several hypotheses for the higher MI rate in diabetic patients treated with insulin compared to patients managed with oral agents. It is most likely that this represents confounding by disease severity, with insulin treatment a marker of more severe disease in this group. Additionally, endothelial dysfunction as an effect of insulin therapy might be responsible for accelerated atherosclerosis in untreated lesions or enhanced restenosis in treated lesions.10,11 Hyperinsulinemia, which also occurs in insulin-requiring diabetic patients treated with insulin, may contribute to the atherogenic process. Although we did not collect information on the etiology of diabetes in insulin-treated patients in the Dynamic Registry, it is likely that the vast majority of patients were likely to have type 2, insulin-resistant, diabetes. The BARI 2 Diabetes (BARI 2D) trial randomized 2368 patients in a 2 × 2 factorial design to medical therapy versus revascularization (with CABG or PCI at the discretion of the clinician) and also randomized to a glucose management strategy with an insulin-sensitizing strategy versus an insulin-providing one.12 The rates of death and freedom from major cardiovascular events at 5 years were not significantly different between the revascularization versus medical therapy group or between the insulin-sensitizing versus insulin-providing strategy. An evaluation of outcomes based on treatment strategies demonstrated no significant difference in MI rates in the PCI stratum between the insulin sensitization and insulin provision groups.13
Only a few studies in the bare metal stent era have focused on the outcomes after PCI of insulin-treated versus oral agent treated diabetic patients.14,15 Our data are similar to the reported experience at Emory in the angioplasty era, in which insulin-requiring diabetic patients had reduced 5-year survival and infarction-free survival compared with non-insulin requiring diabetic patients.14 Despite the fact that these data come from patients treated from 1980–1990, the similarity in regards to higher adverse outcomes in insulin-treated diabetic patients is striking. This highlights the fact that our advancements in medical therapy post PCI as well as improved procedural techniques have not mitigated some of the underlying risks in insulin-treated diabetic patients undergoing PCI.
Several studies have described temporal improvements in outcomes post PCI in patients with DM, despite an increase in the risk profile of these patients. However these analyses focused on all diabetic patients, regardless of treatment. It is clear from our analysis that diabetic patients undergoing PCI who are treated with oral agents as compared with insulin have very different risk profiles. Our data show significant improvement in the outcomes of the oral agent-treated diabetic patients which was no t seen in the patients treated with insulin. The Mid America Heart Institute investigators reported improvements in outcomes from 1980–1999 in diabetic patients undergoing elective PCI that was not seen in the diabetic patients undergoing urgent PCI16. Given the dates evaluated in this analysis, the majority of these patients would not have undergone intracoronary stenting. Furthermore, our group has reported improvements of outcomes in PCI in patients with DM when stent era patients are compared with angioplasty era patients.17 Our study represents a unique, contemporary analysis of temporal trends in outcomes in the stent era of diabetic patients undergoing PCI.
The Dynamic Registry has published data focusing on the outcomes of drug-eluting stents versus bare-metal stents in insulin-treated versus oral agent treated diabetic patients that describes improved efficacy and similar safety of drug-eluting stents in insulin-treated diabetic patients.18 These data support the findings seen in other clinical trials of drug-eluting stents in diabetic patients.19,20 In line with these findings, we observed a lower risk of repeat PCI and revascularization in insulin-treated diabetic patients in Waves 4 and 5 which occurred after the approval of drug-eluting stents. This may therefore represent one important strategy to reduce the repeat PCI rates in all diabetic patients. It is possible that a component of the subsequent MI rate will be reduced in the setting of longer drug-eluting stents placed in insulin-treated diabetic patients, which will require long-term follow up. However, reduction of the overall mortality of insulin-treated patients who require PCI will likely require focus on alternative therapeutic strategies outside of the catheterization laboratory.
The limitations of this study are that it uses registry data as opposed to a randomized trial, such that differences in the demographic characteristics of the groups exist, especially differences between those treated with insulin compared with those treated with oral agents. This analysis was designed to evaluate the temporal trends of the two groups, such that this variability would be expected. While a multivariable analysis was performed to adjust for differences in demographic and angiographic characteristics, such statistical adjustments may not have fully accounted for the patient differences. However, since many diabetic patients have features of risk that would preclude their entry into randomized clinical trials of PCI, we feel that these data offer a critical, real world experience with diabetic patients undergoing PCI. Diabetic phenotyping is somewhat limited in the present data, with no information collected on duration of diabetes, decision making regarding insulin therapy, concomitance of microvascular disease complications, differentiation of type 1 versus type 2 diabetes, and no assessments were made of glycemic control. Therefore, these factors could not be factored into the analyses. Additionally, while the number of insulin-treated patients in each wave is representative of current catheterization laboratory volume, the numbers are smaller than those of oral-agent treated diabetic patients, leading to less power to detect temporal differences in this group. Lastly, important differences exist between the various oral agents presently available for the treatment of diabetes, and the data collection does not allow for analyses according to specific oral agents by drug or by class. Our study highlights the need for further investigation into therapeutic strategies to reduce intermediate-term outcomes in insulin-requiring diabetic patients undergoing PCI.
Appendix
The complete list of variables that were included in the regression models are as follows:
Insulin-treated
Death: age, chronic kidney disease, cardiogenic shock, circumstances/acuity of procedure, attempted lesion supplies collaterals, reason for revascularization, number of significant lesions, and discharged on thienopyridines, ace inhibitors, statins and beta blockers. Myocardial Infarction: history of congestive heart failure, vessel disease, attempted lesion in ostial location, circumstances/acuity of procedure, and attempted lesion supplies collaterals. CABG: cardiogenic shock, discharged on thienopyridines, and ace inhibitors. Repeat PCI: history of prior PCI and number of significant lesions. Repeat Revascularization: Circumstances/acuity of procedure and any total occlusion. MACE: age, cardiogenic shock, number of significant lesions, any total occlusion, circumstances/acuity of procedure, chronic kidney disease, and discharged on thienopyridines, and statins.
Oral Agent-treated
Death: age, history of congestive heart failure, cardiogenic shock, chronic kidney disease, reason for revascularization, attempted torturous lesion, attempted lesion supplies collaterals, circumstances/acuity of procedure, number of significant lesions, attempted calcified lesion, cancer, and discharged on thienopyridines, beta blockers and statins. Myocardial Infarction: prior PCI, attempted calcified lesion, any total occlusion, vessel disease, and peripheral vascular disease. CABG: sex, prior PCI, attempted calcified lesion, number of significant lesions, hypertension, and discharged on thienopyridines. Repeat PCI: age, sex, number of significant lesions, prior PCI, reason for revascularization, peripheral vascular disease, and presence of luminal irregularities. Repeat Revascularization: age, sex, prior PCI, reason for revascularization, number of significant lesions, attempted calcified lesion, vessel disease, chronic kidney disease, and peripheral vascular disease. MACE: sex, number of significant lesions, circumstances/acuity of procedure, prior PCI, history of congestive heart failure, attempted torturous lesion, attempted lesion with thrombus, attempted calcified lesion, attempted Class-C lesion, peripheral vascular disease, and discharged on thienopyridines, statins, and beta-blockers.
References
- 1.Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care. 2004;27:1047–1053. doi: 10.2337/diacare.27.5.1047. [DOI] [PubMed] [Google Scholar]
- 2.Abizaid A, Kornowski R, Mintz G, Hong M, Abizaid A, Mehran R, Pichard A, Kent K, Satler L, Wu H, Popma J, Leon M. The influence of diabetes mellitus on acute and late clinical outcomes following coronary stent implantation. Journal of the American College of Cardiology. 1998;32:584–589. doi: 10.1016/s0735-1097(98)00286-1. [DOI] [PubMed] [Google Scholar]
- 3.Matthew V, Gersh B, Williams B, Laskey W, Willerson J, Tilbury R, Davis B, Holmes D. Outcomes in patients with diabetes mellitus undergoing percutaneous coronary intervention in the current era: A report from the Prevention of REStenosis with Tranilast and its outcomes (PRESTO) trial. Circulation. 2004;109:476–480. doi: 10.1161/01.CIR.0000109693.64957.20. [DOI] [PubMed] [Google Scholar]
- 4.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. doi: 10.1161/01.cir.94.8.1818. [DOI] [PubMed] [Google Scholar]
- 5.Barsness GW, Peterson E, Ohman E. Relationship between diabetes mellitus and long-term survival after coronary bypass and angioplasty. Circulation. 1997;96:2551–2556. doi: 10.1161/01.cir.96.8.2551. [DOI] [PubMed] [Google Scholar]
- 6.Rozenman Y, Sapoznikov D, Gotsman M. Restenosis and progression of coronary disease after balloon angioplasty in patients with diabetes mellitus. Clin Cardiol. 2000:33. doi: 10.1002/clc.4960231207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Elezi S, Kastrati A, Pache J. Diabetes mellitus and the clinical and angiographic outcome after coronary stent placement. Journal of the American College of Cardiology. 1998;32:1866–1873. doi: 10.1016/s0735-1097(98)00467-7. [DOI] [PubMed] [Google Scholar]
- 8.Joseph T, Fajadet J, Jordan C. Coronary stenting in diabetics: immediate and mid-term clinical outcome. Catheter Cardiovasc Interv. 1999;47:279–284. doi: 10.1002/(sici)1522-726x(199907)47:3<279::aid-ccd3>3.0.co;2-p. [DOI] [PubMed] [Google Scholar]
- 9.Williams DO, Holubkov R, Yeh W, Bourassa MG, Al-Bassam M, Block PC, Coady P, Cohen H, Cowley M, Dorros G, Faxon D, Holmes DR, Jr, Jacobs A, Kelsey SF, King SB, III, Myler R, Slater J, Stanek V, Vlachos HA, Detre KM. Coinvestigators ft Percutaneous coronary intervention in the current era compared with 1985–1986. The National Heart, Lung, and Blood Institute Registries. Circulation. 2000;102:2945–2951. doi: 10.1161/01.cir.102.24.2945. [DOI] [PubMed] [Google Scholar]
- 10.Arcaro G, Cretti A, Balzano S, Lechi A, Muggeo M, Bonora E, Bonadonna RC. Insulin causes endothelial dysfunction in humans: Sites and Mechanisms. Circulation. 2002;105:576–582. doi: 10.1161/hc0502.103333. [DOI] [PubMed] [Google Scholar]
- 11.Nishimoto Y, Miyazaki Y, YT Enhanced secretion of insulin plays a role in the development of atherosclerosis and restenosis of coronary arteries: elective percutaneous transluminal coronary angioplasty in pateints with effort angina. Journal of the American College of Cardiology. 1998;1998:1624–1629. doi: 10.1016/s0735-1097(98)00428-8. [DOI] [PubMed] [Google Scholar]
- 12.Brooks MM, Frye RL, Genuth S, Detre KM, Nesto R, Sobel BE, Kelsey SF, Orchard TJ. Hypotheses, Design, and Methods for the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) Trial. The American Journal of Cardiology. 2006;97:9–19. doi: 10.1016/j.amjcard.2006.02.023. [DOI] [PubMed] [Google Scholar]
- 13.Chaitman BR, Hardison RM, Adler D, Gebhart S, Grogan M, Ocampo S, Sopko G, Ramires JA, Schneider D, Frye RL the Bypass Angioplasty Revascularization Investigation 2 Diabetes Study G. The Bypass Angioplasty Revascularization Investigation 2 Diabetes Randomized Trial of Different Treatment Strategies in Type 2 Diabetes Mellitus With Stable Ischemic Heart Disease: Impact of Treatment Strategy on Cardiac Mortality and Myocardial Infarction. Circulation. 2009;120:2529–2540. doi: 10.1161/CIRCULATIONAHA.109.913111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Stein B, Weintraub WS, Gebhart SSP, Cohen-Bernstein CL, Grosswald R, Liberman HA, Douglas JS, Jr, Morris DC, King SB., III Influence of Diabetes Mellitus on Early and Late Outcome After Percutaneous Transluminal Coronary Angioplasty. Circulation. 1995;91:979–989. doi: 10.1161/01.cir.91.4.979. [DOI] [PubMed] [Google Scholar]
- 15.Mathew V, Frye RL, Lennon R, Barsness GW, Holmes DR. Comparison of Survival After Successful Percutaneous Coronary Intervention of Patients with Diabetes Mellitus Receiving Insulin Versus Those Receiving Only Diet and/or Oral Hypoglycemic Agents. American Journal of Cardiology. 2004;93:399–403. doi: 10.1016/j.amjcard.2003.10.031. [DOI] [PubMed] [Google Scholar]
- 16.Marso S, Giorgi L, Johnson W, Huber K, Laster S, Shelton C, McAllister B, Coen M, Rutherford B. Diabetes mellitus is associated with a shift in the temporal risk profile of in hospital death after percutaneous coronary intervention: An analysis of 25,223 patients over 20 years. American Heart Journal. 2003;145:270–277. doi: 10.1067/mhj.2003.56. [DOI] [PubMed] [Google Scholar]
- 17.Freeman A, Abbott J, Jacobs A, Vlachos H, FS, Laskey W, Detre K, Williams D. Marked improvements in outcomes of contemporary percutaneous coronary intervention in patients with diabetes mellitus. Journal of Interventional Cardiology. 2006;19:475–482. doi: 10.1111/j.1540-8183.2006.00211.x. [DOI] [PubMed] [Google Scholar]
- 18.Mulukutla SR, Vlachos HA, Marroquin OC, Selzer F, Holper EM, Abbott JD, Laskey WK, Williams DO, Smith C, Anderson WD, Lee JS, Srinivas V, Kelsey SF, Kip KE. Impact of Drug-Eluting Stents Among Insulin-Treated Diabetic Patients: A Report From the National Heart, Lung, and Blood Institute Dynamic Registry. J Am Coll Cardiol Intv. 2008;1:139–147. doi: 10.1016/j.jcin.2008.02.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sabate M, Jimenez-Quevedo P, Angiolillo DJ, Gomez-Hospital JA, Alfonso F, Hernandez-Antolin R, Goicolea J, Banuelos C, Escaned J, Moreno R, Fernandez C, Fernandez-Aviles F, Macaya C for the DI. Randomized Comparison of Sirolimus-Eluting Stent Versus Standard Stent for Percutaneous Coronary Revascularization in Diabetic Patients: The Diabetes and Sirolimus-Eluting Stent (DIABETES) Trial. Circulation. 2005;112:2175–2183. doi: 10.1161/CIRCULATIONAHA.105.562421. [DOI] [PubMed] [Google Scholar]
- 20.Hermiller JB, Raizner A, Cannon L, Gurbel PA, Kutcher MA, Wong SC, Russell ME, Ellis SG, Mehran R, Stone GW, Investigators T-I. Outcomes with the polymer-based paclitaxel-eluting TAXUS stent in patients with diabetes mellitus: The TAXUS-IV trial. J Am Coll Cardiol. 2005;45:1172–1179. doi: 10.1016/j.jacc.2004.10.075. [DOI] [PubMed] [Google Scholar]