Percutaneous revascularization in diabetic is frequent and the use of drug-eluting stents (DES) is desirable, since they reduce restenosis and the need for repeat revascularization. The objective of this study was to compare the long-term outcomes of diabetic patients treated with and without DES.
MethodsA consecutive cohort of diabetic patients undergoing percutaneous coronary intervention (PCI) between January 2009 and December 2012 in a public tertiary hospital was prospectively followed-up.
ResultsNine hundred and thirty-nine diabetic patients (38.3%) treated with DES and 580 (61.7%) treated with bare metal stents (BMS) were evaluated. The rate of major adverse cardiac events (MACE) in 12.6±3.4months was greater in the BMS group (9.5% vs. 14.8%; RR, 1.56; 95% CI, 1.07-2.27; P=0.02), as well as death (2.8% vs. 6.7%; RR, 2.41; 95% CI, 1.22-4.77; P<0.01), and target vessel revascularization (3.9% vs. 7.2%; RR, 1.85; 95% CI, 1.03-3.35; P=0.04). There were no differences in the incidence of myocardial infarction (1.7% vs. 0.5%; RR, 0.30; 95% CI, 0.07-1.23; P=0.08) or stroke (1.1% vs. 0.2%; RR, 0.15; 95% CI, 0.01-1.37; P=0.07). Multivariate analysis indicated that chronic kidney disease (RR, 2.05; 95% CI, 1.40-2.98; P<0.01) and acute coronary syndrome (RR, 2.08; 95% CI 1.42-3.02; P<0.01) were the only independent predictors of MACE.
ConclusionsIn non-selected diabetic patients the long-term clinical outcome was worse for patients treated with BMS. After adjusting for confounding variables, the use of DES was not an independent predictor of reduced MACE.
Evolução Clínica de Pacientes Diabéticos Tratados porIntervenção Coronária Percutânea Utilizando Stents Com e Sem Eluição de Fármacos
IntroduçãoA revascularização percutânea de diabéticos é frequente e a utilização de stents farmacológicos (SF) é desejável, pois estes reduzem a reestenose e a necessidade de nova revascularização. O objetivo desse estudo foi comparar os resultados clínicos de longo prazo entre diabéticos tratados com e sem SF.
MétodosAnalisou-se uma coorte consecutiva de diabéticos submetidos à intervenção coronária percutânea (ICP) entre janeiro de 2009 e dezembro de 2012, em hospital terciário da rede pública. Esses pacientes foram acompanhados prospectivamente.
ResultadosAvaliamos 939 diabéticos, sendo 359 (38,3%) tratados com SF e 580 (61,7%) tratados com stents não farmacológicos (SNF). A taxa de eventos cardiovasculares adversos maiores (ECAM) em 12,6±3,4 meses foi maior no grupo SNF (9,5% vs. 14,8%; risco relativo – RR=1,56; intervalo de confiança de 95% – IC 95% 1,07-2,27; P=0,02), assim como o óbito (2,8% vs. 6,7%; RR=2,41; IC 95% 1,22-4,77; P<0,01) e a revascularização do vaso alvo (3,9% vs. 7,2%; RR=1,85; IC 95% 1,03-3,35; P=0,04). Não foram observadas diferenças na incidência de infarto do miocárdio (1,7% vs. 0,5%; RR=0,30; IC 95% 0,07-1,23; P=0,08) ou acidente vascular encefálico (1,1% vs. 0,2%; RR=0,15; IC 95% 0,01-1,37; P=0,07). A análise multivariada revelou que a doença renal crônica (RR=2,05; IC 95% 1,40-2,98; P<0,01) e a síndrome coronária aguda (RR=2,08; IC 95% 1,42-3,02; P<0,01) foram os únicos preditores independentes de ECAM.
ConclusõesEm pacientes diabéticos não selecionados, a evolução clínica tardia foi pior para os tratados com SNF. Após o ajuste das variáveis de confusão, o uso de SF não se mostrou preditor independente da redução de ECAM.
Coronary artery disease (CAD) accounts for approximately 75% of cardiovascular deaths in diabetics.1 A considerable number of these patients undergo percutaneous coronary intervention (PCI), whether in the presence of acute conditions or for the treatment of the stable disease; currently, diabetics represent approximately 30% of those percutaneously revascularized patients.2
In these patients, CAD has peculiar characteristics, either due to the association of comorbidities, which increase the clinical risk, or to the presence of more extensive and complex atherosclerotic disease, with smaller vessels and a greater propensity to thrombosis. Specifically regarding PCI, this subgroup has a greater tendency to coronary restenosis after stent implantation, and a higher incidence of other adverse cardiac events in the late evolution, such as death and acute myocardial infarction (AMI).3−5
In turn, drug-eluting stents (DES) have proven their effectiveness in reducing restenosis and the need for additional revascularization procedures when compared to bare metal stents (BMS) in both the overall population and in the diabetic subgroup.6,7 In these latter patients, drug-eluting prostheses have also been widely investigated, demonstrating high success rates, low rates of acute complications, and significant reduction of new revascularizations when compared to BMS.8
However, the results related to major cardiovascular safety outcomes, such as death, AMI, and stroke in the medium- to long-term are still contradictory.7−9 Some international registries have observed reduction of mortality and AMI in favor of DES in diabetic patients, in both early and late stages10 – benefits not observed in other series.11,12 More recent randomized trials comparing these prostheses with CABG in multivessel diabetics also failed to demonstrate favorable results for DES.13,14
In Brazil, where the majority of CAD patients are treated in the public health network, subject to curtailment of funds provided by the Brazilian Unified Health System (Sistema Único de Saúde – SUS), the acquisition of DES for universal use is impractical. In this context, the availability of these devices, with a much higher cost than that of BMS, could occur in the treatment of subgroups with more complex coronary anatomy, such as diabetics.
Thus, this study aimed to investigate the late clinical outcome in diabetic patients treated with PCI, with and without DES implantation.
METHODSFrom January 2009 to December 2011, a consecutive cohort of diabetic patients undergoing PCI in the Instituto Dante Pazzanese de Cardiologia, São Paulo, Brazil, a tertiary public hospital with extensive experience in percutaneous procedures, was analyzed.
The primary objective was to compare the rate of major adverse cardiovascular events (MACE), defined as death, stroke, AMI, and target vessel revascularization (TVR) in diabetic patients treated with and without DES. The secondary objective was to evaluate the rates of occurrence of each clinical event isolated in both groups, as well as to identify possible independent predictors of MACE.
DefinitionsDiabetes mellitus was diagnosed according to the criteria of the American Diabetes Association: presence of fasting glucose values≥126mg/dL, or glycated hemoglobin≥6.5%, or glucose≥200mg two hours after a glucose overload. Death was defined as any death, regardless of cause, and cardiac death as a death secondary to immediate cardiac causes (AMI, heart failure, and fatal arrhythmia), in addition to unwitnessed deaths and/or deaths due to unknown causes.
The diagnosis of AMI was established when periprocedural elevation of CK-MB>3 times the upper limit of normal was observed, in association with symptoms indicative of myocardial ischemia or with the presence of new Q waves on ECG after the procedure, or with changes in regional contractility on imaging exams during the in-hospital stay. Conversely, the occurrence of AMI after discharge was considered in cases of a rise and fall of troponin or CK-MB levels, in association with typical ischemic symptoms or with electrocardiographic changes (new Q waves/ST segment elevation or lowering), or with a need for urgent revascularization, or finally, in face of anatomopathological findings consistent with acute ischemia.
Stroke was defined as the occurrence of cerebral infarction (ischemic stroke) or subarachnoid and brain hemorrhage (hemorrhagic stroke), with symptoms persisting>24 hours or that resulted in death<24 hours.
TVR was defined as the performance of a new percutaneous or surgical procedure after discharge to treat the target vessel, in the presence of angina or of a clinical equivalent of ischemia, or in the presence of functional tests with abnormal results.
Stent thrombosis was defined according to the criteria of the Academic Research Consortium (ARC) as definite if confirmed by angiographic (Trombolysis in Myocardial Infarction [TIMI] flow=0 with stent occlusion by thrombus or TIMI flow 1, 2, or 3 and presence of thrombus) or anatomopathological findings; or as probable, defined as any unexplained death within the first 30 days after PCI or any AMI related to the target vessel without angiographic confirmation and in the absence of any other causes.15
ProcedureThe percutaneous procedure was performed according to the standard practice recommended by national and international guidelines.16,17 The choice of access route (femoral or radial) was at the discretion of the surgeon, as well as the use of glycoprotein IIb/IIIa.
Dual antiplatelet therapy consisted of acetylsalicylic acid at a dose of 100mg/day in patients with chronic use, and 300mg (loading dose) in previously untreated patients. Clopidogrel was the P2Y12 inhibitor most frequently used, at loading doses of 300 to 600mg before the procedure and, subsequently, 75mg/day combined with acetylsalicylic acid 100mg/ day for one year.
The decision to use DES was at the surgeon’s discretion and subject to the availability of the device at the time of the trial, considering the following scenarios: vessels≤2.5mm, lesions>20mm and in-stent restenosis; expectation of non-adherence to dual antiplatelet therapy at one year; presence of hemorrhagic diathesis, or of hematologic diseases, or prior and recent history (< six months) of digestive, urological, or other bleeding that could prevent the use of dual antiplatelet therapy for the recommended time.
Clinical follow-upThe clinical follow-up was prospective, and conducted by clinical visits with review of records, since this tertiary hospital has an outpatient setting for monitoring patients undergoing PCI. Contacts by phone, email, telegram, or reports from physicians who referred patients to this service were used.
Statistical analysisContinuous variables were analyzed as means±standard deviations and compared using the Mann Whitney test. Categorical variables were compared using the chi-squared or Fisher’s exact test, when appropriate.
Analysis of survival free of event was obtained using Kaplan-Meier curves, and comparisons were performed using the log-rank method.
Multivariate regression models were derived from Cox for MACE, and the variables considered for inclusion in the multivariate models were age, gender, insulin use, chronic renal failure (CRF), acute coronary syndrome (ACS), prior PCI, number of vessels treated, use of periprocedural DES, and AMI. The significance level for permanence in the model was adjusted to 0.01.
To perform the analysis, the SPSS, version 20.0, was used. P-values<0.05 were considered significant.
RESULTS939 consecutive diabetic patients were analyzed; 359 patients (38.3%) were treated with DES and 580 patients (61.7%) with BMS. Mean follow-up in both groups was 12.9±3.6months (1–13 months) and 12.4±2.5months (1–14 months), respectively, and 80% of patients had a minimum follow-up of one year.
Demographic and clinical characteristics of these patients revealed a higher proportion of patients with risk factors for atherosclerosis, previous AMI, heart failure, previous PCI, and stable clinical status in diabetic patients treated with DES, whereas patients with chronic kidney disease and ACS were more frequent in the group treated with BMS. Regarding the pharmacological treatment before the procedure, patients treated with DES most often used beta-blockers, statins, and insulin (Table 1).
Clinical and demographic characteristics of patients
| Characteristics | DES(n=359) | BMS(n=580) | P-value |
|---|---|---|---|
| Age, years | 63.0±10.6 | 63.2±10.5 | 0.80 |
| Female, n (%) | 145 (40.4) | 219 (37.8) | 0.42 |
| Hypertension, n (%) | 337 (93.9) | 533 (91.9) | 0.30 |
| Dyslipidemia, n (%) | 314 (87.5) | 453 (78.1) | <0.01 |
| Smoking, n (%) | 204 (56.8) | 62 (10.7) | <0.01 |
| Prior PCI, n (%) | 104 (29.0) | 64 (11.0) | <0.01 |
| Previous CABG, n (%) | 35 (9.7) | 88(11.0) | 0.17 |
| Prior AMI, n (%) | 163 (45.4) | 224 (38.6) | 0.04 |
| Prior stroke or TIA, n (%) | 13 (3.6) | 18 (3.1) | 0.70 |
| Family history of CAD, n (%) | 73 (20.3) | 52 (9.0) | <0.01 |
| COPD, n (%) | 4 (1.1) | 10 (1.7) | 0.58 |
| CRF not requiring dialysis, n (%) | 27 (7.5) | 188 (32.4) | <0.01 |
| POAD, n (%) | 16 (4.5) | 16 (2.8) | 0.19 |
| Heart failure, n (%) | 127 (35.4) | 9 (1.6) | <0.01 |
| Clinical status, n (%) | <0.01 | ||
| Acute coronary syndrome | 36 (10.0) | 174 (30.0) | |
| Stable angina | 218 (60.7) | 268 (46.2) | |
| Asymptomatic | 105 (29.2) | 138 (23.8) | |
| Pharmacological treatment, n (%) | |||
| Insulin | 106 (29.5) | 58 (10.0) | <0.01 |
| Oral hypoglycaemic | 313 (87.2) | 520 (89.7) | 0.24 |
| Acetylsalicylic acid | 357 (99.4) | 568 (97.9) | 0.09 |
| Statins | 357 (99.4) | 537 (92.4) | <0.01 |
| ACEI/BRA | 336 (93.6) | 536 (92.4) | 0.51 |
| Beta blockers | 343 (95.5) | 477 (80.5) | <0.01 |
DES=drug-eluting stents; BMS=bare metal stents; PCI=percutaneous coronary intervention; CABG=coronary artery bypass graft surgery; AMI=acute myocardial infarction; TIA=transient ischemic attack; CAD=coronary artery disease; COPD=chronic obstructive pulmonary disease; CRF=chronic renal failure; POAD=peripheral obstructive arterial disease; ACEI/ARB=angiotensin-converting enzyme inhibitors/ angiotensin receptor blockers.
In the DES group, the left anterior descending artery was the most frequently target vessel (50.7%), and multivessel PCI was more frequent (25.3% vs. 5.7%; P<0.01), resulting in a ratio target vessel per patient/stent per patient also significantly higher in this group. A significantly higher percentage of bifurcation injuries and restenotic lesions was observed in patients treated with DES (Table 2). Zotarolimus- (65%), sirolimus- (31%), paclitaxel- (2.5%), and novolimus (1.5%) -eluting stents were used.
Angiographic and procedural characteristics
| Characteristics | DES(n=359) | BMS(n=580) | P-value |
|---|---|---|---|
| Target vessels, n (%) | <0.01 | ||
| Left anterior descending artery | 182 (50.7) | 186 (32.1) | |
| Left circumflex artery | 64 (17.8) | 103 (17.7) | |
| Right coronary artery | 61 (17) | 200 (34.5) | |
| Left main coronary artery | 1 (0.3) | 9 (1.6) | |
| Saphenous graft, n (%) | 14 (3.9) | 37 (6.4) | 0.10 |
| Bifurcation, n (%) | 79 (22) | 57 (9.8) | <0.01 |
| Total occlusion, n (%) | 10 (2.7) | 17 (2.9) | 0.89 |
| Restenosis, n (%) | 61 (17) | 15 (2.6) | <0.01 |
| Number of vessels treated per patient | 1.5±0.22 | 1.0±0.50 | <0.01 |
| Treatment of two or more vessels, n (%) | 91 (25.3) | 33 (5.7) | <0.01 |
| Number of stents per patient | 1.48±0.5 | 1.07±0.4 | <0.01 |
| Stent caliber, mm | 2.90±4.2 | 3.10±4.3 | 0.11 |
| Stent length, mm | 19.6±9.3 | 20.5±6.6 | 0.09 |
DES=drug-eluting stents; BMS=bare metal stents.
In relation to in-hospital outcomes, the clinical success rate of PCI was similar in both groups (97.2% vs. 95.6%; P=0.22), with no differences in the occurrence of death (0% vs. 0.7%; P=0.30) and stroke (0% vs. 0.2; P=0.99). The group of patients treated with DES showed a higher incidence of elevation of periprocedural myocardial injury markers when compared to those treated with BMS (8.6% vs. 4.0%; P<0.01).
In a clinical follow-up of 12.6±3.4months, the rate of MACE was higher in the group receiving BMS (9.5% vs. 14.8%; RR=1.56; 95% CI: 1.07-2.27; P=0.01). Death by the end of clinical follow-up was also higher in the BMS group (2.8% vs. 6.7%; RR=2.41; 95% CI: 1.22-4.77; P<0.01), as well as death of cardiac origin (1.7% vs. 3.7%; RR=2.78; 95% CI: 1.16-6.68; P=0.01) and TVR (3.9% vs. 7.2%; RR=1.85; 95% CI: 1.03-3.35; P=0.03). Moreover, there were no significant differences between the two groups in the incidence of AMI (1.7% vs. 0.5%; RR=0.30; 95% CI: 0.07-1.23; P=0.07) and stroke (1.1% vs. 0.2%; RR=0.15; 95% CI: 0.01-1.37; P=0.07) in the same trial period (Table 3).
Cardiovascular events at the end of one year
| Event | DES(n=359) | BMS(n=580) | RR | 95% CI | P-value |
|---|---|---|---|---|---|
| MACE, n (%) | 34 (9.5) | 86 (14.8) | 1.56 | 1.07-2.27 | 0.01 |
| Death, n (%) | 10 (2.8) | 39 (6.7) | 2.41 | 1.22-4.77 | <0.01 |
| Cardiac death, n (%) | 6 (1.7) | 27 (3.7) | 2.78 | 1.16-6.68 | 0.01 |
| Stroke, n (%) | 4 (1.1) | 1 (0.2) | 0.15 | 0.01-1.37 | 0.07 |
| AMI, n (%) | 6 (1.7) | 3 (0.5) | 0.3 | 0.07-1.23 | 0.07 |
| TVR, n (%) | 14 (3.9) | 42 (7.2) | 1.85 | 1.03-3.35 | 0.03 |
DES=drug-eluting stents; BMS=bare metal stents; RR=relative risk; 95% CI=95% confidence interval; MACE=main adverse cardiovascular events; AMI=acute myocardial infarction; TVR=target vessel revascularization.
The results of the Kaplan Meier curves regarding survival free of MACE, death, death of cardiac origin, and TVR are detailed in Figure 1, revealing significant differences in favor of DES.
In relation to definite or probable stent thrombosis, this complication had a very low incidence in both groups (0.83% vs. 0%; P=0.06) during the relatively short follow-up period.
A multivariate analysis of pre-specified variables identified only chronic renal disease (CRD) (RR=2.05; 95% CI: 1.40-2.98; P<0.01) and ACS (RR=2.08; 95% CI: 1.42-3.02; P<0.01) as independent predictors for occurrence of MACE in the late evolution of diabetes (Table 4). The use of DES was shown to be a predictor of MACE in this group (RR=1.35; 95% CI: 0.87-2.10; P=0.17).
Predictors of major adverse cardiovascular events in diabetic patients
| RR | 95% CI | P-value | |
|---|---|---|---|
| Age | 1.10 | 0.99-1.03 | 0.08 |
| Gender | 1.00 | 0.69-1.45 | 0.97 |
| Insulin use | 0.96 | 0.60-1.54 | 0.87 |
| Chronic renal failure | 2.05 | 1.40-2.98 | <0.01 |
| Acute coronary syndrome | 2.08 | 1.42-3.02 | <0.01 |
| Previous PCI | 0.80 | 0.48-1.33 | 0.40 |
| Number of vessels treated | 0.74 | 0.44-1.25 | 0.27 |
| Heart failure | 0.61 | 0.32-1.13 | 0.11 |
| DES use | 1.35 | 0.87-2.10 | 0.17 |
| Periprocedural AMI | 1.50 | 0.80-2.92 | 0.19 |
RR=relative risk; 95% CI=95% confidence interval; PCI=percutaneous coronary intervention; DES=drug-eluting stent; AMI=acute myocardial infarction.
In the present study, which included a cohort of untreated, non-selected diabetic patients for PCI in a tertiary public hospital, a reduction of MACE, death, death due to a cardiac cause, and TVR in the group treated with DES was observed. However, after adjusting for confounding variables, the use of DES did not modify the occurrence of combined adverse events. Chronic renal disease and ACS were the only prespecified variables predictors of poorer clinical outcome.
Currently, the technological development of materials available for the completion of PCI, together with the vast experience of the surgeons in the management of high-risk patients and the availability of DES, allowed for the expansion of indications for percutaneous treatment, including patients with CAD of high clinical and anatomical complexity, such as diabetics.16,17
In the present study, it was observed that diabetics treated with DES corroborate this scenario, in that these patients presented more challenging angiographic characteristics, often involving the treatment of left anterior descending artery (with a greater area of myocardium at risk for ischemia), lesions located in bifurcations, and coronary restenoses. Acknowledging the complexity of the lesions treated, a greater number of vessels was approached, resulting in a significantly higher ratio of vessels treated per number of stents in the group of diabetics revascularized with DES. Such features require the use of more efficient prostheses, with elution of drugs that effectively inhibit neointimal proliferation, in order to reduce the recurrence of obstructions due to coronary restenosis, and, hence, the need for new revascularizations in patients already predisposed to these events.
In the literature, several trials comparing the use of DES vs. BMS in diabetic and nondiabetic patients have clearly demonstrated the superiority of the former device in significantly reducing coronary restenosis and, therefore, in a new approach to recurrent lesions, reducing the exposure of patients to the inherent risks of a new procedure, whether percutaneous or surgical.7,11,12
It is also interesting to note, in the present population, the greater proportion of patients using insulin in the DES group (29.5% vs. 10.0%; P<0.01), indicating a greater severity and duration of diabetes. It is known that, in general, these patients may experience a more diffuse atherosclerotic disease and an intense inflammatory process of the coronary arteries and vascular beds in general, which result in greater predisposition to restenosis, arterial thrombosis, and AMI.18−20 A recent brazilian publication21 investigated in-hospital outcomes of nearly 2,000 diabetic patients from the Angiocardio Registry undergoing PCI, divided among those treated with insulin (21%) and those using oral hypoglycemic agents. The researchers found that, despite the higher percentage of women and of chronic kidney disease in the insulin subgroup, the major cardiovascular outcomes were similar in both groups, and the insulin treatment was not an independent predictor for the occurrence of adverse events in the in-hospital phase.
A register held by the National Heart, Lung, and Blood Institute in 2008, involving over 2,500 diabetic patients with and without insulin use treated with DES and BMS, revealed that the use of DES reduced the risk of repeat revascularization and of safety outcomes such as death and AMI. However, the reduction in mortality and in AMI was restricted to diabetics not treated with insulin.21 It is noteworthy that, in the present study, the use of insulin was not a predictor of adverse cardiovascular events in the long-term.
Regarding in-hospital results, the present findings demonstrated equivalent procedural success rates between the two types of stents, as well as the rates of occurrence of death and stroke, confirming the safety and efficacy of DES in the earliest phase after PCI. The higher incidence of elevated markers of peri-PCI myocardial injury probably relates to the greater anatomical complexity of the lesions treated in the DES subgroup, with a larger number of approached vessels and of implanted stents.
Regarding stent thrombosis (definite or probable, according to ARC criteria), this occurrence was very low in both groups (0.83% vs. 0%) in a follow-up period of 12.6±3.4months. These results are similar to the ESSENCEDIABETES II trial conducted by Park et al.,22 who observed rates of subacute and late thrombosis in diabetic patients undergoing implantation of zotarolimus- and sirolimuseluting stents<1% after 12 months of evolution.
Regarding the primary outcome of the trial, it was demonstrated that the occurrence of MACE was significantly lower in diabetic patients treated with DES, and the same occurred with death due to a cardiac cause and TVR. At first glance, these results appear to demonstrate that the use of DES in diabetics is related to better clinical outcome in the long-term, which requires caution in interpreting the data. When comparing the present results to those of other trials, it was observed that the majority of them presented no significant reduction in mortality or in other major cardiovascular outcomes in patients treated with DES, whether or not they were diabetic.23,24
Some trials have observed a reduction of stent thrombosis and AMI, but without a consistent reduction in the mortality rates with the use of this type of prosthesis, even with the second- or third-generation, so called due to their new metal platforms, more biocompatible and less thrombogenic polymers, and drugs with higher power of neointimal inhibition.8,25
Other observational trials and international registries, involving large numbers of diabetic and nondiabetic patients treated with DES, observed a reduction of ominous outcomes such as death and AMI, as well as in the rates of new revascularizations.10,26,27 However, these results deserve reflection and must be viewed with caution, due to a possible selection bias and the relatively short duration of the clinical follow-up. In the present study, for instance, the multivariate analysis of the predictors for the occurrence of MACE took into consideration chronic renal disease and ACS, a fact fully demonstrated in the literature.28 Coincidentally, these variables were significantly more present in patients treated with BMS, which was a disadvantage for this group and could explain the better results obtained in patients undergoing DES implantation. After adjustment of the multivariate analysis, no significant differences were observed in the rates of MACE between patients treated with and without DES at the end of one year of clinical follow-up (RR=1.35; 95% CI: 0.87-2.10; P=0.17).
The choice of BMS implantation in diabetic patients with ACS and renal failure – without doubt, patients at high risk for adverse events – can be questioned. Some possible explanations are: (1) DES are not routinely available in most public hospitals linked to the SUS; (2) the safety and efficacy of BMS for the treatment of patients with acute ischemic conditions, requiring urgent or emergent PCI, is equivalent to the safety and efficacy for DES, since in such cases the priority is the treatment of the culprit lesion, in order to reduce acutely events such as death and AMI; (3) in patients with ACS, the option of BMS use may be preferable, due to issues related to a lack of detailed knowledge of the patient’s clinical history in terms of adherence to dual antiplatelet therapy, or even to the necessity of an early discontinuation of this therapy, due to noncardiac comorbidities; and (4) in chronic renal patients, the use of BMS can be defended in view of a potential need for invasive non-cardiac procedures, such as the implantation of catheters or performing arteriovenous fistulas for dialysis and the possibility of a kidney transplant – situations in which a dual antiplatelet therapy may predispose to major bleeding.
LimitationsThis was a nonrandomized trial, and therefore subject to selection bias; the follow-up period was relatively short for the analysis of the real differences in the occurrence of adverse cardiovascular events in diabetic patients; and no information regarding adhesion and duration of dual antiplatelet therapy was obtained, factors that could affect the incidence of adverse events, especially in patients treated with DES and in those with pre-PCI ACS.
CONCLUSIONSIn a population consisting of unselected diabetic patients treated by PCI, the use of DES was associated with significantly lower rates of major cardiovascular outcomes compared to BMS. However, after adjusting for confounding variables, the use of DES was not an independent predictor of major adverse cardiovascular events in the long-term.
CONFLICTS OF INTERESTThe authors declare no conflicts of interest.
The authors would like to thank the team of statisticians (John Italo França and Hayala Cristina de Souza) for their careful data analysis; Renata Viana, Patricia Paiva, and Mayra Yagihara, from the Clinical Research Center, Section of Coronary Angioplasty, staff; and Mr. Reginaldo Barreto, for the availability, organization, and assistance in the collection of information from the database of the Department of Invasive Cardiology of the Instituto Dante Pazzanese de Cardiologia.