Percutaneous coronary intervention in patients with multivessel disease in Brazil

Introduction: Percutaneous coronary intervention (PCI) in patients with multivessel disease is associated with a lower success rate and a higher incidence of complications. The results of this treatment in Brazil are poorly studied. The objective of this study was to analyze the results of PCI performed in patients with multivessel disease, which were registered in the National Center for Cardiovascular Interventions (CENIC) registry. Methods: Complete electronic records of procedures performed in patients with multivessel disease from 2006 to 2016 were analyzed. Results: A total of 191,127 PCI were submitted to the CENIC registry in the study period, including 80,093 (45.3%) cases classified as multivessel disease. The patients were predominantly male (67.5%) with stable (49.6%), twovessel disease (65%). Type B2/C lesions were present in 70.8% of the cases, with a mean of 1.6 artery treated per patient and 1.7 stent implanted per procedure, 71.6% of which were bare-metal stents. The success rate was 96%. During hospitalization, the occurrence of major adverse cardiac events was 1.5%, and death was the most frequent complication (1.2%). Independent predictive factors of death were age, sex, diabetes mellitus, previous infarction, the extent of coronary disease, the use of glycoprotein IIb/IIIa inhibitors, acute coronary syndrome, emergency interventions and procedures undertaken in the 2006-2008 period. Conclusions: Multivessel PCI has a high success and a low in-hospital complication rate. The identification of characteristics associated with a poor prognosis can be useful for stratification and for the selection of the most appropriate treatment strategy. © 2017 Sociedade Brasileira de Hemodinâmica e Cardiologia Intervencionista. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Autor para correspondência: Avenida do Contorno, 9.530, Barro Preto, CEP 30110-934, Belo Horizonte, MG, Brasil. E-mail: falchettoeduardo@gmail.com (E.B.Falchetto). A revisão por pares é de responsabilidade da Sociedade Brasileira de Hemodinâmica e Cardiologia Intervencionista. Intervenção coronária percutânea em pacientes multiarteriais no Brasil Eduardo Belisario Falchetto*, Jamil Abdalla Saad, Eduardo Kei Marquezini Washizu, Frederico Lemos de Almeida, Efraim Lunardi Flam, Ari Mandil Serviço de Cardiologia Intervencionista do Hospital Felício Rocho, Belo Horizonte, MG, Brasil 37 E. Falchetto et al. / Rev Bras Cardiol Invasiva. 2017;25(1-4):36-41


Introduction
Myocardial revascularization is a treatment modality for patients with coronary artery disease that aims to alleviate symptoms and/ or improve prognosis.The annual risk of cardiac death or non-fatal acute myocardial infarction (AMI) is 4 to 6% among stable patients with moderate or severe myocardial ischemia. 1Similarly, patients with acute coronary syndrome ACS, who exhibit high-risk characteristics, have a better prognosis when subjected to an invasive strategy, aiming at a revascularization procedure. 2When revascularization is indicated, the potential benefits, risks of complications and the patient's preferences must be considered to determine whether to proceed with surgical or percutaneous revascularization.In patients with multivessel coronary disease, this decision is more complex given that results from randomized comparative trials are not generalizable.Furthermore, the safety and efficacy of both of these treatment modalities in brazil are still poorly studied, and the choice is frequently arbitrary. 3o find information for collaborative decision-making, clinical records are very important because their data complement the findings of multicenter randomized trials.Therefore, the results can be assessed for reproducibility on a large scale in other populations and under various circumstances, such as resource availability. 4In 1991, the brazilian Society of Hemodynamics and Interventional Cardiology (SbHCI, acronym in Portuguese) implemented the National Center for Cardiovascular Interventions (CENIC, acronym in Portuguese) as a registry of percutaneous coronary interventions (PCIs) performed in brazil.It was the first national initiative to assess the development and evolution of PCI and has stimulated continuous improvements in the quality of patient care.
The objective of this study was to assess temporal trends in demographic, clinical and anatomical characteristics and in-hospital results of PCI in patients with multivessel disease performed in interventional cardiology centers in brazil that were reported to the CENIC and to evaluate the associations of clinical and anatomical variables with hospital outcomes.

Methods
The process of data acquisition and storage in the CENIC registry has been described elsewhere. 5Sending data to CENIC is voluntary and consists of completing standardized digital forms regarding clinical and angiographic aspects of the procedure, immediate results and complications during hospitalization.Data collection was initiated in 1992.
The present study addressed PCI performed in patients with multivessel disease from 2006 to 2016.No reported procedure was excluded.Multivessel coronary disease was considered when a lesion obstructed > 50% of at least two major epicardial coronary arteries or their side branches.Patients who exhibited lesions obstructing > 50% of two or more side branches of the same artery were classified as one-vessel disease.Angiographic variables were analyzed by visual estimation by the operators, and definitions were based on the Guideline for Percutaneous Coronary Intervention and Adjunct Diagnostic Methods in Interventional Cardiology of the SbHCI. 6emporal trends of the studied variables were assessed by dividing the analysis into three periods: 2006-2008, 2009-2011 and  2012-2016.The procedure was considered successful when a residual lesion obstructed < 50% of an artery in non-stent procedures and < 30% in those that endoprostheses were used.Major adverse cardiac events included death, AMI or emergency revascularization.
Statistical analysis was performed with the Statistical Package for the Social Sciences (SPSS), version 19.Continuous variables were compared with the Chi-square test and, when necessary, the like-lihood-ratio test.Categorical variables were compared by analysis of variance (ANOVA), with bonferroni correction in multiple comparisons.The influence of the variables of interest with respect to mortality was assessed with simple logistic regression.In multiple logistic regression models, the forward selection method determined the independent variables.Variables with high rates of missing data (Killip class, left ventricular dysfunction and presence of collateral circulation) were not included in the multivariate analysis.Statistical significance was considered at 5% (p value < 0.05).

Results
In the studied period (2006-2016), CENIC received data for 191,127 PCI performed in 176,780 patients.Of these, 80,093 (45.3%) cases were classified as multivessel disease and a total of 86,153 procedures were performed, constituting the study sample.Six hundred seventeen SbHCI centers submitted forms, corresponding to 55.8% of all centers with registered authorization to perform PCI at the time of the study.
Table 1 shows the patients' clinical characteristics.There was a discrete but significant increase in age over time (p < 0.0001), and male sex was predominant.The prevalence of smoking decreased progressively (p < 0.0001), and other risk factors, such as arterial hypertension, diabetes mellitus and dyslipidemia, increased (p < 0.0001).For past history, approximately 20% of the patients had previous AMI and PCI, with a progressive increase in the latter (p < 0.001).
Table 2 shows the angiographic characteristics of the procedures.Regarding the extent of coronary artery disease, patients with twovessel disease were predominant (65%), and the left anterior descending artery was the most commonly addressed vessel (38.6%).The ratio of type b 2 /C lesions increased, reaching almost 90% of the lesions treated in the 2012-2016 period, with a paradoxical decrease in the prevalence of unfavorable anatomical characteristics.The mean number of treated arteries per patient was 1.6, and coronary stents were used in approximately 95% of the cases.The success rate of the interventions was high and increased over time (Table 3).
During hospitalization, the rate of major adverse cardiac events was 1.5%, which decreased after the first triennium, and death was the most frequent complication (Table 4).As shown by simple logistic regression, the likelihood of death increased by 5% per year of age, and it was highest among women (Table 5).Multivessel disease, especially three-vessel disease, acute coronary syndrome and left ventricular dysfunction were significantly related to death.The variables that remained predictive of death after multivariate analysis were age, sex, diabetes mellitus, previous AMI, the extent of coronary artery disease, the use of glycoprotein IIb/IIIa inhibitors, acute coronary syndrome, emergency interventions and procedures performed in the 2006-2008 triennium (Table 6).

Discussion
][9][10][11][12][13][14] The studied sample was mainly composed of men of advanced age with multiple comorbidities.Compared to the demographic profiles of patients in 1996 and 1997, the studied population was older (a mean of 5 years) and had a higher probability of previous PCI. 15Furthermore, despite the increasing complexity of clinical and angiographic profiles among treated patients, the rate of complications significantly decreased between 2006 and 2016, especially for death and AMI.
7][18] In several national and international registries,     women constitute approximately one-third of treated patients and exhibit a poorer prognosis than men, possibly due to a higher prevalence of small vessels and the number of comorbidities. 19ichtman et al. also found that the influence of sex does not depend on age and that younger women have an even worse prognosis. 13opes et al. studied the results of PCI in the country with respect to sex by consulting data from the CENIC registry between 1999 and 2007. 20In their analysis, the women were older and had a higher prevalence of diabetes mellitus, had more favorable angiographic characteristics and received implanted stents with smaller diameters more often.The authors showed that women exhibited higher rates of death (1.20% vs. 0.79%; p < 0.0001) and non-fatal AMI (0.54% vs. 0.41%; p < 0.0001) compared to men.The authors concluded that these findings were probably due to more advanced age, smaller vessels and worse vascular remodeling.
In contrast to international registries and studies, data regarding the results of PCI in brazil are scarce.Therefore, advancements are necessary for the development and growth of this field in brazil.Piegas et al. assessed coronary procedures performed between 2005 and 2008 by consulting the unified Health System's Informatics Department database (DATASuS, acronym in Portuguese). 21The 166,514 procedures retrieved within the above period, with no mention of the extent of coronary artery disease, resulted in a reported in-hospital mortality of 2.33%, a mortality of 0.86% in elective cases, and a mortality of 3.25% in emergency procedures.
The first reported series using the CENIC registry assessed the procedures performed immediately after the beginning of data collection, between 1992 and 1993. 5A total of 19,305 balloon PCI were assessed, and in-hospital mortality was 1.8%.Subsequently, cases from 1996 to 1997, immediately after the introduction of coronary stents, were analyzed. 7Of the 25,854 analyzed procedures, 38.5% were performed on patients with multivessel disease, and stents were used in 36% of the cases.The reported mortality was 1.4%.Therefore, this angiographic profile reflects an increase in the number of interventions.In 1996/97, patients with multivessel disease accounted for 38.5% of the sample.In the present study, these patients accounted for 45.3% of the sample.
De Paula et al. published in-hospital results of PCI performed in six brazilian national centers. 22A total of 1,239 patients were consecutively included in 2009.In 61.5% of the cases, the indication for the procedure was acute coronary syndrome.Hospital mortality was 2.3%.Among patients with ST-segment elevation myocardial infarction, the mortality rate was 6.1%; this rate was 2.4% among those with non-ST-segment elevation myocardial infarction and 0.2% among those with a stable presentation.In brazil, the results of PCI performed in eight regional centers were analyzed by Lodi-Junqueira et al. 23 The authors identified the presence of multivessel coronary disease as a predictive factor of hospital death (55.7% of patients had multivessel disease, with 1.9% of deaths among patients with two-vessel disease and 3.9% among those with three-vessel disease).
We found that the mortality rate of PCI in brazil ranged between 1.2% and 2.3%, and it was higher among patients with acute coronary syndrome.These findings are reproducible compared with international data.However, there are also differences.Stent implantation was the standard technique (96.4% of procedures), but the use of drug-eluting stents was low.Although decreasing, the difference in the use of this device in brazil in relation to the world practice, estimated between 80 to 90%, is still unsatisfactory. 9,14The lower use of drug-eluting stents in brazil is related to the current cost of endoprostheses.Considering the possible decline in the rate of cardiovascular events through the use of new-generation drug-eluting stents, [24][25][26][27][28][29] these devices should be implemented more frequently in our country, especially among patients with multivessel disease.
There are some limitations to the use of mortality rates as indicators of quality of care in interventional cardiology.PCI-related mortality is determined as death caused by complications of the procedure, such as perforation or dissection, stent thrombosis, bleeding and kidney failure.Valle et al. found a mortality rate of 1.54% among 5,520 patients undergoing PCI in a single center between 2001 and 2009.The cause of death was determined as secondary to complications of the procedure in only 8.2% of the cases.In the remaining cases, death was due to pre-existing causes or those acquired after the procedure (heart failure, arrhythmias or neurological complications). 17Therefore, predicting cardiovascular outcomes after PCI is more accurate when the general and neurological conditions of a patient are considered in addition to baseline cardiovascular variables given that more than half of deaths were not due to cardiac causes or to the procedure itself. 13he results of PCI in patients with multivessel disease are affected by many factors.Improvements in treatment are related to advancements in technology and antithrombotic therapy, more efficient pharmacological agents and intravascular imaging, greater operator experience and recognition of predictors of complications prior to procedures. 30In fact, international registries show reduced rates of adverse cardiovascular outcomes after the introduction of these improvements, which have remained stable over time. 24ur study has some limitations.The voluntary nature of the CENIC registry and the absence of inclusion criteria at each center may reflect a selection bias.The data were not audited, and the SbHCI does not have a primary lab to perform independent angiographic analyses.Consequently, the patients were not stratified according to an anatomical severity score.Data regarding adjunctive pharmacotherapy with PCI were not analyzed.Similarly, the lengths of hospitalization and the rates of vascular complications, which are frequent in this population, are unknown.The results could be more accurately analyzed if they were stratified according to the type of hospital (secondary or tertiary) and the experience of the operator/ center considering that greater experience is positively correlated with lower complication rates. 9,31

Conclusions
Data from the CENIC registry show that percutaneous coronary intervention as a treatment modality for patients with multivessel disease is safe, efficient and reproducible.National results should be analyzed beyond hospitalization to obtain long-term data.The identification of characteristics associated with a poor prognosis can be useful for stratification and for the selection of the most appropriate treatment strategy.

Table 3
Characteristics of percutaneous coronary interventions

Table 4
Clinical outcomes during hospitalization * Comparison between quartiles.AMI: acute myocardial infarction; MACE: major adverse cardiac events.

Table 6
Predictive factors of in-hospital death (multivariate analysis)