Influence of Angiotensin-Converting-Enzyme Gene Polymorphism on Echocardiographic Data of Patients with Ischemic Heart Failure

Background Association between angiotensin-converting-enzyme (ACE) gene polymorphisms and different clinical and echocardiographic outcomes has been described in patients with heart failure (HF) and coronary artery disease. Studying the genetic profile of the local population with both diseases is necessary to assess the occurrence of that association. Objectives To assess the frequency of ACE gene polymorphisms in patients with ischemic HF in a Rio de Janeiro population, as well as its association with echocardiographic findings. Methods Genetic assessment of I/D ACE polymorphism in association with clinical, laboratory and echocardiographic analysis of 99 patients. Results The allele frequency was: 53 I alleles, and 145 D alleles. Genotype frequencies were: 49.5% DD; 47.48% DI; 3.02% II. Drug treatment was optimized: 98% on beta-blockers, and 84.8% on ACE inhibitors or angiotensin-receptor blocker. Echocardiographic findings: difference between left ventricular diastolic diameters (ΔLVDD) during follow-up: 2.98±8.94 (DD) vs. 0.68±8.12 (DI) vs. -11.0±7.00 (II), p=0.018; worsening during follow-up of the LV systolic diameter (LVSD): 65.3% DD vs. 19.0% DI vs. 0.0% II, p=0.01; of the LV diastolic diameter (LVDD): 65.3% DD vs. 46.8% DI vs. 0.0% II, p=0.03; and of the LV ejection fraction (LVEF): 67.3% DD vs. 40.4% DI vs. 33.3% II, p=0.024. Correlated with D allele: ΔLVEF, ΔLVSD, ΔLVDD. Conclusions More DD genotype patients had worsening of the LVEF, LVSD and LVDD, followed by DI genotype patients, while II genotype patients had the best outcome. The same pattern was observed for ΔLVDD.


Introduction
Heart failure is a complex syndrome, and there is strong evidence that gene polymorphisms play an important role in its pathophysiology and progression. 1,2 In addition, neuro-hormonal activation has a role in heart failure course. Angiotensin-converting-enzyme (ACE), a key player in the renin-angiotensin-aldosterone system, is essential to heart function regulation. 3,4 Angiotensin-converting-enzyme gene polymorphisms (ACEGP) have been associated with heart failure prognosis, and several studies have shown the association of D allele and DD genotype with worse echocardiographic outcomes in patients with systolic dysfunction. 5,6 The DD genotype is associated with higher frequency of acute myocardial infarction in several populations, in addition to major ischemic defects after occlusion of a coronary artery. 7,8 The inclusion criteria were as follows: age over 18 years; heart failure diagnosis according to the Framingham criteria; left ventricular ejection fraction (LVEF) <50% on echocardiography, assessed with the Simpson's method at any time of clinical follow-up; CAD demonstrated on coronary angiography with evidence of significant obstructive disease (≥ 75%) 11 or previous acute myocardial infarction or previous percutaneous coronary angioplasty or surgical myocardial revascularization. The exclusion criteria were as follows: unavailable or inappropriate medical records; non-ischemic etiology of heart failure; and loss to follow-up by the end of the study.
This study was approved by the Ethics Committee of the University, being included in the Brazilian system of Ethics in Research. All patients provided written informed consent before the beginning of the study, which abided by the principles of the Declaration of Helsinki.
The procedures of data analysis and collection from the medical records were blind to the researchers. The genotype was known only at the end of the review of the medical record; therefore, no physician knew that information at the time of the medical visits.
Skin color was observed by the physician, the individuals being classified as white, black, mixed or other (yellow/Asian).

Echocardiographic variables
All patients underwent at least two echocardiographic assessments at different times, undergoing new tests at the clinical discretion of the medical team. Data of the first echocardiography and of another conducted at the end of the follow-up were collected, in two device models, GE Vivid 3 and HD7 Philips, with a 2.75-MHz transducer, the test being performed by a physician blinded to the patients' genotypes.
The following echocardiographic data were assessed: LVEF (Simpson's method); left ventricular systolic and diastolic diameters (LVSD and LVDD, respectively). The methodology to measure echocardiographically the ventricular diameters and muscle thickness followed the rules of the American Society of Echocardiography.
Echocardiographic outcomes were assessed by calculating the differences between the final and initial values of the parameters measured (LVEF, LVSD and LVDD) as follows: variation of the left ventricular ejection fraction (ΔLVEF), variation of the LVSD (ΔLVSD), and variation of the LVDD (ΔLVDD). In addition, objective improvement or worsening of those parameters during follow-up was assessed, with the creation of the following variables: FΔLVEF, for LVEF improvement or worsening during follow-up; FΔLVSD and FΔLVDD, for improvement or worsening of LVSD and LVDD, respectively, during follow-up.

Genetic analysis
Blood samples were collected and stored at 5-15 o C for genetic analysis with DNA extraction, according to the saltingout method, genotyping with polymerase chain reaction, and later classification as DD, DI or II genotypes.

Statistical analysis
All data obtained were analyzed with an IBM PC computer by using the SPSS for Windows statistical program, version 17.0 of 2008. The following tests were used: Tukey, chi-square (χ 2 ), analysis of variance (F) and Pearson correlation. The statistical significance level adopted was 5%. Categorical variables were presented as absolute values and their respective percentages. Continuous variables were presented as mean ± standard deviation. To assess the distribution of the variables studied, skewness analysis was used. Gene and haplotype frequencies were tested for Hardy-Weinberg equilibrium, using ARLEQUIN software, version 2000.

Weight of D allele
In addition to categorizing ACE genotypes into three groups (DD, DI and II) and assessing their relationship with the other variables, an analysis model was elaborated to test the isolated impact of each D allele on echocardiographic findings. Thus, a mathematical model was created to simulate the behavior of the ACE gene codominance, in which each copy of the D allele was assigned weight 1 in the analysis, so that the genotypes had the following weights: 0 (II genotype), 1 (DI genotype) and 2 (DD genotype), depending on the number of D alleles. Therefore, a categorical variable of ACEGP was transformed into a numerical variable (0, 1, 2) to simulate the weight of each copy of the D allele in the echocardiographic findings.

Results
Genetic profile of the sample Regarding the allele frequency, I alleles occurred 53 times, while D alleles, 145 times. Genotype frequencies were 3.02% II, 47.48% DI and 49.5% DD. The genetic profile was tested and showed no deviation from the Hardy-Weinberg equilibrium.

Characteristics of the population
Mean age was 65.4±11.4 years, with a wide range (36 years -94 years). The distribution of skin color was as follows: white, 69.7%; mixed, 16.2%; black, 14.1%. There were no Asians. There were more males (73 men and 26 women) in the population and in the groups with D alleles, but not in the II group. There were more white individuals in all groups, with lower evidence in the DD group, with no statistically significant difference (Table 1). Drug treatment was assessed, and most patients were on ACE inhibitors and beta-blockers. There was no statistically significant variation between the gene groups assessed (Table 1). Figure 1 shows the LVEF findings at the initial and final echocardiographic tests.

Echocardiographic results
Initially most patients (37.38%) were in the LVEF range of 35-45%, being followed by those in the LVEF range of 46-55% (24.24%). On the final echocardiogram, there was  Table 2 shows the mean values of LVEF, LVSD and LVDD on both echocardiograms assessed, without statistical difference between the values found. Figure 2 shows the mean LVEF value changes during follow-up between final and initial echocardiographies in the sample and in the genotype groups.
The changes during follow-up in the echocardiographic parameters, regarding their improvement or worsening, were objectively assessed, and the differences were quantified. Table 2 shows the differences between the two echocardiographic assessments of LVSD, LVDD and LVEF (ΔLVSD, ΔLVDD and ΔLVEF).
The ΔLVDD was positive in the sample and individuals with DD and DI genotypes, showing and increase in LVDD. Patients with II genotype had negative ΔLVDD (mean, -11), evidencing a reduction in LVDD. That ΔLVDD assessment was statistically significant in the analysis between the groups (p=0.018).
The ΔLVSD showed the same trend in the genotype groups and in the sample (increase in the DD and DI genotypes, and decrease in the II genotype), but with no statistical significance.
The ΔLVEF was negative in the sample and individuals with DD genotype, showing a decrease in LVEF, and positive in DI and II genotypes, showing an increase in LVEF. However, differently from ΔLVDD, there was no statistical significance.
To objectively assess whether there was improvement or worsening of the parameters analyzed (LVEF, LVDD and LVSD) during follow-up, FΔLVEF, FΔLVSD and FΔLVDD were obtained.  Worsening of the LVEF was observed in most DD genotype patients (67.3%), in 40.4% of DI genotype patients and in only 33.3% of genotype II patients.

Continuation
Regarding FΔLVSD, worsening was observed in most DD genotype patients (65.4%) and in 40.4% of DI genotype patients, with statistical significance (p=0.010), while all II genotype patients (100.0%) had improvement of that parameter ( Table 3).

Discussion
The allele frequency obtained in this study differs from that of most national and international studies, because we found a lower number of II genotype patients, only 3% of the population. In patients with CAD or heart failure, higher D allele frequency than that in the general population and higher prevalence of the DI genotype have been reported, which differs from the findings in this study, where the DD genotype was the most prevalent.
Remodeling after acute myocardial infarction is a predictor of heart failure and mortality, and the increase in ventricular diameters in patients with heart failure is associated with clinical worsening. The renin-angiotensinaldosterone system and ACE are known to contribute to those processes; thus, some studies assessing ACEGP in those populations have also assessed echocardiographic parameters, similarly to the present study. Higher serum levels of ACE and angiotensin II in patients with DD and DI genotypes can be related to worse outcome for those patients.
Some studies have reported different echocardiographic outcomes for patients with heart failure and CAD, depending on the ACEGP. 12 Nagashima et al. 13 have shown, in patients with old anteroseptal infarction, the higher influence of DD and DI genotypes on left ventricular remodeling as compared with that of II genotype patients. In addition, He et al. 14 have reported that the I/D ACEGP can have an important role in late ventricular remodeling after acute myocardial infarction. Ohmichi et al. 9 have shown that the D allele presence can be a risk factor for the development of heart failure with left ventricular dysfunction after acute myocardial infarction.
The present study found worsening in the LVEF ranges during follow-up, with most patients with ejection fraction values lower than those in the initial test, despite drug treatment. Analyzing the mean values of ejection fraction and ventricular systolic diameters, a trend towards worsening is observed in DD genotype individuals, but with no statistical significance between the ACEGP groups.
However, there was echocardiographic worsening of the mean values of LV diastolic volume in DD genotype patients, with an increase in the ΔLVDD, with statistical significance in the analysis between the ACEGP groups. In addition, the objective analysis of improvement or worsening of the echocardiographic parameters during follow-up evidenced, with statistical significance, more DD genotype patients with worsening, followed by DI genotype patients, while most II genotype patients improved those parameters. This suggests a pattern in which the D allele presence would be associated with worsening of echocardiographic parameters, more evident in the DD genotype group than in the DI genotype group. Assessing the importance of the D allele, there was a significant correlation between its weight and the echocardiographic variables ΔLVEF, ΔLVSD and ΔLVDD, evidencing that, in that population, ACEGP associated with different echocardiographic outcomes, according to the D allele presence and genotypes of that polymorphism. Such results are in accordance with literature reports of higher severity of those patients and worse echocardiographic outcome. 5,12 A Brazilian study conducted in 2005 15 with patients with heart failure of all causes, 63 of ischemic etiology, has shown the trend towards greater left ventricular diameters, mainly the LVSD in patients with the DD genotype, meaning worse outcome for the DD genotype; however, the present study did not find the same statistical impact. Another study 16 has reported the association of the D allele presence with left ventricular dysfunction in patients with acute myocardial infarction, but at a more acute phase of the infarction, differently from that proposed in the present study.
In a study 17 assessing 142 patients with acute myocardial infarction, echocardiographic assessment including the measurement of LVEF and left ventricular diastolic and systolic volumes has shown no statistical difference between the mean values of the tests performed in each genotype group. However, differently from the results of the present study, those authors have reported improvement during follow-up of LVEF and both diameters in patients with the DD genotype, as well as improvement in LVEF during follow-up in DI genotype patients, but not in II genotype patients.
Other studies support the thesis that drug treatment with ACE inhibitors 18 or with beta-blockers 12 has a more positive influence on the echocardiographic parameters of DD genotype patients. The Russian study 18 assessing patients with ischemic heart failure has reported greater improvement in ejection fraction and systolic and diastolic diameters in the DD genotype patients who started treatment with perindopril. In those studies, the rates of ACE inhibitor use were higher than those of this study, which, considering the reports on the benefit of the use of those drugs in D allele patients, could partially explain the difference in results; that, however, cannot be applied when analyzing the use of angiotensin-receptor blockers. 17 The beta-blocker use across the genotypes was very high and very similar (100% DD, 95.7% DI and 100% II). In addition, the rates of use of ACE inhibitors or angiotensin-receptor blockers were high and similar, with no statistically significant difference in treatment according to the genotypes. The same occurs regarding the target dose of ACE inhibitors and beta-blockers, because, although higher doses could lead to a different outcome, the genotype groups studied did not significantly differ regarding the target dose.
Possible limitations of this study include the number of patients, mainly in the II genotype group; however, although several studies have larger samples, 19 genetic studies with smaller numbers of patients have been reported. 20 In addition, several pertinent results were obtained with evident statistical significance. The small number of patients with II genotype might somehow be related to the severity of the population studied, mostly composed by patients with DD and DI genotypes, reported as related to worse outcome. The analysis of genotype subgroups reported in the literature comprises always the three subgroups, without gathering any of them. This study evidenced the correlation of the D allele presence (none in II, one in DI and two in DD) with echocardiographic outcome, emphasizing the importance of analyzing each genotype separately, despite the difference in the number of patients in each genotype group. Another limitation relates to data collection from medical records, which can generate errors, but that was reduced by the fact that the The ACEGP studied associated with the echocardiographic outcome: there were more DD genotype patients with worsening of the LVEF, LVSD and LVDD, followed by DI genotype patients, while II genotype patients had the best outcome. Echocardiographic analysis of the difference between LVDD during follow-up showed the same pattern.

Potential Conflict of Interest
No potential conflict of interest relevant to this article was reported.

Sources of Funding
This study was partially funded by FAPERJ.

Study Association
This article is part of the thesis of master submitted by Gustavo Salgado Duque, from Universidade do Estado do Rio de Janeiro. population was cared for at a university center of teaching and research with experienced professionals.

Conclusions
In a population of 99 patients with ischemic heart failure: The allele and genotype frequencies related to ACEGP found in this study differed from those of the national and international literature. Only 3% of the population had II genotype.