- Email: [email protected]
QRS Widening Rates and Genetic Polymorphisms of Matrix Metalloproteinases in a Cohort of Patients With Chronic Heart Failure
Canadian Journal of Cardiology 30 (2014) 345e351
Clinical Research
QRS Widening Rates and Genetic Polymorphisms of Matrix Metalloproteinases in a Cohort of Patients With Chronic Heart Failure Virgilio Olsen, MS,a Luis E. Rohde, MD, PhD,a,b,c Luis Beck-da-Silva, MD, PhD,b,c Kátia G. Santos, PhD,a,c,d Andreia Biolo, MD, PhD,a,b,c Nadine Clausell, MD, PhD,a,b,c and Michael Andrades, PhDa,c a
Experimental and Molecular Cardiovascular Laboratory, Experimental Research Center, Hospital de Clínicas de Porto Alegre, Porto Alegre, Rio Grande do Sul, Brazil b c
Heart Failure and Cardiac Transplant Unit, Cardiology Division, Hospital de Clínicas de Porto Alegre, Porto Alegre, Rio Grande do Sul, Brazil
Post-Graduate Program in Cardiology and Cardiovascular Sciences, Universidade Federal do Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil d
Laboratory of Human Molecular Genetics, Universidade Luterana do Brasil, Canoas, Rio Grande do Sul, Brazil
ABSTRACT
RESUM E
Background: QRS duration is considered to be an indicator of adverse outcome in patients with heart failure (HF), and genetic polymorphisms may be involved in this conductivity impairment. We studied the prognostic impact of the QRS widening rate (QRS-WR) on patients with HF and the influence of the matrix metalloproteinases gene polymorphisms on the QRS-WR. Methods: This prospective cohort study included 184 patients with left ventricular (LV) systolic dysfunction (LV ejection fraction [LVEF] < 45%). The QRS-WR was calculated as the difference between 2 electrocardiogram assessments (in ms) divided by the time elapsed between each evaluation (months). The MMP-1 -1607 1G/2G, MMP-2 -790G/T and -1575G/A, MMP-3 -1171 5A/6A, MMP-9 -1562 C/T and R279Q, and MMP-12 -82A/G polymorphisms were genotyped using polymerase chain reactionerestriction fragment length polymorphism. Results: Patients were predominantly white (68%) men (67%) in New York Heart Association functional classes I and II (77%). Patients with HF with a QRS-WR 0.5 ms/month had more HF-related deaths and more combined clinical events than those with a QRS-WR < 0.5 ms/ month (P ¼ 0.03 and P ¼ 0.01, respectively). After adjusting for other
e du QRS est conside re e comme un indicateur Introduction : La dure sultats de favorables chez les patients ayant une insuffisance de re ne tiques peuvent être cardiaque (IC) et les polymorphismes ge s dans cette alte ration de la conductivite . Nous avons e tudie implique quences pronostiques du taux d’e largissement du QRS (TEles conse QRS) sur les patients ayant une IC et l’influence des polymorphismes talloprote inases matricielles (MMP : matrix metaldes gènes des me loproteinases) sur le TE-QRS. thodes : Cette e tude de cohorte prospective incluait 184 patients Me ayant une dysfonction systolique ventriculaire gauche (VG; fraction jection VG < 45 %). Le TE-QRS a e te calcule comme e tant la d’e rence entre 2 e valuations d’e lectrocardiogramme (en millisecondes diffe e par le temps e coule entre chaque e valuation (mois). Les [ms]) divise polymorphismes MMP-1 -1607 1G/2G, MMP-2 -790G/T et -1575G/A, MMP-3 -1171 5A/6A, MMP-9 -1562 C/T et R279Q, et MMP-12 -82A/G te ge notype s par le polymorphisme de longueur des fragments de ont e action en chaîne de la polyme rase. restriction et la re sultats : Les patients e taient en majorite des hommes (67 %) cauRe pondant à la classification fonctionnelle I et II de la casiens (68 %) re
Left ventricular (LV) remodelling is a process that involves the loss of cardiomyocytes and substantial changes in extracellular matrix components, culminating in cavity dilatation and a decrease in systolic and diastolic functions. A potentially
harmful consequence of heart remodelling is impairment of the heart electrical conductivity. The restructured tissue may be prone to arrhythmia generation and QRS widening.1 A QRS duration of > 120 ms is an important prognostic indicator in patients with heart failure (HF) because it has been associated with sudden cardiac death and overall mortality.2,3 Alternatively, some authors have postulated that rather than relying on a static cutoff value, changes in QRS duration over time, as observed by serial electrocardiograms, might have a stronger prognostic value.4,5 However, the underlying mechanisms responsible for QRS widening remain obscure. Matrix metalloproteinases (MMPs) are a family of zincdependent enzymes that are responsible for the turnover of
Received for publication September 18, 2013. Accepted November 15, 2013. Corresponding author: Dr Michael Andrades, Experimental and Molecular Cardiovascular Laboratory, Experimental Research Center, Hospital de Clínicas de Porto Alegre, Rua Ramiro Barcelos 2350, Sala 12201, Porto Alegre, Rio Grande do Sul, 90035-003, Brazil. Tel./Fax: þ55-51-33598844. E-mail: [email protected] See page 350 for disclosure information.
0828-282X/$ - see front matter Ó 2014 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cjca.2013.11.014
346
Canadian Journal of Cardiology Volume 30 2014
covariates, the QRS-WR remained an independent predictor of combined clinical events (hazard ratio, 1.6; 95% confidence interval, 1.12.5; P ¼ 0.02). The MMP-1 2G2G genotype was associated with nearly a 2-fold increase in QRS-WR (P ¼ 0.03). Conversely, patients with the MMP-3 5A5A genotype and a nonischemic cause of HF were protected against QRS enlargement (P ¼ 0.03). Conclusions: QRS-WR retains prognostic value in patients with chronic HF receiving guideline-based pharmacologic treatment. MMP gene polymorphisms can influence the rate of QRS enlargement over time.
New York Heart Association (77 %). Les patients ayant une insuffisance cardiaque qui avaient un TE-QRS 0,5 ms/mois subissaient plus cès lie s à l’IC et plus d’e ve nements cliniques combine s que ceux de de qui avaient un TE-QRS < 0,5 ms/mois (P ¼ 0,03 et P ¼ 0,01, respectivement). Après l’ajustement des autres covariables, le TE-QRS dicteur inde pendant des e ve nements cliniques demeurait un pre s (rapport de risque 1,6; intervalles de confiance à 95 % de combine notype MMP-1 2G2G a e te associe à une 1,1e2,5; P ¼ 0,02). Le ge augmentation de près de 2 fois du TE-QRS (P ¼ 0,03). Inversement, notype MMP-3 5A5A et une e tiologie les patients ayant un ge mique e taient prote ge s contre l’e largissement du QRS non ische (P ¼ 0,03). Conclusion : Le TE-QRS conserve sa valeur pronostique chez les patients ayant une IC chronique recevant le traitement pharmacologique selon les recommandations. Les polymorphismes des gènes MMP largissement du QRS avec le temps. peuvent influencer le taux d’e
extracellular matrix components, thereby regulating the balance between collagen synthesis and degradation.6 MMP gene polymorphisms may influence transcriptional activity7,8 and influence HF-related characteristics and outcomes.9-12 However, no previous study has attempted to evaluate the potential impact of MMP gene polymorphisms on the progression of QRS widening in a cohort of patients with chronic HF. Thus, the aims of the present study were to evaluate (1) the rates of QRS widening in a contemporary cohort of patients with chronic HF receiving the standard-of-care guideline-based pharmacologic treatment, (2) the prognostic impact of the QRS-WR in patients with HF, and (3) the impact of common MMP-1, -2, -3, -9, and -12 gene polymorphisms on the progression of QRS widening.
history of myocardial infarction could also be defined as having an ischemic cause if electrocardiography, echocardiography, or nuclear medicine and invasive coronary angiography were consistent with a silent myocardial infarction. Outcome assessment Patients with HF had follow-up visits at the outpatient clinic at our institution, and vital status was evaluated using hospital electronic records, telephone contacts, and a regional death certificate database (from the state health system database). All causes of mortality, HF-related death (defined as sudden unexpected death within 1 hour of the initiation of symptoms or caused by advanced refractory disease), and HFrelated admissions were considered the main outcomes. ECG data
Methods Patients Patients were recruited between July 2003 and November 2007 from the HF and transplant outpatient clinic in a tertiary care university hospital in Porto Alegre, Brazil. We included a total of 316 consecutive patients older than 18 years with HF and LV systolic dysfunction (LV ejection fraction [LVEF] < 45%). Patients who had an implantable device (standard pacemakers or biventricular pacemakers for cardiac resynchronization therapy; 27 patients) or who did not have the minimum 3-month interval between the electrocardiogram (ECG) records (105 patients) were excluded, resulting in a total of 184 patients who remained in this study. Patients underwent clinical and laboratory evaluation consisting of a standardized questionnaire, physical examination, evaluation of echocardiographic parameters, and laboratory tests. An ischemic cause was defined as a previous history of an acute myocardial infarction associated with the expected alterations on the surface electrocardiogram (clear pathologic Q waves in at least 2 contiguous leads), echocardiography (akinetic areas on transthoracic echocardiography), or nuclear medicine scans (fixed perfusion deficits on myocardial perfusion imaging). Coronary angiography was not a requirement to define an ischemic cause, but the majority of patients (>85%) had an invasive evaluation demonstrating coronary lesions. Patients without a clear clinical
All patients in our prospective cohort had an ECG evaluation at the time of inclusion (QRS1). For the current protocol, we searched the records of the patients to identify those who also had an electrocardiogram obtained before inclusion in the study (QRS0). Patients without a previous electrocardiogram (100 patients) or with a previous electrocardiogram < 3 months before QRS1 assessment (5 patients) were excluded. The QRS duration was calculated by automatic analysis (ELI 350; Mortara Instrument Inc, Milwaukee, WI) in ms. The rate of QRS widening (ms/month) was calculated as the difference between the 2 assessments (QRS1 QRS0) divided by the time elapsed between each evaluation (date of QRS1 date of QRS0) in months. For analysis, patients were stratified according to the following QRS-WR: 0.5 ms/month (n ¼ 45) and < 0.5 ms/month (n ¼ 139), as previously suggested.4 Genotyping Genomic DNA was extracted from peripheral blood according to a nonenzymatic method previously described.13 The polymerase chain reaction method was used to amplify fragments containing the MMP-1 -1607 1G/2G (rs1799750), MMP-2 -790G/T (rs243864), MMP-2 -1575G/A (rs243866), MMP-3 -1171 5A/6A (rs3025058), MMP-9 -1562C/T (rs3918242), MMP-9 R279Q (rs17576), and MMP-12 -82A/ G (rs2276109) polymorphisms, using primers and conditions as described in Supplemental Table S1. The digested fragments
Olsen et al. QRS Widening Rates and MMP in HF
347
Table 1. Demographic and clinical characteristics of patients
Age (y) White (%) Male sex (%) Cause of HF Ischemic (%) Idiopathic (%) Hypertensive (%) NYHA (n ¼ 179) Classes I and II (%) Classes III and IV (%) All-cause mortality (%) HF-related deaths (%) Comorbidities AMI (%) Hypertension (%) Diabetes mellitus (%) Current smoker (%) Laboratory variables Creatinine (mg/dL) Sodium (mEq/dL) Hemoglobin (g/dL) Baseline drugs b-blockers (%) ACEi (%)
All patients with HF (N ¼ 184)
QRS-WR < 0.5 ms/mo (n ¼ 139)
QRS-WR 0.5 ms/mo (n ¼ 45)
P value
62 12 125 (67.9) 124 (67.4)
62 12 98 (70.5) 90 (64.7)
59 13 27 (60.0) 34 (75.6)
0.17 0.20 0.20
56 (40) 42 (30) 32 (23)
15 (33) 15 (33) 9 (20)
0.48 0.71 0.83
(75.0) (25.0) (33.1) (15.8)
36 7 14 11
(83.7) (16.3) (31.1) (24.4)
0.30
(37) (66) (35) (14)
9 23 14 4
(20) (51) (31) (9)
0.04 0.07 0.71 0.60
71 (39) 57 (31) 41 (22) 138 41 60 33
(77.1) (22.9) (32.6) (17.9)
60 115 63 23
(33) (63) (34) (13)
102 34 46 22 51 92 49 19
0.86 0.19
1.3 0.4 140.5 3.3 13.0 1.6
1.3 0.4 140.6 3.2 13.1 1.5
1.3 0.5 140.0 3.5 12.7 1.9
0.96 0.31 0.15
161 (87.5) 161 (87.5)
118 (84.9) 123 (88.5)
43 (95.6) 38 (84.4)
0.07 0.44
Data are expressed as the mean standard deviation or as a number (percentage), as appropriate. ACEi, angiotensin-converting enzyme inhibitor; AMI, acute myocardial infarction; HF, heart failure; NYHA, New York Heart Association; QRS-WR, QRS widening rate.
were separated by electrophoresis in an 8% polyacrylamide gel and visualized after silver nitrate staining (MMP-1 and MMP-3 polymorphisms) or in a 2% agarose gel and visualized after ethidium bromide staining under ultraviolet light (MMP-9, MMP-2, and MMP-12 polymorphisms). Statistical analysis Continuous variables with normal distribution were compared using the Student t test and were expressed as the mean standard deviation, whereas continuous variables with asymmetrical distribution were compared using the Mann-Whitney U test and were expressed as the median and 25th-75th percentile. Categorical variables were compared using the c2 test or the Fisher exact test, as appropriate, and were expressed as absolute numbers and percentages. c2 was also used to test the Hardy-Weinberg equilibrium. For the prognostic analysis, Kaplan-Meier survival curves were constructed from the date of entry at the outpatient clinic up to the last registry of follow-up or death and were compared using the log-rank statistics. Cox proportional hazard models were created and adjusted for variables that are recognized as influencing HF-related outcomes. For this analysis, the model was adjusted for functional class, LVEF, and serum creatinine levels. All data were analyzed using SPSS, version 16.0, software for Windows (SPSS, Inc, Chicago, IL). A P value < 0.05 was considered statistically significant.
functional classes I and II (77%), with moderate to severe LV dysfunction (mean LVEF ¼ 31% 8%) and a mean age of 62 12 years. The main causes of HF were ischemic (39%) and idiopathic (31%); more than half of the patients had hypertension (63%), and approximately one third of the patients had experienced a previous myocardial infarction. Patients were receiving the standard-of-care drug treatment; almost 90% were using angiotensin-converting enzyme inhibitors and beta-blockers. When patients were stratified according to QRS-WR, no difference in baseline characteristics was seen, except for a small increase in the presence of myocardial infarction in the QRS-WR 0.5 ms/month group (Table 1). Despite this, previous myocardial infarction was not independently associated with clinical combined events or HF-related deaths (data not shown). Table 1 describes the main clinical characteristics of the cohort. ECG data
Results
As shown in Table 2, the majority of the patients were in sinus rhythm (79%), and approximately one third had left bundle branch block at the time of inclusion. Overall, the median QRS0 duration was 110 ms (25th-75th percentile: 96150 ms), the median QRS1 duration was 121 ms (25th-75th percentile: 100-154 ms), and the median time difference between ECG assessments was 24.4 months (25th-75th percentile: 13.7-55.0 months). The median rate of QRS progression was 0.18 ms/month (25th-75th percentile: 0.00.49 ms/month). No statistically significant differences were observed in the baseline assessment when patients were stratified according to QRS-WR.
Demographic and clinical data of patients with HF
QRS-WR and clinical outcomes
Our HF cohort was predominantly composed of white (68%) male (67%) patients in New York Heart Association (NYHA)
A higher QRS-WR predicted a worse prognosis, whereas both the QRS0 and QRS1 were not related to future clinical
348
Canadian Journal of Cardiology Volume 30 2014
Table 2. ECG and echocardiographic data
Echocardiography LVEF (%) LVDD (cm) LVSD (cm) LA dimension (cm) Electrocardiography Sinus rhythm (%) Atrial fibrillation (%) LBBB (%) QRS0 duration (ms) QRS1 duration (ms) QRS-WR (ms/mo) ECG interval (mo)
All patients with HF (N ¼ 184)
QRS-WR < 0.5 ms/mo (n ¼ 139)
QRS-WR 0.5 ms/mo (n ¼ 45)
31 8 6.6 0.9 5.6 1.0 4.7 0.8
32 8 6.6 0.9 5.6 1.0 4.8 0.8
31 8 6.7 1.0 5.7 1.0 4.7 0.8
79 21 34 110 (96-150) 121 (100-154) 0.18 (0.00-0.49) 24.4 (13.7-55.0)
76 23 33 112 (98-152) 118 (99-152) 0.09 (0.13-0.25) 30.5 (16.3-63.0)
86 14 35 104 (94-136) 134 (112-154) 0.86 (0.64-1.47) 16.1 (9.5-31.7)
P value 0.68 0.33 0.51 0.51 0.20 0.28 0.97 0.10 0.03 < 0.01 < 0.01
Data are expressed as the mean standard deviation, the median (25th-75th percentile), or as a number (percentage), as appropriate. ECG, electrocardiogram; HF, heart failure; LA, left atrial; LBBB, left bundle branch block; LVDD, left ventricular diastolic diameter; LVEF, left ventricular ejection fraction; LVSD, left ventricular systolic diameter; QRS-WR, QRS widening rate.
outcomes. Patients with HF with a QRS-WR 0.5 ms/ month had an increased risk of HF-related death or combined clinical events when compared with patients with a QRS-WR < 0.5 ms/month (log-rank P ¼ 0.03 and 0.01, respectively) (Fig 1). After adjusting for other risk covariates in a Cox proportional hazard model, QRS-WR remained an independent predictor of combined clinical events (HR, 1.6; 95% CI, 1.1-2.5; P ¼ 0.02). However, QRS-WR was not associated with all-cause mortality (log-rank P ¼ 0.95). MMP genotypes, LV remodelling, and QRS-WR Patients were stratified according to the MMP-1 or MMP-3 genotype, and echocardiographic parameters were evaluated. There was no association between MMP genotype and ventricular remodelling (Supplemental Tables S2 and S3). For the analysis of the impact of the MMP genotypes on QRS-WR, patients with HF were stratified according to cause (ischemic and nonischemic), as shown in Table 3. The MMP-1 2G2G genotype was associated with nearly a 2-fold increase in QRS-WR (P ¼ 0.03). Such an effect was not evident when ischemic and nonischemic patients were analyzed separately. Despite the MMP-1 2G2G genotype being highly associated with QRS-WR, it was not associated with clinical combined events (HR, 0.8; 95% CI, 0.5-1.2; P ¼ 0.33). Conversely, patients with a MMP-3 5A5A genotype and a nonischemic cause were highly protected against QRS enlargement. No other MMP polymorphism was associated with QRS-WR, even when the patients were stratified by cause (Table 3). It is worth noting that the observed genotype frequencies were in Hardy-Weinberg equilibrium in the sample studied for all analyzed polymorphisms.
Figure 1. (A) Heart-failureerelated deaths and (B) combined clinical events for all patients according to their QRS-WR. Solid line represents patients with a QRS-WR < 0.5 ms/month (n ¼ 139), and the dashed line represents patients with a QRS-WR 0.5 ms/month (n ¼ 45).
Discussion In the present study, we investigated the rate of QRS widening and its prognostic value in patients with HF, as well as the impact of common MMP polymorphisms on QRS-WR. Our findings indicate that QRS-WR is an independent predictor of HF morbidity and mortality in a contemporary HF cohort. In addition, our results suggest that genetic polymorphisms of MMP-1 and MMP-3 may explain, in part, the interindividual variability of QRS-WR.
Olsen et al. QRS Widening Rates and MMP in HF
349
Table 3. QRS duration according to the MMP genotypes All patients (N ¼ 184) MMP-1 -1607 1G/2G 1G carriers (73/73/73)* 2G2G (27/27/27) P value MMP-2 -1575G/A A carriers (35/40/32) GG (65/60/68) P value MMP-2 -790G/T G carriers (38/44/34) TT (62/56/66) P value MMP-3 -1171 5A/6A 6A carriers (82/82/82) 5A5A (18/18/18) P value MMP-9 -1562C/T T carriers (20/21/19) CC (80/70/81) P value MMP-9 R279Q R carriers (50/52/50) QQ (50/48/50) P value MMP-12 -82A/G AA (84/74/90) AG (16/26/10) P value
Ischemic cause (n ¼ 71)
Nonischemic cause (n ¼ 113)
0.17 (0.04-0.42) 0.29 (0.07-0.63) 0.03
0.20 (0.07-0.40) 0.29 (0.09-0.66) 0.20
0.11 (0.11-0.48) 0.32 (0.04-0.64) 0.09
0.22 (0.00-0.44) 0.17 (0.02-0.53) 0.63
0.26 (0.11-0.44) 0.18 (0.00-0.45) 0.46
0.19 (0.10-0.50) 0.15 (0.08-0.57) 0.95
0.21 (0.00-0.56) 0.17 (0.04-0.50) 0.36
0.26 (0.10-0.45) 0.18 (0.01-0.41) 0.31
0.17 (0.08-0.75) 0.15 (0.08-0.53) 0.75
0.20 (0.00-0.54) 0.14 (0.09-0.31) 0.11
0.21 (0.03-0.40) 0.26 (0.10-0.45) 0.58
0.20 (0.05-0.70) 0.00 (0.44-0.25) 0.03
0.13 (0.00-0.38) 0.20 (0.02-0.52) 0.55
0.15 (0.08-0.25) 0.24 (0.00-0.44) 0.25
0.08 (0.12-0.76) 0.17 (0.08-0.55) 0.93
0.18 (0.00-0.54) 0.20 (0.02-0.44) 0.93
0.29 (0.12-0.44) 0.18 (0.01-0.47) 0.19
0.13 (0.12-0.43) 0.18 (0.00-0.59) 0.47
0.18 (0.02-0.45) 0.23 (0.07-0.65) 0.66
0.26 (0.15-0.43) 0.22 (0.10-0.64) 0.85
0.13 (0.11-0.52) 0.27 (0.15-0.94) 0.62
* Data are expressed as the median (25th-75th percentile) of the QRS widening rate (ms/mo). Numbers in parentheses indicate the genotype (percentage) in all patients, patients with an ischemic cause of HF, and patients with a nonischemic cause of HF, respectively.
A resting electrocardiogram is a basic tool in the initial evaluation of patients with HF. The prognostic role of QRS duration in a single resting electrocardiogram, however, is not consistent in different studies.14-17 Therefore, some authors have suggested that serial ECG monitoring may provide stronger prognostic information than isolated measurements because sequential data on QRS width could portray disease progression, adverse heart remodelling, and progressive impairment of electrical conductivity. Grigioni et al. analyzed QRS intervals in 82 patients with dilated cardiomyopathy and observed that the incidence of cardiac death or heart transplantation was higher among patients with QRS interval widening > 0.5 ms/month after 12 months of follow-up.4 QRS widening remained independently and unfavourably predictive of a worse prognosis after adjustment for baseline risks during multivariate analysis. In the present analysis, we demonstrated that patients with HF are prone to widening of the QRS over a median time of 24 months. Moreover, although isolated QRS0 and QRS1 measurements were not associated with future HF-related events, QRS-WR was independently associated with HF-related combined events after controlling for confounding variables. Despite optimized medical management, QRS widening still occurs over time and the intrinsic causes are unclear. Some mechanisms that are likely to be involved include progressive interstitial fibrosis,18 ischemia in the conduction tissue,5,19 LV dilatation,20 and genetically based influences.21 These findings indicate that substitution of an electrical competent tissue (cardiomyocyte) for another apparently less adequate tissue (fibrosis) may be the basis of QRS widening. MMPs play a pivotal role in heart remodelling, as described in experimental and clinical studies.22-24 Activation of MMPs
promotes changes in the extracellular matrix and an imbalance in collagen content, which can lead to disruptions in cardiomyocyte connectivity and impairment in tissue electrical conductivity. In the present analysis, we evaluated whether common genetic MMP polymorphisms could explain, at least in part, the interindividual variability of QRS-WR. Our results suggest that patients carrying the MMP-1 2G2G genotype had a significantly higher QRS-WR. MMP-1 acts on types I, II, III, VII, and X collagen, but its role in heart remodelling is still under debate. The 2G allele is associated with increased promoter activity of the MMP-1 gene, and previous studies have suggested that MMP-1 may have an impact on LV remodelling after myocardial infarction and the overall risk of future coronary heart disease.25,26 Our group has previously demonstrated that the MMP-1 2G allele was associated with better prognosis in an HF cohort,9 which seems to be contradictory to the increased QRS-WR shown in the present study. However, Velho et al. also demonstrated that after including the QRS in the multivariate analysis, the MMP-1 genotype was no longer associated with prognosis.9 This suggests that the electrical conductivity system and MMP-1 may act by different mechanisms in the course of the disease. Taken together, these findings suggest that the net final effect of isolated nucleotide polymorphisms per se on HF prognosis might be outweighed by other compensatory or regulatory pathways. Also, because most data are based on the analysis of specific predefined polymorphisms (and not genome-wide analysis), prognostic data must be interpreted with caution. Haplotype analysis and other more comprehensive genetic approaches might be able to unravel the overall role of genetic variability on HF prognosis. Although our results suggest that MMP genotypes impact on QRS-WR,
350
this was not translated into a consistent effect on HF prognosis. Increased levels of MMP-3 have been associated with cardiac remodelling in ischemic and nonischemic dilated cardiomyopathy.22,27 Interestingly, we found that the MMP-3 5A5A genotype was associated with a lower QRS-WR only in nonischemic patients, suggesting that the cause of HF can affect the molecular pathways that lead to LV remodelling. Despite the influence on the QRS-WR, neither MMP-1 nor MMP-3 genotypes were associated with ventricular dilatation or myocardial hypertrophy in the present study. To extend our comprehension of the role that the MMP genotype plays in QRS-WR progression, other polymorphisms were included in the analysis, namely MMP-2 -790G/T, MMP-2 -1575G/A, MMP-9 -1562C/T, MMP-9 R279Q, and MMP-12 -82A/G. Gelatinase is a general name given to both MMP-2 (gelatinase A) and MMP-9 (gelatinase B) because of their ability to hydrolyze gelatin. MMP-2 and -9 gene polymorphisms have been associated with HF development.12,28 Two polymorphisms in the promoter region of the MMP-2 gene (-790G/T and -1575G/A) were included in our study. Both variants have already been described as being functional in vitro,29 but their association with HF remains controversial.12,28 The MMP-9 -1562C/T polymorphism alters gene transcription efficiency, whereas the MMP-9 R279Q polymorphism is associated with the MMP-9 catalytic domain; both polymorphisms have been associated with cardiovascular diseases.30,31 Despite the biological importance described for each polymorphism analyzed, we found no association of the MMP-2 -790G/T, MMP-2 -1575G/A, MMP-9 -1562C/T, MMP-9 R279Q, or MMP-12 -82A/G polymorphisms with QRS-WR. Some aspects of our study design should be considered when evaluating the results. We found that QRS-WR is associated with major events in patients with HF and that patients carrying the MMP-1 2G2G genotype and MMP-3 6A allele had an increased QRS-WR. Despite being important, the translation of MMP polymorphisms into enzymatic activity, tissue remodelling, and eventually impairment in electrical conductivity was not the goal of this study. The simultaneous assessment of genetic polymorphisms and serum levels of MMPs/tissue inhibitors of metalloproteinase could be an interesting approach to obtain a more comprehensive view of the processes involved in heart remodelling and electrical impairment. Additionally, our relatively small sample size can limit the statistical power to identify other interactions. For this reason, our data on the impact of genetic polymorphisms on QRS-WR need further validation in an independent cohort. Conclusions We have observed that a QRS-WR 0.5 ms/month was associated with increased HF-related events and mortality in a contemporary cohort of patients with HF and systolic dysfunction. We tried to unravel potential mechanisms that might be partially responsible for QRS widening. The 2G2G genotype of the MMP-1 gene was associated with a higher QRS-WR, whereas the impact of the MMP-3 5A5A genotype was dependent on the cause of HF. Because QRS-WR is an independent predictor of prognosis in HF, further studies are
Canadian Journal of Cardiology Volume 30 2014
needed to elucidate why some patients are particularly prone to QRS widening. Genetic variability in the molecular processes involved in LV remodelling may explain this phenomenon to some extent. Funding Sources This work was supported by Fundo de Incentivo à Pesquisa e Eventos do Hospital de Clínicas de Porto Alegre (FIPE-HCPA), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES). Disclosures The authors have no conflicts to disclose. References 1. Rubart M, Zipes DP. Genesis of Cardiac Arrhythmias: Electrophysiological Considerations. In: Braunwald E, ed. Braunwald’s Heart Disease: A Textbook of Cardiovascular Medicine. Philadelphia: Saunders, 2008: 727-61. 2. Beck-da-Silva L, Goldraich L, Bonzanini L, et al. Pulse pressure and QRS width evaluation as an inexpensive tool for heart failure assessment. Congest Heart Fail 2009;15:222-7. 3. Kalra PR, Sharma R, Shamim W, et al. Clinical characteristics and survival of patients with chronic heart failure and prolonged QRS duration. Int J Cardiol 2002;86:225-31. 4. Grigioni F, Carinci V, Boriani G, et al. Accelerated QRS widening as an independent predictor of cardiac death or of the need for heart transplantation in patients with congestive heart failure. J Heart Lung Transplant 2002;21:899-902. 5. Shamim W, Yousufuddin M, Cicoria M, et al. Incremental changes in QRS duration in serial ECGs over time identify high risk elderly patients with heart failure. Heart 2002;88:47-52. 6. Gajarsa JJ, Kloner RA. Left ventricular remodeling in the post-infarction heart: a review of cellular, molecular mechanisms, and therapeutic modalities. Heart Fail Rev 2011;16:13-21. 7. Medley TL, Kingwell BA, Gatzka CD, Pillay P, Cole TJ. Matrix metalloproteinase-3 genotype contributes to age-related aortic stiffening through modulation of gene and protein expression. Circ Res 2003;92: 1254-61. 8. Langers AM, Verspaget HW, Hommes DW, Sier CF. Single-nucleotide polymorphisms of matrix metalloproteinases and their inhibitors in gastrointestinal cancer. World J Gastrointest Oncol 2011;3:79-98. 9. Velho FM, Cohen CR, Santos KG, et al. Polymorphisms of matrix metalloproteinases in systolic heart failure: role on disease susceptibility, phenotypic characteristics, and prognosis. J Card Fail 2011;17:115-21. 10. Mizon-Gerard F, de Groote P, Lamblin N, et al. Prognostic impact of matrix metalloproteinase gene polymorphisms in patients with heart failure according to the aetiology of left ventricular systolic dysfunction. Eur Heart J 2004;25:688-93. 11. Tang LJ, Chen XF, Zhu M, et al. Matrix metalloproteinase-1, -3, and -9 gene polymorphisms and the risk of idiopathic dilated cardiomyopathy in a Chinese Han population. Clin Biochem 2007;40:1427-30. 12. Hua Y, Song L, Wu N, et al. Polymorphisms of MMP-2 gene are associated with systolic heart failure prognosis. Clin Chim Acta 2009;404:119-23.
Olsen et al. QRS Widening Rates and MMP in HF
351
13. Lahiri DK, Nurnberger JI. A rapid non-enzymatic method for the preparation of HMW DNA from blood for RFLP studies. Nucleic Acids Res 1991;19:5444.
23. Rohde LE, Ducharme A, Arroyo LH, et al. Matrix metalloproteinase inhibition attenuates early left ventricular enlargement after experimental myocardial infarction in mice. Circulation 1999;99:3063-70.
14. Dhar R, Alsheikh-Ali AA, Estes NAM, et al. Association of prolonged QRS duration with ventricular tachyarrhythmias and sudden cardiac death in the Multicenter Automatic Defibrillator Implantation Trial II (MADIT-II). Heart Rhythm 2008;5:807-13.
24. Cadete VJ, Arcand SA, Chaharyn BM, et al. Matrix metalloproteinase-2 is activated during ischemia/reperfusion in a model of myocardial infarction. Can J Cardiol 2013;29:1495-503.
15. Yerra L, Anavekar N, Skali H, et al. Association of QRS duration and outcomes after myocardial infarction: the VALIANT trial. Heart Rhythm 2006;3:313-6. 16. Katritsis DG, Siontis KC, Bigger JT, et al. Effect of left ventricular ejection fraction and QRS duration on the survival benefit of implantable cardioverter-defibrillators: Meta-analysis of primary prevention trials. Heart Rhythm 2013;10:200-6. 17. Exner DV, Birnie DH, Moe G, et al. Canadian Cardiovascular Society guidelines on the use of cardiac resynchronization therapy: evidence and patient selection. Can J Cardiol 2013;29:182-95. 18. Song BG, Yang HS, Hwang HK, et al. Correlation of electrocardiographic changes and myocardial fibrosis in patients with hypertrophic cardiomyopathy detected by cardiac magnetic resonance imaging. Clin Cardiol 2013;36:31-5. 19. Sheridan DJ, Culling W, Penny WJ. Electrophysiological disturbances associated with acute myocardial infarction. Eur Heart J 1986;(7 suppl B): 11-8. 20. Mazzoleni A, Curtin ME, Wolff R, Reiner L, Somes G. On the relationship between heart weights, fibrosis, and QRS duration. J Electrocardiol 1975;8:233-6. 21. Royer A, van Veen TAB, Le Bouter S, et al. Mouse model of SCN5Alinked hereditary Lenègre’s disease: age-related conduction slowing and myocardial fibrosis. Circulation 2005;111:1738-46. 22. Spinale FG, Coker ML, Heung LJ, et al. A matrix metalloproteinase induction/activation system exists in the human left ventricular myocardium and is upregulated in heart failure. Circulation 2000;102: 1944-9.
25. Martin TN, Penney DE, Smith JA, et al. Matrix metalloproteinase-1 promoter polymorphisms and changes in left ventricular volume following acute myocardial infarction. Am J Cardiol 2004;94:1044-6. 26. Ye S, Gale CR, Martyn CN. Variation in the matrix metalloproteinase-1 gene and risk of coronary heart disease. Eur Heart J 2003;24:1668-71. 27. Wilson EM, Moainie SL, Baskin JM, et al. Region- and type-specific induction of matrix metalloproteinases in post-myocardial infarction remodeling. Circulation 2003;107:2857-63. 28. Vasku A, Goldbergova M, Holla LI, et al. Two MMP-2 promoter polymorphisms (-790T/G and -735C/T) in chronic heart failure. Clin Chem Lab Med 2003;41:1299-303. 29. Price SJ, Greaves DR, Watkins H. Identification of novel, functional genetic variants in the human matrix metalloproteinase-2 gene: role of Sp1 in allele-specific transcriptional regulation. J Biol Chem 2001;276: 7549-58. 30. Zhang B, Ye S, Herrmann SM, et al. Functional polymorphism in the regulatory region of gelatinase B gene in relation to severity of coronary atherosclerosis. Circulation 1999;99:1788-94. 31. Zhi H, Wang H, Ren L, et al. Functional polymorphisms of matrix metallopeptidase-9 and risk of coronary artery disease in a Chinese population. Mol Biol Rep 2010;37:13-20.
Supplementary Material To access the supplementary material accompanying this article, visit the online version of the Canadian Journal of Cardiology at www.onlinecjc.ca and at http://dx.doi.org/10. 1016/j.cjca.2013.11.014.
Clinical Research
QRS Widening Rates and Genetic Polymorphisms of Matrix Metalloproteinases in a Cohort of Patients With Chronic Heart Failure Virgilio Olsen, MS,a Luis E. Rohde, MD, PhD,a,b,c Luis Beck-da-Silva, MD, PhD,b,c Kátia G. Santos, PhD,a,c,d Andreia Biolo, MD, PhD,a,b,c Nadine Clausell, MD, PhD,a,b,c and Michael Andrades, PhDa,c a
Experimental and Molecular Cardiovascular Laboratory, Experimental Research Center, Hospital de Clínicas de Porto Alegre, Porto Alegre, Rio Grande do Sul, Brazil b c
Heart Failure and Cardiac Transplant Unit, Cardiology Division, Hospital de Clínicas de Porto Alegre, Porto Alegre, Rio Grande do Sul, Brazil
Post-Graduate Program in Cardiology and Cardiovascular Sciences, Universidade Federal do Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil d
Laboratory of Human Molecular Genetics, Universidade Luterana do Brasil, Canoas, Rio Grande do Sul, Brazil
ABSTRACT
RESUM E
Background: QRS duration is considered to be an indicator of adverse outcome in patients with heart failure (HF), and genetic polymorphisms may be involved in this conductivity impairment. We studied the prognostic impact of the QRS widening rate (QRS-WR) on patients with HF and the influence of the matrix metalloproteinases gene polymorphisms on the QRS-WR. Methods: This prospective cohort study included 184 patients with left ventricular (LV) systolic dysfunction (LV ejection fraction [LVEF] < 45%). The QRS-WR was calculated as the difference between 2 electrocardiogram assessments (in ms) divided by the time elapsed between each evaluation (months). The MMP-1 -1607 1G/2G, MMP-2 -790G/T and -1575G/A, MMP-3 -1171 5A/6A, MMP-9 -1562 C/T and R279Q, and MMP-12 -82A/G polymorphisms were genotyped using polymerase chain reactionerestriction fragment length polymorphism. Results: Patients were predominantly white (68%) men (67%) in New York Heart Association functional classes I and II (77%). Patients with HF with a QRS-WR 0.5 ms/month had more HF-related deaths and more combined clinical events than those with a QRS-WR < 0.5 ms/ month (P ¼ 0.03 and P ¼ 0.01, respectively). After adjusting for other
e du QRS est conside re e comme un indicateur Introduction : La dure sultats de favorables chez les patients ayant une insuffisance de re ne tiques peuvent être cardiaque (IC) et les polymorphismes ge s dans cette alte ration de la conductivite . Nous avons e tudie implique quences pronostiques du taux d’e largissement du QRS (TEles conse QRS) sur les patients ayant une IC et l’influence des polymorphismes talloprote inases matricielles (MMP : matrix metaldes gènes des me loproteinases) sur le TE-QRS. thodes : Cette e tude de cohorte prospective incluait 184 patients Me ayant une dysfonction systolique ventriculaire gauche (VG; fraction jection VG < 45 %). Le TE-QRS a e te calcule comme e tant la d’e rence entre 2 e valuations d’e lectrocardiogramme (en millisecondes diffe e par le temps e coule entre chaque e valuation (mois). Les [ms]) divise polymorphismes MMP-1 -1607 1G/2G, MMP-2 -790G/T et -1575G/A, MMP-3 -1171 5A/6A, MMP-9 -1562 C/T et R279Q, et MMP-12 -82A/G te ge notype s par le polymorphisme de longueur des fragments de ont e action en chaîne de la polyme rase. restriction et la re sultats : Les patients e taient en majorite des hommes (67 %) cauRe pondant à la classification fonctionnelle I et II de la casiens (68 %) re
Left ventricular (LV) remodelling is a process that involves the loss of cardiomyocytes and substantial changes in extracellular matrix components, culminating in cavity dilatation and a decrease in systolic and diastolic functions. A potentially
harmful consequence of heart remodelling is impairment of the heart electrical conductivity. The restructured tissue may be prone to arrhythmia generation and QRS widening.1 A QRS duration of > 120 ms is an important prognostic indicator in patients with heart failure (HF) because it has been associated with sudden cardiac death and overall mortality.2,3 Alternatively, some authors have postulated that rather than relying on a static cutoff value, changes in QRS duration over time, as observed by serial electrocardiograms, might have a stronger prognostic value.4,5 However, the underlying mechanisms responsible for QRS widening remain obscure. Matrix metalloproteinases (MMPs) are a family of zincdependent enzymes that are responsible for the turnover of
Received for publication September 18, 2013. Accepted November 15, 2013. Corresponding author: Dr Michael Andrades, Experimental and Molecular Cardiovascular Laboratory, Experimental Research Center, Hospital de Clínicas de Porto Alegre, Rua Ramiro Barcelos 2350, Sala 12201, Porto Alegre, Rio Grande do Sul, 90035-003, Brazil. Tel./Fax: þ55-51-33598844. E-mail: [email protected] See page 350 for disclosure information.
0828-282X/$ - see front matter Ó 2014 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cjca.2013.11.014
346
Canadian Journal of Cardiology Volume 30 2014
covariates, the QRS-WR remained an independent predictor of combined clinical events (hazard ratio, 1.6; 95% confidence interval, 1.12.5; P ¼ 0.02). The MMP-1 2G2G genotype was associated with nearly a 2-fold increase in QRS-WR (P ¼ 0.03). Conversely, patients with the MMP-3 5A5A genotype and a nonischemic cause of HF were protected against QRS enlargement (P ¼ 0.03). Conclusions: QRS-WR retains prognostic value in patients with chronic HF receiving guideline-based pharmacologic treatment. MMP gene polymorphisms can influence the rate of QRS enlargement over time.
New York Heart Association (77 %). Les patients ayant une insuffisance cardiaque qui avaient un TE-QRS 0,5 ms/mois subissaient plus cès lie s à l’IC et plus d’e ve nements cliniques combine s que ceux de de qui avaient un TE-QRS < 0,5 ms/mois (P ¼ 0,03 et P ¼ 0,01, respectivement). Après l’ajustement des autres covariables, le TE-QRS dicteur inde pendant des e ve nements cliniques demeurait un pre s (rapport de risque 1,6; intervalles de confiance à 95 % de combine notype MMP-1 2G2G a e te associe à une 1,1e2,5; P ¼ 0,02). Le ge augmentation de près de 2 fois du TE-QRS (P ¼ 0,03). Inversement, notype MMP-3 5A5A et une e tiologie les patients ayant un ge mique e taient prote ge s contre l’e largissement du QRS non ische (P ¼ 0,03). Conclusion : Le TE-QRS conserve sa valeur pronostique chez les patients ayant une IC chronique recevant le traitement pharmacologique selon les recommandations. Les polymorphismes des gènes MMP largissement du QRS avec le temps. peuvent influencer le taux d’e
extracellular matrix components, thereby regulating the balance between collagen synthesis and degradation.6 MMP gene polymorphisms may influence transcriptional activity7,8 and influence HF-related characteristics and outcomes.9-12 However, no previous study has attempted to evaluate the potential impact of MMP gene polymorphisms on the progression of QRS widening in a cohort of patients with chronic HF. Thus, the aims of the present study were to evaluate (1) the rates of QRS widening in a contemporary cohort of patients with chronic HF receiving the standard-of-care guideline-based pharmacologic treatment, (2) the prognostic impact of the QRS-WR in patients with HF, and (3) the impact of common MMP-1, -2, -3, -9, and -12 gene polymorphisms on the progression of QRS widening.
history of myocardial infarction could also be defined as having an ischemic cause if electrocardiography, echocardiography, or nuclear medicine and invasive coronary angiography were consistent with a silent myocardial infarction. Outcome assessment Patients with HF had follow-up visits at the outpatient clinic at our institution, and vital status was evaluated using hospital electronic records, telephone contacts, and a regional death certificate database (from the state health system database). All causes of mortality, HF-related death (defined as sudden unexpected death within 1 hour of the initiation of symptoms or caused by advanced refractory disease), and HFrelated admissions were considered the main outcomes. ECG data
Methods Patients Patients were recruited between July 2003 and November 2007 from the HF and transplant outpatient clinic in a tertiary care university hospital in Porto Alegre, Brazil. We included a total of 316 consecutive patients older than 18 years with HF and LV systolic dysfunction (LV ejection fraction [LVEF] < 45%). Patients who had an implantable device (standard pacemakers or biventricular pacemakers for cardiac resynchronization therapy; 27 patients) or who did not have the minimum 3-month interval between the electrocardiogram (ECG) records (105 patients) were excluded, resulting in a total of 184 patients who remained in this study. Patients underwent clinical and laboratory evaluation consisting of a standardized questionnaire, physical examination, evaluation of echocardiographic parameters, and laboratory tests. An ischemic cause was defined as a previous history of an acute myocardial infarction associated with the expected alterations on the surface electrocardiogram (clear pathologic Q waves in at least 2 contiguous leads), echocardiography (akinetic areas on transthoracic echocardiography), or nuclear medicine scans (fixed perfusion deficits on myocardial perfusion imaging). Coronary angiography was not a requirement to define an ischemic cause, but the majority of patients (>85%) had an invasive evaluation demonstrating coronary lesions. Patients without a clear clinical
All patients in our prospective cohort had an ECG evaluation at the time of inclusion (QRS1). For the current protocol, we searched the records of the patients to identify those who also had an electrocardiogram obtained before inclusion in the study (QRS0). Patients without a previous electrocardiogram (100 patients) or with a previous electrocardiogram < 3 months before QRS1 assessment (5 patients) were excluded. The QRS duration was calculated by automatic analysis (ELI 350; Mortara Instrument Inc, Milwaukee, WI) in ms. The rate of QRS widening (ms/month) was calculated as the difference between the 2 assessments (QRS1 QRS0) divided by the time elapsed between each evaluation (date of QRS1 date of QRS0) in months. For analysis, patients were stratified according to the following QRS-WR: 0.5 ms/month (n ¼ 45) and < 0.5 ms/month (n ¼ 139), as previously suggested.4 Genotyping Genomic DNA was extracted from peripheral blood according to a nonenzymatic method previously described.13 The polymerase chain reaction method was used to amplify fragments containing the MMP-1 -1607 1G/2G (rs1799750), MMP-2 -790G/T (rs243864), MMP-2 -1575G/A (rs243866), MMP-3 -1171 5A/6A (rs3025058), MMP-9 -1562C/T (rs3918242), MMP-9 R279Q (rs17576), and MMP-12 -82A/ G (rs2276109) polymorphisms, using primers and conditions as described in Supplemental Table S1. The digested fragments
Olsen et al. QRS Widening Rates and MMP in HF
347
Table 1. Demographic and clinical characteristics of patients
Age (y) White (%) Male sex (%) Cause of HF Ischemic (%) Idiopathic (%) Hypertensive (%) NYHA (n ¼ 179) Classes I and II (%) Classes III and IV (%) All-cause mortality (%) HF-related deaths (%) Comorbidities AMI (%) Hypertension (%) Diabetes mellitus (%) Current smoker (%) Laboratory variables Creatinine (mg/dL) Sodium (mEq/dL) Hemoglobin (g/dL) Baseline drugs b-blockers (%) ACEi (%)
All patients with HF (N ¼ 184)
QRS-WR < 0.5 ms/mo (n ¼ 139)
QRS-WR 0.5 ms/mo (n ¼ 45)
P value
62 12 125 (67.9) 124 (67.4)
62 12 98 (70.5) 90 (64.7)
59 13 27 (60.0) 34 (75.6)
0.17 0.20 0.20
56 (40) 42 (30) 32 (23)
15 (33) 15 (33) 9 (20)
0.48 0.71 0.83
(75.0) (25.0) (33.1) (15.8)
36 7 14 11
(83.7) (16.3) (31.1) (24.4)
0.30
(37) (66) (35) (14)
9 23 14 4
(20) (51) (31) (9)
0.04 0.07 0.71 0.60
71 (39) 57 (31) 41 (22) 138 41 60 33
(77.1) (22.9) (32.6) (17.9)
60 115 63 23
(33) (63) (34) (13)
102 34 46 22 51 92 49 19
0.86 0.19
1.3 0.4 140.5 3.3 13.0 1.6
1.3 0.4 140.6 3.2 13.1 1.5
1.3 0.5 140.0 3.5 12.7 1.9
0.96 0.31 0.15
161 (87.5) 161 (87.5)
118 (84.9) 123 (88.5)
43 (95.6) 38 (84.4)
0.07 0.44
Data are expressed as the mean standard deviation or as a number (percentage), as appropriate. ACEi, angiotensin-converting enzyme inhibitor; AMI, acute myocardial infarction; HF, heart failure; NYHA, New York Heart Association; QRS-WR, QRS widening rate.
were separated by electrophoresis in an 8% polyacrylamide gel and visualized after silver nitrate staining (MMP-1 and MMP-3 polymorphisms) or in a 2% agarose gel and visualized after ethidium bromide staining under ultraviolet light (MMP-9, MMP-2, and MMP-12 polymorphisms). Statistical analysis Continuous variables with normal distribution were compared using the Student t test and were expressed as the mean standard deviation, whereas continuous variables with asymmetrical distribution were compared using the Mann-Whitney U test and were expressed as the median and 25th-75th percentile. Categorical variables were compared using the c2 test or the Fisher exact test, as appropriate, and were expressed as absolute numbers and percentages. c2 was also used to test the Hardy-Weinberg equilibrium. For the prognostic analysis, Kaplan-Meier survival curves were constructed from the date of entry at the outpatient clinic up to the last registry of follow-up or death and were compared using the log-rank statistics. Cox proportional hazard models were created and adjusted for variables that are recognized as influencing HF-related outcomes. For this analysis, the model was adjusted for functional class, LVEF, and serum creatinine levels. All data were analyzed using SPSS, version 16.0, software for Windows (SPSS, Inc, Chicago, IL). A P value < 0.05 was considered statistically significant.
functional classes I and II (77%), with moderate to severe LV dysfunction (mean LVEF ¼ 31% 8%) and a mean age of 62 12 years. The main causes of HF were ischemic (39%) and idiopathic (31%); more than half of the patients had hypertension (63%), and approximately one third of the patients had experienced a previous myocardial infarction. Patients were receiving the standard-of-care drug treatment; almost 90% were using angiotensin-converting enzyme inhibitors and beta-blockers. When patients were stratified according to QRS-WR, no difference in baseline characteristics was seen, except for a small increase in the presence of myocardial infarction in the QRS-WR 0.5 ms/month group (Table 1). Despite this, previous myocardial infarction was not independently associated with clinical combined events or HF-related deaths (data not shown). Table 1 describes the main clinical characteristics of the cohort. ECG data
Results
As shown in Table 2, the majority of the patients were in sinus rhythm (79%), and approximately one third had left bundle branch block at the time of inclusion. Overall, the median QRS0 duration was 110 ms (25th-75th percentile: 96150 ms), the median QRS1 duration was 121 ms (25th-75th percentile: 100-154 ms), and the median time difference between ECG assessments was 24.4 months (25th-75th percentile: 13.7-55.0 months). The median rate of QRS progression was 0.18 ms/month (25th-75th percentile: 0.00.49 ms/month). No statistically significant differences were observed in the baseline assessment when patients were stratified according to QRS-WR.
Demographic and clinical data of patients with HF
QRS-WR and clinical outcomes
Our HF cohort was predominantly composed of white (68%) male (67%) patients in New York Heart Association (NYHA)
A higher QRS-WR predicted a worse prognosis, whereas both the QRS0 and QRS1 were not related to future clinical
348
Canadian Journal of Cardiology Volume 30 2014
Table 2. ECG and echocardiographic data
Echocardiography LVEF (%) LVDD (cm) LVSD (cm) LA dimension (cm) Electrocardiography Sinus rhythm (%) Atrial fibrillation (%) LBBB (%) QRS0 duration (ms) QRS1 duration (ms) QRS-WR (ms/mo) ECG interval (mo)
All patients with HF (N ¼ 184)
QRS-WR < 0.5 ms/mo (n ¼ 139)
QRS-WR 0.5 ms/mo (n ¼ 45)
31 8 6.6 0.9 5.6 1.0 4.7 0.8
32 8 6.6 0.9 5.6 1.0 4.8 0.8
31 8 6.7 1.0 5.7 1.0 4.7 0.8
79 21 34 110 (96-150) 121 (100-154) 0.18 (0.00-0.49) 24.4 (13.7-55.0)
76 23 33 112 (98-152) 118 (99-152) 0.09 (0.13-0.25) 30.5 (16.3-63.0)
86 14 35 104 (94-136) 134 (112-154) 0.86 (0.64-1.47) 16.1 (9.5-31.7)
P value 0.68 0.33 0.51 0.51 0.20 0.28 0.97 0.10 0.03 < 0.01 < 0.01
Data are expressed as the mean standard deviation, the median (25th-75th percentile), or as a number (percentage), as appropriate. ECG, electrocardiogram; HF, heart failure; LA, left atrial; LBBB, left bundle branch block; LVDD, left ventricular diastolic diameter; LVEF, left ventricular ejection fraction; LVSD, left ventricular systolic diameter; QRS-WR, QRS widening rate.
outcomes. Patients with HF with a QRS-WR 0.5 ms/ month had an increased risk of HF-related death or combined clinical events when compared with patients with a QRS-WR < 0.5 ms/month (log-rank P ¼ 0.03 and 0.01, respectively) (Fig 1). After adjusting for other risk covariates in a Cox proportional hazard model, QRS-WR remained an independent predictor of combined clinical events (HR, 1.6; 95% CI, 1.1-2.5; P ¼ 0.02). However, QRS-WR was not associated with all-cause mortality (log-rank P ¼ 0.95). MMP genotypes, LV remodelling, and QRS-WR Patients were stratified according to the MMP-1 or MMP-3 genotype, and echocardiographic parameters were evaluated. There was no association between MMP genotype and ventricular remodelling (Supplemental Tables S2 and S3). For the analysis of the impact of the MMP genotypes on QRS-WR, patients with HF were stratified according to cause (ischemic and nonischemic), as shown in Table 3. The MMP-1 2G2G genotype was associated with nearly a 2-fold increase in QRS-WR (P ¼ 0.03). Such an effect was not evident when ischemic and nonischemic patients were analyzed separately. Despite the MMP-1 2G2G genotype being highly associated with QRS-WR, it was not associated with clinical combined events (HR, 0.8; 95% CI, 0.5-1.2; P ¼ 0.33). Conversely, patients with a MMP-3 5A5A genotype and a nonischemic cause were highly protected against QRS enlargement. No other MMP polymorphism was associated with QRS-WR, even when the patients were stratified by cause (Table 3). It is worth noting that the observed genotype frequencies were in Hardy-Weinberg equilibrium in the sample studied for all analyzed polymorphisms.
Figure 1. (A) Heart-failureerelated deaths and (B) combined clinical events for all patients according to their QRS-WR. Solid line represents patients with a QRS-WR < 0.5 ms/month (n ¼ 139), and the dashed line represents patients with a QRS-WR 0.5 ms/month (n ¼ 45).
Discussion In the present study, we investigated the rate of QRS widening and its prognostic value in patients with HF, as well as the impact of common MMP polymorphisms on QRS-WR. Our findings indicate that QRS-WR is an independent predictor of HF morbidity and mortality in a contemporary HF cohort. In addition, our results suggest that genetic polymorphisms of MMP-1 and MMP-3 may explain, in part, the interindividual variability of QRS-WR.
Olsen et al. QRS Widening Rates and MMP in HF
349
Table 3. QRS duration according to the MMP genotypes All patients (N ¼ 184) MMP-1 -1607 1G/2G 1G carriers (73/73/73)* 2G2G (27/27/27) P value MMP-2 -1575G/A A carriers (35/40/32) GG (65/60/68) P value MMP-2 -790G/T G carriers (38/44/34) TT (62/56/66) P value MMP-3 -1171 5A/6A 6A carriers (82/82/82) 5A5A (18/18/18) P value MMP-9 -1562C/T T carriers (20/21/19) CC (80/70/81) P value MMP-9 R279Q R carriers (50/52/50) QQ (50/48/50) P value MMP-12 -82A/G AA (84/74/90) AG (16/26/10) P value
Ischemic cause (n ¼ 71)
Nonischemic cause (n ¼ 113)
0.17 (0.04-0.42) 0.29 (0.07-0.63) 0.03
0.20 (0.07-0.40) 0.29 (0.09-0.66) 0.20
0.11 (0.11-0.48) 0.32 (0.04-0.64) 0.09
0.22 (0.00-0.44) 0.17 (0.02-0.53) 0.63
0.26 (0.11-0.44) 0.18 (0.00-0.45) 0.46
0.19 (0.10-0.50) 0.15 (0.08-0.57) 0.95
0.21 (0.00-0.56) 0.17 (0.04-0.50) 0.36
0.26 (0.10-0.45) 0.18 (0.01-0.41) 0.31
0.17 (0.08-0.75) 0.15 (0.08-0.53) 0.75
0.20 (0.00-0.54) 0.14 (0.09-0.31) 0.11
0.21 (0.03-0.40) 0.26 (0.10-0.45) 0.58
0.20 (0.05-0.70) 0.00 (0.44-0.25) 0.03
0.13 (0.00-0.38) 0.20 (0.02-0.52) 0.55
0.15 (0.08-0.25) 0.24 (0.00-0.44) 0.25
0.08 (0.12-0.76) 0.17 (0.08-0.55) 0.93
0.18 (0.00-0.54) 0.20 (0.02-0.44) 0.93
0.29 (0.12-0.44) 0.18 (0.01-0.47) 0.19
0.13 (0.12-0.43) 0.18 (0.00-0.59) 0.47
0.18 (0.02-0.45) 0.23 (0.07-0.65) 0.66
0.26 (0.15-0.43) 0.22 (0.10-0.64) 0.85
0.13 (0.11-0.52) 0.27 (0.15-0.94) 0.62
* Data are expressed as the median (25th-75th percentile) of the QRS widening rate (ms/mo). Numbers in parentheses indicate the genotype (percentage) in all patients, patients with an ischemic cause of HF, and patients with a nonischemic cause of HF, respectively.
A resting electrocardiogram is a basic tool in the initial evaluation of patients with HF. The prognostic role of QRS duration in a single resting electrocardiogram, however, is not consistent in different studies.14-17 Therefore, some authors have suggested that serial ECG monitoring may provide stronger prognostic information than isolated measurements because sequential data on QRS width could portray disease progression, adverse heart remodelling, and progressive impairment of electrical conductivity. Grigioni et al. analyzed QRS intervals in 82 patients with dilated cardiomyopathy and observed that the incidence of cardiac death or heart transplantation was higher among patients with QRS interval widening > 0.5 ms/month after 12 months of follow-up.4 QRS widening remained independently and unfavourably predictive of a worse prognosis after adjustment for baseline risks during multivariate analysis. In the present analysis, we demonstrated that patients with HF are prone to widening of the QRS over a median time of 24 months. Moreover, although isolated QRS0 and QRS1 measurements were not associated with future HF-related events, QRS-WR was independently associated with HF-related combined events after controlling for confounding variables. Despite optimized medical management, QRS widening still occurs over time and the intrinsic causes are unclear. Some mechanisms that are likely to be involved include progressive interstitial fibrosis,18 ischemia in the conduction tissue,5,19 LV dilatation,20 and genetically based influences.21 These findings indicate that substitution of an electrical competent tissue (cardiomyocyte) for another apparently less adequate tissue (fibrosis) may be the basis of QRS widening. MMPs play a pivotal role in heart remodelling, as described in experimental and clinical studies.22-24 Activation of MMPs
promotes changes in the extracellular matrix and an imbalance in collagen content, which can lead to disruptions in cardiomyocyte connectivity and impairment in tissue electrical conductivity. In the present analysis, we evaluated whether common genetic MMP polymorphisms could explain, at least in part, the interindividual variability of QRS-WR. Our results suggest that patients carrying the MMP-1 2G2G genotype had a significantly higher QRS-WR. MMP-1 acts on types I, II, III, VII, and X collagen, but its role in heart remodelling is still under debate. The 2G allele is associated with increased promoter activity of the MMP-1 gene, and previous studies have suggested that MMP-1 may have an impact on LV remodelling after myocardial infarction and the overall risk of future coronary heart disease.25,26 Our group has previously demonstrated that the MMP-1 2G allele was associated with better prognosis in an HF cohort,9 which seems to be contradictory to the increased QRS-WR shown in the present study. However, Velho et al. also demonstrated that after including the QRS in the multivariate analysis, the MMP-1 genotype was no longer associated with prognosis.9 This suggests that the electrical conductivity system and MMP-1 may act by different mechanisms in the course of the disease. Taken together, these findings suggest that the net final effect of isolated nucleotide polymorphisms per se on HF prognosis might be outweighed by other compensatory or regulatory pathways. Also, because most data are based on the analysis of specific predefined polymorphisms (and not genome-wide analysis), prognostic data must be interpreted with caution. Haplotype analysis and other more comprehensive genetic approaches might be able to unravel the overall role of genetic variability on HF prognosis. Although our results suggest that MMP genotypes impact on QRS-WR,
350
this was not translated into a consistent effect on HF prognosis. Increased levels of MMP-3 have been associated with cardiac remodelling in ischemic and nonischemic dilated cardiomyopathy.22,27 Interestingly, we found that the MMP-3 5A5A genotype was associated with a lower QRS-WR only in nonischemic patients, suggesting that the cause of HF can affect the molecular pathways that lead to LV remodelling. Despite the influence on the QRS-WR, neither MMP-1 nor MMP-3 genotypes were associated with ventricular dilatation or myocardial hypertrophy in the present study. To extend our comprehension of the role that the MMP genotype plays in QRS-WR progression, other polymorphisms were included in the analysis, namely MMP-2 -790G/T, MMP-2 -1575G/A, MMP-9 -1562C/T, MMP-9 R279Q, and MMP-12 -82A/G. Gelatinase is a general name given to both MMP-2 (gelatinase A) and MMP-9 (gelatinase B) because of their ability to hydrolyze gelatin. MMP-2 and -9 gene polymorphisms have been associated with HF development.12,28 Two polymorphisms in the promoter region of the MMP-2 gene (-790G/T and -1575G/A) were included in our study. Both variants have already been described as being functional in vitro,29 but their association with HF remains controversial.12,28 The MMP-9 -1562C/T polymorphism alters gene transcription efficiency, whereas the MMP-9 R279Q polymorphism is associated with the MMP-9 catalytic domain; both polymorphisms have been associated with cardiovascular diseases.30,31 Despite the biological importance described for each polymorphism analyzed, we found no association of the MMP-2 -790G/T, MMP-2 -1575G/A, MMP-9 -1562C/T, MMP-9 R279Q, or MMP-12 -82A/G polymorphisms with QRS-WR. Some aspects of our study design should be considered when evaluating the results. We found that QRS-WR is associated with major events in patients with HF and that patients carrying the MMP-1 2G2G genotype and MMP-3 6A allele had an increased QRS-WR. Despite being important, the translation of MMP polymorphisms into enzymatic activity, tissue remodelling, and eventually impairment in electrical conductivity was not the goal of this study. The simultaneous assessment of genetic polymorphisms and serum levels of MMPs/tissue inhibitors of metalloproteinase could be an interesting approach to obtain a more comprehensive view of the processes involved in heart remodelling and electrical impairment. Additionally, our relatively small sample size can limit the statistical power to identify other interactions. For this reason, our data on the impact of genetic polymorphisms on QRS-WR need further validation in an independent cohort. Conclusions We have observed that a QRS-WR 0.5 ms/month was associated with increased HF-related events and mortality in a contemporary cohort of patients with HF and systolic dysfunction. We tried to unravel potential mechanisms that might be partially responsible for QRS widening. The 2G2G genotype of the MMP-1 gene was associated with a higher QRS-WR, whereas the impact of the MMP-3 5A5A genotype was dependent on the cause of HF. Because QRS-WR is an independent predictor of prognosis in HF, further studies are
Canadian Journal of Cardiology Volume 30 2014
needed to elucidate why some patients are particularly prone to QRS widening. Genetic variability in the molecular processes involved in LV remodelling may explain this phenomenon to some extent. Funding Sources This work was supported by Fundo de Incentivo à Pesquisa e Eventos do Hospital de Clínicas de Porto Alegre (FIPE-HCPA), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES). Disclosures The authors have no conflicts to disclose. References 1. Rubart M, Zipes DP. Genesis of Cardiac Arrhythmias: Electrophysiological Considerations. In: Braunwald E, ed. Braunwald’s Heart Disease: A Textbook of Cardiovascular Medicine. Philadelphia: Saunders, 2008: 727-61. 2. Beck-da-Silva L, Goldraich L, Bonzanini L, et al. Pulse pressure and QRS width evaluation as an inexpensive tool for heart failure assessment. Congest Heart Fail 2009;15:222-7. 3. Kalra PR, Sharma R, Shamim W, et al. Clinical characteristics and survival of patients with chronic heart failure and prolonged QRS duration. Int J Cardiol 2002;86:225-31. 4. Grigioni F, Carinci V, Boriani G, et al. Accelerated QRS widening as an independent predictor of cardiac death or of the need for heart transplantation in patients with congestive heart failure. J Heart Lung Transplant 2002;21:899-902. 5. Shamim W, Yousufuddin M, Cicoria M, et al. Incremental changes in QRS duration in serial ECGs over time identify high risk elderly patients with heart failure. Heart 2002;88:47-52. 6. Gajarsa JJ, Kloner RA. Left ventricular remodeling in the post-infarction heart: a review of cellular, molecular mechanisms, and therapeutic modalities. Heart Fail Rev 2011;16:13-21. 7. Medley TL, Kingwell BA, Gatzka CD, Pillay P, Cole TJ. Matrix metalloproteinase-3 genotype contributes to age-related aortic stiffening through modulation of gene and protein expression. Circ Res 2003;92: 1254-61. 8. Langers AM, Verspaget HW, Hommes DW, Sier CF. Single-nucleotide polymorphisms of matrix metalloproteinases and their inhibitors in gastrointestinal cancer. World J Gastrointest Oncol 2011;3:79-98. 9. Velho FM, Cohen CR, Santos KG, et al. Polymorphisms of matrix metalloproteinases in systolic heart failure: role on disease susceptibility, phenotypic characteristics, and prognosis. J Card Fail 2011;17:115-21. 10. Mizon-Gerard F, de Groote P, Lamblin N, et al. Prognostic impact of matrix metalloproteinase gene polymorphisms in patients with heart failure according to the aetiology of left ventricular systolic dysfunction. Eur Heart J 2004;25:688-93. 11. Tang LJ, Chen XF, Zhu M, et al. Matrix metalloproteinase-1, -3, and -9 gene polymorphisms and the risk of idiopathic dilated cardiomyopathy in a Chinese Han population. Clin Biochem 2007;40:1427-30. 12. Hua Y, Song L, Wu N, et al. Polymorphisms of MMP-2 gene are associated with systolic heart failure prognosis. Clin Chim Acta 2009;404:119-23.
Olsen et al. QRS Widening Rates and MMP in HF
351
13. Lahiri DK, Nurnberger JI. A rapid non-enzymatic method for the preparation of HMW DNA from blood for RFLP studies. Nucleic Acids Res 1991;19:5444.
23. Rohde LE, Ducharme A, Arroyo LH, et al. Matrix metalloproteinase inhibition attenuates early left ventricular enlargement after experimental myocardial infarction in mice. Circulation 1999;99:3063-70.
14. Dhar R, Alsheikh-Ali AA, Estes NAM, et al. Association of prolonged QRS duration with ventricular tachyarrhythmias and sudden cardiac death in the Multicenter Automatic Defibrillator Implantation Trial II (MADIT-II). Heart Rhythm 2008;5:807-13.
24. Cadete VJ, Arcand SA, Chaharyn BM, et al. Matrix metalloproteinase-2 is activated during ischemia/reperfusion in a model of myocardial infarction. Can J Cardiol 2013;29:1495-503.
15. Yerra L, Anavekar N, Skali H, et al. Association of QRS duration and outcomes after myocardial infarction: the VALIANT trial. Heart Rhythm 2006;3:313-6. 16. Katritsis DG, Siontis KC, Bigger JT, et al. Effect of left ventricular ejection fraction and QRS duration on the survival benefit of implantable cardioverter-defibrillators: Meta-analysis of primary prevention trials. Heart Rhythm 2013;10:200-6. 17. Exner DV, Birnie DH, Moe G, et al. Canadian Cardiovascular Society guidelines on the use of cardiac resynchronization therapy: evidence and patient selection. Can J Cardiol 2013;29:182-95. 18. Song BG, Yang HS, Hwang HK, et al. Correlation of electrocardiographic changes and myocardial fibrosis in patients with hypertrophic cardiomyopathy detected by cardiac magnetic resonance imaging. Clin Cardiol 2013;36:31-5. 19. Sheridan DJ, Culling W, Penny WJ. Electrophysiological disturbances associated with acute myocardial infarction. Eur Heart J 1986;(7 suppl B): 11-8. 20. Mazzoleni A, Curtin ME, Wolff R, Reiner L, Somes G. On the relationship between heart weights, fibrosis, and QRS duration. J Electrocardiol 1975;8:233-6. 21. Royer A, van Veen TAB, Le Bouter S, et al. Mouse model of SCN5Alinked hereditary Lenègre’s disease: age-related conduction slowing and myocardial fibrosis. Circulation 2005;111:1738-46. 22. Spinale FG, Coker ML, Heung LJ, et al. A matrix metalloproteinase induction/activation system exists in the human left ventricular myocardium and is upregulated in heart failure. Circulation 2000;102: 1944-9.
25. Martin TN, Penney DE, Smith JA, et al. Matrix metalloproteinase-1 promoter polymorphisms and changes in left ventricular volume following acute myocardial infarction. Am J Cardiol 2004;94:1044-6. 26. Ye S, Gale CR, Martyn CN. Variation in the matrix metalloproteinase-1 gene and risk of coronary heart disease. Eur Heart J 2003;24:1668-71. 27. Wilson EM, Moainie SL, Baskin JM, et al. Region- and type-specific induction of matrix metalloproteinases in post-myocardial infarction remodeling. Circulation 2003;107:2857-63. 28. Vasku A, Goldbergova M, Holla LI, et al. Two MMP-2 promoter polymorphisms (-790T/G and -735C/T) in chronic heart failure. Clin Chem Lab Med 2003;41:1299-303. 29. Price SJ, Greaves DR, Watkins H. Identification of novel, functional genetic variants in the human matrix metalloproteinase-2 gene: role of Sp1 in allele-specific transcriptional regulation. J Biol Chem 2001;276: 7549-58. 30. Zhang B, Ye S, Herrmann SM, et al. Functional polymorphism in the regulatory region of gelatinase B gene in relation to severity of coronary atherosclerosis. Circulation 1999;99:1788-94. 31. Zhi H, Wang H, Ren L, et al. Functional polymorphisms of matrix metallopeptidase-9 and risk of coronary artery disease in a Chinese population. Mol Biol Rep 2010;37:13-20.
Supplementary Material To access the supplementary material accompanying this article, visit the online version of the Canadian Journal of Cardiology at www.onlinecjc.ca and at http://dx.doi.org/10. 1016/j.cjca.2013.11.014.