We wanted to evaluate the prevalence and prognostic value of the fragmented QRS (fQRS) complex, defined as changes in QRS morphology with various RSR′-patterns in 12-lead electrocardiogram (ECG) in a middle-aged general population. We evaluated the 12-lead ECGs of 10,904 Finnish middle-aged subjects (52% men, mean age 44 ± 8.5 years) with (n = 2,543) and without (n = 8,361) an evidence of cardiac disease drawn from general population and followed them for 30 ± 11 years. Fragmentation of the QRS complex was defined as various RSR′-patterns in at least 2 consecutive leads within the same territory (inferior II, III, aVF; lateral I, aVL, V4 to V6; anterior V1 to V3). Primary end points were death from any cause, cardiac, and arrhythmic deaths. In the total population, fQRS was present in 19.7% (n = 2,147) of subjects, including 15.7% (n = 1714) in inferior leads, 0.8% (n = 84) in lateral leads, and 2.9% (n = 314) in anterior leads. Fragmentation was not associated with increased mortality in subjects without a known cardiac disease. However, fQRS observed in lateral leads in subjects with an evidence of cardiac disease was associated with an increased risk of all-cause (p = 0.001), cardiac (p = 0.001), and arrhythmic (p = 0.004) mortalities. In conclusion, fQRS reflecting minor intraventricular conduction defect is a common finding, especially in the inferior leads, but it is not a sign of increased risk of mortality in subjects without a known cardiac disease. Lateral fQRS, which is less commonly observed in the ECG, is associated with a worse outcome in patients with a known cardiac disease.
Fragmented QRS (fQRS) complex in a 12-lead electrocardiogram (ECG) has been shown to represent inhomogenous activation of the ventricles in the infarcted myocardium. Recent studies have focused on assessing the prognostic significance of fQRS mainly in patients with coronary artery disease (CAD). In general, the presence of fQRS has been shown to be an independent risk factor of cardiac events and lower event-free survival in patients with CAD. As far as we are aware, no studies have investigated the prevalence and significance of fQRS in the general population. Therefore, we set out to study the prevalence and long-term prognostic significance of fQRS in a large Finnish middle-aged general population consisting of 5,703 men and 5,201 women. This study was made following the guidelines of the Declaration of Helsinki.
Methods
Consent was obtained at baseline from patients to participate in the Social Insurance Institution’s coronary heart disease study in Finland from 1966 to 1972. The original cohort consisted of 12,310 subjects (participation rate 89%), aged 30 to 59 years. The coronary heart disease study was a part of a larger prospective Automobile Health Survey study performed using 35 populations drawn from different geographic regions in Finland to obtain a representative sample of Finnish middle-aged general population. All ECGs from the original cohort available were used in this study, thus the final population consisting of 10,904 subjects (52.3% men). The baseline examination was completed as described in the earlier publication including ECG recordings; blood pressure; body mass index and cholesterol measurements; and questions about health habits, present diseases, and use of medication. All symptoms and cardiovascular risk factors were also documented. Because the prognostic significance of fQRS has been previously reported in patients with a structural cardiac disease, we performed the analysis of the prevalence and prognosis separately for those with and without an evidence of cardiac disease. A cardiac disease was defined as the use of cardiac medication, clinical history of previous myocardial infarction (MI), documented CAD or congestive heart failure, or ECG signs of CAD according to Minnesota codes (1.1 to 1.3, 4.1 to 4.3, 5.1 to 5.2, 7.1, and 7.4). Subjects with complete or partial bundle branch block (n = 227) and missing or otherwise unreadable ECGs were excluded (n = 53). To avoid overlapping between fQRS and early repolarization (ER), we also excluded subjects with an ER ECG pattern with a terminal notching of the QRS complex in lateral/inferior leads (n = 215). After exclusion criteria, 8,277 subjects without and 1,975 subjects with evidence of cardiac disease were included.
At baseline, resting 12-lead ECGs were recorded with a paper speed of 50 mm/s. At the time of baseline examinations, ECGs were analyzed by 9 professionals for bundle branch blocks and left ventricular hypertrophy (according to the Sokolow-Lyon criteria), and QT-interval was also measured (corrected for heart rate with Bazett’s formula). For the present study, ECGs were read by 2 independent readers (H.T. and J.T.) blinded to outcome data for the presence of fQRS, and repeatability procedures were performed. Kappa values were 0.81 for fQRS in any lead, 0.82 for inferior, 0.70 for lateral, and 0.89 for anterior fQRS. In those ECGs with conflicting coding, a consensus decision was used.
Our ECG criteria for fQRS followed those described by Das et al in 2006. We divided fQRS into anterior (V1 to V3), inferior (II, III, and aVF), and lateral (I, aVL, and V4 to V6) territories corresponding to the major coronary arteries. Lead aVR was ignored. Changes in the morphology of the QRS complex had to be seen in ≥2 consecutive leads for each territory (inferior, lateral, and anterior), and patients with more than 1 territory involved were also identified. Different RSR′ patterns included ≥1 additional R wave (R′) or notching of the S or R wave.
The baseline measurements were performed from 1966 to 1972. Patients were followed for 30 ± 11 years until the end of 2007. During this follow-up period, <2% of the subjects were lost as a result of moving abroad, but even from these patients, the majority of details of the survival status could be determined. The primary end point was death from cardiac causes, whereas secondary end points were arrhythmic and death from any cause. The causes and rates of death were determined from death certificates from Statistics Finland. We reviewed all cases of cardiac death as separate cardiac and arrhythmic death. Original death causes were evaluated by a committee of qualified and experienced cardiologists unaware of the ECG analysis. Using death certificates and hospital records, deaths were defined either as arrhythmic or as not arrhythmic following the definitions demonstrated in the Cardiac Arrhythmia Pilot Study and described previously by our study group.
All continuous variables are presented as means ± SD. The general regression model was used to compare the age- and sex-adjusted mean values for continuous variables and the prevalence of categorical variables among different groups. We added age, body mass index, systolic blood pressure, QTc and QRS durations, and cholesterol levels as continuous variables and smoking and gender as categorical variables. Two multivariate models were performed to evaluate prognostic significance of fQRS. First model included subjects with no clinical or ECG evidence of cardiac disease. We included subjects with clinical history or ECG signs of cardiac disease to our second analysis. Cox proportional-hazard model was used to calculate hazard ratios and 95% confidence intervals. Subjects with absent fQRS pattern served as the reference group. Kaplan-Meier survival curves were plotted for subjects with no evidence of cardiac disease and fQRS in any lead and also for those with lateral fQRS with clinical and/or ECG evidence of cardiac disease. Curves were compared by means of the log-rank test. Statistical analyses were conducted using SAS software, version 9.1.3 (SAS Institute, Espoo, Finland) and with the Statistical Package for Social Studies, version 14.0. All 2-sided p values <0.05 were considered to represent statistical significance.
Results
Figure 1 shows various morphologies of the fQRS pattern. In Figure 2 , we have an example of 12-lead ECG of a subject with inferior fQRS. Fragmentation of the QRS complex was present in 19.7% (n = 2,147) of the subjects. Inferior fQRS was present in 15.7% (n = 1,714), anterior in 2.9% (n = 316), and lateral in 0.8% (n = 84) of the subjects. There were some differences in the baseline characteristics between the subjects with and without fQRS. In general, subjects with fQRS were more commonly men, somewhat older, had a higher body mass index and systolic and diastolic blood pressure, and had a longer QRS duration than those without fQRS. A demographic comparison of the fQRS and non-fQRS groups is listed in Table 1 . During the follow-up of 30 ± 11 years, a total of 6,159 (56.5%) subjects died. Of these, 1,981 (32.2% of all deaths) died from cardiac causes, and 802 deaths (13.0%) were due to sudden arrhythmic causes.
| Characteristic | fQRS | p Values | |||||||
|---|---|---|---|---|---|---|---|---|---|
| − | + | Inferior | Anterior | Lateral | Total | Inferior | Anterior | Lateral | |
| Number of subjects | 8757 (80.3%) | 2147 (19.7%) | 1714 (15.7%) | 316 (2.9%) | 84 (0.8%) | ||||
| Male ∗ | 50.2% | 61.0% | 62.1% | 55.1% | 63.1% | <0.001 | <0.001 | 0.08 | 0.02 |
| Age (years) † | 43.8 ± 8.4 | 44.9 ± 8.7 | 44.7 ± 8.5 | 44.5 ± 9.0 | 45.6 ± 9.3 | <0.001 | <0.001 | 0.15 | 0.04 |
| Current smoker ‡ | 33.8% | 34.5% | 34.9% | 33.2% | 38.1% | 0.001 | 0.002 | 0.32 | 0.91 |
| Cholesterol mmol/l ‡ | 6.49 ± 1.31 (251 ± 50.7 mg/dl) | 6.53 ± 1.34 (253 ± 51.8 mg/dl) | 6.54 ± 1.36 (253 ± 52.6 mg/dl) | 6.52 ± 1.24 (252 ± 48.0 mg/dl) | 6.68 ± 1.45 (258 ± 56.1 mg/dl) | 0.81 | 0.72 | 0.91 | 0.34 |
| Body mass index (kg/m 2 ) ‡ | 25.8 ± 3.9 | 26.4 ± 3.7 | 26.6 ± 3.6 | 25.6 ± 3.8 | 27.0 ± 4.4 | <0.001 | <0.001 | 0.31 | 0.01 |
| Heart rate (bpm) ‡ | 76 ± 15 | 75 ± 15 | 75 ± 15 | 75 ± 16 | 75 ± 16 | 0.20 | 0.26 | 0.81 | 0.92 |
| Blood pressure (mm Hg) ‡ | |||||||||
| Systolic | 138 ± 21 | 140 ± 22 | 140 ± 22 | 136 ± 20 | 141 ± 25 | 0.29 | 0.02 | 0.01 | 0.66 |
| Diastolic | 82 ± 12 | 83 ± 13 | 84 ± 12 | 81 ± 11 | 83 ± 13 | 0.003 | <0.001 | 0.03 | 0.78 |
| ECG left ventricular hypertrophy ‡ | 31.9% | 28.9% | 30.3% | 30.4% | 32.4% | <0.001 | <0.001 | 0.27 | 0.50 |
| QTc (ms) | 408 ± 27 | 408 ± 28 | 408 ± 28 | 409 ± 29 | 413 ± 31 | 0.60 | 0.97 | 0.73 | 0.11 |
| QRS (ms) | 84 ± 11 | 88 ± 11 | 88 ± 11 | 87 ± 11 | 87 ± 11 | <0.001 | <0.001 | 0.001 | 0.05 |
| JTc (ms) | 322 ± 28 | 319 ± 30 | 320 ± 29 | 320 ± 30 | 325 ± 31 | <0.001 | 0.04 | 0.58 | 0.30 |
Of the subjects included, a total of 8,277 had no clinical or ECG evidence of cardiac disease. Inferior fQRS was present in 1,290 (15.6%), anterior in 237 (2.9%), and lateral in 45 (0.5%) subjects. Mortality rates are listed in Table 2 . Only a marginal increased risk of all-cause mortality was observed in subjects with inferior fQRS, but after multivariate adjustments, the difference did not remain significant. Lateral and anterior fQRS did not increase the risk of any of the end points in subjects considered free of CAD ( Table 2 ). None of the fQRS groups predicted sudden arrhythmic death.
| Variable | fQRS in Any Lead (n = 1518) | p Value | fQRS Inferior (n = 1290) | p Value | fQRS Anterior (n = 237) | p Value | fQRS Lateral (n = 45) | p Value |
|---|---|---|---|---|---|---|---|---|
| Cardiac death | ||||||||
| No. of death | 257 (17%) | 217 (17%) | 35 (15%) | 7 (16%) | ||||
| Univariate RR (95% CI ∗ ) | 1.1 (96–1.27) | 0.17 | 1.1 (0.98–1.31) | 0.09 | 0.9 (0.70–1.40) | 0.88 | 1.1 (0.51–2.24) | 0.87 |
| Multivariate RR † (95% CI ∗ ) | 0.9 (0.81–1.07) | 0.32 | 0.9 (0.79–1.06) | 0.24 | 1.0 (0.75–1.41) | 0.77 | 1.3 (0.53–2.33) | 0.79 |
| Arrhythmic death | ||||||||
| No. of death | 100 (6.8%) | 90 (7.0%) | 14 (5.9%) | 2 (4.4%) | ||||
| Univariate RR (95% CI ∗ ) | 1.1 (0.90–1.39) | 0.33 | 1.2 (0.92–1.45) | 0.22 | 1.1 (0.56–1.63) | 0.88 | 0.74 (0.18–2.95) | 0.67 |
| Multivariate RR † (95% CI ∗ ) | 0.9 (0.75–1.17) | 0.57 | 0.9 (0.74–1.17) | 0.54 | 1.0 (0.61–1.77) | 0.90 | 0.77 (0.19–3.08) | 0.71 |
| Death from any cause | ||||||||
| No. of death | 791 (54%) | 698 (54%) | 123 (52%) | 25 (56%) | ||||
| Univariate RR (95% CI ∗ ) | 1.1 (0.99–1.15) | 0.11 | 1.1 (1.0–1.18) | 0.05 | 1.0 (0.85–1.22) | 0.85 | 1.1 (0.77–1.68) | 0.53 |
| Multivariate RR † (95% CI ∗ ) | 0.9 (0.87–1.02) | 0.15 | 0.9 (0.86–1.01) | 0.09 | 1.0 (0.87–1.26) | 0.61 | 1.1 (0.75–1.66) | 0.60 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
