ECG changes in factory workers exposed to 27.2 MHz radiofrequency radiation
Abstract
To research the effect of 27.2 MHz radiofrequency radiation on electrocardiograms (ECG), 225 female workers operating radiofrequency machines at a shoe factory were chosen as the exposure group and 100 female workers without exposure from the same factory were selected as the control group. The 6 min electric field strength that the female workers were exposed to was 64.0 ± 25.2 V/m (mean ± SD), which exceeded 61 V/m, the International Commission on Non-Ionizing Radiation Protection reference root mean square levels for occupational exposure. A statistical difference was observed between the exposed group and the control group in terms of the rate of sinus bradycardia (χ2 = 11.48, P = 0.003). When several known risk factors for cardiovascular disease were considered, including smoking, age, alcohol ingestion habit, and so on, the exposure duration was not an effective factor for ECG changes, sinus arrhythmia, or sinus bradycardia according to α = 0.05, while P = 0.052 for sinus arrhythmia was very close to 0.05. We did not find any statistical difference in heart rate, duration of the QRS wave (ventricular depolarization), or corrected QT intervals (between the start of the Q wave and end of the T wave) between the exposed and control groups. Occupational exposure to radiofrequency radiation was not found to be a cause of ECG changes after consideration of the confounding factors. Bioelectromagnetics 34:285–290, 2013. © 2012 Wiley Periodicals, Inc.
INTRODUCTION
The effect of electromagnetic fields (EMF) on the circulatory and nervous systems has been the subject of great interest for many years, since electric impulses generated in these systems by outer electric and magnetic fields can theoretically disturb their functions [Bortkiewicz et al., 1996]. Gordon [1966] demonstrated that if the intensity of microwave exposure was 100 µW/cm2 or more, functional changes in the cardiovascular system could be observed, such as a decrease in blood pressure and heart rate (HR), as well as electrocardiogram (ECG) changes. Huber et al. [2003] exposed volunteers to 900 MHz radiofrequency radiation before sleep and observed that the HR of the volunteers decreased during the first stage of sleep and at the moment of awakening. Regarding the long-term effects of radiofrequency radiation, some researchers concluded that radiofrequency electromagnetic radiation (EMR) contributed to adverse health effects on the cardiovascular system [Vangelova et al., 2006]. Due to the lack of sufficient information on exposure and confounding factors, further research is necessary to elucidate the relationship between long-term exposure to radiofrequency radiation and adverse health effects on the cardiovascular system.
The aim of the current study was to investigate whether long-term exposure to high intensity radiofrequency EMF was associated with ECG changes. One advantage that our study had over previous studies was that the exposure level of radiofrequency workers was high and steady [Grajewski et al., 2000]. Another advantage of our current work was that we considered several known risk factors for cardiovascular disease including smoking, age, alcohol ingestion habit, history of past illness, and hereditary disease, which were surveyed by questionnaire and controlled by statistics. We also aimed to evaluate the relationship between exposure duration and ECG changes. Specifically speaking, the goal of our study was to examine whether individuals were susceptible to ECG changes with increased duration of EMF exposure.
MATERIALS AND METHODS
Subjects
In total, 225 female machine workers in a shoe factory who were exposed to radiofrequency radiation took part in the study as the exposed group. Because there were just 10 male workers in the workshop, and females tend to be more sensitive to hazards, male subjects were not utilized in this study. One person was excluded because of a history of hyperthyroidism. A total of 100 female workers were selected as the control group. These women cut or sewed shoe materials at another workshop in the same factory without exposure to radiofrequency radiation or other toxic compounds. In the control group, two people were excluded because they had been exposed to radiofrequency radiation during the past 2 years. Overall, 322 female workers were included in the study: 224 workers in the exposed group and 98 workers in the control group.
Questionnaires
We surveyed the workers for general information such as age, work shift, occupational history, smoking and drinking status, history of past illness, hereditary disease, and so on, by questionnaire. The questionnaire was created by researchers and filled out by physicians in a face-to-face interview with the workers.
Assessment of Exposure
In this study, there were 80 radiofrequency machines in the radiofrequency workshop. The work frequency of the 80 machines was 27.2 MHz (all frequencies of these types of radiofrequency machines are set at 27.2 MHz in China). The highest voltage was 8 kV in 30 out of the 80 machines, and 4 kV for the remaining 50 machines. Generally, one to two workers operated one machine at different sites (Fig. 1). All workers were divided into two shifts, the morning shift or the night shift. The workers from either shift did the same task repeatedly and welded plastic materials or tagged one shoe after another for 8 h/day, 5–6 days/week. They usually took 30 s to finish one welding task and welded several hundreds shoes a day.
1–2 workers operated one machine at different sites.
According to the Chinese National Occupational Health Standard (i.e., measurement of physical hazards in the workplace), at least four operating posts should be selected for exposure-level assessment when the number of the same type of operating posts reaches 10 or more; in this study, all operating posts were the same. We then measured the exposure levels of 27 operating posts at random, which consisted of one-third of all 80 machines in the factory. Among the 27 measured posts, 12 operating posts of 8 kV machines and 15 operating posts of 4 kV machines were selected for exposure assessment.
According to the International Commission on Non-Ionizing Radiation Protection, the reference root mean square (RMS) level for occupational exposure to time-varying electric fields from 10 to 400 MHz is 61 V/m. For frequencies between 100 and 10 GHz, the RMS can be calculated by averaging over any 6-min period (during operation). Therefore, we assessed the exposure level using 6 min RMS of electric field strength. We measured the 6 min RMS electric field strength three times for every post at the time when the workload had been fully engaged, and then the mean value of three measurements was taken as the exposure of the post. Lastly, the exposure level of the exposed group was calculated by averaging the electric field strength of 27 measured posts. All measurements of the 27 operating posts were made using a PMM 8053A portable field meter with an EHP50 three-axial probe (PMM Construction Center for Electronic Radio Measurements, Cisano Sul Neva, Italy). This instrument is capable of taking simultaneous three-axial measurements of radiofrequency electric fields and providing an RMS vector magnitude of the electric field strength. Spot measurements were performed at the height of the chest (near the heart), about 1.2 m from ground because the workers were seated when they were working. In addition to mean 6 min RMS electric field strength, exposure duration is considered to be another important exposure metric. Thus, we obtained information such as exposure duration, work shift duration, and exposure years through use of the questionnaire.
ECG
Electrocardiograms (ECG) were recorded and analyzed by three licensed doctors who had practiced ECG for more than 3 years at the Guangdong Prevention and Cure Centre of Occupational Diseases, Guangzhou, China. Four parameters were derived from the ECG: HR, duration of QRS wave (ventricular depolarization), duration of PR (beginning of the P wave [atrial depolarization] to the beginning of the QRS complex), and duration of QT intervals (between the start of the Q wave and end of the T wave [repolarization of the ventricles]). The duration of the QT interval was corrected (QTc) using the Bazelt equation, which gave the QTc for a HR of 60 [Goldman, 1970].
Data Analysis
In this study, we input the data with Epidata 3.0 (Chinese Center for Disease Control and Prevention, Peking, China). Using the Epidata program, we set the numerical value limit, number of bytes, and so on as a method of quality control. Two people entered the entire dataset at different times, and we looked for the differences between them. Finally, we confirmed the reasons for the differences and corrected any erroneous data.
In this study, percentage was used to describe the components of age, exposure duration, drinking status, and ECG changes in each group (Table 1). For discussing the relationship between exposure duration and ECG changes, all workers were divided into three groups according to the exposure duration: 1, control group; 2, those who had been exposed <2 years; and 3, those that had been exposed ≥2 years. We used binary logistic regression to estimate the association between radiofrequency radiation exposure and ECG changes, taking age and drinking status in the binary logistic regression at the same time. A one-way multivariate analysis of covariance (MANCOVA) was conducted to determine the effect between the duration of exposure to radiofrequency radiation and the four ECG parameters.
| Factors | Control group n (%) | Exposed groups n (%) | ||
|---|---|---|---|---|
| 3 Months to 2 years | More than 2 years | Total | ||
| n | 98 (100) | 145 (100) | 79 (100) | 224 (100) |
| Age (years old) | ||||
| ≤20 | 8 (8.2) | 44 (30.3) | 11 (13.9) | 55 (24.6) |
| 21–30 | 70 (71.4) | 83 (57.2) | 61 (7.7) | 144 (64.3) |
| 31–40 | 20 (20.4) | 18 (12.4) | 7 (8.9) | 25 (11.2) |
| Smoking | 2 (2.0) | 1 (0.7) | 0 | 1 (0.4) |
| Drinking alcohol | 4 (4.1) | 4 (2.8) | 2 (2.5) | 6 (2.7) |
| Number of ECG changes | 18 (18.4) | 35 (24.1) | 22 (27.8) | 57 (25.4) |
| Sinus bradycardia | 8 (8.2) | 16 (11.0) | 15 (19.0) | 31 (13.8) |
| Sinus arrhythmia | 2 (2.0) | 7 (4.8) | 7 (8.9) | 14 (6.3) |
| Others | 8 (8.2) | 14 (9.7) | 5 (6.3) | 19 (8.5) |
RESULTS
In our study, the ages of the workers ranged from 18 to 48, and the mean age of the exposed group (24.8 ± 4.7 years) was lower than that of the control group (27.9 ± 5.8 years). In the exposed group, the 6 min RMS of electric fields strength was 64.0 ± 25.2 V/m (mean ± SD); the minimum was 36.0 V/m and the maximum was 106.1 V/m. The mean was also higher than the reference levels for occupational exposure (61 V/m). In addition, the mean 6 min RMS exposure level in this study was more than double the reference level for general public exposure to electric fields from 10 to 400 MHz (28 V/m). The mean exposure duration of radiofrequency radiation of the subjects was 22.8 ± 17.8 months (mean ± SD). The exposed group was divided into two groups according to exposure duration. In the first group, 145 workers (64.7%) were exposed to radiofrequency radiation for <2 years, and in the second group, 79 workers (35.3%) had been exposed for ≥2 years. In the second group, there were 25 workers (31.6%) who had been exposed for 3 years and above, 14 workers (17.7%) who had been exposed for 5 years and above, and 1 person (1.3%) who had been exposed for 10 years. Only two of the women (2.0%) smoked in the control group and no one smoked in the exposed group. Six people (6.3%) drank alcohol in the control group, and seven people (3.1%) drank alcohol in the exposed group. Eight people (3.6%) in the exposed group and three people (3.1%) in the control group had a history of illness (appendicitis, nephritis, etc.).
The results of this study demonstrated that 18 workers (18.4%) had abnormal ECGs in the control group compared with 57 workers (25.4%) in the exposed group; however, this difference was not statistically significant (P > 0.05). In total, 12 types of ECG abnormalities were observed: sinus arrhythmia, sinus bradycardia, sinus tachycardia, right axis deviation, left ventricular high voltage, right axis deviation, left axis deviation, pre-excitation syndrome, limb lead low voltage, left auricular rhythm, atrial premature rhythm, and ventricular premature rhythm. Sinus arrhythmia was the most common abnormality among the female workers in this study. The rates of sinus arrhythmia were 8.2 and 13.8% in the control group and exposed group, respectively. The results also revealed that 14 workers (6.3%) had sinus bradycardia in the exposed group, while only 2 workers (2.0%) in the control group had sinus bradycardia; however, the differences were not statistically significant.
The rate of ECG abnormalities in the group exposed for the longest duration was 27.8%, which was the highest. The rates were 24.1% in the group exposed for <2 and 18.4% in the control group. No statistically significant differences were observed among the three groups (χ2 = 2.31, P = 0.32). If α = 0.05 was the level of test, the rates of sinus arrhythmia among the three groups were not significantly different (χ2 = 5.1, P = 0.078). The rates were 8.2, 11.0, and 19.0% in the control group, the group exposed for <2 years, and the group exposed for ≥2 years, respectively. The results of this study also revealed statistically significant differences among the rates of sinus bradycardia among the three groups (χ2 = 11.48, P = 0.003).
Three binary logistic regression analyses were conducted to determine the effect of ECG changes, sinus arrhythmia, and sinus bradycardia using exposure duration, age, and drinking status as the covariates. Age was determined to be an effective factor for ECG changes, sinus arrhythmia, and sinus bradycardia. Young people tended to have more ECG changes. Drinking status was not found to be an effective factor for ECG change in the female workers. As can be seen in Table 2, the odds ratio (OR) values were higher with the prolongation of exposure duration; however, exposure duration was not an effective factor for ECG changes, sinus arrhythmia, or sinus bradycardia according to α = 0.05.
| ECG change | Sinus arrhythmia | Sinus bradycardia | ||||
|---|---|---|---|---|---|---|
| OR (95% CI) | P | OR (95% CI) | P | OR (95% CI) | P | |
| Exposure duration | ||||||
| 0 | 1 | 0.32 | 1 | 0.05 | 1 | 0.11 |
| 3 months to 2 years | 1.42 (0.39, 5.23) | 0.60 | 1.42 (0.29, 6.84) | 0.67 | 2.40 (0.22, 25.83) | 0.47 |
| Above 2 years | 1.98 (0.66, 5.91) | 0.22 | 3.12 (0.83, 11.80) | 0.09 | 5.71 (0.73, 44.46) | 0.96 |
| Age (years old) | ||||||
| 19 | 1 | 0.00 | 1 | 0.03 | 1 | 0.04 |
| 20–29 | 0.30 (0.16, 0.59) | 0.00 | 0.33 (0.14, 1.16) | 0.09 | 0.24 (0.07, 0.79) | 0.02 |
| ≥30 | 0.25 (0.10, 0.66) | 0.01 | 0.41 (0.13, 1.34) | 0.14 | 0.16 (0.02, 1.43) | 0.10 |
| Drinking alcohol | 0.57 (0.15, 2.84) | 0.48 | 1.41 (0.25, 6.22) | 0.67 | — | — |
| Constant | 0.87 (0.16, 9.79) | 0.89 | 0.117 (0.01, 3.12) | 0.72 | — | — |
Finally, an MANCOVA was conducted to determine the effect of the level of exposure to radiofrequency radiation on the four ECG parameters using individual exposure duration (in years) as the covariate. Statistical analysis revealed no significant differences among the HR, duration of the QRS wave, PR intervals, or QTc intervals, as shown in Table 3.
| ECG | Control group (mean ± SD) | Exposed groups (mean ± SD) | |
|---|---|---|---|
| 3 Months to 2 years | More than 2 years | ||
| Heart rate (number/min) | 71.3 ± 8.7 | 71.5 ± 9.7 | 69.8 ± 9.2 |
| PR (ms) | 0.146 ± 0.016 | 0.148 ± 0.026 | 0.144 ± 0.016 |
| QRS (ms) | 0.069 ± 0.01 | 0.071 ± 0.01 | 0.072 ± 0.01 |
| QTc (ms) | 0.388 ± 0.03 | 0.383 ± 0.03 | 0.382 ± 0.03 |
DISCUSSION
Concerning the effects of radiofrequency and microwave radiation on the ECG, most studies found that the rate of abnormal ECG among people exposed to EMF was higher than without exposure. The study of Bortkiewicz et al. [1996] indicated that exposure to EMF in an AM broadcast station increased the risk of electrographic disturbances by six times in comparison with the control group. In radioservice workers, this risk was twice as high as that in the control group. Mo et al. [2004] revealed that the rates of ECG changes among workers in a transformer substation were higher than those of the control group. Zhang [1995] analyzed the ECG changes between workers in a microwave station and its administrators, and found that ECG changes occurred in 26.1% of workers in the exposed group compared with 15.6% of those in the control group. In that study, the frequency of microwaves was 3.2–4.2 GHz, the power was 23–30 kW, and the people worked 6–8 h every workday; the level of microwaves in the station was also less than 10 µw/cm2. Cai et al. [2006] compared the rate of ECG changes of 330 soldiers in a radar station with that of 157 soldiers without exposure to radar; the rate of ECG changes in the exposed group was 48.8%, which was significantly higher than the 17.9% in the control group (P < 0.01).
The results of our study demonstrated that the rate of abnormal ECGs in people exposed to high intensity radiofrequency EMF was higher than those without exposure but the difference was not statistically significant. In this study, age, exposure duration, and drinking status were taken through logistic regression analysis at the same time. We observed that age was related to ECG changes but exposure duration and drinking status were not.
Sinus arrhythmia and sinus bradycardia have been the main abnormalities among all the ECG changes observed in most studies. In one study, the effects of microwave radiation from interphones on the ECGs of the occupational population were examined using 165 male security guards who were often exposed to microwave radiation from interphones and 80 male staff members in the same company who were not exposed [Lin et al., 2008]. The results showed that the prevalence of sinus arrhythmia, sinus bradycardia, and the total abnormal rates of ECGs in the exposed group were significantly higher than those in the control group. The study of Mo et al. [2004] revealed that the rate of arrhythmias, such as sinus bradycardia and sinus tachycardia, in the microwave-exposed group was 7.56%, which was more than that observed in the control group (2.25%). Also, in the same study, the rate of abnormal ECGs was correlated with exposure duration.
In the present study, we found that sinus arrhythmia and sinus bradycardia were the most commonly observed ECG abnormalities. The risk of sinus arrhythmia or bradycardia in people exposed to 27.2 MHz EMF for more than 2 years was higher than for those with no exposure. The ORs of sinus arrhythmia and bradycardia in the group exposed for ≥2 years were 3.12 and 5.71, respectively. However, when age was taken into account, the differences were not statistically significant (P < 0.05).
Previous studies have shown that exposure to radiofrequency electromagnetics was associated with blood pressure and HR changes. Braune et al. [1998] demonstrated that the blood pressure of ten 26–36-year-old volunteers increased 1.65–3.33 kPa after exposure to Global System for Mobile Communications (GSM) EMF. They considered that the increase in blood pressure to be associated with changes in sympathetic nerves affected by EMF. In the study of the effects of EMF emitted by GSM900-based mobile phones (MPs) on HR variability (HRV) [Yilmaz and Yildiz, 2010], 16 healthy young volunteers were exposed to EMFs emitted by GSM900-based MPs at two levels—from a very low EMF (MP at stand-by) to a higher EMF (MP at pre-ring, vibrating and ringing). The results showed that the degree of chaos in the HRV signals increased at higher EMF levels compared with low-level EMF. The study of Gadzicka and Bortkiewicz [1997] covered male workers of middle wave broadcast stations (71), radioservices (40), and radio line stations (42). The subjects were aged 21–60 years and the duration of their work with devices generating high frequency EMF ranged between 1 and 42 years. The first group of workers was exposed to EMF at a frequency of 1 MHz, the second at about 150 MHz, and the third group was not exposed (control group). The study revealed that the mean arterial blood pressure and day/night blood pressure variability indicators revealed no significant differences between the groups. However, the daily HR was significantly lower in the workers of middle wave broadcast stations in comparison with the control. The decreased value may suggest an occurrence of disorders in neurovegetative regulation. Johansson's study [2008] explored the effects of HRV during exposure to an EMF from a visual display terminal (VDT). Fifty healthy volunteers, seated in any ordinary working posture in front of a VDT, were examined. The HR was unchanged at the start of EMF exposure. The time-domain measurements indicated a significant decrease in HR and a significant increase in HRV. The aim of another study was to estimate the influence of an MP call on HRV in 32 young healthy people [Andrzejak et al., 2008]. Changes in HRV during the call with an MP could be affected by the EMF but the influence of speaking could not be excluded. A study by Borjanovic et al. [2005] found that of the ECG changes in humans exposed to 50 Hz magnetic fields, the duration of the QTc intervals decreased in the medium exposure group, in which the time-weighted average (TWA) exposure intensity of the magnetic field was equal to 1.18 uT, compared with that in the low exposure group (TWA = 0.067 uT).
As all the above studies demonstrated, radiofrequency radiation such as MPs, VDT, etc. might be related to the HR decrease. However, sufficient information on exposure levels and confounding factors was not available in most of the previously reported studies. In this study, although the exposure level was high and steady, any statistical difference in HR, duration of the QRS wave, or QTc intervals were not observed between the exposed groups and the control group. Some previous studies have concluded that radiofrequency EMF affected the pneumogastric nerve, causing changes in the ECG; however, after considering the confounding factors (age, smoking and drinking status, history of past illness, hereditary disease, etc.), occupational exposure to 27.2 MHz radiofrequency radiation was not the cause of the ECG changes.
