Abstract
The authors investigated the association between cellular phone use and epithelial parotid gland malignancy. The subjects were 136 cases who were treated for this condition at the authors’ hospital from January 1993 to March 2010, and 2051 controls who did not have salivary gland tumours and were admitted to the oral and maxillofacial surgery department during the same period. Logistic analysis was used to examine the relationship between cellular phone use and risk of epithelial parotid gland malignancy and mucoepidermoid carcinoma. Overall, the frequency of cellular phone use was not significantly associated with epithelial parotid gland malignancy. Female gender, advanced age, married status, high income, and smoking were associated with an elevated risk of epithelial parotid gland malignancy, especially mucoepidermoid carcinoma. Residence in a rural area was associated with reduced risk of epithelial parotid gland malignancy. The results suggest a possible dose–response relationship of cellular phone use with epithelial parotid gland malignancy. The authors suggest that the association of cellular phone use and epithelial parotid gland malignancy and mucoepidermoid carcinoma requires further investigation with large prospective studies.
Cellular phones are an important type of wireless communication that meet the needs of fast-paced modern societies, accommodate increasingly mobile populations, and are the most convenient and efficient communication tool for work and social life. Based on data from the International Telecommunication Union , there were 4.6 billion global subscriptions for cellular phones at the end of 2009, and more than 67% of people world-wide are cellular phone subscribers. Cellular phone subscriptions in China have increased dramatically since 2000. According to the Ministry of Industry and Information Technology of the People’s Republic of China , there were 747 million Chinese cellular phone subscribers by the end of 2009, corresponding to an average of 56.3 cellular phone subscribers per 100 inhabitants. China has the largest number of cellular phone subscribers world-wide.
As the prevalence of cellular phone use has increased , there have been concerns about the potential carcinogenic effects of exposure to the electromagnetic fields (EMFs) that are emitted by cellular phones. Numerous epidemiological studies have examined the association of cellular phone use and risk of cancer, and most of these have focused on intracranial tumours such as meningioma, glioma, acoustic neuroma, and pituitary tumour . The INTERPHONE project , a series of epidemiological studies supported by the European Union in which all 13 participating countries followed the same study design , examined the relationship of exposure to radiofrequency fields from cellular phones and tumour risk. In addition, there were four published studies that examined the association of cellular phone use and parotid gland tumours . Given the small number of cases studied, the use of different experimental designs, and the presence of bias, the association of cellular phone use and parotid gland tumours remains controversial.
In the present study, the authors used a retrospective case–control design to investigate the effect of cellular phone use on the risk of epithelial parotid gland malignancy, especially mucoepidermoid carcinoma of the parotid gland. This study provides an important complement to the previously published epidemiological studies of the health effects of cellular phones .
Materials and methods
In this hospital-based case–control study, all cases had histologically or cytologically confirmed epithelial parotid gland malignancies and underwent oral maxillofacial (OMF) surgery in the authors’ department from January 1993 to March 2010. All diagnoses were validated by a single surgeon. Histological typing of epithelial salivary gland tumours was based on the 1991 WHO guidelines .
Controls were individuals who did not have salivary gland tumours but who were treated during the same period as the eligible cases. This group included patients with impacted teeth, maxillofacial trauma, infections, temporomandibular joint disorders, maxillofacial nerve disorders, non-cancerous potentially oral and maxillofacial tumours (without salivary gland tumour involvement), salivary gland infections, congenital cleft lips and palates, or maxillofacial deformities. Patients with OMF malignancies and those with potentially cancerous tumours (including cysts and tumour-like lesions) were excluded from the control group. This included 11 cases with OMF malignancies, six cases with odontogenic keratocysts, two cases with pigmented nevi, one case with odontogenic myxoma, one case with giant cell tumour of the mandible, and one case with osteoclastoma. The authors considered the non-cancerous potentiality of OMF benign tumours in the pilot study for this research. As far as the authors know, there is no positive correlation between cellular phone use and the development of non-cancerous (benign) OMF tumours.
Data were obtained by case registration, and personal or telephone interviews and included patient identification, gender, date of birth, age at initial diagnosis, tumour location, pathological diagnosis, place of residence, marital status, educational background, monthly income, smoking status, and cellular phone exposure data.
Cellular phone use was considered a proxy for exposure dose to EMFs. All cellular phones were classified as first generation (1G), second generation (2G), or third generation (3G). IP phones, satellite phones, professional radio communication phones, car phones, and cordless phones were excluded. Exposure was characterized according to: frequency of use, with regular use defined as at least one call per week for 6 months or more before the time of diagnosis; end date of exposure, defined as the date of the initial histological diagnosis and the associated reference dates of relevant controls; exposure intensity, defined as the duration since the first use of a cellular phone to the time of diagnosis, calculated duration of cellular phone use, average daily use, average daily longest time of a single cal, average daily number of calls, number of calls since first use, and time of calls since first use; and preferred side for cellular phone use amongst regular users. The exposure intensity was based on median and quartiles of controls who were regular users and on previously published long-term exposure data: ≤median, >median to ≤third quartile, >third quartile to ≤10 years, >10 years. The others were based on median and quartiles of controls who were regular users: ≤median, >median to ≤third quartile, and >third quartile.
Statistical methods
The χ 2 test and χ 2 test for linear trend were used to analyse the effect of demographic characteristics and preferred ear for cellular phone use in the cases and controls using SPSS13.0 for Windows (SPSS Inc., Chicago, IL, USA). Unconditional logistic regression analysis, both univariate analysis ( Table 1 ) and multivariate analysis ( Tables 2 and 3 ), were used to calculate odds ratios (ORs) and 95% confidence intervals (95% CI). A p -value less than 0.05 was considered statistically significant.
| Epithelial parotid gland malignancies | Mucoepidermoid carcinoma | |||||||
|---|---|---|---|---|---|---|---|---|
| Case | Control | OR † | 95% CI † | Case | Control | OR † | 95% CI † | |
| n = 136 | n = 2051 | n = 64 | n = 2051 | |||||
| n (%) | n (%) | n (%) | n (%) | |||||
| Frequency of use * | ||||||||
| Never or rarely | 45(33.1) | 893(43.5) | 1 | 20(31.3) | 893(43.5) | 1 | ||
| Regular use | 91(66.9) | 1158(56.5) | 1.559 | 1.080–2.252 | 44(68.8) | 1158(56.5) | 1.697 | 0.993–2.889 |
| Duration since the first use of a cellular phone to the time of diagnosis (years) § | ||||||||
| Never or rarely | 45(33.1) | 893(43.5) | 1 | 20(31.3) | 893(43.5) | 1 | ||
| 0.5–6 * | 67(49.3) | 595(29.0) | 2.235 | 1.510–3.306 | 28(43.8) | 595(29.0) | 2.101 | 1.173–3.764 |
| 7–8 | 6(4.4) | 337(16.4) | 0.353 | 0.149–0.836 | 1(1.6) | 337(16.4) | 0.132 | 0.018–0.991 |
| 9–10 | 3(2.2) | 198(9.7) | 0.301 | 0.093–0.977 | 2(3.1) | 198(9.7) | 0.451 | 0.105–1.945 |
| >10 | 15(11.0) | 28(1.4) | 10.631 | 5.306–21.300 | 13(20.3) | 28(1.4) | 20.73 | 9.379–45.821 |
| Calculated duration of cellular phone use (years) § | ||||||||
| Never or rarely | 45(33.1) | 893(43.5) | 1 | 20(31.3) | 893(43.5) | 1 | ||
| 0.5–6 * | 67(49.3) | 595(29.0) | 2.235 | 1.510–3.306 | 28(43.8) | 595(29.0) | 2.101 | 1.173–3.764 |
| 7–8 | 7(5.1) | 337(16.4) | 0.412 | 0.184–0.923 | 2(3.1) | 337(16.4) | 0.265 | 0.062–1.140 |
| 9–10 | 2(1.5) | 198(9.7) | 0.2 | 0.048–0.833 | 1(1.6) | 198(9.7) | 0.226 | 0.030–1.690 |
| >10 | 15(11.0) | 28(1.4) | 10.631 | 5.306–21.300 | 13(20.3) | 28(1.4) | 20.73 | 9.379–45.821 |
| Average daily use (hours) ¶ | ||||||||
| Never or rarely | 45(33.1) | 893(43.5) | 1 | 20(31.3) | 893(43.5) | 1 | ||
| ≦0.5 | 53(39.0) | 627(30.6) | 1.677 | 1.113–2.528 | 29(45.3) | 627(30.6) | 2.065 | 1.158–3.684 |
| 0.5–2.5 | 30(22.1) | 521(25.4) | 1.143 | 0.711–1.836 | 8(12.5) | 521(25.4) | 0.686 | 0.300–1.568 |
| >2.5 | 8(5.9) | 10(0.5) | 15.876 | 5.978–42.162 | 7(10.9) | 10(0.5) | 31.255 | 10.799–90.456 |
| Average daily longest time of a single call (hours) ¶ | ||||||||
| Never or rarely | 45(33.1) | 893(43.5) | 1 | 20(31.3) | 893(43.5) | 1 | ||
| ≦0.5 | 66(48.5) | 627(30.6) | 2.089 | 1.411–3.093 | 31(48.4) | 627(30.6) | 2.208 | 1.247–3.909 |
| 0.5–2.5 | 25(18.4) | 531(25.9) | 0.936 | 0.567–1.544 | 13(20.3) | 531(25.9) | 1.095 | 0.540–2.220 |
| >2.5 | 0(0.0) | 1(0.0) | 0 | 0.000–0.059 | 1(0.0) | |||
| Average daily no. of calls ¶ | ||||||||
| Never or rarely | 45(33.1) | 893(43.5) | 1 | 20(31.3) | 893(43.5) | 1 | ||
| ≦8 | 77(56.6) | 724(35.3) | 2.111 | 1.442–3.088 | 37(57.8) | 724(35.3) | 2.282 | 1.313–3.966 |
| 40,431 | 11(8.1) | 283(13.8) | 0.774 | 0.394–1.511 | 4(6.3) | 283(13.8) | 0.631 | 0.214–1.862 |
| >10 | 3(2.2) | 151(7.4) | 0.394 | 0.121–1.285 | 3(4.7) | 151(7.4) | 0.887 | 0.260–3.022 |
| No. of calls since first use ¶ | ||||||||
| Never or rarely | 45(33.1) | 893(43.5) | 1 | 20(31.3) | 893(43.5) | 1 | ||
| ≦24,000 | 78(57.4) | 604(29.4) | 2.563 | 1.751–3.752 | 41(64.1) | 604(29.4) | 3.031 | 1.758–5.224 |
| 24,001–42,000 | 12(8.8) | 295(14.4) | 0.807 | 0.421–1.547 | 2(3.1) | 295(14.4) | 0.303 | 0.070–1.303 |
| >42,000 | 1(0.7) | 259(12.6) | 0.077 | 0.011–0.559 | 1(1.6) | 259(12.6) | 0.172 | 0.023–1.291 |
| Time of calls since first use (hours) ¶ | ||||||||
| Never or rarely | 45(33.1) | 893(43.5) | 1 | 20(31.3) | 893(43.5) | 1 | ||
| x ≦ 1350 | 57(41.9) | 595(29.0) | 1.901 | 1.269–2.848 | 30(46.9) | 595(29.0) | 2.251 | 1.267–4.002 |
| 1351–4320 | 26(19.1) | 275(13.4) | 1.876 | 1.136–3.098 | 9(14.1) | 275(13.4) | 1.461 | 0.658–3.246 |
| >4320 | 8(5.9) | 288(14.0) | 0.551 | 0.257–1.183 | 5(7.8) | 288(14.0) | 0.775 | 0.288–2.084 |
| Preferred side of calling | ||||||||
| Never or rarely | 45(33.1) | 893(43.5) | 1 | 20(31.3) | 893(43.5) | 1 | ||
| Ipsilateral | 26(19.1) | 326(15.9) | 1.583 | 0.961–2.607 | 19(29.7) | 326(15.9) | 2.602 | 1.371–4.938 |
| Contralateral | 20(14.7) | 308(15.0) | 1.289 | 0.749–2.217 | 15(23.4) | 308(15.0) | 2.175 | 1.100–4.300 |
| Bilateral | 45(33.1) | 524(25.5) | 1.704 | 1.112–2.612 | 10(15.6) | 524(25.5) | 0.852 | 0.396–1.834 |
# Univariate non-conditional logistic regression; unadjusted for covariates such as gender, age, resident area, marital status, education background, occupation, monthly income, smoking status. Because these covariates except occupation were included in the final statistical model. Occupation was not included in the final model most of them are blank in case group.
† OR, odds ratio; 95% CI, 95% confidence interval; SD, standard deviation.
* Regular use was defined as at least one call per week for 6 months or more before initial diagnosis time.
§ Duration classification were based on median and quartiles amongst controls who are regular users as well as long term exposure in the literature: data divided in to ≤median, >median to ≤third quartile, >third quartile to ≤10 years, >10 years.
¶ Classification based on median and quartiles amongst controls who are regular users: data divided in to ≤median, >median to ≤third quartile, >third quartile.
| B | S.E. | Wald | Sig. | OR † | 95% CI | ||
|---|---|---|---|---|---|---|---|
| Lower | Upper | ||||||
| Frequency of use * X1 | |||||||
| X1 = 0 (never or rarely) | 1.000 | ||||||
| X1 = 1 (regular use) | 0.133 | 0.235 | 0.319 | 0.572 | 1.142 | 0.720 | 1.811 |
| Duration since the first use of a cellular phone to the time of diagnosis (years) § X2 | 39.083 | 0.000 | |||||
| X2 = 0 (never or rarely) | 1.000 | ||||||
| X2 = 1 (0–6) | 0.525 | 0.245 | 4.602 | 0.032 | 1.691 | 1.046 | 2.731 |
| X2 = 2 (7–8) | 1.428 | 0.471 | 9.181 | 0.002 | 4.172 | 3.248 | 5.096 |
| X2 = 3 (9–10) | 1.679 | 0.648 | 6.717 | 0.010 | 5.359 | 4.090 | 6.629 |
| X2 = 4 (>10) | 1.419 | 0.435 | 10.666 | 0.001 | 4.133 | 3.282 | 4.985 |
| Calculated duration of cellular phone use (years) § X3 | 38.629 | 0.000 | |||||
| X3 = 0 (never or rarely) | 1.000 | ||||||
| X3 = 1 (0–6) | 0.524 | 0.245 | 4.580 | 0.032 | 1.689 | 1.045 | 2.728 |
| X3 = 2 (7–8) | 1.307 | 0.447 | 8.539 | 0.003 | 3.695 | 2.818 | 4.571 |
| X3 = 3 (9–10) | 2.041 | 0.765 | 7.121 | 0.008 | 7.699 | 6.200 | 9.199 |
| X3 = 4 (>10) | 1.419 | 0.434 | 10.675 | 0.001 | 4.135 | 1.765 | 9.688 |
| Average daily use (hours) ¶ X4 | 7.312 | 0.063 | |||||
| X4 = 0 (never or rarely) | 1.000 | ||||||
| X4 = 1 (<0.5 h) | 0.197 | 0.257 | 0.590 | 0.443 | 1.218 | 0.736 | 2.016 |
| X4 = 2 (0.5 h < x ≦ 2.5 h) | 0.072 | 0.291 | 0.062 | 0.804 | 1.075 | 0.505 | 1.645 |
| X4 = 3 (>2.5 h) | 1.794 | 0.717 | 6.254 | 0.012 | 6.012 | 1.474 | 24.524 |
| Average daily longest time of a single call (hours) ¶ X5 | 9.203 | 0.010 | |||||
| X5 = 0 (never or rarely) | 1.000 | ||||||
| X5 = 1 (<0.5 h) | 0.418 | 0.248 | 2.843 | 0.092 | 1.518 | 0.934 | 2.468 |
| X5 = 2 (0.5 h < x ≦ 2.5 h) | 0.423 | 0.314 | 1.814 | 0.178 | 1.527 | 0.911 | 2.144 |
| Average daily no. of calls ¶ X6 | 14.488 | 0.002 | |||||
| X6 = 0 (never or rarely) | 1.000 | ||||||
| X6 = 1 (0–8) | 0.445 | 0.244 | 3.332 | 0.068 | 1.560 | 0.968 | 2.516 |
| X6 = 2 (9–10) | 0.527 | 0.421 | 1.572 | 0.210 | 1.694 | 0.870 | 2.519 |
| X6 = 3 (>10) | 1.514 | 0.689 | 4.826 | 0.028 | 4.543 | 3.193 | 5.894 |
| No. of calls since first use ¶ X7 | 22.055 | 0.000 | |||||
| X7 = 0 (never or rarely) | 1.000 | ||||||
| X7 = 1 (≤24,000) | 0.577 | 0.238 | 5.879 | 0.015 | 1.780 | 1.117 | 2.838 |
| X7 = 2 (24,001–42,000) | 0.564 | 0.382 | 2.187 | 0.139 | 1.758 | 1.010 | 2.506 |
| X7 = 3 (>42,000) | 2.732 | 1.030 | 7.035 | 0.008 | 15.363 | 13.344 | 17.382 |
| Time of calls since first use (hours) ¶ X8 | 10.663 | 0.014 | |||||
| X8 = 0 (never or rarely) | 1.000 | ||||||
| X8 (≤1350) | 0.353 | 0.251 | 1.983 | 0.159 | 1.424 | 0.871 | 2.328 |
| X8 (1351–4320) | 0.217 | 0.316 | 0.471 | 0.492 | 1.242 | 0.669 | 2.308 |
| X8 (>4320) | 1.024 | 0.458 | 5.002 | 0.025 | 2.784 | 1.887 | 3.681 |
| Preferred side of calling X9 | 2.608 | 0.456 | |||||
| X9 = 0 (never or rarely) | 1.000 | ||||||
| X9 = 1 (ipsilateral) | 0.136 | 0.312 | 0.190 | 0.663 | 1.146 | 0.621 | 2.114 |
| X9 = 2 (contralateral) | 0.212 | 0.347 | 0.373 | 0.541 | 1.236 | 0.556 | 1.915 |
| X9 = 3 (bilateral) | 0.292 | 0.266 | 1.209 | 0.271 | 1.339 | 0.796 | 2.254 |
Stay updated, free dental videos. Join our Telegram channel
VIDEdental - Online dental courses
