Case-Control Study on Radiology Work, Medical X-ray Investigations, and Use of Cellular Telephones as Risk Factors for Brain Tumors

, Department of Oncology, Orebro Medical Center, SE-701 85 Orebro, Sweden email: , Department of Oncology, Orebro Medical Center, SE-701 85 Orebro, Sweden , Department of Neurology, Orebro Medical Center, SE-701 85 Orebro, Sweden , Department of Oncology, Karolinska Institute and Stockholms Sjukhem, Mariebergsgatan 22, SE-112 35 Stockholm, Sweden

In This Article


The purpose of this investigation was not disclosed to the study subjects. A questionnaire, which assessed different occupational and leisure time exposures, was used to minimize recall bias. All telephone interviews and coding of the data were made blinded to case or control status to reduce observational bias. Only living patients who were judged to be able to respond to the questionnaire were included to obtain as high-quality data as possible. During the study period, use of mobile telephones was not often discussed in the media as a risk factor for brain tumors.

Thirty-seven patients were excluded because their physicians deemed them ineligible. This might have introduced observational bias in the study if a risk factor is related to the prognosis of the disease for those too ill to participate. However, there are no data in the literature indicating that this is the situation.

Information on radiology work was obtained by the lifetime occupational history. The physicians were interviewed about type of work. Two of 3 cases had worked for only a short period at a radiology department. However, both reported work with fluoroscopy, and the dosimeter of the female case patient showed "always maximal exposure." The third case had worked as an anesthesiologist at radiology departments for about 30% of his working time. Thereby, he had participated during angiographies and heart catheterizations of patients. For other occupational categories at radiology departments, no increased risk was found.

In addition, radiotherapy of the head and neck region was associated with an increased risk for brain tumors, which is in accordance with previous studies.[8,9,10] Meningioma has been reported to be the most common tumor associated with radiation,[1,2] and 3 of the 4 cases that had been treated with radiotherapy had meningioma in our study.

Medical x-ray investigation of the head and neck region increased the risk, which is in accordance with other results, although the association is somewhat more controversial than for high-dose radiation.[1] Of the patients who reported medical diagnostic x-ray investigations of the head and neck, 36% had meningioma compared with 23% of all cases in this study. The questionnaire data did not allow us to study any dose-response effect, and the reported investigations were not qualified by use of information in patient records. We tried to get information on dental x-ray examinations, but almost all subjects were exposed, and the number of x-ray examinations was difficult to assess, thus making statistical analysis not meaningful.

In a register study linking census data on occupation with the Swedish Cancer Registry, we did not find an increased risk for brain tumors among physicians.[11] However, we had no data on different specialists, so radiology work could not be studied separately. Ionizing radiation is an established risk factor for brain tumors, as reviewed by Finkelstein[3] and others.[1] Clearly, the report by Finkelstein,[3] other studies,[1] and our results indicate that radiology work, especially fluoroscopy, may be a health hazard and increase the risk for brain tumors. The brain is a part of the body that usually is not shielded during fluoroscopy. Current findings indicate that radiation protection of the head is warranted as precautionary avoidance of exposure but also that further studies to confirm the findings are necessary. Furthermore, studies of dose-response effects should be performed in larger investigations. Also, theoretical calculations of the risk based on dosimeter data on exposure to the brain during fluoroscopy are warranted.

Exposure to extremely low-frequency electromagnetic fields has been suggested to increase the risk for brain tumors. However, in our overview of studies on that topic, we concluded that no consistent association could be found.[12] In the present investigation, no association was found with occupations with potential exposure to electromagnetic fields, such as electrician, electronics work, lineman, or telecommunications work. Nor did use of a video display unit increase the risk.

Some studies have suggested an association with exposure to certain pesticides.[1,13] This was not found in this study. Regarding aspartame, it is difficult to assess total exposure, since it occurs in different types of food, such as beverages, ice cream, cakes, and sweets. However, the highest per capita exposure is from low-calorie drinks, with an estimate of 45% in a Norwegian study.[5] Thus, we assessed only intake of such beverages. For malignant brain tumors, an increased risk was found in the highest-exposure group. This was based on low numbers and must be interpreted with caution. Also, the cutoff dose for dividing the number of exposed controls into 2 groups with equal number was low (6864 centiliters) indicating possible underreporting of intake of low-calorie drinks. However, the mean age of cases and controls was 50 years, and consumption of low-calorie drinks is clearly a more common habit in young subjects. No increased risk was found in a US study on childhood brain tumor and aspartame consumption.[14]

During a mobile phone call, depending on the antenna, highest exposure occurs in the temporal, occipital, and temporoparietal lobes on the same side of the head used for the call. There is a rapid decline of the microwave dose in the brain, and the other side of the brain is only exposed to a low degree. An increased risk for brain tumors in the anatomic areas with highest exposure to microwaves from a cellular telephone has previously been reported from this investigation.[6] In that report, we displayed the results for each hemisphere separately. We have now combined the areas with highest exposure (temporal, occipital, and temporoparietal lobes), the remaining lobes with low exposure (frontal, parietal frontoparietal, and parieto-occipital), and also each hemisphere irrespective of left or right side of the brain. The OR was calculated for ipsilateral, contralateral, or both ipsilateral and contralateral (both sides reported by some subjects) exposure to microwaves from a mobile telephone. An increased risk was only found for ipsilateral exposure in the anatomic area with the highest microwave dose. In a multivariate analysis including other exposures with significantly increased risk, this result was further strengthened.

Obviously, this result was based on low number of exposed subjects and must be interpreted with caution. However, the finding might be of biological relevance. Since patients do not usually have exact information of the anatomic area (lobe) of the tumor, recall bias is less likely to explain the results. All but 1 of these 13 patients had used the NMT system (analogue), and it should be noted that analogue telephones have at least a 3 times higher output power than digital telephones. In the 1980s, only the analogue system was used, and the digital system was introduced in the Swedish market in early 1990s; thus, tumor induction period might also be relevant. Because of the low numbers, it was not meaningful to calculate ORs according to tumor induction time or cumulative exposure in hours for high-exposure area of the brain. Other parts of the brain were also included in multivariate analysis, but the results were similar to the univariate analysis.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: