Gender ratio of offspring and exposure to shortwave radiation among female physiotherapists.

Gender ratio of offspring and exposure to shortwave radiation among female physiotherapists. Scand J Work Environ Health 1994;20:345-8. OBJECTIVES - The goal of this study was to investigate whether the deficit of male births found among the offspring of Danish physiotherapists exposed to shortwave radiation during the first month of their pregnancy could be confirmed among the offspring of physiotherapists from Switzerland. METHODS - A self-administrated questionnaire was mailed (two mailings) to all of the 2846 female members of the Swiss Federation of Physiotherapists. It included questions on the gender and birth weight of all children of the physiotherapists, as well as on the use of shortwave or microwave equip ment during the first month of each pregnancy. The response rate was 79.5%, and the analysis was based on 1781 pregnancies. RESULTS - The gender ratio (the number of males per number of females x 100) was 107 with a 95% confidence interval (95% CI) of 89-127 for the 508 pregnanc ies exposed to shortwave radia tion and 101 (95% CI 90-113) for the 1273 unexposed pregnancies. There was no trend in the gen der ratio with increasing inten sity or duration of exposure. The prevalence of low birthweight (::=; 2500 g) was not related to exposure to shortwave radiation for either the boys or the girls. CONCLUSIONS - No atypical gender ratio was found for the children of female physiotherapists from Switzerland who had been exposed to shortwave radiation at the beginning of pregnancy. The find ings of the Danish study could not be confirmed.

In a case-referent study amon g the female member s of the Union of Danish Physiotherapists, Larsen et al (1, 2) investigated the reproductive hazard s of exposure to high-frequency electromagnetic radiation from shortwave diathermy. No significant association was found between various indices of exposure and congenital malformations, spontaneous abortion or stillbirth and death within one year. However, an association was found between exposure during the first month of pregnancy and a deficit of male births. The gender ratio (number of males per number of females x 100) of offspring was lSI for the physiotherapi sts with no exposure, 48 for those with low expo sure , and 31 for those with high exposure. The I Med ical Branch of Labour Inspect ion, Geneva, Switzerland . 2 Clin ic of Sterilit y and Gynecologic Endocrinology, Universit y Hospital, Geneva, Switzerl and . J School of Physiotherapy, Gen eva, Switzerl and . 4 Physiotherapist, Mezieres, Switzerl and . 5 MRC Epid emi olog y Unit , Penarth , South Wales, Great Britain . 6 Eu ropean Organization for Nuc lea r Research CERN, Gene va, Switzerl and.
trend was significant at the 0.1 % level. In addition there was an association, although not significant, between exposur e and low birth weig ht among the boys, while there was no such association among the girls. To explain the deficit of male births among the offspring of exposed physiotherapists, it was suggested that this type of electromagnetic radiat ion could induce very early abortion at a preclinical stage. This occurrence would mainly involve male embryos, which are considered to be more fragile than female ones.
The aim of the present study was to investi gate whether the findings by Larsen et al could be confirmed among physiotherapists from Switzerland.

Subjects and methods
In Switzerland, about 95% of self-employed physiotherapists and about 60% of salaried physiotherapists are members of the Swiss Federation of Physiotherapists. A questionnaire was sent to all of the 2846 female members of that federation in November 1992 and a seco nd time in the spring of 1993 to those who did not answer the first request. The physiotherapist was questioned on the gender and the birthweight of all her children, whether born alive or dead. Abortions were excluded because the gen-der of the fetus was unknown. The physiotherapist was also asked to report for each pregnancy whether she had been exposed to shortwaves during the first month and whether the equipment had been used by her or any colleague. Further questions asked for estimates of the duration of exposure per week at distances from the equipment of I m or less and of more than I m. The different types of electrodes, as well as the relative percentages of them was also requested. The limit of 1 m was chosen because the electric field strength was shown to fall to 10% at 1.0 m and to 2% at 2.0 m (3). An additional question concerned exposure to microwave radiation.
Measurements in Denmark (3,4) and Geneva (table I) indicated that exposure to electromagnetic radiation is very low for the circuplode electrode. This is also the case for the other types of electrode at a distance of I m or more. The results were thus analyzed separately for the relatively highly exposed subgroup (ie, those exposed to monode, diplode or plate electrodes at a distance of 1 m or less).
For more than one century, the gender ratio of live births in Switzerland has been remarkably stable at around 105. We have assumed that this figure is the "true" gender ratio of the offspring of unexposed physiotherapists. The 95% confidence interval (95% CI) for the gender ratio among various subgroups of physiotherapists has been calculated with the use of the exact binomial confidence limits from Geigy Scientific Tables (5).
From our knowledge, the type of equipment used for shortwave diathermy in Switzerland is similar to that used in Denmark. Larsen & Skotte extensively studied the exposure of Danish physiotherapists giving diathermy (3,4). By a limited series of measurements in Geneva, we were able to verify that the exposure of physiotherapists from Switzerland was likely to be of the same order of magnitude as that of their colleagues in Denmark.
Measurements were carried out in the vicinity of one of the diathermy installations of the School of Physiotherapy, Geneva, being of the type Curaplus 419 operating at 600 W peak power (260 Weff) and a frequency of 27.12 MHz, to determine the potential radiofrequency exposure of therapists during treatment. These measurements were carried out with a Narda model 8608 broadband isotropic radiation monitoring system using a model 8662 B probe which measures the E field.
The measurements were made under normal operating conditions with the electrodes placed on the legs of a patient. Exposure was determined at distances of 30 and 100 em from the electrode next to the patient. The results are given in table 1. These values should be considered as upper limits of exposure since measurements of the H field showed considerably lower values at the same distances. It can be seen that exposure is a strong function of the type of electrode used, particularly at the shorter distance.
Moreover, it can be postulated that, due to the introduction of the circuplode electrode in the early 1970s, exposure of physiotherapists has been lower during the last two decades than beforehand.

Results
As shown in table 2, the gender ratio of offspring was 107 (95% CI 89-127) for physiotherapists exposed to shortwaves at the beginning of pregnancy, whereas it was 101 (95% CI 90-113) for unexposed physiotherapists. The gender ratio was neither related to the type of electrode (on which the intensity of the electromagnetic field mainly depends) (table 3) nor to the duration of exposure (table 4). The prevalence of low birthweight (s2500 g) was not related to exposure among either the boys or the girls (table 5). Among all of the subgroups exposed to shortwaves, and among the physiotherapists exposed to microwaves (table 6), the 95% confidence interval of the gender ratio included the expected population value of 105. There was no significant difference between the exposed and unexposed physiotherapists (chi-square test) in the gender ratio of offspring (tables 2 and 6) or in the prevalence of low birthweights of offspring , Expose d at ,; 1 m from the mono de, diplo de, or plate electr odes. Table 5. Prevalence of low birthweight of ", 2500 g of t he offspring" of phys iotherapists by exposure to shortw aves.  a Subg roup expose d at '" 1 m fro m the monode, dlpl od e, or plate elect rodes. , Too few data to ju stify interval est imat ion. c Short wave equipment used by the subject only. d Short wave equ ipm ent used by co ll eague(s) only.

Discussion
In the present study, there was no suggestion of a defi cit of male births or of an inc reased prevalence of low birthweight of boys among the offspring of phys iotherapists exposed to shortwave radiati on in Switzerland at the beginning of pregnancy. Th is result is contrary to the find ing of Lar sen et al in Denmark . Among Danish physiotherapists the findin g of a signific ant trend in the gender distribution of the offspr ing with increasing exposure to electromagnetic radi ation was strongly influenced by the ext raordin ary gender ratio of 151 in the unexposed group (time-weighted exposure index = 0). The 95% confidence inter val for that ratio is 112 to 206, whereas the gender ratio for all births in Denm ark is 105.8 (l ). Nevertheless, Larsen et al also found a large deficit of male births in their ex pose d group (timeweighted exposure index = I and 2), for whom the gender ratio was only 42. The current study was large enou gh to have a power in excess of 99% to detect, as significant at the I% level, a decre ase in the gender ratio to Larsen ' s value of 42 . However we found no suggestion of any deficit in male birth s. Animal exp eriments with pre gnant rat s or mice exposed to shortwave or micro wave radiation have show n increases in malformati ons and in embr yo lethality, as well as reduction in fertility (I). It remains uncl ear, however, whether these adverse outc omes were only due to the well-known harmful effects of hyperthermia or whether they could be attr ibutable to a possible nonth ermal effect of electromagnetic radiation (1). The gender ratio of the offspring was not reported in these studies.
Tw o pre vious studies (6, 7) have addressed the possible reproductive haz ard of various types of equipment used by female physiotherapists, but none has included an analysis of the gender ratio. In the first study, the frequency of use of shortwave equipment was significantly asso ciated with congenital malform ations in Finland, but onl y in the lower exposure gro up (6). In the seco nd study, there was an association of borderline sig nificance betw een frequent exposure to shortwaves and dead or malformed infants in Swed en , but there was little evidence that this associ ation was causal (7 ). In neither of the studies, however, was the type of electrode taken into accoun t to classify the expos ure of physiotherapists, even though th is is an ext remely important variable.