ID Number		1669
Study Type		In Vivo
Model		Mobile phone use and fertility as measured by sperm count and motility.
Details		Human males (n = 361) from India, Ohio, and New Orleans were subjects in a study investigating correlations between mobile phone use and sperm count and motility (as a measure of fertility). The authors report men using a mobile phone more than 4 hrs per day had reduced sperm counts, reduced motility, and reduced normal forms. In a proposed study, 40,000 Indian participancts will be followed for mobile phone use and gynecological and reproductive endpoints, hormone changes and fertility. The program will also evaluate carcinogenic and genetic endpoints. A recent study collected semen samples (n = 32) and exposed half to 850 MHz GSM using a phone hooked to the network and held at 2.5 cm from the tube with an SAR reported of 1.45 W/kg. The authors report a significant increase in reactive oxygen species and a decrease in sperm motility, viability, and total antioxidant capacity. The authors also published a literature review to examine lifestyle factors that may affect testicular dysfunction and conclude that mobile phone use is one such factor. In a review paper, the authors discuss potential tissue targets and mechanisms that may be associated with testicular damage from RF exposure. AUTHORS' ABSTRACT: Agarwal and Durairajanayagam 2015 (IEEE #6193): The advent of mobile phones has revolutionized communication trends across the globe. As the popularity of mobile phone usage continues to escalate, there is now growing concern about the effects of radiofrequency electromagnetic waves (RFEMW) exposure on biological tissues, such as the brain and testes. Researchers have sought to link the much debated decline in human sperm quality in the last decade, with increased exposure to RFEMW, particularly through mobile phone usage. In a recent systematic review and metaanalysis on the effect of mobile phone RFEMW radiation on sperm quality, Adams et al.1 demonstrated an association between mobile phone exposure and reduced sperm motility and viability, with inconsistent effects on sperm concentration.1 Results from 10 pooled experimental (in vitro) and observational (in vivo) human studies (n = 1492) led these researchers to suggest that exposure to RFEMW radiation from carrying a mobile phone in the trouser pocket negatively impacts sperm quality.
Findings		Effects
Status		Completed With Publication
Principal Investigator		Cleveland Clinic, Cleveland OH, USA
Funding Agency		Nat'l Res Prog, India
Country		INDIA
References		Agarwal, A et al. Fertil Steril, (2009) 92:1318-1325 Agarwal, A et al. Biomed Pharmacother, (2008) 62:550-553 Agarwal, A et al. Fertil Steril., (2009) 92:1318-1325 Agarwal, A et al. Fertility and Sterility, (2008) 89:124-128 Desai, NR et al. Reprod Biol Endocrinol., (2009) 7:114-(9 pages) Agarwal, A et al. International braz j urol., (2011) 37:432-454 Agarwal, A et al. Asian Journal of Andrology., (2015) 17:433-434 Agarwal, A et al. World J Mens Health., (2014) 32:1-17 Deepinder, F et al. Reproductive biomedicine online., (2007) 15:266-270 McGill, J et al. In: Male Infertility: A Complete Guide to Lifestyle and Environmental Factors (du Plessis, S.S., Agarwal, A., Sabanegh Edmund, S.J., Eds.); Springer: New York, NY, USA., (2014) :161-177 Desai, N et al. Fertil Steril., (2009) 92:1626-1631 Durairajanayagam, D et al. Reprod Biomed Online., (2015) 30:14-27
Comments		1.45 W/kg is from Sony-Ericsson website for human head exposure. Statistics: 1) The appropriate statistical test was not performed. A parametric test (e.g., t-test) should have been used to make sure the exposed and control values were compared against normal physiologic values  such a parametric test would have given weight to outliers that would have decreased the strength of the correlation. The statistical test used (Wilcoxon rank sum test) is for non-parametric analysis  or for data that does not need to conform to an expected distribution. Because of that, the magnitude of the p values do not need to correspond to the magnitude of the SD bars. Since the Wilcoxon rank sum test cannot be backwards computed from just the reported mean and s.d.s as with a t-test, it is impossible to check whether the actual math is correct without the raw data. In addition, since the data are paired (a semen sample is divided into two parts, one exposed and the other control) the appropriate non-parametric test would have been the Wilcoxon signed rank test that takes into account this control-exposed pairing  that was not done. Usually, analyzing paired data using a paired test increases the power, i.e. produces a more significant p-value. 2) Table 1 is not internally consistent. One should be able to compute the overall mean from the two subgroup means and the sample sizes - Log ROS+0.001 is correct: the overall mean -1.72 = (23* -1.85 + 9* -1.37)/32 = -1.715. However, Viability and Motility are not correct - viability appears to have been calculated with 7 donors, not 9 and motility (reported as 7 donors) appears to have been calculated with 9. 3) More importantly, the text makes no mention and does not account for missing observations (observed as different sample sizes in the Tables). 4) There is no mention of blinding.