TMobile Study

Mobile Telecommunications and Health

Review of the current scientific research in view of precautionary health protection

April 2000

ECOLOG-Institut

TranslatedbyAndreaKlein

Mobile Telecommunications and Health

Review of the Current Scientific Research in view of Precautionary Health Protection

Commissionedby

TMobilDeTeMobilDeutscheTelekomMobilNetGmbH

Authors

DrKerstinHenniesDrH.PeterNeitzkeDrHartmutVoigt

Withthesupportof

DrGisaKahleAnders

ECOLOGInstitutfrsozialkologischeForschungundBildunggGmbHNieschlagstrasse2630449HannoverTel.05119245646Fax05119245648Emailmailbox@ecologinstitut.de

Hannover,April2000

iii

Contents

1 Introduction 1

1.1 NewTechnologiesandPrecautionaryHealthProtection 1 1.2 TermsofReferenceandStructureoftheReview 3

2 CollatingandInterpretingtheScientificData(Methodology) 5

2.1 CriteriafortheSelectionofPapers 5 2.2 AssessmentCriteria 5

3 PrimaryReciprocalEffectsbetweenHighFrequencyElectromagneticFieldsandBiologicalSystems(BiophysicalandBiochemicalProcesses) 8

3.1 ThermalEffects 8 3.1.1 EffectsofHomogenousWarming 8 3.1.2 MicrothermalEffects 8

3.2 DirectFieldEffects 9 3.2.1 EffectsfromtheElectricalComponentoftheElectromagneticField 9 3.2.2 EffectsfromtheMagneticComponentoftheElectromagneticField 10

3.3 QuantumEffects 10 3.4 OtherEffects 10 3.5 ParticularPropertiesofPulsedElectromagneticFields 11

4 BiologicalPrimaryEffectsofHighFrequencyElectromagneticFieldsEffectsonCellularLevel 12

4.1 GeneToxicity 12 4.2 CellularProcesses 13

4.2.1 GeneTranscriptionandGeneTranslation 13 4.2.2 MembraneFunction 14 4.2.3 SignalTransduction 14 4.2.4 CellCycle 16

4.3 CellTransformationandCellProliferation 16 4.3.1 CellTransformation 17 4.3.2 CellCommunication 17 4.3.3 CellProliferation 17

5 PathoPhysiologicalEffects 19

5.1 ImmuneSystem 19 5.2 CentralNervousSystem 19

5.2.1 BloodBrainBarrier 19 5.2.2 Neurotransmitters 20 5.2.3 Electroencephalogram(EEG) 20 5.2.4 CognitiveFunctions 22

5.3 HormoneSystems 23 5.3.1 StressHormones 23 5.3.2 Melatonin 24

iv

6 PathologicalEffects 26

6.1 ResultsofExperimentalStudies 26 6.1.1 Cancer 26 6.1.2 InfertilityandTeratogenicEffects 27

6.2 ResultsofEpidemiologicalStudies 28

7 Health Risks toHumans Resulting from Exposure to the ElectromagneticFieldsofMobileTelecommunications 33

8 Recommendations 37

8.1 PrecautionaryHealthProtectioninRelationtoExposurestoElectromagneticFieldsofMobileTelecommunications 37

8.2 ScientificStudiesRegardingtheHealthRiskofMobileTelecommunications 38

Literature 40

AppendixA 56

Studiesoftheeffectsofhighfrequencyelectromagneticfieldsonthecellularlevel 56 TableA.1 GenotoxicEffectsofHighFrequencyElectromagneticFields

TableA.2 EffectsofHighFrequencyElectromagneticFieldsonCellularProcesses

TableA.3 EffectsofHighFrequencyElectromagneticFieldsonCellTransformationandCellProliferation

AppendixB 66

Studiesoftheeffectsofhighfrequencyelectromagneticfieldsonthecentralnervoussystem(BloodBrainBarrier) 66 TableB.1 EffectsofHighFrequencyElectromagneticFieldsontheBloodBrain

Barrier

AppendixC 68

StudiesoftheCarcenogenicEffectsofHighFrequencyElectromagneticFieldsinAnimalExperiments 68 TableC.1 AnimalExperimentsRegardingtheCarcinogenicEffectsofHigh

FrequencyElectromagneticFields

AppendixD 71

EpidemiologicalStudiesofthehealthRisksofHFEMFs 71 TableD.1 Overviewoftheresultsofepidemiologicalstudiesregarding

exposuresinthehighfrequencyspectrumandhealthrisks

AppendixE(onlyavailableinGerman) 82

Extractsofourdatabase(EMFbase)

Important research papers relevant to the assessment of health risks resulting fromexposuretotheelectromagneticfieldsofmobiletelecommunicationsundertheaspectofprecautionaryhealthprotection

Mobile Telecommunications and Health 1

1 Introduction

1.1 New Technologies and Precautionary Health Protection

Notechnologycoveringvirtuallyentirecountrieswithitsemissionshaseverbeenrolledoutasquicklyasmobiletelecommunications.Atthesametime,thereareonlyfewdirectstudiesofthepotentialhealthrisksoftypicalmobiletelecommunicationsfrequenciesandmodulationsfortheexposedpopulation.Also,manyoftheexistingstudiesworkedwithhigh intensities,whichwill only be found in rare cases in the real environment.Highintensities of high frequency electromagnetic fields can heat the absorbing tissue andtrigger stress reactions in the body and thuswith rising temperatures lead to thermaldamage.Effects fromhigh intensityhigh frequencyEMFs,alsoknowas thermaleffects,on the centralnervous system, the immune system, the cardiovascular systemand thereproductivesystemincludingteratogeniceffects,havebeenprovenformammalswithamultitudeofexperiments.

TheresultsofstudiesofthethermaleffectsofhighfrequencyEMFsformthebasisoftherecommendationsoftheInternationalCommissiononNonIonizingRadiationProtection(ICNIRP),which, inthepast,werethebasisfortheguidelinessetbythegovernment inGermanyandmanyothercountries.ThebaseguidelinewasanupperlimitontheSpecificAbsorptionRate (SAR), i.e. theamountof energy absorbedby thebody from the fieldwithinagivenunitoftime.

AccordingtoICNIRP,thermaldamagewillnotoccuratSARvaluesofunder4W/kgandexposure levels of 0.4W/kg for professional exposures and 0.08W/kg for the generalpopulationareconsideredsafe.

Paralleltotheexperimentsexaminingthermaleffects,therehavebeenagrowingnumberofstudiesexamining theeffectson thebodyofHFEMFsatsubthermal intensities.Wenowhaveaplethoraofexperimentalstudiesexaminingavarietyofeffectsonalllevelsoftheorganism,rangingfromeffectsonsinglecellstoeffectswhichmanifestthemselvesasreactionsof theentirebody. Inaddition to theexperimental studies, therehavebeenanumberofepidemiological studies inorder toestablish thepossible causal correlationsbetween higher exposures toHF EMFs, for example as found near base stations, andhealthdamageamongstthepopulationgroupswithhigherexposures.

Themobiletelecommunicationssituationreflects,onceagain,thedilemmaalreadyknownfromchemicaltoxicology:Thestudyofpotentialhealtheffectscannotgenerallycompetewith thespeedof technicaldevelopmentand therolloutof theproduct.Theextremelyfastrolloutof themobile telecommunications technologyand theaccompanyingpublicfearofthepotentialdangerofthistechnologyhavestimulatedresearchinsofarthatnowwehavemore studiesexamining theeffectsof frequenciesandmodulationsasused inmobile telecommunicationsonbiological systems.There are also agrowingnumberofexperiments using lower intensities, reflecting the real conditions of exposure in thevicinity of base stations and equipment, so that effects found in the studies can beextrapolated into real life conditions. The number of studies which examine the


physiologicaleffectsofrealmobileexposuresisstillverylow,comparedtothedegreeofpenetrationachievedbythetechnologyandthenumberof(potentially)exposedpersons.TheWHO amongst others, have only recently begun todevelop targeted strategies toexaminethepotentialhealthriskfrommobiletelecommunicationsandresultscanearliestbeexpectedwithinseveralyears.

In the meantime, it is only possible to assess the potential dangers of mobiletelecommunicationsusingtheresultsgeneratedbyuncoordinatedresearch,whichisstillmainly orientated towards topics and criteria of relevant to science only, rather thanaddressingtherequirementsofsocietyasawhole.

Facedwithastateof incompletescientificresearch it isnecessary tochosebetween twofundamentallydifferentassessmenttheorieswhenplanningtoassessthepotentialhealthrisksofnewtechnologies:

Thefirst theory isbasedon the (withoutdoubtcorrect)scholarlyunderstanding that ispracticallyimpossibletoprovethenonharmfulnesstohumanhealthortheenvironmentofatechnology,amaterialoraproduct.Thisunderstandingisinterpretedinsuchawaythatapresuppositionof notguilty isadoptedandany riskshave tobeunequivocallyproven.

Unequivocal proof in this context means the consistent evidence for a biologicalphysiological or an ecological chain of effects, from the biophysical or biochemicalprimary effect through to the physiological effects and the resulting illness or, ifapplicable,theecologicaldamage.

Thistheory,whichisfirmlybasedinscientificconservatism,hastheadvantagethatitwillstand up in court and will not hinder the introduction of new technologies. It ismethodologicallysimple,since it issufficienttoexaminestudieswhicharepresentedasproofwithregardstotheirmethodologicalcorrectnessandtheirvalidityandthentoputall thesereviewedpiecesofevidence together likea jigsaw toproduceawholepicture.The complete whole picture finally constitutes the scientific proof required by thelegislatorsandcourts.

The disadvantage of this theory is obviously the length of time necessary to obtainenough knowledge for a completed chain of proof, duringwhichmany factswill becreated, which may later prove irreversible or only reversible with very high costsattached,suchasinvestmentsandirreversibledamagetohealthandtheenvironment.

Thesecondtheorysolvesthedilemmaofthetimedelay.Itisbasedontheassessmentofthe potential risks of a technology on the basis of existing knowledge. If there aresufficientindicationsthattheremaybedamagingeffects,theprecautionaryprinciplefortheprotectionofhealthandtheenvironmentwillapplyandavoidableexposureswillbeavoideduntilsuchtimewhenthereisenoughknowledgeforawiderintroductionofthetechnology inquestion.Thistheorydraws its justificationnot leastfromtheexperienceswith the introduction of technologies and products (such as asbestos, DDT, CFCs,formaldehyde,woodpreservatives,massXrayscreeningsetc.),whichwerewidelyused,evenmany years after the first clear indications of health and ecological damage hadappeared.When finally sufficient scientificproof for thehealth and ecologicaldamage


was provided, it tookmanymore years until the further usewas finally reduced andbannedthroughthecourtsandinternationalnegotiations.

The advantage of the precautionary principle is of course primarily medical andecological, since exposures are initially limited to a level recognised as safe under theprecautionaryprinciple.But it can also offer economical advantages, because firstly, itmay prevent potentially highly risky investments, but also secondly, because thecommitmenttoandobservanceoftheprecautionaryprinciplewillcreatetrustwithinthegeneralpopulationandthusincreaseacceptancefortheplacingofemittingequipment.

Ontheotherhand,itwillbetheindustryastheownerofemittingequipmentwhohastobear thedisadvantageof thisprinciple,when itbecomesclear that,forprecautionaryreasons,aneconomicallyandtechnicallyperfectlysuitedsitecantbeapproved,ormaybeevenanentiretechnologyhastobeabandoned.

Furthermore, themethodologicaldifficultiesof this theorymustnotbeunderestimated,since it isnotenough toprove thereliabilityofsinglescientificstudies,which is justasessentialunderthispremiseasunderthefirsttheory.Theultimategoalhowever istoremainwiththejigsawanalogytoputtheexistingjigsawpiecestogetherandrecogniseearlyonwhichpicturesmightappearoncetheworkiscompleted.

1.2 Terms of Reference and Structure of the Review

Theaimofthisstudywastheassessmentofthepotentialrisksofelectromagneticfieldsasthey are used for mobile telecommunications with respect to precautionary healthprotection.Tothisaim,thescientificliteraturewasreviewedwithregardstostudyresultswhichmight be of importance to the assessment ofpotential health risks frommobiletelecommunications.

To create a base for later scientific discussion, a list of studieswhich are particularlyimportant inthisrespectshouldbecreated.Onthebasisofthesepapers,thehealthriskfrom exposure to electromagnetic fields from mobile telecommunications should beassessed.Finally,recommendationsforfuturescientificstudiesshouldbeformulated.

The methodological aspects of this examination are presented in Chapter 2. This isfollowedbyareviewofthecurrentscientificknowledgeoftheeffectsofhighfrequencyelectromagneticfields.Thisreviewisstructuredaccordingtothedifferentlevelsofeffects:

biophysicalandbiochemicalprimaryeffectsofHFfieldsonorganicmatterasawholeoratthelevelofcellsandmembranes(Chapter3)

primary biological effects on the cellular level, i.e. on the genetic substance and onintracellularprocessesaswellascelltransformationandcellproliferation(Chapter4)

pathophysiological effects, i.e. physiological effects with possible but not certainnegativehealthimplications(Chapter5)

pathological effects, which means manifested illness and other effects such as thedamage of cognitive functions, which have been found in epidemiological orexperimentalstudies(Chapter6).


The conclusions of all findings are drawn in Chapter 7. In Chapter 8, we makerecommendations forprecautionaryhealthprotectionwith regards to exposures to theelectromagnetic fields of mobile telecommunications and for focal points for furtherresearch.


2 Collating and Interpreting the Scientific Data (Methodology)

2.1 Criteria for the Selection of Papers

Inordertoincludeamaximumofrelevantliterature,weanalysedtheliteraturewehavecataloguedinourowndatabase,EMFbase,aswellasexploringthethreefollowingpaths:

researchinotherrelevantscientificdatabases

complete sifting of at least the last two full years issues of all relevant scientificjournals available in the Central Library of Medicine in Cologne, the TechnicalInformationLibraryinHanover,andtheLibraryoftheMedicalUniversityofHanover

evaluationofallexistingmonographs, reviewsand conference reports related to thesubjectmatter

ThebasicliteratureresearchwasfinishedinFebruary2000.

Literature databases are a convenient research tool, but their value in assessing thecurrent scientificknowledge in a subjectmatter is limitedby thenumberof registeredpublications, inconsistent use of keywords, the changing understanding of certainprocedures,effectsetc.andlastbutnotleast,duetolongtimedelaysbetweenthetimeofpublication and availability in thedatabase. Furthermore,databasesusually only keepabstracts of papers, and those differ often from the full text with regards to thepresentationandinterpretationoftheresults.Ourresearchforthisreviewconfirmedthisobservation, reflecting the results of a study of Pitkin et al. (1999) according towhichalmost 40% of all papers published in the six largest medical journals containedinaccuracies and mistakes in the abstracts. To be at the cutting edge of scientificknowledge,itisnecessarytoresearchcurrentscientificjournalsandfindolderpapersviamonographs and reviews.Reviews are onlyuseful to gain an overview over a subjectmatter and as a source for literature leads. It is inappropriate to use assessments orinterpretations of a review study since some authors of reviewswill have based theirconclusiononabstractsratherthanthefulltextsofthepaperstheydiscuss.

2.2 Assessment Criteria

One subgoal of the present paper was to identify those scientific papers which areparticularly interesting for the assessment of potential health risks caused by theelectromagnetic emissions of mobile telecommunications. (Extracts from our databaseEMFbasewithasummaryoftheresultsofthesepaperscanbefoundinAnnexE.Inthesourcereferences,thesepaperscarryanasterisk*).Onlypeerreviewedpaperspublishedinscientificjournalswereincludedinourreview.WealsoaccordedweighttotheImpactFactor,whichiscalculatedbytheInstituteforScientificInformationinPhiladelphia.Thisfactor isa roughmeasure for theamountof importanceand reputationattributed toascientificjournalinitssubjectmatter.


Thepapers able topass this first filterwere subsequently interpreted according to thefollowingcriteria:

carrierfrequencyorfrequencyrange

mannerofmodulation

modulationfrequencyorfrequencyrange

powerfluxdensity

specificAbsorptionRate

electricfiledstrength

durationofexposure

otherparametersofexposure(suchasotherfields[incl.ELF],ambientandifapplicablebodytemperature,particularformsofmodulation)

sourceofexposureorenvironmentof theexposure (suchasantennaemitting freely,anechoicchamber,transmissionline)

objectofexperiments(human,animal,cellsystem)

examinedpathologicalresults(manifestedillnessandothereffectsonthewholebody)

examinedpathophysiologicaleffects(physiologicaleffectswithapotentialforhealthdamage)

examinedbiologicaleffects(mostlyonthecellularlevel)

examined biophysical and biochemical processes (primary effects on the level ofmolecules,membranesetc.)

methodologyoftheexperiments(proceduresused)

results(includingamentionifourowninterpretationsdifferfromthoseoftheauthor)

statisticalsignificanceoftheresults

appropriatenessof themodel (with regards to the statementsmadeabouteffectsonhumans)

appropriatenessofthemethodology(methodicalweaknessanalysis)

documentationoftheconditionsoftheexperiments(completeness,reproducibility)

contextofotherexperiments(mentionofexperimentswith thesameorcontradictingresults)

meaning(Mainconclusionsdrawnfromtheresults, importancefortheassessmentofhealthrisksforhumans)

Becauseofthedelayofsciencewithregardstotheelectromagneticfrequenciesemittedbymobile telecommunications, a risk analysis cannot be limited to the frequencies and


modulations actually used by this technology. Therefore,we have included all papersexamining carrier frequencies from 100MHz to 10GHz. In the experiments at cellularlevel,butalsoinanimalexperiments,effectshavebeenfoundthatonlyappearatcertainmodulationsorarealotstrongeratthesemodulations(chapter3and4).Atthispointintimeitisnotpossibletodeterminewhetherthemajorityofthefoundeffectsarecausedbythe HF carrier wave or the modulation. This is why we have included all forms ofmodulation into this review.Becauseof thenatureand the importanceof the socalledthermaleffects (chapter3.1)wehavenotsetanexclusion limit forpower fluxdensityandSpecificAbsorptionRate.However,wedidnot includepapers, inwhich theEMFexposure ledtoaconsiderablerise inbodytemperature(>1C)oftheanimalsorhumansubjects.

Whenevaluatingthepapers,wekeptmakingthefollowingobservations:

importantsingleresultsaremaskedforexamplewhendataarepooled

certain observations are dismissed by the authors as blips if they dont fit the(expected/otherwise observed) general trend, without sufficient explanation beingofferedforthisdismissal

singleresultsarenottakenintoaccountforstatisticalreasons,butacommontrendisnotrecognisedornotsufficientlyacknowledged.

In such cases,whenever thiswaspossiblebasedon theexistingdata,weproceeded tomakeourowninterpretations.Whereourevaluationdifferedfromthemainstatementsoftheauthors,itwillbenoted.


3 Primary Reciprocal Effects between High Frequency Electromagnetic Fields and Biological Systems (Biophysical and Biochemical Processes)

3.1 Thermal Effects

3.1.1 Effects of Homogenous Warming

HFelectromagnetic fieldsareabsorbeddependingon the frequencyandpolarisationofthe fields on the one hand and the dimensions and material characteristics of thebiologicalsystemontheotherhand.Theycauseelectriccurrents(dominantintherangeunder 1MHz),polarisation effects andpotentialdifferenceson cellmembranes (in therangebetween1MHzand100MHz)ortriggerrotationaloscillationsofpolarmolecules(mainlywithintheGHzrange).Alltheseprocessescanleadtoawarmingofthebiologicalmaterial if the intensity is sufficient (Ohmic losses in the low frequency range anddielectrical losses in theGHzrange).Theavoidanceofhealthdamagingwarming is thebaseoftheconceptofSAR,expressedby limitingthespecificabsorptionrate,measuredastheenergyabsorptionperunit,toaratewhichwillexcludeoverheatingbasedonthebodysownthermoregulativeprocesses.Forhumans,awholebodyexposureof0.4W/kgcorrespondsapproximatelytohalfthemetabolicbaserate.Inabsenceofheatconductionorotherthermaldissipation,aSARof0.4W/kgwillleadtoatemperatureriseof104K/sek(Foster1996)insofttissuelikemusclesorthebrain.

3.1.2 Microthermal Effects

Thewarmingthroughmicrowavesisfundamentallydifferentfromthewarmingthrougha water bath for example. In the latter case the energy is transmitted by stochasticcollisions.Inmicrowaveheatingitisinthesimplestcasetheelectricalcomponentwhichputs polar molecules within the medium collectively in vibration (3.2.1). Because offriction with the dense ambient medium, the energy is quickly transmitted to thismedium and further dissipated by collisions. When corresponding inner moleculardegreesoffreedomexist,themicrowaveenergycanalsobeabsorbedasaquantumand,inalargemolecule,stored(3.3.).Comparedtoconventionalwarming,theabsorptionofamicrowave quantum is a singular process,which can lead to localisedwarming if theabsorbingmoleculesare suitablydistributed.Liu&Cleary (1995) show ina theoreticalmodelthatatthecellularlevel,membraneboundwatercanleadtofrequencydependentspatialdiscrepanciesindissipationoftheSARandtheinducedHFfields.

Microthermaleffectscanalsobecausedbythenonuniformityofthermalconductivityoftissueatmicroscopiclevel,especiallywhenthewarmingisshort,strongandlocal.Thisisofimportancemainlyfortheevaluationofpulsedfields,becauseinsuchfields,evenatalowaveragepower fluxdensity, theenergyabsorbedduringapulse canbeveryhigh.Radiation in the formof shortpulses can lead to averyhigh rateof temperature rise,which can itself trigger thermoelastic waves, a phenomenon, which is inked to the


acousticperceptionofmicrowaves.AhighpeakSARcanalsotriggerthermallyinducedmembranephenomena(Foster1996).

3.2 Direct Field Effects

3.2.1 Effects from the Electrical Component of the Electromagnetic Field

The electric componentof the electromagnetic field exerts a forceon electrical charges,permanent dipolemoments, induced dipolemoments and highermultipolemoments.The forceson charges create currents,however theseonlyplay a role in the lowerHFrange, causing changes in membrane potentials (stimulation) or thermal effects (seeabove).

Permanent chargedistributions inbiomolecules and cells lead topermanentdipole (orhighermultipole)moments.Theelectricalfieldexertsatorqueondipoles,whichtriestoalign the dipole moment parallel to the field. In alternating fields with not too highfrequencies, the interactions lead to oscillations of the dipoles. In densemedia, theseoscillations are hindered by interactionswith the surrounding particles,which lead toheating(seeabove).Iftheparticlesaretoolargeorthesurroundingparticledensityistoohighorifthefrequencyofthefieldistoohigh,theoscillationscannotdevelop.

The threshold field strengths for the orientation of dipolar cells and other objects ofsimilar size (radius of approx. 1 m) are at 100V/m, the cutoff frequencies inwater(temperature300K)areatcirca0.05Hz,hencefaroutsidetheHFrange.DNAmoleculesandotherbiopolymerscanbeputintooscillationbyfieldswithfrequenciesinthekHzrange.Sphericalproteinmolecules(radiusapprox.5nm)canstillfollowfieldswithfrequenciesup to400kHz,however thisrequires fieldstrengthsof106V/m (Foster1996).Such fieldstrengthsarenotusuallyreachedintheenvironment.

Theinteractionbetweenafieldandaparticlewithaninduceddipolemomentdependsonthe fieldstrength to thepowerof2, thatmeans,acontinuouselectricalalternating fieldinfluences theparticle via a constant torque, however the torque of amodulated fieldfollows the modulation. There is no limitation through a cutoff frequency for theinteractionbetweenafieldandaninduceddipolemoment,howeverforfrequenciesover1MHz, the forces exerted on the cells are very small unless field strengths of severalthousand V/m are used. With such field strengths however, strong dielectrophoreticforcesappear,whichcanleadtocelldeformations,totheorientationofnonsphericalcellsandtothesocalledcoinrolleffect,astringingtogetherofcells.Sincetheinduceddipolemoment depends on the polarizability of the particle and the latter on the size of theparticle, even higher field strengths are needed for smaller bodies than cells(biopolymers).

Electric fields can induce electrical potentials on cell membranes. The size of thesepotentialsdependsontheelectricfieldstrength,thedimensionsofthecell,thefrequencyof the field, the electrical conductivity within and outside of the cell as well as thecapacitanceofthecellmembrane.


With frequencies above 1 MHz the membrane is practically shortcircuited and theinduced membrane potentials become very small. However, theoretical rectificationprocesses and nonlinear phenomena at the cellmembrane have been discussed, andthesecould leadtoan intensificationoftheeffectandtomembranepotentialsthathaveaneffectoncellphysiology.

3.2.2 Effects from the Magnetic Component of the Electromagnetic Field

Withsomeexceptions,biological tissue isnotmagneticand themutualeffectsbetweenthe magnetic component of an electromagnetic field and tissue are generally small.However,thepresenceofmagnetitecrystals,whichhaveastrongcapacitytoabsorbthefrequency rangeof0.5 to10GHzwhich is relevant formobile telecommunications,hasbeen found in the human brain aswell as in the tissue ofmany animals (*Kirschvink1996). Under exposure to amplitude modulated or pulse modulated microwaves, thefrequencyofthecrystalvibrationsvariesaccordingtothemodulationfrequency,andthustransmits it,forexample intheformofanacousticwaveontotheambientmediumandthecellmembrane,whichpossiblyleadstochangesofthepermeabilityofthemembrane(*Kirschvink1996).Theoretical calculations show thatmagnetite transmittedeffects canonlyoccurathighdensitiesofsuperparamagneticparticles(*Dobson&St.Pierre1998).

3.3 Quantum Effects

The quantum energy from radio andmicrowaves in the frequency range between 100MHz to 10GHz is far too low tobreak ionic, covalentorhydrogenbonds.Bohr et al.(*1997) have however shown theoretically, that wring resonances can be triggered inchainmolecules.Thefrequenciesoftheseresonancesareintherangefrom1to10GHzforproteins and 10MHz to 10GHz forDNAmolecules. Thewringmodes ofmoleculesmanifest themselves as torsions in the molecule chain, which can lead to structuralchanges.

The influences ofmicrowaves on structural changes inmolecules have been found inexperimentsusing theproteinLactoglobuline (*Bohr&Bohr 2000).The triggering ofresonantwringmodes can even lead to chainbreaks, sincedue toWhitesTheory, theaddedenergycanbeconcentratedinaverylimitedpartofthemoleculeduringstructuralchanges(*Bohretal.).Inthispart,thechaincanbreak.

3.4 Other Effects

ResonancePhenomena

Whenthefrequencyoftheelectromagneticwavemeetsthenaturalvibrations inthecellstructures or in tissue, it can lead to resonances. Rhythmical fluctuations of signalsubstances, matterexchangeprocesses and Ionconductivity can be found in manyneurones, receptors and cell types. These oscillations can influence the membranepotentialsandswitchcertainstimulionandoff.Anexternalfieldaccordingtotheorycanimprintanexternaloscillationfrequencyontothesestructures.Neuroneswhichhave


beenmodifiedinthiswaycanevensynchronisethefollowingneuronesinthesameway.Thisexternalsynchronisationwouldevenremainafterthedisappearanceoftheexternalstimulus.

IndirectEffects

Inaddition to thepreviouslydescribed triggeringofwringresonances,microwavescanpossiblydamagethegeneticsubstanceviatheformationofhydroxylradicals.Theinputenergyofmicrowaves is sufficient to raise the ratioofoxidants toantioxidants,a selfacceleratingchainreactionoffreeradicalscanleadtodamageoftheDNA(Scott1992,seealsoMaesetal.1995).

3.5 Particular Properties of Pulsed Electromagnetic Fields

In an evaluation of circa 40 studies, in which the biological effects of pulsed highfrequencyfieldsweredirectlycomparedtothoseofcontinuousfieldsofthesamemedianpower density, Postow & Swicord (1996) concluded that in half of the studies, thebiological effectivenessofpulsed fieldswas significantlyhigher.Only in a few studieswere the continuous fields more effective and in the remainder of the studies theeffectivenessofbothwaspracticallythesame.Thestudieswhicharemainlydiscussedinchapter4and5conveyasimilarpicture.

At first glance, the higher biological effectiveness of pulsed electromagnetic fields incomparisontocontinuousfieldsatthesamemedianpowerfluxdensitiescouldhaveanalmosttrivialcause:

The individual pulses of pulse modulated fields have a higher amplitude than thecontinuous fields; thepossible threshold for the triggeringofbiological reactions couldthereforebepassedinthesefieldsduringthedurationofthepulse.

However, numerous experiments found that the biological response depends in acomplicatedmanner on the duration of the pulse and its frequency.Given that someeffectshaveonlybeenobservedatcertainpulsefrequencies,wepresumethatinadditionto the described effect, there are otherswhich can be originally attributed to the lowfrequencymodulation(seealsochapter4).


4 Biological Primary Effects of High Frequency Electromagnetic Fields Effects on Cellular Level

Atthecellular level, it ispossiblethattheremaybedirecteffectsoftheEMfieldonthegeneticmaterial,whichwehavecollatedundertheheadingGeneToxicityandwhichwillmanifestasmutationsifthecellsownrepairmechanismsfail.Ontheotherhand,itisalsopossible that the fields influencecellularprocessessuchasgenetranscriptionandgenetranslation.Furthermore it ispossible that the fieldscan impacton thecellmembranes,the intracellular processes of signal transmission and not least the cell cycle. Just likedirect damage of the genetic substance, a disruption of these processes can lead to atransformationofthecell,todisruptionsofintercellularcommunicationandtoachangedrateofcelldivision,whichcan lead toaslowerorvery importantlywithrespect toapotentialcarcinogeniceffectfastergrowth.

4.1 Gene Toxicity

Abasicquestionfortheevaluationofthepotentialdangersofmobiletelecommunicationiswhethertheelectromagneticfieldsusedaregenotoxic.Ifthefieldshadthepotentialtodamage genetic substance directly, they would not only amplify the effects of othercarcinogenic teratogenic ormutagenic substances, but theywould induce these effectsthemselves.Adirectgenotoxiceffectofelectromagneticfieldswithfrequenciesastheyareusedformobiletelecommunicationshasbeenthoughttobenotlikelyinthepast(Brusicket al. 1998,Moulder et al. 1999, Repacholi 1997, Repacholi 1998, Saunders et al.1991,Verschaeve1995,Verschaeve&Maes1998).Thereasonsforthisassumptionwereontheonehand that thequantum energy contained inEM field in the radio andmicrowaverangewasnotsufficienttobreakmolecularbonds.Thisassumption isno longertenableaftertheexperimentsofBohretal.(*1997)andBohr&Bohr(*2000)(seealsochapter3.3).Ontheotherhand,itwasarguedthattherewasalargenumberofexperimentsshowingnogenotoxiceffects.Our listofpapers inAnnexA,TableA.1showshowever, that themuchdebatedfindingsoftheworkofLai&Singh(*1995),inwhichthedirectdamageofDNA(singlestrandanddoublestrandbreaks)hasbeenproven,havebeenconfirmedbyawhole rangeofother studies, someby the same laboratory,butalsobyothergroups(*Lai&Singh1996,1997, *Phillips1998, *Sarkar1994).A studybyVarma&Traboulay(1977) on the effect ofHF fields on pureDNA had already resulted in hints of directgenotoxiceffects,however,thisexperimentusedarelativelyhighpowerfluxdensityandthereforesignificantwarmingmayhaveoccurred,at least locally.LaiandSingh (*1997)found that the dispensation of melatonin and NTertButylalphaPhenylnitron (PBN)before theEMFexposureprevented theoccurrenceofDNAbreaks.MelatonincapturesfreeradicalsandforPBNithasbeenproventhatitprotectscellsfromcelldeathinducedbyfreeradicals.

InAppendixTableA.1wealsolisttheexperimentofMeltzetal.(*1987)andStaggetal.(*1997)whichexaminedtheinfluencesofEMFfieldsontheDNArepairmechanismsandtheDNAsynthesis.


ThetermchromosomeaberrationsumsupallanomaliesoftheDNAdoublestrandlevelwith respect to chromatids and chromosomes. Examples for structural chromosomeaberrationsare:chromatidandchromosomebreaks,chromatidgaps,acentric fragmentsaswellasdiandtetracentricchromosomes.

Chromosomeaberrationshavebeenobservedinamultitudeofexperimentalconditions,in vivo aswell as in vitro (TableA.1).Maes et al.(*1997) found a rise of chromosomeaberrations in human lymphocytes in workers who were professionally exposed toradiation from mobile equipment, but also in experiments with human blood undercontrolledexposure conditions (GSMbase station,15W/m,exposure timeof2hours).However,thiswastheonlystudysofarwhichusedtheactualfieldsofarealbasestation.

Theincidenceofmicronucleiindicateswhetherthedistributionofchromosomesintothedaughternucleiafteracelldivisionhasbeennormalandcomplete.Anumberofstudieshave proven a higher incidence ofmicronuclei under the influence ofHF EMF fields,which is interpreted as an indication for chromosome damage (Table A.1).With oneexception, the frequencieswere all over 1GHz and inmost cases the intensitieswererelativelyhigh.

For the incidenceofsisterchromatidexchangeasameasure fordamageatDNAsinglestrand level,onlyveryfewstudiesusingtypicalmobilefrequenciesand intensitieshavebeendoneso far (TableA.1).Maesetal.(*1996) found that theradiationofaGSMbasestation (954MHz,217Hz,durationofexposure:2hours) raises thegenotoxiceffectsofMitomycinCsignificantly,demonstratedviathesisterchromatidexchange.

Geneticdamagecan lead tocellmutationwithpossiblydamagingeffects for the livingorganism.Mutationswhichpromotefastercelldivisionwillbediscussed inchapter4.3.TableA.1showsinitslastblocksomestudieswhichfocussedontheevidenceofchangesin the genetic materials which manifest themselves as changed properties within theorganism.

4.2 Cellular Processes

4.2.1 Gene-Transcription and Gene-Translation

The code of the DNA controls protein synthesis in the ribosomes via the RNA. ThecreationofRNA, i.e. the imprintingofgenetic informationhappens in the cellnucleus(transcription).TheencodedinformationistransportedviamessengerRNA(MRNA)totheribosomesandisreadtherewiththehelpofTransferRNA(tRNA).Accordingtothetransmitted code, proteins are subsequently synthesized. This process of synthesis iscalledtranslation.SinceonemRNAchaincanbeusedbyseveralribosomes,therateofsynthesis of the corresponding protein can be a lot higher than that of the mRNA.Mistakesmadeduringthegenetictranscriptioncanthusberaisedtoahigherpowerattheproteinlevel.

In the first block of Appendix Table A.2, we list several recent studies whichdemonstrated changes of gene transcription and translation under the influence ofelectromagnetic fields of mobile telecommunications. Fritze et al. (*1997) observed


changedgenetranscriptionincertainareasofthebrainsofratswhichhadbeenexposedtothefieldofaGSMphoneforfourhours.

Inan invitroexperiment, Ivaschuketal. (*1997)exposedcells toapulsemodulatedHFfield(836.55MHz,TDMA50Hz)andafterwardsextractedandanalysedtheentirecellularRNA.

This showed statistically significant changes with regards to the transcription of theresponse gene cjun (90W/m, duration of exposure: 20minutes), however no changeswith regards to cfos.The resultsof theexperimentsbyGoswamietal. (*1999) foundaevidenceforaninfluenceonthetranscriptionoftheresponsegenecfosbyasimilarfield,whilstforcjunandcmyc,nostatisticallysignificanteffectwasobserved.The intensitiesatwhich effects on gene translationhad been observedwerewell below the values atwhichthermaleffectswouldoccurinmammals.

4.2.2 Membrane Function

Thereisalargenumberofexperimentalevidencethathighfrequencyfields,nonpulsedand pulsed can affect different properties of the ion channels in cellmembranes, forexample in the formofa loweringof the rateof channel formationor the reductionoffrequency of the opening of individual channels (Repacholi 1998). The frequency ofopeningsof ionchannelswhichareactivatedbyacetylcholinewassignificantly loweredbyamicrowavefield(10.75GHz)withapowerfluxdensityofafewW/cm.(*DInzeoetal.1988). Changes of the membranes as a whole have also been observed under theinfluenceofweakfields.Thus,Phelanetal.(*1992)observedthata2.45GHzfield,withapulse modulation of 100 Hz could trigger a phase transition from liquid to solid inmelatonincontainingcellsafteranexposureof1hourataSARof0.2W/kg.

4.2.3 Signal Transduction

Ca2+

ThedivalentCalciumcationCa2+playsan importantrole in thecellsignaltransduction:regulatingtheenergyoutput,thecellularmetabolismandthephenotypicalexpressionofcellcharacteristics.

ThesignalfunctionoftheCa2+isbasedonacomplicatednetworkofcellularchannelsandtransportmechanisms,whichmaintainstheCa2+concentrationwithinthecellata lowerlevel than outside, but which is also linked to dynamic reservoirs. This allows thetransduction of extracellular signals (hormones, growth factors) as Ca2+ peaks in thecytosol, transmitting information encoded in their intensityand frequency. It isknownthat this signal process can be disrupted by a variety of toxic chemicals in theenvironment,whichcanleadtocelldamageandevencelldeath(Kass&Orrenius1999).

StudiesbyBawinetal.(*1975)andBlackmanetal.(*1979)showedveryearlyoninvitroexperimentsthattheCa2+balanceofnervecellsandbraintissuecanbedisruptedbyHFfieldswithlowfrequencyamplitudemodulations.


Bothstudiesworkedwithamplitudemodulated147MHzfields(withintensitiesrangingfrom5 to20W/m2).Themaximumeffectoccurredatamodulation frequencyof16Hz.Experiments conducted byDutta et al. (*1984 *1989) and LinLiu&Adey (*1982) alsoshowedsignificantdependenceonthemodulationfrequencies, insomecasesatSpecificAbsorptionRatesofaslowas0.5W/kg.Equally,Somosyetal.(*1993)foundthataneffectonthedistributionofCainintestinalcellsisonlypossiblewithinafieldmodulatedwithalow frequency. Wolke et al. (*1996) observed in their experiment on myocytes thatexposuretofieldswithmobilelikecarrierfrequenciesof900MHzand1800MHzresultedinlowerintracellularconcentrationsofCa2+forallmodulationfrequencies(16Hz,50Hz,217Hz,30KHz)comparedtoexposurestoacontinuous900MHzfieldornoexposureatall.Astatisticallysignificanteffectwasonlyfoundwiththecombinationofacarrierwaveof900MHzandamodulation frequencyof50Hz.TheSpecificAbsorptionRate for thisexperiment was between 0.01 and 0.034 W/kg, far below the range which might berelevantforthermaleffects.

Enzymes

Protein kinases are enzymes with the property to phosphorylate other enzymes orproteins. Phosphorylation, a covalentmodification by addition of a phosphate group,changestheactivityorfunctionofaprotein.Theproteinkinasesplayanimportantroleinthe transmission of information from the membrane receptors for hormones andcytokines into the interior of the cell, and thus in the regulation ofmany intracellularprocesses such as glucose and lipid metabolisms, protein synthesis, membranepermeability,enzymeintakeandtransformationbyviruses.

Anamplitudemodulated450MHz field iscapableofdecreasing theactivityofproteinkinaseswhicharenotactivatedbycyclicalAdenosinemonophosphate.Byusetal.(*1984)showed that the degree of inactivity depended on the exposure time as well as themodulationfrequency.Maximumeffectsoccurredatexposuretimesof15to30minuteswithamodulationfrequencyof16Hz.

Theenzymeornithinedecarboxylase (ODC)determines thespeedof thebiosynthesisofpolyamines. Polyamines are needed forDNA synthesis and cell growth.ODC is alsoactivated in relation to carcinogenesis.The controlofOCDactivity from theexterior isfacilitatedviaprocessesonthecellmembrane.Byusetal.(*1988)exposedthreedifferentcell types (rathepatomacells,eggcellsof theChinesehamster,humanmelanomacells)foronehour toa450MHz fieldwitha16Hzamplitudemodulationandapower fluxdensity of 10W/m2. The exposure raisedODC activity by a littlemore than 50%. TheheightenedODC activity remained for several hours after the exposure. Similar fieldswitha60Hzanda100Hzmodulationhadnoeffects.Anotherstudy(*Penafieletal.1997)observed heightenedODC activity after the radiation ofL929cells ofmicewith a 835MHzfieldwhichhadbeenamplitudemodulatedat60Hzorpulsemodulatedat50Hz.Noeffects whatsoever were observed with an analogue mobile phone, a frequencymodulationat60Hzandaspeechamplitudemodulation.Thislastfindingconfirmsotherresultsbythesamegroup,accordingtowhichaminimumcoherencetimeof10secondsof the fieldneeds to bepresent for an effect onODC activity tomanifest (*Litovitz etal.1993, 1997, see also Glaser 1998 and Litovitz 1998). The coherence time of speechmodulatedfieldshoweverisshorterthanasecond.


Further importantproof that low frequencymodulationhasadetermining influenceontheeffectsofelectromagneticfieldsonenzymeactivitywasfoundbyDuttaetal.(*1994):Theycomparedtheeffectsofalowfrequencymodulated147Hzfield(0.05W/kg)andacombinedlowfrequencyelectricandmagneticfield(ELFEM,21.2V/97nT).Acontinuoushigh frequency fieldonlyhad a small effect (3.6per cent)on the activityof enolase inEscheriaColi,a16Hzmodulatedfieldledtoanincreaseinactivityofnearly62percent,a60Hzmodulatedfieldledtoadecreaseofactivityof28.5percent.AtELFEMasimilarresponse couldbeobserved: increaseofenzymeactivitybymore than59per centatafrequencyof16Hzanddecreaseof24percentat60Hz.TheresultsoftheexperimentsbyBehari et al.(*1998) point in the same direction. They found that a 30 to 35 day longexposure of rats to amplitudemodulated fields (6.11 9.65W/kg) led to a significantincrease inNa+K+ATPaseactivity,whichwas independent from the carrier frequency,but characteristically dependent on themodulation frequency, because the effectwasalwaysstrongerata16Hzmodulationthanata76Hzmodulation.

4.2.4 Cell Cycle

Anundisruptedsignaltransductionorefficientcellcyclecontrolmechanismswhicharecapableofcorrecting false informationor facilitatingrepairsare theprerequisite forcellcycleprogressionifthegenomicintegrityofthecellistobemaintained(Shackelfordetal.1999).Disturbancesof theDNA replication can lead todetrimentalmutationsandasaconsequence to cell death or in multicellular organisms to cancer. The causes forirregularities in the course of the cell cycle are almost always to be found inmistakesduringsignaltransductionand/orthefailureofcontrolmechanisms.

InAppendixTableA.2.we liststudieswhichexamineddisruptionof thecellcycle.TheonlyinvivoexperimentistheonebyMankowskaetal.(*1979)whichalsousedintensitiesastheyarefoundintheenvironmentofrealemittingequipment.Statisticallysignificantincreases of disrupted metaphases with uni, quadri and hexavalencies weredemonstratedinthisstudyfromapowerfluxdensityof5W/m2.

Clearyetal.(81996)found intheirexperimentthat2.45GHzfieldsareroughlytwiceaseffective as 27MHz fieldswhen it comes to the triggering of cell cycle disturbances.Whilstthe27MHzfieldshadnoinfluenceontheG2/MphaseofeggcellsoftheChinesehamster,disturbancesofallphaseswereobservedina2.45GHzfield.

4.3 Cell Transformation and Cell Proliferation

Invitroexperimentsof theeffectsofhigh frequency fieldson therateordivisionor therate of proliferation of cells, expressed in the proliferation rate and the (neoplastic)transformationofcellscanofferimportantfindingswithregardstopossiblecarcinogeniceffectsofthefields.Theadverseinfluencesofthefieldswhichcouldnotbepreventedbythecellsownrepairmechanismsmanifestthemselvesindisruptedcellproliferationandcelltransformationrates.

TableA.3givesanoverviewofthestudies,inwhichtheeffectsofhighfrequencyfieldsoncelltransformationandcellproliferationrateswerethefocusoftheexaminations.


4.3.1 Cell Transformation

BalzerKubiczek & Harrison (*1985, *1989, *1991) found an increase in neoplastictransformationsincellswhichhadbeenexposedinvitrotoahighfrequencyfieldwithalow frequency pulse. The effect depended on intensity, butwas only observable, if atumourpromoter(TPA)wasaddedaftertheexposure.

Czerskaetal. (*1992) found that low frequencypulsedmicrowave radiation (2.45GHz)increased the rate of transformation of small inactive lymphocytes into large activatedlymphoblasts.Continuousradiationcouldtriggerthiseffectonlyatpowerfluxdensitiesthatalsoledtomeasurablewarming.

However,theexperimentswithpulsedradiationwhichtriggeredthecelltransformationatpowerfluxdensities,forwhichahomogenouswarmingcanberuledout,showedthathomogenouswarmingcannotberesponsibleforthiseffect.

4.3.2 Cell Communication

Disruptedcommunicationbetweentransformedcellsandnormalcellsplaysanimportantrole in tumorpromotion.Cainetal. (*1997)cocultivated transformedcellswithnormalcells.Thecoculturewasexposedfor28daystoaTDMA(50Hz)modulated836.55MHzfield aswell as to the tumor promoter TPA in various concentrations.At power fluxdensitiesof3and30W/m2,whichcorrespondedtoSpecificAbsorptionRatesof1.5and15mW/kg,theydidnotfindastatisticallysignificantdifferenceoffocusformationbetweentheexposedandthecontrolculturesforanyoftheTPAconcentrations.Thedataforthelowestintensity(0.3W/m2/0.15mW/kg)showfortwoofthethreeTPAconcentrationsthattherewasasmallbutstatisticallysignificantdifferenceinthenumberoffoci,andforthelowestTPAconcentrationalsoforthesurfaceanddensityofthefoci.

4.3.3 Cell Proliferation

Anderstam et al. (*1983) found in their experiments with bacteria that some strainsreacted to theexposurewithanamplitudemodulated2.45GHz field (500Hz,35 to100W/kg)withan increasedproliferation.Also for some species, thenumberofmutationsand the frequency of mutations were increased. These results were confirmed byHamneriusetal.(*1985)amongstothers.Grospietschetal.(1995)foundsimilarresultsfor150MHzfieldswithseveralamplitudemodulations.

Clearyetal.(*1990a,b)demonstratedonhumanlymphocytesandonGliomacellsthattherateofcelldivisionwas increasedafterexposurewithacontinuous2.45GHzfield.Inanewer experiment, the same effect could be observed for exposures with a pulsemodulatedfieldofthesamecarrierfrequency(*Clearyetal.1996).

In the firstof the twoexperimentswhichwereconductedwith fieldsdisplayingall thecharacteristics of realpulsedmobile emissions (see alsoTableA.3), an increasedDNAsynthesisratewasobserved,butnofasterproliferationoftheexaminedcellswasfound.(*Staggetal.1997). In thesecondexperiment,atsimilarly low intensities (0.0021W/kg)however, transmitted by a GSM modulated 960 MHz wave, an increase of the cell


proliferationratewasfound(*Velizarovetal.1999).TheEMFexposureinthisexperimentwasconductedat twodifferent temperatures,whichalsoapplied to therelatingcontrolcultures.Theincreaseoftheproliferationrateonlyhappenedintheexposedcellcultures.Similar experiments to prove that microwaves and conventional heat have differenteffects,wereconductedbyLaCaraetal.(*1999)onathermophilebacterium,inwhichtheradiationwith a 10.4GHz field led to an irreversible inactivation of the thermostableenzyme galactosidase, whilst heating in a water bath had no effect. This resultconfirmedtheresultsofSaffer&Profenno(*1992)whichhadworkedwithfrequenciesinthelowerGHzrange.


5 Patho-Physiological Effects

5.1 Immune System

The immune system plays a central role in the protection against infectious microorganismsintheenvironmentand,also,againstseveralkindsofcancercells.Experimentsonhamsters,miceandratsfound,amongstotherthings,thattherewasareductionintheactivityofnaturalkillercellsandanincreaseinmacrophageactivity(seee.g.Yangetal.1983;RamoRaoetal.1983;Smialowiczetal.1983).However,themajorityofexperimentsonlivinganimalswerecarriedoutatpowerfluxdensitylevelsthatproducedanincreaseinbodytemperatureofmorethan1oC.Ontheotherhand,itwasobservedinparallelinvitro experiments, that in vitro heating of macrophages did indeed lead to increasedactivity;theeffectwas,however,weakerthanthatoftheinvivoradiationwhichproducedthesametemperature(RamoRaoetal.1983).

Elekesetal.(*1996)observedthat,afterexposingmiceforaperiodof3hoursperdayoverseveraldaysusingmicrowaves (2.45GHz)withapower fluxdensityof1W/m2 (SAR=0.14W/kg),therewasanincreaseinantibodyproducingcellsinthespleenofabout37%withcontinuousradiationandaround55%withamplitudemodulatedradiation.

In contrast to the in vivo experiments,numerous in vitro experimentswere carried outwith intensities atwhich an effect due towarming can be excluded. Thus, Lyle et al.(*1983)observedaninhibitionofcytotoxicityofTLymphocytesinthemousewitha450MHzfieldthatwasamplitudemodulatedwithvariousfrequenciesintherangebetween3Hzto100Hz.Theeffectthatwasdemonstratedwitharelativelylowpowerfluxdensityof15W/m2wasgreatestatthe60Hzmodulation.Theinhibitionofcytotoxiceffectivenessof the irradiated lymphocytes declined continually for both the lower and highermodulationfrequencies.

The tables inAppendixA list further experimentswith (human) leucocytes inwhichdamagingeffectswereprovenatnonthermalpower fluxdensity levels,especiallyalsowithlowfrequencyamplitudemodulatedfields.

TheworkofMaesetal.(*1995)deservesspecialconsideration. Inan invitroexperimentwith human leucocytes at a GSM base station and also in the examination of thelymphocytesinthebloodofworkerswhowereexposedtothefieldsofthemobilephonebase stations during maintenance work, they found that there was an increase inchromosome damage (chromatid breakage, acentric fragments and some chromosomebreaks).

5.2 Central Nervous System

5.2.1 Blood Brain Barrier

Thebrainofmammalsisprotectedfrompotentiallydangerousmaterialsinthebloodbythe blood brain barrier, a specialized neurovascular complex. The blood brain barrier


functionsasaselectivehydrophobicfilterthatcanonlybeeasilypassedthroughbysmallfatsolublemolecules.Othernonfatsolublemolecules,e.g.glucose,canpassthroughthefilterwiththehelpofcarrierproteinsthathaveahighaffinityforspecificmolecules.

Itisknownthatalargenumberofdisordersofthecentralnervoussystemarecausedbydisturbancesofthebarrierfunctionofthebloodbrainbarrier(*Salfordetal.1994).

Severewarming of the brain can lead to an increased permeability of the bloodbrainbarrier for thosematerialswhosepassage should actually beprevented.The results offirst experiments with high frequency fields of high intensity, which led to a higherpermeability of the blood brain barrier, were then interpreted as a consequence ofwarmingbytheHFradiation.

However,AppendixTableB.1listsawholeseriesofstudiesinwhichagreatlyincreasedpermeability of the blood brain barrierwas produced through pulsed high frequencyfieldsofvery low intensity (*Oscar&Hawkins1977, *Neubaueretal.1990, *Salfordetal.1994, *Fritze et al.1997) amongst others with carrier frequencies and modulationfrequencieswhichcorrespondedtothoseofmobiletelephony(GSM).

5.2.2 Neurotransmitters

Pulsed and continuous high frequency fields of low intensity may lead to chemicalchangesinthebrain.Inabaetal.(*1992)exposedratstoacontinuous2.45GHzfieldwithapowerfluxdensityofbetween50to100W/m2andfoundasignificantreductionintheNoradrenalin content of the Hypothalamus, whilst the two other neurotransmittersDihydroxyphenylaceticacidand5Hydroxyindolaceticacidwere found in theponsandmedulla oblongata in significantly increased concentrations. The radiation did notproducesignificantchangesinthedopamineorserotoninconcentrations.

Laietal.(*1987,1989a,b,seeaboveLaietal.1988)foundalsoinexperimentsusingratsthata2.45GHzfieldmodulatedwith500Hzpulsemodulationinfluencesbrainactivity,especially in the frontal cortex and the hippocampus, via the most importantparasympatheticneurotransmitteracetylcholine.Itcouldbedemonstratedthattheeffectwas related to theexposureduration.A45minuteexposureduration led to significantreductionsincholineuptake,thereductionto20minutesexposureproducedasignificantincrease.Asimilarbehaviourwasfoundinanimalsalsoasareactiontostressthroughthereductionofthefreedomofmovementandthroughacousticwhitenoise.

5.2.3 Electroencephalogram (EEG)

In contrast to theneuroendocrineeffects,which canbarelybemeasureddirectly in thebrainofhumans,EEGstudiescanbecarriedoutrelativelyeasily.Severalvalidstudiesofthatkinddonowexist.

Mostanimalexperimentshavelimitedvalidity,sincetheywerecarriedoutwithrelativelyhighpower fluxdensityvalues (see e.g.Chizhenkova 1988: 2.397MHz, cw, 400W/m2,Chizhenkova&Safroshkina1996:799MHz,cw,400W/m2,Thuroczyetal.1994;2.45GHz,AM16Hz,100W/m2).


Oneof the fewexceptionsare the studiesbyVorobyovetal. (*1997),whoobservedanincreaseon the leftrightsymmetry in theEEG in rats thatwereexposed toa945MHzfield(AM,4Hz,1to2W/m2,withinthefirst20secondsafterthestartoftheexposure.

Early experiments by vonKlitzing (1995)with EEG recording during the exposure ofsubjects topulsedhigh frequency fields, thatwere similar to thoseofmobile telephonefields (150MHz,217Hz,power fluxdensity in thepulse in thebrainata6 cmdepthbelow 102 W/m2), found changes in the awake EEG, these were called into questionbecauseofinsufficientdocumentation.

Inlaterexperimentshowever,acleareffectwasdemonstratedintheawakeandsleepingEEGs.

Reiser et al. (*1995) observed, both with exposures to a 150 MHz field (modulatedfrequency9.6Hz,peakpower0.5mW,4cmdistance,nearfieldconditions)andalso inthefieldofamobiletelephone(902MHz,modulationfrequency217Hz,peakpower8W,40cmdistance),asignificantincreaseintheenergyintheEEGfrequencybandsAlpha,Beta1andBeta2.

ExperimentsbyRschke&Mann(*1997)resultedinnosignificantdifferenceintheEEGsforexposedandshamexposedsubjectsundershortexposureconditions(3.5minutes,900MHz,GSM, 0.5W/m2).However, the peak of approx. 9Hz in the presented averagedpowerdensityspectraofexposedsubjectswasclearlylowerandnarrowerthanfornonexposedsubjects.Thesameauthors(*Mann&Rschke1996)demonstratedagain in thefieldof aGSMmobile telephone (8W,distance 40 cmpower fluxdensity 0.5W/m2), areductionof the time taken to fallasleepanda statistically significant reductionof thedurationandtheproportionoftheREMsleep.Furthermore,thespectralanalysisrevealedanincreasedpowerdensityoftheEEGsignalduringREMsleepaboveallintheAlphafrequencyband.TheREMsuppressiveeffectand thereductionof the time taken to fallasleepwerealsoconfirmedbythesameresearchteam(*Mannetal.1997,*Wagneretal.1998). The study carried out in 1997 also found a significant increase in the cortisolconcentration in thebloodofhumansexposed toa900MHz/217Hz fieldwithapowerfluxdensityvalueof0.2W/m2.SystematicdeviationswerealsoobservedfortheGrowthHormoneandMelatoninlevels,butthesedidnotreachsignificancelevel.

Whilst in thepreviouslycitedstudies,changes in thesleepEEGcouldbedemonstratedonly as a consequenceof the influenceofmobile telecommunications fields for severalhours,Borblyetal.(1999)wereabletodemonstratethatchanges insleepwerealreadyoccurringafter15to30minutesexposure.Thisresearchteamusedalsoa900MHzfield,which couldbe selectivelypulsemodulatedwitheither2,8,217or1736Hz.As in theotherexperiments,astatisticallysignificantreductionintheproportionofREMsleepwasfoundataSpecificAbsorptionRateoflessthan1W/kg.Inaddition,thewakingupphasewasnoticeablyreduced.

Freude et al. (*1998, see alsoHenschel et al. 1999) examined the effectof the radiationfrommobiletelephonesonslowbrainpotentials.

Slow brain potentials are eventcorrelated brain potentials that arise during thepreparation formotoractionand/or informationprocessing.Changes in the slowbrain


potentials give an indication about the influences on specific aspects of humaninformationprocessing.Freudeetal. found that the fieldsofamobile telephone (916.2MHz,217Hz,SAR0.8821.42W/kg,exposure time3 to5minutes) led toa statisticallysignificantdecrease of the slow readinesspotentials for specific tasks, in specific brainareas.

5.2.4 Cognitive Functions

Impairmentsof thebrain,e.g.bymodificationof thecholineuptake,canbeexpected tocauselearningdeficits.Theseweredemonstratedinmanylearningexperiments,inwhichratswerepreviouslyexposedtopulsedmicrowavefields(*Laietal.1989,1994;*Wong&Lai2000,seeaboveDAndrea1999forolderstudies).InthestudybyLaietal.(*1994),ratswereexposedfor45minutestoa500Hzpulsed2.45GHzfieldwithapowerfluxdensityof10W/m2.ThisintensityresultedinameanwholebodySARof0.6W/kg.Followingtheexposure,thestarvedratswereplacedinalabyrinthwithseveralarmsinwhichfoodwasplaced. The researchers measured how effectively the exposed rats and the shamexposedratssearchedthelabyrinthforfood.Fortheexposedgroup,significantlymorefailedattemptswereobserved,i.e.searchingalreadyemptiedlabyrintharms.Theauthorsattributed the lowperformanceof the exposed rats todeficits in spatialmemory.ThehandicapoftheEMFexposurecouldbelevelledoutinafollowupexperiment,inwhichtheratsweregiveneithertheacetylcholineagonistPhysostigminortheopiateantagonistNaltrexone before their exposure. According to the authors, these findings areconfirmationoftheirresultsfrompreviousstudies(seeabove),inwhichtheyhadfoundthathigh frequencyelectromagnetic fields influencecholinergicandendogenousopioidneurotransmittersystemsinthebrainandthatthiseffectcanleadtomemorydeficits.Inthemeantime,theeffecthasbeenconfirmedbyotherexperiments(Mickley&Cobb1998).

Inafurtherexperiment(*Wang&Lai2000),ratsweretrainedoverseveralsessionstofindaplatformsituatedjustunderthewatersurfaceinsidearoundwaterbasin.Subsequently,theywereexposedtopulsedmicrowaveradiationforanhour(2.45GHz,500pulsespersecond,meanpowerfluxdensity2W/m2,meanwholebodySAR1.2W/kg).Testingwasthen carriedout todeterminehow long the exposed ratsneeded to find theplatformfromdifferent startingpositions, compared to the nonexposed ratsor shamexposedrats.The exposed rats clearly required longer for this,as they spent significantly lesstime in the correct quadrant of the water basin. Finally, the recorded traces of theswimminglanesusedbytheexposedanimalsdifferedfromthoseofthecontrolgroups,this suggests thatdifferent strategieswereusedwhen searching for theplatform.Thisresult confirms the findings from other studies that pulsed high frequency fields caninfluencespecificaspectsofmemoryperformance.

Theeffectsofa600MHzfieldonthememoryofratswerealsodemonstratedbyMickleyetal.(*1994).Inthisexperiment,thecapacityoftheanimalstorecognizefamiliarobjectswasmeasuredinrelationtotheradiationtheyreceived.Whilstthenonexposedcontrolanimals and also the animals who were exposed to a SAR of 0.1 W/kg occupiedthemselves for longer with a novel object compared to a familiar object, the higherexposedanimalsspent justasmuchtimeexamininganactuallyfamiliarobjectaswitha


novelobject.Thelimitforthisexposuredependentchangeinbehaviourwasbetween0.1and1.0W/kg

ThelowestSARsofarwhichhasbeenshowntohaveaneffectoncognitivefunctioninginratswas0.072W/kg.However, in thisexperiment,pulseswithapeakofmore than700MW(megawatts)wereused(Raslearetal.1993).ThelowSARinthiscaseresultedonlyfromaveragingovertimewithaverylowpulserepetitionrateof0.125pulsespersecondandapulsewidthofonly80nsec.

Ithasbeen shown in experimentsbyPreece etal. (*1999) that fields like thoseused inmobiletelephonycaninfluencecognitivefunctionsofthebrain.Inthisstudy,36subjectsweresubjectedtoa915MHzfieldofasimulatedmobiletelephone.Thefieldwasoverlaideitherwitha217Hzsinusoidalmodulationora217Hzpulsemodulation.Intheanaloguesimulation thenet forwardpowerwasaboutoneWatt,and in thedigital simulation itwas0.125Watt.Under theconditions Exposure toanalogue field, Exposure todigitalfield or Sham exposurewithout any field, each of the test persons had to carry outseveralteststomeasureabilitytoreactandvarioustestsofmemoryperformance.Inbothexposedgroups therewasa slightbut statistically significantdecrease in reaction time,whichwasmoremarkedforAnalogueexposurethanforDigitalexposure.

5.3 Hormone Systems

5.3.1 Stress Hormones

Environmental pollution can act as a stressor on the body, like physical and mentalstressors, and cause alarm reactions. Such reactions are associated with hormonalchanges.Thepresenceofa stresssituation canbeprovedby thepresenceofhormoneslike adrenocorticotropin [the adrenocorticotrophic hormone] (ACTH), cortisol andcorticosteroneintheblood,andalsotoalesserextentbychangesintheconcentrationofprolactinandgrowthhormone.

Electromagneticfieldscanclearlycausestressreactionsinanimalsusedforexperiments.Thus,theexperimentbyImaidaetal.(*1998a)onratsthatwereexposedforadurationof90minutesdailyoveraperiodof6weekstoafieldwithacarrierfrequencyof929.9MHzanda50Hzpulsemodulation,showedastatisticallysignificantincreaseintheACTHandcorticosteronelevels.ThewholebodySARvalueinthisexperimentwasbetween0.58and0.8W/kg.Theexposureinthe1.439GHzfield,equallywitha50HzpulsemodulationandaSARvaluebetween0.453and0.680W/kghadthesameeffect(*Imaidaetal.1998b).

Chouetal.(*1992)exposedratsinalongtermexperiment(25months)to800MHzpulsemodulated 2.45GHz field that led to a SpecificAbsorption Rate of 0.15 to 0.4W/kg.Alongside other physiological parameters the corticosterone profile was regularlymeasured for the first half year of the experiment.Whilst the hormone profile of theexposed animals and the nonexposed animals were practically identical in the laterstages of the experiment,with the exception of a slight increase in the shamexposedgroupofanimalsinthethirdphaseoftheexperiment,thefirstexaminationafter6weeksexposure showed a statistically significant increase in the corticosterone profile in thebloodoftheexposedanimals.


The authors report that their attempt to replicate this effect produced no statisticallysignificantresults,however,only20animalsweretestedinthissecondexperimentwhilsttheactualseriesofexperimentscontained200animals.

A similarly extensive experiment on rats like that of Chou et al. However, with anunmodulated435MHzfieldshowednodifference intheconcentrationofthehormonesACTH,corticosteroneandprolactinbetween theexposedanimalsand thenonexposedanimals(Toleretal.1988).

The few experiments previously carried out on humans do not yet produce a clearpicture.Mannetal. (*1998)exposed24volunteersubjectswhilstasleep to the fieldofamobiletelephonethatwastransmittedfromaseparateantenna(900MHz,217Hz,0.2W/m2).Blood sampleswerewithdrawnviaa catheterwhilst the subjectswereasleepandthey were analysed for, amongst other things, cortisol and growth hormoneconcentrations.Thereweresystematicdifferencesbetweentheexposedsubjectsandtheshamexposed subjectsduring the courseof thenight forbothhormones,whichonlyreachedstatisticalsignificancelevelsforcortisol.

DeSezeetal.(*1998)examinedtheeffectofaGSMmobiletelephone(900MHz,217Hz)onsubjectswhowereexposed to the field for2hoursperday,5daysperweek foroveramonth.Based on nine blood samplewithdrawals perweek; amongst other things, thechange intheconcentrationsofACTH,growthhormoneandprolactinweredeterminedovertime.

Theauthorsevaluationof theirstudieswasthatatonemonth, intermittentexposure intheradiofrequentfieldfromthemobiletelephonehadnolastingoraccumulativeeffectsonthehormonesecretionsfromtheanteriorlobeofthepituitarygland.Intheirdata,itishowevernoticeablethatthatACTHandprolactinfollowaquitesimilarprofileovertime:the concentrations started at high initial values at the start of the exposure and thendecreased in the following 3 weeks, and they then rose slightly again. The growthhormone concentrations are very high for the firstmeasurementsduring the exposureperiod, they then fall to thepreexposureconcentration levelsandmaintain these levelsuntil theendof theexperiment.Possibly, thesemeasurementsshowa temporarystressreaction,whichreducedinthefollowingweeks.

5.3.2 Melatonin

Thehormonemelatonin,whichisproducedinthepinealgland,functionsasaregulatinghormonal signal that synchronizes theendocrine rhythmsofall thehormoneglands. Itregulates, amongst other things, thedaily cycles ofACTH and the cortisolrelease andtherebyregulatesthedailyrhythmsofmanymetabolicprocesses.

Melatonin also exerts influences (inhibitory) on sexhormones and ithas a stimulatoryeffecton the immune system.Melatoninalso influences specific cancer illnessesvia theregulation of the release of the sex hormones. In addition,melatonin is a free radicalscavenger, inactivating radicals such as OH, which amongst other things can bedangerous for the genetic material. Furthermore, during in vivo experiments, it wasdemonstratedthatmelatoninhinderschangesinDNAproducedbychemicalcarcinogens


anditprotectslymphocytesfromchromosomedamageinhighfrequencyelectromagneticfields(*Lai&Singh1997).

In thepreviouslydescribedexperiments carriedoutby Imaidaetal.(*1998a,b), itwasfound that the experimental animals that were exposed to a pulsemodulated highfrequencyfieldhadareducedmelatoninconcentrations intheblood.Thisfindingcouldnotbe confirmedbyHeikkinenetal. (1999),whoexposedmice for17months toa900MHz fieldwith a 217HzGSM pulsemodulation (SAR: 0.35 to 1.5W/kg). Studies byVollrathetal. (1997)usingratsandhamsterswitha900MHz field (217HzGSM,SAR:0.04to0.36W/kg)couldnotcontributemuchtotheclarificationoftheproblem,sinceinseveral subsets of the experiment statistically significantdifferences between exposedanimalsand nonexposedanimalshadbeen found,butaccording to theauthors theseresultedfrommistakesintheexperimentalorder.

InexperimentsbyMannetal.(*1997seeabove), thestresshormonesweremeasuredaswellastheserummelatoninprofile.Thisshowed,inthecaseoftheexposedhumans,thatfor aperiod ofbetween 3 to 4hours in themiddle of thenight therewas an increasecomparedtothecontrolvalues,butthesewerenotstatisticallysignificantaccordingtotheevaluationoftheauthors.


6 Pathological Effects

6.1 Results of Experimental Studies

6.1.1 Cancer

Carcinogenesis

Carcinogenesisisamultilayeredprocess,atthebeginningofwhichisacertainimpactonthe level of the genetic material. This can be a direct impact (for example ionisingradiation)oranindirectactionviatheproductofareaction(forexampleOHradicals).AdirectorindirectinteractionwithDNAcanleadtodamageoftheDNAorthechromatinstructures (see also Chapter 3). If those damages are not repaired by endogenousprocesses, the damage will be permanent. Thus, the initiated cell can, if theimmunologicalcontrolfails,undertheinfluenceofhormonesandpromotersdevelopintoapreneoplasticfocus,whichcan then lead toamalignant tumor.Thedifferentstepsofcarcinogenesisaresummarisedinthreephases:

Initiation:TriggeringofdamageontheDNAandmutationsoncriticalgenes

Promotion:IncreasedrateofDNAsynthesisandproliferationoftransformedcells

Progression:Transitionofapreneoplasticfocustoamalignanttumor

A physical or chemical pollutant can in principle be effective in all three phases ofcarcinogenesis.

Initiation: Triggering of directDNA damage or of a substancewhich causesDNAdamage,disruptionofrepairprocessesoftheDNA

Promotion:Promotionoftheproliferationoftransformedcells

Progression:Suppressionofimmunereactionsandpromotionoftumorgrowth

ResultsfromAnimalExperiments

Invivoexperimentsusinganimalswithan inbredgeneticpredispositionforcertaintumorillnessesorinwhichanimalswereinjectedwithcancercells,yieldedverydifferentresults(seeAppendixC,TableC.1). In themajorityof thestudies,nocancerpromotingeffectofhighfrequencyelectromagneticfieldscouldbefound,oreffectswereonlyobservedundercertainconditionsofexposure(markedintheTablewithpartly),andeveninthosecasestheywereoftennotstatisticallysignificant.However,itneedstobenotedthatmanystudieswithnegativeresultshadveryshortexposure timesanddurationsof thestudy itself (forexampleChagnaudetal.1999:2weeks,Salfordetal.1993:2to3weeks)andhencetheydonothavemuchrelevance toanswer thequestionwhetherhigh frequencyelectromagneticfieldshavecarcinogenicpotential.

Some longterm studies have yielded results which indicate a carcinogenic or cocarcinogeniceffectofelectromagneticfieldswithmobiletelecommunicationsfrequencies


iftheanimalsareexposedoveralongperiodoftime.(*Repacholietal.1997,*Szmigielskietal.1982and*Szudinskietal.1983).ImportantinthiscontextisalsothestudyofChouetal.(*1992).Thisstudydidnotfindastatisticallysignificantriseintumorsinaparticularorgan.However, theexposedgroupdevelopednotonlyahighernumberof tumors intotal,butalsothenumberofprimarymalignantandmetastaticmalignantneoplasmswassignificantlyhigherintheexposedanimals.Intheirdiscussionoftheresults,theauthorspoint to the fact that the number of the primarymalignant neoplasms in the exposedgroup compared to the control group is four times higher and that this finding isstatisticallysignificant,but thengoon toundermine their findingbyquoting literature,according towhich the tumor incidenceof theexposedgroupshouldstillbewithin thenormalrange.

TheexperimentofToleretal.(*1997)usinganimalswithapredispositionforchesttumorsdid not result in a higher incidence of these, but the number of ovarian tumorswassignificantlyhigherintheexposedgroupcomparedtothecontrols.

The intensities atwhich an increase in tumorswas found in animalswere one to twopowers of ten below the values atwhich onewould expect a triggering of thermaleffects.Accordingtothepresentingresults, lowfrequencymodulationdoesnotseemtoberesponsibleforthecarcinogeniceffect.

6.1.2 Infertility and Teratogenic Effects

Teratogenesis

Teratogeniceffectsofapollutantcanaswiththecarcinogeniceffecteitherbecausedbythe triggering of a genetic defect or a harmful impact on the foetal development. Theformation of a geneticmalformation during its initiation phase is analogous to carcinogenesis,i.e.teratogeniceffectsarealsocausedbydirectorindirectimpactontheDNAanddisruptionsoftheendogenousrepairmechanisms.Laterdamagesofthefoetuscaneitherbe causedbydirecteffectsof thepollutanton the foetusorby reactions to thepollutantwithinthemothersorganism,whichwouldthenbepassedontothefoetus.

ResultsfromAnimalExperiments

Amultitudeofstudieshavedemonstratedthathighbodytemperaturesinmammalsleadtoaspermatotoxicandteratogeniceffect.Sincemanystudiesexaminingsucheffectsfromhigh frequency electromagnetic fields worked with intensities that were capable ofsignificantly raising body temperature, it cannot be excluded that the observedspermatotoxicand teratogeniceffectswerecausedbya thermaleffect, (see forexampleBerman et al. 1982, 1983,Berman&Carter 1984, Jensh et al. 1983a,b,Kowalczuk et al.1983,Laryetal.1983,Nawratetal.1985,Saundersetal.1981,1983,fortheresultsofolderstudies,seeOConnor1980).Theresultsofthesestudiesdonotalwaysappearconsistent,however, this can possibly be explained by a different thermal susceptibility of thedifferentanimalspeciesused.Inratsforexample,alossofthermallydamagedembryosisoftenobserved,whilst thebirthofmalformed animals is rare.Othermammals show awiderbandwidthbetweenteratogenicandlethalexposures.(Verschaeve&Maes1998).


However,therearesomeindicationsintheliteratureforteratogeniceffectsatintensitiesthatcauseno(or,ifatallverysmall)risesintemperature.Magras&Xenos(1997)exposedmiceduringsixmonthstoarealtransmitter.Themicehadoffspringfivetimesduringthisperiodandacontinuousdecreaseinoffspringwasfounddowntoirreversibleinfertility.TheexposureconsistedofseveralradioandTVtransmittersintheVHFandUHFbandsandmeasuredbetween0.00168and0.01053W/m2.A repetitionof this studywouldbedesirableinordertoexcludethattheeffectwasduetoproblemswiththemaintenanceoftheanimalsorthescreeningofthecontrolgroup.

KhillareandBehari(*1998)foundthatmaleratsthathadbeenexposedtoa200MHzfield(power fluxdensity:14.7W/m2,SAR:1.65 to2.0W/kg)duringaperiodof35days for sixdaysperweekandtwohoursperexposuredayandwhichwereafterwardsmatedwithunexposedfemales,producedsignificantlylessoffspringthatthemalesintheunexposedcontrolgroup.

InanexperimentbyAkdagetal.(1999)maleratswereexposedonehoureverydaytoa9.45GHzfield(powerfluxdensity:2.5W/m2,SAR:1.8W/kg)duringdifferentperiodsof13,26, 39 or 52 days corresponding to one, two, three and four cycles of the seminalepithelium.

Attheendofeachexposureperiodthefollowingdataweremeasuredandcomparedtoanunexposedcontrolgroup:numberofsperm in theepididymides,morphologyof thespermandweightofthetesticles,epididymides,seminalvesiclesandprostate.

They found amongst other effects a decrease in the number of sperm (statisticallysignificant in the group exposed for 53 days) and an increase of abnormal sperm(statisticallysignificantinthegroupsexposedfor26,39and52days).

Acoteratogeniceffectundernonthermalexposureswithpower fluxdensitiesof10 to100 W/m2 in combination with cytosine arabinoside (CA) was found in a study byMarcickiewiczetal.(*1986).Intheexperiment,micewereexposedinuterofortwohoursadayto2.45GHzfromthefirsttothe18thdayofthepregnancy.Thefield,whichalonewas not teratogenic, significantly increased the teratogenic effect of CA. A directteratogeniceffectofmicrowaveradiationwitha frequencyof2.45GHzon thebrainsofnewbornratswasfoundbyInalszetal.(*1997).HowevertheauthorsdeclaredthattheSARof2.3W/kgledtoariseofrectaltemperatureof1.0C.

6.2 Results of Epidemiological Studies

MethodologicalRequirements

Inprinciple,epidemiologicalstudiesareaneffectiveinstrumenttoprovepotentialhealthrisksofapollutantunderrealenvironmentalandexposureconditions.Usually,theyarecarriedoutbycomparingstatisticaldataabout the incidenceofan illness inanexposedpopulationasopposed to the incidenceof this illness inanunexposedpopulation.Theexactclassificationofexposurewouldrequirethemetrologicalrecordingofthepollutantfor all participants (exposed and unexposed) during the entire latency period of theillness.Thisisoftennotpracticableandforlonglatencyperiods,whichcanusuallyonlybeaddressedviaretrospectivestudies,inherentlyimpossible.Undersuchcircumstancesit


hastosufficethatsurrogatesareused,forexamplehavingaprofessionwhichislinkedtoacertainexposureortheproximityofthehometoanemittinginstallation.Insomecases,iftheemittinginstallationshavebeenusedforalongtimeinthesamemode,itispossibletoextrapolatepastexposuresfromcurrentmeasurements.

The quality of the exposure classification determines the validity of an epidemiologicalstudy.Possibleweaknesses,whichcanleadtowrongresults,are:

Peopleareclassifiedas exposedor stronglyexposedalthough infactthere isnooronlylittleexposure.Anexamplewithregardstohighfrequencyfieldsistheoftenusedexposureclassificationonthebasisofprofessionalcategories,suchasradaroperatorsor telecommunications engineers, for whom it cannot be excluded that the mainoccupationisadeskjobwithoutexposure.

Itisassumedthatthecontrolgroupiscompletelyunexposed,althoughthepollutantisactually ubiquitous, which will lead to smaller but still potentially significantexposures in the control group.One known example aremains frequencymagneticfields,which affect the immediate neighbours ofpower supply equipment, but stillexist at nonnegligible strengths in houses which are further away from suchequipment.

Botheffectsleadtoalevellingoutbetweentheexposedandunexposedgroupandhencetoanunderestimationoftherealhealthriskposedbythepollutantinquestion.

Another weakness of epidemiological studies can be the presence of unrecognizedconfounders,i.e.otherinfluences,whichalsoaffectthegroupsstudiedandinfluencethedevelopmentoftheillness.Thiscanbeenvironmentalfactors,suchasexposurestootherpollutants,butalsosocioeconomicandbehaviouralfactors.Ifnotallpotentiallyrelevantconfoundersarefactoredin,theresultscanbedistorted,eithertowardsanoverestimationoranunderestimationoftherealrisk.

Thefastdevelopmentofmobiletechnologyhasleadtoadoubledilemmawithregardstothestudyofpotentialrisksthroughepidemiologicalstudies:

Forillnesseslikecancerwithlatencyperiodsofmanyyearsitisstilltooearlytoexpectvalidresults.Ifmobiletelecommunicationsareindeedlinkedtoahigherincidenceofcancer,theillnesswillonlyhavemanifestedinafewpeoplesofar.Thisshouldatleastbevalid for thepartof thepopulationwhoseexposuresare frombasestationsonly.Potentiallyitcouldbedifferentfordirectmobilephoneusers,sincethesearegenerallyexposed to significantlyhigher intensities.Butalso for thisgroup,at thismoment intime,wewouldexpect results fromepidemiological studies tounderestimate the realrisk.

In some years epidemiological studieswillhit adifferent obstacle: oncebase stationscovertheentirecountryandalargeproportionofthepopulationuseamobilephone,itwillbecomedifficulttofindthenecessaryunexposedcontrolgroups.

Given this dilemma, epidemiological studies carried out in the past have a certainvalidity,even if theexposuresarenotexactly thesameas theywouldbe todayand thestudiesdonotalwayscorrespondtotodaysqualitystandards.


TheSelectionofStudies

Atthetimeoffinishingthispresentreporttherewereonlytwoepidemiologicalstudiesofhealthrisksinrelationtoactualexistingmobiletelecommunicationsexposures(*Rothmanat al. 1996, *Hardell et al. 1999).However there are amuch larger number of studiesavailable,inwhichthehealtheffectsofhighfrequencyelectromagneticfieldsinhumanswereexamined(seealsoAppendixD,TableD.1).Justunderaquarterofallresultsrelateto exposureswith low frequencypulse or amplitudemodulatedhigh frequency fields,suchastheyareusedformobiletelecommunications,evenifthecarrierandmodulationfrequenciesareinmostcasesnotidenticalwiththoseofmobiletelecommunications.

InAppendixTableD.1, theexamined illnessesare listedwith theirevaluatedendpoint(incidenceormortality),datadescribingtheexposuresituationisgivenandthequalityofthe exposure classification is assessed. Finally, the result of the study is evaluated asRelativeRisk (RR)which includes therelevantrisk factors in the formofstandardisedmortality rates, standardised morbidity rates and odds ratios, and the statisticalsignificanceisassessed.Foreachstudywelistthevalueforthehighestexposureclassorif therewasa furtherdifferentiationof theexaminedgroups, forexampleaccording tooccupationalgroups,thehighestfoundvalue.

Valuesareconsideredstatisticallysignificant (s.s.) if thevalueRR=1outsideof the95%confidenceintervalorifp1andhowmanyarestatisticallysignificant.

Almostallthestudies,inwhichthetotalcancerriskwithoutanydifferentiationaccordingtotumorformwereexamined,showedariskfactorofRR>1.Halfofthestudiesresultedinstatisticallysignificantriskfactorswithamaximumvalueof2.1,whichcorrespondstoa doubling of the statistical risk to develop cancer from exposure to high frequencyelectromagneticfields.

Asimilarpicturewasfoundinrelationtotumorsofthenervoussystem,especiallybraintumors.Here,themaximumvalueforrelativeriskfoundwas3.4.Elevenofthetotalof15studiesyieldedapositiveresult,morethanhalfofwhichwerestatisticallysignificant.

The incidence of breast cancer in relation to high frequency fieldsmust be examinedseparatelyformenandwomen.Allthreestudiesrelatingtothebreastcancerincidenceinwomenyieldedriskfactorsgreaterthan1,thestatisticallysignificantvalueswere1.15and1.5. Formen, risk factors of up to 2.9were found; however, not allwere statisticallysignificant.

Of the totalof16 results for leukaemiawithout furtherdifferentiationof the illness,13were positive (RR>1),more than half of these resultswere statistically significant.Thehigheststatisticallysignificantvaluefortherelativeriskwas2.85.Amongsttheresultsofthe differentiated studies, the following are notable: lymphatic leukaemia (7 results, 5positive,4statisticallysignificant,RRmaximumvalue:2.74)andacutemyeloicleukaemia(4differentstudies,3positiveresults,2statisticallysignificant,maximumRRvalue:2.89).


Withregards to thecorrelationofhigh frequencyelectromagnetic fields fromradarandother sources and testicular cancer, three studies have been conducted. All lead tostatisticallysignificantriskfactorswithamaximumvalueof6.9.

Thestudiesregardingcardiovasculardiseasesdidnotresult inaclearpicture,not leastbecauseofthemultitudeofthesymptomsexamined.

All four studiesof fertilityproblems in relation to the exposureofmen tomicrowavesindicateincreasedrisk.Intwostudiesstatisticallysignificantriskfactorsofupto2.7werefound.

With regards to irregular courses of pregnancies and malformations in children ofmotherswhichhadbeenexposed tohigh frequency fields, therearea largenumberofstudieswithpositive results,ofwhichonly two fit into the frequency range relevant toour report.Bothof these studies found statistically significantpositive resultswith riskfactorsofupto2.36.

Of the studies of cancer risk of childrenwhose fathers had been exposed to electromagneticfields,onlytwocorrespondtothequalitycriteriarequiredforinclusionintothisreport.Both indicatean increased risk,butonlyone result is statistically significant at avalueofRR=2.3.(Withregardstothecancerriskofchildrenincorrelationtotheexposureoftheirparents,seealsoColt&Blair1998).

Regardingthedisruptionofmotorfunctionsaswellaspsychologicalfunctionsandwellbeing,thereisonlyonevalidstudyforthefrequencybandsrelevanttothisreport,whichyieldedaslightly increasedriskfactor.Howeversinceotherstudiesoftransmitterswithfrequenciesbelow100MHz resulted in serious indicationsof increased risk, indicatingthatthisproblemshouldbegivenmoreattentioninthefuture,wealsoincludedthestudyofZhaoetal. (1994),although itdidntmeetourquality standardswith regards to thestatisticalevaluation.

Unfortunately, the majority of the studies do not state the actual strength of theexposures.MeasurementsareonlyavailablefortheradioandtelevisiontransmitterusedforthestudiesofHockingetal.(1996)andMcKenzieetal.(1998).Themeanpowerfluxdensitiesforall16municipalitiesaffectedbythistransmitterwere3.3103W/m2withintherangefrom2.6104to1.46102W/m2(McKenzieetal.1998).TheICNIRPguidelinesforthegeneral population recommend amaximum value of 2 to 2.51W/m2 for the range offrequencies emitted by this transmitter (64.25 to 527.25MHz). This means that theexposuresinthesestudieswerebelowtheGermanguidelinesbyafactorof104.


Table 6.1 Overview over the results of epidemiological studies with regards to the health risks of high frequency electromagnetic exposures (see also Appendix D, Table D.1)

Illness Number of studies (results)

Studies (results) with RR>1

Statistically significant

results

All illnesses 2 0 0

Cancer, unspecified 6 (7) 5 (6) 3

Brain tumours unspecified and tumours of the nervous system unspecified

14 (21) 10 (15) 6 (7)

Cancer (eyes) 1 1 1

Cancer of the respiratory organs, lung cancer 5 2 1

Chest cancer, men 2 2 0

Breast Cancer, women 3 3 2

Cancer of the lymphatic and blood forming system unspecified

4 4 1

Leukaemia unspecified 12 (16) 9 (13) 5 (7)

Acute leukaemia unspecified 4 4 0

Lymphatic leukaemia unspecified 4 (7) 2 (5) 1 (4)

Acute lymphatic leukaemia 2 2 0

Chronic lymphatic leukaemia 4 4 1

Leukaemia, non lymph. non-myelo 1 (4) 1 (4) 1 (2)

Lymphoma, Hodgkin-Syndrome 5 (7) 3 (4) 1

Testicular cancer 3 (5) 3 (5) 3 (4)

Uterine cancer 1 1 1

Skin cancer 4 3 1

Cardio-vascular diseases 4 (5) 3 (4) 1

Infertility, reduced fertility, men 4 (7) 4 (7) 2 (4)

Infertility, reduced fertility, women 1 1 0

Miscarriages, stillbirths, malformations and other birth defects

2 (3) 2 (3) 2

Cancer, offspring (parental exposure) 2 2 1

Neurodegenerative diseases, Alzheimers 1 1 0

Disruptions of motor and psychological functions and well-being

2 (9) 2 (9) 1 (7)


7 Health Risks to Humans Resulting from Exposure to the Electromagnetic Fields of Mobile Telecommunications

Thetriggeringofanillnesscausedbyan(environmental)pollutantandthedevelopmentofthisillnessareamultiphasedprocess,whichbeginswithabiological,biochemicalorbiophysicalprimaryinteractionofthepollutantwiththebiologicalsystemandendswiththemanifestationoftheillness.Duringthedifferentphasesoftheprocess,thebodysownrepairmechanismscan interveneand impede the furtherdevelopmentof the illness.Anassessment of the potential health risks of electromagnetic fields as they are used formobile telecommunications should therefore be mainly based on studies conducteddirectlyonhumans,becauseextrapolationsfromanimalstudiesoreveninvitrostudiesoncell cultures only have limited validity for effects in humans, due to the difference insusceptibilitiesand the lackoforganic interactions incellcultures.However,due to theethicallimitstotheresearchonhumans,itisunavoidabletouseresultsfromexperimentswithanimals,singleorgansorcells inordertodiscoverthebiologicalandphysiologicalmechanisms.

Cancer

Given the results of the present epidemiological studies, it can be concluded thatelectromagneticfieldswithfrequenciesinthemobiletelecommunicationsrangedoplayarole inthedevelopmentofcancer.This isparticularlynotablefortumoursofthecentralnervoussystem,forwhichthereisonlytheoneepidemiologicalstudysofar,examiningtheactualuseofmobilephones.Themost striking resultof this studywasanobviouscorrelation between the side atwhich the phonewas used and the side atwhich thetumour occurred.The brain tumour incidence howeverwas only slightly increased.A(hypothetical)explanationofsuchafindingcouldforexamplebethatmobilefieldshavea promoting effect on previously initiated (multiple) tumours, triggering a defencemechanisminthebodywhichiscapableofsuppressingunpromotedtumours.

Higherriskswerealsodemonstratedforseveralformsofleukaemia.

Althoughthestudies inrelationtotesticularcancerwereexaminingparticularexposureconditions(emittingequipmentwornpartlyonthebodyathiplevel),giventhehighriskfactorfound,apossibleriskcannotbeexcluded,especiallynotformobileuserswearingthedevices in standbymode on theirbelts.The epidemiological findings for testicularcanceralsoneedtobeinterpretedinconjunctionwiththeresultsofthestudiesoffertilityproblemsoccurringinrelationtohighfrequencyelectromagneticfields.

Theriskfactorsforcancersotherthantesticularcancerareonlymoderatelyincreased,butnot negligible, considering this technology will potentially reach full coverage of theentirepopulation.


Reliableconclusionsaboutapossibledoseresponserelationshipcannotbemadeon thebasis of the present results of epidemiological studies, but an increase of cancer riskcannotbeexcludedevenatpowerfluxdensitiesaslowas0.1W/m2.

In longterm animal experiments, the carcinogenic effect of pulse modulated highfrequency fields was demonstrated for power flux densities of circa 3W/m2 (mouse,exposureduration18months,30minutesperday,SAR(mouse)circa0.01W/kg).

On the cellular level, a multitude of studies found the type of damage from highfrequency electromagnetic fields which is important for cancer initiation and cancerpromotion:

Direct damage on DNA as well as influences on DNA synthesis and DNA repairmechanismsweredemonstrated in in vivo and in vitro experiments for continuous andpulsedfieldsatpowerfluxdensitiesfrom10W/m2and9W/m2respectively.

Chromosomeaberrationsandmicronucleioccurredatpowerfluxdensitiesfrom5W/m2.

NeoplasticcelltransformationandanenhancedcellproliferationweredemonstratedforSpecificAbsorptionRatesofbelow0.5W/kg,andindividualstudiesdemonstratedthattheobviousdisturbanceof thecommunicationbetweencells,which isaprerequisitefor theuninhibitedproliferationofcells that ischaracteristic forcancerdevelopment,occursatjustafewW/m2.

Conclusion:

The resultsof the studies forall stagesof cancerdevelopment from thedamageof thegeneticmaterialviatheuninhibitedproliferationofcellsanddebilitationofthe immunesystem (see below) up to themanifestation of the illness prove effects at power fluxdensitiesof less than1W/m2.For some stagesofcancerdevelopment, intensitiesof0.1W/m2orevenlessmaysufficetotriggereffects.

DebilitationoftheImmuneSystem

Damagingeffectsontheimmunesystemwhichcanaidthedevelopmentofillnessesweredemonstratedinanimalexperimentsatpowerfluxdensitiesof1W/m2(mouse,exposureduration 6 days, 3 hours per day, SAR (mouse) 0.14W/kg). In in vitro experiments onlymphocytes,defectsofthegeneticmaterialweredemonstratedatpowerfluxdensitiesofcirca10W/m2.Thepresenceofstresshormones,whichwhenpermanentcandebilitatetheimmune system, was found to be increased in human experiments from power fluxdensities of 0.2W/m2. In animal experiments (rat) a similar effect was observed at aSpecificAbsorptionRateofcirca0.2W/kg.

Conclusion:

Experimentsonanimalsproveharmfuleffectsontheimmunesystemfromcirca1W/m2;atpowerfluxdensitiesof0.2W/m2highersecretionsofstresshormonesinhumanshavebeendemo

Date post:	10-Jan-2016
Category:	Documents
Upload:	hamdo2
View:	16 times
Download:	1 times

TMobile Study

Documents