Exposure of Kenyan population to aflatoxins in foods with special reference to Nandi and Makueni counties

© The Au ho 2017. Published by Ox o d Uni e si y P ess on behal o Zhejiang Uni e si y P ess. 131 Food Quali y and Sa e y, 2017, 1, 131–137 doi:10.1093/ qs/ yx011 Re iew Ad ance Access publica ion 26 May 2017 Re iew Exposu e o Kenyan popula ion o a la oxins in oods wi h special e e ence o Nandi and Makueni coun ies E. K.Kang’e he*, M.Ga wi i*,**, A. J.Si ma*,***, E. O.Ouko*,***, C. K.Mbu ugu-Muso i*,**, P. M.Ki ala*, G. J.Nduhiu*, J. G.Nde i u*, J. K.Munga u**, V.Hie aniemi****, V.Jou sjoki**** and H. J.Ko honen**** *Depa men o Public Heal h, Pha macology and Toxicology, Nai obi, Kenya, **Jomo Kenya a Uni e si y o Ag icul u e and Technology, Nai obi, Kenya, ***Minis y o Ag icul u e Li es ock and Fishe ies, Ve e ina y Labo a o ies Kabe e, P i a e Bag Kabe e, Nai obi, Kenya and **** Na u al Resou ces Ins i u e Finland, LUKE, Jokioinen, Finland Co espondence o: E. K. Kang’e he, Depa men o Public Heal h, Pha macology and Toxicology, Box 29053– 00625 Nai obi, Kenya, E-mail: [email p o ec ed] Recei ed 19 Feb ua y 2017; Re ised 8 Ma ch 2017; Edi o ial Decision 9 Ma ch 2017. Abs ac A la oxins cause acu e and ch onic heal h and p oduc ion e ec s in humans and animals, espec i ely. This s udy epo s on he exposu e o he Kenyan popula ion o he wo myco oxins in wo coun ies (Nandi and Makueni) using child en younge han 5yea s as he p oxy. Aques ionnai e was adminis e ed, which cap u ed knowledge o he incidences o acu e a la oxin poisoning, ood consump ion pa e ns. Analysis o a la oxin in samples o u ine, b eas milk, maize, so ghum, and mille was done using cELISA and HPLC. Maize and so ghum a e used in he weaning o mula o child en. The di e ence in consump ion o maize- and so ghum-based die s in child en be ween younge han 1 and be ween 1 and 5yea s was signi ican wi h P=0.037 and P=0.002, espec i ely, in Nandi and Makueni. In child en younge han 5yea s, he consump ion o maize anged om 0.1 o 0.25kg pe pe son pe day in Nandi and Makueni wi h an a la oxin exposu e a e o 0.011 and 0.49µg pe kg body weigh (bw ) pe day, espec i ely. The exposu e o a la oxin h ough milk o child en younge han 5yea s was 4×10−4 and 1×10−4 µg pe kg bw pe day in Makueni and Nandi, espec i ely. The exposu e o nu sing child en h ough b eas milk was 6×10−3 and 1×10−6 µg pe kg bw pe day in Makueni and Nandi, espec i ely. Child en younge han 30mon hs in Makueni had 1.4 imes highe le els o a la oxin M1 (AFM1) in u ine han hose o he same age in Nandi. The s un ing and se e e s un ing a es in Makueni and Nandi we e 28.7%, 18.5% and 30.7%, 16.5%, espec i ely. Thus, he e is need o u gen mi iga ion measu es, a cons an su eillance p og am, s ic egula ions, and awa eness c ea ion di ec ed a poo households o educe die a y exposu e o myco oxins. Key wo ds: A la oxins; Con amina ion; Exposu e; Kenya. In oduc ion A la oxins equen ly con amina e s aple oods o many Kenyans. A la oxins a e p ima ily p oduced by Aspe gillus la us and A. pa asi icus. Humans a e exposed o a la oxins om consump- ion o ce eals, nu s, pulses, oo s, and animal sou ce oods om li es ock ed con amina ed eeds. In Kenya, maize is he majo s aple This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion Non-Comme cial License (h p://c ea i ecommons.o g/licenses/by-nc/4.0/), which pe mi s non-comme cial e-use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. Fo comme cial e-use, please con ac [email p o ec ed] 132 E.K. Kang’e he e al., 2017, Vol. 1, No. 2 ood and i is es ima ed ha annual pe capi a consump ion s ands a 97kg (FAO, 2000) compa ed o 76kg pe capi a consump ion in Eas and Sou he n A ica, espec i ely (M’mboyi e al., 2010). Maize con amina ion wi h a la oxins in Kenya has been epo ed by Lewis e al. (2005), Mu u e and Ogana (2005), S osnide e al. (2006), P obs e al. (2010), Daniel e al. (2011), and Mu homi e al. (2012). Human exposu e om milk ob ained om animals ed wi h con- amina ed eeds has been epo ed by Kang’e he e al. (2007) and Kang’e he and Lang’a (2009). Long- e m exposu e (2–3mon hs p io o sampling) o a la ox- ins in humans can be assessed by analysis o albumin adduc s (Wild and Tu ne 2002; Gong e al., 2002, 2004) and p esence o a la oxin M1 (AFM1) p oduc s o AFB1 b eakdown in u ine o sho - e m exposu e. This s udy aimed a es ablishing he exposu e le els o house- holds o a la oxins and e ec s o a la oxin exposu e exempli ied by g ow h indices in child en younge han 5yea s in he Nandi and Makueni coun ies o Kenya. Ma e ials and Me hods Si e and household selec ion The coun ies Nandi and Makueni in Kenya we e pu posely selec ed based on his o y o human acu e a la oxicosis in Makueni (Lewis e  al., 2005) and high incidences o esophageal cance in Nandi (Wakhisi e al., 2005). The sub-loca ions we e iden i ied by a eam o esea che s oge he wi h e e ina ians, ag icul u e ex ension o ice s, and heal h o ice s a he coun y and wa d o ices. The selec ion c i e ion was based on dai y and maize p oduc ion. In he i s ins ance, he sub-coun ies we e selec ed ha i ed he c i- e ia based on he eco ds a he coun y le el. Th ee sub-coun ies we e selec ed in each coun y. A he sub-coun y headqua e s, he wa ds we e selec ed using he same c i e ia o dai y p oduc ion and maize g owing. A he wa d le el, he ex ension o ice s ( e e ina - ians, ag icul u e, and public heal h) selec ed he sub-loca ion ha bes i ed he c i e ia used o selec he wa ds. The households in he sub-loca ion we e lis ed which i ed he c i e ia o ha ing dai y animals, g ew maize and o so ghum and mille , had a child younge han 5yea s, and/o i he spouse was p egnan a he ime o lis ing. Based on he popula ion and numbe o households in he sub- loca ion and he households ha i ed he c i e ia, a p opo ion o he households was andomly selec ed o sampling (Ma in e al., 1987), and he household sample size was co ec ed o ini e popu- la ion acco ding o Daniel (1999). Nandi alls wi hin he ag o-ecological zones o lowe humid highlands o uppe midland and uppe highland zones ((LH2, LH3, UM3 and UM4, Jae zold e al. 2006). I lies be ween 1300 and 2500 m abo e sea le el and ecei es abou 1200–2000mm o ain pe yea . Nandi is bes sui ed o ea cul i a ion, dai y p oduc ion, and g owing maize, whea , and ba ley. Makueni alls wi hin he lowe midland ag o-ecological zones LM3, LM4, and LM5. I ecei es be ween 200 and 1200mm o ain all pe yea , which is un eliable a imes wi h equen d ough s esul ing in c op ailu es. Sampling o u ine U ine samples we e collec ed om child en younge han 5yea s in all he households. Clean bo les we e supplied o he mo he s o collec ing he u ine. The samples we e ozen a ದ20°C un il analysis wi hin 3mon hs. Sampling o b eas milk Nu sing mo he s we e gi en clean bo les in o which hey exp essed milk om hei b eas s. The samples we e ozen a ದ20°C un il hey we e analysed wi hin 3mon hs. Tes ing o a la oxin M1 in u ine using c-ELISA (helica es ki ) To de e mine a la oxin M1 in u ine, 5ml o u ine was cen i uged a 743× g o 10min. Nine hund ed and i y mic oli es o dis illed wa e was pipe ed in o ska on (SKATRON AS LIER, No way. CAT. No 7071) ubes and 50µl o s anda ds o supe na an -u ine was added in o 950µl o dis illed wa e in he ska on ubes and mixed by p iming pipe ing a leas i e imes. Two hund ed mic o- li es o he assay bu e was added in o he mixing well pe pla e and 100µl o he dilu ed s anda ds ( anging om 0 o 40 pp ) and u ine samples was added in o wells o he mixing well pe pla e. The con en s o he mixing well we e shaken using a mic o-shake (DYNATECH) o 2min. One hund ed mic oli es o he mix u e was ans e ed o he an ibody-coa ed Reac ion-Assay Pla e (a la- oxin M1 assay o u ine, Helica Biosys ems Inc, 1527 W. Al on San a Ana, Cali o nia, USA). The samples we e mixed by shaking o 1min (DYNATECH Shake ) and incuba ed a oom empe a u e in he da k o 1h. The pla e was washed h ee imes using phos- pha e-bu e ed saline–Tween-20 (0.05%) bu e using he well-wash (The mo Scien i ic, Finland) machine wi h 3-min in e als be ween he washes. A e d ying, 100µl o conjuga e was added in o each well, mixed gen ly by apping, and incuba ed a oom empe a u e o 15min in he da k. The pla e was hen washed and 100µl o sub- s a e eagen ( e ame hyle hidine) was added in o each well, mixed gen ly by apping, and incuba ed a oom empe a u e o 15min in he da k. The eac ion was s opped by adding 100µl pe well o s op solu ion and he op ical densi y (OD) ead a 450nm wi hin 15min o s opping he eac ion. The le el in each sample was de e mined using he p og am om he ki manu ac u e , which allowed calcula- ions o he le els based on he abso bance eadings. Tes ing o a la oxin M1 in b eas milk using cELISA (Ridasc een es ki ) The analysis o milk o AFM1 was ca ied ou using he com- pe i i e ELISA me hod acco ding o manu ac u e ’s ins uc ions ( -biopha m, Ge many) wi h modi ica ion adop ed om EVIRA (Finnish Food Sa e y Au ho i y). Analysis was done wi hin 3mon hs o sample collec ion. Fi e millil es o milk sample was wa med and cen i uged o 15min a 1011× g.The uppe c eam laye was emo ed and 2.5ml o he de a ed milk ans e ed o a es ube and 5ml o e hyl ace a e (88.1 g/mol; mel ing poin ದ83.6°C) added. This was o exed o 1min and he mix u e cen- i uged a 1011× g o 15min a oom empe a u e. Th ee mil- lil es o he e hyl ace a e laye we e ans e ed in o a clean es ube and e apo a ed o d yness using a s eam o ni ogen. The sample was dilu ed wi h 250μl o he sample dilu ion bu e and 30μl o 70% me hanol:wa e , o exed, and analysed o AFM1 using a compe i i e ELISA acco ding o manu ac u e ’s p ocedu e. The limi o de ec ion was 5ng/l. Le els below his limi we e es i- ma ed using a company p og am ha de e mined quan i ies by ex apola ion. An h opome ic measu emen s Heigh , weigh , and age we e he an h opome ic measu es aken om he child en en olled in he s udy. The heigh was measu ed Exposu e o Kenyan popula ion o a la oxins in oods, 2017, Vol. 1, No. 2 133 using a measu ing ape o he nea es whole cen ime e while he weigh was measu ed o he nea es g am using a Sal e weighing scale. Weigh o age Z-sco e (WAZ), heigh o age Z-sco e (HAZ), and weigh o heigh Z-sco es (WHZ) we e calcula ed acco ding o he median alue o he in e na ional e e ence popula ion ecom- mended by he Na ional Cen e o Heal h S a is ics (NCHS) and Wo ld Heal h O ganiza ion (WHO, 1986). Household ood sou ces and consump ion A ques ionnai e was adminis e ed o he esponden s o 547 house- holds in o al. The esponden s, mainly spouses, we e asked o indica e o each household membe he ype o ood ea en (maize, so ghum, and mille ); sou ce o he ood ype whe he i was home- g own, ma ke sou ced, gi , o elie ; amoun s ea en and equency pe day. The amoun ea en was indica ed by he wi e who dished ou ood o e e y household membe om a empla e displaying a pho- og aph o he ood quan i ies; she would selec he amoun which closely es ima ed he amoun se ed. Da a analysis The da a we e en e ed in sp eadshee s and analysed in SPSS Ve sion 19.0. Explo a ion o he da a was by nume ical summa ies oge he wi h g aphics. Rela ions and associa ions in bo h he opinion/pe - cep ions and expe imen al da a we e u he examined using chi- squa e and - es o he le el o signi icance a 5%. Resul s Food consump ion Ce eals The la ges quan i y o maize, so ghum, and milk consumed by households was home p oduced. Maize is he main s aple ood in Nandi and Makueni. The majo i y o he households (60–90%) elied on homeg own maize excep in Makindu loca ion in Makueni and Kap umo loca ion in Nandi whe e homeg own maize con ib- u ed abou 40% o he household needs. Maize and so ghum was consumed as ugali o po idge and mu hokoi ( adi ionally dehulled maize, cooked oge he wi h beans o pigeon peas). Maize and so - ghum in he o m o uji, gi he i, and ugali we e used o weaning child en. Child en younge han 1yea and be ween 1 and 5yea s a e mo e so ghum uji in Makueni, compa ed o hose o he same age (Figu e1) in Nandi. The di e ence in consump ion be ween he wo age g oups was signi ican wi h P=0.037 and P=0.002 o younge han 1yea and 1–5 yea s, espec i ely. Makueni house- holds we e gi ing uji made om maize o olde child en (1–5yea s) while in Nandi his s a ed wi h child en younge han 1yea . Consump ion o maize in he o m o lou pe pe son pe day di e ed ac oss he si es and wi hin age g oups. Fo child en younge han 5yea s, consump ion anged om 0.1 o 0.25kg pe pe son pe day in Nandi and Makueni, espec i ely. The o e all a e age consump ion o maize in he o m o lou in Nandi and Makueni in e ms o kg pe pe son pe day was 0.18, 0.36 and 0.4 o peo- ple younge han 5, 6–60, and olde han 60yea s, espec i ely. The mos ulne able indi iduals o a la oxin poisoning (younge han 5yea s and olde han 60yea s) in Makueni we e consuming 0.22 and 0.41kg o lou pe day, espec i ely, bu his was no he case in Nandi whe e hose younge han 5 and olde han 60yea s con- sumed 0.13 and 0.37kg, espec i ely. Consump ion o milk Households in bo h Makueni and Nandi had an a e age o 5.9 pe - sons and consumed on a e age 0.38 o milk pe pe son pe day in he wo si es. The a e age consump ion in Makueni and Nandi was 0.32 and 0.44 o milk pe pe son pe day, espec i ely. The annual pe capi a consump ion was 117.9 and 161.9 in Makueni and Nandi, espec i ely. Six y- h ee pe cen o child en less han 1 yea old we e consuming milk as esh, 16% consumed milk mixed wi h po idge, while 3.4% d ank milk wi h ugali and 8.7% ook milk in he o m o e men ed milk, especially in Nandi. A la oxin M1 inu ine A o al o 377 and 362 u ine samples we e collec ed om chil- d en younge han 5yea s in Makueni and Nandi, espec i ely, and analysed using cELISA o AFM1 in u ine. Resul s showed ha 79% and 83% o he u ine samples we e posi i e in bo h Makueni and Nandi, espec i ely (Table1). The e was a signi ican di e - ence ( =12.64, P alue=0.00) in he mean o a la oxin M1 le - els in he u ine samples collec ed om he wo coun ies, hough he e was no signi ican di e ence in he alues be ween u ine samples om child en younge han 30mon hs and hose be ween 30 and 60 mon hs wi hin he coun ies. U ine samples (posi i e and nega i e) om Makueni had a highe mean le el o AFM1 (910.56 [721.31, 1099.78] pa s pe illion (pp )) han hose om Nandi (518. 60 [400.58, 636.62] pp ). The child en younge han Figu e1. Consump ion pa e ns o die s based on maize, so ghum, and mille by child en younge han 60mon hs in Makueni and Nandi coun ies. Ugali, hick po idge p epa ed om maize lou ; Uji, po idge p epa ed om lou (so ghum, mille , and maize singly o in combina ions). 134 E.K. Kang’e he e al., 2017, Vol. 1, No. 2 30 mon hs in Makueni had e y high le els o AFM1 in u ine; being 1.4 imes highe han ha ound in he u ine o child en o he same age in Nandi. Fo child en be ween 30 and 60mon hs in Makueni, he le els o AFM1 in u ine we e 2.3 imes highe han hose omNandi. A la oxin M1 in b eas milk A o al o 67 and 98 human milk samples we e collec ed om Nandi and Makueni, espec i ely, and 86.7% and 56.7% o hese samples we e posi i e o a la oxin M1 in Makueni and Nandi, espec i ely (Table2). The e was a signi ican di e ence ( =30.75, P<0.001) in he mean alues o a la oxin M1 in he b eas milk om Nandi and Makueni. The Makueni samples had a highe mean o 8.46 [8.24, 8.68] pp compa ed o 0.02 [0, 0.061] pp in Nandi. The mean le el in b eas milk samples om Makueni was 423 imes highe han he mean le el ound in he b eas milk samples omNandi. Exposu e h ough maize andmilk Table 3 shows he exposu e o a la oxins in Makueni and Nandi among he a ious age g oups. Makueni households had a highe exposu e a e han he households in Nandi. This is a e lec ion o highe a e age le els o a la oxins in oods om he wo si es. Exposu e a es o a la oxins h ough lou in child en younge han 5yea s in Makueni we e 44 imes mo e han in hei coun e pa s inNandi. G ow h indices S un ing S un ed g ow h is a educed g ow h a e in human de elopmen . I is a p ima y mani es a ion o malnu i ion in ea ly childhood. The s un ing and se e e s un ing a es in Makueni and Nandi we e 28.7, 18.5 and 30.7, and 16.5%, espec i ely. The na ional a e age o s un ing is 26 and 11% o se e e s un ing (KDHS, 2015). The le els o a la oxin in ce eals om households wi h s un ed child en di - e ed signi ican ly be ween he wo coun ies (Table4). The AFM1 le els in u ine om he s un ed child en was no signi ican ly di e - en be ween Makueni andNandi. Unde weigh Weigh - o -age is a composi e index o heigh - o -age and weigh - o -heigh . I akes in o accoun bo h acu e and ch onic malnu i- ion. Child en whose weigh - o -age (WAZ) is below –2 s anda d de ia ions (SD) a e classi ied as unde weigh . Child en whose weigh - o -age is below ದ3 SD a e conside ed se e ely unde - weigh . The unde weigh (WAZ) child en we e 2.9 and 14.6% in Nandi (n = 102) and Makueni (n = 103), espec i ely, while Makueni and Nandi egis e ed a p opo ion o 3.9% each o se e e unde weigh . The na ional a e age o unde weigh is 11 and 2% o se e e unde weigh (KDHS, 2015). The es ima ed le els o unde weigh a e 10.2 and 11% in Makueni and Nandi, espec i ely. Discussion The p esence o AFM1 in u ine indica es ecen exposu e o child en o a la oxin-con amina ed ood because he AFM1 is easily exc e ed om he body. Child en younge han 5yea s we e in oduced o ce eal-based die s a di e en imes in he wo coun ies and hei consump ion o maize-based die s anged om 0.1 o 0.25kg pe pe son pe day in Nandi and Makueni, espec i ely. Lowe con- sump ion o maize-based die s in Nandi is due o consump ion o al e na i e oods om so ghum and inge mille . Nandi compa ed o Makueni is iche in o he ypes o oods. The p opo ion o maize con amina ed wi h a la oxin om bo h coun ies was 68.3 and 80.4% in Nandi and Makueni, espec i ely (Kang’e he e al., unpub- lished da a), wi h Makueni ha ing 24.5% o posi i e samples wi h a la oxin le els abo e 10 ppb le el. So ghum was equally con ami- na ed wi h a la oxin, 66.7 and 88.9% o he samples we e posi i e and 37.1 and 29.9% exceeded he limi s in Nandi and Makueni, espec i ely. Feeding on such con amina ed oods explains he high le els o AFM1 in u ine samples om child en in bo h Nandi and Makueni coun ies. This inding ag ees wi h Gong e al. (2002, 2004) in Benin who no ed ha he le els o a la oxin–albumin adjunc le els we e app oxima ely wo imes highe in ully weaned chil- d en compa ed wi h hose ecei ing a mix u e o b eas milk and solid ood. Human b eas milk samples om Makueni had highe AFM1 le - els han he le els in Nandi samples (mean 8.46 [8.24, 8.68] pp in Makueni compa ed o 0.02 [0, 0.061] in Nandi). The p opo ion o samples wi h le els abo e he 25 pp EU limi was 10.2% in Makueni bu none exceeded his limi (25 pp ) in Nandi. This is a u he e lec- ion o he highe le els o a la oxins ound in ce eals consumed in hese wo a eas whe e mo e samples exceeded he 10 ppb limi in Makueni (25.05 [18.48, 31.62]) han in Nandi 0.98 ([0.86, 1.10]). In Nige ia Adejumo e al. (2013) epo ed 82% o he b eas milk samples we e posi i e and 16% exceeded he 25ng/kg (pp ) limi o AFM1 by he Eu opean Union. The socioeconomic s a us o he mo he s o he child en in he Nige ian s udy was ound o in lu- ence hei die a y exposu e and also ha o he in an s b eas eeding on hem. Abdul azzaq e al. (2003) epo ed ha in he Uni ed A ab Emi a es 92% o he b eas milk samples es ed we e posi i e, while Sadeghi e al. (2009) epo ed a p e alence o 98.1% posi i e in he b eas milk samples aken in Teh an bu only one sample exceeded he EU/USA limi o 25ng/kg limi . The majo sou ce o AFM1 in b eas milk, like AFM1 in child en u ine, is he ecen consump ion o a la oxin-con amina ed oods by he nu sing mo he s. Homeg own maize cons i u ed 75 and 59% o he household consumed maize in Nandi and Makueni, espec i ely. The home- g own maize had mean a la oxin le els o 0.98 [0.86, 1.10] and 25.05 [18.48, 31.62] o Nandi and Makueni, espec i ely (Kang’e he e al., unpublished da a). The o he sou ce o high le els o AFM1 in b eas Table1. A la oxin M1 in posi i e u ine samples om child en unde 30mon hs and younge han 5yea s in Makueni and Nandi coun ies. Coun y Age g oup Mean (pp ) (95% CI) - alue P- alue Makueni Younge han 30mon hs (n=34) 1182.894 (±637.99) 0.562 0.46 30–60mon hs (n=47) 1546.257 (±662.65) Nandi Younge han 30mon hs (n=42) 857.279 (±625.09) 0.481 0.49 30–60mon hs (n=54) 667.294 (±299.06) Exposu e o Kenyan popula ion o a la oxins in oods, 2017, Vol. 1, No. 2 135 milk is consump ion o cow milk con amina ed wi h AFM1. The p es- ence o AFM1 in cow milk is a esul o eeding AFB1 con amina ed eeds o he animals. Spoil maize was used as animal eed ac oss he wo coun ies (23.9% in Makueni and 77.6% in Nandi, Kang’e he e al., unpublished da a) and his ac ed as an exposu e ou e o ani- mals o a la oxins and e en ually exposu e o b eas eeding mo he s and child en weaned on o cow milk o his myco oxin. A la oxin M1 mean le els in he home-p oduced milk we e 22.3 [16.89, 7.67] pp in Makueni and 2.7 [1.92, 3.50] pp inNandi. Conside ing ha a child consumes 776±141g pe day o b eas milk (WHO/UNICEF, 1998), and aking an a e age o 776g pe day and a mean o a la oxin con amina ion o 10.83 and 0.06 pp o b eas milk ob ained om Makueni and Nandi, espec i ely, de e - mined using HPLC, exposu e h ough his ou e would be 6×10−3 and 1×10–6 µg pe kg body weigh (bw ) pe day in Makueni and Nandi, espec i ely. The le els a e lowe han he allowed limi o 19 400 pp pe day in ake conside ing a maximum limi o 25 pp and 776g o milk in ake pe day (Ghiasian and Maghsood, 2012). A la oxin in ake in child en om his ou e would be exc e ed h ough u ine and is consis en wi h hese in ake le els. This s udy con i ms ha malnu i ion is a p oblem in he wo coun ies; Makueni shows a highe pe cen age o child en unde - weigh , 14.6%, and equa ed o Nandi, 2.9%. Makueni also had a highe pe cen age o s un ed child en (28.7%) han Nandi (18.6%). These igu es, al hough lowe han hose epo ed o Kenya (16.4% o unde weigh and 36.2% o s un ing; WHO, 2010), he Makueni (in Eas e n p o ince) igu es a e lowe o s un ing and highe o unde weigh han hose epo ed o he Eas e n p o ince (30.1% s un ing and 12.2% unde weigh , KDHS, 2015). Again, he Nandi (Ri alley p o ince) igu es a e highe o s un ing han he igu es gi en o he en i e Ri alley (14.9%) bu less han he unde weigh igu e (7.3%) o Ri alley p o ince (Kenya Nu i ional p o ile, FAO, 2005), al hough hese alues a e high enough o cause conce n. These igu es a e also compa able wi h he p e alence eco ded by Gong e al. (2002) in Togo and Benin o 33% o s un ing and 29% o unde weigh . Al hough he e idence linking a la oxins o causal o s un ing is s ill weak, a la oxins a e known o con ibu e o s un - ing. I was no ed in a s udy in Togo and Benin (Gong e al., 2002) ha he deg ee o s un ing and unde weigh in child en younge han 5yea s was co ela ed wi h high a la oxin–albumin adduc s ound in he se um o 99% o he child en. In Nandi, child en younge han 30mon hs had highe le els o a la oxin M1 in u ine han hose o e 30 bu less han 60mon hs old, bu his was he e e se in Makueni. The di e ence in he a la oxin M1 le els in u ine o child en younge han 30mon hs and olde han 30mon hs in Makueni and Nandi coun ies was no s a is ically signi ican (  = 0.562, P = 0.456; = 0.481, P = 0.49, espec i ely). The e was no signi ican di e - ence o he AFM1 in u ine o child en younge han 30mon hs in Table2. A la oxin M1 in b eas milk samples (cELISA and HPLC) om b eas eeding mo he s in Nandi and Makueni coun ies. Sample Si e % Posi i e % >25 pp Range o posi i e (pp ) Mean (pp ) [95% CI] - alue P- alue cELISA B eas milk Makueni (n=98) 86.7 10.2 0.215–47.5 8.46 [8.24, 8.68] 19.96 <0.001 Nandi (n=67) 56.7 0.0 0.003–3.7 0.02 [0, 0.061] HPLC B eas milk Makueni (n=18) 22.2 11.1 1.4–152.7 10.83 [6.86, 14.8] 1.4 0.183 Nandi (n=21) 9.5 0.0 0.5–0.8 0.06 [0.043, 0.077] Table3. Exposu e o a la oxins om consump ion o maize and a la oxins in milk in Makueni and Nandi coun ies. Food ype Age in yea s A la oxins Makueni Nandi A e age consump ion (kg pe day) A e age exposu e (µg pe kg bw pe day)* A e age consump ion (kg pe day) A e age exposu e (µg pe kg bw pe day)* Maize lou <5 0.28 0.49 0.17 0.011 6–59 0.59 0.26 0.48 0.008 O e 60 0.60 0.26 0.48 0.008 Milk <5 0.27 0.0004 0.55 0.0001 6–59 0.36 0.0001 0.73 0.00004 O e 60 0.33 0.0001 0.72 0.00004 *The a e age body weigh was based on median measu ed body weigh o 16 500 men and 19 969 women in 13 EU s a es a e applying g ow h ac o s. Fo less han 5yea s, he a e age body weigh was he mean o a e age body weigh o child en less han 3yea s and a e age body weigh o child en be ween 3 and 6yea s ( an Buu en e al., 2012). Table 4. A la oxin le els in ce eals om households wi h s un - ed child en and A la oxin M1 con en in u ine samples om he s un ed child en in he same households in Makueni and Nandi coun ies. Coun y Food ype Mean a la oxin le els in ppb (±95% CI) Nandi Maize (n=37) 0.97 [0.67, 1.27] So ghum (n=9) 27.8 [0, 63.88] Mille (n=6) 1.50 [1.06, 1.94] U ine AFM1 (n=41) 997.3 [0, 17638.82] Makueni Maize (n=98) 41.5 [26.65, 56.35] So ghum (n=46) 20.43 [6.94,33.92] Mille (n=1) 2.00 U ine AFM1 (n=106) 1336 [925.33, 1746.67] AFM1,A la oxin M1 in pa s pe illion. 136 E.K. Kang’e he e al., 2017, Vol. 1, No. 2 Makueni and Nandi ( =0.67, P=0.415), bu he e was a signi ican di e ence in he AFM1 le els in u ine samples o child en olde han 30mon hs and younge han 60mon hs ( =6.126, P=0.015) wi h hose in Makueni ha ing highe le els and double hose inNandi. The a la oxin le els in maize, so ghum, and mille om house- holds wi h s un ed child en and AFM1 in u ine o s un ed child en om he same households we e compa ed o Makueni and Nandi (Table4). Households wi h s un ed child en bu whe e no samples we e analysed o a la oxins we e no aken in o accoun . The mean a la oxin le els in maize om households wi h s un ed child en we e signi ican ly di e en be ween Makueni and Nandi ( =10.75, P<0.01) whe eas a la oxin le els in so ghum and mille om house- holds wi h s un ed child en and u ine om s un ed child en we e no signi ican be ween he wo coun ies. The likelihood o child en who we e s un ed and being exposed o a la oxins (p esence o AFM1 in u ine) was 6.8 imes mo e likely o occu in Makueni compa ed o 2.1 imes in Nandi (OR=6.8 and 2.1, espec i ely) while he p ob- abili y o unde weigh child en being exposed o a la oxins was 2.4 imes mo e likely o occu in Makueni compa ed o 1.9 imes among child en in Nandi (OR=2.4 and 1.9 espec i ely). Conclusions The exposu e o people o a la oxins (o myco oxins) consequen ly s a s ea ly and con inues in o adul hood. The esul s o his s udy emphasize he need o be mo e igilan on a la oxin and con ol. I is impo an o educa e he a me s on be e ca le eeding p ac ices and mo he s on he need o eed he amily wi h good-quali y ce eals o a oid ch onic exposu e o a la oxin h ough he eed-milk chain. Mo he s would be a good a ge o hese capaci y building messages as p epa a ion o he amily ood is he mo he ’s main esponsibili y. Acknowledgemen s We hank he Minis y o Fo eign A ai s o Finland o p o iding inancial assis ance ha enabled his s udy (MFA 24819801); he communi ies in Nandi and Makueni o allowing us o ca y ou he s udy and Go e nmen o ice s in bo h coun ies o hei acili a ion. The iews exp essed in his pape a e no necessa ily hose o he unding agency o ins i u ions aking pa in he s udy, bu hose o he au ho s. Con lic o in e es s a emen . The au ho s decla e ha hey ha e no con lic o in e es wi h he publica ion o his pape . Re e ences Abdul azzaq, Y. M., Osman, N., Yousi , Z. M., Al-Falahi, S. (2003). A la oxin M1 in b eas -milk o UAE women. Annals o T opical Paedia ics 23: 173–179. Adejumo, O., A anda, O., Raiola, A., Somo in, Y., Bandyopadhyay, R., Ri ieni, A. (2013). Co ela ion be ween a la oxin M1 con en o b eas milk, die a y exposu e o a la oxin B1 and socio-economic s a us o lac- a ing mo he s in Ogun S a e, Nige ia. Food and Chemical Toxicology 56: 171–177. CAST. (2003). Myco oxins: Risks in Plan , Animal and Human Sys ems. Task Fo ce Repo No.139 CAST, CAST, Ames, Iowa, USA. Daniel, W. W. (1999). Bios a is ics: A Founda ion o Analysis in he Heal h Sciences, 7 h edn. John Wiley and Sons, New Yo k, p. 504. Daniel, J. H., e al. (2011). Comp ehensi e assessmen o maize a la oxin le els in Eas e n Kenya, 2005–2007. En i onmen al Heal h Pe spec i es 119: 1794–1799. De Onis, M., Blossne , M., Bo ghi, E. (2011). P e alence and ends o s un - ing among p eschool child en, 1990–2020. Public Heal h Nu i ion 15: 142–148. Diene , U. L., Cole, R. J., Sande s, T. H., Payne, G. A., Lee, L. S., Klich, M. A. (1987). Epidemiology o a la oxin o ma ion by Aspe gillus la us. Annual Re iew o Phy opa hology 25:249–270. Food and Ag icul u e O ganiza ion (FAO). (2000). Special Repo C op and Food Supply Si ua ion in Kenya. FAO Co po a e Documen Reposi o y. h p://www. ao.o g/doc ep/004/X7697E/X7697E00.HTM. Food and Ag icul u e O ganiza ion (FAO). (2005). Kenya Nu i ion Coun y P o ile. h p://www. ao.o g/ag/agn/nu i ion/ken_en.s m. Ghiasian, S. A., Maghsood, A. H. (2012). In an s’ exposu e o a la oxin M1 om mo he ’s b eas milk in I an. I anian Jou nal o Public Heal h 41: 119–126. Gong, Y. Y. e al. (2002). C oss-sec ional s udy o die a y a la oxin exposu e and impai ed g ow h in young child en om Benin and Togo, Wes A ica. B i ish Medical Jou nal 325: 20–21. Gong, Y. Y. e al. (2004). Pos weaning exposu e o a la oxin esul s in impai ed child g ow h: alongi udinal s udy in Benin, Wes A ica. En i onmen al Heal h Pe spec i es 112; 1334–1338. Hall, A. J., Wild, C. P. (1994). Epidemiology o a la oxin ela ed disease. In: Ea on D. A. and G oopman J. D. (eds), Human Heal h, Ve e ina y and Ag icul u al Signi icance. San Diego, CA: Academic P ess, 233–258. Hend ickse, R. G. (1997). O sick u keys, kwashio ko , mala ia, pe ina al mo ali y, he oin addic s and ood poisoning: esea ch on he in luence o a la oxins on child heal h in he opics. Annals o T opical Medicine and Pa asi ology 91: 787–793. Jae zold, R., Schmid , H., Ho ne z, B., Shisanya, C. (2006). Fa m Managemen Handbook o Kenya, Vol. 11, Pa B1 and C1. Minis y o Ag icul u e Kenya, Kenya. Kang’e he, E. K., Lang’a , A. K. (2009). An in es iga ion o A la oxin B1 and M1 con amina ion o animal eeds and milk om u ban cen e s in Kenya. A ican Heal h Sciences 9: 218–226. Kang’e he, E. K., M’Ibui, G. M., Randolph, T. F., Lang’a , A. K. (2007). The p e alence o a la oxin M1 and B1 in milk and animal eeds om u ban smallholde dai y p oduc ion in Dago e i di ision, Nai obi – Kenya. Eas A ican Medical Jou nal 84:S83–S86. Kenya Demog aphic and Heal h Su ey (KDHS). (2015). Kenya Na ional Bu eau o S a is ics. Lewis, L. e al. (2005). A la oxin con amina ion o comme cial maize p oduc s du ing an ou b eak o acu e a la oxicosis in Eas e n and Cen al Kenya. En i onmen al Heal h Pe spec i es 113: 1763–1767. Ma in, S. W., Meek, A. H., Willebe g, P. (1987). Ve e ina y Epidemiology: P inciples and Me hods. Iowa S a e Uni e si y P ess, p. 343. Mu homi, J. W., Mu ei hi, B. K., Chemining’WA, G. N., Ga humbi, J. K., Mu i u, E. W. (2012). Aspe gillus and a la oxin B1 con amina ion o maize and maize p oduc s om Eas e n and No h-Ri egions o Kenya. In e - na ional Jou nal o Ag icul u e Sciences 2: 22–34. Mu u e, B. N., Ogana G. (2005). A la oxin le els in maize and maize p oduc s du ing he 2004 ood poisoning ou b eak in Eas e n P o ince o Kenya. Eas A ican Medical Jou nal 82: 275–279. M’mboyi, F., Mugo S., Mwimali M., Ambani L. (2010). Maize P oduc ion and Imp o emen in Sub-Saha an A ica. A ican Bio echnology S akeholde s Fo um. Vol. 25, 2013. Mwihia, J. T., e al. (2008). A la oxin le els in locally g own maize om Mak- ueni Dis ic , Kenya. Eas A ican Medical Jou nal 85: 311–317. Oko h, S. A., Ohingo, M. (2004). Die a y a la oxin exposu e and impai ed g ow h in young child en om Kisumu Dis ic , Kenya. A ican Jou nal o Heal h Sciences 11: 43–54. P obs , C., Schul hess, F., Co y, P. J. (2010). Impac o Aspe gillus sec ion Fla i communi y s uc u e on he de elopmen o le hal le els o a la oxins in Kenyan maize (Zea mays). Jou nal o Applied Mic obiology 108; 600–610. Sadeghi, N., O eisi, M. R., Janna , B., Hajimahmoodi, M., Bonyani, H., Jan- na , F. (2009). Incidence o a la oxin M1 in human b eas milk in Teh an, I an. Food Con ol 20: 75–78. S osnide , H. e  al. (2006). Wo kg oup epo : public heal h s a egies o educing a la oxin exposu e in de eloping coun ies. En i onmen al Heal h Pe spec i es 114: 1989–1903. an Buu en, S., Schönbeck, Y., an Dommelen, P. (2012). Collec ion, colla ion and analysis o da a in ela ion o e e ence heigh s and e e ence weigh s o emale and male child en and adolescen s (0–18yea s) in he EU, as Exposu e o Kenyan popula ion o a la oxins in oods, 2017, Vol. 1, No. 2 137 well as in ela ion o he age o onse o pube y and he age a which di - e en s ages o pube y a e eached in adolescen s in he EU. EFSA Sup- po ing Publica ions, I aly, p. 59. P ojec de eloped on he p ocu emen p ojec CT/EFSA/NDA/2010/01. 2012:EN-255. Vic o a, C. G., e al.; o he Ma e nal and Child Unde nu i ion S udy G oup (2008). Ma e nal and child unde nu i ion: consequences o adul heal h and human capi al. The Lance 371: 340–357. Wakhisi, J., K i ika, P., Buziba, N., Ro ich, J. (2005). Esophageal can- ce in No h Ri Valley o Wes e n Kenya. A ican Heal h Sciences 5: 156–163. Wild, C. P., Tu ne , P. C. (2002). The oxicology o a la oxins as he basis o public heal h decisions. Mu agenesis 17: 471–481. Wo ld Heal h O ganiza ion (WHO). (1986). Use and in e p e a ion o an h o- pome ic indica o s o nu i ional s a us. Bulle in o he Wo ld Heal h O ganiza ion 64: 924–941. Wo ld Heal h O ganiza ion o he Uni ed Na ions (WHO). (2010). Wo ld Heal h S a is ics, pp. 105–106. WHO/UNICEF. (1998). Complemen a y Feeding o Young Child en in De el- oping Coun ies: A Re iew o Cu en Scien i ic Knowledge. (WHO/ NUT/98.1). Gene a: Wo ld Heal h O ganiza ion.