Prenatal Iron Deficiency and Replete Iron Status Are Associated with Adverse Birth Outcomes, but Associations Differ in Ghana and Malawi

The Jou nal o Nu i ion Communi y and In e na ional Nu i ion P ena al I on De iciency and Reple e I on S a us A e Associa ed wi h Ad e se Bi h Ou comes, bu Associa ions Di e in Ghana and Malawi B ie a M Oaks,1Josh M Jo gensen,2Lacey M Baldi iez,4Se h Adu-A a wuah,5Ken Male a,6 Ha ie Ok onipa,2,5John Sadalaki,6Anna La ey,5Pe Asho n,7Ulla Asho n,7,8S ephen Vos i,3Lindsay H Allen,4and Ka h yn G Dewey2 1Depa men o Nu i ion and Food Sciences, Uni e si y o Rhode Island, Kings on, RI; 2P og am in In e na ional and Communi y Nu i ion, Depa men o Nu i ion; 3Depa men o Ag icul u al and Resou ce Economics, Uni e si y o Cali o nia, Da is, CA; 4USDA, Ag icul u al Resea ch Se ice Wes e n Human Nu i ion Resea ch Cen e , Da is, CA; 5Depa men o Nu i ion and Food Science, Uni e si y o Ghana, Legon, Ghana; 6Depa men o Communi y Heal h, Uni e si y o Malawi College o Medicine, Blan y e, Malawi; 7Depa men o Paedia ics, Tampe e Uni e si y Hospi al, Tampe e, Finland; and 8Cen e o Child Heal h Resea ch, Uni e si y o Tampe e Facul y o Medicine and Li e Sciences and Tampe e Uni e si y Hospi al, Tampe e, Finland ABSTRACT Backg ound: P e ious li e a u e sugges s a U-shaped ela ion be ween hemoglobin concen a ion and ad e se bi h ou comes. The e is less e idence on associa ions be ween i on s a us and bi h ou comes. Objec i e: Ou objec i e was o de e mine he associa ions o ma e nal hemoglobin concen a ion and i on s a us wi h bi h ou comes. Me hods We conduc ed a seconda y da a analysis o da a om 2 coho s o p egnan women ecei ing i on-con aining nu i ional supplemen s (20–60 mg e ous sul a e) in Ghana (n=1137) and Malawi (n=1243). Hemoglobin concen a ion and 2 ma ke s o i on s a us [zinc p o opo phy in and soluble ans e in ecep o (sT R)] we e measu ed a ≤20 weeks and 36 weeks o ges a ion. We used linea and Poisson eg ession models and bi h ou comes included p e e m bi h (PTB), newbo n s un ing, low bi h weigh (LBW), and small- o -ges a ional-age. Resul s: P e alence o i on de iciency (sT R >6.0 mg/L) a en ollmen was 9% in Ghana and 20% in Malawi. In ea ly p egnancy, i on de iciency was associa ed wi h PTB (9% compa ed wi h 17%, adjus ed RR: 1.63; 95% CI: 1.14, 2.33) and s un ing (15% compa ed wi h 23%, adjus ed RR: 1.44; 95% CI: 1.09, 1.94) in Malawi bu no Ghana, and was no associa ed wi h LBW in ei he coun y; eple e i on s a us (sT R <10 h pe cen ile) was associa ed wi h s un ing (9% compa ed wi h 15%, adjus ed RR: 1.71; 95% CI: 1.06, 2.77) in Ghana, bu no PTB o LBW, and was no associa ed wi h any bi h ou comes in Malawi. In la e p egnancy, i on de iciency was no ela ed o bi h ou comes in ei he coun y and i on- eple e s a us was associa ed wi h highe isk o LBW (8% compa ed wi h 16%, adjus ed RR: 1.90; 95% CI: 1.17, 3.09) and s un ing (6% compa ed wi h 13%, adjus ed RR: 2.14; 95% CI: 1.21, 3.77) in Ghana, bu was no associa ed wi h bi h ou comes in Malawi. Conclusions: The associa ions o low o eple e i on s a us wi h bi h ou comes a e popula ion speci ic. Resea ch o eplica e and ex end hese indings would be bene icial. These ials we e egis e ed a clinical ials.go as NCT00970866 (Ghana) and NCT01239693 (Malawi). J Nu 2019;149:513–521. Keywo ds: p egnancy, i on de iciency, p e e m bi h, low bi h weigh , newbo n s un ing, anemia, i on s a us, Ghana, Malawi In oduc ion The WHO cu en ly ecommends daily supplemen a ion wi h 30–60 mg/d elemen al i on (+400 μg olic acid) h oughou p egnancy, and in se ings whe e anemia in p egnan women is a se e e public heal h p oblem (p e alence o ≥40%), he daily dose o 60 mg is ecommended o e a lowe dose (1). Al hough his amoun o supplemen a ion clea ly educes ma e nal i on de iciency and anemia, i is ques ionable whe he he e is a bene icial impac on bi h ou comes (2) and he e is conce n ha high ma e nal i on s a us may ha e a nega i e impac on he newbo n (3). Nume ous s udies ha e epo ed a U-shaped ela ion be ween ma e nal hemoglobin (Hb) concen a ions du ing C 2019 Ame ican Socie y o Nu i ion. This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i eco mmons.o g/licenses/by/4.0/), which pe mi s un es ic ed euse, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. Manusc ip ecei ed Ap il 10, 2018. Ini ial e iew comple ed May 29, 2018. Re ision accep ed Oc obe 8, 2018. Fi s published online Janua y 9, 2019; doi: h ps://doi.o g/10.1093/jn/nxy278. 513 Downloaded om h ps://academic.oup.com/jn/a icle-abs ac /149/3/513/5281546 by Raila Melin use on 23 Ap il 2019 p egnancy and ad e se bi h ou comes, wi h bo h low and high Hb concen a ions associa ed wi h highe isks o low bi h weigh (LBW) and p e e m bi h (PTB) (4–7), al hough i is unclea whe he hese associa ions a e due o ma e nal i on s a us o o he mechanisms. Resea ch inco po a ing speci ic bioma ke s o i on s a us is compa a i ely limi ed and has been conduc ed p ima ily in highe -income coun ies whe e women ypically ake lowe doses o i on du ing p egnancy: inc eased isk o ad e se bi h ou comes has been obse ed wi h bo h low (8,9) and high (10–14) ma e nal i on s a us. A ecen e iew no ed ha in s udies ha ha e examined ma e nal i on s a us and bi h ou comes, se um e i in concen a ion is he mos widely used bioma ke o i on s a us (15). Se um e i in concen a ion is an indica o o i on s o es, bu is a ec ed by in lamma ion. Fu he mo e, ma e nal plasma olume no mally expands du ing p egnancy [mainly in he second and hi d imes e s (16,17)] which leads o lowe concen a ions o Hb, e i in, and o he bioma ke s. Thus, ela i ely high concen a ions o Hb o e i in in he second and hi d imes e s could be a ma ke o inadequa e plasma olume expansion, which is associa ed wi h ad e se bi h ou comes. Soluble ans e in ecep o (sT R) and zinc p o opo phy in (ZPP) a e al e na i e ma ke s o i on s a us ha may also be a ec ed by in lamma ion, bu because hey espond o i on de iciency in he opposi e di ec ion o e i in, lowe alues o sT R o ZPP indica e highe i on s a us, which would no be caused by ailu e o plasma olume expansion. Li le esea ch has been conduc ed ega ding ma e nal i on s a us and bi h ou comes in sub-Saha an A ica (15), despi e he high p e alence o ma e nal anemia and i on de iciency in he egion. The aim o his c oss-coun y compa ison was o examine he associa ions o ma e nal Hb concen a ion and i on s a us du ing p egnancy wi h bi h ou comes in Ghana and Malawi. In pa icula , we we e in e es ed in bo h low and high Hb concen a ion and i on s a us, as well as whe he he iming o hese measu emen s (ea ly compa ed wi h la e p egnancy) had any in luence on he a o emen ioned associa ions. Me hods Pa icipan s and s udy design Women included in his coho s udy we e om 2 andomized con olled ials conduc ed in Malawi (NCT01239693) and Ghana (NCT00970866), bo h mala ia-endemic coun ies, as pa o he Suppo ed by Bill & Melinda Ga es Founda ion g an 49817 o he Uni e si y o Cali o nia, Da is and an O ice o Heal h, In ec ious Diseases, and Nu i ion, Bu eau o Global Heal h, US Agency o In e na ional De elopmen (USAID) g an o he Uni e si y o Cali o nia, Da is unde e ms o Coope a i e Ag eemen No. AID-OAA-A-12-00005 h ough he Food and Nu i ion Technical Assis ance III P ojec (FANTA), managed by FHI 360. Au ho disclosu es: BMO, JMJ, LMB, SA-A, KM, HO, JS, AL, PA, UA, SV, LHA, and KGD, no con lic s o in e es . The indings and conclusions con ained wi hin he a icle a e hose o he au ho s and do no necessa ily e lec posi ions o policies o USAID o he Bill & Melinda Ga es Founda ion. Supplemen al Me hods, Supplemen al Figu e 1, and Supplemen al Tables 1– 9 a e a ailable om he “Supplemen a y da a” link in he online pos ing o he a icle and om he same link in he online able o con en s a h ps://academic.oup.com/jn/. Add ess co espondence o BMO (e-mail: [email p o ec ed]). Abb e ia ions used: AGP,α-1-acid glycop o ein; CRP, C- eac i e p o ein; Hb, hemoglobin; HCZ, head ci cum e ence zsco e; IDA, i on de iciency anemia; LAZ, leng h- o -age zsco e; LBW, low bi h weigh ; PTB, p e e m bi h; PVE, plasma olume expansion; SGA, small- o -ges a ional-age; SQ-LNS, small- quan i y lipid-based nu ien supplemen ; sT R, soluble ans e in ecep o ; ZPP, zinc p o opo phy in. In e na ional Lipid-Based Nu ien Supplemen s P ojec (www.ilins. o g). The p ima y objec i e o hese ials was o de e mine he e icacy o small-quan i y lipid-based nu ien supplemen s (SQ-LNSs) o p e en ing malnu i ion in p egnan and lac a ing women and hei in an s. De ails o he s udy me hods ha e been epo ed elsewhe e (18–20). B ie ly, he ials we e simila in design, bu each ial ope a ed independen ly and he e we e some di e ences in inclusion and exclusion c i e ia. In bo h ials, women ≤20 weeks o ges a ion we e andomly assigned o ecei e daily h oughou p egnancy ei he : 1) a 60-mg Fe ( e ous sul a e) +400-μg olic acid capsule; 2)a mul iple mic onu ien capsule (18 mic onu ien s); o 3) a sache o SQ-LNS (118 kcal, 22 mic onu ien s, essen ial a y acids, and p o ein). Bo h he mul iple mic onu ien capsule and SQ-LNS con ained 20 mg Fe ( e ous sul a e) pe daily supplemen . Di e ences in inclusion and exclusion c i e ia included ma e nal age (Ghana ial: ≥18 y o age; Malawi ial: ≥15 y o age) and HIV s a us (Ghana ial: excluded women i hey we e HIV posi i e; Malawi ial: did no exclude women i hey we e HIV posi i e). Women ecei ed in e mi en p e en i e mala ia ea men in acco dance wi h s anda d an ena al ca e ecommenda ions in Ghana and Malawi. A bo h en ollmen and 36 weeks o ges a ion, ained phlebo omis s collec ed blood samples by enipunc u e wi h hepa inized blood collec ion ubes. Blood samples we e ypically collec ed in he mo ning, wi h he majo i y o samples collec ed be ween 0800 and 1200. P e ious s udies sugges ha ime o day does no a ec he eliabili y o ma ke s o i on s a us (21). In Ghana, we e e ed any woman wi h Hb <70 g/L a baseline o <100 g/L a 36 weeks o ges a ion o he hospi al o ea men o anemia, bu allowed hem o emain in he s udy. In Malawi, we excluded women wi h Hb <50 g/L a baseline (<0.1% o women) and e e ed hem o ea men ; women wi h Hb 50–69 g/L a baseline we e ea ed wi h i on supplemen s i ound o be i on de icien acco ding o ZPP plasma concen a ion and we e allowed o emain in he s udy. Bo h ials collec ed sociodemog aphic in o ma ion and ained an h opome is s measu ed ma e nal weigh and heigh a en ollmen . Women om hese ials we e included in his coho s udy i hey had: 1) a leas one measu emen o Hb concen a ion o i on s a us, a ei he en ollmen o 36 weeks o ges a ion, and 2) any newbo n an h opome ic measu emen o an es ima ed p egnancy du a ion. We ha e p e iously epo ed ha supplemen a ion wi h SQ-LNSs o mul iple mic onu ien s esul ed in lowe Hb and i on s a us a 36 weeks o ges a ion han o he i on and olic acid ial a m (22,23). Assigned supplemen g oup in he main ial was no he ocus o he cu en s udy and was con olled o in analyses o he 36-wk measu emen s. Bi h ou comes In Ghana, newbo n measu emen s we e ob ained by ieldwo ke s wi hin 48 h a e bi h o 90.6% o in an s and be ween 3 and 14 d o 9.4% o in an s. In Malawi, ieldwo ke s measu ed bi h weigh usually wi hin 48 h a e bi h, wi h 10.9% o bi h weigh s ob ained be ween 3 and 14 d. All o he newbo n an h opome y o he Malawi ial was comple ed a a clinic isi 1–2 wk a e bi h. All newbo n an h opome ic measu emen s ob ained 3–14 d a e bi h we e back-calcula ed using o mulae desc ibed by he WHO (24). Fieldwo ke s measu ed bi h weigh o he nea es 20 g (Seca 383; Seca GmbH & Co.), leng h o he nea es 0.1 cm (Ghana: Seca 416; Seca GmbH & Co.; Malawi: Ha penden In an ome e ; Hol ain Limi ed), and head ci cum e ence o he nea es 0.1 cm (Sho ape, Weigh and Measu e, LLC). We de e mined ges a ional age a en ollmen by ul asound (Ghana: Aloka SSD 500; Malawi: EDAN DUS 3 Digi al Ul asonic Diagnos ic Imaging Sys em; EDAN Ins umen s Inc.) and da e o bi h was ob ained by in e iewing he mo he and con i med wi h he in an ’s hospi al o heal h ca d when a ailable. Labo a o y analysis Labo a o y echnicians assayed Hb (HemoCue) and mala ia pa a- si emia wi h apid es s (Ghana: Vision Bio ech; Malawi: Clea iew Mala ia Combo; B i ish Biocell In e na ional L d.) in pe iphe al blood and cen i uged he emaining blood samples a 1252 ×g o 15 min a oom empe a u e o sepa a e RBCs and plasma. RBCs we e washed 514 Oaks e al. Downloaded om h ps://academic.oup.com/jn/a icle-abs ac /149/3/513/5281546 by Raila Melin use on 23 Ap il 2019 TABLE 1 Cha ac e is ics o 2 s udy coho s o p egnan women in Ghana and Malawi a ≤20 weeks o ges a ion1 Ghana (n=1137) Malawi (n=1243) P Ma e nal age, y 26.7 ±5.5 25.3 ±6.1 <0.001 Nullipa ous women 33.8 19.7 <0.001 Male e us248.4 48.7 0.88 Ges a ional age, wk 16.1 ±3.3 16.8 ±2.1 <0.001 BMI, kg/m224.7 ±4.2 22.1 ±2.8 <0.001 O e weigh o obese (BMI ≥25 kg/m2) 42.3 12.0 <0.001 Low BMI (<18.5 kg/m2) 2.9 5.6 0.001 Posi i e HIV es —312.5 —3 Posi i e mala ia es 10.2 22.7 <0.001 Household Food Insecu i y Index42.6 ±4.3 4.9 ±4.5 <0.001 1Values a e means ±SDs o pe cen ages; P alues a e o chi-squa e es s (ca ego ical a iables) o es s (con inuous a iables). 2Based on in an sex a bi h. 3HIV+women we e no en olled in he Ghana ial. 4Scale anging om 0 (no ood insecu i y) o 27 (e e y ood insecu i y condi ion occu s o en). 3 imes wi h no mal saline and we used he o iginal A i co e -slides and 3-le el con ol ma e ial o he ZPP measu emen s ob ained using a hema o luo ome e (A i Biomedical, Inc.). Plasma samples we e s o ed a −20◦C and shipped o he USDA Ag icul u al Resea ch Se ice Wes e n Human Nu i ion Resea ch Cen e , Da is, CA whe e sT R (mg/L), α-1-acid glycop o ein [AGP (g/L)], and C- eac i e p o ein [CRP (mg/L)] concen a ions we e de e mined using a Cobas In eg a 400 plus Au oma ic Analyze (Roche Diagnos ic Co p.). The in e - and in a- assay CVs o hese assays we e: 1) sT R in a-assay: <2.2%, in e assay: <1.1%; 2) AGP in a-assay: <2.9%, in e assay: <2.0%; 3) CRP: in a- assay: <3.5%, in e assay: <1.3%. S a is ical analysis We es ed no mali y using he Shapi o–Wilk es . Hb was de e mined o ha e a no mal dis ibu ion and ZPP and sT R concen a ions we e log ans o med. We also analyzed Hb, ZPP, and sT R using ca ego ical a iables. We de ined anemia using a cu o alue o Hb <100 g/L, based on esea ch sugges ing ha his cu o alue is mo e accu a e when de ining anemia in p egnan women o A ican descen (7,25, 26). We de ined high Hb as >130 g/L (2) and ele a ed sT R (p oxy o issue i on de iciency) as >6.0 mg/L. We de i ed he 6.0 mg/L cu o alue o sT R based on p e ious esea ch epo ing ha sT R alues ob ained using he Au oma ic Analyze assay (as used in his s udy) a e on a e age 30% lowe han alues ob ained wi h he ELISA assay (27). The e o e, we educed by 30% he 8.5 mg/L cu o alue used when sT R is de e mined wi h ELISA (28) o ob ain he cu o o ∼6.0 mg/L o ou analysis. We de ined i on de iciency anemia (IDA) as Hb <100 g/L and sT R >6.0 mg/L. A s udy in an adul nonp egnan A ican popula ion indica ed ha sT R concen a ion is in e sely co ela ed wi h issue i on concen a ions and is dec eased in he p esence o i on o e load (29), al hough he e is no such e idence o ZPP. Due o he lack o a published cu o o low sT R, we ca ego ized women as i on eple e i sT R was <10 h pe cen ile based on he dis ibu ion wi hin each coho . Fo each coun y, we used linea eg ession models o examine he associa ions o Hb, ZPP, and sT R wi h bi h ou comes as con inuous a iables [du a ion o ges a ion, bi h weigh , leng h- o -age zsco e (LAZ), and head ci cum e ence zsco e (HCZ)]. Poisson eg ession models we e used o es ima e RR o dicho omous bi h ou comes, including PTB (<37 weeks o ges a ion), LBW (<2.5 kg), small- o - ges a ional-age (SGA) [bi h weigh <10 h pe cen ile by ges a ional age and sex using he INTERGROWTH-21s s anda d (30,31)], and s un ing (LAZ <−2), in associa ion wi h he dicho omous p edic o s o Hb and i on s a us de ined p e iously. PTB was examined only wi h espec o measu emen s o Hb and i on s a us aken a en ollmen because many PTBs occu ed be o e he 36 wk blood d aw. HCZ was no analyzed as a ca ego ical a iable owing o a low numbe o in an s wi h HCZ <−2. We checked all models o U-shaped ela ions using quad a ic e ms and ound a lack o any U-shaped ela ions excep o 1 associa ion (Hb concen a ion a 36 wk and p egnancy du a ion). We conside ed co a ia es o inclusion in he model i hey we e signi ican ly (P<0.1) associa ed wi h he ou come in bi a ia e analyses. Based on p e ious li e a u e, a iables iden i ied a p io i as po en ial con ounding ac o s we e ges a ional age a en ollmen , pa i y (nullipa ous compa ed wi h pa ous), ma e nal age, educa ion le el, household ood insecu i y, household asse index, AGP TABLE 2 P e alence o anemia and i on de iciency and mean concen a ions o Hb, ma ke s o i on s a us, and in lamma ion a ≤20 weeks and 36 weeks o ges a ion in p egnan women in Ghana and Malawi1 ≤20 weeks o ges a ion 36 weeks o ges a ion Ghana (n=1137)2Malawi (n=1243)2PGhana (n=1137)2Malawi (n=1243)2P CRP, mg/L 6.9 ±11.6 8.7 ±17.8 0.003 5.7 ±15.8 6.5 ±14.1 0.27 AGP, g/L 0.65 ±0.21 0.73 ±0.25 <0.001 0.48 ±0.20 0.56 ±0.23 <0.001 Hb, g/L 111 ±12 112 ±16 0.85 117 ±12 111 ±15 <0.001 Anemia (Hb <100 g/L) 14.0 18.6 <0.001 6.0 19.6 <0.001 sT R, mg/L 4.1 ±2.6 4.7 ±2.7 <0.001 4.5 ±1.7 5.6 ±3.0 <0.001 ZPP, μmol/mol heme 45 ±28 53 ±40 <0.001 46 ±23 60 ±41 <0.001 I on de iciency (sT R >6.0 mg/L) 9.3 19.7 <0.001 14.6 33.5 <0.001 IDA (Hb <100 g/L and sT R >6.0 mg/L) 4.3 6.7 0.001 2.9 10.1 <0.001 1Values a e means ±SDs o pe cen ages; P alues a e o chi-squa e es s (ca ego ical a iables) o es s (con inuous a iables). AGP, α-1-acid glycop o ein; CRP, C- eac i e p o ein; Hb, hemoglobin; IDA, i on de iciency anemia; sT R, soluble ans e in ecep o ; ZPP, zinc p o opo phy in. 2Missing da a a en ollmen o Ghana included CRP (n=21), AGP (n=21), sT R (n=21), ZPP (n=2), and IDA (n=21) and o Malawi included CRP (n=7), AGP (n=7), Hb (n=1), sT R (n=7), ZPP (n=52), and IDA (n=7). Missing da a a week 36 o Ghana included CRP (n=156), AGP (n=156), Hb (n=152), sT R (n=156), ZPP (n=157), and IDA (n=159) and o Malawi included CRP (n=140), AGP (n=136), Hb (n=165), sT R (n=136), ZPP (n=187), and IDA (n=183). P ena al i on s a us and bi h ou comes 515 Downloaded om h ps://academic.oup.com/jn/a icle-abs ac /149/3/513/5281546 by Raila Melin use on 23 Ap il 2019 TABLE 3 S anda dized eg ession coe icien s o Hb, sT R, and ZPP wi h p egnancy du a ion and newbo n an h opome ic indica o s in Ghana and Malawi1 Hb sT R ZPP Adjus ed β(95% CI)2PAdjus ed β(95% CI)2PAdjus ed β(95% CI)2P Ea ly p egnancy: <20 wk P egnancy du a ion Ghana 0.05 (−0.01, 0.12) 0.11 0.00 (−0.06, 0.06) 0.92 0.04 (−0.02, 0.10) 0.18 Malawi 0.09 (0.03, 0.14) 0.004 −0.07, (−0.13, −0.02) 0.01 −0.07 (−0.12, −0.01) 0.02 Bi h weigh Ghana 0.03 (−0.03, 0.08) 0.39 0.02 (−0.04, 0.08) 0.48 0.01 (−0.04, 0.07) 0.68 Malawi 0.08 (0.02, 0.15) 0.006 −0.10 (−0.16, −0.04) <0.001 −0.03 (−0.09, 0.03) 0.35 Newbo n LAZ Ghana 0.01 (−0.05, 0.07) 0.66 0.06 (0.00, 0.12) 0.03 0.05 (0.00, 0.11) 0.07 Malawi 0.08 (0.02, 0.14) 0.007 −0.10 (−0.16, −0.04) <0.001 −0.06 (−0.12, −0.003) 0.04 Newbo n HCZ Ghana 0.03 (−0.03, 0.09) 0.35 0.03 (−0.03, 0.09) 0.37 0.04 (−0.02, 0.10) 0.21 Malawi 0.05 (−0.01, 0.11) 0.13 −0.07, (−0.13, −0.01) 0.03 −0.02 (−0.08, 0.04) 0.58 La e p egnancy: 36 wk P egnancy du a ion Ghana —3—30.05 (−0.01, 0.09) 0.10 0.00 (−0.05, 0.04) 0.88 Malawi −0.02 (−0.05, 0.01) 0.27 −0.03 (−0.07, 0.001) 0.055 −0.03 (−0.06, 0.002) 0.07 Bi h weigh Ghana −0.04 (−0.10, 0.02) 0.18 0.10 (0.03, 0.15) 0.002 0.09 (0.02, 0.14) 0.01 Malawi −0.03 (−0.08, 0.03) 0.32 0.001 (−0.06, 0.06) 0.96 0.04 (−0.02, 0.09) 0.24 Newbo n LAZ Ghana −0.05 (−0.10, 0.01) 0.12 0.10 (0.03, 0.15) 0.002 0.07 (0.00, 0.12) 0.04 Malawi 0.001 (−0.05, 0.06) 0.97 0.01 (−0.05, 0.06) 0.85 0.01 (−0.04, 0.07) 0.69 Newbo n HCZ Ghana −0.03 (−0.09, 0.03) 0.33 0.07 (0.01, 0.12) 0.03 0.04 (−0.02, 0.09) 0.18 Malawi −0.05 (−0.10, 0.01) 0.14 0.04 (−0.02, 0.09) 0.21 0.07 (−0.0004, 0.12) 0.052 1Hb, hemoglobin; HCZ, head-ci cum e ence- o -age zsco e; LAZ, leng h- o -age zsco e; sT R, soluble ans e in ecep o ; ZPP, zinc p o opo phy in. 2S anda dized eg ession coe icien s. Adjus ed models included he ollowing co a ia es i signi ican ly (P<0.1) associa ed wi h he ou come: ges a ional age a en ollmen , pa i y, ma e nal age, educa ion le el, household ood insecu i y, household asse index, α-1-acid glycop o ein a he ime he blood sample was d awn, C- eac i e p o ein a he ime he blood sample was d awn, in an sex, ma e nal BMI a en ollmen , ma e nal mala ia a en ollmen , and HIV s a us (Malawi models only). All 36-wk models adjus ed o in e en ion g oup. Speci ics ega ding each adjus ed model a e p o ided in he Supplemen al Me hods. 3U-shaped ela ion; see Figu e 1. a he ime he blood sample was d awn, CRP a he ime he blood sample was d awn, in an sex, baseline ma e nal BMI (in kg/m2), a posi i e apid es o mala ia a en ollmen , and HIV s a us (Malawi models only). Household ood insecu i y was measu ed using a scale anging om 0 (no ood insecu i y) o 27 (e e y ood insecu i y condi ion occu s o en) (32). We c ea ed he household asse index based on ligh ing sou ce, d inking wa e supply, sani a ion acili ies, loo ing ma e ials, adio, ele ision, e ige a o , cell phone, and s o e using p incipal componen s analysis (33). Speci ic mul i a iable models o each ou come a e p o ided in he Supplemen al Me hods.All mul i a iable models examining 36-wk measu emen s o Hb and i on s a us we e also adjus ed o in e en ion g oup. In e ac ions wi h ma e nal age, pa i y, and supplemen g oup we e examined and de e mined o be signi ican a P<0.1, and s a i ied analyses we e pe o med o signi ican in e ac ions. All analyses we e pe o med using SAS e sion 9.4 (SAS Ins i u e). Resul s O he 1320 and 1391 women en olled in he ials in Ghana and Malawi, espec i ely, 1137 and 1243 we e included in his se o analyses, espec i ely. Reasons o exclusion included win p egnancy (Ghana: n=22, Malawi: n=12), misca iage (Ghana: n=37, Malawi: n=10), s illbi h (Ghana: n=29, Malawi: n=26), and loss- o- ollow-up (Ghana: n=95, Malawi: n=100) (Supplemen al Figu e 1). In Malawi, included pa icipan s we e mo e likely o be olde (25 compa ed wi h 24 y, P<0.001); ha e a lowe p oxy SES (−0.04 compa ed wi h 0.18, P<0.002); and be less o en nullipa ous (19.7% compa ed wi h 28.8%, P<0.001), HIV posi i e (12.5% compa ed wi h 17.8%, P=0.029), o anemic (18.6% compa ed wi h 27.6%, P<0.001) han excluded pa icipan s, bu hey had simila BMI, yea s o educa ion, and mala ia a es a en ollmen o hose excluded. In Ghana, excluded s udy pa icipan s did no di e om included pa icipan s in any o he a o emen ioned lis ed cha ac e is ics. S udy pa icipan cha ac e is ics a e p o ided in Table 1. P e alence o anemia (Hb <100 g/L), i on de iciency (sT R >6.0 mg/L), and IDA and mean concen a ions o Hb, sT R, and ZPP a en ollmen and 36 weeks o ges a ion a e p esen ed in Table 2.Mean±SD p egnancy du a ion was 39.3 ±1.9wkand 39.1 ±2.9 wk in Ghana and Malawi, espec i ely. Newbo n mean ±SD bi h weigh was 2982 ±432 g in Ghana and 2970 ±447 g in Malawi. Ad e se bi h ou comes included PTB (Ghana: 8.5%,Malawi: 10.0%),LBW (Ghana: 11.8%,Malawi: 12.8%), SGA (Ghana: 21.8%, Malawi: 29.5%), and s un ing (Ghana: 9.4%, Malawi: 16.0%). Hb and i on s a us a en ollmen —associa ions wi h bi h ou comes A en ollmen , Hb concen a ion in adjus ed linea eg ession analyses was no associa ed wi h any bi h ou comes in 516 Oaks e al. Downloaded om h ps://academic.oup.com/jn/a icle-abs ac /149/3/513/5281546 by Raila Melin use on 23 Ap il 2019 TABLE 4 Risk o ad e se bi h ou comes o women wi h i on de iciency du ing ea ly o la e p egnancy1 Wi hou i on de iciency, n/ o al n(%)2 Wi h i on de iciency, n/ o al n(%)3Adjus ed RR (95% CI)4P Ea ly p egnancy: ≤20 wk PTB Ghana 77/924 (8.3) 10/102 (9.8) 1.13 (0.61, 2.11) 0.70 Malawi 82/912 (9.0) 41/248 (16.5) 1.63 (1.14, 2.33) 0.007 LBW Ghana 110/923 (11.9) 7/102 (6.9) 0.51 (0.24, 1.05) 0.07 Malawi 100/813 (12.3) 35/210 (16.7) 1.24 (0.87, 1.75) 0.23 SGA Ghana 194/889 (21.8) 19/100 (19.0) 0.87 (0.57, 1.33) 0.53 Malawi 195/788 (24.7) 58/205 (28.3) 1.15 (0.89, 1.49) 0.27 Newbo n s un ing Ghana 78/918 (8.5) 7/102 (6.9) 0.81 (0.39, 1.68) 0.56 Malawi 114/762 (15.0) 48/205 (23.4) 1.44 (1.09, 1.94) 0.01 La e p egnancy: 36 wk LBW Ghana 58/753 (7.7) 10/141 (7.1) 0.78 (0.42, 1.44) 0.43 Malawi 53/560 (9.5) 34/327 (10.4) 0.97 (0.65, 1.46) 0.89 SGA Ghana 161/735 (21.9) 28/138 (20.3) 0.82 (0.57, 1.18) 0.28 Malawi 137/546 (25.1) 84/321 (26.2) 1.03 (0.82, 1.29) 0.81 Newbo n s un ing Ghana 42/750 (5.6) 6/141 (4.3) 0.64 (0.28, 1.45) 0.28 Malawi 76/536 (14.2) 38/317 (12.0) 0.76 (0.53, 1.08) 0.13 1PTB: <37 weeks o ges a ion; LBW: <2.5 kg; SGA: bi h weigh <10 h pe cen ile by ges a ional age and sex using he INTERGROWTH-21s s anda d (30); s un ing: leng h- o -age zsco e <−2. LBW, low bi h weigh ; PTB, p e e m bi h; SGA, small- o -ges a ional-age; sT R, soluble ans e in ecep o . 2Womenwi hsT R≤6.0 mg/L. The e e ence g oup excluded women wi h i on- eple e s a us (sT R <10 h pe cen ile). A ≤20 wk, his was <2.49 mg/L o Ghana and <2.65 mg/L o Malawi. A 36 wk, his was <2.86 mg/L o Ghana and <3.08 mg/L o Malawi. 3sT R >6 mg/L. 4Adjus ed models included he ollowing co a ia es i signi ican ly (P<0.1) associa ed wi h he ou come: ges a ional age a en ollmen , pa i y, ma e nal age, educa ion le el, household ood insecu i y, household asse index, α-1-acid glycop o ein a he ime he blood sample was d awn, C- eac i e p o ein a he ime he blood sample was d awn, in an sex, ma e nal BMI a en ollmen , ma e nal mala ia a en ollmen , and HIV s a us (Malawi models only). All 36-wk models adjus ed o in e en ion g oup. Speci ics ega ding each adjus ed model a e p o ided in he Supplemen al Me hods. Ghana bu was associa ed wi h a longe p egnancy du a ion, highe bi h weigh , and highe newbo n LAZ in Malawi (Table 3). Highe sT R, indica ing a lowe i on s a us, was associa ed wi h a sho e p egnancy du a ion, lowe bi h weigh , lowe newbo n LAZ, and lowe HCZ in Malawi, bu was associa ed wi h a highe newbo n LAZ in Ghana. Highe ZPP, also an indica o o lowe i on s a us, was associa ed wi h a sho e p egnancy du a ion and lowe newbo n LAZ in Malawi bu was no associa ed wi h any bi h ou comes in Ghana. Supplemen al Table 1 p o ides a nonnume ic isual o e iew o he associa ions p esen ed in Table 3 and unadjus ed esul s a e p esen ed in Supplemen al Table 2.The e we e no signi ican in e ac ions o Hb concen a ion o i on s a us a en ollmen wi h ma e nal age, pa i y, o supplemen g oup. Anemia (Supplemen al Table 3), i on de iciency (Table 4), and IDA (Table 5) a en ollmen we e all associa ed wi h an inc eased isk o PTB in Malawi; i on de iciency and IDA we e also associa ed wi h inc eased isk o newbo n s un ing. In Ghana, anemia was no associa ed wi h ad e se bi h ou comes; howe e , IDA was associa ed wi h a dec eased isk o LBW (P=0.046) and i on de iciency showed a simila end (P=0.07). High Hb (>130 g/L) was no associa ed wi h ad e se bi h ou comes in ei he Ghana o Malawi (Supplemen al Table 4), al hough eple e i on s a us (sT R <10 h pe cen ile) was associa ed wi h an inc eased isk o newbo n s un ing in Ghana (Table 6). Unadjus ed esul s o anemia, i on de iciency, IDA, high Hb, and i on- eple e s a us a e p esen ed in Supplemen al Tables 5–9. In unadjus ed analyses, he e was a signi ican ly highe isk o LBW o women wi h IDA in Malawi ha was a enua ed and became nonsigni ican in adjus ed esul s. All o he unadjus ed esul s we e simila o adjus ed esul s. Hb and i on s a us a 36 weeks o ges a ion—associa ions wi h bi h ou comes A 36 weeks o ges a ion, a U-shaped ela ion be ween Hb concen a ion and p egnancy du a ion was e iden in Ghana, wi h bo h low and high Hb concen a ions associa ed wi h a sho e du a ion o ges a ion (Figu e 1). Hb concen a ion was no associa ed wi h any o he bi h ou comes. Highe sT R (indica o o lowe i on s a us) was associa ed wi h a highe bi h weigh , LAZ, and HCZ in Ghana bu was no associa ed wi h any bi h ou comes in Malawi. Highe ZPP (indica o o lowe i on s a us) was associa ed wi h a g ea e bi h weigh and LAZ in Ghana bu was no associa ed wi h any bi h ou comes in Malawi (Table 3). Supplemen al Table 1 p o ides a nonnume ic isual o e iew o he associa ions p esen ed in Table 3. Unadjus ed esul s we e simila o adjus ed esul s and a e p esen ed in Supplemen al Table 2. The e we e no signi ican P ena al i on s a us and bi h ou comes 517 Downloaded om h ps://academic.oup.com/jn/a icle-abs ac /149/3/513/5281546 by Raila Melin use on 23 Ap il 2019 TABLE 5 Risk o ad e se bi h ou comes o women wi h IDA du ing ea ly o la e p egnancy1 Wi hou IDA, n/ o al n(%)2Wi h IDA, n/ o al n(%)3Adjus ed RR (95% CI)4P Ea ly p egnancy: ≤20 wk PTB Ghana 92/1089 (8.5) 5/48 (10.4) 1.13 (0.52, 2.47) 0.75 Malawi 104/1181 (8.8) 25/106 (23.6) 2.23 (1.48, 3.37) <0.001 LBW Ghana 132/1088 (12.1) 1/48 (2.1) 0.14 (0.02, 0.97) 0.046 Malawi 129/1051 (12.3) 17/86 (19.8) 1.45 (0.94, 2.25) 0.10 SGA Ghana 234/1051 (21.3) 6/48 (12.5) 0.55 (0.26, 1.17) 0.12 Malawi 306/1051 (29.1) 26/83 (31.3) 1.38 (0.99, 1.91) 0.057 Newbo n s un ing Ghana 101/1083 (9.3) 1/48 (2.1) 0.21 (0.03, 1.45) 0.11 Malawi 147/997 (14.7) 25/85 (29.4) 1.85 (1.30, 2.64) <0.001 La e p egnancy: 36 wk LBW Ghana 81/963 (8.4) 2/27 (7.4) 0.57 (0.18, 1.80) 0.34 Malawi 86/837 (10.3) 10/102 (9.8) 0.85 (0.46, 1.57) 0.60 SGA Ghana 217/941 (23.1) 5/27 (18.5) 0.57 (0.22, 1.43) 0.23 Malawi 251/837 (30.0) 25/101 (24.8) 0.98 (0.70, 1.38) 0.92 Newbo n s un ing Ghana 57/960 (5.9) 2/27 (7.4) 0.73 (0.18, 2.94) 0.66 Malawi 105/810 (13.0) 16/95 (16.8) 1.19 (0.76, 1.86) 0.45 1PTB: <37 weeks o ges a ion; LBW: <2.5 kg; SGA: bi h weigh <10 h pe cen ile by ges a ional age and sex using he INTERGROWTH-21s s anda d (30); s un ing: leng h- o -age zsco e <−2. Hb, hemoglobin; IDA, i on de iciency anemia; LBW, low bi h weigh ; PTB, p e e m bi h; SGA, small- o -ges a ional-age; sT R, soluble ans e in ecep o . 2Womenwi hHb≥100 g/L and sT R ≤6.0 mg/L. 3De ined as Hb <100 g/L and sT R >6.0 mg/L. 4Adjus ed models included he ollowing co a ia es i signi ican ly (P<0.1) associa ed wi h he ou come: ges a ional age a en ollmen , pa i y, ma e nal age, educa ion le el, household ood insecu i y, household asse index, α-1-acid glycop o ein a he ime he blood sample was d awn, C- eac i e p o ein a he ime he blood sample was d awn, in an sex, ma e nal BMI a en ollmen , ma e nal mala ia a en ollmen , and HIV s a us (Malawi models only). All 36-wk models adjus ed o in e en ion g oup. Speci ics ega ding each adjus ed model a e p o ided in he Supplemen al Me hods. in e ac ions o Hb concen a ion o i on s a us a 36 wk wi h ma e nal age, pa i y, o supplemen g oup. Anemia (Supplemen al Table 3), i on de iciency (Table 4), and IDA (Table 5) a 36 wk we e no associa ed wi h any ad e se bi h ou comes in Ghana o Malawi. The e we e also no associa ions in ei he coun y be ween high Hb and ad e se bi h ou comes (Supplemen al Table 4). Reple e i on s a us was no associa ed wi h ad e se bi h ou comes in Malawi. In Ghana, eple e i on s a us was associa ed wi h an inc eased isk o LBW, SGA, and newbo n s un ing (Table 6). Unadjus ed esul s o anemia, i on de iciency, IDA, high Hb, and i on- eple e s a us a e p esen ed in Supplemen al Tables 5–9. All unadjus ed esul s we e simila o adjus ed esul s. Discussion This s udy examined associa ions o ma e nal Hb concen a ion and i on s a us wi h bi h ou comes in 2 coho s o p egnan women in A ica. In Malawi bu no in Ghana, i on de iciency and IDA in ea ly p egnancy we e ela ed o PTB and newbo n s un ing, and anemia in ea ly p egnancy was also ela ed o PTB. By con as , in Ghana i on- eple e s a us in ea ly p egnancy was associa ed wi h newbo n s un ing, and i on de iciency and IDA in ea ly p egnancy we e ela ed o lowe isk o LBW. Anemia, i on de iciency, and IDA in la e p egnancy we e no signi ican ly ela ed o bi h ou comes in ei he coun y, al hough eple e i on s a us in la e p egnancy was ela ed o highe isk o LBW, SGA, and newbo n s un ing in Ghana. In popula ions a isk o anemia o i on de iciency, highe ma e nal Hb concen a ions and i on s a us ha e o en been associa ed wi h be e bi h ou comes (34,35). Highe Hb o highe i on s a us may imp o e he sys emic esponse o in lamma ion and in ec ion, educe he s ess esponse om ch onic hypoxia, and lowe oxida i e s ess ia less e y h ocy e oxida ion (36,37). Howe e , p e ious s udies did no dis inguish anemia o i on de iciency by iming in p egnancy (i.e., ea ly compa ed wi h la e p egnancy). Associa ions wi h bioma ke s in la e p egnancy a e complica ed by he issue o plasma olume expansion (PVE). Ma e nal plasma olume expands du ing no mal p egnancy, esul ing in lowe concen a- ions o Hb and o he bioma ke s. In i on- eple e popula ions, highe Hb concen a ion and i on s a us ha e been associa ed wi h ad e se bi h ou comes (15); howe e , inadequa e PVE is also associa ed wi h ad e se bi h ou comes, hus making i di icul o in e p e associa ions o ad e se bi h ou comes wi h high Hb o e i in (a ma ke o i on s o age). In Ghana and Malawi, we measu ed sT R and ZPP concen a ions as bioma ke s o i on s a us. Fe i in was no included because i s u ili y is limi ed du ing pe iods when he e is li le i on s o age because o high equi emen s, such as he second and hi d imes e s o p egnancy. Du ing such imes, i on s o age may be low bu i on concen a ions o physiological needs may s ill be adequa e and may be mo e accu a ely assessed using o he ma ke s (38). Unlike Hb o 518 Oaks e al. Downloaded om h ps://academic.oup.com/jn/a icle-abs ac /149/3/513/5281546 by Raila Melin use on 23 Ap il 2019 TABLE 6 Risk o ad e se bi h ou comes o women wi h i on- eple e s a us du ing ea ly o la e p egnancy1 Re e ence g oup (sT R ≥10 h pe cen ile), n/ o al n(%)2 Wi h i on- eple e s a us (sT R <10 h pe cen ile), n/ o al n(%)3Adjus ed RR (95% CI)4P Ea ly p egnancy: ≤20 wk PTB Ghana 77/924 (8.3) 10/111 (9.0) 0.88 (0.44, 1.76) 0.71 Malawi 82/912 (9.0) 6/128 (4.7) 0.66 (0.29, 1.48) 0.31 LBW Ghana 110/923 (11.9) 16/111 (14.4) 1.06 (0.66, 1.70) 0.81 Malawi 100/813 (12.3) 11/115 (9.6) 0.81 (0.46, 1.45) 0.80 SGA Ghana 194/889 (21.8) 27/110 (24.6) 1.09 (0.76, 1.55) 0.64 Malawi 195/788 (24.7) 24/111 (21.6) 0.82 (0.57, 1.19) 0.30 Newbo n s un ing Ghana 78/918 (8.5) 17/111 (15.3) 1.71 (1.06, 2.77) 0.03 Malawi 114/762 (15.0) 10/116 (8.6) 0.60 (0.33, 1.11) 0.10 La e p egnancy: 36 wk LBW Ghana 58/753 (7.7) 16/99 (16.2) 1.90 (1.17, 3.09) 0.01 Malawi 53/560 (9.5) 14/99 (14.1) 1.39 (0.81, 2.39) 0.23 SGA Ghana 161/735 (21.9) 34/98 (34.7) 1.51 (1.12, 2.05) 0.01 Malawi 137/546 (25.1) 23/98 (23.5) 0.92 (0.62, 1.36) 0.67 Newbo n s un ing Ghana 42/750 (5.6) 13/99 (13.1) 2.14 (1.21, 3.77) 0.01 Malawi 76/536 (14.2) 12/101 (11.9) 0.73 (0.42, 1.26) 0.26 1PTB: <37 weeks o ges a ion; LBW: <2.5 kg; SGA: bi h weigh <10 h pe cen ile by ges a ional age and sex using he INTERGROWTH-21s s anda d (30); s un ing: leng h- o -age zsco e <−2. LBW, low bi h weigh ; PTB, p e e m bi h; SGA, small- o -ges a ional-age; sT R, soluble ans e in ecep o . 2The e e ence g oup excluded women wi h i on de iciency (sT R >6.0 mg/L). 3sT R <10 h pe cen ile; a ≤20 wk, his was <2.49 mg/L o Ghana and <2.65 mg/L o Malawi. A 36 wk, his was <2.86 mg/L o Ghana and <3.08 mg/L o Malawi. 4Adjus ed models included he ollowing co a ia es i signi ican ly (P<0.1) associa ed wi h he ou come: ges a ional age a en ollmen , pa i y, ma e nal age, educa ion le el, household ood insecu i y, household asse index, α-1-acid glycop o ein a he ime he blood sample was d awn, C- eac i e p o ein a he ime he blood sample was d awn, in an sex, ma e nal BMI a en ollmen , ma e nal mala ia a en ollmen , and HIV s a us (Malawi models only). All 36-wk models adjus ed o in e en ion g oup. Speci ics ega ding each adjus ed model a e p o ided in he Supplemen al Me hods. e i in, lowe sT R and ZPP concen a ions indica e highe i on s a us. We did no measu e PVE in ou coho s and i is possible ha some women we e expe iencing inadequa e PVE, which would ha e concen a ed he bioma ke s. As we did no ha e any signi ican associa ions wi h Hb >130 g/dL, and PVE would ha e biased he associa ions o sT R and ZPP owa ds he null, we ind i unlikely ha inadequa e PVE explains he associa ions be ween highe i on s a us (lowe sT R and ZPP concen a ions) and poo e bi h ou comes in Ghana. Lowe sT R concen a ion can be a sign o impai ed RBC p oduc ion (which is linked o inadequa e PVE) (39) and possibly would ha e biased he esul s away om he null. Howe e , we ind i unlikely ha impai ed RBC p oduc ion would occu mo e o en in he Ghanaian coho han in he Malawian coho . One po en ial biological mechanism ha could explain why highe i on s a us may be de imen al o he e us is non– ans e in bound i on. I on is ypically ca e ully chape oned a ound he body, p edomina ely by ans e in. Unbound i on can esul when he a e o i on in lux in o plasma exceeds he a e o i on acquisi ion by ans e in (40). I is he e o e possible ha highe i on s a us may lead o oxida i e s ess ia unbound i on, which may esul in lipid pe oxida ion and DNA damage o placen al cells (41,42) and impai he sys emic esponse o in ec ion (36), comp omising he g ow h o he e us. The e is some e idence ha a modes inc ease in plasma non– ans e in bound i on can occu a e i on supplemen a ion in nonp egnan women (40), al hough i is unclea whe he his is ele an in ou s udy popula ion. I is unclea why IDA a en ollmen in Ghana was associa ed wi h a lowe isk o LBW, wi h i on de iciency a en ollmen showing a simila nonsigni ican end. Associa ions be ween i on s a us and bi h ou comes may ha e di e ed by coun y owing o di e ences be ween Ghanaian and Malawian women in i on s a us a en ollmen . Ghanaian women had highe i on s a us (mean sT R: 4.1 compa ed wi h 5.6 mg/L; mean ZPP: 44.8 compa ed wi h 54.5 μmol/mol heme) han Malawian women a en ollmen , and a lowe p opo ion o Ghanaian women we e iden i ied as i on de icien (9.3% compa ed wi h 19.7%, sT R >6mg/L). All women in hese 2 coho s ecei ed i on supplemen a ion, wi h app oxima ely wo- hi ds ecei ing 20 mg and one- hi d ecei ing 60 mg Fe/d. Thus he e was a la ge numbe o i on- eple e women in Ghana han in Malawi who ecei ed an amoun o i on supplemen a ion ha may ha e been unnecessa y, and plausibly had nega i e e ec s on he e us. We p e iously demons a ed ha bi h weigh was highe among in an s bo n o women ecei ing SQ-LNSs (which con ained 20 mg Fe) han among in an s o women ecei ing i on and olic acid capsules (which con ained 60 mg Fe), e en hough he P ena al i on s a us and bi h ou comes 519 Downloaded om h ps://academic.oup.com/jn/a icle-abs ac /149/3/513/5281546 by Raila Melin use on 23 Ap il 2019 FIGURE 1 Associa ion be ween p egnancy du a ion and ma e nal Hb concen a ion measu ed a 36 weeks o ges a ion in Ghanaian p egnan women (n=985). Hb, hemoglobin. i on and olic acid g oup had highe mean Hb concen a ion, highe i on s a us, and a lowe p e alence o anemia a 36 weeks o ges a ion (18,22). I is possible ha he e a e o he di e ences be ween he coun ies ha migh explain he di e en associa ions be ween i on s a us and bi h ou comes. S eng hs o his s udy include he use o ul asound o es ima e ges a ional age a en ollmen ; he a ailabili y o a la ge numbe o co a ia es o es and con ol o con ounding, including 2 ma ke s o in lamma ion; and analysis o 2 coho s om di e en egions in A ica whe e simila s udy me hods we e used. Including ma ke s o in lamma ion is impo an , because in lamma ion can lead o ele a ed sT R concen a ions (43). In addi ion, we measu ed i on s a us du ing bo h ea ly and la e p egnancy. Concen a ions o sT R end o be simila o nonp egnan concen a ions du ing he i s imes e , g adually inc ease du ing he second imes e , and peak in he hi d imes e (44). We a e limi ed in he in e p e a ion o ou esul s, because i is unclea whe he low sT R can be used as an indica o o i on- eple e s a us. The e is e idence ha sT R is lowe du ing impai ed e y h opoiesis (39), which complica es he use o i as a ma ke o i on s a us. Fu he esea ch o iden i y an accu a e ma ke o i on- eple e s a us du ing p egnancy is u gen ly needed. O he limi a ions include a delay in bi h an h opome y o some in an s, al hough back-calcula ions we e pe o med acco ding o WHO guidelines in such cases, as well as a possible limi ed gene alizabili y o he indings in Malawi due o di e ences in some cha ac e is ics o included compa ed wi h excluded pa icipan s. Howe e , di e ences be ween included and excluded pa icipan s we e minimal. We es ed mul iple hypo heses bu did no pe o m a s a is ical co ec ion o mul iple hypo hesis es ing because he bi h ou comes a e closely ela ed o each o he (45). The e o e, he e is an inc eased isk ha some indings could be due o chance. Measu emen s a 36 weeks o ges a ion do no include women who expe ienced a misca iage o ga e bi h be o e 36 weeks, he e o e su i o bias may a ec he in e p e a ion o associa ions wi h hese measu emen s. In conclusion, his esea ch p o ides e idence ha he associa ions be ween low o eple e ma e nal i on s a us and bi h ou comes a e popula ion speci ic. Fu u e esea ch o eplica e and ex end hese indings would be bene icial. Acknowledgmen s We hank Jan Pee son, Rebecca Young, and Cha les A nold o s a is ical ad ice; he In e na ional Lipid-Based Nu ien Supplemen s P ojec S ee ing Commi ee (www.ilins.o g) o p o iding leade ship o he s udies; Se i Shahab-Fe dows a he Wes e n Human Nu i ion Resea ch Cen e o labo a o y assis ance; and Ma y A imond o suppo wi h p ojec managemen . The au ho s’ con ibu ions we e as ollows—AL, KGD, KM, LHA, PA, SA-A, SV, and UA: designed he esea ch; SA-A, AL, HO, JS, and KM: conduc ed he esea ch; LMB: pe o med labo a o y analysis; BMO and JMJ: pe o med he s a is ical analysis; PA and KGD: ad ised on he analysis; BMO, JMJ, and KGD: w o e he manusc ip ; and all au ho s: e iewed he d a manusc ip and ead and app o ed he inal manusc ip . Re e ences 1. Wo ld Heal h O ganiza ion. Guideline: Daily I on and Folic Acid Supplemen a ion in P egnan Women. Gene a: Wo ld Heal h O ganiza ion; 2012. 2. Peña-Rosas J, De-Regil L, Ga cia-Casal M, Dowswell T. Daily o al i on supplemen a ion du ing p egnancy. Coch ane Da abase Sys Re 2015;(7):CD004736. 3. B abin L, B abin BJ, Gies S. In luence o i on s a us on isk o ma e nal o neona al in ec ion and on neona al mo ali y wi h an emphasis on de eloping coun ies. Nu Re 2013;71:528–40. 4. S ee P, Alam MA, Wadswo h J, Welch A. Rela ion be ween ma e nal haemoglobin concen a ion and bi h weigh in di e en e hnic g oups. BMJ 1995;310:489–91. 5. Ga n SM, Kea ing MT, Falkne F. Hema ological s a us and p egnancy ou comes. Am J Clin Nu 1981;34:115–17. 6. Gonzales GF, S eenland K, Tapia V. Ma e nal hemoglobin le el and e al ou come a low and high al i udes. Am J Physiol Regul In eg Comp Physiol 2009;297:1477. 7. Chang SC, O’B ien KO, Na hanson MS, Mancini J, Wi e FR. Hemoglobin concen a ions in luence bi h ou comes in p egnan A ican-Ame ican adolescen s. J Nu 2003;133:2348–55. 8. Alwan NA, Cade JE, McA dle HJ, G eenwood DC, Hayes HE, Simpson NA. Ma e nal i on s a us in ea ly p egnancy and bi h ou comes: insigh s om he Baby’s Vascula Heal h and I on in P egnancy s udy. B J Nu 2015;113:1985–92. 9. Scholl TO, Hedige ML, Fische RL, Shea e JW. Anemia s i on de iciency: inc eased isk o p e e m deli e y in a p ospec i e s udy. Am J Clin Nu 1992;55:985–8. 10. Khambalia AZ, Collins CE, Robe s CL, Mo is JM, Powell KL, Tase ski V, Nassa N. High ma e nal se um e i in in ea ly p egnancy and isk o spon aneous p e e m bi h. B J Nu 2015;114: 455–61. 11. Goldenbe g RL, Tamu a T, DuBa d M, Johns on KE, Coppe RL, Negge s Y. Plasma e i in and p egnancy ou come. Obs e Gynecol 1996;175:1356–9. 12. Tamu a T, Goldenbe g RL, Johns on KE, Cli e SP, Hickey CA. Se um e i in: a p edic o o ea ly spon aneous p e e m deli e y. Obs e Gynecol 1996;87:360–5. 13. Lao TT, Tam KF, Chan LY. Thi d imes e i on s a us and p egnancy ou come in non-anaemic women; p egnancy un a ou ably a ec ed by ma e nal i on excess. Hum Rep od 2000;15:1843–8. 14. Scholl TO. High hi d- imes e e i in concen a ion: associa ions wi h e y p e e m deli e y, in ec ion, and ma e nal nu i ional s a us. Obs e Gynecol 1998;92:161–6. 15. Dewey KG, Oaks BM. U-shaped cu e o isk associa ed wi h ma e nal hemoglobin, i on s a us, o i on supplemen a ion. Am J Clin Nu 2017;106:1702S. 16. Ins i u e o Medicine, Subcommi ee on Nu i ional S a us and Weigh Gain Du ing P egnancy. Nu i ion Du ing P egnancy. Washing on, DC: Na ional Academy P ess; 1990. 17. Klebano MA, Shiono PH, Be endes HW, Rhoads GG. Fac s and a i ac s abou anemia and p e e m deli e y. JAMA 1989;262:511–5. 520 Oaks e al. Downloaded om h ps://academic.oup.com/jn/a icle-abs ac /149/3/513/5281546 by Raila Melin use on 23 Ap il 2019 18. Adu-A a wuah S, La ey A, Ok onipa H, Asho n P, Zeilani M, Pee son JM, A imond M, Vos i S, Dewey KG. Lipid-based nu ien supplemen inc eases he bi h size o in an s o p imipa ous women in Ghana. Am J Clin Nu 2015;101:835–46. 19. Asho n P, Alho L, Asho n U, Cheung YB, Dewey KG, Ha junmaa U, La ey A, Nkhoma M, Phi i N, Phuka J, e al. The impac o lipid- based nu ien supplemen p o ision o p egnan women on newbo n size in u al Malawi: a andomized con olled ial. Am J Clin Nu 2015;101:387–97. 20. Asho n P, Alho L, Asho n U, Cheung YB, Dewey KG, Gondwe A, Ha junmaa U, La ey A, Phi i N, Phi i TE, e al. Supplemen a ion o ma e nal die s du ing p egnancy and o 6 mon hs pos pa um and in an die s he ea e wi h small-quan i y lipid-based nu ien supplemen s does no p omo e child g ow h by 18 mon hs o age in u al Malawi: a andomized con olled ial. J Nu 2015;145:1345– 53. 21. Dale JC, Bu i MF, Zinsmeis e AR. Diu nal a ia ion o se um i on, i on-binding capaci y, ans e in sa u a ion, and e i in le els. Am J Clin Pa hol 2002;117:802–8. 22. Adu-A a wuah S, La ey A, Ok onipa H, Asho n P, Zeilani M, Baldi iez LM, Oaks BM, Vos i S, Dewey KG. Impac o small-quan i y lipid- based nu ien supplemen on hemoglobin, i on s a us and bioma ke s o in lamma ion in p egnan Ghanaian women. Ma e n Child Nu 2017;13(2):e12262. 23. Jo gensen JM, Asho n P, Asho n U, Baldi iez LM, Gondwe A, Male a K, Nkhoma M, Dewey KG. E ec s o lipid-based nu ien supplemen s o mul iple mic onu ien supplemen s compa ed wi h i on and olic acid supplemen s du ing p egnancy on ma e nal hemoglobin and i on s a us. Ma e n Child Nu 2018;14(4):e12640. 24. Wo ld Heal h O ganiza ion Mul icen e G ow h Re e ence S udy G oup. WHO Child G ow h S anda ds: Leng h/heigh - o -age, Weigh - o -age, Weigh - o -leng h, Weigh - o -heigh and Body Mass Index- o - age: Me hods and De elopmen . Gene a: Wo ld Heal h O ganiza ion; 2006. 25. Johnson-Spea M, Yip R. Hemoglobin di e ence be ween black and whi e women wi h compa able i on s a us: jus i ica ion o ace-speci ic anemia c i e ia. Am J Clin Nu 1994;60:117–21. 26. Cao C, O’B ien KO. P egnancy and i on homeos asis: an upda e. Nu Re 2013;71:35–51. 27. P ei e CM, Cook JD, Mei ZG, Cogswell ME, Looke AC, Lache DA. E alua ion o an au oma ed soluble ans e in ecep o (sT R) assay on he Roche Hi achi analyze and i s compa ison o wo ELISA assays. Clinica Chimica Ac a 2007;382:112–6. 28. Vande ij e e S, Amsalkhi S, Van Oyen H, Egli I, Mo eno-Reyes R. I on s a us and i s de e minan s in a na ionally ep esen a i e sample o p egnan women. J Acad Nu Die 2013;113:659–66. 29. Khumalo H, Gomo ZAR, Moyo VM, Go deuk VR, Saungweme T, Rouaul TA, Gangaidzo IT. Se um ans e in ecep o s a e dec eased in he p esence o i on o e load. Clin Chem 1998;44:40–4. 30. Villa J, Ismail LC, Vic o a CG, Ohuma EO, Be ino E, Al man DG, Lambe A, Papageo ghiou AT, Ca alho M, Ja e YA, e al. In e na ional s anda ds o newbo n weigh , leng h, and head ci cum e ence by ges a ional age and sex: he Newbo n C oss- Sec ional S udy o he INTERGROWTH-21s P ojec . Lance 2014;384: 857–68. 31. Villa J, Giuliani F, Fen on TR, Ohuma EO, Ismail LC, Kennedy SH. INTERGROWTH-21s e y p e e m size a bi h e e ence cha s. Lance 2016;387:844–5. 32. Coa es J, Swindale A, Bilinsky P. Household Food Insecu i y Access Scale (HFIAS) o Measu emen o Food Access: Indica o Guide (V.3) [In e ne ]. Washing on, DC: Food and Nu i ion Technical Assis ance P ojec , Academy o Educa ional De elopmen ; 2007. A ailable om: h p://www. ao.o g/ ileadmin/use _upload/eu ao- si4dm/doc- a ining/h ias.pd . 33. Vyas S, Kuma anayake L. Cons uc ing socio-economic s a us indices: how o use p incipal componen s analysis. Heal h Policy Plan 2006;21:459–68. 34. Rahman MM, Abe SK, Rahman MS, Kanda M, Na i a S, Bilano V, O a E, Gilmou S, Shibuya K. Ma e nal anemia and isk o ad e se bi h and heal h ou comes in low- and middle-income coun ies: sys ema ic e iew and me a-analysis. Am J Clin Nu 2016;103: 495–504. 35. Scholl TO. Ma e nal i on s a us: ela ion o e al g ow h, leng h o ges a ion, and i on endowmen o he neona e. Nu Re 2014;69:S23– 9. 36. D akesmi h H, P en ice AM. Hepcidin and he i on-in ec ion axis. Science 2012;338:768–72. 37. Allen LH. Biological mechanisms ha migh unde lie i on’s e ec s on e al g ow h and p e e m bi h. J Nu 2001;131:581S–9S. 38. Lynch S. Case s udies: i on. Am J Clin Nu 2011;94:673S–8S. 39. Beguin Y. Soluble ans e in ecep o o he e alua ion o e y h opoiesis and i on s a us. Clin Chim Ac a 2003;329:9–22. 40. B i enham GM, Ande sson M, Egli I, Foman JT, Zede C, Wes e man ME, Hu ell RF. Ci cula ing non- ans e in-bound i on a e o al adminis a ion o supplemen al and o i ica ion doses o i on o heal hy women: a andomized s udy. Am J Clin Nu 2014;100:813–20. 41. Keh e JP. The Habe -Weiss eac ion and mechanisms o oxici y. Toxicology 2000;149:43–50. 42. Casanue a E, Vi e i FE. I on and oxida i e s ess in p egnancy. J Nu 2003;133:1700s–8s. 43. Rohne F, Namas e SM, La son LM, Addo OY, Mei Z, Suchde PS, Williams AM, Sak Ashou FA, Rawa R, Rai en DJ, e al. Adjus ing soluble ans e in ecep o concen a ions o in lamma ion: Bioma ke s Re lec ing In lamma ion and Nu i ional De e minan s o Anemia (BRINDA) p ojec . Am J Clin Nu 2017;106: 372S–82S. 44. Choi JW, Im MW, Pai SH. Se um ans e in ecep o concen a ions du ing no mal p egnancy. Clin Chem 2000;46:725–7. 45. S eine DL. Bes (bu o - o go en) p ac ices: he mul iple p oblems o mul iplici y—whe he and how o co ec o many s a is ical es s. Am J Clin Nu 2015;102(4):721–8. P ena al i on s a us and bi h ou comes 521 Downloaded om h ps://academic.oup.com/jn/a icle-abs ac /149/3/513/5281546 by Raila Melin use on 23 Ap il 2019