Memoirs

Dietary Iron and Risk of Myocardial Infarction in the Rotterdam Study

Description
American Jurnal f Epiemilgy Cpyright 999 by The Jhns Hpkins University Schl f Hygiene an Public Health All rights reserve Vl. 49, N. 5 Printe in U.S.A. Dietary rn an Risk f Mycarial nfarctin in the Rtteram
Categories
Published
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
American Jurnal f Epiemilgy Cpyright 999 by The Jhns Hpkins University Schl f Hygiene an Public Health All rights reserve Vl. 49, N. 5 Printe in U.S.A. Dietary rn an Risk f Mycarial nfarctin in the Rtteram Stuy Kerstin Klipstein-Grbusch, 2 Dieerick E. Grbbee, 3 Jhanna H. en Breeijen, 4 Heiner Being, 2 Albert Hfman, an Jacqueline C. M. Witteman Free irn has been implicate in lipi perxiatin an ischemic mycarial amage, an it has been suggeste that irn is an inepenent risk factr fr mycarial infarctin. The authrs investigate whether ietary irn is assciate with an increase risk f fatal an nnfatal mycarial infarctin in the Rtteram Stuy, a cmmunity-base prspective chrt stuy f 7,983 elerly subjects in Rtteram, the Netherlans. The stuy sample cnsiste f 4,802 participants wh at baseline ha n knwn histry f mycarial infarctin an fr whm ietary ata were available. Frm 990 t 996, 24 subjects ha a mycarial infarctin. N assciatin was bserve between ttal irn intake an risk f mycarial infarctin after ajustment fr age an sex (relative risk fr the highest vs. the lwest tertile f intake = 0.89, 95% cnfience interval (Cl) , p fr tren = 0.640). Heme irn intake was psitively assciate with risk f mycarial infarctin (relative risk fr the highest vs. the lwest tertile f intake =.83, 95% Cl.6-2.9, pfr tren = 0.008) after ajustment fr age an sex, an this assciatin persiste after multivariate ajustment (relative risk =.86, 95% Cl , p fr tren = 0.00). A istinctin between fatal an nnfatal cases f mycarial infarctin inicate that the assciatin f heme irn with mycarial infarctin was mre prnunce in fatal cases. The results suggest that a high ietary heme irn intake is relate t an increase risk f mycarial infarctin an that it may specifically affect the rate f fatality frm mycarial infarctin. Am J Epiemil 999;49:42-8. crnary isease; elerly; irn, ietary; mycarial infarctin Because f the imprtance f xiize lw ensity lipprtein (LDL) in athersclersis (, 2), the ptential fr irn t act as a catalyst in bilgic xiative amage (3, 4), an eclgic bservatins f crnary artery isease an irn status (5), a ptential rle f irn in crnary heart isease seems plausible. S far, the strngest supprting evience that irn is a risk factr fr crnary heart isease stems frm a chrt stuy f eastern Finnish men (6, 7), in which high levels f serum ferritin an ietary irn were psitively assciate with the incience f mycarial infarctin. Mst subsequent prspective stuies investigating whether irn status r ietary irn intake is assciate Receive fr publicatin December 8, 997, an accepte fr publicatin July 23, 998. Abbreviatins: Cl, cnfience interval; LDL, lw ensity lipprtein; SFFQ, semiquantitative f frequency questinnaire. Department f Epiemilgy an Bistatistics, Erasmus University Meical Schl, Rtteram, the Netherlans. 2 Department f Epiemilgy, German nstitute f Human Nutritin, Ptsam-Rehbrucke, Germany. 3 Julius Center fr Patient riente Research, Utrecht University Hspital, Utrecht, the Netherlans. 4 TN Nutritin an F Research nstitute, Zeist, the Netherlans. Reprint requests t Dr. J. C. M. Witteman, Department f Epiemilgy an Bistatistics, Erasmus University Meical Schl, P Bx 738, 3000 DR Rtteram, the Netherlans. with an increase risk f mycarial infarctin r crnary heart isease cul nt supprt the hypthesis that high by irn stres increase the risk f crnary heart isease (8-3), althugh Ascheri et al. (4) reprte an increase risk f nnfatal mycarial infarctin r fatal crnary heart isease with heme irn intake. Hwever, measures f irn status such as serum irn r transferrin saturatin as applie in mst f these stuies generally are nt cnsiere t accurately reflect by irn stres (5). Mst stuies i nt iscriminate between nnfatal an fatal mycarial infarctin, althugh irn may be irectly invlve in the mycarial injury cause by ischemia an reperfusin. nirect evience that irn is invlve in reperfusin injury after an ischemic event has been prvie by results f animal experiments shwing that free raicals are generate after bl flw is restre t ischemic mycarium (6), cntributing t the subsequent mycarial injury (7). Experiments with an irn chelatr (8, 9) r irn verlaing (20) in animals emphasize this ntin. t has subsequently been suggeste that elevate irn stres may aversely affect survival after a mycarial infarctin (2). n the current stuy, we investigate whether ietary irn an heme irn are assciate with the incience an case Dwnlae frm at Pennsylvania State University n September 8, 422 Klipstein-Grbusch et al. fatality f mycarial infarctin in the Rtteram Stuy chrt. MATERALS AND METHDS Stuy ppulatin The Rtteram Stuy is a cmmunity-base prspective chrt stuy f 7,983 persns (respnse rate, 78 percent) age 55 years r ler living in mmr, an urban istrict in Rtteram, the Netherlans. The aim f the stuy is t investigate the incience f an risk factrs fr chrnic an isabling carivascular, neuregenerative, lcmtr, an phthalmic iseases, as escribe elsewhere (22). The stuy was apprve by the Meical Ethics Cmmittee f Erasmus University in Rtteram, an written infrme cnsent was btaine frm all participants. Fllw-up fr crnary heart isease mrtality starte after the baseline survey in 990. Fllw-up infrmatin until April 996 was available fr the present analysis. f the 5,59 subjects fr whm ietary ata were available, 73 ha a mycarial infarctin uring the fllw-up peri. Because f pssible changes in their ietary patterns, subjects wh ha a histry f mycarial infarctin at baseline (n - 357) were exclue frm analysis. This resulte in 4,802 subjects fr the current analysis. f these subjects, 24 ha a first fatal r nnfatal mycarial infarctin. Case ascertainment The fllw-up peri cmprise 3-7 years (mean, 4). nfrmatin n the vital status f participants was btaine at regular intervals frm the municipal health service in Rtteram, an infrmatin n fatal an nnfatal enpints was prvie by the general practitiners wrking in the stuy istrict f mmr. All pssible events reprte by the general practitiners were verifie by research physicians frm the Rtteram Stuy by using the recrs f participating general practitiners an meical specialists. The causes an circumstances f eaths were btaine by using a questinnaire fr general practitiners an by scrutinizing infrmatin frm hspital ischarge recrs, in cases f amittance r referral, shrtly after a eath was reprte by the municipal health service r the general practitiner. verall, fllw-up infrmatin was available fr 94 percent f the chrt. Classificatin f fatal an nnfatal events was base n the nternatinal Statistical Classificatin f Diseases an Relate Health Prblems, Tenth Revisin (23). Fr the present analysis, cases f first nnfatal mycarial infarctin r fatal mycarial infarctin (ces 2-24) were selecte. All events were classifie inepenently by tw research physicians. f they isagree, a cnsensus was reache in a special sessin. Finally, all events were verifie by a carivascular isease expert. n case f iscrepancies, this persn's ecisin was cnsiere final. Measurements nfrmatin n participants' current health status, meical histry, rug use, eucatin, incme, an smking behavir was btaine by using a cmputerize questinnaire uring a hme interview. Height an weight were measure, an by mass inex (weight in kg ivie by height in m 2 ) was calculate as a measure f besity. Sitting bl pressure was measure n the right upper arm by using a ranmzer sphygmmanmeter. The average f tw measurements was use in the analysis. Hypertensin was efine as a systlic bl pressure f at least 40 mmhg an/r a iastlic bl pressure f at least 90 mmhg an/r current use f antihypertensive meicatin. A venepuncture was perfrme, an serum ttal an high ensity lipprtein chlesterl cncentratins were etermine by using an autmate enzymatic prceure. f serum chlesterl levels were 6.5 mml/liter r higher, subjects were cnsiere hyperchlesterlemic. All participants except thse using antiiabetic meicatin were given a 37.5 percent ral glucse slutin (75 g f glucse) while they were in a nnfasting state. Venus glucse levels were measure befre an after 2 hurs. Nn-insulin-epenent iabetes was iagnse if at least ne measurement was at least. mml/liter r if the participant use antiiabetic meicatin. Dietary assessment A semiquantitative f frequency questinnaire (SFFQ) that participants cmplete uring the baseline survey aime at assessing their habitual f intake uring the past year. The questinnaire inclue 70 f items in 3 f grups an questins abut ietary habits, supplementatin, an prescribe iets. SFFQ ata were cnverte t nutrient intake by using the cmputerize Dutch f cmpsitin table (24). Heme irn intake was estimate t be equal t 40 percent f the irn fun in meat, fish, an pultry (25). Data n P-cartene an tcpherl were upate by using an aitinal atabase frm the Netherlans nstitute f Public Health an Envirnmental Prtectin (Y. C. J. Vllebregt, EJM Feskens, unpublishe bservatins, 993). ntake thrugh nutritinal supplements was nt cnsiere, since infrmatin frm bran labels was nt recre with sufficient accuracy. The valiity f the SFFQ was assesse in a Am J Epiemil Vl. 49, N. 5, 999 Dwnlae frm at Pennsylvania State University n September 8, 206 Dietary rn an Mycarial nfarctin 423 subsample f 80 men an wmen age years. Nutrient intakes estimate frm subjects' cmplete SFFQs were cmpare with estimate nutrient intakes in 5 ays f f recrs cllecte ver a -year peri (26). The ability f the SFFQ t aequately rank subjects accring t their ietary intake was emnstrate by Pearsn's crrelatin cefficients ( fr crue ata) an a high egree f classificatin int the same r ajacent quintile (75.8 percent fr crue ata). Fr irn, the Pearsn's crrelatin cefficient between f recrs an the SFFQ was 0.67; ajustment fr sex, age, ttal energy intake (27), an withinpersn variability in aily intake (28) resulte in a value f Data analysis The assciatin between energy-ajuste ietary irn an heme irn intake an risk f mycarial infarctin was examine by using multivariate lgistic regressin. Energy-ajuste ietary intake f irn an heme irn was ivie int tertiles, an the risk f mycarial infarctin bserve in the mile an highest tertiles was cmpare with the risk in the lwest tertile. The initial analysis was ajuste fr age an sex. The analyses were subsequently ajuste fr by mass inex, pack-years f smking, equivalent husehl incme (five categries), highest eucatinal level attaine (five categries), an alchl intake (five categries). n aitin, energy-ajuste ietary intake f P-cartene, vitamin E, vitamin C, ttal fat, saturate fat, an chlesterl was cnsiere. Use f antixiative vitamin supplements efine as use f (3-cartene, vitamin C, vitamin E, r multivitamins was als taken int accunt. Each assciatin was expresse as a relative risk with a 95 percent cnfience interval. Results were cnsiere statistically ifferent at the tw-taile alpha level f Statistical analysis was perfrme by using SAS release 6. sftware (SAS nstitute, nc., Cary, Nrth Carlina). s u 3 B CM i a ccu 75 a 0 3 5 S 2 (0 cav ai a D 3 2 s c &! jl ii 3 E g s is? t 2^ a) CM CM Cg C7) c\i en c c CM en t~- c ^ i t^ (D S N CM t- T- m en i- N c en T- S i- CJ) T^ T-' y- q in c ^ cri q en 5 00 eri c e c cri -a- r- CM in CM tv. c c c ) r-' iv. CM CM q CM h-- c CM in c ^t in m E a) a a.. 5 E Dwnlae frm at Pennsylvania State University n September 8, 206 RESULTS The mean intake f energy-ajuste ttal irn was 9.3 mg/ay in the lwest an 4.3 mg/ay in the highest tertiles. Fr energy-ajuste heme irn, the figures were 0.48 mg/ay an.36 mg/ay, respectively. Since sex an age were nt istribute equally acrss tertiles f ttal an heme irn intake, baseline factrs an selecte nutrient intakes presente in table were ajuste fr sex an age. The percentages f current smkers ecrease acrss tertiles f irn intake, whereas P-cartene an vitamin C intake increase an ttal fat an alchl intake ecrease. Fr heme ca s «s ca. in c B 5 si i 2 a E E tu in i-. en ^t T- en c * in c c.2. T3 D SZ p -* q q T- c c in c c si LU e5b « Q. D ESS Jn C- , «W W =) *!-++ J Epiemil Vl. 49, N. 5, 999 424 Klipstein-Grbusch et al. irn intake, an increase was bserve in the prevalence f hyperchlesterlemia an ttal fat an chlesterl intake acrss tertiles. The percentage f current smkers was cnsierably higher in the highest tertile f heme irn intake. rn supplementatin was negligible in this elerly chrt (0.2 percent) an was nt relate t ietary irn intake. When ajuste fr age an sex, intake f ttal irn was nt assciate with risk f mycarial infarctin. Hwever, intake f heme irn was significantly assciate with an increase risk f mycarial infarctin (table 2). The relative risk fr the highest versus the lwest tertile was.83 (95 percent cnfience interval (C).6-2.9, p fr tren = 0.008). After ajustment fr by mass inex, pack-years f smking, equivalent husehl incme, eucatin, an alchl intake, the bserve assciatin with heme irn remaine (relative risk =.84, 95 percent C , p fr tren ). Ajustment by these factrs an subsequently fr intake f (3-cartene, vitamin E, vitamin C, ttal fat, saturate fat, an chlesterl an fr use f antixiative vitamin supplements i nt materially alter the risk estimate (table 2). When the analysis was repeate withut applying any exclusin criteria, the relative risk was smewhat attenuate but still significantly elevate:.68 (95 percent C , p fr tren = 0.07) fr the highest versus the lwest tertile f heme irn intake. The multivariate ajuste relative risks f mycarial infarctin accring t the presence f ther risk factrs fr mycarial infarctin hypertensin, hyperchlesterlemia, cigarette smking, an iabetes an heme irn intake are shwn in table 3. Fr all risk factrs that were investigate, the increase risk f mycarial infarctin was bserve primarily in the presence f these risk factrs. Relative risk estimates fr fatal (n = 30) an nnfatal (n = 94) cases f first mycarial infarctin an heme irn intake are shwn in table 4. The multivariateajuste risk fr the highest versus the lwest tertile f heme irn was cnsierably elevate fr fatal cases (relative risk = 3.77, 95 percent C , p fr tren = 0.033). Fr nnfatal cases, the relative risk was.59 (95 percent C , p fr tren = 0.064). DSCUSSN n the prspective chrt f the Rtteram Stuy, we fun n evience f a psitive assciatin between ttal irn intake an risk f mycarial infarctin. Hwever, there was a psitive assciatin fr heme irn intake. The assciatins were strngest in thse subjects wh ha ther risk factrs fr mycarial infarctin. The results inicate that the assciatin f heme irn with mycarial infarctin was mre pr!5 c c i Cfl a a S.2 s u s. c 8 5 j c c a 55 csi in 05 i its f ls =5 ~ E. il ill S 3? =5 E E. 8 p q 77 in ^t c CM i- q n c c CM v * en t~- c CM c in (*- cp cq ch c c in v- T- c in ^ ) prq T^ cp un z? q q cq 5 f s i- CM Q. as c c 8. ra 2 c ta CB 3 3 0) J= c 2 & c c c till 3 E g 8f 8 E m s?2«.^ = 3 ^ en's cnjs c = c T S 5 «Qj ^_- is - 8 «- E»- LU «0 t a) ^ J E2ES 5 3 C a * c «S a: Dwnlae frm at Pennsylvania State University n September 8, 206 J Epiemil Vl. 49, N. 5, 999 Dietary rn an Mycarial nfarctin 425 CM S (0 a c (0 5 0) 5 ll 5- E U (0 is E i p ~ 0) C ? c (a *u 5* 3 - 0 li (A U c 8 i- 5 «22! C h in c in ai in T^ T^ f) r-- c C\l ^~ T^ r- h- q q ^ cj CM in i r r^ CM T-^ ^ ( w .5-2 S 9 m j si.e. » w «~ w «t E S C ^ D .^ ) JScJ w T 8 fe ir nunce in fatal than in nnfatal cases f first mycarial infarctin. Befre these results are implemente, sme methlgical issues shul be cnsiere. Ptential bias ue t incmplete fllw-up was unlikely because f the high fllw-up rate (94 percent) achieve. Case ascertainment was facilitate because general practitiners wrking in the stuy istrict f mmr cvere 85 percent f the chrt an recrs were linke t the municipal health service in Rtteram. Since all events were classifie inepenently by tw research physicians an in case f isagreement by a carivascular isease expert, inaccuracies in cing iagnses were minimize. Because subjects with a previus iagnsis f mycarial infarctin may have altere their iets as a cnsequence f isease, thse reprting at baseline that they ha been hspitalize fr mycarial infarctin were exclue frm analysis. Cnfuning by ietary fats an chlesterl was f particular cncern, since these factrs are likely t be crrelate with heme irn. (Age- an sex-ajuste Spearman's crrelatins between heme irn an ttal fat, saturate fat, an chlesterl were 0.4, 0.0, an 0.34 (fr all, p 0.000), respectively.) We attempte t aress this pint by incluing them in the multivariate mel. Hwever, ajustment fr these factrs i nt materially alter the risk estimate. Since heme irn is erive largely frm meat an meat pructs, the bserve assciatin between heme irn intake an mycarial infarctin may be attribute t ther meatrelate factrs besies fat an chlesterl r t unientifie lifestyle habits assciate with meat intake. We attempte t ajust fr lifestyle factrs such as smking an alchl intake, but we cul nt exclue the presence f resiual cnfuning. Hwever, it seems unlikely that a cnfuner cul explain the fact that the bserve assciatin between heme irn an mycarial infarctin was mst marke in the presence f hypertensin, hyperchlesterlemia, iabetes, r cigarette smking. n fact, this is highly cmpatible with the hypthesis that irn plays a rle in prmting LDL chlesterl xiatin. Furthermre, a cnfuner cul nt explain the ifference in the finings cncerning fatal an nnfatal events. Thus, it seems unlikely that cnfuning accunte fr the bserve assciatin between heme irn an mycarial infarctin, althugh we cannt cnclusively ismiss the presence f resiual cnfuning. We fun that heme intake but nt ttal irn intake was assciate with mycarial infarctin. There are several reasns t separate ietary heme irn frm nnheme irn as a ptential risk factr fr crnary isease. rn status is regulate as a result f bth absrptin an biavailability. n general, ietary irn Dwnlae frm at Pennsylvania State University n September 8, 206 Am J Epiemil Vl. 49, N. 5, 999 426 Klipstein-Grbusch et al. TABLE 4. Relative risks an 95 percent cnfience intervals fr nnfatai an fatal mycariai infarctin, by energy-ajuste tertiles f heme irn intake, 4,802 subjects f the Rtteram Stuy, Rtteram, the Netherlans, Nnfatai mycariai infarctin (n = 94) Fatal mycariai infarctin (n = 30) Tertile (lwest)* 2 3 (highest) p fr tren N. f mycariai infarctins Age- an sexajuste FR* % Cl* Multivariateajuste RRt % Cl N. f mycariai infarctins Age- an sexajuste RR % Cl Multivariateajuste RRt % Cl * Cl, cnfience interval; RR, relative risk. t Ajuste fr age; sex; by mass inex; pack-years f smking; equivalent husehl incme (five categries); eucatin (five categries); alchl intake (five categries); categries f energy-ajuste p-cartene, vitamin C, vitamin E, fat, saturate fat, an chlesterl; an use f antixiative vitamin supplements. t Reference categry. absrptin is etermine by tw majr factrs, the rate f erythrpiesis an the size f the by's irn stres. Biavailability is epenent n bth the amunt f meat in the iet, since heme irn is well absrbe, an the balance in the iet between prmters f nnheme irn absrptin (e.g., ascrbic aci, meat, fish) an inhibitrs f irn absrptin (e.g., phytates, sy prtein, plyphenls) (29). The mechanisms f absrptin f nnheme an heme irn iffer markely. At physilgic levels, irn uptake is meiate by a series f receptrs an bining prteins, whereas heme irn is taken up by the mucsal cells; therefre, the irn in these cells is nt expse t the effects f the many ligans in the iet that inhibit irn absrptin. Fr these reasns, nly a slight assciatin f ttal irn with risk f crnary heart isease wul be expecte in a generally well-nurishe ppulatin. Stuies investigating the assciatin between ietary irn an crnary heart isease have yiele incnsistent results. Whereas sme researchers bserve n assciatin between ietary irn intake an crnary heart isease (9, 3), Ascheri et al. (4) reprte a ecrease risk f nnfatai mycariai infarctin r fatal crnary heart isease with ttal ietary irn intake. Hwever, intake f heme irn was assciate with an increase risk f nnfatai mycariai infarctin r fatal crnary heart isease, as was fun in the present analysis. After multivariate ajustment, the inverse assciatin
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x