. ___ ■
`
`The New Englaikd 宀
`Journal of Medicine
`
`• Copyright, I9K5. g the 너心sachuscH、Medic이 Sckict>
`
`\* *ohnne 312
`
`MAY 23, 1985
`
`Number 21
`
`EVIDENCE OF ALUMINUM LOADINQ IN INFANTS RECEIVING INTRAVENOUS THERAPY
`Aileen B, SumiAN, M.I)r (Jordon L. Klein, M.D., Russei丄J. Merritt, M,D,, Ph,D., Nancy L. MYli.er, M,S.,,
`Kim 〇. Weber, B,S.N,, Whj.iam L. Gn4.,M.D., Harish Anand, ~M.D., and Ai.i.en C. Ai.frey, M.D.
`was *10 times higher in infants whd had received at least
`three weeks of intravenous tha^py than in those who had
`received limited intravenous therapy: 20.16±13.4 vs.
`1.98 ±1.44 mg per kilogram 〇| dry weight (P<0.0001).
`Creatinine clearances corrected for weight did not reach
`expected adult value§ until 34 weeks of gestation. Many 贷
`commonly used intravenous solutions are found to' be
`highly contaminated with aluminum.
`We conclude that infantè receiving intravenous therapy
`have aluminum loading, which is reflected in increas^ ,
`urinary excretion and elevated concentratioris in plasm^'
`and bone. Such infants may be at high risk for aluminum
`intoxication secondary to increased parenteral exposure
`and poor renal clearance. (N Eng! J Med 1085： 312:
`1337-43.)
`。
`-
`
`Abstract To investigate the possibility that premahjre
`infants may be-vulnerable to aluminum toxicity acq비ired
`through intravenous feeding, we prospectively studied
`plasma and urinary aluminum concentrations in18 prema­
`ture infants receiving intravenous therapy and in 8 term
`infants receiving no intravenous therapy. We also meas­
`ured bone aluminum concentrations in autopsy specirrtens
`from 23 infant&, including 6 who had received at least'
`three weeks of intravenous therapy.
`Premature infants who received intravenous therapy
`h흤d high plasrrfe and urinary aluminum concentrations, as
`compared with normal controls: plasma' alumirium.
`36.78±45.30 vs. 5.17±3.*1 >9 per liter (mean ±§D,
`P<0.0001); urinary aluminumrcreatinine ratio. 5.4±4.e vs.
`0.64±0.75 (P<0.01). The bone alumin니m goncentr흖tion
`
`.
`
`*
`
` Subsequently, it was *
`lack of normal renal excretidn.
`**
`realized that these patients aisp had fracturing ostco- ,
`malacia? Another indiQaiion of the skeletal toxicity of
`aluminum was the demohsiration of fracturing osteo­
`malacia in animals by parenteral administration of
`aluminum./
`After standards for *ceptable concentrations of
`aluminum in dialysate\had -been established and di­
`alysis units throughout the world began monitoring
`the amount of alumini/m in their water, the incidence
`of dementia and (rafcturing bone disease dropped
`dramaiicaHy」° In st^sequcnl years, a more ins서祐us
`form of bone disease has been described, which is
`sccoridary to excessive exposure to oral alumint^
`'*^ ■
`**
`The eflccis are especially devastating in young chil­
`dren with compromised renal Encu〇n」Z'3 Other
`studies have provided fuhher evidence that parenteral
`nutrition solutions contaminated with aluminum cause
`loading and possibly even bone disease in adults with
`normal renal mnciion」시' No studies concerning alu«
`minum loading have thus far been carried out in pre­
`mature infants who are known to hwc a high inci-
`dcnce" of osteopenia.
`•
`Wc have found high concentrations of aluminum in(
`bone, urine, and plasn|a from infants receiving intra­
`venous therapy. It seems possible that aluminum
`loading may be a factor in the bone disease seen in
`very ill neonates who have reduced rcnaKunction and
`have received long-term parenteral therapy with alu-
`minum-contaminätcd fluids. It is not known whether
`
`、ハ
`
`Although osteopenia in premature infants has
`
`L been well documented, the cause of this compli­
`cation remains to be determined J* It appears to be
`directly related to parenteral therapy, since bone-
`accrelion rates have been found to return to normal
` .. ー・
`after infants were given enteral feedings supplemented
`'
`.
`* '
`I con­
`'
`'
`.
`ノ 爲 with calcium, phosphorus, aqd vitamin D： In
`trast, infants who could not tolerate enteral feeding
`' メ
`had progress" osteopenia and fractures, which were
`resistant to standard therapy」'' This suggests that
`some substance chheF delettd from or given with par­
`enteral therapy may he responsible for premature os­
`teopenia. Since aluminum has been implicated in the
`pathogenesis of vitamin D-rcsisjani osteomalacia in
`human beings* and animals, we elected to study this
`element in premature infants.
`Aluminum loxicity-in human beings has been de­
`scribed in patients receiving hemodialysis; severe de­
`mentia developed, and the patients died as a result of
`excessive exposure to aluminum in their dialysate and
`
`4が
`
`From the Dcpaitments of Pediatrics and Medicine, University of Colorado 겨hd
`Veterans Administration Hospitals, Denver;%¢ Division of Gastmcntcrology 그nd
`Nutrition. Children's Hnspila! of Los Angeles, University of Southern California
`ScE지 of Medicine. Los Angeles; and the Departrnm 인 Pediatrics. Tulane
`University School of Medicine, New Orleans Address reprint requests to Dr,
`Sedman at the Pediatric Nephrology Unit, Box 54, 〇비patient, CI07H, University
`*of Michigan Medical Center, Ann Arbor, MI 4비f)%〇〇0
`Supported in p그π by a ^ranl (RR”69) from the General Clinical Rcsemch
`Center?. Pnigrams of the Division of Research and Resources, National Institutes
`■ of Health: a grant from the Edwan! G Schlieder Education Fund. New prleans;
`and research funds from the Veterans Administration Medical Center.
`
`1
`
`EXELA 2002
`Eton Pharmaceuticals v. Exela Pharma Sciences
`PGR2020-00086
`
`

`

`I t in
`
`THE NEW ENGLAND JOURNAL ()E MEDICINE
`
`May 23, 1905
`z、
`
`r시ah\•시y shnロ시uin high ronccnlralions ofaluminum
`ill plasma and i)<)nr in premature inRuUs can repro­
`duce the loxiciiy that( hrmiie hiding causes in older
`children and adults with chronic renal Hiilure.
`
`Methods
`\\ r nir.isitrrd aliiinunini((ni(mirations in phtsma, urine, or honr
`in five groups of patients, (iroiip I itirhnlcd IH prcmatiHて infants
`Kecks 〇| L；rs(a(i()n} uhn rnpiirnl adinission to the intensive
`(；irr unit and intravrnotis therapy. Plasma and urinary aluminum
`(oiurntratioiK wrrr inrasiircd in thrsp IH infants on two diOrrent
`”<で“si”ns approximately three werks apart： 9 urrr stuciird on the
`fir이 (tuy of life and three weeks later, and 9 were studied at. a »
`ratnlofn tinx' {11 lo 42 days i)f age) and three weeks later. UrinaVy
`anti wnini creatinine levels were measured in infants who were
`more than oiir week eld at the time of the study. ( :r「기wine was n()\
`mrasurrrf in infants less than one week of agr. bee즈use the concen­
`trations *rrr lli(iut;h( to br n rrllrrtinn of (hr mnlhrr's rrralininc
`
`Table 1. Plasma and Urinary Aluminum Levels in 18 Premature Infants ReceXXhg
`Parenteral N비rition.・. "
`■
`-
`
`
`
`Pl.*5MA
`FH卜 DZ； ALVMINttM
`
`UkiNAtY
`UUNAIV
`UHINKIIY
`Aluminum:
`Exchetion
`Aiuminlim or Aluminum CucATtNiNE
`t
`
`No
`
`I
`
`2
`
`3
`
`4
`
`5
`
`6
`
`*Z
`
`1
`
`uA
`
`2K
`
`32
`
`26
`
`32
`
`3D
`
`-
`
`AOE
`
`da、$
`
`1
`22
`I
`22
`1
`IK
`I
`22
`
`Bia TH
`Wfh^ht
`
`X
`
`1070
`
`980
`
`*
`1060
`
`760
`
`1200
`
`1460
`
`1610
`
`HX'liirr
`
`.PRほ4 hr
`
`*
`
`•
`
`5 0
`
`2 9
`
`10 I
`
`,5,3
`
`18
`
`19.5
`
`-
`
`4 3
`
`2.2
`
`4.3
`
`60
`
`0 5
`I
`11.4
`
`and no( that of the infant. Jn 13 infants complete 24-hour samples
`were collected at the thrcc-wcck study; in the remainder of the
`infants 24-hour collections were incomplete and thus could not be
`usrd for calculation of rlrarancrs.
`Group II included eight term infants'who did no^ require in-
`〔ravenous therapy (normal co 기 mh). In four of these infants plasif,
`ma and urinâry alutninum concentrations were measured at one*'
`day and three weeks of age, and in the other four, aluminum con­
`centrations were measured at one visit. Aluminum was also meas­
`ured in 35 umbilical-cord plasma samples, to csta비ish a normal
`base line; these v너ucs are reported with normal control measure­
`ments.
`To estimate the amount of aluminum being given to and retained
`by infantsrwho were receiving only intravenous therapy, we stud­
`ied ä third group of five infants (Group HI) who had been receiv­
`ing intravenous therapy for at least two weeks. Intake and excretion
`of aluminum in the urine ^ere measured for a period of two or
`three days.
`Clirtical data on the three groups of infants arc shown in Tables !,
`2, and 3. Parents were required to sign a standard consent form
`before the infants were enrolled in the pro­
`■
`spective study.
`Autopsy specimens were collected from
`、23 infants (age range, 0 to 7 months) and
`divided into two groups. *Seventeen of the
`infants (Group IV) did not receive pro­
`longed parenteral therapy because they died
`in the emergency or delivery room or with­
`in one week pf hospitalization. Representa»
`tive diagnoses at the time of death included
`the sudden infant death syndrome, severe
`prcmaiuriiy, and congenital cardiac defect.
`The other six infants (Group V) had re­
`ceived, at least three weeks (mean ±S.D.,
`9.6±5.2) of parenteral nutrition before death
`(considered long-term parenteral nutrition).
`All these infants were under 37 weeks of
`gestation and required parenteral nutrition
`secondary to necrotizing enterocolitis, con­
`genital malformation of the gastrointestinal
`tract, or severe feeding intolerance. One
`member of this group was a premature in­
`fant who had been discharged from the hos­
`pital at four months of age but died in the
`cmergcftcy room al six months, with a diag­
`nosis of sudden infant death syndrome. The
`other five infants were inpatients at the time
`of death.
`Plasma and urine samples were collected
`by clinical research nurses or technicians
`trained in study collection techniques. To
`avoid exogenous contamination of plasma
`samples until completion ,of the analysis, the
`skin was carefully cleansed with deionized
`water, and samples were collected in unal­
`tered plastic containers and frozen in plastic
`containers. Urine samples were obtained by
`cleansing the skin as described above and
`applying a standard plastic urinc-collection
`bag. Before 24-hour samples were collected,
`the bag w궁s put in place, and urine was con­
`tinuously grawn from the bag during the 24­
`hour period. .
`Specimens of bone were collected in the
`autopsy room and immeÖtatcly frozen in
`plastic containers. They were later cleanly
`dissected, and the surface thsue that was
`adherent or contaminated was discarded.
`Twelve of the bone specimens were from tbc
`vertebral body, and eleven were from the
`iliac crest. All vertebral specimens were from
`infants who died acutely without prolonged
`
`0-2 .
`
`1.8
`
`1.7
`
`22
`
`t6
`2,1
`크,9
`3.6
`0.9
`9.8
`9.6
`3.4
`13,5
`5 9
`3 1
`14.5
`1,9
`6-7
`II 0
`
`*Aluminum *a、mcBsuinl <kt iwddifterenl occanon$ ^cpanled by in inlcrv»! of three week，. IV denotes ioira venous. B 뷔
`hEi、t milk, and F formula Tif ciinven aluminum vilueü to mt© i)moies per liter. divide by 27.
`
`IV .
`BM/F
`IV
`BM/F
`IV
`BM/F
`IV
`F
`IV
`F
`IV "■
`F
`IV
`F
`IV
`F <
`IV
`ナ
`F
`F
`F
`F
`F
`F )
`f /
`f"
`IV
`!V
`IV -
`IV
`IV
`F
`IV
`F
`IV
`F
`
`1
`
`49
`29
`35
`3
`근 60
`26
`6 -28
`170
`9
`60
`9
`35
`9
`26
`冬｝
`3
`.6
`8
`3
`20
`¢2
`137
`6
`15
`<2
`<2
`<2
`7
`<2
`<2
`20
`35
`9
`20
`3
`4
`22
`20
`13
`32
`46
`16
`13
`25
`3
`3 고
`6
`39
`"42
`26
`i 15
`i 90
`29
`2 야
`32
`23
`20
`4
`72
`9
`106
`6
`
`132 r -
`
`I
`초 5
`1
`.33
`1
`21
`1
`、23
`*•
`24
`42
`78
`17
`3«
`16
`37
`II
`22
`고 2
`45
`30
`5!
`22
`46
`t6
`43
`16
`32
`
`1440
`
`1980
`
`1520
`
`1220
`
`820
`
`2720
`
`2320
`
`75»
`
`990
`
`H0Ü
`
`2W)K
`
`33
`
`34
`
`31
`
`3 그
`
`32
`
`2«
`
`36
`
`37
`
`27
`
`2K
`
`3 그
`
`れ
`
`K.
`
`9
`
`10
`
`11
`
`12
`
`13
`
`)4
`
`15
`
`16
`
`17
`
`IK
`
`<2.0
`
`3 9
`
`4,7
`
`3.5
`
`4,0
`
`3 8
`
`27,4
`
`2
`
`

`

`\'시 112 N(. :M
`
`沪 WIN 니 M
`
`roxuu rv aド「er in vrzWflnoi's therapy — 3EI)m.\n et al.
`
`1339
`
`Table 3- Urinary Aluminum Levels in Five Infants Receiving Intravenous Therapy for at
`Least Two Weeks before the Study.* ,
`
`tNFANI
`No
`
`1
`2
`.3
`4
`5
`
`Aiit
`
`me
`
`5 5
`i 5
`9 0
`4,5
`5 0
`
`Al UMtNUM
`Iniakf
`
`At UMINttM
`mH hr
`
`121
`53
`103
`lOX
`lOH
`
`35
`21
`25
`14
`5
`
`灯¢
`
`7 H
`3、3
`6 5
`6.7
`6 0
`
`HiiMAiro
`Rr HNTHtH
`
`じ* IM니・ Y
`Ai eMINVM CRFATINfNE
`
`や
`
`71
`60
`76
`K7
`95
`
`,
`
`4,4
`,.55
`26
`35
`07
`
`.
`
`•To given *tuminum vilucs to micrun*»!« per liter. JivnJe by 27.
`
`in(i.i\rii<iiis (hrr.iff\ < ^(n( rti(r,in(»ns oi .tJu-
`'nniuiii in v rrtrlirai lxiiir \\err 11111 sttfiiiti
`( antly diflrrrtil hoiit h、\시ヾ in Uiat -i rrst Ihhic
`f2 ,25 *: 1, 10 vs, ! f \ t ing prr kii(mr;ini
`(t| dry いri이H). I hrrrJnrr. ilir rrsulls i)i llir
`iwinr aii.tlssts arr rcpiinrd v」lh¢나ハ(lrsii<na-
`tmn of ilir U pr of botir, Sint r tiortnal Ihhit
`histofngy hax iu)( hrrii wr|| <hi( uiiictHrd it)
`prrnialurr ttilanls, liisHiloyic studies of bone
`\\(Tf not prrfnrinrti in this study.
`Vlasina, Ihhic. and urine were pnnessetl
`and analyzed l(ir ahiniiiiuni < onieiu. as pre-
`\ iously tlescriixxi. by Hainclrss aluinu sibsiirp-
`tton sprt iruplHtloinrlry.
`*'*
` Urinary and se­
`rum crralininr が\¢is were dctcrnnnrtl by
`snuitlant nirlbfxJs-
`(:rcalininr clearances wrre ntrrrcted lor
`ImkU wri^lu al (he time of the 24-hcur uriiir c(»Urvlioiv Mean values
`of atumiiiuni were caiculatrd from I be higher plasma aod urinary
`ciHitnitrations rrmrdrtl ft»r rach child, if two values were obiaHicd.
`Aluniinuni:crraiininr ratios wrrr calculated us micrograms per liter
`n「aluniinum divided t)y milligrams per drcühcr(»f creatinine, («
`(irriyr a number ihai iactorrd out dinrrrners in eoncrntraiion. Pcs-
`i.ible sources(»f aluminum ( Tablrs 4 and 5) were invcsiigaicd by
`(ibiainint; aliquois of nirditations and solünoiiE from pharmacies or
`formula packages. Breast milk was <(iHmr<i from mothers insiruct-
`r(i to clcansr tbc skin carrruUy and express milk directly, into plastic
`nintainrrs.
`,
`Statistical analyses were done by IkiiIi parametric and nonpara・
`meiric metho<ls (Sitideni's l-iesi and the Mann-Whitney U test).
`Values arc expressed as means ±S.D. P values arc derived from
`Sludcnl's t-icst unless otherwise designated.
`
`
`
`Mg per iiicr (I.4± i.7 vs. 0J9±0.I I gmoi per liter)
`(PVO.OO이); aluminumrcrcatininc ratio, 5*4±4.6 vs.
`0.64±0.75 (PV。이). Plasma aluminum levels and
`urinary aluminumrcrcatininc ratios in these two groups
`are summarized in Figure I. The previously published
`normal plasrpa concentration in adults
` is 6±3 gg
`***
`per liter (0.22±〇. 11 卩m이 per literX which is not sig­
`nificantly different from the level in our normal
`infants. Aluminum concentrations in the 35 umbilical-
`cord plasma samples were also not significantly diflcr-
`ent from the levels in normal infants ¢4.5±3.7 /xg per
`liter [0.17±0.I4;imol per liter]). Although toxic con-*
`centrations of aluminum in plasma have not been de-
`nmiiv미y established, a value over 100 mR per liter is
`Results
`consistently found in association with bone and neuro­
`logic abhormahtics」흐'"'시8 Although the mean value
`Table 1 shows the data on the premature infants
`in our infants in Group I was 36 jug per liter. Infants
`with a history of intravenous therapy (Group I), in­
`17 and Id had values of 172 and 136 /ig per liter (6.4
`cluding gestational age, birth weight, and age at the
`and 4.4 /imol per liter), rcspecn^cly, which were well
`time of the plasma and urinary aluminum determina­
`into this presuma비y toxic range.
`*
`tions, along with the type ofintake at the time of meas­
`In 13 Group I patients plasma aluminum conccn-
`urement. 丁able 2 shows the data on normal infants
`' trations were measured while the infants were receiv­
`without parenteral exposure (Group II). Plasma
`ing intravenous therapy and then while they were
`시uminum kv이s and urinary aluminumxreatininc
`receiving formula. Plasma concentrations during in-
`ratios were significantly higher in Group I than in
`Group π： plasma aluminum, 36.78±45.3 vs. 5.17±3.1 * travenous therapy were significantly diflerent fi-om
`the levels during formula feedings:
`36.18±54.57 vs. 8.08±8.2 Ng per
`liter (1.4±2.0 vs. 0.3±〇.31 /xm이
`per liter; P<0.01, Mann-Whitney
`U test). There was no correlation
`hetween the (btal number of days
`of intravenous therapy and subse
`
`*
`quent urinary aluminumrcrcatininc
`ratios (r = 04!〉Interestingly, uri­
`nary aluminum concentrations and
`aluminumxreatinine ratios contin­
`ued to be very high in some iniants
`(TableJ, InfZmis I, 3,,6, and 18),
`as compared with normal values,
`even though the infants had not re­
`ceived intravenous therapy for a
`mean of 13 days and had normâl
`plasma aluminum concentrations.
`In very premature infant^, weight-
`corrected creatinine clearances were
`
`Table 2. Plasma and Urinary Aluminum Levels rn Eight Normal Term Infants without
`Exposure to Parenteral TTierapyノ
`
`Infant
`N。.
`
`Age At
`Study
`
`Ftf.CTNG
`
`Plasma
`ALUMIN 由
`4
`
`U치N시lY
`ALUMINUM
`
`Ukinakv
`AtUMW 바，:Cjuwzne
`
`1
`
`2
`
`5
`
`4
`
`5
`6
`7
`K
`
`6
`10
`3
`<2
`10
`3
`3
`3
`
`*
`
`I day
`23 days
`J day
`21 days
`1 day
`21 days
`1 day
`22 days
`9 mo
`2 mo
`7 mu
`3 nio
`
`H.O
`
`F
`F
`F
`F
`F
`F
`F
`H：O/BM
`F
`HM
`
`IS
`70
`9
`5
`12
`15
`28
`2
`20
`22
`7
`12
`
`2 4
`
`0 6
`
`0 8
`
`0 1
`0,13
`0-2
`06
`0.28
`
`*BM derMrfcs breast milk, tnd F &»rm바，To convert «turritnum vjitues to mkromoin per Jjicr. divide by 27,
`
`3
`
`

`

`I n(i
`
`I HE NEW EN(；EAM)J()l-RNAE OF MEDRUNE
`
`May 23. 1983
`
`Table 4. Levels 〇| Aluminum in Commonly Administered Intravenous S이니ions.，
`
`N(» ()1
`tun Tistm
`
`Al
`
`M
`
`S<H I Ihih
`
`MU
`PiHasMUin phtnphalc ( MKM) mintiPhlcr)
`(3(XKJ mnKil/iitcr|
`Stxiium
`，,
`Calcium glucenaic 시〇りI)
`Heparin (ItXX> unhs/mh
`Heparin (5OÜ0 unils/mh
`Heparin 110.0ÜÜ unii\/mh
`Normal scrum albumin (25*i
`iniralipid
`TPN sututiun ｛泌％ essential
`amino acid)
`5% Dextrose
`,
`Sodium chloride (4000 mmol/liter)
`Putassium chloride (3000 mmoUliter)
`
` )
`
`'
`
`3
`■1
`5
`3
`1
`I
`4
`1
`6
`
`2
`3
`1
`
` I.HÜI
`)6.598
`*
`5：977
`5.056*335
`:6K4±761
`359
`468
`1.82212.503
`--195
`72±59
`
`73±1
`6±4
`6r
`
`Tables 4 and 5 list the concen­
`trations of aluminum in commonly
`used intravenous and oral solutions.
`Calcium and phosphate salts, hepa­
`rin, and normal ser냐m albumin have
`very high concentrations of alumi­
`num, although ihei% is wide vari-
`-ation among lots. Soy and prema­
`ture-infant formulas with the highest
`additives of calcium and phosphate
`salts also have the highest aluminum
`concentrations. Human hreast milk
`has the least aluminum. •
`Among the infants in Groups I
`and π, there was no correlation be­
`tween plasma and urinary aluminum
`and the type of formula they were
`receiving at the time of the study.
`None of our infants were receiving
`soy formula, which has the highest concentration of
`aluminum. It should be noted that normal adult con­
`centrations of urinary aluminum
` wotild give an -
`*^
`aluminumxreatininc ratio ofO.l; thus, the ratio of (城
`in our normal infafhs may reflect the fact that infants
`absorb aluminum diHerently from adults.
`
`•TPN denote» total parenien! nuiniion To convert alumtnum valu«((i mithmuks per hicr. divitlc by 27 Plus-minui»
`valuei> are means 三S D
`
`as low as 20 per cent of those in mature infants
` Weight-corrected creatinine clearances were
`(Fig.
`*2).
`directly correlated 베th gestational age, and by 34
`weeks h쵸d reached normal adult levels (1.7 ml per
`minute per kilogram of body weight, Fig. 3). The
`highest plasma aluminum level in these infants d너
`not correlate with creatinine clearance, suggesting
`that the intake of aluminum is the more import효nl
`variable.
`，
`Group III infants had been receiving intravenous
`therapy for least two weeks when aluminum intake
`and excretion studies were pcr(()rmed: four infants
`were studied for three days, and one was studied for
`two days (Table 3). All five infants had negligible stool
`output. Although stool aluminum was not measured,
`previous studies have shown that f&jpl aluminum
`losses arc small in patients receiving parenteral nutri-
`tion」다 Aluminum intake was calculated from concen­
`trations in matched solutions. The five infants had
`a mean intake of98.6±l 1.56 “g (3.6±0.43 林mol) per
`24 hours, with a mean excretion of 20.0±5.1 jLig
`(0.74±0.19 卩mol) per day, or 78 per cent retention.
`Normal adult urinary excretion of aluminum is 13±6
` The aluminumxreat-
`gg (0.48±〇*22 /xmol) per day.'
`**
`inine ratio was 3.3±0.08, which is significantly difler-
`ent from the value in normal infants: 0.64±0.75
`(P VO・이).
`Bone aluminum content in Group V (infants who
`had received intravenous therapy for over three weeks)
`was significantly higher than in Group IV (those
`who had received little or no intravenous therapy):
`20.16±13.4 vs. 1.98±L44 (0.75±0.49 vs. 〇.〇7±〇.〇5
`mmol) per kilogram of dry weight (P<0.0001, Fjg. 3).
`All except one infant treated with long-term parenter­
`al niJirition died during intravenous therapy. I'he in­
`fant who was discharged and died two months later
`from the sudden infant death syndrome had a bone
`aluminum concentration of 7 mg (0.26 mmol) per kilo­
`gram of dry weight -- three limes the normal level
`in infant hone. The normal bone concentration in
`ad비is'" is 3.3±2.9 mg (0J2±〇.11 mmol) per kilo­
`gram of dry weight.
`
`Discussion
`Eight per cent of the earth's crust is composed of
`aluminum. Human beings are exposed to this clement
`constantly through ingestion of v炉icr, food, and dust
`paru이cs. In fact, it is estimated that an average adult
`.ingests 3 to 5 mg of aluminum per day*니" and ex­
`cretes 14 #tg per day in urine — prestima비y the
`amount absorbed こ— leaving only a sihall level of
`aluminum in the body」탕 Since the gastrointestinal
`tract, skin, and lung provide excellent barriers to alu­
`minum absorption and the kidney is efTicient in
`eliminating any absorbed aluminum, the body burden
`of aluminum in normal persons is exceedingly low.
`Nonetheless, excessive aluminum accumulation can
`
`Ta비θ 5. Levels 이 Aluminum in Commo히y Used Oral Solutions.*
`
`SOIUTION
`
`Glucose water
`Tap water (ColonKJo)
`Well water (Cokgd이
`Breast milk
`Cow's mitk-based formula
`(20 kcal/30 ml)
`Cow's milk-based formula,
`"premature"
`(24 kcal/3U ml)
`Soy formula .
`(20 kcal/3() ml)
`
`-
`
`Muhivit^jmins (liquid)
`Nystatin
`
`No tn
`Lots Tested
`
`Aluminum
`CONTEHT
`Oig/Hi히
`
`'
`
`1
`1
`1
`12
`4
`
`4
`
`4
`
`1
`1
`
`20
`12
`5
`99*687
`266*192
`
`699*321
`
`
`
`'
`
`1478土 103
`
`32
`72
`
`•To convert aluminum vilues to mittimotci per liief. divide by 27 Ptin-minui vduct we
`means ±S.D
`
`4
`
`

`

`\小U Jヽ“ 21
`
`\LI \HヽI M
`
`! 1i V Al 1 i.R IM RA\ KXOCS「HER \卩ヽ’一 SKDMAX ET AL.
`
`I in
`
`D Norm하 tMo기・
`
`□ Nor mot infonti
`
`was 10 times higher than normal in
`.1 small group of infants who died
`after receiving parenteral therapy for
`three \yecks or more.
`'
`The amount of toxicity that res비is
`from parenterally administered alu-
`minunu remains to he determined.
`'Fhe classic manifestations of system­
`ic aluminum intoxication include
`fracturing osteomalacia, encephaiop-
`a'thy, and microcytic hypochromic
` For a variety of reasons.
`*-^'-*
`anemia?
`It is difficult to relate any of ihe ab­
`normalities found in til premature
`infants to aluminum intoxication,
`z\lthough bone disease is common,
`the histologic data required for the
`diagnosis of osteomalacia are non­
`existent for premature infants. In
`addition, metabolic encephalopathy
`is a common complication of* many
`of the other events that occur along
`with prematurity, including hypoxia,
`acidosis, and electrolyte imbalances.
`Finally, premature infants frequent­
`ly require multiple transfusions to
`replace iatrogenic blood loss, pre­
`cluding definitive hematologic evaluation. The facts
`that plasma aluminum levels in two infants exceeded-
`
`Figure 1. Plasma Aluminum Levels (A) 휺nd Urinary Aluminum:Creatinine Rattos (B) in
`Norma* Infants (Geo비p U) and Premature Infants Receiving Intravenous Therapy
`(Group 1). ,
`Bars denote SQ. The normal values in adults are 6±3 尹g per liter and 0.1. respective­
`ly. Al denotes aluminum, and Cr creatinine. To convert aluminum values to micromoles
`per liter, divide by 27.
`
`occ팂r and has been reported with bo나】 parenteral and
`oral exposure. Parenteral exposure has resulted in
`accumulation in patients receiving dialysis with alu-
` and in those re­
`**-^**
`min탆m-coniamma/d dialysate
`ceiving long-ttvm parenteral nutrition with 시umi-
`num-contaminated fluids.새，心 Patients with chronic
`renal (aiiurc receiving large oral loads of aluminum
`in the form of phosphate-binding gels have ako heen
`found to have an increased body burden of alu-
`mhnnn」'い시"
`'
`'
`Our study concerns another type of patient at risk
`for 죠himinum toxicity — the premature infant who is
`receiving intravenous therapy. *Fwo conditions may
`predispose such children to aluminum loading: p^r^n-
`teral aluminum exposure and reduced ren교I function.
`A number'of components of parenteral fluids given to
`premarure infants, such as the calcium and phosphate
`salts used as adm니ves and h탾man albumin, have high
`concentrations of aluminum.
`Evidence of aluminum loading in premature infants
`is based on our finding that aluminum in plasma or
`urine or both was almost uniformly increased in in­
`fants receiving p교rcnteral therapy. Moreover, bal죠nee
`studies showed that the infants excreted only about 23
`per cent of the intravenously administered aluminum.
`Additional evidence for aluminum loading comes from
`the finding that urinary aluminum concentrations re­
`mained well above the range fo나nd in control infants
`for several weeks after the termination of parenteral
`Seeding. This finding s냖ggcsls a slow unloading of tis­
`sue stores()1' aluminum that had accumulated during
`I he period when lllc inHints received parenteral nutri­
`tion. I'he strongest evidence for aluminum loading .
`is the finding that the hone aluminum concentration .ノ
`
`26 27 28 29 30 31 32 33 34 35 36 37
`WEEKS GESTCTION OF PREMATURE -
`INFANTS AT BIRTH ,
`Figure 2. Correlation between CrMlinine Clearance Corrected for
`Body Weight and &Station at Birth.
`Infants started at one fifth the clearance in 흐dutts and h효d normal
`levels at 34 weeks of gestation. Adult clearance Is 1.7 ml
`per minute. *
`
`5
`
`

`

`I il2
`
`THE NEW EN(；LANn jOl RNAE OF XtEDICINE
`
`May 23, 1985
`
`Group IV* infanta without
`prolonged intravenous therapy
`
`Group V- infants with > 3 weeks
`intravenous therapy
`
`50 r
`
`小
`
`P < .0001
`
`BONE ALUMINUM
`Figure 3. Bone Aluminum (Al) in Infants Who Did Mt Receive
`.Intravenous Therapy (Group IV) and in Those Who Had over
`Three Weeks of Intravenous Therapy (Group V).
`Values 츊re means ±S.D. and are express휴d as milligrams p드r
`kilogram 이 dry weight. The norma! adult value is 3.3± 흐.9 mg per
`kilogram. To convert aluminum values to milliE이es per kilogram.
`divide by 27. 〇
`
`100 fig per liter — a value previously associated with
`toxicity in children'：시:시传 一 and that bone alumi­
`num levels were as high as 36 mg per kilogram of dry
`weight suggest that parenteral aluminum loading in
`prcmuiurc ihlanls may have some 이mic시 conse­
`quences.
`We are less certain of the evidence f()r aluminum
`loading in premature infiints g代でn oral al탾minum-
`cnniaining substances. /\lihough we have shown that
`many oral substances given to infants coni츠in alumi­
`num, we believe that our data support the fact that the
`gastrointestinal tract is an adequate barrier in most
`persons, since we did find th교i normal infants had
`plasma aluminum concentrations similar to those re­
`ported in normal adults. We also R)und that bone alu­
`minum concentrations were consistently low in infants
`who had been fed orally but not parenterally bef()rc
`they died. Sull, wc must make note of the laci Oku the
`aluminum:creatinine ratio in normal, botUc니ed in­
`Hints is approximately six limes higher than that in
`normal adults「시 This relaliv이y high ratio s비ggcsls
`that the gastrointestinal barrier to aluminum absorp­
`tion is not fully developed in infants.
`Another (actor to consider is that 효 number of
`formulas, especially soy, have an aluminum content
`that is much higher than that found in human breast
`milk. In an adult, normal dietary intake of aluminum
`is 40 to 50 /ifT per kilogram of body weight per day,
`whereas in an infant onノoy formula, it can be as high
`
`as 250 Mg per kilogram per day. Although it is unlikely
`that this amount of orally ingested aluminum could
`cause any toxicity, a recent report comparing soy and
`milk formulas found decreased mineralization in in­
`fants fed soy preparations." Many other diflerences
`between soy and milk, such as calcium and phospho­
`rus content and availability, may account for the min­
`eralization defect. Although evidence is needed, it is
`not unwarranted to suspect possible aluminum con­
`tamination in infants receiving large oral loads of soy
`formula.
`To conclude, even though the mechanisms arc un­
`known, the evidence is strong that aluminum causes
`skeletal and neurologic toxicity in human beings. It
`seems possible that small children with rapidly matur­
`ing skeletal and nervous systems may be unduly sus­
`ceptible to such toxicity. /\Ithough altered b。n흔 min­
`eralization is the clinical problem that raised our
`concern about aluminum tヾicily in infants, the pos­
`sibility of neurologic toxicity must also be consid­
`ered. With technological advances in medicine, the
`number of premature infants who require long-term
`intravenous support is rapidly expanding, thereby in­
`creasing the potential severity of this problem. The
`source of aluminum contamination in intravenous
`fluids is unknown, but the wid¢ variation in aluminum
`concentrations tn diflercnt^ots of the assorted coqj-
`pounds suggests that contamination may occur during
`the processing of these materials. If this is the case,、
`it may be r이aiiv시y simple to eliminate or reduce alu­
`minum contamination by modifying the manufactur­
`ing process. Until aluminum can be removed from
`these fluids, the amount of parenteral aluminum expo­
`sure leading to toxicity in infants needs to be de­
`termined, and permissible levels of exposure must be
`established.
`
`Wr arr indrlxrd to Dr. Strvë Abrams and Dan Berry for s이u・
`lion and tissue retrieval, to Dr. M.A. Shork and Susan Scr^ika for
`statistical assistance, to Dr. John Benson (Ross laboratories)
`his coninients, and to Ruth Primas f()r preparation of (he manu­
`script. *
`
`•
`
`References
`
`L Minton SD. Stcichcn JJ. Tsang RC. Bone mineral cem히it tn term
`*sge
` infants. J Pediatr 1979; 95;
`alid prcicmn appropnate-for-gestationai
`2374L
`2. Grver F. Lane J. Weiner S. Mazes$ R. An improved technique for accuraie
`dcierniiniiliun of bone mineral content (BMC) and identifying disorder of
`bone mineral metabolism in low-binb-weight infants. Pediatr R« 1983;
`16:163A.
`3. Griscom NT. Craig JN. Neuhauser EBD. Systemic bone disease developing
`in small premature infa지s. Podi금tries 1971; 48:883-95.
`'
`4„ Hinstad! DH. L'Hcurcux PR. Rickets as a complkatton of intravenous hy­
`peralimentation in infants. Pediatr Radiol 1978; 7:211-4,
`5 Klein GL・ Cannon RA. Amc지 이E. Norman AW. Cobum JW. Rickets and
`<»slcopcnia wiih normal 25 OH vitamin D lev히s in infants on parenteral
`nuiritiiin Clin Res 19K0： 28:596A
`6. Alfrcy AC, LcGcndrc GR. Kachny WD The dialysis emxphalopathy
`syndrome: possible aluminum intoxication. N Engl J Med 1976： 294:
`184-8.
`7. Purkinson IS, Ward MK. Feest TG. Fawcett RWP, Kerr DNS. Fracturing
`dialysis osteodystrophy and dialysis encephalofutby; an epidemiologtca]
`survey Lancet 1979; 1:406-9.
`.K Ellis HA. McCanrty JH. Herringion L Bone aluminum in hacmodialyted
`
`6
`
`

`

`\■<.I, 112 N(
`
`TWO SUB I Vi'ES OF Cl'SHiNC'S DISEASE 一 Vz\N CAIH ER AND REEETOFF
`
`1343
`
`patients and in rirfs ingcd *uh uluminum chloride: rvlalionship (o im­
`paired Nine minerali'-aiion J Clin Pathol 1979； 52;K32-44
`り (HXHJman W(；. Ciilligan J. Horst R Shtirl-term aluminum admimsinihon in
`the r기: 비见cい nn bone lurmatmn and rclatuinship t(i rcn바 osteomalacia
`J ( lin Invest 19K4； 73:i7I Kl
`H) Alfrey AC Aluminum Adv Clin Chem 19X3; 23:69-91
`11 Mehla RP Encephalopathy in vhn>nic renal failure appeanng before the
`start of dialy.MS Can Med Assix: J 1979; I2ÜJ ) J 2-4
`12 Andreoli SP” Bergstein JM. Sherrard DJ Aluminum intoxicatton from alu-
`minum-contatning phosphate binder、in thildrcn with azotemia nu( undergo-
`까F dialysis N Engl J Med t 다X4： 가():1079
`4*
`15 Setiman AB. Miller NL. Warady BA. Lum GM. Alfrey AC Alumi­
`num loading in children with chronic rcnaj failure Kidney fnl 1984; 26:
`2이 4
`-
`、
`14 Klein GL. Alfrey AC. 이出er NL. 미 at Aluminum loading during lulul
`parcnicral nutriuun. Am J Clin N비r 1982: 35:1425-9
`15. 〇« SM. Maloney NA. Klein GL. cl aL Aluminum is ass(x;iaicd with low
`Nine formation in patients receiving chnmic parenteral nuirition Ann Inicm
`Med i州3:，*州:，
`16 Sedman AB. Wilkcning GW. Warady BA. Lum GM. Alfrey AC. Enceph­
`alopathy tn childhtxxl secondary ［〇 aluminum loxicily J Pediatr 19H4;
`1O5.K36-X
`
`17 Alfrey AC Ahjmin브m inioxicaiion N Engl J Med 1984; 310 H13,
`5
`U，Polinsky 이S. Gruskin AB Aluminum toxicity tn children with chronic
`renal failurh J Pediatr 1984： 105:758-61
`19. Klein GL. Otl $이. Alfrey AC. 히 al Aluminum as a factor in the bone
`disease of long-term parenteral nutrition. Trans Assoc Am Physicians 1982;
`95:155-(W
`20 Greger JL. Baier MJ Excretion and rcicniion of low or modcraic lev­
`els of aluminum by human subjccts„ Food Chem Toxic시 1983; 21:473­
`7
`■
`21 Gorsky JE. Dietz AA. Spencer H. Osis D Metabolic balance of aluminum
`studied in six men. Clin Chem 1979; 25:1739-43^
`22. Alfrey AC. Afu