Volume 15 October 1999 Number 10

APPLIED NUTRITIONAL INVESTIGATION Nutrition Vol. 15, No. 10, 1999

Whole Blood Manganese Levels in Pregnancyand the Neonate

ALAN SPENCER, GRAD DIP DIET, MPH

From the Department of Nutrition, Gold Coast Hospital, Southport, Queensland, Australia

ABSTRACT

A finding of hypermanganesemia was reported in a pregnant female who received total parenteral nutrition (TPN) for a totalof 27 wk as a result of intractableHyperemesis gravidarum. It was assumed this was the result of the TPN therapy which includedmanganese (Mn) supplementation. However, the Mn levels continued to rise despite ceasing supplemental Mn, and no evidenceof hepatic cholestasis. These high levels were also observed in the 10-d-old neonate, despite a fall in the mother’s levels atparturition. A cross sectional study of 19 females at 34 wk pregnancy showed six (31.5%) were outside the quoted referencerange for Mn levels. This study was undertaken to monitor Mn levels throughout the pregnancy period and in the neonate in theinitial post partum period. Thirty-four pregnant females were randomly recruited from the Gold Coast Hospital antenatal clinic.Whole blood Mn levels were determined for each female on three separate occasions during the trial, at the first visit to the clinic(10–20 wk) and at 25 wk and 34 wk gestation. Mn levels were also determined from blood taken from the baby at neonatalscreening (3–4 d old) by heel prick. There was a significant increase in the mean whole blood Mn levels throughout thepregnancy (P , 0.0025 andP , 0.0005,respectively) and an average three-fold increase in this level in the neonate (P ,0.0005). Thedata confirms a concentrating effect of Mn, as measured by whole blood levels, throughout pregnancy and moredramatically so in the neonate which may indicate a vital role in fetal development.Nutrition 1999;15:731–734. ©ElsevierScience Inc. 1999

Key words: manganese, pregnancy, neonate, total parenteral nutrition (TPN)

INTRODUCTION

Manganese (Mn) is a trace element which has been shown tobe an essential nutrient in several animal species particularlyduring gestation and early infancy.1 Its essentiality is related to itsdiverse function as activator and cofactor in a series of importantenzymatic reactions. Of particular note are the gluconeogenicenzymes, pyruvate carboxylase and isocitrate dehydrogenase; thebiological antioxidant, superoxide dismutase; and the glycosyl-transferases, which are involved in the synthesis of mucopolysac-charides and glycoproteins.1

In humans the recommended daily intake for Mn is estimatedat 2.5–5.0 mg/d. This is obtained from a mixed diet of grains,cereals, and fruits.2

During intravenous feeding Mn (MW5 55) is usually suppliedas a mixture of trace elements which include copper and chro-mium. Some preparations also include zinc and selenium. TheAmerican Medical Association Nutrition Advisory Group hasrecommended a daily parenteral intake of 3–15mmol/d,3 thoughmost authors now recommend an intake around 3–5mmol/d.4

Although pregnancy requirements for Mn have not been estab-

Correspondence to: Alan Spencer, MPH, Department of Nutrition, Gold Coast Hospital, Southport, Queensland 4215, Australia. E-mail:SpencerA@scrha.health.qld.gov.au

Nutrition 15:731–734, 1999©Elsevier Science Inc. 1999 0899-9007/99/$20.00Printed in the USA. All rights reserved. PII S0899-9007(99)00144-6

lished, adapted guidelines have set oral and parenteral intakes atsimilar levels to the non-pregnancy state.5

An incidental finding of hypermanganesemia was reported in a32-y-old pregnant female who received total parenteral nutrition(TPN) for a period of 27 wk as a result of severeHyperemesisgravidarumwhich was uncontrolled by antiemetics.6 It was as-sumed that the elevated whole blood Mn level (676 nmol/L,reference range: 80–240 nmol/L7) found at 28 wk gestation, wasthe result of the 18 wk of TPN therapy, which included Mnsupplementation at a dosage of 18mmol/d for the first 21⁄2 wk and6 mmol/d thereafter. However, the whole blood Mn level contin-ued to rise to 1010 nmol/L at 35 wk gestation (25 wk of TPNtherapy) despite ceasing Mn supplementation at 31 wk gestation,and no evidence of hepatic cholestasis, throughout the TPN ther-apy period. At parturition (37 wk gestation) the whole blood Mnlevels fell to 556 nmol/L and further dropped to 422 nmol/L by 2mo post partum.

These high levels were also observed in the 10-d-old neonate(Mn 5 1090 nmol/L). However, by 2 mo this had dropped downto 216 nmol/L, which was within the reference range. A follow-upat 5 mo post partum revealed the mother and the baby to be well,the baby exhibiting no signs of developmental problems. Nofurther Mn levels were taken at this stage.

To date, there is no set reference range for whole blood Mnlevels during pregnancy or in the neonate. A cross sectional studyundertaken by our hospital of 19 females at 34 wk gestationattending an outpatient antenatal clinic showed: 68.5% (n 5 13)were within the reference range, and the mean level was 221.0698.1 nmol/L. Of those outside the reference range, only one wasmore than twice the upper level.

The major recognized features of Mn toxicity are related to thecentral nervous system, where high whole blood Mn levels areassociated with an abnormal high intensity lesion on T1-weightedmagnetic resonance images in the basal ganglia.8 The toxicitysyndrome is characterized by a Parkinsonian-type disorder result-ing from damage to the extra-pyramidal system and central cate-cholamine depletion.9 Liver damage has also been reported whereMn concentrations reached four times the upper reference range.10

This study was undertaken to monitor whole blood Mn levelsthroughout pregnancy and in the neonate during the initial postpartum period to determine whether the existing reference rangesfor whole blood Mn are appropriate for these population groups.

PATIENTS AND METHODS

Thirty-four pregnant females were randomly recruited from theGold Coast Hospital antenatal clinic. Whole blood Mn levels weredetermined for each female on three separate occasions during thetrial, at the first visit to the clinic (10–20 wk), at 25 wk, and at 34wk gestation. Mn levels were also measured in the capillary bloodof the babies at neonatal screening (3–4 d old).

All blood was collected in low trace element EDTA (Na2)blood collection tubes (Becton Dickinson Vacutainer, FranklinLakes, NJ, USA). Samples were refrigerated and forwarded to thereference laboratory in an esky. Mn was analyzed by atomicabsorption spectrometry using a graphite furnace and Zeemanbackground correction. Within-run precision was established byrepeated sampling (n) of three whole blood controls within the onebatch. As shown in Table I, the coefficient of variation (CV)ranged from 5% at Mn concentrations of 730–935 nmol/L to 8%at 425 nmol/L. Between-run precision was established by analyz-ing three aqueous standards (S1, S2, and S3) once with each batchof Mn assays for “n” batches. As shown in Table II, the CV rangedfrom 6% at a mean Mn concentration of 271 nmol/L to 8.7% at187 nmol/L, and 9.2% at 86 nmol/L.

Statistical Analysis

A pairedt test was used to test whether there was a significantdifference between whole blood Mn levels at 10–20 wk and 25 wkgestation; between 25 and 34 wk gestation; and between 34 wkgestation and that of the neonate.

RESULTS

There was a significant increase in the mean whole blood Mnlevels from 10 to 20 wk (150.46 53.4 nmol/L) to 25 wk (171.6659.7 nmol/L) to 34 wk gestation (230.06 67.8 nmol/L), (P ,0.0025 andP , 0.0005 respectively). This level appears toremain largely within the currently quoted reference range of80–240 nmol/L up until 34 wk gestation at which stage 30.4%(n 5 7) are outside this range. However, in the neonate, there wasan average 3.2 fold increase (range5 1.5–7.7) in mean Mn levels(Mn 5 737.76 209.7 nmol/L,P , 0.0005),(Table III).

DISCUSSION

This study of a cohort of healthy adult females has confirmedthe results of the earlier cited case report that there is a significantincrease in whole blood Mn levels throughout pregnancy. Theproportion of females whose levels were outside the quoted ref-erence range at 34 wk gestation was 30.4% (n 5 7). This figurecompares favourably with that of the cross sectional study of 19females at 34 wk gestation where 31.5% (n 5 6) were outside thisrange.

This increase in Mn levels may be related to the acceleratederythropoiesis associated with pregnancy which results in a 17–25% increase in red cell mass.11 This increase may be even greaterif iron supplements are taken throughout the pregnancy period.However, any increase in red cell mass is normally masked due toa disproportionate increase in plasma volume which begins in the

TABLE I.

WITHIN-RUN PRECISION FOR WHOLE BLOOD MNDETERMINATION USING ZAAS IN THE PERKIN ELMER Z/3030

Level nMean Mn level

(nmol/L) SD (nmol/L) CV (%)

Low 33 425.0 35.0 8.0Medium 33 730.0 37.0 5.0High 33 935.0 43.0 5.0

CV, coefficient of variation; MN, manganese; SD, standard deviations;ZAAS, Zeeman atomic absorption spectroscopy.

TABLE II.

BETWEEN-RUN PRECISION FOR WHOLE BLOOD MNDETERMINATION USING ZAAS IN THE PERKIN ELMER Z/3030

nMean Mn level

(nmol/L) SD (nmol/L) CV (%)

S1 27 187.0 16.2 8.7S2 25 86.0 7.9 9.2S3 26 271.0 16.3 6.0

CV, coefficient of variation; MN, manganese; SD, standard deviations;ZAAS, Zeeman atomic absorption spectroscopy.

WHOLE BLOOD MANGANESE LEVELS IN PREGNANCY AND THE NEONATE732

sixth week of pregnancy and reaches a peak value at 24 wkgestation.11 As a result, pregnancy is normally associated with adilutional anaemia which is characterized by a reduction in thehematocrit and hemoglobin concentration. This phenomenom wasevident among those members of this study group who had serialfull blood counts. Since most of the Mn (;66%) is located on theerythrocytes,12 it would therefore be expected that the whole bloodMn level would be reduced as the pregnancy progressed. Anotherexplanation for the increasing Mn levels is related to the obser-vation of enhanced Mn absorption in iron deficient individuals.13

Reduced iron status in pregnancy and particularly late pregnancymay lead to increased uptake of dietary Mn due to an up-regulatediron absorption, since the intestinal transport mechanism for ironis unable to differentiate between iron and Mn.14 The wide rangeof whole blood Mn levels observed in this study population maybe reflective of different dietary habits and varying degrees of irondeficiency status.

Concern was expressed by the offsite chemical pathologist asto whether the observed three fold increase in the mean Mn levelsin the neonate as compared to the 34 wk gestation female mayhave been related to possible skin contamination due to the dif-ference in blood collection methods (capillary heel prick collec-tion versus venepuncture). This was investigated by analysis ofcord blood taken randomly from 10 babies born at the Gold CoastHospital. The results showed a mean whole blood Mn level of732.3 6 165.2 nmol/L, which confirmed the higher levels ob-served in this study.

Hatano et al.15 also showed similar findings in a cross sectionalstudy of a healthy Japanese population, where the erythrocyte Mnlevel in cord blood was three times as high as in female adults(400 6 67 ng/g Hb versus 1416 20 ng/g Hb respectively). Thishigh level was also observed at 1 mo of age (435.16 118.7 ng/gHb), but decreased significantly after 5 wk.

Alarcon et al.16 also showed a progressive fall with increasingage in serum Mn levels in a cross sectional study of 180 appar-ently healthy Venezuelan infants ranging from 5 d to 12 moold.These observations confirm that of the authors’ case report.

It is most likely that the maintenance of these high levels in theearly neonatal period is due to the high retention rates of Mn.17,18

It remains unclear whether these rates are related to the reducedrate of biliary excretion of Mn, which is the main excretory routefor Mn, due to immaturity of the neonatal liver or to the highaffinity of various organs for Mn in the early neonatal period.19

Previous studies in neonatal and suckling animals showed a re-duced capacity to excrete Mn until day 18–19 of life, particularlywhen given in low doses. However, at higher doses, the rate ofexcretion is enhanced.19 Bilirubin levels were not routinely mea-sured in the neonates of this study population unless they showedsigns of jaundice or had a history of hemolytic disease. However,three were done and, in each case, both total and direct bilirubinlevels were normal.

Although serial hematology investigations were not done withthe neonates, reference data shows that hemoglobin and hemato-crit levels in cord blood are approximately 18 and 24%. in adultfemales, respectively.20 During the first 1–3 d of life, these levelsrise to 132 and 141% in adult females, respectively. By 2 wk ofage, they are back down to cord blood levels. Thereafter, theselevels continue to drop and stabilize at 3–6 mo of age. This fall isbelieved to reflect a gradual adjustment to improved blood oxy-genation that results from the replacement of the placenta by thelungs as a source of oxygenation.11 It is, therefore, possible thatthe rise in Mn levels in the neonate are, in part, a result of thesechanges in hematological indices.

Hatano et al.21 also observed that during the 1–5 wk period ofage when Mn levels were higher, the concentration of Mn inerythrocytes was significantly higher in formula fed than breastfed babies (P , 0.001). At thetransitional and weaning periods,it was still higher in formula fed babies, but the difference was nolonger statistically significant. These higher levels in formula fedbabies are believed to be due to the higher levels of Mn in infantfomulas, notably so in soy-based formulas.14 In Australia, Mn isadded to formulas as Mn sulphate, and the levels in standard feedsvary between 34.6 to 169.0mg/L. These levels are even higher forspecialist feeds (personal communication with Mead Johnson,Australia). This compares to 6.2mg/L quoted by Dorner et al.18 forthe mean Mn concentration of 2339 samples of breast milk.

In this study, 81.8% (n 5 18) were breast fed at the time ofblood collection (refer to Table III). It is unlikely at this stage (day3–4), that there would be a noticeable increase in Mn levels in theformula fed neonates. The mean Mn levels for the two groups,breast fed versus formula fed, was 712 nmol/L and 680 nmol/Lrespectively.

TABLE III.

WHOLE BLOOD MN LEVELS (NMOL/L)* THROUGHOUTPREGNANCY AND IN THE NEONATE

Studygroup

10–20 wkgestation

25 wkgestation

34 wkgestation

Neonate(3–4 d old)

1 190 184 201 NA2 123 188 NA 8583 149 150 202 7354 168 237 256 NA5 129 NA 154 NA6 175 NA 284 6157 159 158 227 5768 230 231 278 NA9 135 188 NA 480

10 195 200 NA NA11 110 108 182 81512 114 122 182 NA13 122 143 222 79014 101 173 NA 430†15 181 193 307 119516 136 137 137 NA17 99 113 133 795†18 190 235 280 NA19 120 NA 218 88520 188 245 344 NA21 366 392 422 63222 144 167 224 852†23 81 90 NA 47624 NA NA NA 59425 182 177 247 59426 188 152 224 645†27 121 138 163 125028 94 118 NA 77629 69 NA NA 57230 134 194 NA NA31 144 173 NA NA32 130 106 198 75533 145 118 204 NA34 150 145 NA 910

* Reference range: 80–240 nmol/L (see ref. 7).† Bottle fed.NA, not available.

WHOLE BLOOD MANGANESE LEVELS IN PREGNANCY AND THE NEONATE 733

In conclusion, this study has shown a progressive increase inwhole blood Mn levels throughout pregnancy. In the 3–4 d oldneonate, there is an average three-fold increase in these levels. Itis unclear as to the mechanism and functional significance of theseincreases at this stage, though possible reasons have been dis-cussed. This work suggests that the current reference ranges forwhole blood Mn in pregnancy and the early neonatal period needto be reviewed.

ACKNOWLEDGMENTS

This study would not have been possible without the support ofthe following people: David Heathcoate, Senior Biochemist, GoldCoast Hospital; Dr. Glenn Gardener, Obstetric Medical Officer,Gold Coast Hospital; Dr. Alan Clague, Chemical Pathologist,Royal Brisbane Hospital; Dr. Deryck Charters, Director of Ob-stetrics, Gold Coast Hospital; and Ms. Sue Neil, Pathology Nurs-ing Coordinator, Gold Coast Hospital.

REFERENCES

1. Tyrala EE. Trace elements. In: Yu VYH, MacMahon RA, eds.Intravenousfeeding of the neonate. Melbourne: Edward Arnold, 1992:78

2. National Research Council recommended daily allowances, 9th ed.Washington,DC: National Academy of Sciences, 1980

3. American Medical Association, Departments of Food and Nutrition. Guidelinesfor essential trace element preparations for parenteral use. JAMA 1979;241:2051

4. Shenkin A, Wardle CA, Roberts NBR, et al. Manganese supplementation duringTPN. JPEN 1996;20:311

5. Kirby DF, Fiorenza V, Craig R. Intravenous nutrition support duringpregnancy—review. JPEN 1988;12:72

6. Spencer A, Stephens H, Richards B, et al. Hypermanganesemia in a long term TPNpatient withHyperemesis gravidarum—A case report. In:Proceedings of the 20thAnnual Scientific Meeting of the Australian Society of Parenteral and EnteralNutrition. Brisbane 28–30 April 1994

7. Pleban PA, Pearson KH. Determination of manganese in whole blood and serum.Clin Chem 1979;25:1915

8. Ono J, Harada K, Kodaka R, et al. Manganese deposition in the brain during longterm total parenteral nutrition. JPEN 1995;19:310

9. Cotzias GC, Papavasiliou MI, Tang LC, et al. Manganese and catecholamines.Adv Neurol 1974;5:235

10. Lustic S, Pitlik SD, Rosenfield JB. Liver damage in self-induced hypermanga-nesemia. Arch Intern Med 1982;142:405

11. Wintrobe MM, Lukens JN, Lee GR. In: Lee GR, Bithell TC, Foerster J, AthensJW, Lukens JN, eds.Wintrobe’s clinical haematology, ninth edition, volume 1.Lea & Febiger, 1993:733

12. Milne DB, Sims RL, Ralston NVC. Manganese content of the cellular compo-nents of blood. Clin Chem 1990;36:450

13. Thomson ABR, Olatunbosun D, Valberg LS. Interrelation of intestinal transportsystem for manganese and iron. J Lab Clin Med 1971;78:642

14. Lonnerdal B. Nutritional aspects of soy formula. Acta Paediatr 1994;402(suppl):105

15. Hatano S, Nishi Y, Usui T. Erythrocyte manganese concerntration in healthyJapanese children, adults and the elderly, and in cord blood. Am J Clin Nutr1983;37:457

16. Alarcon OM, Reinosa-Fuller JA, Silva T, et al. Manganese levels in serum ofhealthy Venezuelan infants living in Merida. J Trace Elem Med Biol 1996;10:210

17. Keen CL, Bell JG, Lonnerdal B. The effect of age on manganese uptake andretention from milk and infant formula in rats. J Nutr 1986;116:395

18. Dorner K, Dziadzka S, Hohn A, et al. Longitudinal manganese and copperbalances in young infants and preterm infants fed on breast milk and adaptedcows milk formulas. Br J Nutr 1989;61:559

19. Ballatori N, Miles E, Clarkson TW. Homeostatic control of manganese excretionin the neonatal rat. Am J Physiol 1987;252:R842

20. Lubin BH. In: Nathan DG, Oski FA, eds.Haematology of infancy and childhood,third edition, volume 2. Philadelphia: WB Saunders Co., 1987:1678

21. Hatano S, Aihara K, Nishi Y, et al. Trace elements (copper, zinc, manganese andselenium) in plasma and erythrocytes in relation to dietary intake during infancy.J Pediatr Gastroenterol Nutr 1985;4:87

WHOLE BLOOD MANGANESE LEVELS IN PREGNANCY AND THE NEONATE734