• No results found

Detection of Phenylketonuria in the Very Early Newborn Blood Specimen


Academic year: 2020

Share "Detection of Phenylketonuria in the Very Early Newborn Blood Specimen"


Loading.... (view fulltext now)

Full text



of Phenylketonuria

in the Very








RN, MS*;


J. Rohr,

MS, RD*;



L. Levy,


From the *Bjochemical Genetics Unit of the Division of Genetics, The Children’s Hospital; the Joseph P. Kennedy, Jr Laboratories of the Neurology Service, Massachusetts

General Hospital; fthe Department of Neurology, Harvard Medical School; and tthe

Newborn Screening Program, State Laboratory Institute, Massachusetts Department of

Public Health, Boston

ABSTRACT. Early hospital discharge ofnewborns is lead-ing to collection of the newborn screening blood specimen during the first day of life in increasing numbers of

newborns. There is concern that neonates with

phenyl-ketonuria who are tested this early may be missed. To

examine this question, the authors screened specimens collected during the first 24 hours oflife from 23 neonates at risk for hyperphenylalaninemia. The blood phenylal-anine level in each of the 6 neonates with phenylketo-nunia and a seventh with mild hyperphenylalaninemia was greater than 2 mg/dL as early as 4 hours of age and 6 mg/dL or greater by 24 hours of age. A newborn

screening phenylalanine cutoff level of 2 mg/dL would

have identified all of these neonates within the f,rst 24 hours of life, but a cutoff level of 4 mg/dL would have missed 2 of the 6 with phenylketonuria before 24 hours oflife. Newborn screening programs should adopt a blood

phenylalanine level of 2 mg/dL as the cutoff for suspicion

of phenylketonuria and request for a second specimen. Breast-fed affected neonates had higher early blood phen-ylalanine elevations than formula-fed neonates, perhaps reflecting the higher protein (phenylalanine) content of

colostrum. Pediatrics 1991;87:240-244; phenylketonuria,

newborn screening, early discharge, breast-feeding,


The filter paper blood specimen for routine new-born screening of phenylketonuria (PKU) and other disorders is usually obtained from the neonate at the time of nursery discharge. In the United States this has traditionally been on the third or fourth day oflife.’ By then PKU is readily identified on the basis of a distinct hyperphenylalaninemia,

Received for publication Apr 30, 1990; accepted Jul 11, 1990.

Reprint requests to (H.L.L.) Gardner 648, The Children’s

Hos-pital, 300 Longwood Aye, Boston, MA 02115.

PEDIATRICS (ISSN 0031 4005). Copyright © 1991 by the

American Academy of Pediatrics.

because the blood phenylalanine level in the phen-ylketonunic neonate rises progressively during the first few days after birth.2’3

There is now a trend toward earlier discharge from the newborn nursery; consequently newborn blood specimens are collected earlier. Discharge often occurs before the third day of life and, with increasing frequency, during the first day. In Ver-mont, for example, 4.7% of neonates are currently discharged before 24 hours of age (C Walters, PhD, written communication, January 16, 1990).

Several studies have concluded that the phenyl-alanine concentration will be below the cutoff level used for the detection of PKU in a substantial number of affected neonates when the blood speci-men is collected during the first 24 hours of life. Holtzman et al4 originally estimated that 16% of these neonates will be missed by newborn screening at a phenylalanine cutoff level of 4 mg/dL. A sub-sequent study by McCabe et al3 indicated that more than 30% of phenylketonuric neonates tested dur-ing the first 12 hours of life and nearly 10% tested from 12 to 24 hours of age will be missed at a cutoff level of 4 mg/dL. However, these estimates are predictions based on linear regression analysis4 or


scores of log transformed blood phenylalanine levels obtained by fluorometnic or column chromat-ographic methods.5 Few levels in either study were obtained from phenylketonunic neonates during first 24 hours of life, and very few were obtained by

testing a filter paper blood specimen as used in

routine newborn screening.


20 1 5 10 Blood phenyl-alanine (mg/dl) 5 0 S S PKU #{163}MHP 0 unaffected S S

. S S



. S

. S


0 0

. S 0

0 4 8 12 16 23 4 5

Hours of age Days of age

blood specimens are collected in the same manner

as in routine newborn screening, and the blood

phenylalanine levels are measured by the Guthnie

assay.5 Previously we reported the initial results.6 We now report experience in an additional seven affected neonates, six with PKU and one with non-PKU mild hyperphenylalaninemia. We also com-pane the early blood phenylalanine levels between breast-fed and formula-fed affected neonates.


We followed all pregnancies in mothers who had

children with PKU on non-PKU mild

hyperphen-ylalaninemia attending the PKU clinic. We an-ranged for collection of filter paper blood specimens

by heel puncture when the newborn was 4, 12, and

24 hours of age and at nursery discharge. The

specimens were mailed or otherwise delivered to

the screening laboratory and tested for PKU by the

Guthnie bacterial inhibition assay.5 Guthnie test

levels read as 2 to 4 mg/dL were designated as 3

mg/dL, 4 to 6 mg/dL as 5 mg/dL, and 6 to 8 mgI dL as

7 mg/dL.

Newborn medical records were obtained for each

affected neonate, and the feeding history was re-viewed for type, frequency, and volume. The protein content of breast milk was estimated from the data of Atkinson et al,7 and the phenylalanine content

of breast milk from Macy and Kelly.8 The protein

and phenylalanine contents of the infant formula were determined from product information.9 Inas-much as the volume consumed by the breast-fed neonates was not measured, we assumed that these

neonates ingested at each feeding a volume equal

to that ingested by the average formula-fed neonate.


During a 6-year period, 23 neonates at risk for

hyperphenylalaninemia were born. Six of these

neonates had PKU, and 1 had non-PKU mild

hy-perphenylalaninemia; each appeared to have the

same degree of metabolic defect as the sibling. The blood phenylalanine levels during the first day of life and subsequent days in the 7 affected neonates are plotted in the Figure.

The six neonates with PKU had blood phenylal-anine levels of 3 mg/dL or greater at either 4 hours

of age (five neonates) or 6 hours of age (one

neo-nate). At 12 hours of age, their levels ranged from 3 mg/dL to 7 mg/dL; by 24 hours of age, the blood phenylalanine concentration in each had risen, with a group range of 6 to 10 mg/dL. By the time of nursery discharge at 2 to 4 days of age, the blood phenylalanine concentration in each was 8 mg/dL on greater. The neonate with non-PKU mild

hyper-Figure. Blood phenylalanine levels obtained by the

Guthrie bacterial inhibition assay in filter paper

speci-mens collected during the first 24 hours of life and within

the first week from neonates at risk for phenylketonuria (PKU) or non-PKU hyperphenylalaninemia. MHP, mild hyperphenylalaninemia.

phenylalaninemia had a blood phenylalanine level of 5 mg/dL at age 4 hours and 7 mg/dL and 8 mgI dL at 12 hours and 24 hours, respectively. At the time of nursery discharge, his level remained at 8


Among the unaffected neonates, early blood

phenylalanine levels in most were less than 2 mgI

dL (Figure). Several of these neonates, however, had increased phenylalanine values during the first

24 hours of age, ranging from 3 mg/dL at age 4

hours to as high as 4 mg/dL at age 12 hours; these values overlap the lower increases in affected

neo-nates. All but one of the unaffected neonates had

normal (<2 mg/dL) values by 3 days of age; the exceptional neonate had a normal value at 7 days of age.

To determine the relationship between protein (phenylalanine) ingestion and early blood phenyl-alanine levels in the affected neonates, protein and phenylalanine intakes were estimated for each dun-ing the first 4 hours, 12 hours, and 24 hours of life (Table 1). At 4 hours of age, three neonates had

not yet fed, but their blood phenylalanine

eleva-tions were in the same range as the neonates who


phenylal-TABLE 1. Estimated Phenylalanine Intakes and Blood Phenylalanine Levels During the First 24 Hours of Life in Seven Affected Neonates

Patient Type of


Type of


4 h 12 h 24 h


Phenylalanine Blood .

Phenylalanine Blood Phenylalanine . Blood

Intake, Phenylalanine, Intake, Phenylalanine, Intake, Phenylalanine,

mg/kgt mg/dL mg/kgt mg/dL mg/kgt mg/dL

KM Classic Formula 5.5 3 12.3 3 23.2 6

JS Classic Formula 2.2 5 9.8 5 19.7 7

LO Atypical Formula 8.5 4 12.8 5 23.9 6


Classic Breast 0 3 8.0 3 23.9 8

KW Classic Breast 7.7 5 15.3 7 30.6 8

KG Classic Breast 0 4 16.4 6 32.9 10

CM MHP Breast 0 5 15.8 7 39.4 8

* PKU, phenylketonunia; MHP, mild hyperphenylalaninemia.

t Cumulative intakes 0-4 h, 0-12 h, and 0-24 h.

TABLE 2. Screening Results During the First


24 Hours of Age in Neonates With

Blood Phenylalanine Level, mg/dL Reference

<2 2 3t 4


2 1

1 2 2





12 3

Present study

* Results are numbers of neonates with the given blood phenylalanine level during the

first 24 hours of age.

t Interpolation of Guthrie 2 to 4 mg/dL.

anine level was among the lowest at 12 hours of age but nose considerably by 24 hours of age.


In this prospective study ofearly newborn screen-ing for PKU, the blood phenylalanine level was above 4 mg/dL by 24 hours of life in each of the six neonates with PKU as well as in the neonate with non-PKU mild hyperphenylalaninemia. Even within the first 12 hours of life, four of the six neonates with PKU and the neonate with mild

hyperphenylalaninemia had blood phenylalanine

levels greaten than 4 mg/dL. Two neonates with PKU, had blood phenylalanine levels during the first 12 hours of life between 2 and 4 mg/dL. Thus, a cutoff level of 4 mg/dL would have missed those two neonates had they been screened within the first 12 hours of life, as predicted by McCabe et al,3 but would not have missed any of the neonates if screened at 24 hours of life. A cutoff level of 2 mgI dL, however, would have identified all of the neo-nates with PKU in our study even as early as within the first 12 hours of life.

Blood phenylalanine levels obtained by newborn screening methods during the first 24 hours of life,

including those in this study, have been

docu-mented for 32 neonates with either PKU or a less severe degree of hyperphenylalaninemia (Table 2).

In 27 the phenylalanine level was greater than 2 mg/dL. Three of the five exceptions were included in the study by McCabe et al,3 and the other two in the study by Szeinbeng and Cohen!2 The blood

phenylalanine level was below 2 mg/dL in three of

these exceptions and at 2 mg/dL in the remaining two. Five other neonates among the 32 had blood phenylalanine levels between 2 mg/dL and 4 mgI dL. Thus, in 31% of early screened neonates with PKU, the blood phenylalanine level during the first

24 hours of life has been recorded at below 4 mgI

dL, and in 10% at below 2 mg/dL.

The blood phenylalanine levels in the breast-fed neonates in our study tended to rise more rapidly during the first 24 hours of life than in the formula-fed neonates (Table 1). This was true for the neo-nate with mild hyperphenylalaninemia as well as

the neonates with PKU, so it cannot be explained

simply by the degree of the metabolic block. The

likely explanation is the relatively high concentra-tion of phenylalanine in colostrum, the breast milk produced during the first three postpartum days. The protein content of colostrum is 2.3 g/dL,7 and the phenylalanine content is 105 mg/dL,8 both con-siderably higher than values of 1.5 g/dL and 57 mg/ dL, respectively, for the proprietary formula used.#{176}

Assuming equal volumes of colostrum and formula

consumed, therefore, the breast-fed neonates


the formula-fed neonates during the first 24 hours of life. Our assumption of 30 mL (1 oz) of breast

milk consumed per feeding during the first day of

life is within the range of 20 to 100 mL per feeding reported by Lawrence.’#{176}

Another possible explanation for the more rapid rise in the blood phenylalanine level among breast-fed neonates is a lower energy intake producing a greater degree of catabolism and proteolytic release of phenylalanine. This is suggested by the greater degree of weight loss in the breast-fed neonates we studied. The energy content of colostrum is 208 kJ/ dL, whereas the infant formula contains 279 kJ,1 dL.1’

It has been reported that breast-feeding delays the rise of blood phenylalanine level in the neonate with PKU.’3 This may be true after the first weeks

of life when mature breast milk is consumed,

be-cause the protein and phenylalanine concentrations

of breast milk gradually decline and the energy

content rises during the first month postpar-turn.7”#{176}” Our data indicate that during the first few days of life, however, a breast-fed neonate with PKU is perhaps more likely than a formula-fed neonate to be identified by newborn screening, even

as early as 24 hours of life. This experience is

consistent with that reviewed by Berry and

Pon-ten,’4 who concluded that PKU can be readily

de-tected at 24 hours of age even in breast-fed neonates because of the higher phenylalanine level in cobs-trum.

With the growing trend of early discharge for

mothers and their newborns, there is a major

con-cern that newborns with metabolic disorders, for which newborn screening is conducted, will be

missed. In that context the data, with regard to

early screening for PKU, seem to be indicating several conclusions.

First, the great majority of neonates with PKU

will have an elevated blood phenylalanine level within the first 24 hours of life. This elevation may be less than 4 mg/dL, however, which is the cutoff level often used in screening. A cutoff level of 2 mg/dL for phenylalanine in identifying neonates for follow-up should detect more than 90% of those with PKU screened during the first 24 hours of life, but a cutoff level of 4 mg/dL would probably miss

30% of these neonates. These conclusions are

con-sistent with the predictions of McCabe et al.3

Con-sequently, as recommended by McCabe et al,3 a

cutoff level of 2 mg/dL should be used in screening for PKU. We would go further and recommend

obtaining second specimens from all neonates

whose blood phenylalanine level is 2 mg/dL or greater, not only from those with a level >2 mg,/dL. The fear that this lower cutoff level will greatly

increase the number of requests for second

speci-mens and overextend the capabilities of the

screen-ing program is unfounded. In Virginia lowering the

cutoff from 4 mg,/dL to >2 mg,/dL resulted in

increasing the rate of requested additional

speci-mens from 0.01% to only 0.03%.’ The higher rate

still represents only 30 requested additional

speci-mens among 100 000 screened newborns. In the

New England Newborn Screening Program, the

recall rate is 0.15% at a cutoff of 2 mg/dL for phenylalanine (Simmons JR, Levy HL;

unpub-lished data, June 30, 1990). Unfortunately, only

nine state newborn screening programs presently use 2 mg/dL or >2 mg/dL as the cutoff phenylala-nine level (Council of Regional Networks for

Ge-netic Services; Comprehensive 1988 Newborn

Screening Report, January 1990).

Second, neonates with non-PKU mild hyper-phenybalaninemia may more often have a normal blood phenylalanine level during the first day of life, and, consequently, may more likely be missed

by newborn screening at that time than neonates

with PKU. Although this was not true for the one such neonate in the present study, it was so for the one neonate with mild hyperphenylalaninemia in our initial group of studied neonates.6 Furthermore, we recently encountered a neonate with mild

hy-perphenylalaninemia in routine screening for PKU

whose blood phenylalanine level was <2 mg,/dL at

25 hours of age when the initial screening specimen was collected. This neonate was identified on the

basis of a phenylalanine level of 8 mg,/dL in a

follow-up specimen collected at 2 weeks of age

(Simmons JR, Levy HL; unpublished data,

Novem-ben 25, 1989). Although non-PKU mild hyperphen-ylalaninemia seems to be benign,’6”7 the ptenin defects with secondary hyperphenylalaninemia may also have a delayed rise in the blood phenyl-alanine level18 and are important to identify in the newborn.

Third, breast-feeding does not result in a delayed rise of the blood phenylalanine level in the neonate with PKU during the first 24 hours of life. To the contrary, the breast-fed neonate tends to have a somewhat higher level of blood phenylalanine dun-ing the first day of life than the formula-fed neo-nate. Thus, breast-feeding seems to offer some pro-tection against missing PKU in the very early new-born specimen.


PKU’9 and a slower clearance of phenylalanine among some hetenozygotes. However, this should not produce a large number of requests for addi-tional specimens for a cutoff level of 2 mg/dL among neonates tested during the first 24 hours of life even at the PKU heterozygote frequency of 2% in the general population, inasmuch as the number of neonates initially tested during the first day of life is still quite low.

This experience in early screening for PKU

should not be misinterpreted as an endorsement for

collecting the newborn blood specimen during the

first day of life. Indeed, not only is early PKU

screening less reliable than in specimens collected later in the newborn period, but the detection of

other disorders may also be compromised with the

very early specimen.2#{176} On the other hand, it is essential that no newborn leave the hospital

with-out a screening blood specimen having been

col-lected, regardless of how early discharge occurs. This may be the only opportunity for the screening specimen to be collected. It is not appropriate med-ical cane to rely on return of the neonate at a later time for collection of the initial newborn specimen. In addition, most disorders for which newborn screening is conducted will be detected by means of a very early specimen. However, the uncertainty of this requires the collection of a second screening blood specimen from all neonates whose initial

specimen is obtained during the first 24 hours of



This work was supported by grant NS-05096 and

con-tract NO1-HO-42809 from the National Institutes of


We thank Deborah Lobbregt for coordinating the col-lection of blood specimens, Deborah Frederick for secre-tanial and editorial assistance, and Jane Simmons for testing the blood specimens.


1. Holtzman NA, Meek AG, Mellits ED, Kallman CH.

Neo-natal screening for phenylketonuria, III: altered sex ratio: extent and possible causes. Pediatrics. 1974;85:175-181

2. Holtzman NA, Mellits ED, Kailman CH. Neonatal screening for phenylketonuria, II: age dependence of initial

phenylal-anine in infants with PKU. Pediatrics. 1974;53:353-357

3. McCabe ERB, McCabe L, Mosher GA, Allen RJ, Berman

JL. Newborn screening for phenylketonuria: predictive

va-lidity as a function of age. Pediatrics. 1983;72:390-398

4. Holtzman NA, McCabe ERG, Cunningham GC, Berry HK. Screening for phenylketonuria. N Engi J Med.


5. Guthrie R, Susi A. A simple phenylalanine method for

detecting phenylketonuria in large populations of newborn

infants. Pediatrics. 1963;32:338-343

6. Meryash DL, Levy HL, Guthrie R, Warner R, Bloom S, Carr JR. Prospective study of early neonatal screening for phenylketonuria. N Engi J Med. 1981;304:294-296

7. Atkinson SA, Bryan MH, Anderson, GH. Human milk: difference in nitrogen concentration in milk from mothers of term and premature infants. J Pediatr. 1978;93:67-69

8. Macy IG, Kelly HJ. Human Milk and cow’s milk in infant

nutrition. In: Kon SK, Cowie AT, eds. Milk: The Mammary

Gland and Its Secretion. New York, NY: Academic Press

Inc; 1961;2:265

9. Pediatric Products Handbook. Evansville, IN: Mead Johnson Nutritional Division; 1986

10. Lawrence RA. Breast-feeding, a Guide for the

MedicaiProfes-sion. St Louis, MO: CV Mosby; 1980

11. Anderson GH, Atkinson SA, Bryan MH. Energy and mac-ronutrient content of human milk during early lactation from mothers giving birth prematurely and at term. Am J Clin Nutr. 1981;34:258-265

12. Szeinberg A, Cohen BE. Early blood sampling in neonatal

programs for the detection of phenylketonuria. Padiatr u

Padol. 1982;17:287-292

13. Binder J, Johnson CF, Saboe B, Krug-Wispe S. Delayed elevation of serum phenylalanine level in a breast-fed child. Pediatrics. 1979;63:334-336

14. Berry HK, Porter UI. Newborn screening for

phenylketo-nuria. Pediatrics. 1982;70:505-506

15. Weiner DL, Canton RM, Mitchell PL, Mamunes P. False positive rate in neonatal phenylketonuria (PKU) screening using a 2 mg/dl phenylalanine (PA) cutoff. Am J Hum Genet. 1981;33:94A

16. Levy HL, Shih VE, Karolkewicz V, et al. Persistent mild

hyperphenylalaninemia in the untreated state: a prospective study. N Engi J Med. 1971:285:424-429

17. Lang MJ, Koch R, Fishier K, Baker R. Nonphenylketonuric hyperphenylalaninemia. AJDC. 1989:143:1464-1466 18. Schaub J, Daumling S, Curtius H-Ch, et al.

Tetrahydrobi-opterin therapy of atypical phenylketonuria due to defective dihydropterin biosynthesis. Arch Dis Child. 1978;53:674-683 19. Scriver CR, Cole DEC, Houghton SA, Levy HL, Grenier A,

Laberge C. Cord-blood tyrosine levels in the full-term

phen-ylketonuric fetus and the ‘justification hypothesis.’ Proc

Nati Acad Sci USA. 1980;77:6175-6178

20. Committee on Genetics, American Academy of Pediatrics. Newborn screening fact sheets. Pediatrics. 1989;83:449-464 21. Committee on Genetics, American Academy of Pediatrics.

New issues in newborn screening for phenylketonuria and congenital hypothyroidism. Pediatrics. 1982;69:104-106

22. Levy HL, Mitchell ML, Ridley SE. Newborn screening.




Lauren Barnico Doherty, Frances J. Rohr and Harvey L. Levy

Detection of Phenylketonuria in the Very Early Newborn Blood Specimen


Updated Information &


including high resolution figures, can be found at:

Permissions & Licensing


entirety can be found online at:

Information about reproducing this article in parts (figures, tables) or in its






Lauren Barnico Doherty, Frances J. Rohr and Harvey L. Levy

Detection of Phenylketonuria in the Very Early Newborn Blood Specimen


the World Wide Web at:

The online version of this article, along with updated information and services, is located on

American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.


Related documents

In this study, a fuzzy knowledge based intelligent model has been developed based on dye concentration, dyeing time and process temperature as input variables and color

In the impact test the energy needed to fracture the element was measured i.e. the ability of a material to withstand fracture was found. In this test specimen

In this work, we focus on buffer overflow, format string, random number generation, shell, race condition vulnerabilities.. Using the three tools, we conduct

Background: Councils of Chiropractic Education (CCE) indirectly influence patient care and safety through their role of ensuring the standards of training delivered by

Also, both diabetic groups there were a positive immunoreactivity of the photoreceptor inner segment, and this was also seen among control ani- mals treated with a

MMRDA has planned to get fund from Japan International Cooperation Agency (JICA) for the construction of Mumbai Metro Line III. The scope of the study is framed as per JICA

Regarding the common antioxidant capacity of walnuts (obtained as a mean of all varieties), ORAC assay results in higher antioxidant capacity ( p ≤ 0,05) of the walnuts

The Pearson r-coefficients were calculated and the Analysis of Variance (ANOVA) method was applied to establish the statistical significance levels (p- values) of the effect