fatty acid profile in patients with phenylketonuria and its relationship with bone mineral density

RESEARCH REPORT

Fatty acid profile in patients with phenylketonuriaand its relationship with bone mineral density

Sergio Lage & María Bueno & Fernando Andrade &

José Ángel Prieto & Carmen Delgado & María Legarda &

Pablo Sanjurjo & Luis Jose Aldámiz-Echevarría

Received: 24 May 2010 /Revised: 21 July 2010 /Accepted: 6 August 2010# SSIEM and Springer 2010

AbstractBackground Patients with phenylketonuria (PKU) undergo arestrictive vegan-like diet, with almost total absence of n-3fatty acids, which have been proposed as potential contrib-utors to bone formation in the healthy population. The PKUdiet might lead these patients to bone mass loss and,consequently, to the development of osteopenia/osteoporosis.Therefore, we proposed to analyze their plasma fatty acidprofile status and its relationship with bone health.Methods We recruited 47 PKU patients for this cross-sectionalstudy and divided the cohort into three age groups (6–10 years,11–18 years, 19–42 years).Wemeasured their plasma fatty acidprofile and bone mineral density (BMD) (both at the femoralneck and the lumbar spine). Seventy-seven healthy controls alsoparticipated as reference values of plasma fatty acids.

Results Docosahexaenoic acid (DHA) and eicosapentae-noic acid (EPA) and total n-3 fatty acids were signifi-cantly diminished in PKU patients compared with healthycontrols. DHA, EPA, and total n-3 fatty acids were alsopositively associated with bone mineral density (r=0.83,p=0.010; r=0.57, p=0.006; r=0.73, p=0.040, respective-ly). There was no association between phenylalanine (Phe),Index of Dietary Control (IDC), calcium, 25-hydroxivitaminD concentrations, daily calcium intake, and BMD.Conclusion Our results suggest a possible influence ofessential fatty acids over BMD in PKU patients. The lack ofessential n-3 fatty acids intake in the PKU diet might affectbone mineralization. Further clinical trials are needed toconfirm the effect of the n-3 essential fatty acids on boneaccrual in a cohort of PKU patients.

Communicated by: John H. Walter

Competing interest: None declared.

S. Lage : F. Andrade : J. Á. Prieto :M. Legarda : P. Sanjurjo :L. J. Aldámiz-Echevarría (*)Division of Metabolism, Department of Paediatrics,Cruces Hospital,Plaza de Cruces s/n,48903 Barakaldo, Vizcaya, Spaine-mail: [email protected]

S. Lagee-mail: [email protected]

F. Andradee-mail: [email protected]

J. Á. Prietoe-mail: [email protected]

M. Legardae-mail: [email protected]

P. Sanjurjoe-mail: [email protected]

M. BuenoDepartment of Paediatrics, Virgen del Rocío Teaching Hospital,Avda. Manuel Siurot s/n,41013 Sevilla, Spaine-mail: [email protected]

P. SanjurjoFaculty of Medicine and Surgery, Basque Country University,Barrio de Sarriena s/n,48940 Leioa, Vizcaya, Spain

C. DelgadoMetabolic Pathologies Laboratory,Virgen del Rocío Teaching Hospital,Avda. Manuel Siurot s/n,41013 Sevilla, Spaine-mail: [email protected]

DOI 10.1007/s10545-010-9189-0J Inherit Metab Dis (201 ) 3 (Suppl 3):S363–S3710 3

/ Published online: 10 September 2010

AbbreviationsBMD bone mineral densityBMI body mass indexDHA docosahexaenoic acidEPA eicosapentaenoic acidLC-PUFA long-chain polyunsaturated fatty acidMUFA monounsaturated fatty acidPKU phenylketonuriaPUFA polyunsaturated fatty acidRDA recommended dietary allowancesSD standard deviationSFA saturated fatty acid

Introduction

Patients with phenylketonuria (PKU) represent a vulnerablegroup who must be carefully monitored for life. Nutritionalfollow-up is needed to maintain phenylalanine (Phe) withinsafety levels and thus prevent neurological damage whileensuring that neurological and physical development isadequate by avoiding the lack of any essential nutrient dueto a deprived diet. The dietary intervention must beimplemented as early as the diagnosis is made and comprisea strict vegetarian diet, with almost total absence of anynatural protein source. These patients are fed on a syntheticprotein substitute (Phe-free amino acid mixtures) andspecial low-protein foods. Despite the increasing varietyof these supplements (Feillet and Agostoni 2010), thedietary intervention is established by selecting one of thesupplements according to patients’ requirements. Thesesynthetic nutritional supplements also provide PKU patientswith vitamins, minerals, and trace elements, which permitthem to reach daily energy and micro- and macronutrientrequirements.

The vegan-like diet prescribed to treat PKU may some-times lead to transient growth retardation (Dobbelaere et al.2003, Dhondt et al. 1995, Schaefer et al. 1994, Verkerk et al.1994) compared with their age- and gender-matched healthycontrols. Concerning skeletal development, the literaturereports various studies focused on analyzing bone health ofthese individuals (Modan-Moses et al. 2007, Pérez-Dueñas etal. 2002, Przyrembel and Bremer 2000, Zeman et al. 1999,Al-Qadreh et al. 1998, Hillman et al. 1996, Allen et al. 1994,McMurry et al. 1992, Carson et al. 1990). All studies found adefect in bone mineralization that caused the appearance ofosteopenia/osteoporosis.

Several factors may lead to the development of osteopenia/osteoporosis: lack of physical activity, deficient nutrientintake, inflammation, and genetic factors. Previous reportsregarding the influence of nutrition over bone accrual in thehealthy population showed an association between fatty acid

intake and bone mineral density (BMD) (Högström et al.2007, Eriksson et al. 2009, Weiss et al. 2005). Thus, n-3long-chain polyunsaturated fatty acids (LC-PUFAs), especiallydocosahexaenoic and eicosapentaenoic acids (DHA and EPA,respectively), enhanced bone formation, and n-6 LC-PUFAscaused bone resorption due to an increase in osteoclastactivity.

Several studies focused on analyzing the fatty acidpattern of PKU patients (PKU; OMIM 261600) undertreatment. Some of them showed a clear depletion ofplasma (Beblo et al. 2007, Galli et al. 1991) or both plasmaand red blood cell n-3 LC-PUFA levels (especially DHAand EPA) (Moseley et al. 2002, Acosta et al. 2001, vanGool et al. 2000, Sanjurjo et al. 1994). As a result, their n-6to n-3 fatty acid ratios were higher than those for healthycontrols. This fact can be viewed as a consequence ofthe restrictive diet followed by these patients, withalmost total absence of natural proteins (fish representsthe main natural source of n-3 fatty acids) and of fatbeing predominantly supplied as plant oils, which containalmost no preformed PUFAs of the n-3 series and highlevels of n-6 fatty acids. In contrast, other authors reportednormal levels of n-3 and n-6 LC-PUFAs in treated PKUpatients (Lavoie et al. 2009, Pöge et al. 1998). Bearing inmind these precedents, our study aimed at evaluating fattyacid profile and bone health of PKU patients, analyzingthe relationship between PUFAs (especially DHA andEPA) and n-6 to n-3 fatty acid ratio and osteopenia/osteoporosis.

Methods

Participants

Forty-seven PKU patients (30 male and 17 female) wereenrolled in this cross-sectional study, which was carried outsimultaneously at Cruces Hospital (Bilbao) and Virgen delRocío Hospital (Sevilla). Bearing in mind the wide agerange of our cohort, we decided to divide it into three agegroups: group 1 (6–10 years), group 2 (11–18 years), andgroup 3 (19–42 years). These patients were diagnosedeither through a neonatal screening program or through lateclinical diagnosis, and a definitive diagnosis was confirmedby genetic analysis. According to the plasma Phe concen-tration measured when the diagnosis is made, PKU can beclassified into three subtypes: classic (Phe >1,200 μmol/L),moderate (Phe 600–1,200 μmol/L), and mild (Phe 360–600 μmol/L). Within our cohort, there were 26 patientswith classic, 11 with moderate, and 10 with mild PKU.None of the patients presented neurological damage.They all were able to function independently in theirdaily activities. None suffered from any other disease or

J Inherit Metab Dis (201 ) 3 (Suppl 3):S363–S3710 3 S364

exposure to medication known to affect bone mineralhealth. Patients undergoing tetrahydrobiopterin (BH4)treatment were excluded, as this treatment allows liberal-ization of the protein-restricted diet at least up to thenormal World Health Organization (WHO) recommenda-tions for age and sex, which entails normalization ofessential PUFA concentrations.

Seventy-seven healthy controls undergoing minorsurgery (65 male and 12 female) were also enrolled in thisstudy to obtain reference values of plasma fatty acids. Thehealthy volunteers ranged in age from 6 to 45 years.Participants were excluded if they smoked, were takingmedication known to alter plasma lipids, or suffered fromany eating or feeding disorder that led them to consumeinadequate amounts of essential fatty acids. The study wasperformed in accordance with the Declaration of Helsinki.Informed consent was obtained from all participants, andthe Ethics Committee at both hospitals approved the studyprotocol.

Nutritional assessment

All participants were asked to fill in a 3-day food diaryrecording the amount of each item of food and drinkconsumed during the 3 days prior to the densitometric andbiochemical study. Thus, the dietician calculated the dailyaverage intake of fat, carbohydrate, energy, and calcium.PKU patients were fed on a nonnatural protein source basedon a Phe-free amino acid mixture, depending on the dietaryPhe tolerance of each of them, which was averaged over aperiod of 6 months. Therefore, this nonnatural proteinsource contributed to 15–93% of the daily total proteinintake. This formula also provided adequate amounts ofvitamins, minerals, and oligoelements. Vegetables and fruitsrepresented the natural protein source. Patients receivedneither n-3 nor n-6 PUFA supplementation from the timetheir treatment was implemented. Plasma Phe concentrationwas the biochemical parameter of choice to evaluatecompliance with the diet. Thus, we defined the cutofflimits for this biochemical parameter to be considered to bewithin safety levels. For patients younger than 18 years, thetarget value of plasma Phe was <360 μmol/L (CampistolPlana et al. 2006); for patients 18 years or older, the targetvalue was <600 μmol/L (Campistol Plana et al. 2006). TheIndex of Dietary Control (IDC) was calculated as themedian Phe levels from the year of the study (minimum of12 Phe values). We also considered Phe level at the momentof densitometric and biochemical study.

Anthropometric measurements

Standing height was measured with a wall-mountedstadiometer. Body weight was measured with a digital

scale with a sensitivity of 100 g. Patients were weighedbarefoot and after overnight fasting. Height, weight, andBMI Z-scores were calculated for each patient by compar-ing them with those of the Spanish reference population(Hernández et al. 2000).

Bone mineral density measurements

BMD at the lumbar spine (L2–L4) and femoral neck weremeasured by dual-energy X-ray absorptiometry (DEXA;Hologic QDR 4500 W v9.8 Elite, Woltham, MA, USA).The same instrument was used at both hospitals. Resultswere expressed in terms of Z-scores [difference betweenBMD of the patient and average BMD of age- and sex-matched controls divided by the standard deviation (SD) ofthe control group] for comparison of patients with young–normal and age-matched populations of the same race andsex. According to the WHO guidelines, a Z-score of –2.5or below implies the presence of osteoporosis, whereas a Z-score of –1 to –2.5 indicates osteopenia (Kanis 1994) in theadult population.

In the case of children and adolescents, the diagnosis ofosteoporosis requires the presence of both a clinicallysignificant fracture history and low bone mineral content(BMC) or BMD. A clinically significant fracture historycan be defined in terms of one or more of the following(Rauch et al. 2008):

& Long-bone fracture of the lower extremities& Vertebral compression fracture& Two or more long-bone fractures of the upper extremities& Low BMC or BMD is defined as a BMC or areal BMD

Z-score less than or equal to –2, adjusted for age,gender, and body size

Biochemical measurements

All biochemical measurements were obtained from fastingmorning plasma samples. Plasma calcium was analyzed usinga photometric assay (modular photometer D-2400) in whichthe working wavelength was fixed at the values of maximumabsorbance for calcium (700 and 505 nm). Plasma Phe wasmeasured by a fluorometric quantitative assay (FluoroskanAscent, Thermo, Spain). Plasma 25-hydroxyvitamin D wasdetermined by a fully automated electrochemiluminescencesystem (Roche Diagnostic GmbH, Mannheim, Germany).Plasma lipids were extracted from blood according to themethod developed by Folch et al. (1957). Phospholipids weresubsequently isolated by thin-layer chromatography (TLCSilica gel 60, 20×20 cm, Merck, Spain). First, a mixture ofheptane, diisopropyl ether, and acetic acid (7/3/0.2v/v) wasemployed as the eluent and then heptane as the secondeluent. After being removed from the silica matrix, the


phospholipids (PL) were transmethylated according toLepage and Roy (1986) using tridecanoic acid as theinternal standard. The fatty acid methyl esters wereseparated on a Hewlett Packard GC 5890 gas chromato-graph using a flame-ionization detector on a capillarycolumn SP 2330 (30 m×0.25 mm, 0.20 μm) (Supelco,Bellefonte, PA, USA), with hydrogen as the carrier gas.The results were expressed as g fatty acid/g total fattyacids in PLs. Plasma total fatty acids were also trans-methylated according to Lepage and Roy (1986) usingtridecanoic acid as the internal standard. The fatty acidmethyl esters were separated and quantified according tothe method described above. The results were expressed asg fatty acid/100 g total fatty acids.

Statistical analyses

Statistical analyses were performed with the statisticalsoftware program SPSS® 16.0 for Windows (StatisticalPackage for the Social Sciences Inc., Chicago, IL, USA).Statistically significant difference was set at p<0.05.Descriptive statistics are presented as mean ± SD. Allvariables were analyzed using the Kolmogorov–Smirnovtest. Differences between variables were tested by using theMann–Whitney test. Bivariate correlations were calculatedby Spearman’s correlation coefficients (r).

Results

Anthropometric data

Patients’ anthropometric data are shown in Table 1. Meanheight Z-score for all age groups was below the average forthe healthy population (–0.5, –0.1 and –0.5, respectively). Asfar as mean weight Z-score is concerned, groups 1 and 3 wereabove the average (0.4 and 0.3, respectively), and group 2was slightly below the average (–0.5). Patients in groups 1and 3 showed a BMI Z-score above the reference values forthe healthy population (0.9 and 0.6, respectively). BMI Z-score of group 2 was slightly below average (–0.5).

Nutritional indices

Tables 1 and 2 summarize nutritional and biochemical data.Although all patients were advised to adhere to the diet, not allof them met the recommended plasma Phe concentrations.Within the infant and adolescent group (age <18 years, n=23),only 12 met the target value (<360 μmol/L) (CampistolPlana et al. 2006); within the adult group (age ≥18 years,n=24), only 11 met the target value (<600 μmol/L)(Campistol Plana et al. 2006). In addition, 36 patientsexceeded the recommended dietary allowance (RDA) for

proteins, whereas three patients reported protein intakebelow RDA. As far as calcium intake is concerned, 33patients exceeded the RDA and 12 reported a daily intakeslightly below RDA.

Although mean plasma 25-hydroxyvitamin D concentra-tions for all groups met the optimum value (>75 nmol/L)(Table 1), four PKU patients had a level below the optimumvalue (56.5, 69.0, 67.1, and 67.3 nmol/L, respectively).

Biochemical parameters and BMD

Osteopenia (Z-score between –1.0 and –2.5) of at least oneskeletal site was detected in 13 patients (28%), whereasosteoporosis (Z-score <–2.5) was detected in six patients(13%). Bivariate correlations between age and Z-score at thelumbar spine revealed a significant negative correlation ingroup 1 (r=–0.85, p=0.007) and a significant positivecorrelation in group 2 (r=0.77, p=0.0002) with Z-score atthe femoral neck. Fatty acid profiles of healthy controls andpatients are presented in Table 3. We found statisticallysignificant differences between essential fatty acid levels ofPKU patients and their age-matched controls within all rangesof age and in both total fatty acids fraction and PL fraction.PKU patients had in general lower plasma concentrations ofEPA, DHA, PUFA, and n-3 fatty acids and higher plasmaconcentrations of oleic acid, monounsaturated fatty acids(MUFAs), n-6 fatty acids, and n-6 to n-3 fatty acid ratio.

Bivariate correlations between fatty acid profile andBMD expressed in terms of Z-scores are presented inTable 4. There were positive correlations between palmiticacid (group 3, TFA, femoral neck), EPA (group 2, PL,lumbar spine; group 3, PL, femoral neck), DHA (group 1,PL, femoral neck; group 3, PL, lumbar spine), PUFA(group 2, PL, femoral neck), total n-3 fatty acids (group 1,PL, femoral neck) and BMD. Interestingly, although MUFA(group 1, PL femoral neck) and PUFA (group 1, PL,femoral neck) did not correlate significantly with BMD,they did indicate a trend towards significance (p=0.05).

Bivariate correlations between daily calcium intake, plasmaPhe, IDC, plasma calcium, plasma 25-hydroxyvitamin Dconcentrations, and BMD expressed in terms of Z-score valuesshowed no significant association between these variables.

Discussion

Treating PKU represents an important nutritional challengethat leads PKU patients to need to be carefully monitored forlife so that they achieve normal neurological and physicaldevelopment. To the best of our knowledge, previous studiesconcerning bone mineralization in PKU patients have beenaimed at investigating how BMD can be affected by protein,Phe, minerals (calcium and phosphorus), and vitamin D daily


intake (Modan-Moses et al. 2007, Al-Qadreh et al. 1998,Zeman et al. 1999, Pérez-Dueñas et al. 2002). The crucialrole played by n-3 fatty acids in promoting bone accretion inthe healthy population (Watkins et al. 2003) and a positiveassociation between n-3 fatty acid levels and bone formation(Eriksson et al. 2009, Högström et al. 2007, Weiss et al.2005) have been reported in the literature. In addition,Koletzko et al. highlighted the need for supplementing PKUpatients with LC-PUFAs so they can achieve optimalneurological function (Koletzko et al. 2009a, Koletzko et al.2009b). However, there is no reference in the literature withrespect to their possible influence over bone mineralizationin a cohort of PKU patients under treatment.

The restrictive diet established for treating PKU, whichis normally life long, may have long-term unfavorableeffects on skeletal development, such as the appearance ofosteopenia/osteoporosis. Osteoporosis is a condition withlate onset, generally presenting in the elderly, sometimespreceded by an asymptomatic period of osteopenia. Bearingin mind that PKU patients are treated from birth, theseconditions may start at an earlier stage as a secondarycomplication of PKU. For this reason, there is a real needfor analyzing and regulating potential contributors to thebone loss observed in these patients (Modan-Moses et al.2007) to avoid a worsening of their delicate health statusbecause of the appearance of osteopenia/osteoporosis.

Fatty acid pattern has been previously analyzed bymeasuring different lipid classes (total plasma lipids,plasma phospholipids, erythrocyte phosphatidylcholine,erythrocyte phosphatidylethanolamine, or total erythrocytelipids). Based on the concordant results of fatty acid profile(DHA, EPA, total n-6, total n-3, and n-6/ n-3) in plasma anderythrocyte lipids observed by other authors (van Goolet al. 2000, Moseley et al. 2002, Sanjurjo et al. 1994, Inniset al. 1992), we concluded that plasma phospholipids wouldrepresent a valid marker for fatty acid status. Moreover,Skeaff et al. (2006) observed that dietary-induced changesin the fatty acid composition of plasma phospholipids andplatelet and erythrocyte phosphatidylcholine as a functionof time were qualitatively similar. Koletzko et al. (2007)and Agostoni et al. (1995) also considered this lipid class tobe a valid biomarker of dietary fatty acid intake in thesetting of clinical trials in which infants with phenylketonuriawere supplemented with essential fatty acids.

Our results regarding fatty acid profile analysis were inaccordance with previous studies (Vilaseca et al. 2010,Beblo et al. 2007, Moseley et al. 2002, Acosta et al. 2001,van Gool et al. 2000, Sanjurjo et al. 1994, Galli et al. 1991)in which the authors proved that PKU patients undertreatment had a different essential fatty acid profile fromthat of their age- and gender-matched controls, especiallyconcerning n-3 fatty acids. The results obtained led them to

Group 1 (6–10years) Group 2 (11–18years) Group 3 (19–42years)

Caloric intake (kcal/day) 1959.9±624.7 1619.2±140.0 2121.8±563.8

Protein intake (g/day) 29.6±17.8 48.7±11.1 63.4±17.8

Phe intake (mg/day) 476.7±337.3 246.8±72.8 412.7±195.5

Fat intake (g/day) 64.0±29.2 47.3±20.4 71.7±41.2

Calcium intake (mg/day) 573.1±471.8 918.9±293.5 987.6±231.1

Carbohydrate intake (g/day) 286.7±92.3 241.7±27.5 310.7±71.8

Table 2 Nutritional andbiochemical indices ofphenylketonuric patientsa

a All values are given asmean ± standard deviation

Group 1(6–10 years)(5 M/3 F)



Age (years) 8.4±1.3 13.7±2.6 29.3±6.6

Height (cm) 122.7±6.6 157.7±9.1 168.6±9.1

Height (Z-score) −0.5±1.4 −0.1±0.4 −0.5±0.9Weight (kg) 29.0±7.2 47.2±10.6 70.2±17.5

Weight (Z-score) 0.4±1.4 −0.5±0.4 0.3±1.3

BMI (kg/m2) 19.1±2.7 18.9±2.1 24.4±4.2

BMI (Z-score) 0.9±1.1 −0.5±0.4 0.6±1.2

BMD (Z-score femoral neck) −0.7±0.8 −1.3±1.5 −1.5±0.9BMD (Z-score lumbar spine) −0.3±0.2 0.1±0.7 −0.7±0.9Plasma Phe (μmol/L) 185.6±110.0 600.5±387.9 746.8±280.4

IDC (μmol/L) 331.2±118.1 537.3±236.1 639.3±200.2

Plasma 25-hydroxivitamin D (nmol/L) 83.1±12.5 84.9±12.9 88.5±14.6

Table 1 Anthropometricsand biochemical indices ofphenylketonuric patientsa

BMI body mass index, BMDbone mineral density, Phephenylalanine, IDC Index ofDietary Controla Values are expressedas mean ± standard deviation


Tab

le3

Fatty

acid

profile

intotalplasmafatty

acidsandin

theph

osph

olipid

fractio

nof

healthycontrolsandph

enylketonu

ricpatientsa

Controlsgroup1

Patientsgroup1

Controlsgroup2

Patientsgroup2

Controlsgroup3

Patientsgroup3

(6–10

years)

(n=16)

(6–10

years)

(n=8)

(11–

18years)

(n=19)

(11–

18years)

(n=17)

(19–

42years)

(n=42)

(19–

42years)

(n=22)

Fatty

acid

profile

TFA

PL

TFA

PL

TFA

PL

TFA

PL

TFA

PL

TFA

PL

Palmiticacid

(16:0)

20.0±1.3

23.5±3.6

20.4±2.0

25.9±0.9

20.2±1.2

25.3±5.8

19.7±1.6

26.3±1.8

19.5±1.0

23.8±2.4

19.7±1.1

25.7±2.1

Oleic

acid

(18:1n-9)

17.3±5.4

12.5±5.1

22.8±6.0b

11.3±2.6

18.4±3.8

11.6±3.9

23.8±4.6b

12.0±1.9

14.2±1.8

9.7±4.5

21.7±4.5b

10.2±2.6

Linoleicacid

(18:2n-6)

29.2±6.6

14.1±4.0

24.6±7.0b

20.9±2.9b

29.2±6.6

14.3±5.1

29.0±4.7b

20.9±3.7b

25.8±14.1

19.7±5.0

30.8±5.4

21.3±6.3

α-Linolenic

acid

(18:3n-3)

0.3±0.1

0.2±0.1

0.3±0.1

0.1±0.0d

0.4±0.2

0.1±0.1

0.4±0.2

0.1±0.0d

0.3±0.2

0.3±0.1

0.4±0.2

0.1±0.0e

AA

(20:4n-6)

7.3±2.6

9.1±6.0

8.7±3.6

9.9±2.8

7.8±2.6

9.3±4.9

6.7±0.8

10.2±1.3

10.0±3.9

9.5±3.5

7.5±1.7

10.2±2.6

EPA

(20:5n-3)

0.5±0.2

1.3±1.0

0.2±0.1c

0.2±0.1b

0.5±0.2

0.8±0.3

0.2±0.1c

0.2±0.1b

0.4±0.1

0.7±0.6

0.2±0.1b

0.2±0.1

DHA

(22:6n-3)

2.4±1.1

4.3±2.4

1.3±0.7b

2.0±0.6b

2.4±0.6

3.6±1.4

1.1±0.4c

2.1±0.6b

3.0±2.3

4.0±1.3

1.2±0.3b

2.1±0.7

SFA

32.7±3.1

46.2±2.5

32.4±5.0

44.7±0.6

32.5±3.5

49.3±8.2

30.3±1.3b

43.1±2.9b

35.5±6.0

44.3±3.6

29.9±1.4b

45.0±5.2

MUFA

23.9±6.0

19.3±5.2

27.8±6.0b

15.8±2.7

23.4±4.0

17.9±5.5

28.5±4.8b

16.7±1.9

20.1±1.0

16.6±7.8

26.1±4.9

14.5±3.2

PUFA

43.3±6.4

34.4±7.4

39.8±6.0

39.5±2.7b

44.0±4.6

32.7±4.7

41.2±5.0

40.2±2.6c

44.3±6.5

39.1±6.9

44.0±4.7

40.5±5.0

Totaln-6

39.4±6.4

27.3±6.3

37.2±6.1

36.3±2.0b

40.1±4.8

26.9±4.1

38.9±4.8

36.7±3.1c

39.8±9.2

33.2±7.2

41.7±4.6

37.1±5.0

Totaln-3

3.7±1.3

6.8±3.9

2.4±1.6b

3.1±0.8

3.7±0.8

5.3±1.7

2.1±0.6c

3.3±0.7b

4.5±2.7

5.7±1.5

2.2±0.5b

3.2±0.9

n-6/n-3ratio

11.6±3.8

5.3±2.7

19.0±7.2b

12.2±3.0b

11.1±2.4

5.4±1.8

19.2±5.5c

11.7±4.0c

11.1±7.2

6.3±2.4

19.9±5.1b

12.6±4.8

AAarachido

nicacid,EPA

eicosapentaeno

icacid,DHAdo

cosahexaenoicacid,MUFA

mon

ounsaturated

fatty

acid,PUFA

polyun

saturatedfatty

acid,SFAsaturatedfatty

acid,TFA

totalfatty

acids,

PLph

osph

olipids

aAllvalues

aremean±standard

deviation(SD)andareexpressedas

g/10

0gtotalplasmafatty

acidsor

g/10

0gplasmaph

osph

olipids

bSignificantly

differentfrom

controlgroup,

p<0.05

cSignificantly

differentfrom

controlgroup,

p<0.001

dSD=0.04

eSD=0.06


conclude that the changes observed in PKU patients’ fattyacid profile were related to the strict diet used for treatingPKU and not to PKU itself. We must point out that otherauthors have reported normal levels of fatty acids in PKUpatients. A previous study highlighted that PKU patientsdid not exhibit signs or symptoms suggestive of essentialfatty acid deficiency despite lower concentrations of fattyacids than control individuals, thereby suggesting thatpatients with PKU had normal and adequate essential fattyacid concentrations (Lavoie et al. 2009). Pöge et al. (1998)found no significant differences in the plasma fatty acidpattern of these patients when compared with the controlgroup. However, one of the PKU groups enrolled in thestudy (aged 1–6 years) had significantly lower levels of redblood cell n-3 fatty acids. Vilaseca et al. (2010) emphasizedthat fatty acids levels of PKU patients undergoing BH4

treatment met reference values for the healthy population.BH4 therapy allowed liberalization of the protein-restricteddiet, which implies an almost free protein diet withcompletely normalized n-3 and n-6 PUFA intake.

With respect to bone health status, our results confirmed adiminished bone mineralization of PKU patients, as otherauthors reported previously (Modan-Moses et al. 2007,Zeman et al. 1999, Allen et al. 1994, Przyrembel andBremer 2000). BMD at the femoral neck showed a positivecorrelation with EPA, DHA, and total n-3 fatty acids. BMD

at the lumbar spine also showed a positive correlation withDHA. In this context, previous studies in healthy individualsalso observed a positive correlation between DHA, total n-3fatty acids, and bone accrual (Eriksson et al. 2009, Högströmet al. 2007, Weiss et al. 2005). In addition, experimentalanimal studies clearly indicated a possible relationshipbetween dietary supplementation with DHA and an increasein calcium absorption in bone (Kruger and Schollum 2005,Haag et al. 2003, Weiler and Fitzpatrick-Wong 2002).Bivariate correlations between MUFA, PUFA, and BMDindicated a trend towards statistical significance (p=0.05).We did not observe any significant correlation between oleicacid and n-6 to n-3 fatty acid ratio and BMD.

It has also been reported in the literature that anuntreated PKU patient showed diminished BMD whencompared with her age- and sex-matched control (Schwahnet al. 1998). The absence of any dietary intervention wouldcause Phe level to be above the safety level. Therefore, thepresence of a defect in bone mineralization could be viewedas a consequence of the toxic effect of a high Pheconcentration. A previous study supports this hypothesis(Porta et al. 2008). The authors found a strict direct positivecorrelation between the blood mean Phe concentration andthe spontaneous osteoclastogenesis in PKU patients. Thus,high plasma Phe levels would cause a greater rate of boneresorption processes and BMD to decrease.

Table 4 Bivariate correlations between total fatty acids or fatty acids measured in the phospholipid fraction and bone mineral density expressedin terms of Z-score in PKU patients

Group 1 (6–10 years) (n=8) Group 2 (11–18 years) (n=17) Group 3 (19–42 years) (n=22)

Z-score(femoral neck)

Z-score(lumbar spine)





Fatty acid TFA PL TFA PL TFA PL TFA PL TFA PL TFA PL

Palmitic acid (16:0) −0.16 −0.50 0.56 0.11 0.18 0.15 0.13 0.07 0.47* −0.10 0.27 −0.06Oleic acid (18:1 n-9) −0.01 −0.41 −0.37 −0.18 −0.32 −0.39 −0.27 −0.09 −0.19 −0.34 −0.22 −0.19Linoleic acid (18:2 n-6) −0.38 −0.19 −0.47 0.59 0.25 0.31 0.10 0.09 0.25 −0.05 0.25 −0.11α-Linolenic acid(18:3 n-3)

−0.10 −0.45 −0.39 0.56 0.27 0.12 0.19 0.07 0.18 0.04 −0.02 −0.14

AA (20:4 n-6) 0.41 0.41 0.60 −0.44 0.00 0.09 0.19 0.25 0.03 −0.06 0.27 0.03

EPA (20:5 n-3) 0.31 0.51 0.61 0.33 0.11 −0.22 0.49* −0.28 0.02 0.57* 0.28 0.06

DHA (22:6 n-3) 0.66 0.83* 0.54 −0.07 0.08 −0.01 0.11 0.15 0.30 0.33 0.28 0.44*

SFA 0.05 −0.12 0.48 0.16 −0.16 −0.24 −0.17 −0.24 0.15 0.34 0.12 0.14

MUFA 0.07 −0.71** −0.14 −0.25 −0.28 −0.40 −0.24 −0.23 −0.23 −0.34 −0.20 −0.17PUFA −0.18 0.71** −0.30 0.25 0.26 0.50* 0.17 0.45 0.27 −0.03 0.23 −0.01Total n-6 −0.18 0.30 −0.27 0.11 0.15 0.41 0.04 0.33 0.21 −0.11 0.22 −0.09Total n-3 0.44 0.73* 0.70 0.11 0.13 0.02 0.21 0.14 0.26 0.34 0.36 0.32

n-6/n-3 ratio −0.66 −0.68 −0.55 −0.14 −0.10 0.06 −0.29 −0.05 −0.24 −0.35 −0.21 −0.41

PKU phenylketonuria, AA arachidonic acid, EPA eicosapentaenoic acid, DHA docosahexaenoic acid, PUFA polyunsaturated fatty acid, MUFAmonounsaturated fatty acid, SFA saturated fatty acid, TFA total fatty acids, PL phospholipids

*p<0.05

**The correlation between these variables indicated a trend towards statistical significance (p=0.05)


Several additional findings in our study should also beconsidered. Other factors known to influence bone miner-alization were analyzed in order to evaluate whether thesefactors could affect BMD of our patients. We found nocorrelation between BMD, plasma Phe concentration, andIDC. This result is consistent with the findings of Zeman etal. (1999) and Pérez-Dueñas et al. (2002). However, otherauthors reported that Phe concentrations above the recom-mended values were associated with a decrease in bonemineral content (McMurry et al. 1992). Furthermore, Baratet al. suggested that variations of plasma Phe concentrationmight contribute to osteopenia in PKU patients (Barat et al.2002). The 25-hydroxyvitamin D levels were mostly withinthe reference range for the healthy population, which can beviewed as a result of the vitamin D supplement comingfrom the formula on which our patients were fed and of ourpatients taking part in normal outdoor activities, given thatthe main source of vitamin D in Mediterranean countries iscutaneous synthesis of UV radiation. Plasma calciumconcentration was within the normal range. This variableshowed no correlation with BMD, as other authorsreported previously (McMurry et al. 1992, Hillman etal. 1996). Unlike a previous report (Pérez-Dueñas et al.2002), we found no association between calcium intakeand BMD. Nutritional indices showed that some patients didnot adhere to the diet. This fact might affect treatmenteffectiveness and highlights the importance of continuingdietary education and support. An adequate amount offormula intake is important for bone accrual, as this is themain source of mineral and trace elements in PKU patients.

Osteopenia of at least one skeletal site was detected in 13patients (28%), whereas osteoporosis was detected in sixpatients (13%). These frequencies of osteopenia andosteoporosis are similar to those observed by other groups(Pérez-Dueñas et al. 2002, Al-Qadreh et al. 1998, Zemanet al. 1999, Barat et al. 2002). Although the progressivereduction in BMD is a common feature of PKU, itspathogenesis remains unclear. Our findings suggest thatthe diminished BMD may be a consequence of the lack ofn-3 essential fatty acids. Several biologically plausiblemechanisms whereby PUFAs may regulate bone metabolismhave been described in the literature in a cohort of healthyindividuals. N-3 fatty acids have been considered to enhancecalcium absorption in the intestine and reduce boneresorption by lowering urinary excretion of calcium (Krugerand Horrobin 1997, Claassen et al. 1995a, Claassen et al.1995b). Disorders interfering with calcium absorption makethe patient more vulnerable to the later complications ofosteopenia/osteoporosis, namely, fractures. Therefore, n-3fatty acids might affect calcium absorption of our patients,and the lack of these nutrients might cause a reduction incalcium absorption effectiveness, and it might reduce BMDby extension. Nevertheless, we must point out that the

associations found between essential fatty acids and BMDdoes not demonstrate causality. Further clinical trialsconcerning n-3 fatty acid supplementation are needed toconfirm the hypothesis and evaluate the possible beneficialrole of these nutritional supplements over bone accretion ofPKU patients. If the treatment proved to be successful, itcould offer a safe, relatively inexpensive, approach topreventing or treating osteopenia/osteoporosis in PKUpatients on a low Phe diet.

Acknowledgements We acknowledge support from O+IKER(Basque Institute for Healthcare Research, Basque Foundation forHealthcare Innovation and Research), Mutua Madrileña Foundation(MUTUAM09/013), and Red SAMID (Red de Salud Materno-Infantily del Desarrollo RD08/0072/0036 – Spanish Research Network forMaternal and Child Health and Development).

Funding This study was funded by Mutua Madrileña Foundation(MUTUAM09/013). The authors confirm independence from thesponsors; the content of the article has not been influenced by the sponsor.

References

Acosta PB, Yannicelli S, Singh R, Eisas LJ 2nd, Kennedy MJ,Bernstein L, Rohr F, Trahms C, Koch R, Breck J (2001) Intakeand blood levels of fatty acids in treated patients withphenylketonuria. J Pediatr Gastroenterol Nutr 33:253–259

Agostoni C, Riva E, Biasucci G, Luotti D, Bruzzese MG, Marangoni F,Giovannini M (1995) The effects of n-3 and n-6 polyunsaturated fattyacids on plasma lipids and fatty acids of treated phenylketonuricchildren. Prostaglandins Leukot Essent Fatty Acids 53:401–404

Allen JR, Humphries IR, Waters DL et al (1994) Decreased bonemineral density in children with phenylketonuria. Am J Clin Nutr59:419–422

Al-QadrehA, Schulpis KH, Athanasopoulou H,Mengreli C, SkarpalezouA, Voskaki I (1998) Bone mineral status in children withphenylketonuria under treatment. Acta Paediatr 87:1162–1166

Barat P, Barthe N, Redonnet-Vernhet I, Parrot F (2002) The impact ofthe control of serum phenylalanine levels on osteopenia inpatients with phenylketonuria. Eur J Pediatr 161:687–688

Beblo S, Reinhardt H, Demmelmair H, Muntau AC, Koletzko B (2007)Effect of fish oil supplementation on fatty acid status, coordination,and fine motor skills in children with phenylketonuria. J Pediatr150:479–484

Campistol Plana J, Lambruschini Ferri N, Vilaseca Buscá MA, Pérez-Dueñas B, Fusté Rich E, Gómez López L (2006) Hiperfenilalanine-mia. In: Sanjurjo P, Baldellou A (eds) Diagnóstico y tratamiento de lasenfermedades metabólicas hereditarias, 2nd edn. Ergon, Madrid,pp 305–319

Carson DJ, Greeves LG, Sweeney LE, Crone MD (1990) Osteopeniaand phenylketonuria. Pediatr Radiol 20:598–599

ClaassenN, Coetzer H, Steinmann CM,KrugerMC (1995a) The effect ofdifferent n-6/n-3 essential fatty acid ratios on calcium balance andbone in rats. Prostaglandins Leukot Essent Fatty Acids 53:13–19

Claassen N, Potgieter HC, Seppa M et al (1995b) Supplementedgamma-linolenic acid and eicosapentaenoic acid influence bonestatus in young male rats: effects on free urinary collagencrosslinks, total urinary hydroxyproline, and bone calciumcontent. Bone 16:385S–392S

Dhondt JL, Largillière C, Moreno L, Farriaux JP (1995) Physical growthin patients with phenylketonuria. J Inherit Metab Dis 18:135–137


Dobbelaere D, Michaud L, Debrabander A et al (2003) Evaluation ofnutritional status and pathophysiology of growth retardation inpatients with phenylketonuria. J Inherit Metab Dis 26:1–11

Eriksson S, Mellström D, Strandvik B (2009) Fatty acid pattern in serumis associated with bone mineralisation in healthy 8-year-oldchildren. Br J Nutr 102:407–412

Feillet F, Agostoni C (2010) Nutritional issues in treating phenylketonuria.J Inherit Metab Dis [Epub ahead of print]

Folch J, Lees M, Sloane Stanley GH (1957) A simple method for theisolation and purification of total lipids from animal tissues. JBiol Chem 226:497–509

Galli C, Agostoni C, Mosconi C, Riva E, Salari PC, Giovannini M(1991) Reduced plasma C-20 and C-22 polyunsaturated fattyacids in children with phenylketonuria during dietary interven-tion. J Pediatr 119:562–567

Haag M, Magada ON, Claassen N, Böhmer LH, Kruger MC (2003)Omega-3 fatty acids modulate ATPases involved in duodenal Caabsorption. Prostaglandins Leukot Essent Fatty Acids 68:423–429

Hernández M, Sánchez E, Sobradillo B (2000) Curvas y tablas decrecimiento. In: Argente J, Carrascosa A, Gracia R, Rodríguez F(eds) Tratado de endocrinología pediátrica y de la adolescencia,2nd edn. Ediciones Doyma, Barcelona, pp 1441–1499

Hillman L, Schlotzhauer C, Lee D et al (1996) Decreased bonemineralization in children with phenylketonuria under treatment.Eur J Pediatr 155:S148–152

Högström M, Nordström P, Nordström A (2007) n-3 Fatty acids arepositively associated with peak bone mineral density and boneaccrual in healthy men: the NO2 Study. Am J Clin Nutr 85:803–807

Innis SM (1992) Plasma and red blood cell fatty acid values asindexes of essential fatty acids in the developing organs ofinfants fed with milk or formulas. J Pediatr 120:S78-S86

Kanis JA (1994) Assessment of fracture risk and its application toscreening for postmenopausal osteoporosis: synopsis of a WHOreport. WHO Study Group. Osteoporos Int 4:368–381

Koletzko B, Beblo S, Demmelmair H, Müller-Felber W, Hanebutt FL(2009a) Does dietary DHA improve neural function in children?Observations in phenylketonuria. Prostaglandins Leukot EssentFatty Acids 81:159–164

Koletzko B, Beblo S, Demmelmair H, Hanebutt FL (2009b) Omega-3LC-PUFA supply and neurological outcomes in children withphenylketonuria (PKU). J Pediatr Gastroenterol Nutr 48(Suppl 1):S2–7

Koletzko B, Sauerwald T, Demmelmair H, Herzog M, von Schenck U,Böhles H, Wendel U, Seidel J (2007) Dietary long-chainpolyunsaturated fatty acid supplementation in infants withphenylketonuria: a randomized controlled trial. J Inherit MetabDis 30:326-332

Kruger MC, Horrobin DF (1997) Calcium metabolism, osteoporosisand essential fatty acids: a review. Prog Lipid Res 36:131–151

Kruger MC, Schollum LM (2005) Is docosahexaenoic acid moreeffective than eicosapentaenoic acid for increasing calciumbioavailability? Prostaglandins Leukot Essent Fatty Acids73:327–334

Lavoie SM, Harding CO, Gillingham MB (2009) Normal fatty acidconcentration in young children with phenylketonuria (PKU).Top Clin Nutr 24:333-340

Lepage G, Roy CC (1986) Direct transesterification of all classes oflipids in a one-step reaction. J Lipid Res 27:114–120

McMurry MP, Chan GM, Leonard CO, Ernst SL (1992) Bone mineralstatus in children with phenylketonuria–relationship to nutritionalintake and phenylalanine control. Am J Clin Nutr 55:997–1004

Modan-Moses D, Vered I, Schwartz G (2007) Peak bone mass inpatients with phenylketonuria. J Inherit Metab Dis 30:202–208

Moseley K, Koch R, Moser AB (2002) Lipid status and long-chainpolyunsaturated fatty acid concentrations in adults and adolescentswith phenylketonuria on phenylalanine-restricted diet. J InheritMetab Dis 25:56–64

Pérez-Dueñas B, Cambra FJ, Vilaseca MA, Lambruschini N,Campistol J, Camacho JA (2002) New approach to osteopeniain phenylketonuric patients. Acta Paediatr 91:899–904

Pöge AP, Bäumann K,Müller E, Leichsenring M, Schmidt H, Bremer HJ(1998) Long-chain polyunsaturated fatty acids in plasma anderythrocyte membrane lipids of children with phenylketonuria aftercontrolled linoleic acid intake. J Inherit Metab Dis 21:373–381

Porta F, Roato I, Mussa A, Repici M, Gorassini E, Spada M, Ferracini R(2008) Increased spontaneous osteoclastogenesis from peripheralblood mononuclear cells in phenylketonuria. J Inherit Metab Dis[Epub ahead of print]

Przyrembel H, Bremer HJ (2000) Nutrition, physical growth, and bonedensity in treated phenylketonuria. Eur J Pediatr 159:S129–135

Rauch F, Plotkin H, DiMeglio L (2008) Fracture prediction and thedefinition of osteoporosis in children and adolescents: the ISCD2007 Pediatric Official Positions. J Clin Densitom 11:22–28

Sanjurjo P, Perteagudo L, Rodríguez Soriano J, Vilaseca A, CampistolJ (1994) Polyunsaturated fatty acid status in patients withphenylketonuria. J Inherit Metab Dis 17:704–709

Schaefer F, Burgard P, Batzler U (1994) Growth and skeletal maturationin children with phenylketonuria. Acta Paediatr 83:534–541

Schwahn B, Mokov E, Scheidhauer K, Lettgen B, Schönau E (1998)Decreased trabecular bone mineral density in patients withphenylketonuria measured by peripheral quantitative computedtomography. Acta Paediatr 87:61–63

Skeaff CM, Hodson L, McKenzie JE (2006) Dietary-induced changes infatty acid composition of human plasma, platelet, and erythrocytelipids follow a similar time course. J Nutr 136:565–569

van Gool CJ, van Houwelingen AC, Hornstra G (2000) The essentialfatty acid status in phenylketonuria patients under treatment. JNutr Biochem 11:543–547

Verkerk PH, van Spronsen FJ, Smit GP, Sengers RC (1994) Impairedprenatal and postnatal growth in Dutch patients with phenylke-tonuria. The National PKU Steering Committee. Arch Dis Child71:114–118

Vilaseca MA, Lambruschini N, Gómez-López L, Gutiérrez A, Moreno J,Tondo M, Artuch R, Campistol J (2010) Long-chain polyunsatu-rated fatty acid status in phenylketonuric patients treated withtetrahydrobiopterin. Clin Biochem 43:411–415

Watkins BA, Li Y, Lippman HE, Feng S (2003) Modulatory effect ofomega-3 polyunsaturated fatty acids on osteoblast function and bonemetabolism. Prostaglandins Leukot Essent Fatty Acids 68:387–398

Weiler HA, Fitzpatrick-Wong SC (2002) Modulation of essential(n-6):(n-3) fatty acid ratios alters fatty acid status but not bonemass in piglets. J Nutr 132:2667–2672

Weiss LA, Barrett-Connor E, von Mühlen D (2005) Ratio of n-6 to n-3 fatty acids and bone mineral density in older adults: the RanchoBernardo Study. Am J Clin Nutr 81:934–938

Zeman J, Bayer M, Stepán J (1999) Bone mineral density in patientswith phenylketonuria. Acta Paediatr 88:1348–1351


fatty acid profile in patients with phenylketonuria and its relationship with bone mineral density

Documents