This is a very informative article shared by my friend Carrie Hall to our Fb group.

Maternal phenylketonuria: low phenylalaninemia might increase the risk of intrauterine growth retardation

By; Raphaël Teissier & Emmanuel Nowak & Murielle Assoun & Karine Mention & Aline Cano & Alain Fouilhoux & François Feillet & Hélène Ogier & Emmanuel Oger & Loïc de Parscau & On behalf of the AFDPHE (Association Française pour le Dépistage et la Prévention des Handicaps de l’Enfant)

Presented in 2012 ( 3 years ago)

Link to study :

https://attachment.fbsbx.com/file_download.php?id=923115184434553&eid=ASvuUhwN3XS0LPhu4fpjqTUit7Ujlm1xjLpvt7DOe7zSbYKhknNs4ReMuzROS0wfFd8&inline=1&ext=1445019346&hash=AStUoRm5pDE6fCa8

or

https://attachment.fbsbx.com/file_download.php?id=923115184434553&eid=ASvuUhwN3XS0LPhu4fpjqTUit7Ujlm1xjLpvt7DOe7zSbYKhknNs4ReMuzROS0wfFd8&inline=1&ext=1445019346&hash=AStUoRm5pDE6fCa8

Please feel free to share!

if the link does not work, please continue below , I have copied pasted the article to this post. This is not written by me. I do not own this piece. I simply am sharing it for education purposes.

Maternal phenylketonuria: low phenylalaninemia might increase the risk of intrauterine growth retardation Raphaël Teissier & Emmanuel Nowak & Murielle Assoun & Karine Mention & Aline Cano & Alain Fouilhoux & François Feillet & Hélène Ogier & Emmanuel Oger & Loïc de Parscau & On behalf of the AFDPHE (Association Française pour le Dépistage et la Prévention des Handicaps de l’Enfant) Received: 6 September 2011 /Revised: 13 April 2012 /Accepted: 16 April 2012 # SSIEM and Springer 2012

Abstract Background Malformations and mental retardation in the offspring of women with Phenylketonuria (PKU) can be prevented by maintaining maternal blood Phenylalanine (PHE) within a target range (120–300 μmol/L) through a PHE-restricted diet. In a former French study, a high and unexpected proportion of intra uterine growth retardation (IUGR) has been reported. Guidelines have been proposed to all French centres caring for maternal PKU since 2002. Objective To confirm IUGR and investigate its causes. The other goals were to assess the follow-up of these pregnancies based on the new guidelines and the pertinence of these recommendations. Design Clinical, biological and ultrasound data of all pregnancies in PKU women in France, from 2002 to 2007 were retrospectively analyzed. Results Data from 115 pregnancies in 86 women with PKU were collected. Ninety percent of women had been informed of the risk of maternal PKU in the absence of a strict diet during pregnancy, 88 % of women had started a diet before conception, and 45 % of infants were born small for gestational age (birth length and/or weight ≤−2 SD). PHE intakes were lower in the group with IUGR from the fifth to the eighth month of pregnancy and duration of time spent at <120 μmol/L during pregnancy was associated with a higher risk of IUGR. Conclusion Hyperphenylalaninemia (HPA) is not the only risk factor for IUGR; PHE lower than 120 μmol/L could also be associated with the IUGR occurence. Even if the monitoring of these pregnancies has been improved since the initiation of guidelines, we would like to stress on the importance of the dietary aspect of the disease. Communicated by: Verena Peters R. Teissier (*) : L. de Parscau Department of Pediatrics, Brest University Hospital, CHRU Morvan, 2 avenue Foch, 29200 Brest, France e-mail: raphael.teissier@chu-brest.fr E. Nowak : E. Oger INSERM CIC 05-02, Brest University Hospital, Brest, France M. Assoun : H. Ogier PKU group of the AFDPHE, Paris, France K. Mention PKU group of the AFDPHE, Lille, France A. Cano PKU group of the AFDPHE, Marseille, France A. Fouilhoux PKU group of the AFDPHE, Lyon, France F. Feillet PKU group of the AFDPHE, Nancy, France L. de Parscau PKU group of the AFDPHE, Brest, France J Inherit Metab Dis DOI 10.1007/s10545-012-9491-0 Introduction Phenylketonuria (PKU; OMIM #261600) is an inherited metabolic disease caused by a defect of the phenylalanine (PHE) hydroxylase. Untreated, it leads to PHE accumulation which is toxic for brain and results in severe learning disability and behavioural problems. Neonatal screening has contributed to the decrease of mental retardation. As a result, the girls with PKU treated by a low PHE intake diet become healthy women and potential mothers. In 1980, Lenke and Levy (1980) reported a high proportion of congenital abnormalities in the offspring born to mother with PKU: congenital heart disease (CHD), mental retardation, facial dysmorphism and intra uterine growth r-etardation (IUGR). Frequency of these malformations was correlated with the maternal blood PHE levels during pregnancy. North American and English studies (Koch et al. 2003; Lee et al. 2005) have demonstrated the efficacy of PHE restricted diet during pregnancy to prevent PKU embryopathy. In France, a former study (Feillet et al. 2004) confirmed these data but highlighted an important proportion of IUGR in the offspring of PKU women despite a PHE-restricted diet. In order to further investigate this observation and to make the monitoring easier, specific guidelines were proposed and diffused to all French centres caring for PKU. The main goal of the study was to confirm IUGR and to investigate its causes. We also aimed to assess monitoring of these pregnancies according to the new guidelines and thus, to investigate the relevance of these recommendations. Patients and methods Patients All known French PKU women, pregnant between January 2002 and December 2007, were included, regardless of their pregnancy outcome. Methods Guidelines communicated to all regional centres in 2001 recommended: – collect of thorough, preconception (2–3 months) data on girls and women with PKU, – control of PHE blood concentration at least weekly with target values between 120 and 300 μmol/L – careful diet monitoring – control of biological nutritional parameters and at least three ultrasound scans during the pregnancy – monthly follow-up – careful newborn examination. A retrospective study of all known pregnancies in France was performed by reviewing medical and dietary records. Medical data included maternal PKU history, onset of PHE restricted diet initiation (pre or post-conception), biological and dietary monitoring, ultrasound scans results, blood PHE concentrations, and birth measurements of newborns. All data were entered into a data base (excel spreadsheet). Data entry was checked for reliability through a data base doubling. PHE indexes The pregnancy and foetus outcomes are related to the maternal plasma PHE during pregnancy. The mean plasma PHE is the most widely used parameter for analysis. However it may not be accurate enough to detect both PHE excess and deficiency. To estimate more precisely the control of PHE levels during pregnancy, we calculated 3 monitoring indexes of excess (IE) or deficiency (ID) for rates above 300 μmol/L and below 120 μmol/L: – I.E-1 and I.D-1: Area under the curve (AUC) with a baseline at 300 μmol/L or Area Above the curve (AAC) with a baseline at 120 μmol/L – I.E-2 and I.D-2: Mean of PHE >300 μmol/L and Mean of PHE <120 μmol/L – I.E-3 and I.D-3: time spent above 300 μmol/L or under 120 μmol/L. In order to study the relationship between PHE excess or deficiency during pregnancy and the occurrence of IUGR, we selected pregnancies (Fig. 1) for which we had birth measurements and an extended PHE rates followed up to one month before delivery. Once these 55 “well documented” pregnancies identified, we calculated the quartiles (Q25, Q50, and Q75) for both I.E-3 and I.D-3 indexes. Then we divided pregnancies into four classes according to the quartiles for I.E-3 (time spent above 300 μmol/L) and I.D-3 (time spent under 120 μmol/L) (Table 1): – Class 1: pregnancies with a high I.E-3 (IE3>Q75) and high I.D-3 (ID3>Q7 5) (both PHE excess and deficiency) – Class 2: pregnancies with a high I.E-3 (IE3>Q75) but not I.D-3 (ID3≤Q75) (PHE excess) – Class 3: non-high IE3 (IE3≤Q75) but high ID3 (ID3> Q75) (PHE deficiency) – Class 4: pregnancies with non-high I.E-3 (IE3≤Q75) and non-high I.D-3 (ID3≤Q75) (well maintained PHE level; reference pregnancies). J Inherit Metab Dis Statistical analysis Means/median were expressed±1 standard deviation (SD) or with range [minimum-maximum]. Fisher’s exact or Khi² tests were used to compare frequencies between groups. Group means/median were compared by using Wilcoxon, Mann Whitney tests or by Student’st-test, when appropriate. Correlations between different data were investigated using Spearman’s correlation coefficient. Statistical analyses were performed using SAS® software package. A p-value ≤0.05 was considered as significant. Results Cohort description 1. PKU mothers profile From January 2002 to December 2007, data from 115 pregnancies in 86 women with PKU were collected. Sixty four (79 %) had a classical PKU, 14 (17 %) a mild PKU and three (4 %) a non-PKU HPA. The phenotype was unknown for five women. Mean weight, height and body mass index (BMI) were not different from the French general population (Charles et al 2009): respectively 59±11 kg, 160.5±5.5 cm and 23±4 kg/m². The birth weight of the offspring was not correlated to the mother’s BMI. Ninety one percent of women with PKU had been informed of the risk of maternal PKU (69/76). Data were not available for ten women. The educational level of mothers with PKU was not different from the general population (INSEE 2007). No pregnancy was reported in patients lost to follow-up, undiagnosed before pregnancy or in teenage girls (mean age 28 years [18.8-39.1]). 2. Pregnancy outcome From 2002 to 2007, 115 pregnancies led to the birth of 91 newborns (Fig. 1). When documented birth measurements were compared to Usher and McLean’s data (Usher and McLean 1969); results showed a low birth weight (≤ −2 SD) in 17 % of the newborns, a low birth length (≤ −2 SD) in 40 % and a microcephaly (≤ −2 SD) in 37 %. According to the International Small for Gestational Age Advisory Board Consensus Development Conference Statement (birth 2002-2007 Maternal PKU 115 pregnancies 91 newborns Elective abortion n = 8 Therapeutic abortion n = 5 Spontaneous abortion n = 5 Intra Uterine Fetal death n = 1 Unknown outcome n = 5 New borns with birth measurements n = 86 Birth weight – 2 SD n = 15/86 Birth length – 2 SD n = 29/72 Microcephaly n = 23/62 IUGR n = 34/75 Birth measurements and PHE monitoring n = 55 Class 1 n = 4 Class 2 n = 18 Class 3 n = 14 Class 4 n = 19 ≤ ≤ Fig. 1 Maternal PKU: Pregnancy outcome in France from 2002 to 2007 and screening process for IUGR/no IUGR comparison. a Four were imputable to maternal PKU and one to congenital toxoplasmosis. b corpus callosum agenesis. c number of cases/ number of documented patients. Class 1, 2, 3 and 4 are explicited in Table 1 Table 1 distribution of patients according to their indices of PHE excess or deficiency and by IUGR I.Excess-3 I.Deficiency-3 IUGR Total Classes 0 1 1 : high IE3 0 (0 %) 4 (100 %) 4 high ID3 2 : high IE3 8 (44 %) 10 (66 %) 18 non high ID3 3: non high IE3 8 (57 %) 6 (43 %) 14 high ID3 4 : non high IE3 17 (89 %) 2 (11 %) 19 non high ID3 Total 33 22 55 IUGR is expressed in number of newborns with birth measurement corresponding to IUGR J Inherit Metab Dis length and/or weight ≤ −2 SD) (Lee et al 2003), 45 % of newborns (n034/75) had an IUGR. This incidence was the same (46 %, n012/26) for the “well controlled” (PHE levels during pregnancy < 300 μmol/L) pregnancies. The main feature of the IUGR was a low birth length (mean −2.7 SD), present in 29 cases out of 34. Only one third of them had an associated low birth weight and two third a microcephaly. To study the correlation between maternal weight gain and the occurrence of IUGR, we calculated the ratio “weight gain during pregnancy/ideal weight gain during pregnancy” by prepregnancy BMI (Carmichael et al 1997; Institute of Medicine (US) and National Research Council (US) Committee to Reexamine IOM Pregnancy Weight Guidelines 2009) and compared this ratio between pregnancies with IUGR (ratio00.71) and pregnancies without (ratio00.72). This difference was not statistically significant. We also compared this ratio between pregnancies with Phe deficiency (class 3) (ratio00.64) and reference pregnancies (class 4) (ratio00.85). This difference was not statistically significant. Two cases of CHD were reported and led to a therapeutic abortion. The corpus callosum agenesis reported in one patient is not a common feature of PKU embryopathy and might be independent of maternal PKU. 3. Dietary monitoring At the first follow-up visit with the intention for pregnancy, 73 % women did not follow any diet, 13.5 % had a mild diet (no diet but controlled PHE intake) and 13.5 % followed a strict diet (data available for 105 pregnancies). A pre-conception diet was started in 88 % of documented pregnancies (n078/89) and a well maintained PHE level (120<[PHE]<300 μmol/L) was obtained in 75 % of documented pregnancies (n065/87). The median delay to normalize the maternal blood PHE after the diet onset was 28 days. It was not significantly correlated to mother’s educational level. The maternal median PHE intake at the diet onset was 350 mg/d [180–2400] and increased from the fourth month of pregnancy. From the first to the ninth month of pregnancy, median protein intake increased from 60 to 70 g/d and energy intake from 2000 from 2500 kcal/d. Protein intake was close to the recommended level for French pregnant woman (110 % in average) (Abadie et al 2001). 4. Blood PHE levels PHE concentrations were measured once or twice a week before conception and during pregnancy. Target range was 120–300 μmol/L. Using the same thresholds as the Maternal Phenylketonuria Collaborative Study (Rouse et al 1997), mean PHE during pregnancy was less than 360 μmol/L in 83 % of pregnancies, between 361 and 600 μmol/L in 10 %, between 601 and 900 μmol/L in 3.5 % and more than 900 μmol/L in 3.5 %. To assess the control of HPA, three indexes were calculated: the mean of PHE excess (IE-2) and the time spent above 300 μmol.L-1/time of follow-up (IE-3) were well correlated to the AUC with a baseline at 300 μmol/L (IE-1): r00.98; [0.96 to 0.99] and r00.97; [0.95 to 0.98] respectively. Likewise three indexes looking for a PHE deficiency were calculated; the mean of PHE deficiency (ID-2) and the time spent under 120 μmol.L-1/time of follow-up (ID-3) were well correlated to AAC with a baseline at 120 μmol/L (ID-1): r00.99; [ 0.98 to 1] and r00.96; [0.94 to 0.98] respectively. 5. Biological monitoring Guidelines recommended assessing blood nutritional parameters before conception and at each trimester of pregnancy. Most of them were in the normal ranges (Table 2). Carnitine was low but remained stable. Among Table 2 Follow-up of plasma nutritional parameters at the three trimesters (T1, T2, T3) of pregnancy in a group of PKU women; results are mean of 20 to 49 measurements depending on parameters T1 T2 T3 p Reference intervals Carnitine (μmol/L) 27.4 24.5 27.9 NS 33.8 – 77.5* Protein (g/L) 72.8 68.8 65.5 < 0.001 65 – 75 Serum alkaline phosphatase (UI/L) 55.4 66 103 < 0.001 30 – 100 Haemoglobin (g/dL) 12.3 11.7 11.3 < 0.001 > 10.6 * Iron (μmol/L) 27.9 29.7 28.3 NS 9 – 29* Ferritin (μg/L) 60.3 45.4 20.3 < 0.001 15 * Vitamin A (μmol/L) 0.89 0.76 0.58 NS 0.35 – 1.75 Vitamin E (μmol/L) 311 404 462 < 0.001 186-372 Vitamin B 12 (pmol/L) 300 262 215 < 0.001 111 – 295 Vitamin D (nmol/L) 69 83 99 < 0.05 37 – 100 Folate (nmol/L) 40 46 40 NS 0.90 – 45 Calcium (mmol/ L) 2.35 2.28 2.29 0.002 2.25 – 2.5 Phosphorus (mmol/L) 1.14 1.07 1.14 NS 0.8 – 1.35 Zinc (μmol/L) 13.2 10.3 10.4 < 0.001 9.2 – 20 Copper (μmol/L) 21.2 26.9 30.7 < 0.001 11 – 31 Magnesium (mmol/L) 0.83 0.8 0.79 NS 0.75 – 1 Selenium (μmol/L) 0.77 0.74 0.74 NS 0.76 – 1.52 Cholesterol (mmol/L) 3.67 4.55 5.59 < 0.05 6.7±0.79 * Triglyceride (mmol/L) 0.91 1.58 1.81 < 0.01 3.4±0.7* T1: first trimester of pregnancy. T2: second trimester of pregnancy. T3: third trimester of pregnancy. Reference intervals for normal adults (SI Units and Conversion Factors (available from: http://www.hhsc.ca/ body.cfm?id01571. Accessed 12/11/2010)) or for women at third trimester of pregnancy* (Cho and Cha 2005; Dubucquoi et al 2005) J Inherit Metab Dis the trace elements, only selenium was slightly low. Ferritin and haemoglobin decreased significantly but remained within the normal range for pregnant women. Physiological hemodilution in the second part of pregnancy may explain some of the decrease of biological parameters. IUGR vs. no IUGR comparison 1. Dietary monitoring To explain IURG incidence, we compared dietary intake between pregnancies with IUGR and pregnancies without. Protein and energy intake were not statistically different between these two populations. Likewise energy and protein intake, we compared amino acid intake between the two populations. There was not a significant difference between the 2 groups. Median PHE intake was lower in the IUGR group from the fifth to the eighth month of pregnancy whereas it was similar during the first half of pregnancy (Fig. 2). Birth length was significantly correlated to PHE intake from the fifth to the eight month (p≤0.05). Weight and head circumference were significantly correlated with the PHE intake from the sixth to the ninth month (p≤0.05). 2. Blood PHE levels a. Control of PHE and IUGR In regard with our target range (120–300 μmol/L) and after selection of “well documented” pregnancies (n055), an IUGR was identified in 42 % (n012/26) of pregnancies for which the rates of PHE during pregnancy were<300 μmol/L. b. Control of PHE excess Birth length, weight and head circumference were inversely correlated with mean PHE excess (> 300 μmol/L) during pregnancy (respectively R0−0.35; p00.006, R0−0.33; p00.004 and R0−0.3; p00.04). Comparing the frequency of IUGR between class 2 (PHE excess) and class 4 (good PHE control) (Table 1), there were significantly more IUGR in the group which spent more time above 300 μmol/L (class 2) (p00.005). c. Control of PHE deficiency Comparing the frequency of IUGR between class 3 (PHE deficiency) and class 4 (well maintained PHE level) (Table 1), there were significantly more IUGR in the group which spent more time below 120 μmol/L (class 3) (p00.04). There was no correlation between indexes of PHE deficiency and the studied birth parameters. Discussion Guideline compliance was not observed in all pregnancies, thus our results must be interpreted with caution. However, it gives a nationwide picture of maternal PKU management during 6 years. Compared to the former study dealing with maternal PKU in France before 2001 (Feillet et al 2004), the main improvements are: a better information of the patients (91 % versus 63 %), a higher pre-conception diet (88 % versus 42 %) and less cardiac malformations of offspring (0.8 % versus 5.2 %). The latter may be due to the dietary balance achieved before conception (Levy et al 2001; Maillot et al 2008). Early onset of low PHE diet is one of the major factors of foetal prognosis. Information of PKU young girls and women is all the more crucial that they use contraception less often than general female population (Waisbren et al 1995). No pregnancy was reported in patients undiagnosed with PKU before pregnancy. Nevertheless, awareness on maternal PKU related complications needs to be raised among the related health care specialists. We would like to stress that the French neonatal screening program did not start before the 1970s. The most striking data was the high proportion of IUGR (45 %), mainly with regard to the birth length. This confirms on a large scale, the report of IUGR with normal psychomotor development after a good monitoring during pregnancy in the former French study (Feillet et al 2004). The Maternal Phenylketonuria International Study and the report from the United Kingdom Registry showed that normal maternal blood PHE during pregnancy was associated with better birth parameters (length, weight and head circumference) (Koch et al 2003; Lee et al 2005). However, in Koch et al study, birth weight among the offspring of non-PKU HPA mothers on diet has been reported to be lower than that of the untreated group (Koch et al 2000). These results suggest that the maternal blood PHE is not the only risk factor for IUGR; other maternal nutritional 0 100 200 300 400 500 600 700 800 900 123456789 months of pregnancy median PHE mg/day with IUGR without IUGR total Fig. 2 Median PHE intakes (mg/day) during pregnancy in the total group (♦), group with IUGR (●) and group without IUGR (▲). Stars indicate when median Phe intakes are significantly different between pregnancies with IUGR and pregnancies without IUGR J Inherit Metab Dis parameters might be involved when PHE levels are well maintained. Comparison with literature is not easy because definitions of IUGR and reference data are not universal. The Maternal Phenylketonuria Collaborative Study (MPKUCS) reported no cases of IUGR in the best controlled group (PHE<360 μmol/L) and only less than 30 % in the group with the highest mean levels of maternal PHE (> 900 μmol/L) (Rouse et al 1997). In our study, a good control of mean PHE level was achieved due to a PHE level, lower than 360 μmol/L in 83 % of the pregnancies. Birth length, weight and head circumference were inversely correlated to indexes of maternal HPA which is known to be an important risk factor for “maternal PKU embryopathy syndrome”. However an IUGR was found in 46 % of our “well controlled” pregnancies (mean PHE <300 μmol/L). Different references for birth anthropometry [Niklasson in MPKUCS (Niklasson et al 1991) and Usher McLean (Usher and McLean 1969) in our study] cannot explain such a discrepancy. Several arguments support the hypothesis that a PHE deficiency could result in an IUGR. Firstly, IUGR could be due to a lower PHE intake during the second part of pregnancy even though we can not exclude that it could also be due to a more severe PKU with lower PHE tolerance. Secondly, birth measurements are positively correlated with the PHE intake during the second part of the pregnancy. And finally, a longer time spent below 120 μmol/l during pregnancy is associated with a higher risk of IUGR. Moreover, this hypothesis is supported by data in animals (Bhasin et al 2009). Bhasin et al showed that a low-protein diet reduces circulating essential amino acids and leads to intrauterine growth restriction. Therefore, it seems important to maintain essential amino acids, including phenylalanine, at adequate levels during pregnancy. The term “PHE deficiency” must be used with caution. Even though our reported Phe plasma concentrations are below the thresholds (120 μmol/L) recommended by different authors, they are not abnormally low compared to normal plasma PHE levels in adults (Hagenfeldt et al 1984). To date, there is no consensus on using lower (60 to 120 μmol/L) and higher (180 to 360 μmol/L) Phe thresholds tolerated during pregnancy. In our study, we used the threshold 120–300 μmol/L used for the first French study by Feillet et al (Feillet et al 2004) in order to compare different results between the two studies, in particular for pregnancies outcome. The lower threshold was not really different from the other studies. Even though English guidelines recommended 60 μmol/L (Report of Medical Research Council Working Party on Phenylketonuria 1993), most English published studies used the threshold of 100 μmol/L (Maillot et al 2008). Moreover, the threshold used by the Maternal Phenylketonuira Collaborative Study (MPKUCS) was also 120 μmol/L. Data on Tyrosine supplementation were insufficient to be analyzed. Even, PKU might result in Tyrosine deficiency which becomes an essential amino acid in this disease, there is no clinical consensus on Tyrosine supplementation in PKU patients (Webster and Wildgoose 2010) and guidelines remain controversial in regard with maternal PKU (Rohr et al 1998; Van Spronsen et al 2001). Meanwhile, fasting perhaps post prandial plasma Tyrosine could be measured during pregnancy to guide the potential use of a supplementation, preventing both Tyrosine excess and deficiency as has been suggested in PKU patients by Van Spronsen (Van Spronsen et al 1996). However, the observation of IUGR in several pregnancies without HPA or other conventional causes of IUGR suggest a multifactorial mechanism. Even though we can not definitely prove that PHE levels lower than 120 μmol/L are responsible for IUGR, management of maternal PKU should not only be focused on HPA prevention but also on PHE deficiency, especially in the second and third trimesters. IUGR is associated with an increased risk for type 2 diabetes, cardiovascular disease, and hypertension, thus it should be a major concern in maternal PKU management (Godfrey and Barker 2000). Even though no other risk factors could be demonstrated here, dietary parameters may be associated with IUGR risk. Following specific diet recommendations for PHE and Tyrosine intakes may prevent dietary deficiencies responsible for IUGR. Furthermore, for a better understanding of the correlation of PKU pregnancy outcomes with PKU pregnancy conditions, we suggest a thorough follow-up of the PKU offspring. Acknowledgments We would like to thank all the doctors for their involvement in this French maternal PKU study: Angers (Dr Berthelot), Bordeaux (Dr Barat), Lille (Dr Dobbelaere, Dr Mention), Lyon (Dr Dubreuil, Dr Fouilhoux, Dr Guffon), Marseille (Dr Cano, Dr Maurin), Nancy (Dr Feillet), Paris (Dr Abadie, Dr Bilette de Villemeur, Dr De Lonlay, Dr Narcy, Dr Ogier, Dr Touati), Rennes (Dr Journel, Dr Odent, Dr Pasquier), Rouen (Dr Dumesnil), Saint-Étienne (Dr Gay), Toulouse (Dr Coustols-Valat); and the«French Association for the Diagnosis and Prevention of Child Handicap» for his support. We are indebted to the dietician. We also thank Brest University Hospital’s Medical Writer (Zarrin Alavi, MSc). Conflict of interest None.