Very premature infants are at an increased risk for low bone mass and metabolic bone disease because they miss out on the period of greatest mineral accretion that occurs during the last trimester of pregnancy (1). Most of these infants cannot tolerate full enteral feedings within the first days or weeks after birth, and nutrients need to be delivered by parenteral nutrition (PN) (2).
The threshold for calcium (Ca)-phosphate (P) precipitation limits the delivery of appropriate amounts of Ca and P by PN (3,4). The range of recommended doses of Ca and P delivered by PN in preterm infants is wide, varying from 40 to 120 mg · kg−1 · day−1 for Ca and from 31 to 71 mg · kg−1 · day−1 for P (5,6). Lower doses may be inadequate for some infants, compromising short- and long-term bone formation.
Besides mineral intake, bone nutrition is dependent on the intake of energy, protein, magnesium, and vitamin D. Other factors such as calciuric drugs, corticosteroids, lack of mechanical stimulation, and immobility may lead to diminished new bone formation and reduced osteoid (2,7).
No simple, portable, reliable, and validated method is available to detect low bone mass and metabolic bone disease in preterm infants. Alkaline phosphatase and serum phosphate are poor predictors of bone mineralization in preterm infants (8).
The standard tool for measurement of bone mineral content and bone density in infants is dual energy x-ray absorptiometry (DXA). The use of this method is limited because it is expensive, involves ionizing radiation, and requires transporting fragile and often sick infants to the device (9).
More recently, quantitative ultrasonography (QUS) has been proposed as a nonionizing, portable, and relatively low-cost method for assessing bone status (10). This method is standardized in neonates and small infants (11), but it has not been validated in this population. In contrast to DXA, which measures the bone size and mineral content, QUS is thought to better diagnose the low mineral density per volume of bone tissue (deficit in mineralization) and reduced bone mass (osteopenia) in preterm infants (12). More specifically, speed of sound (SOS) measured by QUS reflects bone properties such as density, cortical thickness, elasticity, and microarchitecture, all of which are important determinants of bone strength (7,9). In preterm infants, no correlation was found between DXA and SOS measurements, and SOS cannot be used as a surrogate for DXA; however, both methods may be used to complement each other in the assessment of bone health (13).
Almost all of the longitudinal studies assessing bone strength in preterm infants have reported a significant decline in SOS in the first weeks after birth (7,14–22) followed by a later catch-up (15,21,22).
On the basis of the hypothesis that an early and higher parenteral Ca and P intake can prevent the reported short-term bone strength decline after birth in preterm infants, the effects of 2 Ca and P regimens delivered by PN on bone strength were longitudinally assessed from birth to discharge.
PATIENTS AND METHODS
A triple-blinded randomized controlled trial was conducted in patients receiving 1 of the 2 Ca and P regimens by PN. The investigators who performed SOS measurements were not aware of the parenteral Ca and P regimens, the pharmacist who prepared the parenteral solutions was not aware of the SOS measurements, and the clinicians were not aware of SOS measurements or the parenteral Ca and P regimens.
The study was approved by the ethics committee of Hospital Dona Estefânia and informed parental consent was obtained. Registration of race and ethnicity is not permitted under Portuguese law.
Between February 2006 and October 2008, infants born with gestational age (GA) ≤33 weeks consecutively admitted to a tertiary neonatal intensive care unit were recruited. Only those who required PN for at least 1 week were included in the study. Infants with major congenital abnormalities, severe central nervous system disorders, and bone and/or muscular diseases were not included.
The neonates were classified as appropriate-for-gestational age, small-for-gestational age (SGA), and large-for-gestational age (LGA) using the intrauterine growth reference values of Kramer et al (23). The gestational age was calculated from the mothers' last menstrual period, corrected by early obstetric ultrasound.
Parenteral Nutrition Intervention
In most neonates PN was initiated within the first day of life. In some neonates born on Saturday and all of the neonates born on Sunday, PN was initiated up to 48 hours after birth; in these neonates, intravenous intake was limited to glucose and Ca 27 mg · kg−1 · day−1 using calcium gluconate 10%. When initiating PN, the neonates were randomized by the pharmacy service, using a software-generated random numbers table, to receive 1 of 2 parenteral mineral regimens from the first day on PN, within or close to the recommended doses: Ca 45 mg · kg−1 · day−1 (low dose [LD]) (5) or Ca 75 mg · kg−1 · day−1 (high dose [HD]) (6). Because in both groups P was added to PN solutions at a fixed Ca:P ratio (mg) of 1.7:1 (24), the LD group received 26.5 mg · kg−1 · day−1 and the HD group 44.1 mg · kg−1 · day−1. The maximum concentration established in PN solutions was 60 mg/100 mL for Ca (25), and consequently 35.3 mg/100 mL for P. Calcium gluconate 20% (Labesfal, Santiago de Besteiros, Portugal) and sodium glycerophosphate (Glycophos, Fresenius Kabi, Halden, Norway), respectively, were used as sources of Ca and P. Both groups received similar parenteral intake of other nutrients, including 0.3 mEq · kg−1 · day−1 of magnesium (5) and 160 IU · kg−1 · day−1 of vitamin D (26) using a fat-soluble vitamin solution (Vitalipid N Infant, Fresenius Kabi, Uppsala, Sweden). While receiving exclusive PN, energy intake was targeted at 90 to 115 kcal · kg−1 · day−1(25). Amino acids were introduced on the first day receiving PN at the dose of 1.5 g · kg−1 · day−1, increasing 0.5 g · kg−1 · day−1 daily up to 3.2 to 3.8 g · kg−1 · day−1 according to GA (26,27).
Patients were withdrawn from the study and PN was adjusted individually whenever hypercalcemia, hypocalcemia, hyperphosphatemia, or hypophosphatemia occurred, according to Thomas reference values (28). Patients were also dropped from the study if PN was interrupted for any other reason, if fluid restriction ≤130 mL · kg−1 · day−1 (limiting mineral intake) was necessary after the first week after birth, or whenever deceased, discharged home, or transferred to other units.
The same enteral feeding protocol was used in both groups. Minimal enteral feeding was initiated when bowel sounds were audible, and significant abdominal distention and bilious or bloody gastric residuals were absent. Mother's milk was preferred. From 2 weeks after birth, or when mother's milk volume reached 100 mL · kg−1 · day−1, breast milk was fortified with 4.2 g/100 mL of Nutriprem (Nutricia, Cuijk, the Netherlands) (357 kcal; protein 18.7 g, Ca 1545 mg, and P 1070 mg/100 g). When mother's milk was insufficient or unavailable, the preterm formula Miltina Prem (Humana GmbH, Herford, Germany) (80 kcal; protein 2.2 g, Ca 101 mg, and P 52 mg/100 mL) was used. Initially, feeds were administered continuously, and changed to bolus feeding as soon as infants could tolerate it. When enteral feeding exceeded 50% of daily fluid intake, intravenous fat-soluble vitamins were stopped and enteral supplement of approximately 660 · U/day of vitamin D (Vigantol, Merck, Darmstadt, Germany) was initiated.
The intake of Ca, P, Mg, vitamin D, protein, and energy delivered by parenteral and enteral nutrition 24 hours preceding each SOS measurement was estimated from the prescriptions corrected with the records of the effectively delivered volumes of parenteral and enteral nutrition.
Stimulation of Physical Activity and Therapy With Diuretics or Steroids
According to the protocol of the neonatal intensive care unit, early developmental care intervention for preterm neonates is administered by specially trained physiotherapists as soon as they are stable for the purpose of preventing spasticity and contractures and promoting adequate spontaneous activity. Therapy with diuretics and prenatal and postnatal steroids was recorded.
SOS was measured at the tibia using the Sunlight Omnisense 7000P (Sunlight Medical Ltd, Tel Aviv, Israel). The site of measurement on the tibia was determined by identifying the midpoint between the plantar aspect of the flexed foot and the dorsal aspect of the flexed knee (midshaft of the tibia), using an eyeliner pencil provided by the manufacturer. Both legs were used for measurements because previous results had shown no differences between right or left SOS measurements in infants with normal limb movement (29).
The CS probe for preterm infants was used, aligned along and parallel to the bone and moved in a semiarc over the circumference of the site of measurement until a reliable estimate of the SOS was measured. A set of at least 3 repeated measurements was obtained. The software used the 3 most consistent measurements to compute the result (30). This is expressed in meters per second, and displayed together with a z score (units of standard deviations relative to the mean for age- and sex-matched population reference values) based on a cross-sectional reference range for term and preterm infants that was included with the software and reported by Littner et al (11). Measurements were not performed if the infant was felt to be too unstable or if access to the tibia was difficult, for example, due to intravenous cannulation.
The first measurement (baseline SOS) was performed within the first 5 days after birth, reflecting the intrauterine bone status (baseline). Thereafter, SOS was measured weekly until the infant was discharged from the neonatal unit (either home or to another unit).
Measurements were made by 1 of the 3 observers (A.C., L.P., or A.F.F.). The intra- and interobserver coefficients of variation were 1.1% and 1.2%, respectively. The instrument accuracy was 0.25% to 0.50%, according to the manufacturer's data.
SOS measurements were compared with SOS reference values obtained in term and preterm infants in Portugal (31). Low bone strength (SOS <10th centile of the reference values) was the main outcome of the PN intervention.
The sample size was calculated at 85 infants based on power of 80% to detect changes in SOS of 30 m/s (circa 1% variation of the Portuguese reference values for those gestational ages (31)) or more over time (within subjects), based on a minimum of 3 measurements.
Normality of continuous numerical variables (SOS) was tested graphically and using 1-sample Kolmogorov-Smirnov test. As appropriate, data are expressed as mean and standard deviation or medians and quartiles, and compared using Student t test, 1-way analysis of variance, Mann-Whitney U test, or Kruskal-Wallis test. Spearman rank correlation was used to compare associations between variables, and analysis of covariance was performed to establish the level of associations between multiple measured variables. Proportions were compared between variables using either χ2 test or Fisher exact test, as appropriate. Proportions within 1 variable were compared using binomial test. Binary logistic regression analysis was used to identify determinants of the outcome (bone strength) in an explanatory model that included demographic (sex), clinical (GA and birth weight), and nutritional variables (duration of PN, intakes of energy, protein, Ca, P, Mg, and vitamin D). Survival experience (proportion of low bone strength) was compared using Wilcoxon (Gehan) statistic. P values <0.05 were considered significant. Statistical analysis was performed with XL STAT version 7.0 (Addinsoft, France), SPSS 11.0 (SPSS Inc, Chicago, IL), Statcalc (EpiInfo V6, Centers for Disease Control, Atlanta, GA), and Microsoft Excel 2000 (Microsoft Corp, Redmond, WA).
Ninety-one neonates were enrolled and randomized at birth, but 5 in the LD group interrupted PN within the first week after birth due to hypernatremia (2), oliguria (2), and hyperkalemia (1). Thus, only 86 were included in the study, with mean (standard error) GA of 29.6 weeks (2.1) and birth weight of 1262 g (0.356). Forty were assigned to the LD group and 46 to the HD group. The proportion of randomized neonates who dropped out early in the study was significantly higher in the LD group (Fisher exact test, P = 0.026); the reasons for dropouts are not likely to be related to parenteral Ca and P regimens. Both groups were similar in terms of GA, birth weight, and duration of PN, but girls predominated in the LD group (P = 0.004) (Table 1). No significant differences were found between groups regarding treatment with diuretics (LD 6/40 vs HD 7/46), prenatal steroids (LD 31/40 vs HD 33/46), and postnatal steroids (LD 2/40 vs HD 3/46).
Six-weekly SOS measurements were performed from birth to discharge. The number of patients and their postnatal age at each SOS measurement are shown in Table 2. No significant differences between groups were found in the proportion of SGA or LGA at any SOS measurement (data not shown).
Throughout the study period, most of the dropouts (15/23 in LD group and 20/29 in HD group) resulted from transfer to regional neonatal units or discharge based on a satisfactory growth and a favorable clinical outcome. Infants too immature to be transferred or who had been born at this maternity ward covering their residential area remained in the unit. Few infants dropped out from the study due to PN interruption because of electrolyte imbalance or other reasons or because of fluid restriction for management of patent ductus arteriosus or chronic lung disease; in the LD group: fluid restriction (3), hypophosphatemia (2), hypercalcemia (1), severe septicemia (1), and acute renal failure (1); in the HD group: fluid restriction (4), severe septicemia (2), and hypercalcemia (1). No apparent heterogeneity exists between the patients lost to follow-up and those kept in the study, the proportions of low SOS remaining stable in more advanced chronological ages in spite of progressive reduction of the sample size.
Daily intakes of energy and protein (Table 3) and of magnesium and vitamin D (Table 4), representing the sum of parenteral and enteral intake, did not differ significantly between the groups in the 24 hours preceding each SOS measurement. Daily intakes of Ca and P, representing the sum of parenteral and enteral intake, and the percentage of parenteral intake of Ca and P, in the 24 hours preceding each SOS measurement are shown in Table 5.
Regarding early Ca and P intake, it was noticed that PN was not initiated before the measurement 1 in 4 LD and 10 HD neonates because they were born during a weekend.
The HD group received significantly greater Ca and P intake 24 hours preceding measurement 1. Later on, the proportion of parenteral Ca and P intake in both groups decreased steadily until full enteral supply was reached. By measurement 2 (second week after birth) 25% of the infants in both groups still received, respectively, 95.7% and 81% of Ca intake and 96.4% and 82.5% of P intake by PN. By measurement 3 (third week after birth) 25% of the infants in both groups still received, respectively, 43.9% and 24.3% of Ca intake and 46.6% and 25.3% of P intake by PN (Table 5).
By the first SOS measurement (baseline), at a mean of 3.2 days after birth, no statistical differences between groups were found both in the mean SOS (m/s) (Table 6) and in the proportion of low SOS (Table 7), reflecting similar bone status at birth.
Subsequently, a progressively steady and significant decline of the mean SOS was recorded in the LD group from birth to the sixth week of life (Kruskal-Wallis test, P = 0.027) (Table 6 and Fig. 1). In the HD group, the mean SOS never decreased below the mean baseline value, remaining statistically stable (Kruskal-Wallis test, P = 0.976) (Table 6 and Fig. 1).
In the HD group, significantly higher mean SOS values (m/s) were found compared with those in the LD group from the fifth week of life onward (from measurement 5) (Table 6 and Fig. 1), in spite of infants already receiving full enteral feeds (Table 5).
Significantly less frequent low bone strength (SOS <10th centile) was found in the HD group compared with that in the LD group, already from the second week after birth (from measurement 2) (Wilcoxon, P = 0.0002) (Table 7 and Fig. 2), in spite of an increase in the proportion of infants receiving less Ca and P by PN or already receiving full enteral feeds (Table 5).
Binary logistic regression analysis revealed that Ca intake while on PN was the most significant factor on the explanatory model for low bone strength (r2 = 0.63 for the model; P = 0.077 for Ca intake by PN). The potential role of sex cannot be fully elicited, but no confounding due to this factor was highlighted in the analysis.
Despite improvements in nutrition, disturbed mineral metabolism due to delayed full enteral feeding and dependence on PN may have a negative effect on bone health in preterm infants (2). The hypothesis that early higher intake of Ca and P delivered by PN to preterm infants, within the wide range of proposed doses (5,6), may positively affect their bone strength was tested. There is strong evidence that Ca and P delivered by PN to preterm infants do influence bone strength evolution. At about 10 days after birth (measurement 2) 25% of the studied patients in both groups still received >81% of Ca and 82.5% of P intake by PN (Table 5).
The Ca:P ratio (mg) of 1.7:1 in PN solutions was chosen because it seems to promote better mineral retention in preterm infants than either lower or higher ratios (32). Parenteral Ca regimens assigned for both groups were within recommended doses (5). To maintain the fixed Ca:P ratio in PN solutions, the P intake assigned for the LD group was slightly lower (median 26.5 mg · kg−1 · day−1) than the low recommended limit (31 mg · kg−1 · day−1) (5,6). This only occurred by measurement 1, when the LD group had actually received a median intake of P 27.9 mg · kg−1 · day−1 (Fig. 5) due to small additional amounts of enteral feeds administered in some cases from the first day after birth.
The intake of Ca, P, magnesium, vitamin D, protein, and energy delivered by parenteral and enteral routes was assessed by the best estimate during the 24 hours preceding each SOS measurement. This estimate is a limitation of the study because it may not reflect the accumulated intake of nutrients delivered before, namely during the preceding week. Nevertheless, the preceding 24-hour period, as an indicator of the pattern of nutrient intake, proved to reflect a consistent homogeneity between groups except preceding the first SOS measurement, when the HD group received higher Ca and P intake, as expected.
Bone strength was assessed by SOS using the QUS method because this is a convenient, noninvasive method, standardized in term and preterm infants (11). SOS reflects not only bone density but also bone mass and cortical thickness, which are dependent on other nutrients and factors beyond mineral intake (9). To better evaluate the independent effect of 2 different parenteral regimens of Ca and P on bone strength, infants of both groups were submitted to similar parenteral and enteral protocols regarding protein, energy, magnesium, and vitamin D intakes and a similar protocol for early developmental care. The proportion of additional factors potentially affecting bone strength was found to be similar in both groups, including therapy with diuretics, prenatal and postnatal steroids, and the proportions of SGA and LGA infants, who were reported to have different SOS values compared with appropriate-for-gestational age infants (33,34).
No apparent heterogeneity exists between patients lost to follow-up and those kept in the study. Most of the dropouts resulted from transfer to regional neonatal units or home discharge based on a satisfactory growth and a favorable clinical outcome, which were similar in both groups. The difference between proportions of low SOS remains significant in more advanced chronological ages in spite of the progressive reduction of the sample size, which we believe can be interpreted as an indication of the strong significance of the association.
Almost all of the studies assessing bone strength in preterm infants have reported a postnatal SOS decline in the first weeks after birth (7,14–22), which seems to be more pronounced in more immature infants, and dependent on more prolonged PN (17,21). In those studies, daily Ca and P intakes have not been specified as per kilogram of body weight, and the effect of different doses of those minerals on bone strength has not been tested. In the present study, a significant steady decline in the mean SOS was registered in the LD group from birth to discharge. This decline is clinically important because by the sixth week after birth (about 35 weeks of corrected postnatal age) bone strength not only failed to catch up with the values measured close to birth but also continued to decrease. In previous studies, bone strength catch-up has been registered from the end of the second month of corrected postnatal age up to 18 months of life. (15,21,22). In the present study, the mean SOS values in the HD group never decreased below the mean value recorded close to birth. This trial has not been designed to evaluate the time required for bone strength catch-up.
To the best of our knowledge, the independent preventive effect of early high parenteral Ca and P intake on short-term postnatal bone strength decrease has not yet been reported in preterm infants. The finding of less frequent low bone strength in the HD group longitudinally recorded in the study period confirms the positive effect of an early and a high parenteral mineral intake. Interestingly, this positive effect continues after the reduction of parenteral mineral intake or the achievement of full enteral feeding. Prestridge et al (35) also tested the effect of 2 different Ca and P regimens provided by PN in preterm infants and found a greater bone mineral content estimated by single-photon absorptiometry in those receiving early greater mineral intake. This effect persisted several weeks beyond the PN study interval, which the authors have called a “memory effect” from early nutritional intervention. Because premature infants miss the intrauterine period of greater mineral accretion, it may be speculated that these infants are more sensitive to an early higher mineral intake leading to a reduction of early bone demineralization (32), possibly reflecting on bone strength beyond the period of intervention.
To summarize, early assigned parenteral intake of Ca 75 mg · kg−1 · day−1 and P 44 mg · kg−1 · day−1 has significantly contributed to prevent bone strength decrease in preterm infants during the first weeks of life. Longer follow-up to evaluate potential long-term benefits of this intervention on the bone status in this population is needed.
1. Lyon AJ, McIntosh N, Wheeler K, et al
. Radiological rickets in extremely low birthweight infants. Pediatr Radiol 1987; 17:56–58.
2. Rauch F, Schoenau E. Skeletal development in premature infants: a review of bone physiology beyond nutritional aspects. Arch Dis Child Fetal Neonatal 2002; 82:F82–F85.
3. Klein GL. Metabolic bone disease of total parenteral nutrition. Nutrition 1998; 14:149–152.
4. Pereira-da-Silva L, Nurmamodo A, Videira Amaral JM, et al
. Compatibility of calcium and phosphate in four parenteral nutrition solutions for preterm neonates. Am J Health Syst Pharm 2003; 60:1041–1044.
5. Mirtallo J, Canada T, Johnson D, et al
. Safe practices for parenteral nutrition. J Parenter Enteral Nutr 2004; 28:S39–S70.
6. Koletzko B, Goulet O, Hunt J, et al
, for the Parenteral Nutrition Guidelines Working Group. Iron, minerals, and trace elements. J Pediatr Gastroenterol Nutr 2005; 41:S39–S46.
7. Litmanovitz I, Dolfin T, Friedland O, et al
. Early physical activity intervention prevents decrease of bone strength in very low birth weight infants. Pediatrics 2003; 112(1 Pt 1):15–19.
8. Faerk J, Peitersen B, Petersen S, et al
. Bone mineralisation in premature infants cannot be predicted from serum alkaline phosphatase or serum phosphate. Arch Dis Child Fetal Neonatal Ed 2002; 87:F133–F136.
9. Teitelbaum JE, Rodriguez RJ, Ashmeade TL, et al
. Quantitative ultrasound
in the evaluation of bone status in premature and full-term infants. J Clin Densitom 2006; 9:358–362.
10. Foldes AJ, Rimon A, Keinan DD, et al
. Quantitative ultrasound
of the tibia: a novel approach for assessment of bone status. Bone 1995; 17:363–367.
11. Littner Y, Mandel D, Mimouni FB, et al
. Bone ultrasound velocity curves of newly born term and preterm infants. J Pediatr Endocrinol Metab 2003; 16:43–47.
12. McDevitt H, Ahmed SF. Quantitative ultrasound
assessment of bone health in the neonate. Neonatology 2009; 9:2–11.
13. Gianni ML, Mora S, Roggero P, et al
. Quantitative ultrasound
and dual-energy X ray absorptiometry in bone status assessment of ex-preterm infants. Arch Dis Child Fetal Neonatal Ed 2008; 93:F146–F147.
14. Litmanovitz I, Dolfin T, Regev R, et al
. Bone turnover markers and bone strength during the first week of life in very low birthweight premature infants. J Perinat Med 2004; 32:58–61.
15. Ritschl E, Wehmeijer K, de Terlizzi F, et al
. Assessment of skeletal development in preterm and term infants by quantitative ultrasound
. Pediatr Res 2005; 58:341–346.
16. Liao XP, Zhang WL, He J, et al
. Bone measurements of infants in the first 3 months of life by quantitative ultrasound
: the influence of gestational age, season, and postnatal age. Pediatr Radiol 2005; 35:847–853.
17. Tomlinson C, McDevitt H, Ahmed SF, et al
. Longitudinal changes in bone health as assessed by the speed of sound in very low birth weight preterm infants. J Pediatr 2006; 148:450–455.
18. Mercy J, Dillon B, Morris J, et al
. Relationship of tibial speed of sound and lower limb length to nutrient intake in preterm infants. Arch Dis Child Fetal Neonatal Ed 2007; 92:381–385.
19. Ashmeade T, Pereda L, Chen M, et al
. Longitudinal measurements of bone status in preterm infants. J Pediatr Endocrinol Metab 2007; 20:415–424.
20. Fewtrell MS, Loh KL, Chomtho S, et al
. Quantitative ultrasound
(QUS): a useful tool for monitoring bone health in preterm infants? Acta Paediatr 2008; 97:1625–1630.
21. McDevitt H, Tomlinson C, White MP, et al
. Changes in quantitative ultrasound
in infants born at less than 32 weeks' gestation over the first 2 years of life: influence of clinical and biochemical changes. Calcif Tissue Int 2007; 81:263–269.
22. Roggero P, Gianni ML, Orsi A, et al
. Postnatal “speed of sound” decline in preterm infants: an exploratory study. J Pediatr Gastroenterol Nutr 2007; 45:615–617.
23. Kramer MS, Platt RW, Wen SW, et al
. A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics 2001; 108:E35.
24. A.S.P.E.N. Board of Directors and the Clinical Guidelines Task Force. Guidelines for the use of parenteral and enteral nutrition in adult and pediatric patients. Section XII: administration of specialized nutrition support—issues unique to pediatrics. J Parent Enteral Nutr
25. Greene HL, Hambidge KM, Schanler R, et al
. Guidelines for the use of vitamins, trace elements, calcium, magnesium, and phosphorus in infants and children receiving total parenteral nutrition: report of the Subcommittee on Pediatric Parenteral Nutrient Requirements from the Committee on Clinical Practice Issues of the American Society for Clinical Nutrition. Am J Clin Nutr 1988; 48:1324–1342.
26. Koletzko B, Goulet O, Hunt J, et al
. Guidelines on Pediatric Parenteral Nutrition of the European Society of Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) and the European Society for Clinical Nutrition and Metabolism (ESPEN), Supported by the European Society of Pediatric Research (ESPR). J Pediatr Gastroenterol Nutr 2005; 41:S1–S11.
27. Ziegler EE. Nutrient requirements of premature infants. Nestle Nutr Workshop Ser Pediatr Program 2007; 59:161–172.
28. Thomas JL, Reichelderfer TE. Premature infants: analysis of serum during the first seven weeks. Clin Chem 1968; 14:272–280.
29. Eliakim A, Nemet D, Freidland O, et al
. Spontaneous activity in premature infants affects bone strength. J Perinatol 2002; 22:650–652.
30. Weiss M, Ben-Shlomo A, Hagag P, et al
. Discrimination of proximal hip fracture by quantitative ultrasound
measurement at the radius. Osteoporos Int 2000; 11:411–416.
31. Pereira-da-Silva L, Costa AB, Pereira L, et al
. Reference values for bone strength in term and preterm neonates assessed by the quantitative ultrasound
method [abstract]. Early Hum Dev 2008; 84:S110.
32. Pelegano JF, Rowe JC, Carey DE, et al
. Effect of calcium/phosphorus ratio on mineral retention in parenterally fed premature infants. J Pediatr Gastroenterol Nutr 1991; 12:351–355.
33. Littner Y, Mandel D, Mimouni FB, et al
. Bone ultrasound velocity of infants born small for gestational age. J Pediatr Endocrinol Metab 2005; 18:793–797.
34. Littner Y, Mandel D, Mimouni FB, et al
. Decreased bone ultrasound velocity in large-for-gestational-age infants. J Perinatol 2004; 24:21–23.
35. Prestridge LL, Schanler RJ, Shulman RJ, et al
. Effect of parenteral calcium and phosphorus therapy on mineral retention and bone mineral content in very low birth weight infants. J Pediatr 1993; 122:761–768.