- Email: [email protected]
Calcium Absorption in Very Low Birth Weight Infants with and without Bronchopulmonary Dysplasia
Calcium Absorption in Very Low Birth Weight Infants with and without Bronchopulmonary Dysplasia Penni D. Hicks, PhD, RD, Stefanie P. Rogers, MD, Keli M. Hawthorne, MS, RD, Zhensheng Chen, PhD, and Steven A. Abrams, MD Objective To evaluate the effects of early bronchopulmonary dysplasia (BPD) on calcium (Ca) metabolism and growth in very low birth weight (VLBW) infants. Study design A dual-tracer, stable isotope method was used to assess Ca absorption in VLBW infants. Infants with early BPD received energy-dense feedings and mild fluid restriction. Results Sixteen of 41 preterm infants were classified as having early BPD. Fractional Ca absorption (early BPD, 58.4 4.6% versus no early BPD, 50.3 4.0%, P = .2), total Ca absorption (early BPD, 127 14 mg/kg/d versus no early BPD, 104 9 mg/kg/d, P = .9), and Ca retention (early BPD, 99.6 10.0 mg/kg/d versus no early BPD, 91.0 9.8 mg/kg/d, P = .2) were similar among groups. There was no significant difference in weight gain, linear growth, or head circumference growth between groups. Conclusions The ability of VLBW infants with early BPD and fluid restriction to grow and accrete calcium is similar to those without early BPD. The use of high caloric density feedings in VLBW infants with early BPD can help achieve bone and overall growth outcomes close to those achievable in utero. (J Pediatr 2011;158:885-90). See editorial, p 876
B
ronchopulmonary dysplasia (BPD) is a common sequela of preterm birth. Affected infants are more likely to have respiratory disorders in later infancy, to be hospitalized in the first 1 to 2 years after birth, and to have abnormal growth and neurodevelopment. The incidence of BPD is typically 20% to 30% in very low birth weight (VLBW) infants.1 Adequate nutrition is a key aspect of care for the infant with BPD.2 Malnutrition can delay somatic growth and the development of new pulmonary alveoli and can decrease muscle strength, making successful weaning from mechanical ventilation less likely. However, it is difficult to provide adequate nutritional support to VLBW infants who have BPD because it is often necessary to limit fluid intake below 150 to 170 mL/kg/d, which is usually needed when standard feeding strategies are implemented. It is common clinical practice to alter nutrition therapy in infants at risk of BPD early in their course, well before 36 weeks’ postmenstrual age (PMA), when the diagnosis of BPD is formally made. The development of severe bone demineralization may exacerbate respiratory and other problems due to weakness of rib strength needed to support lung expansion. In its most severe form, rickets can develop leading to fractures and growth failure.3-5 Rickets in premature infants is almost exclusively a deficiency of the primary bone minerals calcium (Ca) and phosphorus (Phos) rather than vitamin D deficiency. It is often necessary to use high energy density feedings for infants with BPD. Available mineral absorption data using high caloric density feedings are minimal and the solubility of these feedings may decrease nutrient absorption.6 High caloric density infant feedings using either formula or human milk have been studied minimally. The primary objective of our study was to evaluate the effects of early BPD on Ca metabolism and bone mineral status in VLBW infants. The secondary goal was to evaluate growth and body composition in these infants when provided with different feeding regimens used in the neonatal intensive care unit (NICU).
Methods Infants were recruited from the NICU at Texas Children’s Hospital in Houston, Texas. Eligible infants were born between 23 and 32 6/7 weeks’ gestation, 500 to 1500 g, and free of major congenital anomalies. Infants were excluded from the
BPD Ca NICU Phos PMA VLBW
Bronchopulmonary dysplasia Calcium Neonatal intensive care unit Phosphorus Postmenstrual age Very low birth weight
From the United States Department of Agriculture/ Agricultural Research Service (P.H., S.R., K.H., Z.C., S.A.), Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, TX; and the Section of Neonatology (S.R., S.A.), Department of Pediatrics, Baylor College of Medicine and Texas Children’s Hospital, Houston, TX Supported by the Gerber Foundation. The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2011 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2010.12.034
885
THE JOURNAL OF PEDIATRICS
www.jpeds.com
study if they had necrotizing enterocolitis requiring surgical intervention or were unlikely to survive. The study was approved by the Institutional Review Board for Human Subject Research of Baylor College of Medicine and Affiliated Hospitals. Informed written consent was obtained from the parents. Infants were fed either preterm formula or fortified human milk, based on the mother’s decision to provide human milk (Table I). Infants without early BPD were fed according to standard NICU protocol, which provided 150 to 170 mL/ kg/d of a 24 kcal/oz preterm formula (Similac Special Care, Abbott Nutritionals, Columbus, Ohio) or human milk that was fortified with 4 kcal/oz of a commercial human milk fortifier. Infants at risk of early BPD were provided with preterm formula or fortified human milk at volumes of 130 to #150 mL/kg/d and/or at a caloric density >24 kcal/oz.
Vol. 158, No. 6 The decision for fluid restriction and caloric density of the feeding was made by the individual attending physician caring for each infant. Generally, fluid restriction was instituted based on an infant receiving >40% oxygen and on CPAP or mechanical ventilation. The subject was assigned to the early BPD group at the time of the absorption study, based on the need for respiratory support combined with the use of a caloric density >24 kcal/oz. At the discretion of the parents, either a cow milk–based (Similac Human Milk Fortifier, Abbott Nutritionals, Columbus, Ohio) or a human milk–derived (Protact +, Prolacta Bioscience, Monrovia, California) human milk fortifier was used for fortification of human milk (Table I). Fortification was typically begun when human milk feeds reached 100 mL/kg/d.
Table I. Subject classification Classification Feed type† Early BPD PMA at study PNA at study Respiratory support Respiratory support and caloric density ID No Early BPD (wk) (wk) before study* at study* (kcal/oz) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
No early BPD No early BPD No early BPD No early BPD No early BPD Early BPD Early BPD Early BPD No early BPD Early BPD No early BPD Early BPD No early BPD Early BPD No early BPD No early BPD Early BPD No early BPD No early BPD Early BPD Early BPD Early BPD No early BPD Early BPD No early BPD No early BPD Early BPD No early BPD No early BPD No early BPD No early BPD Early BPD Early BPD Early BPD No early BPD No early BPD Early BPD No early BPD No early BPD No early BPD No early BPD
33.3 33.5 33.6 36.0 36.0 33.6 33.6 34.0 31.1 32.0 33.1 34.6 33.0 33.0 33.0 34.1 37.6 32.6 32.0 34.4 34.4 33.0 33.1 32.3 31.5 32.0 30.2 33.5 31.6 34.6 33.3 31.6 33.6 33.6 33.6 33.6 30.0 33.6 33.6 33.0 33.6
3.3 5.5 3.6 6.0 6.0 2.6 2.6 6.0 3.1 3.0 2.1 5.6 3.0 3.0 3.0 6.1 10.6 5.6 5.0 4.4 4.4 7.0 3.1 6.3 3.5 4.0 6.2 4.5 4.6 10.6 2.3 6.6 10.6 10.6 3.6 3.6 4.0 2.6 2.6 4.0 4.6
RA NCPAP NC RA NCPAP NCPAP ETT NCPAP NC NCPAP RA NCPAP NCPAP NCPAP ETT NC NC NC NC RA RA NCPAP RA NCPAP RA RA RA RA NCPAP ETT NCPAP ETT NCPAP ETT NCPAP NCPAP ETT RA RA RA RA
RA NC RA RA RA NCPAP PPV NCPAP RA NC RA NCPAP RA RA RA NC NC RA NC RA RA NC RA NCPAP RA RA ETT RA NCPAP NCPAP RA NCPAP NC NCPAP NC RA ETT RA RA RA RA
FHM-H 24 FHM-H 25 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 26 FHM-H 26 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 24 Formula 24 Formula 27 Formula 24 Formula 24 FHM-CT 27 FHM-CT 27 Formula 27 FHM-C 24 Formula 27 FHM-C 24 FHM-C 24 FHM-C 25 Formula 24 FHM-C 24 FHM-C 24 Formula 24 Formula 24 FHM-C 25 FHM-C 24 FHM-C 24 FHM-C 24 FHM-C 24 FHM-C 24 FHM-C 25 FHM-C 25 FHM-C 25
Meds at abs study
BPD diagnosis at 36 wk
No No No No No Caffeine Yes Caffeine No Caffeine Yes Caffeine No No Caffeine No Caffeine Yes No Caffeine Yes Caffeine No Caffeine No Furosemide, Thiozide Yes Caffeine No Caffeine Yes Caffeine No Caffeine No Caffeine, Furosemide Yes Caffeine No Caffeine No Caffeine No Caffeine No Yes No Caffeine No Caffeine Yes Caffeine No Caffeine Yes Caffeine Yes Caffeine Yes Caffeine No Caffeine No Thiozide Yes No No Caffeine No Caffeine No Caffeine Caffeine
PNA, postnatal age. *RA indicates room air; NCPAP, nasal continuous positive airway pressure; NC, nasal cannulae; ETT, endotracheal tube; PPV, positive pressure ventilation. All subjects with NCPAP, NC, ETT, and PPV were provided supplemental oxygen. †FHM-H indicates fortified human milk with human milk–derived fortifier, formula preterm formula; FHM-CT, fortified human milk with cow milk–derived fortifier and transitional formula (Neosure); FHM-C, fortified human milk with cow milk–derived fortifier.
886
Hicks et al
ORIGINAL ARTICLES
June 2011 Absorption Study Approximately 2 weeks after infants had reached full enteral feedings (defined as at least 120 mL/kg/d and having parenteral nutrition discontinued) and were determined to be stable enough for the study, Ca absorption was measured by a dualtracer stable isotope method as described previously.7-10 Briefly, 18 hours before the study, one-half volume of a single feeding was mixed with 44Ca (2 mg) and refrigerated until the study began to allow for equilibration. The isotope was mixed with either fortified human milk or preterm formula. The next morning, a urine collection bag (First-Time Urine Specimen Collector, Hollister Incorporated, Libertyville, Illinois) was attached to the infant. Diapers were weighed before and after use to ensure that all urine was collected. A plastic liner with cotton balls was also placed in the diaper to capture any urine that leaked. The urine from the cotton balls was extracted and added to the urine collection. Urine was extracted from the cotton balls for only three infants. The infants were given 46Ca (0.01 mg) intravenously over a period of 1 minute. The feeding with the premixed oral isotope then was given via orogastric tube, and the remaining feeding volume was given immediately. Urine was collected for 24 hours after the isotope administration. Fractional Ca absorption was calculated from the ratio of the fraction of the oral tracer dose to the intravenous tracer dose recovered in the urine during the 24-hour collection. Total Ca absorbed was calculated by multiplying the Ca intake by the fractional absorption. Approximate net Ca retention was determined using estimates for endogenous fecal excretion derived from previous studies in premature infants and the measured urinary excretion.10 Isotopic enrichment was determined using a magnetic sector Finnigan MAT 261 thermal ionization mass spectrometer (Bremen, Germany). Anthropometrics and Bone Health Markers Daily weight and weekly length and head circumference measures were determined by bedside nursing staff, with continual training for proper anthropometric measures. Nutritionrelated labs, serum Ca, Phos, and alkaline phosphatase activity were recorded when measured for routine care. At study onset and at 6 weeks or discharge, all infants received a speed of sound ultrasound (Sunlight Omnisense 7000p, Sunlight Medical, Israel) of the left pretibial area. Infants received a body composition measurement using air displacement plethysmography (PEA POD, Life Measurement Inc, Concord, California) at 6 weeks or discharge if they were medically stable and not receiving oxygen therapy or ventilatory support. Quality control procedures for the speed of sound ultrasound and air displacement plethysmography machines were performed each day before the first subject was tested. The speed of sound ultrasound machine uses a phantom to measure each probe and each probe measurement consists of three cycles. The difference in the speed of sound measured and speed of sound calculated values is plotted over time. The acceptable difference is between 50 and 50 m/s. For the air
displacement plethysmography machine, five tests were performed: hardware analysis, scale calibration, mass calibration, auto run assessment, and volume assessment. All tests must be successfully passed to perform subject measurements. Statistics The primary outcome was the assessment of Ca absorption. A sample size of n = 30, with at least 15 in each group, was chosen such that a significant difference in intake-adjusted fractional Ca absorption of 10%, from a predicted baseline of 55% 8%, could be identified with an 80% power at a P < .05 (null hypothesis being accepted would be <10% difference) between infants with and without early BPD. Differences in fractional Ca absorption and estimated Ca retention were assessed using analysis of variance with presence or absence of early BPD as a covariate. Growth and biochemical variables were assessed weekly and analyzed by repeated-measures analysis of covariance with feeding type (fortified human milk or preterm formula) and presence or absence of early BPD as independent variables and the baseline measurement as a covariate. Speed-of-sound ultrasound was analyzed by two-way analysis of variance with feeding type (fortified human milk or preterm formula) and feed concentration (24 kcal/oz or 27 kcal/ oz) as the independent variables. Growth was compared with standards expected for infants of these gestational ages. A target weight gain of 15 g/kg/d, length gain of 1 cm/wk, and head circumference growth of 1 cm/wk was used to approximate typical intrauterine growth.11,12 All results are presented as mean SEM unless otherwise indicated.
Results Forty-one premature infants were studied (Table II). Of these, 16 infants were classified as having early BPD. As expected, gestational age and birth weight were significantly different with early BPD infants born earlier and weighing less than infants without early BPD. Postmenstrual age and weight at the time of the Ca absorption study were similar. Feeding volumes were significantly lower in early BPD infants, reflecting fluid restriction, but both groups had adequate energy intakes of at least 120 kcal/kg/d. Calcium intake, fractional Ca absorption, total Ca absorption, urinary Ca excretion, and approximate Ca retention were similar among groups (Table III). Using regression analysis, there was no effect of feeding type, fortified human milk or preterm formula, on Ca absorption (P = .3). There was a significant positive correlation between total Ca absorbed and PMA at the time of study (r = 0.5, P = .001), but there was no significant relationship between fractional Ca absorption and PMA (r = 0.3, P = .06) (Figure; available at www.jpeds.com). There was no significant relationship between Ca intake and fractional absorption (r = 0.12, P = .4). At the time of the study, three subjects were taking diuretics and 31 subjects were treated with caffeine. Mean Ca absorption for the three subjects receiving diuretics was
Calcium Absorption in Very Low Birth Weight Infants with and without Bronchopulmonary Dysplasia
887
THE JOURNAL OF PEDIATRICS
www.jpeds.com
Table II. Characteristics of the study population Early BPD (n = 16) Birth weight (g) 909 66 Gestational age (wk) 27.0 0.5 Postmenstrual age at study (wk) 32.7 0.5 Weight at study (g) 1514 98 Energy intake at study (kcal/kg/d) 125 3 Feeding volume at study (mL/kg/d) 145.0 2.3
No early BPD (n = 25)
P value
1230 38 29.1 0.3 33.2 0.2 1668 48 123 2 153.9 1.5
>.001 .004 NS NS NS .003
All results are mean SEM.
78.5% 0.9%, compared with subjects free of diuretics (51.5% 3.1%). Mean urinary Ca excretion value of the subjects on diuretics was 3.7 1.4 mg/kg/d, compared with the subjects free of diuretics (2.3 0.3 mg/kg/d). Removing the three subjects on diuretics from the overall analyses, Ca absorption, urinary Ca excretion, and approximate Ca retention remain similar among groups. Both Ca absorption and urinary Ca excretion values were similar in the subjects on or free of caffeine. Mean Ca absorption for those receiving caffeine was 56.3% 6.0%, compared with the subjects not receiving caffeine (52.5% 3.6%, P = .6). Mean urinary Ca excretion values for subjects on caffeine was 2.4 0.3 mg/ kg/d compared with the subjects not receiving caffeine (2.5 0.6 mg/kg/d, P = .6). There was no significant difference in weight gain, linear growth, or head circumference gain over the study period between groups (Table IV). There was no significant difference in speed of sound ultrasound measurement between groups over the study period. Eight infants with early BPD and 21 infants without early BPD had air displacement plethysmography measurements. There was no significant difference in percent fat mass between groups. During the study, the average alkaline phosphatase activity levels were significantly greater in infants with early BPD than infants without early BPD. Serum Phos levels were significantly less in infants with early BPD than infants without early BPD infants (Table IV). Effects of Feeding Type During the Ca absorption study, 15 subjects were consuming fortified human milk containing the human milk derived fortifier and 26 subjects were consuming either human milk containing the cow milk–derived fortifier or preterm formula (Table V; available at www.jpeds.com). There was no Table III. Calcium absorption outcomes Early BPD
No early BPD
(n =16)
(n = 25)
P value
200 8 50.3 4.0 104 9 2.41 0.33 91.0 9.8
NS NS NS NS NS
Ca intake (mg/kg/d) 210 12 Fractional Ca absorption (%) 58.4 4.6 Total Ca absorption (mg/kg/d) 127 14 Urinary Ca (mg/kg/d) 2.38 0.45 Approximate Ca retention (mg/kg/d) 99.6 10.0 All results are mean SEM.
888
Vol. 158, No. 6 significant difference in Ca absorption among subjects consuming human milk containing the human milk– derived fortifier (47.8% 4.8%) and subjects consuming human milk containing the cow milk–derived fortifier or preterm formula (56.7% 3.9%, P = .1). Calcium intake was significantly lower in those subjects consuming human milk containing the human milk–derived fortifier (159.0 5.5 mg/kg/d), compared with those subjects consuming human milk containing the cow milk–derived fortifier or preterm formula (229.3 5.6 mg/kg/d, P < .01). Calcium retention was also significantly lower in subjects consuming human milk containing the human milk–derived fortifier (63.4 7.7 mg/kg/d) compared with subjects consuming human milk containing the cow milk–derived fortifier or preterm formula (112.3 8.6 mg/kg/d, P < .01). Weight gain, linear growth, and head circumference growth were similar between subjects consuming human milk containing the human milk–derived fortifier and those consuming human milk containing the cow milk–derived fortifier or preterm formula (Table VI; available at www.jpeds.com). There was a significant difference in speed of sound ultrasound measurement between feeding type groups at the beginning of study and the end of study. There was no significant difference in decrease in speed-of-sound ultrasound measure during the study period. Eleven infants consuming human milk containing the human milk–derived fortifier and 18 infants consuming human milk containing the cow milk–derived fortifier or preterm formula had air
Table IV. Growth, laboratory measurements, and body composition Early BPD
Weight gain* (g/kg/d) Linear growth* (cm/wk) Head circumference growth* (cm/wk) Alkaline phosphatase activity (IU/L) Phosphorus (mg/dL) SOS U/S at start of study† (m/s) Postmenstrual age range at start of study† (wk) Postnatal age range at start of study† (wk) SOS U/S at end of studyz (m/s) Postmenstrual age range at end of studyz (wk) Postnatal age range at end of studyz (wk) Fat mass at end of studyx (%) Postmenstrual age range at end of studyx (wk) Postnatal age range at end of studyx (wk)
No early BPD
(n = 16)
(n = 25)
P value
14.6 0.8 1.0 0.1 1.1 0.1
14.6 0.5 1.2 0.1 0.9 0.1
NS NS NS
390 22
286 19
.001
5.9 0.2 2844 38
6.5 0.2 2881 21
.01 NS
28.4-37.4
30.4-33.6
1.4-11.4
3.3-3.6
2787 40 33.4-40.6
2838 27 34.3-39.6
6.3-3.6
7.1-9.3
13.3 1.2 34.6-39.3
12.8 0.7 34.6-38.0
4.6-15.3
3.6-10.0
NS
NS
SOS U/S, speed-of-sound ultrasound. All data are mean SEM unless otherwise indicated. *Gain over study period. †n = 15 Early BPD, n = 21 no early BPD. zn = 13 Early BPD, n = 20 no early BPD. xn = 8 Early BPD, n = 21 no early BPD.
Hicks et al
ORIGINAL ARTICLES
June 2011 displacement plethysmography measurements. There was no significant difference in percent fat mass between groups (Table VI). During the study, the average alkaline phosphatase activity levels and Phos levels were similar in subjects consuming human milk containing the human milk–derived fortifier and those consuming human milk containing the cow milk–derived fortifier or preterm formula (Table VI). Categorizing infants into three feeding types, 15 subjects were consuming human milk containing the human milk– derived fortifier, 17 subjects were consuming human milk containing the cow milk–derived fortifier, and nine subjects were consuming preterm formula. Using post hoc analysis, there were no significant differences in Ca absorption among the three groups. Mean fractional Ca absorption for the group consuming human milk containing the cow milk–derived fortifier was 59.8% 4.9% and for the group consuming preterm formula was 50.9% 6.4% (P = .2).
Discussion We found that when provided with an adequate nutritional intake of at least 120 kcal/kg/d, infants with and without early BPD achieve both bone and overall growth outcomes close to those achievable in utero. The source of extra protein and minerals, whether preterm formula or human milk fortifier, and type of fortifier did not affect these outcomes as long as energy intake was adequate. Because we found no significant differences in feeding type, for analyses we combined those consuming preterm formula with those consuming human milk containing the cow milk–derived fortifier as they both contain cow milk protein and have been found to have similar bioavailability.7 We found no significant difference in Ca intake, fractional Ca absorption, and Ca retention between the groups with early BPD and without early BPD. Overall Ca absorption rates were high and retention of Ca from both fortified human milk and preterm formula reached the expected in utero accretion rate of at least 90 mg/kg/d.13 Although obtaining intrauterine accretion rates remain the goal in premature infants, it is unclear whether it is an appropriate marker of adequacy.14 Bone remodeling occurs in both term and preterm infants after birth, with increased bone reabsorption and a reduction in bone mineral apparent density.15 Calcium retention between 60 and 90 mg/kg/d appears to allow normal mineralization and decreases the risk of fractures and osteopenia.14 Infants consuming human milk containing the human milk–derived fortifier did not meet these calcium retention levels due to the decreased calcium content of this fortifier. Thus, the statistical nonsignificance of our primary outcome, total Ca absorption, may potentially be the result of the lower intake of some infants. Regardless, the observed similar calcium absorption efficiency across a range of feedings and intakes suggests comparable bioavailability of the calcium sources. However, further balance studies at a range of intakes not achieved in this study would be needed to confirm this.
The biochemical markers of bone mineralization we assessed were serum alkaline phosphatase activity and Phos. The early BPD group did have higher alkaline phosphatase activity and lower Phos values than the group without early BPD; however, none of the values were outside the normal limits seen in growing preterm infants. Although there are no values of alkaline phosphatase activity that indicate osteopenia, these values remains far below the threshold alkaline phosphatase activity of 900 IU/L that Backstr€ om16 concluded gave a 88% sensitivity and 71% specificity to detect osteopenia. There are very few studies investigating nutrition in VLBW infants with BPD. A single study in Canada compared the use of 27 kcal/oz formula with 24 kcal/oz formula in infants with BPD who were nearing hospital discharge and were >37 weeks’ PMA. They found that in infants recovering from BPD, growth failure was not related to malabsorption of nutrients but rather inadequate nutrient intakes.17 In the United Kingdom, Fewtrell18 compared high caloric density formula (30 kcal/oz) with 24 kcal/oz formula in infants with BPD. They found no difference in growth between groups, despite higher caloric intake in the nutrient-dense group. Although both groups received volumes above that used in the United States, the outcomes of our study are in agreement with these data as we found no significant difference in growth parameters between groups. Speed-of-Sound ultrasound is an inexpensive, welltolerated bedside tool with good reproducibility.19 In preterm infants, speed-of-sound ultrasound values decrease after birth. Tomlinson showed a median decrease of 3.2% and Fewtrell showed a decrease of 2.6%.4,20 Our data are consistent with these findings. This may reflect less mechanical stimulation after birth, thus lowering bone mass accrual. The positive correlation between speed-of-sound ultrasound and greater gestational age seen in these infants is in agreement with the findings of others and may be due to a greater transfer of Ca that occurs during later gestational age.21-23 The use of speed-of-sound ultrasound in preterm infants to measure bone status has not been validated against dual-energy x-ray absorptiometry; there are no reference standards to refer to, and it has not been found to predict biochemical indicators of metabolic bone disease. However, it may be useful in monitoring response to dietary and exercise interventions or as a predictor of long-term bone health in this population.21 Body composition measurement is an important indicator of nutritional status. It can identify the composition of weight gain and can be a useful tool to evaluate the efficacy of nutritional interventions. Recent availability of air displacement plethysmography measurements have allowed the study of VLBW infants. Both groups had similar percent fat mass when measured at an average PMA of 35 weeks (range of 31 to 39 weeks) and postnatal age range of 4 to 15 weeks. These infants were assessed at markedly different times in their development, but, without clinical standards for body composition in premature infants, we cannot be certain of the clinical relevance of this.
Calcium Absorption in Very Low Birth Weight Infants with and without Bronchopulmonary Dysplasia
889
THE JOURNAL OF PEDIATRICS
www.jpeds.com
There were a number of limitations in our study. We used an extrinsic tracer, as opposed to an intrinsic or biological tracer, in the products used to assess Ca absorption, which assumes a complete exchange has taken place between the tracer and the native calcium. It is unlikely that any significant proportion of the fortified human milk or preterm formula did not equilibrate with the tracer, as this approach has been widely used and validated for milk-based products.24 We also used endogenous fecal excretion rates that were previously determined from a premature infant study; thus, we were only able to report approximate Ca retention. Another potential limitation was that vitamin D status was not assessed in our study. These data support the use of high caloric density feedings in VLBW infants at risk of BPD to achieve both bone and overall growth outcomes close to those achievable in utero. Growth failure and bone mineral deficiencies are not due to inadequate utilization of nutrients but may be due to inadequate intake and early nutritional intervention can reduce the prevalence and severity of these.25 Our results may not be applicable in infants with longstanding BPD, although the feeding of preterm formula is uncommonly continued in our institution after infants are discharged from the hospital or are over about 3 kg. Long-term data are necessary to determine if physiological outcomes are dependent on achieving intrauterine Ca accretion. n Submitted for publication Mar 16, 2010; last revision received Nov 16, 2010; accepted Dec 22, 2010. Reprint requests: Dr Penni D. Hicks, PhD, RD, Children’s Nutrition Research Center at Baylor College of Medicine, 1100 Bates Street, Room 7005, Houston, TX 77030. E-mail: [email protected]
References 1. Van Marter LJ. Epidemiology of bronchopulmonary dysplasia. Semin Fetal Neonatal Med 2009;14:358-66. 2. Bancalari E, Wilson-Costello D, Iben SC. Management of infants with bronchopulmonary dysplasia in North America. Early Hum Dev 2005; 81:171-9. 3. Fewtrell MS, Cole TJ, Bishop NJ, Lucas A. Neonatal factors predicting childhood height in preterm infants: evidence for a persisting effect of early metabolic bone disease? J Pediatr 2000;137:668-73. 4. Tomlinson C, McDevitt H, Ahmed SF, White MP. Longitudinal changes in bone health as assessed by the speed of sound in very low birth weight preterm infants. J Pediatr 2006;148:450-5. 5. Mitchell SM, Rogers SP, Hicks PD, Hawthorne KM, Parker BR, Abrams SA. High frequencies of elevated alkaline phosphatase activity and rickets exist in extremely low birth weight infants despite current nutritional support. BMC Pediatr 2009;9:47. 6. Goss S, Prushko J, Bogner R. Factors affecting calcium precipitation during neutralisation in a simulated intestinal environment. J Pharm Sci 2010;99:4183-91.
890
Vol. 158, No. 6 7. Abrams SA, Schanler RJ, Yergey AL, Vieira NE, Bronner F. Compartmental analysis of calcium metabolism in very low birth infants. Pediatr Res 1994;36:424-8. 8. Yergey AL, Abrams SA, Vieira NE, Eastell R, Hillman LS, Covell DG. Recent studies of human calcium metabolism using stable isotopic tracers. Can J Physiol Pharmacol 1990;68:973-6. 9. Abrams SA. Pubertal changes in kinetics in girls assessed using 42Ca. Pediatr Res 1993;34:455-9. 10. Abrams SA, Esteban NV, Vieira NE, Yergey AL. Dual tracer stable isotopic assessment of calcium absorption and endogenous fecal excretion in low birth weight infants. Pediatr Res 1991;29:615-8. 11. Lubchenco LO, Hansman C, Boyd E. Intrauterine growth in length and head circumference as estimated from live births at gestational ages from 26 to 46 weeks. Pediatrics 1966;37:403-8. 12. Lucas A, Gore SM, Cole TJ, Bamford MF. Multicenter trial on feeding low birthweight infants: effect of diet on early growth. Arch Dis Child 1984;59:722-30. 13. Atkinson SA, Tsang R. Calcium, magnesium, phosphorus and vitamin D. In: Tsang R, Uauy R, Koletzko B, et al, eds. Nutrition of Preterm Infants. Cincinnati, OH: Digital Educational Publishing; 2005. p. 245-75. 14. Rigo J, De Curtis M, Pieltain C, Salle B, Senterre J. Enteral calcium, phosphate, and vitamin D requirements and bone mineralization in preterm infants. Acta Paediatrica 2007;96:969-74. 15. Rigo J, De Curtis M, Pieltain C, Picaud JC, Salle BL, Senterre J. Bone mineral metabolism in the micropremie. Clin Perinatol 2000; 27:147-70. 16. Backstrom MC, Kouri T, Kuusela AL, Siev€anen H, Koivisto AM, Ikonen RS, et al. Isoenzyme of serum alkaline phosphatase and serum inorganic phosphate in metabolic bone disease of prematurity. Acta Paediatr 2000;89:867-73. 17. Brunton JA, Saigal S, Atkinson SA. Growth and body composition in infants with bronchopulmonary dysplasia up to 3 months corrected age: a randomized trial of a high-energy nutrient-enriched formula fed after hospital discharge. J Pediatr 1998;133:340-5. 18. Fewtrell MS, Adams C, Wilson DC, Cairns P, McClure G, Lucas A. Randomized trial of high nutrient density formula versus standard formula in chronic lung disease. Acta Paediatr 1997;86:577-82. 19. Oyafemi OM, Ellis KJ, Griffin IJ, Abrams SA. Bone mineral status assessment by ultrasound in preterm infants. Int J Body Comp Res 2005;3: 141-5. 20. Fewtrell MS, Loh KL, Chomtho S, Kennedy K, Hawdon J, Khahoo A. Qualitative ultrasound: a useful tool for monitoring bone health in preterm infants? Acta Paediatr 2008;97:1625-30. 21. Rubinacci A, Moro GE, Boehm G, de Terlizzi F, Moro GL, Cadossi R. Quantitative ultrasound for the assessment of osteopenia in preterm infants. Eur J Endocrinol 2003;149:307-15. 22. Pereda L, Ashmeade T, Zaritt J, Carver JD. The use of quantitative ultrasound in assessing bone status in newborn preterm infants. J Perinatol 2003;23:655-9. 23. Liao X, Zhang W, He J, Sun J, Huang P. 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 2008;35:847-53. 24. Nickel KP, Martin BR, Smith DL, Smith JB, Miller GD, Weaver CM. Calcium bioavailability from bovine milk and dairy products in premenopausal women using intrinsic and extrinsic labeling techniques. J Nutr 1996;126:1406-11. 25. Bozzetti V, Tagliabue P. Metabolic bone disease in preterm newborn: an update on nutritional issues. Ital J Pediatr 2009;35:20-8.
Hicks et al
ORIGINAL ARTICLES
June 2011
Table V. Nutrient composition of feeds Nutrients per 100 mL Feed type* and caloric density (kcal/oz)
Energy (kcal)
Calcium (mg)
Phosphorus (mg)
Protein (g)
82 85 90 83 89 93 81 91
141 169 149 110 129 149 146 164
80 95 84 54 61 71 81 91
1.9 2.0 2.1 1.9 2.1 2.4 2.4 2.7
FHM-C 24 FHM-C 25 FHM-CT 27 FHM-H 24 FHM-H 25 FHM-H 26 Formula 24 Formula 27
Figure. Data shown are all subjects with and without early bronchopulmonary dysplasia.
*FHM-C 24 indicates fortified human milk with cow milk–derived fortifier, 4 packs of human milk fortifier added (Similac Human Milk Fortifier); FHM-C 25, fortified human milk with cow milk–derived fortifier, 5 packs of human milk fortifier added (Similac Human Milk Fortifier); FHM-CT, fortified human milk with cow milk–derived fortifier and transitional formula, 4 packs of human milk fortifier added (Similac Human Milk Fortifier) plus powdered transitional formula (Neosure); FHM-H, fortified human milk with human milk–derived fortifier (Prolact +), made up to 24, 25, or 26 kcal/oz. Formula preterm formula made up to 24 or 27 kcal/oz (Similac Special Care).
Table VI. Growth, laboratory measurements, and body composition by feeding type
Weight gain* (g/kg/d) Linear growth* (cm/wk) Head circumference growth* (cm/wk) Alkaline phosphatase activity (IU/L) Phosphorus (mg/dL) SOS U/S at start of study† (m/s) SOS U/S at end of studyz (m/s) Fat mass at end of studyx (%)
Fortified human milk containing human milk–derived fortifier
Fortified human milk containing cow milk–derived fortifier or preterm formula
(n = 15)
(n = 26)
P value
14.8 0.5 1.1 0.1 1.0 0.1 299 34 6.7 0.2 2928 28 2869 28 12.0 1.2
14.3 0.8 1.1 0.1 0.9 0.1 341 16 6.1 0.2 2822 24 2776 31 13.5 0.6
NS NS NS NS NS .007 .04 NS
SOS U/S, speed-of-sound ultrasound. All data are mean SEM unless otherwise indicated. *Gain over study period. †n = 15 Fortified human milk containing human milk–derived fortifier; n = 21 fortified human milk containing cow milk–derived fortifier or preterm formula. zn = 15 Fortified human milk containing human milk–derived fortifier; n = 18 fortified human milk containing cow milk–derived fortifier or preterm formula. xn = 11 Fortified human milk containing human milk–derived fortifier; n = 18 fortified human milk containing cow milk–derived fortifier or preterm formula.
Calcium Absorption in Very Low Birth Weight Infants with and without Bronchopulmonary Dysplasia
890.e1
B
ronchopulmonary dysplasia (BPD) is a common sequela of preterm birth. Affected infants are more likely to have respiratory disorders in later infancy, to be hospitalized in the first 1 to 2 years after birth, and to have abnormal growth and neurodevelopment. The incidence of BPD is typically 20% to 30% in very low birth weight (VLBW) infants.1 Adequate nutrition is a key aspect of care for the infant with BPD.2 Malnutrition can delay somatic growth and the development of new pulmonary alveoli and can decrease muscle strength, making successful weaning from mechanical ventilation less likely. However, it is difficult to provide adequate nutritional support to VLBW infants who have BPD because it is often necessary to limit fluid intake below 150 to 170 mL/kg/d, which is usually needed when standard feeding strategies are implemented. It is common clinical practice to alter nutrition therapy in infants at risk of BPD early in their course, well before 36 weeks’ postmenstrual age (PMA), when the diagnosis of BPD is formally made. The development of severe bone demineralization may exacerbate respiratory and other problems due to weakness of rib strength needed to support lung expansion. In its most severe form, rickets can develop leading to fractures and growth failure.3-5 Rickets in premature infants is almost exclusively a deficiency of the primary bone minerals calcium (Ca) and phosphorus (Phos) rather than vitamin D deficiency. It is often necessary to use high energy density feedings for infants with BPD. Available mineral absorption data using high caloric density feedings are minimal and the solubility of these feedings may decrease nutrient absorption.6 High caloric density infant feedings using either formula or human milk have been studied minimally. The primary objective of our study was to evaluate the effects of early BPD on Ca metabolism and bone mineral status in VLBW infants. The secondary goal was to evaluate growth and body composition in these infants when provided with different feeding regimens used in the neonatal intensive care unit (NICU).
Methods Infants were recruited from the NICU at Texas Children’s Hospital in Houston, Texas. Eligible infants were born between 23 and 32 6/7 weeks’ gestation, 500 to 1500 g, and free of major congenital anomalies. Infants were excluded from the
BPD Ca NICU Phos PMA VLBW
Bronchopulmonary dysplasia Calcium Neonatal intensive care unit Phosphorus Postmenstrual age Very low birth weight
From the United States Department of Agriculture/ Agricultural Research Service (P.H., S.R., K.H., Z.C., S.A.), Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, TX; and the Section of Neonatology (S.R., S.A.), Department of Pediatrics, Baylor College of Medicine and Texas Children’s Hospital, Houston, TX Supported by the Gerber Foundation. The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2011 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2010.12.034
885
THE JOURNAL OF PEDIATRICS
www.jpeds.com
study if they had necrotizing enterocolitis requiring surgical intervention or were unlikely to survive. The study was approved by the Institutional Review Board for Human Subject Research of Baylor College of Medicine and Affiliated Hospitals. Informed written consent was obtained from the parents. Infants were fed either preterm formula or fortified human milk, based on the mother’s decision to provide human milk (Table I). Infants without early BPD were fed according to standard NICU protocol, which provided 150 to 170 mL/ kg/d of a 24 kcal/oz preterm formula (Similac Special Care, Abbott Nutritionals, Columbus, Ohio) or human milk that was fortified with 4 kcal/oz of a commercial human milk fortifier. Infants at risk of early BPD were provided with preterm formula or fortified human milk at volumes of 130 to #150 mL/kg/d and/or at a caloric density >24 kcal/oz.
Vol. 158, No. 6 The decision for fluid restriction and caloric density of the feeding was made by the individual attending physician caring for each infant. Generally, fluid restriction was instituted based on an infant receiving >40% oxygen and on CPAP or mechanical ventilation. The subject was assigned to the early BPD group at the time of the absorption study, based on the need for respiratory support combined with the use of a caloric density >24 kcal/oz. At the discretion of the parents, either a cow milk–based (Similac Human Milk Fortifier, Abbott Nutritionals, Columbus, Ohio) or a human milk–derived (Protact +, Prolacta Bioscience, Monrovia, California) human milk fortifier was used for fortification of human milk (Table I). Fortification was typically begun when human milk feeds reached 100 mL/kg/d.
Table I. Subject classification Classification Feed type† Early BPD PMA at study PNA at study Respiratory support Respiratory support and caloric density ID No Early BPD (wk) (wk) before study* at study* (kcal/oz) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
No early BPD No early BPD No early BPD No early BPD No early BPD Early BPD Early BPD Early BPD No early BPD Early BPD No early BPD Early BPD No early BPD Early BPD No early BPD No early BPD Early BPD No early BPD No early BPD Early BPD Early BPD Early BPD No early BPD Early BPD No early BPD No early BPD Early BPD No early BPD No early BPD No early BPD No early BPD Early BPD Early BPD Early BPD No early BPD No early BPD Early BPD No early BPD No early BPD No early BPD No early BPD
33.3 33.5 33.6 36.0 36.0 33.6 33.6 34.0 31.1 32.0 33.1 34.6 33.0 33.0 33.0 34.1 37.6 32.6 32.0 34.4 34.4 33.0 33.1 32.3 31.5 32.0 30.2 33.5 31.6 34.6 33.3 31.6 33.6 33.6 33.6 33.6 30.0 33.6 33.6 33.0 33.6
3.3 5.5 3.6 6.0 6.0 2.6 2.6 6.0 3.1 3.0 2.1 5.6 3.0 3.0 3.0 6.1 10.6 5.6 5.0 4.4 4.4 7.0 3.1 6.3 3.5 4.0 6.2 4.5 4.6 10.6 2.3 6.6 10.6 10.6 3.6 3.6 4.0 2.6 2.6 4.0 4.6
RA NCPAP NC RA NCPAP NCPAP ETT NCPAP NC NCPAP RA NCPAP NCPAP NCPAP ETT NC NC NC NC RA RA NCPAP RA NCPAP RA RA RA RA NCPAP ETT NCPAP ETT NCPAP ETT NCPAP NCPAP ETT RA RA RA RA
RA NC RA RA RA NCPAP PPV NCPAP RA NC RA NCPAP RA RA RA NC NC RA NC RA RA NC RA NCPAP RA RA ETT RA NCPAP NCPAP RA NCPAP NC NCPAP NC RA ETT RA RA RA RA
FHM-H 24 FHM-H 25 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 26 FHM-H 26 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 24 FHM-H 24 Formula 24 Formula 27 Formula 24 Formula 24 FHM-CT 27 FHM-CT 27 Formula 27 FHM-C 24 Formula 27 FHM-C 24 FHM-C 24 FHM-C 25 Formula 24 FHM-C 24 FHM-C 24 Formula 24 Formula 24 FHM-C 25 FHM-C 24 FHM-C 24 FHM-C 24 FHM-C 24 FHM-C 24 FHM-C 25 FHM-C 25 FHM-C 25
Meds at abs study
BPD diagnosis at 36 wk
No No No No No Caffeine Yes Caffeine No Caffeine Yes Caffeine No No Caffeine No Caffeine Yes No Caffeine Yes Caffeine No Caffeine No Furosemide, Thiozide Yes Caffeine No Caffeine Yes Caffeine No Caffeine No Caffeine, Furosemide Yes Caffeine No Caffeine No Caffeine No Caffeine No Yes No Caffeine No Caffeine Yes Caffeine No Caffeine Yes Caffeine Yes Caffeine Yes Caffeine No Caffeine No Thiozide Yes No No Caffeine No Caffeine No Caffeine Caffeine
PNA, postnatal age. *RA indicates room air; NCPAP, nasal continuous positive airway pressure; NC, nasal cannulae; ETT, endotracheal tube; PPV, positive pressure ventilation. All subjects with NCPAP, NC, ETT, and PPV were provided supplemental oxygen. †FHM-H indicates fortified human milk with human milk–derived fortifier, formula preterm formula; FHM-CT, fortified human milk with cow milk–derived fortifier and transitional formula (Neosure); FHM-C, fortified human milk with cow milk–derived fortifier.
886
Hicks et al
ORIGINAL ARTICLES
June 2011 Absorption Study Approximately 2 weeks after infants had reached full enteral feedings (defined as at least 120 mL/kg/d and having parenteral nutrition discontinued) and were determined to be stable enough for the study, Ca absorption was measured by a dualtracer stable isotope method as described previously.7-10 Briefly, 18 hours before the study, one-half volume of a single feeding was mixed with 44Ca (2 mg) and refrigerated until the study began to allow for equilibration. The isotope was mixed with either fortified human milk or preterm formula. The next morning, a urine collection bag (First-Time Urine Specimen Collector, Hollister Incorporated, Libertyville, Illinois) was attached to the infant. Diapers were weighed before and after use to ensure that all urine was collected. A plastic liner with cotton balls was also placed in the diaper to capture any urine that leaked. The urine from the cotton balls was extracted and added to the urine collection. Urine was extracted from the cotton balls for only three infants. The infants were given 46Ca (0.01 mg) intravenously over a period of 1 minute. The feeding with the premixed oral isotope then was given via orogastric tube, and the remaining feeding volume was given immediately. Urine was collected for 24 hours after the isotope administration. Fractional Ca absorption was calculated from the ratio of the fraction of the oral tracer dose to the intravenous tracer dose recovered in the urine during the 24-hour collection. Total Ca absorbed was calculated by multiplying the Ca intake by the fractional absorption. Approximate net Ca retention was determined using estimates for endogenous fecal excretion derived from previous studies in premature infants and the measured urinary excretion.10 Isotopic enrichment was determined using a magnetic sector Finnigan MAT 261 thermal ionization mass spectrometer (Bremen, Germany). Anthropometrics and Bone Health Markers Daily weight and weekly length and head circumference measures were determined by bedside nursing staff, with continual training for proper anthropometric measures. Nutritionrelated labs, serum Ca, Phos, and alkaline phosphatase activity were recorded when measured for routine care. At study onset and at 6 weeks or discharge, all infants received a speed of sound ultrasound (Sunlight Omnisense 7000p, Sunlight Medical, Israel) of the left pretibial area. Infants received a body composition measurement using air displacement plethysmography (PEA POD, Life Measurement Inc, Concord, California) at 6 weeks or discharge if they were medically stable and not receiving oxygen therapy or ventilatory support. Quality control procedures for the speed of sound ultrasound and air displacement plethysmography machines were performed each day before the first subject was tested. The speed of sound ultrasound machine uses a phantom to measure each probe and each probe measurement consists of three cycles. The difference in the speed of sound measured and speed of sound calculated values is plotted over time. The acceptable difference is between 50 and 50 m/s. For the air
displacement plethysmography machine, five tests were performed: hardware analysis, scale calibration, mass calibration, auto run assessment, and volume assessment. All tests must be successfully passed to perform subject measurements. Statistics The primary outcome was the assessment of Ca absorption. A sample size of n = 30, with at least 15 in each group, was chosen such that a significant difference in intake-adjusted fractional Ca absorption of 10%, from a predicted baseline of 55% 8%, could be identified with an 80% power at a P < .05 (null hypothesis being accepted would be <10% difference) between infants with and without early BPD. Differences in fractional Ca absorption and estimated Ca retention were assessed using analysis of variance with presence or absence of early BPD as a covariate. Growth and biochemical variables were assessed weekly and analyzed by repeated-measures analysis of covariance with feeding type (fortified human milk or preterm formula) and presence or absence of early BPD as independent variables and the baseline measurement as a covariate. Speed-of-sound ultrasound was analyzed by two-way analysis of variance with feeding type (fortified human milk or preterm formula) and feed concentration (24 kcal/oz or 27 kcal/ oz) as the independent variables. Growth was compared with standards expected for infants of these gestational ages. A target weight gain of 15 g/kg/d, length gain of 1 cm/wk, and head circumference growth of 1 cm/wk was used to approximate typical intrauterine growth.11,12 All results are presented as mean SEM unless otherwise indicated.
Results Forty-one premature infants were studied (Table II). Of these, 16 infants were classified as having early BPD. As expected, gestational age and birth weight were significantly different with early BPD infants born earlier and weighing less than infants without early BPD. Postmenstrual age and weight at the time of the Ca absorption study were similar. Feeding volumes were significantly lower in early BPD infants, reflecting fluid restriction, but both groups had adequate energy intakes of at least 120 kcal/kg/d. Calcium intake, fractional Ca absorption, total Ca absorption, urinary Ca excretion, and approximate Ca retention were similar among groups (Table III). Using regression analysis, there was no effect of feeding type, fortified human milk or preterm formula, on Ca absorption (P = .3). There was a significant positive correlation between total Ca absorbed and PMA at the time of study (r = 0.5, P = .001), but there was no significant relationship between fractional Ca absorption and PMA (r = 0.3, P = .06) (Figure; available at www.jpeds.com). There was no significant relationship between Ca intake and fractional absorption (r = 0.12, P = .4). At the time of the study, three subjects were taking diuretics and 31 subjects were treated with caffeine. Mean Ca absorption for the three subjects receiving diuretics was
Calcium Absorption in Very Low Birth Weight Infants with and without Bronchopulmonary Dysplasia
887
THE JOURNAL OF PEDIATRICS
www.jpeds.com
Table II. Characteristics of the study population Early BPD (n = 16) Birth weight (g) 909 66 Gestational age (wk) 27.0 0.5 Postmenstrual age at study (wk) 32.7 0.5 Weight at study (g) 1514 98 Energy intake at study (kcal/kg/d) 125 3 Feeding volume at study (mL/kg/d) 145.0 2.3
No early BPD (n = 25)
P value
1230 38 29.1 0.3 33.2 0.2 1668 48 123 2 153.9 1.5
>.001 .004 NS NS NS .003
All results are mean SEM.
78.5% 0.9%, compared with subjects free of diuretics (51.5% 3.1%). Mean urinary Ca excretion value of the subjects on diuretics was 3.7 1.4 mg/kg/d, compared with the subjects free of diuretics (2.3 0.3 mg/kg/d). Removing the three subjects on diuretics from the overall analyses, Ca absorption, urinary Ca excretion, and approximate Ca retention remain similar among groups. Both Ca absorption and urinary Ca excretion values were similar in the subjects on or free of caffeine. Mean Ca absorption for those receiving caffeine was 56.3% 6.0%, compared with the subjects not receiving caffeine (52.5% 3.6%, P = .6). Mean urinary Ca excretion values for subjects on caffeine was 2.4 0.3 mg/ kg/d compared with the subjects not receiving caffeine (2.5 0.6 mg/kg/d, P = .6). There was no significant difference in weight gain, linear growth, or head circumference gain over the study period between groups (Table IV). There was no significant difference in speed of sound ultrasound measurement between groups over the study period. Eight infants with early BPD and 21 infants without early BPD had air displacement plethysmography measurements. There was no significant difference in percent fat mass between groups. During the study, the average alkaline phosphatase activity levels were significantly greater in infants with early BPD than infants without early BPD. Serum Phos levels were significantly less in infants with early BPD than infants without early BPD infants (Table IV). Effects of Feeding Type During the Ca absorption study, 15 subjects were consuming fortified human milk containing the human milk derived fortifier and 26 subjects were consuming either human milk containing the cow milk–derived fortifier or preterm formula (Table V; available at www.jpeds.com). There was no Table III. Calcium absorption outcomes Early BPD
No early BPD
(n =16)
(n = 25)
P value
200 8 50.3 4.0 104 9 2.41 0.33 91.0 9.8
NS NS NS NS NS
Ca intake (mg/kg/d) 210 12 Fractional Ca absorption (%) 58.4 4.6 Total Ca absorption (mg/kg/d) 127 14 Urinary Ca (mg/kg/d) 2.38 0.45 Approximate Ca retention (mg/kg/d) 99.6 10.0 All results are mean SEM.
888
Vol. 158, No. 6 significant difference in Ca absorption among subjects consuming human milk containing the human milk– derived fortifier (47.8% 4.8%) and subjects consuming human milk containing the cow milk–derived fortifier or preterm formula (56.7% 3.9%, P = .1). Calcium intake was significantly lower in those subjects consuming human milk containing the human milk–derived fortifier (159.0 5.5 mg/kg/d), compared with those subjects consuming human milk containing the cow milk–derived fortifier or preterm formula (229.3 5.6 mg/kg/d, P < .01). Calcium retention was also significantly lower in subjects consuming human milk containing the human milk–derived fortifier (63.4 7.7 mg/kg/d) compared with subjects consuming human milk containing the cow milk–derived fortifier or preterm formula (112.3 8.6 mg/kg/d, P < .01). Weight gain, linear growth, and head circumference growth were similar between subjects consuming human milk containing the human milk–derived fortifier and those consuming human milk containing the cow milk–derived fortifier or preterm formula (Table VI; available at www.jpeds.com). There was a significant difference in speed of sound ultrasound measurement between feeding type groups at the beginning of study and the end of study. There was no significant difference in decrease in speed-of-sound ultrasound measure during the study period. Eleven infants consuming human milk containing the human milk–derived fortifier and 18 infants consuming human milk containing the cow milk–derived fortifier or preterm formula had air
Table IV. Growth, laboratory measurements, and body composition Early BPD
Weight gain* (g/kg/d) Linear growth* (cm/wk) Head circumference growth* (cm/wk) Alkaline phosphatase activity (IU/L) Phosphorus (mg/dL) SOS U/S at start of study† (m/s) Postmenstrual age range at start of study† (wk) Postnatal age range at start of study† (wk) SOS U/S at end of studyz (m/s) Postmenstrual age range at end of studyz (wk) Postnatal age range at end of studyz (wk) Fat mass at end of studyx (%) Postmenstrual age range at end of studyx (wk) Postnatal age range at end of studyx (wk)
No early BPD
(n = 16)
(n = 25)
P value
14.6 0.8 1.0 0.1 1.1 0.1
14.6 0.5 1.2 0.1 0.9 0.1
NS NS NS
390 22
286 19
.001
5.9 0.2 2844 38
6.5 0.2 2881 21
.01 NS
28.4-37.4
30.4-33.6
1.4-11.4
3.3-3.6
2787 40 33.4-40.6
2838 27 34.3-39.6
6.3-3.6
7.1-9.3
13.3 1.2 34.6-39.3
12.8 0.7 34.6-38.0
4.6-15.3
3.6-10.0
NS
NS
SOS U/S, speed-of-sound ultrasound. All data are mean SEM unless otherwise indicated. *Gain over study period. †n = 15 Early BPD, n = 21 no early BPD. zn = 13 Early BPD, n = 20 no early BPD. xn = 8 Early BPD, n = 21 no early BPD.
Hicks et al
ORIGINAL ARTICLES
June 2011 displacement plethysmography measurements. There was no significant difference in percent fat mass between groups (Table VI). During the study, the average alkaline phosphatase activity levels and Phos levels were similar in subjects consuming human milk containing the human milk–derived fortifier and those consuming human milk containing the cow milk–derived fortifier or preterm formula (Table VI). Categorizing infants into three feeding types, 15 subjects were consuming human milk containing the human milk– derived fortifier, 17 subjects were consuming human milk containing the cow milk–derived fortifier, and nine subjects were consuming preterm formula. Using post hoc analysis, there were no significant differences in Ca absorption among the three groups. Mean fractional Ca absorption for the group consuming human milk containing the cow milk–derived fortifier was 59.8% 4.9% and for the group consuming preterm formula was 50.9% 6.4% (P = .2).
Discussion We found that when provided with an adequate nutritional intake of at least 120 kcal/kg/d, infants with and without early BPD achieve both bone and overall growth outcomes close to those achievable in utero. The source of extra protein and minerals, whether preterm formula or human milk fortifier, and type of fortifier did not affect these outcomes as long as energy intake was adequate. Because we found no significant differences in feeding type, for analyses we combined those consuming preterm formula with those consuming human milk containing the cow milk–derived fortifier as they both contain cow milk protein and have been found to have similar bioavailability.7 We found no significant difference in Ca intake, fractional Ca absorption, and Ca retention between the groups with early BPD and without early BPD. Overall Ca absorption rates were high and retention of Ca from both fortified human milk and preterm formula reached the expected in utero accretion rate of at least 90 mg/kg/d.13 Although obtaining intrauterine accretion rates remain the goal in premature infants, it is unclear whether it is an appropriate marker of adequacy.14 Bone remodeling occurs in both term and preterm infants after birth, with increased bone reabsorption and a reduction in bone mineral apparent density.15 Calcium retention between 60 and 90 mg/kg/d appears to allow normal mineralization and decreases the risk of fractures and osteopenia.14 Infants consuming human milk containing the human milk–derived fortifier did not meet these calcium retention levels due to the decreased calcium content of this fortifier. Thus, the statistical nonsignificance of our primary outcome, total Ca absorption, may potentially be the result of the lower intake of some infants. Regardless, the observed similar calcium absorption efficiency across a range of feedings and intakes suggests comparable bioavailability of the calcium sources. However, further balance studies at a range of intakes not achieved in this study would be needed to confirm this.
The biochemical markers of bone mineralization we assessed were serum alkaline phosphatase activity and Phos. The early BPD group did have higher alkaline phosphatase activity and lower Phos values than the group without early BPD; however, none of the values were outside the normal limits seen in growing preterm infants. Although there are no values of alkaline phosphatase activity that indicate osteopenia, these values remains far below the threshold alkaline phosphatase activity of 900 IU/L that Backstr€ om16 concluded gave a 88% sensitivity and 71% specificity to detect osteopenia. There are very few studies investigating nutrition in VLBW infants with BPD. A single study in Canada compared the use of 27 kcal/oz formula with 24 kcal/oz formula in infants with BPD who were nearing hospital discharge and were >37 weeks’ PMA. They found that in infants recovering from BPD, growth failure was not related to malabsorption of nutrients but rather inadequate nutrient intakes.17 In the United Kingdom, Fewtrell18 compared high caloric density formula (30 kcal/oz) with 24 kcal/oz formula in infants with BPD. They found no difference in growth between groups, despite higher caloric intake in the nutrient-dense group. Although both groups received volumes above that used in the United States, the outcomes of our study are in agreement with these data as we found no significant difference in growth parameters between groups. Speed-of-Sound ultrasound is an inexpensive, welltolerated bedside tool with good reproducibility.19 In preterm infants, speed-of-sound ultrasound values decrease after birth. Tomlinson showed a median decrease of 3.2% and Fewtrell showed a decrease of 2.6%.4,20 Our data are consistent with these findings. This may reflect less mechanical stimulation after birth, thus lowering bone mass accrual. The positive correlation between speed-of-sound ultrasound and greater gestational age seen in these infants is in agreement with the findings of others and may be due to a greater transfer of Ca that occurs during later gestational age.21-23 The use of speed-of-sound ultrasound in preterm infants to measure bone status has not been validated against dual-energy x-ray absorptiometry; there are no reference standards to refer to, and it has not been found to predict biochemical indicators of metabolic bone disease. However, it may be useful in monitoring response to dietary and exercise interventions or as a predictor of long-term bone health in this population.21 Body composition measurement is an important indicator of nutritional status. It can identify the composition of weight gain and can be a useful tool to evaluate the efficacy of nutritional interventions. Recent availability of air displacement plethysmography measurements have allowed the study of VLBW infants. Both groups had similar percent fat mass when measured at an average PMA of 35 weeks (range of 31 to 39 weeks) and postnatal age range of 4 to 15 weeks. These infants were assessed at markedly different times in their development, but, without clinical standards for body composition in premature infants, we cannot be certain of the clinical relevance of this.
Calcium Absorption in Very Low Birth Weight Infants with and without Bronchopulmonary Dysplasia
889
THE JOURNAL OF PEDIATRICS
www.jpeds.com
There were a number of limitations in our study. We used an extrinsic tracer, as opposed to an intrinsic or biological tracer, in the products used to assess Ca absorption, which assumes a complete exchange has taken place between the tracer and the native calcium. It is unlikely that any significant proportion of the fortified human milk or preterm formula did not equilibrate with the tracer, as this approach has been widely used and validated for milk-based products.24 We also used endogenous fecal excretion rates that were previously determined from a premature infant study; thus, we were only able to report approximate Ca retention. Another potential limitation was that vitamin D status was not assessed in our study. These data support the use of high caloric density feedings in VLBW infants at risk of BPD to achieve both bone and overall growth outcomes close to those achievable in utero. Growth failure and bone mineral deficiencies are not due to inadequate utilization of nutrients but may be due to inadequate intake and early nutritional intervention can reduce the prevalence and severity of these.25 Our results may not be applicable in infants with longstanding BPD, although the feeding of preterm formula is uncommonly continued in our institution after infants are discharged from the hospital or are over about 3 kg. Long-term data are necessary to determine if physiological outcomes are dependent on achieving intrauterine Ca accretion. n Submitted for publication Mar 16, 2010; last revision received Nov 16, 2010; accepted Dec 22, 2010. Reprint requests: Dr Penni D. Hicks, PhD, RD, Children’s Nutrition Research Center at Baylor College of Medicine, 1100 Bates Street, Room 7005, Houston, TX 77030. E-mail: [email protected]
References 1. Van Marter LJ. Epidemiology of bronchopulmonary dysplasia. Semin Fetal Neonatal Med 2009;14:358-66. 2. Bancalari E, Wilson-Costello D, Iben SC. Management of infants with bronchopulmonary dysplasia in North America. Early Hum Dev 2005; 81:171-9. 3. Fewtrell MS, Cole TJ, Bishop NJ, Lucas A. Neonatal factors predicting childhood height in preterm infants: evidence for a persisting effect of early metabolic bone disease? J Pediatr 2000;137:668-73. 4. Tomlinson C, McDevitt H, Ahmed SF, White MP. Longitudinal changes in bone health as assessed by the speed of sound in very low birth weight preterm infants. J Pediatr 2006;148:450-5. 5. Mitchell SM, Rogers SP, Hicks PD, Hawthorne KM, Parker BR, Abrams SA. High frequencies of elevated alkaline phosphatase activity and rickets exist in extremely low birth weight infants despite current nutritional support. BMC Pediatr 2009;9:47. 6. Goss S, Prushko J, Bogner R. Factors affecting calcium precipitation during neutralisation in a simulated intestinal environment. J Pharm Sci 2010;99:4183-91.
890
Vol. 158, No. 6 7. Abrams SA, Schanler RJ, Yergey AL, Vieira NE, Bronner F. Compartmental analysis of calcium metabolism in very low birth infants. Pediatr Res 1994;36:424-8. 8. Yergey AL, Abrams SA, Vieira NE, Eastell R, Hillman LS, Covell DG. Recent studies of human calcium metabolism using stable isotopic tracers. Can J Physiol Pharmacol 1990;68:973-6. 9. Abrams SA. Pubertal changes in kinetics in girls assessed using 42Ca. Pediatr Res 1993;34:455-9. 10. Abrams SA, Esteban NV, Vieira NE, Yergey AL. Dual tracer stable isotopic assessment of calcium absorption and endogenous fecal excretion in low birth weight infants. Pediatr Res 1991;29:615-8. 11. Lubchenco LO, Hansman C, Boyd E. Intrauterine growth in length and head circumference as estimated from live births at gestational ages from 26 to 46 weeks. Pediatrics 1966;37:403-8. 12. Lucas A, Gore SM, Cole TJ, Bamford MF. Multicenter trial on feeding low birthweight infants: effect of diet on early growth. Arch Dis Child 1984;59:722-30. 13. Atkinson SA, Tsang R. Calcium, magnesium, phosphorus and vitamin D. In: Tsang R, Uauy R, Koletzko B, et al, eds. Nutrition of Preterm Infants. Cincinnati, OH: Digital Educational Publishing; 2005. p. 245-75. 14. Rigo J, De Curtis M, Pieltain C, Salle B, Senterre J. Enteral calcium, phosphate, and vitamin D requirements and bone mineralization in preterm infants. Acta Paediatrica 2007;96:969-74. 15. Rigo J, De Curtis M, Pieltain C, Picaud JC, Salle BL, Senterre J. Bone mineral metabolism in the micropremie. Clin Perinatol 2000; 27:147-70. 16. Backstrom MC, Kouri T, Kuusela AL, Siev€anen H, Koivisto AM, Ikonen RS, et al. Isoenzyme of serum alkaline phosphatase and serum inorganic phosphate in metabolic bone disease of prematurity. Acta Paediatr 2000;89:867-73. 17. Brunton JA, Saigal S, Atkinson SA. Growth and body composition in infants with bronchopulmonary dysplasia up to 3 months corrected age: a randomized trial of a high-energy nutrient-enriched formula fed after hospital discharge. J Pediatr 1998;133:340-5. 18. Fewtrell MS, Adams C, Wilson DC, Cairns P, McClure G, Lucas A. Randomized trial of high nutrient density formula versus standard formula in chronic lung disease. Acta Paediatr 1997;86:577-82. 19. Oyafemi OM, Ellis KJ, Griffin IJ, Abrams SA. Bone mineral status assessment by ultrasound in preterm infants. Int J Body Comp Res 2005;3: 141-5. 20. Fewtrell MS, Loh KL, Chomtho S, Kennedy K, Hawdon J, Khahoo A. Qualitative ultrasound: a useful tool for monitoring bone health in preterm infants? Acta Paediatr 2008;97:1625-30. 21. Rubinacci A, Moro GE, Boehm G, de Terlizzi F, Moro GL, Cadossi R. Quantitative ultrasound for the assessment of osteopenia in preterm infants. Eur J Endocrinol 2003;149:307-15. 22. Pereda L, Ashmeade T, Zaritt J, Carver JD. The use of quantitative ultrasound in assessing bone status in newborn preterm infants. J Perinatol 2003;23:655-9. 23. Liao X, Zhang W, He J, Sun J, Huang P. 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 2008;35:847-53. 24. Nickel KP, Martin BR, Smith DL, Smith JB, Miller GD, Weaver CM. Calcium bioavailability from bovine milk and dairy products in premenopausal women using intrinsic and extrinsic labeling techniques. J Nutr 1996;126:1406-11. 25. Bozzetti V, Tagliabue P. Metabolic bone disease in preterm newborn: an update on nutritional issues. Ital J Pediatr 2009;35:20-8.
Hicks et al
ORIGINAL ARTICLES
June 2011
Table V. Nutrient composition of feeds Nutrients per 100 mL Feed type* and caloric density (kcal/oz)
Energy (kcal)
Calcium (mg)
Phosphorus (mg)
Protein (g)
82 85 90 83 89 93 81 91
141 169 149 110 129 149 146 164
80 95 84 54 61 71 81 91
1.9 2.0 2.1 1.9 2.1 2.4 2.4 2.7
FHM-C 24 FHM-C 25 FHM-CT 27 FHM-H 24 FHM-H 25 FHM-H 26 Formula 24 Formula 27
Figure. Data shown are all subjects with and without early bronchopulmonary dysplasia.
*FHM-C 24 indicates fortified human milk with cow milk–derived fortifier, 4 packs of human milk fortifier added (Similac Human Milk Fortifier); FHM-C 25, fortified human milk with cow milk–derived fortifier, 5 packs of human milk fortifier added (Similac Human Milk Fortifier); FHM-CT, fortified human milk with cow milk–derived fortifier and transitional formula, 4 packs of human milk fortifier added (Similac Human Milk Fortifier) plus powdered transitional formula (Neosure); FHM-H, fortified human milk with human milk–derived fortifier (Prolact +), made up to 24, 25, or 26 kcal/oz. Formula preterm formula made up to 24 or 27 kcal/oz (Similac Special Care).
Table VI. Growth, laboratory measurements, and body composition by feeding type
Weight gain* (g/kg/d) Linear growth* (cm/wk) Head circumference growth* (cm/wk) Alkaline phosphatase activity (IU/L) Phosphorus (mg/dL) SOS U/S at start of study† (m/s) SOS U/S at end of studyz (m/s) Fat mass at end of studyx (%)
Fortified human milk containing human milk–derived fortifier
Fortified human milk containing cow milk–derived fortifier or preterm formula
(n = 15)
(n = 26)
P value
14.8 0.5 1.1 0.1 1.0 0.1 299 34 6.7 0.2 2928 28 2869 28 12.0 1.2
14.3 0.8 1.1 0.1 0.9 0.1 341 16 6.1 0.2 2822 24 2776 31 13.5 0.6
NS NS NS NS NS .007 .04 NS
SOS U/S, speed-of-sound ultrasound. All data are mean SEM unless otherwise indicated. *Gain over study period. †n = 15 Fortified human milk containing human milk–derived fortifier; n = 21 fortified human milk containing cow milk–derived fortifier or preterm formula. zn = 15 Fortified human milk containing human milk–derived fortifier; n = 18 fortified human milk containing cow milk–derived fortifier or preterm formula. xn = 11 Fortified human milk containing human milk–derived fortifier; n = 18 fortified human milk containing cow milk–derived fortifier or preterm formula.
Calcium Absorption in Very Low Birth Weight Infants with and without Bronchopulmonary Dysplasia
890.e1