Plasma retinol-binding protein response to vitamin A administration in infants susceptible to bronchopulmonary dysplasia

Plasma retinol-binding protein response to vitamin A administration in infants susceptible to bronchopulmonary dysplasia J a y a n t P. Shenai, MD, M a r g a r e t G. Rush, MD, Mildred T. Stahlman, MD, a n d Frank Chytil, PhD From the Departments of Pediatrics and Biochemistry, Vanderbilt University School of Medicine, Nashville, Tennessee We hypothesized that c h a n g e s in plasma retinol-binding protein (RBP) c o n c e n tration in response to vitamin A administration might be useful for evaluating vitamin A status of very low birth weight infants susceptible to bronchopulmonary dysplasia. We prospectively studied 24 consecutively admitted neonates (birth weight <1350 gm, gestational a g e <31 weeks, ventilator d e p e n d e n t for >24 hours after birth), who were eligible to receive 2000 IU supplemental vitamin A by intramuscular injection on postnatal day I and on alternate days thereafter for 28 days. In addition to serial assessment of vitamin A status, we measured plasma RBP just before and I, 3, and 6 hours after an intramuscular injection of vitamin A (2000 IU/kg retinyl palmitate) on days I and 28. The percent increase in plasma RBP (A-RBP) was high (mean _+ SD: 61 • 37%) and plasma vitamin A and RBP values were low on d a y I, indicative of vitamin A deficiency. Supplemental vitamin A improved vitamin A status of all infants as shown by low A-RBP (mean _+ SD: 8 • 9%) and normal plasma vitamin A and RBP values on day 28. Bronchopulmonary dysplasia was d i a g n o s e d in 12 of 24 infants. Infants with bronchopulmonary dysplasia h a d a higher mean ( • SD) A-RBP on day 28 than those without bronchopulmonary dysplasia (13 • 10% vs 3 • 3%, p <0.01), indicative of persistence of low vitamin A status in infants with lung disease despite supplementation. We c o n c l u d e that the plasma RBP response to vitamin A is a useful indicator of vitamin A status in very low birth weight infants. Although vitamin A supplementation at the d o s a g e used in this study normalizes conventional plasma indexes of vitamin A in very low birth weight infants, the plasma RBP response to vitamin A may continue to reflect persistence of low vitamin A status in the more immature infants with significant lung disease. We suggest that the plasma RBP response to vitamin A may be a useful functional test in such infants. (J PEDIATR1990;116:607-14)

Supported by research grants HL 14214 and ttD09195 from the National Institutes of Health. Presented in part at the Annual Meeting of the Society for Pediatric Research, Washington, D.C., May 3, 1989. Submitted for publication Aug. 2, i989; accepted Oct. 17, 1989. Reprint requests: Jayant P. Shenai, MD, Department of Pediatrics, S-4311 MCN, Vanderbilt University Medical Center, Nashville, TN 37232-2370. 9/23/17485

Vitamin A (retinol) is transported in plasma bound to a specific carrier protein, retinol-binding protein. 1 The human RBP molecule consists of a single polypeptide chain with a molecular weight of 21,000 and a single binding site for one molecule of retinol. 24 RBP is synthesized in the liver "and secreted into plasma as the retinol:RBP complex. 4 In vitamin A-deficient animals, RBP secretion is blocked, leading to its accumulation in the liver and resulting in high

607

608

Shenai et al.

The Journal of Pediatrics April 1990

Table I. Initial clinical characteristics of study subjects (N = 24) Characteristics

Values"

Gestational age (wk) 27.4 _+ 1.9 (25.0-30.0) Birth weight (gm) 949 _+ 215 (640-1320) Head circumference (cm) 24.8 _+ 2.1 (21.5-27.5) Crown-heel length (cm) 35.9 _+ 3.0 (31.0-41.0) Male/female 10/14 White/black 17/7 Singleton/multiple birth 20/4 Inborn/outborn 14/10 Cause of respiratory insufficiency Hyaline membrane disease 17 Pulmonary edema 6 Pneumonia 1 *Plus-minusvaluesare mean + SD (range);othervaluesare absolutenumbers of patients.

BPD Fio2 RBP VLBW

Bronchopulmonarydysplasia Fraction of inspired oxygen Retino[-binding protein Very low birth weight

liver and low plasma concentrations of RBP. 5' 6 When these animals are given vitamin A, mobilization of RBP from the liver occurs, which leads to a rapid decrease in the liver RBP concentration and a concomitant increase in the plasma RBP concentration.5 In vitamin A-sufficient animals, on the other hand, RBP does not accumulate in the liver, and minimal or no change in plasma RBP concentration is seen after vitamin A administration.5 Very low birth weight infants are often born with limited liver reserves of vitamin A 7 and with low plasma concentrations of vitamin A and RBP. 81~ Bronchopulmonary dysplasia occurs commonly in these infants when they are subjected to acute, subacute, and chronic lung injury. 11 Most VLBW infants with development of BPD manifest clinical, biochemical, and histopathologic evidence of vitamin A deficiency.12' 13 Vitamin A supplementation from early postnatal life not only improves their vitamin A status but also appears to promote regenerative healing from lung injury, as evidenced by decreased BPD incidence and morbidity. 14 These observations have led to the implementation in our neonatal intensive care unit of routine vitamin A supplementation in VLBW infants susceptible to BPD. We hypothesized that changes in plasma RBP concentration in response to vitamin A administration might be useful for evaluating the vitamin A status of VLBW infants. We measured plasma RBP concentrations in sequentially obtained blood samples from a group of VLBW infants just before, and then during the first 6 hours after, a single dose

of supplemental vitamin A. We studied this plasma RBP response to vitamin A administration shortly after birth and again after a 28-day period of vitamin A supplementation. On the basis of these studies, we developed a test that may be useful for monitoring the vitamin A status of VLBW infants susceptible to BPD.

METHODS Subjects. Twenty-lour infants born between November 1987 and April 1988 and hospitalized in the neonatal intensive care unit at Vanderbilt Medical Center were the subjects of this study (Table I). These infants were selected prospectively by using criteria similar to those proposed by Bancalari et al. 15 to identify infants at risk for BPD. These criteria included a birth weight < 1350 gm, gestational age <31 weeks, and need for supplemental oxygen (Fio2 >0.3) and mechanical ventilation for at least 24 hours after birth. All infants were appropriate in size for gestational age. Of the 24 infants, 10 were transferred from other hospitals; all outborn infants were younger than 6 hours of age at the time of admission to the unit. Hyaline membrane disease, diagnosed by standard clinical and radiographic criteria, was the predominant cause of respiratory insufficiency in these infants. No infant had major co~genital anomalies. General procedures. Informed consent was obtained from one or both parents of each infant. The study was approved by the Vanderbilt University Committee for the Protection of Human Subjects. The neonatal intensive care unit staff was responsible for all decisions related to clinical management of the infants. The general management of infants with respiratory problems in this unit has been previously described] 6 A single type of mechanical ventilator (Bear Cub, Bear Medical Systems, Inc., Riverside, Calif.) was used throughout the study period. Nutritional procedures. All infants were sustained with intravenously administered gh~cose with added electrolytes, beginning at birth; a protein-dextrose solution was added within the first 4 days after birth, and a fat emalsion within the first 10 days. The vitamins were provided by an aqueous multivitamin preparation (M.V.I. Pediatric, from Armour Pharmaceutical Co., Kankakee, Ill.) added to the protein-dextrose solution. The vitamin A concentration of this parenteral solution was estimated to be 920 IU/dl. Enteral feedings were begun and advanced in all infants as tolerated. Continuous transpyloric infusion was the preferred mode of feeding for ventilator-dependent infants. Intermittent orogastric garage was the choice for non-ventilator-dependent infants until they could be fed orally, Infants were fed either human milk or infant formula. The vitamin A concentrations of these milk samples were estimated to range from 240 to 550 IU/dl. Approximately 1.0 ml of an oral multivitamin preparation (Poly-Vi-Sol, from

Volume 116 Number 4

Mead Johnson & Co., Evansville, Ind.) containing 1500 IU vitamin A was added to the diet after the enteral feeding was established. Enteral feeding Was considered established when the energy intake by the enteral route exceeded 75% of the total energy intake. In addition, all infants received, according to nursery protocol, 2000 IU supplemental vitamin A by intramuscular injection on postnatal day 1 and on alternate days thereafter, a total of 14 injections for 28 days or until the establishment of enteral feeding. A water-miscible preparation (Aquasol A Parenteral, from Armour) containing 50,000 IU/ml vitamin A as retinyl palmitate was used. The unit dose (0.04 ml) of this preparation containing 2000 IU vitamin A was freshly prepared in the pharmacy and administered within 30 minutes to the infant by the nursing staff. The solution was shielded from light at all times. Vitamin A status. Vitamin A intake of each infant was assessed daily. The cumulative vitamin A intake by all routes (intravenous, intramuscular, enteral) in a given 1-week period was calculated to determine the average daily intake of vitamin A (in international units per kilogram per day) during that week. Similarly, the average daily intake of other nutrients, namely, protein, carbohydrate, and fat, and average daily energy intake were calculated for all infants. Approximately 0.7 ml of blood was drawn from each infant by venous puncture on postnatal day 1 before the administration of vitamin A and at weekly intervals thereafter throughout the period of vitamin A supplementation. Blood samples were drawn into heparinized tubes. Plasma was separated from each sample by centrifugation and stored at - 2 0 ~ C until analysis. Plasma vitamin A concentration was determined in duplicate by the fluorometric method described by Thompson et al. Iv Plasma RBP concentration was determined in duplicate by quantitative radial immunodiffusion (LC-Partigen immunodiffusion plate plasma RBP, from Behring Diagnostics, Inc., Somerville, N.J.). Plasma retinol/RBP molar ratio was calculated from the values of plasma concentrations, assuming that the molecular weights of retinol and RBP were 286 and 21,000, respectively. 18 Plasma R B P response to vitamin A administration. In addition to the serial assessment of vitamin A status, we studied the plasma RBP response to vitamin A administration in each infant on postnatal day 1 before initiating vitamin A supplementation and again on postnatal day 28 after the period of vitamin A supplementation. We measured plasma RBP concentrations in sequentially obtained blood samples just before (baseline) and 1, 3, and 6. hours after an intramuscular injection of vitamin A (2000 !U/kg retinyl palmitate). Each sample consisted of approximately 80 ~1 blood drawn by puncture of a warmed heel into two

Plasma R B P response to vitamin A

609

50" 40"

VITAMIN A (.ug/dL)

30' 20" I0O3,0-]

RBP (mg/dL)

1,2" hO'

RETINOL:RBP MOLARRATIO o,8 0,6" 0 7

14

21

28

POSTNATAl_ AGE (days)

Fig. t. Plasma concentrations of vitamin A and RBP, with plasma retinol/RBP molar ratios in infants. Mean (_+SEM) values on postnatal day t reflect values before vitamin A supplementation. All mean values marked with asterisk on postnatal days 7, 14, 21, and 28 were significantlyhigher than those on postnatal day 1 (p <0.001).

heparinized microhematocrit capillary tubes (Fisher Scientific Co., Pittsburgh, Pa.). Plasma was separated from each sample by centrifugation. Plasma RBP concentration was determined in duplicate by enzyme-linked immunosorbent assay19; the coefficient of variation for intraassay variabiIity is reported to range between 4% and 7%. 19This microassay technique was used to minimize blood loss associated with frequent sampling. The percent increase in the plasma RBP concentration (A-RBP) from its baseline value was calculated by the following equation: A-RBP (%) = [RBP ( m a x i m u m ) - R B P (baseline)]/RBP (baseline) • 100. The lowest A-RBP value assigned was zero if there was no change or if there was a decrease in the plasma RBP concentration from the baseline value. The dose of vitamin A used in this test was based on our previous observations related to vitamin A supplementation in VLBW infants]4 The timing of blood samples was based on kinetic studies of RBP secretion in human adults2~ and in rats. 6 The timing of blood samples was confirmed after preliminary kinetic studies in six additional VLBW infants were performed

0 10

Shenai et al.

The Journal of Pediatrics April ]990

70

~-r MEAN*- SEM

60

1

*p < 0.00~

~

I

"

50" /-Day (%)

Vitamin A zo- Injection ~

/

~-~

~

.L

1

~

Day 2 .

0

9 . . . . .

- . . . .

' - - - -

- . . . . . .

TIME

(hours)

.

.

.

.

Fig. 2. Plasma RBP response to vitamin A administration in infants. A-RBP was calculated as percentage of increase in plasma RBP concentration from baseline value at 1, 3, and 6 hours after intramuscular injection of vitamin A (2000 IU/kg retinyl palmirate). All mean (• SEM) values marked with asterisk, reflecting plasma RBP response on day I, were significantly higher than those on day 28 (p <0.001).

with sampling times up to 24 hours after vitamin A injection. Clinical surveillance. The ventilatory status of each infant was determined on postnatal days 1, 7, 14, 21, and 28. The following variables were examined: Fio2, ventilator rate, peak inspiratory Pressure, positive end-expiratory pressure, mean airway pressure, and oxygenation index (Fio2 x Mean airway pressure/Pao2 x 100). Multiple readings of each of these variables obtained in a given 24-hour period were examined, and the mean value was calculated for each variable. The clinical course of each infant was reviewed on postnatal day 28 for the occurrence of BPD. The diagnosis of BPD was based on clinical criteria described by Bancalari et aI., ~5 including the need for supplemental oxygen (Fi02 >0.3) on postnatal day 28, and on chest radiographic criteria described by Edwards. ;I The clinical course of each infant was also reviewed for the occurrence of pulmonary air dissection, symptomatic patent ductus arteriosus, intraventricular hemorrhage, airway infection, and retinopathy of prematurity. Standard clinical and radiographic criteria were used for diagnosis of pulmonary air dissection. The protocol for diagnosis and management of symptomatic patent ductus arteriosus in our unit has been previously described. 22 Intraventricular hemorrhage was diagnosed and graded by serial ultrasonography. 23 Airway infection was diagnosed by clinical and radiographic criteria, with leukocyte and platelet counts, and confirmed by positive serial microbiologic cultures of airway secretions. Retinopathy of prematurity was diagnosed and graded 24 according to the highest stage of retinopathy observed in each ififant during serial examinations conducted until the retina was

completely vascularized at approximately 40 weeks of postconceptional age. Statistical methods. Statistical analysis was performed by using the Statistical Package for the Social Sciences computer system (SSPS Inc., Chicago, Ill.). Pearson chisquare analysis was used for comparing discrete variables. Wilcoxon nonparametric rank sum test was used to compute p values for comparison of means between continuous variables. All variables involving repeated measurements were compared by analysis of covariance. A l l p values were based on two-tailed tests; p <0.05 was assumed to be significant. RESULTS Of 75 VLBW (<1350 gin) infants consecutively admitted to the unit during the 6-month enrollment period, 44 were eligible for inclusion in the study on the basis of their need for supplemental oxygen (Fio2 >0.3) and mechanical ventilation for at least 24 hours after birth. Twenty of these infants were excluded for the following reasons: major congenital anomalies (8 infants), nonviability because of extreme prematurity (7), postnatal age at admission to the unit >24 hours (4), and fulminant sepsis (1). The remaining 24 infants were enrolled in the study. Every infant completed the study l~rotocol. Pulmonary interstitial emphysema complicated the initial clinical course in three infants. Symptomatic patent ductus arteriosus was diagnosed in five infants; four received medical management, including one to four intravenous doses of indomethacin (0.2 mg/kg per dose), and one underwent surgical ligation of the'ductus on postnatal day 5. lntraventricular hemorrhage was diagnosed in eight infants (grade I / I I / I I I / I V : 6 / 1 / O / 1 , respectively); the postnatal age at the time of diagnosis ranged from 1 to 4 days. Vitamin A status. The average daily intake of vitamin A in infants ranged from 1602 to 2591 I U / k g per day (mean _+ SD: 199I _+ 296"IU/kg per day) during the period of vitamin A supplementation. Their mean plasma concentrations of vitamin A were significantly higher on postnatal days 7, 14, 21, and 28 (during the period of supplementation) than on postnatal day 1 (before supplementation) (p <0.001) (Fig. 1). The dosage of vitamin A used in this study raised the plasma concentrations of vitamin A in all infants to levels within the range of 20.0 to 80.0 ~g/dl (0.70 to 2.80 #tool/L) seen in healthy children and adultsfl 5 The plasma vitamin A concentration did not exceed 80.0 ~g/dl (2.80 #tool/L) in any infant. A significant increase in the mean plasma RBP concentration from its initial low value to levels within the range seen in healthy children 26 was seen in the first week of vitamin A supplementation (p <0.001), and this increase was sustained throughout the period of supplementation (Fig. 1). Likewise, a significant increase in the mean plasma retinol/RBP molar ratio was

Volume 116 Number 4

Plasma R B P response to vitamin A

611

T a b l e II. C l i n i c a l o u t c o m e in s t u d y s u b j e c t s Characteristics

Gestational age (wk) Birth weight (gin) Head circumference (cm) Crown-heel length (cm) Ventilatory requirements on day 1 Fio2 Ventilator rate (bpm) PIP (cm H=O) PEEP (cm H20) MAP (cm H20) Oxygenation index (%)? Ventilatory requirements on day 28 Fio2 Ventilator rate (bpm) PIP (cm H20) PEEP (cm H20) MAP (cm H20) Oxygenation index (%)? Airway infection Retinopathy of prematurity Stage I / I I / I I I / I V / V

BPD (n = 12)

N o BPD (n = 12)

p

26.4 849 23.7 34.2

_+ 1.7" +_ 173 _+ 1.6 _+ 2.2

28.3 1048 25.9 37.7

_+ 1.7" _+ 212 _+ 1.9 _+ 2.6

<0.02 <0.02 <0.005 <0.002

0.51 32 22 4.7 8.1 6.1

+ 0.18 _+ 12 _+ 4 _+ 0.6 _+ 1.3 _+ 3.5

0.44 15 t2 4.4 5.9 4.3

_+ 0.22 _+ 18 _+ 11 _+ 1.9 + 3.1 _+ 3.5

NS <0.02 <0.02 NS <0.05 NS

0.70 _+ 0.21 22 _+ 1 i 22 _+ 7 4.3 _+ 1.5 7.3 _+ 2.8 10.9 _+ 7.2 8/12 6/12 0/1/2/1/2

0.26 _+ 0.06 2 -+ 5 3 _+ 6 0.8 _+ 1.7 0.9 _+ 2.1 0.4 _+ 1.0 1/12 2/12 0/2/0/0/0

<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0005 <0.001 <0.001 <0.001

NS, Not significant; bpm, breaths per minute; PIP, peak inspiratory pressure; PEEP,positiveend-expiratory pressure; MAP. mean airway pressure. *Plus-minus values are mean _+SD; other v~tluesare absolute numbers of patients. tOxygenation index = Floe • mean airway pressure/Pao2 • 100.

seen from the initial low value of 0.6 to values ranging from 0.9 to 1.1 during the period of vitamin A supplementation (p <0.001) (Fig. 1). In vitamin A-sufficient persons, RBP is saturated with vitamin A in the plasma and consequently the retinol/RBP molar ratio is approximately 1.0. 27 Plasma R B P response to vitamin A administration. In tests conducted on postnatal day l, before the initiation of vitamin A supplementation, a substantial increase in the plasma RBP concentration from its baseline value was seen in response to vitamin A injection in all infants. Among the time intervals studied, the maximum plasma RBP concentration was seen at 6 hours after injection in 20 of 24 infants and at 3 hours after injection in the remainder. The A-RBP ranged from 24% to 192% (mean _+ SD: 61 + 37%) (Fig. 2). There was a significant negative correlation between the A-RBP values and the plasma concentrations of vitamin A on postnatal day 1 (r = -0.65; p <0.001). There was no significant correlation between the A-RBP values and the plasma concentrations of RBP on postnatal day 1 (r = -0.19; p = 0.38). In tests conducted on postnatal day 28, after the period of vitamin A supplementation, a significant increase in the plasma R B P concentration from its baseline value was seen in respOnse to vitamin A injection in only 9 of 24 infants. Among these infants, the maximum plasma RBP concentration was seen at 6 hours after injection in five and at 3 hours after injection in the remainder. The A-RBP ranged from 0% to 33% (mean _+ SD: 8 _+ 9%)

(Fig. 2). The mean ( + SD) A-RBP on postnatal day 28 was significantly lower than that on postnatal day 1 (8 _+ 9% vs 61 +_ 37%, respectively) (p <0.001). As'o n postnatal day 1, there was a significant negative correlation between the ARBP values and the plasma concentrations of vitamin A on postnatal day 28 (r = -0.61; p <0.001). Likewise, as on postnatal day 1, there was no significant correlation between the A-RBP values and the plasma concentrations of R B P on postnatal day 28 (r -- -0.29; p = 0.25). Clinical outcome. We diagnosed BPD on postnatal day 28 in 12 of 24 infants, using the criteria described. Infants in whom BPD developed were more premature than those without BPD (Table II). The mean ventilatory requirements of infants with BPD were significantly higher than those of infants without BPD throughout the study period for most of the variables analyzed. Episodes of airway infection, confirmed by positive serial microbiologic cultures of airway secretions, occurred more frequently in infants with BPD than in those without BPD. The isolated microorganisms included Staphylococcus epidermidis, Staphylococcus aureus, Enterobacter cloacae, Klebsiella pneumoniae, Pseudomonas aeruginosa, and Escherichia coli. Retinopathy of prematurity was diagnosed in 6 of 12 infants with BPD versus 2 of 12 infants without BPD (p <0.001) (Table II). The postnatal age at the time of diagnosis of retinopathy ranged from 41 to 108 days (mean _+ SD: 64 _+ 22 days), infants with BPD had more advanced

6 12

Shenai et al.

The Journal of Pediatrics April 1990

Table Ill. V i t a m i n A status of study subjects Characteristics

Vitamin A intake (IU/kg per day) Total Intravenous Intramuscular Enteral Plasma concentrations on day 1 Vitamin A (#g/dl) [#tool/L] RBP (mg/dl) [#tool/L] Retinol/RBP ratio Plasma concentrations on day 28 Vitamin A (#g/dl) [~mol/L] RBP (mg/dl) [~mol/L] Retinol/RBP ratio ~-RBP (%) On day 1 On day 28

BPD (n = 12)

1999 • 515 + 1063 • 421 • 16.5 [0.58 1.8 [0.86 0.6

270* 203 320 461

No BPD (n = 12)

p

1984 _+ 383* 312 _+ 139 743 • 354 929 _+ 336

NS <0.01 <0.05 <0.01

_+ 6.9 • 0.24] _+ 0.6 _+ 0.29] • 0.2

16.4 [0.57 1.9 [0.91 0_6

_+ 6.4 _+ 0.22] _+ 0.5 _+ 0.24] _+ 0.2

37.8 _+ 14.7 [1.32 _+ 0.51] 2.6 _+ 0.8 [1.25 • 0.38] 1.0 • 0.3

45.8 [1.60 2.6 [1.25 1.2

_+ 16.3 _+ 0.57] • 0.4 _+ 0.19] _+ 0.3

70 __+46 13 • 10

51 • 24 3 _+ 3

NS NS NS NS NS NS NS <0.01

NS, Not significant. *Plus-minus values are mean _+SD.

retinopathy in comparison with those who did not have BPD. Although the total vitamin A intake during the study period did not vary between the infant groups, the route of its administration was different (Table IIl). Infants with BPD received a significantly higher percentage of total vitamin A intake by the parenteral (intravenous plus intramuscular) route and a significantly lower percentage by the enteral route relative to infants without BPD (parenteral 79% vs 53%, enteral 21% vs 47%, respectively) (p <0.0l). The average intakes of protein, carbohydrate, fat, and energy were similar in the two groups throughout the study period. There were no significant differences between the groups with respect to the mean plasma concentrations of vitamin A and RBP and the mean plasma retinol/RBP molar ratios throughout the study period (Table llI). The mean A-RBP value on postnata~ day 1 in infants with BPD did not differ significantly from that in infants without BPD (p = 0.21) (Table lII). However, the mean A-RBP value on postnatal day 28 was significantly higher in infants with BPD than in infants without BPD (p <0.01). The individual A-RBP values on postnatal day 28 exceeded 8% in 9 of 12 infants with BPD, compared with none from the group without BPD. DISCUSSION The plasma RBP response to vitamin A administration in these VLBW infants studied shortly after birth, before the initiation of vitamin A supplementation, was characterized

by high 2x-RBP values. This response is similar to that resulting from release of liver RBP after vitamin A administration in patients with vitamin A deficiency. 5 This observation, together with the low plasma concentrations of vitamin A and RBP and the low plasma retinol/RBP molar ratios on postnatal day 1, shows that these infants were vitamin A deficient at the beginning of the study. It is possible that these neonates were vitamin A deficient because of deprivation of transplacental vitamin A supply resulting from their delivery at an early gestational age. The plasma RBP response to vitamin A administration after the 28-day period of vitamin A supplementation, on the other hand, was characterized by lower A-RBP values. This observation, together with the normal plasma concentrations of vitamin A and RBP and the normal plasma retinol/RBP molar ratios seen during the period of supplementation, shows that the dosage of vitamin A used in this study did improve the vitamin A status of all these infants. The decrease in the /.X-RBP values concurrent with improving vitamin A status, as well as a significant negative correlation between the A-RBP values and the plasma concentrations of vitamin A, suggests that the plasma RBP response to vitamin A administration is a useful indicator of vitamin A status in VLBW infants. A response characterized by high A-RBP values is compatible with a vitamin A-deficient status, whereas a response characterized by low A-RBP values reflects improved vitamin A status. Similar findings have been reported in preschool children with protein-energy malnutrition and vitamin A deficiency.2s The lack of correlation

Volume 116 Number 4

between the 2x-RBP values and the plasma concentrations of RBP suggests that the magnitude of the plasma RBP response is independent of the baseline plasma concentration of RBP. Whether other nutrient deficiencies influence the plasma RBP response to vitamin A administration in VLBW~ infants remains to be studied. We previously reported that early normalization of plasma vitamin A concentrations results in a nearly 50% reduction in the incidence of BPD and a decrease in its associated morbidity in susceptible VLBW infants who are given vitamin A supplementation, in comparison with unsupplemented control subjects.14 We have demonstrated the same beneficial effect of vitamin A supplementation in this study. Study infants in whom BPD developed, in comparison with those without BPD, were more premature at birth, had more severe initial lung disease and subsequent need for prolonged ventilatory support, and had a higher incidence of complications such as airway infection and retinopathy of prematurity. The overall incidence of retinopathy observed in this study was similar to that reported earlier in vitamin A-supplemented VLBW infants. 14 Infants with BPD in this study did not differ from those without BPD with respect to the total vitamin A intake, the mean plasma concentrations of vitamin A and RBP, or the mean plasma retinol/RBP molar ratios in any of the serial determinations; however, the plasma RBP response to vitamin A administration appeared to differ. The plasma RBP response was characterized by high A-RBP values (>8%) despite vitamin A supplementation and normalization of conventional plasma indexes of vitamin A in most infants with BPD, indicative of the persistence of low vitamin A status in these infants. In contrast, the A-RBP values were low after vitamin A supplementation in all infants without BPD, indicative of normalization of vitamin A status in these infants. These results will require verification with larger numbers of infants. It is possible that the requirement of vitamin A in infants with BPD relative to those without BPD was higher because of the ongoing need for regenerative healing from lung injury. It is also possible that the more immature the infant, and the more injured the lung, the less efficient was the utilization of available vitamin A at a cellular level. Additionally, it is possible that the provision of vitamin A to infants with BPD relative to those without BPD was less than optimum, perhaps because a higher percentage of vitamin A intake was received by these infants via the parenteral route. Previous studies have shown that the parenteral administration of vitamin A is inefficient because of substantial photodegradative and adsorptive losses of the vitamin.2931 These observations lead us to believe that although vitamin A supplementation at the dosage used in this study normalizes conventional plasma indexes of vitamin A in VLBW infants, the plasma RBP response to vi-

P l a s m a R B P response to vitamin A

6 13

tamin A administration may continue to reflect persistence of low vitamin A status in the more immature infants with significant lung disease. We suggest that the plasma RBP response to vitamin A administration may be a useful functional test for identification of such infants and that their vitamin A intake may need modification based on the response. Long-term vitamin A supplementation in VLBW neonates raises the need for monitoring of their vitamin A status. Plasma vitamin A concentratiojas, do not always correlate with liver vitamin A concentrations.7, 32 Likewise, plasma RBP concentrations and plasma retinol/RBP molar ratios, although useful indexes of vitamin A status, can be potentially altered in severe protein malnutrition33 and other disease states. 27 Although liver vitamin A concentrations are more accurate indicators of vitamin A status, 34 these measurements cannot be used for monitoring. Thus there is a need for a diagnostic test that is simple to perform, clinically applicable, and a predictive indicator of vitamin A status. Loerch et al. 35 reported a relative dose-response test in which the increase in plasma concentration of vitamin A 5 hours after an oral dose of retinyl acetate was determined and correlated over a wide range of levels of plasma and liver vitamin A values in rats. This approach has been evaluated in adults with alcoholic hepatic cirrhosis, 36 adults undergoing abdominal surgery, 37 and children subjected to liver biopsy.3s A variation of the relative dose response test has been performed in preterm infants to determine their vitamin A status at discharge from the hospital. 39 No information is available regarding the usefulness of this test as a reliable indicator of the vitamin A status of VLBW infants. The plasma RBP response to vitamin A administration was simple to assess. The dose of vitamin A (2000 IU/kg retinyl palmitate) used was within the range that we have previously reported to be safe for VLBW infants. 14 The intramuscular route of administration avoided the potential difficulties associated with variable absorption after enteral administration and with variable losses in the delivery system after intravenous administration of vitamin A. The blood volume required for testing was within the range commonly used for biochemical monitoring of VLBW infants. More important, the test provided information that could be used for identification of infants with persistence of low vitamin A status despite supplementation and possibly with increased risk for BPD. Additional studies will be important in assessing the value of this diagnostic approach in the management of VLBW infants at risk for severe complications of lung immaturity. We thank Dr. Thomas Hazinski for helpful advice; Mark Hunt and Lucie Chytil for technical assistance in the laboratory; and Dr. Robert Parker for assistance in statistical evaluation.

6 14

Shenai et al.

REFERENCES

1. Kanai M, Raz A, Goodman DS. Retinol-binding protein: the transport protein for vitamin A in human plasma. J Clin Invest 1968;47:2025-44. 2. Rask L, Anundi H, Bohme J, et al. Structural and functional studies of vitamin A-binding proteins. Ann NY Acad Sci 1981;359:79-90. 3. Raz A, Shiratori T, Goodman DS. Studies on the protein-protein and protein-ligand interactions involved in retinol transport in plasma. J Biol Chem 1970;245:1903-12. 4. Goodman DS. Plasma retinol-binding protein. In: Sporn MB, Roberts AB, Goodman DS, eds. The retinoids; vol 2. Orlando, Fla.: Academic Press, 1984:41-88. 5. Muto Y, Smith JE, Milch PO, Goodman DS. Regulation of retinol-binding protein metabolism by vitamin A status in the rat. J BiN Chem 1972;247:2542-50. 6. 'Smith JE, Muto Y, Milch PO, Goodman DS. The effects of chylomicron vitamin A on the metabolism of retinol-binding protein in the rat. J Biol Chem 1973;248:1544-% 7. Shenai JP, Chytil F, Stahlman MT. Liver vitamin A reserves of very low birth weight neonates. Pediatr Res 1985;19:892-3. 8. Brandt RB, Mueller DG, Schroeder JR, et al. Serum vitamin A in premature and term neonates. J PEDIATR 1978;92:101-4. 9. Shenai JP, Chytil F, Jhaveri A, Stahlman MT. Plasma vitamin A and retinol-binding protein in premature and term neonates. J PEDIATR 1981;99:302-5. 10. Bhatia J, Ziegler EE. Retinol-binding protein and prealbumin in cord blood of term and preterm infants. Early Hum Dev 1983;8:129-33. 11. Northway WH Jr, Rosan RC, Porter DY. Pulmonary disease following respirator therapy of hyaline-membrane disease: bronchopulmonary dysplasia. N Engl J Med 1967;276:357-68. 12. Shenai JP, Chytil F, Stahlman MT. Vitamin A status of neonates with bronchopulmonary dysplasia. Pediatr Res 1985;19:185-8. 13. Hustead VA, Gutcher GR, Anderson SA, Zachman RD. Relationship of vitamin A (retinol) status to lung disease in the preterm infant. J PEDIATR 1984;105:610-5. 14. Shenai JP, Kennedy KA, Chytil F, Stahlman MT. Clinical trial of vitamin A supplementation in infants susceptible to bronchopulmonary dysplasia. J PEDIATR 1987;111:269-77. 15. Bancalari E, Abdenour GE, Feller R, Gannon J. Bronchopulmonary dysplasia: clinical presentation. J PEDIATR 1979; 95:819-23. 16. Stahlman MT. Acute respiratory disorders in the newborn. In: Avery GB, ed. Neonatology: pathophysiology and management of the newborn. 3rd ed. Philadelphia: JB Lippincott, 1987:4 t 8-45. 17. Thompson JN, Erdody P, Brien R, Murray TK. Flurometric determination of vitamin A in human blood and liver. Biochem Med 1971;5:67-89. I8. Smith JE, Goodman DS. RetinoI-binding protein and the regulation of vitamin A transport. Fed Proc 1979;38:2504-9. 19. Monji N, Bosin E. Use of enzyme-linked immunosorbent assay technique for quantitation of serum retinol-binding protein. Methods Enzymol 1986;123:85-92. 20. Vahlquist A, Peterson PA, Wibell L. Metabolism of the vitamin A transporting protein complex. 1. Turnover studies in normal persons and in patients with chronic renal failure. Eur J Clin Invest 1973;3:352-62. 21. Edwards DK. Radiographic aspects of bronchopulmonary dysplasia. 3 PEDIATR I979;95:823-9.

The Journal of Pediatrics April 1990

22. Cotton RB, Hickey D, Stahlman MT. Management of premature infants with symptomatic patent ductus arteriosus. In: Stern L, Bard H, Friis-Hansen B, eds. Intensive care in the newborn; vol 4. New York: Masson, 1983:225-31. 23. Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymat and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gin. J PEDIATR 1978;92:529-34. 24. The Committee for the Classification of Retinopathy of Prematurity. An international classification of retinopathy of prematurity. Arch Ophthalmol 1984; 102:1130-4. 25. Underwood BA. Vitamin A in animal and human nutrition. In: Sporn MB, Roberts AB, Goodman DS, eds. The ret~noids; vol 1. Orlando, Fla.: Academic Press, 1984:281-392. 26. Vahlquist A, Rask L, Peterson PA, Berg T. The concentrations of retinol-binding protein, prealbumin, and transferrin in the sera of newly delivered mothers and children of various ages. Scand J Clin Lab invest I975;35:569-75. 27. Smith FR, Goodman DS. The effects of diseases of the liver, thyroid, and kidneys on the transport of vitamin A in human plasma. J CIin Invest 1971;50:2426-36. 28. Large S, Nea[ G, Glover J, Thanangkul O, Olson RE. The early changes in retinol-binding protein and prealbumin concentrations in plasma of protein-energy malnourished children after treatment with retinol and an improved diet. Br J Nutr 1980;43:393-402. 29. Howard L, Chu R, Feman S, Mintz H, Ovesen L, Wolf B. Vitamin A deficiency from long-term parenteral nutrition. Ann Intern Med 1980;93:576-7. 30. Hartline JV, Zachman RD. Vitamin A delivery in total parenteral nutrition solution. Pediatrics 1976;58:448-51. 31. Shenai JP, Stahlman MT, Chytil F. ViiLaminA delivery from parenteral alimentation solution. J PEDIATR 1981;99:661-3. 32. Meyer KA, Popper H, Steigmann F, Walters WH, Zevin S. Comparison of vitamin A of liver biopsy specimens with plasma vitamin A in man. Proc Soc Exp Biol Med 1942;49: 589-91. 33. lngenbleek Y, Van Den Sehrieck HG, De Nayer P, De Visscher M. The rote of retinol-binding protein in protein-calorie malnutrition. Metabolism 1975;24:633-41. 34. Olson JA. Evaluation of vitamin A status in children. World RevNutr Diet 1978;31:130-4. 35. Loerch JD, Underwood BA, Lewis KC. Response of plasma levels of vitamin A to a dose of vitamin A as an indicator of hepatic vitamin A reserves in rats. J Nutr 1979;109:778-86. 36. Mobarhan S, Russell RM, Underwood BA, Wallingford J, Mathieson RD, A1-Midani H. Evaluation of the relative dose response test for vitamin A nutriture in cirrhotics. Am J Clin Nutr 198I;34:2264-70. 37. Amedee-Manesme O, Anderson D, Olson JA. Relation of the relative dose response to liver concentrations of vitamin A in generally weIl-nourished surgical patients. Am J Clin Nutr 1984;39:898-902. 38. Amedee-Manesme O, Mourey MS, Hanck A, Therasse J. Vitamin A relative dose response test: validation by intravenous injection in children with liver disease. Am J Clin Nutr 1987;46:286-9. 39. Woodruff CW, Latham CB, Mactier H, Hewett JE. Vitamin A status of preterm infants: correlation between plasma retinol concentration and retinol dose response. Am J Clin Nutr 1987;46:985-8.