Original Article
Vitamin E in the Preterm Infant: A Forgotten Cause of Hemolytic Anemia Enrique Gomez-Pomar, MD, MSc1 Emily Hatfield, MS, PA-C2 Philip M. Westgate, PhD4 Henrietta S. Bada, MD, MPH1 1 Division of Neonatology, Department of Pediatrics, University of
Kentucky, Lexington, Kentucky 2 Division of Neonatology, Seattle Children’s Hospital, Seattle, Washington 3 Department of Pharmacy, Kentucky Children’s Hospital, Lexington, Kentucky 4 Department of Biostatistics, College of Public Health, University of Kentucky, Lexington, Kentucky
Karen Garlitz, PharmD, BCPPS, BCPS3
Address for correspondence Enrique Gomez-Pomar, MD, MSc, Division of Neonatology, Department of Pediatrics, University of Kentucky, 138 Leader Avenue, Room 018, Lexington, KY 40508-9983 (e-mail: [email protected]).
Abstract
Keywords
► ► ► ►
vitamin E preterm infants hemolytic anemia preterm anemia
Objective Vitamin E deficiency in premature infants has been associated with hemolytic anemia. Its incidence decreased after the supplementation of preterm formulas and parenteral nutrition with vitamin E. Despite this, some infants still develop hemolytic anemia and receive vitamin E. Design Retrospective analysis of 70 infants admitted to a level IV intensive care unit and who developed hemolytic anemia and were treated with vitamin E. Infants were classified into two groups based on whether or not they responded to vitamin E therapy. Statistical methods included the use of descriptive statistics and marginal logistic regression models. Results Low hematocrit and reticulocytosis before vitamin E administration were associated with adequate response to treatment. Thrombocytosis, iron treatment (duration and dose), gestational age, birth weight, and type of feedings were not. Infants who received a short duration of parenteral nutrition and were on oxygen responded to vitamin E therapy. Infants with a hematocrit 26% and reticulocyte of 36.1% were more likely to respond to vitamin E. Conclusion Although formulas and parenteral nutrition are supplemented with vitamin E; some preterm infants may still develop hemolytic anemia. Those with anemia, reticulocytosis, and oxygen requirement may benefit from additional vitamin E.
Vitamin E (Vit E) refers to a family of fat-soluble nutrients that are known as tocopherols,1,2 the main function of which has been identified as free radical scavengers.1–4 The placental transfer of Vit E is minimal compared with the enteral absorption of Vit E from breast milk.5 Compared with term infants, preterm infants have decreased levels of Vit E as identified by Moyer in 19506 who also found that premature infants have poor absorption of Vit E during the first 2 months of life and this finding was also supported in subsequent studies.5,7–9
The function of Vit E as an antioxidant and its decreased level in premature infants was a topic of debate among neonatal health care providers trying to prevent morbidities, such as bronchopulmonary dysplasia, retinopathy of prematurity (ROP), and intraventricular hemorrhage (IVH).1,2 A 2003 Cochrane review2 showed that Vit E plays a role in the prevention of IVH, severe ROP, and blindness but at the expense of an increased risk of sepsis. For this reason, prophylactic Vit E (especially high intravenous doses) is not recommended in premature infants.
received July 24, 2017 accepted after revision September 2, 2017
Copyright © by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI https://doi.org/ 10.1055/s-0037-1607283. ISSN 0735-1631.
Downloaded by: Scott Memorial Library, AISR. Copyrighted material.
Am J Perinatol
Gomez-Pomar et al.
Low plasma Vit E levels have been associated with increased sensitivity to peroxidase damage on the erythrocyte membrane which could lead to hemolysis.10 Oski and Barness in 1967 found signs of hemolysis in a control group of low-birthweight infants at 6 to 11 weeks of age;11 hemolysis decreased after the administration of Vit E. In comparison, infants who received Vit E since birth were found to have increased hemoglobin and decreased reticulocyte (retic) level. The association between thrombocytosis and Vit E deficiency has been described in the literature,4,10,12,13 however, controversy remains as to its clinical significance.2,14 The relationship between Vit E deficiency and hemolysis was further studied12,13,15–19 and the association between greater hemolysis and iron administration was proposed.13,14,19,20 Iron is recommended in preterm infants to prevent anemia of prematurity.21,22 Iron can cause lipid peroxidation which could disrupt the erythrocyte membrane in the presence of an already decreased Vit E level. These results led the American Academy of Pediatrics to recommend Vit E supplementation in all preterm formulas23 as well as parenteral nutrition for preterm infants.4,23,24 After this recommendation was implemented, no new studies regarding the use of Vit E in preterm infants were conducted and one study on Vit E levels by Kositamongkol et al in 2011 reported deficiency in 77.4, 16.1, and 35.7% of 35 very low-birth-weight infants at birth, at the postnatal age to reach full feedings, and at term postconceptional age, respectively.25 With the existing standard recommendation of Vit E supplementation, the need for additional treatment of Vit E deficiency remains controversial2,23 especially in the presence of other confounders, such as lower gestational age and method of supplementation, whether by parenteral nutrition, formula or both. As per our unit clinical practice guideline, hematocrit (hct) and retic count are monitored weekly, and oral Vit E therapy is started when hemolytic anemia and thrombocytosis are noted after exclusion of other causes. The objective of this study is to analyze the effect of Vit E therapy in preterm infants with anemia and hemolysis and the factors associated with successful treatment.
Methods This retrospective study was conducted by reviewing electronic medical records data from a level IV neonatal intensive care unit (NICU). The study was approved by the institutional review board. Inclusion criteria included infants who received Vit E therapy during the study period (2013, 2014, and 2015) and whose gestational age were 326/7 weeks or less. Infants who received a red blood cell transfusion during the treatment period were excluded.
Vitamin E In our NICU, Vit E is administered as part of the parenteral nutrition using Infuvite Pediatric, Baxter Healthcare Corporation, Deerfield, IL. Each 1 mL vial of Infuvite contains Vit E 7 IU (7 mg of dl-α-tocopheryl acetate). Infants < 1 kg, receive 1.5 mL/day and infants 1 to 3 kg receive 3.25 mL/day American Journal of Perinatology
of Infuvite via parenteral nutrition, corresponding to 2.1 IU and 4.55 IU of Vit E per day, respectively. Based on the results of the weekly hct and retic count, Vit E deficiency was considered after ruling out other factors or causes of hemolytic anemia. Vit E therapy was initiated at the discretion of the attending physician. Vit E was administered in a liquid form with a concentration of 15 units/0.3 mL. Since there is not a recommended standardized dose of Vit E,26 a dose of 5 units of Vit E (0.1 mL) per day was given orally for 10 days as per our NICU guideline.
Data Collection Demographic information was collected for all enrolled infants. To analyze the effect of Vit E, we collected hct, retic count, and platelet (plt) levels before and after the treatment. The desired outcome was an increase in hct as a response to treatment with oral Vit E. Infants who responded to Vit E therapy were compared with nonresponders. Information on red blood cell transfusion (number and the time interval between the last transfusion and start of Vit E treatment) was also collected.
Statistical Analysis Statistics were based on the comparisons of the treatment responders and nonresponders. Sample means and standard deviations or medians and interquartile ranges are presented for continuous variables, and frequencies and percentages are given for categorical variables. Groups were also compared via marginal logistic regression models, modeling the odds of the given success. Models were fit using generalized estimating equations27 to ensure the validity of inference due to some subjects contributing multiple observations. After fitting univariate models, multivariate models were fit. Backward elimination was then utilized to find final models. Additional results are also presented with respect to univariate models in which hct, retic, and plt were the predictors of success. Specifically, we present the C statistic, which is an estimate for the area under the receiver operating characteristic curve and thus the predictive accuracy of the variable, and the optimal cutoff value based on the minimum Euclidean distance and maximum sum of sensitivity and specificity. Furthermore, we give sensitivities (Sn), specificities (Sp), positive predictive values (PPV), and negative predictive values (NPV) based on these optimal cutoffs. All tests were two-sided using a 5% significance level, with some exceptions in our backward elimination procedures (see footnote of ►Table 2). Analyses were conducted in SAS version 9.4 (SAS Institute, Cary, NC).
Results During the study period, 76 infants received Vit E and were evaluated. Six infants were excluded, leaving 70 infants for inclusion in the final analysis. A total of 46 infants achieved the desired outcome of an increase in hct. ►Table 1 shows the demographic characteristics and the potential clinical predictors for all of the infants. Clinical factors associated with the decision to treat infants with Vit E included the levels of hct,
Downloaded by: Scott Memorial Library, AISR. Copyrighted material.
Vitamin E in the Preterm Infant
Vitamin E in the Preterm Infant
Gomez-Pomar et al.
Table 1 Demographics and potential clinical predictors of achieving the outcome (increase in hct) in response to Vit E therapy Achieved n ¼ 46
Did not achieve n ¼ 24
p Value
C statistic
GA (wk)
26.6 2.0
27.5 2.2
0.06
0.633
BW (kg)
0.93 0.29
1.02 0.38
0.28
0.540
Hct before Vit E
25.8 2.8
27.6 3.4
0.02
0.663
Reticulocytes before Vit E
8.68 4.55 (n ¼ 42)
5.24 2.33 (n ¼ 21)
0.004
0.754
Plt before Vit E
489 176
576 151 (n ¼ 21)
0.051
0.642
Hct after Vit E
28.2 2.9
26.1 3.1
0.01
0.670
Reticulocytes after Vit E
7.6 2.9 (n ¼ 45)
6.5 2.6 (n ¼ 21)
0.08
0.620
Plt after Vit E
432.6 161.7 (n ¼ 44)
468.8 173.8 (n ¼ 23)
0.42
0.550 0.584
Day of life at start of iron
28.1 12.9
24.1 8.0
0.06
Hct at start of iron
29.3 4.1 (n ¼ 43)
30.9 5.0 (n ¼ 23)
0.14
0.583
Reticulocytes at start of iron
5.66 3.21 (n ¼ 32)
3.60 2.40 (n ¼ 15)
0.08
0.743
Iron start dose (mg) 3
Vitamin E in the Preterm Infant: A Forgotten Cause of Hemolytic Anemia Enrique Gomez-Pomar, MD, MSc1 Emily Hatfield, MS, PA-C2 Philip M. Westgate, PhD4 Henrietta S. Bada, MD, MPH1 1 Division of Neonatology, Department of Pediatrics, University of
Kentucky, Lexington, Kentucky 2 Division of Neonatology, Seattle Children’s Hospital, Seattle, Washington 3 Department of Pharmacy, Kentucky Children’s Hospital, Lexington, Kentucky 4 Department of Biostatistics, College of Public Health, University of Kentucky, Lexington, Kentucky
Karen Garlitz, PharmD, BCPPS, BCPS3
Address for correspondence Enrique Gomez-Pomar, MD, MSc, Division of Neonatology, Department of Pediatrics, University of Kentucky, 138 Leader Avenue, Room 018, Lexington, KY 40508-9983 (e-mail: [email protected]).
Abstract
Keywords
► ► ► ►
vitamin E preterm infants hemolytic anemia preterm anemia
Objective Vitamin E deficiency in premature infants has been associated with hemolytic anemia. Its incidence decreased after the supplementation of preterm formulas and parenteral nutrition with vitamin E. Despite this, some infants still develop hemolytic anemia and receive vitamin E. Design Retrospective analysis of 70 infants admitted to a level IV intensive care unit and who developed hemolytic anemia and were treated with vitamin E. Infants were classified into two groups based on whether or not they responded to vitamin E therapy. Statistical methods included the use of descriptive statistics and marginal logistic regression models. Results Low hematocrit and reticulocytosis before vitamin E administration were associated with adequate response to treatment. Thrombocytosis, iron treatment (duration and dose), gestational age, birth weight, and type of feedings were not. Infants who received a short duration of parenteral nutrition and were on oxygen responded to vitamin E therapy. Infants with a hematocrit 26% and reticulocyte of 36.1% were more likely to respond to vitamin E. Conclusion Although formulas and parenteral nutrition are supplemented with vitamin E; some preterm infants may still develop hemolytic anemia. Those with anemia, reticulocytosis, and oxygen requirement may benefit from additional vitamin E.
Vitamin E (Vit E) refers to a family of fat-soluble nutrients that are known as tocopherols,1,2 the main function of which has been identified as free radical scavengers.1–4 The placental transfer of Vit E is minimal compared with the enteral absorption of Vit E from breast milk.5 Compared with term infants, preterm infants have decreased levels of Vit E as identified by Moyer in 19506 who also found that premature infants have poor absorption of Vit E during the first 2 months of life and this finding was also supported in subsequent studies.5,7–9
The function of Vit E as an antioxidant and its decreased level in premature infants was a topic of debate among neonatal health care providers trying to prevent morbidities, such as bronchopulmonary dysplasia, retinopathy of prematurity (ROP), and intraventricular hemorrhage (IVH).1,2 A 2003 Cochrane review2 showed that Vit E plays a role in the prevention of IVH, severe ROP, and blindness but at the expense of an increased risk of sepsis. For this reason, prophylactic Vit E (especially high intravenous doses) is not recommended in premature infants.
received July 24, 2017 accepted after revision September 2, 2017
Copyright © by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI https://doi.org/ 10.1055/s-0037-1607283. ISSN 0735-1631.
Downloaded by: Scott Memorial Library, AISR. Copyrighted material.
Am J Perinatol
Gomez-Pomar et al.
Low plasma Vit E levels have been associated with increased sensitivity to peroxidase damage on the erythrocyte membrane which could lead to hemolysis.10 Oski and Barness in 1967 found signs of hemolysis in a control group of low-birthweight infants at 6 to 11 weeks of age;11 hemolysis decreased after the administration of Vit E. In comparison, infants who received Vit E since birth were found to have increased hemoglobin and decreased reticulocyte (retic) level. The association between thrombocytosis and Vit E deficiency has been described in the literature,4,10,12,13 however, controversy remains as to its clinical significance.2,14 The relationship between Vit E deficiency and hemolysis was further studied12,13,15–19 and the association between greater hemolysis and iron administration was proposed.13,14,19,20 Iron is recommended in preterm infants to prevent anemia of prematurity.21,22 Iron can cause lipid peroxidation which could disrupt the erythrocyte membrane in the presence of an already decreased Vit E level. These results led the American Academy of Pediatrics to recommend Vit E supplementation in all preterm formulas23 as well as parenteral nutrition for preterm infants.4,23,24 After this recommendation was implemented, no new studies regarding the use of Vit E in preterm infants were conducted and one study on Vit E levels by Kositamongkol et al in 2011 reported deficiency in 77.4, 16.1, and 35.7% of 35 very low-birth-weight infants at birth, at the postnatal age to reach full feedings, and at term postconceptional age, respectively.25 With the existing standard recommendation of Vit E supplementation, the need for additional treatment of Vit E deficiency remains controversial2,23 especially in the presence of other confounders, such as lower gestational age and method of supplementation, whether by parenteral nutrition, formula or both. As per our unit clinical practice guideline, hematocrit (hct) and retic count are monitored weekly, and oral Vit E therapy is started when hemolytic anemia and thrombocytosis are noted after exclusion of other causes. The objective of this study is to analyze the effect of Vit E therapy in preterm infants with anemia and hemolysis and the factors associated with successful treatment.
Methods This retrospective study was conducted by reviewing electronic medical records data from a level IV neonatal intensive care unit (NICU). The study was approved by the institutional review board. Inclusion criteria included infants who received Vit E therapy during the study period (2013, 2014, and 2015) and whose gestational age were 326/7 weeks or less. Infants who received a red blood cell transfusion during the treatment period were excluded.
Vitamin E In our NICU, Vit E is administered as part of the parenteral nutrition using Infuvite Pediatric, Baxter Healthcare Corporation, Deerfield, IL. Each 1 mL vial of Infuvite contains Vit E 7 IU (7 mg of dl-α-tocopheryl acetate). Infants < 1 kg, receive 1.5 mL/day and infants 1 to 3 kg receive 3.25 mL/day American Journal of Perinatology
of Infuvite via parenteral nutrition, corresponding to 2.1 IU and 4.55 IU of Vit E per day, respectively. Based on the results of the weekly hct and retic count, Vit E deficiency was considered after ruling out other factors or causes of hemolytic anemia. Vit E therapy was initiated at the discretion of the attending physician. Vit E was administered in a liquid form with a concentration of 15 units/0.3 mL. Since there is not a recommended standardized dose of Vit E,26 a dose of 5 units of Vit E (0.1 mL) per day was given orally for 10 days as per our NICU guideline.
Data Collection Demographic information was collected for all enrolled infants. To analyze the effect of Vit E, we collected hct, retic count, and platelet (plt) levels before and after the treatment. The desired outcome was an increase in hct as a response to treatment with oral Vit E. Infants who responded to Vit E therapy were compared with nonresponders. Information on red blood cell transfusion (number and the time interval between the last transfusion and start of Vit E treatment) was also collected.
Statistical Analysis Statistics were based on the comparisons of the treatment responders and nonresponders. Sample means and standard deviations or medians and interquartile ranges are presented for continuous variables, and frequencies and percentages are given for categorical variables. Groups were also compared via marginal logistic regression models, modeling the odds of the given success. Models were fit using generalized estimating equations27 to ensure the validity of inference due to some subjects contributing multiple observations. After fitting univariate models, multivariate models were fit. Backward elimination was then utilized to find final models. Additional results are also presented with respect to univariate models in which hct, retic, and plt were the predictors of success. Specifically, we present the C statistic, which is an estimate for the area under the receiver operating characteristic curve and thus the predictive accuracy of the variable, and the optimal cutoff value based on the minimum Euclidean distance and maximum sum of sensitivity and specificity. Furthermore, we give sensitivities (Sn), specificities (Sp), positive predictive values (PPV), and negative predictive values (NPV) based on these optimal cutoffs. All tests were two-sided using a 5% significance level, with some exceptions in our backward elimination procedures (see footnote of ►Table 2). Analyses were conducted in SAS version 9.4 (SAS Institute, Cary, NC).
Results During the study period, 76 infants received Vit E and were evaluated. Six infants were excluded, leaving 70 infants for inclusion in the final analysis. A total of 46 infants achieved the desired outcome of an increase in hct. ►Table 1 shows the demographic characteristics and the potential clinical predictors for all of the infants. Clinical factors associated with the decision to treat infants with Vit E included the levels of hct,
Downloaded by: Scott Memorial Library, AISR. Copyrighted material.
Vitamin E in the Preterm Infant
Vitamin E in the Preterm Infant
Gomez-Pomar et al.
Table 1 Demographics and potential clinical predictors of achieving the outcome (increase in hct) in response to Vit E therapy Achieved n ¼ 46
Did not achieve n ¼ 24
p Value
C statistic
GA (wk)
26.6 2.0
27.5 2.2
0.06
0.633
BW (kg)
0.93 0.29
1.02 0.38
0.28
0.540
Hct before Vit E
25.8 2.8
27.6 3.4
0.02
0.663
Reticulocytes before Vit E
8.68 4.55 (n ¼ 42)
5.24 2.33 (n ¼ 21)
0.004
0.754
Plt before Vit E
489 176
576 151 (n ¼ 21)
0.051
0.642
Hct after Vit E
28.2 2.9
26.1 3.1
0.01
0.670
Reticulocytes after Vit E
7.6 2.9 (n ¼ 45)
6.5 2.6 (n ¼ 21)
0.08
0.620
Plt after Vit E
432.6 161.7 (n ¼ 44)
468.8 173.8 (n ¼ 23)
0.42
0.550 0.584
Day of life at start of iron
28.1 12.9
24.1 8.0
0.06
Hct at start of iron
29.3 4.1 (n ¼ 43)
30.9 5.0 (n ¼ 23)
0.14
0.583
Reticulocytes at start of iron
5.66 3.21 (n ¼ 32)
3.60 2.40 (n ¼ 15)
0.08
0.743
Iron start dose (mg) 3
