AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 7, NUMBER 2
April 1990
TOLERANCE OF FAT EMULSIONS IN VERY LOW BIRTHWEIGHT NEONATES EFFECT OF BIRTHWEIGHT ON PLASMA LIPID CONCENTRATIONS Yves W. Brans, M.D., Donna S. Andrew, M.S., Donna W. Carrillo, B.S.N., M.P.H., Elisabeth B. Dutton, B.S.N., Elizabeth M. Menchaca, B.A., and Belen A. Puleo-Scheppke, B.A., M.P.H.
Plasma concentrations of various lipid fractions (total lipids, free glycerol, true triglycerides, free fatty acids, and cholesterol) were studied in 20 normally grown neonates ranging in birthweightfrom820to1500gm and ingestational age from 28 to 34 weeks. They were subdivided into three weight categories: 750to999,1000to1249, and 1250 to 1500 gm. A lipid emulsion was infused at a constant rate over 24 hours, beginning with an hourly infusion rate of 0.04 gm/kg and increasing each day by 0.04 gm/kg up to a maximum of 0.16 gm/kg. Neonates who weighed less than 1000 gm had higher mean plasma concentrations of total lipids and free glycerol at hourly infusion rates of 0.08 and 0.16 and of triglycerides and free fatty acids at hourly infusion rate of 0.16 gm/kg than their heavier peers. These data suggest that extreme caution be used when administering parenteral fat emulsions to neonates who weigh less than 1000 gm and that we need to monitor plasma closely for signs of hyperlipemia.
Increases in plasma lipid concentrations have been documented in low birthweight neonates receiving various doses of fat emulsions parenterally.1"13 Available data suggest that preterm neonates may not clear fat emulsion particles from plasma as rapidly as older children or adults. Of particular concern are the adverse effects of reduced lipoprotein lipase and restricted adipose tissue mass on the clearance of fat particles in very low birthweight neonates.14 In this report we compare the effects of fat emulsions on plasma lipid patterns in three groups of increasingly lower birthweight neonates. The effect of fat emulsions on intake of fluids and calories and the effect of the various infusion regimens and of hourly rate of lipid infusion on plasma lipid patterns have been reported separately.15-16
MATERIALS AND METHODS
Tolerance to parenterally administered fat emulsions was studied in neonates who weighed 1500 gm or less at birth. Informed parental consent was obtained in all cases and the study was approved by the Institutional Review Board of the University of Texas Health Center at San Antonio. Birthweights were recorded to the nearest 10 gm. Gestational ages were determined from the mother's menstrual history, checked in most cases by sonographic determination of the biparietal diameter, and confirmed by physical examination of the neonate.17 All neonates had normal growth, that is, their birthweights were between the 10th and 90th percentiles for gestational maturity, sex, and race. Separate birthweightgestational age curves were used for Hispanic (Gibbs
Neonatal Research Laboratories, Departments of Pediatrics and Obstetrics and Gynecology, Wayne State University School of Medicine, Detroit, Michigan, and the Perinatal Research Laboratory, Departments of Pediatrics and Obstetrics and Gynecology, The University of Texas Health Science Center, San Antonio, Texas Supported in part by grant HD 15967 from the National Institute of Child Health and Human Development, Bethesda, Maryland, and by KabiVitrum, Alameda, California
114
Copyright © 1990 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 10016. All rights reserved.
Downloaded by: University of British Columbia. Copyrighted material.
ABSTRACT
EFFECT OF BIRTHWEIGHT ON LIPIDS/Brans et al.
recorded but the standard deviation was ignored. The mean values obtained for each infant while receiving a stated rate of lipid infusion were averaged, but the standard deviation within that infant was again ignored. The means analyzed were therefore the mean of the average values for individual neonates who received 0 (1 value), 0.04 (2 values), 0.08 (2 values), 0.12 (2 values), or 0.16 (8 values) gm/kg of lipids per hour (Fig. 1), and the standard error of the mean was calculated from the standard deviations across subjects. The sampling unit used (df) was each subject, not each measurement. In view of the small number of observations in some groups, the data were also analyzed nonparametrically by the Kruskal-Wallis test. Since parametric and nonparametric tests gave identical results, only the former are presented. RESULTS
Downloaded by: University of British Columbia. Copyrighted material.
CE: unpublished data), white Anglo-American,18 and black18 babies. Neonates weighing less than 750 gm at birth or with estimated gestational ages below 27 weeks were excluded from the study. Total parenteral nutrition was started on the third postnatal day with a basic regimen to provide daily 100 to 130 ml/kg of fluids (babies were cared for under radiant warmers), 2.5 gm/kg of crystalline amino acids, as much dextrose as tolerated, electrolytes, and vitamins. Individual intakes were tailored to each neonate's requirements and metabolic tolerance in order to provide maximal caloric intake. All decisions concerning nutrition, save one, were made by the attending staff. The one exception was the means and amount of administration of fat emulsion, which were specified by study protocol. The fat emulsion (Intralipid 20%, KabiVitrum, Alameda, CA) was provided at constant rates over 24 hours, beginning with an hourly infusion rate of 0.04 gm/ kg on the first day of infusion (1 gm/kg/day) and increasing each day by 0.04 gm/kg/hour up to a maximum of 0.16 gm/kg/hour (4 gm/kg/day). As long as the neonate required an arterial umbilical catheter for blood gas monitoring, parenteral fluids, fat emulsion excepted, were infused via this route. The tip of the catheter was always located radiographically below the bottom of the third lumbar vertebra. After removal of the catheter, peripheral veins were used. The fat emulsion was infused through a peripheral vein dedicated to that purpose. Enteral feedings were not provided during the study. The study was discontinued if the plasma was frankly creamy,5 if a serious medical problem interfered with the study (such as surgery or necrotizing enterocolitis), if a breach of protocol occurred, or if the parents requested it. The study period lasted 8 days. Blood samples were obtained through the umbilical catheter, after removal of 3 ml of blood to clear the line of infusion fluids or by heel-stick before beginning the fat infusion and every 12 hours thereafter. Plasma concentrations of total lipids, free glycerol, true triglycerides (that is, measured triglycerides minus free glycerol), free fatty acids, and cholesterol were determined in triplicates or quadruplicate, as previously described.16 Each variable was analyzed by one-way analysis of variance to detect differences between the three groups of birthweights at a given time during the study and, within each weight group, differences in mean plasma lipid values associated with increasing hourly rates of lipid infusion. When main effect differences were detected (F test, p =s 0.05), Duncan's multiple range test was applied to locate differences between means for various groups. A significance level equal to or less than 0.05 was chosen to define statistical significance. Although there was some variability in mean plasma concentrations of the various lipid fractions during infusion of fat at a given hourly rate, these differences were not statistically significant. Each specimen of plasma was analyzed in triplicate or quadruplicate: the mean value was
Twenty neonates entered the study. Two infants failed to complete the first four days of the study: one developed severe hyperlipemia (plasma was frankly creamy by inspection), requiring discontinuation of the lipid infusion, and the other one -I249g
I25O-I5OOJ
— 1500 f
1 S occ
Figure 1. Mean (± SEM) plasma concentrations of total lipids, free fatty acids, glycerol, triglycerides, and cholesterol before infusion and at each rate of lipid infusion. Each bar corresponds to different hourly rates of lipid infusion (from left to right, successively, 0, 0.04, 0.08, 0.12, and 0.16 gm/kg/day). Cross-hatched bars indicate mean values that are significantly higher than mean preinfusion value for that group. Symbols above the bars indicate statistically significant differences at a given time between the 750 to 999 gm and 1000 to 1249 gm groups (asterisk) or between the 750 to 999 gm and the 1250 to 1550 gm groups (dagger).
115
died; their data were excluded from analysis (attrition rate, 10%). Selected characteristics of the 18 remaining neonates are detailed in Table 1. Neonates in the three weight groups had similar mean preinfusion concentrations of the various lipid fractions (Fig. 1). Within weight groups, mean plasma lipid concentrations increased with the hourly rate of lipid infusion, although some differences did not reach statistical significance due to large standard errors of the mean and small number of subjects. At hourly infusion rates of 0.04 gm/kg, there were no statistically significant differences in plasma lipid patterns between the three weight categories. At infusion rates of 0.08 gm/kg/hour, the smaller neonates had significantly higher mean (±SEM) plasma concentrations of total lipids than either of the heavier groups (758 ± 55 mg/dl versus 548 ± 49 and 498 ± 40 mg/dl) and higher mean concentrations of free glycerol (0.60 ± 0.17 mmol/liter versus 0.28 ± 0.03 and 0.24 ± 0.03 mmol/liter). At hourly infusion rates of 0.12 gm/dl, differences between groups did not reach statistical significance. Finally, at infusion rates of 0.16 gm/kg/hour, the smaller neonates had significantly higher mean concentrations of total lipids (1126 ± 150 mg/dl versus 610 ± 58 and 664 ± 76 mg/dl), free glycerol (1.25 ± 0.38 mmol/liter versus 0.44 ± 0.08 and 0.58 ± 0.14 mmol/ liter), triglycerides (1.86 ± 0.39 mmol/liter versus 1.03 ± 0.17 and 1.23 ± 0.18 mmol/liter), and free fatty acids (1.51 ± 0.23 mmol/liter versus 0.80 ± 0.17 and 0.77 ± 0.10 mmol/liter) than heavier neonates. There were no statistically significant differences between groups in mean cholesterol concentrations at any infusion rates. DISCUSSION
This report examines the relationship between birthweight and tolerance to parenteral fat emulsions. Shennan et al14 have shown that immature neonates with gestational ages of 27 to 32 weeks clear a bolus of fat emulsion half as rapidly as their more mature peers of 36 to 40 gestational weeks. They attributed this to a depression of heparin-induced lipoprotein lipase activity and to a decreased mass of adipose tissue. Most of the clearance of fat emulsion particles from the bloodstream occurs in muscle and fat. There is a correlation between the rate of clear-
Table 1.
Characteristics of the Study Groups
750-999 gm
1000-1249 gm
1250-1500 gm
4 2/2
7 3/4
7 7/0
4/0/0 930 ± 40 (820-995) 28 ± 0.2 (27-28) 4 ± 1.6 (2-9)
5/1/1 1150 ± 30 (1000-1240) 29 ± 0.4 (27-31) 4 ± 0.5 (2-6)
7/0/0 1420 ± 30 (1260-1500) 30 ± 0.8 (28-34) 4 ± 0.4 (2-5)
Characteristic
Number Sex: M/F Race: H/VV/B* Birthweight (gm)f Gestational age (weeks)"1" Postnatal age (days)1" at start of lipid infusion 116
ance of fat emulsions and the amount of fat in the body.14 Furthermore, since lipoprotein lipase activity before 27 weeks of gestation is even lower than between 27 and 32 weeks,19 the smaller low birthweight neonates should be at greater risk of hyperlipemia. Since the best estimates of gestational maturity are accurate only to ±2 weeks and metabolic maturity may sometimes lag behind neurologic and somatic maturity, great care should be taken on theoretical grounds when administering fat emulsions to neonates with birthweights less than 1000 gm or estimated gestational maturity of 27 to 29 weeks. Our data suggest that neonates weighing less than 1000 gm tend to have higher mean plasma concentration of total lipids, free glycerol, and free fatty acids than their heavier (or more mature) peers, especially while receiving fat emulsions at an hourly infusion rate of 0.16 gm/kg (or 4 gm/kg per day of continuous infusion). This would indicate that the tiny premature neonate experiences difficulties in clearing or metabolizing lipids, or both, and also in metabolizing glycerol used as an osmotic agent in the emulsion. These data are consistent with information available in the literature and are in agreement with the recommendation of the American Academy of Pediatrics to limit daily intake of parenteral fat emulsions to a maximum of 3 gm/kg, especially in the smaller neonate.20 They also emphasize the need to monitor the plasma of these tiny premature neonates very carefully for early signs of hyperlipemia. Since there are currently no convenient and accurate methods of detecting hyperlipemia and no consensus on what constitutes acceptable levels of plasma lipids in the neonate,16 this issue should be given a high priority for future research. This study is no more than a pilot study, the findings of which demand to be confirmed and expanded by careful investigations of larger numbers of subjects. It does, however, serve a purpose in identifying important methodological questions. First, it identifies differences and trends relating lipid tolerance to both hourly rate of infusion and birthweight or gestational maturity, or both. Second, it stresses the importance of stratifying groups of subjects by weight or maturity: 1000 gm or 28 weeks of gestation seem to mark an important turning point in lipid tolerance, but what then about the much smaller neonate (less than 750 gm or less than 27 weeks)? Occasionally, they do not even tolerate 0.04
*H: Hispanics; W: white Anglo-American; B: black. +Mean ± SEM (range).
Downloaded by: University of British Columbia. Copyrighted material.
AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 7, NUMBER 2 April 1990
EFFECT OF BIRTHWEIGHT O N LIPIDS/Brans et al.
REFERENCES
1. Gustafson A, Kjellmer I, Olegard R, et al: Nutrition in lowbirthweight infants. I. Intravenous injection of fat emulsions. Acta Paediatr Scand 61:149-158, 1972 2. Olegard R, Gustafson A, Kjellmer I, et al: Nutrition in lowbirthweight infants. III. Lipolysis and free fatty acid elimination after intravenous administration of fat emulsion. Acta Paediatr Scand 64:745-751, 1975 3. Andrew G, Chan G, Schiff D: Lipid metabolism in the neonate. I. The effect of Intralipid infusion on plasma triglyceride and free fatty acid concentrations in the neonate. J Pediatr 88:273-278, 1976 4. Reid WD, Robinson H: Atherosclerosis and childhood: Metabolic response to Intralipid infusions in the neonatal period. Adv Exp Med Biol 82:201-203, 1977 5. Griffin E, Breckenridge WC, Kuksis A: Appearance and characterization of lipoprotein X during continuous Intralipid infusions in the neonate. J Clin Invest 64:1703— 1712, 1979 6. Sun SC, Ventura C, Verasestakul S: Effect of Intralipidinduced lipaemia on the arterial oxygen tension in preterm infants. Resuscitation 6:265-270, 1978
7. Gustafson A, Kjellmer I, Olegard R, et al: Nutrition in lowbirth weight infants. II. Repeated intravenous injections of fat emulsions. Acta Paediatr Scand 63:177-182, 1974 8. Pereira GR, Fox WW, Stanley CA, et al: Decreased oxygenation and hyperlipemia during intravenous fat infusions in premature infants. Pediatrics 66:26-30, 1980 9. Hilliard JL, Shannon DL, Hunter MA, et al: Plasma lipid levels in preterm neonates receiving parenteral fat emulsions. Arch Dis Child 58:29-33, 1983 10. Brans YW: Erratum: Values for plasma glycerol, FFA and triglycerides. J Pediatr 105:855, 1984 11. Kao LC, Cheng MH, Warburton D: Triglycerides, free fatty acids, free fatty acids/albumin molar ratio, and cholesterol levels in serum of neonates receiving long-term lipid infusions: Controlled trials of continuous and intermittent regimens. J Pediatr 104:429-435, 1984 12. Cooke RJ, Burckhart GJ: Hypertriglyceridemia during the intravenous infusion of a safflower oil-based fat emulsion. J Pediatr 103:959-961, 1983 13. Berkow SE, Spear ML, Stahl GE, et al: Total parenteral nutrition with Intralipid in premature infants receiving IPN with heparin: Effect of plasma lipolytic enzymes, lipids, and glucose. J Pediatr Gastroenterol Nutr 6:581588, 1987 14. Shennan AT, Bryan MH, Angel A: The effect of gestational age on Intralipid tolerance in newborn infants. J Pediatr 91:134-137, 1977 15. Brans YW, Andrew DS, Carrillo DW, et al: Tolerance of fat emulsions in very-low-birthweight neonates. Am J Dis Child 142:145-152, 1988 16. Brans YW, Andrew DS, Carrillo DW, et al: Tolerance of fat emulsions in very-low-birthweight neonates: Monitoring of plasma lipid concentrations. Am J Perinatol 5:8-12, 1988 17. Dubowitz LMS, Dubowitz V, Goldberg C: Clinical assessment of gestational age in the newborn infant. J Pediatr 77:1-10, 1970 18. Freeman MG, Graves WL, Thompson RL: Indigent Negro and Caucasian birthweight-gestational age tables. Pediatrics 46:9-15, 1970 19. Dhanireddy R, Hamosh M, Sivasubramanian KN, et al: Postheparin lipolytic activity and Intralipid clearance in very low-birth-weight infants. J Pediatr 98:617—622, 1981 20. American Academy of Pediatrics: Use of intravenous fat emulsions in pediatric patients. Pediatrics 81:738-743, 1981
Downloaded by: University of British Columbia. Copyrighted material.
gm/kg/hour of fat. Neonates less than 1000 gm, down to the lower limit of viability clearly need to be focused on. Third, it provides data from which to calculate sample sizes needed to obtain clinically and statistically important differences. Estimated sample size will vary, depending on which variable is deemed most important and on how large a risk of type II error the investigator is willing to take. Fourth, it identifies some methodological problems. Studying very premature neonates, one must expect a minimum attrition rate due to unexpected events, and this estimated attrition rate must be built into the calculation of sample size. Our experience suggests an attrition rate of 10 to 20% should be expected. Finally, it sets the stage for studies of intervention designed to improve lipid tolerance.
117
April 1990
TOLERANCE OF FAT EMULSIONS IN VERY LOW BIRTHWEIGHT NEONATES EFFECT OF BIRTHWEIGHT ON PLASMA LIPID CONCENTRATIONS Yves W. Brans, M.D., Donna S. Andrew, M.S., Donna W. Carrillo, B.S.N., M.P.H., Elisabeth B. Dutton, B.S.N., Elizabeth M. Menchaca, B.A., and Belen A. Puleo-Scheppke, B.A., M.P.H.
Plasma concentrations of various lipid fractions (total lipids, free glycerol, true triglycerides, free fatty acids, and cholesterol) were studied in 20 normally grown neonates ranging in birthweightfrom820to1500gm and ingestational age from 28 to 34 weeks. They were subdivided into three weight categories: 750to999,1000to1249, and 1250 to 1500 gm. A lipid emulsion was infused at a constant rate over 24 hours, beginning with an hourly infusion rate of 0.04 gm/kg and increasing each day by 0.04 gm/kg up to a maximum of 0.16 gm/kg. Neonates who weighed less than 1000 gm had higher mean plasma concentrations of total lipids and free glycerol at hourly infusion rates of 0.08 and 0.16 and of triglycerides and free fatty acids at hourly infusion rate of 0.16 gm/kg than their heavier peers. These data suggest that extreme caution be used when administering parenteral fat emulsions to neonates who weigh less than 1000 gm and that we need to monitor plasma closely for signs of hyperlipemia.
Increases in plasma lipid concentrations have been documented in low birthweight neonates receiving various doses of fat emulsions parenterally.1"13 Available data suggest that preterm neonates may not clear fat emulsion particles from plasma as rapidly as older children or adults. Of particular concern are the adverse effects of reduced lipoprotein lipase and restricted adipose tissue mass on the clearance of fat particles in very low birthweight neonates.14 In this report we compare the effects of fat emulsions on plasma lipid patterns in three groups of increasingly lower birthweight neonates. The effect of fat emulsions on intake of fluids and calories and the effect of the various infusion regimens and of hourly rate of lipid infusion on plasma lipid patterns have been reported separately.15-16
MATERIALS AND METHODS
Tolerance to parenterally administered fat emulsions was studied in neonates who weighed 1500 gm or less at birth. Informed parental consent was obtained in all cases and the study was approved by the Institutional Review Board of the University of Texas Health Center at San Antonio. Birthweights were recorded to the nearest 10 gm. Gestational ages were determined from the mother's menstrual history, checked in most cases by sonographic determination of the biparietal diameter, and confirmed by physical examination of the neonate.17 All neonates had normal growth, that is, their birthweights were between the 10th and 90th percentiles for gestational maturity, sex, and race. Separate birthweightgestational age curves were used for Hispanic (Gibbs
Neonatal Research Laboratories, Departments of Pediatrics and Obstetrics and Gynecology, Wayne State University School of Medicine, Detroit, Michigan, and the Perinatal Research Laboratory, Departments of Pediatrics and Obstetrics and Gynecology, The University of Texas Health Science Center, San Antonio, Texas Supported in part by grant HD 15967 from the National Institute of Child Health and Human Development, Bethesda, Maryland, and by KabiVitrum, Alameda, California
114
Copyright © 1990 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 10016. All rights reserved.
Downloaded by: University of British Columbia. Copyrighted material.
ABSTRACT
EFFECT OF BIRTHWEIGHT ON LIPIDS/Brans et al.
recorded but the standard deviation was ignored. The mean values obtained for each infant while receiving a stated rate of lipid infusion were averaged, but the standard deviation within that infant was again ignored. The means analyzed were therefore the mean of the average values for individual neonates who received 0 (1 value), 0.04 (2 values), 0.08 (2 values), 0.12 (2 values), or 0.16 (8 values) gm/kg of lipids per hour (Fig. 1), and the standard error of the mean was calculated from the standard deviations across subjects. The sampling unit used (df) was each subject, not each measurement. In view of the small number of observations in some groups, the data were also analyzed nonparametrically by the Kruskal-Wallis test. Since parametric and nonparametric tests gave identical results, only the former are presented. RESULTS
Downloaded by: University of British Columbia. Copyrighted material.
CE: unpublished data), white Anglo-American,18 and black18 babies. Neonates weighing less than 750 gm at birth or with estimated gestational ages below 27 weeks were excluded from the study. Total parenteral nutrition was started on the third postnatal day with a basic regimen to provide daily 100 to 130 ml/kg of fluids (babies were cared for under radiant warmers), 2.5 gm/kg of crystalline amino acids, as much dextrose as tolerated, electrolytes, and vitamins. Individual intakes were tailored to each neonate's requirements and metabolic tolerance in order to provide maximal caloric intake. All decisions concerning nutrition, save one, were made by the attending staff. The one exception was the means and amount of administration of fat emulsion, which were specified by study protocol. The fat emulsion (Intralipid 20%, KabiVitrum, Alameda, CA) was provided at constant rates over 24 hours, beginning with an hourly infusion rate of 0.04 gm/ kg on the first day of infusion (1 gm/kg/day) and increasing each day by 0.04 gm/kg/hour up to a maximum of 0.16 gm/kg/hour (4 gm/kg/day). As long as the neonate required an arterial umbilical catheter for blood gas monitoring, parenteral fluids, fat emulsion excepted, were infused via this route. The tip of the catheter was always located radiographically below the bottom of the third lumbar vertebra. After removal of the catheter, peripheral veins were used. The fat emulsion was infused through a peripheral vein dedicated to that purpose. Enteral feedings were not provided during the study. The study was discontinued if the plasma was frankly creamy,5 if a serious medical problem interfered with the study (such as surgery or necrotizing enterocolitis), if a breach of protocol occurred, or if the parents requested it. The study period lasted 8 days. Blood samples were obtained through the umbilical catheter, after removal of 3 ml of blood to clear the line of infusion fluids or by heel-stick before beginning the fat infusion and every 12 hours thereafter. Plasma concentrations of total lipids, free glycerol, true triglycerides (that is, measured triglycerides minus free glycerol), free fatty acids, and cholesterol were determined in triplicates or quadruplicate, as previously described.16 Each variable was analyzed by one-way analysis of variance to detect differences between the three groups of birthweights at a given time during the study and, within each weight group, differences in mean plasma lipid values associated with increasing hourly rates of lipid infusion. When main effect differences were detected (F test, p =s 0.05), Duncan's multiple range test was applied to locate differences between means for various groups. A significance level equal to or less than 0.05 was chosen to define statistical significance. Although there was some variability in mean plasma concentrations of the various lipid fractions during infusion of fat at a given hourly rate, these differences were not statistically significant. Each specimen of plasma was analyzed in triplicate or quadruplicate: the mean value was
Twenty neonates entered the study. Two infants failed to complete the first four days of the study: one developed severe hyperlipemia (plasma was frankly creamy by inspection), requiring discontinuation of the lipid infusion, and the other one -I249g
I25O-I5OOJ
— 1500 f
1 S occ
Figure 1. Mean (± SEM) plasma concentrations of total lipids, free fatty acids, glycerol, triglycerides, and cholesterol before infusion and at each rate of lipid infusion. Each bar corresponds to different hourly rates of lipid infusion (from left to right, successively, 0, 0.04, 0.08, 0.12, and 0.16 gm/kg/day). Cross-hatched bars indicate mean values that are significantly higher than mean preinfusion value for that group. Symbols above the bars indicate statistically significant differences at a given time between the 750 to 999 gm and 1000 to 1249 gm groups (asterisk) or between the 750 to 999 gm and the 1250 to 1550 gm groups (dagger).
115
died; their data were excluded from analysis (attrition rate, 10%). Selected characteristics of the 18 remaining neonates are detailed in Table 1. Neonates in the three weight groups had similar mean preinfusion concentrations of the various lipid fractions (Fig. 1). Within weight groups, mean plasma lipid concentrations increased with the hourly rate of lipid infusion, although some differences did not reach statistical significance due to large standard errors of the mean and small number of subjects. At hourly infusion rates of 0.04 gm/kg, there were no statistically significant differences in plasma lipid patterns between the three weight categories. At infusion rates of 0.08 gm/kg/hour, the smaller neonates had significantly higher mean (±SEM) plasma concentrations of total lipids than either of the heavier groups (758 ± 55 mg/dl versus 548 ± 49 and 498 ± 40 mg/dl) and higher mean concentrations of free glycerol (0.60 ± 0.17 mmol/liter versus 0.28 ± 0.03 and 0.24 ± 0.03 mmol/liter). At hourly infusion rates of 0.12 gm/dl, differences between groups did not reach statistical significance. Finally, at infusion rates of 0.16 gm/kg/hour, the smaller neonates had significantly higher mean concentrations of total lipids (1126 ± 150 mg/dl versus 610 ± 58 and 664 ± 76 mg/dl), free glycerol (1.25 ± 0.38 mmol/liter versus 0.44 ± 0.08 and 0.58 ± 0.14 mmol/ liter), triglycerides (1.86 ± 0.39 mmol/liter versus 1.03 ± 0.17 and 1.23 ± 0.18 mmol/liter), and free fatty acids (1.51 ± 0.23 mmol/liter versus 0.80 ± 0.17 and 0.77 ± 0.10 mmol/liter) than heavier neonates. There were no statistically significant differences between groups in mean cholesterol concentrations at any infusion rates. DISCUSSION
This report examines the relationship between birthweight and tolerance to parenteral fat emulsions. Shennan et al14 have shown that immature neonates with gestational ages of 27 to 32 weeks clear a bolus of fat emulsion half as rapidly as their more mature peers of 36 to 40 gestational weeks. They attributed this to a depression of heparin-induced lipoprotein lipase activity and to a decreased mass of adipose tissue. Most of the clearance of fat emulsion particles from the bloodstream occurs in muscle and fat. There is a correlation between the rate of clear-
Table 1.
Characteristics of the Study Groups
750-999 gm
1000-1249 gm
1250-1500 gm
4 2/2
7 3/4
7 7/0
4/0/0 930 ± 40 (820-995) 28 ± 0.2 (27-28) 4 ± 1.6 (2-9)
5/1/1 1150 ± 30 (1000-1240) 29 ± 0.4 (27-31) 4 ± 0.5 (2-6)
7/0/0 1420 ± 30 (1260-1500) 30 ± 0.8 (28-34) 4 ± 0.4 (2-5)
Characteristic
Number Sex: M/F Race: H/VV/B* Birthweight (gm)f Gestational age (weeks)"1" Postnatal age (days)1" at start of lipid infusion 116
ance of fat emulsions and the amount of fat in the body.14 Furthermore, since lipoprotein lipase activity before 27 weeks of gestation is even lower than between 27 and 32 weeks,19 the smaller low birthweight neonates should be at greater risk of hyperlipemia. Since the best estimates of gestational maturity are accurate only to ±2 weeks and metabolic maturity may sometimes lag behind neurologic and somatic maturity, great care should be taken on theoretical grounds when administering fat emulsions to neonates with birthweights less than 1000 gm or estimated gestational maturity of 27 to 29 weeks. Our data suggest that neonates weighing less than 1000 gm tend to have higher mean plasma concentration of total lipids, free glycerol, and free fatty acids than their heavier (or more mature) peers, especially while receiving fat emulsions at an hourly infusion rate of 0.16 gm/kg (or 4 gm/kg per day of continuous infusion). This would indicate that the tiny premature neonate experiences difficulties in clearing or metabolizing lipids, or both, and also in metabolizing glycerol used as an osmotic agent in the emulsion. These data are consistent with information available in the literature and are in agreement with the recommendation of the American Academy of Pediatrics to limit daily intake of parenteral fat emulsions to a maximum of 3 gm/kg, especially in the smaller neonate.20 They also emphasize the need to monitor the plasma of these tiny premature neonates very carefully for early signs of hyperlipemia. Since there are currently no convenient and accurate methods of detecting hyperlipemia and no consensus on what constitutes acceptable levels of plasma lipids in the neonate,16 this issue should be given a high priority for future research. This study is no more than a pilot study, the findings of which demand to be confirmed and expanded by careful investigations of larger numbers of subjects. It does, however, serve a purpose in identifying important methodological questions. First, it identifies differences and trends relating lipid tolerance to both hourly rate of infusion and birthweight or gestational maturity, or both. Second, it stresses the importance of stratifying groups of subjects by weight or maturity: 1000 gm or 28 weeks of gestation seem to mark an important turning point in lipid tolerance, but what then about the much smaller neonate (less than 750 gm or less than 27 weeks)? Occasionally, they do not even tolerate 0.04
*H: Hispanics; W: white Anglo-American; B: black. +Mean ± SEM (range).
Downloaded by: University of British Columbia. Copyrighted material.
AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 7, NUMBER 2 April 1990
EFFECT OF BIRTHWEIGHT O N LIPIDS/Brans et al.
REFERENCES
1. Gustafson A, Kjellmer I, Olegard R, et al: Nutrition in lowbirthweight infants. I. Intravenous injection of fat emulsions. Acta Paediatr Scand 61:149-158, 1972 2. Olegard R, Gustafson A, Kjellmer I, et al: Nutrition in lowbirthweight infants. III. Lipolysis and free fatty acid elimination after intravenous administration of fat emulsion. Acta Paediatr Scand 64:745-751, 1975 3. Andrew G, Chan G, Schiff D: Lipid metabolism in the neonate. I. The effect of Intralipid infusion on plasma triglyceride and free fatty acid concentrations in the neonate. J Pediatr 88:273-278, 1976 4. Reid WD, Robinson H: Atherosclerosis and childhood: Metabolic response to Intralipid infusions in the neonatal period. Adv Exp Med Biol 82:201-203, 1977 5. Griffin E, Breckenridge WC, Kuksis A: Appearance and characterization of lipoprotein X during continuous Intralipid infusions in the neonate. J Clin Invest 64:1703— 1712, 1979 6. Sun SC, Ventura C, Verasestakul S: Effect of Intralipidinduced lipaemia on the arterial oxygen tension in preterm infants. Resuscitation 6:265-270, 1978
7. Gustafson A, Kjellmer I, Olegard R, et al: Nutrition in lowbirth weight infants. II. Repeated intravenous injections of fat emulsions. Acta Paediatr Scand 63:177-182, 1974 8. Pereira GR, Fox WW, Stanley CA, et al: Decreased oxygenation and hyperlipemia during intravenous fat infusions in premature infants. Pediatrics 66:26-30, 1980 9. Hilliard JL, Shannon DL, Hunter MA, et al: Plasma lipid levels in preterm neonates receiving parenteral fat emulsions. Arch Dis Child 58:29-33, 1983 10. Brans YW: Erratum: Values for plasma glycerol, FFA and triglycerides. J Pediatr 105:855, 1984 11. Kao LC, Cheng MH, Warburton D: Triglycerides, free fatty acids, free fatty acids/albumin molar ratio, and cholesterol levels in serum of neonates receiving long-term lipid infusions: Controlled trials of continuous and intermittent regimens. J Pediatr 104:429-435, 1984 12. Cooke RJ, Burckhart GJ: Hypertriglyceridemia during the intravenous infusion of a safflower oil-based fat emulsion. J Pediatr 103:959-961, 1983 13. Berkow SE, Spear ML, Stahl GE, et al: Total parenteral nutrition with Intralipid in premature infants receiving IPN with heparin: Effect of plasma lipolytic enzymes, lipids, and glucose. J Pediatr Gastroenterol Nutr 6:581588, 1987 14. Shennan AT, Bryan MH, Angel A: The effect of gestational age on Intralipid tolerance in newborn infants. J Pediatr 91:134-137, 1977 15. Brans YW, Andrew DS, Carrillo DW, et al: Tolerance of fat emulsions in very-low-birthweight neonates. Am J Dis Child 142:145-152, 1988 16. Brans YW, Andrew DS, Carrillo DW, et al: Tolerance of fat emulsions in very-low-birthweight neonates: Monitoring of plasma lipid concentrations. Am J Perinatol 5:8-12, 1988 17. Dubowitz LMS, Dubowitz V, Goldberg C: Clinical assessment of gestational age in the newborn infant. J Pediatr 77:1-10, 1970 18. Freeman MG, Graves WL, Thompson RL: Indigent Negro and Caucasian birthweight-gestational age tables. Pediatrics 46:9-15, 1970 19. Dhanireddy R, Hamosh M, Sivasubramanian KN, et al: Postheparin lipolytic activity and Intralipid clearance in very low-birth-weight infants. J Pediatr 98:617—622, 1981 20. American Academy of Pediatrics: Use of intravenous fat emulsions in pediatric patients. Pediatrics 81:738-743, 1981
Downloaded by: University of British Columbia. Copyrighted material.
gm/kg/hour of fat. Neonates less than 1000 gm, down to the lower limit of viability clearly need to be focused on. Third, it provides data from which to calculate sample sizes needed to obtain clinically and statistically important differences. Estimated sample size will vary, depending on which variable is deemed most important and on how large a risk of type II error the investigator is willing to take. Fourth, it identifies some methodological problems. Studying very premature neonates, one must expect a minimum attrition rate due to unexpected events, and this estimated attrition rate must be built into the calculation of sample size. Our experience suggests an attrition rate of 10 to 20% should be expected. Finally, it sets the stage for studies of intervention designed to improve lipid tolerance.
117
