



* Division of Neonatology, Department of Pediatrics
Department of Nutrition Services
Institute of Metabolic Disease, Baylor University Medical Center, Dallas, Texas
| ABSTRACT |
|---|
|
|
|---|
Methods. Eighty preterm infants <1500 g were enrolled in a prospective, double-blind, placebo-controlled study of carnitine supplementation within 96 hours of delivery. Growth, length of hospital stay, and frequency and severity of apnea were the primary outcome measures.
Results. Weight gain and change in length, fronto-occipital head circumference, mid arm circumference, and triceps skinfold thickness were similar between the carnitine-supplemented and placebo groups. The amount and severity of apnea and the overall length of hospitalization were also similar between the 2 groups. The carnitine levels in the supplemented group were significantly higher than in the placebo group at 4 and 8 weeks after study entry.
Conclusion. Although preterm infants <1500 g have low carnitine levels, routine supplementation with carnitine has no demonstrable effect on growth, apnea, or length of hospitalization and thus seems to be unnecessary.
Key Words: carnitine supplementation very low birth weight infants apnea growth
Abbreviations: VLBW, very low birth weight LOS, length of stay NICU, neonatal intensive care unit PCG, pneumocardiogram
| INTRODUCTION |
|---|
|
|
|---|
The fetus probably obtains exogenous carnitine via placental transport.6,7 Carnitine is stored in fetal tissue in increasing amounts over the latter part of gestation, and tissue carnitine stores at birth are directly related to gestational age.8,9 Endogenous synthesis of carnitine is limited in the neonate by low levels of
-butyrobetaine hydroxylase, the enzyme that catalyzes the final step of the carnitine synthetic pathway. Without an exogenous supply of carnitine, preterm infants cannot achieve carnitine homeostasis.
Very low birth weight (VLBW) infants are at high risk of carnitine deficiency when receiving parenteral nutrition or unsupplemented infant formulas, such as soy-based preparations.10 Carnitine can be viewed as a conditionally essential nutrient in the preterm infant, primarily because of immaturity of biosynthetic capability and the dependence on exogenous sources. Enteral infant formulas have carnitine added and provide 2 to 5 mg/kg/d when on full enteral feeds. This is similar to the amounts contained in breast milk.11 Infants receiving parenteral nutrition have no exogenous source of carnitine.
Assessing an infants carnitine status is complicated. Blood levels alone are not indicative of the total metabolic pools of carnitine, eg, those found in red cells, muscles, and liver stores. Full-term neonates experience a decrease in plasma carnitine concentrations in the first 2 weeks of life. Without enteral feeds and carnitine supplementation of parenteral nutrition, preterm infants plasma carnitine levels fall to one third of the value found in full-term infants at birth.12,13
Previous studies of carnitine supplementation have shown varying results, with some studies suggesting benefit and others showing no benefit. One study presented in abstract form demonstrated a beneficial effect on preterm infants with apnea. That study showed less apnea, a shorter mechanical ventilator course, and better growth in the carnitine-supplemented group.14 However, lack of consistent dosing, lack of consistent caloric and protein intake, small sample size, and lack of consistent measures of growth make the studies difficult to interpret.1519 Most studies of carnitine supplementation have been short term and have focused on intravenous lipid tolerance.12,2026
This study was designed to test the hypothesis that carnitine supplementation within 96 hours of delivery in VLBW infants has a beneficial effect on apnea, growth, and length of hospitalization.
| PATIENTS AND METHODS |
|---|
|
|
|---|
of <0.05. The following were assumptions about the primary outcome variables: 1) apnea: to decrease the incidence at 32 weeks gestational age for significant apneic episodes (>20 seconds) from 70% to 30% required a total of 64 patients; 2) length of stay (LOS): to demonstrate a 7-day reduction in LOS required a total of 60 infants; and 3) weight gain: to show an increase in weight gain by 10 g per day required a total of 32 patients. Neonates <1500 g were enrolled in the study between June 1998 and May 2000. Infants with congenital malformations, chromosome abnormalities, inborn errors of metabolism, end-stage liver disease, peritoneal dialysis, or severe intraventricular hemorrhage (grade IV) were excluded from the study. All patients were enrolled at Baylor University Medical Center in Dallas, which has a level 3 neonatal intensive care unit (NICU). The protocol underwent institutional review board approval, and parental informed consent was obtained by the investigators for each subject.
Participants were recruited within 96 hours of birth. The study was masked to the clinical caretakers and investigators. The research pharmacist conducted the randomization using an Excel spreadsheet (Microsoft Corp, Redmond, WA) with randomization into blocks of 10 patients.
Laboratory tests included prestudy plasma free and total carnitine levels and subsequent levels every 4 weeks until discharge from the NICU or 36 weeks postconceptional age. The assay used was stable isotope dilution tandem mass spectrometry of butyl ester derivatives by either fast atom bombardment or electrospray ionization, modified from Kodo et al27 and Rashed et al.28 These laboratory results were not available to the clinical care team. In addition, acyl carnitine profiles in dried blood spots and urine organic acids were analyzed at enrollment to rule out a possible inherited metabolic disorder (none detected).
Patient characteristics were recorded, including birth weight, route of delivery, Apgar scores, sex, gestational age as determined by attending neonatologist, length, fronto-occipital head circumference, mid arm circumference, and triceps skinfold thickness. Weekly measurements of weight, fronto-occipital head circumference, and length were recorded from the bedside NICU flowsheets completed by the registered nurse. The mid arm circumference and triceps skinfold thickness were measured weekly by the research dietitian. A large skinfold caliper was used, and measurements were rounded to the nearest 1 mm. An admission severity of illness score (Score for Neonatal Acute Physiology) was calculated on each infant at 24 hours of life using vital signs and other data from the NICU flowsheet and chart.29
Carnitine Supplementation
Infants were randomly assigned to receive intravenous infusions of L-carnitine (15 mg/kg/d) or no supplementation. Infusions were initiated within 96 hours of birth. When full enteral feeds were established (120150 mL/kg/d), 100 mg/kg/d of L-carnitine was given orally in 4 divided doses. The control infants received an osmotically matched placebo prepared in an identical manner. The placebo was supplied by the research pharmacist and given in 4 doses each day. The bedside nurse did not know whether carnitine or placebo was given.
The carnitine was supplied by Sigma-Tau Pharmaceuticals, Gaithersburg, Maryland.
Nutritional Regimens
Infants were given maintenance fluids initially, usually with 10% dextrose, and parenteral nutrition was started within 72 hours. Fluids and the caloric content of parenteral nutrition and feeds were ordered according to the preference of each attending neonatologist. There was no set protocol.
Evaluation for Apnea
Apnea was evaluated in 4 ways:
34 weeks gestational age.
Statistical Analysis
Categorical variables were analyzed using the Fisher 2-sided exact test for simple 2 x 2 tables. A likelihood ratio was used for larger tables. Continuous variables were analyzed with the Wilcoxon 2-sample test.
| RESULTS |
|---|
|
|
|---|
|
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
The supplemented group had levels that were 4 to 5 times higher than that of the control group. Despite these high levels, no growth advantage was observed in these infants. In 1 study, carnitine levels of almost 200 µmol/L after supplementing 48 mg/kg/d in a group of preterm infants on parenteral nutrition were associated with increased nitrogen excretion, higher metabolic rate, and lower weight gain.21 No other studies have reported adverse effects of carnitine supplementation.
Our study did not demonstrate any beneficial effect on growth as measured by weight, fronto-occipital head circumference, length, mid arm circumference, or triceps skinfold thickness. Cairns and Stalker34 recently reviewed all published articles of carnitine supplementation in parenterally fed neonates, with the primary outcome measure being weight gain. Fourteen trials were found, but 8 were excluded because of lack of randomization. Of the 6 studies in which randomization was considered appropriate, only 1 showed a significant effect in the second week of supplementation, and this effect was noted only in infants of 1001 to 1500 g.18 This beneficial effect on growth was not sustained. None of the other studies showed any effect on weight gain.
Apnea of prematurity is one of the most common disorders of the VLBW infant, affecting >50% of VLBW infants.35,36 Methylxanthines (theophylline and caffeine) are the mainstay of pharmacotherapy for idiopathic apnea of premature infants. All 3 types of apnea, ie, mixed, obstructive, and central, respond to this therapy.37 Despite this treatment, many infants continue to have episodes of apnea with associated bradycardia and hypoxia that are believed to be clinically significant with potentially harmful physiologic changes. The infant who fails to respond adequately to methylxanthines is a challenge for the physician. The use of continuous positive airway pressure may eliminate the need for intubation in many infants, but occasionally intubation and mechanical ventilation are required.
A study by Iafolla and Browning14 reported decreased apnea and the need for mechanical ventilation in carnitine-supplemented infants <33 weeks gestation. The study also demonstrated improved growth in supplemented infants compared with control infants (weight gain of 14 g/kg/d vs 21 g/kg/d). The authors speculated that preterm infants have many of the symptoms of carnitine deficiency, such as apnea, hypotonia, poor weight gain, delayed development, and gastrointestinal reflux. Supplementation with carnitine was associated with decreased apnea and a decreased need for mechanical ventilation. These observations were made despite the concurrent use of xanthines in all of the infants studied. The carnitine-supplemented infants were weaned more quickly from xanthines than the control infants. The caregivers in that study were not blinded to the carnitine supplementation. Based on these observations, we hypothesized that if carnitine does have such a salutatory effect on apnea of prematurity, the effect would be seen despite concurrent use of mechanical ventilation and xanthines. For this reason, no attempt to control the use of xanthines was made. In general, our practice at the time of this study was to use xanthines to facilitate extubation and to trial weaning off xanthines by 33 to 34 weeks gestation.
A weakness of this study design was lack of control over xanthine therapy. Our prestudy power analysis assumed a 70% incidence of apnea. In fact, attributable most likely to xanthine administration the incidence was only 28% (Table 3). Much larger numbers and a very profound effect (28% reduced to 0%) would have been required to show a beneficial effect of carnitine. PCG recordings were not available on all infants. Lack of PCG machines or nonavailability of technicians at time of discharge were the principal reasons for lack of recordings. This occurred in random fashion and affected both placebo and carnitine-supplemented groups equally. Given these factors, we were unable to demonstrate any beneficial clinical effect of carnitine on apnea in our study. The study design does not allow us to conclude that there is no effect. More subtle effects of carnitine on apnea may have been masked by the concurrent use of xanthines. The total number of clinical episodes and of PCG apneic episodes >20 seconds as well as the percentage of periodic breathing in both groups were similar. Likewise, the length of mechanical ventilation was similar in the supplemented and placebo group. This latter observation was dramatically different in the Iafolla study, in which all the supplemented patients were extubated significantly earlier than the placebo patients.
| CONCLUSION |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
We thank the dedicated nursing and respiratory therapy staffs of the NICU and the metabolic technicians in the mass spectrometry laboratory of the Institute of Metabolic Disease at Baylor University Medical Center. Also, thanks to Linda Jennings and David Nicewander for their expertise with study design and statistical analysis.
| FOOTNOTES |
|---|
Reprint requests to (J.W.) Division of Neonatology, Department of Pediatrics, Baylor University Medical Center, 3500 Gaston Ave, Dallas, TX 75246. Email: jonathaw{at}baylorhealth.edu
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. M. Crill and R. A. Helms The Use of Carnitine in Pediatric Nutrition Nutr Clin Pract, April 1, 2007; 22(2): 204 - 213. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. J. Carlson Current Nutrition Management of Infants With Chronic Lung Disease Nutr Clin Pract, December 1, 2004; 19(6): 581 - 586. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||