Date:
Dec 01, 2004

It is becoming increasingly recognized that either maternal or fetal infection is a major element in the cause of Cerebral Palsy. This process may also affect the premature newborn as demonstrated by a recent study performed by Stoll and her colleagues at the NIH National Institute of Child Health and Human Development.1

The authors identified 6093 children born between 1993 and 2001 included in the data base of the Institute. These children weighed less than 1000 gm (a bit over 2 pounds), survived, and were available for study when they were between 18 and 22 months corrected gestational age. Children who had shunts placed in the ventricles of their brains to drain fluid or had major malformations of the brain were excluded.

Nearly two of three children in their study acquired some type of infection during their newborn period. Over 40% of those with some sort of infection developed brain injury resulting in cerebral palsy, cognitive impairments, or both. For cerebral palsy alone, 16% of children who acquired any kind of infection in the post natal period developed cerebral palsy while only 8% of those without infection developed cerebral palsy. Children with infection in their blood (sepsis), infection associated bowel disease, and meningitis were especially at risk. The forms of Cerebral Palsy were not specified in this study.

Comment:

This study is most interesting taken in the context of the present attention to the role of maternal and placental infection and inflammation on the occurrence of Cerebral Palsy. It shows that newborn children with acquired infections, not caused by infection of the mother or of the placenta, are also prone to develop Cerebral Palsy and cognitive defects. It is worth noting that the infection need not affect the brain directly. The chemicals (cytokines) released by the infant to fight infection may be the source of the injury. Adult neurologists may see an analogy between the white matter injury seen in these infants and those seen in multiple sclerosis where it is believed that the inflammatory process begins outside of the brain and is then carried into it in a number of possible ways.

1Stoll, B.J., et al. Neurodevelopmental and growth impairment among extremely low-birth-weight infants with neonatal infection. JAMA, Nov 17 2004; 292 (19):2357-2401

© UCP Research & Educational Foundation, December 2004

Ten years ago, the UCP Research and Educational Foundation funded a pilot research project to study the use of a drug, magnesium sulfate, for the prevention of cerebral palsy in premature infants. As reported in the Foundation’s Research Fact Sheet of February 1995, the results of the pilot study did not show that the drug was effective. However, because it was a pilot study one of whose purposes was to test the research protocol in a limited number of women, the negative result of the intervention was not the final word.

Using the experience of the pilot study, a larger, longer term study of the drug was subsequently initiated in Australia by Dr. Caroline Crowther of the University Of Adelaide, Australia, and her colleagues. They are publishing preliminary information which suggests that magnesium sulfate given intravenously to women about to deliver extremely premature babies, helped reduce brain damage including cerebral palsy, and death in those infants. The study included 1,062 women about to deliver very premature infants in 16 hospitals in Australia and New Zealand. Surviving infants were evaluated at two years of age.

Children of women given the drug and who were born 2 months early (at 30 weeks) had 17 percent less risk of death and cerebral palsy, than women who were given placebo. The protective mechanism is not clear, but it appears the drug may help prevent bleeding in the premature infant brain—a common cause of cerebral palsy. There were no serious harmful effects of the treatment on either the mothers or children. This is a preliminary report to be published in December 2003 in the Journal of the American Medical Association (JAMA).

Comment:

The study’s results are encouraging. The scientists caution that “widespread use of prenatal magnesium sulfate cannot be recommended solely on the basis of the present study.” The review of additional data from this study and from another long term study being done in the USA will be needed before a final conclusion can be reached.

In addition to the value of the information provided, these recent results also point out the danger of jumping to a conclusion from small, short term studies in which chance alone can influence the significance of the results. Clearly, the prior UCP pilot study showing no benefit was not the last word. These skilled investigators wisely advise caution about the conclusions to be drawn at the time from even this larger study. In the reasonably near future, when all the data are available, we should have the full information needed to know if magnesium sulfate given to women in premature labor reduces the occurrence of developmental brain damage in their premature babies.

© UCP Research & Educational Foundation, December 2003

Research continues for developing effective ways of protecting the brain of the newborn infant which has been threatened by a lack of blood and/or oxygen during or shortly after the delivery process (perinatal hypoxia-ischemic brain injury). The immediate insult to the brain is often followed by several hours or a few days of continuing damage. To date, there is no clinically effective treatment to stop the progression of the brain injury; however several experimental treatments are under study including protective drugs and brain cooling. The principal mechanism for minimizing the damage to the threatened brain is to reduce the activity of the injured cells (decrease their metabolic demand). It has been shown in animal and human studies that this can be achieved by lowering the temperature of the brain (hypothermia). Unfortunately, in the human studies this physiologic approach has not been clinically successful because of the undesirable side effects brought on by total body cooling. Why the side effects? Is it the technique that is used?

Dr. J.R. Tooling and his colleagues are exploring this question.¹ Using an animal model of new born piglets exposed to a severe reduction of oxygen for 45 minutes, the investigators considered the effectiveness of using differential cooling in which the brain was profoundly cooled using a head cap, while the remainder of the body was only mildly cooled. Also, the animals were under anesthesia to reduce both total body stress (e.g. cardiac stress) and shivering. The logic is that body stress was increased in reacting to total body cooling and being alert; this indirectly influenced the metabolic activity of threatened brain cells.

The results of his revised method of cooling were “twenty four hours of selected head cooling combined with mild total body cooling during anesthesia is highly protective in the global perinatal hypoxic-ischemic brain injury piglet model.”

Comment:

In an experimental, carefully controlled animal model of lower oxygen availability to the newborn animal’s brain, the immediate use of brain cooling was successful in significantly protecting the threatened brain. The control of total body stress by the use of anesthesia and differential cooling of the brain and body appears to control the factors that defeated the clinical usefulness of brain cooling in the past. If additional studies using animal models demonstrate the same protective effect with no undesirable complications, the stage will be set to try this approach in human infants. Hopefully, this modification in technique will finally provide a useful clinical-intervention to protect the brain of the newborn threatened by a perinatal episode of hypoxia-ischemia.

¹ Tooley, J.R. et. al. Head Cooling with Mild Systemic Hypothermia in Anesthetized Piglets is Neuroprotective. Annuals Neurol 2003: 65-72

UCP research & Educational Foundation, February 2003

Prematurity is an important risk factor for the subsequent development of cerebral palsy. In the premature infant, there are two principle mechanisms responsible for developmental brain damage: hemorrhage into the ventricles (cavities) in the brain; and lack of blood and/or oxygen to a vulnerable part of the developing brain. This latter mechanism is the more common and leads to poor formation of the brain’s “white matter”: the insulating material that surrounds the brain nerve cell processes. This white matter insufficiency disturbance is called PVL, periventricular (around the ventricles) leukomalacia (deficiency of white matter).

Another disturbance associated with prematurity is the respiratory distress syndrome (RDS). It occurs because the premature infant’s lungs do not function properly; this is particularly true in premature infants of less than 31 weeks of gestational (intra-uterine) age. Fortunately, the use of hormone therapy (glucocorticoids) given to the mother prior to delivery often can prevent RDS.

There have been preliminary observations that the occurrence of PVL is also decreased when women in premature labor have been given hormones to prevent RDS in their infants. Do they really work in preventing PVL and, if so, which of the available hormones works better?

To answer these questions, a team of investigators in several hospitals in France did a careful analysis of the records of 883 live born infants.¹ They found there was a substantial reduction of the occurrence of PVL in very premature infants whose mothers had received one of the two hormones commonly used prior to delivery to prevent RDS: betamethasone.

Comment

This is a very important clinical finding to assist in the prevention of cerebral palsy in infants at high risk: the very premature infant. Of the two glucocorticoid hormones now often given to the mother in pre-term labor to prevent respiratory distress syndrome, one was found to be important also for the prevention of cerebral palsy: bethamethasone. This finding adds to our ability to address the problem of preventing brain damage in the very premature infant.

¹Baud, O. et al. Antenatal Glucocorticoid Treatment and Cystic Periventricular Leukomalacia in Very Premature Infants. NEJM 1999; 341: 1190-1196

© UCP Research & Education Foundation, September 2000

Pre-term birth usually associated with low birth weight of the infant is one of the most important risk factors associated with cerebral palsy. Although pre-term birth accounts for only 12% of all live births, it is associated with 40% of infants with cerebral palsy. Thus, the causes of pre-term birth and steps to prevent it are important subjects of cerebral palsy research.

Bacterial infection of the pregnant woman’s vagina (bacterial vaginosis), often with no maternal symptoms of illness, affects about 800,000 pregnant women per year in the U.S.A. Women with bacterial vaginosis are more likely than other women to have a pre-term delivery (less than 37 weeks) or a low birth weight infant (under 5.5 lbs). If treatment of bacterial vaginosis was to reduce the occurrence of pre-term delivery, the risk of infants born with cerebral palsy would be significantly lessened.

A group of investigators evaluated the usefulness of antibiotic treatment of pregnant women who had bacterial vaginosis and a prior history of pre-term delivery.¹The investigators found no benefit of this treatment in either reducing the occurrence of premature delivery or of preventing injury to the infant.

Comment:

Needless to say, the results of this study were a great disappointment. An editorial in the same journal evaluated the reasons for what might appear to have been the reason (s) for the failure. Was the population studied similar to the general population of the USA? Probably not; the population was a very high-risk population. Was the antibiotic given too late in pregnancy? Probably. Should both intravaginal antibiotics and oral antibiotics have been used? Probably. As often happens, failure in a study leads to a sharpening of the methodology that needs to be used and generally leads to a second, more precise study. This may happen.

For the time being, the Public Health Service’s recommendation continues to stand: women who have symptomatic bacterial vaginosis during pregnancy should be evaluated, the specific organism identified and specific therapy used. The research issues are what to do about women who have no symptoms? Are there special groups of infected women who have no symptoms but are at high risk of premature delivery? How are they to be identified? When should they be treated; probably the earlier the better. How should they be treated?

The present unsuccessful study gives us information on how to better answer these questions. We hope the next study will be developed and these questions answered.

¹ Carey, J. Christopher et al. Metronidazale to Prevent Preterm Delivery in Pregnant Women With Asymptomatic Bacterial Vaginosis. NEJM 2000; 8: 534-540

© UCP Research & Educational Foundation, June 2000

In the first trimester (3 mos) of pregnancy, the fetus is completely dependent upon the mother for thyroid hormone. During the second and final trimesters of pregnancy, most fetuses are able to provide some thyroid hormone but also continue to rely upon their mothers for some hormone. Problems of brain development occur when either the mother is unable to provide for the fetus’ needs in the first trimester, or neither the mother nor the fetus can provide for fetal needs during the remainder of pregnancy. Another important issue is the availability of iodine to the mother since iodine is an important requirement to provide for proper thyroid production by both the mother and the developing fetus.

Should all pregnant women routinely be evaluated for thyroid function early in pregnancy? The thyroid status of a large group of pregnant women and their offsprings were evaluated by the authors.1 In those women with low levels of thyroid hormone, the authors show that hypothyroidism (low thyroid function) adversely affects the child’s performance on tests of intelligence, language, reading ability and motor performance. This can be true even when the mother is presenting no symptoms. As a result of their study, the authors suggest that an evaluation of the mother’s thyroid function should become routine early in pregnancy for all women.

In an editorial in the same journal (pages 601-602), Dr. Robert D. Utiger raises questions about the cost/benefit effects of routine examination of thyroid function in all pregnant women. He suggests instead that since iodine is essential for thyroid hormone production, an adequate supply of iodine needs to be made available to all pregnant women. This is particularly true in that a common source of iodine in the U.S. diet is iodized table salt. With efforts to decrease the use of table salt in order to prevent high blood pressure, the consumption of iodine may no longer be adequate — particularly for pregnant women. He suggests iodine needs to be added to other foods and be included in all vitamin products as the first step to prevent developmental brain damage due to maternal hypothyroidism. Thus, he proposes that before a national program of routine thyroid function in pregnant women is initiated, an increased supply of iodine be added to vitamin products used by pregnant women.

Comment:

It has been recognized for a very long time that an adequate supply of thyroid hormone is essential for normal brain development. This is true during the entire period of pregnancy but it is particularly true in the last half of pregnancy and in the early years of infant life. Modest thyroid hormone insufficiency leads to cognitive inadequacies (e.g. learning disabilities); a large insufficiency leads also to neurological deficits — spasticity, poor balance and deafness. In women at risk of a troubled pregnancy, a thorough thyroid function evaluation early in pregnancy would certainly seem prudent. Also, an available source of iodine within the diet or by dietary supplements (e.g. iodine fortified vitamins) should be routine. In addition, attention needs to be given to the thyroid hormone status of the newborn infant; and in selected cases to the infant several weeks after birth to make certain the infant’s thyroid gland continues to function properly.

There are no longer any questions about the absolute requirement of thyroid hormone for adequate brain development. Both pregnant women and their physicians need to recognize this on the “check list” of things to consider to foster normal development of the developing brain.

In summary, a controversy continues to exist about whether all pregnant women should have their thyroid function routinely evaluated early in pregnancy; or, should only women who are “at risk” of poor thyroid function be evaluated? In any case, most experts agree that routinely an iodine supplement is a good idea.

1Haddow J.E. et al. Maternal Thyroid Deficiency During Pregnancy and Subsequent Neuropsychological Development of the Child NEJM 1999; 341:549-555

© UCP Research & Educational Foundation, February 2000

Date:
Sep 01, 1999

A large proportion of full term or nearly full term infants who have oxygen supply difficulties at the time of birth (perinatal hypoxia) demonstrate no indication of developmental brain damage. However a small proportion of them will subsequently demonstrate brain damage (e.g. cerebral palsy). At the time of birth or shortly thereafter, there are no reliable methods for indicating which hypoxic newborns are in danger of developing brain damage. It is important to be able to identify newborns at risk of brain damage as soon as possible if interventions are to be used early enough to safeguard the threatened developing brain.

A recent research paper1 describes the results of a chemical analysis of the newborn’s urine that had oxygen insufficiency during the birthing period. Studies of urine taken within 6 hours of birth were very indicative of the danger of the infant developing brain damage. The urine was tested for two products of energy metabolism, lactate and creatine.

Comment:

About 25% of full term infants who demonstrated signs of fetal distress during birthing (the perinatal period) subsequently develop developmental brain damage; the other 75% do not. How to tell them apart so as to take advantage of the window of opportunity in which an intervention (e.g. cooling) might protect the threatened brain? Most clinical evaluations and laboratory tests presently available have been poor predictors. The report summarized above would seem to be a valuable tool for helping to make that prediction.

However, having identified the newborn at risk of developmental brain damage, the remaining issue is how to protect the infants brain–neuroprotection. The methods presently available are still far from satisfactory. A number of experimental drugs have not worked out; head cooling is again being evaluated; increased oxygen is also again being considered but its potential usefulness is clouded by its sometimes disastrous side effects. Entirely new approaches to the repair of the brain injury are being explored and their very early use may prove effective.

Despite the above, being able to identify the newborn at risk is an important step forward. The success of this laboratory evaluation will undoubtedly lead to even more accurate means of making this assessment. It also opens up possibilities to far better understand what is happening in the threatened brain and thus what might be done to protect it.

1 Huang, C-C et al. Measurement of the Urinary Lactate-Creatinine Ratio for the Early Identification of Newborn/Infants at Risk for Hypoxic-Ischemic Encephalapathy. NEJM 1999; 341:328-335
© UCP Research & Educational Foundation, September 1999

In June 1996, the Little Foundation (the British counterpart of our country’s UCP Research and Educational Foundation) sponsored a research workshop in London, England. Research findings on a variety of subjects relevant to cerebral palsy were presented and discussed. Among them were “hot issues” on the prevention of cerebral palsy. One of the discussants was Alan Leviton, MD, Director of Neuro-epidemiology at the Children’s Hospital in Boston, and a research consultant to our Foundation. Dr. Leviton’s overview was so useful we have translated it into non-technical terms to share with you.

The following research activities were selected by Dr. Leviton from scientific literature published in 1995 and 1996 and presented at the research workshop:

Infection

The hypothesis that maternal infection contributes to the occurrence of cerebral palsy is one that is now receiving a great deal of attention. For infants born at term (around 9 months) and whose mothers had infections at admission to the hospital, there was a three fold increased risk of cerebral palsy when compared to mothers who did not have an infection. For infants born either prematurely or who had very low birth weight (3.3 lbs or less) and whose mothers had an infection at admission to the hospital, there was a 2.3 fold increase of risk of cerebral palsy when compared to mothers who did not have an infection. In other studies it has been demonstrated that 21% of the mothers whose infants developed a brain lesion associated with cerebral palsy, had a fever when admitted to the hospital for delivery; only 5% of mothers who did not have a fever delivered babies with similar brain lesions. Thus, evidence of maternal infection at the time of delivery is an important risk factor for the occurrence of cerebral palsy in the infant. The infection may or may not be producing clinical signs of illness in the mother.

Another set of findings of potential importance: infection of the vagina or uterine cavity before the midpoint of pregnancy is predictive of premature labor and of increased risk of fetal brain damage. Thus, evidence of infection early in pregnancy is an important risk factor for both premature delivery and for fetal brain damage. Finally, how best to diagnose and treat the suspect infections?

Fetal Distress

Over one hundred years ago, the hypothesis was presented that problems in labor and delivery contribute to the occurrence of cerebral palsy. One aspect of this is the hypothesis that oxygen lack during and persisting after a uterine contraction not only damages the brain, but also results in persistent slowing of the fetal heart. These appear to be true. What remains unclear is the issue of cause and effect. Does the decrease in fetal heart rate increase the risk of cerebral palsy or are the increased risk of cerebral palsy and of decreased fetal heart rate both due to a common cause? However, if the decrease fetal heart rate is a risk factor for cerebral palsy and if accelerated delivery of the infant would eliminate this risk factor from having an effect, more than 9,000 infants would have to be delivered by cesarean section to prevent one case of cerebral palsy!

Pregnancy Induced Hypertension

Infants are usually born before the seventh month of pregnancy for one of three reasons: 1) premature labor (infection?) 2) premature rupture of the membranes; or 3) intervention to protect the mother because of pregnancy-induced hypertension (high blood pressure). Studies indicate that the presence of pregnancy-induced hypertension in the mother decreases the probability of cerebral palsy in infants born prematurely or with very low birth weight. Why? Does the treatment given to protect the mother also protect the infant’s brain?

Magnesium Sulfate

Please refer to the Research Fact Sheet of February 1995. A retrospective study of records indicates that women who are given a common drug, magnesium sulfate, to suppress premature labor are less at risk of having an infant with cerebral palsy than those who did not get this drug. Does this drug protect the infant’s brain during labor and delivery? Our Foundation is presently supporting a research project (clinical trial) to obtain the answer to this question.

Low Blood Thyroid Level

Please refer to the Research Fact Sheet of April 1996. The lower the age of the newborn infant, the lower its blood thyroid hormone level during the first week of postnatal life. There are several possible reasons for this and they are being studied. However, we now find that severe levels of thyroid hormone deficiency is associated with a four-fold increase in the risk of cerebral palsy in very low birth weight infants. Would the administration of thyroid hormone to these infants decrease the risk of cerebral palsy? What would be the danger of giving thyroid hormone to these infants? Our Foundation is presently exploring the development of a research project to answer these questions.

Comment

Although not an exhaustive review of research in cerebral palsy, Dr. Leviton’s excellent review has identified a number of the “areas of research excitement”. He is planning to publish more details about these in a scientific journal during the next year. We are pleased to learn that our Foundation is already an active partner in these research areas. We will keep you informed as we learn more about the results of these studies.

© UCP Research & Educational Foundation, January 1997

Date:
Jun 01, 1997

A research study published in the journal Developmental Medicine and Child Neurology1 explores the risk factors associated with the occurrence of periventricular leukomalacia (PVL), one of the most important pathological changes in the premature infant’s brain that causes cerebral palsy.

Periventricular leukomalacia (PVL) is a nerve fiber tract degeneration — leukomalacia — around the cavities in the center of the brain filled with cerebrospinal fluid — periventricular. The degeneration can destroy neural communication pathways between several areas of the brain and/or between the brain and the spinal cord. Cerebral palsy is a major consequence of this destruction.

The study demonstrates that the period of greatest risk for the occurrence of PVL is between the 27th-30th weeks of fetal development, peaking at the 28th week. The full period of pregnancy is 40 weeks; thus the PVL damage occurs most often late in the second trimester or early in the third trimester of pregnancy.

The most frequent events associated with PVL during this highly sensitive period are:

• Intrauterine infection, affecting the membranes surrounding the infant;
• Premature rupture of the membranes;
• A combination of both (the most common occurrence associated with PVL).

The underlying factor is usually an infectious process in the mother’s uterus. The infection releases toxins affecting the membranes surrounding the fetus and also selectively injuring sensitive areas of the developing brain. Not only do these toxins affect the fetal brain, but they may also be responsible for the premature rupture of the membranes and premature birth of the baby.

Thus, an important period of risk to the fetal brain and a potential cause of PVL are being identified.

Comment

Evidence continues to accumulate pointing to intrauterine infection as a very important reason underlying premature delivery, fetal brain damage due to PVL, and the occurrence of cerebral palsy. The mother is often not aware of the infection and usually presents no symptoms of being “ill”. The issue is how to detect this sub-clinical infection and identify the specific organism(s) responsible for it. Doing elaborate tests on all women at 5-7 months of pregnancy searching for a potential infection is not a practical answer. A “screening” method needs to be available which would alert mother and doctor that more elaborate testing is indicated. A number of investigators are exploring the development of a practical method of screening. Also, investigators are studying why these subclinical infections cause PVL. The objectives are to identify pregnant women at risk of having a subclinical intrauterine infection, provide therapy and prevent PVL — eliminating a major cause of cerebral palsy occurring during pregnancy, but prior to delivery.

1Zupan, V et al. Periventricular Leukomalacia: Risk Factors Revisited. Developmental Medicine and Child Neurology 38:12: 1061-1067 (1996).

© UCP Research and Educational Foundation, June 1997

The pregnant woman needs to consider both her own requirements and the requirements of her baby when thinking about nutrition and nutritional supplements. Vitamins are needed during pregnancy for the maintenance of the mother’s health, to support the developing infant’s growth, and to help prevent developmental defects. However, vitamins taken as a dietary supplement are not innocuous and the infant’s genetic makeup and the environment provided by the mother are essential for their appropriate utilization. Excessive amounts of some vitamins may increase the occurrence of a developmental defect, inadequate amounts can predispose to a developmental defect. A great deal has been written about this and physicians usually provide the necessary information to an expectant mother during pre-natal care visits.

During this past year, additional information has become available about two nutrients, both usually recommended by physicians to pregnant women as supplements to a “normal American diet”. However, the best source of any nutrient is food; supplementation by pills can be very useful if done properly, but it can also be dangerous if done poorly.¹

Two nutritional supplements about which new information is available are:

TOO LITTLE – Folate (folic acid): There is now substantial evidence that during early pregnancy (the first several weeks), an additional intake of folic acid (a B vitamin) can help prevent defective development of the spinal cord (spina bifida) and of the brain (anencephaly) in infants with a predisposition to these abnormalities. The problems are: (a) recognition of pregnancy in these critical early weeks and (b) who is predisposed? During the first several weeks of pregnancy, critical areas of the brain and spinal cord of the infant are developing. Sufficient amounts of a nutrient, folic acid, is essential at this time. This is particularly important when there is a family history of either spina bifida or anencephaly.

Since folic acid is important very early in pregnancy, it is recommended by the U.S. Public Health Service that women who are either considering having children or who are at risk of becoming pregnant accidentally discuss their needs for supplemental folic acid with their physicians. Supplementation can be by vitamin pill and/or by diet. Foods rich in folic acid are cereals, lentils, dark green leafy vegetables, chicken and beef liver, and orange juice.

TOO MUCH – Vitamin A: Vitamin A is essential for fetal growth and development, particularly of the nervous system, heart and face. Vitamin A and a Vitamin A source — beta carotene, are usually present in a wide variety of commonly eaten foods. However, too much Vitamin A can lead to serious developmental defects. This is particularly true if excess amounts of Vitamin A in vitamin supplements are taken early in pregnancy. How much is too much? How much is too little? These questions are still being explored. It is suggested by experts that women who plan to become pregnant, are subject to becoming pregnant, or who are pregnant need to discuss their Vitamin A needs with their physician.

Discussion:

At this time, indiscriminate use of vitamin supplements for women who are pregnant can be dangerous to their infants; dangerous because they may not be getting enough of a needed vitamin (folic acid) or too much of a vitamin also needed (Vitamin A). In very general terms, a multi-vitamin tablet containing 0.4 mg of folic acid and no more than 8,000 IU of Vitamin A daily is being recommended by the experts. More important, it is extremely important that women before becoming pregnant or very early in their pregnancies talk to their physicians about their own specific needs and not “vitamin pill pop” indiscriminately.

¹ Rothman, K.J. et al; Teratogenicity of High Vitamin A Intake; NEJM, 1995; 333: 1369-1373; 1414-1415

© UCP Research & Educational Foundation, February 1996