About 20 years ago, Alexander J. Schaffer' coined the words "neonatology" and "neonatologist." "The one," he wrote, "designates the art and science of diagnosis and treatment of disorders of the newborn infant; the other the physician whose primary concern lies in the specialty." The word "neonate" has been with us for at least forty years, as has "neonatal;" both referring to the infant's first four weeks after birth. Schaffer's book, Diseases of the Newborn,' first published in 1960, and Clement A. Smith's Physiology of the Newborn Infant, 2 first published in 1945, served as landmarks for the emerging field of neonatology. The first comprehensive book on the diseases of the newborn was published in 1914 by August von Reuss (1879-1954); an English translation appeared in 1922.3 His book was primarily concerned with diseases but it did outline normal growth and development and the meager knowledge then available about the physiology of the newborn infant. In the past two decades, neonatology has grown from a field with few pioneers to a major subspecialty of pediatrics. But if one looks back only a few decades, interest in the care and feeding of the newborn was quite spotty.
CARE OF NEWBORN INFANT IN THE PAST
The primary provider of care for the healthy infant in all times, until less than a century ago, has been the mother. Midwives, wet nurses, grandmothers, wise old women in general, all joined to help mothers. Physicians were not much help until the late nineteenth century. Most physicians before this time shunned infants, feeling unprepared to cope with their high morbidity and mortality and also feeling incompetent to diagnose or cure them. There were no hospitals for infants, save for a few foundling homes, until 1802. The infant mortality in these homes was appalling. For example, in the late eighteenth century the mortality in the Paris Foundling Hospital, where artificial feeding was practiced, was 85% of the 32,000 infants received; in Dublin, of 10,000 infants admitted to the hospital during the years 1775 to 1796, only 45 survived-in other words, 99.6% died .4 The mortality of foundlings under a year of age in some American almshouses was as high as 97%. Previous to 1869, when the New York Foundling Hospital was opened, the foundlings of the city, numbering at least two thousand a year, had all been committed to the care of poor women in the Almshouse on Blackwell's Island, "a care so wise and tender that when the physician appointed to attend them visited the Almshouse, he was shown one miserable infant which was regarded as a prodigy because it has managed to attain the age of two months.
Even as late as the middle of the nineteenth century, it was unusual for the physician to care for the newborn infant, and not until the child was weaned did his or her medical care become the province of the physician.
Soranus on the Care of the Newborn Infant. If we look back to antiquity, we will find that Soranus (A.D. 98-138) of Ephesus wrote the first definitive treatise on the care of the newborn infant which was included in his On Diseases of Women. This treatise established a precedent, as Radbill has noted, "which was to last down to our own day, of the association of midwifery with diseases of children." This association, according to Still, "was to hamper and obstruct progress in the study of diseases of children until the mischief of it began to be recognized at the end of the last century." The care of the newborn until a century or so ago was essentially that outlined by Soranus in the second century. He was the first medical writer we know of to advise the salting and swaddling of the newborn infant, two customs that have persisted almost to the present era.
Salting and swaddling of the newborn date back hundreds of years before Hippocrates. Their omission was regarded as a mark of uncivilized barbarism as far back as the days of Ezekiel. About 600 B.C., he wrote (16:3): "Canaan is the land of your ancestry and there you were born; an Amorite was your father and a Hittite your mother. This is how you were treated at birth: when you were born, your navel-string was not tied, you were not bathed in water ready for the rubbing. You were not salted as you should have been nor wrapped in swaddling clothes. No one cared for you enough to do these things or, indeed, to have any pity for you."
Soranus was the first writer to mention the method of testing the quality of breast milk by the behavior of a drop of milk placed on one's finger nail. This test continued in use for over 1,600 years, making it probably the oldest continuously performed test in clinical medicine. The first book on diseases of children written in English, the work of Thomas Phaer (1510-1560)8 in 1545, repeats it in words which correspond very closely to Soranus' original description: "That mylke is goode that is whyte and sweete; and when ye droppe it on your nayle and do move your finger; neyther fletheth abrod at every stirring, nor will hange faste ypon your naile, when ye turn it downward, but that whyche is betwene bothe is best."
Swaddling to keep the child from growing crooked or lame was assiduously observed until William Cadogan's (1711-1797)9 An Essay upon Nursing (1748) broke down the barriers of worn-out traditions, following which swaddling and overdressing of the infant gradually disappeared.
Weakness at birth was attributed to prematurity, although it could be due to infirm parents. Rarely did such infants survive the least indisposition. But Isaac Newton (1643-1727), Christopher Wren (1632-1723), and Jonathan Swift (1667-1745), were all premature infants, yet all lived a long and vigorous life.
Birth Weight of the Newborn Infant. Before interest in the care of the newborn could be considered on anything like a firm foundation, something so mundane as knowing the infant's birth weight would be of vital importance, and would eventually be used to distinguish the infant born prematurely from one born at term. One would have thought the birth weight of the newborn to have been common knowledge in past centuries, because for us it is probably the pivotal statistic upon which pediatrics rotates. Such was not the case, because the first English writer known to have mentioned the birth weight was not a physician but rather Dean Swift, the author of Gulliver's Travels. In 1729, in his biting satire A Modest Proposal, he wrote: "I have reckoned upon a medium that a child just born will weigh twelve pounds, and in a solar year, if tolerably nursed, will increase to twenty-eight pounds."1° I believe Swift came upon his erroneously high birth weight from the writings of Francois Mauriceau (1637-1709), the famous seventeenth century French obstetrician and the leading obstetrician of his day. In the fourth edition of his Traite des Maladies des Femmes Grosses (1694), and only in this edition, Mauriceau wrote that "one will find that one (fetus) of nine months ordinarily weighs about 12 pounds ... each; I have even seen some weigh as much as fourteen pounds."
Mauriceau's excessively high birth weight was topped by that quoted by Theophilus Lobb (1678-1763)10 in his Compendium of the Practice of Physick (1747). Lobb, a member of the College of Physicians, and a Fellow of the Royal Society of London, wrote his compendium so that those who were unable to attend his lectures might have the advantage granted to his listeners. In the course of his twenty-fourth and last lecture, Lobb cited a letter, dated August 8, 1747, which he had received from Mr. Phillipson, whom Lobb further described as an ingenious surgeon and man-midwife. Phillipson wrote that "I have according to your desire weighed a child and placenta soon after delivery; the child weighed sixteen pounds and seven ounces, which is a large child but have seen some larger."
William Smellie (1697-1763), a towering figure in English obstetrics, in his Treatise on the Theory and Practice of Midwifery (1752) wrote that "at nine months [the infant would weigh] from ten to twelve, and sometimes sixteen pounds."10 One wonders how Smellie could have been so mistaken. He gave hundreds of courses of lectures attended by thousands of pupils; that none of the thousands who listened to him questioned his figures for an infant's birth weight would point to a widespread indifference concerning human growth during the eighteenth century. This error was perpetuated in several eighteenth century writings and was not challenged until the German obstetrician, Johann Georg Roederer (1727-1763),10 read a paper on December 1, 1753 before the Royal Society of Gottingen. This paper, which received little publicity, may be considered the first to record the correct length and birth weight of the newborn infant.
It was not until 1815 that the first large-scale, accurate report on the weight of newborn infants was published; weights were recorded on 7,077 newborn infants delivered at the Maternity in Paris.10
Emergence of Lying-In Hospitals. One reason, perhaps, for the physician's inattention to the medical needs of the newborn infant until relatively recently was that almost all infants, until the early part of this century, were born at home; only a small number were attended by a physician. This occurred because the lying-in hospital, prior to the latter part of the last century, was viewed as even a worse place for a patient than the general hospital. According to Vogel,11 "Maternity care would be among the last reasons the comfortable classes entered hospitals. In the late nineteenth century, women still considered that childbirth could and should be performed even in the simplest and poorest of homes. Hospitals offered no specialized paraphernalia or contrivances; puerperal fever was rampant in them and maternal mortality was high." For this reason, only the most desperate women entered hospitals to have children. And perhaps the major cause of this desperation was illegitimacy. Small lying-ins, often no more than a few rooms in a tenement or boarding house, kept by midwives or the unscrupulous or untrained, served those seeking to "hide their shame" or having absolutely no alternative. Vogel has also noted that these lying-ins and the baby farms that sometimes accompanied them were seen as accessories to vice and degradation, and as adjuncts to brothels. Licensed and respectable lying-ins existed, but their patients too were "unfortunate women," often "utterly destitute of money and friends."11
Hospitalization for childbirth was linked to the rise of the obstetrician. Once the specter of puerperal fever was largely banished and when hospital-based obstetricians developed new medical skills, and the hospitals' image improved, women of all classes began to use hospitals for difficult birth and gradually came to consider it a safer, more comfortable, and cleaner place than their homes for all deliveries.
Mothers Delivered in Hospital. From 1900 to 1925 the number of prenatal clinics increased markedly and a growing number of American mothers decided to have their babies in hospitals; this trend led to the establishment of newborn nurseries in American hospitals. While less than 5% of American women had delivered in hospitals in 1900, the percentage greatly increased as the century progressed. By 1921, more than half the births in many large American cities took place in hospitals, varying from a high of 85% in San Francisco to a low of 9.2% in New Bedford, Massachusetts.12 The development of newborn nurseries in hospitals and the need for someone to direct them stimulated, more than anything else, an interest in the medical care of the newborn; up until the 1920's the care of the newborn was more likely rendered by obstetricians than by pediatricians. Before 1900, rooming-in was the usual mode of care in maternity wards. However, during the first decade or so of this century the elevated mortality and morbidity from infectious disease in newborns led to the institution of strict isolation techniques and a reduction in the handling of and visiting with infants in newborn nurseries. No longer were infants kept close to their mothers. However, rooming-in practices were revived in the mid 1940's at the Grace-New Haven Unit at Yale University. 
Interest in the neonate was further heightened by the nationwide mortality figures published by the newly-created Children's Bureau (1912).4 These studies, while indicating a favorable trend in lowering the death rate among infants in general, clearly pointed out that the infant mortality in the first month of life had remained practically stationary during the twenty-year period from 1900 to 1920.
By 1940 about 60% of all white births were delivered in hospitals as compared to about 25% of all non-white births; but by 1950 the respective percentages were, for whites, 93%, and for non-whites, 58%. By 1970, almost all white infants, and more than 80% of non-white infants, were delivered in hospitals. The continued increase in the number of hospital deliveries offered physicians a greatly expanding experience in the care of the neonate. Thus by 1950 a few pediatricians were creating the field of neonatology, in order to study the pathogenesis of the disorders that have their onset in utero or during the neonatal period, so as to improve the medical care of the newborn infant.
Care and Feeding of Premature Infants. The beginnings in rational and, eventually, scientific attention to the neonate came largely through the emerging interest in the care and feeding of the premature infant. It is correct to say that in the past two decades, neonatology, as an accepted subspecialty of pediatrics, has come into being around the intensive care premature nursery. Increased efforts to conserve infant lives, as Ethel Dunham (1888-1969)14 had noted, "follow periods of national disaster." The Franco-Prussian War (1870-1871) brought about such a situation in France. There the grave loss of human life at that time aroused the public to the urgency of reducing infant deaths. Foremost in this effort were two Parisian obstetricians, Stephane Tarnier (1828-1897) and Pierre Budin (1846-1907).
Interest in infant welfare in France in 1870 was not entirely motivated by humanitarian concerns; it was primarily the declining birth rate that led to an intensive investigation of the causes of infant mortality and means of prevention. The decline in births in France after 1870 (not matched by either Germany or England) led to a condition viewed by the French as approaching national disaster, because deaths threatened to exceed births. A Commission on Depopulation was appointed to make a thorough study of the whole question. While the decrease in the number of children born overshadowed the danger of deaths through the first year of life, measures to prevent infant mortality formed an important part of the work of the Commission on Depopulation. Among these measures were the establishment of Consultation des Nourissons, Goutte de Lait, Creches, and greatly improved maternity hospitals. These gave France the distinction of being the pioneer in modern, organized maternity and infant hygiene programs.
Burdin who was successively head of obstetrics at the Charite and Maternity Hospitals in Paris was the first to outline the special care needs that still prevail for premature infants. He wrote that "with weaklings we shall have to consider three points: (a) their temperature and their chilling, (b) their feeding, and (c) the diseases to which they are susceptible."
The first and most immediate need was to keep the infant warm. The inability of the premature infant to maintain its body temperature was a well-known fact, and the need for an artificial means to overcome this handicap led to the development of incubators.
Incubators. Budin and Tarnier were aware of the studies of Milne-Edwards and those of Villerme, as early as 1829, that demonstrated the significant effect temperature had on neonatal mortality.14 Both were also aware of the efforts of Paul Denuce (1824-1889)16 of Bordeaux and Carl Crede (1819-1892)17 of Leipzig to maintain an appropriate body temperature in the premature infant by placing the infant in a double-walled heated bath. I would also assume that Denuce and Crede knew the story of Licetus Fortunio (1577-1657) who was said to have measured 5 zoll (inches) at birth and who lived to be 80 years old and who wrote 80 books.
In his comic novel Tristram Shandy, Laurence Sterne (1713-1768) described Fortunio's premature birth and how he was cared for in an incubator as follows:
And for Licetus Fortunio . . . all the world knows he was born a foetus. [He] was no larger than the palm of the hand, but the father, having examined it in his medical capacity, and having found that it was something more than a mere embryo, brought it living to Rapallo, where it was seen by Jerome Bardi and other doctors of the place. They found it was not deficient in anything essential to life, and the father, in order to show his skill, undertook to finish the work of nature and to perfect the formation of the infant by the same artifice as is used in Egypt for the hatching of chickens. He instructed a wet-nurse in all she had to do, and having put his son in an oven, suitably arranged, he succeeded in rearing him, and in making him take on the necessary increase of growth, by the uniformity of the external heat, measured accurately in the degrees of the thermometer, or other equivalent instrument.
The first paper describing, but not illustrating, an incubator developed for use in the care of the premature infant was that of Denuce. In a paper of about 400 words, citing no references, and published in the Journal de Medicine de Bordeaux in 1857, he described a double-walled zinc tub with a space between the two walls that could be filled periodically with warm water. A similar double-walled metal incubator was developed, however, as early as 1835, at the suggestion of Johann Georg von Ruehl (1769-1846), physician-in-ordinary to the Czarina Feodorowna, wife of Czar Paul I, for use at the Imperial Foundling Hospital in St. Petersburg. 19
About 1850, a modification of Ruehl's incubator was in use at the Moscow Foundling Hospital, founded in 1764 by order of Catherine the Great; by the mid-1850's this hospital was said to have had more than 40 of these warming tubs (Waermewannen). As a full description of Ruehl's apparatus and of its proven value in the care of the premature infant did not appear in the medical literature until 1873, Denuce's paper may rightfully be given priority for being the first published report of an incubator which had actually been used and found of value in the care of a premature infant.
In 1884, Crede published the results he had obtained over a period of almost twenty years at the Leipzig Maternity Hospital, in caring for 647 premature or debilitated infants, with a similarly constructed incubator.
Tarnier claimed that he came upon the idea for his incubator as a result of a walk he had in the Jardin d'Acclimation in early 1878 where he saw an incubator (modified from older Egyptian and Chinese models and constructed by Odile Martin) to hatch hens' eggs .21 He immediately had the idea to replace the egg with a premature infant. The first model, based on the chicken incubator, was tried at the Maternite of Port-Royal in 1880. Tarnier inspired his intern, M. A. Auvard (1855-1941),22 to improve the incubator in 1883.
By the use of the incubator, the mortality of premature infants weighing less than 2000 grams, cared for between 1879 and 1882, fell from 66% to 38%. The so-called Tarnier incubator was improved by Madame Henry, midwife-in-chief of the Paris Maternite. The first incubator manufactured in the United States was that of Thomas Morgan Rotch (1849-1914), but it was soon found to be too expensive and cumbersome for general use.
It was Budin who first instituted a special unit for premature infants at the Maternite, toward the end of 1893. He was in charge of this unit until March 1898, at which time he was appointed to head the Clinique Tarnier's premature unit.14
These two hospitals, the Maternite and the Clinique Tarnier, became the first neonatology centers in the world for clinical research and teaching of premature infant care.
The Denuce open incubator was the prototype of the Hess bed, while Tarnier's closed incubator evolved in time to become the Isolette.®
Influence of France and Germany on American Premature Centers. Another major step in the care of the premature infant occurred in 1909 with the founding of Kaiserin Auguste Victoria Haus in Berlin, a center for teaching and research in the prevention of infant mortality. It was here that Leo Langstein (1876-1933) developed a major academic milieu for the investigation of the premature infant. In this hospital, Professor Arvo Ylppo (b. 1887) served as a staff member and participated in research, from 1912 to 1920. Here he made his valuable contributions to knowledge of the pathology of prematurity, prenatal and postnatal growth in weight and length, and the mortality rates of premature infants in relation to birth weight.
The influence of the French and German centers for premature infant care played a role in the development of such centers in other countries. For example, Julius H. Hess (1876-1955), who founded the first premature center in the United States at the Michael Reese Hospital in Chicago, spent several years in Germany and Austria studying the care and feeding of premature infants. In 1922,23 Hess published the first book ever written dealing solely with premature and "congenitally diseased" infants.
Birth Weight vs. Gestational Age. The diagnosis of prematurity on the basis of birth weight was first mentioned by Alexandre Gueniot (1832-1935)14 in 1872 as "not over 2300 gm and maybe below 1500 gm." The criterion of 2,500 gm for distinguishing a premature from a mature infant was used for the first time in 1869 by Nikolay F. Miller (1847-1897),14 physician-in-chief of the Moscow Foundling Hospital. In 1948, the first World Health Assembly adopted a birth weight of 2,500 gm or less as an international definition of prematurity. About a decade later, an expert committee of the World Health Organization recommended that the concept of "prematurity" be replaced with that of "low birth weight." The Committee of the Fetus and Newborn of the American Academy of Pediatrics, in June 1962, endorsed this change.
During recent years, neonatologists have learned that intrauterine growth is not constant and that considerable discrepancy in size related to gestation may occur. Formerly it was held that all infants weighing less than 2,500 gm were premature. However, studies have shown that as many as one-third of infants weighing less than 2,500 gm at birth are of term gestation.
To appraise intrauterine growth with respect to gestation, the standards of Lubchenco et al.,24 first published in 1963, from the 24th to 42nd week of gestation serve well and they have made the initials SGA, AGA and LGA (small for gestational age; average for gestational age; and large for gestational age) as routine and as useful in newborn nurseries as the Apgar score. The use of gestational age in conjunction with birth weight adds greatly to our understanding of the newborn infant, because neonatal mortality is highly correlated with both.
Feeding the Low Birth-Weight Infant. During the first quarter of this century human milk was considered by most pediatricians to be of the utmost importance in feeding premature as well as term infants; a few, however, believed that even breast milk should be diluted and predigested for feeding the infant weighing less than 1,500 gm. If breast milk was unavailable, Hess23 recommended buttermilk, or skimmed milk with added carbohydrates, for the first 3 weeks, at which time some whole boiled milk was added to prevent fat inanition. Clifford G. Grulee (1880-1962)25 favored albumin milk. In a thorough review of the literature, J. B. Griffith (1856-1941)26 in 1912, noted that "breast-fed babies have at least five times the chance of living, than the bottle-fed babies possess."
During the early years of this century the number of calories per kilogram per day required by premature infants was believed to be between 120 and 150. These figures were derived from German studies inaugurated by Max Rubner (1854-1932) and Otto Heubner (1843-1926) in Leipzig in 1894, from their work with the respiration calorimeter designed for infants.
Premature infants were fed by gavage or by using a Breck feeder if they were too weak to nurse. Gavage feeding of debilitated newborn infants was first recommended by Marchant in 1850. In the premature wards of Bellevue Hospital in 1916, breast milk was diluted one-half with whey for the first few days, one ounce being given every one and a half to two hours, depending on the size of the baby. The infants were initially given 120 to 150 calories per kilogram per day.29
Not everyone considered breast milk the ideal food for premature infants. Rotch, at one time, believed that the best way to feed them was by means of a modified milk formula carefully prepared at milk laboratories. He believed that this method of feeding premature infants was "far superior to even breast feeding, and . . . will result in a decided reduction in their mortality."
Diarrheal disturbances in infants fed cow's milk were the subject of many late nineteenth and early twentieth century studies. Since Pasteur's day, and with the increasing sophistication of bacteriological methods, it became obvious that contaminated, unclean milk caused serious digestive disorders. To prevent such disorders, attempts were made to destroy the pathogenic organisms that thrived in milk. Philipp Biedert (1847-1916),31 in 1882, was the first to treat milk by heating it for two hours in a hermetically closed vessel placed in a water bath at 100° C.
As late as 1900, the predominant opinion of members of the American Pediatric Society was that raw milk was best for infant feeding, if a clean supply was available. The early meetings of the American Pediatric Society contained many reports about the pros and cons of raw as opposed to pasteurized milk. Infants fed raw milk were said to have far less chance of developing scurvy than those fed pasteurized milk. However, Abraham Jacobi (1830-1919)32 first proposed the boiling of milk for infant feeding, as early as 1873. He continued teaching the wisdom of this method for the rest of his life, despite the outcries of many skeptics.
During the early 1900's pediatricians were faced with several widely conflicting hypotheses about the cause of nutritional disturbances in infants. Biedert claimed that the difficulty of digesting casein was a prominent cause of infantile digestive disorder. Heinrich Finkelstein (1865-1942)4, on the other hand, claimed that fats were injurious to the digestive tract, and Thomas Morgan Rotch's theory of percentage feeding was devised as a means of offsetting the putative harmfulness of proteids (proteins).
Early vs. Late Feeding. At the beginning of the century, early feeding of premature infants, within the first twelve hours of the infant's birth, was encouraged because of the generally held belief that small infants could not tolerate starvation. Hess23 for example, recommended that the first feeding of milk should be given to premature infants at about 12 hours of age. He stressed that "the necessity of an early supply of food cannot be overemphasized as even the better developed infants do not withstand starvation." This view of early feeding was frequently questioned in the first three decades of this century and, in the 1940's, an era dawned when delaying the first feeding until the second or third day, or even longer in sick or very small infants, became the custom. Two factors led to this change in attitude. The first was to prevent the risk of pneumonia from the infant aspirating milk into the respiratory tract. Supporters of late feeding based their argument primarily upon the valid reason that swallowing is imperfect in the premature infant, and that reflux with its dangers of aspiration pneumonitis, is extremely hazardous.34 The second factor was the emerging belief that premature infants retained an excessive amount of extra-cellular fluid in the early days of life, so that early feeding was perhaps both unnecessary and even stressful to the infant's kidneys. Despite severe loss of weight -- as much as 20% was often noted -- as well as hemoconcentration and hyperosmolality when the first feeding was delayed for 2 to 3 days, or even longer, it was felt that these metabolic disturbances were quickly reversed when feedings were started. Thus it was believed that premature infants could tolerate fluid and calorie restriction without adverse effects.
Delayed feeding practices originated about 1947 in the United States and were soon adopted in the United Kingdom. For example, Crosse in 1952 wrote, "In recent years the survival rate has been improved by giving nothing by mouth for several days." In her nursery, the smallest infants were submitted to the longest periods of starvation; e.g., 4-5 days for infants weighing 2 lbs and less, 3 days for 3 lb infants, and 1-2 days for babies of 4 lbs or more.
The putative advantages of delayed feeding were not accepted by all neonatologists. In Finland, Ylppo questioned the concept of delayed feeding and recommended, as he had always done, that milk should be given as soon as possible after birth. In 1955 Geiss, in Germany, also opposed the starving of premature infants, claiming a lower mortality (28%) in premature infants who were given their first feeding between 12 and 24 hours, compared with infants who were fed initially at 36 hours (41%).
By the early 1960's, some pediatricians began to question the wisdom of delaying the feeding of premature infants. Two developments responsible for these changing attitudes emerged. First, there arose an increasing awareness from a number of clinical studies that some types of cerebral palsy, especially spastic diplegia, might be due to early starvation. And second was the fear that early undernutrition could permanently affect the growth of many organs, particularly the brain. The experimental studies demonstrating the dangers to brain and body growth formed a sound basis for questioning the wisdom of delaying the feeding of premature infants. Those low-birth-weight infants who were starved initially had a higher incidence of neurologic handicap, visual impairment, and mental defect than those who were fed early.
The first detailed clinical study of the effects of feeding milk very early to premature infants was reported by Smallpeice and Davies in 1964. Their results confirmed the value of early milk feeding in reducing the incidence of hypoglycemia, neonatal jaundice, and hypernatremia. By the early 1970's, the practice of early feeding of low birth weight infants was well established. As Davies has written, "the wheel had now turned a complete circle since the beginning of the 20th century."
Drillien42 and Churchill, in well-controlled clinical studies in the early 1960's, offered evidence of the possible relationship between early undernutrition and cerebral palsy. For example, Drillien, from his Edinburgh studies, showed that three-quarters of the infants weighing less than 1,367 gm at birth, who were born in the years 1953 and 1954 (when feedings were delayed), had developed severe neurological handicap. On the other hand, infants of similar birth weight who were born before this time (1948 to 1952) and afterwards (1955 to 1960) had a much lower incidence of neurological handicap (about 30% in each period). Drillien42 believed that delayed feeding practices might have accounted for the very high incidence of handicap in the infants born in 1953 and 1954, since these infants had not received any fluid until the third or fourth day and the introduction of milk was occasionally delayed until the fifth to the ninth day.
Another dramatic shift in feeding practice took place in the United States in the 1940's. The change was influenced by studies conducted in New York, in 1941, when the premature infant's difficulty in absorbing milk fat from the intestine was described in quantitative terms. It was noted that when human milk or a cow's milk formula was fed, a significant fraction of the calories ingested was lost in the stools as unabsorbed fat. But the feeding of half-skimmed milk mixtures led to reduction of fat in the stool.43 The latter practice was adopted quickly. A half-skimmed cow's milk mixture was offered in the first feeding; it was not reserved as an alternative for use in infants who failed to gain weight after a trial on human milk. Not only was the amount of milk protein increased, but the predominant kind of protein was changed, from whey protein in human milk to casein in cow's milk.
Debate resulted about this major shift in infant feeding practice. At issue was whether the weight gain in infants fed the cow's milk mixture was more the result of water and electrolyte retention rather than a primary increase in body tissue.
The current trend in this country now leads to feeding low birth weight infants cow's milk, modified to reduce the fat content and electrolyte load, and with adjustments in protein composition to reach a level of 4 gm per kilogram per day.
Volumes have been written about the relative merits in infant feeding of human milk versus cow's milk, of whole milk versus skimmed milk mixtures, and of simulated human milk products versus the less complex cow's milk or evaporated milk formulas. It is apparent, however, that many low birth weight infants have been raised successfully on breast milk.
Methods of Feeding. In 1951, Royce et a1.44 published the first report of their experience with an indwelling nasogastric tube for feeding small premature infants. Previous to this report, infants too weak to suck well had been fed by gavage or with a dropper. The indwelling nasogastric tube permitted an increase in the frequency of feedings without a concomitant increase in fatigue or in the likelihood of esophageal irritation.
Total intravenous alimentation for low-birth-weight infants using peripheral veins began about 1970.45 Techniques for providing total parenteral nutrition have been extended to newborns. The value of parenteral nutrition for infants unable to tolerate other routes of caloric intake has been amply demonstrated. Applied to newborns with surgical problems of the gastrointestinal tract, according to careful protocols for catheter placement and care, these techniques have proved to be life saving. However, there is no justification for the routine use of parenteral alimentation in low-birth-weight infants. The American Academy of Pediatrics concurs with this view.
There remain a number of unanswered questions concerning nutritional problems. It is clear, however, that much has been accomplished through parenteral nutrition and enteral feeding, by a number of routes, of small, sick, premature infants. For example, infants with bronchopulmonary dysplasia and those with abdominal surgical problems are surviving for longer periods now. Nutrition is no longer the limiting factor in their survival.
Infections. A century or so ago there were two frequently mentioned, and now almost completely forgotten, diseases of the newborn infant; they were Buhl's and Winckel's diseases. The former was also known as acute fatty degeneration of the newborn, and the latter as epidemic hemoglobinuria of the newborn.
Buhl's disease, first described in 1861, was said to be an acute parenchymatous fatty degeneration of the liver, kidney, or heart, combined with hemorrhages into the various organs, or from the umbilicus, intestines or stomach. Most late nineteenth century pediatricians considered this disease to be the result of sepsis; many cases described as omphalitis and hemorrhage from the umbilicus were thought by authorities such as Koplik, Holt, and Rotch to belong to the category of Buhl's disease.
Winckel's disease, first described in the epidemic form in 1879, was characterized by the sudden appearance of cyanosis and icterus with hemoglobinuria. Like Buhl's disease, it was, in all likelihood, due to sepsis. The symptoms in Winckel's cases appeared in the fourth day after birth in apparently healthy and well-developed infants. From the clinical course he described, this disease was certainly a form of sepsis.
Infection as a serious threat to the infant, especially to one born before term, has been recognized by all who have cared for the neonate. Gueniot in 1872, was one of the first to warn of the danger of infection. In outlining the needs of the premature infant, he mentioned not only the nutritional and thermal requirements, but also the importance of proper care of the skin to prevent infection.
Budin, after a severe epidemic of respiratory tract infection among the premature infants at the Maternite in 1896, recognized the defective organization of the nursing service for premature infants in failing to provide for isolation of sick infants. This led him to propose a plan for a special self-contained unit for premature infants, the first of its kind. Budin listed its principles as follows: the grouping together of healthy prematures; the isolation of the sick; the total separation of the wet nurses' infants from contact with the premature infants; the establishment of a milk room where sterilized milk could be heated prior to feeding; an ice chamber for keeping the bottles of sterilized milk cool in summer; and a dressing room for wet nurses so that they could wash their hands and face and put on clean gowns.
By the end of the nineteenth century Budin and others had called attention to the four chief sources of infection -- the skin, umbilicus, and the gastrointestinal and respiratory tracts.
The persistently high mortality and morbidity of newborn infants in the early 1900's, usually from epidemics of diarrhea or respiratory infections, led to strict isolation techniques in newborn nurseries. Most textbooks on hospital care of the newborn, from the 1920's until the mid-1960's, recommended minimal handling, strict isolation, and the exclusion of all visitors from the nursery. Unfortunately, while careful precautions were instituted to reduce the incidence of contamination and cross-infection, no provisions were made for the mother-infant bonding, and we are also now learning about the harmful effects of early separation on the subsequent maternal relationship.
Even today as many as 2% of fetuses may be infected in utero and up to 10% of neonates can acquire an infection during birth or the first month of life.49
At present, the incidence of sepsis in full-term and premature infants is estimated to be 1 to 1.8 per 1,000 live births, and there continues to be an unexplained male predominance. Mortality rates in infants with sepsis range from 20% to 40%; gram-negative enteric bacilli and group B-beta-hemolytic streptococci now account for up to 70% of reported cases.
Newborns in intensive care units have a higher rate of sepsis; meningitis and beta-hemolytic streptococci of the Lancefield group B type have emerged over the last decade as the leading causes of neonatal sepsis and meningitis. The reason for this is not known.
A recent review of fifty years of neonatal sepsis at Yale-New Haven hospital showed that the predominant organism in the 1930's and early 1940's was group A Streptococcus which was replaced by coliform organisms in the late 1940's and early 1950's. During 1955-65 E. coli and Klebsiella were the principal pathogens. From 1966-1978, the most common etiologic agents were E. coli (31.7% of the total), group B Streptococcus (25.2%), Klebsiella (14.6%), and S. aureus (6.3%). Recently other organisms have emerged as neonatal pathogens including Hemophilus, group D Streptococcus, Staphlococcus epidermidis, Citrobacter species, and anaerobic bacteria. Eighty percent of group B Streptococcus infections were early onset.
The overall mortality rate from neonatal sepsis fell from 90% in the early 1930's to 18% in 1978.51
Supportive Therapy. Supportive therapy includes those procedures which might be termed physiological treatment since they are directed toward correcting physiological disturbance rather than toward eliminating the cause of disease. They involve the use of parenteral fluids such as blood, fluid and electrolyte solutions.
The pediatrician practicing in the last quarter of this century, with his modern (usually disposable) needles, syringes, and tubing has little idea of the handicaps the physician faced less than half a century ago in performing a task so mundane as a venipuncture. Requisite technical procedures are inevitably evolved when there are definite therapeutic needs and these develop only with an understanding of the disturbances in physiological processes brought about by disease. This understanding was what was lacking fifty years ago.
The drawing of blood was facilitated by an ingenious and yet rather complicated suction apparatus devised by Kenneth Blackfan (1883-1941)51 in 1912. He noted that "the methods usually employed for obtaining blood from infants are in the majority of instances extremely unsatisfactory. The veins are too small to enter and it is tedious, difficult, and painful procedure to collect the necessary amount of blood by puncturing the fingers or toes."
In the early decades of this century fluids, if prescribed at all, were usually given subcutaneously or rectally. Intraperitoneal injections of saline for use in the newborn infant were first mentioned in the American literature by Blackfan in 1918. He had used this route at the suggestion o£ John Howland, (1873-1926), "Who saw it used as a routine measure on the service of Professor [Archibald] Garrod at St. Bartholomew's Hospital, London."
Intraperitoneal fluids were prescribed extensively in the Harriet Lane Home, between 1916 and 1918, with satisfactory results. For small infants, 100 to 250 ml. was injected and repeated in twelve to twenty-four hours if needed.53
By 1920, intraperitoneal injections, usually of physiological saline, rarely of 5-10% glucose solution, were commonly used throughout the country in the treatment of infants suffering from persistent diarrhea or infantile atrophy. Fluids were also administered through the superior sagittal sinus.
In 1915, Henry F. Helmholtz (1882-1958)54 wrote: "The method of obtaining blood from the veins of the scalp, from the jugular or other veins, is very difficult and open to frequent failures," but "there was one place that is far superior to any other for the purpose of obtaining blood, namely the longitudinal sinus in the area of the anterior fontanel." This route was also sanctioned for the administration of intravenous fluids.
Blood transfusions were rare occurrences in pediatrics in the early part of the century, because Karl Landsteiner (1863-1943) had only recently discovered (1900) isoagglutinins in human blood capable of agglutinating other human red blood cells.
In 1923, James B. Sidbury (1886-1967)55 first described transfusion through the umbilical vein in the treatment of a case of hemorrhage of the newborn. Louis K. Diamond (b. 1902), in 1951, twenty-eight years after Sidbury's report, established the umbilical route as the safest and simplest for exchange transfusion in hemolytic disease of the newborn.
Oxygen inhalation in the newborn premature or debilitated infant was first suggested in a paper written by Tarnier in 1889 and read before the Societe d'Obstetrique et Gynecologic. The first detailed clinical report of the value of oxygen inhalation in premature infants, or in infants with cyanosis, was published in 1891.56 In this report, oxygen inhalation was given only for two hours daily. Oxygen inhalation in newborn nurseries became commonplace, but not routine, about 1920. The shift to routine oxygen treatment occurred in the 1940's. Intragastric oxygen was first suggested by Ylppo in 1917 as a useful method for oxygenating newborn infants. Following clinical reports by Akerren and Fuerstenberg in 1950 and later by Waller and Morris in 1953, this technique was accepted in a few newborn centers as ideal for newborn resuscitation, primarily because of its safety and simplicity. However, despite its acceptance, James, Apgar, Burnard, and Moya showed that there was no evidence of transfer of oxygen from the gastrointestinal tract to the systemic blood in asphyxiated infants when absorption from the lung was prevented by an endotracheal tube.
Neonatal Iatrogenic Diseases. The 1950's and early 1960's were not only the years of early starvation; they were also, as stated in a lead article in the Lancet (1974),61 "the years where modern neonatal iatrogenesis reached a peak when almost every major error in newborn care was widely practiced, at least for a time."
Among the iatrogenic diseases none proved more catastrophic than retrolental fibroplasia. This dreaded retinopathy was first reported by Terry in 1942, about two years after respiratory difficulties in immature infants began to be treated more aggressively with oxygen. Terry considered the condition a congenital anomaly of embryonic blood vessels in the vitreous persisting abnormally and subsequently undergoing a fibrous hyperplasia. He suggested that this fibrous overgrowth incorporated the retina in a mass behind the crystalline lens.
The association between oxygen exposure and this disease was not recognized until Dame Kate Campbell (b. 1899), in Australia in 1951, reported that oxygen was the main causative factor of what was by this time called retrolental fibroplasia. However, a standard American pediatric textbook (Holt and McIntosh's Pediatrics, 12th ed.), as late as 1953, stated that "the oxygen content of the incubator need not exceed 60%, although higher concentrations appeared to do no harm." By 1945, 12% of premature infants with a birth weight of 1,360 gm (3 pounds) or less were blind, and it has been estimated that by the time control of the disease was achieved, eight thousand infants had suffered this fate.
Prior to 1940, the condition was virtually nonexistent. Within ten years, however, retrolental fibroplasia became the largest single cause of child blindness in the United States, greater than all other causes combined. After 1955, when a general restriction of oxygen usage was adopted (not to exceed 40%), the largely iatrogenic condition virtually disappeared, and only sporadic cases, in very low birth weight infants, are now encountered. Immaturity was a major factor; most affected infants had a gestational age of less than 34 weeks. However, even meticulously careful monitoring of oxygen concentrations has not entirely eliminated the condition. Because of the increasing survival rate of smaller preterm infants there appears to be a resurgence of retrolental fibroplasia, because such infants seem particularly susceptible to oxygen damage.
During the upsurge in incidence of retrolental fibroplasia between 1945 and 1950, the condition was paradoxically prevalent in the larger and better-equipped medical centers but spared the smaller rural hospitals. It is now apparent that the larger institutions were the first to utilize the more efficient incubators which permitted the delivery of a higher concentration of oxygen.
Two other iatrogenic diseases of the newborn infant during this period were kernicterus as a result of sulfisoxazole prophylaxis, begun in 1953 to prevent bacterial infections-especially in the premature infant-and the "gray syndrome," caused by excessive doses of chloramphenicol. Kernicterus was found nine times more often among infants who succumbed, after having received a prophylactic course of penicillin and sulfisoxazole, than among those in a control group. It was not until 1959 that sulfonamide drugs (especially sulfisoxazole) were noted to augment the movement of bilirubin into tissues, including brain, by decreasing the available binding sites on albumin necessary to keep bilirubin in the vascular space.
In the late 1950's, in all parts of the country, nurses and physicians observed a strange new disorder in premature infants. It came to be known as the "gray syndrome." The affected infants on the third or fourth day of life developed abdominal distention, vomiting, irregular respiration, and pallor. Cyanosis soon followed, and the victims developed a ghastly gray color of the skin. In a few hours, they were dead. In 1959 it was found that in newborn infants, especially premature ones, usual doses of chloramphenicol, because of reduced rates of conversion to glucuronide and of renal excretion, often led to toxic levels of the drug in the blood with fatal consequences.65
Erythroblastosis fetalis became a disease whose cause was clearly determined in 1932 by Diamond, Blackfan, and Baty when they recognized that hydrops fetalis, icterus gravis, and congenital anemia of the newborn-which had, up to that time, been considered three unrelated conditions-were in fact related and were all due to one basic mechanism, to which they applied the term erythroblastosis fetalis. Its treatment was successfully developed to a great extent in 1946 by Diamond and coworkers by exchange transfusion by umbilical vein catheterization and, since 1958, by the use of phototherapy, after Cremer et al. proved that serum bilirubin from jaundiced infants was highly photosensitive. And finally its prevention was made possible in 1963 with the development of a gamma globulin concentration of anti D-Rho (D) immune globulin (human) (RhoGAM), first reported in this country by Freda for routine post-partum prophylaxis against sensitization of Rh-negative women .s9
Respiratory Distress Syndrome (RDS) was first described as a specific pathological entity in 1953 under the name of hyaline membrane disease (HMD). Prior to that date, infants who had signs and symptoms of this syndrome were diagnosed as having congenital atelectasis. RDS remains a leading cause of infant mortality accounting for at least 20% of all neonatal deaths in the United States. The single most important discovery in advancing our knowledge of RDS was reported in 1959 by Mary Ellen Avery and J. Mead .70 They observed that lungs of infants dying of this disorder, as well as lungs of immature newborns in general, were deficient in surface-active material (surfactant). In the years since 1959, many studies have been published to buttress the hypothesis that the symptoms of RDS reflect an underlying deficiency of pulmonary surfactant.
Among the major advances in the treatment and prevention of RDS developed within the past decade and a half are these:
1. An understanding of the pathophysiology and pathogenesis of RDS, in relationship to pulmonary surfactant.
2. An appreciation of the fundamental aspects of fetal lung development, including the movement of surfactant phospholipids from fetal pulmonary fluid to the amniotic cavity.
3. A method to accelerate fetal lung maturation, especially with maternally administered prenatal glucocorticoids. Many clinical trials with pregnant women in premature labor document a significantly lower incidence of RDS after antenatal glucocorticoid treatment.
4. A knowledge of ventilatory techniques effective in protecting and conserving alveolar surfactant by the continuous application of end expiratory pressure. In infants weighing more than 1500 gm, it is now established that reduced mortality rates result from the use of end expiratory pressure.
5. An ability for prenatal assessment of the risk of developing RDS, by amniocentesis and analysis of the ratio of lecithin to sphingomyelin in amniotic fluid, and
6. The emerging method of postponing delivery.
And finally, in 1980, Japanese investigators offered evidence that endotracheal instillation of artificial surfactant may prove to be a useful treatment for severe hyaline membrane disease .
Survival of Very-Low-Birth-Weight Infants improved steadily. Table I citing infant mortality, by age, in the United States for the years 1930, 1978, and 1979 demonstrates the highly satisfactory downward trends both in neonatal and infant mortality rates.
Within the past several years, a number of reports have suggested that the outcome of very-low-birth weight infants (< 1500 gm) is now better than it was 20-30 years ago. These reports have been encouraging to those in perinatal intensive care which, it has been assumed, through its technological development, has been responsible for this improvement.
However, Jones, Cummins, and Davies of the Departments of Pediatrics and Neonatal Medicine, Hammersmith Hospital, London followed 357 babies weighing 501-1500 gm born alive at Hammersmith Hospital in 1961-1975. Of these infants, 58.8% died in the neonatal period, and 1.4% at 4-25 months of age. These authors noted no significant improvement in neonatal mortality in three five-year periods, 1961-65, 1966-70, and 1971-75, except that more small-for-gestational-age infants survived in the last period than in the first. Also of importance was their finding that "there was no significant improvement in the proportion of handicapped children among the very-low-weight live-born infants throughout the 15 year period, despite increasing complexity of care."
This report was followed by passionate correspondence, largely concerned with the observation that the survival of these infants had not improved over the same period. A number of neonatologists took issue with this report and pointed out that in many centers there has been a remarkable improvement in the survival of small infants, particularly those under 1000 gm, in the past 3 or 4 years. 73,74
The present confusion concerns two problems: the mortality rate and the handicap rate. There seems to be little question that the neonatal mortality rate for very-low-birth weight infants has improved significantly in the past 5 years, both in the U.K. and in the U.S.A. The mortality of infants between 1001 gm and 1500 gm in major centers is now about 15%.
There has been an equally remarkable improvement in the mortality of the smallest infants. Those weighing a kilogram or less now have a neonatal mortality of 50-70% in major units, and even as low as 36% in Cambridge, England.75
The more difficult problem is to assess the numbers of these small infants who are handicapped. Most authors divide those affected as either having a major or a minor handicap, but they do not always agree on definition.
REGIONAL PERINATAL SERVICES
In the late 1960's and early 1970's, reports began to appear in the pediatric and obstetrical literature describing improved outcome in caring for high risk pregnancies.
In nine hospitals throughout the United States and Canada, the introduction of regional neonatal intensive care units resulted in decreases in neonatal mortality from 25-42%. Encouraged by these decreases, the American Medical Association in 1971 urged the adoption of regionalized perinatal programs. Next, the National Foundation-March of Dimes established a joint committee on perinatal health services, which led to the publication of an extremely influential report, Toward Improving the Outcome of Pregnancy (1976). This report became the reference manual for regionalization. It established guidelines which any region could use in implementing a program and warned against duplication of effort, by citing the "staggering cost" of providing comprehensive services to high-risk infants and mothers.
Table 1. Infant Mortality: 1930, 1978, 1979
< 1 year
< 28 days
28 days to 11 months
In 1972, the Robert Wood Johnson Foundation began planning for a national program to support regionalized perinatal care.
The salvage of premature infants has greatly improved because of centrally located intensive care units to which high-risk prematures may be brought. The neonatologist is now aided by a variety of new technological innovations and biochemical testing procedures, such as amniocentesis, sonography, fetal monitoring, L/S ratio, continuous positive airway pressure (CPAP), and mechanical ventilation, which contributed to the significant decrease in neonatal mortality from 20.5 per 1,000 live births in 1950 to 8.7 in 1979, a decrease of 58%. Further, the best results from major centers indicate that less than 10% of surviving infants have major handicaps (a major handicap would prevent a child from attending a normal school, and would include definite cerebral palsy, an intelligence quotient below 70, and severe deafness or visual loss).
Neonatology, a relatively new and currently the fastest growing pediatric subspecialty, has fostered both a biomedical and social revolution during the last decade. The neonatal intensive care unit stands as the product of the highly technical revolution; while the social revolution has made us aware of the newborn's individual psyche and of the enormous importance of almost immediate maternal bonding to the infant's later development.
A new group of children have been created by the survival of infants who are born prematurely and undersized. Each year about 230,000 low-birth weight infants and about an equal number with birth defects are born; many of these would not have survived a decade ago. To reduce the numbers of such infants born each year has become one of neonatology's major challenges. To meet it, neonatologists now know that infant care must begin long before birth and in some cases even before conception. This has led to the emergence of the even newer subspecialty of fetology, and with it such new techniques in fetal diagnosis as amniocentesis, fetoscopy, and ultrasonography.
With the enormous progress made in neonatology during the past decade, and with every indication that this will continue, one can even imagine that by the century's end the neonatologist may then be looking for entirely new medical worlds to conquer.
1. Schaffer A. J.: Diseases of the Newborn. Philadelphia: Saunders, 1960, p. 1.
2. Smith C. A.: Physiology of the Newborn Infant. Springfield: Charles C Thomas, 1945.
3. v. Reuss A.: The Diseases of the Newborn (Trams.). London: J. Bale, 1920.
4. Cone T. E., Jr.: History of American Pediatrics. Boston: Little Brown, 1979, pp. 57-58.
5. Bovaird D., Jr.: The New York Foundling Hospital. Arch. Pediatr. 13:170, 1896.
6. Radbill S. X.: Pediatrics. In Debus, A. G. (ed.): Medicine in Seventeenth Century England. Berkeley: University of California Press, 1974, pp. 237-240.
7. Still G. F.: The History of Paediatrics. London: Oxford University Press, 1931, p. 26.
8. Phair T.: The Boke of Chyldren. Edinburgh & London: E. & S. Livingstone, 1955, p. 19.
9. Cadogan W.: An Essay Upon Nursing London: J. Roberts, 1748, pp. 9-14.
10. Cone T. E., Jr.: De pondere infantum natorum: The history of weighing the newborn infant. Pediatrics 28:490, 1961.
11. Vogel M. J.: The Invention of the Modern Hospital, Boston 1870-1930. Chicago: The University of Chicago Press, 1980, pp. 60-80.
12. Wertz R W., Wertz, D. C.: Lying-In: A History of Childbirth in America. New York: The Free Press, 1977, p. 133.
13. Strong R. A.: Rooming-in the newborn with its lying-in mother. Diplomate 21:301, 1949.
14. Dunham E. C.: Evolution of premature infant care. Ann. Pediatr. Fenniae. 3:170, 1957.
15. Bolt R: A.: The Mortalities of Infancy. In Abt, 1. A. (ed.): Pediatrics, vol 2. Philadelphia: Saunders, 1923, pp. 58-60.
16. Denuce P.: Berceau incubateur pour les enfants nes avant terme. J. Med. Bordeaux 2:723-724, 1857.
17. Crede C.: Ueber Erwarmungsgerathe fur friihgeborene und schwachliche kleine Kinder. Arch. Gynak.24:128-147, 1884.
18. Sterne L.: The Life and Opinions of Tristram Shandy. Baltimore: Penguin Books, 1967, p. 283.
19. Doepp P.: Bemerkungen uber einige Krankheiten der Sauglinge. Analekten caber Kinderkrh 3:150-168, 1835.
20. Marx F. F.: Die Entwicklung der Sauglings-incubatoren. Bonn, Verlag Siering, 1968, p. 28.
21. Huard P., Imbault-Huart, M.-J.: Stephane Tarnier (1828-1897) et les prematures. Gaz. Med. 79:6887, 1972.
22. Auvard A.: De la couveuse pour enfants. Arch Tocol. 10:577, 1883.
23. Hess J. H.: Premature and Congenitally Diseased Infants. Philadelphia: Lea & Febiger, 1922, pp. 170-204.
24. Lubchenco L. O., Hansman C., Dressler M., et al.: Growth as estimated from liveborn birth-weight data at 24 to 42 weeks gestation. Pediatrics 32:793, 1963.
25. Grulee C. G., Bonar B. F.: The Newborn in Clinical Pediatrics, vol. 11. New York: Appleton, 1926, pp. 139-159.
26. Griffith J. P. C.: The ability of mothers to nurse their children. J. Am. Med. Assoc. 59:1874, 1912.
27. Rubner M., Heubner O.: Die Naturliche Ernahrung eines Sauglings. J. Biol. 36:1, 1898.
28. Marchant A.: Soins a dormer aux nouveau-nes. Gaz. Med. (Paris) ser. 3, 16:824, 1851.
29. LaFetra L. E.: The hospital care of premature infants. Trans. Am. Pediatr. Soc. 28:90, 1916.
30. Botch T. M.: Infant Feeding-Weaning. In Keating, J. M. (ed.): Cyclopaedia of the Diseases of Children, Medical and Surgical vol. 1 Philadelphia: Lippincott 1889, pp. 270-329.
31. Biedert P.: Die Kinderndhrung im Sduglingsalter. Stuttgart: F. Enke, 1880, p. 50.
32. Jacobi A.: Infant Diet. New York: Putnam, 1885, p. 90.
33. Davies D. P.: The first feed of low birth weight infants: Changing attitudes in the twentieth century. Arch. Dis. Child. 53:187, 1978.
34. Hansen J. D. L., Smith C. A.: Effects of withholding fluid in the immediate postnatal period. Pediatrics 12:99, 1953.
35. Smith C. A., et al..: Adjustment of electrolytes and water following premature birth (with special reference to edema). Pediatrics 3:34, 1949.
36. Crosse V. M., et al.: The value of human milk compared with other feeds for premature infants. Arch. Dis. Child. 29:178, 1954.
37. Ylppo A.: Premature children should they fast or be fed in the first days of life? Annales Paediatriae Fenniae 1:99-104, 1954.
38. Geiss J.: Ueber Fiitterungs-und umweltbedingte Atenstorungen bei Fruhge-borenen. Zeitschriftf Kinderheilk. 76:261-268, 1955.
39. Churchill J. A.: Weight loss in premature infants developing spastic dysplegia. Obst. and Gynaec. 22:601, 1963.
40. Winick M., Rosso P.: Head circumference and cellular growth of the brain in normal and marasmic children. J. Pediatr. 74:774, 1969.
41. Smallpeice F., Davies P. A.: Immediate feeding of the premature infants with undiluted breast milk. Lancet 2:1349, 1964.
42. Drillien C. M.: The Growth and Development of the Prematurely Born Infant. Edinburgh and London: Livingstone, 1964.
43. Gordon H. H., McNamara H.: Fat excretion of premature infants: effect on fecal fat of decreasing fat intake. Am. J. Dis. Child. 62:328, 1941.
44. Royce S., Tepper C., Watson W., et al.: Indwelling polyethylene nasogastric tube for feeding premature infants. Pediatrics 8:79, 1951.
45. Babson S. G.: Peripheral intravenous alimentation for small premature infants. J. Pediatr. 79:494, 1971.
46. Buhl L.: Die acute Fettegeneration der Neugeborenen. In Hecker, C., and Buhl, L.: Klinik der Geburtskunde. Leipzig: Englemann, vol. 1, 1861, pp. 296-300.
47. Winckel F. C.: Ueber eine bisher nicht beschriebene endemisch aufgetretene Erkrankung Neugeborener. Dtsch. Med. Wschr. 5:303, 415, 431, 447, 1879.
48. Budin P.: Le Nourrisson. Paris: Octave Doin, 1900.
49. Santos J. I., Hill H. R.: Bacterial infections of the neonate. In Kelley V. C. (ed): Practice of Pediatrics. Hagerstown: Harper & Row, 1980, vol. 1, Chap. 84, p. 1.
50. Baker C. J.: Summary of the workshop of perinatal infections due to Group B Streptococcus. J. Infect. Dis. 136:137, 1977.
51. Freeman R. M., et al.: Fifty years of neonatal sepsis at Yale 1928-1978 (Abstract). Pediatric Research 14:598, 1980.
52. Blackfan K. D.: Apparatus for collecting infants' blood for the Wassermann reaction. Am. J. Dis. Child. 4:33, 1912.
53. Blackfan K. D., Maxey, K. F.: The intraperitoneal injection of saline solution. Am. J. Dis. Child. 15:19, 1918.
54. Helmholtz H. F.: The longitudinal sinus as the place of preference in infancy for intravenous aspirations and injections, including transfusion. Am. J. Dis. Child. 10:194, 1915.
55. Sidbury J. B.: Transfusion through the umbilical vein in hemorrhage of the new-born. Am. J. Dis. Child. 25:290, 1923.
56. Bonnaire E.: Inhalations of oxygen in the new-born. Arch. Pediatr. 8:769, 1891.
57. Ylppo A.: Uber Magenatmung beim Menschen. Biochem. Z. 78:273, 1917.
58. Akerren Y., Furstenberg N.: Gastrointestinal administration of oxygen in the treatment of asphyxia in the newborn. I. Obstet. Gynaecol. (Brit. Emp.) 57:705, 1950.
59. Waller H. K., Morris D.: Resuscitation of the newborn with intragastric oxygen. Lancet 2:1757, 1954.
60. James L. S., Apgar V. A., Burnard E. D. et al.: Intragastric oxygen and resuscitation of the newborn. Acta Paediatr. 52:245, 1963.
61. Lancet: (Leading article). The price of perinatal neglect 1:437, 1974.
62. Terry T. L.: Retrolental fibroplasia. Am. J. Ophthalmol. 25:203, 1942.
63. Reece A. B.: Editorial: An epitaph for retrolental fibroplasia. Am. J. Ophthalmol. 40:267, 1955.
64. Silverman W.: Medical inflation. Perspectives in Biology and Medicine 617, 1980.
65. Done A. K.: Perinatal pharmacology. Ann. Rev. Pharmacol. Ther. 6:189, 1966.
66. Diamond L. K., Blackfan K. D., Baty J. M.: Erythroblastosis fetalis and its association with universal edema of the fetus, icterus gravis neonatorum, and anemia of the newborn, J. Pediatr. 1:269, 1932.
67. Diamond L. K., Allen F. H., Thomas W. O., Jr.: Erythroblastosis fetalis. VII: Treatment with exchange transfusion. N. Engl. J. Med. 244:39, 1951.
68. Cremer R. J., Perryman P. W., Richards D. H.: Influence of light on the hyperbilirubinaemia of infants. Lancet 1:1094, 1958.
69. Freda V. J., et al.: Rh disease. How near the end? Hosp. Pract. 13:61, 1978.
70. Avery M. E., Mead J.: Surface properties in relation to atelectasis and hyaline membrane disease. Am. J. Dis. Child. 97:517, 1959.
71. Fujiwara T., et al.: Artificial surfactant therapy in hyaline membrane disease. Lancet 1:55, 1980.
72. James R. A., Cummins M., Davies P. A.: Infants of very low birth weight: a 15-year analysis. Lancet 1:1332, 1979.
73. Dunn P. M., Speidel B. D., Fleming P. J.: Intensive care and the very-lowbirth weight infant. Lancet II:254, 1979.
74. Stewart A., et al.: Prognosis for infants weighing 1000 gm or less at birth. Arch. Dis. Child. 52:97, 1977.
75. Robertson N. R. C.: Intensive care and the very-low-birth weight infant. Lancet 11:362, 1979.