While reference to premature infants is found as early as the Babylonian Talmud, the first documentation did not occur until 1861 with the work of British pediatrician, WJ. Little.1 Sincethat time, numerous investigators have traced the outcome of low birthweight infants with a view toward understanding their development and assessing the impact of neonatal intensive care practices. During the period from the 1920s to the 1950s, most infants with birthweights less than 1500 g died; but today, the maj ority of these infants survive. This article will evaluate the current status of very low birthweight infants (^ 1 500 g), as reported in most recent follow-up studies and consider the significance of these findings for perinatologists and for pediatricians who provide primary medical care for small, premature infants surviving the neonatal period.
A review of outcome studies on infants born over the past 20 years indicates a significant decline in the mortality of very low birthweight infants. In the 1960s and early 1970s, the survival rate of infants with birthweights between 1000 g and 1500 g increased greatly. Since 1975, there has been an important rise in the percentage of survivors who are less than 1000 g (extremely low birthweight).
The vast majority of follow-up studies have shown that low birthweight infants are far more likely to sustain sensorimotor and intellectual handicapping conditions than healthy, full-term infants.2"4 Furthermore, it is practically axiomatic that the smaller the infant at birth, the poorer his potential for normal development.
Therefore, as each succession of studies has proclaimed increased survival in groups of infants of lower and lower birthweight, a fundamental and socially relevant question emerges:
Is increased survival in very low birthweight infants associated with proportional declines in morbidity (ie, handicaps); or, does saving increased numbers of small infants indicate that more low birthweight infants than ever survive with handicapping conditions?
Perhaps the best means of addressing this question is through review of changes in mortality and morbidity among very low birthweight infants since the early 1960s. The task of comparing results from follow-up studies performed at different neonatal centers and at varying periods is laden with problems. A major difficulty lies in the fact that different study populations vary on major variables found to affect survival and morbidity: whether the infant is appropriate for dates or small for dates; if he was transported or in-born; whether his parents are low income or high income. Furthermore, it is not always possible to determine the impact of such factors on a specific population, since many studies do not describe their populations in sufficient detail to do so. Another problem arises from a lack of consistency in definition of morbidity. Thus, some studies define mental deficiency as an intelligence quotient (IQ) or developmental quotient (DQ) of less than 85; while others use scores of 70 c» 80; and still others regard need for special schooling as the definition of intellectual impairment. Comparability on rate of morbidity is also impeded by the fact that the age at which outcome is assessed varies from investigation to investigation. However, in so far as consistency in findings occurs among studies of very low birthweight infants born during the same years, it may be possible to form generalizations regarding the mortality and morbidity of neonates for that time period.
OUTCOME OF VERY LOW BIRTHWEIGHT INFANTS BORN FROM 1960 TO 1975
Premature nurseries throughout the United States have treated sick, low birthweight infants since the 1920s. However, transition to the modern neonatal intensive care unit (NICU) did not occur until the 1960s when major strides in understanding infant physiology and sophisticated instrumentation developed through bioengineering were first applied to the clinical care of the immature newborn. Mechanical ventilators first became available to create constant distention of airway pressure and the development of umbilical-arterial catheters simultaneously enabled close monitoring of oxygen administration, so that infants were neither over- nor under-oxygenated as in previous years. Early provision of intravenous fluids and caloric supplementation, optimal control of the neonate's thermal environment, earlier intubation and ventilation, and active treatment of hyperbilirubinemia with phototherapy and exchange transfusions were, similarly, outgrowths of scientific and technological advances during this period. The subsequent costs of equipment and of highly trained medical personnel necessitated the formation of tertiary care centers and the system of regionalized neonatal intensive care, as we know it, began to develop.
Reflecting the new advances just described, outcome and demographic data on infants born in the 1960s and early 1970s show a significant decline in mortality of very low birthweight infants, in contrast to the two previous decades. Controlling for changes in birth rate, Lee and his associates5 found no improvement in mortality of infants in any 500 g weight category under 2500 g from 1 950 to 1960 and little change in mortality of infants more than 2500 g. However, they noted a significant decline in neonatal mortality during the period from 1965 to 1975, attributable to improvements in treating low birthweight infants. Reviewing mortality of low birthweight babies born in England and Wales from 1953 to 1 979, Pharoah et al6 also found increased survival in low birthweight infants after the mid-1960s, particularly for those in weight groups of 1000 g to 1499 g and 1500 g to 2000 g.
Findings presented by a number of single NICUs also point to decline in mortality of very low and extremely low birthweight children during the 1960 to 1975 period.7 As shown in Table I, 55% of infants under 1500 g survived the neonatal period, including a 70% survival rate for infants between 1000 g and 1500 g. When data were reported for infants under 750 g and a 20% survival rate for infants under 1000 g findings were uniformly discouraging, ranging from 0% to 3% survival.
In contrast to the 1 940s and 1 950s, morbidity decreased significantly during the period from 1960 to 1975. Earlier follow-up work on babies born in the late 1 940s and 1 950s revealed a shockingly high incidence of handicap among very low birthweight babies. Developmental impairment, including central nervous system (CNS) dysfunction (eg, cerebral palsy, hydrocephalus), mental retardation, hearing loss, and blindness due to retrolental fibroplasia (RLF) ranged from 40% to 70%.8"10 Reporting on 63 infants born between 1947 and 1953 with birth weights less than 1500 g, Lubchenco8 found that 25% had IQs less than 80; 48% sustained CNS damage; and 25% developed permanent visual impairment due to RLF. In contrast, representative studies of infants born between 1960 and the early 1970s show the morbidity rate to have dropped to about 10%. "~12 For example, at the University College Hospital in London,11 investigators found that of 95 very low birthweight survivors born between 1966 and 1970, only one suffered from visual deficits due to cataracts, 8% had IQs less than 80; and 4% had central nervous system handicaps. Surveying nearly 20 outcome studies performed worldwide, Stewart, Reynolds, and Lipscomb13 concluded that from 1960 onward, the proportion of healthy survivors tripled, while the proportion of handicapped children remained constant at a figure of 6% to 8%.
Thus, with major advances in neonatal intensive care, outcome of very low birthweight infants was improved considerably, as reflected in reports of decreased mortality rates and proportionally even greater declines in morbidity rates.
OUTCOME OF VERY LOW BIRTHWEIGHT INFANTS BORN FROM 1976 TO 1978
Just as the 1960s were heralded as a period of major improvement in provision of medical care for low birthweight neonates, so the middle 1970s have been viewed as the beginning of another era in neonatology. Truly "perinatal" in orientation, the neonatal intensive care unit has now become an integral component of a network of referring community hospitals, pediatric subspecialists, and pediatric practitioners concerned with recognizing and maximizing conditions of high-risk pregnancies prior to delivery and treating the low birthweight infant most adventitiously following delivery.
ADMISSION AND SURVIVAL OF VERY LOW BIRTHWEIGHT INFANTS IN REPRESENTATIVE NEONATAL INTENSIVE CARE CENTER
The development and implementation of maternal-fetal transports, fetal monitoring, pharmacologic inhibition of premature labor, caesarean section deliveries, and preparation for prompt resuscitation of the newborn all represent a new collaboration of obstetrical and pediatric efforts. Further refinement of neonatal care represented in increased attention to the nutritional requirements of the tiny infant, further miniaturization of equipment, and availability of diagnostic tools such as ultrasonography and CAT scans are also viewed as positive contributions to the development of perinatology. Perhaps, even more important, has been the advent of routine use of continuous positive airway pressure (CPAP) in the ventilation of very immature newborns, making it possible to keep many more of these infants alive.
Along with these newer changes in perinatal care, observers once again noted an increased survival in low birthweight and, particularly, extremely low birthweight infants. Survival for infants between 1000 g and 1500 g increased to 85% around 1976, in comparison to 55% for the 1960 to 1975 period. Survival for infants less than 1000 g jumped to 40% versus 20% in the earlier period (Table 1). Thus, survival rate for infants between 1000 g and 1500 g increased by 15%, but it doubled for infants less than 1000 g. Although increased survival among the extremely low birthweight group largely reflects improvements with newborns between 750 g and WOOg, very recent work has even begun to report increased survival (8% to 20%) among "fetal" infants, less than 800 g and 26 weeks estimated gestational age.14 However, a breakdown of survival by 50 g weight groups still indicates that infants less than 700 g at birth are, by and large, not viable.
At this point it is appropriate to reconsider the question initially posed. Are the newest advances in saving the very smallest infants also increasing the number of children with handicapping conditions? As previously suggested, the answer to this question must be couched in tentative terms, owing to the paucity of satisfactory data on mortality rates, the scarcity of information regarding incidence of handicaps, and difficulties in interpreting this information when it is available due to variability among the low birthweight populations studied.
With these caveats in mind, a comparison of outcome studies of very low birthweight infants born between I960 and 1975 and those born between 1976 and 1978 suggests that latest efforts to save the smallest infants has increased the number of NICU graduates with developmental impairment. The morbidity rates of all infants less than 1500 g appears to have increased only slightly, if at all (Table 2). However, the rates of handicap among infants less than 1000 g appear to have risen about 15% between the 1960 to 1975 and 1976 to 1978 periods.
Even if morbidity rates for all infants with birth weights less than 1500 g had not increased during the past 20 years, the mere increase in survival rates for this group suggests that the number of children with handicaps has risen. Assuming a consistent annual birth rate of 40,000 very low birthweight infants* from 1960 to 1980, with a survival rate of 55% for the earlier period ( 1960 to 1975) and 70% for the later period (1975 to 1980), 22,000 survivors less than 1500 g would have resulted each year in the 1960 to 1975 period in contrast with 28,000 such infants annually since 1975. Positing an average annual morbidity rate of 20% nationwide between 1960 and the present, one would then expect at least 1200 additional handicapped infants to be produced by the neonatal intensive care system in the United States each year after 1975 than in each of the 15 prior years.
As modern neonatal intensive care techniques have made it possible to create survivors out of smaller and smaller infants, neonato logists, pediatricians, ethicists, and society as a whole have begun to question whether there are lower limits on the weight and gestation of infants who can be saved and whether the price of saving such infants may be too great. Most often, this question has been directed to the issue of whether the costs of neonatal intensive care are justified for tiny infants whose chances for intact survival are relatively small.19 One recent article suggests that 700 g is the lower limit for viability and that intensive care should not be attempted with infants below that weight.14
The question of costs must be extended beyond the doors of the neonatal intensive care unit, however, to the emotional and monetary burdens of caring for a handicapped child once he leaves the hospital. A recent study by the Rand Corporation20 estimates that it costs about $5,000 annually to provide special education for a mentally retarded child and $10,000 for a child who is functionally blind. Over a 20-year period, this would cost society $100,000 to $200,000 per child. Perhaps with these concerns in mind, some tertiary care centers practice judicious (although some would argue too liberal) withdrawal of treatment for extremely low birthweight infants. Centers in Canada and England report withdrawal of care for newborns who fail to respond to therapy, show clinical signs of massive intracranial hemorrhage, have severe pulmonary hemorrhage, or are moribund on admission.14'21 Clearly such practices reduce the number of survivors as well as the incidence of handicap among infants of extremely low birthweight.
RATE OF HANDICAP IN REPRESENTATIVE STUDIES OF INFANTS
One wonders, however, about the outcome of such infants in neonatal intensive care units where tiny babies are managed far more aggressively and where survival rates of infants at greatest risk are, therefore, higher.
In sincere and well-intentioned efforts to extend the possibilities of life to even the smallest of newborns, neonatologists point enthusiastically to the increased numbers of healthy survivors among tiny babies. Reporting on series of 37 infants between 700 g and 800 g whp would not have survived had they been born ten years earlier, Britton et al concluded that "every effort should be made in the care of. . .(infants) 700 g to 800 g," since about 50% were developing normally at 18 months of age."14 However, it must also be noted that the remaining 50% of these infants were classified as abnormal, with developmental quotients two to four standard deviations below average. Thus, for every infant between 700 g to 800 g who was judged to be normal, there was another found to be abnormal. Thus, the question ultimately becomes one of whether the contributions of a larger number of healthy individuals saved by neonatal intensive care outweigh the burden of maintaining or remediating a smaller, but increased, number of individuals with handicaps. As Pomerance and his associatesconcluded, "Society. . .must be the ultimate judge. . .as it must pay the bill and reap the benefits and the heartaches as well."'
In fairness, the still nascent field of neonatology continues to achieve new frontiers, both in sustaining life and in ensuring normalcy among survivors. It remains to be seen whether medical technology and knowledge necessary for producing healthy survivors will "catch up" with practices already used to create survival. The long range view is somewhat discouraging. A number of investigations have clearly shown that developmental disorders in premature infants are significantly related to the degree of perinatal asphyxia and the need for assisted ventilation during hospitalization.22'23 Reviewing the outcome of a group of infants less than 1000 g, all of whom required assisted ventilation, one center found that 44% had either severe neurological sequelae and /or low psychometric scores, in comparison to 10% of infants of similar weight who did not require mechanical ventilation.24 By virtue of their extreme immaturity the very smallest babies are most likely to suffer from perinatal asphyxia and to require assisted ventilation; and by deduction, are most likely to sustain handicaps. Neonatologists have concluded that morbidity among the very low birthweight population will be reduced only when perinatal advances enable the prevention of asphyxia immediately following delivery and in the NICU.22'23 In so far as it is not yet possible to accomplish these tasks, we may continue to be faced with higher numbers of handicapped graduates of neonatal intensive care.
Unquestionably, the number of all infants less than 1500 g is small; and of those, the number of infants less than 1000 g is even smaller (40,000 and 18,340, respectively, out of 3.5 million live births annually).' s~18 However, they comprise a disproportionately high percentage of children with handicaps.4'25
The answer to the ethical question - Is saving tiny babies worth it? - is surely a complicated one and certainly not to be decided here. However, it is clear that as more infants with very low birthweights enter the pediatric population, pediatricians must become more expert in recognizing early signs of developmental disorders and in monitoring the development of these high-risk children. The American Academy of Pediatrics, aware of the necessity for this expertise, has funded training courses on diagnosis of developmental disabilities for pediatricians throughout the country. Routine physical evaluation is usually not sufficient for such children, and more time may be required in their assessments and in discussion with their parents than for the typical well-child. The private pediatrician may also find that he/she will have to more closely coordinate his/her efforts with those of follow-up clinics, specialists, and remediative programs in the community. In this era of increasingly sophisticated neonatal intensive care, the general pediatrician is in the position of reaping what the obstetrician and neonatologist sow and must develop and utilize his/her expertise in treating this harvest.
1. Little WJ: On the influence of abnormal parturition, difficult labours, premature birth, and asphyxia neonatorum, on the mental and physical condition of the child, especially in relation to deformities. Transactions of the Obstetrics Society of London 1861-1862: 3:293.
2. DeHirsch K, Jansky L. Langford WS: Comparisons between prematurely and maturely born children at three age levels. AmJ Orthopsychiatry 1966: 36:616-628.
3. Caputo DV, Mandell W: Consequences of low birthweight. Developmental Psychology 1970; 3:363-383.
4. Fisch RO, Graven HJ. Engel RR: Neurological status of survivors of neonatal respiratory distress syndrome. J Pediatr 1968: 3:395-403.
5. Lee K, Panefh N, Gartner LM, et ai: Neonatal mortality: An analysis of recent improvement in the United States. Am J Public Health 1980: 70:1518.
6. Pharoah POD. Albcrman ED: Mortality of low birthweight infants in England and Wales 1953 to 1979. Arch Dis Child 1981; 56:86-89.
7. Thompson T. Reynolds J: The results of intensive care therapy for neonates. J Perirmi Med 1977; 5:59-75.
8. Lubchenco LO, Horner FA, Reed LH, et al: Sequelae of premature birth. Am J Dis Child 1963; 106:135-149.
9. Drillien CM: The incidence of mental and physical handicaps in school-age children of very low birthweight. Pediatrics 1961; 27:452-464.
10. Dann M, Levine SZ, New EV: The development of prematurely bom children with birthweights or minimal postnatal weights of 1.000 grams or less. Pediatrics 1958; 22:1037-1052.
11. Stewart AL, Reynolds EOR: Improved prognosis for infants of very low birthweight. Pediatrics 1974; 54:724-735.
12. Francis-Williams J, Davies P. Very low birthweight and later intelligence. Dev Med Child Neurol 1974; 16:709-728.
13. Stewart AL. ReynoldsEOR, Lipscomb AP: Outcomefor infants of very low birthweight: Survey of world literature. Lancet. 1981; 1:1038-104).
14. Britten SB, Fitzhardinge PM, Ashby S: Is intensive care justified for infants weighing fcss than 801 gm at birth? J Pediatr 1981; 99:937-943.
15. Vital statistics of the US. 1967. vol I: Natality. US Dept of Health. Education, and Welfare (PHS), National Center for Health Statistics. Washington. 1970.
16. Vital statistics of the US. i970. vol I: Natality. US Dept of Health. Education, and Welfare (PHS), National Center for Health Statistics. Rockvillc, Maryland, 1975.
17. Vital statistics of the US. 1975. vol 1: Natality. US Dept of Health, Education, and Welfare (PHS), National Center for Health Statistics. Hyattsville, Maryland, 1978.
18. Advanced report of final natality statistics, 1979. Monthly Vital Statistics Report. September 1981, vol 30, Suppl 2.
19. Pomerance JJ, Ukrainski CT, Ukra T, et al: Cost of living for infants weighing 1000 grams or less at birth. Pediatrics 1978; 61:908-910.
20. Kakalik J, Furry M.Thomas M, et al: The Cost of Special Education. Rand Report No. N-1792-ED, Rand Corporation, Santa Monica. California. 1981.
21. Stewart AL, Turcan DM, Rawlings G. et al: Prognosis for infants weighing 1000 grams or less at birth. Arch Dis Child 1977; 52:97-104.
22. Driscoll JM. Driscoll VT. Steir ME. et al: Mortality and morbidity in infants less than 1001 grams birthweight. Pediatrics 1982; 69:21-26.
23. Fitzhardinge PM. Pape K, Arstikaitis M, et al: Mechanical ventilation of infants less than 1501 grams birthweight: Health, growth and neurologic sequelae. J Pediatr 1976; 88:531-541.
24. Fitzhardinge PM: Follow-up studies in infants treated by mechanical ventilation. Clin Perinatol 1978; 5:451-459.
25. Nelson KB, Ellenberg JH: Neonatal signs as predictors of cerebral palsy. Pediatrics 1979: 64:225-232.
26. Jones RA, Cummins M, Davies PA: Infants of very low birthweight. Lancet 1979;2:1332-1335.
27. Alden ER, Mandelkorn T, Woodrum DE, et al: Morbidity and mortality of infants weighing less than 1000 grams in an intensi ve care nursery. Pediatrics 1972; 50:40-7.
28. Fitzhardinge PM: Early growth and development in low birthweight infants following treatment in an intensive care nursery. Pediatrics 1975; 56: 162172.
29. Hack M, Fanaroff AA. Merkatz IR: The low birthweight evolution of a changing outlook. N Engl J Med 1979: 301:1 162-1 165.
30. Saigal S, Roscnbaum P, Stoskopf B, and Milner R: Follow-up of infants 501 to 1 500 gm birthweight delivered to residents of a geographically defined region with perinatal intensive care facilities. J Pediatr 1982; 100:606-613.
31. Roberton NRC: Intensive care and the very low birthweight infant. Lancet 1979: 1:362.
32. Schechner S: For the 1980s: How small is too small? Clinics in Perinatology 1980:7:135-143.
33. Rothberg AD. Maisels MJ. Bagnato S, et al: Outcome for survivors of mechanical ventilation weighing less than 1250 gm at binh. J Pediatr 1981; 98: 106-11 1.
34. Kumar SP, Anday EK. Sacks LM. et al: Follow-up studies of very low birthweight infants ( 1250 grams or less) born and treated within a perinatal center. Pediatrics 1980; 66:438-*44.
35. Shennan AT, Milligan JE: The growth and development of infants weighing 1000 to 2000 grams at birth and delivered in a perinatal unit. Am J Obstet Gynecol 1980: 136:271-275.
36. Pape KE. Buncic RJ, Ashby S, et al: The status at two years of low birthweight infants bom in 1974 with birthweights of less than 1001 gm. J Pediatr 1978:92:253-260.
37. Stewart AL. Turcan DM. Rawlings G. et ab Arch Dis Child 1977; 52:97-104.
38. Dweck HS, Saxon SA. Benton JW. Cassady G: Early development of the tiny premature infant. Ar» J Dis Child 1973; 126:28-34.
39. Bhat R. Raju TNK. Vidyasagar D: Immediate and long-term outcome of infants less than 1000 grams. Crii Care Med 1978; 6:147-150.
ADMISSION AND SURVIVAL OF VERY LOW BIRTHWEIGHT INFANTS IN REPRESENTATIVE NEONATAL INTENSIVE CARE CENTER
RATE OF HANDICAP IN REPRESENTATIVE STUDIES OF INFANTS