Avroy A. Fanaroff
Infant mortality is an important outcome measure of the health services of a population. In the United States, where there are approximately 4 million births each year, the infant mortality is around 7 per 1000 live births. The highest risk of infant death is within 24 hours of birth, but mortality and morbidity remain high during the neonatal period, from birth to the 28th day of life. In the United States each year, nearly 1% of pregnancies are complicated by fetal death and about 0.5% by neonatal mortality.1-8 The fetus and newborn are most vulnerable during labor, delivery, and the neonatal period because central nervous system injury may result in lifelong morbidity and neurodevelopmental impairment. The perinatal period, from 28 weeks of gestation to the 28th day of life, is the period of greatest mortality. In the modern era, with survival of extremely-low-birth-weight infants, postneonatal mortality also contributes significantly to the infant mortality rate.
The reduction of maternal and infant mortality and the improvement of the health of mothers and infants in the United States are high priorities. Statistical comparisons among countries, states, regions, and individual centers have been hampered by differences in definitions. In order to compare outcomes and plan interventions, it is imperative that standard definitions be utilized9:
Appropriate for gestational age (AGA): An infant with a birth weight between the 10th and 90th percentiles for that gestational age. Those below the 10th percentile are regarded as small for gestational age (SGA), whereas those above the 90th percentile are considered large for gestational age (LGA).
Birth weight: The weight of a neonate determined immediately after delivery or as soon thereafter as feasible, expressed to the nearest gram.
Fetal death: Death before the complete expulsion or extraction from the mother of a product of human conception, fetus and placenta, irrespective of the duration of pregnancy; the death is indicated by the fact that after such expulsion or extraction, the fetus does not breathe or show any other evidence of life, such as beating of the heart, pulsation of the umbilical cord, or definite movement of voluntary muscles. Heartbeats are to be distinguished from transient cardiac contractions; respirations are to be distinguished from fleeting respiratory efforts or gasps. This definition excludes induced termination of pregnancy.
Gestational age: The number of weeks that have elapsed between the first day of the last normal menstrual period (not the presumed time of conception) and the date of delivery, irrespective of whether the gestation results in a live birth or a fetal death.
Infant death: Any death at any time from birth up to, but not including, 1 year of age.
Live birth: The complete expulsion or extraction from the mother of a product of human conception, irrespective of the duration of pregnancy, which, after such expulsion or extraction, breathes or shows any other evidence of life, such as beating of the heart, pulsation of the umbilical cord, or definite movement of voluntary muscles, whether or not the umbilical cord has been cut or the placenta is attached. Heartbeats are to be distinguished from transient cardiac contractions; respirations are to be distinguished from fleeting respiratory efforts or gasps.
Low birth weight: Any neonate, regardless of gestational age, whose weight at birth is less than 2500 g.
Neonatal death: Death of a liveborn neonate before the neonate becomes 28 days old.
Perinatal death: Death from the 28th week of gestation to the 28th day of life. Perinatal mortality measures should be based on specific weight rather than gestational age, and indices of perinatal mortality combine fetal deaths and live births with only brief survival (up to a few days or weeks) on the assumption that similar factors are associated with these losses.
Postterm: Any neonate whose birth occurs from the beginning of the first day (295th day) of the 43rd week following the onset of the last menstrual period.
Preterm: Any neonate whose birth occurs through the end of the last day of the 37th week (259th day) following the onset of the last menstrual period.
Term: Any neonate whose birth occurs from the beginning of the first day (260th day) of the 38th week through the end of the last day of the 42nd week (294th day) following the onset of the last menstrual period.
Regionalization of perinatal care to various levels (see Table 41-1), first introduced in the 1970s, was cost effective and reduced morbidity and mortality. Market forces and economics disrupted regionalization in the 1990s. Nonetheless, the evidence continues to demonstrate that the best outcomes for low-birth-weight infants are achieved when they are delivered at the larger subspecialty centers (formerly known as level III), those with an average neonatal intensive care unit (NICU) daily census in excess of 15. There has been a large increase in both the number of NICUs in community hospitals and the complexity of the cases treated in these units. The mortality among very-low-birth-weight infants was lowest for deliveries that occurred in hospitals with NICUs that had both a high level of care and a high volume of such patients, implying that increased use of such facilities might reduce mortality among very-low-birth-weight infants. Risk-adjusted neonatal mortality for infants born in smaller level III NICUs and in level II+ and level II NICUs (specialty), regardless of size, was not significantly different from that in hospitals without a NICU and was significantly higher than in hospitals with large subspecialty NICUs.10,11
Despite the differences in outcomes, costs for the birth of infants born at hospitals with large subspecialty NICUs were not more than those for infants born at other hospitals with NICUs. Concentration of high-risk subspecialty NICU care has the potential to decrease neonatal mortality without increasing costs.
Infant and neonatal mortality rates are presented in Figure 41-1. Advances in perinatal care have improved the chances for survival of infants with major congenital anomalies in addition to those with cardiorespiratory disorders and other major organ system failures. The outlook for extremely-low-birth-weight and lowgestational-age infants has also improved remarkably.12,13 Many factors influence neonatal mortality. In addition to race, birth weight, gestational age, gender, place of delivery, and intrauterine growth, there are wide variations in population descriptions, in the criteria used for starting or withdrawing treatment, in the reported duration of survival, and in care. The perinatal mortality rate in the United States has consistently declined with an overall decrease of 10% from 1990 to 2003. In 2002 to 2003, the perinatal mortality rate reached its nadir of 6.74 deaths per 1000 live births and fetal deaths. The recent decline is attributable to a drop in late fetal deaths. In 2004, the US fetal mortality rate was 6.20 fetal deaths of 20 weeks of gestation or more per 1000 live births, and fetal deaths were greatest in teenagers, mothers over age 35, and those with multiple fetuses. Race played a major role, and the rate in non-Hispanic black women (11.25 per 1000) far exceeded non-Hispanic white women (4.98 per 1000).
Table 41-1. Levels of In-Hospital Perinatal Care
Fetal and perinatal mortality rates have declined slowly but steadily from 1990 to 2004. Fetal mortality rates for 28 weeks of gestation or more have declined substantially, whereas those for 20 to 27 weeks of gestation have not declined. In 2004, one half of fetal deaths of 20 weeks of gestation or more occurred between 20 and 27 weeks of gestation.6
Hamilton7,8 noted that the neonatal mortality rate had declined steadily until 2001 to 2002, when it increased from 4.5 to 4.7 deaths per 1000 live births. This increase was primarily driven by the birth rate of infants with birth weights below 750 grams and particularly by those with weights below 500 grams aided and abetted by assisted reproduction techniques, which increased the number of multiple births, premature deliveries, and extremely-low-birth-weight infants. In 2004, the neonatal mortality rate declined back to 4.52 per 1000 live births.4
Factors that increase mortality rates several fold include prematurity, no prenatal care, inadequate weight gain in pregnancy, African American ethnicity, and inadequate prenatal care. The importance of race as a determinant of neonatal mortality is shown by the 1997 mortality rates in the United States for infants born to Asian and Pacific Islander mothers (5:1000 live births) followed by white (6.0), American Indian (8.7), and black (13.7) mothers. African American women have many more preterm and very early preterm births, which, in part, explain the doubling of their fetal and infant mortality rates. In general, however, the racial discrepancies in preterm birth and other pregnancy outcomes remain unexplained. Although there has been a noticeable decline in the perinatal mortality rates for the past few decades, the prematurity rate has remained fairly constant, and African American infants continue to have a higher mortality rate than their white counterparts. The relative differences in perinatal and neonatal mortality between different racial/ethnic populations have not substantially changed. Non-Hispanic black newborns, with an infant mortality rate (IMR) of 13.6 per 1000 live births, are twice as likely as non-Hispanic white (IMR 5.7:1000) and Hispanic-black infants (IMR 5.65:1000) to die within the first year of life. Although for the total birth cohort, blacks have a higher mortality risk than whites, the reverse is noted at the lower weight group/gestational age distributions.
Infant mortality rates are higher for pregnancies in which prenatal care is initiated after the first trimester of pregnancy and in infants born to teenagers or to women 40 years of age or older who did not complete high school, were unmarried, or smoked during pregnancy. Infant mortality is also higher for male infants, multiple births, and infants born preterm or at low birth weight. In many instances, the precipitating cause of preterm delivery remains undetected, but risk factors for premature birth include uterine abnormalities, placental bleeding including abruptio placenta associated with cocaine use, maternal chronic illnesses, multifetal gestation, premature rupture of the membranes, chorioamnionitis, and bacterial vaginosis. Bacterial vaginosis, a short cervix, and the presence of fetal fibronectin in the vaginal tract are predictors of preterm delivery, but treatment of the bacterial vaginosis has been ineffective in preventing prematurity. Premature delivery complicates over 10% of births but contributes disproportionately to at least two thirds of the infant deaths and to a significant amount of neonatal and long-term morbidity, which may include cerebral palsy, mental restriction, physical handicap, blindness, and deafness in addition to major and minor school adaptive and learning problems.14-16 Although there has been a substantial decline in the number of medically preventable deaths and deaths from respiratory distress syndrome, the number of deaths from extremely low birth weight has increased relative to other causes; asphyxia, birth trauma, early-onset sepsis, and meconium aspiration syndrome have been reduced to a minimum. Nonetheless, congenital malformation is the leading cause of infant death in the United States and accounts for a much greater proportion of infant mortality than does premature birth. To further reduce neonatal mortality, the incidence of lethal congenital malformations and very-low-birth-weight infants must be addressed; congenital anomalies cause approximately 23% of infant mortality, and short gestation and low birth weight, about 15%.
FIGURE 41-1. Fetal (FMR; fetal deaths per 1000 live births plus fetal deaths), early fetal (EFMR; early fetal deaths [20–27 weeks’ gestation] per 1000 live births plus fetal deaths), late fetal (LFMR; late fetal deaths [28 weeks’ gestation] per 1000 live births plus fetal death), infant (IMR; infant deaths per 1000 live births), neonatal (NMR; neonatal deaths per 1000 live births), and postneonatal (PNMR; postneonatal deaths per 1000 live births) mortality rates: United States, 1990–2004 (final). (Hamilton BE, et al. Pediatrics 2007;119:345–360.)
Perinatal morbidity and mortality are significantly increased in multiple gestation, and the incidence of severe handicap is increased in survivors of multiple gestation, predominantly because of preterm delivery. In the United States, the number and rate of multiple births have risen dramatically. An ever-increasing number of multiple births are related to infertility treatments; the number of triplet and higher-order multiple births jumped 16% between 1996 and 1997, contributing to the increase in the percentage of low-birth-weight infants. Multiple gestation accounts for 26% of deliveries with birth weight below 1500 g. In addition to multiple gestation, aggressive interventions for fetal compromise identified in extremely immature infants contribute to the continuing toll of preterm birth.
CAUSES OF DEATH
Congenital anomalies, disorders relating to short gestation and low birth weight, and sudden infant death syndrome accounted for nearly half of all infant deaths in the United States in 1997.1 Other leading causes of death in the neonatal period include respiratory distress syndrome, newborn complications of pregnancy, newborn complications related to placental disorders, cord accidents, membrane disorders, neonatal infections, intrauterine hypoxia, and birth asphyxia. Advances in neonatal and perinatal care reduced the infant mortality rate due to respiratory distress syndrome from 156.2 per 100,000 live births and 12% of infant deaths in 1979, to 21.3 per 100,000 live births in 2004.1,4
“In the United States in the last 2 decades, despite increasing availability of prenatal care, nutrition supplementation programs, and drugs to stop preterm contractions, the preterm birth rate has increased from 9.5% in 1980 to 11% in 1998.”15 The rate of preterm births in 2004 increased to 12.5%, and the percentage of low-birth-weight births (< 2500 g) increased to 8.4% in 2004, up from 6.7% in 1984. Part of this increase is due to multiple births associated with infertility treatments, but many preterm births occur spontaneously. Multiple births accounted for 3.3% of all births in 2003 but account for at least 25% of births below 1.5 kilograms. The medical or public health strategies used to reduce preterm birth have in general not succeeded.15 However, for women with a previous history of preterm birth, progesterone therapy has reduced the rate of prematurity. Because of their high risk of mortality and serious morbidity, most studies on prematurity (defined as birth at less than 37 weeks) have focused on very preterm infants (birth at < 32 weeks). However, risks for infant death from all causes among singletons born at 32 through 33 gestational weeks were increased over 6-fold in the United States and 15-fold in Canada, whereas among singletons born at 34 through 36 gestational weeks, the relative risks were increased at least 3-fold. Mildly and moderately preterm-birth infants—now referred to as late preterm births—are more numerous than extremely-low-birth-weight infants and are responsible for an important fraction of infant deaths. Intense focus has been placed on this population because much of their morbidity and mortality may be prevented.16,17
There has been little to no success at reducing the incidence of preterm birth, intrauterine growth restriction, and congenital malformations. However, over the past 20 years, the survival of extremely-low-birth-weight infants (< 1500 g) has increased from 74% to 85%.12,13 Survival without major neonatal morbidity, including bronchopulmonary dysplasia, intraventricular hemorrhage, and necrotizing enterocolitis, is 70%. From 1997 to 2002, birth weight–specific survival was 55% for infants weighing 501 to 750 g, 88% for 751 to 1000 g, 94% for 1001 to 1250 g, and 96% for 1251 to 1500 g. More females survived.18 These numbers are very comparable to those in the Vermont-Oxford Network, which accumulates data on more than 20,000 low-birth-weight infants from 1995 to 1996 (eFig. 41-1 ) and to more recent findings (Fig. 41-2). Similar improvements are noted when the data are analyzed by gestational age.12,13 A comprehensive prospective geographically based study of all births between 20 and 25 weeks of gestation from the United Kingdom and Ireland documented mortality, neonatal morbidity, and early neurodevelopmental outcomes consistent with the data from the United States. This cohort was followed to age 6 years, at which time they demonstrated considerable neurodevelopmental disability.19
Although the validity of the Apgar scoring system continues to be challenged, a 5-minute Apgar score below 4 remains a better predictor of mortality than does severe metabolic acidosis (pH < 7.0) measured from the cord blood. Male sex, failure to receive antenatal steroids, persistent bradycardia at 5 minutes, hypothermia, and poor intrauterine growth all independently increase the risk for death.
FETAL GROWTH RESTRICTION
The most common causes of inadequate fetal growth relate to maternal hypertension, malnutrition, and smoking. These disorders may be associated with intrauterine fetal demise, neonatal adaptive problems including severe hypoglycemia, or long-term abnormalities of growth and neurodevelopment. Cigarette smoking remains a major cause of intrauterine growth restriction, preterm birth, fetal and neonatal deaths, and sudden infant death syndrome. Alcohol and drug use may also affect pregnancy outcome. Prenatal alcohol exposure is an important cause of fetal growth restriction, including microcephaly, which results in long-term growth failure in addition to substantial neurodevelopmental delay and mental restriction. Mortality and morbidity are increased among infants born at term whose birth weights are at or below the third percentile for their gestational age. The incidence of intubation at birth, seizures during the first day of life, and sepsis were also significantly increased among term infants with birth weights at or below the third percentile.
FIGURE 41-2. Mortality by birth weight, gestational age, and gender. (Fanaroff AA, Stoll BJ, Wright LL, et al; NICHD Neonatal Research Network. Trends in neonatal morbidity and mortality for very low birthweight infants. Am J Obstet Gynecol. 2007; 196:147.e1–8.)
Gender is a major determinant of survival with the female advantage most noticeable at the youngest gestational ages (Fig. 41-2). For example, the mortality rate of a girl who weighs 700 g at 24 weeks of gestation is 30% to 40%, whereas a boy of equal weight and gestational age has a mortality risk of 50%. In addition to a higher mortality than girls, boys are more likely to need cardiopulmonary resuscitation at delivery and are at greater risk for most adverse neonatal outcomes, including chronic lung disease, intracranial hemorrhage, and nosocomial infections.
Inspection of Figure 41-2 makes it apparent that there is a wide spread in the mortality risk at a given gestation and that the larger infants at each gestational age have a greater chance of surviving. Similarly, there is variability in the morbidity at each gestational age, and babies in the bottom percentiles for weight are more likely to be acidotic, require admission to the intensive care unit, develop major morbidities including respiratory failure, or die (eFig. 41-2 ).
FACTORS REDUCING MORBIDITY
During the past decade, mortality has been reduced by the introduction of surfactant therapy for respiratory distress syndrome in 1990 and the widespread use of antenatal steroids that followed the National Institutes of Health consensus conference in 1994. Before that time, only 20% of women who delivered infants with a birth weight of 1500 g or lower received antenatal corticosteroids, whereas that number now exceeds 80% in most centers in the United States and may be even higher in other parts of the world. Care of infants with birth weights of 750 g or lower has become more aggressive so that the rate of cesarean section has increased from 17% to 44%, and that of delivery room intubation from 54% to 82%.13
Some reports indicate that the survival chances for infants between 501 and 1500 g at birth improved in the 1990s, but concerns about morbidity and the high rates of neurodevelopmental handicap persist. The incidence of bronchopulmonary dysplasia (defined as receiving supplemental oxygen at 36 weeks of postmenstrual age) (22%), proven necrotizing enterocolitis (7%), severe intraventricular hemorrhage (grade III or IV) (11%), and retinopathy of prematurity (76%) among infants with birth weights between 501 and 1500 g fluctuated only slightly between 1991 and 2002. Growth failure (weight < 10th percentile) at 36 weeks’ postmenstrual age decreased from 97% in 1995–1996 to 91% in 1997–2002.13
Although infant mortality rates continue to decrease, specific factors, including prematurity, lack of or inadequate prenatal care, inadequate weight gain in pregnancy, and African American ethnicity, increase mortality rates several fold.
Advances in perinatal care have led to increased survival of children born at the lower limits of viability. Children with birth weights of 1000 g or lower have poorer outcomes relative to normal-birth-weight term-born controls in neurologic and health status, cognitive-neuropsychological skills, school performance, academic achievement, and behavior. Outcomes are highly variable but are related to neonatal medical complications of prematurity and social risk factors.
Factors predictive of poor neurodevelopmental outcome include low birth weight, male gender, the severity of cerebral ultrasound abnormality including periventricular leukomalacia and persistent ventriculomegaly, bronchopulmonary dysplasia, and growth failure. In addition to gross neurologic deficits including cerebral palsy, hydrocephalus, and hemiplegia, these infants are susceptible to growth failure, deafness, and blindness.19-24
Major neonatal mortality increases with decreasing gestational age and birth weight: Survival at 23 weeks of gestation ranges from 2% to 35%; at 24 weeks of gestation, 17% to 62%; and at 25 weeks, 35% to 72%. Major neonatal morbidity also increases with decreasing gestational age and birth weight. At 23 weeks of gestation, chronic lung disease occurs in 57% to 86% of survivors; at 24 weeks, in 33% to 89%, and at 25 weeks, in 16% to 71% of survivors. The rates of severe cerebral ultrasound abnormality range from 10% to 83% at 23 weeks of gestation, 9% to 64% at 24 weeks, and 7% to 22% at 25 weeks. Of 77 survivors at 23 weeks of gestation, 26 (34%) have severe disability (defined as subnormal cognitive function, cerebral palsy, blindness, and/or deafness). At 24 weeks of gestation, the rates of severe neurodevelopmental disability range from 22% to 45%, and at 25 weeks of gestation, from 12% to 35%. The continuing toll of major neonatal morbidity and neurodevelopmental handicap is of serious concern. Neurodevelopmental impairment of extremely-low-birth-weight infants increased in the 1990s.
In Cleveland, Ohio, the outcomes of 114 children with birth weights of 500 to 749 g born in 1990 to 1992 were compared to 112 infants born in 1993 to 1995.20 Twenty-month survival was similar (43% and 38%, respectively). Although the use of antenatal and postnatal steroids increased, the rates of bronchopulmonary dysplasia, subnormal cognitive function at 20 months corrected age, and rate of cerebral palsy also increased. Similar findings have been reported elsewhere. Vohr,21 on the basis of evaluation of a multicenter cohort at 18 to 22 months corrected age, noted that extremely-low-birth-weight infants (< 1 kg) are at significant risk of neurologic abnormalities, developmental delays, and functional delays. Twenty-five percent of the children had an abnormal neurologic examination; 37% had a Bayley II Mental Developmental Index below 70; 29% had a Psychomotor Developmental Index below 70; 9% had vision impairment; and 11% had hearing impairment. Neurologic, developmental, neurosensory, and functional morbidities increased with decreasing birth weight. Factors significantly associated with increased neurodevelopmental morbidity include bronchopulmonary dysplasia, grades III to IV intraventricular hemorrhage/periventricular leukomalacia, the use of steroids for bronchopulmonary dysplasia, necrotizing enterocolitis, and male gender. Factors significantly associated with decreased morbidity include increased birth weight, female gender, higher maternal education, antenatal corticosteroids, and white race.
Wilson22 reported that since 2000, neurodevelopmental impairment has decreased among extremely-low-birth-weight infants in the Cleveland cohort. A variety of perinatal and neonatal factors were associated with the improved outcomes, including increased antenatal steroid use and cesarean section delivery as well as decreased sepsis, decreased severe cranial ultrasound abnormalities, and reduced postnatal steroid use.
Wood and colleagues23 evaluated all children who were born at 25 or fewer completed weeks of gestation in the United Kingdom and Ireland from March through December 1995, at the time when they reached a median age of 30 months. Two hundred eighty-three (92%) of the 308 surviving children were formally assessed. The mean (± standard deviation [SD]) scores on the Bayley Mental and Psychomotor Developmental Indices, referenced to a population mean of 100, were 84 ± 12 and 87 ± 13, respectively. Nineteen percent had severely delayed development (with scores more than 3 SD below the mean), and 11% had scores from 2 SD to 3 SD below the mean. Twenty-eight children (10%) had severe neuromotor disability, seven (2%) were blind or perceived light only, and eight (3%) had hearing loss that was uncorrectable or required aids. This cohort included 49% of survivors with no disability. Physicians and parents anticipating the delivery of extremely-low-birth-weight infants must be aware of these outcomes to make informed decisions as to the advisability of aggressive care at birth and thereafter. Marlow et al19 examined this cohort and observed that impairment of motor, visuospatial, and sensorimotor function, including planning, self-regulation, inhibition, and motor persistence, contributes excess morbidity over cognitive impairment in extremely preterm children and contributes independently to poor classroom performance at 6 years of age.
In summary, there have been marked improvements in the perinatal outcomes so that in the United States in 2004, the expectation of life at birth is 77.6 years for all gender and race groups combined. However, our national perinatal mortality rates still lag behind those of other industrialized countries, especially for minority groups.