- Email: [email protected]
Perinatal correlates and neonatal outcomes of small for gestational age infants born at term gestation
Perinatal correlates and neonatal outcomes of small for gestational age infants born at term gestation Benedict A. Doctor, MD,a Mary Ann O’Riordan, MS,a H. Lester Kirchner, PhD,a Dinesh Shah, MD,b and Maureen Hack, MB, ChBa Cleveland, Ohio OBJECTIVE: We sought to examine the current perinatal correlates and neonatal morbidity associated with intrauterine growth failure among neonates born at term gestation. STUDY DESIGN: We compared 372 small for gestational age (SGA, birth weight <10th percentile) infants born at term gestation to 372 appropriate for gestational age controls (AGA, birth weight 10th to 90th percentile) matched by sex, race, and gestational age within 2 weeks. RESULTS: Compared with AGA controls, significant (P < .05) maternal risk factors for SGA status included single marital status (59% versus 53%), lower prepregnancy weight (144 ± 41 lbs versus 153 ± 40 lbs), lower weight gain during pregnancy (29 ± 15 lbs versus 33 ± 15 lbs), smoking (25% versus 17%), hypertension (14% versus 7%), and multiple gestation (9% versus 2%). Mothers of SGA infants were more likely to undergo multiple (≥3) antenatal ultrasound evaluations (19% versus 7%), biophysical profile monitoring (11% versus 4%), and oxytocin delivery induction (28% versus 16%) (P < .05). Pediatrician attendance was more common among SGA deliveries (50% versus 37%, P < .05). SGA infants had significantly higher rates of hypothermia (18% versus 6%) and symptomatic hypoglycemia (5% versus 1%). These neonatal problems remained significant even when medical or pathologic causes of intrauterine growth failure, including pregnancy hypertension, multiple gestation, and congenital malformations, were excluded. CONCLUSION: Despite higher rates of pregnancy complications among mothers of SGA infants, the rates of neonatal adverse outcomes are low. However, SGA infants remain at risk for hypothermia and hypoglycemia and require careful neonatal surveillance. (Am J Obstet Gynecol 2001;185:652-9.)
Key words: Intrauterine growth failure, small for gestational age, neonatal morbidity, hypoglycemia, hypothermia
Infants born small for gestational age (SGA) after intrauterine growth failure are reported to have higher rates of neonatal mortality and morbidity than normal birth weight infants and to be at greater risk for neurologic and developmental deficits during childhood.1-7 In the 1970s, Lubchenco1 reported that SGA infants born at term gestation had a greater than 6-fold risk for neonatal mortality and a nearly 3-fold risk for neonatal morbidities when compared with normal birth weight infants. These included birth asphyxia, meconium aspiration syndrome, hypoglycemia, and polycythemia. More recent reports have noted lower rates of mortality and morbidity associated with intrauterine growth failure, with complications
From the Departments of Pediatricsa and Obstetrics and Gynecology,b Case Western Reserve University School of Medicine. Supported by the William Randolph Hearst Neonatology Fellowship Endowment Fund at Rainbow Babies and Children’s Hospital (B.A.D.). Received for publication November 8, 2000; revised February 12, 2001; accepted April 24, 2001. Reprint requests: Maureen Hack, MB, ChB, Rainbow Babies and Children’s Hospital, 11100 Euclid Ave, Cleveland, OH 44106. Copyright © 2001 by Mosby, Inc. 0002-9378/2001 $35.00 + 0 6/1/116749 doi:10.1067/mob.2001.116749
652
occurring mainly among infants with severe growth failure.2, 3, 7 Chard8 has suggested that many infants with birth weight less than the 10th percentile for gestational age, who have traditionally been defined as SGA, represent the spectrum of normal growth and development rather than pathology. The correlates and causes of intrauterine growth failure include sociodemographic risk factors, smoking, pregnancy-induced hypertension, multiple gestation, congenital malformations, and intrauterine infections.9-15 However, modern methods of pregnancy and perinatal care may have changed the relative contribution of these causes and their associated neonatal outcomes. For example, infertility treatment has led to an increase in the rate of multiple gestations, and antenatal diagnosis of congenital malformations with subsequent elective pregnancy termination may have decreased the rates of major congenital malformations at term gestation.16, 17 In addition, ultrasound diagnosis of intrauterine growth failure and biophysical profile testing allow the clinician to monitor fetal well-being and to electively deliver an at-risk infant prematurely.18 Thus, infants with intrauterine growth failure who are allowed to progress to term gestation may be a selected group of relatively low-risk infants. Further-
Volume 185, Number 3 Am J Obstet Gynecol
more, anticipatory pediatric care at delivery and monitoring for neonatal complications, such as hypoglycemia, may have decreased these sequelae. We thus sought to examine the current perinatal correlates of intrauterine growth failure and the associated neonatal morbidity among SGA infants born at term gestation in our perinatal center. We furthermore sought to examine the effects of the severity and causes of intrauterine growth failure on neonatal morbidity. Population and methods The population was drawn from a cohort of 4879 infants delivered at MacDonald Womens Hospital, University Hospitals, Cleveland, Ohio between January 1 and December 31, 1997. Of the 4107 infants (84%) who were born at term gestation (37-42 weeks), 435 (10.6%) were recorded in the perinatal data base as having a birth weight of less than the 10th percentile for gestational age.19 Sixty-three of these infants were excluded from the study population: 7 because of errors in the recording of birth weight, 44 because the gestational age was considered to be unreliable, 6 with unknown race, and 6 because maternal or infant charts were unavailable. The study population thus included 372 infants with birth weight less than the 10th percentile for gestational age. Their mean birth weight was 2609 g and mean gestational age was 38.9 weeks; 133 (36%) were male, 143 (38%) Caucasian, 223 (60%) African American, and 6 (2%) of other races. One hundred and sixty-seven of the infants had a birth weight less than the 5th percentile for gestational age.19 Their mean birth weight was 2467 g. To examine the perinatal correlates and neonatal outcomes associated with intrauterine growth failure, a comparison group of infants appropriate for gestational age (AGA) with birth weight between the 10th and 90th percentile for gestational age, was selected by matching each infant with birth weight less than the 10th percentile with an AGA infant of the same race, sex, and gestational age within 2 weeks. We used a birth weight less than the 10th percentile for gestational age for the definition of SGA, as this has traditionally been used.1, 4, 7 Nine and 15 percent of infants in the SGA and AGA groups respectively were multiple births. Gestational age was determined according to the last menstrual period and confirmed by second-trimester ultrasound at 16 to 20 weeks’ gestation. If there was a discrepancy of more than 2 weeks between the determination of gestation according to the last menstrual period and the second-trimester ultrasonography, the ultrasonography determination was used. The last menstrual period was used to determine gestational age if no early ultrasonography had been performed. Maternal and infant birth data were extracted retrospectively from the hospital charts by 3 examiners. Information from the maternal charts included sociodemo-
Doctor et al 653
graphic factors (education, race, age, marital status), past medical and obstetric history, and details pertaining to the current pregnancy, including prepregnancy weight, weight gain during pregnancy, information concerning the timing and findings of any prenatal ultrasonography performed, pregnancy complications, drug use, labor events, and mode of delivery. Hypertension (blood pressure > 140/90 mm Hg) in the current pregnancy included chronic hypertension and pregnancy-induced hypertension, with or without proteinuria. Fetal heart rate abnormalities considered included bradycardia, tachycardia, decreased variability, variable decelerations, and late decelerations on internal or external monitoring. Information extracted from the infant charts included growth parameters, Apgar scores at 1 and 5 minutes, need for delivery room resuscitation, neonatal morbidity, medical procedures, and treatment. Hypothermia was defined as a body temperature <36°C. Transient tachypnea of the newborn was defined as a respiratory rate > 60 breaths per minute for more than 4 hours with the newborn requiring oxygen therapy and with normal or “wet lung” radiographic findings. Pneumonia was defined according to clinical and radiographic evidence. Meconium aspiration was defined in the presence of respiratory distress caused by aspiration of meconium with supportive radiographic findings. Pulmonary hypertension was defined as respiratory distress with echocardiographic evidence of a right-to-left shunt in the absence of a congenital cardiac malformation. Symptomatic hypoglycemia was defined as a blood glucose screen of <40 mg/dL in the presence of symptoms including tachypnea, hypothermia, tachycardia, jitteriness, hypotonia, or lethargy. Necrotizing enterocolitis was defined according to Bell’s criteria.20 Suspected septicemia was defined as clinical or laboratory evidence of infection, or both, treated with antibiotics, but not proven by bacterial or fungal blood cultures. Proven septicemia was defined as positive bacterial or fungal blood cultures in similar settings. The protocol for newborn care at our institution includes pediatric attendance at all high-risk deliveries, including suspected intrauterine growth failure. Large and SGA infants and infants of mothers who are diabetic are screened for hypoglycemia within 1 hour of birth, and early feeding is encouraged. Infants with asymptomatic hypoglycemia (<40 mg/dL by heel-stick screening) are initially fed orally, and capillary blood glucose levels are followed at half-hour intervals until the glucose level has normalized (One Touch, Lifescan, Inc, Milpitas, Calif). If the glucose level does not normalize, if the quantitative blood glucose is < 20 mg/dL or if an infant becomes symptomatic, infants are transferred to the transitional care nursery and treated with intravenous dextrose. Screening for polycythemia is only performed as clinically indicated. In an analysis of the data, the total population of 372 infants with birth weights <10th percentile for gestational age and the subgroup with birth weights <5th percentile
654 Doctor et al
September 2001 Am J Obstet Gynecol
Table I. Maternal demographic, medical, and previous obstetric risk factors of infants with birth weight <10th and <5th percentile for gestational age and their respective controls Birth weight
Sociodemographic risk Age (y, mean ± SD) Marital status (unmarried) Education < high school Prepregnancy weight mean lbs ± SD Height (in, mean ± SD) Primigravid Chronic disease Pulmonary Cardiac Renal Previous obstetrical risk Low birth weight infant at term† Miscarriage/abortion Stillbirth Neonatal death
Birth weight
<10th Percentile (n = 372)
Normal (n = 372)
<5th Percentile (n = 167)
Normal (n = 167)
26.6 ± 6 219 (59%)* 74 (20%)
27.1 ± 6 197 (53%) 74 (20%)
27.1 ± 6 102 (61%)* 33 (20%)
27.6 ± 6 90 (54%) 30 (18%)
144 ± 41* 63.8 ± 3* 106 (28%)
153 ± 40 64.5 ± 3 106 (28%)
140 ± 35* 63.7 ± 3* 46 (28%)
155 ± 42 64.4 ± 2 42 (25%)
53 (14%) 2 (0.5%) 4 (1.1%)
50 (13%) 8 (2%) 10 (3%)
23 (14%) 1 (0.6%) 0
27 (16%) 6 (4%) 5 (3%)
21 (6%)* 156 (42%) 6 (1.6%) 5 (1.3%)
2 (0.5%) 156 (42%) 6 (1.6%) 3 (0.8%)
12 (7%) 74 (44%) 5 (3%) 1 (0.6%)
0 76 (46%) 2 (1.1%) 0
*P < .05 comparing SGA with control. †Low birth weight < 2500 g.
for gestational age were compared with their matched controls (1-to-1 matching) with regard to maternal demographic data, obstetric complications, maternal interventions, neonatal morbidity, and medical therapy. Exact conditional logistic regression was performed to estimate and test the association of predictors of subnormal growth (<5th and <10th percentile).21 In examining potential differences between the SGA group and the matched controls, McNemar’s test was used to compare categorical data, and the paired t test or the Wilcoxon signed rank test was used to compare continuous outcome data. Variables not available on all infants were tested by using regression models controlling for the matching variables (age, sex, race) as the matching was broken. We furthermore examined the neonatal outcomes, controlling for identified medical or pathologic causes of intrauterine growth failure as a group, by exact stratified logistic regression (LogXact 2.1, Cambridge, Mass. Cytel Software Corp. 1996). These included major congenital malformation, multiple birth, and maternal hypertension. A single variable indicating the presence of any of these identified medical causes was included in the models. Results Perinatal correlates and neonatal outcomes at birth weights < 10th percentile for gestational age Maternal demographic, pregnancy and perinatal descriptors. A comparison of maternal demographic, medical, and previous obstetric risk factors between the mothers of SGA infants and controls is shown in Table I. Significantly more of the mothers of the SGA infants were unmarried. They had a lower mean prepregnancy weight and height
and had previously delivered significantly more low birth weight infants at term gestation compared with the control mothers (P < .05) Consideration of the pregnancy-related risk factors revealed that mothers of the SGA infants had a significantly lower weight gain during pregnancy (Table II). The rates of smoking, multiple gestation, pregnancy hypertension, and oligohydramnios were also significantly higher. The mothers of SGA infants were more likely to undergo multiple (≥ 3) antenatal ultrasonographic evaluations, biophysical profile monitoring, and oxytocin delivery induction (Table III). The rates of cesarean section were similar between groups; however, fetal distress was listed as an indication for cesarean section in 24 (45%) of 53 mothers among the SGA group as compared with 11 (20%) of 56 mothers among the AGA group (P < .05, data not shown in table). Neonatal outcomes. A comparison of the birth data and delivery room care of the SGA infants and their matched controls is shown in Table IV. Pediatricians attended significantly more of the deliveries of the SGA infants. The condition of the infants at birth did not differ, with similar rates of low Apgar scores (≤6) at 1 and 5 minutes and of the need for resuscitation. Hypothermia (< 36°C) occurred more frequently among SGA infants both at first examination and during the total hospital course. The rates of transfer of infants to the transitional and intensive care nurseries were similar. In a comparison of neonatal morbidity (Table V), 13 infants (3.5%) among the SGA group had a major congenital malformation compared with 2 (0.5%) of the AGA group. The malformations included 3 SGA infants
Doctor et al 655
Volume 185, Number 3 Am J Obstet Gynecol
Table II. Obstetric risk factors of mothers of infants with birth weights <10th and <5th percentile for gestational age and their respective controls Birth weight
Prenatal care Weight gain (lbs, mean ± SD) Substance abuse Smoking Alcohol Other Pregnancy hypertension Diabetes Oligohydramnios Antenatal ultrasound Any ≥3 Biophysical profile monitoring Urine toxin screen Of these, positive
Birth weight
<10th Percentile (n = 372)
Normal (n = 372)
<5th Percentile (n = 167)
Normal (n = 167)
367 (99%) 29 ± 15*
366 (98%) 33 ± 15
165 (99%) 29 ± 14*
164 (98%) 33 ± 15
92 (25%)* 17 (5%) 32 (9%) 53 (14%)* 11 (3%) 39 (10%)*
62 (17%) 23 (6%) 23 (6%) 27 (7%) 10 (3%) 10 (3%)
45 (27%) 8 (5%)* 18 (11%)* 27 (16%)† 4 (2%) 23 (14%)*
29 (17%) 5 (3%) 7 (4%) 15 (9%) 8 (5%) 3 (1.8%)
340 (91%)* 72 (19%)* 41 (11%)* 50 (13%) 12 (24%)
319 (86%) 27 (7%) 16 (4%) 44 (12%) 15 (34%)
159 (95%)* 38 (23%)* 20 (12%)* 28 (17%) 7 (25%)
145 (87%) 15 (9%) 6 (4%) 18 (11%) 5 (8%)
*P < .05 comparing SGA with control. †P = .065 comparing SGA with control.
Table III. Intrapartum course, labor, and delivery of infants with birth weight <10th and <5th percentile and their respective controls Birth weight
Pregnancy duration (weeks, mean ± SD) Pitocin induction Pitocin augmentation Fetal scalp pH monitoring Fetal distress† Meconium-stained fluid Fetal lie Vertex Breech/complex Mode of delivery Vertex vaginal Spontaneous Vacuum/forceps Breech vaginal Cesarean section
Birth weight
10th Percentile (n = 372)
Normal (n = 372)
<5th Percentile (n = 167)
Normal (n = 167)
38.9 ± 1 103 (28%)* 114 (31%)* 179 (48%) 66 (18%) 20 (5%)*
38.9 ± 1 61 (16%) 196 (53%) 168 (45%) 67 (18%) 6 (1.6%)
38.9 ± 1 58 (35%)* 44 (26%)* 82 (49%) 28 (17%) 7 (4%)
38.9 ± 1 26 (16%) 80 (48%) 71 (43%) 33 (20%) 2 (1.2%)
345 (93%) 24 (6%)*
362 (97%) 10 (3%)
149 (90%) 17 (10%)
159 (95%) 8 (5%)
268 (72%) 44 (12%) 5 (1.3%) 53 (14%)
269 (72%) 40 (11%) 1 (0.3%) 56 (15%)
115 (69%) 16 (10%) 5 (3%) 30 (18%)
123 (74%) 16 (10%) 1 (0.6%) 26 (16%)
*P < .05 comparing SGA with control. †Includes bradycardia, tachycardia, decreased variability, and variable and late decelerations on internal or external fetal heart rate monitoring.
with a chromosomal abnormality (trisomy 21, trisomy 18, trisomy 1/monosomy 13), 3 with major congenital heart malformations (double-outlet right ventricle, tetralogy of Fallot, and critical aortic coarctation), 3 with genitourinary anomalies (utero-pelvic junction obstruction with multicystic kidney, horseshoe kidney, and hypospadias), 1 with Prader-Willi syndrome, 1 Pierre Robin syndrome, 1 with a knee subluxation that required serial casting, and 1 with nonsyndromic multiple major malformations. The 2 AGA infants with major malformations had trisomy 21
and bilateral talipes equinovarus, respectively. Ten percent of the SGA infants were tested for cytomegalovirus; however, none were proven positive. One asymptomatic AGA infant, among the 7 infants with positive maternal Venereal Disease Research Laboratory screens, had a positive serum screen with negative spinal fluid and was treated with a single dose of intramuscular penicillin. The overall rates of respiratory distress did not differ; however, SGA infants had a higher rate of transient tachypnea of the newborn than AGA infants (2% versus
656 Doctor et al
September 2001 Am J Obstet Gynecol
Table IV. Birth data and delivery room care of infants with birth weight < 10th and < 5th percentile for gestational age and their respective controls Birth weight
Birth weight (g), mean ± SD <2500 g Multiple birth Pediatrician attendance at delivery Apgar score ≤6 (1 min) Apgar score ≤ 6 (5 min) Delivery room resuscitation Bag & mask ventilation Intubation for meconium Intubation for ventilation Chest compressions First temperature <36°C Any temperature < 36°C during stay Hospitalization Regular nursery Transitional nursery Intensive care nursery
Birth weight
10th Percentile (n = 372)
Normal (n = 372)
5th Percentile (n = 167)
Normal (n = 167)
2609 ± 206* 106 (29%) 34 (9%)* 186 (50%)*
3310 ± 301 — 8 (2%) 138 (37%)
2467 ± 184* 84 (50%) 25 (15%)* 98 (59%)*
3329 ± 320 — 5 (3%) 67 (40%)
47 (13%) 8 (2%)
42 (11%) 6 (1.6%)
27 (16%) 5 (3%)
19 (11%) 1 (0.6%)
22 (6%) 40 (11%) 9 (2%) 3 (0.8%) 40 (11%)* 101 (28%)*
33 (9%) 44 (12%) 5 (1.3%) 3 (0.8%) 7 (2%) 27 (8%)
12 (7%) 14 (8%) 7 (4%) 3 (1.8%) 26 (17%)* 51 (31%)*
14 (8%) 24 (14%) 2 (1.2%) 2 (1.2%) 3 (1.8%) 11 (7%)
328 (88%) 19 (5%) 24 (6%)
335 (90%) 20 (5%) 17 (5%)
142 (85%) 9 (5%) 16 (10%)
151 (90%) 9 (5%) 7 (4%)
*P < .05 comparing SGA with control.
0.3%, P < .05). Glucose screening was performed more frequently among the SGA infants, although the overall rates of subnormal glucose (< 40 mg/dL) among those screened were similar between groups. Significantly more of the SGA infants had symptomatic hypoglycemia (5% versus 1.1%). The symptoms of hypoglycemia among the 18 symptomatic SGA infants included hypothermia in 8, tachypnea in 5, and jitteriness in 5. Among the 4 symptomatic AGA infants, symptoms included jitteriness in 2, tachypnea in 1, and tachycardia in 1. A comparison of neonatal interventions and therapies performed during the hospital course is shown in Table VI. Eight SGA infants (2%) required endotracheal intubation and assisted ventilation as compared with 2 (0.5%) control infants (P < .05). Reasons for assisted ventilation among the SGA infants included sequelae of a major congenital malformation in 3; severe birth depression in 2; and meconium aspiration, pneumonia, and transient tachypnea in 1 infant each, respectively. The reasons in the 2 AGA infants included severe birth depression and meconium aspiration. Death occurred in 2 neonates, both in the SGA group and both as a result of major congenital malformations. One infant, with a birth weight of 2603 g, was initially discharged home, but readmitted on day of life 8 with critical coarctation of the aorta and died soon after admission. The second infant, with a birth weight of 2300 g, had trisomy 1/monosomy 13 with multiple malformations and died on day of life 2. Outcomes of the subgroups of infants with birth weights less than the 5th percentile for gestational age and between the 5th and less than the 10th percentile. The significant
comparisons, described above, for the total group of SGA infants with birth weight less than the 10th percentile for gestational age were also significant for the subgroup of infants with birth weight less than the 5th percentile for gestational age when compared with their matched controls (Tables II and V). Infants with birth weight less than the 5th percentile also required significantly more supplemental oxygen, intravenous fluids, and antibiotics and had significantly longer hospital stays than their normal birth weight controls (Table VI). Comparison of the 205 mothers of infants with birth weight between the 5th and less than the 10th percentile for gestational age with their matched controls revealed significant differences in maternal height, weight gain during pregnancy, and performance of ≥3 antenatal ultrasonography. Significant differences among the infants pertained only to hypothermia (data not shown). Relationship between causes of intrauterine growth failure and neonatal morbidity. Medical or pathologic causes of intrauterine growth failure were identified in 95 (26%) of the 372 SGA mother-infant pairs, of whom 5 had more than 1 identified cause. Of the 95, there was major congenital malformation in 13, multiple gestation in 34 (32 twin and 2 triplet), and pregnancy hypertension in 53 cases. Mother-infant pairs with more than 1 identified reason for being SGA included 4 with multiple gestation and pregnancy hypertension and 1 with multiple gestation and a major congenital malformation. Among the AGA maternal-infant control pairs, 2 had a major congenital malformation, 8 were multiple gestation, 27 had pregnancy hypertension, and 1 had suspected congenital syphilis. After controlling for these combined medical or
Doctor et al 657
Volume 185, Number 3 Am J Obstet Gynecol
Table V. Neonatal morbidity of infants with birth weight <10th and <5th percentile for gestational age and their respective controls Birth weight
Birth weight
<10th Percentile (n = 372)
Normal (n = 372)
<5th Percentile (n = 167)
Normal (n = 167)
13 (4%)* 6 (1.6%)
2 (0.5%) 13 (3%)
7 (4%)* 3 (1.8%)
0 6 (4%)
8 (2%)* 3 (0.8%) 2 (0.5%) 1 (0.3%) 176 (47%)*
1 (0.3%) 2 (0.5%) 5 (1.3%) 0 87 (23%)
6 (4%)§ 2 (1.2%) 1 (0.6%) 1 (0.6%) 106 (63%)*
1 (0.6%) 0 2 (1.2%) 0 39 (23%)
34 (19%) 20 (11%) 18 (5%)* 2 (0.5%) 107 (29%)
18 (21%) 8 (9%) 4 (1.1%) 1 (0.3%) 97 (26%)
22 (21%) 13 (12%) 10 (6%)* 1 (0.6%) 55 (33%)
9 (23%) 3 (8%) 1 (0.6%) 0 46 (28%)
3 (3%) 2 (0.5%) 92 (25%)* 8.5 ± 4
2 (0.5%) 0 66 (18%) 7.6 ± 4
3 (5%) 2 (4%) 42 (28%) 7.5 ± 4
1 (0.6%) 0 31 (19%) 6.6 ± 3
0
0
0
0
25 (7%) 3 (0.8%) 38 (10%)* 0
19 (5%) 2 (0.5%) 21 (6%) 0
18 (11%)* 2 (1.2%) 28 (17%)* 0
5 (3%) 1 (0.6%) 5 (3%) 0
Congenital anomaly Major Minor† Respiratory distress Transitory tachypnea‡ Pneumonia Meconium aspiration Pulmonary hypertension Glucose screened Of these, glucose <40 mg/dL At <2 h of age Beyond 2 h of age Symptomatic hypoglycemia Seizures Blood count performed Of these Hematocrit >65% Hematocrit >70% Bilirubin check Maximum recorded level (mg/dL, mean ± SD) Necrotizing enterocolitis Septicemia Suspect Proven Cytomegalovirus cultured Culture positive *P < .05 comparing SGA with control. †Including polydactyly. ‡With supplemental oxygen requirement. §P = .06 comparing SGA with control. P = .07 comparing SGA with control.
pathologic causes of intrauterine growth failure by exact stratified logistic regression, the differences between the SGA (birth weight less than the 10th percentile) and AGA groups of infants remained significant for the prevalence of hypothermia (both initially and during the hospital course) and for the rates of symptomatic hypoglycemia (P < .05). However, the group differences in the rate of pediatric attendance at delivery and the need for assisted ventilation no longer remained significant. The differences between the subgroup of infants with birth weight less than the 5th percentile for gestational age and their normal birth weight controls, after controlling for the identified medical or pathologic causes of intrauterine growth failure, remained significant for the prevalence of hypothermia, but not for the prevalence of symptomatic hypoglycemia. Comment We sought to describe the current neonatal morbidity and obstetric correlates of intrauterine growth failure at term gestation in an urban tertiary perinatal center. Our results reveal significantly higher rates of hypothermia
and symptomatic hypoglycemia in the total group of infants with birth weight less than the 10th percentile for gestational age when compared with infants born with birth weights AGA. The increased need for neonatal intervention and therapies pertained mainly to the subgroup of infants with birth weights less than the 5th percentile. Only 2 SGA infants died, both with major congenital malformations. Current perinatal practice includes planned preterm delivery for conditions such as severe preeclampsia and fetal growth retardation. Because we did not prospectively examine the rates of planned preterm delivery for maternal or fetal causes of intrauterine growth failure, we cannot conclude that our findings of relatively low neonatal morbidity and mortality among SGA infants delivered at term gestation are a result of this practice. The perinatal correlates of intrauterine growth failure in our population are consistent with the literature and include single marital status, lower maternal prepregnancy weight and pregnancy weight gain, smoking, pregnancy hypertension, multiple gestation, and major congenital malformation.10-15, 18, As a result of the obstetric
658 Doctor et al
September 2001 Am J Obstet Gynecol
Table VI. Neonatal interventions and therapies of infants with birth weights <10th and <5th percentile for gestational age and their respective controls Birth weight
Supplemental oxygen (days, mean ± SD) Assisted ventilation (days, mean ± SD) Umbilical artery catheterization Umbilical vein catheterization Intravenous fluids Antibiotics† Phototherapy Partial exchange transfusion Diagnostic imaging‡ Total hospital days Mean ± SD
Birth weight
<10th Percentile (n = 372)
Normal (n = 372)
<5th Percentile (n = 167)
Normal (n = 167)
16 (4%) 2.6 ± 3 8 (2%)* 3 ± 4 4 (1.1%) 6 (1.6%) 35 (9%) 27 (7%) 12 (3%)* 1 32 (9%)
10 (3%) 2.2 ± 1 2 (0.5%) 1.5 ± 1 6 (1.6%) 6 (1.6%) 21 (6%) 21 (6%) 3 (0.8%) 0 29 (8%)
12 (7%)* 3.2 ± 4 7 (4%)* 3.3 + 4 4 (2%) 6 (3%) 22 (13%)* 20 (12%)* 5 (3%) 1 19 (11%)
3 (1.8%) 1.7 ± 1 0 0 1 (0.6%) 1 (0.6%) 9 (5%) 6 (4%) 1 (0.6%) 0 12 (7%)
3.54 ± 2
3.44 ± 2
4.02 ± 3*
3.35 ± 2
*P < .05 comparing SGA with control. †Intravenous or intramuscular. ‡Includes chest x-rays, abdominal x-rays, and head ultrasonography.
complications, significantly more antenatal ultrasonography examinations were performed and significantly more deliveries were induced among mothers of SGA infants when compared with mothers of the normal birth weight infants. We defined SGA as birth weight less than the 10th percentile for gestational age because this cut-off has traditionally been used. Comparison of the neonatal morbidity among infants with birth weight less than the 5th percentile for gestational age and between the 5th and 10th percentile reveals that the majority of significant differences in neonatal morbidity are the result of the lower birth weight cut-off of less than the 5th percentile. The only significant difference between infants with birth weights between the 5th and 10th percentile for gestational age and their matched normal birth weight controls was the development of neonatal hypothermia. The differences in the rates of hypothermia and symptomatic hypoglycemia between the SGA infants and their respective controls remained significant even after controlling for identified medical or pathologic causes of intrauterine growth failure, indicating that these neonatal morbidities are related to growth failure per se rather than its cause. The differences in the rates of pediatric attendance at delivery and the need for assisted ventilation, however, no longer remained significant because these interventions are usually indicated in the presence of maternal or fetal pathology such as multiple births, maternal hypertension, and congenital malformations. We sought to describe the overall impact of intrauterine growth failure on neonatal morbidity in our perinatal center irrespective of cause. Comparison of our results with other reports are thus confounded by the inclusion of multiple births and infants with major congenital malforma-
tions and by the varying definitions of SGA between studies.2-4, 7 We did not use race or sex-specific fetal growth curves because we sought to reflect clinical practice. The national reference for fetal growth, which we used, predominantly reflects the fetal growth patterns of white US births.19 This may have resulted in defining some African-American infants as SGA, whereas they might have been defined as AGA if we had used race-specific curves. Male infants are larger than female infants at birth. The preponderance of female infants in our SGA cohort might similarly be explained by our use of the combined gender fetal growth curve. Kramer2 examined neonatal outcomes among 673 (13%) of 5319 term singleton infants without congenital malformations, born between 1980 and 1986 in Montreal, who were classified as having intrauterine growth failure according to a fetal growth ratio of less than 0.85. The rates of neonatal morbidity examined, including birth depression, hypoglycemia, and polycythemia were highest among infants with severe intrauterine growth failure, defined as a fetal growth ratio less than 0.75. Similar to our results, Kramer concluded that the rates of neonatal morbidity were associated with the severity of the intrauterine growth failure rather than its cause. More recently, McIntire3 reported on the effects of intrauterine growth failure among 11,503 term singleton infants without congenital malformations, with birth weights less than the 10th percentile who were born between 1988 and 1996 in Dallas. Significant differences in neonatal morbidity pertained mainly to infants with birth weight less than the 3rd percentile for gestational age. These differences persisted after adjusting for maternal race, parity, and infant sex. Ott,7 in contrast, in a study of 150 SGA and 807 AGA term infants without congenital malformations, found that morbidity risk was related to maternal pregnancy risk factors rather than infant size.
Doctor et al 659
Volume 185, Number 3 Am J Obstet Gynecol
Our study is important and relevant to current practice because it presents recent neonatal outcomes in an urban tertiary perinatal center. Weaknesses include its retrospective observational design, its low statistical power when examining outcomes with low incidence rates, and the fact that the results might not be applicable to other settings. Because of the low incidence of adverse outcomes in both SGA infants and AGA controls, the ability to detect small differences between groups is limited. The study is also limited by the lack of prospective laboratory information on parameters such as glucose, calcium, and hematocrit measurements. Glucose was screened in only 47% of the infants with birth weights less than the 10th percentile in our nursery, most probably because the “Lubchenco” charts were used in our nursery. These standards were normed at the altitude of Denver, Colo, where the 10th percentile birth weight for gestational age cutoff corresponds to Alexander’s 5th percentile.19 Thus, in fact, only half of the infants with birth weight less than the 10th percentile were screened according to nursery protocol. Use of more appropriate growth charts might improve the rates of glucose screening and possibly reduce the rates of symptomatic hypoglycemia at our center. There have been no recent reports of neonatal hypothermia among SGA infants since Sinclair’s review in 1970.21 The occurrence of transient hypothermia in 28% and symptomatic hypoglycemia in 5% of SGA neonates in our population can be considered morbidities of relatively high incidence considering the ongoing anticipatory pediatric surveillance in our nursery. Because hypothermia was listed as a symptom in 8 of the 18 SGA infants with symptomatic hypoglycemia, it is not clear whether the hypoglycemia predisposed to the hypothermia or the opposite. The hypothermia may, however, be partly explained by the relatively large body surface area of SGA infants and the relative lack of an insulating layer of subcutaneous adipose tissue.22 Improved attention to providing an optimal neutral thermal environment for SGA infants immediately after birth and closer ongoing monitoring of their body temperature may possibly prevent the hypothermia we have described. We conclude that while the incidence of the classic morbidities associated with intrauterine growth failure— which include birth asphyxia, meconium aspiration, and polycythemia—are now of low incidence, SGA neonates remain a high-risk population deserving of special neonatal surveillance and care. We thank Dr Mark Schluchter for his statistical expertise and Sue Bergant, Nancy Newman and Bonnie Siner for their research assistance.
REFERENCES
1. Lubchenco LO. The high risk infant: intrauterine growth and neonatal morbidity and mortality. Philadelphia: WB Saunders; 1976. 2. Kramer MS, Olivier M, McLean FG, Willis DM, Usher RH. Impact of intrauterine growth retardation and body proportionality on fetal and neonatal outcome. Pediatrics 1990;86:707-13. 3. McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in relation to morbidity and mortality among newborn infants. N Engl J Med 1999;340:1234-8. 4. Vik T, Markestad T, Ahlsten G, Gebre-Medhin M, Jacobsen G, Hoffman HJ, et al. Body proportions and early neonatal morbidity in small-for-gestational-age infants of successive births. Acta Obstet Gynecol Scand 1997;165:76-81. 5. Strauss RS, Dietz WH. Growth and development of term children born with low birth weight: effects of genetic and environmental factors. J Pediatr 1998;133:167-72. 6. Blair E, Stanley F. Intrauterine growth and spastic cerebral palsy. I. Association with birth weight for gestational age. Am J Obstet Gynecol 1990;162:229-37. 7. Ott WJ. Small for gestational age fetus and neonatal outcome: reevaluation of the relationship. Am J Perinatol 1995;12:396400. 8. Chard T, Yoong A, Macintosh M. The myth of fetal growth retardation at term. Br J Obstet Gynaecol 1993;100:1076-81. 9. Kramer MS. Socioeconomic determinants of intrauterine growth retardation. Eur J Clin Nutr 1998;52(S1):S29-33. 10. Hickey CA, Cliver SP, Goldenberg RL, Kohatsu J, Hoffman HJ. Prenatal weight gain, term birth weight, and fetal growth retardation among high-risk multiparous black and white women. Obstet Gynecol 1993;81:529-35. 11. Cliver SP, Goldenberg RL, Cutter GR, Hoffman HJ, Davis RO, Nelson KG. The effect of cigarette smoking on neonatal anthropometric measurements. Obstet Gynecol 1995;85:625-30. 12. Misra DP. The effect of the pregnancy-induced hypertension on fetal growth: a review of the literature. Paediatr Perinat Epidemiol 1996;10:244-63. 13. Haelterman E, Reart G, Paris-Llado J, Dramaix M, Tchobroutsky C. Effect of uncomplicated chronic hypertension on the risk of small-for-gestational age birth. Am J Epidemiol 1997;145:689-95. 14. Alexander GR, Kogan M, Martin J, Papiernik E. What are the fetal growth patterns of singletons, twins, and triplets in the United States? Clin Obstet Gynecol 1998;41:115-25. 15. Snijders RJM, Sherrod C, Gosden CM, Nicolaides KH. Fetal growth retardation: associated malformations and chromosomal abnormalities. Am J Obstet Gynecol 1993;168:547-55. 16. Ventura SJ, Martin JA, Curtin SC, Mathews TJ, Park MM. Births: final data for 1998. National vital statistics reports; vol 48 no 3. Hyattsville, Md: National Center for Health Statistics. 2000. 17. Ewigman BG, Crane JP, Frigoletto FD, LeFevre ML, Bain RP, McNellis D. Effect of prenatal ultrasound screening on perinatal outcome. N Engl J Med 1993;329:821-7. 18. Kramer MS, Platt R, Yang H, Joseph KS, Wen SW, Morin L, et al. Secular trends in preterm birth: a hospital-based cohort study. JAMA 1998;280:1849-54. 19. Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States National reference for fetal growth. Obstet Gynecol 1996;87:163-8. 20. Bell MJ, Ternberg JL, Feigin RD, Keating JP, Marshall R, Barton L, et al. Neonatal necrotizing enterocolitis: therapeutic decisions based on clinical staging. Ann Surg 1978;187:1-7. 21. Sinclair J. Heat production and thermoregulation in the small for date infant. Pediatr Clin North Am 1970;17:147-58. 22. Kliegman RM. Intrauterine growth retardation. In: Fanaroff AA, Martin RJ, editors. Neonatal-perinatal medicine: diseases of the fetus and infant. 6th ed. St Louis: Mosby; 1997. pp. 203-40.
Key words: Intrauterine growth failure, small for gestational age, neonatal morbidity, hypoglycemia, hypothermia
Infants born small for gestational age (SGA) after intrauterine growth failure are reported to have higher rates of neonatal mortality and morbidity than normal birth weight infants and to be at greater risk for neurologic and developmental deficits during childhood.1-7 In the 1970s, Lubchenco1 reported that SGA infants born at term gestation had a greater than 6-fold risk for neonatal mortality and a nearly 3-fold risk for neonatal morbidities when compared with normal birth weight infants. These included birth asphyxia, meconium aspiration syndrome, hypoglycemia, and polycythemia. More recent reports have noted lower rates of mortality and morbidity associated with intrauterine growth failure, with complications
From the Departments of Pediatricsa and Obstetrics and Gynecology,b Case Western Reserve University School of Medicine. Supported by the William Randolph Hearst Neonatology Fellowship Endowment Fund at Rainbow Babies and Children’s Hospital (B.A.D.). Received for publication November 8, 2000; revised February 12, 2001; accepted April 24, 2001. Reprint requests: Maureen Hack, MB, ChB, Rainbow Babies and Children’s Hospital, 11100 Euclid Ave, Cleveland, OH 44106. Copyright © 2001 by Mosby, Inc. 0002-9378/2001 $35.00 + 0 6/1/116749 doi:10.1067/mob.2001.116749
652
occurring mainly among infants with severe growth failure.2, 3, 7 Chard8 has suggested that many infants with birth weight less than the 10th percentile for gestational age, who have traditionally been defined as SGA, represent the spectrum of normal growth and development rather than pathology. The correlates and causes of intrauterine growth failure include sociodemographic risk factors, smoking, pregnancy-induced hypertension, multiple gestation, congenital malformations, and intrauterine infections.9-15 However, modern methods of pregnancy and perinatal care may have changed the relative contribution of these causes and their associated neonatal outcomes. For example, infertility treatment has led to an increase in the rate of multiple gestations, and antenatal diagnosis of congenital malformations with subsequent elective pregnancy termination may have decreased the rates of major congenital malformations at term gestation.16, 17 In addition, ultrasound diagnosis of intrauterine growth failure and biophysical profile testing allow the clinician to monitor fetal well-being and to electively deliver an at-risk infant prematurely.18 Thus, infants with intrauterine growth failure who are allowed to progress to term gestation may be a selected group of relatively low-risk infants. Further-
Volume 185, Number 3 Am J Obstet Gynecol
more, anticipatory pediatric care at delivery and monitoring for neonatal complications, such as hypoglycemia, may have decreased these sequelae. We thus sought to examine the current perinatal correlates of intrauterine growth failure and the associated neonatal morbidity among SGA infants born at term gestation in our perinatal center. We furthermore sought to examine the effects of the severity and causes of intrauterine growth failure on neonatal morbidity. Population and methods The population was drawn from a cohort of 4879 infants delivered at MacDonald Womens Hospital, University Hospitals, Cleveland, Ohio between January 1 and December 31, 1997. Of the 4107 infants (84%) who were born at term gestation (37-42 weeks), 435 (10.6%) were recorded in the perinatal data base as having a birth weight of less than the 10th percentile for gestational age.19 Sixty-three of these infants were excluded from the study population: 7 because of errors in the recording of birth weight, 44 because the gestational age was considered to be unreliable, 6 with unknown race, and 6 because maternal or infant charts were unavailable. The study population thus included 372 infants with birth weight less than the 10th percentile for gestational age. Their mean birth weight was 2609 g and mean gestational age was 38.9 weeks; 133 (36%) were male, 143 (38%) Caucasian, 223 (60%) African American, and 6 (2%) of other races. One hundred and sixty-seven of the infants had a birth weight less than the 5th percentile for gestational age.19 Their mean birth weight was 2467 g. To examine the perinatal correlates and neonatal outcomes associated with intrauterine growth failure, a comparison group of infants appropriate for gestational age (AGA) with birth weight between the 10th and 90th percentile for gestational age, was selected by matching each infant with birth weight less than the 10th percentile with an AGA infant of the same race, sex, and gestational age within 2 weeks. We used a birth weight less than the 10th percentile for gestational age for the definition of SGA, as this has traditionally been used.1, 4, 7 Nine and 15 percent of infants in the SGA and AGA groups respectively were multiple births. Gestational age was determined according to the last menstrual period and confirmed by second-trimester ultrasound at 16 to 20 weeks’ gestation. If there was a discrepancy of more than 2 weeks between the determination of gestation according to the last menstrual period and the second-trimester ultrasonography, the ultrasonography determination was used. The last menstrual period was used to determine gestational age if no early ultrasonography had been performed. Maternal and infant birth data were extracted retrospectively from the hospital charts by 3 examiners. Information from the maternal charts included sociodemo-
Doctor et al 653
graphic factors (education, race, age, marital status), past medical and obstetric history, and details pertaining to the current pregnancy, including prepregnancy weight, weight gain during pregnancy, information concerning the timing and findings of any prenatal ultrasonography performed, pregnancy complications, drug use, labor events, and mode of delivery. Hypertension (blood pressure > 140/90 mm Hg) in the current pregnancy included chronic hypertension and pregnancy-induced hypertension, with or without proteinuria. Fetal heart rate abnormalities considered included bradycardia, tachycardia, decreased variability, variable decelerations, and late decelerations on internal or external monitoring. Information extracted from the infant charts included growth parameters, Apgar scores at 1 and 5 minutes, need for delivery room resuscitation, neonatal morbidity, medical procedures, and treatment. Hypothermia was defined as a body temperature <36°C. Transient tachypnea of the newborn was defined as a respiratory rate > 60 breaths per minute for more than 4 hours with the newborn requiring oxygen therapy and with normal or “wet lung” radiographic findings. Pneumonia was defined according to clinical and radiographic evidence. Meconium aspiration was defined in the presence of respiratory distress caused by aspiration of meconium with supportive radiographic findings. Pulmonary hypertension was defined as respiratory distress with echocardiographic evidence of a right-to-left shunt in the absence of a congenital cardiac malformation. Symptomatic hypoglycemia was defined as a blood glucose screen of <40 mg/dL in the presence of symptoms including tachypnea, hypothermia, tachycardia, jitteriness, hypotonia, or lethargy. Necrotizing enterocolitis was defined according to Bell’s criteria.20 Suspected septicemia was defined as clinical or laboratory evidence of infection, or both, treated with antibiotics, but not proven by bacterial or fungal blood cultures. Proven septicemia was defined as positive bacterial or fungal blood cultures in similar settings. The protocol for newborn care at our institution includes pediatric attendance at all high-risk deliveries, including suspected intrauterine growth failure. Large and SGA infants and infants of mothers who are diabetic are screened for hypoglycemia within 1 hour of birth, and early feeding is encouraged. Infants with asymptomatic hypoglycemia (<40 mg/dL by heel-stick screening) are initially fed orally, and capillary blood glucose levels are followed at half-hour intervals until the glucose level has normalized (One Touch, Lifescan, Inc, Milpitas, Calif). If the glucose level does not normalize, if the quantitative blood glucose is < 20 mg/dL or if an infant becomes symptomatic, infants are transferred to the transitional care nursery and treated with intravenous dextrose. Screening for polycythemia is only performed as clinically indicated. In an analysis of the data, the total population of 372 infants with birth weights <10th percentile for gestational age and the subgroup with birth weights <5th percentile
654 Doctor et al
September 2001 Am J Obstet Gynecol
Table I. Maternal demographic, medical, and previous obstetric risk factors of infants with birth weight <10th and <5th percentile for gestational age and their respective controls Birth weight
Sociodemographic risk Age (y, mean ± SD) Marital status (unmarried) Education < high school Prepregnancy weight mean lbs ± SD Height (in, mean ± SD) Primigravid Chronic disease Pulmonary Cardiac Renal Previous obstetrical risk Low birth weight infant at term† Miscarriage/abortion Stillbirth Neonatal death
Birth weight
<10th Percentile (n = 372)
Normal (n = 372)
<5th Percentile (n = 167)
Normal (n = 167)
26.6 ± 6 219 (59%)* 74 (20%)
27.1 ± 6 197 (53%) 74 (20%)
27.1 ± 6 102 (61%)* 33 (20%)
27.6 ± 6 90 (54%) 30 (18%)
144 ± 41* 63.8 ± 3* 106 (28%)
153 ± 40 64.5 ± 3 106 (28%)
140 ± 35* 63.7 ± 3* 46 (28%)
155 ± 42 64.4 ± 2 42 (25%)
53 (14%) 2 (0.5%) 4 (1.1%)
50 (13%) 8 (2%) 10 (3%)
23 (14%) 1 (0.6%) 0
27 (16%) 6 (4%) 5 (3%)
21 (6%)* 156 (42%) 6 (1.6%) 5 (1.3%)
2 (0.5%) 156 (42%) 6 (1.6%) 3 (0.8%)
12 (7%) 74 (44%) 5 (3%) 1 (0.6%)
0 76 (46%) 2 (1.1%) 0
*P < .05 comparing SGA with control. †Low birth weight < 2500 g.
for gestational age were compared with their matched controls (1-to-1 matching) with regard to maternal demographic data, obstetric complications, maternal interventions, neonatal morbidity, and medical therapy. Exact conditional logistic regression was performed to estimate and test the association of predictors of subnormal growth (<5th and <10th percentile).21 In examining potential differences between the SGA group and the matched controls, McNemar’s test was used to compare categorical data, and the paired t test or the Wilcoxon signed rank test was used to compare continuous outcome data. Variables not available on all infants were tested by using regression models controlling for the matching variables (age, sex, race) as the matching was broken. We furthermore examined the neonatal outcomes, controlling for identified medical or pathologic causes of intrauterine growth failure as a group, by exact stratified logistic regression (LogXact 2.1, Cambridge, Mass. Cytel Software Corp. 1996). These included major congenital malformation, multiple birth, and maternal hypertension. A single variable indicating the presence of any of these identified medical causes was included in the models. Results Perinatal correlates and neonatal outcomes at birth weights < 10th percentile for gestational age Maternal demographic, pregnancy and perinatal descriptors. A comparison of maternal demographic, medical, and previous obstetric risk factors between the mothers of SGA infants and controls is shown in Table I. Significantly more of the mothers of the SGA infants were unmarried. They had a lower mean prepregnancy weight and height
and had previously delivered significantly more low birth weight infants at term gestation compared with the control mothers (P < .05) Consideration of the pregnancy-related risk factors revealed that mothers of the SGA infants had a significantly lower weight gain during pregnancy (Table II). The rates of smoking, multiple gestation, pregnancy hypertension, and oligohydramnios were also significantly higher. The mothers of SGA infants were more likely to undergo multiple (≥ 3) antenatal ultrasonographic evaluations, biophysical profile monitoring, and oxytocin delivery induction (Table III). The rates of cesarean section were similar between groups; however, fetal distress was listed as an indication for cesarean section in 24 (45%) of 53 mothers among the SGA group as compared with 11 (20%) of 56 mothers among the AGA group (P < .05, data not shown in table). Neonatal outcomes. A comparison of the birth data and delivery room care of the SGA infants and their matched controls is shown in Table IV. Pediatricians attended significantly more of the deliveries of the SGA infants. The condition of the infants at birth did not differ, with similar rates of low Apgar scores (≤6) at 1 and 5 minutes and of the need for resuscitation. Hypothermia (< 36°C) occurred more frequently among SGA infants both at first examination and during the total hospital course. The rates of transfer of infants to the transitional and intensive care nurseries were similar. In a comparison of neonatal morbidity (Table V), 13 infants (3.5%) among the SGA group had a major congenital malformation compared with 2 (0.5%) of the AGA group. The malformations included 3 SGA infants
Doctor et al 655
Volume 185, Number 3 Am J Obstet Gynecol
Table II. Obstetric risk factors of mothers of infants with birth weights <10th and <5th percentile for gestational age and their respective controls Birth weight
Prenatal care Weight gain (lbs, mean ± SD) Substance abuse Smoking Alcohol Other Pregnancy hypertension Diabetes Oligohydramnios Antenatal ultrasound Any ≥3 Biophysical profile monitoring Urine toxin screen Of these, positive
Birth weight
<10th Percentile (n = 372)
Normal (n = 372)
<5th Percentile (n = 167)
Normal (n = 167)
367 (99%) 29 ± 15*
366 (98%) 33 ± 15
165 (99%) 29 ± 14*
164 (98%) 33 ± 15
92 (25%)* 17 (5%) 32 (9%) 53 (14%)* 11 (3%) 39 (10%)*
62 (17%) 23 (6%) 23 (6%) 27 (7%) 10 (3%) 10 (3%)
45 (27%) 8 (5%)* 18 (11%)* 27 (16%)† 4 (2%) 23 (14%)*
29 (17%) 5 (3%) 7 (4%) 15 (9%) 8 (5%) 3 (1.8%)
340 (91%)* 72 (19%)* 41 (11%)* 50 (13%) 12 (24%)
319 (86%) 27 (7%) 16 (4%) 44 (12%) 15 (34%)
159 (95%)* 38 (23%)* 20 (12%)* 28 (17%) 7 (25%)
145 (87%) 15 (9%) 6 (4%) 18 (11%) 5 (8%)
*P < .05 comparing SGA with control. †P = .065 comparing SGA with control.
Table III. Intrapartum course, labor, and delivery of infants with birth weight <10th and <5th percentile and their respective controls Birth weight
Pregnancy duration (weeks, mean ± SD) Pitocin induction Pitocin augmentation Fetal scalp pH monitoring Fetal distress† Meconium-stained fluid Fetal lie Vertex Breech/complex Mode of delivery Vertex vaginal Spontaneous Vacuum/forceps Breech vaginal Cesarean section
Birth weight
10th Percentile (n = 372)
Normal (n = 372)
<5th Percentile (n = 167)
Normal (n = 167)
38.9 ± 1 103 (28%)* 114 (31%)* 179 (48%) 66 (18%) 20 (5%)*
38.9 ± 1 61 (16%) 196 (53%) 168 (45%) 67 (18%) 6 (1.6%)
38.9 ± 1 58 (35%)* 44 (26%)* 82 (49%) 28 (17%) 7 (4%)
38.9 ± 1 26 (16%) 80 (48%) 71 (43%) 33 (20%) 2 (1.2%)
345 (93%) 24 (6%)*
362 (97%) 10 (3%)
149 (90%) 17 (10%)
159 (95%) 8 (5%)
268 (72%) 44 (12%) 5 (1.3%) 53 (14%)
269 (72%) 40 (11%) 1 (0.3%) 56 (15%)
115 (69%) 16 (10%) 5 (3%) 30 (18%)
123 (74%) 16 (10%) 1 (0.6%) 26 (16%)
*P < .05 comparing SGA with control. †Includes bradycardia, tachycardia, decreased variability, and variable and late decelerations on internal or external fetal heart rate monitoring.
with a chromosomal abnormality (trisomy 21, trisomy 18, trisomy 1/monosomy 13), 3 with major congenital heart malformations (double-outlet right ventricle, tetralogy of Fallot, and critical aortic coarctation), 3 with genitourinary anomalies (utero-pelvic junction obstruction with multicystic kidney, horseshoe kidney, and hypospadias), 1 with Prader-Willi syndrome, 1 Pierre Robin syndrome, 1 with a knee subluxation that required serial casting, and 1 with nonsyndromic multiple major malformations. The 2 AGA infants with major malformations had trisomy 21
and bilateral talipes equinovarus, respectively. Ten percent of the SGA infants were tested for cytomegalovirus; however, none were proven positive. One asymptomatic AGA infant, among the 7 infants with positive maternal Venereal Disease Research Laboratory screens, had a positive serum screen with negative spinal fluid and was treated with a single dose of intramuscular penicillin. The overall rates of respiratory distress did not differ; however, SGA infants had a higher rate of transient tachypnea of the newborn than AGA infants (2% versus
656 Doctor et al
September 2001 Am J Obstet Gynecol
Table IV. Birth data and delivery room care of infants with birth weight < 10th and < 5th percentile for gestational age and their respective controls Birth weight
Birth weight (g), mean ± SD <2500 g Multiple birth Pediatrician attendance at delivery Apgar score ≤6 (1 min) Apgar score ≤ 6 (5 min) Delivery room resuscitation Bag & mask ventilation Intubation for meconium Intubation for ventilation Chest compressions First temperature <36°C Any temperature < 36°C during stay Hospitalization Regular nursery Transitional nursery Intensive care nursery
Birth weight
10th Percentile (n = 372)
Normal (n = 372)
5th Percentile (n = 167)
Normal (n = 167)
2609 ± 206* 106 (29%) 34 (9%)* 186 (50%)*
3310 ± 301 — 8 (2%) 138 (37%)
2467 ± 184* 84 (50%) 25 (15%)* 98 (59%)*
3329 ± 320 — 5 (3%) 67 (40%)
47 (13%) 8 (2%)
42 (11%) 6 (1.6%)
27 (16%) 5 (3%)
19 (11%) 1 (0.6%)
22 (6%) 40 (11%) 9 (2%) 3 (0.8%) 40 (11%)* 101 (28%)*
33 (9%) 44 (12%) 5 (1.3%) 3 (0.8%) 7 (2%) 27 (8%)
12 (7%) 14 (8%) 7 (4%) 3 (1.8%) 26 (17%)* 51 (31%)*
14 (8%) 24 (14%) 2 (1.2%) 2 (1.2%) 3 (1.8%) 11 (7%)
328 (88%) 19 (5%) 24 (6%)
335 (90%) 20 (5%) 17 (5%)
142 (85%) 9 (5%) 16 (10%)
151 (90%) 9 (5%) 7 (4%)
*P < .05 comparing SGA with control.
0.3%, P < .05). Glucose screening was performed more frequently among the SGA infants, although the overall rates of subnormal glucose (< 40 mg/dL) among those screened were similar between groups. Significantly more of the SGA infants had symptomatic hypoglycemia (5% versus 1.1%). The symptoms of hypoglycemia among the 18 symptomatic SGA infants included hypothermia in 8, tachypnea in 5, and jitteriness in 5. Among the 4 symptomatic AGA infants, symptoms included jitteriness in 2, tachypnea in 1, and tachycardia in 1. A comparison of neonatal interventions and therapies performed during the hospital course is shown in Table VI. Eight SGA infants (2%) required endotracheal intubation and assisted ventilation as compared with 2 (0.5%) control infants (P < .05). Reasons for assisted ventilation among the SGA infants included sequelae of a major congenital malformation in 3; severe birth depression in 2; and meconium aspiration, pneumonia, and transient tachypnea in 1 infant each, respectively. The reasons in the 2 AGA infants included severe birth depression and meconium aspiration. Death occurred in 2 neonates, both in the SGA group and both as a result of major congenital malformations. One infant, with a birth weight of 2603 g, was initially discharged home, but readmitted on day of life 8 with critical coarctation of the aorta and died soon after admission. The second infant, with a birth weight of 2300 g, had trisomy 1/monosomy 13 with multiple malformations and died on day of life 2. Outcomes of the subgroups of infants with birth weights less than the 5th percentile for gestational age and between the 5th and less than the 10th percentile. The significant
comparisons, described above, for the total group of SGA infants with birth weight less than the 10th percentile for gestational age were also significant for the subgroup of infants with birth weight less than the 5th percentile for gestational age when compared with their matched controls (Tables II and V). Infants with birth weight less than the 5th percentile also required significantly more supplemental oxygen, intravenous fluids, and antibiotics and had significantly longer hospital stays than their normal birth weight controls (Table VI). Comparison of the 205 mothers of infants with birth weight between the 5th and less than the 10th percentile for gestational age with their matched controls revealed significant differences in maternal height, weight gain during pregnancy, and performance of ≥3 antenatal ultrasonography. Significant differences among the infants pertained only to hypothermia (data not shown). Relationship between causes of intrauterine growth failure and neonatal morbidity. Medical or pathologic causes of intrauterine growth failure were identified in 95 (26%) of the 372 SGA mother-infant pairs, of whom 5 had more than 1 identified cause. Of the 95, there was major congenital malformation in 13, multiple gestation in 34 (32 twin and 2 triplet), and pregnancy hypertension in 53 cases. Mother-infant pairs with more than 1 identified reason for being SGA included 4 with multiple gestation and pregnancy hypertension and 1 with multiple gestation and a major congenital malformation. Among the AGA maternal-infant control pairs, 2 had a major congenital malformation, 8 were multiple gestation, 27 had pregnancy hypertension, and 1 had suspected congenital syphilis. After controlling for these combined medical or
Doctor et al 657
Volume 185, Number 3 Am J Obstet Gynecol
Table V. Neonatal morbidity of infants with birth weight <10th and <5th percentile for gestational age and their respective controls Birth weight
Birth weight
<10th Percentile (n = 372)
Normal (n = 372)
<5th Percentile (n = 167)
Normal (n = 167)
13 (4%)* 6 (1.6%)
2 (0.5%) 13 (3%)
7 (4%)* 3 (1.8%)
0 6 (4%)
8 (2%)* 3 (0.8%) 2 (0.5%) 1 (0.3%) 176 (47%)*
1 (0.3%) 2 (0.5%) 5 (1.3%) 0 87 (23%)
6 (4%)§ 2 (1.2%) 1 (0.6%) 1 (0.6%) 106 (63%)*
1 (0.6%) 0 2 (1.2%) 0 39 (23%)
34 (19%) 20 (11%) 18 (5%)* 2 (0.5%) 107 (29%)
18 (21%) 8 (9%) 4 (1.1%) 1 (0.3%) 97 (26%)
22 (21%) 13 (12%) 10 (6%)* 1 (0.6%) 55 (33%)
9 (23%) 3 (8%) 1 (0.6%) 0 46 (28%)
3 (3%) 2 (0.5%) 92 (25%)* 8.5 ± 4
2 (0.5%) 0 66 (18%) 7.6 ± 4
3 (5%) 2 (4%) 42 (28%) 7.5 ± 4
1 (0.6%) 0 31 (19%) 6.6 ± 3
0
0
0
0
25 (7%) 3 (0.8%) 38 (10%)* 0
19 (5%) 2 (0.5%) 21 (6%) 0
18 (11%)* 2 (1.2%) 28 (17%)* 0
5 (3%) 1 (0.6%) 5 (3%) 0
Congenital anomaly Major Minor† Respiratory distress Transitory tachypnea‡ Pneumonia Meconium aspiration Pulmonary hypertension Glucose screened Of these, glucose <40 mg/dL At <2 h of age Beyond 2 h of age Symptomatic hypoglycemia Seizures Blood count performed Of these Hematocrit >65% Hematocrit >70% Bilirubin check Maximum recorded level (mg/dL, mean ± SD) Necrotizing enterocolitis Septicemia Suspect Proven Cytomegalovirus cultured Culture positive *P < .05 comparing SGA with control. †Including polydactyly. ‡With supplemental oxygen requirement. §P = .06 comparing SGA with control. P = .07 comparing SGA with control.
pathologic causes of intrauterine growth failure by exact stratified logistic regression, the differences between the SGA (birth weight less than the 10th percentile) and AGA groups of infants remained significant for the prevalence of hypothermia (both initially and during the hospital course) and for the rates of symptomatic hypoglycemia (P < .05). However, the group differences in the rate of pediatric attendance at delivery and the need for assisted ventilation no longer remained significant. The differences between the subgroup of infants with birth weight less than the 5th percentile for gestational age and their normal birth weight controls, after controlling for the identified medical or pathologic causes of intrauterine growth failure, remained significant for the prevalence of hypothermia, but not for the prevalence of symptomatic hypoglycemia. Comment We sought to describe the current neonatal morbidity and obstetric correlates of intrauterine growth failure at term gestation in an urban tertiary perinatal center. Our results reveal significantly higher rates of hypothermia
and symptomatic hypoglycemia in the total group of infants with birth weight less than the 10th percentile for gestational age when compared with infants born with birth weights AGA. The increased need for neonatal intervention and therapies pertained mainly to the subgroup of infants with birth weights less than the 5th percentile. Only 2 SGA infants died, both with major congenital malformations. Current perinatal practice includes planned preterm delivery for conditions such as severe preeclampsia and fetal growth retardation. Because we did not prospectively examine the rates of planned preterm delivery for maternal or fetal causes of intrauterine growth failure, we cannot conclude that our findings of relatively low neonatal morbidity and mortality among SGA infants delivered at term gestation are a result of this practice. The perinatal correlates of intrauterine growth failure in our population are consistent with the literature and include single marital status, lower maternal prepregnancy weight and pregnancy weight gain, smoking, pregnancy hypertension, multiple gestation, and major congenital malformation.10-15, 18, As a result of the obstetric
658 Doctor et al
September 2001 Am J Obstet Gynecol
Table VI. Neonatal interventions and therapies of infants with birth weights <10th and <5th percentile for gestational age and their respective controls Birth weight
Supplemental oxygen (days, mean ± SD) Assisted ventilation (days, mean ± SD) Umbilical artery catheterization Umbilical vein catheterization Intravenous fluids Antibiotics† Phototherapy Partial exchange transfusion Diagnostic imaging‡ Total hospital days Mean ± SD
Birth weight
<10th Percentile (n = 372)
Normal (n = 372)
<5th Percentile (n = 167)
Normal (n = 167)
16 (4%) 2.6 ± 3 8 (2%)* 3 ± 4 4 (1.1%) 6 (1.6%) 35 (9%) 27 (7%) 12 (3%)* 1 32 (9%)
10 (3%) 2.2 ± 1 2 (0.5%) 1.5 ± 1 6 (1.6%) 6 (1.6%) 21 (6%) 21 (6%) 3 (0.8%) 0 29 (8%)
12 (7%)* 3.2 ± 4 7 (4%)* 3.3 + 4 4 (2%) 6 (3%) 22 (13%)* 20 (12%)* 5 (3%) 1 19 (11%)
3 (1.8%) 1.7 ± 1 0 0 1 (0.6%) 1 (0.6%) 9 (5%) 6 (4%) 1 (0.6%) 0 12 (7%)
3.54 ± 2
3.44 ± 2
4.02 ± 3*
3.35 ± 2
*P < .05 comparing SGA with control. †Intravenous or intramuscular. ‡Includes chest x-rays, abdominal x-rays, and head ultrasonography.
complications, significantly more antenatal ultrasonography examinations were performed and significantly more deliveries were induced among mothers of SGA infants when compared with mothers of the normal birth weight infants. We defined SGA as birth weight less than the 10th percentile for gestational age because this cut-off has traditionally been used. Comparison of the neonatal morbidity among infants with birth weight less than the 5th percentile for gestational age and between the 5th and 10th percentile reveals that the majority of significant differences in neonatal morbidity are the result of the lower birth weight cut-off of less than the 5th percentile. The only significant difference between infants with birth weights between the 5th and 10th percentile for gestational age and their matched normal birth weight controls was the development of neonatal hypothermia. The differences in the rates of hypothermia and symptomatic hypoglycemia between the SGA infants and their respective controls remained significant even after controlling for identified medical or pathologic causes of intrauterine growth failure, indicating that these neonatal morbidities are related to growth failure per se rather than its cause. The differences in the rates of pediatric attendance at delivery and the need for assisted ventilation, however, no longer remained significant because these interventions are usually indicated in the presence of maternal or fetal pathology such as multiple births, maternal hypertension, and congenital malformations. We sought to describe the overall impact of intrauterine growth failure on neonatal morbidity in our perinatal center irrespective of cause. Comparison of our results with other reports are thus confounded by the inclusion of multiple births and infants with major congenital malforma-
tions and by the varying definitions of SGA between studies.2-4, 7 We did not use race or sex-specific fetal growth curves because we sought to reflect clinical practice. The national reference for fetal growth, which we used, predominantly reflects the fetal growth patterns of white US births.19 This may have resulted in defining some African-American infants as SGA, whereas they might have been defined as AGA if we had used race-specific curves. Male infants are larger than female infants at birth. The preponderance of female infants in our SGA cohort might similarly be explained by our use of the combined gender fetal growth curve. Kramer2 examined neonatal outcomes among 673 (13%) of 5319 term singleton infants without congenital malformations, born between 1980 and 1986 in Montreal, who were classified as having intrauterine growth failure according to a fetal growth ratio of less than 0.85. The rates of neonatal morbidity examined, including birth depression, hypoglycemia, and polycythemia were highest among infants with severe intrauterine growth failure, defined as a fetal growth ratio less than 0.75. Similar to our results, Kramer concluded that the rates of neonatal morbidity were associated with the severity of the intrauterine growth failure rather than its cause. More recently, McIntire3 reported on the effects of intrauterine growth failure among 11,503 term singleton infants without congenital malformations, with birth weights less than the 10th percentile who were born between 1988 and 1996 in Dallas. Significant differences in neonatal morbidity pertained mainly to infants with birth weight less than the 3rd percentile for gestational age. These differences persisted after adjusting for maternal race, parity, and infant sex. Ott,7 in contrast, in a study of 150 SGA and 807 AGA term infants without congenital malformations, found that morbidity risk was related to maternal pregnancy risk factors rather than infant size.
Doctor et al 659
Volume 185, Number 3 Am J Obstet Gynecol
Our study is important and relevant to current practice because it presents recent neonatal outcomes in an urban tertiary perinatal center. Weaknesses include its retrospective observational design, its low statistical power when examining outcomes with low incidence rates, and the fact that the results might not be applicable to other settings. Because of the low incidence of adverse outcomes in both SGA infants and AGA controls, the ability to detect small differences between groups is limited. The study is also limited by the lack of prospective laboratory information on parameters such as glucose, calcium, and hematocrit measurements. Glucose was screened in only 47% of the infants with birth weights less than the 10th percentile in our nursery, most probably because the “Lubchenco” charts were used in our nursery. These standards were normed at the altitude of Denver, Colo, where the 10th percentile birth weight for gestational age cutoff corresponds to Alexander’s 5th percentile.19 Thus, in fact, only half of the infants with birth weight less than the 10th percentile were screened according to nursery protocol. Use of more appropriate growth charts might improve the rates of glucose screening and possibly reduce the rates of symptomatic hypoglycemia at our center. There have been no recent reports of neonatal hypothermia among SGA infants since Sinclair’s review in 1970.21 The occurrence of transient hypothermia in 28% and symptomatic hypoglycemia in 5% of SGA neonates in our population can be considered morbidities of relatively high incidence considering the ongoing anticipatory pediatric surveillance in our nursery. Because hypothermia was listed as a symptom in 8 of the 18 SGA infants with symptomatic hypoglycemia, it is not clear whether the hypoglycemia predisposed to the hypothermia or the opposite. The hypothermia may, however, be partly explained by the relatively large body surface area of SGA infants and the relative lack of an insulating layer of subcutaneous adipose tissue.22 Improved attention to providing an optimal neutral thermal environment for SGA infants immediately after birth and closer ongoing monitoring of their body temperature may possibly prevent the hypothermia we have described. We conclude that while the incidence of the classic morbidities associated with intrauterine growth failure— which include birth asphyxia, meconium aspiration, and polycythemia—are now of low incidence, SGA neonates remain a high-risk population deserving of special neonatal surveillance and care. We thank Dr Mark Schluchter for his statistical expertise and Sue Bergant, Nancy Newman and Bonnie Siner for their research assistance.
REFERENCES
1. Lubchenco LO. The high risk infant: intrauterine growth and neonatal morbidity and mortality. Philadelphia: WB Saunders; 1976. 2. Kramer MS, Olivier M, McLean FG, Willis DM, Usher RH. Impact of intrauterine growth retardation and body proportionality on fetal and neonatal outcome. Pediatrics 1990;86:707-13. 3. McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in relation to morbidity and mortality among newborn infants. N Engl J Med 1999;340:1234-8. 4. Vik T, Markestad T, Ahlsten G, Gebre-Medhin M, Jacobsen G, Hoffman HJ, et al. Body proportions and early neonatal morbidity in small-for-gestational-age infants of successive births. Acta Obstet Gynecol Scand 1997;165:76-81. 5. Strauss RS, Dietz WH. Growth and development of term children born with low birth weight: effects of genetic and environmental factors. J Pediatr 1998;133:167-72. 6. Blair E, Stanley F. Intrauterine growth and spastic cerebral palsy. I. Association with birth weight for gestational age. Am J Obstet Gynecol 1990;162:229-37. 7. Ott WJ. Small for gestational age fetus and neonatal outcome: reevaluation of the relationship. Am J Perinatol 1995;12:396400. 8. Chard T, Yoong A, Macintosh M. The myth of fetal growth retardation at term. Br J Obstet Gynaecol 1993;100:1076-81. 9. Kramer MS. Socioeconomic determinants of intrauterine growth retardation. Eur J Clin Nutr 1998;52(S1):S29-33. 10. Hickey CA, Cliver SP, Goldenberg RL, Kohatsu J, Hoffman HJ. Prenatal weight gain, term birth weight, and fetal growth retardation among high-risk multiparous black and white women. Obstet Gynecol 1993;81:529-35. 11. Cliver SP, Goldenberg RL, Cutter GR, Hoffman HJ, Davis RO, Nelson KG. The effect of cigarette smoking on neonatal anthropometric measurements. Obstet Gynecol 1995;85:625-30. 12. Misra DP. The effect of the pregnancy-induced hypertension on fetal growth: a review of the literature. Paediatr Perinat Epidemiol 1996;10:244-63. 13. Haelterman E, Reart G, Paris-Llado J, Dramaix M, Tchobroutsky C. Effect of uncomplicated chronic hypertension on the risk of small-for-gestational age birth. Am J Epidemiol 1997;145:689-95. 14. Alexander GR, Kogan M, Martin J, Papiernik E. What are the fetal growth patterns of singletons, twins, and triplets in the United States? Clin Obstet Gynecol 1998;41:115-25. 15. Snijders RJM, Sherrod C, Gosden CM, Nicolaides KH. Fetal growth retardation: associated malformations and chromosomal abnormalities. Am J Obstet Gynecol 1993;168:547-55. 16. Ventura SJ, Martin JA, Curtin SC, Mathews TJ, Park MM. Births: final data for 1998. National vital statistics reports; vol 48 no 3. Hyattsville, Md: National Center for Health Statistics. 2000. 17. Ewigman BG, Crane JP, Frigoletto FD, LeFevre ML, Bain RP, McNellis D. Effect of prenatal ultrasound screening on perinatal outcome. N Engl J Med 1993;329:821-7. 18. Kramer MS, Platt R, Yang H, Joseph KS, Wen SW, Morin L, et al. Secular trends in preterm birth: a hospital-based cohort study. JAMA 1998;280:1849-54. 19. Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States National reference for fetal growth. Obstet Gynecol 1996;87:163-8. 20. Bell MJ, Ternberg JL, Feigin RD, Keating JP, Marshall R, Barton L, et al. Neonatal necrotizing enterocolitis: therapeutic decisions based on clinical staging. Ann Surg 1978;187:1-7. 21. Sinclair J. Heat production and thermoregulation in the small for date infant. Pediatr Clin North Am 1970;17:147-58. 22. Kliegman RM. Intrauterine growth retardation. In: Fanaroff AA, Martin RJ, editors. Neonatal-perinatal medicine: diseases of the fetus and infant. 6th ed. St Louis: Mosby; 1997. pp. 203-40.