Original Article
725
Outcomes for Symmetrical and Asymmetrical Small for Gestational Age Preterm Infants in Canadian Tertiary NICUs Mingyan Hei, MD1 Shoo K. Lee, MD2,3 Prakesh S. Shah, MD2,3 Amish Jain, MBBS, MRCPCH2,3 for the Canadian Neonatal Network 1 Department of Pediatrics, The Third Xiangya Hospital of Central
South University, Changsha, Hunan, China 2 Department of Pediatrics, Mount Sinai Hospital, Toronto, Ontario, Canada 3 Department of Pediatrics, University of Toronto, Toronto, Ontario, Canada
Address for correspondence Amish Jain, MBBS, MRCPCH, Room 19-231, 19th Floor, Department of Pediatrics, Mount Sinai Hospital, 600 University Avenue, Toronto, Canada M5G 1X5 (e-mail: [email protected]; [email protected]).
Abstract
Keywords
► premature ► small for gestational age ► intrauterine growth retardation ► asymmetrical
Objective This study aimed to compare the outcomes for symmetrical (sSGA) versus asymmetrical (aSGA) small for gestational age (SGA) < 32 weeks preterm infants. Study Design A total of 12,179 eligible infants admitted to the Canadian tertiary neonatal intensive care units (NICUs) over a 7-year-period in the Canadian Neonatal Network database were subcategorized as aSGA (n ¼ 536), sSGA (n ¼ 732), or appropriate for GA (AGA, n ¼ 10,911). Baseline demographics, mortality, and major morbidities were compared using multivariable analysis to calculate adjusted odds ratios (AOR) and 95% confidence interval (CI). Results sSGA had higher GA (p < 0.01) and more antenatal tobacco exposure (p ¼ 0.03). Although both the SGA groups had worse outcomes than AGA controls, aSGA versus sSGA comparison revealed lower odds (AOR [CI]) for mortality and chronic lung disease (CLD) for aSGA in the entire cohort (0.45 [0.27, 0.75] and 0.61 [0.44, 0.84], respectively) as well as for infants < 28 weeks GA (0.50 [0.27, 0.92] and 0.47 [0.29, 0.77], respectively). aSGA infants < 28 weeks GA had comparable outcomes to AGA except higher odds for CLD (1.61 [1.13, 2.27]). sSGA infants needed longer hospital stay and had higher use of oxygen and continuous positive airway pressure than aSGA. Conclusion Among SGA infants < 32 weeks GA, significant intragroup differences exist in the neonatal outcomes and hospital resource utilization depending upon the presence or absence of relative “head sparing.”
A significant proportion of premature infants admitted to neonatal intensive care units (NICUs) have a birth weight (BW) below the 10th percentile for the given gestational age (GA) and are classified as small for gestational age (SGA).1 Previous studies have highlighted that SGA preterm infants are at an increased risk of many short- and long-term complications in comparison to appropriate for gestational
age (AGA) controls. These included higher mortality as well as in-hospital morbidities such as hypothermia, hypoglycemia, respiratory distress syndrome (RDS), chronic lung disease (CLD), intraventricular hemorrhage (IVH), retinopathy of prematurity (ROP), pulmonary hemorrhage, nosocomial infections, and necrotizing enterocolitis (NEC).2–9 Further, they have been reported to require longer hospital stay and have a
received June 2, 2014 accepted after revision September 9, 2014 published online December 23, 2014
Copyright © 2015 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1395479. ISSN 0735-1631.
Downloaded by: University of Pittsburgh. Copyrighted material.
Am J Perinatol 2015;32:725–732.
Outcomes for sSGA vs. aSGA Preterm Infants in Canada higher incidence of long-term neurodevelopment problems.10,11 BW, head circumference (HC), and length are the most commonly performed anthropometric measurements to evaluate growth in preterm neonates.12 Of these, due to its relative inaccuracy and practical difficulties in obtaining reliable measurement in sick neonates, length is infrequently used by clinicians. In clinical practice, growth restriction in SGA preterm infants is commonly labeled as “symmetrical” when both the BW and HC fall below the 10th percentile or “asymmetrical” if only one parameter is below the cutoff, with usually the weight being significantly affected. Asymmetry is thought to be a result of fetal adaptation to a reduced metabolic supply and may signify a relatively short-term insult or better fetal adaptation resulting in relative “brain sparing.” The relative impact of asymmetrical versus symmetrical growth restriction on survival and common prematurity-related morbidities in preterm neonates is not known. Previous analysis using the 5-year data collected by Canadian Neonatal Network (CNN) has identified the increased risk of mortality and morbidities in SGA preterm infants born < 32 weeks gestation and admitted to Canadian tertiary NICUs.11 Our objective of this study was to compare the relative impact of symmetrical (sSGA) versus asymmetrical (aSGA) growth restriction on predischarge mortality and major neonatal morbidities as well as to compare their hospital resource utilization.
Patients and Methods
Hei et al. of SGA infants against recently published GA and genderspecific reference curves derived from the CNN.14 Following classification, the common neonatal outcomes and hospital resource utilization was compared between the groups.
Data Collection De-identified data regarding the baseline characteristics and outcomes for infants were extracted from the CNN database. The extracted data for this study included maternal history and delivery details, infant details, illness severity at birth, use of surfactant, and occurrence of various prematurity-related outcomes as defined below.
Primary Outcome The composite outcome of death before discharge or the occurrence of any of the following major morbidities: severe neurological injury defined as either grade 3 IVH or periventricular leukomalacia, stage 2 NEC, stage 3 ROP or CLD.
Secondary Outcomes Secondary outcomes included death, RDS, severe neurological injury, CLD, stage 2 NEC, nosocomial infection, and stage 3 ROP. The utilization of hospital resources was ascertained by comparing the length of hospital stay and duration of use of mechanical ventilation, continuous positive airway pressure (CPAP) and oxygen therapy.
Study Definitions
Study Design and Settings This was a retrospective study conducted using CNN database. The CNN maintains a national database where all tertiary NICUs in Canada systematically upload predefined clinical data on a daily basis for all neonates being cared for at the respective sites. The data for CNN is collected by trained data abstractors. The data abstractors are not involved in performing anthropometric measurements but have been trained to collect all the relevant data from infant health charts as recorded by the admitting physician. Using this database, all preterm infants born < 32 weeks GA and admitted to a Canadian tertiary NICU between January 1, 2003, and December 31, 2010, were identified.
Definitions in the present study were the same as previously published. In brief, GA was defined as the best obstetric estimate based on early prenatal ultrasound, obstetric examination, and obstetric history or pediatric estimate in a hierarchical fashion. Mortality was defined as death occurring before discharge from the NICU. CLD was defined as an oxygen dependency at 36 weeks corrected GA or at the time of discharge to step down neonatal units. IVH was diagnosed according to the criteria of Papile et al from the worst findings among all head ultrasounds performed during the infants stay in the NICU.15 NEC was defined according to the Bell criteria (stage 2 or higher).16 Nosocomial infection was defined as isolation of organisms from a normally sterile site (blood or cerebrospinal fluid) in a symptomatic infant.
Exclusion Criteria and Classification Exclusion criteria included if infants were deemed moribund on admission, congenital or genetic abnormalities, multiples, and missing information on BW, HC, or gender. Following the collection of anthropometric data, large for GA infants (BW > 90th percentile) were also excluded. Remaining infants included in this study and were divided into three groups: (1) Both HC and BW > 10th percentile (AGA control group), (2) Both HC and BW < 10th percentile (sSGA), and (3) HC > 10th percentile and BW < 10th percentile (aSGA). For the categorization of study infants as AGA or SGA, BWs were compared with previously published GA and gender-specific reference values derived from the Canadian population.13 Similarly, further subcategorization of the SGA group as sSGA versus aSGA was performed by plotting the HCs American Journal of Perinatology
Vol. 32
No. 8/2015
Statistical Analysis Descriptive statistical methods were used to describe the study population. The infant characteristics were compared between the three study groups using the chi-square test for categorical variables and the Student t-test for continuous variables. The comparisons between groups were further conducted when three group comparisons revealed a p < 0.05. To examine the effect of type of SGA, the total and GA-specific risk of outcomes were compared between AGA versus aSGA, AGA versus sSGA, and aSGA versus sSGA using chi-square or Fisher exact test as appropriate. Multiple logistic regression models derived by stepwise procedure were further used to determine the overall and GA-specific (< 28 weeks) effect of the type of SGA. Adjustments were
Downloaded by: University of Pittsburgh. Copyrighted material.
726
made for potential confounders including admission status, GA, Apgar score at 5 minutes, mode of delivery, maternal hypertension, receipt of antenatal corticosteroids, and gravidity. Due to highly positive skew in data on NICU resource usage, the association between SGA type and NICU resource usage was examined using negative binomial regression models adjusted for the same confounders. The data management and statistical analysis were performed using SAS 9.2 (SAS Institute, Inc., Cary, NC). p value of < 0.05 was considered significant without adjustment for multiple comparisons. The odds ratio and 95% confidence interval are reported as appropriate.
Ethics Statement The database has approval from the local research ethics board or equivalent quality improvement committees of each participating site. This project was further approved by the CNN’s executive committee.
Results A total of 26,921 infants with GA ranging from 23 to 32 weeks were admitted to tertiary NICUs in Canada during the study period. Of these, 386 were deemed moribund on admission, 5,480 infants were diagnosed with congenital/genetic anomalies, 6,218 were multiples, and 1,611 had missing information on GA, HC, or gender. These were excluded from the analysis. Overall 1,047 infants were found to be large for GA and hence were also excluded from analysis. Based on our study definitions, the remaining 12,179 eligible infants were categorized as AGA (10,911, 89.59%), sSGA (732, 4.40%), and
Hei et al.
aSGA (536, 6.01%) and their demographics and outcomes were compared. There were significant differences in the baseline characteristics of the three groups (►Table 1). The majority of the SGA infants were inborn patients, were delivered by cesarean section and had hypertensive mothers. sSGA infants were of higher GA and were more likely to be exposed to maternal tobacco use than other two groups. In comparison to the sSGA group, aSGA infants had higher incidence of maternal hypertension. Unadjusted comparison between three groups when stratified based on GA is shown in ►Table 2. Overall, both the SGA groups had worse composite outcome of death or any major morbidity as well as higher incidences of CLD and nosocomial infection as compared with AGA infants. Intergroup comparison of individual outcomes further revealed that sSGA group had a higher mortality rate but a lower incidence of grade 3 IVH while aSGA infants had a lower frequency of RDS in comparison to AGA infants. On multivariable analysis after adjusting for all predefined variables, AGA infants had the lowest odds followed by aSGA group while sSGA infants had the highest odds for the primary outcome as well as for death before discharge and CLD (►Table 3). aSGA group had lower odds for RDS. There was no intergroup difference for NEC or IVH. Both the SGA groups had higher incidence of nosocomial infection. Subgroup analyses of infants < 28 weeks GA revealed that aSGA group had similar outcomes to AGA controls except higher occurrence of CLD while sSGA group showed higher odds of mortality and CLD than both the AGA and aSGA groups (►Table 3). sSGA infants also had a higher incidence of ROP than AGA. No intergroup difference was observed for
Table 1 Characteristics of SGA and Non-SGA infants Characteristic
AGA
No. of infants
10,911
Outborn, N (%)
2,108 (19.36)
b
p-Valuea
aSGA
sSGA
536
732
91 (17.04)
96 (13.13)b
< 0.01 < 0.01
Gestational age, N (%) 23–26 wk
2,009 (18.41)b
99 (18.47)c
58 (7.92)b,c
27–28 wk
2,121 (19.44)
99 (18.47)
132 (18.03)
29–30 wk
2,887 (26.46)
136 (25.37)
217 (29.64)
31–32 wk
3,894 (35.69)
202 (37.69)
325 (44.4)
Birth weight (g), mean (SD)
1,349 (415)b,c
934 (281)b
891 (237)c
< 0.01
Male sex, N (%)
5,995 (54.94)
287 (53.54)
390 (53.28)
0.57
Apgar score at 5 min < 7, N (%)
2,187 (20.29)
125 (23.63)
146 (20.19)
0.17
Maternal age (y), mean (SD)
30.0 (6.1)b
30.8 (5.9)b
30.2 (6.0)
5,415 (49.94)
Maternal hypertension, N (%)
1,952 (18.47)
b,c
Antenatal corticosteroid, N (%)
8,651 (83.17)
436 (84.66)
592 (85.55)
0.19
Maternal tobacco use, N (%)
1,722 (15.79)c
72 (13.43)d
137 (18.72)c,d
0.03
Cesarean section, N (%)
464 (87.22)
b
358 (68.19)
b,d
0.02
b,c
643 (88.45)
c
< 0.01
416 (57.94)
c,d
< 0.01
Abbreviations: AGA, appropriate for gestational age; aSGA, asymmetric small for gestational age; sSGA, symmetric small for gestational age. a The reported p-value based on the comparisons among three groups using chi-square for categorical variables and Student t-test for continuous ones. Comparisons between groups were further conducted when p < 0.05. b,c,d The rates/means with same letters indicates significant difference between two groups. American Journal of Perinatology
Vol. 32
No. 8/2015
727
Downloaded by: University of Pittsburgh. Copyrighted material.
Outcomes for sSGA vs. aSGA Preterm Infants in Canada
Outcomes for sSGA vs. aSGA Preterm Infants in Canada
Hei et al.
Table 2 Comparison of outcomes between aSGA and sSGA groups at different gestational ages Outcomes (%)
GA groups
Mortality or major morbidity
Mortality
Severe neurological injuryd
Stage 2a NEC
Stage 3 ROP
CLD
d
d
d
23–26 wk
27–28 wk
29–30 wk
31–32 wk
aSGA vs. AGA
81.8 vs. 77.2
51.5 vs. 43.4
36.0 vs. 18.9a
11.9 vs. 8.2
38.2 vs. 30.6a
sSGA vs. AGA
89.7 vs. 77.2b
71.2 vs. 43.4a
30.4 vs. 18.9a
16.3 vs. 8.2a
36.2 vs. 30.6c
sSGA vs. aSGA
89.7 vs. 81.8
71.2 vs. 51.5c
30.4 vs. 36.0
16.3 vs. 11.9
36.2 vs. 38.2
aSGA vs. AGA
26.3 vs. 21.2
3.0 vs. 4.5
2.2 vs. 1.7
2.0 vs. 0.7b
6.7 vs. 5.5
1.5 vs. 0.7
7.9 vs. 5.5c
Nosocomial infection
a
c
a
sSGA vs. AGA
39.7 vs. 21.2
sSGA vs. aSGA
39.7 vs. 26.3
10.6 vs. 3.0b
7.4 vs. 2.2b
1.5 vs. 2.0
7.9 vs. 6.7
aSGA vs. AGA
23.0 vs. 26.8
10.2 vs. 14.1
7.6 vs. 6.8
5.7 vs. 5.3
10.4 vs. 12.6
sSGA vs. AGA
15.4 vs. 26.8
10.5 vs. 14.1
6.8 vs. 6.8
7.9 vs. 5.3
8.7 vs. 12.6c
sSGA vs. aSGA
15.4 vs. 23.0
10.5 vs. 10.2
6.8 vs. 7.6
7.9 vs. 5.7
8.7 vs. 10.4
10.6 vs. 4.5
7.4 vs. 1.7
aSGA vs. AGA
12.2 vs. 8.5
4.1 vs. 6.1
6.0 vs. 3.3
2.5 vs. 1.8
5.5 vs. 4.3
sSGA vs. AGA
10.7 vs. 8.5
8.7 vs. 6.1
3.7 vs. 3.3
2.5 vs. 1.8
4.6 vs. 4.3
sSGA vs. aSGA
10.7 vs. 12.2
8.7 vs. 4.1
3.7 vs. 6.0
2.5 vs. 2.5
4.6 vs. 5.5 c
aSGA vs. AGA
27.9 vs. 25.9
8.2 vs. 4.4
1.3 vs. 0.97
4.4 vs. 0.7
sSGA vs. AGA
35.1 vs. 25.9
12.5 vs. 4.4a
4.7 vs. 0.97a
1.4 vs. 0.7
8.2 vs. 10.2
sSGA vs. aSGA
35.1 vs. 27.9
1.4 vs. 4.4
8.2 vs. 10.2
2.7 vs. 3.7
26.8 vs. 19.1
12.5 vs. 8.2 b
aSGA vs. AGA
68.5 vs. 56.8
sSGA vs. AGA
76.3 vs. 56.8†
sSGA vs. aSGA RDS
Total
aSGA vs. AGA
76.3 vs. 68.5 88.8 vs. 89.4
47.8 vs. 29.4
4.7 vs. 1.3 a
61.7 vs. 29.4 61.7 vs. 47.8
b
81.6 vs. 81.7
23.5 vs. 11.6
a
19.9 vs. 11.6 19.9 vs. 23.5 63.0 vs. 61.9
10.2 vs. 10.2
7.8 vs. 3.7 7.8 vs. 2.7
24.7 vs. 19.1a
b
29.2 vs. 39.6
24.7 vs. 26.8 c
58.3 vs. 63.0
b
sSGA vs. AGA
96.5 vs. 89.4
86.9 vs. 81.7
64.8 vs. 61.9
38.2 vs. 39.6
59.6 vs. 63.0
sSGA vs. aSGA
96.6 vs. 88.8
86.9 vs. 81.6
64.8 vs. 63.0
38.2 vs. 29.2
59.6 vs. 58.8
aSGA vs. AGA
33.3 vs. 30.6
14.1 vs. 17.9
16.9 vs. 8.7c
12.4 vs. 2.6a
17.7 vs. 12.3a
sSGA vs. AGA
22.4 vs. 30.6
32.6 vs. 17.9a
18.0 vs. 8.7a
8.6 vs. 2.6
16.8 vs. 12.3a
22.4 vs. 33.3
c
18.0 vs. 16.9
8.6 vs. 12.4
16.8 vs. 17.7
sSGA vs. aSGA
32.6 vs. 14.1
Abbreviations: aSGA, asymmetric small for gestational age; sSGA, symmetric small for gestational age; AGA, appropriate for gestational age; RDS, respiratory distress syndrome; NEC, necrotizing enterocolitis; CLD, chronic lung disease; ROP, retinopathy of prematurity. Note: Chi-square or Fisher exact test were used as appropriate. a p < 0.001. b p < 0.05. c p < 0.01. d Considered as a major morbidity.
RDS, NEC, IVH, and nosocomial infection. AGA infants had relatively lower hospital resource utilization followed by aSGA group while sSGA infants had the longest duration of hospital stay and use of oxygen and CPAP (►Table 4).
Comparison with Literature
Discussion Main Results This is the first study reporting the relative impact of severity of growth restriction on neonatal outcomes among premature neonates. We report that although SGA preterm infants < 32 weeks GA are at an increased risk of adverse in-hospital outcomes than their appropriately grown counterparts, within the SGA group, aSGA group is associated with lower odds of mortality and respiratory complications in comparison to sSGA group. In addition, among SGA infants < 28 weeks, American Journal of Perinatology
the majority of adverse odds for in-hospital outcomes were restricted to the sSGA group. Finally, we report that the length of hospital stay and need for respiratory support was significantly longer for sSGA infants than aSGA infants.
Vol. 32
No. 8/2015
Although several studies have now established the increased burden of mortality and morbidities in SGA preterm infants, they included all the SGA infants as one cohort based on the definition of BW < 10th percentile.2–10 Though this approach is useful, it may not tell the whole story given the differences in causes and types of SGA. In the present study, we attempted to further explore this population by subdividing the SGA group into symmetrical and asymmetrical based on simultaneous presence or absence of significant restriction in brain growth, respectively, and found important
Downloaded by: University of Pittsburgh. Copyrighted material.
728
0.68 (0.53, 0.89)
Nosocomial infection
All infants
0.54 (0.43, 0.68)
1.14 (0.95, 1.37)
0.41 (0.32, 0.50)
0.40 (0.26, 0.63)
0.73 (0.49, 1.08)
0.90 (0.66, 1.24)
0.27 (0.19, 0.39)
0.40 (0.33, 0.49)
AGA vs. sSGA
0.79 (0.57, 1.11)
0.72 (0.55, 0.95)
0.61 (0.44, 0.84)
0.56 (0.30, 1.05)
0.97 (0.56, 1.67)
0.93 (0.59, 1.45)
0.45 (0.27, 0.75)
0.65 (0.49, 0.88)
aSGA vs. sSGA
1.08 (0.76, 1.54)
0.89 (0.55, 1.46)
0.62 (0.44, 0.88)
0.74 (0.45, 1.21)
0.94 (0.54, 1.65)
1.12 (0.71, 1.77)
0.77 (0.49, 1.22)
0.73(0.52, 1.02)
AGA vs. aSGA
0.74 (0.53, 1.04)
0.62 (0.36,1.08)
0.29 (0.20, 0.43)
0.45 (0.27, 0.75)
0.78 (0.44, 1.37)
1.43 (0.86, 2.37)
0.39 (0.25, 0.60)
0.31(0.21, 0.45)
AGA vs. sSGA
Infants < 28 wk GA
0.68 (0.43, 1.10)
0.70 (0.34, 1.43)
0.47 (0.29, 0.77)
0.61 (0.31, 1.22)
0.83 (0.38, 1.79)
1.27 (0.66, 2.43)
0.50 (0.27, 0.92)
0.43(0.26, 0.70)
aSGA vs. sSGA
29 (13–54)
Length of hospital stay
35 (16 63)
g
39 (20 64)
1 (0–7)
2 (0–12)g
1 (0–12)
sSGA
h
< 0.0001
0.18
0.024
0.44
p-Valueb
0.15 (0.042)
0.32 (0.038)f
0.57 (0.06)f f
0.35 (0.07)f
0.70 (0.078)f
Estimate (se)
AGA vs. sSGA
0.54 (0.054)f
d
0.14 (0.061)e
0.20 (0.09)e
Estimate (se)
c
AGA vs. aSGA
Multivariable analyses a
0.12 (0.052)e
0.14 (0.10)
0.33 (0.089)f
0.42 (0.12)f
Estimate (se)
aSGA vs. sSGA
American Journal of Perinatology
Vol. 32
Hei et al.
No. 8/2015
Downloaded by: University of Pittsburgh. Copyrighted material.
Linear model was used for length of hospital in log scale and zero-inflated negative binomial models were used for other outcomes. The covariates included were gender, gestational age, Apgar score at 5 min, delivery type, admission status, maternal hypertension, smoking status, and steroid use. b Based on the comparison among three groups using the Mood median test. Comparisons between groups were further conducted when p < 0.05. c Estimated coefficients. d Standard error. The reported estimates were based on the final model derived by back forward selection procedures. e p < 0.05. f p < 0.001. g The medians (p < 0.01) indicate significant difference between the two groups. h The medians p < 0.05) indicate significant difference between the two groups.
a
1 (0–5)
Duration of ventilation
2 (0–11)
3 (0–12)g,h
2 (0–11)g
Duration of CPAP g,h
1 (0–12)
aSGA
1 (0–8)
AGA
Median test (IQ range)
Duration of oxygen
Use of resources
Table 4 Hospital resource utilization
Abbreviations: AGA, appropriate for gestational age; aSGA, asymmetric small for gestational age; CI, confidence interval; CLD, chronic lung disease; GA, gestational age; NEC, necrotizing enterocolitis; NICU, neonatal intensive care unit; OR, odds ratio; RDS, respiratory distress syndrome; ROP, retinopathy of prematurity; sSGA, symmetric small for gestational age. Notes: Multiple logistic regression models were used. Covariates included were gender, GA, Apgar score at 5 min, delivery mode, admission status, maternal hypertension, smoking status, and steroid use. The ORs were estimated based on the final model derived by stepwise procedure. a Considered as a major morbidity.
1.58 (1.27, 1.97)
0.67 (0.51, 0.88)
0.72 (0.45, 1.15)
0.76 (0.50, 1.14)
0.97 (0.68, 1.38)
RDS
CLD
a
Stage 3 ROP
a
Stage 2a NECa
Severe neurological injury
0.60 (0.39, 0.92)
Mortality
a
0.61 (0.49, 0.78)
AGA vs. aSGA
Mortality or major morbidity
Outcomes
Table 3 Comparison of outcomes: adjusted OR (95% CI) after multivariable analyses
Outcomes for sSGA vs. aSGA Preterm Infants in Canada 729
Outcomes for sSGA vs. aSGA Preterm Infants in Canada differences. This has not been reported previously. Although not directly comparable to our study population, similar associations between the proportionality of growth restriction and adverse antenatal or perinatal outcomes as well as respiratory health during 1st year of life have been previously reported in term SGA infants.17–21 The biological mechanisms responsible for relative increase in mortality and morbidities among SGA preterm infants are not well defined. Most experimental studies utilizing different animal models of in utero placental insufficiency have focused on the effect on lung maturation and surfactant production and have shown conflicting results. One explanation for these observed differences could be the heterogeneity in the experimental methods used in these studies. Experiments which used models of relatively acute fetal hypoxemia and at a relatively higher GA have shown an increase in fetal cortisol levels and an associated rise in surfactant production.22 On the other hand, a more recent study using a sheep model of chronic fetal hypoxia (created by preconception removal of endometrial caruncles from the uterus, which results in the formation of smaller placenta and growth-retarded fetuses) has demonstrated worse lung maturity and reduced surfactant production.23 On similar lines, one of the potential explanations for the observed differences in respiratory outcomes between sSGA and aSGA infants in our study could be the differences in the severity or duration of fetal hypoxia. It’s plausible that aSGA infants suffered from a more acute but relatively late onset insult while sSGA infants faced a more chronic placental insufficiency. Unfortunately, due to lack of data we are unable to confirm the antenatal timing of onset of growth restriction to test this hypothesis. However, the observation that relatively more aSGA infants had hypertensive mothers and were delivered at a lower GA along with the fact that the majority of babies in both the SGA groups were delivered by cesarean section points toward higher severity of fetal insult in the aSGA group necessitating emergency delivery. Further, experimental studies investigating the effect of intrauterine growth retardation (IUGR) secondary to placental insufficiency on the postnatal pulmonary function have shown that several weeks after birth IUGR animals continue to demonstrate impaired pulmonary function, abnormal airway structure, as well as worse lung and airway compliance.24,25 These effects along with impaired lung function at the time of birth may explain our findings of an increased incidence of respiratory complications and a longer duration of respiratory support for SGA infants compared with AGA and for sSGA infants compared with aSGA. We also observed an increase in mortality among sSGA infants. Unfortunately, our database does not collect information regarding specific cause of death. The fact that other major “life-threatening” morbidities (IVH, NEC, and nosocomial infections) were similar between the two SGA groups suggests that the increased mortality observed in our study was likely related to the respiratory complications. Others have described the patterns of mortality among SGA preterm infants and have also reported higher mortality with most deaths occurring in the first 4 weeks of life and cardiorespiratory failure being reported as the most common American Journal of Perinatology
Vol. 32
No. 8/2015
Hei et al. cause of death.26,27 Our study suggests that preterm SGA is not a homogenous group and the prognosis may vary depending on the antenatal severity and duration of growth restriction in spite of similar BWs. We hypothesize that infants who had relatively preserved brain growth are likely to have been exposed to less severe or shorter insults and hence may have better outcomes than constitutionally small preterm SGA infants. Ponderal index (PI) is the most common previously described parameter used to subcategorize growth restriction in term SGA infants.17–21 It is calculated from infants weight and crown–heel length (CHL) (PI ¼ [weight {grams}/[CHL {cm}]3 100). The resultant value is then plotted on a reference curve and SGA infants with PI < 10th percentile is classified as asymmetric while others are considered as symmetric. PI has not been validated for use in extreme preterm infants. In addition, studies using PI for term SGA infants has provided contrasting results with some reporting worse perinatal outcomes in infants classified as “asymmetric” while other suggesting “symmetric” SGA infants to be worse. In addition, the majority of these studies is older than 20 years and were conducted in resource-limited settings, where incidence of SGA in term infants is higher and the etiology for SGA may be different (placental dysfunction vs. maternal nutritional status). Another major limitation was that in most of these studies GA was determined based on postnatal examination. This could have resulted in many well late preterm infants being misclassified as sSGA. A study evaluating the use of PI or CHL < 10th percentile (i.e., sSGA) reported that the SGA term infants had worse perinatal outcomes irrespective of the definition used for classification.15 The rationale of using these indices was to account for the timing of in utero event as fetal compromise during the second trimester is likely to affect both the weight and length (symmetrical) while third trimester insult will affect the weight more than the length (asymmetrical). In this study, we choose to subdivide the SGA population based on HC < 10th percentile. The use of HC as a classifying parameter has an advantage over the length as during the period of growth restriction “diving reflex” would initiate the preservation of blood flow to the brain. In addition, measurement of length at birth may have significant methodological issues and high interobserver variability.
Study Limitations In addition to the well-known limitations of retrospective database-driven studies, there are several other limitations of our study. First, although the majority of SGA is expected to be a result of placental dysfunction we are unable to confirm this as specific data on the etiology were not available. This is even more important for the sSGA group as many factors, such as congenital infections or genetic abnormalities, can also result in symmetrical growth restriction. Even though we excluded all such known cases, it is possible that some of the SGA infants included in this study had an undiagnosed underlying genetic etiology. However, the high percentage of hypertensive mothers in both groups suggests placental dysfunction to be the primary etiology. Second, due to lack of data we were
Downloaded by: University of Pittsburgh. Copyrighted material.
730
Outcomes for sSGA vs. aSGA Preterm Infants in Canada
2 Bardin C, Zelkowitz P, Papageorgiou A. Outcome of small-for-
3
4
5
6
7
8
Conclusion In conclusion, among extreme preterm SGA infants, significant differences exist in the neonatal outcomes as well as hospital resource utilization depending upon the presence or absence of relative “head sparing.” This study highlights the importance of a more comprehensive anthropometric assessment of growth restriction while evaluating and counseling families of growth-restricted infants. These results may help in targeting interventions and counseling families. Future studies should investigate the impact of symmetry of growth restriction on long-term neurodevelopmental outcomes.
Acknowledgments The authors would like to thank the Canadian Institutes of Health Research team in Maternal-Infant Care (MiCare) for providing organizational support to the Canadian Neonatal Network. Authors would also like to acknowledge the contribution of Mr. Xiang Y. Ye, Statistician at the MiCare Research Center, Mount Sinai Hospital, Toronto, Canada, for his assistance with the analysis for this project. MiCare Research Centre at Mount Sinai Hospital, Toronto, Ontario, is financially supported by Ontario Ministry of Health and Long-Term Care. Dr. Mingyan Hei’s fellowship training in Neonatal Intensive Care Unit of Mount Sinai Hospital was supported by the policy of “125-Distinguished Clinicians of the Third Xiangya Hospital of Central South University, China.”
9
10
11
12 13
14
15
16
17
18
19
20
References 1 Zaw W, Gagnon R, da Silva O. The risks of adverse neonatal
outcome among preterm small for gestational age infants according to neonatal versus fetal growth standards. Pediatrics 2003; 111(6 Pt 1):1273–1277
731
21
gestational age and appropriate-for-gestational age infants born before 27 weeks of gestation. Pediatrics 1997;100(2):E4 Bernstein IM, Horbar JD, Badger GJ, Ohlsson A, Golan A; The Vermont Oxford Network. Morbidity and mortality among verylow-birth-weight neonates with intrauterine growth restriction. Am J Obstet Gynecol 2000;182(1 Pt 1):198–206 Garite TJ, Clark R, Thorp JA. Intrauterine growth restriction increases morbidity and mortality among premature neonates. Am J Obstet Gynecol 2004;191(2):481–487 Lal MK, Manktelow BN, Draper ES, Field DJ; Population-based study. Chronic lung disease of prematurity and intrauterine growth retardation: a population-based study. Pediatrics 2003; 111(3):483–487 Piper JM, Xenakis EM, McFarland M, Elliott BD, Berkus MD, Langer O. Do growth-retarded premature infants have different rates of perinatal morbidity and mortality than appropriately grown premature infants? Obstet Gynecol 1996;87(2):169–174 Regev RH, Lusky A, Dolfin T, Litmanovitz I, Arnon S, Reichman B; Israel Neonatal Network. Excess mortality and morbidity among small-for-gestational-age premature infants: a population-based study. J Pediatr 2003;143(2):186–191 Reiss I, Landmann E, Heckmann M, Misselwitz B, Gortner L. Increased risk of bronchopulmonary dysplasia and increased mortality in very preterm infants being small for gestational age. Arch Gynecol Obstet 2003;269(1):40–44 Simchen MJ, Beiner ME, Strauss-Liviathan N, et al. Neonatal outcome in growth-restricted versus appropriately grown preterm infants. Am J Perinatol 2000;17(4):187–192 Kok JH, den Ouden AL, Verloove-Vanhorick SP, Brand R. Outcome of very preterm small for gestational age infants: the first nine years of life. Br J Obstet Gynaecol 1998;105(2):162–168 Qiu X, Lodha A, Shah PS, et al; Canadian Neonatal Network. Neonatal outcomes of small for gestational age preterm infants in Canada. Am J Perinatol 2012;29(2):87–94 Lockwood CJ, Weiner S. Assessment of fetal growth. Clin Perinatol 1986;13(1):3–35 Kramer MS, Platt RW, Wen SW, et al; Fetal/Infant Health Study Group of the Canadian Perinatal Surveillance System. A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics 2001;108(2):E35 Barbier A, Boivin A, Yoon W, et al; Canadian Neonatal Network. New reference curves for head circumference at birth, by gestational age. Pediatrics 2013;131(4):e1158–e1167 Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978; 92(4):529–534 Bell MJ, Ternberg JL, Feigin RD, et al. Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging. Ann Surg 1978;187(1):1–7 Balcazar H, Haas J. Classification schemes of small-for-gestational age and type of intrauterine growth retardation and its implications to early neonatal mortality. Early Hum Dev 1990;24(3): 219–230 Caulfield LE, Haas JD, Belizán JM, Rasmussen KM, Edmonston B. Differences in early postnatal morbidity risk by pattern of fetal growth in Argentina. Paediatr Perinat Epidemiol 1991;5(3): 263–275 Kramer MS, Olivier M, McLean FH, Willis DM, Usher RH. Impact of intrauterine growth retardation and body proportionality on fetal and neonatal outcome. Pediatrics 1990;86(5):707–713 Vik T, Markestad T, Ahlsten G, et al. Body proportions and early neonatal morbidity in small-for-gestational-age infants of successive births. Acta Obstet Gynecol Scand Suppl 1997;165:76–81 Vik T, Vatten L, Markestad T, Ahlsten G, Jacobsen G, Bakketeig LS. Morbidity during the first year of life in small for gestational age infants. Arch Dis Child Fetal Neonatal Ed 1996;75(1):F33–F37 American Journal of Perinatology
Vol. 32
No. 8/2015
Downloaded by: University of Pittsburgh. Copyrighted material.
unable to adjust for the severity of fetal growth restriction (FGR). This may be an important confounder as reported in a study in term infants where the differences in outcome based on proportionality became less significant after adjustments for the severity of FGR.19 Third, we do not have long-term neurodevelopmental outcome data on these infants which would be an important subject for future studies. Fourth, due to lack of data on serial growth parameters, we could not assess the groups for differences in postnatal growth and its influence on neonatal outcomes. Finally, we acknowledge that using an arbitrary cutoff of < 10th percentile will result in misclassification for some infants. For instance, an infant with BW at 9th and HC at the 11th percentile will be classified as asymmetric while an infant with BW at 2nd and HC at 9th will be called symmetric, although the former is more symmetric than the later. Nevertheless, we did find important differences in the outcomes of infants classified using HC < 10th percentile cutoff suggesting that a degree of misclassification may be random.
Hei et al.
Outcomes for sSGA vs. aSGA Preterm Infants in Canada
Hei et al.
22 Gagnon R, Langridge J, Inchley K, Murotsuki J, Possmayer F.
25 Joyce BJ, Louey S, Davey MG, Cock ML, Hooper SB, Harding R.
Changes in surfactant-associated protein mRNA profile in growth-restricted fetal sheep. Am J Physiol 1999;276(3 Pt 1): L459–L465 23 Orgeig S, Crittenden TA, Marchant C, McMillen IC, Morrison JL. Intrauterine growth restriction delays surfactant protein maturation in the sheep fetus. Am J Physiol Lung Cell Mol Physiol 2010; 298(4):L575–L583 24 Wignarajah D, Cock ML, Pinkerton KE, Harding R. Influence of intrauterine growth restriction on airway development in fetal and postnatal sheep. Pediatr Res 2002;51(6):681–688
Compromised respiratory function in postnatal lambs after placental insufficiency and intrauterine growth restriction. Pediatr Res 2001;50(5):641–649 26 Giapros V, Drougia A, Krallis N, Theocharis P, Andronikou S. Morbidity and mortality patterns in small-for-gestational age infants born preterm. J Matern Fetal Neonatal Med 2012;25(2):153–157 27 Bartels DB, Kreienbrock L, Dammann O, Wenzlaff P, Poets CF. Population based study on the outcome of small for gestational age newborns. Arch Dis Child Fetal Neonatal Ed 2005;90(1): F53–F59
Downloaded by: University of Pittsburgh. Copyrighted material.
732
American Journal of Perinatology
Vol. 32
No. 8/2015
725
Outcomes for Symmetrical and Asymmetrical Small for Gestational Age Preterm Infants in Canadian Tertiary NICUs Mingyan Hei, MD1 Shoo K. Lee, MD2,3 Prakesh S. Shah, MD2,3 Amish Jain, MBBS, MRCPCH2,3 for the Canadian Neonatal Network 1 Department of Pediatrics, The Third Xiangya Hospital of Central
South University, Changsha, Hunan, China 2 Department of Pediatrics, Mount Sinai Hospital, Toronto, Ontario, Canada 3 Department of Pediatrics, University of Toronto, Toronto, Ontario, Canada
Address for correspondence Amish Jain, MBBS, MRCPCH, Room 19-231, 19th Floor, Department of Pediatrics, Mount Sinai Hospital, 600 University Avenue, Toronto, Canada M5G 1X5 (e-mail: [email protected]; [email protected]).
Abstract
Keywords
► premature ► small for gestational age ► intrauterine growth retardation ► asymmetrical
Objective This study aimed to compare the outcomes for symmetrical (sSGA) versus asymmetrical (aSGA) small for gestational age (SGA) < 32 weeks preterm infants. Study Design A total of 12,179 eligible infants admitted to the Canadian tertiary neonatal intensive care units (NICUs) over a 7-year-period in the Canadian Neonatal Network database were subcategorized as aSGA (n ¼ 536), sSGA (n ¼ 732), or appropriate for GA (AGA, n ¼ 10,911). Baseline demographics, mortality, and major morbidities were compared using multivariable analysis to calculate adjusted odds ratios (AOR) and 95% confidence interval (CI). Results sSGA had higher GA (p < 0.01) and more antenatal tobacco exposure (p ¼ 0.03). Although both the SGA groups had worse outcomes than AGA controls, aSGA versus sSGA comparison revealed lower odds (AOR [CI]) for mortality and chronic lung disease (CLD) for aSGA in the entire cohort (0.45 [0.27, 0.75] and 0.61 [0.44, 0.84], respectively) as well as for infants < 28 weeks GA (0.50 [0.27, 0.92] and 0.47 [0.29, 0.77], respectively). aSGA infants < 28 weeks GA had comparable outcomes to AGA except higher odds for CLD (1.61 [1.13, 2.27]). sSGA infants needed longer hospital stay and had higher use of oxygen and continuous positive airway pressure than aSGA. Conclusion Among SGA infants < 32 weeks GA, significant intragroup differences exist in the neonatal outcomes and hospital resource utilization depending upon the presence or absence of relative “head sparing.”
A significant proportion of premature infants admitted to neonatal intensive care units (NICUs) have a birth weight (BW) below the 10th percentile for the given gestational age (GA) and are classified as small for gestational age (SGA).1 Previous studies have highlighted that SGA preterm infants are at an increased risk of many short- and long-term complications in comparison to appropriate for gestational
age (AGA) controls. These included higher mortality as well as in-hospital morbidities such as hypothermia, hypoglycemia, respiratory distress syndrome (RDS), chronic lung disease (CLD), intraventricular hemorrhage (IVH), retinopathy of prematurity (ROP), pulmonary hemorrhage, nosocomial infections, and necrotizing enterocolitis (NEC).2–9 Further, they have been reported to require longer hospital stay and have a
received June 2, 2014 accepted after revision September 9, 2014 published online December 23, 2014
Copyright © 2015 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1395479. ISSN 0735-1631.
Downloaded by: University of Pittsburgh. Copyrighted material.
Am J Perinatol 2015;32:725–732.
Outcomes for sSGA vs. aSGA Preterm Infants in Canada higher incidence of long-term neurodevelopment problems.10,11 BW, head circumference (HC), and length are the most commonly performed anthropometric measurements to evaluate growth in preterm neonates.12 Of these, due to its relative inaccuracy and practical difficulties in obtaining reliable measurement in sick neonates, length is infrequently used by clinicians. In clinical practice, growth restriction in SGA preterm infants is commonly labeled as “symmetrical” when both the BW and HC fall below the 10th percentile or “asymmetrical” if only one parameter is below the cutoff, with usually the weight being significantly affected. Asymmetry is thought to be a result of fetal adaptation to a reduced metabolic supply and may signify a relatively short-term insult or better fetal adaptation resulting in relative “brain sparing.” The relative impact of asymmetrical versus symmetrical growth restriction on survival and common prematurity-related morbidities in preterm neonates is not known. Previous analysis using the 5-year data collected by Canadian Neonatal Network (CNN) has identified the increased risk of mortality and morbidities in SGA preterm infants born < 32 weeks gestation and admitted to Canadian tertiary NICUs.11 Our objective of this study was to compare the relative impact of symmetrical (sSGA) versus asymmetrical (aSGA) growth restriction on predischarge mortality and major neonatal morbidities as well as to compare their hospital resource utilization.
Patients and Methods
Hei et al. of SGA infants against recently published GA and genderspecific reference curves derived from the CNN.14 Following classification, the common neonatal outcomes and hospital resource utilization was compared between the groups.
Data Collection De-identified data regarding the baseline characteristics and outcomes for infants were extracted from the CNN database. The extracted data for this study included maternal history and delivery details, infant details, illness severity at birth, use of surfactant, and occurrence of various prematurity-related outcomes as defined below.
Primary Outcome The composite outcome of death before discharge or the occurrence of any of the following major morbidities: severe neurological injury defined as either grade 3 IVH or periventricular leukomalacia, stage 2 NEC, stage 3 ROP or CLD.
Secondary Outcomes Secondary outcomes included death, RDS, severe neurological injury, CLD, stage 2 NEC, nosocomial infection, and stage 3 ROP. The utilization of hospital resources was ascertained by comparing the length of hospital stay and duration of use of mechanical ventilation, continuous positive airway pressure (CPAP) and oxygen therapy.
Study Definitions
Study Design and Settings This was a retrospective study conducted using CNN database. The CNN maintains a national database where all tertiary NICUs in Canada systematically upload predefined clinical data on a daily basis for all neonates being cared for at the respective sites. The data for CNN is collected by trained data abstractors. The data abstractors are not involved in performing anthropometric measurements but have been trained to collect all the relevant data from infant health charts as recorded by the admitting physician. Using this database, all preterm infants born < 32 weeks GA and admitted to a Canadian tertiary NICU between January 1, 2003, and December 31, 2010, were identified.
Definitions in the present study were the same as previously published. In brief, GA was defined as the best obstetric estimate based on early prenatal ultrasound, obstetric examination, and obstetric history or pediatric estimate in a hierarchical fashion. Mortality was defined as death occurring before discharge from the NICU. CLD was defined as an oxygen dependency at 36 weeks corrected GA or at the time of discharge to step down neonatal units. IVH was diagnosed according to the criteria of Papile et al from the worst findings among all head ultrasounds performed during the infants stay in the NICU.15 NEC was defined according to the Bell criteria (stage 2 or higher).16 Nosocomial infection was defined as isolation of organisms from a normally sterile site (blood or cerebrospinal fluid) in a symptomatic infant.
Exclusion Criteria and Classification Exclusion criteria included if infants were deemed moribund on admission, congenital or genetic abnormalities, multiples, and missing information on BW, HC, or gender. Following the collection of anthropometric data, large for GA infants (BW > 90th percentile) were also excluded. Remaining infants included in this study and were divided into three groups: (1) Both HC and BW > 10th percentile (AGA control group), (2) Both HC and BW < 10th percentile (sSGA), and (3) HC > 10th percentile and BW < 10th percentile (aSGA). For the categorization of study infants as AGA or SGA, BWs were compared with previously published GA and gender-specific reference values derived from the Canadian population.13 Similarly, further subcategorization of the SGA group as sSGA versus aSGA was performed by plotting the HCs American Journal of Perinatology
Vol. 32
No. 8/2015
Statistical Analysis Descriptive statistical methods were used to describe the study population. The infant characteristics were compared between the three study groups using the chi-square test for categorical variables and the Student t-test for continuous variables. The comparisons between groups were further conducted when three group comparisons revealed a p < 0.05. To examine the effect of type of SGA, the total and GA-specific risk of outcomes were compared between AGA versus aSGA, AGA versus sSGA, and aSGA versus sSGA using chi-square or Fisher exact test as appropriate. Multiple logistic regression models derived by stepwise procedure were further used to determine the overall and GA-specific (< 28 weeks) effect of the type of SGA. Adjustments were
Downloaded by: University of Pittsburgh. Copyrighted material.
726
made for potential confounders including admission status, GA, Apgar score at 5 minutes, mode of delivery, maternal hypertension, receipt of antenatal corticosteroids, and gravidity. Due to highly positive skew in data on NICU resource usage, the association between SGA type and NICU resource usage was examined using negative binomial regression models adjusted for the same confounders. The data management and statistical analysis were performed using SAS 9.2 (SAS Institute, Inc., Cary, NC). p value of < 0.05 was considered significant without adjustment for multiple comparisons. The odds ratio and 95% confidence interval are reported as appropriate.
Ethics Statement The database has approval from the local research ethics board or equivalent quality improvement committees of each participating site. This project was further approved by the CNN’s executive committee.
Results A total of 26,921 infants with GA ranging from 23 to 32 weeks were admitted to tertiary NICUs in Canada during the study period. Of these, 386 were deemed moribund on admission, 5,480 infants were diagnosed with congenital/genetic anomalies, 6,218 were multiples, and 1,611 had missing information on GA, HC, or gender. These were excluded from the analysis. Overall 1,047 infants were found to be large for GA and hence were also excluded from analysis. Based on our study definitions, the remaining 12,179 eligible infants were categorized as AGA (10,911, 89.59%), sSGA (732, 4.40%), and
Hei et al.
aSGA (536, 6.01%) and their demographics and outcomes were compared. There were significant differences in the baseline characteristics of the three groups (►Table 1). The majority of the SGA infants were inborn patients, were delivered by cesarean section and had hypertensive mothers. sSGA infants were of higher GA and were more likely to be exposed to maternal tobacco use than other two groups. In comparison to the sSGA group, aSGA infants had higher incidence of maternal hypertension. Unadjusted comparison between three groups when stratified based on GA is shown in ►Table 2. Overall, both the SGA groups had worse composite outcome of death or any major morbidity as well as higher incidences of CLD and nosocomial infection as compared with AGA infants. Intergroup comparison of individual outcomes further revealed that sSGA group had a higher mortality rate but a lower incidence of grade 3 IVH while aSGA infants had a lower frequency of RDS in comparison to AGA infants. On multivariable analysis after adjusting for all predefined variables, AGA infants had the lowest odds followed by aSGA group while sSGA infants had the highest odds for the primary outcome as well as for death before discharge and CLD (►Table 3). aSGA group had lower odds for RDS. There was no intergroup difference for NEC or IVH. Both the SGA groups had higher incidence of nosocomial infection. Subgroup analyses of infants < 28 weeks GA revealed that aSGA group had similar outcomes to AGA controls except higher occurrence of CLD while sSGA group showed higher odds of mortality and CLD than both the AGA and aSGA groups (►Table 3). sSGA infants also had a higher incidence of ROP than AGA. No intergroup difference was observed for
Table 1 Characteristics of SGA and Non-SGA infants Characteristic
AGA
No. of infants
10,911
Outborn, N (%)
2,108 (19.36)
b
p-Valuea
aSGA
sSGA
536
732
91 (17.04)
96 (13.13)b
< 0.01 < 0.01
Gestational age, N (%) 23–26 wk
2,009 (18.41)b
99 (18.47)c
58 (7.92)b,c
27–28 wk
2,121 (19.44)
99 (18.47)
132 (18.03)
29–30 wk
2,887 (26.46)
136 (25.37)
217 (29.64)
31–32 wk
3,894 (35.69)
202 (37.69)
325 (44.4)
Birth weight (g), mean (SD)
1,349 (415)b,c
934 (281)b
891 (237)c
< 0.01
Male sex, N (%)
5,995 (54.94)
287 (53.54)
390 (53.28)
0.57
Apgar score at 5 min < 7, N (%)
2,187 (20.29)
125 (23.63)
146 (20.19)
0.17
Maternal age (y), mean (SD)
30.0 (6.1)b
30.8 (5.9)b
30.2 (6.0)
5,415 (49.94)
Maternal hypertension, N (%)
1,952 (18.47)
b,c
Antenatal corticosteroid, N (%)
8,651 (83.17)
436 (84.66)
592 (85.55)
0.19
Maternal tobacco use, N (%)
1,722 (15.79)c
72 (13.43)d
137 (18.72)c,d
0.03
Cesarean section, N (%)
464 (87.22)
b
358 (68.19)
b,d
0.02
b,c
643 (88.45)
c
< 0.01
416 (57.94)
c,d
< 0.01
Abbreviations: AGA, appropriate for gestational age; aSGA, asymmetric small for gestational age; sSGA, symmetric small for gestational age. a The reported p-value based on the comparisons among three groups using chi-square for categorical variables and Student t-test for continuous ones. Comparisons between groups were further conducted when p < 0.05. b,c,d The rates/means with same letters indicates significant difference between two groups. American Journal of Perinatology
Vol. 32
No. 8/2015
727
Downloaded by: University of Pittsburgh. Copyrighted material.
Outcomes for sSGA vs. aSGA Preterm Infants in Canada
Outcomes for sSGA vs. aSGA Preterm Infants in Canada
Hei et al.
Table 2 Comparison of outcomes between aSGA and sSGA groups at different gestational ages Outcomes (%)
GA groups
Mortality or major morbidity
Mortality
Severe neurological injuryd
Stage 2a NEC
Stage 3 ROP
CLD
d
d
d
23–26 wk
27–28 wk
29–30 wk
31–32 wk
aSGA vs. AGA
81.8 vs. 77.2
51.5 vs. 43.4
36.0 vs. 18.9a
11.9 vs. 8.2
38.2 vs. 30.6a
sSGA vs. AGA
89.7 vs. 77.2b
71.2 vs. 43.4a
30.4 vs. 18.9a
16.3 vs. 8.2a
36.2 vs. 30.6c
sSGA vs. aSGA
89.7 vs. 81.8
71.2 vs. 51.5c
30.4 vs. 36.0
16.3 vs. 11.9
36.2 vs. 38.2
aSGA vs. AGA
26.3 vs. 21.2
3.0 vs. 4.5
2.2 vs. 1.7
2.0 vs. 0.7b
6.7 vs. 5.5
1.5 vs. 0.7
7.9 vs. 5.5c
Nosocomial infection
a
c
a
sSGA vs. AGA
39.7 vs. 21.2
sSGA vs. aSGA
39.7 vs. 26.3
10.6 vs. 3.0b
7.4 vs. 2.2b
1.5 vs. 2.0
7.9 vs. 6.7
aSGA vs. AGA
23.0 vs. 26.8
10.2 vs. 14.1
7.6 vs. 6.8
5.7 vs. 5.3
10.4 vs. 12.6
sSGA vs. AGA
15.4 vs. 26.8
10.5 vs. 14.1
6.8 vs. 6.8
7.9 vs. 5.3
8.7 vs. 12.6c
sSGA vs. aSGA
15.4 vs. 23.0
10.5 vs. 10.2
6.8 vs. 7.6
7.9 vs. 5.7
8.7 vs. 10.4
10.6 vs. 4.5
7.4 vs. 1.7
aSGA vs. AGA
12.2 vs. 8.5
4.1 vs. 6.1
6.0 vs. 3.3
2.5 vs. 1.8
5.5 vs. 4.3
sSGA vs. AGA
10.7 vs. 8.5
8.7 vs. 6.1
3.7 vs. 3.3
2.5 vs. 1.8
4.6 vs. 4.3
sSGA vs. aSGA
10.7 vs. 12.2
8.7 vs. 4.1
3.7 vs. 6.0
2.5 vs. 2.5
4.6 vs. 5.5 c
aSGA vs. AGA
27.9 vs. 25.9
8.2 vs. 4.4
1.3 vs. 0.97
4.4 vs. 0.7
sSGA vs. AGA
35.1 vs. 25.9
12.5 vs. 4.4a
4.7 vs. 0.97a
1.4 vs. 0.7
8.2 vs. 10.2
sSGA vs. aSGA
35.1 vs. 27.9
1.4 vs. 4.4
8.2 vs. 10.2
2.7 vs. 3.7
26.8 vs. 19.1
12.5 vs. 8.2 b
aSGA vs. AGA
68.5 vs. 56.8
sSGA vs. AGA
76.3 vs. 56.8†
sSGA vs. aSGA RDS
Total
aSGA vs. AGA
76.3 vs. 68.5 88.8 vs. 89.4
47.8 vs. 29.4
4.7 vs. 1.3 a
61.7 vs. 29.4 61.7 vs. 47.8
b
81.6 vs. 81.7
23.5 vs. 11.6
a
19.9 vs. 11.6 19.9 vs. 23.5 63.0 vs. 61.9
10.2 vs. 10.2
7.8 vs. 3.7 7.8 vs. 2.7
24.7 vs. 19.1a
b
29.2 vs. 39.6
24.7 vs. 26.8 c
58.3 vs. 63.0
b
sSGA vs. AGA
96.5 vs. 89.4
86.9 vs. 81.7
64.8 vs. 61.9
38.2 vs. 39.6
59.6 vs. 63.0
sSGA vs. aSGA
96.6 vs. 88.8
86.9 vs. 81.6
64.8 vs. 63.0
38.2 vs. 29.2
59.6 vs. 58.8
aSGA vs. AGA
33.3 vs. 30.6
14.1 vs. 17.9
16.9 vs. 8.7c
12.4 vs. 2.6a
17.7 vs. 12.3a
sSGA vs. AGA
22.4 vs. 30.6
32.6 vs. 17.9a
18.0 vs. 8.7a
8.6 vs. 2.6
16.8 vs. 12.3a
22.4 vs. 33.3
c
18.0 vs. 16.9
8.6 vs. 12.4
16.8 vs. 17.7
sSGA vs. aSGA
32.6 vs. 14.1
Abbreviations: aSGA, asymmetric small for gestational age; sSGA, symmetric small for gestational age; AGA, appropriate for gestational age; RDS, respiratory distress syndrome; NEC, necrotizing enterocolitis; CLD, chronic lung disease; ROP, retinopathy of prematurity. Note: Chi-square or Fisher exact test were used as appropriate. a p < 0.001. b p < 0.05. c p < 0.01. d Considered as a major morbidity.
RDS, NEC, IVH, and nosocomial infection. AGA infants had relatively lower hospital resource utilization followed by aSGA group while sSGA infants had the longest duration of hospital stay and use of oxygen and CPAP (►Table 4).
Comparison with Literature
Discussion Main Results This is the first study reporting the relative impact of severity of growth restriction on neonatal outcomes among premature neonates. We report that although SGA preterm infants < 32 weeks GA are at an increased risk of adverse in-hospital outcomes than their appropriately grown counterparts, within the SGA group, aSGA group is associated with lower odds of mortality and respiratory complications in comparison to sSGA group. In addition, among SGA infants < 28 weeks, American Journal of Perinatology
the majority of adverse odds for in-hospital outcomes were restricted to the sSGA group. Finally, we report that the length of hospital stay and need for respiratory support was significantly longer for sSGA infants than aSGA infants.
Vol. 32
No. 8/2015
Although several studies have now established the increased burden of mortality and morbidities in SGA preterm infants, they included all the SGA infants as one cohort based on the definition of BW < 10th percentile.2–10 Though this approach is useful, it may not tell the whole story given the differences in causes and types of SGA. In the present study, we attempted to further explore this population by subdividing the SGA group into symmetrical and asymmetrical based on simultaneous presence or absence of significant restriction in brain growth, respectively, and found important
Downloaded by: University of Pittsburgh. Copyrighted material.
728
0.68 (0.53, 0.89)
Nosocomial infection
All infants
0.54 (0.43, 0.68)
1.14 (0.95, 1.37)
0.41 (0.32, 0.50)
0.40 (0.26, 0.63)
0.73 (0.49, 1.08)
0.90 (0.66, 1.24)
0.27 (0.19, 0.39)
0.40 (0.33, 0.49)
AGA vs. sSGA
0.79 (0.57, 1.11)
0.72 (0.55, 0.95)
0.61 (0.44, 0.84)
0.56 (0.30, 1.05)
0.97 (0.56, 1.67)
0.93 (0.59, 1.45)
0.45 (0.27, 0.75)
0.65 (0.49, 0.88)
aSGA vs. sSGA
1.08 (0.76, 1.54)
0.89 (0.55, 1.46)
0.62 (0.44, 0.88)
0.74 (0.45, 1.21)
0.94 (0.54, 1.65)
1.12 (0.71, 1.77)
0.77 (0.49, 1.22)
0.73(0.52, 1.02)
AGA vs. aSGA
0.74 (0.53, 1.04)
0.62 (0.36,1.08)
0.29 (0.20, 0.43)
0.45 (0.27, 0.75)
0.78 (0.44, 1.37)
1.43 (0.86, 2.37)
0.39 (0.25, 0.60)
0.31(0.21, 0.45)
AGA vs. sSGA
Infants < 28 wk GA
0.68 (0.43, 1.10)
0.70 (0.34, 1.43)
0.47 (0.29, 0.77)
0.61 (0.31, 1.22)
0.83 (0.38, 1.79)
1.27 (0.66, 2.43)
0.50 (0.27, 0.92)
0.43(0.26, 0.70)
aSGA vs. sSGA
29 (13–54)
Length of hospital stay
35 (16 63)
g
39 (20 64)
1 (0–7)
2 (0–12)g
1 (0–12)
sSGA
h
< 0.0001
0.18
0.024
0.44
p-Valueb
0.15 (0.042)
0.32 (0.038)f
0.57 (0.06)f f
0.35 (0.07)f
0.70 (0.078)f
Estimate (se)
AGA vs. sSGA
0.54 (0.054)f
d
0.14 (0.061)e
0.20 (0.09)e
Estimate (se)
c
AGA vs. aSGA
Multivariable analyses a
0.12 (0.052)e
0.14 (0.10)
0.33 (0.089)f
0.42 (0.12)f
Estimate (se)
aSGA vs. sSGA
American Journal of Perinatology
Vol. 32
Hei et al.
No. 8/2015
Downloaded by: University of Pittsburgh. Copyrighted material.
Linear model was used for length of hospital in log scale and zero-inflated negative binomial models were used for other outcomes. The covariates included were gender, gestational age, Apgar score at 5 min, delivery type, admission status, maternal hypertension, smoking status, and steroid use. b Based on the comparison among three groups using the Mood median test. Comparisons between groups were further conducted when p < 0.05. c Estimated coefficients. d Standard error. The reported estimates were based on the final model derived by back forward selection procedures. e p < 0.05. f p < 0.001. g The medians (p < 0.01) indicate significant difference between the two groups. h The medians p < 0.05) indicate significant difference between the two groups.
a
1 (0–5)
Duration of ventilation
2 (0–11)
3 (0–12)g,h
2 (0–11)g
Duration of CPAP g,h
1 (0–12)
aSGA
1 (0–8)
AGA
Median test (IQ range)
Duration of oxygen
Use of resources
Table 4 Hospital resource utilization
Abbreviations: AGA, appropriate for gestational age; aSGA, asymmetric small for gestational age; CI, confidence interval; CLD, chronic lung disease; GA, gestational age; NEC, necrotizing enterocolitis; NICU, neonatal intensive care unit; OR, odds ratio; RDS, respiratory distress syndrome; ROP, retinopathy of prematurity; sSGA, symmetric small for gestational age. Notes: Multiple logistic regression models were used. Covariates included were gender, GA, Apgar score at 5 min, delivery mode, admission status, maternal hypertension, smoking status, and steroid use. The ORs were estimated based on the final model derived by stepwise procedure. a Considered as a major morbidity.
1.58 (1.27, 1.97)
0.67 (0.51, 0.88)
0.72 (0.45, 1.15)
0.76 (0.50, 1.14)
0.97 (0.68, 1.38)
RDS
CLD
a
Stage 3 ROP
a
Stage 2a NECa
Severe neurological injury
0.60 (0.39, 0.92)
Mortality
a
0.61 (0.49, 0.78)
AGA vs. aSGA
Mortality or major morbidity
Outcomes
Table 3 Comparison of outcomes: adjusted OR (95% CI) after multivariable analyses
Outcomes for sSGA vs. aSGA Preterm Infants in Canada 729
Outcomes for sSGA vs. aSGA Preterm Infants in Canada differences. This has not been reported previously. Although not directly comparable to our study population, similar associations between the proportionality of growth restriction and adverse antenatal or perinatal outcomes as well as respiratory health during 1st year of life have been previously reported in term SGA infants.17–21 The biological mechanisms responsible for relative increase in mortality and morbidities among SGA preterm infants are not well defined. Most experimental studies utilizing different animal models of in utero placental insufficiency have focused on the effect on lung maturation and surfactant production and have shown conflicting results. One explanation for these observed differences could be the heterogeneity in the experimental methods used in these studies. Experiments which used models of relatively acute fetal hypoxemia and at a relatively higher GA have shown an increase in fetal cortisol levels and an associated rise in surfactant production.22 On the other hand, a more recent study using a sheep model of chronic fetal hypoxia (created by preconception removal of endometrial caruncles from the uterus, which results in the formation of smaller placenta and growth-retarded fetuses) has demonstrated worse lung maturity and reduced surfactant production.23 On similar lines, one of the potential explanations for the observed differences in respiratory outcomes between sSGA and aSGA infants in our study could be the differences in the severity or duration of fetal hypoxia. It’s plausible that aSGA infants suffered from a more acute but relatively late onset insult while sSGA infants faced a more chronic placental insufficiency. Unfortunately, due to lack of data we are unable to confirm the antenatal timing of onset of growth restriction to test this hypothesis. However, the observation that relatively more aSGA infants had hypertensive mothers and were delivered at a lower GA along with the fact that the majority of babies in both the SGA groups were delivered by cesarean section points toward higher severity of fetal insult in the aSGA group necessitating emergency delivery. Further, experimental studies investigating the effect of intrauterine growth retardation (IUGR) secondary to placental insufficiency on the postnatal pulmonary function have shown that several weeks after birth IUGR animals continue to demonstrate impaired pulmonary function, abnormal airway structure, as well as worse lung and airway compliance.24,25 These effects along with impaired lung function at the time of birth may explain our findings of an increased incidence of respiratory complications and a longer duration of respiratory support for SGA infants compared with AGA and for sSGA infants compared with aSGA. We also observed an increase in mortality among sSGA infants. Unfortunately, our database does not collect information regarding specific cause of death. The fact that other major “life-threatening” morbidities (IVH, NEC, and nosocomial infections) were similar between the two SGA groups suggests that the increased mortality observed in our study was likely related to the respiratory complications. Others have described the patterns of mortality among SGA preterm infants and have also reported higher mortality with most deaths occurring in the first 4 weeks of life and cardiorespiratory failure being reported as the most common American Journal of Perinatology
Vol. 32
No. 8/2015
Hei et al. cause of death.26,27 Our study suggests that preterm SGA is not a homogenous group and the prognosis may vary depending on the antenatal severity and duration of growth restriction in spite of similar BWs. We hypothesize that infants who had relatively preserved brain growth are likely to have been exposed to less severe or shorter insults and hence may have better outcomes than constitutionally small preterm SGA infants. Ponderal index (PI) is the most common previously described parameter used to subcategorize growth restriction in term SGA infants.17–21 It is calculated from infants weight and crown–heel length (CHL) (PI ¼ [weight {grams}/[CHL {cm}]3 100). The resultant value is then plotted on a reference curve and SGA infants with PI < 10th percentile is classified as asymmetric while others are considered as symmetric. PI has not been validated for use in extreme preterm infants. In addition, studies using PI for term SGA infants has provided contrasting results with some reporting worse perinatal outcomes in infants classified as “asymmetric” while other suggesting “symmetric” SGA infants to be worse. In addition, the majority of these studies is older than 20 years and were conducted in resource-limited settings, where incidence of SGA in term infants is higher and the etiology for SGA may be different (placental dysfunction vs. maternal nutritional status). Another major limitation was that in most of these studies GA was determined based on postnatal examination. This could have resulted in many well late preterm infants being misclassified as sSGA. A study evaluating the use of PI or CHL < 10th percentile (i.e., sSGA) reported that the SGA term infants had worse perinatal outcomes irrespective of the definition used for classification.15 The rationale of using these indices was to account for the timing of in utero event as fetal compromise during the second trimester is likely to affect both the weight and length (symmetrical) while third trimester insult will affect the weight more than the length (asymmetrical). In this study, we choose to subdivide the SGA population based on HC < 10th percentile. The use of HC as a classifying parameter has an advantage over the length as during the period of growth restriction “diving reflex” would initiate the preservation of blood flow to the brain. In addition, measurement of length at birth may have significant methodological issues and high interobserver variability.
Study Limitations In addition to the well-known limitations of retrospective database-driven studies, there are several other limitations of our study. First, although the majority of SGA is expected to be a result of placental dysfunction we are unable to confirm this as specific data on the etiology were not available. This is even more important for the sSGA group as many factors, such as congenital infections or genetic abnormalities, can also result in symmetrical growth restriction. Even though we excluded all such known cases, it is possible that some of the SGA infants included in this study had an undiagnosed underlying genetic etiology. However, the high percentage of hypertensive mothers in both groups suggests placental dysfunction to be the primary etiology. Second, due to lack of data we were
Downloaded by: University of Pittsburgh. Copyrighted material.
730
Outcomes for sSGA vs. aSGA Preterm Infants in Canada
2 Bardin C, Zelkowitz P, Papageorgiou A. Outcome of small-for-
3
4
5
6
7
8
Conclusion In conclusion, among extreme preterm SGA infants, significant differences exist in the neonatal outcomes as well as hospital resource utilization depending upon the presence or absence of relative “head sparing.” This study highlights the importance of a more comprehensive anthropometric assessment of growth restriction while evaluating and counseling families of growth-restricted infants. These results may help in targeting interventions and counseling families. Future studies should investigate the impact of symmetry of growth restriction on long-term neurodevelopmental outcomes.
Acknowledgments The authors would like to thank the Canadian Institutes of Health Research team in Maternal-Infant Care (MiCare) for providing organizational support to the Canadian Neonatal Network. Authors would also like to acknowledge the contribution of Mr. Xiang Y. Ye, Statistician at the MiCare Research Center, Mount Sinai Hospital, Toronto, Canada, for his assistance with the analysis for this project. MiCare Research Centre at Mount Sinai Hospital, Toronto, Ontario, is financially supported by Ontario Ministry of Health and Long-Term Care. Dr. Mingyan Hei’s fellowship training in Neonatal Intensive Care Unit of Mount Sinai Hospital was supported by the policy of “125-Distinguished Clinicians of the Third Xiangya Hospital of Central South University, China.”
9
10
11
12 13
14
15
16
17
18
19
20
References 1 Zaw W, Gagnon R, da Silva O. The risks of adverse neonatal
outcome among preterm small for gestational age infants according to neonatal versus fetal growth standards. Pediatrics 2003; 111(6 Pt 1):1273–1277
731
21
gestational age and appropriate-for-gestational age infants born before 27 weeks of gestation. Pediatrics 1997;100(2):E4 Bernstein IM, Horbar JD, Badger GJ, Ohlsson A, Golan A; The Vermont Oxford Network. Morbidity and mortality among verylow-birth-weight neonates with intrauterine growth restriction. Am J Obstet Gynecol 2000;182(1 Pt 1):198–206 Garite TJ, Clark R, Thorp JA. Intrauterine growth restriction increases morbidity and mortality among premature neonates. Am J Obstet Gynecol 2004;191(2):481–487 Lal MK, Manktelow BN, Draper ES, Field DJ; Population-based study. Chronic lung disease of prematurity and intrauterine growth retardation: a population-based study. Pediatrics 2003; 111(3):483–487 Piper JM, Xenakis EM, McFarland M, Elliott BD, Berkus MD, Langer O. Do growth-retarded premature infants have different rates of perinatal morbidity and mortality than appropriately grown premature infants? Obstet Gynecol 1996;87(2):169–174 Regev RH, Lusky A, Dolfin T, Litmanovitz I, Arnon S, Reichman B; Israel Neonatal Network. Excess mortality and morbidity among small-for-gestational-age premature infants: a population-based study. J Pediatr 2003;143(2):186–191 Reiss I, Landmann E, Heckmann M, Misselwitz B, Gortner L. Increased risk of bronchopulmonary dysplasia and increased mortality in very preterm infants being small for gestational age. Arch Gynecol Obstet 2003;269(1):40–44 Simchen MJ, Beiner ME, Strauss-Liviathan N, et al. Neonatal outcome in growth-restricted versus appropriately grown preterm infants. Am J Perinatol 2000;17(4):187–192 Kok JH, den Ouden AL, Verloove-Vanhorick SP, Brand R. Outcome of very preterm small for gestational age infants: the first nine years of life. Br J Obstet Gynaecol 1998;105(2):162–168 Qiu X, Lodha A, Shah PS, et al; Canadian Neonatal Network. Neonatal outcomes of small for gestational age preterm infants in Canada. Am J Perinatol 2012;29(2):87–94 Lockwood CJ, Weiner S. Assessment of fetal growth. Clin Perinatol 1986;13(1):3–35 Kramer MS, Platt RW, Wen SW, et al; Fetal/Infant Health Study Group of the Canadian Perinatal Surveillance System. A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics 2001;108(2):E35 Barbier A, Boivin A, Yoon W, et al; Canadian Neonatal Network. New reference curves for head circumference at birth, by gestational age. Pediatrics 2013;131(4):e1158–e1167 Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978; 92(4):529–534 Bell MJ, Ternberg JL, Feigin RD, et al. Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging. Ann Surg 1978;187(1):1–7 Balcazar H, Haas J. Classification schemes of small-for-gestational age and type of intrauterine growth retardation and its implications to early neonatal mortality. Early Hum Dev 1990;24(3): 219–230 Caulfield LE, Haas JD, Belizán JM, Rasmussen KM, Edmonston B. Differences in early postnatal morbidity risk by pattern of fetal growth in Argentina. Paediatr Perinat Epidemiol 1991;5(3): 263–275 Kramer MS, Olivier M, McLean FH, Willis DM, Usher RH. Impact of intrauterine growth retardation and body proportionality on fetal and neonatal outcome. Pediatrics 1990;86(5):707–713 Vik T, Markestad T, Ahlsten G, et al. Body proportions and early neonatal morbidity in small-for-gestational-age infants of successive births. Acta Obstet Gynecol Scand Suppl 1997;165:76–81 Vik T, Vatten L, Markestad T, Ahlsten G, Jacobsen G, Bakketeig LS. Morbidity during the first year of life in small for gestational age infants. Arch Dis Child Fetal Neonatal Ed 1996;75(1):F33–F37 American Journal of Perinatology
Vol. 32
No. 8/2015
Downloaded by: University of Pittsburgh. Copyrighted material.
unable to adjust for the severity of fetal growth restriction (FGR). This may be an important confounder as reported in a study in term infants where the differences in outcome based on proportionality became less significant after adjustments for the severity of FGR.19 Third, we do not have long-term neurodevelopmental outcome data on these infants which would be an important subject for future studies. Fourth, due to lack of data on serial growth parameters, we could not assess the groups for differences in postnatal growth and its influence on neonatal outcomes. Finally, we acknowledge that using an arbitrary cutoff of < 10th percentile will result in misclassification for some infants. For instance, an infant with BW at 9th and HC at the 11th percentile will be classified as asymmetric while an infant with BW at 2nd and HC at 9th will be called symmetric, although the former is more symmetric than the later. Nevertheless, we did find important differences in the outcomes of infants classified using HC < 10th percentile cutoff suggesting that a degree of misclassification may be random.
Hei et al.
Outcomes for sSGA vs. aSGA Preterm Infants in Canada
Hei et al.
22 Gagnon R, Langridge J, Inchley K, Murotsuki J, Possmayer F.
25 Joyce BJ, Louey S, Davey MG, Cock ML, Hooper SB, Harding R.
Changes in surfactant-associated protein mRNA profile in growth-restricted fetal sheep. Am J Physiol 1999;276(3 Pt 1): L459–L465 23 Orgeig S, Crittenden TA, Marchant C, McMillen IC, Morrison JL. Intrauterine growth restriction delays surfactant protein maturation in the sheep fetus. Am J Physiol Lung Cell Mol Physiol 2010; 298(4):L575–L583 24 Wignarajah D, Cock ML, Pinkerton KE, Harding R. Influence of intrauterine growth restriction on airway development in fetal and postnatal sheep. Pediatr Res 2002;51(6):681–688
Compromised respiratory function in postnatal lambs after placental insufficiency and intrauterine growth restriction. Pediatr Res 2001;50(5):641–649 26 Giapros V, Drougia A, Krallis N, Theocharis P, Andronikou S. Morbidity and mortality patterns in small-for-gestational age infants born preterm. J Matern Fetal Neonatal Med 2012;25(2):153–157 27 Bartels DB, Kreienbrock L, Dammann O, Wenzlaff P, Poets CF. Population based study on the outcome of small for gestational age newborns. Arch Dis Child Fetal Neonatal Ed 2005;90(1): F53–F59
Downloaded by: University of Pittsburgh. Copyrighted material.
732
American Journal of Perinatology
Vol. 32
No. 8/2015
