Sie sind auf Seite 1von 5

Journal of Perinatology (2014) 34, 669–672

© 2014 Nature America, Inc. All rights reserved 0743-8346/14


www.nature.com/jp

ORIGINAL ARTICLE
Impact of duration of rupture of membranes on outcomes of
premature infants
MW Walker1, AH Picklesimer3, RH Clark1,2, AR Spitzer2 and TJ Garite2

OBJECTIVE: The primary aim of the study was to determine how the risk of adverse outcomes was related to the duration of
the latency period and gestational age at birth following preterm premature rupture of the fetal membranes (PPROM).
STUDY DESIGN: Retrospective review of infants discharged from 330 neonatal intensive care units. We defined four subgroups
based on gestational age: 23 to 25, 26 to 28, 29 to 31 and 32 to 34 weeks. Each gestational age group was evaluated by duration of
ROM: o24 h, 1 to 7 days, 8 to 14 days, 15 to 21 days, 21 to 28 days and >28 days and compared with a referent group (PPROM of
>24 h but o 7 days).
RESULT: In all, 239 808 non-anomalous infants 23 to 34 weeks’ gestational age were identified; 37 233 (15.5%) had rupture of
membranes (ROM) >24 h. Compared with a reference group (PPROM of >24 h but o 7 days), the risk of mortality for PPROM
of 8 to 14, 15 to 21 and 21 to 28 days varied depending on gestational age at birth. Only PPROM >28 days was consistently
associated with increased mortality and decreased likelihood of survival without morbidity in all gestational age subgroups.
CONCLUSION: PPROM for >28 days is associated with an increased risk of death and morbidity.
Journal of Perinatology (2014) 34, 669–672; doi:10.1038/jp.2014.73; published online 24 April 2014

INTRODUCTION care unit (NICU)-associated morbidities in infants admitted to


Preterm premature rupture of the fetal membranes (PPROM) US neonatal intensive care units who were born to mothers with
before 37 weeks’ gestational age is a relatively common occu- varying durations of PPROM. Our context is neonatal and we
rrence during pregnancy, complicating 2 to 4% of all singleton evaluated outcomes based on gestational age at birth rather than
pregnancies and 7 to 20% of twin gestations in the United at gestational age at rupture of membranes. Our results inform
States.1–5 PPROM occurring at earlier gestational ages is typically neonatal counseling based on gestational age and duration of
associated with worse fetal and neonatal outcomes because of the rupture of membranes. We confined our cohort to the highest risk
increased risk of subsequent chorioamnionitis and/or premature group of infants admitted to the NICU, namely infants ⩽ 34 weeks’
delivery. Neonatal survival rates in pre- and periviable PPROM gestational age at birth. The primary aim of the study was to
(o 24 weeks gestational age) have been reported to range from determine how the risk of adverse outcomes was related to the
as low as 24% to as high as 70%.6–10 Most of these studies were duration of the latency period and gestational age at birth.
limited to examinations of small populations of infants in single
institutional settings.
The interval between PPROM and delivery is known as latency.
If PPROM occurs before 32 to 34 weeks’ gestational age, efforts METHODS
can be made to prolong latency as long as there is no evidence of The study comprised a descriptive review of infants from a network of 330
advanced labor, intrauterine infection or significant vaginal NICUs in the United States that reported to the Pediatrix Clinical Data
bleeding, and the fetal status is reassuring. Treatment with Warehouse. Health-care professionals providing care to these infants used
a proprietary software system (BabySteps, MEDNAX, Sunrise, FL) to
broad-spectrum antibiotics can prolong latency after PPROM at generate clinical admission, daily progress and discharge notes. These
these earlier gestational ages, allowing time for administration of data were stored in a consolidated national data set and de-identified for
corticosteroids while reducing morbidities associated with both research purposes. The data set had no protected health information
maternal and fetal infections.11 The benefits of prolonging latency and was compliant with the Health Insurance Portability and Account-
include not only the opportunity to administer corticosteroids but ability Act of 1996 regulations. The Greenville Health System, Greenville,
also minimizing complications of prematurity by achieving a more South Carolina, and the Western Institutional Review Board approved this
advanced gestational age at the time of delivery. The risks of pro- research.
longing latency include, however, the development of subclinical or Duration of rupture of membranes for all infants is reported as a
overt chorioamnionitis, leading to both short- and long-term maternal demographic profile feature based on the review of the maternal
records at the time of delivery. Data on gestational age represented the
maternal and neonatal morbidities.11–17 The impact of duration of
best estimates from both obstetrical data and neonatal examination
the latency period on neonatal outcomes is not clear, particularly for findings. We examined all recorded comorbidities, adverse effects,
pregnancies experiencing PPROM at the limits of viability.18,19 concomitant treatments, other procedures and respiratory support for
The aim of this study was to examine a large multicenter, each neonate’s hospital course. We evaluated the use of vasopressors
multiyear database to provide a more accurate risk assessment of (dopamine, dobutamine and/or epinephrine within the first 3 days after
outcomes, including mortality and significant neonatal intensive birth) as a surrogate marker of significant postnatal illness.

1
Division of Neonatology, Children’s Hospital, Greenville Health System, Greenville, SC, USA; 2Departments of Education, Research and Quality Improvement, Pediatrix-Obstetrix
Center for Research and Education, Sunrise, FL, USA and 3Department of Maternal-Fetal Medicine, Greenville Health System, Greenville, SC, USA. Correspondence: Dr MW Walker,
Division of Neonatology, Children’s Hospital, Greenville Health System, Greenville, SC, USA.
E-mail: whit_walker@pediatrix.com
Received 2 December 2013; revised 7 March 2014; accepted 19 March 2014; published online 24 April 2014
Prolonged rupture of the membranes and outcome
MW Walker et al
670
Data selection mature (32 vs 33 weeks), minimally smaller (1.78 vs 1.79 kg), more
We included inborn infants with an estimated gestational age at birth often exposed to antenatal steroids (81% vs 62%), more often a
between 23 and 34 completed weeks that were discharged from the singleton birth (87% vs 64%) and less often delivered by cesarean
hospital of birth between 1 January 1997 and 1 July 2012. We excluded section (43% vs 67%). Overall mortality was decreased and survival
infants transferred in from another hospital (out born), infants transferred without morbidity was increased in infants whose membranes
out of the birth hospital and infants with major anomalies. were ruptured for >24 h but o 7 days compared with infants
whose membranes were ruptured o 24 h. The absolute differ-
Data analysis ences were small: mortality—3.5% vs 3.9%, AOR = 0.8, confidence
Comparisons of continuous variables (e.g., gestational age) were evaluated interval (CI) = 0.7 to 0.9; P o0.001; survival without morbidity—-
with analysis of variance. Categorical variables (e.g., race and sex) were 86% vs 85%, AOR = 1.13, CI = 1.07 to 1.19; P o0.01. Similar trends
evaluated with two-tailed χ2 tests or Fisher's exact test (2 by 2 tables). were seen within each gestational age group. The demographic
Nonparametric data were assessed with Kruskal–Wallis analysis of variance. and maternal treatment characteristics of the patients with
The linear trend test and the Cochran–Armitage trend test were performed rupture of membranes o 24 h were so distinctive that we did
to evaluate time-related changes. not use them as our reference group.
Multivariable logistic regression was used to calculate the adjusted odds Compared with the infants with PPROM 1 to 7 days, infants with
ratios (AOR) and 95% confidence intervals (CIs) for the variables associated
PPROM >7days were smaller (1.68 vs 1.82 kg), more often exposed
with an increased risk of death or morbidity. Because mortality is a
competing variable with morbidities (i.e., if you die you cannot develop a to antenatal steroids (89% vs 77%) and less often delivered by
late morbidity), ORs for each individual morbidity were not calculated, but cesarean section (40% vs 49%). Gestational age (32 vs 32) and
instead a variable was created for survival without major morbidity (stage 3 singleton birth (89% vs 86%) were similar for infants with PPROM
or 4 retinopathy of prematurity, grade 3 or 4 intraventricular hemorrhage, >7 days and those with PPROM 1 to 7 days. Within each
necrotizing enterocolitis that required medical or surgical treatment, gestational age group the demographic and maternal treatment
oxygen use at 36 weeks postmenstrual age for infants born at 32 weeks or characteristics of the patients with PPROM 1 to 7 days were similar
less and at 28 days for infants >32 weeks) and the AORs for survival and the differences were small compared with the other PPROM
without morbidity were determined. The variables in the model for groups (Supplementary Tables). For this reason, we used the
calculating the AORs included estimated gestational age, birth weight,
PPROM 1 to 7 days as our reference group.
gender, multiples/singleton fetal status, reported exposure to antenatal
steroids, year of discharge and duration of latency in each PPROM group.
For the data in the Supplementary Tables, the logistic model evaluating Medications and support. Compared with the infants with PPROM
ORs for death and survival without morbidity in each gestational age 1 to 7 days, and when corrected for gestational age, birth weight,
group was considered an independent sample population.
gender, multiples, antenatal steroids and discharge year using
In our original design, we used two reference groups for calculation of
the odds of death and the odds of survival without morbidity: PPROM logistic modeling, infants with PPROM >7 days were more often
o24 h and the PPROM 1 to 7 days group (Supplementary Tables 1–4). The treated with vasopressors (11% vs 6%, AOR = 1.8, 95%tile CI
reason for using two reference groups was that infants born to mothers = 1.7 to 2.0), more often required ventilator support (37% vs
with ruptured membranes o24 h were significantly different from infants 24%, AOR = 1.7, 95%tile CI = 1.6 to 1.8) and were more often
with rupture of membranes 24 h or more. Premature infants born within treated with surfactants (35% vs 23%, AOR = 1.6, 95%tile CI
24 h of rupture of membranes may represent a group that was delivered = 1.5 to 1.7). The Supplementary Tables show that the treatment
for fetal distress or emergent medical indications like pre-eclampsia. differences are most pronounced in infants in the more mature
However, our database cannot always define the exact rationale behind gestational age groups (>26 weeks) and in the infants exposed to
each delivery. Because of this potential selection bias, we have not
the rupture of membranes >28 days.
included the data on ORs for this reference group compared with the other
PPROM groups.
Mortality and survival without morbidity. Compared with the
infants with PPROM 1 to 7 days, and when corrected for
RESULTS gestational age, birth weight, gender, multiples, antenatal steroids
Study population and discharge year, infants with more prolonged rupture of
Between January 1997 and July 2012, 239 808 non-anomalous, membranes were more likely to die: 15 to 21 days (4.7% vs
inborn, infants between 23 and 34 weeks’ gestational age were 3.0%, AOR = 1.5, CI = 1.2 to 1.8); 22 to 28 days (5.5% vs
reported to the Clinical Data Warehouse; 37 233 (15.5%) were 3.0%, AOR = 1.6, CI = 1.2 to 2.2) and >28 days (12.6% vs 3.0%,
reported to have rupture of membranes >24 h before delivery. To AOR = 5.7, CI = 4.7 to 6.9). There was no significant difference in
understand the impact of duration of ruptured membranes within mortality for infants with PPROM 1 to 7 days compared with those
specific estimated gestational ages, the infant populations were with PPROM 8 to 14 days (3.8% vs 3.0%). The largest mortality
broken down into four subsets based on gestational age at birth: difference was observed when comparing infants with PPROM of
23 to 25 weeks (n = 12 595; Supplementary Table 1), 26 to 28 weeks 1 to 7 days with those with PPROM >28 days and this was true
(n = 24 459; Supplementary Table 2), 29 to 31 weeks (n = 4949; across all gestational age groups (Supplementary Tables and
Supplementary Table 3) and 32 to 34 weeks (n = 153 263; Figure 1)
Supplementary Table 4). Compared with the infants with PPROM 1 to 7 days and when
Each subset of premature infants within their respective gesta- corrected for gestational age, birth weight, gender, multiples,
tional age groups was evaluated by duration of rupture of mem- antenatal steroids and discharge year, infants with more
branes: o24 h (n = 202 585, 84.5%), 1 to 7 days (n = 25 194, 10.5%), prolonged rupture of membranes were less likely to survive
8 to 14 days (n = 6507, 2.7%), 15 to 21 days (n = 2501, 1.0%), 21 to without morbidity: 15 to 21 days (83% vs 88%, AOR = 0.9,
28 days (n = 1067, 0.4%) and >28 days (n = 1954, 0.8%). In multi- CI = 0.7 to 1, P = 0.03); 22 to 28 days (78% vs 88%, AOR = 0.6,
variate analysis, the 1 to 7 days group was used as the reference CI = 0.5 to 0.7, P o 0.001) and >28 days (71% vs 88%, AOR = 0.3,
group to which the other PPROM groups were compared. CI = 0.3 to 0.4, Po 0.001). There was no significant difference in
survival without morbidity for infants with PPROM 1 to 7 days
compared with those with PPROM 8 to 14 days (85% vs 88%). The
Comparison of infants by gestational age and duration of rupture largest difference was observed when comparing infants with
of membranes PPROM of 1 to 7 days with those with PPROM >28 days and this
Demographics. Compared with infants with rupture of mem- was true across all gestational age groups (Supplementary Tables
branes o24 h, infants with PROM (⩾24 h) were slightly less and Figure 1)

Journal of Perinatology (2014), 669 – 672 © 2014 Nature America, Inc.


Prolonged rupture of the membranes and outcome
MW Walker et al
671
17
16 PPROM Groups; Reference Group is PPROM >1 and <=7 days

Adjusted Odds Ratio For Mortality


15 8 to14 days 15 to 21 days 22 to 28 days > 28 days
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
23 to 25
26 to 28
29 to 31
32 to 34
All Patients

23 to 25
26 to 28
29 to 31
32 to 34
All Patients

23 to 25
26 to 28
29 to 31
32 to 34
All Patients

23 to 25
26 to 28
29 to 31
32 to 34
All Patients
EGA Group

2
PPROM Groups; Reference Group is PPROM >1 and <=7 days
Adjusted Odds Ratio For Survival

1.8
8 to14 days 15 to 21 days 22 to 28 days > 28 days
1.6
without Major Morbidity

1.4
1.2
1
0.8
0.6
0.4
0.2
0
23 to 25
26 to 28
29 to 31
32 to 34
All Patients

23 to 25
26 to 28
29 to 31
32 to 34
All Patients

23 to 25
26 to 28
29 to 31
32 to 34
All Patients

23 to 25
26 to 28
29 to 31
32 to 34
All Patients

EGA Group
Figure 1. Impact of PPROM on the odds of mortality (a) and survival without morbidity (b) within each gestational age group. Within each
gestational age group, the impact of each PPROM group on the outcome is compared with the reference group PPROM 1 to 7 days. (a) AOR
for mortality by duration of rupture of membranes within each gestational age group. (b) AOR for survival without morbidity (stage 3 or 4
retinopathy of prematurity, grade 3 or 4 intraventricular hemorrhage, necrotizing enterocolitis that required medical or surgical treatment,
oxygen use at 36 weeks postmenstrual age for infants born at 32 weeks or less and at 28 days for infants >32 weeks) by duration of rupture of
membranes within each gestational age group. In the logistic model, each gestational age group was considered an independent sample
population. The variables in the model for calculating the AORs for each outcome included gestational age, birth weight, gender, multiples,
antenatal steroids, discharge year and duration of PPROM group.

DISCUSSION Our findings support the existing literature and suggest that the
In premature infants admitted to our NICUs over the past 15 years, ability to prolong latency for at least 1 to 7 days will allow time for
approximately 15% had prolonged rupture of the membranes, but treatment of the fetus with antenatal steroids and antibiotics (or
only 33% of these infants had rupture of the membranes for longer other additional therapies) and will result in improveds outcomes
than 1 week (3.1% of the total sample). Within all gestational groups, (especially crucial in pre- and periviable cases of PPROM).11,20 Most
the best outcomes were in the patients with PPROM of 1 to 7 days. importantly, the risk of bronchopulmonary dysplasia, vertically
Latency of >24 h and less than a week appeared to provide the transmitted neonatal infection and hospital length of stay did not
greatest benefit in the most immature GA group (Supplementary appear to increase as PPROM duration extended to 2 weeks. Our
Tables). As prolonged rupture of the membranes continued beyond data did not allow us to determine if the improved outcomes of
21 days, however, these benefits began to disappear and mortality these infants is due to a treatment effect or a difference in
and death or morbidity began to increase significantly. In all maternal health and reason for delivery. There likely is some 'self-
gestational age groups, survival and survival without mortality was selection' of pregnancies that are able to progress to greater
lowest in those infants where PPROM was >28 days. gestational age that partially explains the differences in outcomes.

© 2014 Nature America, Inc. Journal of Perinatology (2014), 669 – 672


Prolonged rupture of the membranes and outcome
MW Walker et al
672
Our findings of equivalent or better morbidity and mortality REFERENCES
outcomes in the majority of our premature infants with relatively 1 Ananth CV, Vintzileos AM. Epidemiology of preterm birth and its clinical subtypes.
long latency periods (7 to 21 days) following PPROM conflicts with J Matern-Fetal Neonatal Med 2006; 19: 773–782.
some previous reports of increased adverse short- and long-term 2 Ananth CV, Vintzileos AM. Maternal–fetal conditions necessitating a medical
outcomes in these infants.18,21–24 While there have been reports of intervention resulting in preterm birth. Am J Obstet Gynecol 2006; 195:
increased risk of adverse developmental outcomes in infants born 1557–1563.
3 Caughey AB, Robinson JN, Norwitz ER. Contemporary diagnosis and management
following PPROM, the largest study of infants with long-term
of preterm premature rupture of membranes. Rev Obstet Gynecol 2008; 1: 11–22.
follow-up (5 years) revealed minimally increased risk of adverse 4 Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of
long-term developmental outcome compared with control infants. preterm birth. Lancet 2008; 371: 75–84.
Mura et al.25 concluded that infants born after PPROM are not at 5 Pakrashi T, Defranco EA. The relative proportion of preterm births complicated by
greater risk for cognitive impairment compared with infants born premature rupture of membranes in multifetal gestations: a population-
preterm owing to another cause. Their conclusion was supported based study. Am J Perinatol 2013; 30: 69–74.
by the finding of an increased risk of cognitive impairment in 6 Williams O, Hutchings G, Debieve F, Debauche C. Contemporary neonatal out-
infants born after PPROM with a short latency period (o 3 days) come following rupture of membranes prior to 25 weeks with prolonged oligo-
compared with those with a longer latency period. Our results hydramnios. Early Hum Dev 2009; 85: 273–277.
7 Waters TP, Mercer BM. The management of preterm premature rupture of the
support the suggestion made by Mura et al.25 that decisions
membranes near the limit of fetal viability. Am J Obstet Gynecol 2009; 201:
concerning early induction of labor or cesarean delivery in women 230–240.
with preterm rupture of membranes are complex and need to 8 Taylor J, Garite TJ. Premature rupture of membranes before fetal viability. Obstet
consider gestational age, fetal well-being and maternal health as Gynecol 1984; 64: 615–620.
primary indicators for delivery. Neither our study nor that reported 9 Everest NJ, Jacobs SE, Davis PG, Begg L, Rogerson S. Outcomes following pro-
by Mura et al.25 can easily inform how long is too long, but our longed preterm premature rupture of the membranes. Arch Dis Child Fetal Neo-
data suggest that >28 days in all gestational age groups is natal Ed 2008; 93: F207–F211.
associated with poor outcomes. 10 Soylu H, Jefferies A, Diambomba Y, Windrim R, Shah PS. Rupture of membranes
The primary strength of this study is the large sample size. before the age of viability and birth after the age of viability: comparison of
outcomes in a matched cohort study. J Perinatol 2010; 30: 645–649.
However, the problems associated with this retrospective case-
11 Kenyon S, Boulvain M, Neilson JP. Antibiotics for preterm rupture of membranes.
series design include the inability to account for site variability (e. Cochrane Database Syst Rev 2010; 8: CD001058.
g., delivery room management choices at the limits of viability, 12 Kuypers E, Ophelders D, Jellema RK, Kunzmann S, Gavilanes AW, Kramer BW.
postnatal life support strategies, etc), small patient populations at White matter injury following fetal inflammatory response syndrome induced by
the extremes of viability and/or very prolonged premature rupture chorioamnionitis and fetal sepsis: lessons from experimental ovine models. Early
of the membranes and inability to capture non-NICU-admitted Hum Dev 2012; 88: 931–936.
fetuses or newborns who may have died of stillbirth, miscarriage or 13 Ahn HM, Park EA, Cho SJ, Kim YJ, Park HS. The association of histological chor-
death in the delivery room. We also chose not to report outcomes ioamnionitis and antenatal steroids on neonatal outcome in preterm infants born
of infants born at >34 weeks’ EGA because of their much smaller at less than thirty-four weeks' gestation. Neonatology 2012; 102: 259–264.
14 Bersani I, Thomas W, Speer CP. Chorioamnionitis—the good or the evil for neo-
risk of morbidity and mortality. Maternal charts were also
natal outcome? J Matern-Fetal Neonatal Med 2012; 25(Suppl 1): 12–16.
abstracted by pediatric health-care providers in the course of 15 Horvath B, Grasselly M, Bodecs T, Boncz I, Bodis J. Histological chorioamnionitis is
routine care of the infants admitted to the NICU and not by associated with cerebral palsy in preterm neonates. Eur J Obstet Gynecol Reprod
trained data abstractors. It is possible that some details of the Biol 2012; 163: 160–164.
maternal course, such as maternal steroid treatment, were under- 16 Martinelli P, Sarno L, Maruotti GM, Paludetto R. Chorioamnionitis and prematurity:
reported. In addition, we do not have data on incidence of a critical review. J Matern-Fetal Neonatal Med 2012; 25(Suppl 4): 29–31.
anhydramnios or oligohydramnios, and other maternal/obstetric 17 Moscuzza F, Belcari F, Nardini V, Bartoli A, Domenici C, Cuttano A, et al.
comorbidities that might affect outcomes. Nonetheless, the Correlation between placental histopathology and fetal/neonatal outcome:
observations noted in this group of infants with preterm PPROM chorioamnionitis and funisitis are associated to intraventricular haemorrage and
retinopathy of prematurity in preterm newborns. Gynecol Endocrinol 2011; 27:
appear to be important for determining optimal management
319–323.
strategy and for the counseling of families. 18 Manuck TA, Maclean CC, Silver RM, Varner MW. Preterm premature rupture of
membranes: does the duration of latency influence perinatal outcomes? Am J
Obstet Gynecol 2009; 201: 414–416.
CONCLUSIONS 19 Melamed N, Hadar E, Ben-Haroush A, Kaplan B, Yogev Y. Factors affecting the
This large, retrospective case report series of premature infants duration of the latency period in preterm premature rupture of membranes.
born o 34 weeks estimated gestational age with PPROM J Matern-Fetal Neonatal Med 2009; 22: 1051–1056.
demonstrated that the majority of infants with prolonged latency 20 Waters TP, Mercer B. Preterm PROM: prediction, prevention, principles. Clin Obstet
Gynecol 2011; 54: 307–312.
periods following PPROM do not have an increased morbidity and
21 Melamed N, Ben-Haroush A, Pardo J, Chen R, Hadar E, Hod M, et al. Expectant
mortality risk in the NICU compared with similar gestational age management of preterm premature rupture of membranes: is it all about
infants born to mothers with o24 h of PROM. There was an gestational age? Am J Obstet Gynecol 2011; 204: 48.
increased risk of morbidity and mortality in infants at all 22 Mercer BM, Crouse DT, Goldenberg RL, Miodovnik M, Mapp DC, Meis PJ, et al. The
gestational ages born to mothers with PPROM but only to a antibiotic treatment of PPROM study: systemic maternal and fetal markers and
clinically significant degree with PPROM that extended beyond 28 perinatal outcomes. Am J Obstet Gynecol 2012; 206: 145–149.
days. Nevertheless, optimizing maternal–fetal management with a 23 Verma U, Goharkhay N, Beydoun S. Conservative management of preterm
conservative bent toward prolonging the pregnancy as long as premature rupture of membranes between 18 and 23 weeks of gestation—-
possible seems pragmatic (even in previable PPROM situations). maternal and neonatal outcome. Eur J Obstet Gynecol Reprod Biol 2006; 128:
119–124.
More prospective studies and long-term developmental outcome
24 Nourse CB, Steer PA. Perinatal outcome following conservative management of
analysis are warranted in this high-risk group of patients. mid-trimester pre-labour rupture of the membranes. J Paediatr Child Health 1997;
33: 125–130.
25 Mura T, Picaud JC, Larroque B, Galtier F, Marret S, Roze JC, et al. Cognitive
CONFLICT OF INTEREST impairment at age 5 years in very preterm infants born following premature
The authors declare no conflict of interest. rupture of membranes. J Pediatr 2013; 163: 435–440.

Supplementary Information accompanies the paper on the Journal of Perinatology website (http://www.nature.com/jp)

Journal of Perinatology (2014), 669 – 672 © 2014 Nature America, Inc.


Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.

Das könnte Ihnen auch gefallen