OBJECTIVE: Our aim was to assess the effect of labor on the risk and course of transient tachypnea of the newborn (TTN) in term neonates from a contemporary, population-based cohort.
METHODS: We analyzed perinatal characteristics of term singleton newborns (gestational age [GA] of ≥37 completed weeks) who were born between January 2001 and December 2005 in the federal states of Hesse and Saarland (Germany). TTN was diagnosed on the basis of International Classification of Diseases, 10th Revision codes.
RESULTS: Of a total of 275 459 births, 239 971 fulfilled the inclusion criteria of GA of ≥37 completed weeks and singleton live birth. Among those, 13 346 term infants were admitted for neonatal care and 1423 were diagnosed as having TTN. The overall incidence of TTN was 5.9 cases per 1000 singleton live births in our study cohort. Elective cesarean section, low GA, male gender, and low birth weight were associated with TTN. The duration of oxygen supplementation for newborns with TTN was associated inversely with the duration of labor (r = −0.151; P = .028).
CONCLUSIONS: Our study indicates that TTN is strongly related to elective cesarean section and low GA. Furthermore, the absence of exposure to labor contractions is associated with increased risk and severe course of TTN at term, with longer duration of oxygen supplementation.
WHAT'S KNOWN ON THIS SUBJECT:
ECS is known as a main risk factor for TTN, because of its crucial role in perinatal lung fluid clearance. Our study emphasizes and outlines the impact of labor before birth on respiratory outcomes in TTN.
WHAT THIS STUDY ADDS:
Our study indicates that TTN is strongly related to ECS and low GA. Furthermore, the absence of exposure to labor contractions is associated with increased risk and severe course of TTN at term, with longer duration of oxygen supplementation.
Postnatal respiratory complications among term infants are common. The most commonly reported cause of neonatal respiratory distress is transient tachypnea of the newborn (TTN), with an estimated incidence of 1% to 2% of all newborns.1 The disorder is reported to be benign and self-limiting, with resolution usually occurring by 2 to 5 days of age. In severe courses of TTN, complications such as pneumothorax, need for extracorporeal membrane oxygenation, and death have been reported.2 Known risk factors for TTN include lower gestational age (GA) and birth weight, male gender, and elective cesarean section (ECS).3
In recent decades, the rates of cesarean section (CS), especially those performed electively at term and partly at maternal request, have shown an increasing trend.4 In Germany, according to data supplied by the Federal Office for Quality Assurance, the overall CS rate for all singleton deliveries increased constantly from 21.6% to 29.3% between 2001 and 2007.5 The major clinical reasons for performing a CS have not changed and include breech presentation, suspected fetal distress, failure to progress in labor, and previous CS. The fifth most common indication for a CS has changed and now is reported to be “maternal request.” In the literature, the risks and benefits of ECS are controversial. Some authors favor planned CS in attempts to prevent the fear of childbirth associated with labor and delivery, urinary and fecal incontinence, and pelvic prolapse. CS-related maternal morbidity and mortality rates in the German population have decreased markedly in recent years. However, ECS remains major surgery, with potential important surgical complications for the mother that lead to extended hospital stays.6 The impact of ECS on neonatal respiratory outcomes was summarized recently.2 Epidemiological studies reported that ECS increased the risk of neonatal respiratory morbidity and rates of admission to the NICU, compared with GA-matched newborns after spontaneous vaginal delivery. The aim of this study was to assess whether maternal characteristics, mode of birth, labor before birth, and perinatal factors were associated with the risk and course of TTN among term singleton newborns in a large, representative, cohort study in Germany, which has not yet been evaluated. Furthermore, neonatal outcomes were addressed with respect to the prognosis of respiratory failure in TTN, depending on labor before birth. To the best of our knowledge, no studies have investigated the impact of labor on the severity of TTN courses.
Study Population and Data Sources
We analyzed perinatal databases recorded by the national neonatal audit program of the federal states of Hesse and Saarland in Germany, for singletons born at GA of ≥37 weeks between January 2001 and December 2005. Recorded GAs were based on maternal last menstrual period or early prenatal ultrasound examination findings. The register provides detailed information for >98% of all infants born alive in Hesse and Saarland. The perinatal birth registers include data about the mother, the current pregnancy and delivery, and the postnatal course, as documented by the obstetrician in charge of the birth by using an evaluated standardized questionnaire.7 We also used neonatal data sets including data for all infants who were admitted to a neonatal center in Hesse or Saarland within the first 10 postnatal days. Neonatal data were documented by the hospital to which the infant was transferred and were reported to the neonatal audit center with another questionnaire. The neonatal questionnaire includes information on all diagnoses, on the basis of International Classification of Diseases, 10th Revision codes. We excluded all sets of data that did not coincide completely in all of these variables and those that were incomplete. Data from infants with congenital malformations affecting the cardiorespiratory system, chromosomal aberrations, depression at birth (Apgar score at 5 minutes of <7 or umbilical artery pH of <7.10), fetal hydrops, or mothers with alcohol or drug abuse also were excluded from the study. CS was defined as elective when surgery was performed before the onset of labor, with all other cases defined as secondary CS.
Identification of TTN Cases
The data sets for all newborns who were transferred to a NICU because of respiratory problems were reviewed thoroughly. TTN was defined on the basis of characteristic clinical signs presenting within 24 hours after birth and radiographic findings, in agreement with most authors.1,6,8 We identified newborns with TTN on the basis of respiratory distress diagnoses (International Classification of Diseases, 10th Revision code P22), as reported to the Federal Office for Quality Assurance register by the responsible neonatologist. Cases with suspicion of TTN were validated through the absence of other morbidities affecting respiration, particularly perinatal infection, persistent pulmonary hypertension, and meconium aspiration syndrome. Finally, we paired mother-infant data sets of all eligible cases of TTN on the basis of coinciding variables documented on the perinatal and neonatal questionnaires, including date and time of birth, zip code, gender, and birth weight. Therefore, deliveries were divided into groups without labor (no-labor group) and with labor (labor group). Outcome variables of interest were diagnosis of TTN, oxygen therapy, application of continuous positive airway pressure therapy, mechanical ventilation, antibiotic treatment, extracorporeal membrane oxygenation therapy, incidence of pulmonary air leaks, duration of hospitalization, and death. For qualitative comparison with national data, the German Neonatal and Perinatal Database was used.5
For statistical analyses, we used SPSS 15 (SPSS, Chicago, IL). Continuous variables were compared by using t tests when data were distributed normally; otherwise, the nonparametric Mann-Whitney U test was used. Categorical variables were analyzed by using Pearson's χ2 test or Fisher's exact test, as appropriate. Correlation studies were performed by using Pearson's correlation coefficient. The statistical significance level was set at P ≤ .05 for all analyses with 2-tailed comparisons. Results of multivariate regression analyses (including the confounders maternal age, parity, smoking, diabetes mellitus, mode of delivery, GA, birth weight, and gender) were expressed as odds ratios (ORs) and 95% confidence intervals (CIs).
During the study period, the perinatal birth registers included data for 275 459 deliveries, including 259 312 in Hesse between 2001 and 2005 and 16 147 in Saarland between 2004 and 2005 (Fig 1). Among these deliveries, 239 971 (87%) fulfilled the inclusion criteria of GA of ≥37 weeks and singleton live birth. Of these term newborns, 13 346 (5.6%) were transferred to a neonatal ward. After application of the exclusion criteria (congenital anomalies affecting the cardiorespiratory system or chromosomal aberrations, n = 1871; asphyxia, n = 1603; fetal hydrops, n = 8; mothers with alcohol or drug abuse, n = 512), data for the remaining 9352 newborns were matched with maternal data. Variables within the maternal and neonatal data sets coincided for 6843 (73.2%) of those 9352 infants. There were no significant differences between all term infants and those with matched perinatal data sets, with respect to basic maternal and neonatal variables (data not shown). Of the 9352 admitted neonates, postnatal respiratory problems were found as leading causes for neonatal admission for a total of 2984 newborns (31.9%), including TTN (n = 1423), meconium aspiration syndrome (n = 223), persistent pulmonary hypertension of the newborn (n = 24), and infection (n = 1314). The overall incidence of TTN was 5.9 cases per 1000 singleton live births in our study cohort.
The main characteristics of the study cohort and newborns with TTN are summarized in Table 1. Compared with the national German perinatal data for term singletons born between 2001 and 2005, study infants with TTN were more likely to have an inappropriate birth weight for GA (small for GA: 15.8% vs 9.6%; large for GA: 13.5% vs 10.6%; P < .001) and more often were male (60.3% vs 51.4%; P < .001).
The mode of birth was closely related to GA. Nearly 70% of all ECSs were performed at GAs of 37 to 38 completed weeks. The main indications for ECSs were as follows: previous CS (31.7%), breech presentation (17.6%), cephalopelvic disproportion (13.8%), and maternal disorders (11.7%). For 11.6% of ECSs, the indications for surgery included other malpresentations, placental and umbilical cord complications, and pregnancy disorders. In 13.6% of cases, however, the indication for ECS was not defined, and we might have evidence to suggest that CS at maternal request may account for this figure.
The TTN rate was linked inversely to GA and decreased with advancing GA from 37 completed weeks onward (Fig 2). The overall frequency of ECS was 42.2% among newborns with TTN, compared with 9.2% among term singletons in the German population (P < .001) (Table 1). When data were analyzed according to the presence or absence of labor and GA separately, the rate of TTN was greatly increased at each GA from 37 to 40 weeks for the no-labor group, compared with the labor group (Fig 2). In the labor group, the prevalence of TTN remained nearly unchanged between 10.6% and 13.0% from week 42 to week 37; in the no-labor group, there was a significant increase in the incidence of TTN with decreasing GA from 40 to 37 weeks (P for trend < .01). ECSs at GAs of <40 weeks showed significantly greater risk for TTN, compared with those at GAs of >41 weeks (37 weeks of gestation: OR: 4.8 [95% CI: 3.7–6.3]; 38 weeks of gestation: OR: 3.7 [95% CI: 2.9–4.8]; 39 weeks of gestation: OR: 3.4 [95% CI: 2.3–4.7]; 40 weeks of gestation: OR: 2.0 [95% CI: 1.3–3.0]). Infants born vaginally or through secondary CS showed no increased risk for TTN.
Data on clinical outcomes of newborns with TTN were compared between the labor and no-labor groups, as summarized in Table 2. Nearly one half of the infants with TTN needed supplemental oxygen, irrespective of the mode of delivery. However, the absence of labor before birth was significantly associated with longer duration of oxygen supplementation (P < .02) and higher rates of continuous positive airway pressure therapy (P < .05) and mechanical ventilation (P < .002). Infants with TTN after ECS were more likely to develop pulmonary air leaks, compared with newborns after previous labor (2.8% vs 1.0%; P < .05). No cases of extracorporeal membrane oxygenation therapy and no neonatal deaths were recorded.
The duration of oxygen supplementation also was correlated inversely with the duration of labor before birth (r = −0.151; P = .028). This correlation remained significant with controlling for GA, birth weight, and gender of the newborn (P = .046). Multivariate regression analysis of outcome variables, controlled for perinatal confounders, confirmed a significant impact on duration of oxygen therapy for ECS (OR: 2.2 [95% CI: 1.1–17.9]; P < .05). Other outcome variables were found not to be influenced by confounders and mode of birth.
The present study indicates the positive impact of labor on respiratory outcomes among infants with TTN, in agreement with previous studies; the absence of labor in infants with TTN was associated with higher rates of continuous positive airway pressure therapy, mechanical ventilation, and complications from pulmonary air leaks and longer duration of oxygen supplementation, in comparison with infants with TTN who were delivered after having been exposed to uterine contractions. Multivariate regression analyses, with correction for perinatal confounders, established that the absence of labor increased the duration of oxygen supplementation by a factor of 2. The duration of labor was linked significantly to the duration of oxygen supplementation (r = −0.151; P = .028). The length of hospital stay for infants presenting with TTN did not differ on the basis of labor.
We suggest that both the incidence and severity of TTN are affected by labor before birth. TTN is the most common cause of neonatal respiratory distress in term neonates. Delayed resorption of fetal lung fluid after delivery is considered the main pathophysiological factor in TTN.8,9 Defects in active pulmonary epithelial sodium transport,10 mild pulmonary immaturity and transient surfactant deficiency,11 and myocardial left-heart failure attributable to asphyxia have been discussed as the main potential pathogenic factors operative in dysfunction of postnatal lung liquid clearance and TTN pathophysiological processes.12 Studies on genetic risk factors for TTN demonstrated that certain functional polymorphisms in the β-adrenergic receptor–encoding genes are associated with higher TTN risk.13,–,15 Rapid clearance of fetal lung fluid during postnatal lung adaptation is largely correlated with the onset of labor before birth.16 Labor enhances the release of catecholamines in maternal and fetal circulation, resulting in β-adrenergic receptor–mediated upregulation of surfactant synthesis and transepithelial sodium ion transport, with subsequent fluid reabsorption, in the neonatal lung. Infants delivered through ECS often are deprived of this labor-related physiological stress response pattern at birth and consequently experience failure of postnatal respiratory transition.17 According to our study results from a large, population-based, birth cohort, the rate of TTN was 0.59% (1423 cases per 239 971 infants), which is comparable to the previous reports by Morrison et al18 (5.7 cases per 1000 infants) and Riskin et al19 (7.2 cases per 1000 infants). The inclusion of neonates with TTN with GAs of ≥35 weeks might explain the higher incidence of TTN in the study by Riskin et al.19
We found that TTN risk was associated strongly with the mode of delivery and GA. The overall rate of ECS was substantially higher in newborns with TTN, compared with term singletons in the German population (42.2% vs 9.2%; P < .001). Other risk factors for TTN were male gender and inappropriate birth weight for GA. Our results agree with previously reported studies that demonstrated the association of ECS with higher risk for neonatal respiratory morbidity and the importance of timing of ECS to GA of ≥39 weeks, to decrease the TTN risk.18,20,–,25 In line with the report by Richardson et al,26 however, ECS performed even at a GA of 40 completed weeks was associated with a significant increase in the TTN rate in our study, which demonstrates the beneficial effects of labor in the mechanisms mediating postnatal lung liquid clearance.
Our study has several limitations. The main limitation is that the perinatal registers used for this study may contain coding inaccuracies and data entry errors, which are inherent in all population-based analyses. In our study, 26.8% of all data sets for infants who were admitted to the neonatal ward could not be connected with maternal data (for newborns with TTN, the matching rate was 76%). We are aware of the possibility that the results might be biased by the high loss rate. However, basic validation analyses of the main maternal and neonatal characteristics and outcome variables for all 9352 neonates and the 6843 matched newborns demonstrated no significant differences, which indicated that the findings are robust.
Despite these limitations, the strengths of our study are the large number of term newborns from a 5-year, population-based cohort, the maternal and obstetric variables available for analysis, and the fact that infants with TTN were selected accurately after exclusion of other predetermined causes of respiratory distress, such as perinatal infection, meconium aspiration syndrome, asphyxia, and congenital anomalies. Our main concern was the comparability of a large number of newborns with validated TTN differentiated according to the mode of delivery.
The main finding of the present study is that the mode of birth and the absence of labor are strongly related to the development of TTN in newborns, irrespective of GA, even at term. Our analyses are in good accordance with previous data indicating ECS, low GA, male gender, and inappropriate birth weight as main risk factors for TTN. Furthermore, the presence of labor before birth was shown to improve respiratory outcomes for infants with TTN. To the best of our knowledge, this is the first study to demonstrate that not only the presence or absence but also the duration of labor before birth positively influences the course of TTN, with a shorter need for supplemental oxygen therapy. No correlations with other outcome variables were found. Our results clearly indicate ECS as the main risk factor for TTN and the crucial role of labor in respiratory outcomes in TTN.
According to the analyses of our data and previously reported data, it seems that awaiting the onset of labor before delivery reduces significantly the risk of postnatal development of TTN and improves the clinical course of respiratory failure. Every pregnant woman justifiably wishes a short labor and an uneventful delivery without complications. Therefore, a growing number of mothers tend toward ECS. During the decision-making process, women should be given reliable information about the risks and benefits associated with CS. When no clear medical indication or benefit to the mother or fetus is given and ECS is planned, we suggest, on the basis of our findings, that CS be postponed until a GA of 39 completed weeks whenever possible.
- Accepted October 5, 2009.
- Address correspondence to Erol Tutdibi, MD, Children's University Hospital of Saarland, Center of Pediatrics and Neonatology, Kirrberger Strasse, 66421 Homburg/Saar, Germany. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
- TTN =
- transient tachypnea of the newborn •
- ECS =
- elective cesarean section •
- GA =
- gestational age •
- CS =
- cesarean section •
- OR =
- odds ratio •
- CI =
- confidence interval
- 5.↵Federal Office for Quality Assurance.[BQS Bundesgeschäftsstelle Qualitätssicherung gGmbH]Available at: www.bqs-outcome.de/2008/ergebnisse/leistungsbereiche/geburtshilfe/index_html. Accessed December 29, 2009
- Schneider H,
- Husslein P,
- Schneider KTM
- Welsch H,
- Wischnik A
- Spillman T,
- Cotton DB,
- Golunski E
- Landmann E,
- Schmidtpott M,
- Tutdibi E,
- Gortner L
- Barker PM,
- Olver RE
- Hansen AK,
- Wisborg K,
- Uldbjerg N,
- Henriksen TB
- Copyright © 2010 by the American Academy of Pediatrics