SEARCH

SEARCH BY CITATION

Keywords:

  • allergic sensitization;
  • asthma;
  • children;
  • delivery;
  • wheezing

Abstract

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Financial disclosure
  7. Conflicts of interest
  8. References

To cite this article: Keski-Nisula L, Karvonen A, Pfefferle PI, Renz H, Büchele G, Pekkanen J. Birth-related factors and doctor-diagnosed wheezing and allergic sensitization in early childhood. Allergy 2010; 65: 1116–1125.

Abstract

Background:  To investigate the associations between clinical obstetric factors during birth and doctor-diagnosed wheezing and allergic sensitization during early childhood.

Methods:  We followed 410 Finnish women from late pregnancy until 18 months age of their children. All children were delivered at term. Doctor-diagnosed wheezing among children was established by questionnaires, while specific immunoglobulin E antibodies to inhalant and food allergens were measured in 388 children at 1 year of age. Data on maternal obstetric variables were recorded at the time of delivery.

Results:  Children of mothers with longer duration of ruptured fetal membranes before birth had significantly higher risk of doctor-diagnosed wheezing during early childhood compared to those children with shorter period of ruptured fetal membranes (III vs I quartile; aOR 6.65, 95% CI 1.99–22.18; P < 0.002 and IV vs I quartile; aOR 3.88, 95% CI 1.05–14.36, P < 0.043). Children who were born by Cesarean delivery had significantly less allergic sensitization at the age of 1 year compared to those who were born by vaginal route (16.0%vs 32.2%; aOR 0.34, 95% CI 0.14–0.80; P < 0.013). Furthermore, allergic sensitization tended to be more common in children with longer duration of labor before birth. No other birth-related obstetric factors, such as induction, the type of fetal membrane rupture during birth or quality of amniotic fluid were associated significantly with the examined outcomes.

Conclusion:  The longer duration of the ruptured fetal membranes possibly reflected the higher risk of intrapartum infection at birth, and further increased the risk of doctor-diagnosed wheezing among offspring.

Abbreviations
aOR

adjusted Odds ratio

CI

confidence interval

kU/l

kilo units per liter

sIgE

Specific immunoglobulin E

Specific maternal factors and complications during pregnancy and delivery may increase the risk of asthma and wheezing among offspring (1–12). Perinatal effects have less and more diverse significance in the development of allergic sensitization and hay fever (2, 12–14). However, only a few studies have evaluated the clinical intrapartum obstetric risks besides mode of delivery in the development of asthma and allergic diseases among offspring (1, 2, 13, 15, 16). Additionally, children born by any type of Cesarean delivery have often been analyzed together even though the microbial contact with maternal vaginal microbiota is also possible via nonelective Cesarean delivery in cases with ruptured fetal membranes and prolonged labor (1, 4, 16–18). Some of the inconsistency between different studies may be explained by these variations, although the most recent studies have confirmed an association between Cesarean delivery and the risk of the later development of asthma among offspring (3, 4, 14, 19).

Kumar et al. (7) showed that maternal chorioamnionitis was associated with an increased risk of early wheezing and physician-diagnosed asthma in those born preterm, but not eczema or food allergy. Further, we demonstrated recently in another birth cohort that specific intrauterine bacterial growth relating to possible subclinical intrauterine infection at the time of Cesarean increased the risk of asthma three- to sevenfold among adolescents (2). There have been some studies evaluating of the significance of prenatal maternal and early neonatal antibiotics in the developing of allergic diseases and asthma among offspring, (20, 21) but thus far we know only one study has focused on the intrapartum period. This study showed in a large population-based data that maternal antibiotics during delivery associated with both transient early wheezing and persistent wheezing among 6- to 7-year-old children (22).

It is apparent that young children with wheezing have strong correlation between both lung function and immune responses in early life and the subsequent risk of persistent wheezing and later asthma (23, 24). As the significance of birth-related variables may also vary between the different ages of the study population, we wanted to evaluate whether there was any association between birth-related obstetric factors and the prevalence of doctor-diagnosed wheezing and specific allergic sensitization among offspring in early childhood.

Patients and methods

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Financial disclosure
  7. Conflicts of interest
  8. References

Our study population consisted of the Finnish participants in an ongoing international birth cohort study (PASTURE; Protection against allergy study in rural environments), which is evaluating the association between a farming environment and the development of allergic diseases among offspring. This cohort has been prospectively followed up from the third trimester of pregnancy and is reported on in more detail in earlier reports (25). The second half of the cohort (the extended cohort) consisted of pregnant women with estimated delivery at Kuopio University Hospital between May 2004 and May 2005, with no selection by the occupation or area of living; however, subjects living in apartments were excluded as to make the building stock comparable between the two parts of the cohort. Otherwise, the inclusion and exclusion criteria as well as study protocol were the same. In this study, we excluded five children who were delivered with <37 gestational weeks. All the children in the study were born between September 2002 and May 2005 with gestational ages between 37–42 weeks. No differences were detected between the Cesarean delivery rates between the Finnish PASTURE cohort and the extended cohort (13.1%vs 11.8%).

Midwives collected intrapartum information during delivery using standardized questionnaires. Information on clinical factors included gestational age, prostaglandin induction before delivery, the duration of labor and ruptured membranes before birth, the type of fetal membrane rupture during vaginal delivery (spontaneous or artificial), the use of oxytocin during labor, the quality of amniotic fluid at delivery (clear or meconish) and the mode of delivery (spontaneous, assisted vaginal, elective Cesarean or nonelective Cesarean). Neonatal data included birth weight and 1- and 5-min Apgar scores. Maternal and paternal allergic disease was defined as the mother or father being recorded as having doctor-diagnosed asthma, hay fever or allergic eczema at any time in life in the questionnaire data gathered during and after pregnancy. Similarly, information of older biological siblings was collected on asthma, hay fever, allergic eczema and respiratory symptoms and diseases, such as recurrent obstructive or asthmatic (wheezy) bronchitis.

Venous blood samples at the age of 12 months from the offspring were analyzed for specific immunoglobulin E (sIgE) to 19 common allergens, by using the Allergy Screen Test Panel for Atopy (Mediwiss Analytic, Moers, Germany) (26). The 13 inhaled allergens tested were two house dust mites (Dermatophagoides pteronyssinus and D. farinae), seven pollens (alder, birch, European hazel, grass pollen mixture, rye, mugwort and plantain), cat, horse and dog dander and the mold Alternaria alternata. The six food allergens were hen’s egg, cow’s milk, peanut, hazelnut, carrot and wheat. Allergic sensitization was defined as sIgE concentration of equal to or greater than 0.35 kU/l.

The first questionnaire was administered during the third trimester of pregnancy, and follow-up data were collected from the mothers by self-administered questionnaires when the children were 2, 12 and 18 months old. In the Finnish PASTURE cohort, however, the 2-month data were obtained by interviews. Questionnaires were based on the International Study of Asthma and Allergies in Childhood’s standardized Questionnaire (27) and the details of the questionnaires have been described elsewhere (28). In brief, information was collected on respiratory symptoms and diseases, such as parent-reported doctor-diagnosed obstructive or asthmatic (wheezy) bronchitis was asked using the question: ‘Has your child had an obstructive or asthmatic bronchitis diagnosed by a doctor?’. Doctor-diagnosed asthma was inquired on by a corresponding question. Because of the small number of children with doctor-diagnosed asthma at <18 months, doctor-diagnosed obstructive or asthmatic bronchitis and asthma were combined into ‘doctor-diagnosed wheezing’. The questionnaire at the age of 12 months covered the period 2–12 months; the questionnaire at the age of 18 months covered the period 12–18 months. The present analysis includes 410 mother–children pairs with available data from questionnaires concerning doctor-diagnosed wheezing (n = 405) or data on sIgE- levels at the age of 1 year (n = 388) and also data on birth-related factors.

Comparisons of the proportions between doctor-diagnosed wheezing and allergic sensitization and proportions of clinical outcomes in the univariate analysis were performed with the chi-square test or Fisher exact test and for the continuous variables with the Mann–Whitney U-test. Logistic regression analysis was used to investigate the relationships between doctor-diagnosed wheezing and the allergic sensitization and the adjusted effects of various predictors including the clinical variables. In logistic analyses, we tested the following potential confounders: gender (female/male), maternal age at delivery (≤28, 29–33 vs >33 years), maternal parity (0, 1 vs≥2), maternal weight gain during pregnancy (<12, 12–16 vs >16 kg), maternal smoking during pregnancy (earlier smoker but not during pregnancy, smoker also during pregnancy, including all mothers who smoked at least the first weeks of pregnancy vs never smoked), paternal smoking during first year (no vs yes), maternal education (basic, middle vs academic), living on a farm during pregnancy (no vs yes), parental allergic diseases (including either maternal or paternal allergic diseases: no vs yes), gestational age (37–39 vs 40–42 gestational weeks), birth weight in tertiles (≤3480, 3481–3820, vs >3821 grams), Apgar scores at 5 min (≤7 vs >7), neonatal antibiotic administration during the first postnatal week (no vs yes), breastfeeding during the first 2 months (total, partial vs no), day care attendance during 1st year (no vs yes), presence of dogs or cats in the home during 1st year (no vs yes) and the cohort (PASTURE vs extended). In the final logistic models, variables were entered simultaneously and were treated as categorical variables. The type of cohort, farming, parental allergy, parity, maternal smoking during pregnancy, birth weight, gender, breast feeding and day care attendance were included in all analysis. Further, other variables were included in the regression analysis if they were significant at a P-value of <0.10 in the univariate analysis or satisfied a change-in-estimate criterion (≥10% change in the odds ratio of specific examined variable). In addition, the mode of delivery (all vaginal and nonelective Cesarean vs elective Cesarean) was included in the final analysis when evaluating prostaglandin and oxytocin induction, as the mode of delivery has a strong association with these factors. If in any case the proportion of missing values exceeded 5%, the remaining data were included in the analysis as a separate group. The cut-off level of significance was 0.05.

Results

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Financial disclosure
  7. Conflicts of interest
  8. References

Within the first 18 months, 43 of 405 (10.6%) children had had doctor-diagnosed wheezing [obstructive or asthmatic bronchitis in 9.9% (40/405) and asthma in 2.2% (9/405), respectively]. The prevalence of allergic sensitization against any examined allergen was 30.2% (117/388), and the prevalences of allergic sensitization against any inhalant and against any food allergens were 21.4% (83/388) and 16.2% (63/388), respectively. In total, 99 children of 117 (84.6%) were allergic to one to two specific allergens and 18 (15.4%) were allergic to three or more than three different allergens. The highest allergic sensitizations were against cat (17.8%, 69/388) and cow’s milk (11.1%, 43/388).

Table 1 shows the clinical characteristics of the study population in relation to cohorts. As expected, there were significant differences in many maternal and environmental factors between two cohorts. Thus, the type of cohort was included as a confounder in all following multivariate analysis.

Table 1.   Basic characteristics of study population in relation to different cohorts
 Finnish PASTUREExtended cohortP-value
  1. Data are presented as mean (SD) or number (%).

  2. P-values are determined by Mann–Whitney U-test or chi-square or Fisher’s test.

  3. *Including only families with equal to or more than two children.

  4. †Including doctor-diagnosed obstructive or asthmatic bronchitis and/or asthma.

Maternal age at delivery32.1 (4.8)30.1 (4.9)0.0001
Maternal parity1.3 (1.0)0.9 (1.0)0.0001
Nulliparous45 (21.8)94 (46.1)0.0001
Maternal weight gain during pregnancy, kg13.9 (5.3)14.4 (5.5)0.183
Maternal smoking during pregnancy30 (14.6)36 (17.6)0.396
Maternal asthma12 (5.8)14 (6.9)0.666
Maternal hay fever47 (22.8)45 (22.1)0.854
Maternal allergic eczema25 (12.1)55 (27.0)0.0001
Maternal education
 Basic77 (37.4)60 (29.4)0.087
 Academic32 (15.5)49 (24.0)0.031
 Paternal asthma11 (5.3)14 (6.9)0.519
 Paternal hay fever26 (12.6)43 (21.1)0.022
 Paternal allergic eczema22 (10.7)30 (14.7)0.221
 Paternal current smoking59 (28.6)54 (26.5)0.623
 Older siblings with asthma, hay fever, allergic eczema or recurrent wheezing*81 (50.3)58 (52.7)0.696
Area of residence during pregnancy
 Countryside129 (62.6)55 (27.0) 
 Village or town77 (37.4)149 (73.0)0.0001
 Living on a farm during pregnancy108 (52.4)14 (6.9)0.0001
 Male gender101 (49.0)105 (51.5)0.621
 Birth weight, g3672 (477)3661 (463)0.974
 Gestational age, weeks39.6 (1.1)39.7 (1.1)0.284
 Apgar scores, 5 min9.0 (0.5)9.0 (0.5)0.319
 Antibiotics during 1st neonatal week17 (8.3)17 (8.3)0.940
Breast feeding at the age of 2 months
 No15 (7.3)22 (10.8) 
 Partial44 (21.4)69 (33.8) 
 Total147 (71.4)113 (55.4)0.004
Maternal smoking after pregnancy26 (12.6)41 (20.1)0.041
Presence of dogs or cats in the home during 1st year142 (68.9)88 (43.1)0.0001
Day care attendance during 1st year4 (1.9)3 (1.5)0.713
Doctor-diagnosed wheezing† <18 months21/201 (10.4)22/204 (10.8)0.912
Atopic sensitization at 1 year49/192 (25.5)68/196 (34.7)0.049
Total206 (100)204 (100) 

Table 2 shows associations between maternal and neonatal clinical factors and the prevalence of doctor-diagnosed wheezing and allergic sensitization among offspring in the univariate analysis. Higher maternal parity and maternal hay fever, lower Apgar scores and administration of neonatal antibiotics during first postnatal week increased significantly the risk of doctor-diagnosed wheezing. Similar associations were not detected between allergic sensitization and these factors. Only day care attendance was associated with the higher risk of allergic sensitization. However, there were only seven children (1.7%) among the total study population who attended day care during the first year.

Table 2.   Basic characteristics of study population in relation to wheezing and allergic sensitization
 Doctor-diagnosed wheezing* < 18 monthsAllergic sensitization at 1 year
NoYesP-valueNoYesP-value
  1. Data are presented as mean (SD) or number (%).

  2. P-values are determined by Mann–Whitney U-test or chi-square or Fisher’s test.

  3. *Including doctor-diagnosed obstructive or asthmatic bronchitis and/or asthma.

  4. †Including only families with equal to or more than two children.

Maternal age at delivery30.9 (4.9)32.0 (4.7)0.22831.1 (4.9)31.2 (5.1)0.617
Maternal parity1.0 (1.0)1.6 (1.0)0.00011.1 (1.0)1.1 (1.0)0.973
Nulliparous131 (36.2)7 (16.3)0.00991 (33.6)42 (35.9)0.659
Maternal weight gain during pregnancy, kg14.2 (5.5)14.2 (4.5)0.86314.0 (5.7)14.7 (5.0)0.264
Maternal smoking during pregnancy58 (16.0)7 (16.3)0.96541 (15.1)22 (18.8)0.368
Maternal asthma22 (6.1)4 (9.3)0.41520 (7.4)5 (4.3)0.253
Maternal hay fever77 (21.3)15 (34.9)0.04465 (24.0)21 (17.9)0.189
Maternal allergic ezcema71 (19.6)9 (20.9)0.83856 (20.7)21 (17.9)0.538
Maternal education
 Basic vs other117 (32.3)18 (41.9)0.21086 (31.7)47 (40.2)0.108
 Academic vs other73 (20.2)7 (16.3)0.54555 (20.3)19 (16.2)0.351
Paternal asthma22 (6.1)3 (7.0)0.73912 (4.4)11 (9.4)0.057
Paternal hay fever62 (17.1)7 (16.3)0.88941 (15.1)24 (20.5)0.193
Paternal allergic ezcema48 (13.3)4 (9.3)0.63139 (14.4)11 (9.4)0.178
Paternal current smoking103 (28.5)10 (23.3)0.47376 (28.0)31 (26.5)0.754
Older siblings with asthma, hay fever, allergic ezcema or recurrent wheezing†114 (49.4)23 (63.9)0.10598 (54.4)36 (48.0)0.348
Area of residence during pregnancy
 Countryside162 (44.8)20 (46.5) 119 (43.9)55 (47.0) 
 Village or town200 (55.2)23 (53.5)0.826152 (56.1)62 (53.0)0.573
Living on a farm during pregnancy110 (30.4)10 (23.3)0.33379 (29.2)36 (30.8)0.749
Male177 (48.9)27 (62.8)0.085129 (47.6)64 (54.7)0.199
Birth weight, g3648 (460)3834 (531)0.0413648 (473)3742 (475)0.060
Gestational age, weeks39.6 (1.1)39.4 (1.0)0.24239.6 (1.1)39.7 (1.1)0.178
Apgar scores, 5 min9.0 (0.4)8.8 (0.8)0.0279.0 (0.6)9.0 (0.3)0.488
Antibiotics during 1st neonatal week26 (7.2)8 (18.6)0.01120 (7.4)11 (9.4)0.500
Breast feeding at the age of 2 months
 No30 (8.3)6 (14.0) 27 (10.0)10 (8.5) 
 Partial97 (26.8)14 (32.6) 69 (25.5)36 (30.8) 
Total235 (64.9)23 (53.5)0.267175 (64.6)71 (60.7)0.544
Maternal smoking after pregnancy60 (16.6)7 (16.3)0.96144 (16.2)21 (17.9)0.678
Presence of dogs or cats in the home during lst year207 (57.2)19 (44.2)0.105153 (56.5)62 (53.0)0.528
Day care attendance during 1st year7 (1.9)0 (0)1.001 (0.4)5 (4.3)0.011
Total362 (100)43 (100) 271 (100)117 (100) 

In crude and adjusted analyses, the duration of ruptured membranes before birth was associated significantly with the doctor-diagnosed wheezing (III vs I quartile; aOR 6.65, 95% CI 1.99–22.18; P < 0.002 and IV vs I quartile; aOR 3.88, 95% CI 1.05–14.36, P < 0.043) (Table 3). No significant differences were detected in the crude or adjusted analysis between other intrapartum clinical variables and the prevalence of doctor-diagnosed wheezing among offspring (Table 3).

Table 3.   Intrapartal estimates and the prevalence of doctor-diagnosed wheezing among offspring within the first 18 months
VariableDoctor-diagnosed wheezing*n/N (%)Crude OR (95%CI)Adjusted OR† (95%CI)
  1. P-value for unadjusted analysis and adjusted analysis obtained from the trend test (Wald) in logistic regression models.

  2. *Doctor-diagnosed obstructive or asthmatic bronchitis and/or asthma in the age of 12 and/or 18 months’ follow-up.

  3. †The following variables were included in analysis (see Methods): gender, maternal age and parity at delivery, maternal smoking and weight gain during pregnancy, maternal education, living on a farm during pregnancy, gestational age, birth weight, breast feeding, parental allergy, paternal smoking, neonatal antibiotic administration after birth, Apgar scores at 5 min, presence of dogs or cats in the home during 1st year, day care attendance during 1st year and study cohort. Also in the evaluation of prostaglandin and oxytocin, the mode of delivery was included.

  4. ‡Excluded those delivered by elective Cesarean operation.

Type of delivery
 Spontaneous vaginal delivery37/333 (11.1)11
 Assisted vaginal delivery 2/22 (9.1)0.80 (0.18–3.56)1.04 (0.19–5.66)
 Elective Cesarean operation 2/29 (6.9)0.59 (0.14–2.59)0.39 (0.07–2.08)
 Nonelective Cesarean operation 2/21 (9.5)0.84 (0.19–3.76) P = 0.9010.86 (0.17–4.33) P = 0. 742
Type of delivery
 Vaginal delivery39/355 (11.0)11
 Cesarean operation 4/50 (8.0)0.71 (0.24–2.06) P = 0.5230.60(0.18–1.95) P = 0.598
Type of ruptured fetal membrane during delivery‡
 Spontaneously ruptured membranes19/164 (11.6)11
 Artifical ruptured membranes20/182 (11.0)0.94 (0.48–1.84)0.77 (0.37–1.62)
 Not known 1/30 (3.3)0.26 (0.03–2.04) P = 0.4430.30 (0.04–2.45) P = 0.471
Duration of ruptured membranes
 I quartile (<45 min) 4/101 (4.0)11
 II quartile (45–155 min)12/107 (11.2)3.06 (0.95–9.83)3.24 (0.96–10.91)
 III quartile (156–343 min)18/97 (18.6)5.53 (1.80–16.99)6.65 (1.99–22.18)
 IV quartile (>344 min) 9/100 (9.0)2.40 (0.71–8.06) P = 0.0163.88 (1.05–14.36) P = 0.020
Duration of labor
 I quartile (< 208 min)11/102 (10.8)11
 II quartile (208–356 min)12/102 (11.8)1.10 (0.46–2.63)1.75 (0.66–4.66)
 III quartile (356–590 min)13/102 (12.7)1.21 (0.51–2.84)2.19 (0.80–6.00)
 IV quartile (> 590 min) 7/99 (7.1)0.63 (0.23–1.70) P = 0.5931.61 (0.47–5.54) P = 0.479
Any use of oxytocin (augmentation or induction) during delivery
 No17/178 (9.6)11
 Oxytocin26/227 (11.5)1.23 (0.64–2.34) P = 0.5381.29 (0.61–2.73) P = 0.504
Any use of prostaglandin (induction) before delivery
 No34/353 (9.6)11
 Prostaglandin 9/52 (17.3)1.96 (0.88–4.37) P = 0.0991.37 (0.53–3.54) P = 0.512
Quality of amniotic fluid
 Normal39/350 (11.1)11
 Meconium-stained 4/55 (7.3)0.63 (0.21–1.83) P = 0.3900.45 (0.14–1.42) P = 0.172
Total43/405 (10.6)  

Table 4 shows that there was a significantly less allergic sensitization detected among children who were born by Cesarean section compared with children who were born vaginally even after adjustments (16.0%vs 32.2%, aOR 0.34, 95% CI 0.14–0.80; < 0.013). Furthermore, the risk of allergic sensitization tended to be increased with the longer duration of labor before birth (III vs I quartile; aOR 1.95, 95% CI 0.98–3.88; P < 0.056 and IV vs I quartile; aOR 1.89, 95% CI 0.89–4.01, P < 0.097)(Table 4). No significant associations were detected between specific inhalant or food allergic sensitization and intrapartum clinical factors (Table 5).

Table 4.   Intrapartal estimates and allergic sensitization among offspring at the age of 1 year
VariableAllergic sensitization*n/N (%)Unadjusted OR (95% CI)Adjusted OR† (95% CI)
  1. P-value for unadjusted analysis and adjusted analysis obtained from the trend test (Wald) in logistic regression models.

  2. *Measured by specific IgE levels ≥ 0.35 kU/l

  3. †The following variables were included in analysis (see Methods): gender, maternal age and parity at delivery, maternal smoking and weight gain during pregnancy, maternal education, living on a farm during pregnancy, gestational age, birth weight, breast feeding, parental allergy, paternal smoking, neonatal antibiotic administration after birth, Apgar scores at 5 min, presence of dogs or cats in the home during 1st year, day care attendance during 1st year and study cohort. Also in the evaluation of prostaglandin and oxytocin, the mode of delivery was included.

  4. ‡Excluded those delivered by elective Cesarean operation.

Type of delivery
 Spontaneous vaginal delivery102/316 (32.3)11
 Assisted vaginal delivery 7/22 (31.8)0.98 (0.39–2.48)0.91 (0.34–2.44)
 Elective Cesarean operation 4/29 (13.8)0.34 (0.11–0.99)0.36 (0.12–1.09)
 Nonelective Cesarean operation 4/21 (19.0)0.49 (0.16–1.51) P = 0.1540.30 (0.08–1.13) P = 0.103
Type of delivery
 Vaginal delivery109/338 (32.2)11
 Cesarean operation 8/50 (16.0)0.40 (0.18–0.88) P = 0.0230.34 (0.14–0.80) P = 0.013
Type of ruptured fetal membrane during delivery‡
 Spontaneously ruptured membranes 47/155 (30.3)11
 Artifical ruptured membranes 55/177 (31.1)1.04 (0.65–1.65)1.05 (0.64–1.73)
 Not known11/27 (40.7)1.58 (0.68–3.66) P = 0.5571.92 (0.79–4.67) P = 0.350
Duration of ruptured membranes
 I quartile (< 45 min)21/96 (21.9)11
 II quartile (45–155 min) 37/102 (36.3)2.03 (1.08–3.82)1.73 (0.90–3.33)
 III quartile (156–343 min)29/95 (30.5)1.57 (0.82–3.01)1.42 (0.71–2.85)
 IV quartile (> 344 min)30/95 (31.6)1.65 (0.86–3.16) P = 0.1751.43 (0.70–2.90) P = 0.440
Duration of labor
 I quartile (< 208 min) 23/101 (22.8)11
 II quartile (208–356 min)28/97 (28.9)1.38 (0.73–2.61)1.66 (0.85–3.25)
 III quartile (356–590 min)33/94 (35.1)1.84 (0.98–3.44)1.95 (0.98–3.88)
 IV quartile (> 590 min)33/96 (34.4)1.78 (0.95–3.33) P = 0.2081.89 (0.89–4.01) P = 0.235
Any use of oxytocin (augmentation or induction) during delivery
 No 46/168 (27.4)11
 Oxytocin 71/220 (32.3)1.26 (0.81–1.97) P = 0.2991.17 (0.72–1.90) P = 0.525
Any use of prostaglandin (induction) before delivery
 No 97/336 (28.9)11
 Prostaglandin20/52 (38.5)1.54 (0.84–2.82) P = 0.1631.47 (0.77–2.81) P = 0.248
Quality of amniotic fluid
 Normal 96/334 (28.7)11
 Meconium-stained21/54 (38.9)1.58 (0.87–2.86)1.63 (0.87–3.07)
P-value 0.1340.130
Total117/388 (30.2)  
Table 5.   Intrapartal estimates and the prevalence of allergic sensitization against inhalalant and food allergens among offspring at the age of 1 year
VariableInhalalant allergic sensitization*n/N (%)Adjusted†OR (95%CI)Food allergic sensitization*n/N (%)Adjusted†OR (95%CI)
  1. P-value for unadjusted analysis and adjusted analysis obtained from the trend test (Wald) in logistic regression models.

  2. *Measured by specific IgE levels ≥ 0.35 kU/l.

  3. †The following variables were included in analysis (see Methods): gender, maternal age and parity at delivery, maternal smoking and weight gain during pregnancy, maternal education, living on a farm during pregnancy, gestational age, birth weight, breast feeding, parental allergy, paternal smoking, neonatal antibiotic administration after birth, Apgar scores at 5 min, presence of dogs or cats in the home during 1st year, day care attendance during 1st year and study cohort. Also in the evaluation of prostaglandin and oxytocin, the mode of delivery was included.

  4. ‡Excluded those delivered by elective Cesarean operation.

Type of delivery
 Spontaneous vaginal delivery71/316 (22.5)153/316 (16.8)1
 Assisted vaginal delivery5/22 (22.7)0.93 (0.30–2.85)4/22 (18.2)0.99 (0.30–3.30)
 Elective Cesarean operation4/29 (13.8)0.69 (0.22–2.18)4/29 (13.8)0.91 (0.29–2.90)
 Nonelective Cesarean operation3/21 (14.3)0.47 (0.12–1.76) P = 0.6512/21 (9.5)0.40 (0.08–2.09) P = 0.753
Type of delivery
 Vaginal delivery76/338 (22.5)157/338 (16.9)1
 Cesarean operation7/50 (14.0)0.57 (0.24–1.35) P = 0.2016/50 (12.0)0.67 (0.25–1.75) P = 0.409
Type of ruptured fetal membrane during delivery‡
 Spontaneously ruptured membranes31/155 (20.0)125/155 (16.1)1
 Artifical ruptured membranes40/177 (22.6)1.06 (0.61–1.84)26/177 (14.7)1.12 (0.59–2.13)
 Not known8/27 (29.6)1.80 (0.69–4.73) P = 0.4798/27 (29.6)3.26 (1.14–9.26) P = 0.082
Duration of ruptured membranes
 I quartile (<45 min)18/96 (18.8)111/96 (11.5)1
 II quartile (45–155 min)22/102 (21.6)0.98 (0.47–2.05)24/102 (23.5)2.05 (0.90–4.69)
 III quartile (156–343 min)21/95 (22.1)1.06 (0.50–2.25)14/95 (14.7)1.32 (0.53–3.30)
 IV quartile (> 344 min)22/95 (23.2)1.15 (0.54–2.48) P = 0.97514/95 (14.7)1.13 (0.44–2.87) P = 0.278
Duration of labor
 I quartile (< 208 min)16/101 (15.8)115/101 (14.9)1
 II quartile (208–356 min)22/97 (22.7)1.68 (0.80–3.52)15/97 (15.5)1.19 (0.52–2.74)
 III quartile (356–590 min)22/94 (23.4)1.76 (0.81–3.79)14/94 (14.9)0.97 (0.41–2.30)
 IV quartile (> 590 min)23/96 (24.0)1.83 (0.79–4.25) P = 0.41619/96 (19.8)1.33 (0.53–3.30) P = 0.873
Any use of oxytocin (augmentation or induction) during delivery
 No33/168 (19.6)126/168 (15.5)1
 Oxytocin50/220 (22.7)1.18 (0.69–2.01) P = 0.55437/220 (16.8)1.09 (0.59–2.03) P = 0.777
Any use of prostaglandin (induction) before delivery
 No68/336 (20.2)154/336 (16.1)1
 Prostaglandin15/52 (28.8)1.47 (0.73–2.95) P = 0.2809/52 (17.3)1.12 (0.49–2.60) P = 0. 785
Quality of amniotic fluid
 Normal67/334 (20.1)154/334 (16.2)1
 Meconium-stained16/54 (29.6)1.64 (0.83–3.23) P = 0.1539/54 (16.7)0.98 (0.43–2.24) P = 0.958
Total83/388 (21.4) 63/388 (16.2) 

To control the interaction between the mode of delivery and duration of ruptured membranes before birth, we analyzed separately those children who were delivered spontaneously by vaginal route (N = 333). Among this selected population, the association between the duration of ruptured membranes and doctor-diagnosed wheezing among offspring was stronger than in total study population (III vs I quartile; aOR 9.78, 95% CI 1.99–48.14; P < 0.005 and IV vs I quartile; aOR 4.93, 95% CI 0.88–27.44, P < 0.08). In addition, as the duration of ruptured membranes correlates significantly also with maternal parity, we secondly selected women with one or more earlier deliveries (N = 267) in the multivariate analysis excluding nulliparous mothers. In this separate population, similar significant association was detected between duration of ruptured membranes and doctor-diagnosed wheezing among offspring (III vs I quartile; aOR 5.75, 95% CI 1.58–20.94; P < 0.008 and IV vs I quartile; aOR 3.84, 95% CI 0.89–16.54, P < 0.07).

We further evaluated the joint association of the parental allergy and the duration of ruptured fetal membranes with the prevalence of doctor-diagnosed wheezing. While the numbers were low in terms of statistical testing, an increased risk of doctor-diagnosed wheezing was detected among both children of allergic and nonallergic parents with longer duration of ruptured membranes before delivery (aORs 4.16, 13.14 and 6.52 for allergic parents and 6.30, 4.29 and 9.83 for nonallergic parents in comparison with duration of ruptured membranes between II, III and IV vs I quartiles). In contrast, children of allergic parents delivered by Cesarean had significantly lower risk of allergic sensitization compared to those born vaginally (aOR 0.18, 95% CI 0.046–0.68; P < 0.012). Birth by Cesarean delivery was not associated with decreased risk of allergic sensitization in children of nonallergic parents (aORs 0.68 (95% CI 0.21–2.26; P = 0.530).

Discussion

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Financial disclosure
  7. Conflicts of interest
  8. References

The present results show that the duration of ruptured membranes before birth was significantly associated with the risk of doctor-diagnosed wheezing among offspring. Surprisingly, the risk was significantly increased with as short a duration as 3 h of ruptured fetal membranes before birth, which is quite a short period in terms of normal vaginal delivery and a very banal finding in the everyday labor ward. Other birth-related variables such as duration of labor or induction had less effect as regards the development of doctor-diagnosed wheezing at early childhood.

Adjustment for neonatal antibiotic administration did not reduce the observed association between longer duration of ruptured fetal membranes and increased risk of wheezing. There are several potential explanations for the finding. First, if the risk associated with the duration of ruptured membranes is explained by the possible intrauterine infection, newborns with congenitally acquired low grade infection or pathogenic microbial colonization may be free of infective symptoms and therefore are not administered antibiotics right after birth. This is especially true for term and near-term newborns. Secondly, it is possible that the longer duration of ruptured fetal membranes could be a surrogate for some other unrecognized maternal or birth-related risk factor. However, we did include into the final logistic multivariate analysis many other confounders such as maternal parity and parental allergy, gender, birth weight, gestational age and neonatal Apgar scores in addition to the neonatal administration of antibiotics. Entering of all these variables together into the analysis, the detected association between the duration of ruptured fetal membranes and doctor-diagnosed wheezing was even stronger.

The information on maternal microbiota before birth might better elaborate on the infectious and inflammatory mechanisms that link the duration of ruptured fetal membranes to doctor-diagnosed wheezing. Unfortunately, we did not record microbiologic information in this study. To the best of our knowledge, there are no earlier studies on the significance of the maternal vaginal microbiota at the time of birth or its possible long-term effects on the development of wheezing or asthma among offspring, especially in those children who are exposed to the maternal flora during delivery. However, we recently showed that specific intrauterine bacteria that commonly originate from vaginal microflora at the time of Cesarean delivery were associated with a higher risk of asthma among adolescent offspring (2). Further, Benn et al. (29) evaluated the relevance of maternal vaginal microflora in early pregnancy and its association with the development of wheezing and asthma among offspring at early ages. They showed that maternal vaginal colonization with Ureaplasma urealyticum during early pregnancy was associated with infant wheezing, while specific vaginal growth of staphylococci was associated with asthma during the fifth year of life. The transmission of maternal vaginal microbiota to the fetus is rare at elective Cesarean deliveries, which are commonly performed before the onset of labor and with intact membranes. Thus, in children born by elective Cesarean deliveries, maternal vaginal microbiota might have less significance in the development of wheezing and asthma among offspring. Indeed, the lowest prevalence of doctor-diagnosed wheezing in the present study was among those children who were delivered by elective Cesarean. Certainly, we need more long-term follow-up prospective cohort studies to test this interesting hypothesis of the association between maternal vaginal microbiota, intrapartum subclinical infections and the risk of wheezing or asthma later among offspring.

Surprisingly, in the present study, children who were delivered by the vaginal route had the significantly highest prevalence of allergic sensitization. The risk of allergic sensitization tended to be increased also with the longer duration of labor before birth in contrast to our own earlier results (13), but in line with Vonk et al. (1). Present results contradict the hygiene hypothesis theory and run counter to some of those earlier studies showing an increased risk of allergic sensitization in children born by Cesarean (14, 17, 30–32). However, in our study, number of children who were delivered by Cesarean was low, and therefore, our results as regards those delivered by Cesarean may yield less reliable conclusions.

Some limitations to this approach should be acknowledged. First, the follow-up period was relatively short, extending to the first 18 months in our cohort, and the doctor-diagnosed wheezing was based solely on questionnaire data. Rowe et al. (33) suggested earlier that specific IgE levels at 1 year were not representative of later sensitization but might be transient physiological finding during early childhood. It is thus possible that our results mirror rather the effects of intrapartum factors on the immunologic maturation and not solely later sensitization and risk of allergic diseases. However, our study is still ongoing and we will follow this cohort further with objective measurements of asthma and with allergic sensitization as these children grow up.

Secondly, the number of infants delivered by Cesarean was also somewhat lower than in the general population in our area (the Cesarean section rate for general obstetric practice in our hospital area during the study period was 17.0–18.6%) (34). This may hamper the evaluation of the association between the mode of delivery and different allergic outcomes and may cause some selection bias. It is thus possible that excluding women who were living in blocks of flats lowered the Cesarean delivery rate as these women are more likely to be younger and nulliparous and have a higher probability of having a nonelective Cesarean following trial by labor.

In conclusion, we found a significant association between the duration of ruptured fetal membranes before birth and doctor-diagnosed wheezing in early childhood among offspring. This can be explained by the fact that as the duration of rupture of membranes increases, the probability of ascending infections culminating in clinical and subclinical chorioamnionitis and neonatal infections proportionately increases. According to this hypothesis, our result agrees well with earlier results of the higher risk of wheezing and asthma among children whose mothers had chorioamnionitis or pathogenic intrauterine microbial invasion at the time of birth and a related higher risk of congenital infection (2, 7). Further, the prevalence of allergic sensitization was least among children born by Cesarean delivery. However, the number of children delivered by Cesarean was low, and thus these results should be observed with caution.

References

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Discussion
  6. Financial disclosure
  7. Conflicts of interest
  8. References