Results
A total of 430 infants were screened, and 289 were enrolled, 146 in the
Pre-E group and 143 in the normotensive comparison cohort. Figure 1
illustrates the derivation of the study cohort. The demographic and
clinical features of the Control and Pre-E (non-severe and severe)
cohorts are summarized in Table 1. There were no differences among
groups for sex and maternal race; however, as expected, GA, birth
weight, and birth length of the Severe Pre-E infants were significantly
lower compared to infants in the Control and Non-severe Pre-E groups.
Severe Pre-E infants were also more likely to have fetal growth
restriction (FGR), to be small of gestional age (SGA), and to be born to
women treated with antenatal steroids (ANS). No significant differences
were observed among the groups in terms of need for mechanical
ventilation, surfactant therapy, development of bronchopulmonary
dysplasia, history of maternal smoking, and second-hand smoke exposure.
A similar pattern of clinical features was seen when comparisons were
restricted to the infants who had IPFT obtained (data not shown).
IPFTs were performed at a mean corrected-age of 7.99 months (SD 2.69,
range=4.14-19.53), and the results are summarized in Table 2.[add
information to address Laura’s comment of drop out] DLCO, DLCO/VA, and
VA were not different for the Control and Pre-E group, even when the
latter was divided into the Non-Severe and Severe Pre-E groups. In
addition, Hemoglobin (Hb) concentration did not differ among the 3
groups (11.5±1.2, 11.3±1.4, and 11.7±1.2; p > .3728). While
FVC did not differ among any of the groups, the forced expiratory flows
(FEF50; FEF25-75) tended to be higher in the Pre-E group compared to
Controls, but didn’t reach statistical significance (FEF50: 483 vs. 445,
p < 0.057; FEF25-75: 442 vs. 413, p < 0.098 ).
However, when Pre-E was divided by severity, the Non-severe Pre-E cohort
had significantly higher FEFs compared to the Controls and Severe Pre-E
infants (Table 2). In addition, FEV0.5 was higher in the Non-severe
Pre-E group, although the difference did not reach statistical
significance (p = 0.058)
There were 234 infants (Control N=108; Pre-E N = 126) whose caregivers
responded to at least one survey in the first 6 months of life, and
again after 6-months of life. The demographics and clinical features of
this group were similar to those of the entire cohort and the proportion
of respondents was similar in both the control group and pre-E groups.
Among all infants with respiratory quesstonairres, 98 (42%) had at
least 1 episode of wheezing reported in the first year of life. To
determine potentially significant confounding covariates, we initially
performed bivariate analysis of wheezing outcomes (No Wheeze vs Wheeze)
(Table 3). There were 81 out of a total of 126 Pre-E infants with no
wheezing (64%), and this was significantly greater than that of the
Control group (55 out of 108, 51%, P=0.039). When the Pre-E subjects
were divided by severity, there was not a significant difference in the
proportion of subjects with No Wheeze among the three groups (Control,
Non-Severe Pre-E, and Severe Pre-E). Preterm birth, lower birth weight,
lower birth length, and lower GA, maternal smoking and history of asthma
were also associated with a higher prevalence of wheezing. An analysis
restricted to only those infants who also had IPFT revealed a similar
pattern. Using covariates related to risk of wheeze selected from the
bivariate analysis (Table 3) and additional variables based on clinical
relevance (listed in the Methods), we then performed multi-variable
logistic regression modeling for the outcome of Wheeze vs No Wheeze. The
odds ratio (OR) of Wheeze was significantly lower in the combined Pre-E
group compared to Controls (0.47; p < 0.009). When the Pre-E
group was divided into Non-Severe and Severe Pre-E, each of the Pre-E
groups had OR for Wheeze of less than 1.0; however, compared to the
Control group, only the Severe Pre-E group, which composed the majority
of Pre-E subjects, had significantly lower risk for wheeze (Table 4).
Family history of asthma, maternal smoking, and antenatal steroids were
all included in the logistic model based upon their associated increased
risk of Wheeze in the bivariate analysis (Table 3), only antenatal
steroids, which was more frequent in the Severe Pre-E group, was
statistically significant, while family history of asthma and maternal
smoking were still associated with an increased risk of wheeze with p
values < 0.10 (Table 1). Very few infants were hospitalized
for respiratory illness following discharge, precluding a detailed
analysis of this respiratory outcome.
With all subjects grouped together(Control, non-severe and severe
Pre-E), higher FEF25-75 was associated with lower risk of wheeze (OR =
0.74 (0.55, 1.00); p = 0.048); however, when analyzed as 3 separate
groups (Control, Non-Severe, Severe Pre-E), the relationships between
FEF25-75 and the risk of wheeze (OR: 0.61 (0.35, 1.05), 1.57 (0.63,
3.92), 0.73 (0.47, 1.14)) were not significant for any of the groups (p
values: 0.076, 0.339, 0.164).