Table 2.
Log binomial regression analyses: Preterm birth < 30 weeks occurring with and without Bronchopulmonary Dysplasia (BPD) by overall biomarker groupings
|
Preterm Birth < 30 Weeks
|
||
|---|---|---|
|
BPD(−)
*
|
BPD(+)
**
|
|
| n (%) RRAdj (95% CI)† |
n (%) RRAdj (95% CI)‡ |
|
| Sample | ||
| (n = 662,890) | 1,682 (0.3) | 246 (0.04) |
| No abnormal biomarkers § | ||
| (n = 513,480) | 1,066 (0.2) | 140 (0.03) |
| Reference | Reference | |
| Any “High” Biomarker (MoM ≥ 95th Percentile) | ||
| AFP (n = 27,680) | 267 (1.0) | 52 (0.2) |
| 4.5 (3.9, 5.1) | 6.7 (4.8, 9.2) | |
| hCG (n = 31,519) | 190 (0.6) | 38 (0.1) |
| 2.8 (2.4, 3.3) | 4.1 (2.8, 5.8) | |
| uE3 (n = 22,649) | 98 (0.4) | 11 (0.05) |
| 2.0 (1.6, 2.5) | 1.7 (0.9, 3.1) | |
| Any “Low” Biomarker (MoM ≤ 5th Percentile) | ||
| AFP (n = 28,596) | 33 (0.1) | 10 (0.03) |
| 0.6 (0.4, 0.8) | 1.3 (0.7, 2.5) | |
| hCG (n = 31,481) | 86 (0.3) | 10 (0.03) |
| 1.3 (1.1, 1.6) | 1.2 (0.6, 2.3) | |
| uE3 (n = 27, 255) | 122 (0.5) | 23 (0.1) |
| 2.2 (1.8, 2.6) | 3.1 (2.0, 4.8) | |
RRAdj, Adjusted Relative Risk; 95% CI, Confidence Interval; AFP, alpha-fetoprotein; MoM, Multiple of the Median; hCG, human choronic gonatotropin; uE3, unconjugated estriol.
Due to an increased likelihood that this subgroup might include infants with BPD based on a less restrictive definition, preterm infants (< 30 weeks) were excluded if they were on oxygen at 28 days and/or at 36-weeks.
Bronchopulmonary dysplasia (BPD) was defined as Gestational Age between 25 wks + 0 days and 29 wks + 6 days (inclusive); Birth weight < 1500 grams; No birth defects/congenital anomalies; No surgeries except circumcision or PDA ligation; Having NEC is not an exclusion but NEC surgery is an exclusion; In the hospital and on continuous oxygen at 36 weeks; On ventilator for ≥ 3 days.
Binomial analyses included all maternal characteristics found to be predictive of preterm birth < 30 weeks without BPD (maternal age > 34 years and race/ethnicity = Hispanic or Black (yes versus no), parity = 1 (yes versus no), and SGA birth (yes versus no).
Binomial analyses included all maternal characteristics found to be predictive of BPD (race/ethnicity = Black or Pacific Islander(yes versus no) and parity = 1 (yes versus no)) and SGA birth (yes versus no).
AFP, hCG, and uE3 MoMs all between the 5th and 95th percentiles (AFP > 0.58, < 1.80; hCG > 0.41, < 2.38; uE3 > 0.59, < 1.57).