Objective
Endogenous digoxin-like factor (EDLF) has been linked to vasoconstriction, altered membrane transport, and apoptosis. Our objective was to determine whether increased EDLF in the cord sera of preterm infants was associated with an increased incidence of necrotizing enterocolitis (NEC).
Study Design
Cord sera from pregnant women enrolled in a randomized trial of MgSO 4 for fetal neuroprotection were analyzed for EDLF using a red cell Rb + uptake assay in which the inhibition of sodium pump-mediated Rb + transport was used as a functional assay of EDLF. Specimens were assayed blinded to neonatal outcome. Cases (NEC, n = 25) and controls (neonates not developing stage 2 or 3 NEC, n = 24) were matched by study center and gestational age. None of the women had preeclampsia. Cases and controls were compared using the Wilcoxon test for continuous and the Fisher exact test for categorical variables. A conditional logistic regression analysis was used to assess the odds of case vs control by EDLF level.
Results
Cases and controls were not significantly different for gestational age, race, maternal steroid use, premature rupture of membranes, or MgSO 4 treatment. In logistic models adjusted for treatment group, race, premature rupture of membranes, and gestational age, cord sera EDLF was significantly associated with development of NEC ( P = .023).
Conclusion
These data demonstrated an association between cord sera EDLF and NEC.
Early, preterm deliveries result in infants at risk for major medical complications and death. Among the most serious problems is necrotizing enterocolitis (NEC), which represents a leading cause of neonatal mortality and morbidity. NEC can be characterized by biochemical and morphological changes, but the primary cause for this disease is poorly understood and the immediate mechanisms involved are not fully worked out. The incidence is inversely related to gestational age and birthweight.
NEC is characterized by one or more of the following: reduced bowel motility, bacterial infection, a marked inflammatory response perhaps representing activation of toll-like receptor 4, or difficulty in digestion (especially synthetic infant formula), which results in necrosis of a portion of the intestine. It often occurs suddenly and progresses rapidly and can prove lethal. In severe cases, surgical resection of the bowel is required, and this may result in long-term complications, including neurodevelopmental deficits.
Endogenously produced inhibitors of the sodium pump have been implicated in a number of diseases, especially hypertensive disorders. It is also well documented that the endogenous digitalis-like factors (EDLFs) are present in the placenta, cord serum, and neonatal blood. There is evidence that levels of these factors are proportionately higher in the neonatal circulation and cord blood the more premature the delivery. The EDLFs by definition block ion transport and consequently nutrient transport coupled to the sodium pump but can also cause apoptosis of exposed cells in a cell type and dose dependent manner. Moreover, there is preliminary evidence that some complications of prematurity are reduced in response to antibody Fab treatment that targets EDLFs in the maternal circulation just prior to delivery. Thus, we hypothesized that EDLFs may participate in NEC and that higher EDLF levels are associated with a higher incidence of NEC.
Materials and Methods
Patient population
Cord sera were obtained as part of a previously completed study, the Beneficial Effects of Antenatal Magnesium Sulfate Trial carried out under the auspices of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network. This multicenter trial tested whether prophylactic magnesium sulfate given to women, for whom preterm delivery was imminent, reduced the risk of death or moderate to severe cerebral palsy in their children.
The study enrolled pregnant women presenting from 24.0 to 31.6 weeks’ gestation with advanced preterm labor, premature rupture of the membranes (pPROM), or indicated deliveries. Women were randomized to receive either intravenous magnesium sulfate or placebo.
Cord sera analyzed in this nested case-control study were selected by the Maternal-Fetal Medicine Units (MFMU) Network’s independent Biostatistics Center at George Washington University to produce a set of specimens matched for maternal characteristics. Specimens included cord sera from 28 neonate cases with NEC and 27 matched controls (by center and gestational age) selected for further analysis.
Gestational age categories were the following: 24-26 weeks, 27-29 weeks, and 30-32 weeks. Cases were identified as women with a singleton pregnancy delivering prior to 32 weeks’ gestation, during which the neonate experienced stage 2 or 3 NEC. Controls were mothers with the same characteristics but during which the neonate was live born, survived to hospital discharge, and experienced no bronchopulmonary dysplasia, severe necrotizing enterocolitis, or intraventricular hemorrhage (IVH) grades III or IV. Serum volume was sufficient to allow EDLF levels to be obtained on 25 of the controls and 24 of the cases. Subjects with EDLF values between 0% and 2% were reassigned a value of zero. Baseline characteristics are presented in Table 1 . None of the selected subjects had preeclampsia. Researchers analyzing these specimens for EDLF were blinded to case or control status during the analysis and data evaluation.
| Characteristic | Controls (n = 24) | NEC cases (n = 25) | P value |
|---|---|---|---|
| Race | |||
| African American | 13 (54%) | 12 (48%) | 1.0 |
| Hispanic | 3 (12.5%) | 4 (16%) | |
| Caucasian | 8 (33%) | 9 (36%) | |
| Labor type | |||
| Spontaneous | 13 (54%) | 16 (64%) | .45 |
| Induced | 7 (29%) | 3 (12%) | |
| No labor | 2 (8%) | 2 (8%) | |
| Spontaneous, augmented | 2 (8%) | 4 (16%) | |
| Preeclampsia | 0 (0%) | 0 (0%) | n/a |
| Any maternal steroid use | 24 (100%) | 25 (100%) | n/a |
| pPROM | 23 (96%) | 21 (84%) | .35 |
| Active treatment group, MgSO 4 | 11 (46%) | 13 (52%) | .78 |
| Gestational age at delivery, wks | 28 ± 2 | 28 ± 2 | .66 |
Measurement of EDLF
A functional assay of EDLF was used. This method has been described in detail. The approach can be briefly summarized as follows: sodium pump (SP)-dependent Rb + uptake into red blood cells from nonpregnant, healthy controls is measured in the presence of serum from nonpregnant, healthy controls (as a negative, no inhibition control, ie, maximal uptake) or in the presence of cord sera from the controls and cases studied. A known inhibitor of the SP, ouabain, is also added to a fraction of the same cells at a concentration (10 –3 M) sufficient to inhibit SP-dependent Rb + uptake completely (positive control). The reduction in SP-dependent Rb + uptake is calculated as a percentage of maximal SP-dependent Rb + uptake. Note Rb + is not found in detectable levels in cells or the circulation but is an ion handled equivalently to K + , the natural substrate of the SP, allowing one to monitor uniquely the rate of the SP mediated Rb + entry as well as its specific inhibition.
Using an RbCl standard of known concentration (40.4 μg/L), the relative SD within run was 1.46% and run to run was 2.67%. The specimens were assayed in duplicate and the coefficient of variation for duplicates was 5.4%.
Statistical analysis
Fisher exact test and Wilcoxon rank sum test were used to compare characteristics between categorical and continuous variables, respectively. The primary results are presented as medians with quartiles because of the skewness of the data, although means and SEM are also provided in the text. A conditional logistic regression was used to assess the odds of case/control status by EDLF level, adjusting for treatment group, race, and pPROM and stratified by gestational age (categorized as cited in previous text) at delivery. In evaluating the 2 groups, 23 of 25 of the cases and 22 of 24 of the controls experienced labor, and hence, labor status was not assessed in the statistical models. Finally, EDLF values were considered as a function of the severity of NEC using a Spearman correlation analysis. All P values were 2 sided. A value of P < .05 was considered significant.
Results
There were no significant differences in maternal characteristics between women having premature infants who did not develop NEC and those who did ( Table 1 ). All women with control infants and all the women with infants who developed NEC received steroids. By study design, none of the included specimens were obtained from pregnancies complicated by preeclampsia.
Also, neonatal characteristics were considered for cases and controls ( Table 2 ). None of the characteristics were significantly different; however, the length of hospitalization was on average approximately 35 days longer for infants with NEC ( P = .058) as might be predicted.
| Characteristic | Controls (n = 24) | NEC cases (n = 25) | P value |
|---|---|---|---|
| Infant weight, g | 1177 ± 295 | 1200 ± 366 | .75 |
| Female born | 11 (46%) | 8 (32%) | .39 |
| Length of stay in NICU, d | 40.9 ± 27.4 | 75.8 ± 73.3 | .058 |
Association of NEC with cord serum EDLF
The cord levels of ELDF in those infants developing NEC compared with control infants not developing major neonatal complications were significantly different (control median with quartiles: 11.95% [0, 25.9] vs cases: 26.2% [16.2, 51.6] EDLF inhibition, P = .0467; means ± SEM for these datasets are: controls, 16.1 ± 3.7 vs cases, 30.7 ± 5.3% EDLF inhibition) ( Figure ). The distribution of these EDLF values for controls and cases is further described in Table 3 . Cord serum EDLF was detectable by the assay in samples of all cases and in all but 3 of the controls. Cord serum EDLF levels were not statistically different between controls receiving active Mg 2+ treatment compared with controls receiving placebo. This was also true for NEC positive cases (data not shown).
