Objective
Endogenous digitalis-like factors (EDLFs) are elevated in women with preeclampsia, and the use of an anti-digoxin antibody Fab (DIF) in women with preeclampsia who were remote from term reduced maternal blood pressure and preserved renal function. The objective was to determine whether DIF treatment in women with severe preeclampsia in association with positive EDLFs in maternal serum improves maternal-perinatal outcomes.
Study Design
This was a planned secondary analysis from a randomized, placebo-controlled, double-blind study of DIF in women with severe preeclampsia with positive EDLF status that was managed expectantly between 23 weeks 5 days and 34 weeks’ gestation (19 women received placebo, and 17 women received DIF). Primary outcome variables were a change in creatinine clearance and the use of antihypertensives. Secondary outcomes were maternal and perinatal complications.
Results
Women with positive EDLFs who received DIF had an attenuated decline in creatinine clearance from baseline compared with placebo (–4.5 ± 12.9 vs –53.2 ± 12.6 mL/min; P = .005). In this same group, the use of antihypertensives (the other primary outcome) was lower but not significantly so (41% vs 63%; P = .12). However, women who were treated with DIF had a lower rate of pulmonary edema (1/17 vs 6/19 women; P = .035) and lower rates of neonatal intraventricular hemorrhage (DIF: 0/17 women vs placebo: 5/19 women; P = .015).
Conclusion
In women with severe preeclampsia who were remote from term who were EDLF positive, the use of DIF was associated with improved maternal and neonatal outcome. These findings suggest the need for a large multicenter trial that would evaluate the benefits of DIF in the treatment of women with severe preeclampsia who are remote from term and with positive EDLF status.
The reported incidence of preeclampsia ranges from 3–5% of all gestations. This incidence is expected to increase because of the rising prevalence of several risk factors for preeclampsia (such as maternal obesity, gestational diabetes mellitus, chronic hypertension, advanced maternal age at time of pregnancy, and multifetal gestation). Pregnancies that are complicated by severe preeclampsia at <34 weeks’ gestation are associated with high rates of maternal and perinatal complications, and the rates of these complications are dependent on gestational age at time of onset and on the type of management that is used (immediate delivery vs expectant management). Treatment of women with severe preeclampsia at <34 weeks’ gestation is aimed at keeping the mother and fetus safe, and delivery of an infant who will survive and not require prolonged or intensive neonatal care. Recent studies have suggested that expectant treatment is possible in a select group of women with severe preeclampsia between 24 and 34 weeks’ gestation and that such treatment improves neonatal outcome; however, in very early preeclampsia (<24 weeks’ gestation), there were associated increased rates of maternal complications such as hemolysis, elevated liver enzymes, and low platelet count (HELLP) syndrome, pulmonary edema, deterioration in renal function, and eclampsia.
Endogenous digitalis-like factors (EDLFs) represent a family of circulating inhibitors of the sodium pump (SP). Such SP inhibition causes direct vasoconstriction and has been linked to an increased blood pressure in essential and experimental hypertension. EDLF is also elevated in the circulation of women with preeclampsia. The addition of a specific commercially available anti-digoxin antibody fragment, antigen binding (Fab) in vitro reduced the inhibitory effects of EDLF on the SP. This same Fab has also been shown to have an antihypertensive effect in animal models of hypertension that are thought to be mediated by EDLF. Previously, the administration of digoxin immune Fab (DIF) to women with preeclampsia reduced maternal blood pressure and preserved or improved renal function. Indeed, these data led to a randomized, double-blind, placebo controlled trial (Digibind Efficacy Evaluation in Preeclampsia [DEEP] trial) of DIF in women with severe preeclampsia. This study demonstrated a benefit from DIF treatment on renal function in all women who were enrolled in the trial. The trial, however, found no benefit regarding pregnancy prolongation or improved maternal outcome. The primary analysis of the DEEP trial did not take into account the EDLF status of the women. It is possible, even likely, that DIF is only efficacious in women who are positive for EDLF.
When circulating EDLF was measured subsequently at baseline in the 51 subjects who were enrolled in the DEEP trial, approximately 20% had no detectable circulating EDLF. Given the proposed mechanism of action (ie, binding and inactivating EDLF), DIF administration should have no effect in those women who did not have circulating EDLF, and their inclusion in the intent-to-treat analyses may have diminished the apparent effect of DIF on women with preeclampsia with measurable EDLF.
In this article, we present the results of a planned secondary analysis from the DEEP trial in which DIF effect was evaluated in women who were EDLF positive. The objective of this study was to determine whether DIF treatment in women with severe preeclampsia in association with positive EDLF activity in maternal serum improves maternal and perinatal outcomes and whether the effect is dependent on circulating EDLF levels at the time of enrollment.
Materials and Methods
Original study design and SP inhibition assay of EDLF
A detailed description of the DEEP study has been published. Briefly, all participants were pregnant women who fulfilled the American College of Obstetricians and Gynecologists criteria for severe preeclampsia. Institutional review board approval was obtained at each study site, and all of the women provided informed, signed consent before participation. Other eligibility criteria included a pregnancy 23 weeks 5 days to 34 weeks’ gestation and expected delivery of the fetus within 72 hours as judged by the primary physician. The intent of the original study was to test the efficacy of DIF (Digibind; GlaxoSmithKline, Research Triangle Park, NC) on 2 primary endpoints: change in creatinine clearance (CrCl) as a measure of renal function and antihypertensive use as a measure of improvement or deterioration of hypertension. Women were assigned randomly to DIF or placebo that was given intravenously every 6 hours for up to 8 total doses. DIF was administered to 24 of the 51 women who were included in the study. Patients, physicians, and laboratory personnel were blinded to treatment arm, and all clinical parameters, including EDLF status, were compiled before unblinding of the study. Treatment of preeclampsia in study patients, which included the use of antihypertensive drugs and the timing of delivery, was determined by clinical condition as judged by the primary physician.
As part of the original study, EDLF in plasma was measured by the plasma’s ability to inhibit the SP of red blood cells that were obtained freshly from normal, nonpregnant volunteers. This assay, which measures SP-mediated uptake of Rb + ion from an artificial medium into red cells in the presence and absence of inhibitor, has been described previously and validated in other studies. If there is EDLF present, then less Rb + is taken up into the cytosol of the red blood cells. EDLF activity was determined in triplicate at baseline (before drug or placebo), and at 12, 24, and 48 hours (t = 0, 12, 24, and 48 hours). Results of the Rb + uptake assay at baseline were used to classify patients as being EDLF negative or EDLF positive. Among the 51 women who were enrolled in the original trial, samples for EDLF were available and evaluated in 46 subjects; 10 of the women (22%) were negative, and 36 of the women (78%) were positive. This secondary analysis focuses on the 36 women who were EDLF positive, among whom 19 received placebo and 17 received DIF.
The primary outcome variables were the same as in the primary analysis (ie, change in renal function (CrCl) and the use of antihypertensives. Use of antihypertensive medications was defined as (1) first use of antihypertensive medication during the treatment phase, (2) an increase in antihypertensive medications during the treatment phase in women who were already receiving antihypertensive medications at the time of entry into the study, or (3) delivery necessitated by persistent severe hypertension.
Secondary outcomes were clinical, and laboratory markers of maternal (eg, pulmonary edema, HELLP syndrome, blurred vision), fetal (eg, persistent nonreassuring fetal status, which included hypoxia, that was indicated by reduced fetal blood flow, as measured by uterine artery Doppler imaging, or worsening biophysical profile; fetal heart tracing abnormalities that included tachycardia, bradycardia, a decrease in beat-to-beat variability, or an abnormal pattern such as variable decelerations or late decelerations), and neonatal complications (eg, neonatal birthweight, respiratory distress syndrome, and intraventricular hemorrhage [IVH]).
Statistical analysis
The analysis was performed by a contract research organization (Covance Inc, Princeton, NJ) in a blinded fashion.
Continuous data were analyzed by analysis of covariance with baseline (pretreatment) value, treatment group, gestational age, and study site in the model or by logistic regression analysis with treatment, gestational age, and center in the model. Categoric data were analyzed by Barnard Exact Test. A probability value < .05 was considered statistically significant. Change in CrCl was calculated as the difference in milliliters per minute of treatment time minus the screening value.
Results
Table 1 compares the demographic characteristics at enrollment between the 2 treatment groups in the EDLF positive women with preeclampsia. There were no significant differences between the women who received DIF or placebo in any of the variables that were analyzed. There were also no significant differences in maternal age, parity, race, body mass index, mean arterial pressure, or gestational age at screening between EDLF positive and negative women.
| Parameter | Placebo (n = 19) | Digoxin immune Fab (n = 17) |
|---|---|---|
| Maternal age, y a | 25 ± 5.1 | 26 ± 6.5 |
| Median parity, n b | 1 (1.2) | 1 (0.9) |
| African American, n (%) | 7 (37%) | 6 (35%) |
| Body mass index, kg/m 2 a | 33 ± 6.4 | 35 ± 7.5 |
| Mean arterial pressure, mm Hg a | 111 ± 2.5 | 110 ± 1.9 |
| Gestational age at screening, d a | 202 ± 4.3 | 197 ± 4.7 |
Effect of DIF on circulating EDLF activity
DIF treatment of EDLF positive women, when compared with the initial DEEP analysis, which included both EDLF positive and negative subjects, demonstrated larger and more statistically significant reductions in circulating EDLF levels as compared with pretreatment EDLF levels ( Figure 1 ). In the analysis of all subjects (ie, inclusion of both EDLF positive and negative subjects), only the 12- 24 hour difference was significant (+11.0% SP activity recovery, P = .03).
Effect of DIF on primary outcome measures
Figure 2 , A, compares the effects of DIF treatment to placebo on change in CrCl. EDLF positive women with severe preeclampsia who were receiving DIF had a significantly smaller drop in CrCl from baseline as compared with the women receiving placebo. In addition, renal function deterioration in the placebo group was related positively to the EDLF level (women with >30% SP inhibition, P = .032; Figure 2 , B). Compared with the all-subjects analysis, the difference between treatment and placebo for use of antihypertensives was greater, although the differences did not reach statistical significance (EDLF positive: 41% DIF vs 63% placebo [ P = .12]; all subjects: 46% DIF vs 52% placebo [ P = .4]).
Effect of DIF on secondary outcomes
Table 2 compares latency period from study entry, gestational age at delivery, birthweight, changes in fetal heart rate tracing, rate of nonreassuring fetal status, neonatal respiratory distress syndrome, neonatal IVH, and death between the 2 study groups. DIF-treated EDLF-positive women with preeclampsia had a delivery latency period that was 26 hours longer than placebo-treated women, but this difference was not statistically significant ( P = .17). The rates of IVH in infants, regardless of birthweight, and the rate of IVH in infants with birthweights <1250 g were significantly lower in neonates of women who received DIF ( P = .015 for all infants; P = .012 for infants <1250 g).
