Objective
The purpose of this study was to evaluate the surveillance characteristics that precede stillbirth in growth-restricted fetuses that receive integrated Doppler and biophysical profile scoring (BPS).
Study Design
Nine hundred eighty-seven singleton pregnancies that were complicated by fetal growth restriction had multivessel Doppler scans (umbilical and middle cerebral arteries [MCA], ductus venosus, and umbilical vein) and BPS. Surveillance findings were compared between live births and stillbirths.
Results
Forty-seven stillbirths occurred in 2 clusters, 37 at <34 weeks of gestation and 10 thereafter. Before 34 weeks of gestation, stillbirths had parallel escalation of umbilical artery and ductus venosus Doppler findings followed by abnormal BPS. At ≥34 weeks of gestation, only a decline in MCA pulsatility index was observed, and 75% of stillbirths were unanticipated by the BPS.
Conclusion
Before 34 weeks of gestation, multivessel Doppler abnormality anticipates an abnormal BPS and subsequent stillbirth. After 34 weeks of gestation, stillbirths occur after MCA brain-sparing in a shorter interval than predicted by a normal BPS. Recognition of these differences in clinical behavior requires consideration for the planning of monitoring intervals in preterm and term fetal growth restriction.
The prevention of stillbirth is a primary goal of antenatal surveillance. In placenta-based fetal growth restriction (FGR), characteristic Doppler and biophysical profile findings can herald fetal deterioration. Therefore, surveillance is instituted to identify fetal risks for hypoxia or stillbirth that may require intervention through delivery. The Growth Restriction Intervention Trial demonstrated that expectant management carries an increased risk of stillbirth, which suggests that surveillance may fail to identify FGR fetuses who are at greatest risk. Current surveillance strategies use either multivessel arterial and venous Doppler studies or biophysical parameters.
Irrespective of the surveillance strategy used, there may be several reasons for surveillance failure that include gestational age differences in the manifestation of cardiovascular compromise and the inability to predict the rate of clinical deterioration when only dynamic biophysical fetal variables are assessed. Because fetal deterioration that results from placental dysfunction is associated with cardiovascular and behavioral abnormalities, we previously suggested the integration of Doppler and biophysical profile scoring (BPS) for surveillance of FGR pregnancies. However, one of the prerequisites for meaningful integration is to understand the clinical progression to fetal compromise when both surveillance methods are applied concurrently. In this context, stillbirth could be considered the most certain endpoint of fetal compromise, the anticipation of which is the central motivator for fetal surveillance. Accordingly, it was the aim of this study to evaluate the surveillance characteristics that precede stillbirth in growth-restricted fetuses who receive integrated Doppler and biophysical surveillance.
Materials and Methods
This was a retrospective, institutional review board–approved study of FGR singleton pregnancies that were monitored with a prospectively defined standard integrating Doppler and BPS surveillance at 2 referral centers for high-risk pregnancies between January 2000 and December 2012. The criteria that defined FGR included any of the following events: (1) abdominal circumference <5th percentile, (2) sonographically estimated fetal weight <10th percentile according to local reference standards, (3) fetal head-to-abdominal circumference ratio >90th percentile by local reference values, or (4) abdominal circumference growth rate <11 mm in a 14-day interval. In all pregnancies, gestational age was determined by sure last menstrual period that was confirmed by sonogram at <20 weeks of gestation. Exclusion criteria were chromosomal or structural anomalies as determined prenatally or at birth, multiple gestation, fetal infection, and/or unavailability of outcome variables.
All patients had combined Doppler and BPS surveillance. The monitoring intervals and criteria for administration of steroids and delivery planning were at the discretion of the managing obstetrician and the local standard of care. Ultrasound equipment that was capable of high-resolution grey scale pulsed-wave and color-Doppler modes with 4, 5, or 8 MHz-sector probes was used. The following testing parameters were determined after 23 weeks of gestation.
Pulsed-wave Doppler measurements of the umbilical artery (UA) at the mid-cord, middle cerebral artery (MCA), and ductus venosus (DV) were analyzed with the use of the pulsatility index (PI) to quantify arterial Doppler waveforms and the PI for veins to quantify the DV waveform. Absolute PI values were converted to z-scores; UA and DV PI z-scores >2, and MCA z-scores ≤2 (brain sparing) were considered abnormal. End-diastolic velocity in the UA was assessed qualitatively as positive, absent, or reversed; for the DV, atrial systolic velocity was assessed qualitatively as forward or absent/reversed. Umbilical venous pulsations in the mid-cord were also noted. It was the convention that was used to record the best value for all Doppler measurements.
A 30-minute nonstress test was performed, and heart rate reactivity was graded by gestational age criteria as defined in the latest iteration of BPS : at 24-29 weeks of gestation, two 10-beat accelerations, each sustained for 10 seconds; at 30-36 weeks of gestation, two 15-beat accelerations, each sustained for 15 seconds; >36 weeks of gestation, two 20-beat accelerations, each sustained for 20 seconds. Two points were given if these criteria were met. No points were given if these cutoffs were not met in a 30-minute period. The performance of a nonstress test was optional when additional biophysical parameters were normal.
Dynamic fetal variables (tone, movement, fetal breathing) were determined according to the Manning criteria. If criteria were not met, a score of 0 was given. The full 5-component BPS was considered normal for a score of 8 or 10. A score of 6 or 8 with oligohydramnios (defined as maximum vertical amniotic fluid pocket <2 cm) was considered equivocal. Scores <6 or 6 with oligohydramnios were abnormal. When the combination of dynamic variables and maximum vertical pocket yielded a normal score of 8 of 8 the performance of the nonstress test was at the discretion of the managing physician.
The results of all fetal surveillance ultrasound scans, which included gestational age at ultrasound scan, interval to next examination, and interval to delivery, were ascertained. Stillbirth was defined as intrauterine fetal death at >24 0/7 weeks of gestation and was considered unexpected if it occurred within 1 week of a normal BPS. At delivery, gestational age, mode and indication, Apgar scores, and UA cord blood gases were noted.
Distribution of continuous Doppler variables for serial examinations was evaluated for each group by regression analysis with a curve of best fit that was determined by the F-test. The proportional distribution of categoric variables was examined with χ 2 or Fisher exact testing. Continuous variables were analyzed by Mann-Whitney U test. SPSS software (version 21.0; SPSS Inc, Chicago IL) was used for all statistics. The Bonferroni correction was used to adjust probability values to account for several statistical analyses that were performed simultaneously on a single dataset.
A probability value of < .003 was considered significant. A linear regression was done that compared changes in Doppler index between both groups, with data placed into subgroups by 7-day intervals from delivery. Only 1 examination per patient was taken for each time point.
Results
A total of 1185 patients had Doppler and BPS surveillance for FGR. Of these, 198 women were excluded because the exclusion criteria or study endpoints could not be ascertained. Nine hundred eighty-seven patients who met the inclusion criteria had 4426 surveillance visits. The median number of examinations per patient was 4 (range, 1–27 examinations), with a median interval of 6 days between examinations (range, 1–100 days). The median number of days between first and last examination was 37 (range, 1–100 days). Most of the women were white and primiparous who delivered by cesarean section for abnormal fetal testing results ( Table 1 ). There were a total of 47 stillbirths and 39 perinatal deaths before discharge that resulted in a perinatal mortality rate of 9.2%. The rate of stillbirth and neonatal deaths was similar between centers (27 vs 20 and 21 vs 18, respectively). The proportional distribution of stillbirths was significantly related to gestational age at delivery (χ 2 : 248.715; P < .001) and occurred in 2 clusters, with most of them (37/47 stillbirths; 79%) being <34 weeks of gestation ( P < .05; Figure 1 ). Most of our stillbirths had severe growth restriction with 100% of the <34-week group and 75% of the ≥34-week group having birthweights at less than the first percentile ( P = .014 and .010, respectively). Subsequent analysis was performed separately for patients who delivered before and after 34 weeks of gestation.
| Characteristics | Measure |
|---|---|
| Maternal | |
| Age, y a | 27.9 (16–45) |
| Parity, n (%) | |
| 0 | 483 (48.9) |
| 1 | 262 (26.5) |
| 2 | 122 (12.3) |
| 3 | 35 (3.5) |
| ≥4 | 31 (3.1) |
| Race, n (%) | |
| White | 536 (54.3) |
| African American/Afro Caribbean | 422 (42.7) |
| Asian/other/not reported | 16/4/7 (1.6/0.4/0.7) |
| Prepregnancy conditions, n (%) | |
| Systemic lupus erythematosus | 12 (1.2) |
| Chronic hypertension | 118 (11.9) |
| Pregestational diabetes mellitus | 28 (2.8) |
| Delivery and postpartum | |
| Indication for delivery, n (%) | |
| Abnormal fetal testing | |
| Nonreassuring fetal heart rate | 102 (10.3) |
| Oligohydramnios | 42 (4.25) |
| Nonreassuring biophysical profile score | 58 (5.8) |
| Nonreassuring Doppler parameters | 130 (13.7) |
| Abnormal Doppler and biophysical profile score | 33 (3.3) |
| Maternal hypertensive disorders | |
| Preeclampsia | 79 (8.0) |
| HELLP (hemolysis, elevated liver enzymes, and low platelet count syndrome) | 59 (5.9) |
| Eclampsia | 2 (0.2) |
| Placental abruption | 21 (2.1) |
| Elective delivery for fetal growth delay | 64 (6.4) |
| Spontaneous onset of labor | 117 (11.8) |
| Other | 4 (0.4) |
| Fetal death/fetal intracerebral bleed | 46/1 (4.7/0.1) |
| Mode of delivery, n (%) | |
| Vaginal | 366 (37.0) |
| Cesarean | 618 (62.6) |
| Indications for cesarean delivery | |
| Unfavorable cervix remote from delivery | 308 (31.2) |
| Abnormal fetal heart rate in labor | 114 (11.5) |
| Breech presentation | 29 (2.9) |
| Positive oxytocin challenge test | 15 (1.5) |
| Patient preference/previous cesarean delivery | 42 (4.2) |
| Failure to progress/failed induction | 16 (1.6) |
| Placental abruption | 2 (0.2) |
| Elevated viral load | 7 (0.7) |
| Gestational age at delivery, wk b | 34.1 (24.1–43.0) |
| Birthweight, g b | 1643 (320–4850) |
| 5-minute Apgar score <7, n (%) | 50 (5.0) |
| pH, n (%) | |
| <7.20 | 188 (19.0) |
| <7.00 and/or base excess ≤12 | 31 (3.1) |
| Antepartum death | 47 (4.7) |
| Neonatal death before discharge | 39 (3.9) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
