Objective
The objective of this study was to compare the temporal sequence of fetal brain hemodynamic changes in near-term small-for-gestational-age fetuses as measured by spectral Doppler indices or by fractional moving blood volume.
Study Design
Cerebral tissue perfusion measured by fractional moving blood volume, cerebroplacental ratio, anterior cerebral artery, and middle cerebral artery pulsatility indices were weekly performed in a cohort of singleton consecutive small-for-gestational-age fetuses with normal umbilical artery delivered after 37 weeks of gestation.
Results
A total of 307 scans were performed on 110 small-for-gestational-age fetuses. Mean gestational age at diagnosis and at delivery was 35.7 and 38.6 weeks, respectively. The proportion of fetuses with abnormal fractional moving blood volume, cerebroplacental ratio, anterior cerebral artery-pulsatility index, and middle cerebral artery-pulsatility index values was 31.3%, 16.8%, 17.2%, and 10.8% at 37 weeks of gestation and 42.7%, 23.6%, 20.9%, and 16.4% before delivery.
Conclusion
The presence of brain redistribution in small-for-gestational-age fetuses was detected earlier and in a higher proportion of fetuses using cerebral tissue perfusion rather than spectral Doppler indices.
Small fetuses with normal umbilical artery (UA) Doppler are considered 1 end of the spectrum of the normal population. However, recent evidence suggests that a substantial proportion of them have true intrauterine growth restriction (IUGR) as suggested by a poorer perinatal outcome and an increased prevalence of abnormal neurobehavioral and neurodevelopmental tests, both neonatally and in childhood. Because identification of this subgroup of small-for-gestational-age (SGA) fetuses with true milder forms of growth restriction cannot be determined by UA Doppler, direct fetal signs, such as the assessment of brain redistribution, have been proposed. Middle cerebral artery (MCA) pulsed Doppler has long constituted the clinical standard for the diagnosis of brain redistribution. Up to 15% of term SGA have a reduced pulsatility index (PI) in the MCA, and this is associated with poorer perinatal outcome and with an increased risk of abnormal neurobehavior at birth, along with suboptimal neurodevelopmental outcome at 2 years of age.
Aside from the MCA, other spectral Doppler indices reflecting brain circulation changes, such as the anterior cerebral artery (ACA) PI or the cerebroplacental ratio (CPR), have been proposed for clinical detection of brain redistribution in growth-restricted fetuses. However, these indices have only been evaluated in groups of fetuses with early-onset IUGR, and, consequently, their behavior in SGA fetuses with normal UA PI is unknown. In contrast to indices based on the Doppler flow patterns of brain arteries, in recent years a different approach to assess fetal brain circulation has been proposed. Fractional moving blood volume (FMBV) uses power Doppler to estimate quantitatively brain tissue perfusion. This method has been validated in animal models and has been demonstrated to be a reproducible and potentially more sensitive method for the detection of brain blood flow redistribution in fetal growth restriction. In a recent study on SGA fetuses, increased brain tissue perfusion by FMBV predicted abnormal neonatal neurobehavioral performance with better accuracy than pulsed Doppler evaluation of the MCA.
Sequential evolution of previously reported brain hemodynamic parameters, based on either spectral or power Doppler, has not been evaluated in SGA near-term fetuses. This information is of clinical relevance, since parameters offering earlier detection of SGA fetuses with true forms of growth restriction might allow timely delivery in a larger number of cases. We hypothesized that evaluation of brain tissue perfusion by FMBV could be an earlier sign of brain redistribution rather than those parameters based on spectral Doppler evaluation of brain arteries. In this study, we evaluated the longitudinal changes of brain tissue perfusion measured by FMBV in relation to the changes in the MCA, CPR, and ACA Doppler indices.
Materials and Methods
Subjects
A prospective cohort was created of consecutive cases of singleton fetuses with estimated fetal weight below the 10th percentile according to local standards, born beyond 37 weeks of gestation corrected by first-trimester ultrasound, between December 2007–July 2009. Exclusion criteria were as follows: (1) congenital malformations and chromosomal abnormalities; (2) UA PI above the 95th percentile during follow-up; and (3) confirmed birthweight above the 10th percentile according to local standards. The protocol was approved by the hospital ethics committee and written consent was obtained for the study from all the women. A total of 60 fetuses included in this study had been included in a previous series on SGA.
Ultrasound and Doppler measurements
Prenatal Doppler ultrasound examinations were performed weekly between diagnosis and delivery by an experienced operator (R.C.M.) using a Siemens Sonoline Antares (Siemens Medical Systems, Malvern, PA) ultrasound machine equipped with a 6–2 MHz linear curved-array transducer. Doppler recordings were performed in the absence of fetal movements and voluntary maternal suspended breathing. Pulsed Doppler parameters were performed automatically from 3 or more consecutive waveforms, with the angle of insonation as close to 0° as possible. A high-pass wall filter of 70 Hz was used to record low blood flow velocities and avoid artifacts. UA PI was performed from a free-floating cord loop. The MCA PI was obtained in a cross-sectional view of the fetal head, at the level of its origin from the circle of Willis. The CPR was calculated as a ratio of the MCA PI divided by the UA PI. For the ACA PI, the Doppler gate was placed in its first segment, immediately after the origin of the ACA from the internal carotid artery. Normal UA was considered as a PI below the 95th percentile. The MCA PI, ACA PI, or CPR values below the fifth percentile were considered indicative of cerebral blood flow redistribution. In all cases, the last examination was performed within 1 week of delivery. Labor induction was programmed at term for cases with preeclampsia or an estimated fetal weight below the third percentile by cervical ripening. Delivery was attended by a staff obstetrician.
Cerebral blood perfusion
Using power Doppler ultrasound, frontal tissue perfusion was evaluated weekly in a sagittal view of the fetal head. In a midsagittal view of the fetal brain, the power Doppler color box was placed to include all the frontal area of the brain. Five consecutive high-quality images with no artifacts were recorded using the following fixed US settings: gray-scale image for obstetrics, medium persistence, wall filter of 1, gain level of 1, and pulsed repetition frequency of 610 Hz. All images were examined offline and FMBV was estimated with the MATLAB software 7.5 (The MathWorks, Natick, MA) as previously described. The mean FMBV from all 5 images was considered as representative for that specific case and expressed as percentage. The region of interest (ROI) was delineated as described elsewhere: anteriorly by the internal wall of the skull, inferiorly by the base of the skull, and posteriorly by an imaginary line drawn at 90° from the origin of the ACA and parallel to an imaginary line in the front of the face and crossing at the origin of the internal cerebral vein ( Figure 1 ). Increased frontal perfusion was considered as an FMBV above the 95th percentile according to local standards.
Statistical analysis
The longitudinal changes were analyzed by Kaplan-Meier survival analysis, in which the endpoint was defined as an abnormal Doppler value (MCA PI, CPR, and ACA PI below the fifth centile or FMBV above the 95th percentile). The McNemar test was used to compare qualitative data. Statistical analysis was performed using the Statistical Package for Social Sciences (SPSS 15.0; SPSS, Inc, Chicago, IL) statistical software.
Results
During the study period a total of 307 scans were performed on 110 SGA fetuses. UA, MCA, and ACA were successfully obtained in all examinations, whereas frontal brain perfusion could not be obtained in 4 examinations because of the degree of engagement of the fetal head into the pelvis.
The mean gestational ages at first and last scan were 35.7 (range, 29.4–38.4) and 38.6 (range, 37.0–41.9) weeks, respectively. The median interval between the last examination and delivery was 2 (range, 0–8) days. Table 1 shows the maternal and neonatal clinical characteristics of the population.
| Characteristic | SGA, n = 110 |
|---|---|
| GA at inclusion, wk | 35.7 (2.0) |
| GA at last scan, wk | 37.8 (1.4) |
| Maternal age, y | 32.0 (5.4) |
| Low socioeconomic class, a % | 46.4 |
| Primiparity, % | 57.3 |
| Nonwhite ethnicity, % | 13.6 |
| Smoking, % | 26.4 |
| 1–10 cigarettes/d | 18.2 |
| 10–19 cigarettes/d | 1.8 |
| ≥20 cigarettes/d | 6.4 |
| Preeclampsia, % | 7.3 |
| Labor induction, % | 59.1 |
| Cesarean section, % | 31.8 |
| GA at delivery, wk | 38.6 (1.3) |
| Birthweight, g | 2394 (260) |
| Birthweight percentile | 4.5 (3.0) |
| 5-min Apgar score <7, % | 0 |
| Admission to the neonatal unit, % | 6.4 |
| Stay in the neonatal unit, d | 1.1 (2.5) |
a Routine occupations, long-term unemployment, or never worked (UK National Statistics Socio-Economic Classification).
At the first scan, the proportion of cases with abnormal MCA PI, CPR, ACA PI, and FMBV was 3.6% (n = 4), 5.5% (n = 6), 2.7% (n = 3), and 9.1% (n = 10), respectively. No significant differences were observed between these proportions. At the last examination before delivery, the proportion of increased FMBV (42.7%) was significantly higher than the proportion of abnormal MCA PI (16.4%; P < .01), abnormal CPR (23.6%; P < .01), and abnormal ACA PI (20.9%; P < .01).
For survival analysis, cases with abnormal spectral or power Doppler at first scan were excluded, leaving a final population of 96 fetuses that were longitudinally analyzed, in whom a total of 249 scans were performed (median, 2; range, 2–5). Figure 2 shows the survival graph of the Doppler parameters throughout the study period, plotted against gestational age, which could be interpreted as the remaining proportion of normal MCA PI, ACA PI, CPR, and FMBV at each week of gestational age. At 37 weeks, the proportion of abnormal values was 10.8% (95% confidence interval [CI], 4.1–17.4) for the MCA PI, 16.8% (95% CI, 8.7–24.9) for the CPR, 17.2% (95% CI, 9.3–25.4) for the ACA PI, and 31.3% (95% CI, 21.5–41.0) for the FMBV. Similarly, the first quartile survival time (when a quarter of the population had abnormal Doppler) occurred at 39.14 weeks (95% CI, 38.1–40.2) for the MCA, at 38.3 weeks (95% CI, 37.0–39.5) for the CPR, 38.3 weeks (95% CI, 37.0–39.5) for the ACA, and 36.7 weeks (95% CI, 36.0–37.4) for the FMBV. Figure 3 depicts the changes in the proportion of abnormal Doppler between diagnosis and delivery for each parameter.
