To compare uterine artery volume blood flow (Quta), vascular resistance (Ruta), pulsatility index (Uta PI), and the fraction of maternal cardiac output (CO) distributed to the uteroplacental circulation in pregnant women with and without a previous caesarean section.
Cross-sectional observational study.
University hospital in Norway.
Thirty-two pregnant women with previous caesarean section and 32 matched controls.
Ultrasonography was used to measure uterine artery diameter and blood flow velocity between 22+0 and 23+6 weeks of gestation. Impedance cardiography was used to assess maternal haemodynamics.
Main outcome measures
Quta, Ruta, Uta PI, and the fraction of maternal CO distributed to the uteroplacental circulation.
The mean Quta was 356.26 ± 213.72 ml/minute in cases and 456.41 ± 209.70 ml/minute in controls (P = 0.038). Ruta was significantly (P = 0.026) higher among cases compared with controls (0.32 ± 0.20 versus 0.22 ± 0.14 mmHg/ml/minute), but the Uta PI did not differ between the groups (0.93 ± 0.23 versus 0.92 ± 0.47; P = 0.929). The fraction of maternal CO distributed to the uteroplacental circulation was 5.75 ± 3.68% in cases and 8.45 ± 5.02% in controls (P = 0.014).
Uterine artery volume blood flow and the fraction of maternal cardiac output distributed to the uteroplacental circulation are lower, and uterine vascular resistance (but not Uta PI) is higher, in women with previous caesarean section compared with the control group.
The rate of caesarean section is increasing worldwide, with associated adverse consequences. It is well documented that caesarean section increases the immediate postoperative risk of infection, haemorrhage and thromboembolism, compared with vaginal delivery.[1-3]
In subsequent pregnancies there is also an increased risk of uterine rupture, placenta praevia, placental abruption, and placenta accreta.[3-6] A recent meta-analysis reported that caesarean delivery is associated with a 23% increase in the risk of stillbirth, compared with vaginal delivery. Adverse neonatal outcomes, such as prematurity and low birthweight, are also more frequent among women with a history of previous caesarean section.[3, 5, 8]
It is not quite clear why caesarean section increases the risk of complications in subsequent pregnancies, and there are probably different mechanisms leading to adverse outcomes. One of the most plausible explanations is that the condition leading to caesarean section is responsible for the poor outcome; however, it must also be considered that the surgical procedure itself may add to the risk. Devascularisation of the uterine tissue and scarring may lead to impaired implantation, placental development, growth, and function. Blood flow to the uterus and placenta is likely to be affected as a result of fibrosis and scar formation. There are some studies reporting changes in uterine artery Doppler pattern in the postpartum period,[9-12] and a few of them have included women who had delivered by caesarean section.[10, 12] However, to our knowledge no studies have investigated volume blood flow to the uteroplacental circulation in subsequent pregnancy after a previous caesarean section.
We hypothesized that uteroplacental haemodynamics is altered in pregnant women with a history of previous caesarean section. Our objective was to compare the uterine artery volume blood flow (Quta), the uterine artery vascular resistance (Ruta), the uterine artery pulsatility index (Uta PI), and the fraction of maternal cardiac output (CO) distributed to the uteroplacental circulation in pregnant women with previous caesarean section, with their controls matched for age and parity.
This is part of a continuing cross-sectional study on maternal haemodynamics and fetoplacental circulation in normal and complicated pregnancies approved by the Regional Committee for Medical Research Ethics, North Norway. Women attending their routine antenatal scan at 17–20 weeks of gestation were informed about the study and asked to participate. Informed written consent was obtained from all participants.
In the database of our continuing study, we had 36 women with previous caesarean section among a total of 618 (363 primigravida and 255 parous women) participants. We were unable to find an appropriate match for one woman who was 44 years of age, and three women had incomplete data. A matching control was found for the remaining 32 women based on parity (exact) and age (±2 years). Of the 32 women with previous caesarean sections included in the analysis, 29 had one previous caesarean section, two had two previous caesarean sections, and one had three previous caesarean sections. The primary indications for previous caesarean section were: abnormal presentation (12); dystocia (9); hypertensive diseases of pregnancy (7); fetal distress (3); and maternal request (1).
The diameter measurement of the uterine artery was obtained just proximal to its apparent crossover with the external iliac artery on colour Doppler imaging. The power Doppler angiography mode was used to identify the uterine artery and delineate its borders. The scale of Doppler intensity was set at maximum and the gain was optimised to avoid any overestimation of the uterine artery diameter. Three measurements were taken and then averaged. The blood flow velocity was measured using pulsed-wave Doppler, according to the method described previously. The Quta was calculated as: Quta (ml/minute) = CSA (cm2) × TAV (cm/sec) × 60, where CSA is the cross-sectional area of the uterine artery and TAV is the time-averaged intensity-weighted mean velocity. The total uteroplacental blood flow was calculated as the sum of the left and right uterine artery volume blood flows. Impedance cardiography (ICG) (Philips, Böblingen, Germany) was used to measure maternal stroke volume, heart rate, and mean arterial blood pressure (MAP), and the CO and systemic vascular resistance (SVR) were automatically computed by the ICG machine. The body surface area (BSA) was calculated using the woman's height and weight. The cardiac index (CI) and systemic vascular resistance index (SVRI) were calculated as CO/BSA and SVR/BSA, respectively. The Ruta (mmHg/ml/minute) was calculated as: MAP (mmHg)/Quta (ml/minute). Uta PI was calculated as: (peak systolic velocity – end-diastolic velocity)/time-averaged maximum velocity.
The data are presented as means ± SDs, median (range) or number (%), as appropriate. Comparison between cases and their matched controls was performed using paired-samples Student's t–tests (spss statistics 20; IBM) for continuous variables and Fisher's exact test for categorical variables. The chi-square test was used to check differences in caesarean section rate between two groups in the current pregnancy. An independent-sample Student's t–test was used for subgroup analysis of cases with and without a previous history of hypertensive disorder of pregnancy, in order to compare differences in maternal haemodynamic parameters.
The baseline characteristics of the participants and neonatal outcomes are presented in Table 1. There were no statistically significant differences between the groups. In the current pregnancy, five women developed hypertensive complications among cases and four among controls (P = 1.0). The caesarean section rate among cases was 50% (16/32), compared with 6% (2/32) among controls (P = 0.0003). Among cases there was one unexplained intrauterine fetal death in gestational week 36 (the autopsy was inconclusive). This was a healthy woman who had a previous caesarean section at term for face presentation in labour.
Table 1. Baseline characteristics of the study population
Data are presented as n (%), median (range), or mean ± SD, as appropriate.
Parameters of maternal haemodynamics and uteroplacental blood flow are presented in Figure 1 and Table 2. CO, CI, SVR, and SVRI were similar among groups; however, Quta and the fraction of maternal CO distributed to the uteroplacental circulation were significantly lower in cases compared with controls. Ruta was higher in women with a history of previous caesarean section, although Uta PI was similar in both groups.
Table 2. Maternal haemodynamics and uterine artery blood flow
Mean arterial pressure (mmHg)
81.68 ± 9.50
80.23 ± 8.39
Cardiac output (l/minute)
6.43 ± 1.78
6.12 ± 1.15
Cardiac index (l/minute/m2)
3.44 ± 0.73
3.34 ± 0.47
Systemic vascular resistance (dyne seconds/cm5)
1017.81 ± 284.17
1020.63 ± 176.56
Systemic vascular resistance index (dyne seconds/m2/cm5)
Fraction of maternal cardiac output distributed to the utero-placental circulation (%)
5.75 ± 3.68
8.45 ± 5.02
We found significant differences in uterine artery blood flow between women with and without previous caesarean section at 22+0–23+6 weeks of gestation. The Quta and the fraction of maternal CO distributed to the uterus and placenta were significantly lower in women with a previous caesarean section, compared with controls. Although Uta PI was similar among groups, non-invasively measured Ruta was significantly higher among the cases.
Strength and limitations
To our knowledge this is the first study that describes volume blood flow to the uterus in pregnancies with a history of previous caesarean section. Although there are methodological limitations related to the non-invasive measurement of volume blood flow in small arteries, the Quta measurement using the Doppler technique has been validated. A relatively small sample size is another limitation; however, the cases and controls were carefully matched and the differences in outcome observed between groups are significant. Although we did not match cases and controls for previous obstetric history, a subgroup analysis of cases comparing women with a previous history of hypertensive disorders of pregnancy with normotensive women did not show significant differences in MAP, CO, Quta, Ruta, or Uta PI in the current pregnancy.
Our results indicate that uterine blood flow is reduced in subsequent pregnancy following a caesarean section. This could be one of the pathophysiological explanations for the increased frequency of adverse pregnancy outcomes reported in the literature among women with a history of previous caesarean section. The studies investigating uterine artery Doppler indices in the postpartum period report a significant increase in Uta PI during the first week after delivery, followed by a period of 4–6 weeks where the values are quite stable, before increasing again to reach non-pregnant levels by approximately 12 weeks after delivery.[9-11] Nakai et al. examined women longitudinally during the first 9 days postpartum after caesarean section and found no significant change in resistance index. There are no other studies that investigate uterine artery blood flow longitudinally following caesarean section, and uterine artery Doppler indices in subsequent pregnancy in women with a previous history of caesarean section have not been reported.
The Quta has been reported previously in uncomplicated pregnancies,[13, 14, 16, 17] and in pregnancies complicated by intrauterine growth restriction.[18, 19] Reduced Quta has been found in pregnancies with small for gestational age fetuses,[18, 19] but no correlation was found between Quta and birthweight in uncomplicated pregnancies with appropriate for gestational age fetuses. Quta is not the only determining factor for fetal growth, but when the volume blood flow is reduced below a certain level fetal growth is likely to be affected. The reduced uteroplacental blood flow found in women with a previous caesarean section could contribute to an increased risk of adverse pregnancy outcome, such as low birthweight.
The Uta PI is commonly used as a surrogate for resistance, but it does not always correlate with or reflect Ruta. Therefore, the measurement of volume blood flow and vascular resistance may provide additional information on subtle haemodynamic changes that may be present in pregnancies after a previous caesarean section. We did not find a significant difference in Uta PI values between cases and controls; however, Quta and Ruta differed between the groups. As the MAP was not significantly different between groups, it is plausible that the observed difference in Ruta arises from the difference in volume blood flow. In spite of similar CO and SVR values, the fraction of CO distributed to the uterus was less among pregnant women who had a previous caesarean section, indicating that the blood flow to the scarred uterus is altered.
Studies on outcomes of pregnancies following caesarean section compared with pregnancies subsequent to vaginal deliveries are difficult to interpret because of confounding by conditions leading to the first caesarean section. Hemminki et al. and Daltveit et al. found a higher risk of adverse fetal outcomes after the exclusion of women with obstetric problems, indicating that caesarean section in itself is likely to affect subsequent pregnancies.[21, 22] Carlsson Wallin et al. found an increased risk of low Apgar scores and perinatal death in pregnancies after a previous caesarean section, but concluded that pre-existing conditions were the most likely explanation. Smith et al. also found an increased risk of stillbirth in subsequent pregnancies following caesarean section, and this risk persisted after adjustment for obstetric history. As caesarean section increases the risk of placental problems such as placenta praevia and accreta, it is reasonable to hypothesise that caesarean section independently increases the risk of adverse outcomes in subsequent pregnancies. To study the isolated effect of caesarean section, pregnancies following caesarean section for non-medical reasons could be compared with pregnancies following normal vaginal birth, but we are not aware of any such studies. Our study indicates that uteroplacental blood flow is reduced in women with previous caesarean section, but further studies are needed to clarify the pathophysiological mechanisms leading to adverse pregnancy outcomes in women with previous caesarean section.
Uterine artery volume blood flow and the fraction of maternal cardiac output distributed to the uterine arteries are lower, and the uterine vascular resistance is higher, in pregnant women with a scarred uterus, compared with their controls.
Disclosure of interests
None to declare.
Contribution to authorship
KF and CW recruited the participants, collected data, and performed the ultrasound examinations. ÅV performed ICG examinations. GA conceived and designed the study. KF and GA performed statistical analysis, interpreted the results, and wrote the article. All authors approved the final version.
Details of ethics approval
The study was approved by the Regional Committee for Medical Research Ethics in North Norway (ref. no. 5.2005.1386).
This study was funded by the Regional Health Authority of North Norway.