Screening for Preeclampsia with Uterine Artery Dopplers
Abstract & Commentary
By John C. Hobbins, MD, Professor, Department of Obstetrics and Gynecology, University of Colorado School of Medicine, Aurora, is Associate Editor for OB/GYN Clinical Alert.
Dr. Hobbins reports no financial relationships relevant to this field of study.
Synposis: Data from a large trial in patients at low risk for preeclampsia show that uterine artery waveforms analysis has little value as a screening method for preeclampsia.
Source: Myatt L, et al. The utility of uterine artery Doppler velocimetry in prediction of preeclampsia in a low-risk population. Obstet Gynecol 2012;120:815-822.
Ever since stuart campbell introduced the concept of using ultrasound waveforms of the uterine artery to screen for preeclampsia (PE), single studies and meta-analyses have emerged to address its efficacy — with inconsistent results. Investigators in the NICHD perinatal network recently decided to study the screening potential of uterine artery Dopplers in a low-risk patient population, the results of which were published in the October 2012 issue of Obstetrics & Gynecology.1
In this study, 2188 second-trimester nulliparous patients who had no risk factors for PE were selected. All had uterine artery waveforms obtained prior to 21 weeks (average of 16 weeks). Interestingly, these patients were members of a cohort group that was formed as part of a randomized clinical trial to determine if antioxidant supplementation prevented PE. Uterine artery waveforms were obtained in standardized fashion with Doppler ultrasound, and results were analyzed based on the presence or absence of a notch in a diastolic portion of the waveform or the demonstration of increased resistance in the spiral arteries by a high average resistance index (RI) according to gestational age. PE was diagnosed clinically by standard clinical definitions.
PE occurred in 165 of the 2188 patients screened (7.5%) — a surprisingly high rate in this population. Using a threshold RI of the 75th percentile, the sensitivity for PE was 43% (95% confidence interval [CI], 35-51) with a specificity of 67% (95% CI, 65-69) for all forms of PE. However, an increased RI was more predictive of early onset PE (< 34 weeks) with a sensitivity of 78%, a specificity of 66%, and an increase in odds ratio (OR) of 6.9. The presence of notches did not correlate with the overall incidence of PE, but was correlated with early PE. The authors concluded that the method has "a poor sensitivity for prediction of PE overall in a well-characterized, low-risk nulliparous population." However, the authors indicated that the method did predict "trophoblastic invasion of spiral arteries of a magnitude that compromises utero-placental flow that gives early onset disease."
Commentary
Remodeling of the spiral arteries occurs in two waves. During the first wave (before about 14 weeks), the decidual (surface) portion of the spiral arteries in the basal plate of the placenta normally are invaded by the adjacent trophoblast. The second wave occurs in the second trimester and involves deeper invasion into the myometrial portion of the spiral arteries, resulting in a wide-open pathway that allows blood to flow freely into the inter-villus space. In PE, this invasion does not occur, especially the second wave of invasion. Since uterine artery waveforms represent an upstream reflection of the status of the spiral artery bed, they have the potential to provide information to predict PE. However, study results have been variable, depending on the population studied. The largest meta-analysis was undertaken by the World Health Organization and involved 43 studies (22 were in low-risk patients and 18 in high-risk patients).2 The analysis included 42,000 patients. The likelihood ratio for PE overall was 4.2 for a high uterine artery RI and 3.5 to 6.6 for a single notch or double notch, respectively. Most importantly, the likelihood ratio for early PE (< 34 weeks) was 7.94 for a single notch and 15.9 for bilateral notches. Another very recent small study in high-risk pregnancies dealt with using uterine artery waveforms in the first and second trimesters.3 It showed that an increased RI in the second trimester was more predictive of PE (overall) (36%; OR = 2.9) and early PE (18.2%; OR = 11.9) than first trimester waveforms. First trimester waveforms picked up only one case of PE (overall) and no cases of early PE if the second trimester Dopplers were normal later. However, if both first and second trimester waveforms were abnormal, the risk of PE (overall) was 57% (OR = 6.7) and for early PE was 28% (OR = 29.1).
From the data available, there is little evidence that screening a low-risk population with uterine artery Dopplers is useful. However, in high-risk patients, the evidence does suggest a reasonable ability of uterine artery waveform analysis to predict overall PE and a very good ability to predict severe, early PE.
Actually, the question of whether to screen for PE becomes moot if there is nothing we can do to prevent it in susceptible patients. Here there is evidence that increased surveillance and stepped-up prenatal care can improve outcome.4 Also, most importantly, some studies show that we may be able to decrease overall PE and early PE with carefully timed administration of low-dose aspirin (ASA).
In 2003, Coomarasamy et al published a meta-analysis of 14 studies using low-dose ASA to prevent PE in women with risk factors.5 They found that treating with low-dose ASA resulted in a statistically significant, but non-dramatic, drop in PE (OR = 0.86), perinatal death (OR = 0.79), and preterm birth (OR = 0.86), and a sizable increase in average infant weight of 215 g. The same group did a meta-analysis of five studies to evaluate the use of low-dose ASA in patients with abnormal uterine artery waveforms and found a reduction in overall PE by 50%.6
Perhaps the most significant contribution recently came from Montreal. Bujold analyzed outcome data from 22 randomized trials in which ASA randomization was begun after 16 weeks and compared these results with those from 12 studies in which randomization was begun before 16 weeks.7 There was a modest, non-significant, reduction in PE (0.81) in the > 16-week group compared with a significant reduction (0.47) in PE in those in whom ASA was begun < 16 weeks. Most importantly, there was a highly significant drop in early, severe PE (0.09) in the < 16 week group, suggesting that if one were to optimize the ability of ASA to prevent severe PE, it should be on board before 16 weeks.
Summary Points 1. Using uterine artery waveforms to screen for PE in a low-risk population is not effective. 2. All high-risk patients might benefit from low-dose ASA (84 mg OD) after 12 weeks. These would include:
3. If all of the above patients are on, or will be on, low-dose ASA, there would be no need to do uterine artery analysis in the first trimester. 4. Second trimester uterine artery waveforms may be useful in identifying those high-risk patients who have the greatest chance for PE, and the pattern of management can be adjusted accordingly. 5. If the second trimester uterine artery waveforms are normal, these patients can be managed in the usual low-risk fashion and given some reassurance that their risk of severe PE is very low. 6. There are no data to determine how long patients should be on ASA, but I have been discontinuing ASA after 34 weeks (some anesthesiologists are reluctant to offer epidurals to patients on low-dose ASA), and by that time it probably has done its job. |
Since ultrasound evaluation of uterine artery waveform analysis involves increased cost and expertise, would it not make sense to finesse the ultrasound and simply prescribe low-dose ASA for all high-risk patients? Since there is no evidence that ASA delivered in this dosage causes harm, I think an argument could be made for this approach. There are two benefits of the ultrasound step that could impact the outcome and the psyche of the patient. As mentioned above, there is evidence that enhanced perinatal care can benefit patients at risk for PE4 (and this would include those with abnormal uterine artery waveforms). Also, having a negative uterine artery result can have very beneficial spinoff. Every study so far has shown an extremely low chance of PE developing after a reassuring result (negative predictive values of 98-100%). In the vast majority of high-risk patients, the second trimester waveforms will be normal (80% in the most recent study).3 In these patients, we should be able to loosen up on our surveillance and give the patient some reasonable reassurance that she will not have PE, and most importantly, a severe form of it — an experience that usually leaves an indelible impression on every patient having previously experienced it.
References
- Myatt L, et al. The utility of uterine artery Doppler velocimetry in prediction of preeclampsia in a low-risk population. Obstet Gynecol 2012;120:815-822.
- Conde-Agudelo A, et al. World Health Organization systematic review of screening tests for preeclampsia. Obstet Gynecol 2004;104:1367-1391.
- Herraiz I, et al. Predictive value sequential models of uterine artery Doppler in pregnancies at high risk for pre-eclampsia. Ultrasound Obstet Gynecol 2012;40:68-74.
- Menzies J, et al. Instituting surveillance guidelines and adverse outcomes in preeclampsia. Obstet Gynecol 2007;110:121-127.
- Coomarasamy A, et al. Aspirin for prevention of preeclampsia in women with historical risk factors: A systematic review. Obstet Gynecol 2003;101:1319-1332.
- Coomarasamy A, et al. Aspirin for prevention of preeclampsia in women with abnormal uterine artery Doppler: A meta-analysis. Obstet Gynecol 2001;98(8Pt1):861-866.
- Bujold E, et al. Prevention of preeclampsia and intrauterine growth restriction with aspirin started in early pregnancy: A meta-analysis. Obstet Gynecol 2010;116:402-414.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content