Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 1.
Published in final edited form as: Ultrasound Obstet Gynecol. 2023 Mar;61(3):325–332. doi: 10.1002/uog.26100

Cost-effectiveness of ultrasound before non-invasive prenatal screening for fetal aneuploidy

Rentabilidad de la ecografía antes del cribado prenatal no invasivo de la aneuploidía fetal

胎儿非整倍体无创产前筛査前进行超声检査的成本效益

A N BATTARBEE 1,2, N L VORA 3, E E HARDISTY 3, D M STAMILIO 4
PMCID: PMC10577524  NIHMSID: NIHMS1935033  PMID: 36273429

Abstract

Objective

To determine the cost-effectiveness of first-trimester ultrasound before fetal aneuploidy screening with cell-free DNA (cfDNA) compared with screening by cfDNA alone.

Methods

A decision analytic model was constructed for 400 000 pregnant individuals with advanced maternal age who desired first-trimester aneuploidy screening with cfDNA in the USA, to compare two screening strategies: (1) cfDNA only and (2) ultrasound performed within 4 weeks before cfDNA. Input parameters included probability of fetal aneuploidy, cfDNA performance, desire for diagnostic testing, pregnancy outcomes, and pregnancy and lifetime costs and utilities. The primary outcome measure was the incremental cost-effectiveness ratio (ICER), in terms of cost in 2020 US dollars (USD) per quality-adjusted life year (QALY) gained. Secondary outcomes included procedure-related loss, pregnancy termination, live birth with aneuploidy, live birth with structural anomaly and stillbirth. Discounting was performed at 3% per year with an estimated maternal lifespan of 81 years starting at the age of 35 years. One-way, multiway and Monte Carlo probabilistic sensitivity analyses were performed. All base-case estimates and ranges of uncertainty were derived from the literature. The willingness-to-pay threshold was set at 100 000 USD per QALY.

Results

In the base-case analysis, ultrasound before cfDNA screening was more cost-effective than cfDNA screening without pretest ultrasound, with an ICER of 12 588 USD and higher net monetary benefit (24 241 vs 20 466). The strategy involving ultrasound before cfDNA was more costly by 544 USD but also more effective (by 0.04 QALY) compared with cfDNA alone. Base-case results were robust in sensitivity analyses with the strategy involving ultrasound before cfDNA always remaining the most cost-effective approach with the highest net monetary benefit.

Conclusion

First-trimester ultrasound before cfDNA is a more cost-effective strategy for non-invasive prenatal aneuploidy screening compared with cfDNA alone. © 2022 International Society of Ultrasound in Obstetrics and Gynecology.

Keywords: aneuploidy, cell-free DNA, cost-effectiveness, non-invasive, prenatal, screening, ultrasound

INTRODUCTION

All pregnant individuals should be offered the option of prenatal genetic screening and diagnostic testing1,2. Historically, routine first-trimester screening for fetal chromosomal abnormalities or aneuploidy has been performed by a combination of maternal serum analyte testing and ultrasound to measure nuchal translucency (NT) thickness2. In 2011, non-invasive prenatal screening with techniques to analyze cfDNA fragments was introduced in the USA as another screening strategy to detect fetal aneuploidy3. Since then, cfDNA analysis has been validated as an acceptable and superior method of fetal aneuploidy screening compared with standard first-trimester screening, particularly for the common aneuploidies, such as trisomy 21, trisomy 18, trisomy 13 and sex chromosome abnormalities2.

Beyond NT measurement, the standard first-trimester ultrasound examination is used to confirm the location, viability and gestational age of a pregnancy, as well as evaluate for multiple gestation and determine chorionicity. In contrast to the more traditional first-trimester screening approach, cfDNA analysis does not require an ultrasound assessment in order to provide a result. Clinical practice varies in terms of the type and timing of ultrasound performed for pregnant people who desire cfDNA screening. However, there are certain situations in which a baseline ultrasound would be clinically useful before cfDNA screening, as incorrect dating, fetal demise, multifetal gestation or major structural anomaly could alter the appropriateness, timing or interpretation of test results4–6. Nevertheless, it is uncertain whether the added cost of ultrasound results in a more cost-effective screening program as the prevalence of abnormal or unexpected ultrasound findings is relatively low4,5. Thus, the objective of this study was to determine the cost-effectiveness of first-trimester ultrasound before non-invasive prenatal screening with cfDNA. We hypothesized that first-trimester ultrasound before cfDNA would be more costly but more cost-effective compared with cfDNA alone.

METHODS

This was a cost-effectiveness analysis comparing two screening strategies for fetal aneuploidy with cfDNA: first-trimester ultrasound before cfDNA screening vs cfDNA screening alone. For the base-case decision analytic model, we assumed a population of pregnant individuals with advanced maternal age (AMA) since most of the published data for model estimates are derived from AMA populations, historically the first population that was offered cfDNA screening. Ultrasound was defined as a basic evaluation of the fetus in the first trimester within 4 weeks prior to cfDNA screening, that confirmed gestational dating and evaluated viability, presence of a multifetal gestation, NT thickness and absence or presence of major fetal anomaly. CfDNA screening included basic screening for the most common fetal trisomies involving chromosomes 21, 18 and 13 and sex chromosome abnormalities. In order to compare these two strategies, we developed a decision tree from the societal perspective, using relevant estimates for cost, utilities and pregnancy outcomes.

Based on the USA 2018 annual birth census7 and an approximate 15% rate of AMA, we assumed a theoretical cohort of 400000 AMA pregnant people who desired first-trimester screening for fetal aneuploidy with non-invasive cfDNA. The effectiveness of each strategy (ultrasound before cfDNA screening vs cfDNA screening only) was assessed in terms of quality-adjusted life years (QALYs) and average health costs incurred over a person’s lifetime. Probability estimates were obtained from the literature, using the most rigorous data available for base-case estimates in the following hierarchical order: meta-analysis, randomized controlled trial, cohort study and case–control study (Table 1). Additional studies were used to inform ranges for sensitivity analyses. Utility estimates were obtained from published valuations of health states. Costs were derived from either the literature, the Healthcare Cost and Utilization Project (HCUP) (https://www.hcup-us.ahrq.gov/) or third party insurance reimbursement, and are presented in 2020 US dollars (USD). Cost estimates from years preceding 2020 were adjusted using the consumer price index calculator on the U.S. Bureau of Labor Statistics website (https://www.bls.gov/data/inflation_calculator.htm). Base-case estimates for cost for delivery, first- or second-trimester termination of pregnancy care (dilation and evacuation) and prenatal care were derived from HCUP data, and ranges for sensitivity analyses were established based on highest and lowest local (North Carolina) insurance reimbursement values. In calculating average cost estimates for delivery, we used a weighted average by delivery mode and assumed a 30% Cesarean section rate and 70% vaginal delivery rate. Delivery costs were specific for live birth vs stillbirth. Cost estimates for all other procedures, ultrasound scans and genetic tests were based on average local insurance reimbursement for Medicaid, Tricare and private insurance, and ranges for sensitivity analyses were established based on highest and lowest insurance reimbursement values (data for reimbursement-based cost estimates are available upon request from the corresponding author). Estimates for costs included procedural and professional fee costs. Lifetime costs of Down syndrome and fetal anomalies were derived from the literature8,9. We assumed a maternal age of 35 years upon inclusion to the theoretical cohort and we discounted costs and utilities at 3% per year over an estimated maternal lifespan of 81 years10.

Table 1.

Model input parameters used in base-case analysis of two screening strategies for fetal aneuploidy using cell-free DNA (cfDNA)

Model parameter Base case Range Reference
Probability of pregnancy complication
 First-trimester miscarriage 0.065 0.0558–0.0763 4
 Major fetal structural anomaly 0.022 0.0159–0.0297 15
 Not appropriate candidate for cfDNA 0.0718 0.0616–0.0832 4
 Incorrect dating* 0.773 0.7010–0.8349 4
 Procedure-related loss with CVS 0.0139 0.0076–0.0202 16
 Procedure-related loss with amniocentesis 0.0091 0.0073–0.0109 16
 Fetal abnormality on second-trimester ultrasound with aneuploidy 0.88 0.50–0.95 17, 18
 Elective TOP 0.742 0.614–0.933 19–21
 Aneuploid stillbirth 0.049 0.025–0.074 22
 Euploid stillbirth 0.006 0.003–0.015 23, 24
Probability of fetal aneuploidy
 General population 0.0028 0.0007–0.0286 25
 First-trimester fetal demise 0.594 0.501–0.703 26–28
 Major fetal anomaly 0.201 0.071–0.520 29
 Multiple gestation 0.0057 0.0014–0.0571 25
Probability of failed cfDNA test (no result)
 Euploid fetus 0.035 0.029–0.040 30
 Aneuploid fetus 0.044 0.019–0.152 30
 First-trimester fetal demise 0.240 0.1940–0.5763 31
 Incorrect dating 0.274 0.1872–0.3748 32
 Multiple gestation 0.132 0.0623–0.2364 33
Test characteristics when result returned
 Sensitivity of cfDNA, singleton gestation 0.997 0.658–0.999 34
 Specificity of cfDNA, singleton gestation 0.96 0.93–0.98 34
 Sensitivity of cfDNA, multiple gestation 0.994 0.316–0.998 34
 Specificity of cfDNA, multiple gestation 0.92 0.86–0.96 34
 Sensitivity of serum aneuploidy screening 0.893 0.797–0.947 35
 Specificity of serum aneuploidy screening 0.946 0.933–0.957 35
Probability of patient desire for invasive testing
 After fetal anomaly seen on ultrasound 0.650 0.325–0.975 4
 After positive cfDNA result 0.86 0.60–0.96 36
 After failed cfDNA test (no result) 0.86 0.60–0.96 36
 After positive serum aneuploidy screening 0.624 0.472–0.776 36, 37
 After multiple gestation seen on ultrasound 0.10 0.03–0.19 36
Cost See Methods
 cfDNA test (USD) 450.00 400.00–500.00
 First-trimester ultrasound (USD) 138.91 55.00–247.00
 Second-trimester ultrasound (USD) 217.44 96.59–403.08
 Serum aneuploidy screening (USD) 14.80 9.32–18.03
 CVS (USD) 242.00 111.00–393.00
 Amniocentesis (USD) 238.00 116.00–391.00
 Prenatal care (USD) 927.29 596.00–1545.00
 D&C for first-trimester loss or TOP (USD) 381.24 158.00–519.00
 D&C for second-trimester loss or TOP (USD) 477.56 62.50–738.76
 Delivery of live fetus (USD) 6234.49 3117.00–9351.00
 Stillbirth (USD) 4943.97 2472.00–7416.00
 Life of individual with Down syndrome (USD) 914158.00 457079.00–1371237.00 8
 Life of individual with structural anomaly (USD)† 535868.00 295 075.00–852 147.00 9
Utilities for pregnancy outcomes
 Miscarriage 0.880 0.702–1.000 38
 Procedure-related pregnancy loss 0.744 0.459–1.000 38
 TOP 0.771 0.487–1.000 38
 Uncomplicated euploid live birth 1.00 0.93–1.000 38, 39
 Euploid live birth with fetal anomaly 0.75 0.45–0.83 38
 Euploid live birth with FP cfDNA 0.928 0.778–1.000 38
 Aneuploid live birth 0.480 0.175–0.785 38
 Aneuploid live birth after FN cfDNA (disutility) 0.30 0.09–0.43 39
Open in a new tab
*

As reason for not being appropriate candidate for cfDNA.

†

Representative anomalies included neural tube defect, cardiac defect and omphalocele. CVS, chorionic villus sampling; D&C, dilation and curettage; FN, false-negative; FP, false-positive; TOP, termination of pregnancy; USD, US dollar.

We performed a cost-effectiveness analysis and reported incremental cost-effectiveness ratio (ICER) and net monetary benefit (NMB). Our primary outcome was the ICER, which is the cost per 1 additional QALY gained with one strategy compared to the other. For this study, we set the willingness-to-pay threshold at 100 000 USD per QALY gained, meaning that a strategy was cost-effective if it had an ICER < 100 000 USD. NMB for each strategy is also reported since this value measures the cost-effectiveness of a strategy given a certain willingness-to-pay, with the highest NMB representing the most cost-efficient strategy. Secondary outcomes included procedure-related loss, pregnancy termination, live birth with aneuploidy, live birth with structural anomaly and stillbirth.

In our decision analytic model, each pregnant person could follow one of two strategies: ultrasound before cfDNA screening or cfDNA screening alone (Figure 1). These strategies are described as follows. In the strategy comprising ultrasound before cfDNA screening, pregnancies were assumed to have undergone an ultrasound scan within 4 weeks preceding the genetic screen4. We also assumed initiation of prenatal care in the first trimester. An appropriate candidate for cfDNA screening was defined as having a singleton live pregnancy without an anomaly (e.g. increased NT thickness > 3.0 mm, cystic hygroma or neural tube defect)1,4. Patients with increased fetal NT or structural anomalies were not considered to be appropriate candidates for cfDNA screening in this model as they should be counseled about increased risk for aneuploidy and recommended to have diagnostic testing. At the next branch point in the model, appropriate candidates undergo cfDNA screening and, if the result is positive (high risk) or non-informative due to low fetal fraction, are offered prenatal genetic diagnostic testing with chorionic villus sampling (CVS) or, if CVS is declined, amniocentesis performed at a later gestational age when appropriate. Pregnant people who elect to have a diagnostic test have a procedure-specific risk for complication that leads to pregnancy loss. Regardless of the cfDNA screen result or decision for diagnostic testing, pregnant individuals with ongoing pregnancy have chances for a live birth or stillbirth with or without aneuploidy. Pregnancies with fetal demise, multiple gestation or fetal anomaly diagnosed on ultrasound do not undergo cfDNA screening and are redirected for correct management based on the specific ultrasound diagnosis. This includes either management of a missed miscarriage, delayed cfDNA screening for incorrect pregnancy dating, offering of prenatal genetic diagnostic testing for an anomalous fetus or alternative prenatal genetic screening for multiple gestations (e.g. maternal multiple serum marker screening). For cfDNA screening candidates with a normal fetal ultrasound, once a person selects a screening or diagnostic test, they follow one of the pathways described above.

Figure 1.

Figure 1

Open in a new tab

Abbreviated decision tree of screening strategy encompassing first-trimester ultrasound (US) examination within 4 weeks prior to cell-free DNA (cfDNA) aneuploidy screening. This is a simplified and truncated decision tree demonstrating key decision nodes, risk–benefit trade-offs and outcomes. It is not representative of the full decision model or the entire breadth of probability pathways. The decision node (□) choosing between a screening strategy of fetal US prior to cfDNA fetal aneuploidy screening and a strategy of cfDNA screening without antecedent US is identical to that of the full model. Amnio, amniocentesis; CVS, chorionic villus sampling; FF, fetal fraction.

In the strategy comprising cfDNA screening only, pregnancies were assumed to not have had a fetal ultrasound prior to screening. In this strategy, all patients undergo cfDNA screening given that ultrasound information is not available regarding gestational age, fetal viability, fetal anatomy or gestation type (singleton vs multiple). Patients with incorrect pregnancy dating that resulted in testing too early in gestation are offered cfDNA screening later at the correct gestational age (> 9.5 weeks). Pregnant individuals who are later found to have a fetal anomaly on second-trimester ultrasound and a negative cfDNA screen are offered genetic amniocentesis. Pregnant people later found to have a multiple gestation on fetal anatomy ultrasound and a negative cfDNA screen are offered genetic amniocentesis, as individuals with a multiple gestation have a risk of non-informative cfDNA. Similar to the other strategy, once a person selects a screening or diagnostic test, they follow one of the pathways described above.

One-way sensitivity analyses were performed varying influential variables, such as the probability that a patient was not an appropriate candidate for cfDNA screening, the probability that a patient had a failed cfDNA test due to incorrect dating, and the lifetime cost and utility of a live birth with trisomy 21, among others. Select variables were also used for two-way and three-way sensitivity analyses. The most influential variables for sensitivity analyses were identified via Tornado diagrams of ICER and NMB. Monte Carlo probabilistic sensitivity analysis was performed with 1000 iterations to assess the robustness of the base-case results. The Monte Carlo sensitivity analysis varies randomly all model variables across the specified ranges of uncertainty for a series of iterations, and generates an estimate of a 95% CI for cost-effectiveness and incremental cost-effectiveness with illustrative scatterplots. Ranges of uncertainty around probability and utility point estimates were based on published 95% CI when available from a single study or meta-analysis or by a range of values found in the literature. In the rare event of limited data, ranges of uncertainty were defined from 50% below to 50% above the point estimate. The underlying distribution (i.e. normal, triangular, beta estimate of normal, or uniform) for the range of uncertainty for each variable was selected based on the data source, the observed distribution and data type (probability, cost or utility). All computations were performed using TreeAge Pro 2017 (Tree Age Inc, Williamstown, MA, USA). This study was exempt from Institutional Review Board review.

RESULTS

In a theoretical cohort of 400 000 pregnancies undergoing first-trimester cfDNA screening for aneuploidy, the base-case analysis indicated that cfDNA screening only (i.e. without preceding ultrasound) would result in 38 more procedure-related losses, 87 more live births with aneuploidy, 35 more live births with a fetal structural anomaly, four more stillbirths and 93 fewer pregnancy terminations, compared with the strategy encompassing ultrasound before cfDNA screening (Table 2). Ultrasound before cfDNA screening was more costly by 543.68 USD (average discounted cost of 5215.95 USD vs 4672.27 USD) but was more effective (average QALY of 0.295 vs 0.251) compared with cfDNA screening alone (Table 3). Both strategies had favorable cost-effectiveness ratios at 17 706.99 USD per QALY for ultrasound before cfDNA screening and 18 586.28 USD per QALY for cfDNA only. In the base-case analysis, ultrasound before cfDNA screening was more cost-effective than cfDNA alone with an ICER of 12 588.83 and higher NMB of 24 241.03.

Table 2.

Outcomes in theoretical cohort of 400 000 pregnant individuals with advanced maternal age desiring cell-free DNA (cfDNA) aneuploidy screening, based on screening strategy involving first-trimester ultrasound (US) before cfDNA and strategy encompassing cfDNA only

Screening strategy Procedure-related loss Pregnancy termination LB with aneuploidy LB with structural anomaly Stillbirth
cfDNA only 473 1696 922 6501 2275
US before cfDNA 435 1789 835 6466 2271
Difference between cfDNA only and US before cfDNA 38 −93 87 35 4
Open in a new tab

Data are given as n. LB, live birth.

Table 3.

Cost-effectiveness analysis of first-trimester cell-free DNA (cfDNA) screening for aneuploidy, based on screening strategy involving first-trimester ultrasound before cfDNA and strategy encompassing cfDNA only

Strategy Cost (USD) Effectiveness (QALY) Cost-effectiveness ratio (USD/QALY) ICER (USD/QALY gained) NMB*
Ultrasound before cfDNA 5215.95 0.295 17706.99 12588.83 24 241.03
cfDNA only 4672.27 0.251 18 586.28 — 20 466.01
Open in a new tab
*

Assuming willingness-to-pay of 100 000 US dollars (USD) per quality-adjusted life year (QALY). ICER, incremental cost-effectiveness ratio; NMB, net monetary benefit.

Sensitivity analyses were performed to assess the robustness of the base-case results and identify the extent to which the results were affected by changes in key model parameters using the plausible range of uncertainty for each variable (Table 1). One-way sensitivities were evaluated with variability in NMB and in ICER displayed for each variable in a tornado plot to identify the most influential variables on the results of the base-case scenario (Figure 2; ICER tornado plot available upon request). The tornado plots indicated that a large majority of uncertainty was attributable to variables such as the lifetime cost of structural defect or aneuploidy, the cost of live birth, the utility of miscarriage and euploid live birth, and the probability of an aneuploid fetus in the absence of a fetal anomaly on ultrasound (Figure 2). There were no instances in which cfDNA alone was the preferred strategy in these one-way analyses. Combinations of the select influential variables listed above and depicted in the tornado plot were also evaluated in two-way and three-way sensitivity analyses and again the base-case results were not altered, with ultrasound prior to cfDNA screening being consistently the more cost-effective strategy.

Figure 2.

Figure 2

Open in a new tab

Tornado diagram of one-way sensitivity analyses showing impact of selected variables on net monetary benefit (NMB) of screening strategy encompassing first-trimester ultrasound (US) prior to cell-free DNA (cfDNA) aneuploidy screening, in theoretical cohort of 400 000 pregnancies. Values in parentheses are uncertainty ranges. Willingness-to-pay threshold was 100 000 US dollars. Dx, prenatal diagnostic testing; EV, expected value (i.e. base-case value) delineated by vertical line; Prob, probability; TOP, termination of pregnancy.

Monte Carlo simulation of 1000 trials was performed in which all model variables were randomly varied across uncertainty ranges with a willingness-to-pay threshold of 100 000 USD per QALY (Figure 3). Figure 3 shows the incremental cost-effectiveness scatterplot generated from the Monte Carlo sensitivity analysis and the ellipse represents the estimated 95% CI for the ICER. As all of the datapoints in the scatterplot are in the upper right quadrant (quadrant I) or lower right quadrant (quadrant IV), the graph demonstrates that the strategy of routine first-trimester ultrasound before non-invasive fetal aneuploidy screening is always more effective than a strategy that does not incorporate antecedent fetal ultrasound, and in all but one iteration it is more costly. The ICER is always below the willingness-to-pay threshold; thus, the strategy incorporating fetal ultrasound prior to cfDNA screening is always more cost-effective than cfDNA alone and is always cost-effective given a willingness-to-pay threshold of 100 000 USD. This was also true when a willingness-to-pay threshold of 50 000 USD was used (data not shown).

Figure 3.

Figure 3

Open in a new tab

Monte Carlo simulation scatterplot showing incremental cost-effectiveness of screening strategy encompassing first-trimester ultrasound prior to cell-free DNA (cfDNA) aneuploidy screening vs cfDNA screening alone. Ellipse represents estimated 95% CI for incremental cost-effectiveness ratio (ICER). ICER is below willingness-to-pay (WTP) threshold of 100 000 US dollars (USD).

DISCUSSION

In this cost-effectiveness analysis of fetal aneuploidy screening, we found that, although both strategies were cost-effective using a willingness-to-pay threshold of 100 000 USD, routine first-trimester ultrasound before non-invasive prenatal screening with cfDNA is a more cost-effective strategy compared to cfDNA screening without preceding ultrasound. Importantly, the base-case results favoring ultrasound before non-invasive prenatal screening remained robust in sensitivity analyses and were not sensitive to uncertainty in any variable or combination of variables including those that differ by maternal age. The preferred strategy of ultrasound before cfDNA screening was consistently more cost-effective than cfDNA screening alone across all univariable and multivariable sensitivity analyses.

In contrast to first-trimester aneuploidy screening which includes measurement of maternal serum analytes and NT thickness on ultrasound, cfDNA screening does not require a preceding ultrasound examination to obtain a result. However, first-trimester ultrasound can be useful in confirming that the timing of cfDNA screening is appropriate and that the results are accurately interpreted. Additionally, it remains a key component of routine prenatal care. In a retrospective study of 41 884 patients who had prenatal care at a tertiary academic institution and were exposed to one of three serially implemented screening strategies from January 2010 to March 2018, patients who were offered cfDNA screening were almost 70% less likely to undergo any first-trimester ultrasound examination compared with patients who were offered either first-trimester aneuploidy screening alone or parallel testing with first-trimester aneuploidy screening and cfDNA screening (40.9% vs 68.7%; OR, 0.32 (95% CI, 0.24–0.40))11. This decrease in ultrasound uptake was consistent for all types of ultrasound examinations including for dating, viability or assessment of nuchal translucency.

The American College of Obstetricians and Gynecologists and Society for Maternal–Fetal Medicine state that formal NT measurement for aneuploidy risk is not necessary but an ultrasound may be useful prior to cfDNA screening as previous studies have demonstrated that certain findings on ultrasound may affect the timing, appropriateness or interpretation of cfDNA2,12. For example, in a retrospective study of 2337 patients with AMA in a single tertiary center in the USA, 377 (16.1%) patients had at least one ultrasound finding at the time of cfDNA screening that would have altered the provider’s counseling regarding the aneuploidy screening or testing strategy4. These included major structural anomaly, incorrect dating, multiple gestation and non-viable pregnancy. Similarly, an Australian study of 6207 patients found that 598 (9.6%) had a finding on ultrasound before cfDNA screening that had the potential to change management13. While an abnormal cfDNA screening result would prompt an ultrasound assessment, which will identify these findings that would have been detected on prescreening ultrasound, this could result in delays in diagnostic testing and may increase the overall cost if cfDNA screening needs to be repeated or was not indicated initially.

Our study has several strengths. To our knowledge, this is the first study to evaluate the cost-effectiveness of ultrasound before cfDNA. We utilized a societal perspective with evidence-based estimates of probabilities, costs and outcome values (utility) and performed a thorough review of the literature to derive variable estimates and ranges of uncertainty. This is a comprehensive decision analytic model that incorporates the relevant clinical scenarios and decisions required in choosing a strategy of cfDNA fetal aneuploidy screening. This model organizes and makes apparent a complex series of trade-offs among competing risks and benefits, which is necessary to select an optimal healthcare strategy. Moreover, the results were not altered following rigorous multivariable sensitivity analyses; therefore, our findings are likely to be generalizable to the USA obstetric population.

Our study is not without limitations. The cost-benefit ratio may change as cfDNA becomes less costly, which could affect the cost-effectiveness model. While we created a detailed model to organize the complex series of trade-offs, we were limited in our ability to capture the endless spectrum of complexities in clinical medicine. Moreover, we did not consider the impact of a detailed early anatomic survey as this is not performed routinely in all practices at present. Because of the initial restriction of cfDNA screening to just high-risk patients (e.g. AMA or with prior fetal aneuploidy) in clinical practice and availability of published research data, many base-case estimates of risk were based on studies of these high-risk populations and, therefore, it may be argued that our findings are not representative of the general obstetric population. However, we used wide ranges of uncertainty for probability estimates in the sensitivity analyses, and the model results were not altered. For example, even when simultaneously adjusting estimates of probabilities for miscarriage, fetal structural anomalies and fetal aneuploidy to the lowest values in multivariable and probabilistic sensitivity analyses, which would represent the younger maternal population, results remained unchanged, with ultrasound before cfDNA prenatal screening being consistently the more cost-effective strategy. Thus, our results are very likely applicable to the general obstetric population that is seeking screening for fetal aneuploidy irrespective of maternal age14. Lastly, prenatal genetic screening is evolving rapidly. When this decision model was first conceived, cfDNA screening was not recommended for multiple gestations. Although some cfDNA test platforms are now available for twin gestations, interpretation of results in twin pregnancies is still challenging and test performance differences in multiple gestations, such as a higher failure rate, may impact patient decision in selecting a test40. Therefore, we believe that the more recent availability of cfDNA fetal aneuploidy screening for twin gestations does not invalidate our decision model and that a pretest ultrasound demonstrating a multiple gestation is still informative for clinical decision-making and test interpretation.

In summary, a strategy involving routine first-trimester ultrasound before non-invasive fetal aneuploidy screening with cfDNA is more cost-effective than a strategy encompassing cfDNA without antecedent fetal ultrasound. While the base-case model assumed a population of AMA desiring first-trimester aneuploidy screening, our results were robust in sensitivity analyses that approximated both high-risk and low-risk maternal populations. Thus, we recommend performing routinely fetal ultrasound prior to cfDNA fetal aneuploidy screening to prevent unindicated testing, improve interpretation of results, and minimize delay in care caused by inappropriate screening.

CONTRIBUTION.

What are the novel findings of this work?

In this cost-effectiveness analysis of a screening strategy involving first-trimester ultrasound prior to cell-free DNA (cfDNA) vs cfDNA screening alone, we found that both strategies were cost-effective using a willingness-to-pay threshold of 100000 US dollars, however, routine first-trimester ultrasound before cfDNA was more cost-effective.

What are the clinical implications of this work?

We recommend performing routinely a fetal ultrasound scan prior to cfDNA fetal aneuploidy screening in order to prevent unindicated testing, improve interpretation of results and minimize delay in healthcare caused by inappropriate screening.

ACKNOWLEDGMENT

A.N.B. was supported by K23HD103875 from the NICHD.

REFERENCES

  • 1.Screening for fetal chromosomal abnormalities. ACOG Practice Bulletin No. 226. American College of Obstetricians and Gynecologists. Obstet Gynecol 2020; 136: e48–69. [DOI] [PubMed] [Google Scholar]
  • 2.Rose NC, Kaimal AJ, Dugoff L, Norton ME. Screening for Fetal Chromosomal Abnormalities: ACOG Practice Bulletin, Number 226. Obstet Gynecol 2020; 136: e48–e69. [DOI] [PubMed] [Google Scholar]
  • 3.Norton ME, Jacobsson B, Swamy GK, Laurent LC, Ranzini AC, Brar H, Tomlinson MW, Pereira L, Spitz JL, Hollemon D, Cuckle H, Musci TJ, Wapner RJ. Cell-free DNA Analysis for Noninvasive Examination of Trisomy. N Engl J Med 2015; 372: 1589–1597. [DOI] [PubMed] [Google Scholar]
  • 4.Vora NL, Robinson S, Hardisty EE, Stamilio DM. Utility of ultrasound examination at 10–14 weeks prior to cell-free DNA screening for fetal aneuploidy. Ultrasound Obstet Gynecol 2017; 49: 465–469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Reiff ES, Little SE, Dobson L, Wilkins-Haug L, Bromley B. What is the role of the 11- to 14-week ultrasound in women with negative cell-free DNA screening for aneuploidy? Prenat Diagn 2016; 36: 260–265. [DOI] [PubMed] [Google Scholar]
  • 6.Reimers RM, Mason-Suares H, Little SE, Bromley B, Reiff ES, Dobson LJ, Wilkins-Haug L. When ultrasound anomalies are present: An estimation of the frequency of chromosome abnormalities not detected by cell-free DNA aneuploidy screens. Prenat Diagn 2018; 38: 250–257. [DOI] [PubMed] [Google Scholar]
  • 7.Martin JA, Hamilton BE, Osterman MJK, Driscoll AK. Births: Final data for 2018. Natl Vital Stat Reports 2019; 68: 1980–2018. [PubMed] [Google Scholar]
  • 8.Kageleiry A, Samuelson D, Duh MS, Lefebvre P, Campbell J, Skotko BG. Out-of-pocket medical costs and third-party healthcare costs for children with Down syndrome. Am J Med Genet A 2017; 173: 627–637. [DOI] [PubMed] [Google Scholar]
  • 9.Waitzman NJ, Romano PS, Scheffler RM. Estimates of the economic costs of birth defects. Inquiry 1994; 31: 188–205. [PubMed] [Google Scholar]
  • 10.Arias E, Xu J. United States Life Tables, 2018. Natl Vital Stat Rep 2020; 69: 1–45. [PubMed] [Google Scholar]
  • 11.Doulaveris G, Igel CM, Estrada Trejo F, Fiorentino D, Rabin-Havt S, Klugman S, Dar P. Impact of introducing cell-free DNA screening into clinical care on first trimester ultrasound. Prenat Diagn 2022; 42: 254–259. [DOI] [PubMed] [Google Scholar]
  • 12.Society for Maternal-Fetal Medicine (SMFM).Norton ME, Biggio JR, Kuller JA, Blackwell SC. The role of ultrasound in women who undergo cell-free DNA screening. Am J Obstet Gynecol 2017; 216: B2–B7. [DOI] [PubMed] [Google Scholar]
  • 13.Brown I, Fernando S, Menezes M, da Silva Costa F, Ramkrishna J, Meagher S, Rolnik DL. The importance of ultrasound preceding cell-free DNA screening for fetal chromosomal abnormalities. Prenat Diagn 2020; 40: 1439–1446. [DOI] [PubMed] [Google Scholar]
  • 14.Rose NC, Barrie ES, Malinowski J, Jenkins GP, McClain MR, LaGrave D, Leung ML; ACMG Professional Practice and Guidelines Committee. Systematic evidence-based review: The application of noninvasive prenatal screening using cell-free DNA in general-risk pregnancies. Genet Med 2022; 24: 1379–1391. [DOI] [PubMed] [Google Scholar]
  • 15.Rao RR, Valderramos SG, Silverman NS, Han CS, Platt LD. The value of the first trimester ultrasound in the era of cell free DNA screening. Prenat Diagn 2016; 36: 1192–1198. [DOI] [PubMed] [Google Scholar]
  • 16.Salomon LJ, Sotiriadis A, Wulff CB, Odibo A, Akolekar R. Risk of miscarriage following amniocentesis or chorionic villus sampling: systematic review of literature and updated meta-analysis. Ultrasound Obstet Gynecol 2019; 54: 442–451. [DOI] [PubMed] [Google Scholar]
  • 17.Egan JFX, Malakh L, Turner GW, Markenson G, Wax JR, Benn PA. Role of ultrasound for Down syndrome screening in advanced maternal age. Am J Obstet Gynecol 2001; 185: 1028–1031. [DOI] [PubMed] [Google Scholar]
  • 18.Anderson NG, Luehr B, Ng R. Normal obstetric ultrasound reduces the risk of Down syndrome in fetuses of older mothers. Australas Radiol 2006; 50: 429–434. [DOI] [PubMed] [Google Scholar]
  • 19.Natoli JL, Ackerman DL, Mcdermott S, Edwards JG. Prenatal diagnosis of Down syndrome: A systematic review of termination rates (1995–2011). Prenat Diagn 2012; 32: 142–153. [DOI] [PubMed] [Google Scholar]
  • 20.Kazerouni NN, Cunier B, Malm L, Riggle S, Hodgkinson C, Smith S, Tempelis C, Lorey F, Davis A, Jelliffe-Pawlowski L, Walton-Haynes L, Roberson M. Triple-marker prenatal screening program for chromosomal defects. Obstet Gynecol 2009; 114: 50–58. [DOI] [PubMed] [Google Scholar]
  • 21.Palomaki GE, Kloza EM, Haddow JE, Williams J, Knight GJ. Patient and health professional acceptance of integrated serum screening for Down syndrome. Semin Perinatol 2005; 29: 247–251. [DOI] [PubMed] [Google Scholar]
  • 22.Groen H, Bouman K, Pierini A, Rankin J, Rissmann A, Haeusler M, Yevtushok L, Loane M, Erwich JJHM, de Walle HEK. Stillbirth and neonatal mortality in pregnancies complicated by major congenital anomalies: Findings from a large European cohort. Prenat Diagn 2017; 37: 1100–1111. [DOI] [PubMed] [Google Scholar]
  • 23.MacDorman MF, Reddy UM, Silver RM. Trends in Stillbirth by Gestational Age in the United States, 2006–2012. Obstet Gynecol 2015; 126: 1146–1150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.GBD 2015 Child Mortality Collaborators. Global, regional, national, and selected subnational levels of stillbirths, neonatal, infant, and under-5 mortality, 1980–2015: a systematic analysis for the Global Burden of Disease Study 2015. 2016; 388: 1725–1774. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Morris JK, Wald NJ, Mutton DE, Alberman E. Comparison of models of maternal age-specific risk for Down syndrome live births. Prenat Diagn 2003; 23: 252–258. [DOI] [PubMed] [Google Scholar]
  • 26.Levy B, Sigurjonsson S, Pettersen B, Maisenbacher MK, Hall MP, Demko Z, Lathi RB, Tao R, Aggarwal V, Rabinowitz M. Genomic imbalance in products of conception: Single-nucleotide polymorphism chromosomal microarray analysis. Obstet Gynecol 2014; 124: 202–209. [DOI] [PubMed] [Google Scholar]
  • 27.Pylyp LY, Spynenko LO, Verhoglyad NV, Mishenko AO, Mykytenko DO, Zukin VD. Chromosomal abnormalities in products of conception of first-trimester miscarriages detected by conventional cytogenetic analysis: a review of 1000 cases. J Assist Reprod Genet 2018; 35: 265–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Soler A, Morales C, Mademont-Soler I, Margarit E, Borrell A, Borobio V, Muñoz M, Sánchez A. Overview of Chromosome Abnormalities in First Trimester Miscarriages: A Series of 1,011 Consecutive Chorionic Villi Sample Karyotypes. Cytogenet Genome Res 2017; 152: 81–89. [DOI] [PubMed] [Google Scholar]
  • 29.Kagan KO, Avgidou K. Translucency thickness and chromosomal defects. Obstet Gynecol 2006; 107: 6–10. [DOI] [PubMed] [Google Scholar]
  • 30.Kaimal AJ, Norton ME, Kuppermann M. Prenatal testing in the genomic age: Clinical outcomes, quality of life, and costs. Obstet Gynecol 2015; 126: 737–746. [DOI] [PubMed] [Google Scholar]
  • 31.Clark-Ganheart CA, Fries MH, Leifheit KM, Jensen TJ, Moreno-Ruiz NL, Ye PP, Jennings JM, Driggers RW. Use of cell-free DNA in the investigation of intrauterine fetal demise and miscarriage. Obstet Gynecol 2015; 125: 1321–1329. [DOI] [PubMed] [Google Scholar]
  • 32.Pergament E, Cuckle H, Zimmermann B, Banjevic M, Sigurjonsson S, Ryan A, Hall MP, Dodd M, Lacroute P, Stosic M, Chopra N, Hunkapiller N, Prosen DE, McAdoo S, Demko Z, Siddiqui A, Hill M, Rabinowitz M. Single-nucleotide polymorphism-based noninvasive prenatal screening in a high-risk and low-risk cohort. Obstet Gynecol 2014; 124: 210–218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Del Mar Gil M, Quezada MS, Bregant B, Syngelaki A, Nicolaides KH. Cell-free DNA analysis for trisomy risk assessment in first-trimester twin pregnancies. Fetal Diagn Ther 2014; 35: 204–211. [DOI] [PubMed] [Google Scholar]
  • 34.Gil MM, Accurti V, Santacruz B, Plana MN, Nicolaides KH. Analysis of cell-free DNA in maternal blood in screening for aneuploidies: updated meta-analysis. Ultrasound Obstet Gynecol 2017; 50: 302–314. [DOI] [PubMed] [Google Scholar]
  • 35.Prats P, Rodríguez I, Comas C, Puerto B. Systematic review of screening for trisomy 21 in twin pregnancies in first trimester combining nuchal translucency and biochemical markers: A meta-analysis. Prenat Diagn 2014; 34: 1077–1083. [DOI] [PubMed] [Google Scholar]
  • 36.Nicolaides KH, Chervenak FA, McCullough LB, Avgidou K, Papageorghiou A. Evidence-based obstetric ethics and informed decision-making by pregnant women about invasive diagnosis after first-trimester assessment of risk for trisomy 21. Am J Obstet Gynecol 2005; 193: 322–326. [DOI] [PubMed] [Google Scholar]
  • 37.Chetty S, Garabedian MJ, Norton ME. Uptake of noninvasive prenatal testing (NIPT) in women following positive aneuploidy screening. Prenat Diagn 2013; 33: 542–546. [DOI] [PubMed] [Google Scholar]
  • 38.Kuppermann M, Norton ME, Thao K, O’Leary A, Nseyo O, Cortez A, Kaimal AJ. Preferences regarding contemporary prenatal genetic tests among women desiring testing: Implications for optimal testing strategies. Prenat Diagn 2016; 36: 469–475. [DOI] [PubMed] [Google Scholar]
  • 39.Kuppermann M, Nease RF, Gates E, Learman LA, Blumberg B, Gildengorin V, Washington AE. How do women of diverse backgrounds value prenatal testing outcomes? Prenat Diagn 2004; 24: 424–429. [DOI] [PubMed] [Google Scholar]
  • 40.Judah H, Gil MM, Syngelaki A, Galeva S, Jani J, Akolekar R, Nicolaides KH. Cell-free DNA testing of maternal blood in screening for trisomies in twin pregnancy: updated cohort study at 10–14 weeks and meta-analysis. Ultrasound Obstet Gynecol 2021; 58: 178–189. [DOI] [PubMed] [Google Scholar]

RESOURCES