Background
Cell-free DNA screen failures or “no calls” occur in 1-12% of samples and are frustrating for both clinician and patient. The rate of “no calls” has been shown to have an inverse relationship with gestational age. Recent studies have shown an increased risk for “no calls” among obese women.
Objective
We sought to determine the optimal gestational age for cell-free DNA among obese women.
Study Design
We performed a retrospective cohort study of women who underwent cell-free DNA at a single tertiary care center from 2011 through 2016. Adjusted odds ratios with 95% confidence intervals for a “no call” were determined for each weight class and compared to normal-weight women. The predicted probability of a “no call” with 95% confidence intervals were determined for each week of gestation for normal-weight and obese women and compared.
Results
Among 2385 patients meeting inclusion criteria, 105 (4.4%) had a “no call”. Compared to normal-weight women, the adjusted odds ratio of a “no call” increased with increasing weight class from overweight to obesity class III (respectively: adjusted odds ratio, 2.31; 95% confidence interval, 1.21–4.42 to adjusted odds ratio, 8.55; 95% confidence interval, 4.16–17.56). A cut point at 21 weeks was identified for obesity class II/III women at which there is no longer a significant difference in the probability of a “no call” for obese women compared to normal weight women. From 8-16 weeks, there is a 4.5% reduction in the probability of a “no call” for obesity class II/III women (respectively: 14.9%; 95% confidence interval, 8.95–20.78 and 10.4%; 95% confidence interval, 7.20–13.61; P trend < .01).
Conclusion
The cut point of 21 weeks for optimal sampling of cell-free DNA limits reproductive choices. However, a progressive fall in the probability of a “no call” with advancing gestational age suggests that delaying cell-free DNA for obese women is a reasonable strategy to reduce the probability of a “no call”.
Introduction
Despite controversy as to the utility of cell-free DNA (cfDNA) as a universal first-trimester screening approach, cfDNA has been popularized and is being utilized as aneuploidy screening for both high- and low-risk women throughout the United States. The majority of women who undergo cfDNA will have a clinically useful result of positive or negative; however, 1-12% of women and their clinicians will be frustrated by a cfDNA report without a result or a “no call.” It is well established that the risk of a “no call” falls with advancing gestational age in association with an increase in the circulating fetal fraction of cfDNA in maternal serum. Additionally, maternal weight has been shown to be inversely related to the fetal fraction of circulating cfDNA. A low fetal fraction, usually <4%, can result in a “no call”.
Recently, the significance of a “no call” has come to light as high rates of aneuploidy have been reported among pregnancies with a “no call” with odds ratios (ORs) for aneuploidy ranging from 2.5-9.2, depending on whether the fetal fraction is taken into account. With increasing use of cfDNA for aneuploidy screening and increasing rates of maternal obesity, the “no call” has become a common problem proving to hold increasing clinical significance. With the inverse relationship between gestational age and “no call” risk and the direct relationship between maternal weight and “no call” risk, the question remains as to whether delaying cfDNA sampling in obese women might reduce the risk of a “no call”. Therefore, our objective was to determine the optimal gestational age for cfDNA in obese women to minimize the risk of a “no call”.
Materials and Methods
We conducted a retrospective cohort study of prospectively collected data from the perinatal genetics database at Mercy Hospital St Louis. Institutional review board approval was obtained prior to beginning the study. Mercy Hospital St Louis serves as a large community tertiary referral center in St Louis, MO. Our obstetric service performs approximately 8500 deliveries annually with 34,895 births and 35,937 neonates born during the 5-year study period.
The perinatal database is maintained prospectively by 3 board-certified genetic counselors. A total of 5602 patients underwent complete genetic counseling at Mercy Hospital St Louis during the study period and were offered traditional serum screening, cfDNA screening, and diagnostic testing. In all, 2390 patients chose cfDNA and, while it is unknown whether insurance or rebates paid for their testing, all patients were accepting of their out-of-pocket costs. Women who received a “no call” result were offered diagnostic testing in response to the result, but given there is no current standard as to the best management of a “no call”, the options of traditional serum screening as well as a cfDNA redraw were also discussed and the patients were counseled extensively by the genetic counselors as to the risks and benefits of each approach. Maternal demographic characteristics, date of cfDNA, gestational age at time of cfDNA, modality of cfDNA utilized, and result were collected and entered into the database. Gestational age was based on the best obstetrical estimate. The study included all women who underwent cfDNA through our perinatal genetics department from Nov. 30, 2011, through March 15, 2016. Five women missing either height and/or weight data were excluded from the study.
The patient’s height and weight at the time of cfDNA was collected from the electronic medical record. Weight was either obtained from the patient’s primary obstetrician’s office record if she had an appointment within 3 days of her cfDNA or self-reported. Weight and height were used to calculate body mass index (BMI) and women were classified based on the World Health Organization obesity classification system. BMI ≤18.5 kg/m 2 was classified as underweight, BMI 18.5-24.9 kg/m 2 as normal weight, BMI 25-29.9 kg/m 2 as overweight, BMI 30-34.9 kg/m 2 as class I obesity, BMI 35-39.9 kg/m 2 as class II obesity, and BMI ≥40 kg/m 2 as class III obesity.
Maternal demographic information was compared among women with a reportable result and women with a “no call”. Univariate analysis was performed using χ 2 for categorical variables, Student t test for parametric continuous variables, and Mann-Whitney U test for nonparametric continuous variables. Variables with multiple groups were compared using the nonparametric test for trend described by Cuzick.
Logistic regression was used to determine the OR with 95% confidence intervals (CIs) for a “no call” for each BMI category. Multivariable logistic regression was used to control for maternal age. Adjusted OR (aOR) with 95% CIs for a “no call” compared to normal-weight women were then determined for each BMI category. The predicted probability of a “no call” with 95% CIs was calculated throughout gestation stratified by weight class and compared between normal-weight women and overweight/obese women.
Materials and Methods
We conducted a retrospective cohort study of prospectively collected data from the perinatal genetics database at Mercy Hospital St Louis. Institutional review board approval was obtained prior to beginning the study. Mercy Hospital St Louis serves as a large community tertiary referral center in St Louis, MO. Our obstetric service performs approximately 8500 deliveries annually with 34,895 births and 35,937 neonates born during the 5-year study period.
The perinatal database is maintained prospectively by 3 board-certified genetic counselors. A total of 5602 patients underwent complete genetic counseling at Mercy Hospital St Louis during the study period and were offered traditional serum screening, cfDNA screening, and diagnostic testing. In all, 2390 patients chose cfDNA and, while it is unknown whether insurance or rebates paid for their testing, all patients were accepting of their out-of-pocket costs. Women who received a “no call” result were offered diagnostic testing in response to the result, but given there is no current standard as to the best management of a “no call”, the options of traditional serum screening as well as a cfDNA redraw were also discussed and the patients were counseled extensively by the genetic counselors as to the risks and benefits of each approach. Maternal demographic characteristics, date of cfDNA, gestational age at time of cfDNA, modality of cfDNA utilized, and result were collected and entered into the database. Gestational age was based on the best obstetrical estimate. The study included all women who underwent cfDNA through our perinatal genetics department from Nov. 30, 2011, through March 15, 2016. Five women missing either height and/or weight data were excluded from the study.
The patient’s height and weight at the time of cfDNA was collected from the electronic medical record. Weight was either obtained from the patient’s primary obstetrician’s office record if she had an appointment within 3 days of her cfDNA or self-reported. Weight and height were used to calculate body mass index (BMI) and women were classified based on the World Health Organization obesity classification system. BMI ≤18.5 kg/m 2 was classified as underweight, BMI 18.5-24.9 kg/m 2 as normal weight, BMI 25-29.9 kg/m 2 as overweight, BMI 30-34.9 kg/m 2 as class I obesity, BMI 35-39.9 kg/m 2 as class II obesity, and BMI ≥40 kg/m 2 as class III obesity.
Maternal demographic information was compared among women with a reportable result and women with a “no call”. Univariate analysis was performed using χ 2 for categorical variables, Student t test for parametric continuous variables, and Mann-Whitney U test for nonparametric continuous variables. Variables with multiple groups were compared using the nonparametric test for trend described by Cuzick.
Logistic regression was used to determine the OR with 95% confidence intervals (CIs) for a “no call” for each BMI category. Multivariable logistic regression was used to control for maternal age. Adjusted OR (aOR) with 95% CIs for a “no call” compared to normal-weight women were then determined for each BMI category. The predicted probability of a “no call” with 95% CIs was calculated throughout gestation stratified by weight class and compared between normal-weight women and overweight/obese women.
Results
Of the 2385 patients meeting inclusion criteria, 105 (4.4%) had a “no call”. In all, 55.7% of women in the study were overweight or obese with more than a quarter (27%) classified as class ≥I obesity. On average, women with a “no call” were older than women with a reportable result by 2 years (age 36 years; IQR, 31–38 vs age 34 years; IQR, 29–37; P = .03). Women with a “no call” had a higher BMI than women with a reportable result, on average, the BMI of the “no call” group was in the class I obesity range compared to the reportable result group, which was in the overweight range (BMI 33.39 vs 27.71 kg/m 2 , P < .05). Gestational age at the time of sampling, indication for testing, and modality of testing did not differ between groups ( Table 1 ). The average gestational age at the time of initial cfDNA was between 12-13 weeks ( Table 1 ). The most common indication category for cfDNA was low risk (35.7%, n = 852) closely followed by advanced maternal age (32.7%, n = 780). Panorama served as the cfDNA modality for the majority of the study population (72.4%, n = 1726).
| Maternal characteristics | Reportable n = 2280 | No call n = 105 | P |
|---|---|---|---|
| Maternal age at delivery, y | 34 (29–37) | 36 (31–38) | .030 |
| AMA | 915 (40.13) | 53 (50.48) | .035 |
| GA at initial draw, wk | 12.71 (11.57–17.57) | 12.57 (11.57–14.71) | .43 |
| BMI kg/m 2 , mean (SD) | 27.71 (6.84) | 33.39 (8.44) | <.05 |
| Multiple gestation | 70 (3.07) | 2 (1.90) | .50 |
| Indication | P trend | ||
| Low risk | 824 (36.14) | 28 (26.67) | .66 |
| AMA alone | 735 (32.23) | 45 (42.86) | |
| Ultrasound finding | 200 (8.72) | 10 (9.52) | |
| Abnormal serum screen | 452 (19.82) | 20 (19.05) | |
| Other a | 69 (3.02) | 2 (1.90) | |
| Modality | P trend | ||
| Panorama | 1642 (72.02) | 84 (80.00) | .09 |
| Harmony | 380 (16.67) | 11 (10.48) | |
| Maternity 21 | 44 (1.93) | 6 (5.71) | |
| Verifi | 214 (9.39) | 4 (3.81) | |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
