Transcript
283jumonline.qxp:Layout 1
2/14/09
2:26 PM
Page 321
CME
Article
Second-Trimester Genetic Sonography After First-Trimester Combined Screening for Trisomy 21 Joseph R. Wax, MD, Michael G. Pinette, MD, Angelina Cartin, Jacquelyn Blackstone, DO
Objective. The purpose of this study was to evaluate the trisomy 21 screening performance of the first-trimester combined test followed by second-trimester genetic sonography. Methods. This retrospective cohort study included all women with singleton pregnancies undergoing combined screening followed by genetic sonography at 17 to 21 weeks from January 1, 2005, to January 31, 2008. Combined test trisomy 21 risks were multiplied by positive or negative likelihood ratios based on the second-trimester sonographic findings to determine the final trisomy 21 risk. Sonography was evaluated as the second part of (1) a stepwise sequential test applied to combined screen-negative pregnancies and (2) an integrated test applied to all combined screen patients regardless of the latter results. A final trisomy 21 risk of 1:270 or higher was considered screen-positive. Results. A total of 2231 pregnancies underwent combined screening, which detected 7 of 8 Down syndrome cases (87.5%) at a 9.6% screen-positive rate. A total of 884 of these patients (39.6%), including 2 having fetuses with Down syndrome, had genetic sonography. Combined screening detected 1 of these trisomy 21 fetuses (50%) at a 15.7% screen-positive rate. Integrated ultrasound-based aneuploidy screening detected both trisomy 21 cases (100%) at a 22.7% screen-positive rate, whereas stepwise sequential ultrasound-based aneuploidy screening also detected both trisomy 21 fetuses (100%) but at a 28.3% screen-positive rate (P < .0001). Conclusions. Second-trimester genetic sonography after first-trimester combined screening may improve trisomy 21 detection at the expense of increasing screen-positive rates. Key words: aneuploidy screening; first trimester; sonography; trisomy 21.
Abbreviations CI, confidence interval; EIF, echogenic intracardiac focus; IUBAS, integrated ultrasound-based aneuploidy screening; LR, likelihood ratio; NSF, nuchal skin fold; SSUBAS, stepwise sequential ultrasound-based aneuploidy screening
Received October 15, 2008, from the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Maine Medical Center, Portland, Maine USA. Revision requested December 1, 2008. Revised manuscript accepted for publication December 9, 2008. Address correspondence to Joseph R. Wax, MD, MMC Ob/Gyn Associates, 887 Congress St, Suite 200, Portland, ME 04102 USA. E-mail:
[email protected]
CME
Article includes CME test
F
irst-trimester combined screening, either alone or as part of sequential or integrated tests, is widely accepted and effective in detecting Down syndrome and other chromosomal abnormalities.1–3 Before the advent of first-trimester aneuploidy screening, second-trimester multiple–serum marker testing was the standard Down syndrome screening test.1 In experienced hands, second-trimester sonography reliably modifies the serum-based trisomy 21 risk, improving overall screening efficiency.4–6 However, there is a paucity of data on adjusting a combined screen’s trisomy 21 risk with second-trimester sonography.7–9 Therefore, we evaluated the trisomy 21 screening performance of the firsttrimester combined test followed by second-trimester sonography.
© 2009 by the American Institute of Ultrasound in Medicine • J Ultrasound Med 2009; 28:321–325 • 0278-4297/09/$3.50
283jumonline.qxp:Layout 1
2/14/09
2:26 PM
Page 322
Second-Trimester Genetic Sonography for Trisomy 21
Materials and Methods This retrospective cohort study received an Institutional Review Board exemption. We included all women with singleton pregnancies undergoing combined screening at a fetal crownrump length of 45 to 84 mm and second-trimester sonography in a single maternal-fetal medicine practice from January 1, 2005, to January 31, 2008. All sonographers and sonologists were credentialed by either the Fetal Medicine Foundation or the Nuchal Translucency Quality Review Program. The combined screen threshold for trisomy 21 at the laboratory we used was a risk 1:220 or higher. The combined screen results were available at the time of second-trimester sonography. For the purpose of this study, the combined screen was the first part, and second-trimester sonography represented the second component of either an integrated or stepwise sequential test. In the integrated approach, the quantitative combined screen trisomy 21 risk for all patients was multiplied by a likelihood ratio (LR) based on the second-trimester sonographic findings. Normal sonographic findings were conservatively assigned a negative LR of 0.50.5 Positive findings and their respective LRs included the following: a thickened nuchal skin fold (NSF) of 6 mm or greater, LR = 17; a short humerus observed/ expected ratio of 0.89 or less,10 LR = 7.5; a short femur observed/expected ratio of 0.91 or less,10 LR = 2.7; fetal echogenic bowel grade 2 or 3,11 LR = 6.1, an echogenic intracardiac focus (EIF), LR = 2.8; and renal pelvis dilatation of 4 mm or greater, LR = 1.9.12 This value, the final sonographically adjusted trisomy 21 risk, was considered screen-positive if 1:270 or higher, that of a 35-year old woman. Congenital heart disease and duodenal atresia were considered screen-positive. In the stepwise sequential approach, the quantitative combined test trisomy 21 risk in combined screen-negative patients was multiplied by the appropriate second-trimester sonographically determined LR. This value, the final trisomy 21 risk, was considered screen-positive if 1:270 or higher. Congenital heart disease and duodenal atresia were considered screen-positive. Down syndrome cases were ascertained from our prenatal genetic database, including prenatal and newborn chromosomal testing. A normal new322
born physical examination was a surrogate for a normal karyotype when chromosomal analysis was not performed. Trisomy 21 detection and screen-positive rates with 95% confidence intervals (CIs) were compared for combined screening, integrated ultrasound-based aneuploidy screening (IUBAS), and stepwise sequential ultrasound-based aneuploidy screening (SSUBAS). Data were analyzed with descriptive statistics and χ2 and Fisher exact tests.
Results The study patients’ characteristics are noted in Table 1. There were 8 fetuses with trisomy 21 among the 2231 women undergoing combined screening, a rate of 3.6 per 1000. A total of 214 (9.6%; 95% CI, 8.4%–10.8%) combined screens were positive for trisomy 21, detecting 7 (87.5%; 95% CI, 52.9%–97.8%) affected offspring. Among the 884 women (39.6%) undergoing combined screening and second-trimester sonography in our practice, 139 (15.7%; 95% CI, 13.3%–18.1%) were combined screen-positive. There were 2 fetuses with trisomy 21 in this group, of which 1 (50%; 95% CI, 9%–91%) was combined screen-positive (trisomy 21 risk, 1:5). This fetus had an atrioseptal defect found by second-trimester sonography. Sonographic findings by combined test results are noted in Table 2. With the exception of 3 fetuses, all markers were observed in isolation. In these 3 fetuses, positive LRs were multiplied by each other to provide the final LR. Application of second-trimester sonography as the second component of an integrated test detected both fetuses with trisomy 21 (100%; 95% CI, 29%– 100%) at a 22.7% (95% CI, 20%–25.5%) screenpositive rate. Thus, 1 additional case of trisomy 21 was found for 62 more positive screens (7%). This fetus had an isolated EIF. In this case, the combined screen risk of 1:445 was negative but was raised by the EIF to 1:159, or positive. Application of second-trimester sonography as the second part of a stepwise sequential test detected both fetuses with trisomy 21 (100%; 95% CI, 29%–100%) at a 28.3% (95% CI, 25.3%– 31.2%) screen-positive rate. Thus, 1 additional case of trisomy 21 was found for 111 more positive screens (12.6%). The 3 screening algorithms’ efficiencies are compared in Table 3. J Ultrasound Med 2009; 28:321–325
283jumonline.qxp:Layout 1
2/14/09
2:26 PM
Page 323
Wax et al
Discussion This investigation confirms that first-trimester combined screening is highly efficient at detecting trisomy 21 in a general obstetric population. The patients undergoing genetic sonography were primarily high risk on the basis of maternal age, suggesting that second-trimester genetic sonography after combined screening, as the second part of either an integrated or a stepwise sequential test, substantially raises the screenpositive rate in this group of patients. In our study, IUBAS detected an extra case of trisomy 21 for an additional 62 positive screens, whereas SSUBAS detected an extra case of trisomy 21 for 111 additional positive screens. Because only 1 case of trisomy 21 was not detected by combined screening, the potential benefit of follow-up genetic sonography appears to be outweighed by the increased screen-positive rate. However, the small number of fetuses with trisomy 21 in our series limits the strength of this conclusion. The screening efficiency of second-trimester sonography after combined testing will depend on the incidence of trisomy 21 in the population, efficiency of combined screening, selected second-trimester sonographic markers, prevalence of these markers, and assigned LRs. Our findings illustrate an important consequence of a highly effective first-trimester screening program. With most cases of trisomy 21 detected before secondtrimester sonography, few affected fetuses remain, resulting in decreased positive predictive values and higher false-positive rates after secondary sonographic screening. This discussion raises several important points regarding the LRs assigned to second-trimester sonographic findings. We used positive LRs for isolated markers derived by a meta-analysis of published studies, the results of which were consistent with another recent report.12,13 All fetuses except 3 with markers had them as isolated findings. Therefore, we think that our approach was valid and reasonable. Moreover, the number of trisomy 21 cases precluded calculating meaningful LRs from our own data set. Importantly, these LRs were derived from studies performed before the advent of first-trimester aneuploidy screening. Therefore, it is likely that the positive LRs overestimate risk increases and J Ultrasound Med 2009; 28:321–325
the negative LRs underestimate risk reduction when applied to a population previously screened by combined testing. As a result, second-trimester genetic sonographic LRs may need to be adjusted to account for this phenomenon and to avoid the high screen-positive rates evidenced by this study. These LR modifications will require large numbers of affected and unaffected fetuses to ensure narrow CIs. Finally, it remains unclear whether secondtrimester sonographic markers are independent of first-trimester serum markers and nuchal translucency measurements. Although 1 group did not evaluate these relationships, they speculated that such associations were unlikely or negligible.9 Until actual data are available, we recommend caution in this regard. Few studies have evaluated second-trimester sonography after aneuploidy screening that included first-trimester components. Krantz et al9 used previously published data to mathematically model the effect of second-trimester sonography after combined screening. In this model, when genetic sonography was used as the second part of a stepwise sequential screen, trisomy 21 detection increased to 94.6% from 88.5% by combined screening alone, with an accompanying increase in the screen-positive rate from 4.2% to 5.4%. When genetic sonography was used as the second part of a contingent sequential test for intermediate-risk combined screen results, Down syndrome detection and screen-positive rates were 93.3% and 4.9%, respectively. These authors concluded that sec-
Table 1. Characteristics of Women Undergoing Combined Screening and Second-Trimester Ultrasound (n = 884) Characteristic
Maternal age, y, mean ± SD Age ≥35 y, n (%) Gestational age at combined screen, wk, mean ± SD Indications for second-trimester sonography Maternal age ≥35 y, n Abnormal aneuploidy screen results, n Routine anatomy survey, n Other, n Gestational age at genetic sonography, wk, mean ± SD Nuchal translucency, mm, mean ± SD Invasive prenatal diagnosis, n (%) Chorionic villus sampling, n (%) Amniocentesis, n (%)
Value
36.7 ± 3.2 811 (91.7) 12.5 ± 0.7 818 102 33 23 19 ± 0.9 1.46 ± 0.49 97 (11) 13 (1.5) 84 (9.5)
323
283jumonline.qxp:Layout 1
2/14/09
2:26 PM
Page 324
Second-Trimester Genetic Sonography for Trisomy 21
Table 2. Second-Trimester Sonographic Findings After First-Trimester Combined Screening Marker, n (%)
None Any NSF ≥6 mm Short humerus Short femur Fetal echogenic bowel EIF Pyelectasis ≥4 mm Congenital heart disease (atrioseptal defect) Multiple markers
All Patients (n = 884)
771 113 3 6 6 10 67 14 1
Combined Screen-Positive Patients (n = 139)
(87.3) (12.7) (0.3) (0.7) (0.7) (1.1) (7.6) (1.6) (0.1)
118 21 1 3 0 3 7 3 1
6 (0.7)
Combined Screen-Negative Patients (n = 745)a
(84.9) (15.1) (0.7) (2.1) (0) (2.1) (5) (2.1) (0.7)b
653 92 2 3 6 7 60 11 0
3 (2.1)
(87.7) (12.3) (0.3) (0.4) (0.8) (1) (8)b (1.5) (0)
3 (0.4)
a
P = .007 versus combined screen-positive patients. One Down syndrome case each.
b
ond-trimester genetic sonography could serve as an effective screening test after first-trimester combined screening.9 A retrospective cohort study of secondtrimester genetic sonography in a general obstetric population after fully integrated testing found a reduction in the trisomy 21 detection rate from 83% to 75%, whereas the screen-positive rate decreased from 2.5% to 1.8% (cutoff, 1:150) or 2.4% (cutoff, 1:250). Alternatively, applying genetic sonography only to fully integrated screen-negative patients maintained the detection rate but raised the frequency of screen-positive findings by 0.5%. The authors recommended not using genetic sonography to modify integrated test results, offering amniocentesis to patients with positive integrated screen findings or those with negative integrated screen findings when a fetal anomaly but not a soft marker was identified.8 A large multicenter investigation evaluated the effect of a second-trimester (20- to 22-week) NSF of 6 mm or greater on Down syndrome detection after first-trimester combined screening. This
approach improved detection rates from 79.6% to 89.7% with corresponding screen-positive rates of 2.7% and 4.2%. Thus, an NSF of 6 mm or greater detected 5 of 9 fetuses with trisomy 21 (55.6%) missed by the combined screen, for a 1.4% rise in positive screens. The investigators suggested that this 2-step approach represented a reasonable Down syndrome screen.7 In summary, our study shows that secondtrimester genetic sonography applied to firsttrimester combined screen trisomy 21 risks results in unacceptably high screen-positive rates in a primarily high-risk population. We therefore suggest counseling these patients on the basis of the combined screen trisomy 21 risk, offering prenatal diagnosis to screen-positive patients and screen-negative patients when a major anomaly is identified by second-trimester sonography. Future studies should evaluate the importance of aneuploidy markers in general as well as high-risk populations prescreened by first-trimester aneuploidy screening. These investigations should revise and use revised
Table 3. Comparative Efficiency of Screening Algorithms for Trisomy 21 Algorithm
Combined Population (N = 2231) Sample (n = 884) IUBAS (n = 884) SSUBAS (n = 884)
Fetuses With Trisomy 21, n
n
8 2 2 2
7 1 2 2
Detection Rate % (95% CI)
87.5 50.0 100 100
(52.9–97.8) (9–91) (29–100) (29–100)
n
Screen-Positive Rate % (95% CI)
214 139 201 250
9.6 15.7 22.7 28.3
(8.4–10.8) (13.3–18.1)a (20–25.5) (25.3–31.2)
a
P < .0001 versus IUBAS and SSUBAS.
324
J Ultrasound Med 2009; 28:321–325
283jumonline.qxp:Layout 1
2/14/09
2:26 PM
Page 325
Wax et al
sonographically based LRs to reflect the diminished positive and improved negative predictive values after first-trimester screening. This report and others serve as models on how to then reevaluate the role of second-trimester genetic sonography after first-trimester aneuploidy screening.
13.
Nyberg DA, Souter VL, El-Bastawissi A, Young S, Luthhardt F, Luthy DA. Isolated sonographic markers for detection of fetal Down syndrome in the second trimester of pregnancy. J Ultrasound Med 2001; 20:1053–1063.
References 1.
American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 77: screening for fetal chromosomal abnormalities. Obstet Gynecol 2007; 109:217– 227.
2.
Wapner R, Thom E, Simpson JL, et al. First-trimester screening for trisomies 21 and 18. N Engl J Med 2003; 349: 1405–1413.
3.
Malone FD, Canick JA, Ball RH, et al. First-trimester or second-trimester screening, or both, for Down’s syndrome. N Engl J Med 2005; 353:2001–2011.
4.
Pinette MG, Egan JFX, Wax JR, Blacktone J, Cartin A, Benn PA. Combined sonographic and biochemical markers for Down syndrome screening. J Ultrasound Med 2003; 22:1185–1190.
5.
Pinette MG, Garrett J, Salvo A, et al. Normal midtrimester (17–20 weeks) genetic sonogram decreases amniocentesis rate in a high-risk population. J Ultrasound Med 2001; 20:639–644.
6.
Benn PA, Kaminsky LM, Ying J, Borgida AF, Egan JFX. Combined second-trimester biochemical and ultrasound screening for Down syndrome. Obstet Gynecol 2002; 100:1168–1176.
7.
Rozenberg P, Bussières L, Chevret S, et al. Screening for Down syndrome using first-trimester combined screening followed by second-trimester ultrasound examination in an unselected population. Am J Obstet Gynecol 2006; 195: 1379–1387.
8.
Weisz B, Pandya PP, David AL, Huttly W, Jones P, Rodeck CH. Ultrasound findings after screening for Down syndrome using the integrated test. Obstet Gynecol 2007; 109:1046–1052.
9.
Krantz DA, Hallahan TW, Macri VJ, Macri JN. Genetic sonography after first-trimester Down syndrome screening. Ultrasound Obstet Gynecol 2007; 29:666–670.
10.
Nyberg DA, Resta RG, Luthy DA, Hickok DE, Williams MA. Humerus and femur length shortening in the detection of Down’s syndrome. Am J Obstet Gynecol 1993; 168:534– 538.
11.
Slotnick RN, Abuhamad AZ. Prognostic implications of fetal echogenic bowel. Lancet 1996; 347:85–87.
12.
Smith-Bindman R, Hosmer W, Feldstein VA, Deeks JJ, Goldberg JD. Second-trimester ultrasound to detect fetuses with Down syndrome: a meta-analysis. JAMA 2001; 285:1044–1055.
J Ultrasound Med 2009; 28:321–325
325
