|
|
||||||||
|
by the American Institute of Ultrasound in Medicine J Ultrasound Med 23:1619-1627 0278-4297 Phenotypic Characteristics of Absent and Hypoplastic Nasal Bones in Fetuses With Down SyndromeDescription by 3-Dimensional Ultrasonography and Clinical SignificanceDepartment of Obstetrics and Gynecology, Wayne State University, Detroit, Michigan USA (L.F.G., M.L.S., P.D., M.M., T.C.); Perinatology Research Branch, National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland USA (J.E., R.R.); Division of Fetal Imaging, William Beaumont Hospital, Royal Oak, Michigan USA (W.L.); and Fetal Diagnostic Center, Pasadena, California USA (G.R.D.). Address correspondence and reprint requests to Roberto Romero, MD, Perinatology Research Branch, National Institute of Child Health and Human Development, Department of Obstetrics and Gynecology, Wayne State University/Hutzel Hospital, 4707 St Antoine Blvd, Detroit MI 48201 USA. E-mail: warfiela{at}mail.nih.gov.
Objective. To determine the frequency and clinical significance of bilateral and unilateral hypoplastic nasal bones for the detection of Down syndrome by 3-dimensional ultrasonography. Methods. Thirty-seven volumes of the fetal skull from fetuses with Down syndrome and 37 from fetuses without abnormalities were analyzed by 1 investigator blinded to fetal karyotype. The maximum intensity projection algorithm was used to reconstruct nasal bones. Ossification patterns were identified in anteroposterior and profile views. Sensitivity, false-positive rates (FPRs), and likelihood ratios (LRs) for detection of Down syndrome were calculated. Results. After exclusions (coronal acquisition [n = 11], hand in front of the face [n = 4], poor imaging [n = 2], incomplete follow-up [n = 2], and anomalies detected after delivery [n = 2]), 53 volumes were analyzed (26 fetuses with Down syndrome and 27 without abnormalities; median gestational age, 21 6/7 weeks [interquartile range, 19 6/725 2/7 weeks]). Rendered profile views revealed absent nasal bones in 18.9% (10 of 53) of the fetuses, and, among these, 90% (9 of 10) had Down syndrome (sensitivity, 34.6% [9 of 26]; FPR, 3.7% [1 of 27]; LR, 9.3 [95% confidence interval (CI), 1.368.7]). Three ossification patterns were identified in anteroposterior views: (1) normally developed, (2) delayed ossification, and (3) absent nasal bones. Sensitivity, FPR, and LR of absent nasal bones for detecting Down syndrome were 34.6% (9 of 26), 3.7% (1 of 27), and 9.0 (95% CI, 1.368.7), respectively. Sensitivity, FPR, and LR of delayed ossification for detecting Down syndrome were 42.3% (11 of 26), 22% (6 of 27), and 1.83 (95% CI, 0.84.4). Conclusions. Absence of nasal bones is associated with the highest risk of Down syndrome. Delayed ossification is associated with a lower risk of Down syndrome than absent nasal bones. These ossification patterns may be indistinguishable on 2-dimensional ultrasonography.
Key Words: abnormalities Down syndrome nasal bone 3-dimensional ultrasonography Abbreviations: CI, confidence interval FPR, false-positive rate LR, likelihood ratio 3DUS, 3-dimensional ultrasonography 2DUS, 2-dimensional ultrasonography
Absence or hypoplasia of the nasal bones has been proposed as a diagnostic sign for identifying fetuses at high risk for Down syndrome.119 During the first trimester of pregnancy, 60% to 80% of fetuses with Down syndrome have absent nasal bones, in contrast to 0.2% to 2.5% of those with a normal karyotype.2,5,8,10,13,15,16,18,20 The proportion of fetuses with Down syndrome who have absent nasal bones during the second trimester is lower than that observed in the first trimester (37%41%), whereas the prevalence of absent nasal bones among fetuses without abnormalities varies depending on maternal ethnicity.16,18,21 For example, the prevalence of absent nasal bones is highest among euploid fetuses of African Caribbean patients (5.8%9.0%), followed by Asian (0.88%5.0%) and Caucasian (2.2%2.6%) patients.18,21 Cicero et al4 found that the nasal bone length of fetuses with Down syndrome was not significantly different from that of fetuses without abnormalities during the first trimester of pregnancy and, therefore, that nasal bone hypoplasia would not be a useful marker during this period. In the second trimester, however, the combination of absent and hypoplastic nasal bones as a single category may be present in 60% to 100% of fetuses with Down syndrome but only 1.2% to 20% of fetuses without abnormalities.3,6,7,12,19 With the exception of 2 studies,12,22 fetal nasal bones mostly have been evaluated by 2-dimensional ultrasonography (2DUS) using a midline sagittal view of the facial profile. The technique for adequate examination of the nasal bones with 2DUS imaging requires a perfect midline sagittal profile view,6,2326 enough image magnification to produce an increment of 0.1 mm for each movement of the measurement calipers, and an angle of approximately 45° between the transducer and an imaginary line passing through the fetal profile. This task is not technically easy, as showed in a study by Cicero et al,24 who evaluated the number of scans required to become proficient in examination of the fetal nasal bones. Fifteen sonographers, all with previous experience in nuchal translucency screening, were asked to obtain midsagittal images of the facial profile to show the nasal bones. The median number of examinations required to become competent in imaging the fetal nasal bones was 80, ranging from 40 to 120 scans. To overcome these difficulties and make the examination less operator dependent, 3-dimensional multiplanar imaging has been proposed as an alternative method for examining the fetal nasal bones.12,22 Recently, several studies have evaluated the postmortem radiographic or histopathologic appearance of nasal bones in fetuses with Down syndrome.2733 Two of these studies31,33 suggested that hypoplastic nasal bones may be asymmetrical (ie, one nasal bone may have an increased proportion of bone tissue when compared with the contralateral side) or unilateral. The purpose of this study was to determine the frequency and clinical significance of bilateral and unilateral hypoplastic nasal bones for the detection of Down syndrome by 3-dimensional ultrasonography (3DUS).
Thirty-seven volume data sets of the fetal skull acquired during 3DUS examination of fetuses with a confirmed diagnosis of Down syndrome were identified by searching the image databases of the Perinatology Research Branch, National Institute of Child Health and Human Development (Detroit, MI), and the Division of Fetal Imaging, Department of Obstetrics and Gynecology, William Beaumont Hospital (Royal Oak, MI). Each fetus with Down syndrome was matched for maternal ethnicity and gestational age at the time of the examination to a control fetus with no congenital abnormalities at the time of second-trimester ultrasonography. Fetuses with Down syndrome were referred for 3DUS either because of an abnormal fetal karyotype or because of abnormal ultrasonographic findings observed at the time of 2DUS. The final diagnosis of Down syndrome was established by prenatal karyotype. Neonatal records of fetuses considered to have no abnormalities at the time of second-trimester ultrasonography were reviewed. Two fetuses with congenital anomalies identified at birth (pleural effusion and 4p syndrome) and 2 with incomplete neonatal follow-up were excluded from the final analysis. All patients were enrolled in protocols approved by the Institutional Review Boards of the National Institute of Child Health and Human Development and Wayne State University, and by the Human Investigation Committee of William Beaumont Hospital. Patients involved signed written informed consents before participating in the study.
Three-dimensional ultrasonography was performed with Voluson 730 systems (GE Medical Systems, Kretztechnik, Zipf, Austria), and volume data sets were acquired by longitudinal sweeps of the fetal face using motorized curved array transducers (25 and 48 MHz). One volume data set for each fetus was included in the study. All data sets had identifiers removed and were analyzed by 1 author (L.F.G.), who was blinded to fetal and neonatal outcomes. Three-dimensional reconstruction of the nasal bones in anteroposterior and profile projection views was performed with the maximum intensity projection algorithm, as described in Figure 1
Nasal bone ossification patterns were identified in the anteroposterior and profile projection views, and the volume data sets were classified according to the identified patterns. A second analysis of the volume data sets was performed by the same observer 1 week after the first evaluation, as well as by another observer who was also blinded to fetal outcome. Intraobserver and interobserver agreement were evaluated with the coefficient of concordance, which was interpreted as follows: chance (0), poor (00.19), fair (0.200.39), moderate (0.400.59), substantial (0.600.79), and almost perfect (0.801.0) agreement.34 Sensitivity, false-positive rates (FPRs), and likelihood ratios (LRs) for detecting Down syndrome were calculated for nasal bone ossification patterns identified in the anteroposterior and profile projection views.
Fifty-four patients were Caucasian (73%), 18 were African American (24.3%), and 2 were Asian (2.7%). After exclusion of 21 volume data sets (coronal acquisitions [n = 11], hand in front of the face [n = 4], poor imaging [n = 2], unsuspected congenital anomalies detected in the neonatal period [n = 2], and incomplete follow-up [n = 2]), 26 volumes from fetuses with Down syndrome and 27 from fetuses without abnormalities were available for final analysis. Volumes acquired from the lateral aspect of the skull (coronal acquisitions) were excluded from the final analysis because only 1 nasal bone was effectively imaged in this acquisition plane. The median gestational age at the time of examination was 21 6/7 weeks (interquartile range, 19 6/725 2/7 weeks).
Three-dimensional Reconstruction of the Nasal Bones
Profile Projection View In the profile projection view, 2 ossification patterns were identified: (1) nasal bone present (Figures 1B
Figure 4
Intraobserver and Interobserver Agreement Almost perfect intraobserver agreement was observed for both projection views. The coefficients of concordance were 0.87 for nasal bone ossification patterns visualized in the profile projection view and 0.92 for patterns visualized in the anteroposterior projection view. Agreement between different observers, however, was moderate, with a coefficient of concordance of 0.56 for both profile and anteroposterior projection views.
Nasal Bone Ossification Patterns Detected by 3DUS as Markers for Down Syndrome
Of note, 3 of the 26 fetuses with Down syndrome had no ultrasonographic markers for this condition at the time of the examination. Among these fetuses, 2 had complete absence of the nasal bones, whereas 1 had normal nasal bones.
Anteroposterior Projection Views
Thirty-five percent (9 of 26) of the fetuses with Down syndrome and 3.7% (1 of 27) of the fetuses without abnormalities had absent nasal bones (LR, 9.0 [95% CI, 1.368.7]). Hypoplastic nasal bones occurred in 42.3% (11 of 26) of the fetuses with Down syndrome and 22.2% (6 of 27) of the fetuses without abnormalities (LR, 1.83 [95% CI, 0.84.4]). The only case of a unilateral nasal bone was observed in a fetus with Down syndrome. Normal nasal bones were observed in 23% (6 of 26) of the fetuses with Down syndrome and 74.1% (20 of 27) of the fetuses without abnormalities (LR, 0.30 [95% CI, 0.150.65]). If absent and delayed ossification nasal bones were combined into a single category, 20 of 26 fetuses with Down syndrome would have been correctly identified (sensitivity, 76.9%), for an FPR of 25.9% (7 of 27) (LR, 2.97 [95% CI, 1.525.81]).
This study describes volume-rendered views of the nasal bones in fetuses with Down syndrome. We confirmed observations made in previous postmortem studies using either histologic sections33 or radiographs31 that nasal bone hypoplasia in fetuses with Down syndrome is characterized by asymmetrical distribution or unilateral absence of nasal bone tissue. Furthermore, our findings suggest that hypoplastic nasal bones do not fuse in the midline. Complete absence of the nasal bones was associated with the highest risk of trisomy 21, and this finding was best depicted on volume-rendered profile projection views of the fetal skull (sensitivity, 34.6%; FPR, 3%; LR, 9.0 [95% CI, 1.368.7]; Figures 3B
The nasal bones develop from paired independent ossification centers located in a membrane that covers the cartilaginous nasal capsule.35 In a study of 62 human fetuses without gross abnormalities, the earliest histologic observation of the nasal bones occurred in a fetus with a crown-rump length of 42 mm ( Several studies have proposed that delayed maturation of the nasal bones is a possible explanation for the increased prevalence of absent or hypoplastic nasal bones in fetuses with Down syndrome.29,36 Using an accepted animal model for trisomy 21, Ludwig et al36 compared development of the eye, ear, and nose in mouse embryos with trisomy 16 against siblings with a normal karyotype at exactly matched gestational ages. Development of the nose and sensory structures of the otic vesicle and skull ossification were delayed in mouse embryos with trisomy 16. Keeling et al27 studied radiographs of midsagittal tissue blocks of the skeletons of fetuses with trisomy 21 at autopsy (1224 weeks of gestation) and found that malformation or agenesis of the nasal bones was present in 61.3% (19 of 31) of the fetuses. These findings were independently confirmed by other investigators, who found the nasal bones to be hypoplastic or absent in 25% to 58.8% of fetuses with trisomy 21, regardless of gestational age at the time of autopsy.2831 A study by Tuxen et al31 is noteworthy because, in addition to the observation of bilateral absence of the nasal bones in 24% (8 of 33) of the fetuses with Down syndrome, unilateral absence of the nasal bones was detected in 6.1% (2 of 33) of the cases. Four studies reported postmortem histopathologic findings of nasal bones in fetuses with Down syndrome, and nasal bone absence or hypoplasia was observed in all of them.3033 Among these studies, Rustico et al33 used transverse rather than sagittal histologic sections through the nasal crown and, therefore, was able to visually compare the appearance of the paired nasal bones. Nasal bone histologic sections from 3 fetuses with Down syndrome were compared with those obtained from 4 fetuses with a normal karyotype (mean gestational age, 16 weeks). Semiquantitative analysis showed that the proportion of the nasal bone ranged from 35% to 50% of the total volume of osteocartilaginous tissue in fetuses with a normal karyotype, whereas, in fetuses with Down syndrome, only 10% to 20% of bone tissue was present. In at least 1 of the fetuses with trisomy 21, the ossification centers were notably asymmetrical. The observation that hypoplastic nasal bones do not fuse in the midline has important clinical implications. If true midline sagittal views of the facial profile are obtained by either 2DUS or 3DUS, how can the examiner be sure that the nasal bones are truly absent as opposed to hypoplastic? Because the risk for Down syndrome in fetuses with hypoplastic nasal bones is considerably lower when compared with complete absence of the nasal bones, genetic counseling based on a sagittal profile view alone may be associated with an artificially increased FPR and could result in an elevated number of unwarranted invasive procedures to obtain a fetal karyotype. A limitation of this study is that the observations reported herein do not necessarily apply to the first trimester of pregnancy. In addition, this was a case-control study with a 1:1 case-control ratio. Despite the low number of control fetuses for each case, we think that our observations represent adequate estimates of the risk for Down syndrome. Our LR of 9.0 for complete absence of the nasal bones agrees with the lowest reported LR for detection of Down syndrome during the second trimester of the pregnancy.7 Furthermore, an increase in the number of control fetuses would most likely have resulted in lower FPRs without changing the sensitivity and, therefore, higher LRs. In conclusion, complete absence of nasal bones can be shown on volume-rendered sagittal views of the fetal skull with the maximum intensity projection algorithm. This ossification pattern is associated with a high risk of Down syndrome. Hypoplastic nasal bones are more likely to be identified by 3DUS with anteroposterior views of the fetal skull. This ossification pattern is associated with a lower risk of Down syndrome when compared with complete absence of nasal bones. Importantly, hypoplastic nasal bones may be indistinguishable from complete absence of the nasal bones on true midsagittal views of the facial profile obtained by 2DUS or multiplanar 3DUS.
Received July 13, 2004, from the Department of Obstetrics and Gynecology, Wayne State University, Detroit, Michigan USA (L.F.G., M.L.S., P.D., M.M., T.C.); Perinatology Research Branch, National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland USA (J.E., R.R.); Division of Fetal Imaging, William Beaumont Hospital, Royal Oak, Michigan USA (W.L.); and Fetal Diagnostic Center, Pasadena, California USA (G.R.D.). Revision requested July 21, 2004. Revised manuscript accepted for publication August 2, 2004. We thank Patrick Shoff for assistance with preparation of the images.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |