Early detection and diagnosis of fetal neural tube

Size: px

Start display at page:

Download "Early detection and diagnosis of fetal neural tube"

Chastity McCoy
5 years ago
Views:

1 RESEARCH PAPERS Fetal Spine Morphology and Maturation During the Second Trimester Sonographic Evaluation Roy A. Filly, MD, Gary F. Simpson, MOt Gregory Linkowski, MO- We reviewed high+resolution reahime sonographic images (especially selected for a lack of technical arti, facts) of the fetal spine of 46 normal fetuses between menstrual weeks of development. The degree of visible posterior neural arch ossification was graded and compared with the gestational age. Recognizable early ossification of the laminae was seen in the cervical re e gion of all fetuses studied. Similar ossification followed sequentially in the thoracic, lumbar, and sacral regions; however, sonographically recognizable ossification of the laminae in the lumbar spine was delayed until weeks and in the sacral spine until 25 weeks or more. Awareness of these features will help to prevent falsepositive diagnoses of dysraphism and, conversely, hopefully augment our ability to detect such lesions. KEY WORDS: sonography, fetal anatomy; sonography, spine ossification; sonography, fetal development. (j Ultra SOl/lid Med 6:631, 1987) Early detection and diagnosis of fetal neural tube defects is rapidly becoming an integral part of modern obstetrical care. With the advent of maternal serum alpha,fetoprotein (MSAFP) screening, as well as the concurrent development of sophisticated high-resolution ultrasound technology, the potential currently exists to diagnose the vast majority of neural tube defects before the 20th week of pregnancy. These changes in obstetrical care mandate that the morphology of the fetal spine be well understood by sonographers and sonologists. The sequence of development of ossification centers in the fetal vertebral column has been extensively studied in the past with radiographic l - J and histologic methods. 4 Each vertebra usually has three primary ossification centers, one for Received January 23, 1987, (rom the "Department of Radiology, University of California School o( Medicine. San Francisco, Califor nia; and the tdepartment of Fetal Medicine and Surgery, Good 50]4 maritan Medical Center, Phoenix, Arizona. and Reproductive Genetics, The Genetics Center and Southwest Biomedical RfI$earch Center, Scottsdale, Arizona. Revised manuscript accepted for publica. tion May Address correspondence and reprint requests to Dr. Filly: University of California School of Medicine. Department o(radiology, L374, San Francisco, CA 94143, the body (centrum) and one on each side of the posterior neural arch. The centra are ossified first in the lower thoracic and upper lumbar regions followed by progressive ossification in both the cephalic and caudal directions. By contrast and in general, the ossification centers for the posterior neural arch appear in a more standard cephalocaudal direction. 4 The posterior neural arch first begins to ossify (sonographically recognizable highamplitude reflections) at the base of the transverse processes. Ossification proceeds from this center to progressively include the laminae and pedicles. The progression of ossification of the laminae is the more important in the diagnosis of neural tube defects. Spina bifida, the most consistently demonstrable lesion in open spinal defects, is recognized sonographically by an abnormal outward flaring of the posterior neural arch ossification centers (Fig. 1). Currently, there is a paucity of published data 5-7 regarding ultrasonically determined morphologic development of the posterior arch ossification centers in various regions of the human fetal spine during the second trimester of pregnancy. To further evaluate this important anatomy, we made sequential observations of the posterior ossification centers of the fetal spine to determine the degree of 1987 by the American Institute of Ultrasound in Medicine. J Ultrasound Med 6: /87/$3.50

632 FETAL SPINE MORPHOLOGY J Ultrasound Med 6:631-636, 1987 gestational age.

2 632 FETAL SPINE MORPHOLOGY J Ultrasound Med 6: , 1987 gestational age. Only cases exhibiting the highest quality images of the spine were analyzed in order to decrease the influence of technical factors on the desired observations. All examinations in which the fetus assumed a supine position were excluded regardless of technical quality. Most of the examinations (30 fetuses) were concentrated at weeks, when prenatal sonograms are most commonly requested to exclude spina bifida. The interpretation of the sonograms was blinded as to gestational age at the time of the examination. All ob- Figure 1 Example of the importance of outward splayirig of the laminae (L) for diagnosis of spinal neural tube defects. No myelomeningocele or skin defect can be appreciated in this fetus due to severe oligohydramnios secondary to obstructive uropathy. The dorsal aspect of the Cetus lies against the plaa centa (P). However, the spina bifida is readily apparent. C, centrum. sonographically visible ossification of the laminae and the effect of fetal position on the sonologists' perception of the posterior ossification centers. Figure 3 A, axial sonogram of a lumbar spinal segment in a menstrua] week fetus. Primary ossification centers (POC) are present in the posterior neural arch; however, there is no visible extension into the laminae. C, ossification center of the centrum; arrows, neurocentral synchondrosis. B, same fetus at menstrual weeks. Minimal but recognizable progression of ossification into the lamina (L) is present. Posterior neural arch ossification progresses circumferentially from its "starting point" at the bases of the transverse processes (the primary ossification centers). MATERIALS AND METHODS Fortywsix normal fetuses from 18 to 27 menstrual weeks were examined using a highly focused phased-array realatime ultrasound system (Acuson 128) employing 3.,5. and S.O-MHz linear transducers. Gestational age was confirmed by concordance (± 1 week) of age by last menstrual period with the sonographically documented Figure 2 Axial sonogram of a lumbar spinal segment in a menstrual week fetus. Ossification has extended well A into the laminae (L). C, ossificati.on center of the centrum; arrows, neurocenlra] synchondrosis. B

J Ultrasound Med 6:631-636, 1987 FILLY ET AL 633 Figure 4 Axial sonogram of an upper sacral spinal segment in a 19-20 menstrual week fetus.

3 J Ultrasound Med 6: , 1987 FILLY ET AL 633 Figure 4 Axial sonogram of an upper sacral spinal segment in a menstrual week fetus. The posterior arch ossification centers (POC) are identifiable as small round foci of highamplitude reflections with no visible ossification of the laminae. IW, iliac wings. servations were made on axial images. The degree of development of posterior ossification centers was graded as follows: +, ossification had extended well into the laminae of the posterior neural arch (Fig. 2); ±, minimal but recognizable ossification of the laminae (Fig. 38); -, posterior neural ossification centers that are identifiable as small, round foci of high ~ampjitude reflections with no visible ossification of the laminae (Fig. 4). Finally, the fetal position (prone vs. decubitus) was tabulated. RESULTS Table 1 summarizes the results of our study. The range of gestational ages studied, the degree of posterior arch ossification extending into the laminae in the various fetal spine regions, and the position of the fetal spine at Figure 5 A, axial sonogram of a lower cervical spinal se ment it) a 19- io merstrual week fetus. Virtually all fetuses I this age have ea'sily identifiable ossification of the laminae (; of the cervical spine. 5, scapula. B, axial sonogram through mid~ thoracic spinal segment in the same fetus. The laminae (I have easily visible ossification. Arrow, neurocentral synchon drosis. Table 1: Gestational Results Position of Fetal Age (wk) Spine Cervical Decubitus ++++ Prone Decubitus Prone Decubitus Prone Decubitus Decubitus Decubitus ++ Prone Decubitus Decubitus Prone Decubitus ++ Regions of the Fetal Spine Thoracic Lumbar Sacral +±±± ±±- ±± ± ± ± ±±± ± ± ± ±± ± ± +±±

634 FETAL SPINE MORPHOLOGY J Ultrasound Med 6:631-636, 1987 6 Figure 8 Axial sonogram of an upper sacral spinal segment in a 26-27 menstrual week fetus.

4 634 FETAL SPINE MORPHOLOGY J Ultrasound Med 6: , Figure 8 Axial sonogram of an upper sacral spinal segment in a menstrual week fetus. Easily visible ossification of the laminae (L) of the sacral spine is not seen consistently until the late second trimester. B, bladder; IW, iliac wing. 7 Figure 6 Axial sonogram of a lower thoracic spinal segment in a menstrual week fetus. Early ossification of the lamina (I..) is just becoming visible. Compare with the fetus in Fig. 5 of the same age but with more advanced ossification of the thoracic laminae. R, rib; 5, stomach. Figure 7 Axial sonogram of a lower thoracic spinal segment in a menstrual week fetus. Compare the more advanced ossification of the laminae (I..) with Fig. 6. R, rib; S, stomach. the time of examination are depicted. Easily identifiable ossi6cation of the laminae was visible in the cervical region in all fetuses by menstrual weeks (Fig. 5), while thoracic ossification of the laminae was only par~ tially visible during the (Fig. 6) menstrual week period. There was no ossification in the laminae observed in the lumbar (Fig. 3A) or sacral (Fig. 4) regions of any fetuses examined prior to 19 menstrual weeks. The thoracic vertebrae consistently demonstrated partial ossification of the laminae in the range of weeks (Fig. 7), while the lumbar region was delayed until weeks (Fig. 2). The sacral spine revealed no evidence of ossification in the laminae prior to 22 weeks. Only after 25 weeks was there consistently recognizable ossification of the arch in the sacral region (Fig. 8). Fetal position (either prone or decubitus) did not appear to affect appreciably the ability to discern the degree of neural arch ossification (Fig. 9), although the prone position usually resulted in the clearest images (Figs. 2, 8, and 10). To some extent, the study further documents cephalocaudal maturation of the posterior neural arch of the fetal spine. As can be seen in Table 1, ossification of the laminae occurs earlier in the proximal spine and pro gt'esses caudally as the fetus matures. Figure 9 Axial sonograms through an upper sacral spinal segment in both the prone (A) and decubitus (B) fetal positions. This menstrual week fetus Changed positions during the course of the examination. Note that the change in position did not appreciably alter the abijity to discern the degree of ossification of the posterior arch (POC).

J Ultrasound Med 6:631-636, 1987 FILLY ET AL 635 Figure 10 Axial sonograms through a lower thoracic (A) and a mid-lumbar (B) spinal segment. Posterior neural arch ossification is well developed.

5 J Ultrasound Med 6: , 1987 FILLY ET AL 635 Figure 10 Axial sonograms through a lower thoracic (A) and a mid-lumbar (B) spinal segment. Posterior neural arch ossification is well developed. The laminae (L) are nearly apposed in the posterior midline. ES, erector spinal muscle group; e, ossification of the centrum; se, spinal cord; R, rib; K, kidney; N, neurocentral synchondrosis. C, 0, mid-sagittal sonograms of the fetal spine. SP, cartilaginous spinous processes; arrows, delimit dural sac; se, spinal cord with visible central canal; em, conus medullaris; L, laminae. DISCUSSION With routine fetal surveillance techniques such as MSAFP screening and obstetrical ultrasound being ad ~ vocated, the demand for high-resolution scanning of the fetal spine will continue to increase. This study has at tempted to improve our understanding of the sonographic depiction of the varying degrees of maturation of spinal ossification at those stages of development when, as practitioners, we are most frequently called upon to assess normalcy of the fetal spine. One of the potential pitfalls of lack of this knowledge has been previously demonstrated. 8 When normal cephalocaudal maturation of the fetal spine was not taken into account mis+ diagnoses resulted; further, errors in the ultrasound lit~ erature tend to augment the potential for misdiagnosis, For example, one lengthy report describes the most common finding in spina bifida as "absent" posterior ossification centers' when, in fact, the centers are present but abnormally formed, a point discussed later in this paper. While there are some exceptions, ossification in the neural arch first appears at the base of the transverse process,j Early posterior neural arch ossification then progresses anteriorly into the pedides, also contributing a portion of the vertebral body, and posteriorly into the laminae. Additionally, craniocaudal extension into the articular processes and lateral extension into the transverse process occurs. As previously noted, the critical observation in the diagnosis of open spinal neural tube defects is the demonstration of spina bifida (seen as an outward flaring of the posterior arch ossification centers) (Fig. 1). The ideal situation to confirm normalcy '. would be to observe the antithesis (Figs. 2, SA, 8, and 10) of this pathologic state (i.e., inward angulation of the laminae). This indeed is the case when visible ossification is present in the normal laminae. Unfortunately, the current study demonstrates that there is insufficient ossification of the laminae to perceive inward angulation of the posterior neural arch ossification centers to confirm normalcy of the fetal spine during the crucial stage of gestation when this sonographic diagnosis must be

636 FETAL SPINE MORPHOLOGY J Ultrasound Med 6:631-636, 1987 made (18-22 menstrual weeks) (Figs. 3A, 4, 6, and 9).

6 636 FETAL SPINE MORPHOLOGY J Ultrasound Med 6: , 1987 made (18-22 menstrual weeks) (Figs. 3A, 4, 6, and 9). This is particularly important when considering the most common location of such lesions (i.e., lumbar and sacral regions). We did not test the influence of sex (male, female) on the sonographically visualized appearance of the posterior neural arch. It is clear from prior research that throughout the fetal skeletal system female fetuses mature more rapidly than males. l.2 This indicates that male fetuses, with less weli-established spinal maturation, will be even more difficult to evaluate in early pregnancy than females. We further recognize that our method of confirming gestational age, while relatively accurate, is not precise. Fortunately, the method employed should be most accurate at the earliest gestational ages evaluated. From a clinical diagnostic perspective, the younger fe.. tuses (18-22 weeks) are the more important. Roentgenographic studies have been performed on abortuses and have concentrated predominantly on the time of appearance of spinal ossification centers. 1-3 The time of appearance of ossification centers is not cephalocaudal. The appearance of the cenrum varies dramaticaliy from this concept; however, the sonographicaliy more important posterior arch ossification centers do generally adhere, with some variation, to the cephalocaudal maturation rule. Variation in the cephalocaudal "rule" may also be present in the ossification of the laminae, but not observed in our study since our major interest was in general trends of ossification. Additionally, we employed a much lower resolution system (ultrasonography of living fetuses) than earlier studies (roentgenography on abortuses). Further to this point, our description of progress of ossification in the laminae of the sacral spine is also in the "general" sense. In point of fact, our observations in the sacral spine were largely confined to Flecker showed that the earliest appearance of the 53 and 54 posterior arch ossification centers were I8! weeks for female fetuses and 20 weeks for male fetuses. Indeed, ossification of the neural arch of S5 does not even begin until well into the third trimester. We do not feel that these inaccuracies in our study will impact significantly when our results are applied to clinical practice. ]n conclusion, awareness of changes occurring with maturation of the fetal spine will help to prevent false diagnoses of spinal dysraphism based on a misguided inability to document posterior neural arch ossification. Knowledge of the degree of ossification of the laminae at various stages of fetal development, especially between the 18th and 22nd weeks when early sonographic evaluation for open spinal defects is most often requested, will hopefully aid sonologists when they perform the difficult task of diagnosing spina bifida. REFERENCES 1. Flecker H: Time of appearance and fusion of ossification centers as observed by roentgenographic methods. AJR 47:97, O'Rahilly R, Meyer DB: Roentgenographic investigation of the human skeleton during early fetal life. AJR 76:455, Bagnall KM, Harris PF, James RM: A radiographic study of the human fetal spine. 2 The sequence of development of ossification centers in the vertebral column. J Anat 124:791, Noback CR, Robertson GG: Sequences of appearance of ossification centers in the human skeleton during the first five prenatal months. Am J Anat 89:1, Cochlin DL: Ultrasound of the fetal spine. Clin Radiol 33:641, Bagnall KM, Harris PF, Jones PRM: A radiographic study of variations of the human fetal spine. Anat Rec 208:265, deelejalde MM, Elejalde BR: Visualization of the fetal spine: A proposal of a standard system to increase reliabil~ ity. Am J Med Genet 21!445, Dennis MA, Drose JA, Pretorius DH, et al: Normal fetal sacrum simulating spina bifida: "Pseudodysraphism." Ra ~ diology 155:751, 1985

ISUOG Basic Training. Examining Fetal Anatomy from Longitudinal Sections Titia Cohen-Overbeek, The Netherlands

ISUOG Basic Training. Examining Fetal Anatomy from Longitudinal Sections Titia Cohen-Overbeek, The Netherlands ISUOG Basic Training Examining Fetal Anatomy from Longitudinal Sections Titia Cohen-Overbeek, The Netherlands Learning objectives 2 & 3 At the end of the lecture you will be able to: describe how to obtain