THE ANATOMICAL RECORD 227:363-372 (1990)
Ultrasound Investigation of Fetal Human Upper Respiratory Anatomy VICTORIA P. WOLFSON AND JEFFREY T. LAITMAN Departments of Cell BiologylAnatomy (V.P.W., J.T.L.) and Otolaryngology (J.T.L.), Mount Sinai School of Medicine, New York, New York
ABSTRACT
Although the human upper respiratory-upper digestive tract is an area of vital importance, relatively little is known about either the structural or functional changes that occur in the region during the fetal period. While investigations in our laboratory have begun to chart these changes through the use of postmortem materials, in vivo studies have been rarely attempted. This study combines ultrasonography with new applications of video editing t o examine aspects of prenatal upper respiratory development. Structures of the fetal upper respiratory-digestive tract and their movements were studied through the use of ultrasonography and detailed frame-by-frame analysis. Twenty-five living fetuses, aged 18-36 weeks gestation, were studied in utero during routine diagnostic ultrasound examination. These real-time linear array sonograms were videotaped during each study. Videotapes were next analyzed for anatomical structures and movement patterns, played back through the ultrasound machine in normal speed, and then examined with a frame-by-frame video editor (FFVE) to identify structures and movements. Still images were photographed directly from the video monitor using a 35 mm camera. Results show that upper respiratory and digestive structures, as well as their movements, could be seen clearly during normal speed and repeat frame-by-frame analysis. Major structures that could be identified in the majority of subjects included trachea in 20 of 25 fetuses (80%); larynx, 76%; pharynx, 76%. Smaller structures were more variable, but were nevertheless observed on both sagittal and coronal section: piriform sinuses, 76%; thyroid cartilage, 36%; cricoid cartilage, 32%; and epiglottis, 16%. Movements of structures could also be seen and were those typically observed in connection with swallowing: fluttering tongue movements, changes in pharyngeal shape, and passage of a bolus via the piriform sinuses to esophagus. Fetal swallows had minimal laryngeal motion. This study represents the first time that the appearance of upper airway and digestive tract structures have been quantified in conjunction with their movements in the living fetus. The human upper respiratory-digestive tract subsumes the three vital functions of respiration, deglutition, and phonation. Although it is an area of great clinical importance, being the site of many developmental pathologies, its fetal and perinatal anatomy and function have not been fully explored. While recent studies on postmortem fetal material have begun t o document developmental change (Magriples and Laitman, 1987; Moore and Laitman, 1989), the nature of postmortem examinations does not allow for precise functional interpretation. Full clarification of positional relationships of structures similarly requires observation of the fetal upper airway within a natural setting. Because of inherent obstacles of in utero examination of live human fetuses, this has rarely been attempted. With the advent and development of clinical obstetric ultrasonography, however, it has become possible to observe the living prenatal human. Knowledge of both fetal and postnatal laryngeal de0 1990 WILEY-LISS, INC.
velopment is currently founded on postmortem studies and postnatal clinical radiologic examinations. These studies indicate that the larynx of both third-trimester (27 weeks to term when based on a 40 week clinical pregnancy) human fetuses (Magriples and Laitman, 1987; Moore and Laitman, 1989) and newborn infants are placed high in the neck. The position of the larynx at these ages, from epiglottic tip to caudal border of the cricoid cartilage, usually corresponds to the level of the first to third or fourth cervical vertebrae (News, 1949; Roche and Barkla, 1965; Crelin, 1973; Laitman and Crelin, 1976, 1980; Laitman et al., 1978; Sasaki et al.,
Received June 15, 1989; accepted November 29, 1989. Address reprint requests to Dr. Jeffrey T. Laitman, Department of Cell BiologyiAnatomy, Mount Sinai School of Medicine, Box 1007, 1 Gustave Levy Place, New York, NY 10029-6574.
364
V.P. WOLFSON AND J.T. LAITMAN
1977). This high laryngeal position permits the formation of a “two-tube system,” a virtual separation between the respiratory and digestive tracts. The epiglottis can interlock with the soft palate, creating a direct passage for respiration. During swallowing, a liquid bolus can flow laterally around the larynx via the isthmus faucium, through the piriform sinuses, to the esophagus without interrupting the direct path of air in the functionally separate respiratory tract. This two-tube structure permits the separation of the respiratory and digestive tracts, allowing for simultaneous breathing and swallowing (MOSS, 1965; Crelin, 1976; Sasaki, 1977; Laitman and Crelin, 1980). The high position of the fetal human larynx is maintained for some time after birth, before the larynx begins its descent to the permanent lower adult position, usually corresponding to the third through the lower border of the sixth cervical vertebrae (Roche and Barkla, 1965; Laitman and Crelin, 1976; Sasaki, 1977; Crelin, 1987). Although the exact timing of this structural descent is unknown, radiographic studies have shown that a t approximately 4 to 6 months of age functional changes occur that permit the infant increased ease of oral-tidal respiration (Laitman and Crelin, 1980; Laitman, 1988). This functional change also falls within the time period coincident with the peak occurrence of sudden infant death syndrome (SIDS) (Golding e t al., 1985; Hoffman e t al., 19881, which has suggested to some a possible relationship (Sasaki, 1979; Laitman, 1986; Beckwith, 1988). Permanent structural descent of the larynx begins in the second year of life, as the larynx begins its gradual descent into the neck (Roche and Barkla, 1965; Crelin, 1976; Laitman and Crelin, 1980). Examination of the fetal and perinatal upper respiratory tract has several limitations because of both limited access to postmortem material and problems examining in vivo anatomy and function. For example, i t has, until recently, been largely impossible to explore movements of fetal anatomic structures in vivo. Cineradiographic studies of infants and young children have begun to offer some insight into postnatal activity (Ardran and Kemp, 1970; Laitman et al., 1977; Bosma, 1985) but have not been of value for in utero investigation because of the potential danger for the mother and fetus (Stewart, 1973; Wagner et al., 1985). Invasive techniques, such a s esophageal pressure studies (direct manometry; e.g., Gryboski, 1965), nasal occlusion studies (e.g., Swift and Emery, 1973; Rodenstein e t al., 1985) or intra-amniotic contrast studies (e.g., Davis and Potter, 1946; Pritchard, 1965) are similarly impractical. As a result, fetal upper respiratory and digestive activities have been largely unexplored. The introduction of high-resolution ultrasonography and its routine clinical use have now made it possible to study these areas noninvasively. This study has combined anatomic investigation with new techniques of ultrasound study, permitting detailed in utero examination as well as a sequential analysis of movement. It is a preliminary investigation of the uses of these sonographic techniques to further understanding of the components of the upper respiratory and digestive tracts and of the interactions of these systems in the living human fetus.
MATERIALS AND METHODS
The upper respiratory tracts of 25 fetuses were examined using routine diagnostic obstetric ultrasound. Sonograms were performed for such reasons a s dating pregnancy, observing fetal growth, diagnosing multiple pregnancies, localizing placenta, detecting congenital abnormalities, assessing amniotic fluid, and determining fetal position (Bennett, 1985; Filkins and Hadlock, 1985; Bowie, 1986). No distinguishable fetal respiratory-digestive tract abnormalities were observed in our sample. Ultrasound is considered a safe means of examining the mother while inflicting no damage on the fetus (Wagner et al., 1985). Tabulations of fetal age determination are given in Table 1. Fetal gestational age in this sample ranged between 18 and 36 weeks based on a 40 week clinical pregnancy. Gestational age is determined by 1)maternal last menstrual period; and 2) the Mount Sinai U1trasound Computer Program, a composite age dating system based on the growth parameters of biparietal diameter, femur length, head circumference, and abdominal circumference as compared with a normally distributed population (Berkowitz, unpublished program). Last menstrual period is a relatively accurate assessment of fetal gestational age when the date is known and the mother’s cycle is regular (23 weeks, taking into account conditions under which the mother’s cycles are irregular and the date is not necessarily accurate; Bowie, 1986). Ultrasound determination of gestational age is most precise during the first trimester (1-12 weeks gestation, + 1 week), less so during the second trimester (13-26 weeks gestation, 2 2 weeks), and least during the third trimester (27 weeks gestation to term, 2 3 weeks; Hohler, 1984; Bowie, 1986; Hadlock et al., 1986). It should be noted that, if based on actual date of conception, fetal ages could be 2 weeks younger than reported in Table 1. Study of the upper respiratory tract in human fetuses has been restricted by the inherent limitations of ultrasound equipment. Image selection, for example, is traditionally made by freezing the moving image on the ultrasound console monitor during the examination. Still images are then made from that frozen frame. Because not all frames can be photographed, a large amount of image material is lost. In addition, the ability to scrutinize precise movements that occur on the screen is sacrificed. The advent of advanced video editing systems now makes it possible to review a n ultrasound examination in its entirety. Re-examination of sonograms on a frame-by-frame basis filmed to appear a t the original speed of movement (“real-time” movement) vastly increases the number of still images available and improves the ability to observe the finer aspects of movement (Goto and Kato, 1983; Cooper et al., 1985). The final still image may be recorded with equipment ranging from cameras to computer terminals (Shawker et al., 1984; Thieme et al., 1985). Structures of the fetal upper respiratory-digestive tract and their movement were studied via ultrasonography and sequential still-frame analysis. All sonograms were performed by a perinatologist (with a t least one member of the investigating team in attendance) using a JVC BR 64000 ultrasound scanner with a linear array transducer. These real-time images were
ULTRASOUND STUDY OF FETAL ANATOMY
365
TABLE 1. Fetal dimensions Subject No. 1
Age by date (weeks)' 17
Age by sonogram (weeks) 19.0
Femur length (cm) 2.9
Biparietal diameter (cm) 4.2
Head circumference (cm) 16.1
Abdominal circumference (cm) 14.1
17 +
2.4 2.3 3.2
3.9 4.0 4.5
14.5 14.9 17.7
12.2 12.9 15.7
+
2 3 4
182 -
17.5 18.4 19.9
5 6
20 + 21 +
20.4 21.3
3.3 3.7
4.7 5.0
17.6 19.2
16.1 15.7
7 8 9 103
22 22 + 23 23
+
22.4 22.9 23.3
4.2 3.8 4.0
5.3 5.6 5.8
19.9 20.5 21.2
17.7 17.0 16.8
11
23 +
12 13
-
Position Coronal, 314 view, sagittal 314 Supine, sagittal Coronal, supine. prone Prone, sagittal 314 sagittal, supine Sagittal 314, transverse Sagittal, 314 Coronal
-
-
23.3
3.7
5.6
22.0
18.0
Supine, sagittal
24 +
22.9
3.9
5.9
21.1
17.7
Prone, 314
-2
24.8
4.8
6.2
22.9
18.6
5.0 5.1 4.6 5.4 5.4 6.0 5.7
6.6 6.8 6.3 7.0 8.0 7.5 7.9
24.7 26.1 23.6 25.8 29.1 28.4 28.6
22.9 22.4 20.5 24.7 25.1 26.3 27.5
Prone, 314, transverse Coronal Coronal 314, transverse Prone 314 314 Supine, sagittal transverse Prone, sagittal 314 Prone Coronal, 314, sagittal SagGtal
26 + 26 +
-
-
14 15 16 17 18 19 20
272 -
29 29 29 +
26.7 27.4 25.4 28.4 29.1 29.1 31.1
21 22 23 24
29 + 32 + 32 36
31.2 21.0 34.4 34.8
5.7 3.6 6.6 6.8
7.9 5.2 8.4 8.4
29.9 19.1 31.4 31.1
26.1 15.0 32.3 32.0
25
36
35.1
7.0
8.5
32.1
30.5
+
Comments
Choroid plexus cysts Choroid plexus cysts Decreased size for dates
Twin A Twin B
Macrosomia
'Date of the last menstrual period as reported by mother. Ages are based on a 40 week clinical pregnancy. Actual dates of conception may be 2 weeks later. 'Fetal age by date of mother's last menstrual period unknown. 3Fetal dimensions unavailable.
recorded onto a 0.50 inch video cassette linked directly to the ultrasound unit. Tapes were then played back through the ultrasound unit a t normal speed (30 frames per second) and then analyzed for anatomic features and movement patterns. Tapes were then viewed using a Sony BVU800 frame-by-frame video editor (FFVE) for closer assessment of anatomy and specific movement patterns. Still-frame images were then selected and photographed directly from a 17 inch monitor using a 35 mm camera with a 55 mm lens, lending greater clarity and frame selection to the still ultrasound image t h a n has been previously observed. Attempts were made during clinical examination to visualize fetuses in both midsagittal and coronal planes. Observation of live, moving, and generally uncooperative fetuses, however, often makes exact positioning impossible. RESULTS Results are summarized in Tables 2 and 3 and in Figures 1through 4. Table 2 shows the anatomic structures commonly visualized during real-time observation and repeated FFVE and normal speed video scru-
tiny. Figures 1 through 4 illustrate structures visible during ultrasound examination. The fluid-filled vascular conduits of the heart and carotid arteries were seen in all 25 subjects. The vertebral column and portions of the basicranium were readily identified in all subjects because of the highly recognizable morphology and extremely high echogenicity (density). Specific regions of the basicranium, such as the petrous part of the temporal bone or contours of the foramen magnum, were clearly identified in 20% and 32% of subjects, respectively. The bony nasal septum was clearly viewed on coronal section (32%, Fig. 4). Tooth buds were clearly observed in 48% of subjects (Fig. 1). Portions of the larynx were seen in 76% of subjects, defined by anatomical shape and location and by high echogenicity. Well-defined thyroid cartilage was seen in 36% of subjects a s a large, clearly defined, highly echogenic structure in the anterior portion of the neck. It was identified in both sagittal and coronal sections (Figs. 2, 4) but was most clear on coronal section. The cricoid cartilage was also seen (32%of subjects) in both sagittal section in the posterior portion of the larynx
V.P. WOLFSON AND J.T.LAITMAN
366
TABLE 2. Fetal anatomic structures SubVerteject Carotid bra1 Basi- Petrous Foramen Nasal Tooth Tracheal ThyEpi- Phar- Piriform no. Heart art. col. cranium' bone magnum septum buds rings Larynx' roid Cricoid glottis ynx sinus + + + + + + 1 + + + + + + 2 + + + + + + + + + 3 + + + + + + + + + + + 4 + + + + + + + 5 + + + + + + + + 6 + + + + + + + + + + + + 7
+
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
+ + + + + + + + + + + + + + + + + +
+ + + + + + + + + + + + + + + + + +
+
+ +
+
+ + + + + + + + + + + + + + + + + +
+ + + + + + + + + + + + + + + + + +
+
+ + +
+ +
+ + +
+ +
+ + + + + +
+
+ + + + + + + + + +
+ + + + + + + +
+ -4-
+
+
+ +
+ +
+
+ + +
+ +
+
+
+
+ + +
+
+ +
+
+
+ +
+ +
+
+
+
+ +
+
+ +
+ +
+ +
+ +
+
+
+ + + + + + +
+ + + + + + +
'Portions of basicranium seen but not distinguishable as to specific bones. 'Portions of larynx seen but not distinguishable as to individual cartilages.
TABLE 3. Fetal upper respiratory movements Subject No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Heart
Carotid
Mandible
Tongue
Esophagus
Bolus
Pharynx
Swallow
Hiccough
+ + + + + + + + + + + + + + + + + + + + + + + + +
+ + + + + + + + + + + + + + + + + + + + + + + + +
+ + + +
+ + + + + +
+ + + + + + + + + + +
+ + +
+ + + +
+ + + + +
+
+ +
+ + +
+ +
+
+
+ + +
(Fig. 2) and coronal section (Figs. 3, 4). The epiglottis was viewed in 16% of subjects. It was most clearly and frequently seen on coronal section as a small echogenic structure projecting into the anechoic (area of low den-
+ + +
+ + + + + + + +
+
+
+
+ + +
sity) lumen of the pharynx, posterior to the oral cavity (Figs. 3, 41, surrounded laterally by the anechoic piriform sinuses (Fig. 3). The smaller laryngeal cartilages, the arytenoids, corniculates, and cuneiforms could not
ULTRASOUND STUDY OF FETAL ANATOMY
367
Fig. 1 . Sonogram of a n 18 week gestational age fetus, parasagittal view (subject No. 3). b, basicranium; c, cricoid cartilage; e, eyeball; h, heart; 1, lumen of the larynx; p, pharynx; t, tooth b u d tr, tracheal ring; v, vertebral column. The small size of this fetus permits a large area view of the head, neck, and upper thorax.
be clearly identified. The soft palate was also not clearly seen. This was probably due to its being both of similar density and in close proximity to the tongue, thus rendering the structures indistinguishable. Tracheal rings were frequently seen structures (80%) because of their clearly identifiable morphology, size,
and position in the neck. The trachea is the largest and most dense structure in the anterior portion of the middle to lower neck region, and its echogenic cartilages contrast sharply with its anechoic fluid-filled lumen. The pharynx was also frequently observed (76%, especially on coronal section; Fig. 3). The piriform sinuses
368
V.P. WOLFSON AND J.T. LAITMAN
Fig. 2. Sonogram (A) and diagram (B) of a 29 week gestational age fetus, parasagittal view (subject No. 21). c, cricoid cartilage; e, epiglottis; 1, lumen of the larynx; p, pharynx; s, location of soft palate; t, thyroid cartilage; to, location of tongue; tr, tracheal rings, v, vertebral column.
were also seen as lateral projections of the pharynx (76%),at times seen clearly surrounding the epiglottis (Fig. 3). Again, this was most clearly observed on coronal section. Movements commonly seen during real-time and repeat normal speed observations are summarized in Table 3. The rhythmic flow of blood through the heart and
carotid arteries was observed in all fetuses and is, indeed, a feature routinely sought during all ultrasound examinations. Most movements of upper respiratory structures were typically observed in connection with swallowing. Fluttering tongue movements and esophageal passage of a liquid bolus were movements typically seen. Swallowing motions began with prelimi-
ULTRASOUND STUDY OF FETAL ANATOMY
369
D
Fig. 3. A,B,C: Progressively dorsal coronal sonograms; D: composite diagram of a 23 week gestational age fetus (subject No. 10). c, cricoid cartilage; e, epiglottis; h, heart; p, pharynx; ps, piriform sinus; tl, tracheal lumen. Dark hatching in D shows region of the eyes and
orbits as can be seen in A and B; light shading indicates pharynx and piriform sinuses a s can be seen in B and C. Note the high position of the larynx, particularly the epiglottis, in C.
370
V.P. WOLFSON AND J.T. LAITMAN
Fig. 4. Sonogram (A) and diagram (B) of a 26 week gestational age fetus, coronal view (subject No. 15).c, cricoid cartilage; e, epiglottis; n, nasal cavityinasopharynx; ns, nasal septum; 0,orbit; t, thyroid carti-
lage; tl, tracheal lumen; t r , tracheal ring. This view well illustrates both the larger laryngeal cartilages and the high position of the larynx itself.
ULTRASOUND STUDY O F F E T A L ANATOMY
nary fluttering movements of the tongue, followed by narrowing and constriction of the pharyngeal space. Minimal vertical movements of the larynx were observed. Of particular interest was a n episode of repeated spasmodic movements of the diaphragm and larynx, reminiscent of that seen during adult “hiccoughs)’ (Table 3). Identification of upper respiratory structures did not seem to be related to fetal age, as no discernible agerelated patterns in the percentages of structures seen were found. At least within this limited sample, the ability to identify fetal anatomy seems to be related more to fetal position than to age.
371
in this region nevertheless suggest a n interlocking between the soft palate and epiglottis. In all instances in which the epiglottis was identified (16% of specimens) the structure was observed very high in the throat. Furthermore, in our eight observed cases of “swallows” and 11 instances of “pharyngeal movements,” the larynx was seen to exhibit minimal vertical movement. This absence of movement is compatible with a larynx positioned high in the neck during swallowing and is totally unlike the adult condition in which the Iarynx exhibits considerable movement as the epiglottis folds backwards during deglutition (Crelin, 1976; Laitman and Crelin, 1980; Shawker et al., 1984). These two sets of observations, coupled with the lack of any observaDISCUSSION tions indicating the contrary, strongly suggest the inBecause of inherent obstacles of in vivo fetal, infant, terlocking of the epiglottis and soft palate. This interand early childhood study, anatomic investigations of locking suggests the creation of separate respiratory the human upper respiratory region have involved and digestive tracts, i.e., the “two-tube’’ system, prenamainly postmortem material. Postmortem studies of tally. the fetus and infant have contributed a preliminary The clear identification of a number of upper digesunderstanding of upper airway development. In addi- tive structures allowed for observations of the sequence tion, clinical and investigational techniques such as of movements found in a fetal swallow. These movecineradiography, preabortion intra-amniotic injection ments included the passage of a bolus of amniotic fluid of radiocontrast medium, and esophageal manometry from the oral cavity into the pharynx; its continued have delineated the rudiments of function (e.g., Davis caudal movement indicated by narrowing of the phaand Potter, 1946; Gryboski, 1965; Pritchard, 1965; ryngeal space; passage of the bolus around the larynx Ardran and Kemp, 1970; Laitman et al., 1977; Laitman via the piriform sinuses (rather than around and over and Crelin, 1980; Bosma, 1985). These techniques, the larynx, as in adults); and limited up-down larynhowever, are impractical for extensive fetal study. It is geal movement. Esophageal movement and limited thus necessary to find a method th at allows for obser- peristalsis were also seen, supporting previous observation of the living fetus in order to understand fully vations of a n early onset of gastrointestinal function the development and functioning of the upper respira- (Abromovich e t al., 1979; Gryboski and Walker, 1983). The further advancement of ultrasound and recordtory and digestive tracts. With the advent of high-resolution ultrasonography, it has become possible to ing techniques, such as those used in this study, will study noninvasively the specifics of anatomy and to provide a n understanding not only of normal anatombegin to observe function within the fetus’ normal mi- ical development but also of pathology. For example, better understanding of this region in the living fetus lieu. This study examined living fetuses to quantify the could be of great value in diagnosing neonatal and inincidence of appearance of upper respiratory and diges- fant upper respiratory pathologies. Ultrasound obsertive tract structures and to chart their movements. Re- vations indicating unusual differences in density of sults indicate that large portions of the larynx and laryngeal cartilages could be suggestive of develpharynx can be visualized in the majority of middle to opmental weakness of laryngeal cartilages (laryngolate second- and early third-trimester fetuses using the malacia). Similarly, peculiarities in the shape of the combination of ultrasonography and repeat, normal pharyngeal lumen could suggest muscular flaccidity, speed, video examination. Specific identification of the which may later be involved with the respiratory individual laryngeal cartilages is, however, variable aspects of SIDS. The prenatal appreciation of such under real-time conditions. Through our application of compromising pathologies would greatly aid in the FFVE to videotaped ultrasound examinations, it has planning and implementation of perinatal treatment. been possible to see the larger laryngeal cartilages, Clearly, the advent of imaging techniques such as ulsuch as the epiglottis, thyroid, and cricoid. While prior trasonography enables exploration of the growth and ultrasound studies have visualized some aspects of the development of the upper respiratory region far in exupper respiratory region (Bowie and Clair, 1982; Coo- cess of previous capabilities. per e t al., 1985; Staudach, 1987), they have not focused ACKNOWLEDGMENTS on specific aspects of laryngeal anatomy. In addition, The authors express their appreciation to Dr. u. none have attempted to quantify the incidence of appearance of laryngeal structures in the live, in utero Chitkara, Department of Obstetrics, Gynecology and fetus. Through the use of ultrasound and FFVE we Reproduction Science of Mount Sinai School of Mediattempted to do both, and, in addition, we found it pos- cine, for ongoing help, advice, and generous access to sible to produce high-quality views of regions such as ultrasound equipment; Dr. T. Berkowitz, for sharing her expertise on fetal gestational age assessment; and the pharynx and trachea. It was not possible to delineate precisely the struc- G. Maffei, N. Katz, and P.J. Gannon for advice and ture of the soft palate. This was due to both normal invaluable technical assistance in photography. We difficulties produced by differing fetal positions and the also thank Drs. A. Eden and J.S. Reidenberg for both soft palate’s thin structure and similarity in density their constant support throughout our work and their and close proximity to the tongue. The movements seen comments on the manuscript. The ongoing support of
V.P. WOLFSON AND J.T. LAITMAN
372
the Departments of Cell Biology/Anatomy and Otolaryngology of the Mount Sinai School of Medicine is greatly appreciated. This project was supported in part by NIH grant T35 DK07420. LITERATURE CITED Abromovich, D.R., H. Garden, L. Jandial, and K.R. Page 1979 Fetal swallowing and voiding in relation to hydramnios. Obstet. Gynecol. Surg., 54t15-20. Ardran, G.M., and F.H. Kemp 1970 The nasal and cervical airway in sleep in the neonatal period. Am. J. Roentgenol., 108t537-542. Beckwith, J.B. 1988 Intrathoracic petechial hemorrhages: A clue to the mechanism of death in sudden infant death syndrome? Ann. N.Y. Acad. Sci., 533t37-47. Bennett, J.B. 1985 Ultrasonography for detection of fetal disorders in the first and second trimesters of pregnancy. In: Human Prenatal Diagnosis. K. Filkins and J.F. Russo, eds. Marcel Dekker, New York, pp. 183-198. Berkowitz, T. Mount Sinai Ultrasound Fetal Age Assessment Computer Program. Mount Sinai School of Medicine, Department of Obstetrics, Gynecology and Reproduction Science, New York (unpublished). Bosma, J.F. 1985 Postnatal ontogeny of performances of the pharynx, larynx and mouth. Am. Rev. Respir. Dis., 131:SlO-S15. Bawie, J.D. 1986 Ultrasound fetal measurements and IUGR. Clin. Diagn. Ultrasound, 19:ll-30. Bowie, J.D., and M.R. Clair 1982 Fetal swallowing and regurgitation: Observation of normal and abnormal activity. Radiology, 144: 877-878. Cooper, C., B.S. Mahony, J.D. Bowie, T.O. Albright, and P.W. Callen 1985 Ultrasound evaluation of the normal fetal upper airway and esophagus. J. Ultrasound Med., 4t343-346. Crelin, E.S. 1973 Functional Anatomy of the Newborn. Yale University Press, New Haven. Crelin, E.S. 1976 Development of the upper respiratory system. Ciba Clin. Symp., 28:3-30. Crelin, E.S. 1987 The Human Vocal Tract Anatomy, Function, Development and Evolution. New York: Vantage. Davis, M.E., and E.L. Potter 1946 Intrauterine respiration of the human fetus. J.A.M.A. 131t1194-1201. Filkins, K., and F.P. Hadlock 1985 Ultrasonography for the detection of fetal disorders in the third trimester of pregnancy. In: Human Prenatal Diagnosis. K. Filkins and J.F. Russo, eds. Marcel Dekker, New York, pp. 199-269. Golding, J., S. Limerick, and A. Macfarlane 1985 Sudden Infant Death: Patterns, Puzzles and Problems. University of Washington, Seattle. Goto, S., and T.K. Kato 1983 Early fetal movements are useful for estimating the gestational weeks in the first trimester of pregnancy. Ultrasound Med. Biol., S2t577-582. Gryboski, J.D. 1965 The swallowing mechanism of the neonate: I. Esophageal and gastric motility. Pediatrics, 35t445-452. Gryboski, J.D., and W.A. Walker 1983 Gastrointestinal Problems in the Infant. W.B. Saunders, Philadelphia. Hadlock, F.P., R.L. Deter, and R.B. Harrist 1986 Estimation of gestational age in the third trimester. Clin. Diagn. Ultrasound, 19: 1-10, Hohler, C.W. 1984 Ultrasound estimation of gestational age. Clin. Obstet. Gynecol., 27:314-326. Hoffman, H.J., K. Damus, L. Hillman, and E. Krongrad 1988 Risk factors for SIDS Results of the National Institute of Child Health
and Human Development SIDS Cooperative Epidemiological Study. Ann. N.Y. Acad. Sci., 533:13-30. Laitman, J.T. 1986 On J.J. McKenna’s “An anthropological perspective on the sudden infant death syndrome: The role of parental breathing cues and speech breathing adaptations.” Med. Anthropol., 10:65-68. Laitman. J.T. 1988 Prenatal and infant transitions in the development of the human upper respiratory tract. Assoc. Res. Otolaryngol. Abstr., 11:160-161. Laitman, J.T., and E.S. Crelin 1976 Postnatal development of the basicranium and vocal tract region in man. In: Symposium on Development of the Basicranium. J.F. Bosma, ed. U S . Government Printing Office, Washington, D.C., pp. 206-220. Laitman, J.T., and E.S. Crelin 1980 Developmental change in the upper respiratory system of human infants. Perinatol. Neonatol., 4r15-22. Laitman, J.T., E.S. Crelin, and G.J. Conlogue 1977 The function of the epiglottis in monkey and man. Yale J. Biol. Med., 50:43-48. Laitman, J.T., R.C. Heimbuch, and E.S. Crelin 1978 Developmental change in a basicranial line and its relationship to the upper respiratory system in living primates. Am. J . Anat., 152t467482. Magriples, U., and J.T. Laitman 1987 Developmental change in the position of the fetal human larynx. Am. J . Phys. Anthropol., 72: 463-472. Moore, S.M., and J.T. Laitman 1989 Development of the fetal human upper respiratory tract during the second trimester. Am. J . Phys. Anthropol., 78t274-275. Moss, M.L. 1965 The veloepiglottic sphincter and obligate nose breathing in the neonate. Brief Clin. Lab. Observ., 67:330-331. Negus, V.E. 1949 The Comparative Anatomy and Physiology of the Larynx. Grune & Stratton, New York. Pritchard, J.A. 1965 Deglutition by normal and anencephalic fetuses. Obstet. Gynecol., 25t289-297. Roche, A.F., and B.D. Barkla 1965 The level of the larynx during childhood. Ann. Otol. Rhinol. Laryngol., 74t645-654. Rodenstein, D.O., N. Perlmutter, and D.C. Stanescu 1985 Infants are not obligatory nasal breathers. Am. Rev. Respir. Dis., 131t343347. Sasaki, C.T. 1977 Physiology of the larynx. In: Otolaryngology,Vol. 3. Harper & Row Publishers, Hagerstown, MD, pp. 1-21. Sasaki, C.T. 1979 Development of the laryngeal function: Etiologic significance of the sudden infant death syndrome. Laryngoscope, 89t1964-1982. Sasaki, C.T., P.A. Levine, J.T. Laitman, and E.S. Crelin 1977 Postnatal descent of the epiglottis in man. Arch. Otolaryngol., 103:169171. Shawker, T.H., B. Sonies, T.E. Hall, and B.F. Baum 1984 Ultrasound analysis of tongue, hyoid and larynx activity during swallowing. Invest. Radiol., 19t82-86. Staudach, A. 1987 Sectional Fetal Anatomy in Ultrasound. SpringerVerlag, New York. Stewart, A.M. 1973 Cancer a s a cause of abortions and stillbirths: The effect of these early deaths on the recognition of radiogenic leukaemias. Br. J . Cancer, 27t465-472. Swift, P.G.F., and J.L. Emery 1973 Clinical observations on response to nasal occlusion in infancy. Arch. Dis. Child. 48t947-951. Thieme, G.A., R.R. Price, and A.E. James 1985 Ultrasound instrumentation and its practical applications. In: The Principles and Practice of Ultrasonography in Obstetrics and Gynecology. R.C. Sanders and A.E. James, eds. Appleton-Century-Crofts,Norwalk, CT, pp. 23-59. Wagner, L.K., R.G. Lester, and L.R. Saldana 1985 Exposure of the Pregnant Patient to Diagnostic Radiations: A Guide to Medical Management. J.B. Lippincott Company, Philadelphia.
Ultrasound Investigation of Fetal Human Upper Respiratory Anatomy VICTORIA P. WOLFSON AND JEFFREY T. LAITMAN Departments of Cell BiologylAnatomy (V.P.W., J.T.L.) and Otolaryngology (J.T.L.), Mount Sinai School of Medicine, New York, New York
ABSTRACT
Although the human upper respiratory-upper digestive tract is an area of vital importance, relatively little is known about either the structural or functional changes that occur in the region during the fetal period. While investigations in our laboratory have begun to chart these changes through the use of postmortem materials, in vivo studies have been rarely attempted. This study combines ultrasonography with new applications of video editing t o examine aspects of prenatal upper respiratory development. Structures of the fetal upper respiratory-digestive tract and their movements were studied through the use of ultrasonography and detailed frame-by-frame analysis. Twenty-five living fetuses, aged 18-36 weeks gestation, were studied in utero during routine diagnostic ultrasound examination. These real-time linear array sonograms were videotaped during each study. Videotapes were next analyzed for anatomical structures and movement patterns, played back through the ultrasound machine in normal speed, and then examined with a frame-by-frame video editor (FFVE) to identify structures and movements. Still images were photographed directly from the video monitor using a 35 mm camera. Results show that upper respiratory and digestive structures, as well as their movements, could be seen clearly during normal speed and repeat frame-by-frame analysis. Major structures that could be identified in the majority of subjects included trachea in 20 of 25 fetuses (80%); larynx, 76%; pharynx, 76%. Smaller structures were more variable, but were nevertheless observed on both sagittal and coronal section: piriform sinuses, 76%; thyroid cartilage, 36%; cricoid cartilage, 32%; and epiglottis, 16%. Movements of structures could also be seen and were those typically observed in connection with swallowing: fluttering tongue movements, changes in pharyngeal shape, and passage of a bolus via the piriform sinuses to esophagus. Fetal swallows had minimal laryngeal motion. This study represents the first time that the appearance of upper airway and digestive tract structures have been quantified in conjunction with their movements in the living fetus. The human upper respiratory-digestive tract subsumes the three vital functions of respiration, deglutition, and phonation. Although it is an area of great clinical importance, being the site of many developmental pathologies, its fetal and perinatal anatomy and function have not been fully explored. While recent studies on postmortem fetal material have begun t o document developmental change (Magriples and Laitman, 1987; Moore and Laitman, 1989), the nature of postmortem examinations does not allow for precise functional interpretation. Full clarification of positional relationships of structures similarly requires observation of the fetal upper airway within a natural setting. Because of inherent obstacles of in utero examination of live human fetuses, this has rarely been attempted. With the advent and development of clinical obstetric ultrasonography, however, it has become possible to observe the living prenatal human. Knowledge of both fetal and postnatal laryngeal de0 1990 WILEY-LISS, INC.
velopment is currently founded on postmortem studies and postnatal clinical radiologic examinations. These studies indicate that the larynx of both third-trimester (27 weeks to term when based on a 40 week clinical pregnancy) human fetuses (Magriples and Laitman, 1987; Moore and Laitman, 1989) and newborn infants are placed high in the neck. The position of the larynx at these ages, from epiglottic tip to caudal border of the cricoid cartilage, usually corresponds to the level of the first to third or fourth cervical vertebrae (News, 1949; Roche and Barkla, 1965; Crelin, 1973; Laitman and Crelin, 1976, 1980; Laitman et al., 1978; Sasaki et al.,
Received June 15, 1989; accepted November 29, 1989. Address reprint requests to Dr. Jeffrey T. Laitman, Department of Cell BiologyiAnatomy, Mount Sinai School of Medicine, Box 1007, 1 Gustave Levy Place, New York, NY 10029-6574.
364
V.P. WOLFSON AND J.T. LAITMAN
1977). This high laryngeal position permits the formation of a “two-tube system,” a virtual separation between the respiratory and digestive tracts. The epiglottis can interlock with the soft palate, creating a direct passage for respiration. During swallowing, a liquid bolus can flow laterally around the larynx via the isthmus faucium, through the piriform sinuses, to the esophagus without interrupting the direct path of air in the functionally separate respiratory tract. This two-tube structure permits the separation of the respiratory and digestive tracts, allowing for simultaneous breathing and swallowing (MOSS, 1965; Crelin, 1976; Sasaki, 1977; Laitman and Crelin, 1980). The high position of the fetal human larynx is maintained for some time after birth, before the larynx begins its descent to the permanent lower adult position, usually corresponding to the third through the lower border of the sixth cervical vertebrae (Roche and Barkla, 1965; Laitman and Crelin, 1976; Sasaki, 1977; Crelin, 1987). Although the exact timing of this structural descent is unknown, radiographic studies have shown that a t approximately 4 to 6 months of age functional changes occur that permit the infant increased ease of oral-tidal respiration (Laitman and Crelin, 1980; Laitman, 1988). This functional change also falls within the time period coincident with the peak occurrence of sudden infant death syndrome (SIDS) (Golding e t al., 1985; Hoffman e t al., 19881, which has suggested to some a possible relationship (Sasaki, 1979; Laitman, 1986; Beckwith, 1988). Permanent structural descent of the larynx begins in the second year of life, as the larynx begins its gradual descent into the neck (Roche and Barkla, 1965; Crelin, 1976; Laitman and Crelin, 1980). Examination of the fetal and perinatal upper respiratory tract has several limitations because of both limited access to postmortem material and problems examining in vivo anatomy and function. For example, i t has, until recently, been largely impossible to explore movements of fetal anatomic structures in vivo. Cineradiographic studies of infants and young children have begun to offer some insight into postnatal activity (Ardran and Kemp, 1970; Laitman et al., 1977; Bosma, 1985) but have not been of value for in utero investigation because of the potential danger for the mother and fetus (Stewart, 1973; Wagner et al., 1985). Invasive techniques, such a s esophageal pressure studies (direct manometry; e.g., Gryboski, 1965), nasal occlusion studies (e.g., Swift and Emery, 1973; Rodenstein e t al., 1985) or intra-amniotic contrast studies (e.g., Davis and Potter, 1946; Pritchard, 1965) are similarly impractical. As a result, fetal upper respiratory and digestive activities have been largely unexplored. The introduction of high-resolution ultrasonography and its routine clinical use have now made it possible to study these areas noninvasively. This study has combined anatomic investigation with new techniques of ultrasound study, permitting detailed in utero examination as well as a sequential analysis of movement. It is a preliminary investigation of the uses of these sonographic techniques to further understanding of the components of the upper respiratory and digestive tracts and of the interactions of these systems in the living human fetus.
MATERIALS AND METHODS
The upper respiratory tracts of 25 fetuses were examined using routine diagnostic obstetric ultrasound. Sonograms were performed for such reasons a s dating pregnancy, observing fetal growth, diagnosing multiple pregnancies, localizing placenta, detecting congenital abnormalities, assessing amniotic fluid, and determining fetal position (Bennett, 1985; Filkins and Hadlock, 1985; Bowie, 1986). No distinguishable fetal respiratory-digestive tract abnormalities were observed in our sample. Ultrasound is considered a safe means of examining the mother while inflicting no damage on the fetus (Wagner et al., 1985). Tabulations of fetal age determination are given in Table 1. Fetal gestational age in this sample ranged between 18 and 36 weeks based on a 40 week clinical pregnancy. Gestational age is determined by 1)maternal last menstrual period; and 2) the Mount Sinai U1trasound Computer Program, a composite age dating system based on the growth parameters of biparietal diameter, femur length, head circumference, and abdominal circumference as compared with a normally distributed population (Berkowitz, unpublished program). Last menstrual period is a relatively accurate assessment of fetal gestational age when the date is known and the mother’s cycle is regular (23 weeks, taking into account conditions under which the mother’s cycles are irregular and the date is not necessarily accurate; Bowie, 1986). Ultrasound determination of gestational age is most precise during the first trimester (1-12 weeks gestation, + 1 week), less so during the second trimester (13-26 weeks gestation, 2 2 weeks), and least during the third trimester (27 weeks gestation to term, 2 3 weeks; Hohler, 1984; Bowie, 1986; Hadlock et al., 1986). It should be noted that, if based on actual date of conception, fetal ages could be 2 weeks younger than reported in Table 1. Study of the upper respiratory tract in human fetuses has been restricted by the inherent limitations of ultrasound equipment. Image selection, for example, is traditionally made by freezing the moving image on the ultrasound console monitor during the examination. Still images are then made from that frozen frame. Because not all frames can be photographed, a large amount of image material is lost. In addition, the ability to scrutinize precise movements that occur on the screen is sacrificed. The advent of advanced video editing systems now makes it possible to review a n ultrasound examination in its entirety. Re-examination of sonograms on a frame-by-frame basis filmed to appear a t the original speed of movement (“real-time” movement) vastly increases the number of still images available and improves the ability to observe the finer aspects of movement (Goto and Kato, 1983; Cooper et al., 1985). The final still image may be recorded with equipment ranging from cameras to computer terminals (Shawker et al., 1984; Thieme et al., 1985). Structures of the fetal upper respiratory-digestive tract and their movement were studied via ultrasonography and sequential still-frame analysis. All sonograms were performed by a perinatologist (with a t least one member of the investigating team in attendance) using a JVC BR 64000 ultrasound scanner with a linear array transducer. These real-time images were
ULTRASOUND STUDY OF FETAL ANATOMY
365
TABLE 1. Fetal dimensions Subject No. 1
Age by date (weeks)' 17
Age by sonogram (weeks) 19.0
Femur length (cm) 2.9
Biparietal diameter (cm) 4.2
Head circumference (cm) 16.1
Abdominal circumference (cm) 14.1
17 +
2.4 2.3 3.2
3.9 4.0 4.5
14.5 14.9 17.7
12.2 12.9 15.7
+
2 3 4
182 -
17.5 18.4 19.9
5 6
20 + 21 +
20.4 21.3
3.3 3.7
4.7 5.0
17.6 19.2
16.1 15.7
7 8 9 103
22 22 + 23 23
+
22.4 22.9 23.3
4.2 3.8 4.0
5.3 5.6 5.8
19.9 20.5 21.2
17.7 17.0 16.8
11
23 +
12 13
-
Position Coronal, 314 view, sagittal 314 Supine, sagittal Coronal, supine. prone Prone, sagittal 314 sagittal, supine Sagittal 314, transverse Sagittal, 314 Coronal
-
-
23.3
3.7
5.6
22.0
18.0
Supine, sagittal
24 +
22.9
3.9
5.9
21.1
17.7
Prone, 314
-2
24.8
4.8
6.2
22.9
18.6
5.0 5.1 4.6 5.4 5.4 6.0 5.7
6.6 6.8 6.3 7.0 8.0 7.5 7.9
24.7 26.1 23.6 25.8 29.1 28.4 28.6
22.9 22.4 20.5 24.7 25.1 26.3 27.5
Prone, 314, transverse Coronal Coronal 314, transverse Prone 314 314 Supine, sagittal transverse Prone, sagittal 314 Prone Coronal, 314, sagittal SagGtal
26 + 26 +
-
-
14 15 16 17 18 19 20
272 -
29 29 29 +
26.7 27.4 25.4 28.4 29.1 29.1 31.1
21 22 23 24
29 + 32 + 32 36
31.2 21.0 34.4 34.8
5.7 3.6 6.6 6.8
7.9 5.2 8.4 8.4
29.9 19.1 31.4 31.1
26.1 15.0 32.3 32.0
25
36
35.1
7.0
8.5
32.1
30.5
+
Comments
Choroid plexus cysts Choroid plexus cysts Decreased size for dates
Twin A Twin B
Macrosomia
'Date of the last menstrual period as reported by mother. Ages are based on a 40 week clinical pregnancy. Actual dates of conception may be 2 weeks later. 'Fetal age by date of mother's last menstrual period unknown. 3Fetal dimensions unavailable.
recorded onto a 0.50 inch video cassette linked directly to the ultrasound unit. Tapes were then played back through the ultrasound unit a t normal speed (30 frames per second) and then analyzed for anatomic features and movement patterns. Tapes were then viewed using a Sony BVU800 frame-by-frame video editor (FFVE) for closer assessment of anatomy and specific movement patterns. Still-frame images were then selected and photographed directly from a 17 inch monitor using a 35 mm camera with a 55 mm lens, lending greater clarity and frame selection to the still ultrasound image t h a n has been previously observed. Attempts were made during clinical examination to visualize fetuses in both midsagittal and coronal planes. Observation of live, moving, and generally uncooperative fetuses, however, often makes exact positioning impossible. RESULTS Results are summarized in Tables 2 and 3 and in Figures 1through 4. Table 2 shows the anatomic structures commonly visualized during real-time observation and repeated FFVE and normal speed video scru-
tiny. Figures 1 through 4 illustrate structures visible during ultrasound examination. The fluid-filled vascular conduits of the heart and carotid arteries were seen in all 25 subjects. The vertebral column and portions of the basicranium were readily identified in all subjects because of the highly recognizable morphology and extremely high echogenicity (density). Specific regions of the basicranium, such as the petrous part of the temporal bone or contours of the foramen magnum, were clearly identified in 20% and 32% of subjects, respectively. The bony nasal septum was clearly viewed on coronal section (32%, Fig. 4). Tooth buds were clearly observed in 48% of subjects (Fig. 1). Portions of the larynx were seen in 76% of subjects, defined by anatomical shape and location and by high echogenicity. Well-defined thyroid cartilage was seen in 36% of subjects a s a large, clearly defined, highly echogenic structure in the anterior portion of the neck. It was identified in both sagittal and coronal sections (Figs. 2, 4) but was most clear on coronal section. The cricoid cartilage was also seen (32%of subjects) in both sagittal section in the posterior portion of the larynx
V.P. WOLFSON AND J.T.LAITMAN
366
TABLE 2. Fetal anatomic structures SubVerteject Carotid bra1 Basi- Petrous Foramen Nasal Tooth Tracheal ThyEpi- Phar- Piriform no. Heart art. col. cranium' bone magnum septum buds rings Larynx' roid Cricoid glottis ynx sinus + + + + + + 1 + + + + + + 2 + + + + + + + + + 3 + + + + + + + + + + + 4 + + + + + + + 5 + + + + + + + + 6 + + + + + + + + + + + + 7
+
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
+ + + + + + + + + + + + + + + + + +
+ + + + + + + + + + + + + + + + + +
+
+ +
+
+ + + + + + + + + + + + + + + + + +
+ + + + + + + + + + + + + + + + + +
+
+ + +
+ +
+ + +
+ +
+ + + + + +
+
+ + + + + + + + + +
+ + + + + + + +
+ -4-
+
+
+ +
+ +
+
+ + +
+ +
+
+
+
+ + +
+
+ +
+
+
+ +
+ +
+
+
+
+ +
+
+ +
+ +
+ +
+ +
+
+
+ + + + + + +
+ + + + + + +
'Portions of basicranium seen but not distinguishable as to specific bones. 'Portions of larynx seen but not distinguishable as to individual cartilages.
TABLE 3. Fetal upper respiratory movements Subject No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Heart
Carotid
Mandible
Tongue
Esophagus
Bolus
Pharynx
Swallow
Hiccough
+ + + + + + + + + + + + + + + + + + + + + + + + +
+ + + + + + + + + + + + + + + + + + + + + + + + +
+ + + +
+ + + + + +
+ + + + + + + + + + +
+ + +
+ + + +
+ + + + +
+
+ +
+ + +
+ +
+
+
+ + +
(Fig. 2) and coronal section (Figs. 3, 4). The epiglottis was viewed in 16% of subjects. It was most clearly and frequently seen on coronal section as a small echogenic structure projecting into the anechoic (area of low den-
+ + +
+ + + + + + + +
+
+
+
+ + +
sity) lumen of the pharynx, posterior to the oral cavity (Figs. 3, 41, surrounded laterally by the anechoic piriform sinuses (Fig. 3). The smaller laryngeal cartilages, the arytenoids, corniculates, and cuneiforms could not
ULTRASOUND STUDY OF FETAL ANATOMY
367
Fig. 1 . Sonogram of a n 18 week gestational age fetus, parasagittal view (subject No. 3). b, basicranium; c, cricoid cartilage; e, eyeball; h, heart; 1, lumen of the larynx; p, pharynx; t, tooth b u d tr, tracheal ring; v, vertebral column. The small size of this fetus permits a large area view of the head, neck, and upper thorax.
be clearly identified. The soft palate was also not clearly seen. This was probably due to its being both of similar density and in close proximity to the tongue, thus rendering the structures indistinguishable. Tracheal rings were frequently seen structures (80%) because of their clearly identifiable morphology, size,
and position in the neck. The trachea is the largest and most dense structure in the anterior portion of the middle to lower neck region, and its echogenic cartilages contrast sharply with its anechoic fluid-filled lumen. The pharynx was also frequently observed (76%, especially on coronal section; Fig. 3). The piriform sinuses
368
V.P. WOLFSON AND J.T. LAITMAN
Fig. 2. Sonogram (A) and diagram (B) of a 29 week gestational age fetus, parasagittal view (subject No. 21). c, cricoid cartilage; e, epiglottis; 1, lumen of the larynx; p, pharynx; s, location of soft palate; t, thyroid cartilage; to, location of tongue; tr, tracheal rings, v, vertebral column.
were also seen as lateral projections of the pharynx (76%),at times seen clearly surrounding the epiglottis (Fig. 3). Again, this was most clearly observed on coronal section. Movements commonly seen during real-time and repeat normal speed observations are summarized in Table 3. The rhythmic flow of blood through the heart and
carotid arteries was observed in all fetuses and is, indeed, a feature routinely sought during all ultrasound examinations. Most movements of upper respiratory structures were typically observed in connection with swallowing. Fluttering tongue movements and esophageal passage of a liquid bolus were movements typically seen. Swallowing motions began with prelimi-
ULTRASOUND STUDY OF FETAL ANATOMY
369
D
Fig. 3. A,B,C: Progressively dorsal coronal sonograms; D: composite diagram of a 23 week gestational age fetus (subject No. 10). c, cricoid cartilage; e, epiglottis; h, heart; p, pharynx; ps, piriform sinus; tl, tracheal lumen. Dark hatching in D shows region of the eyes and
orbits as can be seen in A and B; light shading indicates pharynx and piriform sinuses a s can be seen in B and C. Note the high position of the larynx, particularly the epiglottis, in C.
370
V.P. WOLFSON AND J.T. LAITMAN
Fig. 4. Sonogram (A) and diagram (B) of a 26 week gestational age fetus, coronal view (subject No. 15).c, cricoid cartilage; e, epiglottis; n, nasal cavityinasopharynx; ns, nasal septum; 0,orbit; t, thyroid carti-
lage; tl, tracheal lumen; t r , tracheal ring. This view well illustrates both the larger laryngeal cartilages and the high position of the larynx itself.
ULTRASOUND STUDY O F F E T A L ANATOMY
nary fluttering movements of the tongue, followed by narrowing and constriction of the pharyngeal space. Minimal vertical movements of the larynx were observed. Of particular interest was a n episode of repeated spasmodic movements of the diaphragm and larynx, reminiscent of that seen during adult “hiccoughs)’ (Table 3). Identification of upper respiratory structures did not seem to be related to fetal age, as no discernible agerelated patterns in the percentages of structures seen were found. At least within this limited sample, the ability to identify fetal anatomy seems to be related more to fetal position than to age.
371
in this region nevertheless suggest a n interlocking between the soft palate and epiglottis. In all instances in which the epiglottis was identified (16% of specimens) the structure was observed very high in the throat. Furthermore, in our eight observed cases of “swallows” and 11 instances of “pharyngeal movements,” the larynx was seen to exhibit minimal vertical movement. This absence of movement is compatible with a larynx positioned high in the neck during swallowing and is totally unlike the adult condition in which the Iarynx exhibits considerable movement as the epiglottis folds backwards during deglutition (Crelin, 1976; Laitman and Crelin, 1980; Shawker et al., 1984). These two sets of observations, coupled with the lack of any observaDISCUSSION tions indicating the contrary, strongly suggest the inBecause of inherent obstacles of in vivo fetal, infant, terlocking of the epiglottis and soft palate. This interand early childhood study, anatomic investigations of locking suggests the creation of separate respiratory the human upper respiratory region have involved and digestive tracts, i.e., the “two-tube’’ system, prenamainly postmortem material. Postmortem studies of tally. the fetus and infant have contributed a preliminary The clear identification of a number of upper digesunderstanding of upper airway development. In addi- tive structures allowed for observations of the sequence tion, clinical and investigational techniques such as of movements found in a fetal swallow. These movecineradiography, preabortion intra-amniotic injection ments included the passage of a bolus of amniotic fluid of radiocontrast medium, and esophageal manometry from the oral cavity into the pharynx; its continued have delineated the rudiments of function (e.g., Davis caudal movement indicated by narrowing of the phaand Potter, 1946; Gryboski, 1965; Pritchard, 1965; ryngeal space; passage of the bolus around the larynx Ardran and Kemp, 1970; Laitman et al., 1977; Laitman via the piriform sinuses (rather than around and over and Crelin, 1980; Bosma, 1985). These techniques, the larynx, as in adults); and limited up-down larynhowever, are impractical for extensive fetal study. It is geal movement. Esophageal movement and limited thus necessary to find a method th at allows for obser- peristalsis were also seen, supporting previous observation of the living fetus in order to understand fully vations of a n early onset of gastrointestinal function the development and functioning of the upper respira- (Abromovich e t al., 1979; Gryboski and Walker, 1983). The further advancement of ultrasound and recordtory and digestive tracts. With the advent of high-resolution ultrasonography, it has become possible to ing techniques, such as those used in this study, will study noninvasively the specifics of anatomy and to provide a n understanding not only of normal anatombegin to observe function within the fetus’ normal mi- ical development but also of pathology. For example, better understanding of this region in the living fetus lieu. This study examined living fetuses to quantify the could be of great value in diagnosing neonatal and inincidence of appearance of upper respiratory and diges- fant upper respiratory pathologies. Ultrasound obsertive tract structures and to chart their movements. Re- vations indicating unusual differences in density of sults indicate that large portions of the larynx and laryngeal cartilages could be suggestive of develpharynx can be visualized in the majority of middle to opmental weakness of laryngeal cartilages (laryngolate second- and early third-trimester fetuses using the malacia). Similarly, peculiarities in the shape of the combination of ultrasonography and repeat, normal pharyngeal lumen could suggest muscular flaccidity, speed, video examination. Specific identification of the which may later be involved with the respiratory individual laryngeal cartilages is, however, variable aspects of SIDS. The prenatal appreciation of such under real-time conditions. Through our application of compromising pathologies would greatly aid in the FFVE to videotaped ultrasound examinations, it has planning and implementation of perinatal treatment. been possible to see the larger laryngeal cartilages, Clearly, the advent of imaging techniques such as ulsuch as the epiglottis, thyroid, and cricoid. While prior trasonography enables exploration of the growth and ultrasound studies have visualized some aspects of the development of the upper respiratory region far in exupper respiratory region (Bowie and Clair, 1982; Coo- cess of previous capabilities. per e t al., 1985; Staudach, 1987), they have not focused ACKNOWLEDGMENTS on specific aspects of laryngeal anatomy. In addition, The authors express their appreciation to Dr. u. none have attempted to quantify the incidence of appearance of laryngeal structures in the live, in utero Chitkara, Department of Obstetrics, Gynecology and fetus. Through the use of ultrasound and FFVE we Reproduction Science of Mount Sinai School of Mediattempted to do both, and, in addition, we found it pos- cine, for ongoing help, advice, and generous access to sible to produce high-quality views of regions such as ultrasound equipment; Dr. T. Berkowitz, for sharing her expertise on fetal gestational age assessment; and the pharynx and trachea. It was not possible to delineate precisely the struc- G. Maffei, N. Katz, and P.J. Gannon for advice and ture of the soft palate. This was due to both normal invaluable technical assistance in photography. We difficulties produced by differing fetal positions and the also thank Drs. A. Eden and J.S. Reidenberg for both soft palate’s thin structure and similarity in density their constant support throughout our work and their and close proximity to the tongue. The movements seen comments on the manuscript. The ongoing support of
V.P. WOLFSON AND J.T. LAITMAN
372
the Departments of Cell Biology/Anatomy and Otolaryngology of the Mount Sinai School of Medicine is greatly appreciated. This project was supported in part by NIH grant T35 DK07420. LITERATURE CITED Abromovich, D.R., H. Garden, L. Jandial, and K.R. Page 1979 Fetal swallowing and voiding in relation to hydramnios. Obstet. Gynecol. Surg., 54t15-20. Ardran, G.M., and F.H. Kemp 1970 The nasal and cervical airway in sleep in the neonatal period. Am. J. Roentgenol., 108t537-542. Beckwith, J.B. 1988 Intrathoracic petechial hemorrhages: A clue to the mechanism of death in sudden infant death syndrome? Ann. N.Y. Acad. Sci., 533t37-47. Bennett, J.B. 1985 Ultrasonography for detection of fetal disorders in the first and second trimesters of pregnancy. In: Human Prenatal Diagnosis. K. Filkins and J.F. Russo, eds. Marcel Dekker, New York, pp. 183-198. Berkowitz, T. Mount Sinai Ultrasound Fetal Age Assessment Computer Program. Mount Sinai School of Medicine, Department of Obstetrics, Gynecology and Reproduction Science, New York (unpublished). Bosma, J.F. 1985 Postnatal ontogeny of performances of the pharynx, larynx and mouth. Am. Rev. Respir. Dis., 131:SlO-S15. Bawie, J.D. 1986 Ultrasound fetal measurements and IUGR. Clin. Diagn. Ultrasound, 19:ll-30. Bowie, J.D., and M.R. Clair 1982 Fetal swallowing and regurgitation: Observation of normal and abnormal activity. Radiology, 144: 877-878. Cooper, C., B.S. Mahony, J.D. Bowie, T.O. Albright, and P.W. Callen 1985 Ultrasound evaluation of the normal fetal upper airway and esophagus. J. Ultrasound Med., 4t343-346. Crelin, E.S. 1973 Functional Anatomy of the Newborn. Yale University Press, New Haven. Crelin, E.S. 1976 Development of the upper respiratory system. Ciba Clin. Symp., 28:3-30. Crelin, E.S. 1987 The Human Vocal Tract Anatomy, Function, Development and Evolution. New York: Vantage. Davis, M.E., and E.L. Potter 1946 Intrauterine respiration of the human fetus. J.A.M.A. 131t1194-1201. Filkins, K., and F.P. Hadlock 1985 Ultrasonography for the detection of fetal disorders in the third trimester of pregnancy. In: Human Prenatal Diagnosis. K. Filkins and J.F. Russo, eds. Marcel Dekker, New York, pp. 199-269. Golding, J., S. Limerick, and A. Macfarlane 1985 Sudden Infant Death: Patterns, Puzzles and Problems. University of Washington, Seattle. Goto, S., and T.K. Kato 1983 Early fetal movements are useful for estimating the gestational weeks in the first trimester of pregnancy. Ultrasound Med. Biol., S2t577-582. Gryboski, J.D. 1965 The swallowing mechanism of the neonate: I. Esophageal and gastric motility. Pediatrics, 35t445-452. Gryboski, J.D., and W.A. Walker 1983 Gastrointestinal Problems in the Infant. W.B. Saunders, Philadelphia. Hadlock, F.P., R.L. Deter, and R.B. Harrist 1986 Estimation of gestational age in the third trimester. Clin. Diagn. Ultrasound, 19: 1-10, Hohler, C.W. 1984 Ultrasound estimation of gestational age. Clin. Obstet. Gynecol., 27:314-326. Hoffman, H.J., K. Damus, L. Hillman, and E. Krongrad 1988 Risk factors for SIDS Results of the National Institute of Child Health
and Human Development SIDS Cooperative Epidemiological Study. Ann. N.Y. Acad. Sci., 533:13-30. Laitman, J.T. 1986 On J.J. McKenna’s “An anthropological perspective on the sudden infant death syndrome: The role of parental breathing cues and speech breathing adaptations.” Med. Anthropol., 10:65-68. Laitman. J.T. 1988 Prenatal and infant transitions in the development of the human upper respiratory tract. Assoc. Res. Otolaryngol. Abstr., 11:160-161. Laitman, J.T., and E.S. Crelin 1976 Postnatal development of the basicranium and vocal tract region in man. In: Symposium on Development of the Basicranium. J.F. Bosma, ed. U S . Government Printing Office, Washington, D.C., pp. 206-220. Laitman, J.T., and E.S. Crelin 1980 Developmental change in the upper respiratory system of human infants. Perinatol. Neonatol., 4r15-22. Laitman, J.T., E.S. Crelin, and G.J. Conlogue 1977 The function of the epiglottis in monkey and man. Yale J. Biol. Med., 50:43-48. Laitman, J.T., R.C. Heimbuch, and E.S. Crelin 1978 Developmental change in a basicranial line and its relationship to the upper respiratory system in living primates. Am. J . Anat., 152t467482. Magriples, U., and J.T. Laitman 1987 Developmental change in the position of the fetal human larynx. Am. J . Phys. Anthropol., 72: 463-472. Moore, S.M., and J.T. Laitman 1989 Development of the fetal human upper respiratory tract during the second trimester. Am. J . Phys. Anthropol., 78t274-275. Moss, M.L. 1965 The veloepiglottic sphincter and obligate nose breathing in the neonate. Brief Clin. Lab. Observ., 67:330-331. Negus, V.E. 1949 The Comparative Anatomy and Physiology of the Larynx. Grune & Stratton, New York. Pritchard, J.A. 1965 Deglutition by normal and anencephalic fetuses. Obstet. Gynecol., 25t289-297. Roche, A.F., and B.D. Barkla 1965 The level of the larynx during childhood. Ann. Otol. Rhinol. Laryngol., 74t645-654. Rodenstein, D.O., N. Perlmutter, and D.C. Stanescu 1985 Infants are not obligatory nasal breathers. Am. Rev. Respir. Dis., 131t343347. Sasaki, C.T. 1977 Physiology of the larynx. In: Otolaryngology,Vol. 3. Harper & Row Publishers, Hagerstown, MD, pp. 1-21. Sasaki, C.T. 1979 Development of the laryngeal function: Etiologic significance of the sudden infant death syndrome. Laryngoscope, 89t1964-1982. Sasaki, C.T., P.A. Levine, J.T. Laitman, and E.S. Crelin 1977 Postnatal descent of the epiglottis in man. Arch. Otolaryngol., 103:169171. Shawker, T.H., B. Sonies, T.E. Hall, and B.F. Baum 1984 Ultrasound analysis of tongue, hyoid and larynx activity during swallowing. Invest. Radiol., 19t82-86. Staudach, A. 1987 Sectional Fetal Anatomy in Ultrasound. SpringerVerlag, New York. Stewart, A.M. 1973 Cancer a s a cause of abortions and stillbirths: The effect of these early deaths on the recognition of radiogenic leukaemias. Br. J . Cancer, 27t465-472. Swift, P.G.F., and J.L. Emery 1973 Clinical observations on response to nasal occlusion in infancy. Arch. Dis. Child. 48t947-951. Thieme, G.A., R.R. Price, and A.E. James 1985 Ultrasound instrumentation and its practical applications. In: The Principles and Practice of Ultrasonography in Obstetrics and Gynecology. R.C. Sanders and A.E. James, eds. Appleton-Century-Crofts,Norwalk, CT, pp. 23-59. Wagner, L.K., R.G. Lester, and L.R. Saldana 1985 Exposure of the Pregnant Patient to Diagnostic Radiations: A Guide to Medical Management. J.B. Lippincott Company, Philadelphia.
