Histupotholug!j 1992. 20, 221-227
Piriform sinus fistula and the ultimobranchial body A.MIYAUCH1, F.MATSUZUKA*, K.KUMA* & S.KATAYAMA7 Second U~purtnientof Surger!y, Kugawa Medical School, * Kctma Uospital, Kobe and Kuro National Hospital. japan
t Department of Clinical Pathology,
Date oi submission 22 August 199 1 Accepted for publication 8 Octobcr 199 1
MIYA[JCHI
A . , MATSlIZIJKA F . , KIJMA K . & K A T A Y A M A S .
(1992) Histopcitholog!/ 20, 221-227
Piriform sinus fistula and the ultimobranchial body Piriform sinus fistulae are an underlying abnormality common in patients with acute suppurative thyroiditis. The fistulae arise from the hypopharynx, and end in or adjacent to the thyroid lobe. These congenital fistulae seem to be remnants of one of the pharyngeal pouches in embryonic development, but their exact origin is still controversial. Kesected specimens of the thyroid glands and fistulae from 1 5 patients were examined immunohistochemically with rabbit antisera to human calcitonin and thyroglobulin. The fistulae were lined by squamous. columnar or ciliated epithelium. and sometimes formed branches in the thyroid lobe. Near the branches solid cell nests existed. MLKOUS glands. follicular structures and thymic tissue were found in the fistula. The follicular structures stained for thyroglobulin. Immunostaining for calcitonin revealed aggregates of many C cells in the thyroid near the fistula. A few calcitonin-positive cells were also found in the fistula. These findings, along with the anatomical relation of the fistulae to major structures of the neck, strongly suggest that the fistulae are remnants related to the ultimobranchial body, and that the fistulae trace the migration route of the ultimobranchial body to the thyroid gland. Keywords: piriform sinus fistula, acute suppurative thyroiditis. C cells, follicular cells, solid cell nests, ultimobranchial body
Introduction
Materials and methods
Piriform sinus fistulae are a common route of infection in acute suppurative thyroiditis’ The fistulae arise from the apex of the piriform sinus of the hypopharynx. and end in or adjacent to the thyroid lobe. allowing bacterial infection to develop in or around the thyroid lobe’ Although the inflammation is readily treated by antibiotics with or without an incision and drainage. recurrence of inflammation is common. Removal of the fistulae, however, prevents the recurrence. These congenital fistulae seem to be remnants of one of the pharyngeal pouches in embryonic development. However, the exact origin of such fistulae is still controversial’ Here we describe aggregation of C cells, which are derivatives of the ultimobranchial bodyh in the thyroid around the fistulae revealed by immunohistochemical studies, and discuss the origin of the fistulae.
’.
’.
’.
’,
Address for correspondence: Dr A.MiyauchL Second Department of Surgery. Kagawa Medical School. Miki-cho. Kita-gun. Kagawa 76 1 07. Japan.
Resected fistulae and thyroid tissue from 15 patients were examined. The patients were eight females and seven males, aged 7- 5 5 years. with a mean of 19.3 years at the time of surgery. They had had one or more episodes of acute suppurative thyroiditis, and they had a piriform sinus fistula as demonstrated by a barium meal examination. Serum calcium levels were normal in all of them. All of the fistulae were on the left side. At surgery, the fistulae were found to be attached to or to enter the thyroid lobe at the medial aspect, one-third to one-fourth distance caudal to the upper pole of the lobe in 11 patients. The fistulae ended medial to the lobe in three patients and lateral to it in one. In cases with the fistula attached to the thyroid lobe. part of the lobe was resected together with the fistula, and in the other cases an incisional biopsy of the thyroid was performed as well as a fistulectomy. Formalin-fixed and paraffin-embedded specimens of the fistulae and the thyroid tissue were examined immunohistochemically with rabbit anti-human calci221
222
A.Miynuchi et al.
tonin serum (Dakopatts. Copenhagen, Denmark) using the avidin-biotin-peroxidase complex techniquex. Sections of the fistulae were also examined for the presence of follicular cells with rabbit anti-thyroglobulin serum (Dakopatts). Sections of medullary thyroid carcinoma and normal thyroid served as positive controls: these showed negative staining when primary antisera were replaced by non-immune serum.
Results The fistulae were lined by squamous. columnar or ciliated epithelium (Figure l a ) . Six of them formed branches, branching within the thyroid lobe in four cases. Mucous glands, follicular structures and thymic tissue were found in the fistulae in five, three, and one case respectively. The epithelium lining the fistulae was negative for calcitonin. However. a few calcitoninpositive cells were found in the base of the lining epithelium in three cases. In one case a cluster of
calcitonin-positive cells was also found under the epithelium (Figure l b ) . The follicular structures found in three fistulae stained for thyroglobulin (Figure l c ) . Thyroid specimens showed various degrees of interstitial fibrosis and evidence of the regeneration of damaged thyroid follicles such as uneven-sized follicles or solid clusters of follicular cells without colloid only in the vicinity of the focus of the previous inflammation. Histological examination demonstrated the piriform sinus fistulae to enter the thyroid lobe in six cases. and to branch there in four of these (Figure 3 ) . Near the branches of the fistulae, solid cell nests were found in three cases (Figure 4). lmmunohistochemical studies revealed many calcitonin-positive cells in the thyroid near the fistula in eight cases, while no cells reactive for calcitonin could be found in the remaining cases. C cells occupied thyroid follicles to partially or completely encircle their inner surface (Figure 2b). When the specimens were stained with haematoxylin and eosin, these C cells were incons-
Figure 1. Piriform sinus fistula: a stratified squamous epithelium and submucosal mucous glands and follicular structures: b calcitoninpositive cells as a single cell in the base of the epithelium and a cluster of cells under the epithelium: c thyroglobulin-positive follicular cells in the wall.
Piriform siniisfistula atid irltimobrattrhial body
223
Figure 2 . Thyroid specimen from a patient with a piriform sinus fistula. a C cells are inconspicuous on haematoxylin and eosin staining.
h Immunostaining reveals numerous calcitonin-positive cells.
picuous. After observing the immunostained slides. they could, however. be recognized as having abundant clear cytoplasm and a round nucleus (Figure l a ) . Five of the eight cases with calcitonin-positive cells were cases in which the fistula entered the thyroid lobe. C cells were found near the fistula. In a case with a fistula branching in the thyroid lobe, numerous C cells surrounded the brunches of the fistula (Figure 3). C cells were distributed densely only near the branches. Most of the C cells were located within the thyroid follicles. However, they also presented as small solid clusters without colloid or even as single cells (Figure 3c), and a few calcitonin-positive cells were seen in the base of the lining epithelium of the branches. Solid cell nests found near the branches were accompanied by many C cells (Figure 4).
Discussion An increase in the number of C cells (C cell hyperplasia)
has been reported in young members of kindreds of familial medullary thyroid carcinoma'. in patients with chronic hypercalcaemia due to primary hyperparathyroidism'" and in a patient with chronic thyroiditis' I , and also as a n age-related phenomenon'?. None of these was the case in our patients. C cells partially or completely occupied thyroid follicles lining their inner surface, or they formed small solid clusters. These appearances resembled the focal, diffuse, or nodular C cell hyperplasia reported as early C cell proliferative changes in familial medullary thyroid carcinoma' 3. However, unlike C cell hyperplasia preceding familial medullary thyroid carcinoma, the C cells in our patients were not hypertrophic and did not show nuclear atypism. Although the thyroid gland could not be examined thoroughly in its entirety, the high concentration of C cells reported in this paper seemed to be a phenomenon localized to the thyroid near the piriform sinus fistula indicating close relation of the fistulae to C cells.
224
A.Miyauchi et al.
Figure 3 . Thyroid specimen with branches of a piriform sinus tistula: a tubular structures with a lining of ciliated epithelium are surrounded by lymphocytes and thyroid follicles: b aggregation of numerous C cells around the branches of the fistula: c C’ cells occupying thyroid follicles ur forming solid clusters.
Piriforiii sinus fistula and ultiniobruricliial body
22 5
Figure 4. A solid cell nest (a: short arrow), found in the vicinity of branches (long arrows) of a piriform sinus fistula, is shown at higher power in b. c Many C cells surround the solid cell nest.
Piriform sinus fistulae seem to be remnants of one of the pharyngeal pouches in embryonic development. Fistulae related to either the third or the fourth pharyngeal pouch are expected to drain into the piriform sinus14.Piriform sinus fistulae run caudal to the external branch of the superior laryngeal nerve (the nerve of the fourth branchial arch)”. and cranial to the recurrent laryngeal nerve (the nerve of the sixth branchial arch)’ Because the fifth pouch is rudimentary in the human, this anatomic relationship suggests a fourth pouch origin for the fistula’,4. However, as we previously pointed out’, the superior parathyroid gland, the major derivative of the fourth pharyngeal pouch’’, had no relation to the fistula and so its presence could not be explained by this theory. At first we thought that the attachment of the fistulae to the thyroid lobe was secondary to adhesions resulting from inflammation, and that cases with fistulae penetrating the lobe were exceptional. However, our series of cases confirmed the close relation of the fistulae to the
’.
thyroid’. In most of our patients the fistulae were attached at a fairly constant location on the thyroid lobe, suggesting some embryological inevitability for this anatomy. The fifth pharyngeal pouch is rudimentary in the human embryo. It is incorporated into the fourth pouch and appears as a diverticulum from the ventral part ofthe fourth pouch. Later, this diverticulum becomes elongated and develops into an ultimobranchial bodyI5-l7. This entire outpocketing of the pharynx is called the caudal pharyngeal complex’’, which expands into a dorsal bulbar portion and ventral elongated part. These are connected with the pharynx via a narrow duct called the pharyngo-branchial duct IV, which disappears with subsequent development. The dorsal portion loses its connection with the pharynx and develops into the superior parathyroid gland. Although the fate of the ventral component of the fourth pouch proper is uncertain, it is believed that it gives rise to a small amount of thymic t i s s ~ e l ~When , ~ ~ .the ventral elongated part of
226
A.Mi!yaiidii et al.
the caudal pharyngeal complex develops into the ultimobranchial body, it becomes markedly thickened and the cavity begins to shrink, although traces of the cavity may persist for some time'' 17. The ultimobranchial body then migrates caudally and laterally to fuse with the lateral lobe of the developing thyroid, and subsequently disseminates to give rise to the C cells'5-". In the adult thyroid, the C cells are concentrated in the middle third of the lateral lobes, along their longitudinal axes'". In most of our cases the site of attachment of the piriform sinus fistula to the thyroid lobe was the most likely location at which a remnant related to the ultimobranchial body would be expected. The expected relationship of such a remnant with the major anatomical structures is consistent with that of the piriform sinus fistula. Compact cell nests are seen in the lateral lobe of the thyroid in man. They are considered to be of ultimobranchial derivation" ", because C cells are usually found in their vicinity and in these nests proper, and because the distribution of the nests is similar to that of C cells. In the present study, compact cell nests were found near the branches of the piriform sinus fistula, and were accompanied by many C cells. This also suggests a close relationship between piriform sinus fistulae, C cells, solid cell nests and the ultimobranchial body. Kecently. Williams & Harach' reported thyroglobulin-containing follicular structures as well as glandular nodules composed of calcitonin-positive cells in cystic structures found in the region of the upper parathyroid in five patients with a maldescended thyroid. They concluded that the cysts were remnants of the ultimobranchial body and that the ultimobranchial body contributes not only C cells but also part of follicular cells to the thyroid in man. They also described considerable variation in the cyst lining, i.e. squamous. columnar and ciliated epithelium, and glandular structures resembling salivary gland ducts and thymic tissue in the cyst wall. We found thyroglobulin-positive follicles in the piriform sinus fistulae. The lining epithelium of the fistulae and accessory tissues of the fistula were quite similar to the branchial remnants described by Williams & Harach?j . The histological and immunohistochemical findings in the present studies, along with the anatomy of the fistulae. strongly suggest that the fistulae are remnants of the communicating ducts between the embryonic pharynx and the ultimobranchial body. and that the fistulae follow the path of migration of the ultimobranchial body to the thyroid gland. We interpret the aggregation of C cells near the fistula as representing a state before the complete dissemination of C cells to the thyroid. Using this hypothesis, the fact that tiny fragments of thymic tissue or parathyroid tissue were
occasionally found along the fistulae'.3 is also readily explained. Kingsbury''.'' described a marked variability between right and left sides in the development of the human ultimobranchial body. He also described human embryos with a poorly developed ultimobranchial body on the right side. Growth of the ultimobranchial body is often restricted or absent on the right side in the lower vertebrates including reptileslx. These findings may elucidate the left-sided predominance of the fistulae.
References 1. Takai S . Miyauchi A. Matsuzuka F. Kunia K. Kosaki G. Internal fistula as a route of infection in acute suppurative thyroiditis. Iaiiref 1979: i; 751-752. 2. Miyauchi A. Matsuxuka P. Takai S. Kuma K. Kosaki G. Piriform sinus fistula. A route of infection in acute suppurative thyroiditis. Arch. Surg. 1 9X 1 : 1 16; 66-69. 3 . Miyauchi A. Matsuzuka F. Kuma K. Takai S. Piriform sinus fistula, a n underlying abnormality common in patients with acute suppurative thyroiditis. World /. Surg. 1990: 14; 400-405. 4. Tucker HM. Skolnik MI.. Fourth branchial cleft (pharyngeal pouch) remnant. Trrrris Am. Arcid. Opfithcrlrnol. Otol. 197 3: 7 ; ; 368-371. 5. Miller D. Hill JL. Sun CC. O'Brien 1)s. Haller JA. Jr. The diagnosis and management of pyriform sinus fistulae in infants and young children. 1. Pediot. Surg. 1 9 8 3 : 18; 377-381. 6 . Pearse AGE. Carvalheira AF. Cytochemical evidence for an ultimobranchial origin of rodent thyroid C cells. h'ntirrr 1967: 214; 929-930. 7. Moseley IM.Matthews EW. Breed RH. Galante L. Tse A. MacIntyre I. The ultimobranchial origin of calcitonin. Lntiwt 196X: i: 1081 10. 8 . Hsu SM. Raine L. Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques. A comparison between ABC and unlabeled antibody (PAP) procedures. J . HistoclrPin. Cytocfwn. 1981 : 29; 5;;-580. 9 . Wolfe HJ. Melvin KEN. Cervi-Skinner SJ et nl. C-cell hyperplasia preceding niedullary thyroid carcinoma. N. ErigI. /. Mrd. 1973: 289; 437-441. 10. 1,iVolsi VA. Feind CR, LoCerfo P, Tashjian AH. ]r. Demonstration by immunoperoxidase staining of hyperplasia of parafollicular cells in the thyroid gland in hyperparathyroidisni. J. Clin. Endocriiiol. R f ~ ~ t a1973: b. 37; 550-559. 1 1. Libbey NP. Nowakowski KJ. Tucci JR. C-cell hyperplasia of the thyroid in a patient with goitrous hypothyroidism and Hashimoto's thyroiditis. Am. /. Sirrg. Pntfiol. 1989: 13; 71-77. 12. Gibson WGH. Peng T. Croker BP. Age-associated C-cell hyperplasia in the human thyroid. Am. J. Patliol. 1982: 106; 388-393. 1 3 . DeLellis KA. Nunnemacher G. Wolfe HJ. C-cell hyperplasia. An ultrastructural analysis. Lab. Inwst. 1977: 36; 2 37-248. 14. Proctor B. Lateral vestigial cysts and tistulae of the neck. Luryngosrope 1955; 65; 355-401. 1 5 . Moore KL. ed. Tf7e Developing Human. 2nd ed. Philadelphia: Saunders. 1977: 156-1 65. 16. Kingsbury BF. On the fate of the ultimobranchial body within the h u man thyroid gland. Annt. Rec. 1935: 61; 155-167. 17. Sugiyama S. The embryology of the human thyroid gland including ultimobranchial body and others related. Ado. Atlrrt. Ernbryol. Cell B i d . 1 971 : 44; 1-1 1 1.
Pirifornr sinus fistitla and ultimobranchial body
18. Kingsbury RF. The question of a lateral thyruid in mammals with special reference to man. Arn. 1. Anat. 1939: 65; 3 33-359. 19. Langman J . ed. Medical Ertihryo1og.y. 4th ed. Baltimore: Williams & Wilkins. 19 8 1 : 2 74-2 76. 20. Wolfe HJ. Voelkel EF, Tashjian AH, ]r. Distribution of calcitonincontaining cells in the normal adult human thyroid gland. A correlation of morphology with peptide content. 1. Clirr. Errdocrinol. Metab. 1974: 38; 688-694. 21. ]anzer RC. Weber E, Hedinger C. The relation between solid cell
22 7
nests and C cells of the thyroid gland. Cell. Tissue Res. 1 9 79; 197; 295-312. 22. Harach HR. Solid cell nests of the thyroid. An anatomical survey and immunohistochemical study for the presence of thyroglobulin. Arta Anat. 1985; 122; 249-253. 23. Williams ED. Harach HR. The ultimobranchial gland and congenital thyroid abnormalities i n man. ]. Puthol. 1989: 1 59; 1 3 5141.
Piriform sinus fistula and the ultimobranchial body A.MIYAUCH1, F.MATSUZUKA*, K.KUMA* & S.KATAYAMA7 Second U~purtnientof Surger!y, Kugawa Medical School, * Kctma Uospital, Kobe and Kuro National Hospital. japan
t Department of Clinical Pathology,
Date oi submission 22 August 199 1 Accepted for publication 8 Octobcr 199 1
MIYA[JCHI
A . , MATSlIZIJKA F . , KIJMA K . & K A T A Y A M A S .
(1992) Histopcitholog!/ 20, 221-227
Piriform sinus fistula and the ultimobranchial body Piriform sinus fistulae are an underlying abnormality common in patients with acute suppurative thyroiditis. The fistulae arise from the hypopharynx, and end in or adjacent to the thyroid lobe. These congenital fistulae seem to be remnants of one of the pharyngeal pouches in embryonic development, but their exact origin is still controversial. Kesected specimens of the thyroid glands and fistulae from 1 5 patients were examined immunohistochemically with rabbit antisera to human calcitonin and thyroglobulin. The fistulae were lined by squamous. columnar or ciliated epithelium. and sometimes formed branches in the thyroid lobe. Near the branches solid cell nests existed. MLKOUS glands. follicular structures and thymic tissue were found in the fistula. The follicular structures stained for thyroglobulin. Immunostaining for calcitonin revealed aggregates of many C cells in the thyroid near the fistula. A few calcitonin-positive cells were also found in the fistula. These findings, along with the anatomical relation of the fistulae to major structures of the neck, strongly suggest that the fistulae are remnants related to the ultimobranchial body, and that the fistulae trace the migration route of the ultimobranchial body to the thyroid gland. Keywords: piriform sinus fistula, acute suppurative thyroiditis. C cells, follicular cells, solid cell nests, ultimobranchial body
Introduction
Materials and methods
Piriform sinus fistulae are a common route of infection in acute suppurative thyroiditis’ The fistulae arise from the apex of the piriform sinus of the hypopharynx. and end in or adjacent to the thyroid lobe. allowing bacterial infection to develop in or around the thyroid lobe’ Although the inflammation is readily treated by antibiotics with or without an incision and drainage. recurrence of inflammation is common. Removal of the fistulae, however, prevents the recurrence. These congenital fistulae seem to be remnants of one of the pharyngeal pouches in embryonic development. However, the exact origin of such fistulae is still controversial’ Here we describe aggregation of C cells, which are derivatives of the ultimobranchial bodyh in the thyroid around the fistulae revealed by immunohistochemical studies, and discuss the origin of the fistulae.
’.
’.
’.
’,
Address for correspondence: Dr A.MiyauchL Second Department of Surgery. Kagawa Medical School. Miki-cho. Kita-gun. Kagawa 76 1 07. Japan.
Resected fistulae and thyroid tissue from 15 patients were examined. The patients were eight females and seven males, aged 7- 5 5 years. with a mean of 19.3 years at the time of surgery. They had had one or more episodes of acute suppurative thyroiditis, and they had a piriform sinus fistula as demonstrated by a barium meal examination. Serum calcium levels were normal in all of them. All of the fistulae were on the left side. At surgery, the fistulae were found to be attached to or to enter the thyroid lobe at the medial aspect, one-third to one-fourth distance caudal to the upper pole of the lobe in 11 patients. The fistulae ended medial to the lobe in three patients and lateral to it in one. In cases with the fistula attached to the thyroid lobe. part of the lobe was resected together with the fistula, and in the other cases an incisional biopsy of the thyroid was performed as well as a fistulectomy. Formalin-fixed and paraffin-embedded specimens of the fistulae and the thyroid tissue were examined immunohistochemically with rabbit anti-human calci221
222
A.Miynuchi et al.
tonin serum (Dakopatts. Copenhagen, Denmark) using the avidin-biotin-peroxidase complex techniquex. Sections of the fistulae were also examined for the presence of follicular cells with rabbit anti-thyroglobulin serum (Dakopatts). Sections of medullary thyroid carcinoma and normal thyroid served as positive controls: these showed negative staining when primary antisera were replaced by non-immune serum.
Results The fistulae were lined by squamous. columnar or ciliated epithelium (Figure l a ) . Six of them formed branches, branching within the thyroid lobe in four cases. Mucous glands, follicular structures and thymic tissue were found in the fistulae in five, three, and one case respectively. The epithelium lining the fistulae was negative for calcitonin. However. a few calcitoninpositive cells were found in the base of the lining epithelium in three cases. In one case a cluster of
calcitonin-positive cells was also found under the epithelium (Figure l b ) . The follicular structures found in three fistulae stained for thyroglobulin (Figure l c ) . Thyroid specimens showed various degrees of interstitial fibrosis and evidence of the regeneration of damaged thyroid follicles such as uneven-sized follicles or solid clusters of follicular cells without colloid only in the vicinity of the focus of the previous inflammation. Histological examination demonstrated the piriform sinus fistulae to enter the thyroid lobe in six cases. and to branch there in four of these (Figure 3 ) . Near the branches of the fistulae, solid cell nests were found in three cases (Figure 4). lmmunohistochemical studies revealed many calcitonin-positive cells in the thyroid near the fistula in eight cases, while no cells reactive for calcitonin could be found in the remaining cases. C cells occupied thyroid follicles to partially or completely encircle their inner surface (Figure 2b). When the specimens were stained with haematoxylin and eosin, these C cells were incons-
Figure 1. Piriform sinus fistula: a stratified squamous epithelium and submucosal mucous glands and follicular structures: b calcitoninpositive cells as a single cell in the base of the epithelium and a cluster of cells under the epithelium: c thyroglobulin-positive follicular cells in the wall.
Piriform siniisfistula atid irltimobrattrhial body
223
Figure 2 . Thyroid specimen from a patient with a piriform sinus fistula. a C cells are inconspicuous on haematoxylin and eosin staining.
h Immunostaining reveals numerous calcitonin-positive cells.
picuous. After observing the immunostained slides. they could, however. be recognized as having abundant clear cytoplasm and a round nucleus (Figure l a ) . Five of the eight cases with calcitonin-positive cells were cases in which the fistula entered the thyroid lobe. C cells were found near the fistula. In a case with a fistula branching in the thyroid lobe, numerous C cells surrounded the brunches of the fistula (Figure 3). C cells were distributed densely only near the branches. Most of the C cells were located within the thyroid follicles. However, they also presented as small solid clusters without colloid or even as single cells (Figure 3c), and a few calcitonin-positive cells were seen in the base of the lining epithelium of the branches. Solid cell nests found near the branches were accompanied by many C cells (Figure 4).
Discussion An increase in the number of C cells (C cell hyperplasia)
has been reported in young members of kindreds of familial medullary thyroid carcinoma'. in patients with chronic hypercalcaemia due to primary hyperparathyroidism'" and in a patient with chronic thyroiditis' I , and also as a n age-related phenomenon'?. None of these was the case in our patients. C cells partially or completely occupied thyroid follicles lining their inner surface, or they formed small solid clusters. These appearances resembled the focal, diffuse, or nodular C cell hyperplasia reported as early C cell proliferative changes in familial medullary thyroid carcinoma' 3. However, unlike C cell hyperplasia preceding familial medullary thyroid carcinoma, the C cells in our patients were not hypertrophic and did not show nuclear atypism. Although the thyroid gland could not be examined thoroughly in its entirety, the high concentration of C cells reported in this paper seemed to be a phenomenon localized to the thyroid near the piriform sinus fistula indicating close relation of the fistulae to C cells.
224
A.Miyauchi et al.
Figure 3 . Thyroid specimen with branches of a piriform sinus tistula: a tubular structures with a lining of ciliated epithelium are surrounded by lymphocytes and thyroid follicles: b aggregation of numerous C cells around the branches of the fistula: c C’ cells occupying thyroid follicles ur forming solid clusters.
Piriforiii sinus fistula and ultiniobruricliial body
22 5
Figure 4. A solid cell nest (a: short arrow), found in the vicinity of branches (long arrows) of a piriform sinus fistula, is shown at higher power in b. c Many C cells surround the solid cell nest.
Piriform sinus fistulae seem to be remnants of one of the pharyngeal pouches in embryonic development. Fistulae related to either the third or the fourth pharyngeal pouch are expected to drain into the piriform sinus14.Piriform sinus fistulae run caudal to the external branch of the superior laryngeal nerve (the nerve of the fourth branchial arch)”. and cranial to the recurrent laryngeal nerve (the nerve of the sixth branchial arch)’ Because the fifth pouch is rudimentary in the human, this anatomic relationship suggests a fourth pouch origin for the fistula’,4. However, as we previously pointed out’, the superior parathyroid gland, the major derivative of the fourth pharyngeal pouch’’, had no relation to the fistula and so its presence could not be explained by this theory. At first we thought that the attachment of the fistulae to the thyroid lobe was secondary to adhesions resulting from inflammation, and that cases with fistulae penetrating the lobe were exceptional. However, our series of cases confirmed the close relation of the fistulae to the
’.
thyroid’. In most of our patients the fistulae were attached at a fairly constant location on the thyroid lobe, suggesting some embryological inevitability for this anatomy. The fifth pharyngeal pouch is rudimentary in the human embryo. It is incorporated into the fourth pouch and appears as a diverticulum from the ventral part ofthe fourth pouch. Later, this diverticulum becomes elongated and develops into an ultimobranchial bodyI5-l7. This entire outpocketing of the pharynx is called the caudal pharyngeal complex’’, which expands into a dorsal bulbar portion and ventral elongated part. These are connected with the pharynx via a narrow duct called the pharyngo-branchial duct IV, which disappears with subsequent development. The dorsal portion loses its connection with the pharynx and develops into the superior parathyroid gland. Although the fate of the ventral component of the fourth pouch proper is uncertain, it is believed that it gives rise to a small amount of thymic t i s s ~ e l ~When , ~ ~ .the ventral elongated part of
226
A.Mi!yaiidii et al.
the caudal pharyngeal complex develops into the ultimobranchial body, it becomes markedly thickened and the cavity begins to shrink, although traces of the cavity may persist for some time'' 17. The ultimobranchial body then migrates caudally and laterally to fuse with the lateral lobe of the developing thyroid, and subsequently disseminates to give rise to the C cells'5-". In the adult thyroid, the C cells are concentrated in the middle third of the lateral lobes, along their longitudinal axes'". In most of our cases the site of attachment of the piriform sinus fistula to the thyroid lobe was the most likely location at which a remnant related to the ultimobranchial body would be expected. The expected relationship of such a remnant with the major anatomical structures is consistent with that of the piriform sinus fistula. Compact cell nests are seen in the lateral lobe of the thyroid in man. They are considered to be of ultimobranchial derivation" ", because C cells are usually found in their vicinity and in these nests proper, and because the distribution of the nests is similar to that of C cells. In the present study, compact cell nests were found near the branches of the piriform sinus fistula, and were accompanied by many C cells. This also suggests a close relationship between piriform sinus fistulae, C cells, solid cell nests and the ultimobranchial body. Kecently. Williams & Harach' reported thyroglobulin-containing follicular structures as well as glandular nodules composed of calcitonin-positive cells in cystic structures found in the region of the upper parathyroid in five patients with a maldescended thyroid. They concluded that the cysts were remnants of the ultimobranchial body and that the ultimobranchial body contributes not only C cells but also part of follicular cells to the thyroid in man. They also described considerable variation in the cyst lining, i.e. squamous. columnar and ciliated epithelium, and glandular structures resembling salivary gland ducts and thymic tissue in the cyst wall. We found thyroglobulin-positive follicles in the piriform sinus fistulae. The lining epithelium of the fistulae and accessory tissues of the fistula were quite similar to the branchial remnants described by Williams & Harach?j . The histological and immunohistochemical findings in the present studies, along with the anatomy of the fistulae. strongly suggest that the fistulae are remnants of the communicating ducts between the embryonic pharynx and the ultimobranchial body. and that the fistulae follow the path of migration of the ultimobranchial body to the thyroid gland. We interpret the aggregation of C cells near the fistula as representing a state before the complete dissemination of C cells to the thyroid. Using this hypothesis, the fact that tiny fragments of thymic tissue or parathyroid tissue were
occasionally found along the fistulae'.3 is also readily explained. Kingsbury''.'' described a marked variability between right and left sides in the development of the human ultimobranchial body. He also described human embryos with a poorly developed ultimobranchial body on the right side. Growth of the ultimobranchial body is often restricted or absent on the right side in the lower vertebrates including reptileslx. These findings may elucidate the left-sided predominance of the fistulae.
References 1. Takai S . Miyauchi A. Matsuzuka F. Kunia K. Kosaki G. Internal fistula as a route of infection in acute suppurative thyroiditis. Iaiiref 1979: i; 751-752. 2. Miyauchi A. Matsuxuka P. Takai S. Kuma K. Kosaki G. Piriform sinus fistula. A route of infection in acute suppurative thyroiditis. Arch. Surg. 1 9X 1 : 1 16; 66-69. 3 . Miyauchi A. Matsuzuka F. Kuma K. Takai S. Piriform sinus fistula, a n underlying abnormality common in patients with acute suppurative thyroiditis. World /. Surg. 1990: 14; 400-405. 4. Tucker HM. Skolnik MI.. Fourth branchial cleft (pharyngeal pouch) remnant. Trrrris Am. Arcid. Opfithcrlrnol. Otol. 197 3: 7 ; ; 368-371. 5. Miller D. Hill JL. Sun CC. O'Brien 1)s. Haller JA. Jr. The diagnosis and management of pyriform sinus fistulae in infants and young children. 1. Pediot. Surg. 1 9 8 3 : 18; 377-381. 6 . Pearse AGE. Carvalheira AF. Cytochemical evidence for an ultimobranchial origin of rodent thyroid C cells. h'ntirrr 1967: 214; 929-930. 7. Moseley IM.Matthews EW. Breed RH. Galante L. Tse A. MacIntyre I. The ultimobranchial origin of calcitonin. Lntiwt 196X: i: 1081 10. 8 . Hsu SM. Raine L. Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques. A comparison between ABC and unlabeled antibody (PAP) procedures. J . HistoclrPin. Cytocfwn. 1981 : 29; 5;;-580. 9 . Wolfe HJ. Melvin KEN. Cervi-Skinner SJ et nl. C-cell hyperplasia preceding niedullary thyroid carcinoma. N. ErigI. /. Mrd. 1973: 289; 437-441. 10. 1,iVolsi VA. Feind CR, LoCerfo P, Tashjian AH. ]r. Demonstration by immunoperoxidase staining of hyperplasia of parafollicular cells in the thyroid gland in hyperparathyroidisni. J. Clin. Endocriiiol. R f ~ ~ t a1973: b. 37; 550-559. 1 1. Libbey NP. Nowakowski KJ. Tucci JR. C-cell hyperplasia of the thyroid in a patient with goitrous hypothyroidism and Hashimoto's thyroiditis. Am. /. Sirrg. Pntfiol. 1989: 13; 71-77. 12. Gibson WGH. Peng T. Croker BP. Age-associated C-cell hyperplasia in the human thyroid. Am. J. Patliol. 1982: 106; 388-393. 1 3 . DeLellis KA. Nunnemacher G. Wolfe HJ. C-cell hyperplasia. An ultrastructural analysis. Lab. Inwst. 1977: 36; 2 37-248. 14. Proctor B. Lateral vestigial cysts and tistulae of the neck. Luryngosrope 1955; 65; 355-401. 1 5 . Moore KL. ed. Tf7e Developing Human. 2nd ed. Philadelphia: Saunders. 1977: 156-1 65. 16. Kingsbury BF. On the fate of the ultimobranchial body within the h u man thyroid gland. Annt. Rec. 1935: 61; 155-167. 17. Sugiyama S. The embryology of the human thyroid gland including ultimobranchial body and others related. Ado. Atlrrt. Ernbryol. Cell B i d . 1 971 : 44; 1-1 1 1.
Pirifornr sinus fistitla and ultimobranchial body
18. Kingsbury RF. The question of a lateral thyruid in mammals with special reference to man. Arn. 1. Anat. 1939: 65; 3 33-359. 19. Langman J . ed. Medical Ertihryo1og.y. 4th ed. Baltimore: Williams & Wilkins. 19 8 1 : 2 74-2 76. 20. Wolfe HJ. Voelkel EF, Tashjian AH, ]r. Distribution of calcitonincontaining cells in the normal adult human thyroid gland. A correlation of morphology with peptide content. 1. Clirr. Errdocrinol. Metab. 1974: 38; 688-694. 21. ]anzer RC. Weber E, Hedinger C. The relation between solid cell
22 7
nests and C cells of the thyroid gland. Cell. Tissue Res. 1 9 79; 197; 295-312. 22. Harach HR. Solid cell nests of the thyroid. An anatomical survey and immunohistochemical study for the presence of thyroglobulin. Arta Anat. 1985; 122; 249-253. 23. Williams ED. Harach HR. The ultimobranchial gland and congenital thyroid abnormalities i n man. ]. Puthol. 1989: 1 59; 1 3 5141.
