- Email: [email protected]
Morphologic changes in breast biopsies after duct endoscopy
The Breast (2001) 10, 149–154 # 2001 Harcourt Publishers Ltd doi:10.1054/brst.2000.0172, available online at http://www.idealibrary.com on
ORIGINAL ARTICLE
Morphologic changes in breast biopsies after duct endoscopy F. Tresserra*, R. Fa´bregas*, J. Torrent*, P. J. Grases, C. Ara*, M. Izquierdo and A. Fernandez-Cid* Departments of 1Pathology and 2Gynecology, Institut Universitari Dexeus, Paseo de la Bonanova 69, 08017 Barcelona, Spain S U M M A R Y . Duct endoscopy is a recent technique used for a direct view of the breast ductal system. The aim of this study is to determine any morphological changes in breast tissue attributable to low-pressure irrigation with saline solution that the technique requires. A total of 26 breast biopsies from patients who underwent ductal endoscopy before surgery were compared with 26 breast specimens from the retroareolar region. Breast specimens from duct endoscopy showed more frequent epithelial detachment (73%), epithelial loss (35%), periductal clefts (77%), stromal disaggregation (46%) and displacement of epithelial cells into the stroma (27%) than the control group in which epithelial detachment was seen in 4% of patients, periductal clefts in 15%, and stromal disaggregation in 15%. Epithelial loss and epithelial displacement where not seen in the control group. Although low-pressure fluid perfusion used for duct endoscopy induced morphological changes in breast tissue, these can easily be distinguished from malignancy, and are most likely to occur as the result of duct rupture. # 2001 Harcourt Publishers Ltd
In order to establish the effects of fluid pressure in the mammary duct system, morphologic changes in 26 surgically removed specimens from patients in whom duct endoscopy had been performed, were compared with 26 control breast specimens obtained from the retroareolar region.
INTRODUCTION Recently, new diagnostic techniques have been applied for the diagnosis of breast pathology. One of these ancillary methods is duct endoscopy or ductoscopy, which permits a direct view of the interior of the mammary ducts.1–7 Berna et al.,1 as well as Makita et al.,2 used a rigid fibrescope to examine women with nipple discharge. Love et al.4,5 used it to identify early changes in the ductal epithelium in cases of breast cancer. Okazaki et al.3,6,7 introduced the flexible silicafibrescope that could be introduced into peripheral sites. The technique also allows microbiopsies2 or cytologic samples to be obtained.4,5 The introduction of the endoscope into the ductal system requires a dilating agent. Most authors have used low-pressure irrigation with saline solution,1,5 but air injection has also been tested.5 Although mammary ducts are distensible to certain limits, they can rupture, probably due to excess pressure.5
MATERIALS AND METHODS The case group was composed of pathologic specimens from 26 patients who underwent ductal endoscopy of breast lactiferous ducts before surgical excision. These procedures were performed at our institution during a 4-year period. The mean age of these patients was 47.4+10.4 years (range 17–66 years). For endoscopy, the duct was gradually dilated by a blunt conical-tipped lacrimal dilator. Then the flexible endoscope (FAS002, Conceptus Inc., San Carlos, CA, USA) with an outer diameter of 0.45 mm, was introduced into the duct through a 0.7 mm sheath, while lowpressure irrigation with saline solution was applied, reaching a depth of 2–3 cm. After the procedure a selective microductectomy or a wide excision was performed in each case.
Address correspondence to: Francisco Tresserra, Department of Pathology, Institut Universitari Dexeus, Paseo de la Bonanova 69, 08017 Barcelona, Spain. Tel/Fax: 34 93 211 8390; E-mail: [email protected]
149
150
The Breast
As a control group, 26 surgically removed specimens from the retroareolar region of the breast during the same period, were used. The mean age of these patients was 45.2+10.3 years (range 23–67 years). All surgical specimens were fixed for at least 24 hours in 10% formalin and paraffin embedded. A mean of 5.7+3.1 slides (range 2–15), in the post-duct endoscopy specimens (PDEE) and a mean of 4.6+2.2 slides (range 1–11) for each case of the control group were stained with hematoxlin and eosin. All the slides were reviewed looking for histological changes attributable to the fluid pressure of endoscopy, and were classified as: . Epithelial detachment: When there was detachment of the epithelium from the underlying myoepithelial cell layer . Epithelial loss: When the ductal epithelium lining was totally or partially absent . Periductal clefts: Defined as an optically empty linear space located around the ducts or lobules with an onion skin appearance. These changes were only taken into account when there where five or more concentrical clefts . Stromal disaggregation: When the stroma or fat tissue showed breaks or deposition of eosinophilic fluid with a gelatinous appearance . Epithelial displacement: When small clusters of epithelial cells without glandular arrangement were seen within stromal fissures. The diagnosis, weight and size of the specimens as well as the endoscopic time elapsed were also recorded. To compare values of each variable, the Fisher’s exact test for quantitative variables and Kruskal–Wallis test for numeric variables were applied. A difference was regarded as significant if P was less than 0.05.
RESULTS The most frequent pathological diagnosis in PDEE (Table 1) was fibrocystic (benign breast) change in 13 cases (50%), duct ectasia in 11 cases (42%) and intraductal papilloma in 11 cases (42%). Other diagnoses were intraductal carcinoma (three cases), usual duct hyperplasia (two cases), papillomatosis (two cases) and nipple adenoma (one case). The mean time elapsed during duct endoscopy was 20+5.3 minutes (ranging from 10–25 minutes). In the control group, duct ectasia was seen in 16 cases (62%), intraductal papilloma in 14 cases (54%) and fibrocystic change in 11 cases (42%). Other diagnoses were usual duct hyperplasia (five cases) and papillomatosis (one case). There was a non-
Table 1
Clinical characteristics of cases and controls
Parameter Mean age Diagnosis Papilloma Duct ectasia Fibrocystic changes Non-atipical duct hyperplasia Intraductal carcinoma Papillomatosis Nipple adenoma
Cases (n=26)
Controls (n=26)
P
47.1+10.4
45.2+10.3
NS
11 (42%) 11 (42%) 13 (50%) 2 (8%) 3 (11%) 2 (8%) 1 (4%)
14 (54%) 16 (62%) 11 (42%) 5 (19%) 0 (0%) 1 (4%) 0 (0%)
NS NS NS NS NS NS NS
NS – Non-significant. Table 2
Pathological characteristics of cases and controls
Parameter
Cases (n=26)
Controls (n=26)
P
Specimen weight (g) Specimen size (cm) Number of slides Epithelial detachment Epithelial loss Stromal disaggregation Periductal clefts Epithelial displacement
23.7+17.3 5+1.6 5.7+3.1 19 (73%) 9 (35%) 12 (46%) 20 (77%) 7 (27%)
18.7+10.1 4.6+1.7 4.6+2.2 1 (4%) 0 (0%) 4 (15%) 4 (15%) 0 (0%)
NS NS NS P50.001 P50.001 P50.05 P50.001 P50.05
NS – Non-significant.
statistically significant difference in the distribution of diagnosis between both groups. The mean specimen weight in PDEE was 23.7+17.3 g and in the control group was 18.7+10.1 g. The mean specimen size in PDEE was 5+1.6 cm and in the control group 4.6+1.7 cm. There were no statistically significant differences in weight and size between both groups (Table 2). In 19 of 26 PDEE (73%) there was detachment of the ductal epithelium from the underlying myoepithelial cell layer (Fig. 1), but the epithelium remained within the duct; an optically empty space between the epithelium and myoepithelium in a rail-road appearance could be seen. This feature was more prominent in those ducts with apocrine metaplasia and in ducts with intraductal carcinoma. Epithelial detachment was only seen in one case of the control group. The ductal epithelium was totally or partially lost in nine of 26 PDEE (35%) (Fig. 2). In these ducts, only the myoepithelial cell layer could be seen. This change was not seen in the control group. Epithelial detachment, as well as epithelial loss, were features occurring predominantly in the large duct system. Stromal disaggregation was seen in 12 of 26 PDEE (46%) (Fig. 3), and only in four cases of the control group. In these cases the stroma, or the fatty tissue,
Morphologic changes in breast biopsies after duct endoscopy
Fig. 1 Epithelial detachment. Clear space between ductal epithelium and myoepithelial cell layer with a rail-road appearance (hematoxylineosin6400).
Fig. 3
151
Fig. 2 Epithelial loss. Duct wall with a segment showing intact epithelial cells adjacent to an area lacking epithelial layer but preserved myoepithelial cells. (hematoxylin-eosin6400).
Stromal disaggregation. Breaks in fat tissue and gelatinous appearance of the stroma (hematoxylin-eosin6200).
showed breaks or had a gelatinous appearance due to eosinophilic fluid deposition. Twenty of 26 PDEE (77%) (Fig. 4) showed periductal clefts distributed concentrically in five or more spaces around the duct in an onion skin fashion. This finding was also seen in four cases in the control group. Periductal clefts were more often seen in the small duct
system and around some lobular units. None of the three cases of intraductal carcinoma had periductal clefts seen around involved ducts. In seven cases of PDEE (27%) (Fig. 5), small clusters of epithelial cells were seen within stromal fissures. In three of these cases the epithelium was squamous with keratinization. In the other four the epithelium was of
152
The Breast
Fig. 4
Periductal clefts. Stromal clefts surrounding the duct in an onion skin fashion (hematoxylin-eosin640).
Fig. 5
Epithelial displacement. Fragment of squamous epithelium inside of an stromal cleft (hematoxylin-eosin640).
ductal type and did not show glandular arrangement or nuclear atypia, hence stromal infiltration was excluded. The clusters of displaced epithelium were always close to
a large duct, but in no case a break communicating the duct and the epithelium could be demonstrated after serial sections.
Morphologic changes in breast biopsies after duct endoscopy Only five cases of PDEE did not show any of these features; one case showed one, seven cases two, five cases three, four cases four, and four cases showed all features. Other changes only occasionally seen in the PDEE group included: intraductal hemorrhage (four cases), hemosiderin laden macrophages located between the ductal epithelium and the myoepithelium (two cases), and intraductal cornified squamous cells (one case). All the features described were focally distributed and only present in a few slides. There was no relationship between the morphological changes and the time elapsed during the procedure.
DISCUSSION Collecting mammary ducts open at the tip of the nipple. Larger ducts form the lactiferous sinuses near or within the nipple. Distally, segmental, subsegmental ducts and finally the terminal duct-lobular unit are present.8 The cellular component of the ductal system comprises an inner single layer of columnar epithelial cells and a discontinuous outer layer of myoepithelial cells that are attached to the basement membrane by hemidesmosomes. Epithelial cells insert directly into the basement membrane wherever there is a gap between myoepithelial cells but no well-developed hemidesmosomes have been identified at this level. Both epithelial and myoepithelial cells adhere to the adjacent cell by desmosomes and well-formed tight junctions.9 Duct endoscopy allows the direct view of the mammary ducts, reaching subsegmental ducts.1–7 These ducts show an internal lustrous and smooth surface. In cases with cancer, the surface appears white and slightly elevated, forming a bridging structure. Papillary lesions appear as solid nodules, often reddish if there is haemorrhage.3 Mammary ducts are quite distensible but the fluid perfusion used during duct endoscopy may rupture them.5 The first step in duct rupture is probably epithelial detachment and loss. The site of detachment is between epithelial cells and myoepithelium, where adherence is minimal.9 This explains the persistence of the myoepithelial layer when the epithelium shrinks. In those ducts with an increase in the epithelial cell population (i.e. intraductal carcinoma) the detachment is more prominent and it is also present between different epithelial cell layers. These changes occur mainly in large ducts in which the epithelial surface is larger than in small ducts and lobular units. Lymphocytes and macrophages have been found in the space between epithelial and myoepithelial cells in 5–15% of resting breasts.10 Although we have only seen this feature in two cases, the macrophages contained
153
hemosiderin, and they probably had been pulled through a discontinuity in the epithelium induced by fluid pressure. Periductal clefts probably represent an early stage of ductal rupture. This finding was more frequent in small ducts with low distension capacity. In those cases with intraductal carcinoma in which a desmoplastic stromal reaction can be seen around small ducts,11 no periductal clefting could be elicited, probably due to higher stromal resistance. Although periductal clefts may represent duct rupture, a communication between epithelial gaps and periductal clefts could not be demonstrated, even after serial sectioning. Evidence of duct rupture includes stromal disaggregation and epithelial displacement. Love et al.5 used dye to identify the endoscopied duct, and found dye extravasation in the stroma. In our cases an eosinophilic fluid was seen in the stroma. Epithelial displacement has been previously observed with other diagnostic techniques in breast pathology, such as stereotactic biopsy.12 The viability and biologic significance of displaced epithelium remains unknown. In cases of intraductal carcinoma, these clusters of malignant cells lying within the stroma could suggest infiltrating carcinoma and even be the source of invasive growth. In our series the fact that three cases showed squamous keratinizing epithelium ruled out adenocarcinoma. In the other four cases, the displaced epithelium was of ductal type but lacked glandular arrangement and nuclear atypia. The squamous epithelium as well as the intraductal cornified squamous cells may proceed from desquamated cells from the nipple or nearby lactiferous ducts. Duct endoscopy is a useful tool for the diagnosis of breast pathology but some morphologic changes secondary to the use of dilating media should be taken into account when determining the histological diagnosis.
References 1. Berna J D, Garcia-Medina V, Kuni C C. Ductoscopy: a new technique for ductal exploration. Eur J Radiol 1991; 12: 127–129. 2. Makita M, Sakamoto G, Akiyama F et al. Duct endoscopy and endoscopic biopsy in the evaluation of nipple discharge. Breast Cancer Res Treat 1991; 18: 179–187. 3. Okazaki A, Okazaki M, Asaishi K et al. Fiberoptic ductoscopy of the breast: a new diagnostic procedure for nipple discharge. Jpn J Clin Oncol 1991; 21: 188–193. 4. Barsky S H, Ko J, Freiman C J, Love S M. Pathologic analysis of breast duct endoscoped mastectomies. Mod Pathol 1996; 9: 15A (Abstract 67). 5. Love S M, Barsky S H. Breast-duct endoscopy to study stages of cancerous breast disease. Lancet 1996; 348: 997–999. 6. Okazaki A, Okazaki M, Hirata K, Tsumanuma T. Progress of ductoscopy of the breast (English abstract). Nippon Geka Gakkai Zasshi 1996; 97: 357–362.
154
The Breast
7. Okazaki A, Hirata K, Okazaki M, Svane G, Azavedo E. Nipple discharge disorders: current diagnostic management and the role of fiber-ductoscopy. Eur Radiol 1999; 9: 583–590. 8. Elston C W, Ellis I O. Normal structure and developmental abnormalities. In: Elston C W, Ellis I O (eds.) The Breast. Edinburgh: Churchill Livingstone 1998: 1–19. 9. Stirling J W, Chandler J A. The fine structure of normal, resting terminal ductal-lobular unit of the female breast. Virchow Arch A Path Anat Histol 1976; 372: 205–226.
10. Ferguson D J P. Intraepithelial lymphocytes and macrophages in the normal breast. Virchow Arch [Pathol Anat] 1985; 407: 369–378. 11. Tavassoli F A. Pathology of the breast. Norwalk: Appleton & Lange 1992; 229–261. 12. Liberman L, Vuolo M, Dershaw D D et al. Epithelial displacement after stereotactic 11-gauge directional vacuum-assisted breast biopsy. Am J Roentgenol 1999; 172: 677–681.
ORIGINAL ARTICLE
Morphologic changes in breast biopsies after duct endoscopy F. Tresserra*, R. Fa´bregas*, J. Torrent*, P. J. Grases, C. Ara*, M. Izquierdo and A. Fernandez-Cid* Departments of 1Pathology and 2Gynecology, Institut Universitari Dexeus, Paseo de la Bonanova 69, 08017 Barcelona, Spain S U M M A R Y . Duct endoscopy is a recent technique used for a direct view of the breast ductal system. The aim of this study is to determine any morphological changes in breast tissue attributable to low-pressure irrigation with saline solution that the technique requires. A total of 26 breast biopsies from patients who underwent ductal endoscopy before surgery were compared with 26 breast specimens from the retroareolar region. Breast specimens from duct endoscopy showed more frequent epithelial detachment (73%), epithelial loss (35%), periductal clefts (77%), stromal disaggregation (46%) and displacement of epithelial cells into the stroma (27%) than the control group in which epithelial detachment was seen in 4% of patients, periductal clefts in 15%, and stromal disaggregation in 15%. Epithelial loss and epithelial displacement where not seen in the control group. Although low-pressure fluid perfusion used for duct endoscopy induced morphological changes in breast tissue, these can easily be distinguished from malignancy, and are most likely to occur as the result of duct rupture. # 2001 Harcourt Publishers Ltd
In order to establish the effects of fluid pressure in the mammary duct system, morphologic changes in 26 surgically removed specimens from patients in whom duct endoscopy had been performed, were compared with 26 control breast specimens obtained from the retroareolar region.
INTRODUCTION Recently, new diagnostic techniques have been applied for the diagnosis of breast pathology. One of these ancillary methods is duct endoscopy or ductoscopy, which permits a direct view of the interior of the mammary ducts.1–7 Berna et al.,1 as well as Makita et al.,2 used a rigid fibrescope to examine women with nipple discharge. Love et al.4,5 used it to identify early changes in the ductal epithelium in cases of breast cancer. Okazaki et al.3,6,7 introduced the flexible silicafibrescope that could be introduced into peripheral sites. The technique also allows microbiopsies2 or cytologic samples to be obtained.4,5 The introduction of the endoscope into the ductal system requires a dilating agent. Most authors have used low-pressure irrigation with saline solution,1,5 but air injection has also been tested.5 Although mammary ducts are distensible to certain limits, they can rupture, probably due to excess pressure.5
MATERIALS AND METHODS The case group was composed of pathologic specimens from 26 patients who underwent ductal endoscopy of breast lactiferous ducts before surgical excision. These procedures were performed at our institution during a 4-year period. The mean age of these patients was 47.4+10.4 years (range 17–66 years). For endoscopy, the duct was gradually dilated by a blunt conical-tipped lacrimal dilator. Then the flexible endoscope (FAS002, Conceptus Inc., San Carlos, CA, USA) with an outer diameter of 0.45 mm, was introduced into the duct through a 0.7 mm sheath, while lowpressure irrigation with saline solution was applied, reaching a depth of 2–3 cm. After the procedure a selective microductectomy or a wide excision was performed in each case.
Address correspondence to: Francisco Tresserra, Department of Pathology, Institut Universitari Dexeus, Paseo de la Bonanova 69, 08017 Barcelona, Spain. Tel/Fax: 34 93 211 8390; E-mail: [email protected]
149
150
The Breast
As a control group, 26 surgically removed specimens from the retroareolar region of the breast during the same period, were used. The mean age of these patients was 45.2+10.3 years (range 23–67 years). All surgical specimens were fixed for at least 24 hours in 10% formalin and paraffin embedded. A mean of 5.7+3.1 slides (range 2–15), in the post-duct endoscopy specimens (PDEE) and a mean of 4.6+2.2 slides (range 1–11) for each case of the control group were stained with hematoxlin and eosin. All the slides were reviewed looking for histological changes attributable to the fluid pressure of endoscopy, and were classified as: . Epithelial detachment: When there was detachment of the epithelium from the underlying myoepithelial cell layer . Epithelial loss: When the ductal epithelium lining was totally or partially absent . Periductal clefts: Defined as an optically empty linear space located around the ducts or lobules with an onion skin appearance. These changes were only taken into account when there where five or more concentrical clefts . Stromal disaggregation: When the stroma or fat tissue showed breaks or deposition of eosinophilic fluid with a gelatinous appearance . Epithelial displacement: When small clusters of epithelial cells without glandular arrangement were seen within stromal fissures. The diagnosis, weight and size of the specimens as well as the endoscopic time elapsed were also recorded. To compare values of each variable, the Fisher’s exact test for quantitative variables and Kruskal–Wallis test for numeric variables were applied. A difference was regarded as significant if P was less than 0.05.
RESULTS The most frequent pathological diagnosis in PDEE (Table 1) was fibrocystic (benign breast) change in 13 cases (50%), duct ectasia in 11 cases (42%) and intraductal papilloma in 11 cases (42%). Other diagnoses were intraductal carcinoma (three cases), usual duct hyperplasia (two cases), papillomatosis (two cases) and nipple adenoma (one case). The mean time elapsed during duct endoscopy was 20+5.3 minutes (ranging from 10–25 minutes). In the control group, duct ectasia was seen in 16 cases (62%), intraductal papilloma in 14 cases (54%) and fibrocystic change in 11 cases (42%). Other diagnoses were usual duct hyperplasia (five cases) and papillomatosis (one case). There was a non-
Table 1
Clinical characteristics of cases and controls
Parameter Mean age Diagnosis Papilloma Duct ectasia Fibrocystic changes Non-atipical duct hyperplasia Intraductal carcinoma Papillomatosis Nipple adenoma
Cases (n=26)
Controls (n=26)
P
47.1+10.4
45.2+10.3
NS
11 (42%) 11 (42%) 13 (50%) 2 (8%) 3 (11%) 2 (8%) 1 (4%)
14 (54%) 16 (62%) 11 (42%) 5 (19%) 0 (0%) 1 (4%) 0 (0%)
NS NS NS NS NS NS NS
NS – Non-significant. Table 2
Pathological characteristics of cases and controls
Parameter
Cases (n=26)
Controls (n=26)
P
Specimen weight (g) Specimen size (cm) Number of slides Epithelial detachment Epithelial loss Stromal disaggregation Periductal clefts Epithelial displacement
23.7+17.3 5+1.6 5.7+3.1 19 (73%) 9 (35%) 12 (46%) 20 (77%) 7 (27%)
18.7+10.1 4.6+1.7 4.6+2.2 1 (4%) 0 (0%) 4 (15%) 4 (15%) 0 (0%)
NS NS NS P50.001 P50.001 P50.05 P50.001 P50.05
NS – Non-significant.
statistically significant difference in the distribution of diagnosis between both groups. The mean specimen weight in PDEE was 23.7+17.3 g and in the control group was 18.7+10.1 g. The mean specimen size in PDEE was 5+1.6 cm and in the control group 4.6+1.7 cm. There were no statistically significant differences in weight and size between both groups (Table 2). In 19 of 26 PDEE (73%) there was detachment of the ductal epithelium from the underlying myoepithelial cell layer (Fig. 1), but the epithelium remained within the duct; an optically empty space between the epithelium and myoepithelium in a rail-road appearance could be seen. This feature was more prominent in those ducts with apocrine metaplasia and in ducts with intraductal carcinoma. Epithelial detachment was only seen in one case of the control group. The ductal epithelium was totally or partially lost in nine of 26 PDEE (35%) (Fig. 2). In these ducts, only the myoepithelial cell layer could be seen. This change was not seen in the control group. Epithelial detachment, as well as epithelial loss, were features occurring predominantly in the large duct system. Stromal disaggregation was seen in 12 of 26 PDEE (46%) (Fig. 3), and only in four cases of the control group. In these cases the stroma, or the fatty tissue,
Morphologic changes in breast biopsies after duct endoscopy
Fig. 1 Epithelial detachment. Clear space between ductal epithelium and myoepithelial cell layer with a rail-road appearance (hematoxylineosin6400).
Fig. 3
151
Fig. 2 Epithelial loss. Duct wall with a segment showing intact epithelial cells adjacent to an area lacking epithelial layer but preserved myoepithelial cells. (hematoxylin-eosin6400).
Stromal disaggregation. Breaks in fat tissue and gelatinous appearance of the stroma (hematoxylin-eosin6200).
showed breaks or had a gelatinous appearance due to eosinophilic fluid deposition. Twenty of 26 PDEE (77%) (Fig. 4) showed periductal clefts distributed concentrically in five or more spaces around the duct in an onion skin fashion. This finding was also seen in four cases in the control group. Periductal clefts were more often seen in the small duct
system and around some lobular units. None of the three cases of intraductal carcinoma had periductal clefts seen around involved ducts. In seven cases of PDEE (27%) (Fig. 5), small clusters of epithelial cells were seen within stromal fissures. In three of these cases the epithelium was squamous with keratinization. In the other four the epithelium was of
152
The Breast
Fig. 4
Periductal clefts. Stromal clefts surrounding the duct in an onion skin fashion (hematoxylin-eosin640).
Fig. 5
Epithelial displacement. Fragment of squamous epithelium inside of an stromal cleft (hematoxylin-eosin640).
ductal type and did not show glandular arrangement or nuclear atypia, hence stromal infiltration was excluded. The clusters of displaced epithelium were always close to
a large duct, but in no case a break communicating the duct and the epithelium could be demonstrated after serial sections.
Morphologic changes in breast biopsies after duct endoscopy Only five cases of PDEE did not show any of these features; one case showed one, seven cases two, five cases three, four cases four, and four cases showed all features. Other changes only occasionally seen in the PDEE group included: intraductal hemorrhage (four cases), hemosiderin laden macrophages located between the ductal epithelium and the myoepithelium (two cases), and intraductal cornified squamous cells (one case). All the features described were focally distributed and only present in a few slides. There was no relationship between the morphological changes and the time elapsed during the procedure.
DISCUSSION Collecting mammary ducts open at the tip of the nipple. Larger ducts form the lactiferous sinuses near or within the nipple. Distally, segmental, subsegmental ducts and finally the terminal duct-lobular unit are present.8 The cellular component of the ductal system comprises an inner single layer of columnar epithelial cells and a discontinuous outer layer of myoepithelial cells that are attached to the basement membrane by hemidesmosomes. Epithelial cells insert directly into the basement membrane wherever there is a gap between myoepithelial cells but no well-developed hemidesmosomes have been identified at this level. Both epithelial and myoepithelial cells adhere to the adjacent cell by desmosomes and well-formed tight junctions.9 Duct endoscopy allows the direct view of the mammary ducts, reaching subsegmental ducts.1–7 These ducts show an internal lustrous and smooth surface. In cases with cancer, the surface appears white and slightly elevated, forming a bridging structure. Papillary lesions appear as solid nodules, often reddish if there is haemorrhage.3 Mammary ducts are quite distensible but the fluid perfusion used during duct endoscopy may rupture them.5 The first step in duct rupture is probably epithelial detachment and loss. The site of detachment is between epithelial cells and myoepithelium, where adherence is minimal.9 This explains the persistence of the myoepithelial layer when the epithelium shrinks. In those ducts with an increase in the epithelial cell population (i.e. intraductal carcinoma) the detachment is more prominent and it is also present between different epithelial cell layers. These changes occur mainly in large ducts in which the epithelial surface is larger than in small ducts and lobular units. Lymphocytes and macrophages have been found in the space between epithelial and myoepithelial cells in 5–15% of resting breasts.10 Although we have only seen this feature in two cases, the macrophages contained
153
hemosiderin, and they probably had been pulled through a discontinuity in the epithelium induced by fluid pressure. Periductal clefts probably represent an early stage of ductal rupture. This finding was more frequent in small ducts with low distension capacity. In those cases with intraductal carcinoma in which a desmoplastic stromal reaction can be seen around small ducts,11 no periductal clefting could be elicited, probably due to higher stromal resistance. Although periductal clefts may represent duct rupture, a communication between epithelial gaps and periductal clefts could not be demonstrated, even after serial sectioning. Evidence of duct rupture includes stromal disaggregation and epithelial displacement. Love et al.5 used dye to identify the endoscopied duct, and found dye extravasation in the stroma. In our cases an eosinophilic fluid was seen in the stroma. Epithelial displacement has been previously observed with other diagnostic techniques in breast pathology, such as stereotactic biopsy.12 The viability and biologic significance of displaced epithelium remains unknown. In cases of intraductal carcinoma, these clusters of malignant cells lying within the stroma could suggest infiltrating carcinoma and even be the source of invasive growth. In our series the fact that three cases showed squamous keratinizing epithelium ruled out adenocarcinoma. In the other four cases, the displaced epithelium was of ductal type but lacked glandular arrangement and nuclear atypia. The squamous epithelium as well as the intraductal cornified squamous cells may proceed from desquamated cells from the nipple or nearby lactiferous ducts. Duct endoscopy is a useful tool for the diagnosis of breast pathology but some morphologic changes secondary to the use of dilating media should be taken into account when determining the histological diagnosis.
References 1. Berna J D, Garcia-Medina V, Kuni C C. Ductoscopy: a new technique for ductal exploration. Eur J Radiol 1991; 12: 127–129. 2. Makita M, Sakamoto G, Akiyama F et al. Duct endoscopy and endoscopic biopsy in the evaluation of nipple discharge. Breast Cancer Res Treat 1991; 18: 179–187. 3. Okazaki A, Okazaki M, Asaishi K et al. Fiberoptic ductoscopy of the breast: a new diagnostic procedure for nipple discharge. Jpn J Clin Oncol 1991; 21: 188–193. 4. Barsky S H, Ko J, Freiman C J, Love S M. Pathologic analysis of breast duct endoscoped mastectomies. Mod Pathol 1996; 9: 15A (Abstract 67). 5. Love S M, Barsky S H. Breast-duct endoscopy to study stages of cancerous breast disease. Lancet 1996; 348: 997–999. 6. Okazaki A, Okazaki M, Hirata K, Tsumanuma T. Progress of ductoscopy of the breast (English abstract). Nippon Geka Gakkai Zasshi 1996; 97: 357–362.
154
The Breast
7. Okazaki A, Hirata K, Okazaki M, Svane G, Azavedo E. Nipple discharge disorders: current diagnostic management and the role of fiber-ductoscopy. Eur Radiol 1999; 9: 583–590. 8. Elston C W, Ellis I O. Normal structure and developmental abnormalities. In: Elston C W, Ellis I O (eds.) The Breast. Edinburgh: Churchill Livingstone 1998: 1–19. 9. Stirling J W, Chandler J A. The fine structure of normal, resting terminal ductal-lobular unit of the female breast. Virchow Arch A Path Anat Histol 1976; 372: 205–226.
10. Ferguson D J P. Intraepithelial lymphocytes and macrophages in the normal breast. Virchow Arch [Pathol Anat] 1985; 407: 369–378. 11. Tavassoli F A. Pathology of the breast. Norwalk: Appleton & Lange 1992; 229–261. 12. Liberman L, Vuolo M, Dershaw D D et al. Epithelial displacement after stereotactic 11-gauge directional vacuum-assisted breast biopsy. Am J Roentgenol 1999; 172: 677–681.