- Email: [email protected]
Mucosal Ultrastructure of Continent Cecal Reservoir for Urine and its Ileal Nipple Valve 2–9 Years After Construction
0022-534 7 /90/1432-0372$02.00/0 THE JOURNAL OF UROLOGY Copyright© 1990 by AMERICAN UROLOGICAL ASSOCIATION, INC.
Vol. 143, February Printed in U.S.A.
MUCOSAL ULTRASTRUCTURE OF CONTINENT CECAL RESERVOIR FOR URINE AND ITS ILEAL NIPPLE VALVE 2-9 YEARS AFTER CONSTRUCTION BIRGITTA CARLEN, ROGER WILLEN
AND
WIKING MANSSON*
From the Departments of Pathology and Urology, University Hospital, Lund, Sweden
ABSTRACT
Biopsy specimens from continent cecal reservoir for urine and from its ileal nipple valve were studied with electron microscopy after two to nine years of function in ten patients. In the colonic mucosa from the reservoir there was shortening of microvilli, in some cases with random orientation and numerical reduction-changes unrelated to the time from reservoir construction. Filamentous core rootlets were also randomly oriented and numerically diminished. Glycocalyceal bodies were present in most cases. Mucosal edema and reduced numbers of goblet cells were found in six cases and increased amount of collagen in two. In the ileal nipple valve mucosa there were no microvillous changes, but metaplastic formation of glycocalyceal bodies was interpreted as adaptation to physiologic conditions comparable with those in the reservoir's colonic mucosa. Collagen increase was found in two of the nipple valves. Neurogenic processes, enterochromaffin cells and Paneth cells were always well preserved in normal amounts in the cecal as well as the ileal mucosa. (J. Ural., 143: 372-376, 1990) Incorporation of bowel segments into the urinary tract is increasingly common. The reasons include the rising incidence of bladder carcinoma and new techniques for reconstructing the urinary tract. Although ileal and colonic conduits remain the most widely used methods for urinary diversion, interest in continent diversion and bladder replacement is rapidly growing. Collection and storage of urine in a receptacle formed from bowel may have important physiologic implications, however, involving transport of water and electrolytes across the mucosa, absorption of urinary constituents and secretion of various substances. Furthermore, the contact of urine with the bowel wall may influence the dynamic behavior of the isolated bowel segment. The mucosa of ileum used as a conduit or reservoir for urine undergoes progressive villous atrophy. 1- 3 The colonic mucosa, by contrast, was found at light microscopy to be little affected in patients with ureterosigmoidostomy, sigmoid conduit or continent cecal reservoir. 4• 5 The ultrastructure of colonic mucosa in contact with urine does not seem to have been previously studied, however, though it may be important for the interpretation of physiologic events. The present report concerns the subcellular structure of the mucosa of continent cecal reservoir for urine and of its ileal nipple valve. MATERIALS AND METHODS
The mucosal ultrastructure of the cecal reservoir for urine and its ileal outlet was studied in eight men and two women who had undergone cystourethrectomy and urinary diversion via a continent cecal reservoir because of bladder carcinoma two to nine years (mean 6.5 years) previously. Preoperative radiotherapy (45 Gy in four weeks or 20 Gy in one week) had been given to eight patients. At the time of cystectomy the patients' age range was 38 to 64 (mean 53) years. The technique for construction of the cecal reservoir was previously described. 6 • 7 Briefly, an ileocecal segment was isoAccepted for publication October 6, 1989. *Requests for reprints: Dept. of Urology, University Hospital, 8-221 85 Lund, Sweden. Supported by grants from the Maud and Birger Gustavsson Fund, Stockholm and the John and Augusta Persson Fund, Medical Faculty, University of Lund, Sweden.
lated, the ureters implanted into the cecum and a continenceproviding intussuscepted ileal nipple valve constructed in the reservoir outlet. The nipple base was anchored to the abdominal wall and a flush ileal stoma fashioned. Two of the ten patients in the present study currently wear a collecting device-one who has occasional slight leakage of urine and empties the reservoir through a small opening in the device, and one with gross leakage. A third patient is awaiting revisional surgery and meanwhile the reservoir is drained via a catheter. The remaining seven patients are continent and empty the reservoir every four to five hours by introducing a plastic or silicone catheter. Mucosal biopsies of the cecal reservoir and the ileal nipple valve were performed via a cystoscope. Urine for culture was obtained on the same occasions. Control biopsies from ileum and cecum were obtained from three patients aged 57, 58 and 70 years, who underwent ileocecal resection because of severe constipation or irritable colon or for augmentation of the urinary bladder. Transmission Electron Microscopy. The biopsy specimens were immediately fixed in 2% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.2), postfixed in 2% osmium tetroxide in S-collidine buffer (pH 7.2), dehydrated in ethanol and embedded in agar resin 100. A semithin section was cut and examined in a light microscope. A representative area was chosen and six ultrathin sections, approximately 50 nm, were cut on an LKB Ultrotome III and contrasted with uranyl acetate and lead citrate. 8 The grids were examined in a Zeiss EM 10 electron microscope working at 60 kV. Each ultrathin section from the cecal reservoir mucosa contained c. 130 enterocytes, 30 goblet cells and 10 crypts. The corresponding figures from the mucosa of the ileal outlet were 200, 70 and 5. In each ultrathin section from the control cecal mucosa c. 150 enterocytes, 40 goblet cells and 10 crypts were present, while the control ileal mucosa contained 320, 80 and 10, respectively. Each biopsy (= ultrathin section) was scanned from the surface to the lamina muscularis mucosa. The microvilli and the filamentous core rootlets were examined with regard to length, orientation and number. As the findings were uniform within each case, 30 microvilli from several enterocytes were used for measurement of length. Presence of glycocalyceal bodies and R-bodies was noted. The jupctional complexes were
372
373
MUCOSAL ULTRASTRUCTURE OF CONTINENT CECAL RESERVOIR TABLE
Electron Microscopy Microvilli length, range (mean) µ,m. random orientation reduced number Filamentous core rootlets random orientation reduced number Glycocalyceal bodies Edema Inflammation mucosa submucosa Goblet cells numerically reduced Collagen increased N eurogenic processes Enterochromaffin cells Paneth cells
1. Ultrastructural changes in ileocecal mucosa Cecal Reservoir Patients Colon Ileum
Control Group Colon
Ileum
= 3)
= 2)
(n = 10)
(n = 9)
0.2-0. 7 (0.5) 3 5
0.7-1.6 (1.3) 0 0
0.7-1.0 (0.9) 0 1
1.6-2.0 (1.8) 0 0
2 5 8 6
1 1 5 4
0 0 3 0
0 0 0 0
1 3 6 2
0 1 1 2
0 1 0 0
1 0 1 0
well preserved present present
FIG. 1. Biopsy specimen of cecal reservoir mucosa nine years postoperatively. Microvilli are markedly reduced in number and height. Preserved glycocalyceal bodies (arrow) and R-body (thick arrow) containing vesicles and rod-like profiles. Xl6 000.
studied, and also the nucleus and its position in the enterocyte. The degree of edema and inflammation in the mucosa and submucosa was evaluated. The number and functional status of the goblet cells were recorded, as were observations on collagen, neurogenic processes, enterochromaffin cells and Paneth cells. Urine culture. Quantitative urine cultures were made in the routine microbiologic laboratory. RESULTS
The main results obtained at transmission electron microscopy from the ten patients in the cecal reservoir group and in the three control patients are surveyed in table 1. The appreciable shortening of the microvilli in the cecal mucosa exposed to urine as compared with the controls is illustrated in figs. 1 and 2. Shortening of microvilli was unrelated to time after construction of the reservoir: Indeed, the patient with the least mean height (0.2 µm.) had the shortest observation time. Nor was random orientation or numerical reduction of microvilli (three and five cases, respectively) associated with follow-up time. Glycocalyceal bodies were present in all but two speci-
(n
(n
well preserved present present
FIG. 2. 'Control' colon. Microvilli of normal height, with prominent filamentous core rootlets (arrowhead) and glycocalyceal bodies (arrow). Some of center microvilli are partly covered with mucus. X 16 0000.
mens from colonic mucosa, and it was also possible to demonstrate R-bodies (fig. 1), structures from which glycocalyceal bodies are thought to develop. 9 In some specimens the filamentous core rootlets extending into the cytoplasm and forming the terminal web showed random orientation and, in those with reduced number of microvilli, numerical diminution. The length of the filamentous core rootlets seemed to be reduced in some cases, but angular distribution of the rootlets made precise measurements difficult. More or less advanced edema was found in six of the ten cecal reservoir cases (table 1), but none in control specimens (fig. 3 vs. fig. 4). Inflammatory cell reaction was found in the mucosa in one of the six edematous specimens and in the submucosa in three. The amount of edema seemed to correlate to the degree of acute and chronic inflammatory cell response together with mast cells and some granular cells. Urine cultures yielded no bacterial growth in three patients, but heavy growth of Escherichia coli in five and of Klebsiella pneumoniae in two patients. Of the six patients with edema in the cecal reservoir wall, three had positive urine culture. However, in no case were bacteria found on the surface or in deeper sections from the cecal reservoir. Numerical reduction of goblet cells (six cases)
374
CARLEN, WILLEN AND MANSSON
numerical reduction was found in these cases or in the control group. Glycocalyceal bodies were found in five of the nine ileal biopsies from the reservoir group (fig. 5), but in none of the controls (fig. 6), and in one of the former there was random
FIG. 3. Cecal mucosa eight years after reservoir construction, showing interstitial edema. Intact junctional complexes at cell apices (arrows). To the right a goblet cell filled with mucigen granules is seen. X 2 000.
FIG. 5. Mucosa of ilea! nipple eight years after construction of cecal reservoir for urine. Microvilli with filamentous core rootlets (arrowhead) and glyocalyceal bodies (arrows). Well-preserved tight junction (T) near lumen. Intermediate junction (I) and desmosome (D) are also demonstrated. X25 000.
FIG. 4. Well preserved 'control' colonic mucosa. No edema. X 2 500.
was unrelated to time after urinary diversion. The two patients with increased submucosal collagen (table 1) had received preoperative radiotherapy, and in one of them there was poor reservoir function, with gross leakage of urine. In the biopsy specimens from the ileal nipple valves (table 1) the microvilli were of normal length. No random orientation or
FIG. 6. Ilea! mucosa, 'control'. Well-preserved microvilli with filamentous core rootlets (arrowheads), but no glycocalyceal bodies. X25 000.
MUCOSAL ULTRASTRUCTURE OF CONTINENT CECAL RESERVOIR
orientation and reduced number of filamentous core rootlets. Edema was prominent in four of the nine ileal valve specimens, one of which also showed some submucosal inflammation. Goblet cells were numerically reduced in one of these nine specimens. Enhancement of collagen bundles in deeper layers of the ileum was demonstrated, and in two specimens there was increase of collagen in the mucosal area. The enterocytes in all specimens were of normal height with nuclei of normal appearance and position. Structures such as neurogenic processes, enterochromaffin cells and Paneth cells all were well preserved in cecal reservoir as well as in ileal outlet. The junctional complexes between the cells, with their tight junctions and desmosomes, were always found to be intact (fig. 3, 5). DISCUSSION
The aim of the present study was to visualize ultrastructural changes in the mucosa of the cecal segment used as a continent urinary reservoir, and of its ileal nipple valve, after two to nine years of reservoir function and to compare the findings with a control group. The most striking changes were in the luminal part of the mucosa. In previous studies with scanning electron microscopy10 we found well defined borders of the cecal enterocytes and the microvilli appeared to be well preserved. Transmission electron microscopy, however, now revealed considerable changes in the luminal mucosa. The microvilli thus were shortened (from c. 1 to 0.5 µm.) and remarkably scanty (fig. 1). The microvilli are fundamentally important for absorption, as they amplify the surface area of the intestine's absorptive cells. As all absorbed substance must pass through this area, reduction in length and number of microvilli may detrimentally affect absorption. In the Kock pouch for urine there is progressive reduction in villous height leading to complete avillous areas together with severe loss of microvilli. 2·3 Our study showed no correlation between changes in microvilli and postoperative observation time. Previous radiotherapy may be a cause of microvillous changes. 1 In our two patients who had not received radiotherapy, however, the microvilli were very short (0.3 µm.) and the filamentous core rootlets, extending into the enterocytes from the microvilli, showed random orientation and numerical reduction. Presently we do not know if these changes, which seem to be less pronounced in cecal than in ileal receptacles for urine, have any clinical significance. It is tempting to regard them as an advantageous adaptation to a new environment, resulting in decreased absorptive capacity for urinary constituents. 2·11 Gradually reduced absorption of L-phenylalanine from the Kock pouch has been found. 12 The intact junctional complexes between the cecal enterocytes at their apices are also important in this context, indicating a preserved barrier separating the luminal content from the tissue fluid in the intercellular space. On the other hand, it cannot be ruled out that the increase of inflammatory cells in the cecal reservoir wall, mainly lymphocytes and eosinophils, 5 reflects enhanced mucosal permeability. While the mucosal microvilli were flattened in the cecal reservoir, those in the ileal outlet were largely unchanged. That lack of change, however, was not incompatible with observations of pronounced reduction in numbers and length of ileal mucosal microvilli in contact with urine. 1-3 In our cases the biopsy specimens were taken from the inside of continenceproviding nipple valves, which should contain no urine. Glycocalyceal bodies are small, discrete structures enmeshed in the glycocalyx on the surface of the microvilli. These mem brane-bound, spherical vesicular bodies 13 are found in normal and pathologic human colonic and rectal mucosa. 14 The combination of microvilli with prominent core rootlets and glycocalyceal bodies has been used as a distinct marker for certain varieties of adenocarcinoma and in metaplastic intestinal-type epithelium. 15-17 Whether or not glycocalyceal bodies occur in healthy small bowel has been disputed. 9·14·15 We found these
375
bodies in most of the biopsy specimens from cecal reservoir as well as from ileal nipple valve. They have also been observed in ileal conduits.' The functional implications of glycocalyceal bodies are not clear, since the exact manner of their formation is unknown. Nor have the functional properties of these structures been identified. Possibly their occurrence in ileal isolates functioning as conduit or reservoir outlet expresses adaptation to a physiologic situation comparable with that in colonic mucosa. Edema resulted in cell spacing in the luminal mucosa in some patients, and seemed to be associated with the inflammatory cell reaction. Concomitant bacteriuria may have been the cause, although such correlation was not found in a study of ileal conduits.' Most of our patients had positive urine cultures, but in no case did we find bacteria adhering to the cecal mucosal surface. Scanning electron microscopy of cecal reservoir biopsies gave similar results, 10 in analogy with a study that failed to demonstrate mucosa-adherent bacteria in ileal conduits. 18 The continuing production of immunoglobulins19 and mucins may be of importance in this context. Mucins are conjugated glycoproteins, and construction of cecal urinary reservoir usually is followed by a shift from sulfomucins to sialomucins, which we have interpreted as a secondary, nonspecific phenomenon.5 The two patients with increased amounts of collagen in the submucosa of the cecal reservoir had received preoperative radiotherapy, which has been suggested as a cause of fibrosis in ileal conduits. 1 Fibrotic stricture of ileal conduit can occur without preceding irradiation, however. Progressive fibrosis may be detrimental to the function of a reservoir as an expandable receptacle for urine. The mucosal fibrosis found in the ileal outlet of the reservoir in two cases may be explained as an ischemic effect caused by mesenteric exclusion and intermittent trauma from catheterization. Preformed anatomic structures such as neurogenic processes, enterochromaffin cells and Paneth cells were all clearly present in normal amounts and shapes in cecal as well as in ileal mucosa. Enterochromaffin cells can be found throughout the gastrointestinal tract, where they produce a number of hormones. In the colon they secrete serotonin, somatostatin and several other polypeptides with hormonal activity. 20 The Paneth cells were previously observed to remain unchanged in ileal conduits. 21 Although their exact functional role has not been established, Paneth cells are known to be capable of producing lysozyme, which can digest bacterial cell walls and phagocytize microorganisms. 22 Together with immunoglobulins, especially secretory IgA, 19 the products of Paneth cells may be important as host defense mechanisms against urinary tract infection in a cecal reservoir for urine. REFERENCES 1. Lindell, 0., Makinen, J., Nickles, J. and Lehtonen, T.: Mucosa!
2.
3.
4. 5. 6.
morphology of ilea! conduits with particular reference to urinary infections. Eur. Urol., 12: 201, 1986. Hansson, H.-A., Kock, N. G., Norlen, L., Philipson, B., Trasti, H. and Ahren, C.: Morphological observations in pedicled ilea! grafts used for construction of continent reservoirs for urine. Scand. J. Urol. Nephrol., Suppl., 49: 49, 1978. Philipson, B. M., Kock, N. G., Hiickenstriim, T., Norlen, L. J., Ahren, C. and Hansson, H.-A.: Ultrastructural and histochemical changes in ilea! reservoir mucosa after long-term exposure to urine. A study in patients with continent urostomy (Kock pouch). Scand. J. Gastroenterol., 21: 1235, 1986. Berg, N. 0., Fredlund, P., Mansson, W. and Olsson, S.-A.: Surveillance colonoscopy and biopsy in patients with ureterosigmoidostomy. Endoscopy, 19: 60, 1987. Mansson, W. and Willen, R.: Mucosa! morphology and histochemistry of the continent cecal reservoir for urine. J. Urol., 139: 1199, 1988. Mansson, W., Colleen, S. and Sundin, T.: Continent caecal reservoir in urinary diversion. Brit. J. Urol., 56: 359, 1984.
376
CARLEN, WILLEN AND MANSSON
7. Mansson, W. and Colleen, S.: The continent cecal reservoir for urine. Sem. Urol., 5: 63, 1987. 8. Reynolds, E. S.: The use of lead citrate at high pH as an electronopaque stain in electron microscopy. J. Cell. Biol., 17: 208, 1963. 9. Ghadially, F. N.: Diagnostic electron microscopy of tumours. 2nd ed. Butterworths, London, pp. 334, 1985. 10. Mansson, W.: The continent cecal urinary reservoir. In: Bladder Reconstruction and Continent Urinary Diversion. Ed. by L. R. King, A. R. Stone and G. D. Webster. Chicago, Yearbook Medical Publishers, chapt. 15, p. 209, 1987. 11. Philipson, B. M., Kock, N. G., Jagenburg, R., Ahren, C., Norlen, L., Robinson, J. W. L. and Menge, H.: Functional and structural studies of ilea! reservoirs used for continent urostomy and ileostomy. Gut, 24: 392, 1983. 12. Akerlund, S., Jagenburg, R., Kock, N. G. and Philipson, B. M.: Absorption ofL-phenylalanine in human ilea! reservoirs exposed to urine. Urol. Res., 16: 321, 1988. 13. Schnitka, T. K.: Current concepts of the pathogenesis and pathology of inflammatory lesions of the intestine. Can. Med. Ass. J., 91: 7, 1964. 14. Stone, J., Mukherjee, T. M. and Hesher, R.: C-bodies and R-bodies in the epithelial cells of normal and diseased human rectum. Arch. Path. Lab. Med., 101: 436, 1977.
15. Marcus, P. B.: Glycocalyceal bodies and their role in tumor typing. J. Submicroscopic Cytol., 13: 403, 1981. 16. Marcus, P. B., Martin, J. H., Green, R. H. and Krouse, M. A.: Glycocalyceal bodies and microvillous core rootlets. Arch. Path. Lab. Med., 103: 89, 1979. 17. Mukherejee, T. M.: The role of electron microscopy in the diagnosis of neoplastic cells in effusion fluids. J. Submicroscopic Cytol., 14: 717, 1982. 18. Bruce, A. W., Reid, G., Chan, R. C. Y. and Costerton, J. W.: Bacterial adherence in the human ilea! conduit: A morphological and bacteriological study. J. Urol., 132: 184, 1984. 19. Mansson, W., Colleen, S., Low, K., Mardh, P.-A. and Lundblad, A.: Immunoglobulins in urine from patients with ilea! and colonic conduits and reservoirs. J. Urol., 133: 713, 1985. 20. Grube, H. and Forsmann, G.: Morphology and function of the entero-endocrine cells. Horm. Metab. Res., 11: 603, 1979. 21. Garner, J. W., Goldstein, A. M. B. and Cosgrove, M. D.: Histologic appearance of the intestinal urinary conduit. J. Urol., 114: 854, 1979. 22. Erlandsen, S. L. and Chase, D. G.: Paneth cell function: phagocytosis and intracellular digestion of intestinal microorganisms. I. Hexamita Muris. J. Ultrastruct. Res., 41: 296, 1972.
Vol. 143, February Printed in U.S.A.
MUCOSAL ULTRASTRUCTURE OF CONTINENT CECAL RESERVOIR FOR URINE AND ITS ILEAL NIPPLE VALVE 2-9 YEARS AFTER CONSTRUCTION BIRGITTA CARLEN, ROGER WILLEN
AND
WIKING MANSSON*
From the Departments of Pathology and Urology, University Hospital, Lund, Sweden
ABSTRACT
Biopsy specimens from continent cecal reservoir for urine and from its ileal nipple valve were studied with electron microscopy after two to nine years of function in ten patients. In the colonic mucosa from the reservoir there was shortening of microvilli, in some cases with random orientation and numerical reduction-changes unrelated to the time from reservoir construction. Filamentous core rootlets were also randomly oriented and numerically diminished. Glycocalyceal bodies were present in most cases. Mucosal edema and reduced numbers of goblet cells were found in six cases and increased amount of collagen in two. In the ileal nipple valve mucosa there were no microvillous changes, but metaplastic formation of glycocalyceal bodies was interpreted as adaptation to physiologic conditions comparable with those in the reservoir's colonic mucosa. Collagen increase was found in two of the nipple valves. Neurogenic processes, enterochromaffin cells and Paneth cells were always well preserved in normal amounts in the cecal as well as the ileal mucosa. (J. Ural., 143: 372-376, 1990) Incorporation of bowel segments into the urinary tract is increasingly common. The reasons include the rising incidence of bladder carcinoma and new techniques for reconstructing the urinary tract. Although ileal and colonic conduits remain the most widely used methods for urinary diversion, interest in continent diversion and bladder replacement is rapidly growing. Collection and storage of urine in a receptacle formed from bowel may have important physiologic implications, however, involving transport of water and electrolytes across the mucosa, absorption of urinary constituents and secretion of various substances. Furthermore, the contact of urine with the bowel wall may influence the dynamic behavior of the isolated bowel segment. The mucosa of ileum used as a conduit or reservoir for urine undergoes progressive villous atrophy. 1- 3 The colonic mucosa, by contrast, was found at light microscopy to be little affected in patients with ureterosigmoidostomy, sigmoid conduit or continent cecal reservoir. 4• 5 The ultrastructure of colonic mucosa in contact with urine does not seem to have been previously studied, however, though it may be important for the interpretation of physiologic events. The present report concerns the subcellular structure of the mucosa of continent cecal reservoir for urine and of its ileal nipple valve. MATERIALS AND METHODS
The mucosal ultrastructure of the cecal reservoir for urine and its ileal outlet was studied in eight men and two women who had undergone cystourethrectomy and urinary diversion via a continent cecal reservoir because of bladder carcinoma two to nine years (mean 6.5 years) previously. Preoperative radiotherapy (45 Gy in four weeks or 20 Gy in one week) had been given to eight patients. At the time of cystectomy the patients' age range was 38 to 64 (mean 53) years. The technique for construction of the cecal reservoir was previously described. 6 • 7 Briefly, an ileocecal segment was isoAccepted for publication October 6, 1989. *Requests for reprints: Dept. of Urology, University Hospital, 8-221 85 Lund, Sweden. Supported by grants from the Maud and Birger Gustavsson Fund, Stockholm and the John and Augusta Persson Fund, Medical Faculty, University of Lund, Sweden.
lated, the ureters implanted into the cecum and a continenceproviding intussuscepted ileal nipple valve constructed in the reservoir outlet. The nipple base was anchored to the abdominal wall and a flush ileal stoma fashioned. Two of the ten patients in the present study currently wear a collecting device-one who has occasional slight leakage of urine and empties the reservoir through a small opening in the device, and one with gross leakage. A third patient is awaiting revisional surgery and meanwhile the reservoir is drained via a catheter. The remaining seven patients are continent and empty the reservoir every four to five hours by introducing a plastic or silicone catheter. Mucosal biopsies of the cecal reservoir and the ileal nipple valve were performed via a cystoscope. Urine for culture was obtained on the same occasions. Control biopsies from ileum and cecum were obtained from three patients aged 57, 58 and 70 years, who underwent ileocecal resection because of severe constipation or irritable colon or for augmentation of the urinary bladder. Transmission Electron Microscopy. The biopsy specimens were immediately fixed in 2% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.2), postfixed in 2% osmium tetroxide in S-collidine buffer (pH 7.2), dehydrated in ethanol and embedded in agar resin 100. A semithin section was cut and examined in a light microscope. A representative area was chosen and six ultrathin sections, approximately 50 nm, were cut on an LKB Ultrotome III and contrasted with uranyl acetate and lead citrate. 8 The grids were examined in a Zeiss EM 10 electron microscope working at 60 kV. Each ultrathin section from the cecal reservoir mucosa contained c. 130 enterocytes, 30 goblet cells and 10 crypts. The corresponding figures from the mucosa of the ileal outlet were 200, 70 and 5. In each ultrathin section from the control cecal mucosa c. 150 enterocytes, 40 goblet cells and 10 crypts were present, while the control ileal mucosa contained 320, 80 and 10, respectively. Each biopsy (= ultrathin section) was scanned from the surface to the lamina muscularis mucosa. The microvilli and the filamentous core rootlets were examined with regard to length, orientation and number. As the findings were uniform within each case, 30 microvilli from several enterocytes were used for measurement of length. Presence of glycocalyceal bodies and R-bodies was noted. The jupctional complexes were
372
373
MUCOSAL ULTRASTRUCTURE OF CONTINENT CECAL RESERVOIR TABLE
Electron Microscopy Microvilli length, range (mean) µ,m. random orientation reduced number Filamentous core rootlets random orientation reduced number Glycocalyceal bodies Edema Inflammation mucosa submucosa Goblet cells numerically reduced Collagen increased N eurogenic processes Enterochromaffin cells Paneth cells
1. Ultrastructural changes in ileocecal mucosa Cecal Reservoir Patients Colon Ileum
Control Group Colon
Ileum
= 3)
= 2)
(n = 10)
(n = 9)
0.2-0. 7 (0.5) 3 5
0.7-1.6 (1.3) 0 0
0.7-1.0 (0.9) 0 1
1.6-2.0 (1.8) 0 0
2 5 8 6
1 1 5 4
0 0 3 0
0 0 0 0
1 3 6 2
0 1 1 2
0 1 0 0
1 0 1 0
well preserved present present
FIG. 1. Biopsy specimen of cecal reservoir mucosa nine years postoperatively. Microvilli are markedly reduced in number and height. Preserved glycocalyceal bodies (arrow) and R-body (thick arrow) containing vesicles and rod-like profiles. Xl6 000.
studied, and also the nucleus and its position in the enterocyte. The degree of edema and inflammation in the mucosa and submucosa was evaluated. The number and functional status of the goblet cells were recorded, as were observations on collagen, neurogenic processes, enterochromaffin cells and Paneth cells. Urine culture. Quantitative urine cultures were made in the routine microbiologic laboratory. RESULTS
The main results obtained at transmission electron microscopy from the ten patients in the cecal reservoir group and in the three control patients are surveyed in table 1. The appreciable shortening of the microvilli in the cecal mucosa exposed to urine as compared with the controls is illustrated in figs. 1 and 2. Shortening of microvilli was unrelated to time after construction of the reservoir: Indeed, the patient with the least mean height (0.2 µm.) had the shortest observation time. Nor was random orientation or numerical reduction of microvilli (three and five cases, respectively) associated with follow-up time. Glycocalyceal bodies were present in all but two speci-
(n
(n
well preserved present present
FIG. 2. 'Control' colon. Microvilli of normal height, with prominent filamentous core rootlets (arrowhead) and glycocalyceal bodies (arrow). Some of center microvilli are partly covered with mucus. X 16 0000.
mens from colonic mucosa, and it was also possible to demonstrate R-bodies (fig. 1), structures from which glycocalyceal bodies are thought to develop. 9 In some specimens the filamentous core rootlets extending into the cytoplasm and forming the terminal web showed random orientation and, in those with reduced number of microvilli, numerical diminution. The length of the filamentous core rootlets seemed to be reduced in some cases, but angular distribution of the rootlets made precise measurements difficult. More or less advanced edema was found in six of the ten cecal reservoir cases (table 1), but none in control specimens (fig. 3 vs. fig. 4). Inflammatory cell reaction was found in the mucosa in one of the six edematous specimens and in the submucosa in three. The amount of edema seemed to correlate to the degree of acute and chronic inflammatory cell response together with mast cells and some granular cells. Urine cultures yielded no bacterial growth in three patients, but heavy growth of Escherichia coli in five and of Klebsiella pneumoniae in two patients. Of the six patients with edema in the cecal reservoir wall, three had positive urine culture. However, in no case were bacteria found on the surface or in deeper sections from the cecal reservoir. Numerical reduction of goblet cells (six cases)
374
CARLEN, WILLEN AND MANSSON
numerical reduction was found in these cases or in the control group. Glycocalyceal bodies were found in five of the nine ileal biopsies from the reservoir group (fig. 5), but in none of the controls (fig. 6), and in one of the former there was random
FIG. 3. Cecal mucosa eight years after reservoir construction, showing interstitial edema. Intact junctional complexes at cell apices (arrows). To the right a goblet cell filled with mucigen granules is seen. X 2 000.
FIG. 5. Mucosa of ilea! nipple eight years after construction of cecal reservoir for urine. Microvilli with filamentous core rootlets (arrowhead) and glyocalyceal bodies (arrows). Well-preserved tight junction (T) near lumen. Intermediate junction (I) and desmosome (D) are also demonstrated. X25 000.
FIG. 4. Well preserved 'control' colonic mucosa. No edema. X 2 500.
was unrelated to time after urinary diversion. The two patients with increased submucosal collagen (table 1) had received preoperative radiotherapy, and in one of them there was poor reservoir function, with gross leakage of urine. In the biopsy specimens from the ileal nipple valves (table 1) the microvilli were of normal length. No random orientation or
FIG. 6. Ilea! mucosa, 'control'. Well-preserved microvilli with filamentous core rootlets (arrowheads), but no glycocalyceal bodies. X25 000.
MUCOSAL ULTRASTRUCTURE OF CONTINENT CECAL RESERVOIR
orientation and reduced number of filamentous core rootlets. Edema was prominent in four of the nine ileal valve specimens, one of which also showed some submucosal inflammation. Goblet cells were numerically reduced in one of these nine specimens. Enhancement of collagen bundles in deeper layers of the ileum was demonstrated, and in two specimens there was increase of collagen in the mucosal area. The enterocytes in all specimens were of normal height with nuclei of normal appearance and position. Structures such as neurogenic processes, enterochromaffin cells and Paneth cells all were well preserved in cecal reservoir as well as in ileal outlet. The junctional complexes between the cells, with their tight junctions and desmosomes, were always found to be intact (fig. 3, 5). DISCUSSION
The aim of the present study was to visualize ultrastructural changes in the mucosa of the cecal segment used as a continent urinary reservoir, and of its ileal nipple valve, after two to nine years of reservoir function and to compare the findings with a control group. The most striking changes were in the luminal part of the mucosa. In previous studies with scanning electron microscopy10 we found well defined borders of the cecal enterocytes and the microvilli appeared to be well preserved. Transmission electron microscopy, however, now revealed considerable changes in the luminal mucosa. The microvilli thus were shortened (from c. 1 to 0.5 µm.) and remarkably scanty (fig. 1). The microvilli are fundamentally important for absorption, as they amplify the surface area of the intestine's absorptive cells. As all absorbed substance must pass through this area, reduction in length and number of microvilli may detrimentally affect absorption. In the Kock pouch for urine there is progressive reduction in villous height leading to complete avillous areas together with severe loss of microvilli. 2·3 Our study showed no correlation between changes in microvilli and postoperative observation time. Previous radiotherapy may be a cause of microvillous changes. 1 In our two patients who had not received radiotherapy, however, the microvilli were very short (0.3 µm.) and the filamentous core rootlets, extending into the enterocytes from the microvilli, showed random orientation and numerical reduction. Presently we do not know if these changes, which seem to be less pronounced in cecal than in ileal receptacles for urine, have any clinical significance. It is tempting to regard them as an advantageous adaptation to a new environment, resulting in decreased absorptive capacity for urinary constituents. 2·11 Gradually reduced absorption of L-phenylalanine from the Kock pouch has been found. 12 The intact junctional complexes between the cecal enterocytes at their apices are also important in this context, indicating a preserved barrier separating the luminal content from the tissue fluid in the intercellular space. On the other hand, it cannot be ruled out that the increase of inflammatory cells in the cecal reservoir wall, mainly lymphocytes and eosinophils, 5 reflects enhanced mucosal permeability. While the mucosal microvilli were flattened in the cecal reservoir, those in the ileal outlet were largely unchanged. That lack of change, however, was not incompatible with observations of pronounced reduction in numbers and length of ileal mucosal microvilli in contact with urine. 1-3 In our cases the biopsy specimens were taken from the inside of continenceproviding nipple valves, which should contain no urine. Glycocalyceal bodies are small, discrete structures enmeshed in the glycocalyx on the surface of the microvilli. These mem brane-bound, spherical vesicular bodies 13 are found in normal and pathologic human colonic and rectal mucosa. 14 The combination of microvilli with prominent core rootlets and glycocalyceal bodies has been used as a distinct marker for certain varieties of adenocarcinoma and in metaplastic intestinal-type epithelium. 15-17 Whether or not glycocalyceal bodies occur in healthy small bowel has been disputed. 9·14·15 We found these
375
bodies in most of the biopsy specimens from cecal reservoir as well as from ileal nipple valve. They have also been observed in ileal conduits.' The functional implications of glycocalyceal bodies are not clear, since the exact manner of their formation is unknown. Nor have the functional properties of these structures been identified. Possibly their occurrence in ileal isolates functioning as conduit or reservoir outlet expresses adaptation to a physiologic situation comparable with that in colonic mucosa. Edema resulted in cell spacing in the luminal mucosa in some patients, and seemed to be associated with the inflammatory cell reaction. Concomitant bacteriuria may have been the cause, although such correlation was not found in a study of ileal conduits.' Most of our patients had positive urine cultures, but in no case did we find bacteria adhering to the cecal mucosal surface. Scanning electron microscopy of cecal reservoir biopsies gave similar results, 10 in analogy with a study that failed to demonstrate mucosa-adherent bacteria in ileal conduits. 18 The continuing production of immunoglobulins19 and mucins may be of importance in this context. Mucins are conjugated glycoproteins, and construction of cecal urinary reservoir usually is followed by a shift from sulfomucins to sialomucins, which we have interpreted as a secondary, nonspecific phenomenon.5 The two patients with increased amounts of collagen in the submucosa of the cecal reservoir had received preoperative radiotherapy, which has been suggested as a cause of fibrosis in ileal conduits. 1 Fibrotic stricture of ileal conduit can occur without preceding irradiation, however. Progressive fibrosis may be detrimental to the function of a reservoir as an expandable receptacle for urine. The mucosal fibrosis found in the ileal outlet of the reservoir in two cases may be explained as an ischemic effect caused by mesenteric exclusion and intermittent trauma from catheterization. Preformed anatomic structures such as neurogenic processes, enterochromaffin cells and Paneth cells were all clearly present in normal amounts and shapes in cecal as well as in ileal mucosa. Enterochromaffin cells can be found throughout the gastrointestinal tract, where they produce a number of hormones. In the colon they secrete serotonin, somatostatin and several other polypeptides with hormonal activity. 20 The Paneth cells were previously observed to remain unchanged in ileal conduits. 21 Although their exact functional role has not been established, Paneth cells are known to be capable of producing lysozyme, which can digest bacterial cell walls and phagocytize microorganisms. 22 Together with immunoglobulins, especially secretory IgA, 19 the products of Paneth cells may be important as host defense mechanisms against urinary tract infection in a cecal reservoir for urine. REFERENCES 1. Lindell, 0., Makinen, J., Nickles, J. and Lehtonen, T.: Mucosa!
2.
3.
4. 5. 6.
morphology of ilea! conduits with particular reference to urinary infections. Eur. Urol., 12: 201, 1986. Hansson, H.-A., Kock, N. G., Norlen, L., Philipson, B., Trasti, H. and Ahren, C.: Morphological observations in pedicled ilea! grafts used for construction of continent reservoirs for urine. Scand. J. Urol. Nephrol., Suppl., 49: 49, 1978. Philipson, B. M., Kock, N. G., Hiickenstriim, T., Norlen, L. J., Ahren, C. and Hansson, H.-A.: Ultrastructural and histochemical changes in ilea! reservoir mucosa after long-term exposure to urine. A study in patients with continent urostomy (Kock pouch). Scand. J. Gastroenterol., 21: 1235, 1986. Berg, N. 0., Fredlund, P., Mansson, W. and Olsson, S.-A.: Surveillance colonoscopy and biopsy in patients with ureterosigmoidostomy. Endoscopy, 19: 60, 1987. Mansson, W. and Willen, R.: Mucosa! morphology and histochemistry of the continent cecal reservoir for urine. J. Urol., 139: 1199, 1988. Mansson, W., Colleen, S. and Sundin, T.: Continent caecal reservoir in urinary diversion. Brit. J. Urol., 56: 359, 1984.
376
CARLEN, WILLEN AND MANSSON
7. Mansson, W. and Colleen, S.: The continent cecal reservoir for urine. Sem. Urol., 5: 63, 1987. 8. Reynolds, E. S.: The use of lead citrate at high pH as an electronopaque stain in electron microscopy. J. Cell. Biol., 17: 208, 1963. 9. Ghadially, F. N.: Diagnostic electron microscopy of tumours. 2nd ed. Butterworths, London, pp. 334, 1985. 10. Mansson, W.: The continent cecal urinary reservoir. In: Bladder Reconstruction and Continent Urinary Diversion. Ed. by L. R. King, A. R. Stone and G. D. Webster. Chicago, Yearbook Medical Publishers, chapt. 15, p. 209, 1987. 11. Philipson, B. M., Kock, N. G., Jagenburg, R., Ahren, C., Norlen, L., Robinson, J. W. L. and Menge, H.: Functional and structural studies of ilea! reservoirs used for continent urostomy and ileostomy. Gut, 24: 392, 1983. 12. Akerlund, S., Jagenburg, R., Kock, N. G. and Philipson, B. M.: Absorption ofL-phenylalanine in human ilea! reservoirs exposed to urine. Urol. Res., 16: 321, 1988. 13. Schnitka, T. K.: Current concepts of the pathogenesis and pathology of inflammatory lesions of the intestine. Can. Med. Ass. J., 91: 7, 1964. 14. Stone, J., Mukherjee, T. M. and Hesher, R.: C-bodies and R-bodies in the epithelial cells of normal and diseased human rectum. Arch. Path. Lab. Med., 101: 436, 1977.
15. Marcus, P. B.: Glycocalyceal bodies and their role in tumor typing. J. Submicroscopic Cytol., 13: 403, 1981. 16. Marcus, P. B., Martin, J. H., Green, R. H. and Krouse, M. A.: Glycocalyceal bodies and microvillous core rootlets. Arch. Path. Lab. Med., 103: 89, 1979. 17. Mukherejee, T. M.: The role of electron microscopy in the diagnosis of neoplastic cells in effusion fluids. J. Submicroscopic Cytol., 14: 717, 1982. 18. Bruce, A. W., Reid, G., Chan, R. C. Y. and Costerton, J. W.: Bacterial adherence in the human ilea! conduit: A morphological and bacteriological study. J. Urol., 132: 184, 1984. 19. Mansson, W., Colleen, S., Low, K., Mardh, P.-A. and Lundblad, A.: Immunoglobulins in urine from patients with ilea! and colonic conduits and reservoirs. J. Urol., 133: 713, 1985. 20. Grube, H. and Forsmann, G.: Morphology and function of the entero-endocrine cells. Horm. Metab. Res., 11: 603, 1979. 21. Garner, J. W., Goldstein, A. M. B. and Cosgrove, M. D.: Histologic appearance of the intestinal urinary conduit. J. Urol., 114: 854, 1979. 22. Erlandsen, S. L. and Chase, D. G.: Paneth cell function: phagocytosis and intracellular digestion of intestinal microorganisms. I. Hexamita Muris. J. Ultrastruct. Res., 41: 296, 1972.