- Email: [email protected]
The effect of fetal urine on the development of thebowel in gastroschisis
The E f f e c t o f F e t a l U r i n e on the D e v e l o p m e n t o f the B o w e l in G a s t r o s c h i s i s By P. Kl{Jck, D. Tibboel, A. W. M. van der Kamp, and J. C. Molenaar Rotterdam, The Netherlands 9 A gastroschisis model was successfully developed in the chicken embryo. The embryologic anatomy of this laboratory animal enabled the inducement of an abdominal wall defect, whereby the eviscerated abdominal contents were not exposed to fetal urine. A total of 999 embryos underwent a surgical intervention at an early developmental stage, from the 5 t h - 8 t h day of the incubation period. Twenty-four hour mortality was 7%. Surgery carried out on the 5th day resulted in the largest number of survivors, 25% of the induced lesions had healed and gastroschisis did not occur. The characteristic picture of gastroschisis only evolved when the herniated bowel was exposed to urine components. Histologic studies were carried out with the aid of various staining techniques to determine the development and distribution of the enteric ganglia in experimental gastroschisis. Contrary to simular studies that would point to damage of enteric ganglion cells as being responsible for the delay in intestinal motility, no ganglionic injury was noted in our bowel studies. INDEX WORDS: Gastroschisis; experimental fetal surgery; chick embryo.
EFERENCES to gastroschisis have appeared in the literature with increasing frequency from the early forties. Many aspects of this congenital anomaly have been discussed at length and various experimental studies have been reported. A great deal still remains to be investigated. One particular aspect that has been noted from the earliest reports is a delayed intestinal motility without signs of an organic obstruction. H Several pathogenetic factors have been proposed to account for this delay in the onset of peristalsis. Two of these are edema of the bowel wall due to profuse serosal granulation tissue and imbedment of the intestines in a dense gelatinous matrix, 2 and damage to the enteric ganglion cells of the exposed bowel. ~ Following the experimental studies of Hailer and coworkers, s no further investigation has been carried out concerning the pathophysiology of the bowel in gastroschisis. The purposes of this paper are: (A) To report the development of a successful experimental model of gastroschisis in a chick embryo, and (B)
R
Journal of Pediatric Surgery, Vol. 18, No. 1 (February),1983
to report the results of histologic studies of the bowel in this experimental model. The chick embryo was chosen as the laboratory animal. In previous experimental fetal surgery we have substantiated that this is a valuable laboratory animal, with the inherent advantage of low cost plus the fact that no anesthesia and only a limited surgical apparatus is required. 6 The chick embryo has an allantoic cavity connected with the embryonic cloaca via the allantoic stalk, totally separate from the amniotic cavity which contains the embryo immersed in amniotic fluid. Consequently, the amniotic fluid of the chick embryo does not contain any cloacal products. An induced evisceration of the abdominal contents should be feasible without involving either the allantoic stalk or cavity. This would ensure that the eviscerated bowel would be exposed to the amniotic fluid only, ruling out any effect of fetal urine on the development of the bowel. MATERIALS AND METHODS An abdominal wall defect was induced in chicken embryos (Gallus domesticus) on the 5th, 6th, 7th, and 8th day of incubation. Hatching normally occurs on the 21st day. The operating procedure was as follows (see Fig. 1): The air pocket is opened and the membrane lining the floor of the air pocket is removed. The embryo is now visible, covered by a number of membranes. In a 5-day-old embryo the allantoic cavity is too small for surgical intervention. It is easy to bypass this cavity and to open just the amniotic cavity. Following the large vessels that enter the embryo ventrally, a lesion with a diameter of approximately 2.5 mm is made in the ectoderm with tweezers, at the level of the ectodermic transition to the body stalk. No intraabdominal structures are involved. In a 6-day-old embryo the amniotic cavity can also
From the Erasmus University, School of Medicine, Rotterdam, the Netherlands. Presented before the XX1X Annual Congress of the British Association of Paediatric Surgeons, Madrid, Spain, July 21-23, 1982. Address reprint requests to Professor Dr. J.C. Molenaar, Department of Pediatric Surgery, Sophia Children's Hospital Gordelweg 160, 3038 GE Rotterdam, Netherlands. 9 1983 by Grune & Stratton, Inc. 0022-3468/83/18014)01 I$01.00/0
47
48
KLUCK ET AL.
Fig. 1, Schematic drawing of chicken embryo. 1 = embryo; 2 = amniotic cavity; 3 = allantoic cavity; 4 = air pocket.
be reached via the allantoic cavity. By the 7th and 8th days of incubation, however, entry to the amniotic cavity can only be achieved via the allantois. Between the 16th and 19th day of incubation the embryo is studied for any gross signs of an abdominal wall defect and a concomitant picture of gastroschisis. A microscopic study of the bowel and mesentery is carried out after fixation in formaldehyde. Paraffin sections are stained with Masson's trichrome. Bodian and Nissl staining is used for neurohistological investigation.
RESULTS
An assessment was made of 445 operations to determine the mortality within 24 hr of the intervention. This amounted to 7%. Table 1 shows that the number of embryos that are alive on the 16th-19th day of incubation is largest (59%) for the group that underwent an intervention on the 5th day. This number decreases to 27% in the group that was operated on the 8th day of incubation. In the embryos where the lesion had also involved the allantois, irregardless of intent, there was a high incidence of mortality (at least 25%) on the 10th-! lth day. During the experiments it was noted that the
embryos with an abdominal wall defect on the 16th-19th day, were smaller in size than embryos that exhibited normal development at the same stage of incubation. This observation was not quantified. At examination on the 16th-19th day, the results of the intervention could be classified in three categories. The lesion induced in the abdominal wall following the procedure described above, appeared to have healed in 13% of the total number of embryos involved. This was mostly the case in the embryos that underwent the intervention on the 5th day of incubation. In an approximately equal number of embryos (12% of the total number), an abdominal wall defect had occurred with herniation of intestines that had a normal appearance. Sometimes the bowel had a slightly edematous aspect, but the bowel loops could be differentiated easily. A concomitant cyst was frequently noted. This cyst contained allantoic fluid, consisting of cloacal products. This was probably the result of irreversible damage to the allantoic stalk. The third and smallest category (6% of total number) also presented with an abdominal wall defect with protruding intestines. In these embryos the eviscerated abdominal contents resembled the picture of gastroschisis as described by Bernstein in 1940: "firmly matting together clusters of bowel loops; in fact, the entire ectopia presents usually as a large solid mass, consisting of densely adherent abdominal viscera with leathery consistency. ''7 (see Fig. 2)
In these cases the skin and the feathers of the chick were frequently covered with cloacal products (urine sediment). The assessment in Table 1 clearly demonstrates that the gastroschisis type of anomaly only occurred when the intervention had involved opening the allantois.
Table 1. Results of Experimental Surgery Operation/ Day of Incubation A 5 A 6 9 6 B 7 B 8 Total
No. of Embryos 94 110 43 125 627 999
Examination/ Day of Incubation
No. of Live Embryos
16-19 16-19 16-19 16-19 18-19
56 (59%) 43 (39%) 14(32%) 34 (26%) 169 (27%) 316 (32%)
Abdominal Wall Defect Only
Gastroschisis
32 33 10 t7 30 122
0 0 3 7 54 64
(34%) (30%) (23%) (13%) (5%) (12%)
(0%) (0%) (7%) (5%) (8.5%) (6%J
Lesion Healed 24 (25%) 10 (9%) 1 (2%) 10 (8%) 85 (13.5%) 130 (13%)
Assessment of number of embryos operated on the 5th, 6th, 7th, and Bth day of incubation with gross pathologic findings towards the end of the incubation period. A = allantoic cavity not involved; B = allantoic cavity opened at operation.
BOWEL STUDIES IN GASTROSCHISIS
49
Gastroschisis in a chick embryo. Normally develFig. 2. oped umbilical cord is held by forceps.
In the chicks that presented with just a defect of the abdominal wall through which normal looking bowel protruded, histological studies of the intestines sometimes revealed only a slight subserous edema. The villi showed a normal structure, the development of the muscular layers was unaffected, and both autonomic plexuses were present. In the "'gastroschisis type of defect," the intestines were generally packed in a cluster, matted together and covered with fibrinous tissue which seemed to stem from the serosa (see Fig. 3). The mesentery could not be differentiated.
Fig. 4. Cross section of fibrinous tissue covering intestines. G T = granulation tissue; L M = lamina muscularis.
The whole picture closely resembled the classical description of the histological findings in human neonates with gastroschisis. 2 (see Fig. 4) The ganglion of Remak, which in birds constitutes the sympathetic chain in the mesentery, as well as the enteric plexus, showed a normal morphologic development. This was established by visualizing the perikarya by means of Nissl staining. The neurofibrils, which also appeared normal, were revealed with the aid of the Bodian staining technique. DISCUSSION
Fig. 3. Section of matted bowel loop covered by f i brinous tissue.
Studies of human embryos have revealed that the pathogenesis of gastroschisis must hail from the initial developmental phase of the human embryo (5th-8th week of gestation).8'9 Consequently, a rigorous animal model to study this anomaly, can only be achieved by some kind of surgical intervention at an early developmental stage of the embryo. This is technically difficult to realize in mammals. That is probably the reason why the experimental studies of gastroschisis described in the literature, were all carried out in the second half of gestation of the animal under study, involving either rabbits or lambs. 5'~~ In contrast, in the chicken embryo it is technically feasible to carry out an intervention
50
KLUCK ET AL.
at an early developmental stage, with standardized reproducibility. 12 As an experimental model for gastroschisis, the chick embryo has an added advantage. In humans, the allantois never develops into a reservoir for the secretion of waste products from the definite urinary system as it does in the chick embryo. In h u m a n s the urine is secreted right into the amniotic cavity and absorbed by the amniotic fluid in which the fetus is immersed. In case of gastroschisis, the eviscerated abdominal contents have therefore been exposed for a considerable length of time to urine components in the amniotic fluid. In the chick embryo, however, it is possible to eliminate the effect of fetal urine on the development of the bowel, by avoiding opening the allantois when making a lesion in the abdominal wall of the experimental model. W e only achieved this in the earliest developmental stage, on the 5th and on the 6th day of incubation. O n the 7th and 8th day, and selectively on the 6th day, the allantoic cavity was always involved in the surgical intervention. The allantoic lesion would sometimes heal spontaneously. In those cases the typical gastroschisis picture
did not emerge, and the gross as well as the microscopic aspect of the bowel remained normal. This demonstrates the important part played by urine components contained in the amniotic fluid in the occurrence of gastroschisis. The high incidence of mortality on the 1 0 t h 1 lth day of incubation exhibited in our experiments when the allantoic cavity had been opened, m a y also be due to the harmful effect of urine on the embryo with an abdominal wall defect. According to Romanoff, kidney function comes into effect around that developmental stage in the avian e m b r y o ) 3 C o n t r a r y to Haller's findings, 5 our histological studies in this experimental model, did not reveal any abnormality of the neuronal plexuses. Extensive histochemical studies will have to be carried out to determine the differentiation of the neuronal components. This would throw light on the delay in intestinal motility. ACKNOWLEDGMENT
We wish to thank Alice Goslinga Ribbink, Translator/ Stylistic Editor, for her assistance in preparing this manuscript.
REFERENCES
1. Watkins DE: Gastroschisis. Virginia M Monthly 70:42-44, 1943 2. Moore TC, Stokes GE: Gastroschisis. Surgery 33:112120, 1953 3. Rubin SZ, Martin DJ, Ein SH: A critical look at delayed intestinal motility in gastroschisis. Can J Surg 21:414-416, 1978 4. Touloukian R J, Spackman TJ: Gastrointestinal function and radiographic appearance following gastroschisis repair. J Pediatr Surg 6:427-433, 1971 5. Hailer JA, Kehrer BH, Shaker I J, Shermata DW, Wyllie RG: Studies of the pathophysiologyof gastroschisis in fetal sheep. J Pediatr Surg 9:627~32, 1974 6. Tibboel D, Molenaar JC, van Nie CJ: New perspectives in fetal surgery: the chicken embryo. J Pediatr Surg 14:438439, 1979
7. Bernstein P: Gastroschisis, a rare teratological condition in the newborn. Arch Pediatr 57:505-513, 1940 8. De Vries PA: The pathogenesis of gastroschisis and omphalocele. J Pediatr Surg 15:245-251, 1980 9. Miintener M: Zur Genese der Omphalocele und Gastroschisis. Z Kinderch 8:380-390, 1970 10. Sherman N J, Asch M J, Isaacs H, et al: Experimental gastroschisis in the fetal rabbit. J Pediatr Surg 8:165-169, 1973 11. Aoki Y, Ohsio T, Komi N: An experimental study on gastroschisis using fetal surgery. J Pediatr Surg 15:252-256, 1980 12. Tibboel D, van der Kamp AWM, Molenaar JC: The effect of experimentally induced intestinal perforation at an early developmental stage. J Pediatr Surg 16:1017-1019, 1981 13. Romanoff AL: The Avian Embryo. New York, Macmillan, 1960
EFERENCES to gastroschisis have appeared in the literature with increasing frequency from the early forties. Many aspects of this congenital anomaly have been discussed at length and various experimental studies have been reported. A great deal still remains to be investigated. One particular aspect that has been noted from the earliest reports is a delayed intestinal motility without signs of an organic obstruction. H Several pathogenetic factors have been proposed to account for this delay in the onset of peristalsis. Two of these are edema of the bowel wall due to profuse serosal granulation tissue and imbedment of the intestines in a dense gelatinous matrix, 2 and damage to the enteric ganglion cells of the exposed bowel. ~ Following the experimental studies of Hailer and coworkers, s no further investigation has been carried out concerning the pathophysiology of the bowel in gastroschisis. The purposes of this paper are: (A) To report the development of a successful experimental model of gastroschisis in a chick embryo, and (B)
R
Journal of Pediatric Surgery, Vol. 18, No. 1 (February),1983
to report the results of histologic studies of the bowel in this experimental model. The chick embryo was chosen as the laboratory animal. In previous experimental fetal surgery we have substantiated that this is a valuable laboratory animal, with the inherent advantage of low cost plus the fact that no anesthesia and only a limited surgical apparatus is required. 6 The chick embryo has an allantoic cavity connected with the embryonic cloaca via the allantoic stalk, totally separate from the amniotic cavity which contains the embryo immersed in amniotic fluid. Consequently, the amniotic fluid of the chick embryo does not contain any cloacal products. An induced evisceration of the abdominal contents should be feasible without involving either the allantoic stalk or cavity. This would ensure that the eviscerated bowel would be exposed to the amniotic fluid only, ruling out any effect of fetal urine on the development of the bowel. MATERIALS AND METHODS An abdominal wall defect was induced in chicken embryos (Gallus domesticus) on the 5th, 6th, 7th, and 8th day of incubation. Hatching normally occurs on the 21st day. The operating procedure was as follows (see Fig. 1): The air pocket is opened and the membrane lining the floor of the air pocket is removed. The embryo is now visible, covered by a number of membranes. In a 5-day-old embryo the allantoic cavity is too small for surgical intervention. It is easy to bypass this cavity and to open just the amniotic cavity. Following the large vessels that enter the embryo ventrally, a lesion with a diameter of approximately 2.5 mm is made in the ectoderm with tweezers, at the level of the ectodermic transition to the body stalk. No intraabdominal structures are involved. In a 6-day-old embryo the amniotic cavity can also
From the Erasmus University, School of Medicine, Rotterdam, the Netherlands. Presented before the XX1X Annual Congress of the British Association of Paediatric Surgeons, Madrid, Spain, July 21-23, 1982. Address reprint requests to Professor Dr. J.C. Molenaar, Department of Pediatric Surgery, Sophia Children's Hospital Gordelweg 160, 3038 GE Rotterdam, Netherlands. 9 1983 by Grune & Stratton, Inc. 0022-3468/83/18014)01 I$01.00/0
47
48
KLUCK ET AL.
Fig. 1, Schematic drawing of chicken embryo. 1 = embryo; 2 = amniotic cavity; 3 = allantoic cavity; 4 = air pocket.
be reached via the allantoic cavity. By the 7th and 8th days of incubation, however, entry to the amniotic cavity can only be achieved via the allantois. Between the 16th and 19th day of incubation the embryo is studied for any gross signs of an abdominal wall defect and a concomitant picture of gastroschisis. A microscopic study of the bowel and mesentery is carried out after fixation in formaldehyde. Paraffin sections are stained with Masson's trichrome. Bodian and Nissl staining is used for neurohistological investigation.
RESULTS
An assessment was made of 445 operations to determine the mortality within 24 hr of the intervention. This amounted to 7%. Table 1 shows that the number of embryos that are alive on the 16th-19th day of incubation is largest (59%) for the group that underwent an intervention on the 5th day. This number decreases to 27% in the group that was operated on the 8th day of incubation. In the embryos where the lesion had also involved the allantois, irregardless of intent, there was a high incidence of mortality (at least 25%) on the 10th-! lth day. During the experiments it was noted that the
embryos with an abdominal wall defect on the 16th-19th day, were smaller in size than embryos that exhibited normal development at the same stage of incubation. This observation was not quantified. At examination on the 16th-19th day, the results of the intervention could be classified in three categories. The lesion induced in the abdominal wall following the procedure described above, appeared to have healed in 13% of the total number of embryos involved. This was mostly the case in the embryos that underwent the intervention on the 5th day of incubation. In an approximately equal number of embryos (12% of the total number), an abdominal wall defect had occurred with herniation of intestines that had a normal appearance. Sometimes the bowel had a slightly edematous aspect, but the bowel loops could be differentiated easily. A concomitant cyst was frequently noted. This cyst contained allantoic fluid, consisting of cloacal products. This was probably the result of irreversible damage to the allantoic stalk. The third and smallest category (6% of total number) also presented with an abdominal wall defect with protruding intestines. In these embryos the eviscerated abdominal contents resembled the picture of gastroschisis as described by Bernstein in 1940: "firmly matting together clusters of bowel loops; in fact, the entire ectopia presents usually as a large solid mass, consisting of densely adherent abdominal viscera with leathery consistency. ''7 (see Fig. 2)
In these cases the skin and the feathers of the chick were frequently covered with cloacal products (urine sediment). The assessment in Table 1 clearly demonstrates that the gastroschisis type of anomaly only occurred when the intervention had involved opening the allantois.
Table 1. Results of Experimental Surgery Operation/ Day of Incubation A 5 A 6 9 6 B 7 B 8 Total
No. of Embryos 94 110 43 125 627 999
Examination/ Day of Incubation
No. of Live Embryos
16-19 16-19 16-19 16-19 18-19
56 (59%) 43 (39%) 14(32%) 34 (26%) 169 (27%) 316 (32%)
Abdominal Wall Defect Only
Gastroschisis
32 33 10 t7 30 122
0 0 3 7 54 64
(34%) (30%) (23%) (13%) (5%) (12%)
(0%) (0%) (7%) (5%) (8.5%) (6%J
Lesion Healed 24 (25%) 10 (9%) 1 (2%) 10 (8%) 85 (13.5%) 130 (13%)
Assessment of number of embryos operated on the 5th, 6th, 7th, and Bth day of incubation with gross pathologic findings towards the end of the incubation period. A = allantoic cavity not involved; B = allantoic cavity opened at operation.
BOWEL STUDIES IN GASTROSCHISIS
49
Gastroschisis in a chick embryo. Normally develFig. 2. oped umbilical cord is held by forceps.
In the chicks that presented with just a defect of the abdominal wall through which normal looking bowel protruded, histological studies of the intestines sometimes revealed only a slight subserous edema. The villi showed a normal structure, the development of the muscular layers was unaffected, and both autonomic plexuses were present. In the "'gastroschisis type of defect," the intestines were generally packed in a cluster, matted together and covered with fibrinous tissue which seemed to stem from the serosa (see Fig. 3). The mesentery could not be differentiated.
Fig. 4. Cross section of fibrinous tissue covering intestines. G T = granulation tissue; L M = lamina muscularis.
The whole picture closely resembled the classical description of the histological findings in human neonates with gastroschisis. 2 (see Fig. 4) The ganglion of Remak, which in birds constitutes the sympathetic chain in the mesentery, as well as the enteric plexus, showed a normal morphologic development. This was established by visualizing the perikarya by means of Nissl staining. The neurofibrils, which also appeared normal, were revealed with the aid of the Bodian staining technique. DISCUSSION
Fig. 3. Section of matted bowel loop covered by f i brinous tissue.
Studies of human embryos have revealed that the pathogenesis of gastroschisis must hail from the initial developmental phase of the human embryo (5th-8th week of gestation).8'9 Consequently, a rigorous animal model to study this anomaly, can only be achieved by some kind of surgical intervention at an early developmental stage of the embryo. This is technically difficult to realize in mammals. That is probably the reason why the experimental studies of gastroschisis described in the literature, were all carried out in the second half of gestation of the animal under study, involving either rabbits or lambs. 5'~~ In contrast, in the chicken embryo it is technically feasible to carry out an intervention
50
KLUCK ET AL.
at an early developmental stage, with standardized reproducibility. 12 As an experimental model for gastroschisis, the chick embryo has an added advantage. In humans, the allantois never develops into a reservoir for the secretion of waste products from the definite urinary system as it does in the chick embryo. In h u m a n s the urine is secreted right into the amniotic cavity and absorbed by the amniotic fluid in which the fetus is immersed. In case of gastroschisis, the eviscerated abdominal contents have therefore been exposed for a considerable length of time to urine components in the amniotic fluid. In the chick embryo, however, it is possible to eliminate the effect of fetal urine on the development of the bowel, by avoiding opening the allantois when making a lesion in the abdominal wall of the experimental model. W e only achieved this in the earliest developmental stage, on the 5th and on the 6th day of incubation. O n the 7th and 8th day, and selectively on the 6th day, the allantoic cavity was always involved in the surgical intervention. The allantoic lesion would sometimes heal spontaneously. In those cases the typical gastroschisis picture
did not emerge, and the gross as well as the microscopic aspect of the bowel remained normal. This demonstrates the important part played by urine components contained in the amniotic fluid in the occurrence of gastroschisis. The high incidence of mortality on the 1 0 t h 1 lth day of incubation exhibited in our experiments when the allantoic cavity had been opened, m a y also be due to the harmful effect of urine on the embryo with an abdominal wall defect. According to Romanoff, kidney function comes into effect around that developmental stage in the avian e m b r y o ) 3 C o n t r a r y to Haller's findings, 5 our histological studies in this experimental model, did not reveal any abnormality of the neuronal plexuses. Extensive histochemical studies will have to be carried out to determine the differentiation of the neuronal components. This would throw light on the delay in intestinal motility. ACKNOWLEDGMENT
We wish to thank Alice Goslinga Ribbink, Translator/ Stylistic Editor, for her assistance in preparing this manuscript.
REFERENCES
1. Watkins DE: Gastroschisis. Virginia M Monthly 70:42-44, 1943 2. Moore TC, Stokes GE: Gastroschisis. Surgery 33:112120, 1953 3. Rubin SZ, Martin DJ, Ein SH: A critical look at delayed intestinal motility in gastroschisis. Can J Surg 21:414-416, 1978 4. Touloukian R J, Spackman TJ: Gastrointestinal function and radiographic appearance following gastroschisis repair. J Pediatr Surg 6:427-433, 1971 5. Hailer JA, Kehrer BH, Shaker I J, Shermata DW, Wyllie RG: Studies of the pathophysiologyof gastroschisis in fetal sheep. J Pediatr Surg 9:627~32, 1974 6. Tibboel D, Molenaar JC, van Nie CJ: New perspectives in fetal surgery: the chicken embryo. J Pediatr Surg 14:438439, 1979
7. Bernstein P: Gastroschisis, a rare teratological condition in the newborn. Arch Pediatr 57:505-513, 1940 8. De Vries PA: The pathogenesis of gastroschisis and omphalocele. J Pediatr Surg 15:245-251, 1980 9. Miintener M: Zur Genese der Omphalocele und Gastroschisis. Z Kinderch 8:380-390, 1970 10. Sherman N J, Asch M J, Isaacs H, et al: Experimental gastroschisis in the fetal rabbit. J Pediatr Surg 8:165-169, 1973 11. Aoki Y, Ohsio T, Komi N: An experimental study on gastroschisis using fetal surgery. J Pediatr Surg 15:252-256, 1980 12. Tibboel D, van der Kamp AWM, Molenaar JC: The effect of experimentally induced intestinal perforation at an early developmental stage. J Pediatr Surg 16:1017-1019, 1981 13. Romanoff AL: The Avian Embryo. New York, Macmillan, 1960