- Email: [email protected]
Erythropoietin restores bowel damage and hypoperistalsis in gastroschisis
Journal of Pediatric Surgery (2006) 41, 352 – 357
www.elsevier.com/locate/jpedsurg
Erythropoietin restores bowel damage and hypoperistalsis in gastroschisis Aykut Ozdamar a, Koray Topcu a, Mukaddes Gumustekinb, Duygu Gurelc, Ayse Gelalb, Erdener Ozerc, Basak Ucana, Gunyuz Temir a, Aytac Karkiner a, Irfan Karaca a, Munevver Hosgora,* a
Department of Pediatric Surgery, Dr. Behcet Uz Children’s Hospital, Izmir, Turkey Department of Pharmacology, Dokuz Eylul University, Izmir, Turkey c Department of Pathology, Dokuz Eylul University, Izmir, Turkey b
Index words: Gastroschisis; Erythropoietin; Gastrointestinal contractility
Abstract Background and Purpose: Despite the decreased mortality in gastroschisis (Gx), patients experience postoperative intestinal hypoperistalsis, malabsorption, and shortened bowel length. The trophic effects of recombinant human erythropoietin (rEpo) in the developing small bowel have been reported, increasing the length and height of the villi, and villous surface area. This study investigated the effects of rEpo on intestinal malfunction in the chick embryos with Gx. Methods: Thirteen-day-old fertilized chicken eggs were used to create Gx model. Study groups included the following: group 1, control; group 2, Gx-only; group 3, Gx + 0.075% saline exchange; group 4, Gx + 10 IU rEpo exchange; group 5, Gx + 20 IU rEpo exchange. The bowels were evaluated by in vitro muscle strip technique, and the response was expressed as a percentage of the maximum carbachol-evoked contraction (E max). In addition, parasympathetic ganglion cells per 10 plexuses and villi height were determined by light microscopy. Results were evaluated statistically by Mann-Whitney U, v 2, and Fisher’s Exact test tests. Results: Saline exchange had no effect on ganglion cell number ( P = .63) and villi height ( P = .10). In group 4, ganglion cell number was not increased ( P = .82), but villi height increase was significant ( P = .03). In Gx + 20 IU rEpo group, both the number of ganglia ( P = .0001) and villi height ( P = .002) were significantly increased. The decrease in contractility in group 2 ( P = .0121) was significantly reversed by rEpo 20 IU treatment ( P = .0216), no significant difference was obtained in groups 3 ( P = .0809) and 4 ( P = .1516) compared with group 2. Conclusion: These data suggest that rEpo has prokinetic effects on hypoperistalsis and restores bowel damage in Gx. D 2006 Published by Elsevier Inc.
Presented at the 52nd Annual Congress of British Association of Paediatric Surgeons, Dublin, Ireland, July 12-15, 2005. 4 Corresponding author. Koruturk Mh Simsek Sk. Yagmur Sitesi K: 1D:1, 35330 Izmir, Turkey. Tel.: +90 232 4895656/5303; fax: +90 232 4892315. E-mail address: [email protected] (M. Hosgor). 0022-3468/$ – see front matter D 2006 Published by Elsevier Inc. doi:10.1016/j.jpedsurg.2005.11.012
Gastroschisis (Gx) is a congenital full-thickness cleft in the anterior abdominal wall, usually to the right side of the umbilical insertion, resulting in protrusion of abdominal contents through the cleft [1]. The characteristic picture of Gx includes congestion, edema, fibrous coating, and
Erythropoietin restores bowel damage and hypoperistalsis in gastroschisis adhesive matting of the protruding bowel loops, which occurs presumably from direct exposure to the amniotic fluid and from partial lymphatic and venous obstruction in the mesentery [2]. Severe intestinal damage is also evidenced by intestinal dilatation, smooth muscle thickening, mesenteric shortening, villous atrophy, great loss of contractility, and decreased mucosal absorption [3]. The incidence of Gx is increasing throughout the world, and it occurs in about 1 of every 4000 live births [4]. Although neonatal intensive care and surgical techniques have significantly improved, a number of patients experience prolonged intestinal hypoperistalsis and malabsorption after surgical repair and may predispose to necrotizing enterocolitis [5]. Erythropoietin (Epo) is a growth factor produced by the kidney in response to anemia [6]. In previous studies, the presence of Epo in human milk and functional Epo receptors in fetal and postnatal small intestines has been proposed because Epo has a role in growth and development of the gastrointestinal tract [7]. It has been shown that Epo binds to enterocytes and stimulates small intestinal growth by increasing length, villous surface area, villi height, and ileal crypt depth [8]. Moreover, it was also reported that Epo protects neurons against ischemia-induced cell death and acts as an angiogenic growth factor for intestinal mesentery microvascular endothelial cells [9,10]. The present study was undertaken using a chick embryo model to determine if intraamniotic recombinant human erythropoietin (rEpo) administration would prevent intestinal damage and enhance bowel contractility in Gx.
1. Materials and methods 1.1. Experimental procedures All experimental procedures were approved by Ethical Committee of the Dr. Behcet Uz Children’s Hospital.
353
Fig. 2 Macroscopic view of Gx after AAF exchange with 0.075% physiological saline.
Thirteen-day-old fertilized chick eggs (Gallus domesticus) were incubated at 37.58C in 80% humidity. Operative procedures for all groups were performed on the 13th day of incubation, using previously published methods [11]. The eggs were divided into 5 groups. In the first group (control) (n = 10), through an eggshell window, the amnioallantoic membrane was opened to create a common cavity, and thus amnioallantoic fluid (AAF) mixture was created, which resembled human amniotic fluid. In the second group (Gx-only; n = 13), Gx was created. During creation of the Gx model, after opening amnioallantoic membrane, a 2.5mm defect was created with delicate tweezers in the umbilical stalk near the abdominal wall, and intestinal loops were exteriorized out of the abdomen. In groups 3 (n = 12), 4 (n = 11), and 5 (n = 15), after creating Gx as described, a catheter was placed into the amnioallantoic cavity at the end of the procedure. AAF exchange was performed with only 0.075% physiological saline in group 3, with 10 IU rEpo (Eprex, Cilag, Zug, Switzerland) solution diluted with 0.075% saline in group 4, and with 20 IU rEpo solution diluted with 0.075% saline in group 5. Exchange in these groups was performed once a day with a volume corresponding to 10% of AAF, 1 mL at the 15th and 16th days and 0.5 mL at the 17th day before hatching as previously described. On the 18th day of hatching, all embryos were killed.
1.2. Light microscopy
Fig. 1 Gx-only group: macroscopic changes including thickening, fibrous peel on the serosal surfaces and fibrous adhesions between bowel loops.
Tissue specimens were taken from intestine of each chick embryos and fixed in 10% formalin. Three sections from a 1-cm portion of the intestinal segments of each embryo were selected randomly. Intestines were examined macroscopically and microscopically. All microscopic examinations were carried out by the same pathologist (EO) under light microscopy in a blinded manner. The numbers of parasympathetic ganglia were counted in 10 plexuses of each section (total, 30 plexuses). The villi height per 10 villi from each
354
A. Ozdamar et al.
Fig. 3 Macroscopic view of Gx after AAF exchange with 10 IU rEpo.
section were measured using a computer assisted image analyzer system consisting of a microscope (Labophot-2, Nikon, Tokyo, Japan) equipped with a high-resolution video camera (VKC220E; Hitachi, Tokyo, Japan). The images were processed using an IBM-compatible personal computer, high-resolution video monitor (Sony, Tokyo, Japan), and image analysis software (BS 200Docu Version 2.0, BAB Imaging Systems, Ankara, Turkey). The number of ganglia was reported as mean value per 10 plexuses. Results were evaluated statistically by v 2 and Fisher’s Exact test tests.
1.3. Contractility studies Contractility studies were performed using in vitro muscle strip technique during the first 2 hours after removal. Similar bowel segments selected for ganglion cell count were also selected randomly for contractility studies. Mesentery tissue was carefully removed, and the bowel was dissected into strips 10 mm. Full-thickness muscle strips were subsequently mounted in a 20-mL jacketed organ bath filled with Tyrode solution composed of (in mmol/L) NaCl,
Fig. 4
Macroscopic view of Gx + 20 IU rEpo exchange.
Fig. 5 Histological view of Gx-only group. Note decreased villi height and ganglion cell number (arrows) and the fibrous peel (fp) (hematoxylin-eosin, original magnification 100).
136.8; KCl, 2.7; MgCl2, 1.1; CaCl2, 1.8; NaH2PO4, 0.4; NaHCO3, 11.9; and glucose, 5.6 (pH 7.4). The Tyrode solution was maintained at 378C and bubbled with 95% O2 + 5% CO2. A resting tension of 0.5 g was applied, and the strips were allowed to equilibrate for 60 minutes. The preparations were washed approximately every 15 minutes during equilibration period. Isometric tension changes were measured using a Grass Instrument 7FT 03E isometric transducers connected a computer-based data acquisition and analysis software system (MAY Polwin 97, Acquisition Software, Ankara, Turkey). Carbamylcholine chloride (carbachol [Cch] 10ÿ9 to 10ÿ4 mol/L) (Sigma, St. Louis, Mo) was applied directly to the organ baths, and contractions were recorded isometrically. Cumulative concentrationresponse curve was constructed for each sample in this study. Each dose of the cumulative curve was added when the effect of preceding one had reached its maximum.
Fig. 6 Histological view of Gx + 20 IU rEpo Exchange. Note that villi architecture is restored and number of ganglion cells is increased (arrow) (hematoxylin-eosin, original magnification 100).
Erythropoietin restores bowel damage and hypoperistalsis in gastroschisis Table 1 groups
The number of ganglia and villi height in study
Study groups
Number of ganglia
Villi height (lm)
Gx-only Gx + % 0.075 saline Gx + 10 IU rEpo Gx + 20 IU rEpo Control
68.84 69.02 73.60 90.00 77.80
52.83 67.62 69.67 75.29 70.05
F F F F F
33.98 28.24 22.82 31.13 9.01
F F F F F
14.24 19.09 24.23 16.49 10.10
Each strip was used in 1 experiment only. The same procedure was done in all groups. The response was expressed as a percentage of the maximum Cch-evoked contraction (E max) in the control group. Isometric tension measurements were expressed as a mean value F SD. Maximal response to Cch in the Gx-only group was compared with the other groups using MannWhitney U test. P b .05 was considered significant, and n denotes the number of strips.
2. Results
Table 3 groups
355
Maximum Cch-evoked contraction (E max) in study
Study groups
E max
Gx-only Gx + 0.075% saline Gx + 10 IU rEpo Gx + 20 IU rEpo Control
75.81 98.92 100.8 108.01 100
F F F F F
39.05 38.17 54.9 49.284 27.344
Values are mean F SD. The maximal responses are expressed as percentage of the E max obtained for control group. n denotes the number of strips. 4 P = .0216 Gx + 20 IU rEpo vs Gx-only group. 44 P = .0121 Control group vs Gx-only group.
compared with Gx-only group ( P = .63 and P = .82, respectively). Saline exchange also had no effect on villi height ( P = .10). Although the number of ganglia has not increased significantly after exchange with rEpo 10 IU, villi height were significantly increased compared with Gx-only group ( P = .03). In Gx + 20 IU rEpo group, both the number of ganglia ( P = .0001) and villi height ( P = .002) were significantly increased (Table 2).
2.3. Contractility findings
2.1. Macroscopic findings In Gx-only group, macroscopic changes included thickening, fibrous peel on the serosal surfaces and fibrous adhesions between bowel loops (Fig. 1). In all exchange groups, there was no fibrous peel formation or intestinal wall thickening (Figs. 2 -4).
2.2. Microscopic findings In Gx-only group, thickened serosal surface, fibrin deposition in the serosa, an inflammatory reaction, and edema were evident during microscopic evaluation (Fig. 5). These findings were not observed in any exchange groups (Fig. 6). In all groups, ganglia were morphologically normal. However, the number of ganglion cells and villi height were significantly decreased in Gx-only group compared with control group ( P = .0001 and P = .001, respectively) (Table 1). The mean number of parasympathetic ganglia per 10 plexuses were increased in groups 3 and 4 (69.02 F 28.24 and 73.60 F 22.82, respectively); however, these results were not statistically significant when
In randomized order, 1 to 5 full-thickness muscle strips were used from each embryo (group 1, n = 30; group 2, n = 29; group 3, n = 15; group 4, n = 20, group 5, n = 25). Cch (10ÿ9 to 10ÿ4 mol/L) caused contraction in a concentration-dependent manner to a maximum effect at 10ÿ4 mol/L (E max) in all groups. The E max value was decreased significantly in Gx-only group compared with control group ( P = .0121). Exchange with rEpo 20 IU has significantly improved contractility in fetal bowel strips with Gx ( P = .0216). Although exchange with 0.075% saline and with rEpo 10 IU have increased the E max values, the differences were not statistically significant ( P = .0809 and P = .1516, respectively) (Table 3). The results are summarized in Fig. 7.
Table 2 Statistical results of comparison of the number of ganglia and villi height Study groups
The number Villi height of ganglia ( P) ( P)
Gx-only Gx-only Gx-only Gx-only
.63 .82 .0001 .0001
vs vs vs vs
Gx + 0.075% Saline Gx + 10 IU rEpo Gx + 20 IU rEpo control
.10 .03 .002 .001
Fig. 7 Maximal contractile response to 10ÿ4 mol/L Cch. The maximal responses are expressed as percentage of the E max obtained for control group. Values are expressed mean F SD (n = 15-30). *P b .05 Gx-only vs Gx + 20 IU rEpo E max; **P b .05 Gxonly vs control E max.
356
3. Discussion Despite prenatal diagnosis of Gx and immediate appropriate neonatal care, significant postoperative morbidity still occurs as a consequence of bowel damage in utero [12]. Amniotic fluid exchange or amnioinfusion have been reported to decrease bowel damage in experimental animal studies as well as in human Gx [13,14]. The majority of these studies have primarily focused on prevention of the intestinal morphological damage. However, there are limited and sometimes conflicting literature information on pathogenesis and treatment of the intestinal malfunction seen in infants with Gx [2,15]. The present study demonstrated that intraamniotic rEpo administration restores, in a dosedependent manner, not only morphological damage but also intestinal dysfunction associated with Gx. In the current study, exposure of the intestines to AAF in Gx-only group has also resulted in bowel wall thickening, fibrous peel on the serosal surfaces, and fibrous adhesions between bowel loops as described in previous reports [16]. After exchange with either only physiological saline or rEpo + saline, normal appearing intestines that was not thickened with no visible fibrous peel or adhesion were observed, these findings supported that AAF exchange decreases the concentration of harmful chemicals responsible for intestinal morphological damage, mainly serosal peel formation and bowel wall thickening as previously reported in other clinical and experimental studies [17]. Using a fetal lamb model of Gx, Haller et al [18] have proposed that the disordered peristalsis in the eviscerated bowel results from atrophy of the ganglion cells; however, all of the studies later performed on different animal models did not confirm this finding [19]. In addition, in our study, ganglia were morphologically normal in all groups. In a recent report, the number of parasympathetic ganglia was found to be diminished in Gx, and this decrease was suggested as a secondary phenomenon caused by serosal inflammation or apoptosis [20]. Moreover, these investigators have reported that the decrease in the number of ganglia can be prevented by AAF exchange with physiological saline. We have also observed a significant decrease in the number of ganglia in Gx-only group, but AAF exchange with physiological saline has not increased the number of ganglion cells. In contrast, a dose-dependent increase in the number of ganglia was found after AAF exchange with rEpo. It is well known that rEpo decreases apoptotic death of neuronal cells and enterocytes [21]. Besides its neuroprotective and neurotrophic effects, it was also reported that rEpo acts in a manner consistent with known anti-inflammatory agents such as glucocorticoids [9]. Therefore, it can be speculated that by exerting its neuroprotective, anti-inflammatory, and antiapoptotic effects, rEpo have prevented the decrease in the number of ganglion cells. The intestinal dysfunction in Gx is believed to result from both loss of contractility and poor absorptive capacity because of mucosal villous atrophy and immaturity of the
A. Ozdamar et al. mucosal enzymatic systems [22]. The etiology of these changes in Gx has been attributed to a variety of causes, including both amniotic fluid exposure and intrauterine ischemia caused by gradual constriction of the bowel at the abdominal wall defect [23]. In the present study, bowel contractility and villi height were significantly lower in Gx-only group than normal. Physiological saline exchange had no significant effect on either bowel contractility or villi height. In contrast, AAF exchange with rEpo significantly improved contractility in a dose-dependent manner; in addition, histological examination of the bowel mucosa showed longer villi in both rEpo-treated groups. In previous rat studies, it has been shown that rEpo significantly increases intestinal length and the absorptive surface of the microvilli, primarily by increasing the villi length [8]. Moreover, Epo is an angiogenic growth factor that stimulates intestinal mesentery microvascular endothelial cell proliferation and increases intestinal blood flow [10]. In this respect, our results suggested that intraamniotic rEpo administration in Gx acts not only as a trophic factor but also as a prokinetic agent on a variety of nonhematopoietic cell types, including enterocytes, endothelial cells, smooth muscle cells, and neuronal cells; cell types that are present in the developing bowel [24]. In conclusion, the data presented here constitute a basis for examining the effectiveness and appropriate dose of rEpo in further studies for treatment of all aspects of intestinal malfunction in human Gx.
References [1] Oyachi N, Laksmanan J, Ross MG, et al. Fetal gastrointestinal motility in a rabbit model of gastroschisis. J Pediatr Surg 2004; 39:366 - 70. [2] Guo W, Swaniker F, Fonkalsrud EW, et al. Effect of intraamniotic dexamethasone administration on intestinal absorption in a rabbit gastroschisis model. J Pediatr Surg 1995;30:983 - 7. [3] Langer JC, Longaker MT, Crombleholme TM, et al. Etiology of intestinal damage in gastroschisis I: effects of amniotic fluid exposure and bowel constriction in a fetal lamb model. J Pediatr Surg 1989; 24:992 - 7. [4] Baerg J, Kaban G, Tonita J, et al. Gastroschisis: a sixteen year review. J Pediatr Surg 2003;38:771 - 4. [5] Jayanthi S, Seymour P, Puntis JWL, et al. Necrotizing enterocolitis after gastroschisis repair: a preventable complication? J Pediatr Surg 1998;33:705 - 7. [6] Ledbetter DJ, Juul SE. Erythropoietin and the incidence of necrotizing enterocolitis in infants with very low birth weight. J Pediatr Surg 2000;35:178 - 82. [7] Juul SE, Joyce AE, Zhao Y, et al. Why is erythropoietin present in human milk? Studies of erythropoietin receptors on enterocytes of human and rat neonates. Pediatr Res 1999;46:263 - 8. [8] Juul SE, Ledbetter DJ, Joyce AE, et al. Erythropoietin acts as a trophic factor in neonatal rat intestine. Gut 2001;49:182 - 9. [9] Brines ML, Ghezzi P, Keenan S, et al. Erythropoietin crosses the blood-brain barrier to protect against experimental brain injury. Proc Natl Acad Sci U S A 2000;197:10526 - 31. [10] Ashley RA, Dubuque SH, Dvorak B, et al. Erythropoietin stimulates vasculogenesis in neonatal rat mesenteric microvascular endothelial cells. Pediatr Res 2002;51:472 - 8.
Erythropoietin restores bowel damage and hypoperistalsis in gastroschisis [11] Aktug T, Erdag G, KargV A, et al. Amnioallantoic fluid exchange for the prevention of intestinal damage in gastroschisis: an experimental study in chicken embryos. J Pediatr Surg 1995;30:384 - 7. [12] Thakur A, Chiu C, Quiros-Tejeira RE, et al. Morbidity and mortality of short bowel syndrome in infants with abdominal wall defects. Am Surg 2002;68:75 - 9. [13] Burc L, Volumenie JL, de Lagausie P, et al. Amniotic fluid inflammatory proteins and digestive compounds profile in fetuses with gastroschisis undergoing amnioexchange. Br J Obstet Gynaecol 2004;111:292 - 7. [14] Dommergues M, Ansker Y, Aubry MC, et al. Serial transabdominal amnioinfusion in the management of gastroschisis with severe oligohydramnios. J Pediatr Surg 1996;31:1297 - 9. [15] Langer JC, Bramlett G. Effect of prokinetic agents on ileal contractility in a rabbit model of gastroschisis. J Pediatr Surg 1997;32:605 - 8. [16] Aktug T, Ucan B, Olguner M, et al. Amnioallantoic fluid exchange for prevention of intestinal damage in gastroschisis II: effects of exchange performed by using two different solutions. Eur J Pediatr Surg 1998; 8:308 - 11. [17] Aktug T, Demir N, Akgur FM, et al. Pretreatment of gastroschisis with transabdominal amniotic fluid exchange. Obstet Gynecol 1998;91:821 - 3.
357
[18] Haller JA, Kehrer BH, Shaker IJ, et al. Studies of the pathophysiology of gastroschisis in fetal sheep. J Pediatr Surg 1974; 9:627 - 32. [19] Tibboel D, Klqck P, van der Kamp AWM, et al. The development of the characteristic anomalies found in gastroschisis — experimental and clinical data. Z Kinderchir 1985;40:355 - 60. [20] Sencan A, Gumustekin M, Gelal A, et al. Effects of amnio-allantoic fluid exchange on bowel contractility in chick embryos with gastroschisis. J Pediatr Surg 2002;37:1589 - 93. [21] Sakanaka M, Wen T, Matsuda S, et al. In vivo evidence that erythropoietin protects neurons from ischemic damage. Proc Natl Acad Sci U S A 1998;95:4635 - 40. [22] Albert A, Margarit J, Julia V, et al. Morphology and mucosal biochemistry of gastroschisis intestine in urine free amniotic fluid. J Pediatr Surg 2003;38:1217 - 20. [23] Steinbrecher HA, Hanna M, Burge DM. Gastroschisis bowel in an intact exomphalos: implications for etiology and possible prevention. J Pediatr Surg 1996;31:342 - 3. [24] Kling PJ. Roles of erythropoietin in human milk. Acta Paediatr Suppl 2002;438:31 - 5.
www.elsevier.com/locate/jpedsurg
Erythropoietin restores bowel damage and hypoperistalsis in gastroschisis Aykut Ozdamar a, Koray Topcu a, Mukaddes Gumustekinb, Duygu Gurelc, Ayse Gelalb, Erdener Ozerc, Basak Ucana, Gunyuz Temir a, Aytac Karkiner a, Irfan Karaca a, Munevver Hosgora,* a
Department of Pediatric Surgery, Dr. Behcet Uz Children’s Hospital, Izmir, Turkey Department of Pharmacology, Dokuz Eylul University, Izmir, Turkey c Department of Pathology, Dokuz Eylul University, Izmir, Turkey b
Index words: Gastroschisis; Erythropoietin; Gastrointestinal contractility
Abstract Background and Purpose: Despite the decreased mortality in gastroschisis (Gx), patients experience postoperative intestinal hypoperistalsis, malabsorption, and shortened bowel length. The trophic effects of recombinant human erythropoietin (rEpo) in the developing small bowel have been reported, increasing the length and height of the villi, and villous surface area. This study investigated the effects of rEpo on intestinal malfunction in the chick embryos with Gx. Methods: Thirteen-day-old fertilized chicken eggs were used to create Gx model. Study groups included the following: group 1, control; group 2, Gx-only; group 3, Gx + 0.075% saline exchange; group 4, Gx + 10 IU rEpo exchange; group 5, Gx + 20 IU rEpo exchange. The bowels were evaluated by in vitro muscle strip technique, and the response was expressed as a percentage of the maximum carbachol-evoked contraction (E max). In addition, parasympathetic ganglion cells per 10 plexuses and villi height were determined by light microscopy. Results were evaluated statistically by Mann-Whitney U, v 2, and Fisher’s Exact test tests. Results: Saline exchange had no effect on ganglion cell number ( P = .63) and villi height ( P = .10). In group 4, ganglion cell number was not increased ( P = .82), but villi height increase was significant ( P = .03). In Gx + 20 IU rEpo group, both the number of ganglia ( P = .0001) and villi height ( P = .002) were significantly increased. The decrease in contractility in group 2 ( P = .0121) was significantly reversed by rEpo 20 IU treatment ( P = .0216), no significant difference was obtained in groups 3 ( P = .0809) and 4 ( P = .1516) compared with group 2. Conclusion: These data suggest that rEpo has prokinetic effects on hypoperistalsis and restores bowel damage in Gx. D 2006 Published by Elsevier Inc.
Presented at the 52nd Annual Congress of British Association of Paediatric Surgeons, Dublin, Ireland, July 12-15, 2005. 4 Corresponding author. Koruturk Mh Simsek Sk. Yagmur Sitesi K: 1D:1, 35330 Izmir, Turkey. Tel.: +90 232 4895656/5303; fax: +90 232 4892315. E-mail address: [email protected] (M. Hosgor). 0022-3468/$ – see front matter D 2006 Published by Elsevier Inc. doi:10.1016/j.jpedsurg.2005.11.012
Gastroschisis (Gx) is a congenital full-thickness cleft in the anterior abdominal wall, usually to the right side of the umbilical insertion, resulting in protrusion of abdominal contents through the cleft [1]. The characteristic picture of Gx includes congestion, edema, fibrous coating, and
Erythropoietin restores bowel damage and hypoperistalsis in gastroschisis adhesive matting of the protruding bowel loops, which occurs presumably from direct exposure to the amniotic fluid and from partial lymphatic and venous obstruction in the mesentery [2]. Severe intestinal damage is also evidenced by intestinal dilatation, smooth muscle thickening, mesenteric shortening, villous atrophy, great loss of contractility, and decreased mucosal absorption [3]. The incidence of Gx is increasing throughout the world, and it occurs in about 1 of every 4000 live births [4]. Although neonatal intensive care and surgical techniques have significantly improved, a number of patients experience prolonged intestinal hypoperistalsis and malabsorption after surgical repair and may predispose to necrotizing enterocolitis [5]. Erythropoietin (Epo) is a growth factor produced by the kidney in response to anemia [6]. In previous studies, the presence of Epo in human milk and functional Epo receptors in fetal and postnatal small intestines has been proposed because Epo has a role in growth and development of the gastrointestinal tract [7]. It has been shown that Epo binds to enterocytes and stimulates small intestinal growth by increasing length, villous surface area, villi height, and ileal crypt depth [8]. Moreover, it was also reported that Epo protects neurons against ischemia-induced cell death and acts as an angiogenic growth factor for intestinal mesentery microvascular endothelial cells [9,10]. The present study was undertaken using a chick embryo model to determine if intraamniotic recombinant human erythropoietin (rEpo) administration would prevent intestinal damage and enhance bowel contractility in Gx.
1. Materials and methods 1.1. Experimental procedures All experimental procedures were approved by Ethical Committee of the Dr. Behcet Uz Children’s Hospital.
353
Fig. 2 Macroscopic view of Gx after AAF exchange with 0.075% physiological saline.
Thirteen-day-old fertilized chick eggs (Gallus domesticus) were incubated at 37.58C in 80% humidity. Operative procedures for all groups were performed on the 13th day of incubation, using previously published methods [11]. The eggs were divided into 5 groups. In the first group (control) (n = 10), through an eggshell window, the amnioallantoic membrane was opened to create a common cavity, and thus amnioallantoic fluid (AAF) mixture was created, which resembled human amniotic fluid. In the second group (Gx-only; n = 13), Gx was created. During creation of the Gx model, after opening amnioallantoic membrane, a 2.5mm defect was created with delicate tweezers in the umbilical stalk near the abdominal wall, and intestinal loops were exteriorized out of the abdomen. In groups 3 (n = 12), 4 (n = 11), and 5 (n = 15), after creating Gx as described, a catheter was placed into the amnioallantoic cavity at the end of the procedure. AAF exchange was performed with only 0.075% physiological saline in group 3, with 10 IU rEpo (Eprex, Cilag, Zug, Switzerland) solution diluted with 0.075% saline in group 4, and with 20 IU rEpo solution diluted with 0.075% saline in group 5. Exchange in these groups was performed once a day with a volume corresponding to 10% of AAF, 1 mL at the 15th and 16th days and 0.5 mL at the 17th day before hatching as previously described. On the 18th day of hatching, all embryos were killed.
1.2. Light microscopy
Fig. 1 Gx-only group: macroscopic changes including thickening, fibrous peel on the serosal surfaces and fibrous adhesions between bowel loops.
Tissue specimens were taken from intestine of each chick embryos and fixed in 10% formalin. Three sections from a 1-cm portion of the intestinal segments of each embryo were selected randomly. Intestines were examined macroscopically and microscopically. All microscopic examinations were carried out by the same pathologist (EO) under light microscopy in a blinded manner. The numbers of parasympathetic ganglia were counted in 10 plexuses of each section (total, 30 plexuses). The villi height per 10 villi from each
354
A. Ozdamar et al.
Fig. 3 Macroscopic view of Gx after AAF exchange with 10 IU rEpo.
section were measured using a computer assisted image analyzer system consisting of a microscope (Labophot-2, Nikon, Tokyo, Japan) equipped with a high-resolution video camera (VKC220E; Hitachi, Tokyo, Japan). The images were processed using an IBM-compatible personal computer, high-resolution video monitor (Sony, Tokyo, Japan), and image analysis software (BS 200Docu Version 2.0, BAB Imaging Systems, Ankara, Turkey). The number of ganglia was reported as mean value per 10 plexuses. Results were evaluated statistically by v 2 and Fisher’s Exact test tests.
1.3. Contractility studies Contractility studies were performed using in vitro muscle strip technique during the first 2 hours after removal. Similar bowel segments selected for ganglion cell count were also selected randomly for contractility studies. Mesentery tissue was carefully removed, and the bowel was dissected into strips 10 mm. Full-thickness muscle strips were subsequently mounted in a 20-mL jacketed organ bath filled with Tyrode solution composed of (in mmol/L) NaCl,
Fig. 4
Macroscopic view of Gx + 20 IU rEpo exchange.
Fig. 5 Histological view of Gx-only group. Note decreased villi height and ganglion cell number (arrows) and the fibrous peel (fp) (hematoxylin-eosin, original magnification 100).
136.8; KCl, 2.7; MgCl2, 1.1; CaCl2, 1.8; NaH2PO4, 0.4; NaHCO3, 11.9; and glucose, 5.6 (pH 7.4). The Tyrode solution was maintained at 378C and bubbled with 95% O2 + 5% CO2. A resting tension of 0.5 g was applied, and the strips were allowed to equilibrate for 60 minutes. The preparations were washed approximately every 15 minutes during equilibration period. Isometric tension changes were measured using a Grass Instrument 7FT 03E isometric transducers connected a computer-based data acquisition and analysis software system (MAY Polwin 97, Acquisition Software, Ankara, Turkey). Carbamylcholine chloride (carbachol [Cch] 10ÿ9 to 10ÿ4 mol/L) (Sigma, St. Louis, Mo) was applied directly to the organ baths, and contractions were recorded isometrically. Cumulative concentrationresponse curve was constructed for each sample in this study. Each dose of the cumulative curve was added when the effect of preceding one had reached its maximum.
Fig. 6 Histological view of Gx + 20 IU rEpo Exchange. Note that villi architecture is restored and number of ganglion cells is increased (arrow) (hematoxylin-eosin, original magnification 100).
Erythropoietin restores bowel damage and hypoperistalsis in gastroschisis Table 1 groups
The number of ganglia and villi height in study
Study groups
Number of ganglia
Villi height (lm)
Gx-only Gx + % 0.075 saline Gx + 10 IU rEpo Gx + 20 IU rEpo Control
68.84 69.02 73.60 90.00 77.80
52.83 67.62 69.67 75.29 70.05
F F F F F
33.98 28.24 22.82 31.13 9.01
F F F F F
14.24 19.09 24.23 16.49 10.10
Each strip was used in 1 experiment only. The same procedure was done in all groups. The response was expressed as a percentage of the maximum Cch-evoked contraction (E max) in the control group. Isometric tension measurements were expressed as a mean value F SD. Maximal response to Cch in the Gx-only group was compared with the other groups using MannWhitney U test. P b .05 was considered significant, and n denotes the number of strips.
2. Results
Table 3 groups
355
Maximum Cch-evoked contraction (E max) in study
Study groups
E max
Gx-only Gx + 0.075% saline Gx + 10 IU rEpo Gx + 20 IU rEpo Control
75.81 98.92 100.8 108.01 100
F F F F F
39.05 38.17 54.9 49.284 27.344
Values are mean F SD. The maximal responses are expressed as percentage of the E max obtained for control group. n denotes the number of strips. 4 P = .0216 Gx + 20 IU rEpo vs Gx-only group. 44 P = .0121 Control group vs Gx-only group.
compared with Gx-only group ( P = .63 and P = .82, respectively). Saline exchange also had no effect on villi height ( P = .10). Although the number of ganglia has not increased significantly after exchange with rEpo 10 IU, villi height were significantly increased compared with Gx-only group ( P = .03). In Gx + 20 IU rEpo group, both the number of ganglia ( P = .0001) and villi height ( P = .002) were significantly increased (Table 2).
2.3. Contractility findings
2.1. Macroscopic findings In Gx-only group, macroscopic changes included thickening, fibrous peel on the serosal surfaces and fibrous adhesions between bowel loops (Fig. 1). In all exchange groups, there was no fibrous peel formation or intestinal wall thickening (Figs. 2 -4).
2.2. Microscopic findings In Gx-only group, thickened serosal surface, fibrin deposition in the serosa, an inflammatory reaction, and edema were evident during microscopic evaluation (Fig. 5). These findings were not observed in any exchange groups (Fig. 6). In all groups, ganglia were morphologically normal. However, the number of ganglion cells and villi height were significantly decreased in Gx-only group compared with control group ( P = .0001 and P = .001, respectively) (Table 1). The mean number of parasympathetic ganglia per 10 plexuses were increased in groups 3 and 4 (69.02 F 28.24 and 73.60 F 22.82, respectively); however, these results were not statistically significant when
In randomized order, 1 to 5 full-thickness muscle strips were used from each embryo (group 1, n = 30; group 2, n = 29; group 3, n = 15; group 4, n = 20, group 5, n = 25). Cch (10ÿ9 to 10ÿ4 mol/L) caused contraction in a concentration-dependent manner to a maximum effect at 10ÿ4 mol/L (E max) in all groups. The E max value was decreased significantly in Gx-only group compared with control group ( P = .0121). Exchange with rEpo 20 IU has significantly improved contractility in fetal bowel strips with Gx ( P = .0216). Although exchange with 0.075% saline and with rEpo 10 IU have increased the E max values, the differences were not statistically significant ( P = .0809 and P = .1516, respectively) (Table 3). The results are summarized in Fig. 7.
Table 2 Statistical results of comparison of the number of ganglia and villi height Study groups
The number Villi height of ganglia ( P) ( P)
Gx-only Gx-only Gx-only Gx-only
.63 .82 .0001 .0001
vs vs vs vs
Gx + 0.075% Saline Gx + 10 IU rEpo Gx + 20 IU rEpo control
.10 .03 .002 .001
Fig. 7 Maximal contractile response to 10ÿ4 mol/L Cch. The maximal responses are expressed as percentage of the E max obtained for control group. Values are expressed mean F SD (n = 15-30). *P b .05 Gx-only vs Gx + 20 IU rEpo E max; **P b .05 Gxonly vs control E max.
356
3. Discussion Despite prenatal diagnosis of Gx and immediate appropriate neonatal care, significant postoperative morbidity still occurs as a consequence of bowel damage in utero [12]. Amniotic fluid exchange or amnioinfusion have been reported to decrease bowel damage in experimental animal studies as well as in human Gx [13,14]. The majority of these studies have primarily focused on prevention of the intestinal morphological damage. However, there are limited and sometimes conflicting literature information on pathogenesis and treatment of the intestinal malfunction seen in infants with Gx [2,15]. The present study demonstrated that intraamniotic rEpo administration restores, in a dosedependent manner, not only morphological damage but also intestinal dysfunction associated with Gx. In the current study, exposure of the intestines to AAF in Gx-only group has also resulted in bowel wall thickening, fibrous peel on the serosal surfaces, and fibrous adhesions between bowel loops as described in previous reports [16]. After exchange with either only physiological saline or rEpo + saline, normal appearing intestines that was not thickened with no visible fibrous peel or adhesion were observed, these findings supported that AAF exchange decreases the concentration of harmful chemicals responsible for intestinal morphological damage, mainly serosal peel formation and bowel wall thickening as previously reported in other clinical and experimental studies [17]. Using a fetal lamb model of Gx, Haller et al [18] have proposed that the disordered peristalsis in the eviscerated bowel results from atrophy of the ganglion cells; however, all of the studies later performed on different animal models did not confirm this finding [19]. In addition, in our study, ganglia were morphologically normal in all groups. In a recent report, the number of parasympathetic ganglia was found to be diminished in Gx, and this decrease was suggested as a secondary phenomenon caused by serosal inflammation or apoptosis [20]. Moreover, these investigators have reported that the decrease in the number of ganglia can be prevented by AAF exchange with physiological saline. We have also observed a significant decrease in the number of ganglia in Gx-only group, but AAF exchange with physiological saline has not increased the number of ganglion cells. In contrast, a dose-dependent increase in the number of ganglia was found after AAF exchange with rEpo. It is well known that rEpo decreases apoptotic death of neuronal cells and enterocytes [21]. Besides its neuroprotective and neurotrophic effects, it was also reported that rEpo acts in a manner consistent with known anti-inflammatory agents such as glucocorticoids [9]. Therefore, it can be speculated that by exerting its neuroprotective, anti-inflammatory, and antiapoptotic effects, rEpo have prevented the decrease in the number of ganglion cells. The intestinal dysfunction in Gx is believed to result from both loss of contractility and poor absorptive capacity because of mucosal villous atrophy and immaturity of the
A. Ozdamar et al. mucosal enzymatic systems [22]. The etiology of these changes in Gx has been attributed to a variety of causes, including both amniotic fluid exposure and intrauterine ischemia caused by gradual constriction of the bowel at the abdominal wall defect [23]. In the present study, bowel contractility and villi height were significantly lower in Gx-only group than normal. Physiological saline exchange had no significant effect on either bowel contractility or villi height. In contrast, AAF exchange with rEpo significantly improved contractility in a dose-dependent manner; in addition, histological examination of the bowel mucosa showed longer villi in both rEpo-treated groups. In previous rat studies, it has been shown that rEpo significantly increases intestinal length and the absorptive surface of the microvilli, primarily by increasing the villi length [8]. Moreover, Epo is an angiogenic growth factor that stimulates intestinal mesentery microvascular endothelial cell proliferation and increases intestinal blood flow [10]. In this respect, our results suggested that intraamniotic rEpo administration in Gx acts not only as a trophic factor but also as a prokinetic agent on a variety of nonhematopoietic cell types, including enterocytes, endothelial cells, smooth muscle cells, and neuronal cells; cell types that are present in the developing bowel [24]. In conclusion, the data presented here constitute a basis for examining the effectiveness and appropriate dose of rEpo in further studies for treatment of all aspects of intestinal malfunction in human Gx.
References [1] Oyachi N, Laksmanan J, Ross MG, et al. Fetal gastrointestinal motility in a rabbit model of gastroschisis. J Pediatr Surg 2004; 39:366 - 70. [2] Guo W, Swaniker F, Fonkalsrud EW, et al. Effect of intraamniotic dexamethasone administration on intestinal absorption in a rabbit gastroschisis model. J Pediatr Surg 1995;30:983 - 7. [3] Langer JC, Longaker MT, Crombleholme TM, et al. Etiology of intestinal damage in gastroschisis I: effects of amniotic fluid exposure and bowel constriction in a fetal lamb model. J Pediatr Surg 1989; 24:992 - 7. [4] Baerg J, Kaban G, Tonita J, et al. Gastroschisis: a sixteen year review. J Pediatr Surg 2003;38:771 - 4. [5] Jayanthi S, Seymour P, Puntis JWL, et al. Necrotizing enterocolitis after gastroschisis repair: a preventable complication? J Pediatr Surg 1998;33:705 - 7. [6] Ledbetter DJ, Juul SE. Erythropoietin and the incidence of necrotizing enterocolitis in infants with very low birth weight. J Pediatr Surg 2000;35:178 - 82. [7] Juul SE, Joyce AE, Zhao Y, et al. Why is erythropoietin present in human milk? Studies of erythropoietin receptors on enterocytes of human and rat neonates. Pediatr Res 1999;46:263 - 8. [8] Juul SE, Ledbetter DJ, Joyce AE, et al. Erythropoietin acts as a trophic factor in neonatal rat intestine. Gut 2001;49:182 - 9. [9] Brines ML, Ghezzi P, Keenan S, et al. Erythropoietin crosses the blood-brain barrier to protect against experimental brain injury. Proc Natl Acad Sci U S A 2000;197:10526 - 31. [10] Ashley RA, Dubuque SH, Dvorak B, et al. Erythropoietin stimulates vasculogenesis in neonatal rat mesenteric microvascular endothelial cells. Pediatr Res 2002;51:472 - 8.
Erythropoietin restores bowel damage and hypoperistalsis in gastroschisis [11] Aktug T, Erdag G, KargV A, et al. Amnioallantoic fluid exchange for the prevention of intestinal damage in gastroschisis: an experimental study in chicken embryos. J Pediatr Surg 1995;30:384 - 7. [12] Thakur A, Chiu C, Quiros-Tejeira RE, et al. Morbidity and mortality of short bowel syndrome in infants with abdominal wall defects. Am Surg 2002;68:75 - 9. [13] Burc L, Volumenie JL, de Lagausie P, et al. Amniotic fluid inflammatory proteins and digestive compounds profile in fetuses with gastroschisis undergoing amnioexchange. Br J Obstet Gynaecol 2004;111:292 - 7. [14] Dommergues M, Ansker Y, Aubry MC, et al. Serial transabdominal amnioinfusion in the management of gastroschisis with severe oligohydramnios. J Pediatr Surg 1996;31:1297 - 9. [15] Langer JC, Bramlett G. Effect of prokinetic agents on ileal contractility in a rabbit model of gastroschisis. J Pediatr Surg 1997;32:605 - 8. [16] Aktug T, Ucan B, Olguner M, et al. Amnioallantoic fluid exchange for prevention of intestinal damage in gastroschisis II: effects of exchange performed by using two different solutions. Eur J Pediatr Surg 1998; 8:308 - 11. [17] Aktug T, Demir N, Akgur FM, et al. Pretreatment of gastroschisis with transabdominal amniotic fluid exchange. Obstet Gynecol 1998;91:821 - 3.
357
[18] Haller JA, Kehrer BH, Shaker IJ, et al. Studies of the pathophysiology of gastroschisis in fetal sheep. J Pediatr Surg 1974; 9:627 - 32. [19] Tibboel D, Klqck P, van der Kamp AWM, et al. The development of the characteristic anomalies found in gastroschisis — experimental and clinical data. Z Kinderchir 1985;40:355 - 60. [20] Sencan A, Gumustekin M, Gelal A, et al. Effects of amnio-allantoic fluid exchange on bowel contractility in chick embryos with gastroschisis. J Pediatr Surg 2002;37:1589 - 93. [21] Sakanaka M, Wen T, Matsuda S, et al. In vivo evidence that erythropoietin protects neurons from ischemic damage. Proc Natl Acad Sci U S A 1998;95:4635 - 40. [22] Albert A, Margarit J, Julia V, et al. Morphology and mucosal biochemistry of gastroschisis intestine in urine free amniotic fluid. J Pediatr Surg 2003;38:1217 - 20. [23] Steinbrecher HA, Hanna M, Burge DM. Gastroschisis bowel in an intact exomphalos: implications for etiology and possible prevention. J Pediatr Surg 1996;31:342 - 3. [24] Kling PJ. Roles of erythropoietin in human milk. Acta Paediatr Suppl 2002;438:31 - 5.