Alatas, Fatima S.; Masumoto, Kouji; Esumi, Genshiro; Nagata, Kouji; Taguchi, Tomoaki
Duodenal atresia (DA) is a well-known intestinal disease, which frequently causes intestinal obstruction in newborns, and it is commonly associated with other congenital anomalies (1,2). DA accounts for 25% to 40% of all cases with intestinal atresia (IA) (3). The frequency of DA in Japan is reported to be 1 in 3000 to 5000 live births.
Various surgical procedures to anastomose the proximal dilated site to the distal site, such as duodenoduodenostomy and duodenojejunostomy, have been introduced with promising results (4–7). Successful surgical repair has also been reported, with a mortality rate as low as 3% to 5% after correction, in addition to an excellent long-term survival rate (8–10). Even after successful surgery, however, the proximal site can continue to be severely dilated with hypoperistalsis, resulting in intestinal dysmotility problems in later life (11). Dysmotility problems in DA appear to be caused by a dilatation of the intestine proximal to the obstructed site, which has not been adequately resected. Intestinal dysmotility often leads to functional obstruction characterized by marked dilatation resulting from ineffective peristalsis (12–14). These findings have been supported by previous manometric studies, which found a reduction in the intraluminal manometric pressure and a transit disturbance in the dilated proximal intestines (15,16).
Similar to other types of IA, in DA, hyperplasia and hypertrophy of the smooth muscle are found in varying degrees in the proximal site of obstruction, whereas these same conditions are rarely observed at the distal site of the obstruction (17,18). Chick studies have demonstrated several abnormalities in the intramural nervous system, muscular elements, and the interstitial cells of Cajal (ICC) in the proximal dilated segment of the IA. These findings were found not only in human samples of patients with IA but also in a chick IA model (11,19). There are no published data describing the differences between the proximal and distal sections of obstructed sites regarding the intramural components in patients with DA. In the present study, we investigated the morphologic differences in the enteric nervous system, the ICC, and smooth muscle, between the regions proximal and distal to the obstructed site in neonates with DA, to enhance our understanding of motility problems in patients with DA.
Twelve resected duodenal samples obtained from neonates with DA who were delivered at Kyushu University Hospital (Fukuoka, Japan) were used in the present study after obtaining the approval of the university ethics committee. The subjects’ gestational ages were 34 to 40 weeks, and the duodenal samples were obtained at the primary operation during the subjects’ first to third days after birth. The 0.5-cm anterior walls of both the proximal and distal segments apart from the obstructed site were collected as samples. Age-matched duodenal samples of controls were obtained from 2 patients without gastrointestinal disease at an autopsy (congenital diaphragmatic hernia). The number of control material samples was insufficient because of the difficulty in obtaining normal controls for the present study. Formalin-fixed, paraffin-embedded tissues were cut into 4-μm-thick slices and were processed for immunohistochemistry.
Duodenal specimen slices were stained with hematoxylin and eosin to evaluate the presence of the submucosal and myenteric plexus and the smooth muscle layer before performing immunohistochemical staining.
All of the specimens were immunohistochemically stained using the standard avidin-biotin complex method. The primary antibodies were a polyclonal antibody to S-100 protein (code no. 422091 Nichirei Co Ltd, Tokyo, Japan) as a general neuronal marker and an antibody to c-kit protein (CD-117, diluted 1:100; DakoCytomation, Carpinteria, CA) as a marker of ICC, and a monoclonal antibody to α-smooth muscle actin (α-SMA, clone IA4, diluted 1:200; Sigma Immunochemicals, St Louis, MO) as a general muscle marker (Table 1).
In brief, after deparaffinization in xylene and dehydration in 100% alcohol, slides were treated with 3% H2O2 in methanol to block endogenous peroxidase activity. For antigen retrieval, the slides were subjected to 10 minutes of microwave treatment in citric acid buffer (pH 6.0). After cooling to room temperature, the slides were incubated with an undiluted blocking solution (Histofine, SAB-PO [MULTI] Kit; Nichirei) containing goat serum albumin. After rinsing with phosphate-buffered saline (PBS), the slides were incubated with a primary antibody (Table 1). The slides were rinsed twice with PBS and incubated for 10 minutes with an undiluted biotinylated secondary antibody (Histofine). Slides were then rinsed again twice with PBS followed by incubation for 10 minutes with undiluted peroxidase-conjugated streptavidin (Histofine). In all of the duodenal specimens stained, peroxidase was detected by diaminobenzidine tetrahydrochloride (Histofine, DAB Kit, Nichirei) with purified water for 5 minutes. Finally, the slides were rinsed with running tap water and counterstained with hematoxylin, dehydrated through a graded alcohol series, and washed with xylene.
Evaluation and Analysis
The morphologic differences were evaluated among the proximal and distal segments of the obstructed sites of DA samples and compared with those of the controls. In addition, the differences between the segments were quantitatively presented. For quantitative evaluation, all of the sections were photographed using light microscopy using 4×, 10×, 20×, and 40× magnifications. The quantification of the immunoreactivities of c-kit were evaluated from each slide by measuring the length of the c-kit-positive area, and the immunoreactivities of S-100 were evaluated from each slice by measuring the length of each stained ganglion or plexus. The α-SMA antibody immunoreactivities were quantified by measuring the thickness of the longitudinal muscle layer, the circular muscle layer, and the muscularis mucosae in 3 different locations. The mean values of the 3 area measurements are presented in Results. All of the measurements were taken using the ImageJ version 1.43s software program (National Institutes of Health, Bethesda, MD) in an area 4080 × 3072 pixels wide, and then were converted to micrometers according to each magnification. The differences between the c-kit-positive area, the length of the neuronal cells, and the width of the mucosal muscle layers of the proximal and distal segments were compared using the Student t test. Differences between the width of the circular muscle layer and the longitudinal muscle layer of the proximal and distal segments were compared using the Mann-Whitney U test. P < 0.05 were considered to be statistically significant. All of the statistical analyses were performed with the SPSS statistical software program (SPSS Inc, Chicago, IL).
The present study included 9 patients with type III and 3 patients with type I DA. All of the patients had been antenatally diagnosed as having DA during the prenatal period; therefore, all of the patients were treated in our department after the immediate diagnosis for the confirmation of DA following birth. In all of the patients, a longitudinal incision and a transverse duodenoduodenostomy (diamond-shape anastomosis) or membranous resection was performed 1 to 3 days after birth.
S-100 Protein Staining
In the control samples, immunoreactivity to S-100 antibody was observed in the myenteric, submucosal plexuses, and nerve fibers distributed throughout the entire bowel wall layers. Auerbach plexus between the circular and longitudinal muscle layers was clearly labeled by S-100 protein immunostaining. Several positive fibers were also detected in the muscularis mucosae and in the villus of the lamina propria (data not shown). In the proximal segment of the obstructed sites in the cases with DA, an abnormal nervous distribution showing S-100-positive immunoreactivity was observed (Fig. 1A, B). In the myenteric plexus, the number and size of the S-100-positive plexus were smaller than those in the distal segments and the controls. In addition, a small number of ganglion cells were also observed. The ganglion cells were small and immature compared with those of the distal segments and the controls. Nerve fibers observed between the circular and the longitudinal musculature not only were fewer in number but also were composed of smaller fibers. The size and length of the ganglionic cells were also smaller than those of the distal segments and the controls, especially in the area where hypertrophic musculature was observed (Fig. 1A, B). In contrast, the nervous distribution of the distal segment of obstructed site was undistinguishable from those of controls (Fig. 1C, D). Quantitative analyses showed a significantly shorter ganglion length and plexus of the proximal segments than was observed in the distal segments of the obstructed site (proximal 203.43 ± 103.49 μm, distal 297.67 ± 136.58 μm, P = 0.002, Table 2). There was no significant difference in the length of the ganglion and myenteric plexus between the distal segments of the obstructed site and control tissues (P = 0.134).
In the control samples, a homogenous immunoreactivity to the α-SMA antibody was observed in all of the muscle layers: the muscularis mucosae, circular muscle layer, and the longitudinal muscle layers (data not shown). In the proximal segment of the obstructed site in samples of patients with DA, a moderately to severely hypertrophic area was observed in the circular and longitudinal muscle layers, particularly in the circular muscle layers, compared with those of the distal segments or the controls (Fig. 2A, B, C). The innermost layer of the circular muscle layer in the proximal segments was also thinner and was nearly undetectable in some sections, compared with those in the distal segments or the controls. In addition, in all of the patients, unusual ectopic muscle bundles were found around the submucosal connective tissue near the innermost layer of the circular muscle layers (Fig. 2B). These smooth muscle bundles originated from the muscle bundle in the thin innermost circular muscle layer. In addition, the muscularis mucosae were also found to be hypertrophic (Fig. 2C). In contrast, in the distal segments of the obstructed site in DA, the staining pattern of α-SMA antibody was similar to that observed in the control samples (Fig. 2D, E, F). There was no significant difference in quantitative analyses of the thickness of the circular muscle layers, the longitudinal muscle layers, and the muscularis mucosae between the distal segments of obstructed sites and control tissues (P = 0.646, P = 598, and P = 0.395, respectively). In contrast, quantitative analyses revealed a significant difference in the thickness of the muscularis mucosae of the proximal segments compared with the distal segments of the obstructed site (proximal 36.46 ± 13.51 μm, distal 12.52 ± 6.06 μm, P < 0.001, Table 2). Quantitative analyses also revealed a significant difference between the circular muscle layers of the proximal and distal segments of the obstructed site (proximal 492.91 μm [range 263.19–733.16 μm], distal 164.94 μm [range 135.61–199.37 μm], P < 0.001, Table 2). There was also a significant difference between the proximal and distal segments of the longitudinal muscle layers (proximal 317.64 μm [range 110.73–369.18 μm], distal 96.28 μm [range 73.83–121.20 μm], P < 0.001, Table 2).
In the control samples, c-kit-positive cells were observed between the intermuscular space of the circular and longitudinal muscle layers, particularly around Auerbach plexus (data not shown). A small number of positive cells also were localized to the circular and longitudinal muscle layers and in the innermost layer of the circular muscle layers. In the proximal segment, the number of c-kit-positive cells was markedly decreased (Fig. 3A, B). Moreover, in some samples, the ICC were barely detectable, even around the myenteric plexus. The positive cells were bipolar in shape. Macrophage-like cells positive for c-kit staining were also observed within the muscularis propria and the submucosal area. Unlike the proximal segments, the distribution pattern and c-kit immunoreactivity in the ICC of the segment distal to the obstructed site were similar to those of the control samples (Fig. 3C, D; P = 0.133). These cells formed a network with cell-cell contacts and their shape was multipolar. The quantitative analyses revealed a significantly smaller c-kit-positive area in the proximal segments compared with the distal segments of the obstructed site (proximal 933.45 μm2/mm2 [range 297.87–3149.62 μm2/mm2], distal 12006.42 μm2/mm2 [range 2473.79–22458.1 μm2/mm2], P < 0.001, Table 2).
As a part of the normal growth process in the embryo from the fourth to the seventh week of gestation (embryo length 8.6–14.5 mm), the epithelial cells of the duodenum begin to proliferate and completely plug the lumen (solid phase). Therefore, between the 15- and 65-mm stages (from the end of the 7th week to the 12th week of gestation), a process of vacuolization, coalescence of vacuoles, and recanalization occurs. DAs, stenosis, and intraluminal webs are believed to be caused by insults resulting in recanalization failure during the lengthening and rotation of the primitive foregut. A resultant obstruction is then believed to cause the dilatation in the proximal duodenum (20–22).
Using histochemical techniques, we have herein provided the first study of the potential differences in abnormalities of the smooth muscle, nervous system, and ICC between the segments proximal and distal to the obstructed site in samples from patients with DA. These 3 enteric components, which are involved in peristaltic activity, were immunohistochemically analyzed.
Although the etiology of DA is not the same as that in other IAs, the morphologic changes of the intestine in DA are thought to be similar. The morphologic change of the proximal segment seems to depend on the postobstructive dilatation during the fetal period. Similarly, in other forms of IA, morphologic change in the proximal segments also depends on the change after the formation of the obstruction.
In an experimental model of IA, several studies showed that the dilatation of the proximal segment induces the involution and lysis of the ganglion cells after initial hyperplasia of the myenteric ganglia has occurred and irreversible distension continues to develop (23). Another possible cause of nerve alteration could be ischemic influence during fetal life through vascular disruption as shown in IA models (19,23). In the present study, we observed that in the area in which muscular layers are severely dilated, the distribution of ganglion and plexus is also less than that of the area with moderately dilated muscular layers. This finding supported a previous IA study (19), which observed marked abnormal neuronal changes in the distended proximal segments. In contrast, the neuronal distribution in the distal segment was close to normal in these studies, as in our study of human DA. The influence of muscular distention on the neuronal cell alteration is also shown in our previous case report of IA (24). In this case report, an improvement in numbers of neuronal cells and fiber distribution between primary operation of IA at 2 days after birth and second operation for the reconstruction of the dilated proximal segment at 6 months of age was observed (24). These abnormalities are probably a result of the developmental delay in the nervous system or of the dilatation of the proximal segment in DA, as mentioned in IA studies, which also found an alteration of enteric nerves in a severely dilated area of the proximal segment (11).
When examining smooth muscle morphology in the present study, we observed that the muscle layers in the proximal segments of the obstructed site are moderately to severely hypertrophic, unlike in the distal segments, which are indistinguishable from those of the control samples. This muscular hypertrophy has been well documented and results from a compensatory process following obstruction during the prenatal period, and it is localized exclusively to the circular musculature of the distended proximal intestinal segment (23,25,26). As mentioned in the literature, dysmotility of the intestine is often encountered in a severely dilated muscle of the segment proximal to the obstructed site. To prevent recurrence of the dilated segment, the markedly dilated duodenum must be completely reconstructed surgically. Even after reconstruction surgery, however, a proximal segment consisting of a hypertrophic area remains because of the difficulty in visualizing the healthy area and minimizing the resected segment. Previous investigators have shown that the contractile pressure in the dilated proximal intestine of an IA model is lower than that in normal intestines. Moreover, physiologic studies of humans and animal models of IA have also shown a decrease in the motor activity in both the proximal and distal segments. These studies suggest that low contractile pressure was involved in the postoperative dysmotility (17). In addition, the existence of unusual muscle bundles, which have an oblique configuration, likely contributes to the disturbed bowel rhythms. Similar findings have been described in our previous study of cases with IA (17). Therefore, the existence of both hypertrophy of the circular muscle layers and unusual muscle bundles in the submucosal layers likely contributes to the development of motility disorders later in life, even after a successful initial operation. Additional procedures, such as intestinal plication or tapering the dilated intestine, are sometimes needed to produce efficient peristalsis of the proximal intestines.
Of particular interest with regard to the occurrence of unusual muscle bundles, our previous study of human IA showed that the smooth muscle bundles which emerged from the innermost layer of the circular musculature could be of either an oblique or vertical orientation to the long axis of the intestines, stretching toward mucosae, forming a coarse, irregular meshwork in the submucosa (17). Based on the chronologic view of our previous study of myogenesis in chick embryos, it is supposed that these muscle bundles may be a remnant of early developmental stages during the formation of the muscularis mucosae (27). Another study also proposed that a possible explanation for this phenomenon is that the ectopic muscle bundles are a secondary reaction of muscle cells to the chronic and progressive dilatation of the proximal segments. Moreover, these bundles also could indicate a secondary regressive reaction, which the proliferating reaction of regressive smooth muscle cells undergo when they first emerge in the inner layer of the circular muscle layer; thereafter, these smooth muscle cells protrude from the inner layer of the circular muscle layer to the layer of muscularis mucosae, according to the rules of normal development. The real causality regarding the development of ectopic muscle bundles in the proximal segment of the DA remains unclear, and the proposed reasons are based mostly on experimental IA, which is probably not appropriate for DA because of differences in the underlying etiology (27,28).
The small intestine exhibits rhythmic and phasic contractions that form the basis for propagating and segmental contraction. These rhythmic and phasic contractions are generated by the ICC surrounding the myenteric plexus (ICC-MY) between the longitudinal and circular muscular layers and the ICC lining the septa separating the CM bundles (ICC-SEP). Each ICC-MY and ICC-SEP generate a spontaneous electrical slow-wave pacemaker activity that is actively propagated through the ICC network, in addition to regulating smooth muscle membrane potential and mediating enteric neurotransmission. The loss or abnormalities of ICC have been described in a variety of human motility disorders, including hypoganglionosis, Hirschsprung disease, and jejunal and ileal atresia but not in DA (29–31). A reduction in the distribution of pacemaking cells has also been reported in a dilated colon of 2 neonates with atresia of the colon (32). In the present study, we observed that the proximal segments of the obstructed site showed not only a decreased immunoreactivity to c-kit protein but also a markedly reduction in the number of ICC. The distribution of ICC also showed a discrete distribution without connection of ICC cells and a bipolar shape. This finding may be associated with the reduction of pacemaker activity and enteric neurotransmission, thus resulting in hypoperistalsis of the proximal segments. Several physiologic studies also showed that peristalsis and spontaneous contraction were disturbed in the proximal segments. Therefore, the abnormality of ICC in the proximal segments may lead to postoperative dysmotility in DA.
In the present study, abnormalities of the enteric nerves, smooth muscle cells, and ICC were predominantly observed in the proximal segments. It has been pointed out that a tight connection existed between the ICC, enteric nervous system, and the smooth muscle to produce a synchronized and sustainable contraction of the duodenum (30). Therefore, the observed abnormalities in these 3 enteric components of the proximal duodenum suggest that duodenal motility disorders may occur later in postnatal life.
1. Masumoto K, Arima T, Nakatsuji T, et al. Duodenal atresia with a deletion of midgut associated with left lung, kidney, and upper limb absences and right upper limb malformation. J Pediatr Surg
2. Choudhry MS, Rahman N, Boyd P, et al. Duodenal atresia: associated anomalies, prenatal diagnosis and outcome. Pediatr Surg Int
3. Sajja SBS, Middlesworth W, Niazi M, et al. Duodenal atresia associated with proximal jejuna perforations: a case report and review of literature. J Pediatr Surg
4. Waever E, Nielsen OH, Arnbjörnsson E, et al. Operative management of duodenal atresia. Pediatr Surg Int
5. Weber TR, Lewis E, Mooney D, et al. Dudenal atresia: a comparison of techniques of repair. J Pediatr Surg
6. Kimura K, Mukohara N, Nishijuma E, et al. Diamond-shaped anastomosis for duodenal atresia: an experience with 44 patients over 15 years. J Pediatr Surg
7. Adzick NS, Harrison MR, deLorimier AA. Tapering duodenoplasty for megaduodenum associated with duodenal atresia. J Pediatr Surg
8. Escobar MA, Ladd AP, Grosfeld JL, et al. Duodenal atresia and stenosis: long-term follow-up over 30 years. J Pediatr Surg
9. Dalla Vecchia LK, Grosfeld JL, West KW, et al. Intestinal atresia and stenosis: a 25-year experience with 277 cases. Arch Surg
10. Grosfeld JL, Rescorla FJ. Duodenal atresia and stenosis: reassessment of treatment and outcome based on antenatal diagnosis, pathologic variance, and long-term follow-up. World J Surg
11. Masumoto K, Suita S, Nada O, et al. Alterations of the intramural nervous distributions in a chick intestinal atresia model. Pediatr Res
12. Sai Prasad TRR, Bajpai M. Intestinal atresia. Indian J Pediatr
13. Nixon HH. Intestinal obstruction in the newborn. Arch Dis Child
14. de Lorimier AA, Harrison MR. Intestinal plication in the treatment of atresia. J Pediatr Surg
15. Takahashi A, Tomomasa T, Suzuki N, et al. The relationship between disturbed transit and dilated bowel, and manometric findings of dilated bowel in patients with duodenal atresia and stenosis. J Pediatr Surg
16. Cezard JP, Cargill G, Faure C, et al. Duodenal manometry in postobstructive enteropathy in infants with a transient enterostomy. J Pediatr Surg
17. Masumoto K, Suita S, Taguchi T, et al. The occurrence of unusual smooth muscle bundles expressing smooth muscle actin in human intestinal atresia. J Pediatr Surg
18. Doolin EJ, Ormsbee HS, Hill JL. Motility abnormality in intestinal atresia. J Pediatr Surg
19. Masumoto K, Suita S, Nada O, et al. Abnormalities of enteric neurons, intestinal pacemaker cells, and smooth muscle in human intestinal atresia. J Pediatr Surg
20. Melek M, Edirne YE. Two cases of duodenal obstruction due to a congenital web. World J Gastroenterol
21. Ando H, Kaneko K, Ito F, et al. Embryogenesis of pancreaticobiliary maljuction inferred from development of duodenal atresia. J Hepatobiliary Pancrear Surg
22. Boyden EA, Cope JG, Bill AH. Anatomy and embryology of congenital intrinsic obstruction of the duodenum. Am J Surg
23. Tepas JJ, Wyllie RG, Shermeta DW, et al. Comparison of histochemical studies of intestinal atresia in the human newborn and fetal lamb. J Pediatr Surg
24. Masumoto K, Akiyoshi J, Nagata K, et al. Chronological change in intramural components in severe proximally dilated jejuna atresia: an immunohistochemical study. J Pediatr Gastroenterol Nutr
25. Harndy MH, Man DWK, Bain D, et al. Histochemical changes in intestinal atresia and its implications on surgical managements: a preliminary report. J Pediatr Surg
26. Pickard LR, Santoro S, Wyllie RG, et al. Histochemical studies of experimental fetal intestinal obstruction. J Pediatr Surg
27. Masumoto K, Nada O, Suita S, et al. The formation of the chick ileal muscle layers as revealed by alpha-smooth muscle actin immunohistochemistry. Anat Embryol
28. Baglaj SM, Czernik J, Kuryszko J, et al. Natural history of experimental intestinal atresia: morphological and ultrastructural study. J Pediatr Surg
29. Jones MP, Bratten JR. Small intestinal motility. Curr Opin Gastroenterol
30. Lee HT, Hennig GW, Fleming NW, et al. The mechanism and spread of pacemaker activity through myenteric interstitial cells of Cajal in human small intestine. Gastroenterology
31. Lee HT, Hennig GW, Fleming NW, et al. Septal interstitial cells of Cajal conduct pacemaker activity to excite muscle bundles in human jejunum. Gastroenterology
32. Sutcliffe J, King SK, Clarke MC, et al. Reduced distribution of pacemaking cells in dilated colon. Pediatr Surg Int