|Year : 2017 | Volume
| Issue : 4 | Page : 187-192
Role of synaptophysin in the intraoperative assessment of quadrantic innervation of the proximal doughnut in Hirschsprung disease
Lokendra Yadav1, MK Babu1, Kanishka Das2, Suravi Mohanty1, P Divya1, Gowri Shankar1, Usha Kini1
1 Department of Pathology, St John’s Medical College and Hospital, St John’s National Academy of Health Sciences, Bengaluru 560034, Karnataka, India
2 Department of Paediatric Surgery, St John’s Medical College and Hospital, St John’s National Academy of Health Sciences, Bengaluru 560034, Karnataka, India
|Date of Web Publication||17-Nov-2017|
Department of Pathology, St John’s Medical College and Hospital, St John’s National Academy of Health Sciences, Bengaluru 560034, Karnataka
Source of Support: None, Conflict of Interest: None
Background. Symptoms may persist in a retained aganglionic segment of the colon after corrective (pull-through) surgery in Hirschsprung disease (HD). Thus, it is important to assess the proximal doughnut for innervation abnormalities intraoperatively by frozen sections stained with conventional haematoxylin and eosin stain and supported by rapid acetylcholinesterase (AChE) histochemistry. When the doughnut is proximal to the sigmoid colon, AChE is not useful and requires ratification by yet another rapid technique and hence this study.
Methods. Two pathologists independently evaluated fresh doughnuts from the proximal bowel clinically assumed to be of normal innervation intraoperatively and chosen for anastomosis in patients with HD along with controls using AChE and synaptophysin (SY) immunohistochemistry.
Results. From 38 patients with HD, 28 doughnuts (63.7%) showed normal innervation with intense SY activity in the mucosa, the muscularis and the ganglion cells. The circumferential aganglionic doughnuts (abnormal innervation) (n= 6, 13.6%) showed neither SY-positive fibres in the mucosa nor in the muscularis. The abnormal transition zone doughnuts (n=10, 22.7%) showed involvement of three quadrants of the doughnut in one, two quadrants in three and one quadrant in six with decreased SY-positive fibres in the muscularis and scattered ganglion cells with a statistically significant measure of agreement of (κ=0.973) between the two.
Conclusion. The pattern, intensity and distribution of SY-positive fibres in the muscularis propria of the doughnut of the proximal bowel chosen intraoperatively for anastomosis in HD can identify sectors with abnormal innervation allowing the surgeon to seek normal innervation status more proximally to avoid complications.
|How to cite this article:|
Yadav L, Babu M K, Das K, Mohanty S, Divya P, Shankar G, Kini U. Role of synaptophysin in the intraoperative assessment of quadrantic innervation of the proximal doughnut in Hirschsprung disease. Natl Med J India 2017;30:187-92
|How to cite this URL:|
Yadav L, Babu M K, Das K, Mohanty S, Divya P, Shankar G, Kini U. Role of synaptophysin in the intraoperative assessment of quadrantic innervation of the proximal doughnut in Hirschsprung disease. Natl Med J India [serial online] 2017 [cited 2018 Dec 19];30:187-92. Available from: http://www.nmji.in/text.asp?2017/30/4/187/218669
| Introduction|| |
Hirschsprung disease (HD) presents with neonatal intestinal obstruction or childhood constipation and failure to thrive. The histological diagnosis of HD rests on the absence of ganglion cells, presence of hypertrophy–hyperplasia of parasympathetic nerve bundles and increased acetylcholine esterase (AChE) activity on frozen rectal biopsies.,,,, The successful siting of a temporizing colostomy/ileostomy or a definitive pull-through surgery is dependent upon precise identification of the proximal extent of the aganglionic colon or the distal extent of normally innervated colon (proximal doughnut).7 An inaccurate definition of the histological transition zone with a retained aganglionic segment of bowel or quadrantic discrepancies in circumferential innervation in the level of bowel used for stoma/pull-through results in persistence of symptoms.
Haematoxylin and eosin (H&E) stain and AChE histochemistry have been used on one-point seromuscular biopsies and on circumferential ring bowel biopsies (doughnut) for rapid intraoperative analysis and histological levelling. As typical AChE-positive neural patterns are not well expressed in tissues proximal to the sigmoid colon, a search is on for an alternative intraoperative marker to assess neural innervation status.
Synaptophysin (SY), a 38-KD membrane protein specific for the synaptic vesicles in the central and peripheral nervous system, is the main constituent of AChE storage compartments, and an important neuromuscular junction marker.,,,,,, Its use as a marker for cholinergic structures and for normal neuromuscular junction has been extrapolated in this study on the neural innervation of doughnuts from the proximal end of the aganglionic segment (proximal doughnut) during histological levelling of HD by estimating the difference in the density of synapses between abnormally innervated (transitional zone/aganglionic segment) and normally innervated ganglionated doughnuts.
| Methods|| |
We conducted this prospective cross-sectional study at a tertiary referral hospital for HD and allied disorders over a 5-year period from 2011 to 2015. A suspicion of HD on clinical findings and contrast enema was confirmed on a conventional rectal biopsy using improvised rapid AChE histochemistry., After confirmation of rectal aganglionosis, proximal seromuscular point biopsies (depending on the radiological and gross transition zones) were obtained till a normal innervation was reported; at this juncture a doughnut of bowel was sent intraoperatively for quadrantic evaluation and final levelling by AChE histochemistry and SY immunohistochemistry (IHC). This study was approved by the institutional ethics board.
Doughnuts from proximal ganglionated segment were considered as control doughnuts. However, doughnuts from resections for habitual constipation, discontinuous aganglionosis and those where AChE histochemistry was not done were excluded.
The doughnuts sent for evaluation were received well soaked in 0.9% normal saline. They were cut into four quadrants (when big)/two (if small) and retained as one when the luminal diameter was small. They were marked and directly fresh frozen at –20 °C in the cryostat; 10 μ cryostat sections were cut from each sector and were concurrently stained with rapid H&E for routine morphology, modified rapid AChE staining technique with rubeanic acid for enzyme histochemistry and IHC for SY marker. The frozen tissue remains were processed thereafter, for paraffin sections.
All the stained sections were labelled, coded and evaluated independently by two pathologists (MKB, UK).
Assessment by H&E
The sections were studied to assess the regular distribution of ganglia in the myenteric plexus in all four quadrants (quadrantic) and for presence of hypertrophic nerve bundles.
A control doughnut was considered to have normal quadrantic innervation if the ganglion cells were regularly arranged equidistant from one another best appreciated in the myenteric neural plexuses and showed no hypertrophic nerve bundles in either myenteric or submucosal plexus. A doughnut was considered abnormal when it exhibited irregular ganglion distribution (either hypoganglionic/ aganglionic) in myenteric plexus involving one or more quadrants along with hypertrophic nerve bundles in either myenteric or submucosal plexuses. Decrease in the number of ganglion cells constitutes a stretch of myenteric plexus that contain only occasional ganglion cells surrounded by minimal neuropil and is synonymous with the descriptive terminology of ‘hypoganglionosis’. The two terminologies, namely ‘irregular distribution of ganglion cells’ and ‘decrease in the number of ganglion cells’, used for abnormal innervation often go hand in hand.
The transition zone between histologically normoganglionic segment and aganglionic bowel, was characterized by hypoganglionosis/aganglionosis around part of the circumference of the bowel seen along with hypertrophic nerve bundles. Aganglionosis, when present in all four quadrants (entire circumference), indicated aganglionic segment and not transition zone.
The innervation status was assessed using routine H&E stain, AChE histochemistry and SY IHC on the doughnuts.
Assessment by AChE histochemistry
The findings on H&E were ratified using AChE histochemistry by the modified rapid improvised AChE staining technique of Kini et al.,, for those tissues from sigmoid colon and distal to it by the demonstration of hypertrophic nerve bundles in the submucosa and ganglion cells in the neural plexuses. The mucosal staining patterns if identified namely A and B when AChE activity was increased and equivocal pattern when the activity was not increased were noted. They are detailed as follows:
Positive pattern A: The presence of AChE-positive nerve fibres in the interglandular spaces throughout the thickness of the mucosa, indicate aganglionosis and hence abnormal innervation.
Positive pattern B: The presence of AChE-positive nerve fibres in the muscularis mucosa and the lowermost portion of the lamina propria below the region of the basal crypts, also indicate aganglionosis and hence abnormal innervation.
Equivocal pattern: Irregular focal presence of nerve fibres in the mucosa, muscularis mucosa or submucosa with no increase in AChE activity, indicate normal innervation.
Negative pattern: Absence of stained nerve fibres in the mucosa with or without occasional obscure nerve twigs in the muscularis mucosa and submucosa calls for repeat staining.
Assessment by SY IHC
IHC staining, was done on frozen sections of 10 μ taken on poly-L-lysine (PLL) coated slides and fixed for 10 minutes in 10% formalin. They were brought to phosphate buffer saline (PBS) and incubated with rabbit polyclonal anti-SY antibody (Clone AP10443 of Genova Scientific) for 20 minutes at 37 °C. After a quick PBS wash, the corresponding secondary antibody added was left for 30 minutes; development of peroxidase was achieved by freshly prepared 3-3 ' diaminobenzidine tetrahydrochloride (DAB, DAKO REALTM EnVision™ Detection System) in a dilution of 20 μl in 1 ml of DAKO REALTM substrate buffer containing hydrogen peroxidase for 2 minutes. The total time of the staining procedure was 60 minutes.
As the study was done during reporting of frozen sections and as morphometric analyses is time-consuming and not practical, an approximate density of synapses was assessed subjectively and scored as 0 to 2 (0=absent, 1=reduced and 2=numerous).
The approximate density of synapses at 10× were assessed subjectively in the inner and outer muscularis propria, and scored on a three-tier system of 0 to 2 (0=absent, 1=reduced and 2=abundant). Staining of neuroendocrine cells in the mucosa was considered as internal control. Density of synapses seen in the control doughnut was read as 2. When the density of synapses in test doughnut was less than half of the lowest seen in control doughnut, it was scored as 1=reduced.
[Table 1] shows the comparison between the staining patterns with H&E stain, AChE histochemistry and SY IHC on ganglionic, transition and aganglionic sections.
|Table 1: Comparison between the staining patterns with H&E stain, AChE histochemistry and SY immunohistochemistry on ganglionic, transition and aganglionic doughnuts on frozen sections|
Click here to view
Statistical analyses were done using SPSS software version 20. The Student t-test, and Fisher’s exact tests were used for comparison between the two groups. The correlation between parameters was done using by Spearman’s rank coefficient of correlation. Cohen’s Kappa was used to analyse the agreement between the two groups. Values of p<0.05 were considered significant.
| Results|| |
Thirty-eight patients of HD confirmed with AChE histochemistry on rectal biopsy (4 term neonates, 20 infants and 14 older children; 30 males and 8 females, M:F: 3.75:1) were included in the study. The contrast enema was suggestive of HD in 26 and inconclusive in 12 [Table 2]. Two patients, a neonate and a two-year-old, had presented acutely with ileal perforation and peritonitis and the rest presented with chronic constipation. Thirty-six (95%) had a history of delayed passage of meconium (24–48 hours in 22, beyond 48 hours in 14). One child had trisomy 21 and none had culture proven sepsis.
Intraoperative histological and histochemical assessment
The levelling biopsies done before doughnut examination classified the cases as recto-sigmoid HD (RSHD) in 28, long segment colonic HD (LSHD) in 7 and total colonic aganglionosis (TCA) in 3. Barium enema images in recto-sigmoid HD assisted in identifying the transition zone. Mapping seromuscular (levelling) biopsies taken from the right and left colon and appendix/terminal ileum confirmed the diagnosis of LSHD and TCA.
Forty-four doughnuts were received for quadrantic innervation studies from 38 patients of confirmed HD; 28 from 28 patients were from the sigmoid and descending colon and showed normal innervation; 10 from 7 patients were from transverse colon and 6 from 3 patients were from ileum, respectively and were grouped as having abnormal innervation.
A minimum of four complete H&E stained frozen sections were studied from both controls and tests on each of the doughnuts along with AChE stained sections and SY immunostains. By the time AChE histochemical stain was studied along with H&E stain, the sections stained with SY immunostain were ready (in 60 minutes), which was comparable to the time taken for AChE enzyme histochemistry. As SY specifically stains ganglion cells in addition to synapses in the muscle layers irrespective of orientation of the doughnut, it is a useful immunostain for reporting frozen sections.
Normal innervation (n=15 control doughnuts, 28 test doughnuts [63.6%]) H&E stain
Twenty-eight test doughnuts from RSHD that showed normal innervation were characterized by regularly spaced myenteric ganglia with distinct ganglion cells [Figure 1]a and no hypertrophic nerve bundles.
|Figure 1: Normally innervated doughnut: Note (a) ganglia regularly placed in the myenteric plexus (frozen section, H&E stain x100) highlighted by AChE stain in (b). Note SY-positive fibres in the cross-section of muscularis in (c) (x200) and longitudinal section in (d) (x400).|
Click here to view
Equivocal staining pattern was noted with the AChE enzyme histochemistry which ratified the uniform distribution of ganglion cells in the myenteric plexus noted in H&E. Ganglion cells with their cytoplasm stained dark brown to black and staining occasional submucosal nerve fibres/nerve twigs in between the muscle fibres [Figure 1]b. No staining pattern was noted in the submucosa.
Distinct SY immunoreactivity was highlighted in both the layers of muscularis propria in these doughnuts. The ganglion cells showed granular cytoplasmic positivity, thus making them easily identifiable. The spindle shaped SY-positive fibres identified in the muscularis propria were linear, elongated wavy structures located parallel to and between the muscle fibres in longitudinal sections and were numerous, dot-like, best appreciated in cross- sections [Figure 1]c and [Figure 1]d.
The inner circular muscle layers were more densely innervated by the SY fibres than the longitudinal muscle layer. The SY fibres were of better intensity in cross-sections than in longitudinal sections. These fibres were also noted in the muscularis mucosa but were shorter in length. They were seen ascending into the lamina propria between the crypts reaching up to the surface epithelial lining—a pattern reminiscent of the classical AChE pattern seen in aganglionic segment of HD. Distinct neuroendocrine cells located near the base and mid-portion of the crypts were highlighted by SY. The blood vessels with muscle layer showed SY-positive fibres interspersed in the muscle wall. The nerve bundles that accompanied these vessels were also stained positive.
Abnormal innervation [n=16 doughnut from 10 patients (n=36.3%, 7 LSHD, and 3 TCA] H&E stain [Figure 2] and [Figure 3]
|Figure 2: Abnormally innervated doughnut on frozen section shows (a) and (b) myenteric plexus with the cross-section of a hypertrophic nerve bundle (thick arrow) and a small cluster of ganglion cells (hypoganglionic) (thin arrow) in rapid H&E stain (x 100) and with AChE enzyme histochemistry, respectively (x100). Note the reduced number of SY-positive fibres in the cross-section (c) (x200) and longitudinal section in (d) of muscularis propria (x400).|
Click here to view
|Figure 3: Quadrantic distribution of abnormal doughnuts in recto- sigmoid Hirschsprung disease (RSHD), long segment HD (LSHD) and total colonic aganglionosis (TCA)|
Click here to view
Sixteen doughnuts from 10 patients showed abnormal innervation characterized by absence or decrease in the number of ganglion cells and presence of hypertrophic nerve bundles in both the submucosal and myenteric plexuses to involve a variable degree of the circumference of the doughnut [Figure 2]a and were correlated with AChE and SY staining [Figure 2]b,[Figure 2]c,[Figure 2]d. Abnormal innervation involved the entire circumference (all four quadrants) in 6 doughnuts (37.5%, 4 from LSHD and 2 from TCA), three quadrants in 1 doughnut from LSHD 6.25%, two quadrants in 3 doughnuts (18.75%, 2 from LSHD and 1 from TCA) and one quadrant in 6 doughnuts (6.25%, 3 each from LSHD and TCA). One abnormal doughnut showed hypertrophic nerve bundles in the submucosa but regularly distributed ganglion cells in the myenteric plexus in all four quadrants.
The 16 abnormal doughnuts showed hypoganglionosis involving one or more quadrants with hypertrophic nerve bundles and/or aganglionosis of one or more quadrants but not all four and confirmed by AChE wherein they exhibited increased AChE activity in the submucosa by staining the hypertrophic nerve bundles [Figure 2]b. No mucosal staining patterns were noted.
Six of the 16 doughnuts with abnormal innervation characterized by aganglionosis involving one or more but less than four quadrants showed no SY-positive fibres in both the muscle layers of muscularis propria in the affected sectors. The lamina propria in addition showed no SY-positive fibres with occasional identifiable neuroendocrine cells in the mucosa.
The doughnuts with abnormal innervation involving one or more quadrants (transitional zone) showed fewer SY-positive fibres in the muscle layer [Figure 2]c and [Figure 2]d than the doughnuts with normal innervation and stained the occasional scattered ganglion cells. The adjacent normally ganglionated and innervated quadrants showed near normal distribution of SY-positive fibres similar to that noted in the doughnuts with normal innervation.
The only doughnut with submucosal hypertrophic nerve bundles showed SY-positive fibres in both the muscle layers but fewer SY-positive fibres in the lamina propria. The abnormal thick hypertrophic nerve bundles seen in submucosa were SY-positive.
When the first doughnut was abnormal as in 10 patients, the operating surgeon elected to submit a second doughnut (5 cm proximal to the first) for evaluation. Only four second doughnuts from LSHD showed normal innervation whereas 3 LSHD and all 3 TCA required a third doughnut, which was normally innervated in all four quadrants.
The above histological and histochemistry findings on 15 control and 38 test doughnuts were concordant between AChE and SY stains as well as between the two pathologists with 1.000 being the measure of agreement of kappa and p<0.005.
Twenty-eight of 38 patients (74%) who underwent a definitive pull-through after demonstration of normal innervation in the proximal doughnut were followed up for a mean duration of 40 months. They were asymptomatic and thriving well. They had 2–3 normal bowel movements daily with formed stools and no persistent obstructive symptoms. Initially, six of the remaining ten children had constipation which settled with regular anal dilatation. However, five of these had intermittent soiling, especially nocturnal and are on bowel management programme to facilitate scheduled rectal evacuation and social continence.
One among the three children who had a transitional pull- through and was re-operated had presented with abdominal distension and constipation 3 months thereafter. His constipation has improved with botulinum injection. The other two are doing well. One child in the series died due to an unrelated cause a year later.
| Discussion|| |
The assessment of ganglionic innervation by mapping seromuscular biopsies in HD determines the extent of affected bowel and intraoperative biopsy helps to determine precisely the distal limit of normal innervation before a bowel stoma/pull-through surgery is done. The minute biopsy specimens are conventionally examined using H&E staining. In certain circumstances, it is difficult to accurately identify ganglion cells in H&E sections, e.g. transition zone, where it may be hypoganglionic or the ganglion cells may appear morphologically immature and hence not have classical histology. Further, a one point seromuscular (levelling) biopsy may not represent a normal innervation along the entire circumference of the doughnut due to the differential migratory nature of neural crest cells which has a leading edge. These problems are accentuated at the transition zone when one or more of the quadrants show hypertrophic nerve bundles and scattered ganglion cells. We have reported that such transition zones need to be meticulously identified by histochemistry and managed appropriately for optimal results.
This study investigated SY as an immunohistochemical marker for ganglion cells (positive staining) as well as for synapses by the demonstration of SY-positive fibres. We hypothesize that, when there is absence of or decreased number of ganglion cells, SY must be proportionally lesser in number if not absent.
SY is one of the important neuromuscular junction markers. The 38-kD membrane protein is specific for synaptic vesicles in the central and peripheral nervous system and responsible for normal neuromuscular junction and neurotransmission.,, Expression of SY by IHC detects intestinal ganglion cells and helps in mapping the ganglionic bowel and the synapses.
Our suspicion that absence of ganglion cells is accompanied by a decrease/absence of synapses is highlighted in this study. The classical positive findings of AChE seen in HD are AChE- positive nerve pattern in contrast to negative patterns with SY IHC characterized by decreased number of SY-positive fibres at the muscle. Thus, the ganglionated segment showed intense SY immunoreactivity while AChE showed no activity and vice versa in the aganglionated segment. Similar findings were reported by Kobayashi and Dzienis–Koronkiewicz. It is supported by a quantitative estimation of AChE and SY immunoreactivity in mucosal/submucosal and muscle layers of ganglionic and aganglionic intestinal segments in 8 patients of HD by Wiedenmann. Using monoclonal antibody 171B5, which selectively labels synaptic vesicles, Yamataka found a similar neuronal innervation pattern in the mucosa and submucosa in aganglionic rectal specimens: only reduced numbers of unorganized synapses were seen in the lamina propria, none in the muscularis mucosae, and a few in the submucosa. Similarly in both non-HD rectal and colonic specimens, many synapses arranged in neural plexuses were found in the lamina propria, a moderate number in the muscularis mucosae and dense clusters in the submucosal plexus.
SY IHC as a labelling method highlights the morphology of ganglion cells as well as indirectly reflects their functional status by demonstrating synapses at the level of muscle fibres on frozen sections. The pattern and density of SY-positive fibre distribution in circular and longitudinal muscularis layer of colonic biopsies is an additional tool to map the ganglionic–aganglionic interface. The described pattern with SY correlates with the information from concurrent AChE staining. In patients with ambiguous AChE staining, and when doughnuts are not well-oriented, SY IHC staining on frozen sections is a valuable support to study neural networks in the bowel in HD during intraoperative consultation.
| References|| |
Montedonico S, Piotrowska AP, Rolle U, Puri P. Histochemical staining of rectal biopsies as the first investigation with chronic constipation. Pediatr Surg Int
Babu MK, Kını U, Das K, Alladı A, D’Cruz AJ. A modified technique for the diagnosis of Hirschsprung disease from rectal biopsies. Natl Med J India
Kini U, Das K, Babu MK, Mohanty S, Divya P, Saleem KM. Role of rapid modified acetylcholinesterase histochemistry in the diagnosis of Hirschsprung’s disease. Indian J Path Microbiol
Kini U, Babu M, Mohanty S, Divya P, Das K, Nandeesh BN, et al
. Role of rapid improvised acetylcholinesterase histochemistry in the diagnosis of Hirchsprung’s disease: A 14-year prospective study of 1890 rectal biopsies at a referral center in India. Mod Pathol
Yadav L, Kini U, Das K, Mohanty S, Puttegowda D. Calretinin immunohistochemistry versus improvised rapid acetylcholinesterase histochemistry in the evaluation of colorectal biopsies for Hirschsprung disease. Indian J Pathol Microbiol
Amiel J, Sproat-Emison E, Garcia-Barcelo M, Lantieri F, Burzynski G, Borrego S, et al
. Hirschsprung disease, associated syndromes and genetics: A review. J Med Genet
Haricharan RN, Georgeson KE. Hirschsprung disease. Semin Pediatr Surg
Kapur RP, Kennedy AJ. Transitional zone pull through: Surgical pathology considerations. Semin Pediatr Surg
Patt S, Stoltenburg-Didinger G. Immunohistochemical studies using synaptophysin in intestinal biopsies in Hirschsprung’s disease [article in German]. Zentralbl pathol
Kobayashi H, Hirakawa H, Puri P. Is intestinal neuronal dysplasia a disorder of the neuromuscular junction? J Pediatr Surg
Munakata K, Tomita R, Kutosu Y. Preliminary immunohistochemical new findings in the myenteric plexus of patients with intestinal neuronal dysplasia type B. Eur J Pediatr Surg
Wiedenmann B, Riedel C, John M, Ahnert-Hilger G, Stoltenburg G, Waldschmidt J, et al
. Qualitative and quantitative analysis of synapses in Hirschsprung’s disease. Pediatr Surg Int
Wiedenmann B, Franke WW. Identification and localization of synaptophysin, an integral membrane glycoprotein of Mr 38000 characteristic of presyanaptic vesicles. Cell
Kobayashi H, Li Z, Yamataka A, Lane GJ, Miyano T. Overexpression of neural cell adhesion molecule (NCAM) antigens on intestinal smooth muscles in hypoganglionosis: Is hypoganglionosis a disorder of the neuromuscular junction? Pediatr Surg Int
Kobayashi H, Miyano T, Yamataka A, Lane GJ, Fujimoto T, Puri P. Use of synaptophysin polyclonal antibody for the rapid intraoperative immunohistochemical evaluation of functional bowel dosorders. J Pediatr Surg
Barshack I, Fridman E, Goldberg I, Chowers Y, Kopolovic J. The loss of calretinin expression indicates aganglionosis in Hirschsprung’s disease. J Clin Pathol
Das K, Kını U, Babu MK, Mohanty S, D’Cruz AJ. The distal level of normally innervated bowel in long segment colonic Hirschsprung’s disease. Pediatr Surg Int
Dzienis-Koronkiewicz E, Debek W, Chyczewski L. Use of synaptophysin immunohistochemistry in intestinal motility disorders. Eur J Pediatr Surg
Yamataka A, Miyano T, Kimura K, Arai T, Nishiye H, Sueyoshi N. Innervation of aganglionic intestines of Hirschsprung’s disease examined by mucosal antibody 171B5. Lancet
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2]