Intractable Constipation With a Decrease in
Substance P-lmmunoreactive Fibres: Is It a Variant of Intestinal
Neuronal Dysplasia?
By J.M. Hutson, C.W. Chow, and J, Borg
Melboume, Australia
 

- After Hirschsprung's disease was ruled out for 25 children who had severe chronic constipation, the authors studied the distribution of immunoreactivity for substance P (SP) and vasoactive intestinal peptide (VIP) in the intestinal wall, using immunofluorescence. SP and VIP immunoreactive. ity identify excitatory and inhibitory nerve fibres, respectively.  Full-thickness rectal biopsy specimens were unsatisfactory, so seromuscular biopsies of the caecum, transverse colon, and sigmoid colon were obtained (by laparoscopy and laparotomy; n = 10 patients).  SP-immunoreactive fibres were markedly reduced in seven, with concomitant reduction of VIP-immunoreactive fibres in four.  In two other patients, there was no obvious reduction in SP- or VIP-immunoreactive fibres.  In a patient who subsequently was found to have multiple endocrine neoplasia type 2b, the myenteric plexus was markedly hyperplastic, with an increase in nerve cells and nerve fibres.  VJP-immunoreactive fibres were increased, but SP-immunoreactive fibres were markedly decreased. Surgical options included proximal stoma, Matone operation, and subtotal colectomy with preservation of the rectum. Three children with subtotal colectomy have had improvement over short-term follow-up.  The combination of seromuscular laparoscopic biopsies and immunofluorescence demonstration of neuropeptides may identify new variants of intestinal neuronal dysplasia than can be treated successfully with surgery.
Copyright 0 1996 by W 8.  Saunders Company

INDEX WORDS: Intestinal neuronal dysplasia, constipation, laparoscopy, substance P, vasoactive intestinal peptide.

CHRONIC CONSTIPATION rarely is so severe that multiple hospital admissions are required for bowel washouts.  For such a presentation, Hirschsprung's disease usually is suspected, but if biopsies cannot confirm aganglionosis, the patient may be left with no diagnosis and often unsatisfactory or irrational therapy.  Secondary psychological disturbance may occur because of frustration and the disruption to family life.
Because children with this condition have profound functional obstruction, we hypothesised that
 

new enteric-nerve staining methods may help to identify patients who have abnormal innervation.  The excitatory and inhibitory nerve fibres within colonic muscles can be identified with antibodies against

substance P (SP) and vasoactive intestinal peptide (VIP), respectively.'
 

MATERIALS AND METHODS

Immunoreactivity for SP and VIP was studied in colonic specimens obtained from 25 patients with severe functional obstruction of the colon, for whom the rectal biopsy results were negative for Hirschsprung's disease.  In the first 15 patients, full-thickness rectal biopsy specimens were obtained but found to be inadequate for neuropeptide analysis.  The last 10 patients, who are described herein, had a laparoscopic scromuscular biopsy of the caecum, transverse colon, and sigrnoid colon.  Children with constipation managed successfully without washouts were not included in the study,
The controls were colectomy specimens from a 6-year-old girl who had familial adenomatous polyposis coli (FAPC) and from several patients with Hirschsprung's disease.  The FAPC colon showed numerous small mucosal polyps, but the muscularis propria and myenteric plexus were normal.  Colonic specimens from the patients with Hirschsprung's disease were from proximal areas that had no histological abnormality These patients had no functional impairment after surgery.
Fresh tissue was placed in Zamboni fixative, at 4'C, for 1 week.  The fixative was removed from the tissue with dimethylsulphoxide, was washed in phosphate-buttered saline (PBS), and was stored in phosphate-buffered sucrose solution at 4'C for at least 24 hours.
The tissue was placed in a 50% solution of freezing embedding medium (OCT) for 1 to 3 hours before being snap-frozen in OCT in liquid nitrogen-cooled isopentane.  Cryostat sections were cut (8 gm each) and mounted on polylysine-coated slides.  The sections were dried overnight at 4'C.
Staining was performed in a humified chamber at room temperature, and all washes were in PBS.  The sections were incubated with 10% normal sheep serum for 30 minutes.  Excess serum was removed, and the sections were incubated overnight with antiserum against the neuropeptides VIP (Accurate Chemical & Scientific Corp, Westbury, NY) and SF (Auspep Pty Ltd, Parkville, Australia),
The sections were washed and then incubated for 2 hours with Fluorescein (DTAF)-conjugated AffiniPure donkey antirabbit IgG (Jackson ImmunoRescarch @bortories, West Grove, PA).  The sections were washed, stained with 1% Chicago Blue for 1 minute, washed again, and mounted in buffered glycerol (pH 8.6).
The children who had abnormal innervation were given four treatment options: explanation alone, proximal diverting stoma (temporary or permanent), Malone antegrade catheterizable enema (ACE) procedure, and subtotal colectomy with preservation of the rectum (with or without previous stoma).
 
 
 

 RESULTS

In the control specimens, immunofluoresence demonstration of VIP showed clearly a defined myenteric plexus and numerous nerve fibres among the circular fibres of the muscularis propria (Fig1A). Nerve fibres were less prominent and more variable in the longitudinal muscle. With SP, the myenteric plexus also was well demonstrated. The pattern of SP-immunoreactive nerve fibres was similar to that of VIP, but the fibres were finer and less numerous (Fig 1B). Because there was some variation in fibre numbers from the different regions of the colon, only moderate to marked changes in the number or pattern of nerve fibres were considered abnormal.
Of the 10 patients from whom laproscopic seromuscular biopsy specimens were obtained (Table1.) a moderate to marked decrease in SP fibres was noted in seven (Fig 1C.), although the myenteric plexus

Fig 1 lmmunofluorescence demonstration of neuropeptides in niiiscula,is propria of the colon, circular fibres at left, myenteric plexus and longitudinal fibres at right, (A, B) Control FAPC patient @A) VIP. (B) SP (C DJ Child with chronic constipation with marked eduction in @C@ VIP- and (0) SP-associated fibres (E, F) Child with MEN2b with rtiarked hyperplasia of the rnyenteric plexus, (E) an increase in VIP ininiiinoreactive fibres, and (F) absence of SP-imniunoreactive fibres (Original miagnification,* 128
 
 
 
 
 
 

Table 1. Summary of Clinical and Pathological Features

remained well defined.  VIP-immunoreactive fibres were reduced in four (Fig ID) and were normal in three.  In two of the other patients, there was no conspicuous abnormality in VIP or SP staining, although the nerve fibres were moderately thickened in one patient.  In the last patient, there was a variable and often extreme increase in nerve cells and nerve fibres in the myenteric plexus, with focal irregular extension between the muscle fibres.  VIP-immunoreactive fibres were moderately increased (Fig 1E).  However, there were only a few SP-immunoreactive fibres among the muscle fibres (Fig IF).  This patient later had typical features of multiple endocrine neoplasia type 2b (MEN 2b), including mucosal neuromas and corneal nerve fibres.  Molecular studies showed a defect in the RET gene.  Sections stained with hematoxylin and eosin showed a normal number of neurons in the myenteric plexus in all cases except the patient with MEN2b, in whom the plexus had a marked increase in neurons and nerve fibres, with the pattern of ganglioneuromatosis.
All 10 children had severe chronic constipation despite careful exclusion of Hirschsprung's disease.  Most also had multiple admissions for bowel washouts, and some had had empirical surgery previously.  Three patients have undergone subtotal colectomy with preservation of the lower half of the rectum, and one has had a Malone operation.  The other six have declined further treatment or are awaiting surgery.  Although the follow-up period is too short to fully assess outcome, there has been considerable improvement for the three patients who had subtotal colectomy (follow-up of 6 to 18 months); they have spontaneous bowel actions and have had minimal medical therapy.

DISCUSSION

Children with severe chronic constipation usually undergo a rectal biopsy.  The exclusion of Hirschsprung's disease by histochemistry and paraffin sections poses a problem.  What disease does the child

have, and what is the appropriate management?  In most cases no obvious diagnosis is made, and the patient has prolonged medical management that can be extremely distressing to the patient and the family.
Intestinal neuronal dysplasia (IND) has been proposed as the diagnosis in many such cases.2,3 UnfortUnately, the entity of IND remains highly controversial.4,5 The histological features of IND can occur alone or in association with various entities such as Hirschsprung's disease, neurofibromatosis, ganglioneuromatoSiS,6 or even mechanical lesions such as volvulus and intussusception.7 Moreover, even in patients with symptoms, these often improve with age." Thus, there is great reluctance in some centres to use this designation as a specific disease entity with a specific form of management.  But the term is very popular with other groups.9 If a plan of treatment that includes surgery is offered, it is more satisfactory to have a specific diagnosis.  It is in this context that "IND" is used herein, ie, an all-embracing term for cases of chronic constipation for which Hirschsprung's disease has been excluded and for which there is a suggestion of abnormal innervation.
Assessment by semiquantiatative or quantitative conventional histology has been unsatisfactory, prompting study with modem histochemical or immunohistochemical methods.  In recent years, at least 30 transmitters have been found in the enteric nervous system.  In humans, VIP is a transmitter of inhibitory fibres, and SP is one of excitatory fibres.' There have been relatively few studies on the distribution of neuropeptides in the colon in children.  Abnormalities have been noted in patients with Hirschsprung's disease, but little has been written with respect to chronic constipation.  Initially it was hoped that rectal biopsy specimens would be suitable for this study.  However, the specimens often are too shallow or from too low a site.  The mucosal and submucosal fibres are involved with secretion rather than motility.  Also, around the internal sphincter, SP-immunoreactive fibres normally are very rare or are absent."'
 

INTESTINAL NEURONAL DYSPLASIA

Therefore, the seromuscular biopsy specimens were taken via laparoscopy, which provided satisfactory samples of the longitudinal and circular fibres and the myenteric plexus.
Specimens from patients with FAPC and Hirschsprung's disease are not optimal controls but were the only ones available of the colon.  Specimens from mechanical conditions such as volvulus and intussusception have been found to be unsuitable, probably because of the irregular decrease in neuropeptides that is associated with ischaemia.
Various immunohistochemical techniques have been used recently to identify IND and Hirschsprung's disease.11-14 VIP and SP were shown to be absent in the aganglionic segments.11,14 Monocional antibodies against neurofilament, which is a structural protein in axons, also were not effective in identifying some cases with IND or its variants.13 Puri and FujiMotOI2 raised specific monoclonal antibodies against glial fibrillary acidic protein, and were able to distinguish cases that had some forms of IND.
The major abnormality noted in the present study was the marked reduction in SP-immunoreactive fibres in the muscularis propria in a large number of children with severe chronic constipation.  Interestingly, even in the patient with MEN 2b, who showed

marked hyperplasia of the myenteric plexus, SP fibres were virtually absent even though they were markedly increased in the mucosa in the subsequent colectomy specimen.  SP-immunoreactive fibres were not similarly reduced in a variety of other conditions including post-necrotising enterocolitis stricture and colostomy (own data).  It is tempting to propose that the reduction of SP-immunoreactive fibres is the basis of the functional impairment.  SP has been shown to be reduced in Hirschsprung's disease, but the cause of this reduction is not clear.  It may be developmental or the result of earlier disease.  It is unlikely to be a result of long-term treatment, because some of the younger patients who had marked fibre reduction had only had short periods of medical therapy.  All these possibilities should be clarified by a long-term follow-up study.  Short term, it appears that subtotal colectomy has overcome the symptoms.9
In conclusion, we report a novel way to investigate cases of severe chronic constipation, and preliminary results suggest an abnormality of innervation of the muscle coat, with favourable outcome after surgery.

ACKNOWLEDGMENT

The authors thank Professor John Furness for his generous help and advice during the study.
 
 
 
 

Annotation

Deficiency of substance P-immunoreactive nerve fibres in children with intractable constipation:
A form of intestinal neuronal dysplasia8

JM HUTSON, CW CHOW, MR HURLEY, S UEMUR, JM WHEATLEY and
AG CATTO-SMITH

F Douglas Stephens Surgical Research Unit and Departments of Anatomical
Pathology and Gastrenterology, Royal Childrens Hospital, Parville,
Victoria
 

Chronic constipation is common in childhood and usually responds to dietary manipulation, attention to toileting and laxative treatment.  Most paediatricians in practice have seen numerous such patients, and will have developed management approaches to both the physical problem and to the family dynamics.  Very occasionally, they will be confronted with a child with intractable constipation that is resistant to all of the usued treatment measures.  In this case, Hirschsprung's disease needs to be considered and excluded by a rectal suction biopsy, even though it would be rare to have such a late presentation.
What can be done when the rectal biopsy has excluded Hirschsprung's disease but the symptoms are refractory to treatment?  Many paediatricians would try prescribing regular enemas administered either at home or in hospital.  Attematively, the child might be hospitalized for bowel irrigation with polyethylene glycol (Golyte@, Stafford Muller, Australia) or a similar agent given orally or via a nasogastric tube.  Sooner or later the doctor may wonder whether the parent or child was sabotaging the treatment is this a case of Munchausen syndrome or Munchausen by proxy?  Once the possibility of psychogenic problems is considered, a referral to a psychiatrist may be arranged.  Failure to identify an organic disease tends to push both doctor and parents towards psychiatric explanations, and everyone becomes progressively more frustrated.
Recent evidence suggests that a proportion of children with refractory constipation have an abnormality of intestinal innervation resulting in dysfunctional colonic motility.  This has been termed intestinal neuronal dysplasia (IND) by investigators Meier-Ruge et a/.' and Schdrii.2
The existence of this entity remains controversial as standard histological and clinical criteria are not yet available.' The most common variant of IND, described primarily in European literature, is type B (Table 1).  It is defined on rectal biopsy by the finding of increased numbers of ganglion cells in the submucosal plexus.' Histochemical stains for succinate
 

Correspondence: Professor JM Hutson, General Surgery, Royal Children's Hospital, Parkville Vic. 3052, Australia.
JM Hutson, FRACS, Professor and Head, Department of Surgery.  CW Chow, FRCPA, Pathologist MR Hurley, BMedSci student S Uemura, MD, Research Fellow.  JM Wheatley FRACS, Research Fellow.  AG Catto_ Smith, FRACP, Gastroenterologist

Accepted for publication 30 September 1996.
 
 

Total intestinal

Uftra-short segment

Hypogenests
Hypoganglionosis
Neuronal hypogenesis
Dysgenesis
IND type A (deficient sympathetic innervation)
IND type B Increased number of small ganglion cells)

AcQuired
Chagas disease

Combined forms

IND B With Hirschsprung's disease or hypoganglionosis or neuronal hypogenesis

dehydrogenase, lactate dehydrogenase and acetylcholinesterase have been used to quantify nerve cells in the ganglia.  In IND type B there are twice as many nerve ceits per ganglion as in normal children (approximately 10 cells per ganglion vs four cells per ganglion).  Individual nerve cells are also reduced in size (20% smalleo when compared with normal, mature ganglion cells.  The functional implications of these morphological differences, however, remain unknown.
In our own laboratory, we have identified a group of children with IND using different criteria." Following the exclusion of Hirschsprung's disease on rectal mucosal biopsy, we then proceeded to ootain seromuscular biopsies of the colon at laparoscopy.
It was felt that direct sampling of the colonic muscle would yield more information about colonic motility than would rectal mucosal biopsies. lmmunofluorescence was then performed, evaluating the tissue for the presence of neurotransmitters associated with excitatory nerves (Substance P; SP) and inhibitory nerves (vasoactive intestinal peptide; VIP).  These represent functional markers for colonic contraction (SP) and relaxation (VIP) necessary for forward propulsion.s Two variants of IND were seen: one group of children with an isolated deficiency of SP-labelling in the colonic nerves, and a second

group with deficient staining for both SP and VIP.  Preliminary results of double labelling with a structural axonal marker (neurone specific enolase) and either one of the functional markers, suggests that the number of axonal processes is normal, despite the lack of neuropeptide transmitter (MR Hurley and S Uemura, unpubl. data, 1996).
This new diagnosis of 'IND with SP deficiency' has coped some clinical problems, but raised numerous questions.  It has reassured many families (and doctors) that their child is suffering from an organic problem, thereby relieving a lot of frustration and uncertainty.  It has also allowed a more rational approach to treatment, introducing surgical options which probably would not have been considered otherwise.  These have included procedures to reduce colonic length, create a stoma, or facilitate colonic irrigation; all with the aim of reducing colonic transit time.  Subtotal colectomy with retention of the rectum has been tried with some success in a small number of patients.  This is considered where a distinct abnormal bowel segment can be identified by neuropeptide staining.  A proximal colostomy has been fashioned in a subgroup with more extensive colonic dysmotility.  Although both of these options reduce colonic 'length' and potentially colonic transit time, they differ in the matter of preserving the anorectum for defaecation.  In some children, colonic irrigation has been facilitated by the creation of a catheterizable stoma via the appendix ('Malone' operations Again, this has been quite successful to date, but only tried in a small number of children.  After this procedure, most parents and children are able to manage their washouts at home rather than being admitted to hospital.  The small stoma needs no bag and is less disfiguring than a colostomy.  Both the volume of fluid needed and the time taken to complete the washout are reduced as the fluid is placed directly into the caecum.  The long-term outcome of these procedures will need careful evaluation.  Each carries the potential for significant morbidity.  This must be weighed against the significant physical and psychological morbidity associated with intractable constipation and its therapy already being endured by these children.  This is all the more important in view of the increasing numbers of children being referred for these investigations.
How does IND with SP deficiency relate to IND-B as diagnosed by suction rectal biopsy?  Are they separate forms of IND or merely the same disease identified differently'.? At present there are no answers to these questions.  Further, is SP deficiency congenital or acquired?  Is it an isolated anomaly or related to other abnormalities in the enteric or central nervous system?  One patient with SP deficiency had multiple endocrine neoplasia type 11 B with a recognizable anomaly in the RET gene.  As RET mutations are also reported in some 20% of patients with Hirschsprung's diseasdi perhaps IND patients should be screened for similar mutations.
Despite these uncertainties, the provision of a diagnostic label in children with intractable constipation has freed families from the tyranny of medical ignorance and given them new hope and enthusiasm for coping with their problem.  Both a parent support group and a support group for the children themselves have been started with the aim of raising public and medical awareness of their plight.
How should paediatricians respond to this new information?  Should they consider such a diagnosis in their own patients with chronic constipation?  Since laparoscopic biopsy (or laparotomy) is required, careful selection is crucial.  Until simpler modes of identification are available, one should refer only those children with chronic, intractable constipation which has
 
 
 
 

Table2 Indications for performing colonic biopsy and neurotransmitter studies

been resistant to standard aggressive therapy.  In our own group, about 73% of selected children were found to have SP deficiency when we included those who had repeated enemas at home or hospital, admissions for colonic irrigation and a previously normal rectal biopsy (Table 2).  It is important to recognize the possibility of missing a short segment of aganglionosis, such that repeat rectal biopsies may be needed, particularly in those children with persistent constipation since birth.  Additionally if severe constipation persists despite intensive treatment, these children likewise need to be re-evaluated.
What does the future hold in this area?  Our knowledge of gut motility and of the enteric nervous system continues to expand.  One of the major unanswered questions at this point is how intestinal neuronal dysplasia, as defined by substance P neurotransmitter deficiency, relates to ]NO defined on rectal biopsy in other centres. in addition to further study of neurotransmitter function, there are other abnormalities reported in patients with intestinal dysmotility disorders.  These are worth considering in patients with possible intestinal neuronal dyspiasia, and include a decrease in the neural cell adhesion molecules, synaptophysin and growth-associated protein-43,8 and an abnormal increase in calcitonin gene-related peptide (CGRP) immunoreactive fibres"- If sufficient tissue is available for study, electron microscopy can be performed, looking for such rare entities as mitochondrial disease" and neuronal inclusion disease" which have been implicated in some instances of intestinal pseudo-obstruction.