Does the Distal Rectal Muscle in Anorectal Malformations Functional Properties of a Sphincter? By Henrik Hedlund and Albert0 New Hyde Park, New 0 Smooth muscle strips from the distal rectum of 11 cloaca1malformations, were studied in vitro to assess the motility response to electrical field stimulations (EFS) and to pharmacological stimulation with adrenergic and cholinergic agonists. EFS induced a nonadrenergic. noncholinergic inhibition in most strips. Acetylcholine caused either a modest contraction, no response, or a relaxation. Following atropine administration, acetylcholine caused a nonadrenergic and tetrodotoxin-resistant relaxation. The a-adrenergic agonist phenylephrine induced contractions in all strips. The response was abolished by a-adrenoceptar blockade with phentolamine.
but was resistant to
atropine and tetrodotoxin. @-Adrenergic stimulation caused a relaxation that was abolished by propranolol. Function of the distal rectal smooth muscle, resected during correction of anorectal malformations, shows similarities to the function reported previously on normal anal smooth muscle evaluated in vitro. @ 1990 by W.9. Saunders
Company.
INDEX WORDS: lmperforate anus: anal sphincter function.
A
NAL RESTING TONE is regarded as an important factor for the maintenance of anal continence. Diminished anal tone that is usually seen after treatment for high anorectal malformations might contribute to the incontinence found in these patients.’ However, it is not known if this low anal tone reflects a developmental abnormality or if the surgical treatment is responsible.2*3 Normal anal resting tone is maintained mainly by the internal anal sphincter (IAS), which is a condensation of the distal rectal smooth muscle with characteristic morphology and function.4-6 The functional characteristics of IAS smooth muscle in vitro are a nonadrenergic, noncholinergic relaxation in response to pharmacological ol-adrenergic stimulation, and a relaxation in response to /3-adrenergic stimulation. With cholinergic drugs, both relaxation and contraction have been reported.6-” Morphological studies in pigs with imperforate anus suggested a sphincter-like configuration of the distal rectal smooth muscle.” However, there are no previous data regarding the functional properties in vitro of the rectal smooth muscle adjacent to the fistula in imperforate anus and cloaca malformations. The aim of this study was to assess the function in vitro of smooth muscle obtained from the rectum-hstula transition zone of patients treated for this condition. Journal of Pediarric Surgery, Vol 25, No 9 (Saptember),
Pefia
York and Bronx, New
patients who underwent surgery for imperforate anus and
1990: pp 985-989
Have the
York
MATERIALS
AND
METHODS
Smooth muscle strips from the fistula site of 11 babies, representing a wide spectrum of anon&al malformations, were studied (Table 1, Fig 1). The operation was performed according to the standard procedure described by deVries and Pefia’j and PetIa,14ie, the fistula and the rectum were exposed via a posterior sagittal midline incision. In cases with a very high anomaly, this approach was combined with a laparotomy. Multiple traction sutures were placed along a longitudinal incision of the posterior rectal wall and in the mucosa of the anterior rectal wall at the site of transition between the rectum and the fistula. Needle-tip cautery was used for the mobilization of the rectum, creating a plane of dissection through the common wall, between the rectum and urethra. The fistula was closed, the rectum pulled down, the striated muscle complex was reconstructed around the mobilized distal bowel, and a cutaneous anoplasty was performed. Excess tissue adjacent to the traction sutures was excised and used for the functional studies. Motility
Recordings
The tissue was placed in a cool Krebs solution and brought fresh to the laboratory, where the mucosa was carefully removed and 8 to 12 x 2 mm transverse strips of smooth muscle prepared. The number of strips from each patient varied between one and four. Tissue that had been visibly damaged by the cautery was discarded. The location of the strips is indicated in Fig 1. Each strip was mounted in a standard tissue bath between a rigid support and an isotonic transducer (Metropac, Gould, Cleveland, OH) operating a polygraph (Gould). The bath contained a Krebs solution of the following composition (g/L): glucose 1.8, Mg 0.1, KCI 0.34, NaCl 7.0. NaPO, 0.28, NaCo, 1.26, and CaClO.55. The solution was oxygenated with 95% 0, and 5% CO, and the temperature maintained at 37OC. Electrical field stimulation was applied with a pair of platinum electrodes connected to a modified Grass S44 stimulator (Grass Instrument Co, Quincy, MA). A l-g load was applied to the strips and allowed to equilibrate for approximately 1hour before the experiment. The following drugs were used: acetylcholine (ACh), phenylefrine (PHE), isoproterenol (ISO), tetrodotoxin (TTX), atropine sulphate (ATR), and propranolol from Sigma Chemical Co (St Louis, MO) and phentolaminemesylat (Regitin) from CIBA-Geigy (Summit, NJ). From the Department of Pediatric Surgery, Schneider Children’s Hospital of Long fsland Jewish Medical Center, New Hyde Park, NY, and Clinical Campus of Albert Einstein College of Medicine, Bronx, NY. Supported in part by the Swedish Medical Research Council. Dr Hedlund was a Research Fellow at the Schneider Children’s Hospital. LIJMC, at the time of this study. Presented at the 38th Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Chicago, Illinois. October 21-23.1989. Address reprint requests to Albert0 Peiia. MD, Chief; Pediatric Surgery, Schneider ChildrenS Hospital. Long Island Jewish Medical Center, New Hyde Park, NY 11042. D 1990 by W.B. Saunders Company. 0022-3468/90/2509-0013$03.00/0
985
HEDLUND AND PEtiA
986
Table 1. Anorectal Malformations Studied
Patients
of Strips
Bulbourethral fistula
3
9
Prostaticfistula
3
6
Bladderfistula
1
2
No fistula
1
2
Cloaca1malformation
3
6
11
25
No.
Diagnosis
Total
of
No.
strips there was either a contraction (three strips), a biphasic response (three strips), or no response (two strips). The relaxation response was resistant to cholinergic and adrenergic receptor blockade with atropine ( 1O-4 mol/L), propranolol ( lop4 mol/L) and phentolamine (10e4 mol/L), respectively. Further the relaxation was abolished by TTX ( lop6 mol/L), indicating that the effect was mediated by nonadrenergic, noncholinergic inhibitory neurons. In contrast, contractions in response to EFS were resistant to TTX, suggesting an unspecific smooth muscle activation.
RESULTS
Strips that showed no response, either to the drugs tested or to electrical field stimulation (EFS), were excluded because there was no indication that they contained viable smooth muscle. Thus a total of 25 strips are included in this report (Table 1). Due to technical problems, there were two patients whose strips could not be tested for response to EFS. In all the other patients each type of stimulus was tested in at least one strip. Spontaneous Motility
Apart from a spontaneous shortening, which occurred in some strips 20 to 40 minutes after they had been mounted, there were few spontaneous changes of tone. Responses to EFS
EFS over a wide range of parameters (1 to 20 Hz, 1 millisecond, 10 to 40 V) induced relaxations in 12 of 20 strips, representing nine patients (Fig 2). In the other
Responses to ACh
ACh, at concentrations ranging from 10-j to 10e2 mol/L, induced contractions in 12 of 23 strips, including all 11 patients. A biphasic or inhibitory response was obtained in four strips, and seven strips did not respond. Usually a high concentration ( 10e3 or 10e2 mol/L) was needed to elicit the contraction, and even then the response amplitude was low, compared with the adrenergic contractions (see below) obtained in the same experiments (Figs 3 and 4). Muscarinic receptor blockade with atropine (10e4 mol/L) abolished the contraction in response to ACh. In the presence of atropine (10e4 mol/L), ACh (10m3 to low2 mol/L) instead induced a relaxation (14 of 16 strips tested). This atropine resistant relaxation was resistant to Lu-adrenoceptor and P-adrenoceptor blockade as well as to TTX (Fig 3). Responses to PHE
The selective a-adrenoceptor agonist PHE caused contractions in all strips, at concentrations ranging from lo-’ to 10m3 mol/L (Fig 4). The response amplitude varied considerably between the strips. Still the response was usually of a larger amplitude than contractions in response to similar concentrations of ACh (Fig 4). The response was abolished by phentolamine ( 10e4 mol/L) but resistant to atropine and TTX, confirming that the effect was caused by activation of smooth muscle a-adrenoceptors. Responses to IS0
Nineteen strips were tested with the selective p-adrenoceptor agonist ISO. At concentrations ranging from lo-’ to lo-’ mol/L relaxation was obtained in 17 strips and two strips did not respond. The effect was abolished by propranolol confirming a p-adrenergic mechanism. DISCUSSION Fig 1. Lateral view of the distal rectum end reotourethral fistule. The shaded ares indicates the bowel region from which the strips were obtained.
The present study was undertaken assuming that there are established functional criteria for normal anal smooth muscle. Anal smooth muscle has been
ANAL SPH1NCTER FUNCTION
EFS
1 Fig 2. Recording showing typical relaxation in response to EFS: 20 Hz, 1 millisecond, 30 V. Note that the response is resistant to Atropine IATR) lo-’ mol/L. and to fl-adrenergic blockade with propranolol (PROPRI 1O-4 mol/L, and abolished by tetrodotoxin (TTXI 10~6mol/L.
2mm
1
ATR
reported to show different functional features in vitro in comparison with smooth muscle from other parts of the large bowel, namely a nonadrenergic, noncholinergic relaxation in response to EFS, a contraction in response to a-adrenergic stimulation and a relaxation in response to ACh.7W11,15 It was not possible to obtain anal smooth muscle from normal healthy children, and a single report (1982) on anal smooth muscle function in children was limited to patients with Hirschsprung’s disease and chronic constipation.* Therefore, our conclusions about the potential sphincter function of the distal rectal smooth muscle in patients with anorectal malformations will have to be based on the assumption that infant and adult anal smooth muscle have similar functional characteristics. In the present study, the response to EFS was usually a pure relaxation (12/20 strips). In the most extensive study of adult anal smooth muscle, all strips responded to EFS with a nonadrenergic, noncholinTime
(W4
mot/L)
PROPR
(lo-”
n&L)
TTX
(10
” mol/L)
ergic relaxation.7 In children with chronic constipation and Hirschsprung’s disease,* anal smooth muscle responded either with relaxation (56%) or with contraction (44%). Such a predominantly inhibitory response pattern to nerve activation with EFS contrasts with colonic smooth muscle function, which shows both cholinergic and noncholinergic neurogenic contractions in response to EFS.15 Contractions in response to EFS previously found in anal smooth muscle were considered to be caused by the direct stimulation of the smooth muscle.@ This is also the most likely explanation for the EFS-induced contractions in this study, since they were resistant to the nerve paralyzing agent TTX. In view of the. strong excitatory and inhibitory responses to pharmacological adrenoceptor stimulation that were found in this study, the absence of an adrenergic motility response to EFS might seem surprising. However, a similar phenomenon was also noted in normal anal smooth muscle in vitro and may be
(mln)
hnl-
Zmm
I
T ACh
7 T
ACh
ATR
(10
4 mol/L)
ATR
(lo-”
mol/L)
TTX
(10
mol/L)
Fig 3. Recording showing a typical contractile response to a--high conce’ntration IlO-’ mol/L) of acetylcholine (ACh). Note that the response b reversed into a relaxetion by atropine (ATM treatment. Also note that the ACh-induced relaxation ir resistant to tetrodotoxin iTTX).
988
2mm
HEDLUND AND PEfiA
I ni’\l’, T
ACh
t
r
T
PHE
PHE
PHE
ATR
( 1Om4 mol/L)
explained by a dominance of inhibitory nonadrenergic, noncholinergic neurons, which are not available for selective pharmacological blockade and which have a lower activation threshold than the adrenergic nerves.6,‘6 ACh induces contraction in most types of visceral smooth muscle. The anal canal is the only part of the gastrointestinal tract where smooth muscle relaxation to ACh has been reported.7M10Burleigh et al’ and Paskins et al8 thus reported predominately relaxation responses in normal adults. However, Parks et al” reported predominantly contraction and Friedman” that the tissue did not respond to ACh. Anal smooth muscle from children with Hirschsprung’s disease and chronic constipation was reported to contract in response to ACh, but it is not known if this excitatory effect reflects a functional disturbance or the age of the patients.8 The dual response to ACh that was observed in this study fits well into the complex response pattern that has been described for anal smooth muscle. The mechanism of the inhibitory ACh effect on anal smooth muscle is not completely understood, and might involve both direct effects on muscarinic smooth muscle receptors and an indirect cholinergic activation of other neurons.6 The relative resistance of the relaxation response to atropine treatment, which was observed in this study, has previously only been demonstrated in anal smooth muscle of primates.17 The selective cu-adrenoceptor agonist PHE invariably caused contraction in this study. This is in accordance with the results in anal smooth muscle, in which contractions have been an almost constant response to cw-adrenergic stimulation.*-” This excitatory effect, as well as the relaxation response to /3-adrenergic stimulation, are likely to be caused by a direct activation of adrenergic smooth muscle receptors. Thus, morphological studies have demonstrated a dense adrenergic innervation of the anal smooth muscle.5 In contrast, colonic adrenergic nerves impinge mainly on excitatory intramural neurons, via inhibitory cY-adrenoceptors.5*‘8
PHENT
(tom4
mol/L)
Fig 4. Recording showing a typical contraction in response to phenylephrine JPHE) (lO_’ mol/L). The response to a higher concentration (IO-’ mol/L) of acetycholine (ACh) is shown for comparison. Note that the PHE-induced contraction is unchanged by atropine IATR) but abolished by phentolamine (PHENT).
Thus, the results suggest that the distal rectal smooth muscle of patients with anorectal malformation responds to electrical and pharmacological stimuli in a similar fashion to that observed in normal anal smooth muscle evaluated in vitro. However, additional functional and morphological studies have to be performed in order to confirm that this tissue has an anal type of innervation. Such studies might also show how far into the rectum an “anal type” smooth muscle extends. Even if further functional and morphological studies will confirm that the distal rectal smooth muscle in patients with anorectal malformations is similar to normal anal smooth muscle, it is not obvious what the clinical implications of such a finding should be. For the past few years we have tried to minimize the rectal dissection during primary repair of anorectal malformations, thereby creating as little excess tissue as possible. This was also done in order to preserve the extrinsic innervation that, among other factors, might be of importance for the maintenance of anal smooth muscle tone.” However, the anal pressure zone, which is supposed to correspond to the IAS is very short in infants, and there is no reason to believe that a corresponding zone is longer in patients with anorectal malformations. Efforts to preserve this structure and its extrinsic nerve supply may lead to insufficient bowel mobilization with the subsequent risk of dehiscence, retraction, or anal stenosis. In addition the rectum and the urethra (or vagina) share a common wall that is very thin distally. Thus, even if the entire length of the bowel is preserved, the smooth muscle will be missing in the ventral aspect of the distal segment. It is recognized that the preservation and reconstruction of the striated muscle, which is accomplished with the posterior sagittal anorectoplasty, is not sufficient to achieve complete continence in all patients.20 Although many factors, such as poor sensation and bowel motility disturbances, might contribute to these prob-
989
ANAL SPHINCTER FUNCTION
lems, it is possible that the preservation of the distal rectal smooth muscle might contribute to an improved functional result in patients with anorectal malformations.1’3*2’ In conclusion, these functional studies in vitro, suggest the presence of an IAS-type of structure located at the most distal part of the bowel in anorectal malformations. However, the results must be interpreted with care, because they were obtained in a limited number of tissue strips and because suitable
control tissue was not available. Therefore, further functional studies are needed, and, in addition, histopathologic evidence is necessary to confirm this concept. Whether this concept will prove to have clinical implications remains to be seen. ACKNOWLEDGMENT The skillful technical assistance of George Rodriguez and the manuscript corrections by Dr Peter Shrock are gratefully acknowledged.
REFERENCES 1. lwai N, Ogits S, Kida M, et al: A clinical and manometric correlation for assessment of postoperative continence in imperforate anus. J Pediatr Surg 14:538-543, 1979 2. Stephens FD, Smith ED: Anorectal Malformations in Children. Chicago, IL, Year Book, 197 I, p 252 3. Penninckx FMA, Kerremans RPJ: Internal sphincter saving in imperforate anus with or without fistula. Int J Colorect Dis 1:23-32, 1986 4. Frenckner B, Ihre T: Influence of autonomic nerves on the internal anal sphincter in man. Gut 17:306-312, 1985 5. Baumggarten HG, Holstein AF, Stelzner F: Differences in the innervation of the large intestine and internal anal sphincter in mammals and human. Verh Anat Ges 66:43-47, 1971 6. Burleigh DE, D’Mello A: Neural and pharmacological factors affecting motility of the internal anal sphincter. Gastroenterology 84:409-417, 1983 7. Burleigh DE, D’Mello A, Parks AG: Responses of isolated internal anal sphincter to drugs and electrical field stimulation. Gastroenterology 77:484-490, 1979 8. Paskins Jr, Clayden GS, Lawson JON: Some pharmacological responses of isolated internal sphincter strips from chronically constipated children. Stand J Gastroenterol 17: 155-l 56,1982 (suppl 71) 9. Bass DD, Ustach TJ, Schuster MM: In vitro pharmacological differentiation of sphincteric and non-sphincteric muscle. Johns Hopkins Med J 127:185-191, 1970 10. Friedman CA: The action of nicotine and catecholamines on the human internal anal sphincter. Am J Dig Dis 13:428-431,1968 11. Parks AC, Fishlock DJ, Cameron JDG, et al: Preliminary
investigation of the pharmacology of the human internal anal sphincter. Gut 103674677, 1969 12. Lambrecht W, Lierse W: The internal sphincter in anorectal malformations: Investigations in neonatal pigs. J Pediatr Surg 11:1160-l 168,1987 13. deVries P, Pefla A: Posterior sagittal anorectoplasty. J Pediatr Surg 5:638-643, 1982 14. Pefia A: Surgical management of anorectal malformations. A unified concept. Pediatr Surg Int 3:82-93, 1988 15. Stockley HL, Bennet A: The intrinsic innervation of human sigmoid colonic muscle, in Daniel EE, (ed): Proceedings of the IC-TH International Symposium on Gastrointestinal Motility. Vancouver, Canada, Mitchell, 1978, pp 165-176 16. Crema A, Del Tacca M, Frigo GM, et al: Presence of a non-adrenergic, non-cholinergic inhibitory system in the human colon. Gut 9:633-637, 1968 17. Rayner V: Characteristics of the internal anal sphincter and the rectum of the vervet monkey. J Physiol286:383-399, 1979 18. Furness JB, Costa F: The adrenergic innervation of the gastrointestinal tract. Ergeb Physiol Biol Chem Exp Pharmacol 69:l-51, 1974 19. Carlstedt A, Nordgren S, Fasth S, et al: Sympathetic nervous influence in the internal anal sphincter and rectum in man. Int J Colorect Dis 3:90-95, 1988 20. Peiia A: Posterior sagittal anorectoplasty: Results in the management of 332 cases of anorectal malformations. Pediatr Surg Int 3:94-104, 1988 2 1. Frenckner B: Use of the recta-urethral fistula for reconstruction of the anal canal in high anal atresia. 2 Kinderchir 40:312-314, 1985
but was resistant to
atropine and tetrodotoxin. @-Adrenergic stimulation caused a relaxation that was abolished by propranolol. Function of the distal rectal smooth muscle, resected during correction of anorectal malformations, shows similarities to the function reported previously on normal anal smooth muscle evaluated in vitro. @ 1990 by W.9. Saunders
Company.
INDEX WORDS: lmperforate anus: anal sphincter function.
A
NAL RESTING TONE is regarded as an important factor for the maintenance of anal continence. Diminished anal tone that is usually seen after treatment for high anorectal malformations might contribute to the incontinence found in these patients.’ However, it is not known if this low anal tone reflects a developmental abnormality or if the surgical treatment is responsible.2*3 Normal anal resting tone is maintained mainly by the internal anal sphincter (IAS), which is a condensation of the distal rectal smooth muscle with characteristic morphology and function.4-6 The functional characteristics of IAS smooth muscle in vitro are a nonadrenergic, noncholinergic relaxation in response to pharmacological ol-adrenergic stimulation, and a relaxation in response to /3-adrenergic stimulation. With cholinergic drugs, both relaxation and contraction have been reported.6-” Morphological studies in pigs with imperforate anus suggested a sphincter-like configuration of the distal rectal smooth muscle.” However, there are no previous data regarding the functional properties in vitro of the rectal smooth muscle adjacent to the fistula in imperforate anus and cloaca malformations. The aim of this study was to assess the function in vitro of smooth muscle obtained from the rectum-hstula transition zone of patients treated for this condition. Journal of Pediarric Surgery, Vol 25, No 9 (Saptember),
Pefia
York and Bronx, New
patients who underwent surgery for imperforate anus and
1990: pp 985-989
Have the
York
MATERIALS
AND
METHODS
Smooth muscle strips from the fistula site of 11 babies, representing a wide spectrum of anon&al malformations, were studied (Table 1, Fig 1). The operation was performed according to the standard procedure described by deVries and Pefia’j and PetIa,14ie, the fistula and the rectum were exposed via a posterior sagittal midline incision. In cases with a very high anomaly, this approach was combined with a laparotomy. Multiple traction sutures were placed along a longitudinal incision of the posterior rectal wall and in the mucosa of the anterior rectal wall at the site of transition between the rectum and the fistula. Needle-tip cautery was used for the mobilization of the rectum, creating a plane of dissection through the common wall, between the rectum and urethra. The fistula was closed, the rectum pulled down, the striated muscle complex was reconstructed around the mobilized distal bowel, and a cutaneous anoplasty was performed. Excess tissue adjacent to the traction sutures was excised and used for the functional studies. Motility
Recordings
The tissue was placed in a cool Krebs solution and brought fresh to the laboratory, where the mucosa was carefully removed and 8 to 12 x 2 mm transverse strips of smooth muscle prepared. The number of strips from each patient varied between one and four. Tissue that had been visibly damaged by the cautery was discarded. The location of the strips is indicated in Fig 1. Each strip was mounted in a standard tissue bath between a rigid support and an isotonic transducer (Metropac, Gould, Cleveland, OH) operating a polygraph (Gould). The bath contained a Krebs solution of the following composition (g/L): glucose 1.8, Mg 0.1, KCI 0.34, NaCl 7.0. NaPO, 0.28, NaCo, 1.26, and CaClO.55. The solution was oxygenated with 95% 0, and 5% CO, and the temperature maintained at 37OC. Electrical field stimulation was applied with a pair of platinum electrodes connected to a modified Grass S44 stimulator (Grass Instrument Co, Quincy, MA). A l-g load was applied to the strips and allowed to equilibrate for approximately 1hour before the experiment. The following drugs were used: acetylcholine (ACh), phenylefrine (PHE), isoproterenol (ISO), tetrodotoxin (TTX), atropine sulphate (ATR), and propranolol from Sigma Chemical Co (St Louis, MO) and phentolaminemesylat (Regitin) from CIBA-Geigy (Summit, NJ). From the Department of Pediatric Surgery, Schneider Children’s Hospital of Long fsland Jewish Medical Center, New Hyde Park, NY, and Clinical Campus of Albert Einstein College of Medicine, Bronx, NY. Supported in part by the Swedish Medical Research Council. Dr Hedlund was a Research Fellow at the Schneider Children’s Hospital. LIJMC, at the time of this study. Presented at the 38th Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Chicago, Illinois. October 21-23.1989. Address reprint requests to Albert0 Peiia. MD, Chief; Pediatric Surgery, Schneider ChildrenS Hospital. Long Island Jewish Medical Center, New Hyde Park, NY 11042. D 1990 by W.B. Saunders Company. 0022-3468/90/2509-0013$03.00/0
985
HEDLUND AND PEtiA
986
Table 1. Anorectal Malformations Studied
Patients
of Strips
Bulbourethral fistula
3
9
Prostaticfistula
3
6
Bladderfistula
1
2
No fistula
1
2
Cloaca1malformation
3
6
11
25
No.
Diagnosis
Total
of
No.
strips there was either a contraction (three strips), a biphasic response (three strips), or no response (two strips). The relaxation response was resistant to cholinergic and adrenergic receptor blockade with atropine ( 1O-4 mol/L), propranolol ( lop4 mol/L) and phentolamine (10e4 mol/L), respectively. Further the relaxation was abolished by TTX ( lop6 mol/L), indicating that the effect was mediated by nonadrenergic, noncholinergic inhibitory neurons. In contrast, contractions in response to EFS were resistant to TTX, suggesting an unspecific smooth muscle activation.
RESULTS
Strips that showed no response, either to the drugs tested or to electrical field stimulation (EFS), were excluded because there was no indication that they contained viable smooth muscle. Thus a total of 25 strips are included in this report (Table 1). Due to technical problems, there were two patients whose strips could not be tested for response to EFS. In all the other patients each type of stimulus was tested in at least one strip. Spontaneous Motility
Apart from a spontaneous shortening, which occurred in some strips 20 to 40 minutes after they had been mounted, there were few spontaneous changes of tone. Responses to EFS
EFS over a wide range of parameters (1 to 20 Hz, 1 millisecond, 10 to 40 V) induced relaxations in 12 of 20 strips, representing nine patients (Fig 2). In the other
Responses to ACh
ACh, at concentrations ranging from 10-j to 10e2 mol/L, induced contractions in 12 of 23 strips, including all 11 patients. A biphasic or inhibitory response was obtained in four strips, and seven strips did not respond. Usually a high concentration ( 10e3 or 10e2 mol/L) was needed to elicit the contraction, and even then the response amplitude was low, compared with the adrenergic contractions (see below) obtained in the same experiments (Figs 3 and 4). Muscarinic receptor blockade with atropine (10e4 mol/L) abolished the contraction in response to ACh. In the presence of atropine (10e4 mol/L), ACh (10m3 to low2 mol/L) instead induced a relaxation (14 of 16 strips tested). This atropine resistant relaxation was resistant to Lu-adrenoceptor and P-adrenoceptor blockade as well as to TTX (Fig 3). Responses to PHE
The selective a-adrenoceptor agonist PHE caused contractions in all strips, at concentrations ranging from lo-’ to 10m3 mol/L (Fig 4). The response amplitude varied considerably between the strips. Still the response was usually of a larger amplitude than contractions in response to similar concentrations of ACh (Fig 4). The response was abolished by phentolamine ( 10e4 mol/L) but resistant to atropine and TTX, confirming that the effect was caused by activation of smooth muscle a-adrenoceptors. Responses to IS0
Nineteen strips were tested with the selective p-adrenoceptor agonist ISO. At concentrations ranging from lo-’ to lo-’ mol/L relaxation was obtained in 17 strips and two strips did not respond. The effect was abolished by propranolol confirming a p-adrenergic mechanism. DISCUSSION Fig 1. Lateral view of the distal rectum end reotourethral fistule. The shaded ares indicates the bowel region from which the strips were obtained.
The present study was undertaken assuming that there are established functional criteria for normal anal smooth muscle. Anal smooth muscle has been
ANAL SPH1NCTER FUNCTION
EFS
1 Fig 2. Recording showing typical relaxation in response to EFS: 20 Hz, 1 millisecond, 30 V. Note that the response is resistant to Atropine IATR) lo-’ mol/L. and to fl-adrenergic blockade with propranolol (PROPRI 1O-4 mol/L, and abolished by tetrodotoxin (TTXI 10~6mol/L.
2mm
1
ATR
reported to show different functional features in vitro in comparison with smooth muscle from other parts of the large bowel, namely a nonadrenergic, noncholinergic relaxation in response to EFS, a contraction in response to a-adrenergic stimulation and a relaxation in response to ACh.7W11,15 It was not possible to obtain anal smooth muscle from normal healthy children, and a single report (1982) on anal smooth muscle function in children was limited to patients with Hirschsprung’s disease and chronic constipation.* Therefore, our conclusions about the potential sphincter function of the distal rectal smooth muscle in patients with anorectal malformations will have to be based on the assumption that infant and adult anal smooth muscle have similar functional characteristics. In the present study, the response to EFS was usually a pure relaxation (12/20 strips). In the most extensive study of adult anal smooth muscle, all strips responded to EFS with a nonadrenergic, noncholinTime
(W4
mot/L)
PROPR
(lo-”
n&L)
TTX
(10
” mol/L)
ergic relaxation.7 In children with chronic constipation and Hirschsprung’s disease,* anal smooth muscle responded either with relaxation (56%) or with contraction (44%). Such a predominantly inhibitory response pattern to nerve activation with EFS contrasts with colonic smooth muscle function, which shows both cholinergic and noncholinergic neurogenic contractions in response to EFS.15 Contractions in response to EFS previously found in anal smooth muscle were considered to be caused by the direct stimulation of the smooth muscle.@ This is also the most likely explanation for the EFS-induced contractions in this study, since they were resistant to the nerve paralyzing agent TTX. In view of the. strong excitatory and inhibitory responses to pharmacological adrenoceptor stimulation that were found in this study, the absence of an adrenergic motility response to EFS might seem surprising. However, a similar phenomenon was also noted in normal anal smooth muscle in vitro and may be
(mln)
hnl-
Zmm
I
T ACh
7 T
ACh
ATR
(10
4 mol/L)
ATR
(lo-”
mol/L)
TTX
(10
mol/L)
Fig 3. Recording showing a typical contractile response to a--high conce’ntration IlO-’ mol/L) of acetylcholine (ACh). Note that the response b reversed into a relaxetion by atropine (ATM treatment. Also note that the ACh-induced relaxation ir resistant to tetrodotoxin iTTX).
988
2mm
HEDLUND AND PEfiA
I ni’\l’, T
ACh
t
r
T
PHE
PHE
PHE
ATR
( 1Om4 mol/L)
explained by a dominance of inhibitory nonadrenergic, noncholinergic neurons, which are not available for selective pharmacological blockade and which have a lower activation threshold than the adrenergic nerves.6,‘6 ACh induces contraction in most types of visceral smooth muscle. The anal canal is the only part of the gastrointestinal tract where smooth muscle relaxation to ACh has been reported.7M10Burleigh et al’ and Paskins et al8 thus reported predominately relaxation responses in normal adults. However, Parks et al” reported predominantly contraction and Friedman” that the tissue did not respond to ACh. Anal smooth muscle from children with Hirschsprung’s disease and chronic constipation was reported to contract in response to ACh, but it is not known if this excitatory effect reflects a functional disturbance or the age of the patients.8 The dual response to ACh that was observed in this study fits well into the complex response pattern that has been described for anal smooth muscle. The mechanism of the inhibitory ACh effect on anal smooth muscle is not completely understood, and might involve both direct effects on muscarinic smooth muscle receptors and an indirect cholinergic activation of other neurons.6 The relative resistance of the relaxation response to atropine treatment, which was observed in this study, has previously only been demonstrated in anal smooth muscle of primates.17 The selective cu-adrenoceptor agonist PHE invariably caused contraction in this study. This is in accordance with the results in anal smooth muscle, in which contractions have been an almost constant response to cw-adrenergic stimulation.*-” This excitatory effect, as well as the relaxation response to /3-adrenergic stimulation, are likely to be caused by a direct activation of adrenergic smooth muscle receptors. Thus, morphological studies have demonstrated a dense adrenergic innervation of the anal smooth muscle.5 In contrast, colonic adrenergic nerves impinge mainly on excitatory intramural neurons, via inhibitory cY-adrenoceptors.5*‘8
PHENT
(tom4
mol/L)
Fig 4. Recording showing a typical contraction in response to phenylephrine JPHE) (lO_’ mol/L). The response to a higher concentration (IO-’ mol/L) of acetycholine (ACh) is shown for comparison. Note that the PHE-induced contraction is unchanged by atropine IATR) but abolished by phentolamine (PHENT).
Thus, the results suggest that the distal rectal smooth muscle of patients with anorectal malformation responds to electrical and pharmacological stimuli in a similar fashion to that observed in normal anal smooth muscle evaluated in vitro. However, additional functional and morphological studies have to be performed in order to confirm that this tissue has an anal type of innervation. Such studies might also show how far into the rectum an “anal type” smooth muscle extends. Even if further functional and morphological studies will confirm that the distal rectal smooth muscle in patients with anorectal malformations is similar to normal anal smooth muscle, it is not obvious what the clinical implications of such a finding should be. For the past few years we have tried to minimize the rectal dissection during primary repair of anorectal malformations, thereby creating as little excess tissue as possible. This was also done in order to preserve the extrinsic innervation that, among other factors, might be of importance for the maintenance of anal smooth muscle tone.” However, the anal pressure zone, which is supposed to correspond to the IAS is very short in infants, and there is no reason to believe that a corresponding zone is longer in patients with anorectal malformations. Efforts to preserve this structure and its extrinsic nerve supply may lead to insufficient bowel mobilization with the subsequent risk of dehiscence, retraction, or anal stenosis. In addition the rectum and the urethra (or vagina) share a common wall that is very thin distally. Thus, even if the entire length of the bowel is preserved, the smooth muscle will be missing in the ventral aspect of the distal segment. It is recognized that the preservation and reconstruction of the striated muscle, which is accomplished with the posterior sagittal anorectoplasty, is not sufficient to achieve complete continence in all patients.20 Although many factors, such as poor sensation and bowel motility disturbances, might contribute to these prob-
989
ANAL SPHINCTER FUNCTION
lems, it is possible that the preservation of the distal rectal smooth muscle might contribute to an improved functional result in patients with anorectal malformations.1’3*2’ In conclusion, these functional studies in vitro, suggest the presence of an IAS-type of structure located at the most distal part of the bowel in anorectal malformations. However, the results must be interpreted with care, because they were obtained in a limited number of tissue strips and because suitable
control tissue was not available. Therefore, further functional studies are needed, and, in addition, histopathologic evidence is necessary to confirm this concept. Whether this concept will prove to have clinical implications remains to be seen. ACKNOWLEDGMENT The skillful technical assistance of George Rodriguez and the manuscript corrections by Dr Peter Shrock are gratefully acknowledged.
REFERENCES 1. lwai N, Ogits S, Kida M, et al: A clinical and manometric correlation for assessment of postoperative continence in imperforate anus. J Pediatr Surg 14:538-543, 1979 2. Stephens FD, Smith ED: Anorectal Malformations in Children. Chicago, IL, Year Book, 197 I, p 252 3. Penninckx FMA, Kerremans RPJ: Internal sphincter saving in imperforate anus with or without fistula. Int J Colorect Dis 1:23-32, 1986 4. Frenckner B, Ihre T: Influence of autonomic nerves on the internal anal sphincter in man. Gut 17:306-312, 1985 5. Baumggarten HG, Holstein AF, Stelzner F: Differences in the innervation of the large intestine and internal anal sphincter in mammals and human. Verh Anat Ges 66:43-47, 1971 6. Burleigh DE, D’Mello A: Neural and pharmacological factors affecting motility of the internal anal sphincter. Gastroenterology 84:409-417, 1983 7. Burleigh DE, D’Mello A, Parks AG: Responses of isolated internal anal sphincter to drugs and electrical field stimulation. Gastroenterology 77:484-490, 1979 8. Paskins Jr, Clayden GS, Lawson JON: Some pharmacological responses of isolated internal sphincter strips from chronically constipated children. Stand J Gastroenterol 17: 155-l 56,1982 (suppl 71) 9. Bass DD, Ustach TJ, Schuster MM: In vitro pharmacological differentiation of sphincteric and non-sphincteric muscle. Johns Hopkins Med J 127:185-191, 1970 10. Friedman CA: The action of nicotine and catecholamines on the human internal anal sphincter. Am J Dig Dis 13:428-431,1968 11. Parks AC, Fishlock DJ, Cameron JDG, et al: Preliminary
investigation of the pharmacology of the human internal anal sphincter. Gut 103674677, 1969 12. Lambrecht W, Lierse W: The internal sphincter in anorectal malformations: Investigations in neonatal pigs. J Pediatr Surg 11:1160-l 168,1987 13. deVries P, Pefla A: Posterior sagittal anorectoplasty. J Pediatr Surg 5:638-643, 1982 14. Pefia A: Surgical management of anorectal malformations. A unified concept. Pediatr Surg Int 3:82-93, 1988 15. Stockley HL, Bennet A: The intrinsic innervation of human sigmoid colonic muscle, in Daniel EE, (ed): Proceedings of the IC-TH International Symposium on Gastrointestinal Motility. Vancouver, Canada, Mitchell, 1978, pp 165-176 16. Crema A, Del Tacca M, Frigo GM, et al: Presence of a non-adrenergic, non-cholinergic inhibitory system in the human colon. Gut 9:633-637, 1968 17. Rayner V: Characteristics of the internal anal sphincter and the rectum of the vervet monkey. J Physiol286:383-399, 1979 18. Furness JB, Costa F: The adrenergic innervation of the gastrointestinal tract. Ergeb Physiol Biol Chem Exp Pharmacol 69:l-51, 1974 19. Carlstedt A, Nordgren S, Fasth S, et al: Sympathetic nervous influence in the internal anal sphincter and rectum in man. Int J Colorect Dis 3:90-95, 1988 20. Peiia A: Posterior sagittal anorectoplasty: Results in the management of 332 cases of anorectal malformations. Pediatr Surg Int 3:94-104, 1988 2 1. Frenckner B: Use of the recta-urethral fistula for reconstruction of the anal canal in high anal atresia. 2 Kinderchir 40:312-314, 1985
