Address for Correspondence Keith J. Lindley, Divison of Neurogastroenterology and Motility, Department of Pediatric Gastroenterology, Great Ormond Street Hospital for Sick Children and UCL Institute of Child Health, Great Ormond Street, WC1N 3HZ London, UK. Tel: +44 0 20 7405 9200 ext 5971; fax: +44 0 20 7813 8382; e-mail: firstname.lastname@example.org
Background Severe pediatric slow transit constipation (STC) is commonly due to intrinsic colonic neuromuscular disease. We sought to correlate neuromuscular histological phenotypes in pediatric STC with colonic manometric phenotypes using high-resolution manometry (HRM). We tested the hypothesis that failure of motor quiescence (FQ) between bisacodyl-induced high amplitude propagating sequences (HAPSs) might predict neuromuscular pathology.
Methods Eighteen children (10 males, median age: 7.5 years) with refractory STC underwent stationary colonic HRM before segmental colonic resection. Six age-matched constipated children with normal colonic transit served as controls. Colonic resection specimens underwent histopathological analysis. Conventional manometric parameters and area under the curve (AUC) during a 1-min period following bisacodyl-induced HAPSs [PBAUC1], as measure of FQ, were calculated.
Key Results Numbers of postbisacodyl HAPSs in descending and sigmoid segments were lower in patients than controls (P < 0.01, respectively). Low amplitude propagating sequences (LAPSs) were common prebisacodyl in controls and rare in STC (P < 0.001), whereas postbisacodyl LAPS were more common in STC (P < 0.001). Postbisacodyl, both retrograde propagating contractions and bursts of contractions were present in STC patients only (P < 0.001 and P < 0.01). Postbisacodyl simultaneous pressurization was seen only in STC (P < 0.05 and P < 0.001, in descending and rectosigmoid segments). Histological abnormalities were present in 17/18. Fourteen were neurogenic, one neuro-myogenic, and two myogenic. In segments with HAPS, PBAUC1 was predictive of colonic neuropathy using a cutoff of 205 mmHg.s-1 (Sensitivity 100%, specificity 86%, PPV92%, NPV100%).
Conclusions & Inferences PBAUC1 is increased in multiple colonic segments in neuropathic pediatric STC and constitutes a sensitive and specific biomarker of neuropathy.
Chronic Constipation, a common complaint in children, accounts for ∼3% of all general pediatric outpatient visits and up to 25% of all cases seen in a pediatric gastroenterology tertiary center.1 A proportion of these will suffer slow transit constipation (STC), which may respond poorly to conventional medical therapies.2,3 In the most severe cases the severity of symptoms and the lack of efficacy of medical treatment may eventually lead to surgical resection of a portion of colon.4
Histological evaluation of the colon of children with severe refractory STC demonstrates a high prevalence of enteric neuropathology in both the resected colon and the residual colon implying the presence of a generalized neuromuscular pathology, which can involve interstitial cells of Cajal (ICCs), enteric nerves, or smooth muscle cells.5,6 Although functional studies provide evidence of a generalized disturbance of colonic motility in this subgroup of patients, so far no colonic manometric biomarkers that predict the type of neuromuscular abnormalities, have been identified.5,7
Bisacodyl is a diphenylemethane laxative that is known to induce, when topically infused in the colonic lumen, powerful peristaltic waves, namely high amplitude propagating sequences (HAPSs) at 3–4 min intervals.8 This response does not occur after lidocaine has been applied to the colonic mucosa, suggesting a direct action of bisacodyl on the enteric neurons.9 Thus, topical bisacodyl has been employed as a possible test for assessing the integrity of intrinsic neural function.10,11 Moreover, it has been shown by using mucosal electromyography that bisacodyl acting within as little as 3 min increases the sporadic propulsive spiking activity corresponding to HAPSs, and suppresses the rhythmic stationary myoelectrical activity between the occurrence of propulsive spiking activity inducing an electrical quiescence.12
The aim of this study was to correlate neuromuscular histological phenotypes in pediatric STC with colonic manometric phenotypes using high-resolution colonic manometry and to seek a manometric biomarker of generalized colonic neuromuscular dysfunction testing the hypothesis that failure of motor quiescence (FQ) between bisacodyl-induced HAPSs might predict colonic neuropathy.
Eighteen children with STC (eight females and 10 males, median age 7.5 years, range 4–15) were enrolled in this study between January 2008 and December 2011 in the setting of a tertiary referral center for intestinal motility disorders (Supplementary Table S1). All suffered severe, treatment-refractory constipation unresponsive intensive medical treatment including stimulant laxatives (senna, sodium picosulfate, and bisacodyl) and high dose of polyethylene glycol (PEG, >1 g day−1), and showed a STC as verified by transit studies. Slow transit constipation was considered refractory if the patient showed the presence of ≤2 defecations in the toilet per week and at least 1 episode of fecal incontinence per week despite the optimum medical management. Total and segmental colonic transit time was measured using radio-opaque markers.13 The children swallowed one capsule a day for 3 days, and on the 4th day, a plain abdominal radiograph was taken. The numbers of markers present in this radiograph was counted, and previously determined bony landmarks were used to locate each colonic segment. The patient with constipation was considered to have slow colonic transit when there was delay in transit through the right colon, the left colon, or both, with the pellets spread throughout the colon. They were considered to have outlet obstruction when the delay occurred in the rectosigmoid with the stagnation of the pellets in the rectum. All patients showed at plain X-ray performed at day 4, >50% of markers retained in the right colon. All but one had undergone anorectal manometry (ARM) with presence of rectoanal inhibitory reflex (RAIR) and normal anal sphincteric pressures. The remaining child had previously undergone suction rectal biopsy which had ruled out the presence of aganglionic segment. All had undergone ileocolonoscopy to exclude mucosal disease. Following colonic lavage for the colonoscopy a repeat transit study in a disimpacted bowel confirmed the presence of STC. Exclusion criteria included previous gastrointestinal surgery, structural abnormalities of the gastrointestinal tract, neurological impairment, Hirschsprung’s disease, and generalized pseudo-obstruction.
All patients underwent colonic manometry. If this was felt to be pathological they were referred for surgical opinion. Where a defunctioning stoma or a segmental colonic resection was subsequently performed, either a 10–12 cm sleeve of colon (17 patients) or numerous large full-thickness biopsies (one patient) taken distal to the stoma were sent for histological examination.
No healthy controls were available. A number of patients were referred for colonic manometry by other tertiary pediatric gastroenterology centers because of lack of satisfactory response to medical treatment. Interestingly, after endoscopic examination there was a considerable improvement in symptoms that were controlled by generally modest amounts of laxative. Moreover, transit studies in these patients following the manometric investigation confirmed normal colonic transit with hold up above the rectal outlet. These patients were defined as ‘disease controls’, and comprised of six children (four females and two males, median age 9 years, range 3–10) all of whom showed the presence of an RAIR at ARM.
Custom built open-tipped 20-lumen PVC manometric catheters (Dentsleeve) with recording lumena at 2.5-cm intervals were used. The lumena were perfused with air-free distilled water (0.15 mL min−1). The system yielded a pressure rise to a distal occlusion of >500 mmHg. Intraluminal pressures were recorded by external pressure transducers and pressure signals were digitized and stored on a PC computer for analysis using commercially available software (MMS, Netherlands; stationary manometry and analysis software v8.19).
Patients discontinued laxative therapies at least 72 h prior to study. Colonic lavage was undertaken with balanced polyethylene glycol – electrolyte solution (Kleen Prep) – according to standard hospital procedures.
Patients were sedated with short acting anesthesia (propofol) and the manometry catheter was inserted using a colonoscope with the tip fixed to the wall of the right colon (usually cecum) using hemostatic clips. The latter were secured to the catheter using a 3-0 silk tie that was looped through a small hole at the tip of the catheter.
The catheter was if necessary withdrawn until the distal recording site was approximately 5 cm above the internal anal sphincter. The final position of the catheter was checked by X-ray prior to the start of the study and again at the end of the study.
A standard stationary recording protocol was used. The recording was started at least 1 h after catheter placement when the child was fully awake. Colonic motor activity was recorded for at least 2 h during fasting, and for 1 h after each of two incremental intraluminal infusions of bisacodyl (0.2 mg kg−1 and 0.4 mg kg−1 max 15 mg). The colonic motor response to the drug was assessed during the 60-min period following each instillation of bisacodyl.
Recordings were reviewed using both conventional pressure/time plots and spatiotemporal plots.14 The following were identified: (i) high amplitude propagated sequences (HAPSs), defined as having a peak amplitude of ≥75 mmHg and propagating for ≥15 cm; (ii) low amplitude propagating sequences (LAPSs), defined as having a peak amplitude <50 mmHg and propagating migrating for ≥15 cm; HAPSs and LAPSs were further classified as antegrade or retrograde, depending upon the direction of propagation; (iii) retrograde propagating contractions (RPC), migrating orad across ≤15 cm with a velocity ≥ 0.2 ≤ 12 cm s−1; (iv) simultaneous contractions, defined as pressure waves occurring simultaneously at least 10 cm apart; (v) discrete random bursts of phasic pressure waves, with a peak amplitude >8 mmHg, frequency ≥3 per min, and a duration ≥3 min; (vi) isolated pressure waves, with a peak amplitude ≥5 mmHg and a duration ≥30 s. The numbers, duration, peak amplitude, peak velocity, and variations pre- and postbisacodyl administration of each parameter were calculated. Phasic pressure activity was quantified as area under the curve (AUC) or motility index (MI). Motility index, i.e., [loge (sum of amplitudes × number of contractions + 1)], was calculated 30 min before and 30 min after bisacodyl administration. Where bisacodyl administration brought about sequences of HAPS, to assess the FQ between the HAPSs, the AUC during a 1-min period following the HAPSs [PBAUC1] was calculated, as a means of quantifying phasic pressure activity. If the HAPSs occurred within 1 min they were discharged. For convenience the colon was divided into four segments comprising ascending, transverse, descending, and rectosigmoid regions.
Histology and immunohistochemistry
Where surgery was performed, colonic specimens (10–15 cm length distal to the site of stoma formation) were either snap frozen, fixed in glutaraldehyde or fixed in formaldehyde. Formalin fixed tissue embedded in paraffin were sectioned (4 μm) and stained with hematoxylin and eosin for routine evaluation. Immunohistochemical analyses using either formalin fixed or cryostat sectioned frozen tissues using commercial antibodies with appropriate controls (Supplementary Table S2).14,15 Sections were reviewed by two individuals (VVS and DR) blinded to the clinical data and manometric findings.
The biopsy/resection was defined as neurogenic, if: (i) the myenteric plexus formed an almost continuous network of neural tissue with an increase in glial elements instead of well-separated ganglia discreetly connected to each other; (ii) the ganglia/neural tissue was misplaced within the smooth muscle of the muscularis propria instead of being in the intermuscular space or when ectopic neutrons/groups of neurons/ganglia were noted within the muscularis propria; (iii) an increase in nerve fibers was seen in the lamina propria of the mucosa; (iv) a decrease in neural tissue in the myenteric plexus with small sparsely spaced ganglia with a reduced number of neurons. The biopsy/resection was defined as myogenic, if: (i) an additional or rudimentary smooth muscle layer was noted; (ii) abnormal immunostaining for actin with or without myocyte vacuolation was seen. If the coexistence of both neuropathic and neuropathic abnormalities was observed, the biopsy was classified as mixed. Finally, the estimation of myenteric ICC network was performed by immunostaining for CD117 (c kit), and a significant reduction in ICC number (usually more than 50%) in addition to an abnormal cell morphology was considered abnormal .
Differences among groups and within group in the mean values of each parameter were analyzed by Kruskal–Wallis one-way analysis of variance by ranks, followed by post hoc analyses using the Dunn’s Multiple. The Mann–Whitney U-test was applied to compare independent samples and the Wilcoxon signed rank test for relative samples. Chi-squared test or Fisher exact test were used when appropriate. Sensitivity, specificity, positive predictive value and negative predictive value of PBAUC1 using as cutoff the highest value measured in any colonic segment of control patients were determined. All statistical tests were two tailed using 0.05 level of significance. Analysis was done by running the Prism software version 5.00 (GraphPad, San Diego, CA, USA). Data are expressed as mean values ± standard deviation (SD), unless otherwise stated.
The study protocol was defined in accordance with the Declaration of Helsinki, and was approved by the ethical committee of our Institution. Written consent was obtained from both parents of all enrolled patients before the procedures; children older than 12 years signed a statement of assent.
Eighteen children with confirmed STC underwent colonic manometry during the study period. Six children who had undergone colonic manometry were subsequently shown in the post procedure transit study to have normal colonic transit time. These individuals have been termed ‘disease controls’.
Propagating activity: HAPSs During 1 h after bisacodyl, HAPSs were seen in all colonic segments of all control subjects (Table 1) (Fig. 1). In STC post bisacodyl HAPSs were not detected in all colonic segments, and the mean number significantly varied throughout the colon showing an aborad decrease and a significant difference between descending and ascending colon (P < 0.05), and rectosigmoid and the remaining colonic segments (P < 0.05 vs descending; P < 0.001 vs ascending and transverse) (Table 1). In four patients HAPS activity was not seen in any colonic segment (Fig. 1). When considered as two groups no differences were found between patients and controls in the number of HAPSs in the ascending and transverse colon, whereas the number of descending and rectosigmoid HAPSs were lower in patients than controls (P < 0.01 and P < 0.001, respectively) (Table 1).
Table 1. Characteristics of manometric parameters in various colonic segments in patients with STC before and after bisacodyl administration (mean ± SD)
*The frequency of HAPS is referred to the number of times each sequence has been detected in each region.
†P < 0.001 vs controls.
‡P < 0.05 vs Ascending by Kruskal–Wallis one-way analysis of variance by ranks.
§P < 0.05 vs Descending by Kruskal–Wallis one-way analysis of variance by ranks.
¶P < 0.001 vs Ascending and Transverse by Kruskal–Wallis one-way analysis of variance by ranks.
**P < 0.001 vs Descending by Kruskal–Wallis one-way analysis of variance by ranks.
††P < 0.05 vs Ascending, Transverse, and Descending by Kruskal–Wallis one-way analysis of variance by ranks.
‡‡P < 0.01 vs controls.
§§P < 0.05 vs controls.
0.1 ± 0.4
0.4 ± 0.1
0.1 ± 0.4
0.2 ± 0.1
0.3 ± 1.2
3.5 ± 1
3.9 ± 2.5
4.7 ± 1
3.1 ± 2.8
4.1 ± 0.4
1.6 ± 2.5‡‡,‡
3.3 ± 1
4.5 ± 5.2
0.5 ± 0.9†
4.0 ± 5.5
0.4 ± 0.8†
3.5 ± 4.3
0.1 ± 0.5†
3 ± 3.5
2.4 ± 2.7§
2.6 ± 3.5§
3 ± 2.8§
0.2 ± 0.5¶**§§
0.1 ± 0.2
0.1 ± 0.3
0.1 ± 0.2
1.3 ± 0.6†
1.2 ± 0.5†
1.3 ± 0.5†
2.3 ± 0.1‡‡,††
0.7 ± 0.8
0.2 ± 0.4
0.7 ± 0.5
0.6 ± 0.9
0.7 ± 0.8
0.2 ± 0.4
0.7 ± 0.8
0.5 ± 0.8
1.4 ± 0.7‡‡
1.3 ± 0.7‡‡
1.3 ± 0.5‡‡
2.4 ± 1.2†,††
0.3 ± 0.5
0.5 ± 0.7
1.3 ± 1.4
0.4 ± 0.6
0.3 ± 0.5
0.6 ± 0.9
0.7 ± 0.7
0.6 ± 0.7
0.5 ± 0.5
0.7 ± 0.8
0.6 ± 0.7
0.2 ± 0.4
0.6 ± 0.9
0.6 ± 0.9
0.2 ± 0.4
1.2 ± 1.9§§
4.9 ± 2.4¶**†
5.5 ± 2.1
2.5 ± 3.2
5.5 ± 3.0
3.2 ± 3.1
3.9 ± 1.9
2.4 ± 3.5
3.9 ± 1.9
2 ± 3.5
9.8 ± 0.1
8.3 ± 3.3
10 ± 1.1
8 ± 3.2
9.9 ± 0.8
7.4 ± 3.2
9.8 ± 0.9
2.8 ± 3.4¶**†
Propagating activity: LAPSs Prebisacodyl LAPSs were detected in all four colonic segments in controls, but were uncommon in patients with STC with significant differences in the mean number between the two groups in all colonic segments (Table 1). Postbisacodyl LAPSs were ubiquitously absent in controls, whereas in patients the mean number significantly varied throughout the colon showing an aborad decrease and a significant difference between rectosigmoid and the remaining colonic segments (P < 0.001). There was a significant difference between patients and controls in the mean number of ascending, transverse, and descending LAPSs (P < 0.05), whereas no difference was found in the rectosigmoid segment (Table 1).
Low amplitude propagating sequences were the only propagative activity seen in any colonic segment in four patients with absent HAPSs. Low amplitude propagating sequences were absent in two patients in whom HAPSs were also absent.
Propagating activity: retrograde propagating contractions (RPC) No RPC were detected in the control group pre/postbisacodyl (Table 1). Postbisacodyl an increase in retrograde activity was commonly found in all colonic segments except the rectosigmoid of the patients (P < 0.05). Overall, the mean values throughout the colon were significantly higher in patients than controls (P < 0.001).
Non-propagating activity: tonic activity (bursts) Burst activity before bisacodyl administration was detected in all colonic segments of controls, whereas in patients they were most common in the transverse and rectosigmoid segments (Table 1). Postbisacodyl bursts were absent in controls, but commonly seen in patients. Postbisacodyl burst activity varied throughout the colon in patients showing a marked increase in rectosigmoid compared with other colonic segments (P < 0.05) (Table 1).
Non-propagating activity: isolated pressure waves In both patients and controls, isolated pressure waves were occasionally found in all colonic segments both before and after bisacodyl administration and no differences were found between patients and controls in both periods.
Simultaneous contractions Before bisacodyl administration, simultaneous contractions were not seen in patients or controls (Table 1). Post bisacodyl, they were never seen in controls, whereas in patients they were seen predominantly in the more distal colon with a significant difference between rectosigmoid and the other colonic segments (P < 0.001) (Table 1). Differences between patients and controls achieved statistical significance in descending (P < 0.05) and rectosigmoid segments (P < 0.001) (Table 1). Simultaneous contractions in the distal colon were seen in the rectosigmoid of all patients in whom HAPSs were present in the more proximal colonic segments.
Motility index Prebisacodyl MI was not statistically different in patients vs controls (Table 1). Post bisacodyl MI increased in all colonic segments in controls and the mean values did not significantly vary throughout the colon (Table 1). In patients the mean MI of the group increased significantly in the right-sided colonic segments and descending colon, whereas no increase was observed in the rectosigmoid segment (P < 0.001 vs ascending, transverse, and descending) (Table 1). Post bisacodyl MI values in the rectosigmoid were lower in patients than controls (P < 0.001) (Table 1).
Histological abnormalities were identified in 17 patients and were classified as neurogenic,14 myogenic, 2 and combined neuro-myogenic.1 The myogenic abnormalities consisted of the presence of an intermittent, poorly formed additional muscle layer in one individual and the presence of abnormal accumulation of actin in vacuole like structures within the external muscle layer in another (‘Polka dot myopathy’). The morphologic abnormalities identified are described in Table 2 (Figs 2–3). Neuropathic abnormalities were the predominant pathology in STC. Manometric features of myopathy and neuropathy differed markedly (see below).
Table 2. Histological abnormalities detected in each patient and their manometric correlates
Sigmoid and rectum spasmodic tonic activity post-Bis induced HAPS
Scattered ectopic neurons in MuscP, reduced ICC
Failure of quiescence between post-Bis HAPS
Rudimentary ICC network
Failure of quiescence and retrograde LAPC between post-Bis HAPC
Ganglia spaced apart and decrease in myenteric neurons
Failure of quiescence between post bisacodyl HAPS and retrograde HAPS
Ganglia encased in MLs, increase in fibers in LP
Failure of quiescence between post-Bis
Foamy cells in c-ML, neural abnormalities, increased fibers in LP
Failure of quiescence between post-Bis
Rudimentary ICC network
Thinner L-ML, reduced ICC and ICC lack in proximal end
Failure of quiescence between post-Bis (only 1 HAPS)
Ectopic neurons in C-ML
Failure of quiescence between post-Bis
Pseudo-vacuolation of external muscle layer on H&E representing subcellular accumulations of actin and desmin (‘Polka dot myopathy’)
Manometric biosignature of colonic neuropathy
A variety of manometric features was seen in STC patients that was absent in the controls. These features were heterogeneous and not seen in all patients (Table 2). Overall FQ of low amplitude contractile activity between bisacodyl-stimulated HAPSs was the most common abnormality being present in 12/18 patients (Fig. 4). This was quantified as the area under the curve in the 1-min period following a bisacodyl-stimulated HAPSs in segments where HAPSs were present (PBAUC1).
Failure of motor quiescence of non-propagative activity between bisacodyl-induced HAPS was present on both sides of the colon in individuals with STC. In the right colon this activity was usually evident between normal amplitude HAPSs, whereas in the descending and rectosigmoid segments it was still evident even if the HAPSs were attenuated. Of the 17 patients with histologically confirmed abnormalities, five patients (two with a myopathy and three with absence of ICCs or a rudimentary ICCs network) failed to show any HAPS activity in any colonic segment. Of the remaining 13 patients 12 had colonic neuropathic change, although in one patient this segmental being was restricted to the distal rectosigmoid segment. Calculation of PBAUC1 was not possible in five patients with absent HAPS. Comparing the 11 patients with non-segmental neuropathic change with controls demonstrated the PBAUC1 to be significantly higher in patients as compared with controls in both right and left colonic segments (Ascending: 355.1 ± 131.9 vs 101.5 ± 65.5, P < 0.001; Transverse: 368.2 ± 267.9 vs 110.2 ± 26.2, P < 0.001; Descending: 324.1 ± 286.3 vs 91.5 ± 37.7 P < 0.001).
The highest measured PBAUC1 in any colonic segment of controls was 205 mmHg s−1. If values greater than this are presumed to pathological then the PBAUC1 was highly predictive of the presence of colonic neuropathy, with a sensitivity of 100%, specificity of 84%, PPV of 92%, and NPV of 100%. The child with segmental neuropathic change had PBAUC1 values of <205 mmHg s−1 in colonic regions without evidence of neuropathic change. The FQ biosignature as quantified by PBAUC1 appears to be a reliable and discriminatory predictor of generalized colonic neuropathic change. Furthermore, high-resolution manometric tracings enabled an appreciation that much contractile activity between bisacodyl-stimulated HAPS was retrograde in children with neuropathic disorders.
This study correlates colonic neuromuscular histological phenotype with high-resolution manometric findings in pediatric STC and tests the hypothesis that failure of quiescence of contractile activity between bisacodyl-induced HAPS is a marker of colonic neuropathy. It demonstrates that manometric methods can distinguish generalized and segmental disorders of colonic neuromuscular pathology and that the presence of a generalized colonic neuropathy can be predicted with a high degree of certainty using the novel variable PBAUC1, which is a quantitative expression of the failure of suppression of contractile activity between bisacodyl-induced HAPS.
The distinction of individuals with STC from those with defecatory disorder can be difficult.16 However, we believe that the criteria adopted for the purposes of this study were appropriate as they predicted neuromuscular pathology in 17/18 patients and manometric abnormality in 18/18 patients, which appears more sensitive than the retention of >25% of markers 5 days after marker ingestion.17
Studies of adults with STC using high-resolution methods have demonstrated reduced numbers of HAPS, increased numbers of retrograde LAPS, and absence of the early morning increase in HAPS activity seen around the time of waking.18,19 In a larger study, Rao et al. studied 21 adults with STC using standard 24-h colonic manometry, and they utilized the response to meal ingestion, response to morning waking, and HAPS frequency as diagnostic criteria for classifying the patients as having neuropathy or myopathy. Ten patients were classified as having neuropathy on the basis of absence of at least two aforementioned manometric features, whereas five patients were classified as having myopathy on the basis of marked attenuation of at least two colonic motor responses. The authors did not report the histopathological fundings in the resected colon, hampering any correlation between the latter and the manometric findings. Moreover, the study reported no data from the ascending colon. On the other hand, in a recent study van den Berg et al. studied colonic specimens from surgical resections or full-thickness biopsy from 13 children (Hirschsprung’s disease = 4, chronic intestinal pseudo-obstruction = 1; intractable constipation = 8) who had undergone standard colonic manometry before surgery. Morphological changes reflecting neuropathies, mesenchymopathies, and myopathies, depending on the predominant involvement of enteric neurons, cells of Cajal, or smooth muscle cells, did not correlate with particular features of colonic manometric recording, suggesting the presence of heterogenous disease and the absence of specific biomarkers in children with intractable constipation.7 In agreement with van den Berg et al., in our cohort we did not find any correlation between standard manometric features and histological findings. In the majority of our patients, we found manometric abnormalities in the left colon, such as decrease in the frequency of HAPSs, increase in simultaneous and retrograde propagating contractions. However, although these motor abnormalities have been found in the left colon in a subgroup of children after surgery for Hirschsprung, they have been also identified in adults with defecatory disorder, suggesting that these parameters are not useful in discriminating between different subgroups of patients with chronic constipation.20–22
The early age of onset of STC in the pediatric population is likely to be a marker of the high prevalence of abnormalities of the intrinsic intestinal neuromusculature of the colon in these children.5,23,24 Given the high prevalence of neuronal abnormalities in STC we chose to study the effects of bisacodyl stimulation on colonic motor activity. The mechanism of action of bisacodyl is incompletely understood, but an intact afferent neural limb of the enteric neuromusculature is necessary for its action.25 Studies with intraluminal electrodes have shown that non-propagative colonic myoelectrical activity is suppressed during bisacodyl stimulation, but propagative activity at a velocity equivalent to that seen in manometric recordings is increased.12 It seemed to us that bisacodyl stimulation would be a good method to evaluate the combined integrity of the afferent and efferent limbs of the enteric nervous system of the colon to screen for the presence of the generalized abnormalities of the colonic intrinsic nervous system which are prevalent in pediatric STC. We show here that in pediatric STC there is commonly an absence of motor quiescence between closely apposed HAPSs as measured by PBAUC1 and that PBAUC1 values >205 mmHg s−1 are highly predictive of colonic neuropathology. In most individuals these manometric abnormalities were generalized in keeping with generalized colonic neuropathology, although in the one individual with segmental neuropathology the abnormalities were restricted to the affected segment. PBAUC1 cannot be measured in the absence of HAPS and hence this measurement cannot be measured in the distal colon of individuals where this segment is aperistaltic. Similarly in the three individuals with myopathic or neuro-myopathic disorders, HAPS were absent and PBAUC1 could not be measured. The manometric appearances of primary colonic myopathy and absence of an ICCs network were similar with little discernible contractile activity. Finally, using traditional pediatric catheter design with 10–15 cm distance between the recording ports a significant proportion of pressure waves can be missed, whereas as in our study increasing the spatial resolution with a greater number of sites (20 cm) at shorter distance (2.5 cm) the full spectrum of colonic phasic activity can be appreciated.26
In addition to providing a description of colonic neuromuscular phenotype/manometric phenotype relationships we describe two novel neuromuscular disorders, namely the segmental abnormality of the myenteric plexus migrating through the muscle layers and the presence of polka dot smooth muscle actin immunostaining within apparent vacuoles in the myocytes of the external colonic muscle layer.
This study is not without drawbacks. The study lacks truly healthy pediatric control subjects for obvious ethical reasons. Although the presence of normal colonic manometry in the ‘disease control group’ allowed a comparison with the patient group, we cannot rule out for certain that the controls had normal histology.
An understanding of whether early surgical intervention is associated with better outcome is urgently required in pediatric STC, together with selection criteria for both patients and the type of surgery required. Our findings of a reliable predictor of colonic neuropathic disorder could allow the early detection of colonic neuropathy and very early intervention which might prevent the development of segmental distal colonic dilatation and the need for segmental surgical resection.
No funding declared.
All authors have no conflict of interest.
VG – performance of investigations, analysis of results, preparation of synopsis data, first draft of manuscript. OB – Recruitment of patients, analysis of results, revised critically the manuscripts, and approved the final version of the manuscript. VVS – performance of histopathologic methods, revised critically the manuscripts, and approved the final version of the manuscript. DR – performance of histopathologic methods, revised critically the manuscripts, and approved the final version of the manuscript. JK – Recruitment of patients, revised critically the manuscripts, and approved the final version of the manuscript. NS – Recruitment of patients, revised critically the manuscripts, and approved the final version of the manuscript. NT – Performance of investigations, revised critically the manuscripts, and approved the final version of the manuscript. JC – Surgical management, revised critically the manuscripts, and approved the final version of the manuscript. KJL – Guarantor of manuscript, study design, supervisor of studies, analysis of data, revision of manuscript.