Pan-Colonic Pressurizations Associated With Relaxation of the Anal Sphincter in Health and Disease: A New Colonic Motor Pattern Identified Using High-Resolution Manometry

Objectives:Only a few studies have applied high-resolution manometry (HRM) to the study of colonic motility in adults and none of them have concurrently evaluated colonic and anal motor activity. The aim of the study was to evaluate colonic and anal motor activity by means of HRM in healthy subjects. As the present study revealed the presence of a new colonic motor pattern (pan-colonic pressurizations) in healthy subjects, three additional studies were conducted: the first and the second to exclude that this motor event results from an artifact due to abdominal wall contraction and to confirm its modulation by cholinergic stimulation, and the third, as pilot study, to test the hypothesis that this colonic pattern is defective in patients with chronic constipation refractory to current pharmacological treatments.Methods:In both volunteers and patients the HRM catheter was advanced proximally during colonoscopy.Results:In all subjects, pressure increases of 15±3 mm Hg and 24±4 s simultaneously occurring in all colonic sensors (pan-colonic pressurizations) and associated with anal sphincter relaxation were identified. Subjects had 85±38 pan-colonic pressurizations, which increased significantly during meal (P=0.007) and decreased afterward (P=0.01), and were correlated with feelings of and desire to evacuate gas. The mean number of propagating sequences was 47±39, and only retrograde increased significantly postprandially (P=0.01). Pan-colonic pressurizations differed from strain artifacts and significantly increased after prostigmine. In patients pan-colonic pressurizations were significantly reduced as compared with volunteers.Conclusions:Pan-colonic pressurizations associated with relaxations of the anal sphincter represent a new colonic motor pattern that seems to be defective in patients with treatment-refractory chronic constipation and may have a role in the transport of colonic gas and in the facilitation of the propagating sequence-induced colonic transport.


INTRODUCTION
Normal activity of the colon, refl ected by bowel function, is critical for health and quality of life. Chronic idiopathic constipation is a functional disorder of the gastrointestinal tract, defi ned by the Rome III criteria as a condition with multiple symptoms, including straining, lumpy or hard stools, sensation of incomplete evacuation, sensation of anorectal obstruction, and less than three bowel movements per week ( 1 ). Th e condition is common, with prevalence between 4 and 20% in cross-sectional communitybased surveys ( 2 ). It has been demonstrated that the impact of chronic constipation on quality of life for patients is comparable to that of organic conditions, such as chronic obstructive pulmonary disease, diabetes, and depression ( 3 ). Alterations of colonic motility have been implicated in the pathophysiology of chronic idiopathic constipation ( 1 ). Previous studies, applying manometry catheters with 7.5 cm spacing, have demonstrated that colonic motor activity in health is represented mainly by non-propagated activity, with propagated sequences of low amplitude and high amplitude occurring, respectively, only in numbers of 2.4±0.1 and 0.4±0.1 per hour ( 4,5 ). Both non-propagated and propagated activities have been found to be reduced in patients with chronic idiopathic constipation ( 6 ). However, as there are no quantitative or qualitative data that unequivocally diff erentiate normal from abnormal colonic function ( 7,8 ), the role of colonic manometry in the management of these patients is unclear ( 8 ), and, at the moment, it is recommended to confi rm colonic inertia (absence of motor response to meal and of high-amplitude propagating sequences (HAPSs) aft er pharmacological stimulation, i.e., bisacodyl) in patients with slow transit not responding to pharmacological treatments selected for surgery ( 8 ).
Th e introduction of high-resolution manometry (HRM) to the study of esophageal motility has improved the understanding of esophageal dysmotility and has led to a new classifi cation of patients in diff erent subtypes with diff erent treatment outcomes ( 9,10 ). In this technique, the number of sensors recording the intraluminal pressure has been increased to one every 1 cm compared with one every 5 cm with conventional manometry. HRM has only recently been introduced in colonic studies ( 11,12 ) and it was found that the application of HRM to the study of colonic activity increases the accuracy in detecting low-amplitude propagating sequences ( 12 ).
At present only few studies have applied HRM to the study of colonic motility in adult humans, and these have been performed mainly in a single center using a fi ber-optic catheter (12)(13)(14). As normal colonic function requires the coordination between colonic and anal motor activity, and as none of the previous studies with HRM simultaneously assessed colonic and anal motor activity in adults, the aim of the present study was to use HRM to evaluate colonic motility and its relationship with anal sphincter activity during fasting and fed state in healthy subjects. Our hypothesis was that HRM could refi ne our understanding about normal colonic motility. As the present study revealed the presence of a new colonic motor pattern in healthy subjects, three additional studies have been performed: the fi rst and the second to exclude that this colonic event results from an artifact due to abdominal wall contraction during strain and to confi rm that it is modulated by cholinergic stimulation, and the third, as pilot study, to test the hypothesis that this colonic pattern is defective in patients with chronic constipation refractory to laxatives, prucalopride, lubiprostone, or linaclotide referred to our motility clinic for evaluation of possible colectomy. Th e choice of the cholinergic stimulus was based on the fact that prostigmine has been demonstrated to induce colonic contractions in previous studies ( 15,16 )

Healthy subjects and patients
In the main and additional studies 1 and 2, we studied respectively ten (30±11 years, 5 females), three (28±2 years, 2 females) and four (42±5 years, 2 females) healthy subjects with a normal bowel habit defi ned by presence of bowel movements between two per day to one every two days. None of the subjects had organic or functional disease aff ecting the gastrointestinal system. None had previous abdominal surgery other than appendectomy and none was taking laxatives or other medications. All the subjects of the main study also participated to a study with oral administration of bisacodyl and demonstrated at least one HAPS (data not shown) ( 17 ). In additional study 3 we studied ten patients (41±14 years, all females) with refractory chronic constipation as defi ned according to Rome III criteria with a mean duration of 10±5 years of symptoms. All the patients present a slow colonic transit as confi rmed using radiopaque markers ( 18 ), none had organic diseases that aff ect the gastrointestinal system or a history of previous abdominal surgery other than appendectomy, and they all did not report any chronic intermittent abdominal pain or discomfort. All had received treatment with polyethylene glycol (either alone, 13.8 mg twice or more a day, or in combination with bisacodyl or sodium picosulfate), prucalopride (2 mg/day), lubiprostone (24 μ g b.i.d.) or linaclotide (290 μ g/die) respectively for at least one month, in case of good tolerance, without satisfactory response. All the patients were using intermittent water enemas to defecate. Each healthy subject and patient gave his/her written informed consent prior to the study, which was approved by the Ethics Committee of the University Hospitals Leuven.

Experimental procedure of the main study
On the day of the experiment each healthy subject presented to the Motility Unit of the Gastroenterology Department, University Hospital of Leuven, in the early morning to undergo, aft er a bowel preparation with water enemas, a colonoscopy-assisted positioning of the colonic HRM catheter as previously described ( 19 ). A 10-F solid state HRM catheter comprising 40 pressure sensors spaced 2.5 cm apart (Unisensor AG, Switzerland) was used ( Supplementary Figure 1 online). Under conscious sedation (midazolam 5 mg intravenous (IV) and pethidine 50 mg (IV)) the colonoscope with the attached HRM catheter was advanced as far as possible into the colon. Aft er reaching the desired location, the HRM catheter was clipped to the colonic mucosa, the colonoscope was gradually removed while maximally aspirating air from the colon, and the external part of the catheter was taped to the buttock of the subject. Th e position of the tip of the HRM catheter and the absence of gas distending the colon was confi rmed fl uoroscopically aft er colonoscopy and at the end of measurement. Th e subjects were then asked to lie on a bed, and the external end of the catheter was connected to a recording system (Medical Measurement Systems, Enschede, Netherlands). Th e colonic pressure recording was started and continued for 4 h before and 3 h aft er a standardized meal (313.8 kcal, two slices of white bread with cheese and butter, and a vanilla pudding (11.6 g protein (14.7%), 13.8 g fat (39.7%), and 35.8 g carbohydrates (45.6%)). Th e choice of the caloric content of the meal was based on a usual Belgian lunch. During the fi rst 2 h of recording they were allowed to sleep (to recover from sedation) while in the remaining period they were asked to be awake. During the periods that the subjects were awake, they were asked to score every 15 min their feeling of abdominal gas, desire to evacuate gas, desire to defecate, urgency to defecate, abdominal discomfort or pain, or any other sensations on a 100 mm visual analog scale (VAS).

Experimental procedure of additional study 1: concomitant evaluation of colonic motility and abdominal wall electromyography activity in healthy subjects
Th is study was conducted as the main one but, during the colonic motility measurement, the electromyography (EMG) activity was concurrently recorded at two diff erent sites corresponding to the lower rectum on both sides of the abdomen. At each site, EMG activity was recorded by means of bipolar Ag-AgCl surface electrodes (Kendall, Covidien, Quebec, Canada) placed 3 cm apart. Th e electrodes were connected to an EMG recording system (Solar Wireless, Medical Measurement Systems, Enschede, Netherlands). Th e correct location of the electrodes was checked by recording EMG responses to maximal voluntary abdominal contraction.

Experimental procedure of additional study 2: effect of cholinergic modulation on colonic motility in healthy subjects
In this study the colonic pressure recording was conducted for 2 h during the recovery period, 1 h during basal period, and 1 h aft er the administration of the acetylcholinesterase inhibitor prostigmine 0.5 mg IV (Prostigmine, Meda Pharma, Brussels, Belgium). Blood pressure and heart rate were monitored every 15 min aft er prostigmine administration. During the period aft er administration of prostigmine each subject was instructed to report the onset of any sensations they might have felt during the recording period by means of an event marker and to score their feeling of the same sensations evaluated during the main study on the VAS.
Experimental procedure of additional study 3: pilot study to evaluate the presence and characteristics of pan-colonic pressurizations in a group of patients with chronic constipation refractory to current pharmacological treatments Th e experimental procedure was the same as in the main study in healthy subjects, but in the patients an additional recording period was added at the end of the study, to evaluate during a 2 h period the response to intra-colonic administration of bisacodyl (Dulcolax, Boehringer Ingelheim, Brussels, Belgium; 2 capsules, 10 mg, dissolved in 10 ml of saline) through an infusion tube (Entral, RT10/100, Eurosteriel Medical, Dronten, Netherlands) attached at the tip of the HRM catheter. During the period aft er intra-colonic administration of bisacodyl, the patients were only asked to use the event marker to report their sensations.

Data analysis
As a fi rst step artifacts (such as catheter movements, cough, body movements, and strain) were identifi ed and removed. Th en the colon was divided in three anatomical regions: right colon (from cecum to splenic fl exure), left colon (from splenic fl exure to sigmoid colon) and rectum. On the basis of the fl uoroscopic images, each recording sensor was then allocated to its closest anatomical region. Th e sensors located at the level of the anal sphincter were assigned to the anal sphincter region.
Potential pressure waves (monophasic pressure change with an onset to peak amplitude >5 mm Hg) were identifi ed and those occurring in three or more adjacent channels defi ned as propagating sequences, according to previous defi nition ( 5,13 ). Propagating sequences were qualifi ed as anterograde (anally propagating) or retrograde (orally propagating) ( 5,13 ) and further sub-classifi ed as HAPSs (if the amplitude of two of the propagating pressure waves was ≥100 mm Hg and that of another one was ≥80 mm Hg) ( 20 ) or as low-amplitude propagating sequences in the remaining cases. Th en low-amplitude sequences were defi ned as single, cyclic (repetitive propagating sequences), long distance, or simultaneous (three or more pressure waves occurring simultaneously in adjacent recording sites) as previously reported ( 13,21 ). Non-propagating activity (pressure changes not associated morphologically or temporally with pressure patterns recorded in adjacent sensors) were also recognized according to previous defi nitions ( 5 ).
Number of propagating and simultaneous sequences, amplitude (mm Hg), duration (seconds), propagation extent (based on the number of sensors involved in the sequence, cm), as well as the colonic site of origin (right colon, left colon, or rectum) were evaluated. For each colonic motor pattern, the relation with the anal sphincter relaxation was evaluated. Applying the defi nitions previously reported during prolonged recording of the anal sphincter activity ( 22,23 ), the anal sphincter relaxation was defi ned as a drop in the resting anal sphincter pressure (of at least 25 mm Hg) with subsequent quick return to basal values over the following 30-60 s. Th e onset and the end, the percent of relaxation (from baseline to nadir) and the duration (seconds) of the anal relaxation were evaluated. A colonic motor pattern was defi ned as associated with sphincter relaxation if the onset of the anal sphincter relaxation started in the time period of the duration of the colonic motor event. In the main study resting sphincter pressure was also calculated as the mean pressure recorded by the sensors located at the level of the anal sphincter every 15 min. In addition an analysis of the colon motor activity was performed using computer soft ware calculating the colonic motility index (MI=(ln(number of peaks×the sum of peak amplitudes+1))) ( 24 ). Th is analysis was used as overall measure of the colonic motor activity and an approximate index of non-propagating activity. In the main and additional study 3, the VAS score of diff erent sensations was also averaged per period of 15 min starting from the moment the subjects were awake aft er the recovery period.

Statistical analysis
Th e outcome parameters were compared by means of Wilcoxon signed-rank test when paired and by means of Mann-Whitney test when unpaired. Diff erences were considered to be signifi cant at the 5% level and Bonferroni correction was applied to multiple comparisons. To analyze the relationship between diff erent sensations and colonic motor patterns, multiple regression analysis was performed with sensations as the dependent variables and motor patterns as the independent variable. Mixed model analysis was used to analyze the eff ect of diff erent recording periods and of different regions of the colon on the MI, with time as a continuous independent variable and region as an independent categorical variable. All data are presented as mean±s.d.

Main study
Th e manometry catheter tip was clipped to the proximal right colon in 8 out of the 10 subjects and to the transverse colon in the remaining subjects. In none of the cases the catheter was displaced at the fl uoroscopic control at the end of the study. Table 1 reports the MI of each region of the colon during the diff erent recording periods. Using mixed models analysis, the MI, which was used as overall measure of the colonic motor activity, increased from recovery to preprandial and to postprandial conditions in each region of the colon ( P =0.003) with a greater eff ect in the left colon ( P =0.04 aft er Bonferroni's correction).

Colonic motor patterns .
During the recording, besides the propagating sequences and non-propagating activity, a never reported colonic motor pattern was noted in each subject and here named pan-colonic pressurization. Low-amplitude propagating sequences: In all but one subject, low-amplitude propagating sequences with a mean total number of 47±39 were recognized during the entire recording period. Th e number of both anterograde and retrograde sequences tended to increase from recovery to preprandial conditions (respectively, 1±2 vs. 8±4 for anterograde, P =0.16; and 0.6±1 vs. 2±2 for retrograde, P =0. 17), and only the number of retrograde propagating sequences was signifi cantly increased in the postprandial period (14±7 for retrograde, P =0.01; and 7±5 for anterograde).
See Table 2 and Supplementary File online for more information on low-amplitude propagating sequences, long-distance lowamplitude propagating sequences, and non-propagating activity. High-amplitude propagating sequence: Only one subject displayed two HAPSs, which occurred 1 h aft er the consumption of the meal ( Supplementary Figure 2C online). Th e two HAPS appeared 90 s apart from each other and their origin was in the middle and distal transverse colon with an extent of propagation, respectively, of 22.5 and 30 cm. Th e mean amplitude of the fi rst was 172 mm Hg and that of the second 196 mm Hg. Both HAPS were associated with the occurrence of strong feeling of desire to defecate. Th e relationship with anal sphincter activity was evaluated only for the fi rst HAPS as during the second the subject requested the commode to defecate. Th e amplitude of anal relaxation was 48% and its duration 30 s. Pan-colonic pressurizations: In all subjects the presence of pressure increases simultaneously occurring in all colonic sensors (pan-colonic pressurizations) and associated with anal sphincter relaxation were identifi ed. Figure 1 shows a part of a tracing with three pan-colonic pressurizations associated with anal sphincter relaxation in one subject. Th e number of the pan-colonic pressurizations was 85±38 per subject during the entire recording time. Th e fi rst pan-colonic pressurization appeared within 40±39 min aft er the start of recording and then they appeared with variable frequencies ranging from 1 every 3-4 min in most of the cases to 1 per minute in some cases ( Figure 2 ). Th e number of pan-colonic pressurizations did not diff er between recovery and preprandial conditions (6±4/30 min vs. 8.6±7/30 min, P =0.57) but it significantly increased during the meal (19±9/30 min, P =0.007) and decreased during the postprandial period (2±2/30 min, P =0.01) as compared with preprandial period. During meal, the frequency of these pan-colonic pressurizations was of 0.83±0.36 events per minute. Th e amplitude and duration of each pressure increases were, respectively, 15±3 mm Hg and 24±4 s, and at detailed analysis, the anal sphincter relaxations started 5±1 s aft er the starting of the pan-colonic pressurization and lasted 23±4 s with a relaxation of 56±23% ( Figure 1 ).
See Supplementary File online for subtypes of pan-colonic pressurizations, relationship between pan-colonic pressurizations and low-amplitude propagating sequences or non-propagating activity, and for anal sphincter activity. Figure 3a shows the average VAS score for the sensation of feeling of gas, desire to evacuate gas, desire to defecate, urgency to defecate, and discomfort/pain reported by each subject during the

Pan-Colonic Pressurizations in Humans
displaced at the fl uoroscopic control at the end of the study. In all subjects pan-colonic pressurizations associated with anal sphincter relaxations were recorded with a mean number of 94±52 per subjects. None of these pan-colonic events was associated with a concomitant increase in the abdominal EMG activity ( Figure 4a ). In the same subjects a mean number of 51±29 artifacts due to cough, movements, and strain were recorded. In all cases these artifacts were associated with simultaneous pressure increases in all the HRM sensors, including those located at the level of the anal sphincter and with concomitant increase in the abdominal EMG activity ( Figure 4a ). Figure 4b shows part of tracing in one subject with artifacts due to movement and one due to strain and with a pan-colonic pressurization. We recorded 50±1 low-amplitude propagating sequences and no HAPS in these subjects.
recording period from the moment they awakened aft er the recovery. As regards to the relationship between sensations and diff erent colonic motor patterns, at multivariate analysis, there was a significant positive relationship between pan-colonic pressurizations and both feeling of gas ( β =0.45±0.21; P =0.03) and desire to evacuate gas ( β =0.60±0.24; P =0.01) ( Figure 3b ). No relationship was found between any of the other sensations and pan-colonic pressurizations, or between any sensations and propagating sequences.

Additional study 1: concomitant evaluation of the colonic motility and of the abdominal wall EMG activity in healthy subjects
Th e manometry catheter tip was clipped to the proximal right colon in two out of the three subjects and to the transverse colon in the remaining subject. In none of the cases was the catheter Considering the low number of propagating sequences recorded during sleeping conditions, the characteristics of these contractions were not reported in the table. The amplitude, duration, and propagation extent did not differ between the different periods of recording both for anterograde (all P >0.20) and retrograde sequences (all P >0.26), and between the two different kinds of sequences (all P >0.10).

Right colon
Pan

Additional study 2: effect of cholinergic modulation on pancolonic pressurizations in healthy subjects
Th e manometry catheter tip was clipped to the proximal right colon in all four subjects. Th e administration of prostigmine did not aff ect blood pressure (106±1.5 vs. 103±2.5 mm Hg for systolic pressure, P =0.33; and 59±3 vs. 59±6 mm Hg for diastolic pressure, P =0.97) or heart rate (61±1.5 vs. 55±2 b.p.m., P =0.08). None of the subjects developed cholinergic side eff ects. In all the subjects pan-colonic pressurizations associated with anal sphincter relaxations were recorded. Th e number of pan-colonic pressurizations did not diff er between the recovery period and the preprandial period, while a signifi cant increase occurred aft er the administration of prostigmine ( P =0.01) ( Figure 5a ). Aft er prostigmine, the majority (72±12%) of the pan-colonic pressurizations was associated with a number of sensations, which attributed a mean VAS score of 59±17 to feeling of gas, 57±24 for desire to evacuate gas, 6±3 for abdominal pain/discomfort, 0.54±0.22 for desire to defecate, and 0 for urgency to defecate. One subject had an episode of fl atus associated with a pan-colonic pressurization at the end of the recording period aft er prostigmine. Figure 5b reports an

Pan-Colonic Pressurizations in Humans
of pan-colonic pressurizations were represented by interrupted ones. Th e majority of low-amplitude propagating sequences were anterograde (75±10%). Supplementary Figure 6 online reports the average VAS score for sensations reported by each patients during the recording period from the moment they awaked aft er the recovery. Aft er a mean of 12±11 min from the intra-colonic administration of bisacodyl, atypical HAPSs as in Figure 6a appeared in three patients, repetitive pan-colonic pressurizations as in Figure 6b,c appeared in two patients, and a combination of repetitive pan-colonic pressurizations and HAPSs in another patient ( Figure 6d ). In two cases the occurrence of repetitive pan-colonic pressurizations coincided with partial expulsion of the catheter ( Figure 6b and d ). Four patients did not show any change in motor pattern in response to the drug. Th e occurrence of repetitive pan-colonic pressurizations was associated with pain while HAPSs were connected with discomfort.
example of the response to prostigmine in one subject. No HAPSs and 20±2 low-amplitude propagating contractions were recorded in the subjects.

Additional study 3: pilot study to evaluate the presence of pan-colonic pressurizations in a group of patients with chronic constipation refractory to pharmacological treatments
Th e manometry catheter tip was clipped to the proximal right colon in eight patients and to the transverse colon in the remaining patient. Th e mean number of both pan-colonic pressurizations and low-amplitude propagating sequences was signifi cantly reduced in patients with chronic refractory constipation as compared with healthy subjects (19±21, P =0.009; and 6±10, P =0.01, respectively), none of these increased aft er the meal and no HAPSs were recorded in this patient group. In four patients no pan-colonic pressurizations were recorded and in one patient 45%

Pan-Colonic Pressurizations in Humans
of the HRM catheter in two cases. Th ese data are in line with previous studies evaluating, by means of the conventional manometry, the motor activity of the rectum and of the anal sphincter or of the colon during prolonged recording periods, which have reported a lower frequency of spontaneous anal relaxations ( 22 ) and of simultaneous colonic pressure rises in patients with chronic constipation compared with healthy subjects ( 32 ). Moreover, they are in line two previous studies, respectively, applying the conventional and the HRM in adult and in children constipated patients, which noted the occurrence of simultaneous pressure increases aft er administration of bisacodyl ( 33,34 ). In the fi rst of the two studies, these pressure increases were also associated with transport of radioisotope ( 33 ). All these fi ndings seem to suggest that pan-colonic pressurizations, which represent a frequent pattern in health, are defective in this subgroup of patients, but in some cases can be elicited by drug stimulation and even participate in the process of expulsion. However, these data need to be confi rmed in larger group of patients, also to assess whether in these patients the pharmacological modulation of pan-colonic pressurizations may represent a new therapeutic target. Th e mechanism underlying the generation of pan-colonic pressurizations associated with anal sphincter relaxations needs to be determined. However, the comparison of their characteristics with those of the contractile events reported by several basic research papers shows that they share many features of the rhythmic contractility described to be generated by the human colon in vitro (35)(36)(37)(38). Th is activity has been suggested to represent a normal response of the colon to even minimal distension and to have a role in facilitating the contact between the colonic wall and the intraluminal content to favor absorption, mixing but also propulsion ( 38 ). With regard to their possible role in propulsion, the fact that in the present study these pan-colonic events overlap with propagating sequences both in anterograde and retrograde directions, suggest that they may have a role in maintaining the colonic wall in a "tonic" contraction and facilitate the movement of colonic content induced by propagating sequences. Th is hypothesis has been suggested by a previous paper combining colon transit scintigraphy and conventional manometry ( 27 ). Moreover, as in previous ( 5,30,33,39 ) and in the present study, both spontaneous and prostigmine-induced pan-colonic pressurizations were correlated with sensation of gas and desire to evacuate gas, and some of these were associated with the actual occurrence of fl atus, these observations may also suggest a potential venting function of pan-colonic pressurizations associated with anal sphincter relaxation similar to what demonstrated for transient lower esophageal sphincter relaxations. In this regard the "tonic" contraction induced by pancolonic pressurizations and the associated anal sphincter relaxation could be suffi cient to induce the movement of gas inside the colon. Given that the colon is main part of the gut where fermentation occurs and it contains about 60% of gas of the gut ( 40 ), it is indeed conceivable that a colonic venting system exists. Ongoing studies combining HRM catheter with impedance segments are evaluating whether these pan-colonic events are indeed associated with gas transport. Finally, even if the present experimental conditions (ambulatory setting and request to score sensations every 15 min)

DISCUSSION
Th e present study describes for the fi rst time the occurrence of a new colonic motor pattern in humans characterized by pancolonic pressurizations associated with anal sphincter relaxation. In the present experimental conditions the pan-colonic pressurizations represent the main colonic motor pattern, increase significantly during the meal and decrease aft erwards, in concert with increased occurrence of mainly retrograde propagating colonic contractions. Th ey diff er from artifacts due to abdominal wall contraction during strain and are signifi cantly increased by the administration of acetylcholinesterase inhibitor prostigmine. In patients with chronic constipation not responding to laxatives, prucalopride, lubiprostone, or linaclotide, the pan-colonic pressurizations are signifi cantly reduced as compared with healthy subjects.
Previous studies applying conventional manometry have reported the occurrence of simultaneous sequences in the colon in healthy subjects ( 21,(25)(26)(27)(28). Th eir presence was to some extent also described in a study combining conventional manometry with colonic barostat measurements, as pressure increases simultaneously involving the intra-colonic barostat balloon in the proximal colon and the manometry recording points below ( 29 ). However, in the past, some authors have attributed these simultaneous pressure increases to the occurrence of artifacts of abdominal wall contraction during strain. In the present study, the concomitant evaluation of colonic and anal sphincter activity has allowed the distinction of these pan-colonic pressurizations, which are associated with anal sphincter relaxation, from artifacts or straining events, which are associated with simultaneous pressure increases also in the sensors located at the anal sphincter level. Th e concomitant evaluation of colonic motility and of abdominal EMG activity has further confi rmed this distinction. Moreover the fact that the number of pan-colonic pressurizations is modulated by the same factors also infl uencing the other colonic motor patterns (meal) and by prostigmine, which has been previously reported to increase colonic tone and induce contractions recorded by the electronic barostat ( 15,16 ), suggest they are generated by the activity of colonic wall.
Previous studies evaluating the motor activity of the rectum and of the anal sphincter during prolonged recording periods have described the occurrence of spontaneous relaxations of the anal sphincter with a frequency in fasting, prandial, and postprandial period similar to the frequency of the anal sphincter relaxations here found to be associated with pan-colonic pressurizations ( 22,23,30,31 ). In some of these studies the association with the presence of pressure increases in rectum was noticed ( 30,31 ), but their association with synchronous pan-colonic motor activity has not been described before.
In the present study we studied the colonic motility and the eff ect of intra-colonic administration of bisacodyl for the fi rst time in a group of chronic constipation patients, not responding to any of the current pharmacological therapies. Th is selected population of patients shows a signifi cantly reduced number of pan-colonic pressurizations, which became the main motor pattern aft er bisacodyl in some patients and was associated with partial expulsion could have enhanced the awareness of these pan-colonic events, which are likely to remain unperceived under physiological conditions, the fact that they can be perceived (as shown more clearly in the present study with prostigmine, and the fact their reduction coincides with almost complete absence of abdominal sensations in the present subgroup of patients) could suggest that they are involved in the induction of abdominal symptoms. Ongoing studies are evaluating their role in the subgroup of patients with constipation and abdominal pain.
With respect to propagating colonic contractions, the data of the present study in healthy subjects confi rm those of previous studies performed with colonic fi ber-optic HRM ( 12,13 ). Th ere are some numerical diff erences in occurrence of motor events in the present paper, which may be related to the meal used (low-caloric in the present study) and to the resolution of the catheter (2.5 cm interval between measuring points as compared with 1 cm with the fi beroptic catheter). On the other hand, the studies applying the fi beroptic catheter, to date, did not report the presence of pan-colonic pressurizations, while recent studies, applying solid-state HRM catheters, also reported the presence of simultaneous colonic pressure rises in children and adults ( 34,41,42 ). With regard to the patients with chronic constipation, the data of the present study are less suitable for comparison with those of Dinning et al. ( 14 ) as these are conducted in diff erent populations (we selected patients not responding to any available pharmacological treatments) and we also included bisacodyl stimulation.
Th e relative short duration of the recording period in the present study as compared with the 24 h measurements reported in literature ( 21,26,28 ), as well as the fact that the measurements have been conducted aft er sedation and colonoscopy and in non-ambulatory subjects, are limitations, which are common to most other colonic manometry studies ( 13,14,25 ). Another potential limitation lies in the recording of the anal sphincter activity by means of 2.5 cm spaced sensors, which in theory may not allow the discrimination between real anal sphincter relaxations and movement of the catheter. However, the fact that the subjects did not move during the study, that the catheter was fi xed to the buttock during the study, and that the anal sphincter pressure decreases and spontaneously comes back to pre-relaxation pressure values all argue in favor of a physiologically relevant and accurate measure. Moreover, a recent study, only published in abstract form and applying HRM to the anorectum, has reported events with similar characteristics of rectal pressurizations and anal sphincter relaxations ( 43 ).
In conclusion, the present study demonstrated for the fi rst time the occurrence of pan-colonic pressurizations associated with relaxations of the anal sphincter as a new colonic motor pattern, which seems to be defective in patients with chronic constipation not responding to any pharmacological treatment. Th e mechanism behind the pan-colonic pressurizations as well as their function remains to be determined, but they are the main colonic motor pattern, increase during meal and with cholinergic stimulation. Pan-colonic pressurizations associated with relaxations of the anal sphincter may have a role in the transport of colonic gas and in the facilitation of the propagating sequences-induced colonic transport.