Pancolonic spatiotemporal mapping reveals regional deficiencies in, and disorganization of colonic propagating pressure waves in severe constipation


Philip Dinning, Department of Gastroenterology, St George Hospital, Kogarah, NSW 2217, Australia.
Tel: +61 2 91132817; fax: +61 2 91133993;


Background  The morphology, motor responses and spatiotemporal organization among colonic propagating sequences (PS) have never been defined throughout the entire colon of patients with slow transit constipation (STC). Utilizing the technique of spatiotemporal mapping, we aimed to demonstrate ‘manometric signatures’ that may serve as biomarkers of the disorder.

Methods  In 14 female patients with scintigraphically confirmed STC, and eight healthy female controls, a silicone catheter with 16 recording sites spanning the colon at 7.5 cm intervals was positioned colonoscopically with the tip clipped to the cecum. Intraluminal pressures were recorded for 24 h.

Key Results  Pan-colonic, 24 h, spatiotemporal mapping identified for the first time in STC patients: a marked paucity of propagating pressure waves in the mid-colon (P = 0.01), as a consequence of a significant (P < 0.0001) decrease in extent of propagation of PS originating in the proximal colon; an increase in frequency of retrograde PS in the proximal colon; a significant reduction in the spatiotemporal organization among PS (P < 0.001); absence of the normal nocturnal suppression of PS.

Conclusions & Inferences  Pan-colonic, 24 h, spatiotemporal pressure mapping readily identifies characteristic disorganization among consecutive PS, regions of diminished activity and absent or deficient fundamental motor patterns and responses to physiological stimuli. These features are all likely to be important in the pathophysiology of slow transit constipation.


Constipation, characterized by infrequent and/or difficult defecation, is a common disorder affecting 12–19% of the general population.1 In severe slow transit constipation (STC), the principal pathophysiological mechanism is thought to be dysfunctional or deficient colonic propulsive motor patterns.2 Motor activities of the large bowel, unlike those in the proximal gut, must cater for: prolonged storage in order to facilitate absorption of water and electrolytes; slow and stepwise transport of fecal content; and relatively infrequent evacuation of a substantial proportion of its total content. These discrete functions are likely to be associated with specific colonic motor patterns, the measurement of which is best accomplished using colonic manometry. In health, manometry has identified propagating activity, defined as propagating sequences (PS; alternatively called propagating contractions: PC), and confirmed that these motor patterns are temporally linked to defecation3 and intra-luminal transit of colonic content.4,5

The characteristics of PS subserving distinct colonic functions appear important: PS associated with defecation tend to be higher in amplitude and generally display a characteristic stereotypical relationship among consecutive PS.3,6 Nevertheless, low amplitude PS can at times move stool substantial distances along the colon.4,5 More recently, regional variation in PS activity4,6,7 has been demonstrated. Also, through the novel application of spatiotemporal mapping,8 which readily permits visual interpretation of colonic motor patterns, consecutive PS have been shown to be linked in an organized fashion, presumably assisting with continuity of flow.9

In STC, relatively few manometric studies have been published since the first report two decades ago.10 The majority have only recorded motor activity from sites distal to the mid-transverse colon, and many are confined to the descending or sigmoid colon only.11–13 From these studies, alterations in PS frequency, and specifically decreased frequency of high amplitude PS (HAPS; alternatively called high amplitude propagating contractions) has been implicated in the pathogenesis of STC.10,14–17 However, an emphasis solely on indices of PS frequency and amplitude may be simplistic, given those contemporary studies which suggest that spatiotemporal organization of PS activity9 may be as important if not more relevant to transit and stool expulsion.

The morphological characteristics, responses to physiological stimuli and spatiotemporal organization among PS have never been defined throughout the entire colon of patients specifically with STC. This study aimed to establish detailed spatiotemporal maps of PS activity from the cecum to the anorectum in STC. Specifically, we hypothesized that: (i) derangements in motor patterns underpinning STC are multifactorial and that deficient, disrupted or disorganized spatiotemporal patterning among consecutive colonic PS exists in this condition; and (ii) pan-colonic, 24 h spatiotemporal pressure mapping reveals one or more recognizable manometric ‘signatures’ that can serve as biomarkers of the disease.


Patients and controls

As severe slow transit constipation in adults almost exclusively affects females,18 the study included only female subjects. Pan-colonic manometry was carried out in 14 patients with STC (mean age 43 ± 16) at St George Hospital, Sydney, Australia. Because of the inability of the NSW Health Department to allocate resources to cater for lengthy inpatient physiological studies in healthy humans, a collaboration was set up with the Barts and The London School of Medicine and Dentistry, London, UK where control data were acquired from eight healthy female volunteers (mean age 37 ± 11). The London-based studies used the same catheters and measurement techniques. The initial two studies were supervised by the primary author to ensure adherence to the protocol.

Patient selection

Eligibility criteria  Patients participating in the study were referrals to St George Hospital, Sydney, for management of intractable constipation. For inclusion in the study, patients had to fulfill all of the following: (i) female, aged 18–75; (ii) Rome III criteria for constipation;19 (iii) passage of a ‘complete’ bowel movement on less than 3 days per week, for at least 2 of 3 weeks. A complete bowel movement was defined as expulsion of stool following which the individual felt that evacuation was complete.20 These data were derived from 3-week stool diaries that detailed, on a daily basis, stool frequency and form,21 and sense of complete evacuation (yes/no); (iv) delayed colonic transit confirmed by isotope colonic transit study;22,23 (v) normal anorectal manometry, with no evidence of strain-related paradoxical sphincter contraction or an inability to expel a rectal balloon;24 (vi) normal evacuation proctogram, with no functional (e.g. pelvic floor dyssynergia) or anatomical (e.g. significant rectocele with retention of contrast, occluding intussusception) impediment to the expulsion of the radio-opaque contrast;25 (vii) failed symptomatic response to standard therapies including laxatives, dietary modification and exercise; and (vii) normal colonoscopy within 5 years of enrollment (with the exception of colonic melanosis coli and non-malignant colonic polyps).

Ineligibility criteria  These includes the following: (i) metabolic, neurogenic or endocrine disorder(s) known to cause constipation (e.g. hypercalcemia, hypothyroidism, diabetes, multiple sclerosis, Parkinson’s, scleroderma); (ii) consumed drugs which list constipation as a potential side effect deemed to be clinically relevant by the referring physician (e.g. calcium channel blockers); (iii) prior abdominal radiotherapy; (iv) prior abdominal surgery (except cholecystectomy, appendicectomy, inguinal hernia repair); (v) current or planned pregnancy; and (vi) current or prior history of malignancy.


All healthy subjects had a normal bowel habit, defined as between three bowel movements a day and one bowel movement every 3 days, with no symptoms of rectal evacuatory difficulty or any other gastrointestinal symptoms. None had a history of metabolic, neurogenic or endocrine disorder(s) known to cause constipation, were not taking regular medications including laxatives, or had a history of prior abdominal surgery, other than appendicectomy. Pregnancy was excluded in all subjects prior to enrollment by urinary human chorionic gonadotropin (HCG) testing. Studies were not performed in any particular phase of the menstrual period. All participants had given written, informed consent and the studies were approved of in Australia by the Human Ethics Committees of the South Eastern Area Health Service, Sydney and the University of New South Wales (05/122), and in London by the Redbridge & Waltham Forest Local Research Ethics Committee (07/H0701/71).

Colonic manometric technique

The water-perfused catheters used in these studies have been described in detail previously.3,26 Briefly, in all studies, a 21 lumen (16 recording sideholes, spaced at 7.5 cm intervals), 4.5 m long extruded silicone-perfused manometric catheter assembly (Dentsleeve, Wayville, SA, Australia) was used. The catheter was rendered radio-opaque with a barium-filled core. Each recording lumen was perfused with distilled water driven by a low compliance pneumohydraulic perfusion pump at 0.15 mL min−1 (Dentsleeve). At St George Hospital, pressures were measured with 16 external pressure transducers (Abbott Critical Care Systems, North Chicago, IL, USA), with recorded signals digitized at 10 Hz by preamplifiers (AqcKnowledge III Software, BIOPAC Systems, Inc., Santa Barbara, CA, USA). In London, this was achieved in a similar fashion through a customized modular manometric system (Solar Measurement System; software version 8.7b; Medical Measurement Systems, Enschede, The Netherlands).

Experimental protocol

On the day prior to colonoscopic insertion of the manometry catheter, patients were allowed clear fluids only and advised to take their usual laxatives. Colonic preparation followed institutional policy; in patients, by oral administration of 2 L of a polyethylene glycol (Golytely; Braintree Laboratories, Braintree, MA, USA), and in healthy controls by oral administration of two Bisacodyl tablets and 250 mL magnesium citrate.

In all subjects on day 1, after an overnight fast, and under conscious sedation with intravenous fentanyl (50–100 mg i.v.) and midazolam (2–5 mg i.v.), the catheter was advanced to the cecum under colonoscopic guidance, pulled by a nylon loop held in an endoscopic snare. In all patients and controls, bowel preparation had led to an empty colon. The nylon loop on the catheter tip was secured to a cecal fold using two hemoclips (Olympus America, Melville, NY, USA)27,28 and the colonoscope removed, extracting as much air as possible.

After recovery from sedation (1–2 h), all subjects were transferred to a private room where they were free to move for the remainder of the day, fed a standard meal at 18:00 and slept overnight. Recordings were not commenced until 08:00 the following morning (day 2), at least 40 h after ingestion of purgatives, and 22 h after catheter placement, to eliminate the impact of administered drugs and of air insufflated during the endoscopic procedure on manometric measures. Recording was continued until the following morning (day 3), following which the catheter was removed by gentle traction, as described previously.26,28 In patients, all laxatives were discontinued throughout the study period.

On day 2, subjects ate standard meals at 09:00 (breakfast: 300 kcal, 15% protein, 34% fat, 51% carbohydrate) and at 12:00 and 18:00 (lunch and dinner: 1000 kcal, 24% protein, 43% fat, 33% carbohydrate). Subjects were instructed to refrain from eating between meals, and were encouraged to consume each meal within 20 min. During the daytime, subjects were told not to sleep, and either sat semi-erect in bed, or in a chair, where they were allowed to read, watch television, or use a personal computer. If defecation occurred, subjects used a commode, but remained attached to the recording system and the time was recorded.

Data analysis

To eliminate the possibility of inter-observer variation in the labeling of PS and HAPS, the principal author (P.G.D.) analyzed the data in all subjects.

Propagating sequences: definitions, classification and characteristics  As described previously,3,7 for the purpose of analysis, we divided the colon into 16 regions (region 1 = cecum, region 4 = hepatic flexure, region 8 = splenic flexure, region 12 = proximal sigmoid colon, region 16 = rectum). Recording sideholes were assigned to the colonic region within which they lay, with sidehole 1 always in region 1 as it was clipped to the cecum and the final sidehole in the rectum/anus. Definitions of PS and HAPS have also been described in detail previously.3,7 Briefly, a PS was defined as an array of three or more pressure waves recorded in adjacent recording sites in which the conduction velocity between the onset of the major upstroke of each wave within that sequence lay between 0.2 and 12 cm s−1. If these criteria for propagation were met, the individual pressure waves within the sequence were termed propagating pressure waves (PPW). The frequency of PPW represents the ‘density’ of the component PS pressure waves per colonic region.

Propagating sequences were further qualified by the terms antegrade or retrograde, depending upon the direction of propagation. A PS was classified as a HAPS if the amplitude of at least one component propagating pressure wave was >116 mmHg as defined previously in healthy subjects.3,7

Spatiotemporal pressure mapping and quantification of spatiotemporal organization among propagating sequences  Spatiotemporal maps of antegrade and retrograde PS activity for each subject over 24 h were created. The details of construction of these maps have been described in detail elsewhere.8

We have previously reported a phenomenon that characterizes the normal relationships among consecutive PS, which we have termed ‘regional linkage.’9 A PS was deemed regionally linked to the PS immediately preceding it if the two PS originated from different colonic regions but where the segments of colon traversed by each PS overlapped.

Diurnal variation in propagating sequence frequency  The antegrade and retrograde PS and HAPS frequency per hour in the 8 h epoch between 22:00 and 06:00 (nocturnal period) was compared to the frequency per hour of these events recorded between 06:00 and 22:00. These epochs were chosen because they matched the epochs used in previous analyses.7

Colonic meal response  The 2-h epoch prior to and after the 1000 kcal lunch on day 1 was subdivided into four 30 min periods.7 In each of these periods PS and HAPS were detailed and compared.

Defecation  The PS and HAPS associated with any episodes of spontaneous defecation were flagged for subsequent analysis. As determined in previous studies, in which we systematically examined the temporal relationship between PS and stool expulsion, a PS or HAPS was defined to be associated with defecation if it occurred within 20 min prior to stool expulsion.3,6

Statistical analysis

A Mann–Whitney U-test was used to make inferences about potential differences regarding PS and HAPS characteristics (frequency, amplitude, velocity, site of origin, and extent of propagation) between patients and controls. The comparisons between these variables were made for the total colon, and for the proximal colon (ascending and transverse colon) and distal colon (descending and sigmoid colon). A non-parametric test for between-group comparisons was used because the normal distribution of the data was not always apparent. The same test was used to compare the regional linkage that existed between the two groups. Comparisons between the basal and the postprandial PS characteristics within subjects were performed using a paired t-test. Comparison between the delta values (basal–postprandial) between the patients and the control groups were performed using the Mann–Whitney U-test. All data are expressed as mean ± SD. A P-value of less than 0.05 was considered statistically significant.


Symptom duration and isotope retention

The age, duration of symptoms, patient satisfaction with weekly bowel motions and isotope retention at 72 h are detailed in Table 1.

Table 1.   Patient characteristics. Feeling of complete evacuation and laxative use were obtained from the patients’ 3 weeks stool diary. Isotope retention was calculated from the entire colon on day 3 (normal <9%)
Patient No.AgeDuration of symptoms (years)Feeling of complete evacuation (days/week)Isotope retention (72 h)

Pan-colonic manometry: 24 h spatiotemporal organization of propagating sequences

In all patients and controls, colonoscopic catheter placement was achieved without complication. Twenty-four hour manometric studies were completed in all subjects. Some subjects upon catheter removal reported a transient lower abdominal ‘unease,’ but there were no complications reported.

Twenty-four hour spatiotemporal maps revealed consistent and striking differences in frequency, distribution, extent and polarity of colonic propagating sequences in patients with STC compared with controls (Figs 1 and 2). Although the frequency of antegrade PS in the proximal and distal colon was similar between groups, there was a notable increase in the frequency of retrograde PS in patients. While substantial numbers of PS originated both in the proximal and distal colon of patients, these events only propagated over short distances, manifest as loss of regional linkage among PS along the colon, with a virtual absence of propagating pressure waves within a relatively adynamic mid-colonic zone. The proportion of regionally linked antegrade PS in patients (40 ± 7%) was significantly lower when compared with controls (59 ± 9%; P < 0.001). The patient group also demonstrated a marked reduction in the amplitude of pressure waves throughout the colon. Additional features readily appreciated from spatiotemporal maps of the STC patients included lack of the normal nocturnal suppression of PS and the absence of the normal meal response.

Figure 1.

 Twenty-four hour, pan-colonic spatiotemporal maps of colonic propagating sequences (PS) in (A) a healthy control and (B) a female patient with STC. Within each map each colored ridge represents an antegrade (green) or retrograde (red) PS. The antegrade PS originate at the orad end of the green ridges, and retrograde PS originate at the anal end of the red ridges. The proximal margin of each antegrade and retrograde ridge indicates the precise site and time of origin of that PS. The axial length of the ridge indicates the extent of propagation of that particular PS. On each map the yellow-hatched line indicates the timing of the 1000 kcal lunch; the white-hatched line denotes the time of defecation; and the blue-hatched line represents the location of the mid-colon (splenic flexure). The pink shading highlights the nocturnal period (22:00–06:00). The features immediately apparent from these spatiotemporal maps that distinguish patients from controls are: the marked paucity of PS in the mid-colon as a consequence of a significant decrease in the extent of propagation of antegrade PS originating in the proximal colon; a lack of HAPS; an increased frequency of retrograde PS in the proximal colon; a lack of nocturnal suppression of PS; and an absent meal response.

Figure 2.

 Twenty-four hour spatiotemporal maps in four healthy controls and four patients with slow transit constipation (STC). The abnormalities in propagating wave characteristics that have been detailed in Fig. 1 are displayed in each of the patients with STC.

Diurnal variation in propagating sequences

All controls demonstrated normal nocturnal suppression of antegrade PS, with a mean decrease of 54 ± 26% from a daytime frequency of 5 ± 3 PS h−1 to a night time frequency of 2 ± 2 PS h−1; (P = 0.01) (Figs 1 and 2). This pattern was not displayed in patients (day: 6 ± 4 vs night: 6 ± 4 PS h−1; P = NS) (Figs 1 and 3).

Figure 3.

 Frequency of antegrade propagating sequences per hour in each of the controls (n = 8) and patients (n = 14) during day (06:00–22:00) and night (22:00–06:00) periods. In healthy controls, there was a significant decrease (P < 0.01) in propagating activity at night. This pattern was not observed in the majority of patients.

Propagating sequence characteristics

The overall frequency of antegrade PS did not differ between groups (Table 2). When compared with controls, the amplitude of PS in patients was significantly lower (P < 0.0001) (Fig. 4A; Table 2) and the extent of propagation of PS originating in the proximal colon was significantly reduced (P = 0.0007) (Fig. 4A; Table 2). As noted above the majority of PS originating in the ascending colon of patients did not extend beyond the distal transverse colon (Figs 1 and 4A) resulting in a markedly reduced density of PPW in the mid-colon of patients (52 ± 24 PPW/24 h) in comparison with controls (123 ± 52 PPW/24 h; P = 0.01).

Table 2.   Antegrade and retrograde propagating sequence characteristics
 Control (n = 8)Patient (n = 14)
Right ColonLeft ColonTotalRight ColonLeft ColonTotal
  1. For ease of presentation all significant differences are presented as P < 0.05.

  2. *A significant difference (P < 0.05) between the left and right colon within controls or patients.

  3. A significant difference (P < 0.05) between control and patient groups for the same region.

  4. A significant difference (P < 0.05) between antegrade and retrograde characteristics within controls or patients for the same region.

Antegrade Propagating SequencesFrequency/24 h58 ± 34*38 ± 3096 ± 6072 ± 4462 ± 43135 ± 72
Propagating pressure wave frequency/24 h258 ± 153207 ± 151465 ± 284243 ± 158236 ± 157479 ± 251
Amplitude (mmHg)62 ± 14*78 ± 2268 ± 1531 ± 832 ± 631 ± 6
Extent of propagation (cm)34 ± 6*21 ± 229 ± 521 ± 617 ± 220 ± 4
Retrograde Propagating SequencesFrequency/24 h13 ± 1215 ± 1727 ± 2456 ± 6721 ± 2178 ± 74
Propagating pressure wave frequency/24 h59 ± 5365 ± 61101 ± 95197 ± 24048 ± 29246 ± 246
Amplitude (mmHg)29 ± 534 ± 832 ± 627 ± 626 ± 527 ± 4
Extent of propagation (cm)18 ± 320 ± 319 ± 319 ± 320 ± 418 ± 2
Figure 4.

 Regional variation in the frequency, amplitude and extent of propagation of antegrade propagating sequences (PS) overall (panel A), and of high amplitude PS (panel B) in controls (n = 8) and patients (n = 14). The vertical bars show the frequency distribution of PS grouped according to the site of origin. The horizontal bars show the mean extent of propagation according to the site of origin. The solid blue (control) and red (patient) lines indicate the mean amplitude of the component pressure waves at each colonic region. (A) When compared with controls, note the significant reduction in pressure wave amplitude (P < 0.0001) and the extent of propagation of PS throughout the colon in patients (P = 0.001). (B) High amplitude PS frequency is markedly reduced in patients (P < 0.0001). High amplitude PS generated in the proximal colon of patients extend significantly shorter distances along the colon when compared with controls (P < 0.0001).

When compared with controls, the overall frequency of retrograde PS was significantly higher in patients (P = 0.03) (Figs 1 and 2; Table 2). This was particularly evident in the proximal colon of patients, where a fourfold increase in retrograde PS was observed in comparison with control subjects (56 ± 67 vs 13 ± 12/24 h in controls; P = 0.04) (Figs 1 and 2; Table 2). There was no significant difference in the amplitude, velocity or extent of propagation of retrograde PS between patients and controls (Table 2).

High amplitude propagating sequences

Antegrade HAPS were recorded in all controls but in only 10 (71%) patients (Figs 1 and 2). In controls, HAPS were initiated five times more frequently in the proximal colon than in the distal colon (P < 0.0001) (Figs 1 and 4B). This regional difference was not seen in patients. Overall, patients exhibited a significant reduction in the frequency (P < 0.0001), amplitude (P < 0.0001), and extent of propagation of HAPS (P < 0.0001) (Figs 1 and 4B) when compared with controls (Table 3).

Table 3.   High amplitude propagating sequence characteristics
Right ColonLeft ColonTotalRight ColonLeft ColonTotal
  1. For ease of presentation all significant differences are presented as P < 0.05.

  2. *A significant difference (P < 0.05) between the left and right colon within controls or patients.

  3. A significant difference (P < 0.05) between control and patient groups for the same region.

  4. Note that a PS was classified as a HAPS if the amplitude of at least one component propagating pressure wave was >116 mmHg. Many pressure waves in the sequence do not exceed 116 mmHg and this reduces the average amplitude.

Frequency/24 h17 ± 7*3 ± 321 ± 8.22 ± 32 ± 44 ± 6
Amplitude (mmHg)104 ± 22147 ± 29125 ± 1975 ± 3268 ± 1278 ± 21
Extent of propagation (cm)43 ± 6*21 ± 438 ± 536 ± 1520 ± 1633 ± 16

Colonic meal response

A postprandial increase in HAPS frequency was recorded in seven of eight controls but in only 2 of 14 (14%) patients (P = 0.003) in response to a 1000 kcal meal (Figs 1 and 2). One healthy control defecated immediately prior to lunch and this event was preceded by HAPS. In this subject, HAPS frequency was the same prior to and after the meal.

The meal-related increase in HAPS frequency seen in controls (3.3 ± 2.8 HAPS/2 h) was significantly blunted in the patient group (0.1 ± 0.7 HAPS/2 h; P = 0.01). The increase in HAPS frequency observed was not specific to any particular 30 or 60-min postprandial epoch.


No patient defecated during the recording period, whereas all controls did (P < 0.0001), mostly passing moderate amounts of watery stool.26 Timing of defecation in one volunteer was unknown based on diary entries, and this subject was removed from this analysis only. Twenty episodes of defecation were recorded in the remaining seven volunteers, and stool expulsion was deemed to be directly associated with 54 (35%) of the 154 HAPS recorded in these subjects (Figs 1 and 2).


Using a novel technique to visually display the spatiotemporal distribution of PS throughout the entire colon over a 24-h period, this study has demonstrated a number of potentially important new phenomena in patients with STC. To our knowledge, this is also the first study to report abnormalities of motor patterns in the proximal colon of such patients. A major finding is a relatively adynamic region around the splenic flexure, which is somewhat ‘disconnected’ from the adjacent proximal and distal colon, and in which there is a marked paucity of propagating pressure waves. This phenomenon is a consequence of a marked reduction in the extent of propagation of antegrade PS in the proximal colon, and poor regional linkage among consecutive PS throughout the colon. Additional important new findings from this study include (i) an increased frequency of proximal colonic retrograde PS; and (ii) absence of the normal nocturnal suppression of antegrade PS in patients with STC. Finally, this study also confirms a number of previously reported findings: (i) an absent meal response in most patients;14,16,29,30 and (ii) reduced frequency (absent in 30%), extent and amplitude of HAPS.10,14–16.

Colonic PS are important for luminal propulsion and defecation,3–5 these marked disturbances in distribution, polarity and organization of PS are likely to help explain the delay in colonic transit and may represent important markers of dysregulated colonic motor function in patients with STC. For example, the patients demonstrated a significant increase in proximal colonic retrograde PS frequency, a characteristic we have shown previously in patients with symptomatically defined obstructed defecation.6 As these motor patterns are capable of propelling content orally,5 this increased retrograde frequency coupled with the short extent of propagation of antegrade proximal colonic PS and the overall reduced linkage between proximal and distal colonic PS may ultimately result in the delayed emptying of the proximal colon observed in these patients.

Severe constipation is a chronic condition with major morbidity and associated healthcare burden.1 In an attempt to elucidate the pathophysiological mechanisms underlying chronic constipation, measurement of in vivo colonic motor function is fundamental. However, direct assessment (as opposed to indirect measurement through transit studies) of human colonic motility poses substantial methodological challenges. Hence, our understanding of the physiology of motor patterns and pathophysiology of constipation remains relatively primitive. While ‘colonic’ dysfunction has been discussed in previous publications,11–13 for the most part, they only detailed activity from sites distal to the splenic flexure. We have shown previously in healthy subjects that there is regional variation in the distribution and frequency of PS between the proximal and the distal colon,6,7 and also co-ordination of PS activity across these regions.9 Pan-colonic manometric assessment is thus essential to determine similar parameters in patients with STC.

The underlying cause of colonic dysmotility displayed in these patients is unknown; however, central factors are candidate mechanisms. In contrast to previous studies,14,16 our data indicate that the normal nocturnal inhibition of antegrade PS is largely absent in STC patients. As it is likely that the diurnal variation in motor patterns is modulated by the central nervous system, and that the waking-related stimulation of PS is immediate,31 the observed attenuated sleep response may indicate a central neuropathic cause in such patients.16.

The discrepancy in nocturnal findings between our study and previous ones is likely to reflect the differences in recording technique. For example, Rao et al. demonstrated nocturnal inhibition of colonic motor patterns,16 however, that study reported no data from the ascending colon. The study used a generalized ‘area under the curve’ measurement to define motility at night, and did not measure individual PS. In the study by Hagger et al.,14 data were recorded from the ascending colon, however, their catheter had only five sensors spaced at 15 cm and such spacing would miss the majority of all propagating activity recorded here. Perhaps because of this a Motility Index was used to define nocturnal activity. Hence, neither study is directly comparable to ours.

Additional mechanisms that may underpin the dysmotility displayed by these patients include feedback inhibition from a distal colonic ‘obstruction,’ and intrinsic dysregulation. While feedback mechanisms secondary to distal ‘obstruction’ (by a retained mass of hard stool) may inhibit proximal colonic motility in the unprepared colon, this is an unlikely explanation for the findings of the current study, because prior bowel preparation had removed solid colonic content. An intrinsic dysregulation is a stronger possibility. In constipated patients, a diminished density of interstitial cells of Cajal has been implicated,32 as has a decrease in the overall population of glial cells.33 Such defects could feasibly account for the striking truncation of propagation of PS and the spatiotemporal disorganization among consecutive PS.9

The reason for the marked quiescence of propagating activity around the splenic flexure is unclear, but it is interesting to note that this represents the junction between the midgut and the hindgut.34 These two embryonically distinct regions have different blood and neural supplies and have been shown to differ in the expression of genes and antigens.35,36 In patients with STC, this region may represent a site of disrupted neural supply. This may be of particular etiological importance in those who report constipation from early childhood (all in this study). Whether similar dysmotility is present in patients who develop their symptoms of STC after pelvic surgery or childbirth37 remains unknown.

The ability of any test of colonic function, including colonic manometry, to help identify specific biomarkers of disease that can differentiate sub-types of constipation and ultimately guide and improve treatment and predict outcomes in these patients still remains elusive. The recent American Neurogastroenterology and Motility Society consensus paper38 stated that There are no published quantitative data of phasic contractility that unequivocally differentiate normal colonic function from colonic inertia. Although we report highly significant quantitative differences between patients and controls, that consensus statement remains true. Notwithstanding, the possibility remains that subsets of patients identified by intraluminal manometry (e.g. with combinations of these identifiable phenomena such as poor linkage, absent nocturnal suppression and absent HAPS meal response) might be predictive of underlying pathology or therapeutic outcome.9,16

There are a number of potential criticisms with this study. These include age differences, and the possibility of geographic confounders being introduced as the studies were carried out in different sites. We have recently addressed these issues in detail.26 Protocols (except bowel preparation) and equipments were standardized between the groups, and the principal author conducted studies at both sites. Although presence or absence of stool is known to influence recorded motor activity,26 all subjects in this study were seen to have an empty colon at colonoscopy, and given the time allowed before recordings commenced, it is improbable that any differences recorded between the groups were due to differences in bowel preparation.

This study has demonstrated the utility of a novel technique to condense and display in a readily interpretable format, a number of new and potentially important disturbances in the spatiotemporal organization of PS in patients with STC. Given the importance of PS to normal transit and defecation, these abnormalities are pathophysiologically relevant. Whether these markers of dysmotility, or combinations of them, might prove to be useful as true disease ‘biomarkers’ will require further systematic evaluation.

Acknowledgments and disclosures

Philip Dinning, Michal Szczesniak and Linda Hunt are supported by NH&MRC, Australia. The authors have no competing interests to declare.

All authors have approved the submitted draft. Phil Dinning was involved in study concept and design, analysis and interpretation, draft and critical review of manuscript; Linda M Hunt and Sergio Fuentealba were involved in acquisition of data in Sydney; Natalia Zarate was involved in acquisition of data in London and manuscript review; Sahar Mohammed was involved in acquisition of data in London; Michal Szczesniak developed spatiotemporal maps and analysis of data; David Lubowski, Sean Preston, and Peter Fairclough were involved in placement of colonic catheters in Sydney and London and critical review of the manuscript. Peter Lunniss and Mark Scott were involved in study supervision in London and critical manuscript review. Ian Cook was involved in study supervision in Sydney and critical manuscript review.