- Top of page
Colonic motility disorders are common conditions. However, our understanding of normal, and, consequently, pathological motor function of the colon remains limited, mainly due to the relative inaccessibility of this organ for study. Investigation of colonic motility may encompass one or more of the four separate components (myoelectric activity, phasic and tonic contractile activity and movement of intraluminal content) using electrophysiological, manometric or transit studies. Although transit studies provide the best ‘functional’ appreciation of colonic motor activity, and are the only techniques used in contemporary clinical practice, manometric methods are becoming increasingly popular, as they allow a direct study of colonic contractile activity over prolonged periods. To date, the majority of studies have been limited to the pelvic colon by a retrograde (per rectal) approach; however, recent technological advances have facilitated ‘pan-colonic’ investigation. This review concentrates on manometry of the human colon proximal to the sigmoid, and includes evaluation of both phasic and tonic motor activity, by utilization of perfused-tube and solid-state manometric catheters, and also the electronic barostat. Methodological techniques, experimental protocols and the analysis and interpretation of recorded data are critically explored, and a contemporary classification of colonic contractile activities is presented.
Colonic functions and motility
- Top of page
The mammalian colon has evolved to perform six major functions:
absorption of water and electrolytes;
absorption of short-chain fatty acids and other bacterial metabolites produced by resident symbiotic micro-organisms;
salvage of carbohydrates and bile salts by absorption;
propulsion of the colonic contents in a net aborad direction;
storage of the residual (waste) faecal material up to the time of defaecation;
rapid emptying of a variable part of the colon during defaecation.
Appropriate motor activity is required to sub-serve these functions. The motility patterns are complex and variable. Firstly, co-ordinated activity between the terminal ileum, caecum and proximal colon is needed to deliver chyme from the distal small bowel into the large bowel in a manageable way. Colonic contents are then transported aborally towards the rectum for eventual evacuation. To aid maximal absorption during this time, the principal motility requirement is slow propulsion with extensive mixing, to allow uniform contact with the colonic mucosa. The growth of microflora is, in turn, facilitated by the slowness and orientation of the mixing movements. Intraluminal contents become progressively more solid as most of the water is absorbed, and colonic motor activity must therefore have the ability to mix and propel materials of a more viscous consistency than that of the small intestine, whose contents are mostly liquid. Finally, specific co-ordinated motor activities are required for the temporary storage of faecal material and its rapid, semi-voluntary expulsion (defaecation).1–4
Is there a clinical need for tests of colonic motility?
- Top of page
Disturbances of colonic motor activity may interfere with colonic function, thus disrupting normal patterns of defaecation. Some diseases characterized by disordered colonic motility are highly prevalent conditions: faecal incontinence and constipation afflict 3% of the general population,5–7 whereas irritable bowel syndrome accounts for 50% of patients referred to Gastroenterology Outpatient Clinics, and affects ∼20% of the population;8,9 diverticular disease is estimated to affect 30% of the population over 60 years of age and 60% of those over 80.10 Others, although less common, may be incapacitating and cause untold suffering (e.g. chronic intestinal pseudo-obstruction, functional diarrhoea). Although many patients will respond to simple therapeutic measures aimed at treating their symptoms, the pathophysiology of such ‘colonic motility disorders’ remains poorly understood. In those patients with intractable symptoms, clinical tests are indicated to try to elucidate the cause of colonic dysfunction, and perhaps also to define patterns of bowel dysfunction in sub-groups of patients in order to better direct therapeutic strategies. Currently, however, our understanding of motor activity of the large intestine lags far behind that of the upper gastrointestinal (GI) tract. This is related to the relative inaccessibility of the organ for study (the majority of studies have, to date, been performed in the left-side only),11–26 the unappealing nature of its contents, regional differences in colonic structure and function, and the somewhat confusing and contradictory data yielded from the multiplicity of techniques that have been employed historically.
Measurement of colonic motility
- Top of page
Colonic motor activity involves four main components:
(underlying) myoelectric activity;
phasic contractile activity;
tonic contractile activity;
Specific methods are available for the assessment of each separate component (Table 1), but no single investigation gives information regarding all four types of activity. Tests may be combined, however, to provide integrated information (i.e. simultaneous measurement of intraluminal pressure changes and movement of intracolonic contents).27–29
Table 1. Methods available for the investigation of colonic motor activity in man
|Phasic contractile activity||Intraluminal manometry|
|Barostat volume recording|
|Tonic contractile activity||Barostat volume recording|
|Transit||Radio-opaque marker studies|
Investigation of colonic myoelectric activity by electromyography (EMG), although popular in the 1970s and 1980s,15–17,30–32 has been superceded by other techniques and is now rarely used.33 Electromyography recordings, yielded via tube-mounted electrodes in contact with, or attached acutely to the colonic mucosa under endoscopic control, do have the advantage of ease of application and relative inexpense. However, controversy exists as to how easily complex EMG signals can be interpreted in terms of colonic wall movements or changes in intraluminal pressure. Although some studies of colonic motility using combined manometry/EMG have shown that nearly 100% of myoelectric ‘spike bursts’ were associated with mechanical activity of the colonic wall,16,31 other authors have disputed such a clear association,32,34 and reported poor correlation between recorded electrical and mechanical events.32 Furthermore, EMG recordings are subject to interference from other electrical sources, both external to the body or internal (e.g. ECG activity), and artefact or variation in signal strength caused by probe movement or displacement from the mucosa.1,34,35 Human colonic electrical activities are reviewed in various other articles.1–3,35
Evaluation of colonic transit is an indirect measure of the contractile activities effecting movement of intraluminal content. In current clinical practice, two techniques exist for the assessment of colonic (or whole gut) transit, both of which involve irradiation of the subjects: radio-opaque markers36–38 and radionuclide scintigraphy.39–42 These investigations are used primarily in patients with constipation,43 and provide objective data regarding the subjective complaint of infrequent defaecation. A simple radio-opaque marker study, with plain abdominal film(s) taken 3–5 days later, is adequate for detecting transit abnormalities.36–38 However, for accurate assessment of segmental colonic transit, scintigraphy is required.39–41 A comprehensive description of such methods is beyond the scope of this manuscript; these can be found in detail elsewhere.43–46
Although transit studies probably provide the best ‘functional’ appreciation of colonic motility (i.e. movement of intracolonic contents), the relative radiation risks (however small) associated with these investigations can be avoided by use of colonic manometry. This review will concentrate on contemporary manometric methods of evaluating colonic motor function, specifically phasic and tonic colonic contractile activities. These include traditional intraluminal manometry (via perfused-tube and solid-state catheters) and use of the barostat.
- Top of page
Changes in intracolonic pressure, resulting from phasic contractions of (predominantly) circular colonic muscle can be detected using intraluminal manometry. However, only lumen-occluding contractions, or contractions generating a common-cavity phenomenon (in a segment filled with a continuous column of content, a contraction occurring anywhere along the length will pressurize the entire segment) will be recorded by this method.34
Three aspects of phasic contractile activity can be evaluated:
characteristics of single pressure waves (the phasic contractile unit);
the temporo-spatial organization of groups of single pressure waves (i.e. motility ‘patterns’);
distinguishable time periods or phases of activity (e.g. the periodicity of recurring motility patterns, the postprandial period, etc.).
Unfortunately, there is no accepted standardized method (nor equipment) for the manometric evaluation of colonic contractile activity, and recorded measurements therefore vary.
Water-perfused catheters Multilumen water-perfused catheters are made from extruded polyvinylchloride (PVC), silicone rubber, or similar (e.g. Dentsleeve Pty Ltd, Wayville, South Australia; Arndorfer Medical Specialities, Greendale, WI, USA; Lectromed, Letchworth, UK). Silicone catheters, especially, are very flexible, which is important for patient tolerability. A manometric side-hole or recording port can be dedicated to each individual lumen, thus providing the capability for multichannel studies. Such assemblies are enormously versatile, as the number, position and orientation of recording sites can be modified according to the application. For perfusion and recording purposes, each lumen is coupled via an external interposed strain gauge pressure transducer, to a minimally compliant, pneumohydraulic capillary perfusion system. Perfusion, at a constant flow rate, is achieved by use of distilled water from a reservoir maintained at a high constant pressure.47,48 Contractions of the colonic wall occlude the manometric ports, thus impeding the flow of perfusate. Resistance to flow is transmitted as pressure change to the strain gauges and the degree of resistance depends upon the amplitude and duration of the motor event.
Perfusate flow rates ranging from 0.1 to 0.6 mL min−1 have been reported.49–57 This observation is important, as in vitro studies have shown that at flow rates <0.2 mL min−1, water-perfused manometry may be unreliable in detecting contractile activity, particularly in the presence of luminal contents.58 However, as the rise rate of colonic pressure waves is slow (e.g. in relation to oesophageal contractions), others have argued that detection of pressure changes is well within the capabilities of systems employing flow rates of 0.1–0.3 mL min−1.48,59,60 Another important consideration is that water-perfused systems will instil fluid into the colon; the greater the number of manometric ports, and the higher the rate of perfusion, the greater the fluid load. Although the colon can effectively absorb large volumes of water,61 prolonged studies using water-perfused technology may present the organ with a considerable quantity of perfusate (≥3 L);53,62 whether this disturbs basal motor activity is unclear.
Aside from their versatility, water-perfused assemblies have the advantages of simplicity, relatively inexpensive components and applicability to the measurement of motor activity in several regions.48 Importantly, they are fully autoclavable, enabling simple sterilization. The major disadvantage of the system is that the catheter is linked to a (usually static) pneumohydraulic infusion pump and recording equipment, which almost invariably precludes ambulatory study. A novel, portable perfused manometry system has recently been described for the study of small intestinal motility in mobile subjects,63 and may be applicable to study of the colon.
Solid-state catheters Multiple microtransducers (strain-gauges), can be incorporated within the design of flexible, so-called ‘solid-state’ catheters (e.g. Gaeltec Ltd, Dunvegan, UK; Konigsberg Instruments, Pasadena, CA, USA; Fig. 1). Each strain-gauge may be linked to a miniature flexible pressure-sensitive diaphragm, and form one arm of a Wheatstone bridge-type electronic circuit via fine connective wiring linked to the amplification/recording system. This provides a simple means of measuring small changes in resistance with a high degree of precision. With no applied pressure (strain), the strain-gauge, which can be regarded as a variable resistor, is held in balance with other fixed resistors within the circuit. Application of pressure (colonic contraction/intraluminal pressure change) will cause deformation of the diaphragm, which in turn alters the resistance of the strain-gauge; this upsets the balance of the bridge circuit, and results in current flow. The greater the strain applied (i.e. the greater the magnitude of contraction, the larger the change in intraluminal pressure, and the bigger the deformation of the pressure-sensitive diaphragm) the greater the current flow.
Figure 1. ‘Solid-state’ manometric recording equipment. (A) Six-channel solid-state catheter attached to a portable recorder. Microtransducer pressure sensors are arrowed. (B) For colonoscopic-assisted intubation, a strong silk thread attached to the tip of the catheter can be grasped by biopsy forceps. (C) Plain radiograph of the catheter in situ in the left colon. The pressure sensors are radio-opaque and are clearly visible (arrowed).
Download figure to PowerPoint
The main advantage with solid-state catheters is that transmitted signals can be recorded by portable digital recorders (e.g. MicrodigitrapperTM; Synectics Medical, Enfield, Middlesex, UK; Flexilog 3000; Oakfield Instruments Ltd, Eynsham, Oxon, UK; Fig. 1A) with large memory capabilities, that allow measurements to be made over a prolonged period from totally ambulant subjects. Recordings may thus take place outside the confines of the laboratory. In addition, compared with water-perfused systems, solid-state technology enables a much higher frequency response.64 However, solid-state manometry is considerably more expensive, transducers are less robust, and the maximum number of recording sensors available is notably reduced.
Catheter design The location of each manometric port/solid-state sensor is the choice of the investigator, and is dependent, to some degree, on the number of recording sites available. Currently, a 4-mm-diameter water-perfused multilumen catheter has the capacity for up to 21 manometric side-holes (Dentsleeve Pty Ltd), whereas the number of solid-state sensors able to be incorporated within an equivalent-sized catheter currently appears limited to a maximum of eight.59 Regardless of catheter type, however, incorporation of ≥6 recording sites may be considered sufficient for the reliable appreciation of pancolonic motility.49–52,54–57,65
Obviously, catheter design is related to its application: if manometric information from the ‘whole’ colon is required, then recording sites may be spaced equidistantly to span the length of the large intestine; if the study segment is shorter, however, the position and spacing of recording sites can be modified accordingly. For perfused-tube catheters, each manometric port (and perhaps also the catheter length) should be marked to allow spatial localization of the recording sites by fluoroscopy following intubation (solid-state sensors are naturally radio-opaque; see Fig. 1C). This is usually achieved by filling the lumen immediately distal to the side hole with a radiodense substance (e.g. small metal ‘slugs’).
Unfortunately, the inherent flexibility of design has meant that standardization is lacking. Various catheter configurations, from studies reported over the past 2 decades, are shown in Table 2.
Table 2. Catheter configurations for colonic manometric recordings
|Authors||Year||References||Route||Bowel preparation||Recording length (short/ prolonged)||Total catheter length (cm)||‘Active’ catheter length (cm)||Number of recording sites||Spacing (cm)||Perfusion rate/ lumen (mL/ min−1)||Catheter diameter (mm)|
|(a) Studies utilizing perfused-tube catheters|
|Kerlin et al.||1983||78||p.o.+ ||No||P||425||60||3||Distal 2 ports 20 cm apart + 40 cm||?||4.5|
| || || ||p.r.|| || ||>25||5||2||5|| || |
|Spiller et al.||1986*||79||p.o.||No||S||>240||66||9||Distal 6 ports 4 cm apart + 8 + 28 + 10 cm||0.1||6.0|
|Bassotti et al.||1987–2001||49,50,93,95,110, 145,179,180||p.r.||Yes||P||?||84||8||12||0.1||4.5|
|Bazzocchi, Reddy Snape et al.||1989–1991*||27,51,52,171||p.r.||Yes||S||>140||105||8||15||0.5||?|
|Kamm et al.||1992*||28||p.r.||Yes||S||?||81||11||Distal 3 ports 0.5 cm apart + Proximal 8 ports 10 cm apart||0.5||4.0|
|Di Lorenzo et al.||1993–2000†||56,57,88,181,182||p.r.||Yes||S||?||75||6||15||0.1||?|
|Cook et al.||1994–2000*||29,62||p.r.||Yes||P||200||110||12||10||0.15||5.9|
|Lémann et al.||1995||67||p.n.||No||P||450||60||10||Distal 4 ports 10 cm apart + Proximal 6 ports 5 cm apart||0.15||6.0|
|McKee, Finlay et al.||1999–2001||54,55||p.r.||Yes||P||190||84||8||12||0.6||?|
|Bampton, Dinning et al.||1999–2001||53,60,82||p.n.||No||P||500||112.5||16||7.5||0.15||3.5|
|Leroi et al.||2000||75||p.r.||Yes||P||?||70||8||10||0.5||?|
|Authors||Year||References||Route||Preparation||Recording length (short prolonged)||Recording type (static ambulatory)||Total catheter length (cm)||‘Active’ catheter length (cm)||Number of recording sites||Spacing (cm)||Catheter diameter (mm)|
|(b) Studies utilizing solid-state catheters|
|Soffer et al.||1989||83||p.n.||No||P||A||420||90||3||45||2.5|
|Garcia et al.||1991‡||183||p.colostomy||No||P||s||?||20||3||10||6.0|
|Crowell et al.||1991||73||p.r.||Yes||P||A||100||10||3|| 5||2.7|
|Herbst et al.||1997||77||p.r.||Yes (rectal)||P||A||?||72.5||6||Distal 2 sites 10 cm apart + proximal 3 sites 17.5 cm apart||3.5|
|Rao et al.||1998–2001||65,118||p.r.||Yes (rectal)||P||A||?||53||6||Distal to proximal = 7 + 11 + 10 + 10 + 15 cm||6.0|
|Smout et al.||2002–2003||147,184,185||p.r.||Yes||P||A||230||50||6||10||3.3|
|Hagger et al.||2002||84||p.n. + p.r.||No||P||A||330||60||5||15||3.0|
| || || || || || || ||195||60||5||15||3.0|
Bowel preparation To position the catheter tip in the proximal colon by a retrograde approach requires colonoscopic manipulation, which obviously necessitates prior bowel preparation. This method undoubtedly affects ‘basal physiological conditions’,49 in that the native colon is never ‘empty’. Although some reports suggest that basal motor activity is unchanged by bowel cleansing,66,67 several studies have shown that specific individual contractile activities in the prepared bowel differ from those of the unprepared bowel [e.g. high-amplitude propagating contraction (HAPC) frequency –see below].66–68 Similarly, cleansing of the bowel has been shown to alter regional colonic transit.69,70 However, although bowel preparation may be seen as a disadvantage, it does standardize at least one methodological aspect, namely the removal of (differing) colonic contents, thus rendering inter-subject comparisons more meaningful.71,72
To cleanse the bowel, ingestion of 1–4 L of a standard polyethylene glycol colonic lavage ± tap water enemas can be used, until the faecal effluent is clear.27,29,73–76 Alternative regimens include sodium picosulphate54 or phosphate enemas.77
By contrast, for studies confined to the distal left-side of the colon, a flexible sigmoidoscope can be used to site the manometry catheter, and bowel preparation is typically avoided.25,50 Intubation by an antegrade route also has the advantage that bowel preparation need not be used; in addition, the catheter may not be expelled or significantly displaced during defaecation.
Antegrade intubation Although measurement of colonic motor activity has been achieved via both perfused-tube53,60,62,67,78–82 or solid-state catheters83,84 passed through the nose or mouth, intubation of the colon by this route is most suited to multilumen perfused-tube assemblies, because of the difficulties involved in the manufacture of solid-state catheters of sufficient length. In order to traverse the majority of the large intestine, catheters introduced by this approach must be 450–500 cm long.
The main modification to catheter design is to secure a small (maximum ∼10-mL capacity) latex balloon to the tip, which facilitates catheter progression once in the small bowel. The catheter must have a central lumen to allow inflation/deflation of the balloon. To further assist antegrade movement of the catheter by gravity, a tip weight (or articulated weights) may also be included in the catheter design.
Intubation should commence in the morning, with the subject fasted overnight. Nasal intubation is preferable, as this will allow the subject to eat without fear of biting the catheter. A local anaesthetic spray should be applied to the nose and hypopharynx to minimize discomfort. A water-soluble gel applied liberally to the tip of the catheter will facilitate intubation. After the catheter is advanced (with the subject repeatedly sipping small volumes of water) to a distance of ∼70 cm from the nose, the position of the tip should be checked on fluoroscopy to see that it lies adjacent to the pylorus. Often, the tube will curl backwards in the fundus, and will have to be manipulated, under fluoroscopic control, into the correct position. Once the orientation of the catheter is acceptable, and there is a reasonable loop of tubing in the stomach to allow distal migration, the subject should be instructed to lie on their right-hand side to aid passage of the tip through the pyloric sphincter. Usually, this should take no more than a couple of hours (and often takes considerably less time). On very rare occasions, where the tube fails to pass beyond the pylorus, it can be pushed through by using a gastroscope. An alternative method of intubating the pylorus is to endoscope the subject first, and deploy a guide wire through the biopsy channel into the duodenum. After withdrawing the endoscope, the catheter can be ‘run-down’ the guide wire into place.
Once the tip of the catheter is into the duodenum, the balloon should be inflated to ∼5–7 mL. This will stimulate small bowel peristaltic activity, which will effect distal propagation of the probe. The subject should advance the catheter at a rate of ∼40 cm h−1. The position of the catheter should be checked periodically (approximately every 2 h). It may be prudent to deflate the balloon to a diameter of 0.5 cm once the tip nears the terminal ileum to prevent impaction at the ileocolonic junction. When the balloon has passed into the caecum, it should be re-inflated to a diameter of ∼2 cm. In normal subjects, pancolonic positioning of the catheter assembly should usually take no longer than 36 h.60,80
The subject need not be present in the laboratory during the entire latter part of this intubation procedure, and may be allowed to go home once the catheter is well into the small bowel.
Retrograde intubation Two colonoscopically assisted methods are available:
1 The catheter is introduced in tandem with the colonoscope. Prior to intubation, a strong thread should be tied to the tip of the catheter; this can be grasped by biopsy forceps or a polypectomy snare passed via the biopsy channel, so that the manometry catheter lies snugly parallel to the colonoscope (Fig. 1B) during its introduction. Care should be taken to allow sufficient thread length so that the tip of the biopsy forceps can be fully withdrawn into the aperture of the biopsy channel, thus reducing any risk of perforation. Once the colonoscope–catheter assembly has been manipulated under direct vision to the desired site (hopefully the caecum, to allow pancolonic recording), the biopsy forceps can be opened, and the colonoscope carefully removed, thus leaving the catheter in situ.29,49,50,65,75 As much insufflated air as possible should be extracted. The final position of the catheter (and recording sensors) can be checked fluoroscopically.
A modification of this procedure is to grasp the catheter thread with an endoscopic haemostatic clipping device, rather than biopsy forceps. When the tip of the catheter is at the desired location, the clip can be ‘fired’ onto colonic mucosa. This process detaches the clip from its introducer within the biopsy channel, leaving it, and the catheter thread attached to the internal colonic wall. This may reduce catheter migration, thus enhancing evaluation of temporo-spatial relationships of colonic contractile patterns.23,85
2 During colonoscopy, a guide wire can be passed via the biopsy channel into (desirably) the right colon. The colonoscope can then be carefully withdrawn, leaving the guide wire in situ. The manometric catheter (with an available large internal lumen to accommodate the guide wire) can then be pushed, under fluoroscopic control, over the guide wire into the colon. Once in place, the guide wire can then be withdrawn.27,51,74,77,86–89 The advantage of this technique is that it can be combined with a diagnostic colonoscopy.
- Top of page
As with most aspects related to colonic manometric investigation, there is a lack of consensus regarding the optimum design of studies. There are no prospective data available, however, to suggest superiority of one particular study protocol over another.90
In general terms, the length of the recording period is governed by the aims of the study, and the type of catheter used. As colonic motor events are relatively infrequent, and may be subject to circadian variations,72,91‘physiological’ studies should be carried out, wherever possible, over a prolonged period to better evaluate changes in motility patterns with time. A recording period of 24 h should be seen as the minimum duration of such studies. Conversely, short-duration (<8 h) recordings are usually adequate to register the effect of a given stimulus (e.g. drug administration) on basal motor activity.
For studies involving perfused-tube technology, the subject needs to be in close proximity to the pneumohydraulic infusion system and recording apparatus, and is thus confined to the laboratory. Because they are required to remain relatively immobile for the duration of the recording period (to limit movement-related artefact), subjects should be made as comfortable as possible, either in a reclined chair, or a head-up supine position. Although numerous prolonged (24 h) studies have been reported using this technique,49,50,53,60,82 yielding important data, comparative immobilization over an extended period may be regarded as non-physiological,65 and thus static recordings are usually limited to a maximum of 6–8 h. As described above, solid-state technology enables subject mobility, and thus studies utilizing such equipment have generally been performed over 24 h.
Medications which may affect intestinal motility should be discontinued, if possible, for 96 h before starting the study (although 48 h is probably adequate).59 If sedation/anaesthesia has been used during the intubation procedure, recovery from these effects needs to be complete before recording starts. Length of recovery varies widely depending upon the drugs used: for short-acting benzodiazepines, a 1-h recovery period is sufficient;92 opiates should be avoided if feasible, but if used, recovery period needs to be extended to 16 h.86
Prolonged (≥24 h) recordings (‘physiological studies’) Once the recording catheter is in the required position, an equilibration period of 2–5 h65 may be allowed to minimize the effects of intubation, enema preparation and administration of sedative medications (see above). Recordings should therefore start at approximately mid-day, and allow for a sufficient period of both diurnal and nocturnal activity. The study should ideally include ingestion of two or three meals and a period of sleep. To reduce inter-subject variability, subjects may be instructed to go to bed and rise at the same time, and be given identical meals at equivalent times (mid-afternoon, evening and morning), although it may be ‘more physiological’ to allow the subjects to eat and sleep ad libitum.
Meal composition should be derived from the subject's usual diet, be culturally appropriate, and in excess of 400 kcal.59,90 The standard ‘major’ meal size commonly reported is 1000 kcal,16,49,50,53,60,62,65,84,93 although a smaller caloric load (e.g. 600 kcal) may invoke an equivalent response.94 Breakfast is typically smaller, being 400–450 kcal.65,84 Individual meal components will influence colonic motility, and may also be standardized; inclusion of >40% of fat has been advocated,53,60,62,80,92,95 as this ingredient appears to provide the major stimulus.96,97 Ingestion of amino acid and protein-rich meals has been shown to inhibit colonic motor activity.97 Free access to water should be allowed (maximum 1.5 L 24 h−1),65 but ingestion of alcohol and smoking are prohibited due to their effects on colonic motor function.98–101
Subjects who are ambulant may be allowed to go home, where they should be encouraged to engage in their normal daily activities. For all prolonged studies, subjects should be instructed to depress the event marker, contained on most recorders, to identify events such as eating, walking and sleeping, or to indicate the occurrence of symptoms such as abdominal discomfort, urge to defaecate and passage of flatus. They should also be provided with a diary sheet to manually record the nature and time of occurrence of these events.
Short-duration (<8 h) recordings (‘provocation’ studies) Short-term, usually static recordings are traditionally used to assess the effect of a given stimulus (e.g. pharmacological studies) on motor activity. Following a fasted baseline period (typically 30–60 min), the stimulus (e.g. drug, meal, etc.) should be administered, and the recording continued until basal activity returns, although this is not always feasible.
The applied stimulus may be used in the context of a provocation test, to achieve a ‘known’ colonic response. Such provocation stimuli may also be used to induce a patient's symptoms (e.g. abdominal pain) that may rarely arise spontaneously in the context of a static study.
Stimuli that have been used to elicit a quantifiable colonic response include:
1Response to food. This is the most powerful normal ‘physiological’ stimulus, and may be particularly useful in clinical studies, where meal-related symptoms are commonly reported. The importance of meal composition and caloric load are discussed above. Following ingestion, a sustained increase in colonic motor activity (phasic and tonic) will normally be effected within 10 min;16,17,102,103 this is the colonic response to eating, or gastro-colonic response (not reflex), and comprises early and late phases. The early response is most intense in the distal colon compared with proximal regions of the viscus,27,104 and phasic contractile activity and tone are significantly increased in comparison with preprandial values for 20–60 min.16,27,49,78,86,92,96 A second peak in colonic postprandial activity (the late postprandial response) occurs after ∼50–110 min,17,27 and lasts for up to 3 h.78,105 In provocation studies involving the response to food, the experimental protocol should therefore take into account the duration of the postprandial response and its constituent phases. Increased motor activity following a meal may be regarded as in index of the integrity of the neurohumoral control of colonic motility.2
2Response to bisacodyl. Intraluminal instillation of bisacodyl (5–10 mg)18,28,106–109 typically provokes a stimulatory effect on colonic motility, notably the induction of high-amplitude propagated contractions.18,28,75,108 This response is absent after application of lidocaine to the mucosal surface of the colon,18 suggesting that bisacodyl has a direct action on the nerve plexuses, and can thus be used to assess the integrity of intrinsic neural function. In clinical studies, an absent response has been shown to be associated with myenteric plexus damage.18,28
3Cholinergic stimulation. Cholinergic agents such as edrophonium chloride (10 mg i.v.)109,110 or neostigmine (0.5 mg i.m.; 0.01 mg kg−1 s.c.)55,111 have typically been used to provoke an increase in colonic motor activity; an impaired response may be suggestive of abnormal cholinergic innervation,110 which is known to play a pivotal role in the regulation of colonic motor function.
4Balloon distension. This is often used to assess visceral sensation in functional bowel disorders112–114 (see Barostat section below), but has also been used to elicit colonic peristaltic activity,115,116 given the assumption that intraluminal distension provides a physiological stimulus. The response is not consistent, however,116 and further studies are required to optimize methodology (e.g. required balloon size, distension volume, site of distension, etc.).
5Response to stress. This may be especially relevant in functional bowel disorders, where symptoms are exacerbated under condition of ‘stress’. Various stressors have been used to induce an increase in colonic contractile activity, either psychological (e.g. stressful interview,117 dichotomous listening test,118 fright119 or anger,117 etc.), or physical (e.g. cold-water immersion test).118,120
- Top of page
Interpretation of a manometric recording involves interpretation of the patterns of pressure change (contractions) and interpretations of their quantity, both in terms of magnitude and numbers.1 In 1962, Ritchie, Ardran and Truelove13 described two-channel recordings of colonic motor activity as containing ‘almost every imaginable form of pressure change’. They also stated that ‘although deflections on the two channels were occasionally similar..., the tracings were usually different. Sometimes they were even different in rhythm’ (e.g. Fig. 2). As a consequence of this inherent complexity, which is further confused by the influence of differing methodologies on the recorded pressure signal,1 there is still a lack of consensus regarding the standardization of criteria used to define individual contraction characteristics, or delineate separate colonic contractile activities. Such criteria provide the foundation for the interpretation of colonic motility data.
Figure 2. Colonic phasic contractile activity. This 20-min tracing from the left colon of a healthy volunteer clearly illustrates the complexity of segmental phasic contractile activity.
Download figure to PowerPoint
Recorded activity should ideally be analysed both manually (i.e. visual observation)27,29,49,50,53,55,83 and with the aid of a dedicated computer software analysis program.63,121–124 Automated analysis considerably increases the clinical potential by removing subjective bias and providing detailed, objective and repeatable quantification of motility patterns of interest. A recent study has shown that 97% of all pressure peaks detected visually were recognized by the computer, and 92% of automatically detected peaks were also observed by the human investigators.124 However, computer-derived analyses are only as good as the algorithms used for contraction recognition. These embodied algorithms are, of course, developed by different human ‘experts’, and until they are universally standardized and validated, the results of such automated analysis will only be truly comparable between centres utilizing the same software. It is also important to remember that qualitative (mostly visual) and quantitative (mostly computer-derived) analyses are not exclusive but complementary.
The inclusion of pressure waves for subsequent analysis basically comprises three main components:
In comparison with manometric recordings from the upper GI tract, colonic motility tracings contain more shifts in baseline and more complex waveforms;124 therefore, contraction-recognition parameters, based on criteria adopted for the analysis of oesophageal contractions, as has been proposed previously,116 may be inadequate.124 For each recording channel, pressure events are identified relative to the channel-specific baseline, and thus baseline changes must be compensated for; this can be achieved either at appropriate times (visual analysis)53 or by applying a protocol that enables a ‘running’ baseline to be determined (automated analysis).124
For all traces, it is vital to exclude pressure changes unrelated to contractile activity (i.e. artefact). Computerized artefact filtering, or smoothing, will, for example, remove all peaks <5 s in duration occurring simultaneously (to approximately the nearest ±0.3 s) in the majority of channels. Such ‘noise’ is resultant from instrumentation-related artefact, gross body movement and changes in intra-abdominal pressure (straining, respiration). Care must be taken not to exclude true ‘synchronous’ contractions, however.
The phasic contraction is the basic unit of contractile activity throughout the GI tract. In the human colon, a phasic pressure wave (contraction) may generally be defined as a predominantly monophasic pressure change with a discernible onset, peak and offset, which does not have the features of a pressure increase associated with artefact.53 However, colonic contractions may also be multiphasic,126 or composed of phasic on tonic changes,29,62 which must be taken into account. Some workers have indeed classed pressure events <30 s in duration as phasic contractions, and those >30 s in duration as tonic contractions.88
The threshold for inclusion as a ‘true’ contraction very much defines the quantity of contractile information available for analysis, and provides one major source of discrepancy between reported studies. Typically, pressure waves are only included if they are >5 mmHg in amplitude (relevant to the baseline) and >4 s in duration,29,54,60,65,102 although greater amplitude thresholds have been employed (e.g. >10 mmHg,67 or >20 mmHg49).
Qualitative analysis is used for contraction recognition and recognition of the various contractile activities.27,29,49,50,53,65,83 Automated analysis allows the quantification of contraction characteristics within each contractile activity or ‘pattern’ (e.g. duration, amplitude and frequency). Analysis is usually performed for each channel of recording (thus allowing investigation of regional variation) over the entire study period. For comparative purposes, the recording is often divided into equal time epochs (typically 10 min or 1 h).
Common analysis parameters include:
Type of contractile activities. Given the complexity of the recorded signal, certain ‘easily recognizable’ waveforms (e.g. high-amplitude propagated contractions, rectal motor complexes) have, understandably, received the greatest attention to date. However, other, relatively neglected, contractile patterns (e.g. low-amplitude propagated contractions) may be of equal functional importance, and thus the whole spectrum of contractile activities should be considered. Further criteria for the identification of various contractile patterns are described in their relevant sections below.
Number of contractions;
Contraction duration (the time between the onset of the major upstroke of the wave and return to baseline);
Contraction amplitude (by subtracting the baseline pressure from the peak wave pressure);
Contraction origin (the site at which a propagated contraction is first recognized);
Contraction frequency (within a ‘pattern’);
Velocity of contraction propagation
(considered from the time between wave peaks or wave onsets in adjacent recording channels; the latter has been reported as an inferior method, given the slow and irregular nature of some colonic wave upstrokes;124
Direction of propagation (orad or aborad);
Length of propagation (determined by number of sites over which the wave propagated).
Quantitative (‘global’ appreciation of contractile activity)
Because readily recognizable ‘patterns’ of colonic phasic contractile activity are difficult to distinguish, and are subject to inter- and intra-subject variability, most authors have attempted to globally quantify phasic activity by summing all pressure events over a given time period. This can be performed for individual recording channels, or by grouping adjacent channels, which provides an appreciation of motor activity in whole colonic segments.4 Calculated parameters include:
number of peaks per unit time;
percentage or duration of time occupied by contractile activity;
area under the pressure curve (AUC or ‘activity index’);
motility index (MI), which integrates the duration and amplitude of pressure events with time.
To study variations in motor activity (secondary to physiological, pharmacological or psychological stimulation), a basal MI can be taken as the mean MI during a given period of fasting (typically 30–60 min) and expressed at 100%. Changes in MIs following or during an event can then be expressed as a percentage of the basal MI.
Events registered by the subject (by depressing the event marker) will be displayed on the recorded trace. The correlation between a given symptom and specific pressure waveforms (i.e. contraction types) can be visually assessed. This method of analysis is especially pertinent in clinical studies.125
Colonic contractile activities (intraluminal catheter recordings)
- Top of page
No classification system regarding the morphology of contractile components of human colonic motility has yet found universal agreement. In rudimentary terms, human colonic motility involves segmenting (mixing) and propulsive contractile activity; some contractions may be organized into recognizable ‘patterns’. On the basis of this simple classification, identifiable contractile activities have been tabulated (Table 3), and are described in detail below. The classification system is based on seminal studies by Adler et al.,126 and Spriggs et al.,11 published more than half a century ago, and has been modified in accordance with combined manometric/radiological studies,13,127,128 contemporary studies of pancolonic motility,29,53,65 and a recent categorization proposed by Bassotti et al.91
Table 3. Colonic contractile activities
|Type of activity||Propagation||Occurrence||Contraction type||Contraction characteristics (Normal values)|
|Amplitude (mmHg)||Duration (s)||Velocity (cm sec−1)||Frequency (contn min−1)|| Incidence (n h−1)|
|Phasic contractions (Traditional manometric studies)|
| ||Non-propagated or propagated over short distances||Single||Short-duration (Type I)||5–60||15–60||–||–||20–50|
|Long-duration (Type II)||5–60||<15||–||–|| |
|Bursts (Type III)|| || || || || || |
|Rhythmic*||–||1–30 min||–||3 or 6–8||<3|
| ||Propagated over long distances||Single or within propagating sequences||LAPCs / PS||20–80||∼10||1–2||–||1.9–5|
|HAPCs / HAPS (Type IV)||100–180||10–20||0.7–1.5||–||0.2–0.6|
| Organized i.e. ‘patterns’|
| ||Non-propagated or propagated over short distances||Periodic bursts||CMCs / PCMA*||–||6–8 min||–||3–6||0.6–2|
|RMCs / PRMA*||15–60||3–30 min||–||2–3 or 6–8||0.2–0.7d|
| || || || || ||0.8–1.4n|
|Tonic contractions (Barostat studies)|
| Sustained|| || ||Baseline volume change†||% change from basal (100–200 mL)||Minutes to hours||–||–||–|
| Non-sustained||PVEs‡||% change from basal (100–200 mL)||Seconds||–||0.6–4||35–100|
Colonic motility: circadian variation
- Top of page
One of the most important aspects of human colonic motility is that under ‘normal’ conditions, the incidence of each type of contractile activity is subject to diurnal variation (Figs 3 and 4). Prolonged manometric studies49,65,78 have consistently shown that colonic motor activity is influenced by:
Figure 3. Prolonged recording of colonic motility. This compressed 19.5 h tracing [recorded via two solid-state catheters, one introduced via an antegrade (per oral) approach and the other by a retrograde (per rectal) route] clearly shows that colonic motor activity is subject to circadian variation. There is a marked reduction in contractile activity at night (Adapted from Ref. 84, with permission).
Download figure to PowerPoint
Figure 4. Circadian variation in colonic motility. The 24 h profile of the incidence of intraluminal pressure waves (phasic contractions) and the area under the curve (AUC; motility index) of pressure waves in 22 healthy volunteers (Reproduced from Ref. 65, with permission).
Download figure to PowerPoint
ingestion of a meal.
During sleep (either nocturnal, or during daytime naps), contractile activity is markedly decreased or abolished for long periods;49,50,62,65,83,105 however, such motor quiescence may be interrupted by transient arousal or rapid eye movement (REM) sleep.62 Morning awakening, or indeed forced arousal from sleep, enhances the frequency of colonic contractions;49,62,65,129 it has been suggested that this mechanism ‘primes’ the colon to empty early in the morning.50 The increase in colonic contractile activity following a meal is one of the most consistent findings in colonic motor physiology (see above).16,17,27,49,65,78,83 A similar circadian trend exists with regard to human colonic muscle tone, being maximal after feeding and least during nocturnal sleep.86,92 The circadian variation exhibited by individual contraction types or patterns is documented below.
The incidence of colonic contractile activities is also influenced by the level at which they are recorded, being more pronounced in the sigmoid compared with the proximal, transverse or descending colon.67 Comprehensive reviews on the effects of various stimuli on the frequency of recorded contractile patterns at differing levels of the large intestine can be found elsewhere.1–3,91,102,130,131
‘Normal values’ described below (and in Table 3) generally relate to the mid-colon under basal (fasted) conditions, unless otherwise stated.
Non-propagated or propagated phasic contractions Prolonged intraluminal pressure recordings have revealed that human colonic motility is primarily characterized by periods of phasic contractile activity, often alternating with phases of motor quiescence. Phasic contractions predominantly occur in irregular ‘bursts’, some of which are ‘rhythmic’ (i.e. contractions occur at a regular frequency), and some of which are ‘arrhythmic’. They also occur as sporadic, isolated events.49,67,78,79,83 For the most part, segmenting phasic pressure waves, whether isolated or in bursts, are localized and non-propagating and are associated with the gradual movement of intraluminal contents. The direction of flow is relative to the ‘gradient’ of motor activity, with movement from areas of high activity to areas of low activity.27 Less frequently, segmenting phasic contractions propagate over short distances, which may effect bolus movement over the length of bowel involved.27 The function of such phasic activity is to continually turn over and mix intraluminal contents for the optimal absorbtion of water and electrolytes. Contractile activity gradients may also provide a functional obstacle slowing up colonic transit, for example, at the rectosigmoid junction.108,133
Phasic contractions may be sub-classified on the basis of duration, and are considered to be of either long duration or short duration. This categorization is consistent with Type I and II waves of early classification systems.11,126,134
•Identification and definition. Long-duration contractions may be defined as pressure waves 15–60 s in duration, whereas short-duration contractions are pressure events 5 (threshold) –15 s in duration.2,11 Short-duration contractions may often be superimposed on long-duration contractions.2,11 The presence of two types of phasic contraction in the colon is unique, compared with the remainder of the GI tract.2 The significance of this phenomenon is not known. Isolated, non-propagating pressure waves are defined as those events occurring (seemingly) randomly without any associated pressure activity within the same channel or adjacent recording channels for at least 30 s. Contractile bursts are defined as discrete, random phases of continuous phasic pressure activity lasting ≥3 min; those bursts containing pressure waves occurring at a regular frequency (≥2 waves min−1) are defined as rhythmic, whereas those comprised of contractions occurring at irregular frequencies are defined as arrhythmic.
Propagation of segmenting contractions is defined as phasic pressure waves that migrate either aborally (antegrade propagation) or orally (retrograde propagation) across short distances only (≤3 consecutive channels at a velocity of 0.2–12 cm s−1), compared with propulsive contractile events (see below). Recent studies have suggested that pressure waves occurring synchronously in at least three, but not all channels (where they would be classified as artefact), should be included in analysis and defined as simultaneous pressure waves.29,65
•Normal values. There is considerable inter- and intra-subject variability in relation to recorded phasic contractile activity that is probably dependent not only on obvious factors such as varying methodologies, but also the nature of the intraluminal contents, and the emotional and metabolic state of the subject under study.4 Typically, however, colonic phasic contractions occur 20–50 times per hour,13,54,65,118 and are 5–60 mmHg in amplitude (mean ∼20 mmHg), although occasional, sporadic contractions of greater magnitude occur. Bursts of phasic activity, of frequencies varying from two to eight cycles per min, occur <3 times per hour, and account for ∼6% of all colonic contractions.135 Such bursts are generally arrhythmic in nature (frequency 2–4 contractions per min), although rhythmic frequencies (predominantly 2–3 contractions per min, or 6–8 contractions per min) may also be recorded.49 A gradient exists from the proximal to distal colon with regard to the presence of rhythmic contractile activities; only 1% of phasic contractions in the ascending colon occur at a regular frequency, increasing to 12% in the transverse colon, 38% in the descending colon and 49% in the sigmoid.135 However, there may be no identifiable temporal organization (i.e. periodicity) to these contractile bursts.135
Long-duration contractions account for >90% of recorded colonic activity,11 and the majority of phasic pressure events (either isolated or in bursts) are non-propagated. A recent study has suggested that synchronous phasic pressure waves may comprise the most common pattern of phasic activity.65
In terms of ‘globally’ quantifying phasic contractile activity, an AUC or MI has usually been reported; this is particularly applicable to non-propagated activity.29,62 Unfortunately, different methods exist for the calculation of AUC or MI, and as a consequence, reported values are hugely disparate (range: 30–5000 mmHg min−1),27,29,51,54,80,89,92,118,136 making comparisons between studies extremely difficult. Studies presenting variation from the basal motility index are more comparable, and have disclosed that colonic phasic contractile activity follows a circadian trend (Fig. 4).65
High-amplitude propagated contractions The rapid propulsion, or ‘mass movement’ of intraluminal colonic contents was first described in the feline colon by Cannon, approximately 1 century ago.137 Subsequent radiological investigations in man showed that colonic mass movements similarly occurred both in health and disease;138–140 Barclay eloquently described that propulsion of contrast occurred ‘so quickly that the eye could hardly follow the movement’.140 However, reports were rare due to the serendipitous nature of recording these extremely infrequent and short-lasting events; Holzknecht proposed that only three or four such movements would occur during the course of a day.139
With the advent of manometric recording techniques, used in combination with cinefluoroscopy or scintigraphy, it is generally accepted that the major motor correlates of colonic mass movements are rapidly migrating, high-amplitude, long-duration contractions,13,52,127,128 generally termed HAPCs.
Functionally, it is now well established that the primary role of HAPCs is to propel intraluminal contents rapidly over large distances52,127,128 (Fig. 5), and there is a clear association with faecal expulsion28,60,141–144 (Fig. 6).
Figure 5. Colonic ‘mass movement’: simultaneous assessment of intraluminal pressure change and transit. The lower five tracings represent a 3-min period of intracolonic pressure activity (as recorded by perfused tube manometry), over a 60-cm study segment proximal to the splenic flexure. A high-amplitude (>100 mmHg) propagating contraction (HAPC), originating in the ascending colon, migrates rapidly (∼1 cm s−1) to the splenic flexure; this is concomitant with a marked shift in intraluminal contents, as seen by movement of the radiolabelled marker (99m Tc, previously instilled in the caecum) from the transverse to the descending colon [three serial (1 min) scintiscans, shown in the upper part of the figure].
Download figure to PowerPoint
Figure 6. High-amplitude, propagated contractile (HAPC) activity in association with defaecation. In this healthy volunteer, propagating sequences involving four HAPCs, occurring in the left colon over a 10-min period, lead to defaecation; the second and third are associated with the call to stool (urge), and the fourth immediately precedes stool expulsion. Normal phasic contractile activity is almost indistinguishable, as the amplitude scale has been adjusted to illustrate the full waveform and magnitude (max. > 250 mmHg) of these events. A ‘straining’ pattern can be noted around the time of stool expulsion, where pressure change occurs simultaneously at all 16 perfused-tube manometric recording sites (Generously supplied by Dr ML Kennedy; see Ref. 60).
Download figure to PowerPoint
•Identification and definition. As mass movement waves are the most dynamic colonic motor pattern, they have received the greatest attention in recent years. Given their amplitude, HAPCs are easily differentiated from single propagating or non-propagating phasic contractions, or organized groups of contractions, and their visual identification is generally straightforward.49,50,67,72 However, colonic propulsive activity is now known to be effected by a spectrum of propagated (and perhaps also non-propagating) wave sequences29,53,145,146 and the criteria used to define HAPCs (notably in terms of threshold amplitude) are widely disparate (Table 4); this may account for the somewhat contradictory nature of published studies.
Table 4. Definitions used for identification of HAPCs
|Authors||Year||References||Amplitude (mmHg)||Duration (s)||Minimum propagation distance (cm)|
|Bassotti et al.||1992–1993||72,93||>50||–||10|
|Rao et al.||1998||118||>50||–||–|
|Leroi et al.||2000||75||>50||–||>10|
|Spiller et al.||1986||79||>60||>10||>4|
|Vassallo et al.||1992||89||>60||10–30||–|
|Lémann et al.||1995||67||>60||–||–|
|Jouet et al.||1998||80||>60||–||10|
|Crowell et al.||1991||73||>75||–||–|
|Malcolm & Camilleri||2000||164||>75||>20||–|
|Bazzocchi et al.||1990||51||>80||–||–|
|Furukawa et al.||1994||62||>90||–||20|
|Hagger et al.||2002||84||>90||–||15|
|Di Lorenzo et al.||1993||181||>100||–||>30|
|von der Ohe et al.||1994||156||>100||10–30||5|
|Brown et al.||1999||54||>100||–||24|
|De Schryver et al.||2002||146||>100*||–||20|
|Rao et al.||2001||65||≥105||≥14||–|
|Bampton et al.||2001||53||≥116||–||15|
|Herbst et al.||1997||77||>136||–||35|
•Normal values. The recorded incidence of HAPCs (and all other contractile activities) is related to the criteria used to define them (Table 4). Studies using a high (>100 mmHg) threshold value generally show a lower incidence than those with a lower cut-off. Consequently, the range of published frequencies is large (4.4–23/24 h).29,49,50,53,67,73,75,77,79,146 Rao et al.,65 who set the thresholds for classification as an HAPC at ≥105 mmHg in amplitude and ≥14 s in duration (these values represent >95% confidence limits for ‘normal’ propagating pressure waves), reported a frequency of 10.3 HAPCs per 24 h. More recently, De Schryver et al.,146 using visually detected HAPCs as reference, set an amplitude threshold of 100 mmHg in two channels, and 80 mmHg in one channel for the automated detection of HAPCs, which yielded a sensitivity of 92%, and a specificity of 99%. With such detection criteria, the mean number of HAPCs recorded per subject was 7.5 per 24 h . Notably, this frequency is equivalent to that reported from studies where simple visual identification alone was employed (∼6 out of 24 h),50,73,93 suggesting that for this type of contractile activity at least, complex analysis is not always a necessity.
Reported mean HAPC amplitudes range from ∼105 to 180 mmHg,49,50,65,67,73,75,146 although contractions may exceed 350 mmHg in magnitude.77 Sensitization by cathartics may increase the number of HAPCs recorded: Lémann et al.67 showed that the frequency of HAPCs was increased from 14.4 to 24 h in the unprepared colon to 23–24 h in the prepared colon. In addition, HAPC incidence shows diurnal variation: they occur most frequently following morning awakening and in the late postprandial period, and are usually absent at night.91 High-amplitude propagated contractions often occur in ‘groups’ or bursts (i.e. PS)29,50,53,60,67,93 (Fig. 6).
Propagation of HAPCs is dependent on enteric neural and smooth muscle continuity,147,148 and occurs predominantly in an antegrade fashion (>95%), although retrograde movement has also been reported.73 Contraction characteristics do not differ between HAPCs migrating either aborally or orally.73
High-amplitude propagated contractions are usually associated with some subjective feeling, typically abdominal discomfort, boborygmi or urge to defaecate49,50,67 (Fig. 6). In terms of defaecation itself, the traditional concept that HAPCs are temporally associated with spontaneous or stimulated faecal expulsion, as has been shown in man28,142 and other mammalian species,141,142,144 has been expanded upon recently by Bampton et al.,60 using multilumen perfused tubes passed per nasally. They have proposed that PS (see below) are the motor correlate of defaecation, some of which are initiated up to 1 h prior to stool evacuation. However, in the 15 min immediately preceding defaecation, ∼65% of PS include an HAPC.60
Low-amplitude propagated phasic contractions The function of LAPCs is poorly understood. Relative to their magnitude, they are likely to effect lesser propulsive movement of intraluminal contents than HAPCs. It has also been reported that LAPCs, in comparison with HAPCs, are involved in the transport of less viscus colonic contents, for example, fluid or gas (flatus).29,62,91,145,149
If the criteria for propagation (see above) are met, LAPCs are defined as waves of amplitude <50 mmHg.145 They occur 45–120 times per 24 h,29,65,145 and are typically 5–40 mmHg in amplitude102 (mean: ∼20 mmHg).145 Incidence is significantly greater during the day than at night,29,62,145 and increases following meals and after waking.29,62,65,145 LAPCs may be invoked by colonic distension.116
Propagating sequences These may be classed as either high amplitude (when components of the array fulfil the criteria for an HAPC) or non-high amplitude. Bampton et al.53 have shown that high amplitude PSs propagate further than non-high amplitude PSs, but the propagation velocity of high-amplitude PSs is significantly slower. The mean amplitude of sequences propagating over long distances (>22.5 cm) is significantly greater than those propagating over shorter (<22.5 cm) distances.53
Defaecation is accompanied by PSs. The frequency of antegrade PSs has been shown to increase approximately threefold in the hour prior to faecal expulsion (defined as the ‘pre-expulsive phase’), compared to the preceding hour.60 In the early component (15–60 min prior to defaecation) of the pre-expulsive phase, the site-of-origin of PSs is initially proximal to the splenic flexure, with subsequent PSs arising progressively more distally; this situation is reversed in the late component (<15 min prior to defaecation), when the site-of-origin of PSs is initially distal to the transverse colon, with subsequent PSs arising progressively more proximal. The final PSs occurring immediately prior to stool expulsion often arise in the caecum.60
Colonic motor complexes As most organs of the upper GI tract exhibit periodic groups of phasic contractions, typified by the migrating motor complex (MMC),150 it is not unreasonable to assume that similar cyclical motility patterns are present in the colon. This is indeed true in the majority of mammalian species, most notably the dog, where colonic motility is primarily characterized by regular contractile ‘bursts’, between which are periods of motor quiescence. These ‘colonic motor complexes’ (CMCs)144 are both migratory and non-migratory. However, single long-duration phasic contractions within a complex do not propagate, or only propagate over a short distance, even if the CMC migrates over a considerable length of bowel.2 The phasic contractions therefore fulfil their mixing function, with a net slow distal propulsion and the recurring complexes do not periodically strip the colon clean of its contents in a short space of time (such as the small bowel MMC).
•Identification and definition. Colonic motor complexes are defined as bursts of phasic pressure activity, of amplitude >8 mmHg, and duration ≥3 min that recur at periodic intervals.53 Frequency of contractions within each CMC is typically regular (Fig. 7).
Figure 7. Periodic motor activity in the human large intestine. This figure illustrates the occurrence of cyclical bursts of phasic contractile activity, recurring approximately every 30 min, alternating with periods of motor quiescence, in: (A) the right colon of the author (‘colonic motor complexes’; as recorded by perfused-tube manometry), and (B) the rectum of a healthy volunteer (‘rectal motor complexes’; as recorded via a solid-state manometric catheter). Neither subject underwent prior bowel preparation.
Download figure to PowerPoint
•Normal values. Previously, prolonged studies of human large intestinal motility from sites proximal to the sigmoid colon generally failed to disclose any regular periodicity to ‘bursts’ of rhythmic phasic activity.49,65,72 It is possible, however, that recording conditions may have influenced the occurrence of CMCs. In dogs, Sarna tested the hypothesis that cleansing the colon disturbed normal motor activity; he showed that CMC activity was disrupted for 3 h following colonic lavage, and was replaced with continuous but irregular contractions.68 More recently, both Bampton et al.53 and Hagger et al.,84 in studies involving antegrade intubation of the unprepared human colon, have reported the presence of colonic motor complex activity, with CMCs lasting ∼6 min, and recurring approximately once or twice per hour.53,84
Rectal motor complexes In humans, periodic contractile activity does predominate in the rectum (Fig. 7), especially during the nocturnal period, when movement-related artefact and central nervous system influences are minimized. The ‘rectal motor complex’ (RMC; or ‘periodic rectal motor activity’, PRMA65) was formally described in 198919 (although an RMC was clearly depicted in a publication by Spriggs et al.,11 almost 40 years earlier), and has been characterized to a greater extent over the last decade.20,21,24,26,65,151 Information regarding periodic colorectal motor activity (i.e. RMCs and CMCs) may, like the small bowel MMC, reflect the integrity of enteric neuromotor function.152
•Identification and definition. It is generally accepted that RMCs should be defined as regular bursts of phasic pressure waves lasting ≥3 min, with a contraction frequency of ≥2 min−1.20,21,26,65,77,151
•Normal values. The circadian trend for RMCs is reversed in comparison with all other contractile activities, which likely reflects underlying control mechanisms. Rectal motor complexes occur more frequently at night (0.8–1.4 h−1) than during the day (0.2–0.7 h−1);19,20,65,151 overall mean nocturnal cycling time is ∼40 min,19–21,151 whereas daytime RMCs recur approximately once every hour.19,118,152 Mean RMC amplitude has been reported to range from 15 to 60 mmHg.19–22,118,151 The duration of each RMC ranges from 3 to 30 min.20–22,118,151 Rao et al.151 showed that ∼80% of RMCs occur within 5 min of more proximal colonic activity. Rectal motor complexes are temporally distinct from the small bowel MMC.20,21
- Top of page
It has been argued that a critical degree of luminal occlusion, which is the fundamental requirement for effecting intraluminal pressure change, is a rare or infrequent event in a relatively large-diameter viscus such as the colon, and thus the use of thin intraluminal catheter assemblies (water-perfused or solid-state) for monitoring phasic contractile activity is questionable.48 An alternative recording technique is to employ an air-filled catheter-mounted balloon or bag which conforms to the internal outline of the segment of gut in which it is situated. This allows movement of the gut wall to be followed as closely as possible, and any change in the size of the lumen within the study segment is reflected as a change in intrabag volume. To record volume changes in this way, a constant intrabag pressure needs to be maintained, and this is provided by a mechanical barostat or distension device.153 In simple terms, the barostat rapidly aspirates air from the bag when the study segment contracts, and injects air when the segment relaxes. The volume of air aspirated/injected is proportional to the magnitude of contraction/relaxation.
In such a pressure-clamped system, aside from the measurement of phasic activity, more sustained changes in intrabag volume reflect changes in intraluminal pressure attributable to alterations in wall tone. Passage of chyme along the gut requires that distal segments be programmed to accommodate contents arriving from more proximal loci; this mechanism of ‘receptive relaxation’ is possible by relaxation of the gut wall, or decrease in tone. Conversely, a sustained increase in tone will reduce the capacity for storage and may be beneficial to the propulsion of intraluminal contents. Measurement of tonic activity cannot be evaluated by conventional manometry or electromyography. Other modalities of motor function that can be investigated using the barostat relate to mechanical properties of the bowel wall, such as compliance and tension.
- Top of page
The computer-driven mechanical barostat,154,155 maintains a constant pressure within the air-filled bag by means of a feedback mechanism, consisting of a strain gauge linked by an electronic relay to an air-injection/aspiration system. The volume of air within the bag is computed from the known excursion of the bellows within the reservoir system.
To ensure optimum fidelity of the system in adapting to changes in gut wall tone, the ‘pressure window’ (the level at which the barostat responds to a change in bag pressure), should be set between 0.25 and 0.5 mmHg above and below the preset (operating) pressure (see below). The barostat is able to correct for changes in bag pressure almost instantaneously: pump activation is typically <5 ms, and flow rate is ∼40 mL s−1.
Barostat bags should be manufactured from ultra-thin (∼40 μm) polyethylene (supermarket freezer/sandwich bags provide a suitable, cheap source), rather than latex. Provided that the range of volumes used for the study remains below 90% of the maximum volume of the bag, polyethylene can be regarded as infinitely compliant, in that its own properties have no influence on the internal pressure (i.e. large volumes can be accommodated without an increase in intrabag pressure, until the volume injected is >90% of the maximum bag volume).153 The bag is thus required to be oversized, that is, of significantly greater diameter than the colonic study segment. The shape of the bag should be related to the organ studied; for the colon, a cylindrical bag of fixed length should be used. Although dimensions vary between reported studies (Table 5), consensus agreement is that the bag should be ∼10 cm in length, and >500 mL in capacity.
Table 5. Catheter configurations for colonic manometry-barostat assemblies
|Authors||Year||References||Site of recording (barostat bag[s])||‘Active’ catheter length (cm)||Number of manometric recording sites||Spacing (cm)||Balloon dimensions||Balloon operating pressure (mmHg)|
|Single barostat balloon|
|Steadman et al. Vassallo et al.||1991 1992||8689||Ascending, transverse or descending||18||3||2 for port orad to balloon, 2 + 5 for ports aborad to balloon||10 cm (L)||8–12|
|Steadman et al.||1992||87||Proximal descending||44||6||2 for single port orad to balloon, 2 + 5 + 5 + 10 + 10 for ports aborad to balloon||10 cm (L)||8–13|
|Bradette et al.||1994||112,186||Descending colon||N/A||0||N/A||10 cm (L); 8 cm (D); 700 mL||10–14|
|von der Ohe et al.||1994||156||Transverse and proximal descending||26||4||2 for port orad to balloon, 2 + 5 + 5 for ports aborad to balloon||10 cm (L); 500 mL||13–17|
|O'Brien et al.||1997*||136||Proximal transverse||∼80||8||10 (2 orad to balloon, 6 aborad)||-||12–14|
|Bharucha et al. Law et al.||1997 2002||157159||Proximal descending||∼40||6||5 (1 orad, 5 aborad to proximal balloon)||10 cm (L); 600 mL||6–14|
|Jouet et al. Coffin et al.||1998 1999†‡||8081||Ascending / hepatic flexure and descending||∼34||6||2 + 7 + 12 orad and aborad to balloon||10 cm (L); 9.4 cm (D); 450 mL||10–13|
|Mollen et al.||1999||74||Lower descending||∼40||6||5 (1 orad to balloon, 5 aborad)||–||–|
|Soffer et al.||2000‡||161||Lower descending||30||4||5 (2 orad to balloon, 2 aborad)||10 cm (L); 1000 mL||14 (median)|
|Dual barostat balloons|
|Ford et al.||1995||92||Transverse + sigmoid||∼90||6||10 (1 port orad to transverse balloon, 1 port aborad to sigmoid balloon, 4 ports between balloons)||7 cm (D)||8–14|
|Sims et al. Björnsson et al.||1995–2002‡||187158,163||Proximal descending + distal descending||30||0||N/A||8 cm (L); 400 mL||10–11 (mean)|
|Sun et al.|| ||162|| || || || || || |
|Malcom & Camilleri||2000||164||Proximal descending + rectum||∼40||6||5 (1 orad, 5 aborad to proximal balloon)||10 cm (L); 600 mL||–|
|Huge et al. Kreis et al.||2000‡ 2002‡||76188||Proximal to anastomosis following left-sided resection||30||4||2.5 orad and aborad to both proximal and distal balloon||7.5 cm (L)||12–16|
The bag is mounted over a standard manometric catheter by sealing it air-tight at both ends. Two manometric ports need to be dedicated to the bag, one for inflation/deflation and one for pressure monitoring. The lumen of the former needs to be of large enough diameter to minimize resistance to flow along it (typically 1.5–3.2 mm).86,153,156 Other manometric ports sited proximal or distal to the bag can be used for the traditional measurement of intraluminal pressure changes (Table 5).
A pneumobelt, applied around the abdominal wall at the level of the lower costal margin157 may help exclude respiratory excursion- or movement-related (e.g. coughing) artefact during subsequent data analysis.
Measurement of phasic and tonic activity Once the effects of sedative/anaesthetic medications (if used) have worn off, and excess insufflated air has been removed from the colon (in the case of colonoscopy), the barostat bag should be inflated to near maximum to unfold it, and then fully deflated.
With the bag in the desired location, a 30–60-min equilibration period should first be allowed. The minimal distending pressure, defined as that pressure required to overcome intra-abdominal pressure (i.e. sufficiently high to keep the bag from collapsing), needs then to be determined. This is the intrabag pressure at which respiratory excursions are regularly recorded as changes in barostat volume; this measurement is usually performed a priori in every subject.153 The barostat operating pressure required to record intrabag volume variations can then be set;153 this is defined as a pressure level 2 mmHg above the minimal distending pressure. In most studies reported thus far, the operating pressure is between 8 and 17 mmHg74,76,80,81,86,92,112,136,157,158 (Table 5).
Each recording should start with the subject fasted. Volume changes should then be monitored in the fasting (basal) state for a minimum period of 30–60 min.74,81,89,136,153 When the effect of a given stimulus is being investigated, it is recommended that the recording continue until the baseline value is reached again.153 This may not, however, be feasible in all circumstances. Following a meal, a 1.5–3-h postprandial recording period has been reported as adequate.92,156 In pharmacological studies, the length of recording time postdrug administration is dependent, to some extent, upon the drug half-life. For the duration of recording, data should be acquired at 4–8 Hz. Dual barostat-bag studies may allow investigation of reflex (e.g. colo-colonic) mechanisms.159
• Data processing Following computerized exclusion of PVEs (Fig. 9), intrabag volumes can be averaged over a given time increment (typically each minute – the ‘barostat minute volume’,86,136 or every 10 min)81,92 from the start of the recording. For each individual study period (e.g. fasted, postprandial, postdrug administration), a mean volume can be calculated by averaging the incremental volumes recorded during that period.
Because there is large inter-subject variation with regard to the baseline volume (see below), changes in colonic tone, induced by a physiological stimulus (e.g. meal, sleep), drug, disease or otherwise, are usually expressed as a percentage of the individual basal fasted value (taken as 100%).
• Normal values The baseline colonic volume depends upon the site of recording, and also on methodological differences. Mean baseline volume for the proximal, mid-colon and distal colon have been reported as ∼125–200 mL,80,81 150–250 mL,89,92,136,156 and 60–160 mL,74,80,81,92,157,159,161–163 respectively. Percentage change from the baseline volume reflects change in colonic tone; a maximal decrease in volume (reflecting an increase in tone) is observed following ingestion of a meal,80,86,161 whereas sleep induces a maximal increase in bag volume, reflecting a decrease in tone.86
Measurement of wall compliance and tension and colonic sensation
- Top of page
In the same study session, mechanical properties of the colonic wall can be evaluated. As gut sensorimotor functions are inextricably linked, measurement of wall compliance and/or tension can be performed in conjunction with assessment of visceral sensation, which is the other major application for the barostat.153 Such measurements may precede or follow investigation of phasic and tonic volume changes; however, if the former is desired, a conditioning distension should be performed first, as compliance curves will be more reproducible thereafter.157
Colonic compliance is defined as the ability of the colonic wall to stretch in response to an imposed force. The barostat bag should be inflated using an isobaric stepwise (ramp) distension protocol,113 and using corresponding values, the relationship between pressure and volume can then be drawn, with the former plotted on the X-axis, and the latter on the Y-axis. The resultant sigmoid curve reflects:
an initial reflex relaxation (accommodation), during which change in wall tension does not effect volume change;
a linear phase, related in part to the elasticity of the wall;
a final plateau phase.
Various mathematical manipulations have been applied to express a single value for compliance from this curve,153,157,160,161 although no method has yet found universal approval. The colon is less compliant (i.e. the wall is ‘stiffer’) than the rectum, which has a reservoir function; rectal compliance is ∼13 mL mmHg−1.165 Compliance of the transverse colon is reported to be higher than that of the sigmoid (7.6 mL mmHg−1vs 4.1 mL mmHg−1).92 However, results are influenced by distending technique.112,153,160
It is clear that neither pressure or volume, as determined by the barostat, provide a direct measure of wall tone,166 and it has been postulated that the level of wall tension within hollow organs determines the level of perception during distension studies.167 A measurement of circumferential wall tension can be derived using the law of Laplace, defined as the product of transmural pressure (the difference between intra- and extra-luminal pressure) and the radius of the viscus.160 However, this formula may be too simplistic, in that it assumes that the bag takes on a pure shape (spherical or cylindrical) and is not deformed in situ, the wall of the bowel is infinitely thin, and that the pressure external to the viscus is known and evenly distributed.166 Although such studies may be more applicable to tubular organs such as the colon (compared with the stomach or rectum),165 as yet, little data are available on colonic wall tension and its influence on function. More detailed critical reviews of this subject are available.113,160,166,168
At present, the barostat is probably most widely used for the assessment of visceral sensation,112,153 notably in the context of functional bowel disorders.113,114,169 A detailed description of the methods used to determine visceral sensory perception is outside the concept of this manuscript (the evaluation of motor function); several comprehensive reviews of this topical subject can be found elsewhere.112–114,153,170
- Top of page
Simultaneous assessment of intraluminal pressure changes (manometry) and colonic transit (scintigraphy) provides a powerful combination to assess the functional significance of colonic pressure events or motility patterns27–29,51,52,77,172 (e.g. Fig. 5). Such studies have superceded those combining radiological assessment of colonic wall movements with manometry, which would now be considered unethical in terms of radiation exposure.13,127 The technique involves instillation of a suitable radionuclide tracer into the colon, usually via one lumen of a multilumen perfused-tube catheter,27,29,51 during simultaneous manometric measurement of contractile activity. The radiopharmaceutical invariably used is 99mTc (technetium) bound to diethylenetriaminepentaacetic acid (DTPA)28,51 or sulphur colloid;29,77 infusion into the sigmoid colon,27,28,51 right colon28 and caecum29 has been reported. For the purpose of analysis, recorded data are integrated in real-time to assess the influence of contractile activities on tracer movement. This can either be done ‘frame by frame’,27,29 or, for more gross movement, by the creation of regions of interest around colonic segments, which enables time-activity curves to be generated; these provide quantification of the amount of radioisotope, as a function of time, within each defined region.28,171
With regard to acquisition of scintigraphic data, increasing the temporal resolution allows the temporal relationship between tracer movement and defined pressure patterns to be linked with an increasing degree of certainty.29 Conversely, the poorer the temporal resolution, and the wider the spacing between manometric ports, the less definitive the association between intraluminal pressure events and flow. A further concern is that use of a liquid isotope marker may not necessarily be extrapolated to the movement of solid or semi-solid material;29 this requires formal investigation.
Using such techniques, seminal studies by Moreno-Osset et al., Bazzocchi et al. and Reddy et al.27,51,52,171 showed that, in control subjects, movement of instilled radionuclide tracer occurred during both non-propagated and propagated motor activity. Movement in the presence of non-propagated contractile activity was slow, with the direction of flow governed by a pressure differential between adjacent colonic segments (i.e. from areas of high motor activity to those of lower motor activity). Rapid distal bolus movement occurred coincident with aborad-propagating contractions.
More recently, Cook et al.29 have provided contradictory evidence to the classically accepted theory that ‘mass intraluminal movement’ is related to high-amplitude propagated contractile activity alone. They reported that two-thirds of PS (containing either LAPCs or HAPCs) were non-propulsive, and a third of the remainder were only partially propulsive. Although the majority of discrete movements of colonic contents were associated with pressure events, 40% of movements occurred with no discernible temporal relationship to contractile activity. Only 28% of isotopic movements were shown to be coincident with propagated contractile activity, and major movement was often related to LAPCs rather than HAPCs.29
Probably the most technically challenging situation applicable to a combined study of manometry and transit is defaecation. The major limitation is that catheters introduced per rectum may be significantly displaced or even expelled during the act of defaecation itself.65 Previous transit studies have shown that both the descending colon and rectum variably empty during faecal expulsion,28,172–174 but there may be concurrent mass movement of intracolonic contents more proximally.28,174,175 With the advent of improved multilumen perfused tube assemblies,29,53,60,82 the potential for studies of this type is greatly enhanced; fascinating data regarding the manometric correlates of defaecation are already emerging.53,60
- Top of page
The manometric evaluation of colonic function has certainly benefited our knowledge of colonic motor physiology and pathophysiology. Increased understanding of this complex biological system is concomitant with advances in methodology. Improved extrusion techniques have allowed increased numbers of manometric lumens to be included in thin perfused-tube catheters, which enables more recording sites to span the entire colon, thus giving a clearer picture of contractile temporo-spatial relationships. Solid-state technology has given us the tools to carry out prolonged recordings of colonic motility in totally ambulant subjects under better physiological conditions. The barostat enables study of a component of colonic motility (tonic activity) that cannot be measured by other traditional methods. Use of the barostat has shed further light on our understanding of the dissociation between phasic and tonic contractile activities, and also variations in regional tonic activity. The combination of barostat and multilumen perfused-tube manometry currently provides the optimal means to detect all phasic contractile activity, and provides a means to evaluate variations in wall tone. In addition, the development of dedicated computer programs for the automated analysis of recorded signals will hopefully improve the clinical potential of this technique, allowing for better comparison of results between institutions and providing a greater pool of normative data.
Despite such technological advances, the inherent difficulties in accessing more proximal regions of the colon means that our knowledge of the organization of contractile activity throughout the large intestine is limited. Moreover, the physiological/functional significance of those contractile activities identified by manometry remains uncertain; this is of fundamental importance to our understanding of defaecation disorders, and hopefully will prompt further investigation by simultaneous assessment of intraluminal pressure changes and transit. Studies of colonic motility are time-intensive and technologically challenging, notably in terms of data interpretation and analysis. Consequently, there is a relative paucity of data on normal colonic motor function, and until this is expanded upon, our understanding of motility disorders affecting the large bowel will remain inadequate; in terms of colonic motor activity, we are currently unable with any certainty to allow an individual to be clearly placed within, or be differentiated from, the normal population, which is the end-point of any useful clinical test.34
In conclusion, therefore, the clinical value of colonic manometry remains, as yet, unproven, and studies should, at present, be limited to specialist centres. The major goals for all involved in this field must therefore be to standardize (and simplify) many aspects of study, including types of equipment, experimental design, nomenclature and taxonomy of colonic contractile activities, and methods of data analysis. Achieving these goals will undoubtedly encourage others to enter this fascinating, but challenging field, and provide the impetus to develop colonic manometric techniques into recognized clinical tools. This may already be the case in certain paediatric conditions, where colonic manometry has been reported to help guide clinical decision-making.56,57,88 In adults, it would likewise be hoped that colonic manometry will soon be recognized as one of a battery of tests that may allow patients, based on identifiable pathophysiology (i.e. abnormal motility),178 to be more reliably assigned to diagnostic sub-groups, thus providing a more directed therapeutic approach.