Chronic constipation in adults: Contemporary perspectives and clinical challenges. 1: Epidemiology, diagnosis, clinical associations, pathophysiology and investigation

Chronic constipation is a prevalent disorder that affects patients' quality of life and consumes resources in healthcare systems worldwide. In clinical practice, it is still considered a challenge as clinicians frequently are unsure as to which treatments to use and when. Over a decade ago, a Neurogastroenterology & Motility journal supplement devoted to the investigation and management of constipation was published (2009; 21 (Suppl.2)). This included seven articles, disseminating all themes covered during a preceding 2‐day meeting held in London, entitled “Current perspectives in chronic constipation: a scientific and clinical symposium.” In October 2018, the 3rd London Masterclass, entitled “Contemporary management of constipation” was held, again over 2 days. All faculty members were invited to author two new review articles, which represent a collective synthesis of talks presented and discussions held during this meeting.


| INTRODUC TI ON
doctor about their symptoms. 19 The same report also suggested that between 2017 and 2018, almost 200 people were admitted to hospital each day as a result of CC, equating to >160 000 bed days per year, and treatment costs in excess of £160 million, including >£70 million for unplanned admissions and >£90 million for laxatives. 19

| Diagnosis and symptom assessment
Constipation is generally considered a symptom-based disorder, with subtypes able to be defined according to the use of diagnostic criteria, such as the Rome IV criteria, 20 which are advocated together with a limited number of tests to rule out other diagnoses, as well as for ensuring eligibility for clinical trials. 21 The Rome IV criteria allow categorization of disorders of CC into four subtypes: (a) functional constipation (FC), (b) irritable bowel syndrome with constipation (IBS-C), (c) opioid-induced constipation (OIC), and (d) functional defecation disorders (FDDs), the latter including inadequate defecatory propulsion and dyssynergic defecation. [20][21][22] IBS-C is characterized by a combination of pain and constipation; FC by the presence of constipation symptoms without predominant pain and/or bloating (ie, criteria for IBS is not met); OIC as new or worsening symptoms of constipation when initiating, changing, or increasing opioid therapy; and FDD as symptoms compatible with IBS-C or FC in combination with objective signs of disturbed rectal evacuation on diagnostic testing. [20][21][22] Conversely, based on the presence or absence of detectable physiological abnormalities on diagnostic testing, at least three subtypes of CC (which may overlap) have been described: slow-transit constipation (STC), evacuation disorders (ED, which encompasses structural or functional obstructive phenomena that impede stool expulsion), and normal transit constipation. More widespread use of physiological tests to better define clinical/physiological phenotypes could, in the future, pave the way for improved management of constipation. For example, establishing a diagnosis of a FDD implies the need to use a specific treatment such as biofeedback therapy.
Other available diagnostic tools include the Bristol Stool Form Scale, which is a 7-point scale (ranging from separate hard lumps to liquid consistency with no solid pieces) used extensively in clinical practice and research for stool form measurement. 23 This non-expensive and widely available instrument has been shown to be a reliable surrogate marker for whole-gut and colonic transit, 24 and has been adapted into several languages and been modified for use in children.
In CC, as in many other chronic diseases where objective findings correlate poorly with reported symptoms, patient-reported outcomes (PROs) are of great importance to evaluate the effectiveness of treatments and disease progression over time. Tools that currently exist to evaluate PRO measures, developed through literature review and input from patient focus groups, include the Patient-Reported Outcomes Measurement Information System (PROMIS) GI symptom item bank, 25 which captures symptoms in 8 domains, including constipation, 26 and the CC Symptom Severity Measures. 27 The Measure Yourself Medical Outcome Profile (MYMOP), a patient-generated outcome measure allowing patients to select the problems that are the most important to them, has also been used in patients with CC. 28 Additionally, several specific constipation questionnaires have been developed such as the Patient Assessment of Constipation Symptoms (PAC-SYM), a 12-item self-report instrument divided into abdominal, rectal, and stool domains. 29 PAC-SYM has been used in several clinical trials and is considered a reliable and valid tool in adult patients. Other validated scores, such as the Cleveland Clinic constipation score (CCCS) 30 and the Knowles-Eccersley-Scott-Symptom (KESS) score 31 have been developed to identify subtypes of CC more from a clinical than a mechanistic perspective.

| Perceptions of constipation
Despite the existence of formal diagnostic criteria for constipation disorders (eg, Rome criteria), 20,32 evidence suggests that patients and clinicians often diagnose CC more pragmatically, based on the assessment of symptoms they consider important for a diagnosis. Indeed, a study showed that general practitioners (GPs) did not typically use the Rome criteria in clinical practice, and focused only on stool frequency and consistency to diagnose CC. 33 Similarly, a recent cross-sectional study in 2557 members of the general population, 411 general practitioners and 365 specialist gastroenterology doctors demonstrated that only 46%-58% of the general population and 39%-73% of clinicians correctly identified FC when provided with case studies of patients meeting the Rome IV criteria for functional constipation. 34 The same study also highlighted differences in symptoms perceived to be important for a diagnosis of constipation; for example, infrequent bowel movements were most frequently reported as important for a diagnosis by specialist gastroenterologists and colorectal surgeons, compared with less than a third of constipated and non-constipated members of the general population. 34 Moreover, this and other studies indicate that symptoms outside of the Rome criteria, such as pain during defecation and spending a long time on the toilet without passing a stool, are used by the general population to define CC, confirming differences in perceptions of CC between the general population and clinicians. [34][35][36] Such differences may impact patients' clinical care, including diagnosis and treatment, reinforcing the need to also use PRO measures in clinical practice to assess patients' individual needs.

| Disorders with chronic constipation: are they distinct entities?
Accumulating clinical and mechanistic evidence suggests that the different disorders of CC exist on a spectrum rather than being distinct entities, as highlighted in the most recent version of the Rome diagnostic criteria for functional bowel disorders. 20,37 It is further acknowledged that it is sometimes difficult to distinguish one from another, as overlap commonly exists, and that transition from one functional bowel disorder or from one predominant symptom to another is frequently seen. Specifically, considerable overlap between IBS-C and FC exists when mutual exclusivity is suspended, [38][39][40][41] and transition from FC to IBS-C, and vice versa, is common. 41,42 Also, when reviewing studies assessing the pathophysiology of IBS-C and FC, a considerable overlap can be seen, even though certain abnormalities, for example, visceral hypersensitivity seems to be more prominent in IBS, and others, for example, abnormal colonic motility, seem to be more related to FC ( Figure 1). 43

| Pediatric chronic constipation and transition to adult medical care
There are similarities between pediatric and adult constipation (eg, hard stools, with painful and infrequent bowel movements, often accompanied by symptoms of bloating and abdominal pain). 44 However, in contrast to adults, children more often present with coexistent fecal incontinence, caused by overflow of soft stools passing around a rectal fecal mass. Moreover, children rarely complain of the sense of incomplete evacuation or obstruction, or the requirement of manual maneuvers to defecate. Further, differences in response to conventional strategies such as biofeedback therapy and pharmacotherapy, and different surgical outcomes following neuromodulation and antegrade continence enema (ACE) surgery, suggest that childhood functional constipation may be a different entity from adult functional constipation. 44 However, transition to adult care is of fundamental clinical importance, since a long-term follow-up study (median follow-up of 11 years) showed that 25% of children still experience symptoms of constipation as adults and that many continue to have severe symptoms. 45 A recent UK guideline on transition of adolescent and young persons with chronic GI conditions from pediatric to adult care recommends the use of structured transition programs to improve GI disease control, which better ensures adherence to medications, clinic attendance, and clinical outcomes. 46 In addition, such programs may improve psychological outcome and health-related quality of life, and may reduce adverse outcomes such as hospitalization and surgery. Currently, however, outpatient transition clinics exist for patients with inflammatory bowel disease, but are lacking for patients with functional gastrointestinal disorders, including constipation. 47 2.7 | Areas for future research 1. More information about the costs of constipation to health services, and society as a whole, globally.
2. Further exploration of means to facilitate the transition from adolescents with CC to presentation in adult gastroenterology care.
F I G U R E 1 Schematic drawing demonstrating the symptom-based spectrum of functional constipation (FC) and irritable bowel syndrome with constipation (IBS-C), and biomarkers that may be used to discriminate these conditions from each other. From Whitehead et al 43

| Fecal incontinence
FI affects 8%-12% of the adult population [48][49][50] and can have a devastating negative impact on quality of life. 51 Though long-considered to predominantly affect females secondary to obstetric-related anorectal injury, recent epidemiological studies show prevalence is equivalent between genders, 48,49 indicating that pathoetiological factors other than traumatic childbirth must play a role. 52,53 Loose stools and fecal urgency are key, well-recognized risk factors for FI, 48,54,55 but in both pediatric 56,57 and geriatric populations, 58,59 a major underlying cause for FI (in >80% of patients) is considered to be constipation. 60 Unfortunately, this relationship has been grossly neglected in the general adult population. However, recent data indicate that in a sizeable proportion of patients (up to 69%), significant symptoms of FI and CC coexist [61][62][63][64][65] ; with recognition of this frequently overlooked by the referring clinician. 65 Pathophysiology of concurrent FI and constipation is undoubtedly multifactorial, though coexistence suggests some commonality of underlying mechanisms (see Figure 2). 60 evacuation disorder. 68 Other important pathophysiological mechanisms include "overflow" secondary to fecal impaction (eg, with a F I G U R E 2 Schematic drawing highlighting the multifactorial pathophysiological mechanisms common to coexistent fecal incontinence (FI) and chronic constipation (CC) 65,67,70,73 megarectum or severe evacuation disorder, often allied to hyposensitivity), 69,70 or where FI results from incomplete rectal evacuation whether by structural 65,71,72 or functional cause, 73 often in the presence of abnormal anal sphincter function/structure. 65 Acknowledgment of coexistent FI and constipation has major implications regarding management. If FI is indeed secondary to underlying constipation, then intervention directed to improving constipation symptoms and efficacy of evacuation should be considered first-line treatments. Several studies have demonstrated significant improvements or resolution of symptoms of FI when causes of evacuatory dysfunction have been addressed (eg, after surgical repair of rectocele and/or intussusception, [74][75][76] and following colorectal irrigation). 77

| Connective tissue disorders
Constipation is present in up to 50% of patients with connective tissue disorders, be they inflammatory (eg, systemic sclerosis: SSc) or non-inflammatory (eg, hEDS), 78,79 and this is more common in patients who have systemic involvement. 78,80 GI symptoms can precede the systemic manifestations and therefore the diagnosis of a connective tissue disorder.
SSc is characterized by autoimmune-mediated neuropathy, myopathy, and fibrosis within the GI tract. Constipation is most common in those patients with upper GI involvement and can be due to slow gastrointestinal transit 81 or anorectal dysfunction. 82 In severe cases, gut dysmotility caused by the underlying pathology can lead to chronic intestinal pseudo-obstruction. Prucalopride, stimulant laxatives, and in refractory cases, neostigmine can be effective for slow-transit constipation in SSc. 83 Fiber worsens bloating and should be avoided. 84 In terms of anorectal dysfunction, anal hypotension, a reduced or undetectable recto-anal inhibitory reflex 85 and increased rectal sensitivity to balloon distension 86 are typically seen on diagnostic testing. Over time, symptoms of diarrhea and FI can develop due to the development of small intestinal bacterial overgrowth and atrophy of the internal anal sphincter, respectively, both consequences of the underlying connective tissue disorder. 87 Ehlers-Danlos syndrome (EDS) is a non-inflammatory connective tissue disorder characterized by joint hypermobility, tissue fragility, and musculoskeletal symptoms. hEDS is the most common subtype of EDS and is the only one in which the etiology and genetic marker have not been identified. The prevalence of constipation is higher in hEDS than in other EDS subtypes. 88 Constipation can be present from early life, 89 with progression to an alternating bowel habit in some. 90 Symptoms of an ED are very common, 91 and there is a high prevalence of rectal hyposensitivity, 89 dyssynergic defecation, and rectal morphological abnormalities. 62 Colonic transit may be delayed, 88 though this may be secondary to a coexistent ED.
Treatment is holistic, involving lifestyle advice, opiate withdrawal, and laxatives/irrigation therapy as appropriate for the underlying pathophysiology. Surgery for constipation in SSc and hEDS is relatively contraindicated because of the risk of anesthetic complications, wound problems, and postoperative ileus. 92,93

| Following colorectal surgery
Colorectal cancer is the third most common cancer, with 1.8 million new cases diagnosed worldwide in 2018 94 ; the majority of these will be treated surgically, and 50%-60% of patients now survive longterm (>10 years). New symptoms of bowel dysfunction are common post-operatively. For example, symptoms such as FI, fecal urgency, constipation, fragmentation of stool, and frequent bowel movements constitute a major problem following low anterior resection (LAR), 95 which is performed in up to 80% of patients undergoing surgery for rectal cancer. 96 These symptoms are collectively referred to as the low anterior resection syndrome (LARS), with 40%-50% of patients having long-term LARS to an extent that it significantly impairs their quality of life. 97,98 In patients undergoing surgery for colon cancer, a recent study showed that 21% suffered from LARSlike symptoms post-operatively 99 ; this is a similar proportion to the number of patients reporting a sense of incomplete rectal emptying after sigmoid colectomy. 100 The reasons behind the poor functional results after both rectal and colon cancer surgery have yet to be established, though tumor height and location, gender, and preoperative radiotherapy are important factors. Pathophysiology is considered to be multifactorial, 96

| Opioid-induced constipation
Opioids are associated with substantial adverse effects, including those arising from the GI tract such as nausea, vomiting, and constipation, collectively referred to as opioid-induced bowel dysfunction (OIBD). The most prevalent form of OIBD is opioid-induced constipation (OIC) which occurs in up to 87% of patients with pain related TA B L E 1 The Rome IV diagnostic criteria for opioid-induced constipation 1. New, or escalating, symptoms of constipation when initiating, changing, or increasing opioid therapy that must include 2 or more of the following: A Straining during more than one-quarter of defecations B Lumpy or hard stools (BSFS 1-2) more than one-quarter of the time C Sensation of incomplete evacuation more than one-quarter of the time D Sensation of anorectal blockage/obstruction in more than onequarter of defecations E Manual maneuvers to facilitate more than one-quarter of defecations F Fewer than three spontaneous bowel movements per week 2. Loose stools rarely present without the use of laxatives to cancer, although rates are approximately 50% in those with noncancer pain. [106][107][108] OIC is associated with reduced work productivity and quality of life, yet remains underdiagnosed. [109][110][111] The Rome IV criteria define OIC as a change in bowel habit or defecatory patterns, in comparison with normal following initiation, alteration, or an escalation in opioid therapy, see Table 1. 20 Two recent cross-sectional studies comparing symptoms and results of diagnostic testing in constipated patients either currently taking or not taking opioids have shown that opioid use is associated with increased symptom severity, diminution in quality of life, and a greater incidence of rectal hyposensitivity, functional EDs/dyssynergic defecation, and delayed whole-gut transit. 112,113

| Mood disorders
The role of psychological factors has been extensively evaluated in the context of functional gastrointestinal disorders (FGIDs).
Traumatic events, childhood physical, and sexual abuse are independently associated with a higher incidence of FGIDs. 114 The personality trait of neuroticism is particularly associated with constipation. 115 In addition, it is well established that FGIDs are linked with an increased prevalence of concomitant disorders of anxiety and depression although there is controversy as to the directionality of this association. In a large prospective study, Koloski et al demonstrated that higher levels of anxiety, but not depression, conferred an approximate 10% increase in risk of developing a FGID over the subsequent 12 years. 116 In a further study, Jones et al reported that the median time period between diagnosis of an affective disorder and FGID was 3.5 years, compared with a median time period of 1.8 years between a FGID and a diagnosis of an affective disorder. 117 However, this study failed to demonstrate such an association for constipation per se.

| Areas for future research
1. Further evaluation of the cause and effect relationship between constipation and fecal incontinence.

| Colonic dysmotility
Abnormalities of both colonic transit and contractility are commonly associated with CC (see Figure 3). Tests of gut transit (see below) can diagnose a patient with normal or slow-transit constipation; the latter is typically characterized by delayed movement of intraluminal content through the ascending and transverse colon. 9 In patients F I G U R E 3 Schematic of principal (overlapping) pathophysiological mechanisms in chronic constipation found to have delayed transit localized to the distal colon, this may be associated with an ED, though data are conflicting as to whether the transit delay is secondary to the ED or vice versa, or indeed they are independent. 9,118 In patients with STC, abnormal colonic contractility has been characterized to an extent, and in comparison with healthy adults, these patients have a reduction in (a) number of high-amplitude propagating contractions, a propulsive motor pattern associated with mass movement and defecation [119][120][121][122]  and (c) pre-and post-prandial colonic pressurizations, synchronous pressure waves recorded across all recording channels, hypothesized to be associated with gas transport. 122 In constipated patients with normal transit constipation or a distal colonic delay, abnormalities of motor activity remain poorly described.

| Upper gut dysmotility
Esophageal, gastric, and small bowel motility abnormalities have also been described in patients with CC. A recent study showed that in patients reporting overlapping symptoms of dyspepsia and constipation, those diagnosed with STC were significantly more likely to have a coexistent delay in gastric emptying when compared to those with normal transit constipation. 124 In another study of 91 STC patients, 31 were diagnosed with delayed gastric emptying and F I G U R E 4 Representative example of a meal response in the descending and sigmoid colon of a healthy adult (A) and a patient with STC (B). In the top two images (A and B), the entire recording is shown 2 h prior to and after the meal. A rapid increase in colonic activity can be seen in the healthy subject after the meal is given (blue line); this response is not evident in the patient. In (C) and (D), an expanded section of the meal response is shown from the area inside the red-hatched boxes in the top two images. In (C), the retrograde cyclic motor pattern is evident (purple arrow). In the expanded section of the patient trace (D), 2-3 cpm activity can be seen, but the cyclic propagating motor pattern is not evident. In this section of the trace, two retrograde short single motor pattern can be seen (purple arrows). The spatiotemporal pressure plots of (C) and ( 9 had delayed small bowel transit. 125 Manometry studies have additionally shown esophageal and small bowel contractile dysfunction in constipated patients with normal or delayed colonic transit. [126][127][128] Whether such findings represent reflex inhibition of proximal GI motility or a shared primary disorder of the enteric nervous system is unknown.

| Evacuation disorders
Although the medical literature is littered with synonyms (eg, defecation disorder, outlet obstruction, obstructive defecation disorder, and obstructed defecation), "evacuation disorder" (ED) is now the accepted term to describe the clinical and/or laboratory features relating to an individual's inability to satisfactorily expel stool. 1,4,9,12 Clinically, the majority of patients with CC complain of symptoms suggestive of an ED, with straining the most commonly reported

| Sensory dysfunction
Normal defecation requires a conscious sensation of rectal filling and urge to defecate. It is therefore not surprising that impaired or reduced rectal sensation (rectal hyposensitivity: RH; practically defined as a diminished perception to rectal mechanical distension, manifest as elevated sensory thresholds), is associated with disordered defecation. Observational studies indicate RH is found in up to 60% of constipated patients, 10% of patients with fecal incontinence, and 27% of individuals with symptoms of both. 70,145 In the largest study published to date (2876 patients), 25% of patients were found to have RH, with a linear relationship existing between the number of elevated sensory thresholds to rectal distension and constipation severity. 146 The etiology of RH is uncertain, but a number of mechanisms have been proposed. In some patients, in whom there is documented disruption of the afferent pathway (eg, due to pelvic nerve damage or spinal cord injury), 147,148 there is a clear cause-effect relationship with development of RH. For example, 78% of patients with complete spinal cord injury and hindgut dysfunction, and 43% of individuals with incomplete lesions have RH. 149,150 RH is also an important mechanism in patients with constipation following stroke 151 and associated with multiple sclerosis. 152 In others, behavioral inattention to defecation (ie, voluntary withdrawal of attention from rectal sensations and/or habitual suppression of the desire to defecate) is a likely factor. 153 However, in the majority of patients with chronic constipation, it remains unclear whether RH is a primary pathology leading to increasing severity of symptoms, whether chronic constipation itself results in the development of RH, or if indeed RH is an epiphenomenon. 113,154 With regard to pathophysiology of RH in CC, this is considered to be either due to dysfunction of the afferent pathway ("primary" RH), as a result of altered rectal wall biomechanics (ie, increased capacity or hypercompliance ("secondary" RH: see Figure 5), or both. 155 Symptomatically, RH is associated with "no urge constipation" 156,157 and is more common in individuals meeting the Rome IV criteria for F I G U R E 5 Rectal pressure-volume relationships determined through use of the electromechanical barostat (phasic isobaric distension protocol) in constipated patients with rectal hyposensitivity (RH) to balloon distension, constipated patients with normal rectal sensation (NS), and healthy volunteers (HV). Both rectal capacity (reflected by elevated distension volumes) and rectal compliance (steeper slope of the linear section of the curve) are increased in patients with RH. Adapted from Gladman et al 228 (with permission from Wolters Kluwer) FC (60%) rather than IBS-C (2%). 145 RH appears to be linked primarily to ED 138 and particularly with "functional" rather than a mechanical (anatomical) obstruction to defecation. 12,32,70,138,158 RH impacts CC via two key mechanisms: (a) through its association with functional ED either directly, due to co-incident/ corresponding efferent dysfunction (ie, concurrent reduced rectal contractility), 154,159 or indirectly via the development of large, hard, and difficult to evacuate stools due to fecal retention and desiccation secondary to reduced awareness; and (b) due to colonic transit delay via inhibitory feedback loops triggered by chronic rectal distension. 160,161

| Genetic factors and enteric neuropathies
The question of whether CC (especially STC) occurs as a result of an enteric neuropathy, and whether this might be genetically deter-  166 It seems likely that a search for a genetic etiology will be consigned to genome-wide association studies, which have determined weak susceptibility factors in IBS-C, 167 or to recognition that epigenetic factors may be more important.
The question of whether an enteric neuropathy underpins the transit disturbance in STC is mired by issues of defining neuropathy histologically. This issue is part technical (right specimen, adequate sampling, right preparation, right staining, etc.) and part interpretive, the latter being especially problematic when neuronal quantification is attempted. 168 Thus early reports using sledge microtome thick sections and silver staining 169,170 should be discounted in favor of modern approaches. 171 With the exception of a single high-quality study from Germany, 172 it is fair to summarize that there is no strong evidence for neuropathy based either on cytoskeletal evidence of cell degeneration or of quantifiable neuronal loss (based on 12 studies from the modern era). 171

| Dysbiosis
Numerous case-control studies have now compared the gastrointestinal microbiome between CC, IBS-C, and healthy controls. In general, studies in adults report lower bifidobacteria and bacteroides in constipation, with some also reporting lower lactobacilli, although these findings are not consistently demonstrated in pediatric patients. 173 One noteworthy study using 16S ribosomal RNA gene sequencing to measure both stool and mucosal microbiome reported marked differences in mucosal microbiome at both the family level (lower proportions of Comamonadaceae and Odoribacteraceae, higher Flavobacteriaceae and Caulobacteraceae) and genus level (higher Flavobacterium and Mycoplana, lower Delftia and Odoribacter). 174 Multivariate analysis (adjusting for age, body mass index, diet) showed that although stool microbiome composition was significantly associated with colonic transit time, it was mucosal microbiome composition that was significantly associated with constipation even after adjusting for transit time. 174 In terms of microbiome metabolites, there is inconsistent evidence regarding differences in short-chain fatty acids (SCFAs) between constipated and healthy subjects; one challenge being that slower colonic transit time can reduce stool SCFA by increasing their colonic absorption rather than decreasing their production. 175 Meanwhile, positive methane breath tests to a carbohydrate challenge have been shown in some (but not all) 176 studies to be more common in people with STC than both normal transit constipation and healthy controls, [176][177][178] and also in IBS-C compared with diarrhea-predominant IBS. 179 Nevertheless, there appears to be no correlation between methane production and constipation symptom severity. 180   2. There is recognized overlap between ED and delayed whole gut/ colonic transit. The question as to which is primary and which is secondary (or do they coexist?) remains an area of debate and warrants further research.

While the presence of RH can negatively impact on treatment
outcomes, 182 RH itself has been proposed as a therapeutic target. 183 Specific sensory bowel retraining therapy has been shown to improve sensory dysfunction (in up to 92% of individuals) with corresponding improvement in symptoms. 184,185 However, further high-quality controlled studies are required and this is an orphan area for drug development.
4. Further studies are also required to define the overall clinical impact of RH in hindgut dysfunction.
5. An understanding of dysbiosis and its effects on nerves either directly or via longer term epigenetic changes in nerves or glia are needed. Fundamental studies are required to understand how enteric neurons survive, if they turnover and whether they do so from glia or neuronal progenitors.

| FUN C TIONAL DIAG NOS TIC INVE S TIG ATIONS
Advanced diagnostic studies of colonic, rectal, and anal function are recommended in patients in whom organic disease has been excluded, who have failed first-line conservative therapies, such as optimization of stool consistency, bowel habit training, and lifestyle advice, and who are also refractory to standard pharmacological treatments. 186,187 The aim of investigation is to provide clinically relevant measurements that explain the cause of symptoms, identify

Screening, advanced, or experimental
Resources required a

Radio-opaque markers
Screening + Delayed whole-gut transit Scintigraphy Advanced +++ Delayed colonic transit Delayed regional GI and whole-gut transit (extension of technique)

Wireless motility capsule
Advanced ++ Delayed regional GI and whole-gut transit Regional GI dysmotility; dysbiosis/ altered colonic fermentation?

3D-Transit capsule
Experimental ++ Delayed regional GI and whole-gut transit

Tests of gut contractility
Colonic manometry pathology, and guide effective management. No one technique provides a complete description of defecation; instead, a combination of tests to evaluate structure, and motor and sensory function are generally employed. 129 Unfortunately, inconsistency in approach exists due to conflicting data on the usefulness of these investigations for decision-making, variability in local expertise, and resource availability. Below is a description of the use, merits, and pitfalls of the most commonly employed techniques (see also Table 2).

| Tests of gut transit
In patients with CC, tests of transit may be useful in those who report infrequent defecation. Although traditional radiological methods (radio-opaque markers [ROM] and scintigraphy) tend to focus on quantification of colonic transit, it is now appreciated that dysmotility is not necessarily restricted to the colon (see above) and therefore newer techniques (wireless motility capsule [WMC] and 3D-Transit method) have the ability to assess pan-enteric function.
Radio-opaque markers testing is considered a screening investigation, and is indicated to differentiate between normal and slow whole-gut transit (often reported as "colonic" transit time, though this is overestimated, as oro-cecal transit is a mean of 8 h, even in healthy volunteers). 188 The test is inexpensive and easy to perform, and results correlate with stool form; for example, prolonged transit time is associated with hard stools. Several ROM protocols exist; the simplest involves the intake of markers (typically 20-50) at a single time-point, followed by a single abdominal X-ray, usually after 120 h 189 ; transit is defined as abnormal if >20% of markers are retained at the time of the X-ray (see Figure 6A). 189 This provides a surrogate of disease severity that is more accurate than subjective assessment of fecal loading on an X-ray film. 190 Alternatively, markers can be taken on consecutive days, enabling assessment of mathematically-derived whole-gut and regional colonic transit times. 191,192 However, such calculations are based on the assumption that transit time is a continuous variable; recent studies employing other methods (eg, WMC and 3D-Transit) have demonstrated, in sizeable healthy volunteer studies, that whole-gut transit times are, in fact, clustered at intervals separated by approximately 24 h. 188,193 Scintigraphy is recognized as the "gold-standard" method for assessing colonic transit time, but availability is limited to a few specialist centers. 194 The technique involves following the progress of a radioisotopic chemical (eg, 111 Indium) through the GI tract using a gamma camera and taking serial scintigrams. 195 Diagnosis of delayed colonic transit is determined by position of the geometric center of the isotope mass at given time points. The test can be extended to include assessment of gastric and small bowel transit also. 196 The WMC (SmartPill) and 3D-Transit system (Motilis Medica, SA) are ingestible capsule devices. The WMC is commercially available, and measures pH, temperature, and intraluminal pressure as it traverses the gut. 192,197 Total and regional gastrointestinal transit times are determined from stereotypical changes in pH and temperature (see Figure 6B); however, accurate information on segmental colonic transit times cannot be determined. The test has been validated against ROM and scintigraphy. 197,198 At present, the 3D-Transit system is an experimental investigation only; through tracking of the location and orientation of ingested electromagnetic capsules, it offers the potential to assess total and regional GI transit times, segmental colonic transit times, 193 and also colorectal motility patterns. 199

| Tests of colonic motility
Colonic manometry can be employed for advanced investigation of colonic motility in highly selected patients. 187 It is generally used to determine the presence or absence of colonic motor patterns in response to physiological and chemical stimuli. 200 Several protocols exist, but most commonly, after bowel preparation, a catheter incorporating >20 recording sensors, spaced between 10-30 mm apart, is placed into the colon with the aid of a colonoscope. 201 With the subject awake, data is collated over a 1-2 h period before and after a high-calorie meal. This can then be followed by assessing the colonic response to intraluminal infusion of a stimulant laxative (eg, Bisacodyl). 202 Though information on colonic motility patterns and propagating sequences can be acquired, standardization of the technique is in its infancy. 202 Catheter types, the number of sensors and the spacing between them, types of meals and the analysis and interpretation of results all differ among different centers. Further, due to a limited number of studies in healthy subjects, a clear definition of "normal" colonic motility is lacking. Nevertheless, high-resolution techniques are revealing motor patterns that were previously undetected using conventional technology. 203 The use of magnetic resonance imaging (MRI) to investigate colonic motility is experimental at present, but allows for the direct visualization of either colon wall 204 or the contents of the lumen. 205 Using cine MRI and post-processing techniques, 206 it is possible to track colon wall motion prior to and following a laxative challenge 207 and to visualize how the luminal contents are moving, using spatiotemporal maps of luminal contraction and dilation. Further, assessment of colonic wall motion can be combined with other MRI measures, for example, transit 208 and volume, 209 to provide an objective view of the colon, both in terms of anatomy and function.

| Tests of evacuation
Tests of evacuation are useful in patients with CC, especially those who report symptoms of ED. The balloon expulsion test (BET) and defecography are direct measures of the ability to expel rectal contents, whereas high-resolution anorectal manometry (HR-ARM) is an indirect test of evacuation, which, in this setting, provides information primarily on recto-anal co-ordination.
The BET is a widely used, simple, inexpensive, office-based test that is easy to interpret. It is used as a screening investigation of evacuation, and provides quantitative information on the time taken to expel a 50 ml water-filled rectal balloon. The upper limit of normality is generally accepted to be between 1-3 min, and expulsion times that exceed this suggest impaired evacuation. 210 Novel evacuation patterns are also described using preload-afterload pressure analysis. 225  The recognized gold-standard for determining visceral sensation to distension is the electromechanical barostat, 226,229 a computer-controlled piston connected to a distensible bag with a volume larger than the viscus being examined (ie, the bag is effectively infinitely compliant). The bag is inflated and deflated automatically to maintain a constant pressure within the rectum while volume is continuously recorded. 230 Barostat recordings are not affected by the recoil properties of the bag, hence permitting the determination of rectal wall bio-elastic properties (eg, compliance and capacity: see Figure 5). 226,228,231,232 The test is reproducible across laboratories and between patients. 229 2. To further develop comprehensive testing to assess the relative importance of colonic motility, rectal motor-sensory function, and patient behavior at the level of the individual patient.

| Tests of rectal sensation
3. Large, preferably multicenter studies in healthy adults are needed to established normal ranges of colonic motility patterns in response to standardized stimuli (meals, sleep, chemical stimulation).
4. MRI "motility" techniques need to be further developed to cover all segments of the colon, thus avoiding the need for more invasive tests.

Given the acknowledged major overlap between health and CC, 223
and poor agreement between results of different tests, 144,224 the optimum method(s) for the diagnosis of dyssynergic defecation requires re-evaluation.
6. Assessment of the impact of the London Classification for disorders of anorectal function; it is hoped that the use of a common language to describe results of diagnostic tests, standard operating procedures, and a consensus classification system will bring muchneeded standardization. This is projected to facilitate co-operation between centers and the performance of multicenter studies, a key requirement if progress is to be made toward improved diagnosis and individually tailored therapy of patients with these conditions.

| SUMMARY AND CON CLUS IONS
Chronic constipation is common, problematic for the sufferer and complex for the physician. This review has sought to coalesce up-todate information on several key aspects of CC. The content should provide an educational resource for the reader with a clinical or research interest in this disease area. It also frames the background for the sister review addressing therapy. It is hoped that some of the many areas of future research outlined will be addressed by a future generation of readers.