BACKGROUND AND AIMS:

Because symptoms alone do not identify pathophysiology or differentiate subgroups of constipation, diagnostic tests are generally recommended. However, their utility is not known. We performed a systematic review of diagnostic tests commonly used in constipation.

METHODS:

We searched the English literature using MEDLINE and PUBMED databases from 1966 to 2004 for studies in adults published as full manuscripts whose methodological quality was above a minimum score.

RESULTS:

No studies assessed the routine use of blood tests or abdominal x-ray. One retrospective endoscopic study showed that cancer and polyp detection rate was comparable to historical controls. Two studies of barium enema were unhelpful in diagnosis of constipation. Physiological studies showed differences in study population, methodology, and interpretation, and there was no gold standard. Ten colonic transit studies showed prevalence of 38–80% in support of slow transit constipation. Nine anorectal manometry studies showed prevalence of 20–75% for detecting dyssynergia. Nine studies of balloon expulsion showed impaired expulsion of 23–67%. Among 10 defecography studies, abnormalities were reported in 25–90% and dyssynergia in 13–37%.

CONCLUSIONS:

Evidence to support the use of blood tests, radiography, or endoscopy in the routine work up of patients with constipation without alarm features is lacking. Colonic transit, anorectal manometry, and balloon expulsion tests reveal physiologic abnormalities in many selected patients with constipation, but no single test adequately defines pathophysiology. Large, well-designed, prospective studies are required to examine the utility of these tests.

INTRODUCTION

Constipation is a symptom-based disorder. It is either a consequence of neuromuscular dysfunction that affects the colon and anorectum or it is secondary to a variety of conditions (¹). In the absence of secondary causes, functional constipation is often defined as persistent symptoms of difficult, infrequent, or incomplete evacuation of stools. Although physicians have often used the term constipation to describe infrequent bowel movements (²), patients have used this term to describe a broader set of complaints including hard stool, straining at defecation, feeling of incomplete evacuation, use of digital maneuvers to assist defecation, and abdominal discomfort or fullness (^{2,3,4,5,6,7,8}).

In order to obviate some of the inherent deficiencies of defining this heterogeneous clinical condition, a consensus definition was proposed by a group of experts (Rome II) in 1998 (⁹). The Rome II committee defined functional constipation as the occurrence of two or more of the following symptoms for at least 12 wk, which need not be consecutive, in the preceding 12 months and in the absence of structural or biochemical explanation:

1. Straining 1/4 of defecation;
2. Lumpy or hard stools 1/4 of defecation;
3. Sensation of incomplete defecation 1/4 of defecation;
4. Sensation of anorectal obstruction/blockage 1/4 of defecation;
5. Manual maneuvers to facilitate 1/4 of defecation (example: digital evacuation, support of the pelvic floor);
6. Less than three defecations per week.

Additionally, loose stools should not be present, there must be insufficient criteria for irritable bowel syndrome, and the criteria do not apply when a patient is taking laxatives (⁹). In contrast, dyssynergic defecation is defined on the basis of the aforementioned symptom criteria for functional constipation (⁹) together with the presence of the following abnormalities on physiologic testing; manometric, EMG, or radiologic evidence of inadequate propulsive forces or failure to relax the pelvic floor muscles during repeated attempts to defecate, evidence of adequate propulsive forces, evidence of incomplete evacuation, or evidence of impaired expulsion forces (^3,8).

Although, symptom-based definitions help to identify patients with this condition, they provide little pathophysiological information. Furthermore, symptoms alone do not appear to differentiate between the subgroups of patients with constipation (^3,4,5,6,7,8). For instance, patients with normal colonic transit and normal anorectal manometry have been shown to report symptoms that are similar to those with slow colonic transit and dyssynergia (^6,7,8). Finally, a diagnosis of dyssynergia requires both symptoms and physiological tests (^3,8). Hence, physiological tests may aid diagnosis, potentially identify subtypes of constipation, and facilitate treatment.

On the basis of the underlying pathophysiology, at least three subtypes have been recognized, although there is a significant overlap between these groups (^10,11). Slow transit constipation may be due to a primary dysfunction of the colonic smooth muscle (myopathy) or its nerve innervation (neuropathy) or could be secondary to dysfunction of the pelvic floor such as dyssynergic defecation (^12,13,14,15). Dyssynergic or obstructive defecation (¹¹) also known as anismus (¹⁶) is characterized by significantly difficult or an inability to expel stool from the anorectum. Many patients with dyssynergic defecation may also have prolonged colonic transit (^8,17,18). A third subtype may exist and is comprised of patients with complaints of constipation but with normal transit and normal pelvic floor function (^10,19). Many of these patients may have irritable bowel syndrome, and abdominal pain is often a prominent symptom with altered bowel habit (^10,20).

A number of tests have been proposed for the diagnosis of constipation and its pathophysiological subgroups and to rule out structural and biochemical disorders (^1,3,5,21), but the utility of these various tests in the management of patients with constipation is incompletely understood. Furthermore, there has been no systematic review of the utility of diagnostic tests that may help to differentiate either patients with functional constipation from those with organic disorders or the subtypes of patients with functional constipation.

METHODS

Literature Search, Data Extraction, and Analysis

Searches of the online databases PUBMED and MEDLINE were performed and relevant manuscripts published in English between 1966 and 2004 were reviewed. The search was confined to adults with constipation and hence articles related to children were excluded. The search terms used included constipation, colonic diseases, anorectal manometry, colonic transit, balloon expulsion, defecography, proctography, colonoscopy, barium enema, defecation, and systemic or local disorders that cause constipation for example, diabetes, colon cancer, or hemorrhoids and Rome criteria. All abstracts were screened and potentially relevant articles were researched and bibliographies were reviewed.

In order to determine the utility of diagnostic tests that are commonly used in the evaluation of patients with constipation, standard forms were used to abstract data regarding study design and the following inclusion criteria were used:

POPULATION

Patients with constipation or matched controls. The definition of constipation has been extremely variable or inadequately described in most studies and there is no single symptom or test that defines this condition. Therefore, we included any study which stated that patients with constipation were examined.

EXPOSURE

A diagnostic test for constipation. The diagnostic tests sought included the following, which have been widely recommended (^{1,3,5,8,21,22}); hematology and biochemical profiles, plain abdominal x-ray, barium enema, flexible sigmoidoscopy or colonoscopy, colonic transit study, anorectal manometry, balloon expulsion test, and defecography.

OUTCOME

The number of patients with organic disease (for blood tests, radiography, endoscopy) or an abnormal physiologic test. Potentially relevant articles were reviewed by two authors independently (SSCR and RO). Those meeting the inclusion criteria were reviewed for a quantitative assessment of the study methodology using previously described criteria and data were extracted for the review (^23,24).

The quality of each study was then scored using the study characteristics described in Table 1, and by following previously validated methodology (^23,24). The data were extracted about: (a) diagnostic evaluation performed; (b) prevalence of confirmed organic GI disease; (c) frequency in which an alternative diagnosis was confirmed to explain symptoms; or (d) prevalence of abnormal results on physiologic testing. Additional data were extracted as described in the tables. We only included studies that received a methodological score of 3 or above (²³). Because study methodology was not uniform, the results obtained were extremely diverse and it was not possible to combine and perform a metaanalysis. Therefore, data are presented in a tabular form. Discrepancies between the data extracted by the two authors were reconciled by mutual consensus.

Table 1 - Qualitative Assessment of Methodologic Quality [Adapted from (23, 24)].

Full table

RESULTS

Role of Laboratory Blood Tests

We were unable to identify any study that examined the utility of routine hematological and biochemical tests and thyroid functions tests in the evaluation of patients with constipation. Some studies stated that metabolic disorders that cause constipation were excluded. Thus, there is little evidence to support or reject the utility of these screening blood tests.

Role of Plain Abdominal x-Ray

We were unable to identify any study that examined the utility of this test in the evaluation of patients with constipation, and met our inclusion criteria, although 112 articles were identified by our search. Thus, there is little evidence to support or reject their use in the routine evaluation of constipated patients.

Role of Flexible Sigmoidoscopy or Colonoscopy

We found 162 articles related to colonoscopy and constipation but only a single study was identified that met our inclusion criteria (Tables 2 and 3). In this large retrospective study of 563 patients with constipation undergoing endoscopic evaluation (358 had colonoscopy and 205 had sigmoidoscopy), 1.6% of patients were found to have colon cancer and 14.4% had colonic adenomas (²⁵). In those undergoing colonoscopy, the yield of cancer and adenoma was 2.0% and 19.3%, respectively, and for those having sigmoidoscopy the yield was 1.0% and 5.9%, respectively. No control group was studied but the authors suggested that their yield for colon polyps or cancer was similar to the rates seen in studies of screening in asymptomatic populations (^26,27).

Table 2 - Qualitative Assessment of Methodology for Selected Studies on Lower Endoscopy and Barium Enema in Patients with Constipation.

Full table

Table 3 - Summary of Studies Evaluating the Prevalence of Constipation Using Lower Endoscopy and Barium Enema.

Full table

Barium Enema

There are insufficient studies to evaluate the clinical utility of barium enema in adult patients with constipation: Only two articles were identified that fulfilled our inclusion criteria (Tables 2 and 3), although our research revealed 207 articles pertaining to this test and constipation. In one retrospective study of 62 individuals with chronic constipation (defined as 3 stools per week) who had a barium enema before anorectal myectomy, no patient was identified with an "organic lesion" or a narrowed segment (²⁸). In another retrospective study, 791 patients with a definite diagnosis (including normal) on barium enema were assessed (²⁹). Constipation was reported in 22% of this population and was just as likely to be present in those with an abnormal study as in those with a normal study (OR = 0.94, CI = 0.61–1.44) (²⁹). Furthermore, in a patient presenting with a change in bowel habit, often regarded as an important and an alarm feature, a barium enema by itself could not exclude organic disease (²⁹).

Physiologic Testing

Physiologic testing with colonic transit study, anorectal manometry, balloon expulsion test, or defecography was performed in patients mostly referred to tertiary care centers with severe chronic idiopathic constipation and who generally had negative evaluations and a lack of response to therapy. Thus, the studies assessing the utility of these tests identify the proportion of patients who demonstrate various physiologic abnormalities. These studies were all from clinical series but none of them employed a gold or reference standard, only rarely was a blinded assessment performed, and very few were prospective series of consecutive patients. Although adequate details of the diagnostic tests were provided in most instances, the details regarding the population studied were variable and these populations also exhibited a range of symptoms, e.g., infrequent stools, difficulty with evacuation without infrequent stools. Selected characteristics for each test and the qualitative scores of each study are included in Tables 2, 3, 4, 5, 6, 7, 8, 9, 10, 11.

Table 4 - Qualitative Assessment of Selected Studies of Colonic Transit in the Diagnostic Evaluation of Patients with Constipation.

Full table

Table 5 - Summary of Colonic Transit Studies in the Diagnostic Evaluation of Patients with Functional Constipation.

Full table

Table 6 - Qualitative Assessment of Methodology for Selected Studies on Anorectal Manometry in Patients with Constipation.

Full table

Table 7 - Summary of Studies Using Anorectal Manometry in the Diagnostic Evaluation of Patients with Functional Constipation.

Full table

Table 8 - Qualitative Assessment of Methodology for Selected Studies on Balloon Expulsion Test in Patients with Constipation.

Full table

Table 9 - Summary of Studies Using Balloon Expulsion Test in the Diagnostic Evaluation of Patients with Functional Constipation.

Full table

Table 10 - Qualitative Assessment of Methodology for Selected Studies on Defecography in Patients with Constipation.

Full table

Table 11 - Summary of Studies Using Defecography in the Diagnostic Evaluation of Patients with Functional Constipation.

Full table

Colonic Transit Study

Colonic transit time has been measured using two general methods; ingestion of radio-opaque markers followed by performing plain x-rays of the abdomen taken at variable times (^30,31) or by using radioisotopes and scintigraphy (^32,33,34). This review focused on the use of radio-opaque marker technique for the assessment of colonic transit because they are widely used in clinical practice for the diagnosis of functional constipation in adults. Although scintigraphic studies have been shown to be validated, reliable, and reproducible, they have been used in a few specialized, tertiary care centers and are expensive. In contrast, the radio-opaque marker technique has been more widely used and is cheaper.

Ten studies of colonic transit assessment were identified in functional constipation (^{6,8,35,36,37,38,39,40,41,42}) that fulfilled our selection criteria out of 309 articles encountered in our search. Five of these studies were of medium-high quality: i.e., they scored in the mid-to-high quartile as described previously (^23,24). Details of these studies are summarized in Tables 4 and 5. The studies also had differences in the definition of constipation and importantly there was no gold standard. The colonic transit study was also the gold standard. Thus, these studies examined the proportion of patients with constipation who showed an abnormal test.

Furthermore, no uniform methodology was employed by the investigators to perform the colonic transit study (Tables 4 and 5). For example, some studies prescribed a fiber intake of 25 g (^35,37) whereas others (^8,38,39) did not impose any dietary restrictions during the test. Likewise, some investigators administered 10 radio-opaque markers daily for 6 days (³⁸), others a single capsule with 24 markers (^8,42); some obtained x-rays daily for 8 days (^35,36), whereas others used a single x-ray at 120 hr (^8,40), (Tables 4 and 5). The interpretation of test results has also been variable; some authors defined slow transit as retention of 44% of markers (⁴⁰) whereas many others used retention of 20% markers as abnormal study (^8,42). Consequently, an abnormal study (i.e., evidence to support the presence of slow transit constipation) varied from 38 to 80%. One study assessed the reproducibility and concluded that colonic transit time measurements were reproducible in 35/51 (69%) patients (⁴²). However, the correlation coefficient for reproducibility was higher in patients with normal transit constipation (0.60) compared to those with either slow transit constipation (0.12) or dyssynergia (0.21).

Anorectal Manometry

Although our search revealed 742 articles related to constipation and anorectal manometry, only nine studies (Tables 6 and 7) were identified that fulfilled our selection criteria (^{8,17,39,40,43,44,45,46,47}). Analysis of these studies revealed significant differences in test methodology, both with regards to test performance and interpretation of data. Also, the concept of dyssynergia as a cause of constipation has only evolved over the last decade, and there is no generally acceptable method of identifying this condition. There is some agreement in published studies that the mere presence of a dyssynergic or obstructive pattern of defecation seen on either manometry or electromyography studies does not qualify for a diagnosis of dyssynergic defecation (^3,8,11,48). Rome II diagnostic criteria requires the presence of symptoms of functional constipation together with either manometric, EMG, or radiologic evidence of paradoxical anal contraction, adequate expulsion force, and impaired evacuation (³). Some patients may exhibit impaired expulsion force and this may also constitute dyssynergia (^8,17). Overall, two studies (^39,43) had a lower positive yield for dyssynergia (20–30%), whereas the remaining seven studies had a yield varying from 44 to 75% (Tables 6 and 7). In two studies, manometric tests yielded new information that was not detected clinically and this formed the basis for a change in the management of 65–67% of patients (^46,47).

Balloon Expulsion Test

Nine studies were identified (^{8,12,40,48,49,50,51,52,53}) that met our inclusion criteria out of 45 articles and provided information on the proportion of patients with abnormal balloon expulsion (Tables 8 and 9). However, the methodology for balloon expulsion has not been standardized. Several techniques have been used including, 25 ml (⁵⁰) or 50 ml balloons (^5,49,50), 18 mm spheres (⁵⁰), silicon-filled artificial stool (⁵⁴), or weights attached to a pulley to assess the extra force required to expel a metal sphere in the lying position (⁵⁵). In a recent study, the balloon was filled with water until the subject reported a desire to defecate and the mean volume of fluid infused into the balloon to achieve this sensation was 250 ml (⁴⁸). In contrast, other studies have used balloons with fixed volumes of water (^5,49,50) or air (⁵¹). Consequently, a comparative interpretation of the reported results of this test is problematic. Also the upper limit of the time taken by a patient to expel the device and for the test to be categorized as abnormal has either been variable or not defined. Some have used greater than 1 min (⁴⁸), and others 5 min (^5,50) as the upper limit for abnormal testing. The prevalence of a positive test varied between 23 and 67%. Three studies used a control group. The prevalence of difficulty with evacuating balloons in healthy controls was 0–16% (Tables 8 and 9).

Defecography

Ten studies were identified (^{8,40,41,43,52,53,56,57,58,59}) out of 190 articles related to the use of defecography in patients with constipation. The overall methodological scores were low (Tables 10 and 11). The prevalence of normal studies varied between 10 and 75%, and likewise the prevalence of abnormal findings which supported a diagnosis of constipation varied between 25 and 90%. The prevalence of dyssynergia varied between 13 and 37%. Although most studies used a combination of findings including a decreased anorectal angle and/or impaired evacuation of barium contrast to diagnose dyssynergia some have used other definitions (Tables 10 and 11). Also, it has been reported that the normal values of rectal emptying vary considerably (^5,43). Furthermore, the percentage of rectal evacuation during defecography does not correlate with symptoms (⁵). In one study, there was no difference in the evacuation patterns between controls and patients with normal or slow transit constipation or dyssynergia (⁴³). Also, the utility of identifying other abnormalities such as a rectocele or mucosal intussusception was unclear and their prevalence was also quite variable (Tables 10 and 11). Some investigators have regarded small rectoceles (2cm) as inconsequential (⁵⁶) whilst others have reported rectoceles of all sizes (⁵⁸).

DISCUSSION

It is generally recommended that the first step in the evaluation of a patient with constipation is to exclude an underlying organic cause (^1,3,5,22). A complete blood count, biochemical profile, serum calcium, blood glucose, and thyroid function tests are routinely recommended in the evaluation of patients with constipation (^1,3,5,22). However, there is little evidence to support or reject the utility of these screening blood tests. Based on history and physical examination and if there is a high index of suspicion, additional tests, such as serum protein electrophoresis, urine porphyrins, serum parathyroid hormone, and serum cortisol levels, may be requested to identify myeloma, porphyria, hyperparathyroidism, and Addison's disease (¹). It must be recognized that these conditions are rare causes of constipation and judicious use of these tests rather than routine use is recommended. Also, most of the published literature is from tertiary care centers. Whether many patients with underlying metabolic disorders were screened by a primary care physician and excluded before a referral was made to a tertiary care center cannot be discounted. Finally, a lack of evidence does not necessarily refute the potential benefit of a diagnostic approach.

An evaluation of the distal colonic mucosa through a flexible sigmoidoscopy may reveal evidence for chronic laxative use such as melanosis coli or other mucosal lesions such as solitary rectal ulcer syndrome, inflammation, or malignancy. However, the role of endoscopic evaluation in the evaluation of constipation is controversial. A recent position paper by the American Gastroenterology Association recommended that a structural evaluation of the colon is appropriate for patients with constipation (¹). In contrast, constipation was not included in the recommended indications for colonoscopy or flexible sigmoidoscopy by the American Society of Gastrointestinal Endoscopy (⁶⁰). In a single, large retrospective study of 563 patients with constipation undergoing colonic surveillance for colon cancer, the range of neoplasia found and the polyp detection rate was comparable to that expected in asymptomatic historical controls (²⁵).

Thus, chronic constipation alone without alarm symptoms, such as rectal bleeding, obstruction, abdominal mass, or weight loss, may not be an appropriate indication for lower endoscopy. However, age-appropriate colorectal cancer screening should be performed with colonoscopy in patients presenting with constipation (^26,27). A prospective study of age and gender matched constipated patients and controls is required to answer this question definitively.

A plain x-ray of the abdomen may provide evidence for an excessive amount of stool in the colon from either proximal stasis or fecal impaction (⁶¹), but is a poor indicator of colonic transit time. There has been no systematic evaluation of the use of this test in the diagnosis of constipation. If colonoscopy has not been performed, a barium enema may be useful for excluding colonic pathology. Patients with constipation may have a redundant sigmoid colon, a megacolon or megarectum, or colonic obstruction from volvulus or neoplasia (^5,61). The presence of Hirschsprung's disease can also be detected by a barium enema, although manometry and histology are required to confirm this diagnosis (^62,63). In children, barium enema appears to be a good initial screening test as it correlates well with manometry and biopsy (⁶³) for identifying Hirschsprungs disease. However, no appropriate study of barium enema was available to assess its utility in adult patients with constipation. In one series of highly selected patients with chronic constipation who were studied prior to an anorectal myectomy, a barium enema did not aid in the diagnosis (²⁸). However, this group of patients provides virtually no useful information about the general population of patients with constipation or even severe, chronic constipation, since they all were undergoing a myectomy, suggesting other disorders had been excluded. Another study suggested that the likelihood of having constipation was similar in patients with abnormal and normal barium enemas (²⁹), but again, this study did not assess the clinical utility of barium enema in patients with constipation. Furthermore, in a patient presenting with a change in bowel habit, an important and alarming symptom, a barium enema by itself did not increase the diagnostic yield (²⁹). Thus, there is insufficient evidence to support the routine use of barium enema in the diagnostic evaluation of a patient presenting with functional constipation.

An assessment of colonic transit time can provide a better understanding of the rate of stool movement through the colon. This is also more reliable because a patient's recall of stool habit is often inaccurate (¹⁹). Typically, this would consist of administering a single capsule containing radio-opaque markers and obtaining a plain x-ray of the abdomen. However, several techniques have been described for performing this test (²¹). This includes the single capsule technique (⁶⁴) and the multiple capsules or multiple x-ray technique (^30,1,35,39). The validity of multiple capsule technique has been questioned (⁶⁵). However, for routine clinical purposes a single capsule technique is sufficient. Two-thirds of patients with dyssynergic defecation may also exhibit slow transit (^17,66). In some patients with constipation, the colonic transit time may be normal. In these subjects, it is important to exclude pelvic floor dysfunction as a mechanism for their symptoms (^17,66) and to look for psychosocial dysfunction (⁴³).

Only 10 studies were identified that assessed colonic transit in constipation. However, an abnormal study (i.e., evidence to support the presence of slow transit constipation) varied from 38 to 80%. There were significant differences in the study methodology and in the interpretation of test results, which may in part account for the discrepancy in the published results. It was reassuring to note that the studies are reproducible at least in patients with normal transit constipation (⁴²). These observations underscore the need for a well-designed, prospective controlled trial to define the exact role of the colonic transit test in the assessment of patients with constipation. Also, there is a need to standardize the test for clinical purposes. A colonic transit study provides an objective and confirmatory evidence for the presence of normal or slow transit through the colon (⁵) but by itself does not differentiate between the subtypes of constipation.

Anorectal manometry provides a comprehensive assessment of the pressure activity in the rectum and anal sphincter region together with an assessment of rectal sensation, rectoanal reflexes, and rectal compliance (^5,11,46,67). It is important to recognize that there are significant intercenter differences with regards to the methodology, test performance, and test interpretation (^8,67). Most recently, an international consensus panel under the auspices of the American and European motility societies has proposed uniform standards for performing and interpreting anorectal manometry (⁶⁷).

Manometry helps to detect abnormalities such as Hirschsprung's disease and dyssynergic defecation. The absence of rectoanal inhibitory reflex is pathognomonic for Hirschsprung's disease (^62,63). When a subject attempts to defecate, normally, there is a rise in intrarectal pressure, which is synchronized with a fall in anal sphincter pressure due to relaxation of the puborectalis and external anal sphincter (^8,11). This maneuver is under voluntary control and is primarily a learned response. The inability to perform this coordinated maneuver represents the chief pathophysiologic abnormality in patients with dyssynergic defecation (^{8,11,17,43,44}). This inability may be due to impaired expulsion forces, paradoxical anal contraction or impaired anal relaxation, or a combination of these mechanism(s) (^3,5,8,52). During attempted defecation, some subjects may not produce a normal relaxation largely because of the laboratory conditions (^17,68,69). Hence, the occurrence of this pattern alone should not be considered as diagnostic of dyssynergic defecation (see diagnostic criteria). In addition, rectal sensory testing has revealed that the threshold for first sensation or a desire to defecate may be impaired in 60% of patients with dyssynergic defecation (¹⁷).

Although several studies of anorectal manometry have been published, only nine studies were of medium-high quality. Analysis of these studies revealed significant differences in test methodology, both with regards to test performance and interpretation of data (Tables 6 and 7). Also, the concept of dyssynergic defecation as a cause of constipation has only been recognized over the last decade, and there is no generally acceptable method of identifying this condition. The prevalence of dyssynergia varied from 20 to 75% (Tables 6 and 7). In one study, the prevalence of dyssynergia was 63% with manometry, 38% with electromyography, and 36% with defecography (⁴⁴). Thus, the sensitivity for detecting this abnormality may vary depending on the type of test. In a few studies, manometric tests yielded new information, such as the presence of dyssynergia or impaired rectal sensation (⁴⁶), that was not detected clinically and this formed the basis for a change in the patients' management (^39,46,47). Because manometry only detects the presence of a dyssynergic pattern of defecation, and the diagnosis of dyssynergia requires more than the mere presence of this pattern (^3,8), one should interpret the results of manometry with caution. Furthermore, there is no gold standard and the test itself is regarded as the gold standard. Thus, anorectal manometry provides confirmatory evidence for the diagnosis of dyssynergic defecation in a patient with functional constipation and may pave the way for biofeedback therapy. However, further systematic studies in a large group of constipated patients and controls are required using standardized methodology to validate the use of this test in the diagnosis of constipation.

A balloon expulsion test provides an assessment of the patient's ability to defecate within the laboratory environment. Nine studies were identified that met inclusion criteria and were of medium quality. However, the methodology for balloon expulsion was not uniform. Consequently, the interpretation of these results is problematic. Also, the normal range of balloon expulsion time has not been defined. The prevalence of a positive test in favor of constipation varied between 23 and 67%. Only one study used a control group. Whether the test influenced the management of these patients or resulted in a change in outcome is not known. A recent study suggested a specificity of 89%, negative predictive value of 97%, sensitivity of 88%, and positive predictive value of 67% for this test (⁴⁸). This suggests that a normal test would at least exclude dyssynergia. However, this observation is confounded by other studies that found many patients with dyssynergia could expel the balloon (⁸) and that this test alone is insufficient to make a diagnosis of dyssynergia. The balloon expulsion test provides a simple, bedside assessment of the ability to expel a simulated stool in a patient with functional constipation. Although, the failure to expel a balloon suggests the possibility of dyssynergia, a normal test does not exclude this possibility. Hence, the results of this test should be interpreted alongside the results of other tests of anorectal function.

Defecography provides useful information regarding the anatomical and functional changes of the anorectum. In patients with dyssynergic defecation, the test may reveal poor activation of the levator muscles, prolonged retention of contrast material or inability to expel the barium, or other abnormalities such as a rectocele, mucosal intussusception, rectal prolapse, or excessive perineal descent. Although defecography may reveal several abnormalities in patients with constipation, there is a significant overlap of findings between patients and healthy controls and poor correlation of symptoms with defecographic findings (^5,52). Interestingly, in one study, the presence of rectocele or mucosal intussusception was unrelated to voiding of barium paste (⁵²). However, the authors concluded that defecography had a positive predictive value of 90% for detecting dyssynergia (⁵²). In contrast, another well-designed study found that the test had poor ability to discriminate between the subtypes of constipation (⁵⁶). Another study, which examined colonic transit and defecography in the same group of patients, concluded that a delayed rectosigmoid transit may help to identify patients who may benefit with a defecography (⁴¹). A most recent study found that defecography did not confer any additional diagnostic benefit over and above what was obtained from anorectal manometry, colonic transit study, and balloon expulsion test (⁸). This finding reaffirms the recommendations of the AGA technical task force on tests of colorectal function (⁵). Because of these inherent deficiencies, defecography should be regarded as an adjunct to clinical and manometric assessment of anorectal function and should not be relied upon as a sole test for assessing defecatory dysfunction (⁵).

CONCLUSIONS

A systematic review of studies assessing the utility of diagnostic tests in patients with constipation revealed no methodologically sound studies. Our review has identified several pitfalls. Firstly, no single test appears to provide a pathophysiological basis for constipation. Often, several tests are required to identify the underlying mechanism(s). Also, the methodology used for each test of colorectal function was not uniform and there were marked differences between the published studies. Consequently, the interpretation of data has been quite variable and abstracting this information does not provide clear guidelines. Secondly, the inclusion criteria for patients with constipation was either not defined or when available there was significant inter-study differences. Thirdly, a reference or gold standard test has not been developed and is unlikely to be developed, because constipation is a heterogeneous condition with multiple pathophysiological mechanisms. The pretest probability of identifying an organic gastrointestinal disease is largely based on performing colonoscopy/barium enema and blood tests but the available evidence does not support the routine use of these tests for constipation symptoms in the absence of alarm symptoms.

An optimal study design to assess the accuracy of a diagnostic test requires that a study be performed prospectively, with a blinded comparison of the test, and a reference gold standard in 100% of a consecutive series of patients from the relevant clinical population (²⁴). Most of the studies identified in this systematic review were less than optimal. The studies did not assess the full spectrum of patients that are likely to be encountered in the real world, although it may be argued that physiologic testing is generally restricted to patients with persistent symptoms refractory to diet and simple laxatives. In such a group, a significant proportion of patients had an abnormal test. The lack of gold standard may reflect the heterogeneity of this condition. However, only through large, well-designed prospective studies with adequate follow-up, and by using more uniform methodology (⁶⁷), would it be possible to critically examine the diagnostic yield and accuracy of the tests of colorectal function. There is little or no evidence to support the use of hematological and biochemical tests, x-rays, or endoscopy in the routine management of constipated patients without alarm features. There is good evidence to support the use of physiological tests, such as colonic transit and anorectal manometry, to define the pathophysiologic subtypes and aid treatment.

References

1. Locke, GR, III, Pemberton, JH, Phillips, SF. AGA technical review on constipation. Gastroenterology 2000;119: 1161–1178.
2. Herz, MJ, Kahan, E, Zalevski, S, et al. Constipation: A different entity for patients and doctors. Fam Pract 1996;13: 156–159.
3. Whitehead, WE, Wald, A, Diamant, N, et al. Functional disorders of the anorectum. Gut 1999;45: 55–59.
4. Glia, A, Lindberg, F, Nilsson, LH, et al. Clinical value of symptom assessment in patients with constipation. Dis Colon Rectum 1999;42: 1401–1408.
5. Diamant, ND, Kamm, MA, Wald, A, et al. AGA technical review on anorectal testing techniques. Gastroenterology 1999;116: 735–760.
6. Grotz, RL, Pemberton, JH, Talley, NJ, et al. Discriminant value of psychological distress, symptom profiles, and segmental colonic dysfunction in outpatients with severe idiopathic constipation. Gut 1994;35: 798–802.
7. Rao, S, Tuteja, AK, Vellema, T, et al. Dyssynergic defecation: Demographics, symptoms, stool Patterns, and quality of life. J Clin Gastroenterol 2004;38: 1–6.
8. Rao, SSC, Mudipalli, RS, Stessman, M, et al. Investigation of the utility of colorectal function tests and Rome II criteria in dyssynergic defecation (Anismus). Neurogastroenterol Motil 2004;16: 1–8.
9. Thompson, WG, Longstreth, GF, Drossman, DA, et al. Functional bowel disorders and functional abdominal pain. Gut 1999;45: 1143–1147.
10. Mertz, H, Naliboff, B, Mayer, E. Physiology of refractory chronic constipation. Am J Gastroenterol 1999;94: 609–615.
11. Rao, SSC. Dyssynergic defecation. Disorders of the anorectum. Gastroenterol Clin N Am 2001;31: 97–114.
12. Preston, J, Lennard-Jones, JE. Severe chronic constipation of young women: Idiopathic slow transit constipation? Gut 1986;27: 41–48.
13. Rao, SSC, Chamberlain, M, Hatfield, R, et al. Ambulatory 24 hour colonic motility in health and in patients with constipation. Am J Gastroenterol 2004;99: 2405–2416.
14. Surrenti, E, Rath, DM, Pemberton, JH, et al. Audit of constipation in a tertiary referral gastroenterology practice. Am J Gastroenterol 1995;90: 1471–1475. | PubMed | ChemPort |
15. Knowles, CH, Scott, SM, Wellmer, A, et al. Sensory and autonomic neuropathy in patients with idiopathic slow-transit constipation. Br J Surg 1999;86: 54–60.
16. Preston, DM, Lennard-Jones, J. Anismus in chronic constipation. Dig Dis Sci 1985;30: 413–418. | Article | PubMed | ChemPort |
17. Rao, SSC, Welcher, K, Leistikow, J. Obstructive defecation: A failure of rectoanal coordination. Am J Gastroenterol 1998;93: 1042–1050. | Article | PubMed | ChemPort |
18. Koch, A, Voderholzer, WA, Klauser, AG, et al. Symptoms in chronic constipation. Dis Colon Rectum 1998;40: 902–906.
19. Ashraf, W, Park, F, Lof, J, et al. An examination of the reliability of reported stool frequency in the diagnosis of idiopathic constipation. Am J Gastroenterol 1996;91: 26–32. | PubMed | ChemPort |
20. Wald, A, Hinds, JP, Caruana, BJ. Psychological and physiological characteristics of patients with severe idiopathic constipation. Gastroenterology 1989;97: 932–937. | PubMed | ChemPort |
21. Rao, SSC. Manometric evaluation of constipation, part I. Gastroenterologist 1998;4: 145–154.
22. Lembo, A, Camilleri, M. Chronic constipation. N Engl J Med 2003;349: 1360–1368. | Article | PubMed | ChemPort |
23. Cash, BD, Schoenfeld, Chey, WD. The utility of diagnostic tests in irritable bowel syndrome patients: A systematic review. Am J Gastroenterol 2002;97: 2812–2819. | Article | PubMed |
24. Lijmer, JG, Mol, BW, Heisterkamp, S, et al. Empirical evidence of design-related bias in studies of diagnostic tests. JAMA 1999;282: 1061–1066. | Article | PubMed | ISI | ChemPort |
25. Pepin, C, Ladabaum, U. The yield of lower endoscopy in patients with constipation: Survey of a university hospital, a public county hospital, and a veterans administration medical center. Gastrointest Endosc 2002;56: 325–332.
26. Rex, DK, Lehman, Ga, Ulbright, TM, et al. Colonic neoplasia in asymptomatic persons with negative fecal blood tests: Influence of age, gender, and family history. Am J Gastroenterol 1993;88: 801–802.
27. Lieberman, DA, Prindiville, S, Weiss, DG, et al. Risk factors for advanced colonic neoplasia and hyperplastic polyps in asymptomatic individuals. JAMA 2003;290: 2959–2967.
28. Patriquin, H, Martelli, H, Devroede, G. Barium enema in chronic constipation: Is it meaningful? Gastroenterology 1978;75: 619–622.
29. Gerson, DE, Lewicki, AM, McNeil, BJ, et al. The Barium enema: Evidence for proper utilization. Radiology 1979;130: 297–301.
30. Metcalf, AM, Phillips, SF, Zinsmeister, AR, et al. Simplified assessment of segmental colonic transit. Gastroenterology 1987;92: 40–47. | PubMed | ChemPort |
31. Martelli, H, Devroede, G, Arhan, P, et al. Mechanisms of idiopathic constipation: Outlet obstruction. Gastroenterology 1978;75: 623–631.
32. Mclean, MB, King, DW, Talley, NA, et al. The utilization of colon transit scintigraphy in the diagnostic algorithm for patients with chronic constipation. Dig Dis Sci 1999;44: 41–47.
33. Camilleri, M, Zinsmeister, AR. Towards a relatively inexpensive, noninvasive, accurate test for colonic motility disorders. Gastroenterology 1992;103: 36–42.
34. Bonapace, ES, Maurer, AH, Davidoff, S, et al. Whole gut transit scintigraphy in the clinical evaluation of patients with upper and lower gastrointestinal symptoms. Am J Gastroenterol 2000;95: 2838–2847.
35. Wald, A. Colonic transit and anorectal manometry in chronic idiopathic constipation. Arch Intern Med 1986;146: 1713–1716.
36. Ducrotte, P, Rodomanska, B, Weber, J, et al. Colonic transit time of radioopaque markers and rectoanal manometry in patients complaining of constipation. Dis Colon Rect 1986;29: 630–634.
37. Chaussade, S, Khyari, A, Roche, H, et al. Determination of total and segmental colonic transit time in constipated patients results in 91 patients with a new simplified method. Dig Dis Sci 1989;34: 1168–1172.
38. Karlbom, U, Pahlman, L, Nilsson, S, et al. Relationships between defecographic findings, rectal emptying, and colonic transit time in constipated patients. Gut 1995;36: 907–912.
39. Halverson, AL, Orkin, BA. Which physiologic tests are useful in patients with constipation? Dis Colon Rectum 1998;41: 735–739.
40. Glia, A, Lindberg, G, Nilsson, H, et al. Constipation assessed on the basis of colorectal physiology. Gastroenterology 1998;33: 1273–1279.
41. Prokesch, RW, Breitenseher, MJ, Kettenbach, J, et al. Assessment of chronic constipation: Colon transit time versus defecography. Eur J Radiol 1999;32: 197–203.
42. Nam, YS, Pikarsky, AJ, Wexner, SD, et al. Reproducibility of colonic transit study in patients with chronic constipation. Dis Colon Rectum 2001;44: 86–92. | Article | PubMed | ChemPort |
43. Wald, A, Caruana, BJ, Freimanis, MG, et al. Contributions of evacuation proctography and anorectal manometry to evaluation of adults with constipation and defecatory difficulty. Dig Dis Sci 1990;35: 481–487.
44. Ger, GC, Wexner, SD, Jorge, MN, et al. Anorectal manometry in the diagnosis of paradoxical puborectalis syndrome. Dis Colon Rectum 1993;36: 816–825.
45. Wexner, SD, Jorge, MN. Colorectal physiological tests: Use of abuse of technology? Eur J Surg 1994;160: 167–174.
46. Rao, SSC, Patel, RS. How useful are manometric tests of anorectal function in the management of defecation disorders. Am J Gastroenterol 1997;92: 469–75.
47. Vaizey, CJ, Kamm, MA. Prospective assessment of the clinical value of anorectal investigations. Digestion 2000;61: 207–214.
48. Minguiz, M, Herreros, B, Sanchiz, V, et al. Predictive value of the balloon expulsion test for excluding the diagnosis of pelvic floor dyssynergia in constipation. Gastroenterology 2004;126: 57–62. | PubMed |
49. Barnes, PR, Lennard-Jones, JE. Balloon expulsion from the rectum in constipation of different types. Gut 1985;26: 1049–1052.
50. Bannister, JJ, Timms, JM, Barfield, LJ, et al. Physiological studies in young women with chronic constipation. Int J Colorect Dis 1986;1: 175–182.
51. Fleshman, JW, Dreznik, Z, Cohen, E, et al. Balloon expulsion test facilitates diagnosis of pelvic floor outlet obstruction due to nonrelaxing puborectalis muscle. Dis Colon Rectum 1992;35: 1019–1025.
52. Halligan, S, Thomas, J, Bartram, C. Intrarectal pressures and balloon expulsion related to evacuation proctography. Gut 1995;37: 100–104.
53. Schouten, WR, Briel, JW, Auwerda, JJ, et al. Anismus: Fact or fiction? Dis Colon Rectum 1997;40: 1033–1041.
54. Pelsang, RE, Rao, SSC, Welcher, K. FECOM: A new artifical stool for assessing defecation. Am J Gastroenterol 1999;94: 183–186.
55. Pezim, ME, Pemberton, JH, Levin, KE, et al. Parameters of anorectal and colonic motility in health and in severe constipation. Dis Colon Rectum 1993;36: 484–491.
56. Turnbull, GK, Bartram, CI, Lennard-Jones, JE, et al. Radiologic studies of rectal evacuation in adults with idiopathic constipation. Dis Colon Rectum 1988;31: 190–197.
57. Karlbom, U, Eeg-Olofsson, EK, Graf, W, et al. Paradoxical puborectalis contraction is associated with impaired rectal evacuation. Int J Colorect Dis 1998;13: 141–147.
58. Jones, HJS, Swift, RI, Blake, H. A prospective audit of the usefulness of evacuating proctography. Ann R Coll Surg Engl 1998;80: 40–45.
59. Karlbom, U, Nilsson, S, Pahlman, L, et al. Defecographic study of rectal evacuation in constipated patients control subjects. Radiology 1999;210: 103–108.
60. Rex, DK, Bond, JH, Winawer, S, et al. Quality in the technical performance of colonoscopy and the continuous quality improvement process for colonoscopy: Recommendations of the U.S. multi-society task force on colorectal cancer. Am J Gastroenterol 2002;97(6):1296–1308. | Article | PubMed | ISI |
61. Barloon, TJ, Lu, CC. Diagnostic imaging in the evaluation of constipation in adults. Am Fam Phys 1997;56: 513–520.
62. Anuras, S, Hade, JE, Soffer, E, et al. Natural history of adult Hirschsprung's disease. J Clin Gastroenterol 1984;6: 205–210.
63. Reid, JR, Buonomo, C, Moreira, C, et al. The barium enema in constipation: Comparison with rectal manometry and biopsy to exclude Hirschsprung's disease after the neonatal period. Pediatr Radiol 2000;30: 681–684.
64. Evans, RC, Kamm, MA, Hinton, JM, et al. The normal range and a simple diagram for recording whole gut transit time. Int J Colorectal Dis 1992;7: 15–17.
65. Bouchoucha, M, Thomas, RS. Error analysis of classic colonic transit time estimates. Am J Physiol Gastrointest Liver Physiol 2000;279: G520–G527.
66. Kacasick, S, Ehrlich, SM: Is constipation a disorder of defecation or impaired motility? Am J Roentgenol 1996;166: 63–66.
67. Rao, SSC, Azpiroz, F, Diamant, N, et al. Minimum standards of anorectal manometry. Neurogastroenterol Mot 2002;14: 553–559.
68. Duthie, GS, Barolo, DCC. Anismus: The cause of constipation? Results of investigation and treatment. World J Surg 1992;16: 831–835.
69. Rao, SSC, Hatfield, R, Soffer, E, et al. Manometric tests of anorectal function in healthy adults. Am J Gastroenterol 1999;94: 773–783.