Abdominal bloating and pain are among the most common symptoms in gastro-intestinal disease and are also experienced in normal populations (1). Defaecation disorders and anal complaints, mostly in form of haemorrhoids, are also occurring with a high frequency. However, it is yet not clear if there is a coherence or a causality between a broader spectrum of abdominal symptoms and physical signs and those from the rectum and anal area (2). The present study was therefore undertaken to test a hypothesis of coexistence between abdominal and recto-anal symptoms and physical signs.

MATERIAL AND METHODS

Six hundred and forty-five patients were referred from general practitioners to an outpatient clinic and ambulatory service from 12th September 1988 to 19th January 1999. They had been referred for various abdominal symptoms, defaecation disorders including bleeding and such signs as abdominal distension or haemorrhoids. A random sample of 251 patients was drawn from the above material for a descriptive analysis, representing a 10-year-period. From the records, a standardized questionnaire was completed for each patient, covering 19 selected symptoms ( Figure 1 ), and any previous abdominal and anal operations. Each patient also underwent a standardized physical examination, including digital exploration of the ano-rectum followed by an initial ano-rectoscopy without cleansing to observe eventual rectal constipation. Otherwise, further investigations were dictated by the patients' symptoms and signs, for example colon barium enema, abdominal ultrasound and biochemical tests. Patient treatment was by a fluid and dietary regimen, rich in fibre and low in fat as advocated by the Danish Nutrition Council. The diet was supplemented by Ispagula HUSK (Ratje Frøskaller, Kastrup, Denmark) 5-10 g/day and by cisapride, 5 mg twice daily to 20 mg, three times daily. Finally, symptoms and signs were reassessed two months after start of the treatment. Surgery was done when necessary.

Statistics: The database was prepared in cooperation with UNI C, the Danish Electronic Data Processing Center for Research and Education (Copenhagen) which subsequently carried out the statistical analyses. The present study was observational and the analyses, therefore, will be more of an explorative nature than testing specific hypotheses, although this also will occur. The frequencies of symptoms, physical signs and investigations were determined. Principal component analysis or factor analysis were chosen as the main statistical explorative tool. Although factor analysis is not primarily designed for binary variables, it could be used in this context (3). The method analyses the pattern of correlation coefficients between the variables and combine groups of highly correlated variables into a smaller number of independent (uncorrelated) new variables explaining a major part of the variation (4). In the present study, the factor analysis could in its best indicate if certain symptoms and signs (abdominal and recto-anal) are correlated in a way that shows that they belong to the same underlying disease-dimension. Thus, the correlation of a variable (symptom, sign or investigation) with the individual factor was expressed as a factor loading or correlation coefficient, which could be between -1 and +1; the greater numeric correlation, the greater the association to the factor. However, factor loadings of numeric <0.3 are not considered of any practical importance. Also, important findings of symptoms and signs were cross-tabulated. A p-value of 0.05 or less was considered significant.

RESULTS

Of the 251 patients included in the study, 159 were female (63%) with a mean age of 49.9 years (range17-85) and 92 were male patients, mean age 51.7 years (range 24-89). In Figure 1 the symptoms are listed with declining frequencies. Most patients reported solid stool with defaecation at least once daily and with relative ease. The most common abdominal symptoms were bloating, abdominal pressure, epigastric discomfort, and episodes of colic pain.

Proctalgia were experienced by 28% and 38% reported occasional bleeding. Halitosis was experienced by 14% of the patients; 9% had episodes of feeling feverish, together with sudden nausea, and dizziness, requiring the patient to lay down for a while until spontaneous relief. Appendectomy had been performed in 16% and cholecystectomy in 7% of the patients.

The physical signs were dominated by right iliac fossa tenderness (58%) a palpaple faecal mass (i.e. additional faecal reservoir) (42%), meteorism (33%) and epigastric tenderness (15%). Tenderness in the left fossa was proven in 10% and a mass in 4% of the patients. Ano-rectoscopy, done at the patient's first visit, showed a reservoir of solid faeces in the ano-rectum in 62%, and haemorrhoids grade 2 or more in 51%. This examination revealed polyps in 20 patients and a tumour in three patients. Of the 251 patients, 138 underwent a barium enema which revealed left-sided diverticulosis in 42%, right-sided diverticulosis in 16%, while a tumour was identified in one patient. Colon elongatum was demonstrated in 16 patients.

The data was studied further by cross-tabulation of the variables. It is shown in Table 1 that the occurrence of main abdominal symptoms and signs were not related to age. In contrast, patients with haemorrhoids and diverticulosis were significantly older than patients without these diseases. From Table 2 it is seen that female patients significantly more often than men experienced the dominant symptoms and signs; there was no sex-difference in occurrence of haemorrhoids or diverticulosis. Since a palpable abdominal mass and faeces in the rectum represent faecal reservoirs (fermenting gas production), these variables were cross-tabulated with bloating to test for a coexistence ( Table 3 ). Although faeces in the ano-rectum occurred together with bloating in 62% of patients and 50% had bloating, faeces and a palpable abdominal mass, no statistical significant differences were found.

The structure of the data was explored further by factor analysis of the main symptoms, physical signs and paraclinical investigations - a total of 45 variables. The factor analysis was able to reduce these variables to 17 factors with an eigenvalue greater than one explaining 68% of the variance in all 45 variables. The nine factors having the largest eigenvalues are shown in Table 4 , which is a rotated component loading matrix. Each loading expresses the correlation between the variable and that particular factor. Bloating, abdominal pressure and colics, epigastric discomfort and tenderness all showed substantial correlations to factor A. Variation in stool consistency and repetetiveness correllated to factor B, as did repetitiveness, incomplete feeling and difficult emptying and proctalgia to factor C. A proven right-sided abdominal mass correlated with tenderness, meteorism, fever-episodes and bad breath, and with incomplete defaecation and difficulty (D). In another factor (E), epigastric and right fossa tenderness correlated with obesity, the latter also been correlated with bad breath, weight gain, polyps and haemorrhoids (G). Abdominal pressure and colics were loaded significantly in another factor (F) with meteorism and fever episodes, negatively with faeces in the rectum. Difficult and infrequent defaecation was associated with fever episodes (I), and incomplete defaecation with proctalgia and bleeding, negatively with the ease of defaecation (J). In another factor (not shown) familiary colon cancer and meteorism were loaded significantly.

Finally the outcome of the treatment was tested on the dominating symptoms (from Figure 1, Table 5 ). Thus, a symptom could have disappeared or still be present or could still be absent or have arrived. The number of patients eligible for analysis were reduced, because some patients did not show up for the control or had moved away. Bloating, abdominal pressure and colics and epigastric discomfort were all reduced significantly. The process of defaecation was overall improved by significant reductions in incomplete- and repetitiveness, and now with ease and no pain and less liquid faeces. The results of treatment on physical signs and investigations were also tested. The presence of an abdominal mass in the right fossa fell from 42% to 17% (p<0.05), the associated tenderness from 58% to 27% (p<0.05) and meteorism from 33% to 18% (p<0.05). Ano-rectal constipation was reduced significantly from 62% to 19% (p<0.05). Cisapride was administered to 159 patients (64%), minor headache or dizziness developing in 7%.

Surgery was performed as follows: 76 patients with haemorrhoids grade 2 or more underwent banding or a Milligan-Morgan procedure, further four patients had an anal-fissure operated, and 20 patients had ano-rectal polyps removed; four patients underwent operation for a carcinoma.

DISCUSSION

In general, the connection between symptoms, clinical signs and disease has been established empirically. In contrast, factor analysis was used in this study and identified significant correlations between different symptoms and signs. Thus, a dominant factor consisted of abdominal mass, tenderness, meteorism, fever episodes and halitosis, two other factors of defaecation disorders. Not only a right-sided mass, pressure and tenderness but also epigastric discomfort and tenderness were included in a factor. All the more, a high proportion of patients was identified with additional faecal accumulation in the right colon and rectum, so-called additional reservoirs, irrespective of daily emptying. The patients did not fulfil criteria of constipation with at least two defaecations per week (5), but have nevertheless great accordance in other symptoms. A meaningful interpretation would be that of functional retention of faeces in various colo-rectal segments, called hidden constipation. Female patients experienced significantly more often than males, the dominant symptoms and signs which are usual in functional bowel disease (6). These symptoms were not related to age, in contrast to organic changes like diverticula and haemorrhoids which occurred significantly more often with increasing age. Demonstration of colonic pathology relied on barium enema, in contrast to the present domination of flexible endoscopy. A high proportion of patients at a higher age had grade 2 haemorrhoids and a faecesfull rectum. This favours (rectal) constipation as being involved in displacement of the anal canal i.e. haemorrhoidal disease, in accordance with an earlier study (7). In all, approximately one half of the patients had a fully developed hidden constipation syndrome.

Bowel symptoms are common in industrial populations with the irritable bowel syndrome having prevalence rates from 6.6% to 25.0% (6). However, our patients did not have abdominal pain re-lieved by defaecation, and although they experienced bloating and some reported repetitiveness and incompleteness, they did not fulfil established criteria for having IBS (6).

In the present study a rather high proportion of patients had underwent appendectomy and cholecystectomy. Recent evidence seems to show that formation of gallstones are related to slow transit or an "indolent intestine" (8, 9). Analyses in this study seem to show difficulties in transportation of faeces, resulting in formation of additional faecal reservoirs in the right-sided colon and rectum. At this stage, the condition is solely functional. However, if it remains for years, an overloaded large bowel seems to lead into organic colorectal disease. Because of a daily emptying, neither the patient nor the physician at first suspect the symptoms of being originated from the large bowel. However, some events would often have pointed towards an overloaded colon like defaecation urge during a meal, and soiling, or a more putrid smell from faeces and flatus. Also more frequent defaecations during workfree weekends seem to occur, probably representing a natural release of the overload. It follows from this that a situation of bowel overload could start insidiously before detected after years of aggravation or finding of a definite pathological lesion. It is also clear that the frequency of defaecation does not reflect the overload or hidden constipation.A constipated colon bears in itself several potential risks. Intraluminal pressure-rising could lead to increasing numbers of diverticula and polyps and malignancies are more frequent (10, 11, 12), as in this study with four malignancies detected out of 251 patients. It should also be noted that the proven additional faecal reservoirs of the right colon and rectum correspond well with the more frequent occurrence of malignancies in these parts of the bowel. On this line it was recently observed that patients with the irritable bowel syndrome were more likely to have a subsequent diagnosis of a colorectal tumour (13). Thus, hidden constipation could be "The common cause" of tumours (benign, malign), diverticular disease and appendicitis, which Burkitt searched for (14). The condition is suspected to be widespread in western populations with many people not seeking medical advice, and thereby having an increased risk of these diseases over the years.

The present study was interventional and not a therapeutic trial of different regimens. Thus, an established bowel stimulatory treatment was initiated with a diet rich in fibre and low in fat, an increased fluid intake and supplementary fleawort-seed, since high intake of dietary fibre is associated with more rapid transit time (15, 16) and also is associated with a lower frequency of bowel cancer. Also, the patients were encouraged to exercise (17). Cisapride was used in varying doses, since it increases bowel transit in normal volunteers, in chronic constipation (18, 19) and in constipation dominant irritable bowel patients (20). Thus bloating, abdominal pressure and colics were reduced significantly, as were defaecation disorders, constipation-fever episodes and proctalgia. The physical signs were reduced to a lesser degree, although still significant. This means that the effect of the propulsive regimen verified in itself the preexisting faecal overload (retention) in reservoirs in the individual patient. The dietary regimen and increased fluid intake had to continue while adjuvant cisapride often could be reduced to a single daily dose of 10-20 mg, or ceased. Recently, cisapride has been withdrawn from the market because of the risk of fatal cardiac arrhytmia.

In conclusion, this study has shown correlation between abdominal and recto-anal symptoms and physical signs. Especially, additional faecal reservoirs were demonstrated in the right colon and rectum i.e. a constipated colon in spite of daily emptying. This hidden constipation gives rise to bloating, pain, and tenderness, and a palpable mass in the right iliac fossa. Also, hidden constipation seems over the years to bear a cumulative risk of developing gallstones, diverticula, polyps, haemorrhoids and malignancy, thus being a link between functional and organic disease. A combined prokinetic regimen should therefore be initiated and continued.

Return

Return

Return

Return

Return

Return