Search

A retrospective cohort of 823 patients with ulcerative colitis who resided at the time of diagnosis in one of three defined geographical areas (West Midlands region, Oxford region, England and Stockholm County, Sweden) was assembled. The patients were first seen at named hospitals in these areas and the diagnosis of ulcerative colitis established within five years of onset of symptoms between 1945-1965. All patients were 15 years of age or more at onset of disease and were followed for a minimum of 17 years and a maximum of 38 years. Ninety seven per cent completeness of follow up was achieved. Examining the colorectal cancer risk in the series relative to the risk in the general population by standardised morbidity ratios, there was an eight fold increased risk of cancer in the series as a whole. Dividing the series by extent of colitis, extensive colitis patients showed a 19 fold increase in risk. A four fold increased risk was shown in the remainder of the series (left sided colitis, proctitis and extent unknown). Life table analyses in extensive colitis gave cumulative risks of 7.2% (CI 3.6-10.8) at 20 years from onset of disease and 16.5% (CI 9.0-24.0) at 30 years from onset. No significant effect of age at onset, sex or referral centre could be detected. Examination of the data by interval from onset to cancer and by actual age at development of cancer suggests that patients who develop colorectal cancer will do so in a distribution around 50 years of age independent of duration of disease in adult onset ulcerative colitis (greater than 15 years at onset of disease). An inverse relationship was shown between age at onset of disease and interval from onset of disease to cancer. Further age specific rates for cancer increased up to 50 years and decreased thereafter. These results suggest that extensive colitis patients have a genetic predisposition to colorectal cancer and that longstanding inflammation is not of primary importance in the initiation/promotion of cancer in this disease.

Background—The appropriate medical treatment of patients with ulcerative colitis is determined largely by the severity of symptoms. Hospital assessment of the severity of disease activity includes investigation of laboratory indices and sigmoidoscopic assessment of mucosal inflammation.Aims—To develop a simplified clinical colitis activity index to aid in the initial evaluation of exacerbations of colitis.Methods—The information for development of the simple index was initially evaluated in 63 assessments of disease activity in patients with ulcerative colitis where disease activity was evaluated using the Powell-Tuck Index (which includes symptoms, physical signs, and sigmoidoscopic appearance). The new index was then further evaluated in 113 assessments in a different group of patients, by comparison with a complex index utilising clinical and laboratory data, as well as five haematological and biochemical markers of disease severity.Results—The newly devised Simple Clinical Colitis Activity Index, consisting of scores for five clinical criteria, showed a highly significant correlation with the Powell-Tuck Index (r=0.959, pr=0.924, pConclusions—This new Simple Colitis Activity Index shows good correlation with existing more complex scoring systems and therefore could be useful in the initial assessment of patients with ulcerative colitis.

The surgical resection rates among parous women with distal ileal and colonic Crohn's disease have been compared with resection rates among distal ileal (n = 197) and colonic (n = 332) Crohn's disease patients. Thirty of 44 parous women with distal ileal Crohn's disease and 28 of 44 with colonic Crohn's disease had achieved their first pregnancy on average 8 years and 6.2 years respectively before the diagnosis of Crohn's disease was established. Resections for each patient were negatively correlated with parity in both groups. (Distal ileal disease (p = 0.034, rs = 0.3207), colonic disease (p = 0.051, rs = -0.2960)). Patients with distal ileal Crohn's disease and a history of pregnancy at diagnosis (n = 30, mean follow up = 15 years) had fewer resections/patient when compared with the published resection group: mean (SD); 1.17 (0.65) v 1.57 (1.05), p = 0.006. Patients with colonic Crohn's disease and a history of pregnancy at diagnosis (n = 28, mean follow up = 16.5 years) had fewer resections/patient when compared with the published resection group: mean (SD); 0.68 (0.77) v 1.05 (0.77), p = 0.019. In summary, patients with distal ileal and colonic Crohn's disease, who had been pregnant in the past subsequently need fewer surgical resections. Pregnancy could influence the natural history of Crohn's disease either by decreasing immune responsiveness or by retarding fibrous stricture formation, which is the commonest indication for surgical intervention.

The increased prevalence of gall stones in Crohn's disease is thought to be related to depletion of the bile salt pool due either to terminal ileal disease or after ileal resection. This study was designed to examine whether this hypothesis is correct and explore alternative explanations. Two hundred and fifty one randomly selected patients (156 females, 95 males, mean age 45 years) were interviewed and screened by ultrasonography to determine the prevalence of gall stones in a large population of patients with Crohn's disease. Sixty nine (28%) patients had gall stones proved by ultrasonography (n = 42), or had had cholecystectomy for gall stone disease (n = 27). The risk factors for the development of gall stones including sex, age, site, and duration of disease, and previous intestinal resection were examined by multivariate analysis. Age and duration of disease were positive risk factors for gall stones and were covariables. The site of disease and of previous intestinal resection did not predispose to gall stones. Previous surgery was an independent risk factor for the development of gall stones, the risk increasing with number of laparotomies. It is suggested that mechanisms other than ileal dysfunction may predispose to gall stones. Postoperative gall bladder hypomotility with biliary sludge formation may be precursors of gall stone formation in patients with Crohn's disease.

Neutrophil superoxide anion production was measured in healthy subjects and in patients with quiescent and active Crohn's disease using superoxide dismutase inhibitable cytochrome C reduction. Three stimuli were used: phorbol 12-myristate 13-acetate (PMA1), phorbol 20-oxo-20-deoxy 12-myristate 13-acetate (PMA2), and Candida albicans in serum. Normal neutrophils produced significantly more superoxide anion than Crohn's disease neutrophils with both PMA1 (mean (SD) 9.6 (2.2) v 8.6 (1.8) nmol/10(6) cells/5 minutes, p = 0.04) and PMA2 (1.8 (0.8) v 0.8 (0.77) nmol/10(6) cells/5 minutes, p = 0.00004). With C albicans in serum, normal and Crohn's disease neutrophils produced similar amounts of superoxide anion (4.4 (1.5) v 4.3 (1.7) nmol/10(6) cells/30 minutes, not significant). Results were independent of disease activity. Superoxide anion production by PMA-stimulated Crohn's disease neutrophils is significantly lower than by normal neutrophils.

A previously fit 23 year old adult male who presented with a sudden onset of profound autonomic neuropathy, for which no cause could be found, is described. The patient subsequently developed ischaemic enterocolitis that ultimately necessitated colectomy and subtotal enterectomy. Potential neural and humoral mechanisms are discussed.

A 31 year old man from Gambia, resident in the United Kingdom for two years, presented with a two month history of unproductive cough, malaise, weight loss, non-specific abdominal pain, and episodic diarrhoea. Acid alcohol fast bacilli were identified in his sputum, together with Strongyloides stercoralis larvae and Giardia lamblia cysts in his stools. This case illustrates that latent strongyloidiasis can become overt in the presence of tuberculosis, and the diagnosis of strongyloidiasis must be borne in mind in patients who have previously resided in endemic regions.

Three patients with Crohn’s disease are described who were treated by ileal resection for intestinal obstruction. Histological examination of the resected specimen in each case established the diagnosis of adenocarcinoma of the small intestine complicating Crohn’s disease. This diagnosis should be considered in patients with longstanding macroscopic Crohn’s disease who present with severe or recurrent symptoms. The diagnosis may not be apparent on routine radiological examination or even macroscopically at laparotomy.

The risk of developing colorectal cancer has been compared in two identically selected cohorts of patients with extensive Crohn's colitis (n = 125) and extensive ulcerative colitis (n = 486). In both groups the effects of selection bias have been reduced wherever possible. There was an 18-fold increase in the risk of developing colorectal cancer in extensive Crohn's colitis and a 19-fold increase in risk in extensive ulcerative colitis when compared with the general population, matched for age, sex, and years at risk. The absolute cumulative frequency of risk for developing colorectal cancer in extensive colitis was 8% at 22 years from onset of symptoms in the Crohn's disease group and 7% at 20 years from onset in the ulcerative colitis group. The relative risk of colorectal cancer was increased in both ulcerative colitis and Crohn's disease among those patients whose colitis started before the age of 25 years. Whether the absolute risk is greater in the younger age group or merely reflects that the expected number of carcinomas increases with age is uncertain. While there is an increased risk of developing colorectal cancer in extensive colitis the number of patients with Crohn's disease who actually develop colorectal cancer is small because many patients with extensive Crohn's colitis undergo colectomy early in the course of their disease to relieve persistent symptoms unresponsive to medical treatment.

The colorectal cancer risk in Crohn's disease eliminating all known biases was assessed in a cohort of 281 patients with Crohn's disease who resided in the West Midlands at the time of diagnosis, and were first seen within five years of onset of symptoms between 1945-1975. All patients were 15 years of age or more at onset and were followed up from 12-35 years (total 5213 person years at risk (PYR)). The colorectal cancer risk in the series compared with the risk in the general population was computed by applying sex and age specific PYRs to the date of death or end of the study period 31 December 1991. There were six colonic and two rectal cancers. Six of the eight colorectal cancers were diagnosed 20 or more years after the onset of Crohn's disease. The relative risk (RR) of colorectal cancer for the series as a whole was 3.4 (p < 0.001), with a fivefold excess in the colon, but no significant excess in the rectum. Patients with extensive colitis showed an 18-fold increase in risk (RR = 18.2, p < 0.001), which decreased with increasing age at onset. This study shows that there is a statistical excess risk of developing colorectal cancer in patients who develop their Crohn's disease at a young age of onset (less than 30 years of age).

Diffuse jejunoileitis is an uncommon, but important manifestation of Crohn's disease because of the associated high morbidity and challenges in medical management. Despite this there have been no studies of prognosis or management of diffuse jejunoileitis for nearly 20 years. This study analysed the outcome among 34 patients (20 women and 14 men) with diffuse jejunoileitis diagnosed between 1960 and 1991 including clinical features, medical and surgical management, death rates, current status, and prognosis. Diffuse jejunoileitis presents in younger patients (mean age at diagnosis 26.4 years) than those with distal ileal Crohn's disease (mean age at diagnosis 33.3 years). Nearly all presented with clear cut abdominal symptoms including a combination of colicky abdominal pain (91%), weight loss (62%), and diarrhoea (53%). Most patients had severe symptoms reflected by the fact that 77% had been treated with corticosteroids for periods of more than six months at some stage during their follow up. The mean follow up from diagnosis was 16 years. Twenty eight patients (82%) had at least one operation for diffuse jejunoileitis and two thirds of the patients (n = 21) required two or more operations. The frequency of surgical intervention was particularly high in the younger patients (r = 0.71, p < 0.001). The proportion of patients requiring surgery was highest in the first year after diagnosis. The annual operative rate was 15% for the first 10 years and then it fell to 5.2% in years 11-15, and 2.6% in years 16-20. The data suggest that the disease does burn itself out over time. The increasing use of strictureplasty for short strictures and the minimal use of resection has eliminated problems associated with the short small bowel syndrome. The longterm prognosis of this patients is good. Only two patients died (one of perforation of the jejunum and another of an unrelated bronchogenic carcinoma). After a mean interval from diagnosis of 16 year, 24 of 32 living patients are well and symptom free. Only eight have abdominal symptoms of whom three are receiving corticosteroid treatment and one azathioprine. The combination of anti-inflammatory drugs with the relief of recurrent obstructive symptoms by strictureplasty can together produce a good longterm prognosis in most patients with diffuse jejunoileitis.

The ability of normal and Crohn's disease neutrophils to kill Candida albicans has been studied using neutrophils isolated from peripheral blood and suspended in phosphate buffered saline at 5 x 10(6) cells per ml. C albicans was grown to a stationary phase in broth culture and suspended in phosphate buffered saline at 10(7) organisms/ml. Neutrophils and Candida were then incubated together at 37 degrees C in a shaking water bath in the presence of fresh serum. At 30 and 60 minutes samples were withdrawn, neutrophils lysed, and Candida survival assessed by colony counting. Results were compared with control suspensions of Candida incubated with serum alone. After 30 and 60 minutes in the presence of autologous serum normal neutrophils had killed significantly more Candida than Crohn's disease neutrophils (mean (SD) 61.0 (16.7)% v 40.5 (16.2)% at 30 minutes, p less than 0.0001; 83.2 (7)% v 70.8) 16)% at 60 minutes, p less than 0.005). The results did not alter significantly when normal neutrophils were incubated with Candida in the presence of Crohn's disease serum instead of normal serum. When Crohn's disease neutrophils were incubated with Candida in the presence of normal serum instead of autologous serum there was some improvement in candidacidal ability at 30 minutes (48.9 (20.6)% v 40.5 (16.2)%, p less than 0.03) but not at 60 minutes. Phagocytosis, measured using a radiometric assay, was normal. Neutrophils from patients with Crohn's disease have an impaired ability to kill this granuloma provoking organism. It is not due to serum inhibitors or defective phagocytosis.

Intestinal disaccharidase and dipeptidase activities were measured in mucosal biopsies from the proximal jejunum in 20 patients with Crohn's disease apparently confined to the distal ileum or large bowel, 14 patients with ulcerative colitis, and 14 healthy volunteers who acted as controls. The dissecting microscopy and histological appearance of the biopsies were normal (Gd 0-1) except for two which showed grade 2 changes. tbiopsy morphometry showed a reduction of jejunal mucosal surface area and an increase in mucosal volume in patients with Chron's disease when compared with the other two groups. The mucosal enzymes studies demonstrated that patients with Crohn's disease had a significant reduction in brush-border enzymes (disaccharidase) but no change in cytoplasmic enzyme activity (dipeptidases). The enzyme levels in patients with ulcerative colitis did not differ from the healthy controls. The reduction of brush-border enzymes with normal cytoplasmic enzymes in the presence of abnormal morphometry is further evidence of the concept of Crohn's disease as a diffuse lesion of the gastrointestinal tract. It also suggests that there is either specific damage to the microvilli or some other abnormality such as impairment of enzyme synthesis.

Fifty male patients with proctitis were examined and the clinical, microbiological, serological and proctological features compared with 51 known male homosexuals attending the genitourinary (GU) clinic at the same hospital. The homosexuals had a short history of bowel symptoms, minor sigmoidoscopic and histological changes on rectal biopsy and many positive serological markers of sexually transmitted infection. There was some evidence of sexually transmitted disease in the IBD patients and three were homosexuals. IgG antibodies were positive for chlamydia trachomatis (n = 10) and hepatitis A (n = 7). One had a positive screening test for syphilis. Stool examination and rectal swab cultures were positive in two patients for cryptosporidium and cytomegalovirus respectively. Gastroenterologists must be aware of the possibility of specific infection in IBD patients and a clinical history should include sexual preferences and practices. If homosexuality is admitted, specific infection must be sought and excluded.

One hundred and twenty seven patients undergoing elective surgery for inflammatory bowel disease were divided into three groups according to their preoperative ideal body weight (less than 80%, 80-90%, and greater than 90%). The groups were well matched in respect of age, sex, corticosteroid therapy, pre-existing sepsis, peroperative antimicrobial chemotherapy, and resection site. None received peroperative nutritional support. The postoperative outcome was similar in each of the three nutritional groups including the incidence of postoperative sepsis, duration of hospital stay, and mortality. Serial peroperative changes in weight, fat, and muscle bulk were assessed by anthropometric measurements in 21 of these patients. The deficits in weight, fat, and muscle bulk were similar at 10 and 21 days postoperatively in the three groups. At 84 days those malnourished preoperatively had recovered their nutritional status faster than the well nourished patients. We conclude that in these patients undergoing elective resection for inflammatory bowel disease preoperative weight loss did not adversely affect the postoperative outcome.

In patients with Crohn's disease involving the rectum (n=25), there was an inverse relationship between rectal capacity and the degree of proctitis. However, in patients with Crohn's disease not involving the rectum (n=22) the rectal capacity was similar to that of normal controls (n=20). The frequency of defaecation was not related to the degree of proctitis or to the pressure of a colectomy and ileorectal anastomosis. Control subjects had a significantly lower frequency of defaecation than patients with Crohn's disease irrespective of involvement of the rectum.

The association of coeliac disease and inflammatory bowel disease is rare, as only three individual cases have been reported. Four additional cases of the association are described. A review of the prevalence figures for the two disorders suggests that this is more than a chance association.

The long-term prognosis including operation rates, the incidences of recurrent disease, morbidity and mortality and current status has been analysed in a group of 360 patients with Crohn's colitis grouped according to the primary site of macroscopic disease at diagnosis. The group has been followed from diagnosis for a mean period of 14·9 years. The overall operation rate was 76 per cent. Prolonged spontaneous or drug induced remission occurred at all sites: right-sided disease (11 per cent), extensive colonic disease (21 per cent) and left-sided disease (38 per cent). The cumulative reoperation rates at 5 and 10 years after right hemicolectomy were 26 and 46 per cent, after colectomy and ileorectal anastomosis 46 and 60 per cent, and after panproctocolectomy 10 and 21 per cent, respectively. There was a twofold excess mortality rate from related Crohn's disease deaths during the period of review, but the mortality rate has fallen with time. There have only been 11 related deaths in the last decade, of which eight were probably unavoidable. The current status of most patients is good, although treatment has included a permanent stoma in less than half (41 per cent) the patients still under review. Currently all but 14 patients are well and symptom-free and only 16 are receiving specific medical treatment. Until the aetiology of Crohn's colitis is understood, if medical treatment has failed to resolve symptoms, appropriate surgical treatment in experienced hands is an effective way of restoring patients with chronic persistent symptoms to good health.

Finger clubbing, measured objectively by using the hyponychial angle, was present in 75 out of 200 (38%) patients with Crohn's disease, 15 out of 103 (15%) with ulcerative colitis, and two out of 24 (8%) with proctitis. In Crohn's disease and ulcerative colitis the hyponychial angle was significantly correlated with both disease activity and the extent of fibrosis in the resected specimens from 47 surgically treated patients. The prevalence of finger clubbing in patients with macroscopic disease within the area of the gut innervated by the vagus nerve was significantly higher than that in patients in whom the disease was confined to the distal colon and rectum. Finger clubbing in patients with Crohn's disease tended to regress after resection of macroscopic disease. It is concluded that finger clubbing is significantly commoner in Crohn's disease than ulcerative colitis. The focal stimuli for finger clubbing include mucosal inflammatory change and fibrosis mediated by the vagus and possibly other autonomic pathways acting as the afferent arc of a finger-clubbing reflex.

The amount of lithocholate input and the size of the total lithocholate exchangeable pool were measured using isotope dilution in three patients ingesting chenodeoxycholic acid for gallstone dissolution and two healthy control subjects. Because the specific activity decay curve was biexponential in appearance, input was calculated using a stochastic analysis employing the Stewart-Hamilton equation. By this method, the lithocholate input and the size of the exchangeable pool in gallstone patients during chenic acid ingestion was four to five times that observed in the healthy control subjects. It was calculated that about one-fifth of the newly formed lithocholate was absorbed in both gallstone patients and healthy control subjects. The extent of sulphation of absorbed lithocholate was then measured in seven gallstone patients whto had been ingesting chenodeoxycholic acid for gallstone dissolution by defining the chemical form of radioactivity in bile after intravenous administration of labelled lithocholate. Similar studies were carried out in eight healthy control subjects, but, in addition the extent of sulphation of injected lithocholylglycine and lithocholyltaurine was also defined. When lithocholate was injected, the majority of radioactivity in bile (50-60%) was present as sulphated conjugates, both in gallstone patients and healthy control subjects. The degree of sulphation was greater for glycine than taurine conjugates, and these findings, which suggested preferential sulphation of the glycine conjugates of lithocholate, were confirmed by showing that injected lithocholylglycine was sulphated to a much greater extent than injected lithocholyltaurine. These findings indicate that in patients receiving chenic acid there is effective sulphation of lithocholate, especially its glycine conjugates, despite a considerable increase in lithocholate absorption; they may provide an explanation for the lack of hepatotoxicity of chenic acid in man.

In a study of a group of 325 blood donors between the ages of 18 and 65 years, the total leucocyte and polymorph counts of the females aged 50 to 65 years were found to be significantly lower than those of the corresponding group of male donors. In the whole series of female donors, there were significantly lower total leucocyte and lymphocyte counts with increasing age and a less significant (P [unk] 0·1) fall in the polymorph count. No age variation in the male counts was found. Analysis of large numbers of leucocyte counts performed on hospital patients has confirmed the sex difference in the total leucocyte count found in the blood donors. The factors responsible for the differences are discussed.

A comparison is made of the immediate and long term mortality of colectomy and ileostomy between 73 patients who had Crohn's colitis and 442 who had ulcerative colitis. The immediate mortality in Crohn's disease is 4%. In ulcerative colitis it is 10%, chiefly because of the higher proportion of emergency operations. The late mortality in both groups is 10%, chiefly as a result of recurrence of Crohn's disease or the sequellae of colonic malignancy present at the time of colectomy for ulcerative colitis. A further comparison is made between the postoperative course of the 64 surviving patients with Crohn's disease and a comparable sample of 65 patients who had an ileostomy for ulcerative colitis in the same era. There was a similar incidence of postoperative septic complications in the two groups (35%). The readmission rate was twice as high in the Crohn's disease patients. Ileostomy reconstruction for mechanical complication was needed in 21 patients with Crohn's disease compared with 6 with ulcerative colitis. Further ileal resection was required for recurrent disease on another 25 occasions in the patients with Crohn's disease but never in those with ulcerative colitis. Long therm review graded the clinical status as excellent or good in 70% of those with Crohn's disease compared with 95% with ulcerative colitis.

Restorative proctocolectomy has become a realistic alternative to construction of a permanent stoma for patients with ulcerative colitis and familial polyposis coli. The operation restores intestinal continuity but continent patients have to open their bowels 4-5 times during the day, and over half have to rise at night to defaecate. The potential complications are considerable, but their frequency decreases with experience with the operation. This article reviews the place and results of ileoanal pouch anastomosis.

Eight adult female patients suffering from abdominal pain and chronic diarrhea which was often incapacitating and frequently nocturnal, had dramatic relief on a gluten-free diet and return of symptoms after gluten challenge. Previous nonspecific measures and a milk-free diet were ineffective. Multiple jejunal biopsies showed minor, but significant changes in cellularity which returned to normal on the gluten-free diet. Apart from a slight increase in jejunal cellularity, no immunological abnormalities were found after gluten challenge. Steatorrhea or other biochemical defects, common in celiac disease, were not found. It was concluded that these patients had a gluten-sensitive diarrhea, but had no evidence of celiac disease.

A total of 19 enterovesical fistulas were recorded in a series of 799 patients with Crohn's disease (2.4%). The origin of the fistulas was: ileum (9), colon (6) and four were complex involving the small and large bowel. Only 13 patients presented with urinary symptoms: pneumaturia (9), haematuria (1) and urinary tract infection (3). Four fistulas were identified incidentally during contrast radiology, one fistula was identified during a laparotomy and one further fistula developed after a previous resection for Crohn's disease. Four patients were managed conservatively and all are asymptomatic, but it is not known whether the fistula has healed. Twelve fistulas were resected: 9 healed, 2 recurred and 1 patient died following resection for a malignant fistula complicating Crohn's disease. Early in the series three patients were managed by bypass or defunction of the fistula. In all cases the sepsis persisted resulting in mortality. Persistent symptomatic fistulas should be treated by resection of the affected segment of bowel with primary anastomosis if appropriate. The defect in the bladder should be closed over an indwelling catheter which should not be removed until there is radiological confirmation that the bladder defect has healed satisfactorily.

Adenocarcinoma of the small intestine is an uncommon complication of Crohn's disease. We report the clinical and pathological details of three cases diagnosed between 1968 and 1980 with a review of 58 cases from the literature. Of the 61 cases, 41 tumours occurred in the ileum, 18 in the jejunum, one in the duodenum and ileum, and one in the ileum and colon. Eighteen occurred in bypassed intestinal loops. The prognosis was poor: 44 patients (72%) had died with a mean interval of only 7.9 months from the diagnosis of cancer.

Protein-energy malnutrition is common in patients with inflammatory bowel disease requiring surgery. Increased morbidity from sepsis may be associated with pre-operative weight loss due to associated diminished cellular immune defence mechanisms. Eighteen patients with Crohn's disease requiring elective abdominal surgery were divided into three nutritional groups before and after surgery, defined by their weight loss. The groups were well matched for other clinical variables. Delayed hypersensitivity skin reactivity to multiple antigens was measured before operation and during convalescence. The delayed hypersensitivity skin reactivity was depressed in the malnourished patients both pre-and post-operatively, although clinical outcome from surgery was not affected by nutritional status.

The spectrum and incidence of liver disease is described among a large series of patients with inflammatory bowel disease. The incidence of significant liver disease identified by the presence of serial biochemical abnormalities of liver function was 8.2 per cent. Transient peri-operative changes in liver function tests are common and usually relate to underlying intra-abdominal sepsis. Percholangitis, sometimes termed portal triaditis, is one of the commoner lesions, and is usually associated with extensive colitis and improves with resection of the underlying bowel disease. Cirrhosis of the liver is an important but uncommon complication and is usually associated with extensive long-standing disease. Stenosing cholangitis and biliary tract carcinoma are both important though rare associations. They are both associated with extensive disease of long-standing, but resection of the underlying inflammatory bowel disease does not necessarily protect the individual from these complications. Although stenosing cholangitis is a diffuse lesion of the biliary tree it is important to exclude strictures of the extra-hepatic biliary tree which may be amenable to surgical correction. Hepatic dysfunction is rarely the sole indication for advising surgery for the underlying bowel disease but the identification of the nature of the hepati- dysfunction provides a rational basis for such a decision and opportunities for the surgical correction of the hepatic lesion itself.

Studies were carried out to assess the stability of the label in [11,12-(3)H(2)]chenodeoxycholic acid and [11,12-(3)H(2)]lithocholic acid during enterohepatic cycling and thus define the suitability of these tracers for isotope dilution studies of bile acids in man. To do this, [11,12-(3)H(2)]-chenodeoxycholic acid and [24-(14)C]chenodeoxycholic acid were administered simultaneously to six healthy adults. The (3)H/(14)C ratio of biliary chenodeoxycholic acid and the specific activity of biliary chenodeoxycholic acid were determined to permit calculation of pool size and turnover rate by the conventional isotope dilution procedure of Lindstedt (1957. Acta Physiol. Scand. 40: 1). Excretion of label in urine was measured as well as the (3)H/(14)C ratio of fecal bile acids. Similar studies were carried out with [11,12-(3)H(2)]lithocholic acid and [24-(14)C]lithocholic acid. With [11,12-(3)H(2)]chenodeoxycholic acid, 10-15% of the label was lost during the first 24 hours, after which the (3)H/(14)C ratio of biliary chenodeoxycholic acid remained constant. The loss of (3)H caused a consistent overestimate of pool size by 10-15%, but since the fractional turnover rate of the two isotopes was identical, the calculated rate of synthesis was 10-15% greater when determined using (3)H. The loss of (3)H was attributed to the presence of (3)H in positions other than 11 and 12 occurring during the labeling procedure when the unsaturated precursor was exposed to carrier-free tritium gas. The (3)H/(14)C ratio of feces was consistently below that of bile, suggesting bacterial removal of additional (3)H during distal intestinal passage. In contrast, [11,12-(3)H(2)]lithocholic acid lost (3)H continuously, so that by 2 days after administration, the (3)H/(14)C ratio had fallen to nearly half of that of the administered mixture. It is concluded that isotope dilution studies may be carried out with one preparation of [11,12-(3)H(2)]chenodeoxycholic acid with an error of 10-15%; possibly a more stable preparation can be prepared with a gentler tritiation procedure. One preparation of [11,12-(3)H(2)]lithocholic acid was unsatisfactory for isotope dilution studies in man.

Cancer morbidity at all sites has been studied in a series of 676 patients with ulcerative colitis under long-term review, of whom more than two-thirds had extensive disease, and the level and pattern of risk over time examined. Age-, sex-, and site-specific incidence rates were used to compute the number of cancers that might have been expected to occur. A highly significant excess of cancers was observed overall but the excess was due entirely to cancers of the digestive system. In women there was no excess or deficit of cancers outside the digestive system. In men there was a small deficit of cancers of the respiratory system. An overall 11-fold excess colorectal cancer risk was found in the series compared with that in a relevant general population after patient-years at risk had been corrected for surgical resection and patients with colorectal cancer at their first referral had been corrected for surgical resection and patients with colorectal cancer at their first referral had been excluded. When these data were expressed in an actuarial form the cumulative probability of developing colorectal cancer in the series was 8% (3.5-13%) at 25 years, after the diagnosis of ulcerative colitis had been established. The relative risk of developing colorectal cancer was highest in those patients developing colitis before the age of 30 years, and the relative risk fell as the age at diagnosis of their colitis increased. The pattern of risk of colorectal cancer over time suggests that there is an association between cancer and colitis in susceptible individuals and that the level of risk is related to age at onset of colitis.

1. Propranolol and folic acid absorption were studied separately in healthy subjects, non-coeliac patients and patients with treated and untreated coeliac disease. 2. The surface pH of jejunal biopsy samples was measured with a pH electrode. 3. When compared with values found in healthy subjects, plasma propranolol levels were elevated in coeliac disease and, in contrast, serum folic acid levels were depressed after oral administration of the drug. Jejunal surface pH was more alkaline in the coeliac groups than in the healthy and non-coeliac subjects. 4. Pharmacokinetic analysis of the plasma drug levels allowed evaluation of the rate constants associated with absorption and elimination. The absorption rate constant was decreased for folic acid in the coeliac group, but increased for propranolol. These changes correlated with variation in surface pH. 5. Although the changes in drug disposition in treated and untreated coeliac disease are the result of several factors, it is suggested that jejunal surface pH may affect the rate of absorption and therefore the plasma concentration-time profile of drugs which undergo dissociation.

The hypothesis that abnormal fibronectin metabolism in Crohn's disease could be an important mechanism leading to stricture formation or postoperative infection was tested in three related studies. (1) Lower concentrations of plasma fibronectin (p less than 0.05) were found in 20 patients with small and large bowel Crohn's disease (mean 0.24 g/l) compared with 13 patients with more limited disease confined to only small or only large bowel (mean 0.27 g/l) or 20 healthy controls (mean 0.29 g/l). (2) In 25 patients followed for 10 days after operation for Crohn's disease, there was a significant fall in fibronectin concentrations of 43% (p less than 0.01, Wilcoxon's rank-sum test). This fall was maximal on the second postoperative day and was more marked in patients undergoing more major operative procedures. (3) The predictive value of plasma fibronectin for subsequent stricture formation or progression was studied for one year; during which 10 patients developed strictures requiring operative treatment. Higher plasma fibronectin concentrations were related to stricture formation, although there was not a complete, predictive relationship. In this study we found that plasma fibronectin concentrations were low in patients with extensive or severe Crohn's disease, fall after operation and may be related to the risk of stricture formation. This relationship is unlikely to be of clinical value, although it shows the potential significance of fibronectin in the pathogenesis of strictures.

We report the outcome of 63 consecutive patients with Crohn9s colitis treated by colectomy and ileorectal anastomosis between 1951-1981. There were no operative deaths. Serious postoperative complications occurred in 10 patients. The mean follow up since colectomy was 9.5 years. At 10 years the cumulative reoperation rate was 48% and the cumulative recurrence rate (based upon operative, radiological, and sigmoidoscopic evidence) was 64%. Ten patients have died during follow up. Two-thirds of the patients still under review have an intact ileorectal anastomosis. These results show that colectomy and ileorectal anastomosis is a useful operation in young patients with chronic ill health, diarrhoea, and anaemia in whom there is minimal anorectal disease. For suitable patients, the operation avoids the morbidity of a permanent stoma, pelvic dissection, and rectal excision. Recurrence developing in the ileum proximal to the anastomosis can often be resected and intestinal continuity preserved, whereas anorectal recurrence usually requires proctectomy.

The course, treatment and outcome of Crohn's disease affecting 174 patients diagnosed more than 20 years ago and followed for 20 to 45 years are reported. Fifty patients died, 23 related to Crohn's disease, against an expected 33, giving an increased mortality risk of 1.5 for the series. Fistulae of varying type occurring in 38 per cent of patients were successfully treated usually by excision and had no significant adverse effect upon the long term prognosis. The incidence of suicide and dementia appeared to be increased. Amyloid disease, obstructive uropathy and osteomalacia were other important complications. Definitive surgical operations averaged 2.5 per patient and after the first five years following diagnosis varied between 5 and 7 per cent of those at risk each year for the next 20 years. Recurrence rates were similar after any of the first five operations on a patient for the first 10 years after any operation. There was no evidence that the chance of operation was increased or made to recur more frequently by the performance of such an operation. All but two of the 121 living patients (three lost to follow-up) with a mean duration since diagnosis of 28 years were leading a normal life free from physical restrictions from Crohn's disease though three were mentally handicapped, 21 had radiological abnormality (recurrence), mostly longstanding, and 16 had less than 200 cm small intestine remaining. Specific therapy (corticosteroids 58 patients, sulphasalazine 11 patients and azathioprine four patients) was used at some time in only one third of the series so that such treatment did not appear necessary to produce good results in the majority. It is concluded that the indications for corticosteroid therapy are limited and that at present the joint efforts of physician and surgeon with adequate supportive therapy and well timed surgical resection produce the best results. An optimistic attitude to the eventual outcome is justified.

A series of 513 patients with Crohn's disease, who were followed for 1 to more than 35 yr by one unit, experienced a twofold increased risk of dying compared with a matched group drawn from the general population (102 observed; 51.8 expected; p0.001). The overall risk was similar for men and women and was greatest in the younger patients within a few years of diagnosis. The relative risk of dying decreased as the age at diagnosis of their Crohn's disease increased and also fell as the period of time from diagnosis increased. There was a small but significant excess of death from tumors of the digestive organs (9 observed; 4 expected; p0.05). There was also a significant excess of deaths from suicide in women (3 observed; 0.4 expected; p0.01). The major burden of mortality lay in causes directly attributable to Crohn's disease or to associated complications within the digestive system (46 observed; 1.41 expected; p0.001).

We have examined the causes of death and the changing pattern of mortality over time among 769 patients with Crohn's disease under long-term review between 1944 and 1984. One hundred and fifty-six patients have died. The cause of death could not be classified in five. The deaths were divided into those related directly to Crohn's disease and unrelated deaths from incidental causes. The cause and age at death among 59 unrelated deaths was similar to that expected in the general population. The common causes of related deaths were sepsis, digestive tract cancer, pulmonary embolus and metabolic disorders. There were 30 deaths after surgery, usually from sepsis after emergency surgical treatment. There have been fewer deaths in recent years and the cause of death has altered. Sepsis and electrolyte imbalance have declined in importance and cancer of the digestive tract is now the most common cause of related death. Occasional death from sepsis occurs in elderly patients. Elective rather than emergency surgical treatment and the routine use of antimicrobial prophylaxis before surgery has probably reduced mortality. With the exception of cancer of the digestive tract the appropriate medical and surgical treatment of patients with Crohn's disease can eliminate excess mortality previously associated with the disorder.

Lithocholate, a secondary bile acid, is hepatotoxic in many animal species including nonhuman primates. The induced histologic changes resemble those observed in patients with hepatic damage associated with inflammatory bowel disease. Accordingly, we have examined the hypothesis that lithocholate is of etiologic importance in causing this association by measuring serum and biliary lithocholates in inflammatory bowel disease patients with and without liver disease. Serum and biliary lithocholates and isolithocholates were normal in all patients. Because defective sulfation in the nonhuman primate which allows lithocholate to accumulate in the enterohepatic circulation is thought to be responsible for inducing liver damage and because secondary bile acids are reduced after colectomy and in established liver disease, we examined thae capacity of all patients to sulfate labeled lithocholate. Effective sulfation of lithocholate was demstrated in all groups. Despite the hepatotoxic effects observed in nonhuman primates, we have found no evidence so far to implicate lithocholate as an etiologic factor in inflammatory bowel disease and hepatic dysfunction nor have we detected other potentially hepatotoxic bile acids in these patients.

A recent mortality study of patients with ulcerative colitis showed significant deficits in deaths from cardiovascular disease and from respiratory disease in males. The reasons for the observed deficits have been examined by studying the risk factors in a consecutive series of 103 patients with ulcerative colitis using matched controls. The patients and controls were assessed for age, sex, social class, alcohol and tobacco consumption. Height, weight and blood pressure were recorded and fasting blood samples were taken for serum lipid estimation. Patients with ulcerative colitis had significantly lower systolic and diastolic blood pressures and the changes were particularly marked in patients with extensive colitis and after panproctocolectomy. After matching for age, sex and social class the patients with ulcerative colitis smoked significantly less than controls. Serum lipid measurements were similar in both groups. The lower blood pressure may be related to sodium and water depletion particularly in ileostomy patients and those with extensive colitis.

The prevalence of use of oral contraception before the onset of disease was established in 100 consecutive women attending follow up clinics for inflammatory bowel disease. A significant excess of women with Crohn9s disease confined to the colon had taken oral contraceptives in the year before developing symptoms (10/16 (63%] compared with women with small-intestinal Crohn9s disease (12/49 (24%); p less than 0.02) and women with ulcerative colitis (3/35 (9%); p less than 0.0005). When the patient groups were matched for age and year of onset of disease usage of oral contraception before the onset of disease was still more common among women with isolated colonic Crohn9s disease (9/12, 75%) than among those with ulcerative colitis (2/12 (17%); p less than 0.02) and was also more common than would be expected from reported figures for oral contraception in England and Wales (31.4% of women aged under 41; p less than 0.005). A survey of current patient records showed that isolated colonic disease was at least twice as common among women with Crohn9s disease (63/218, 29%) compared with men (25/181, 14%; p less than 0.001). These data support the suggestion made previously that oral contraceptives may predispose to a colitis that resembles colonic Crohn9s disease.

The crude five year survival of patients with colorectal cancer complicating ulcerative colitis in a large series of patients under long term review has, for the first time, been compared with the survival of patients with colorectal cancer in the general population (West Midlands region) from which the colitic patients were drawn. Thirty five cases of colorectal cancer were diagnosed in 676 patients with ulcerative colitis between 1944 and 1976. An actuarial five year survival curve was computed for the colitic and non-colitic patients with colorectal cancer. In ulcerative colitis patients with cancer the five year survival was 33.5% (range 16.9-50.1%) compared with 32.6% (28.2-37.0%) in the non-colitic cancer patients drawn from the relevant general population. Overall the prognosis is much better than earlier reports suggest. It is perhaps disappointing that in a closely monitored group the outcome is only as good as that in the general population. Surveillance programmes should improve the outcome in those patients with ulcerative colitis who accept the need for regular review.

A 10 year follow-up of 109 patients with histologic Crohn's disease and anal lesions is reported. Fourteen patients (13 percent) have died, 7 from unrelated disorders. Ten required excision of the rectum, but only 5 for perianal disease (4.5 percent). Of the remaining 85 patients, 61 have been followed up to proctoscopy and rectal biopsy. Anal skin tags were still evident in 25 of 37 patients (68 percent), but new tags have appeared in only 2 patients. Ten of 53 fissures (19 percent) were still present at 10 years, and there were no new fissures. Seven of 21 patients (33 percent) still had fistulas but were asymptomatic; the remainder of the fistulas had healed spontaneously (8) or after operation (6). New fistulas have appeared in five patients. None of the patients have been in continent. These results indicate that perianal manifestations of Crohn's disease pursue a relatively benign course and are rarely an indication for proctectomy.

Clinical observations in developing countries and experimental studies in animals suggest that preoperative weight loss adversely affects postoperative morbidity. Since it is not clear whether these findings can be applied to surgical series in European countries, we have studied 106 surgically-treated patients with inflammatory bowel disease. Patients were divided into three groups according to the degree of preoperative weight loss. The groups were well matched in all other respects. Postoperative morbidity was monitored independently. The outcome in all three groups was similar so that in these patients, at least, preoperative weight loss did not adversely affect the postoperative outcome.

The incidence of Clostridium difficile cytotoxin has been studied in 69 consecutive patients with inflammatory bowel disease complicated by severe diarrhoea or ileostomy flux during 74 admissions to hospital. The cytotoxin was identified in only four patients, all of whom had received antimicrobials. Clostridium difficle, but not cytotoxin, was identified in 10 of 43 admissions. This followed antimicrobial prophylaxis to cover a recent operation in two patients, and five were on long-term sulphasalazine. Only three patients with Clostridium difficile had not received an antimicrobial within one month of the study. Isolation of Clostridium difficile alone is of doubtful pathological significance, as it spontaneously disappeared without treatment in all patients.

Two hundred and two patients with Crohn's disease have been examined during the year 1984 to assess the frequency of perianal disease. One hundred and ten have had evidence of perianal complications (54%). In 30% of patients with perianal disease, the anal manifestations preceded any evidence of intestinal disease. Perianal disease was associated with pain in only 39%. Operations for perianal disease rarely achieved healing and were frequently associated with complications. Attempts to lay open a fistula-in-ano caused healing in only one of 12 cases and 6 developed incontinence. A high proportion of patients with Crohn's ulcers and strictures required proctectomy (87%). Proctectomy was performed in 27 patients with perianal disease of whom only 8 (30%) had primary healing of the proctectomy wound compared with complete healing in all 9 patients having a proctectomy without perianal disease (p less than 0.01). These results imply that patients with perianal Crohn's disease should be treated conservatively and that proctectomy, particularly in patients with rectal strictures, is associated with very high incidence of persistent perineal sinus.

Summary We report a patient with ulcerative colitis who underwent colectomy when aged 25 years for multifocal high grade dysplasia which had developed within 2 years of diagnosis, and only 5 1/2 years after his first symptoms. The implications of this case for the management of ulcerative colitis are discussed.