572 Proc. roy. Soc. Med. Volume 68 September 1975
An Epidemiological Survey of Crohn's Disease in Northern Ireland by W G Humphreys FRCS (The Queen's University of Belfast, Belfast, BT12 6BJ)
Regional ileitis is now accepted as a distinct clinical and pathological entity following the classical description of 13 cases in 1932 by Crohn et al. With increasing interest in the regional and racial incidence of this disease it is essential that groups of patients are defined by strict criteria. This paper presents the results of a retrospective survey of Crohn's disease in Northern Ireland for the years 1966-73 inclusive. It has been suggested that Crohn's disease is rare in the Irish (Edwards 1964). No previous survey has been carried out to confirm or refute this suggestion. A group of patients suffering from chronic Crohn's disease was defined by the following criteria:
16
in the latter half there was a large migration of population within the Belfast area. General practitioner and specialist facilities are available to the whole population and it was thought unlikely that patients would have moved into the towns to be nearer the treatment facilities. It has been assumed that the majority of patients with chronic Crohn's disease present or are referred to hospital for investigation and treatment (Barber et al. 1962, de Dombal et al. 1974). Most cases were accepted into the survey on the basis of a positive histological diagnosis or on a histological diagnosis suggestive of Crohn's disease with characteristic clinical or radiological findings. A few cases were accepted on their clinical and radiological findings alone. A total of 299 cases were traced with a provisional diagnosis of Crohn's disease; 135 were excluded because the diagnosis could not be substantiated when strict criteria were applied; 159 cases fulfilled the criteria for the study. In 5 other cases the records could not be traced. 3.0 7
(1) Clinical history of colicky abdominal pain, diarrhoea and weight loss. (2) Typical histological features in biopsy or resected specimens. (3) Characteristic radiological findings.
§ 2.C
8 Z
1.0
z
Data were obtained from the hospital case notes and these were traced through Hospital Central Fig 2 Incidence of new cases o Crohn's disease per 100 000 of the population at risk per year. The Records and the histopathology biopsy reference histograms represent the total incidence; the incidence in Belfast and Londonderry. both systems for males andfemales is shown separately by the Northern Ireland is a circumscribed area of dottedand solid lines 5459 square miles with a population of 1 536 000 in 1971 (Census). This figure was taken as the A number of cases were excluded from the mean population for the study as it was the most survey, i.e. those who were non-residents of accurate figure available. The population was Northern Ireland, those diagnosed and treated numerically stable throughout this period though outside Northern Ireland and patients with acute terminal ileitis. Patients with a recrudescence of 30 pre-1966 Crohn's disease or those with previously diagnosed ulcerative colitis subsequently reFEMALE designated Crohn's disease were omitted. 1ZMALE 20
zE
0-9
10-19
20-29.
30-39
40-49
X
AGE (YEARS)
Fig
Age (by decades)
atfirst presentation ofmales
and females with Crohn's disease
Results Of 159 cases examined in detail, 68 (43%) were male and 91 (57%) were female. Fig 1 shows the number of cases of Crohn's disease by sex and age (in decades) at the first presentation. It can be seen that there was a peak in the third decade with a preponderance of males and a second peak in the 5th to 7th decades with a marked preponderance of females. When the agespecific incidence rates per 100 000 of the population at risk was calculated there were true peaks in the 3rd and 7th decades (Fig 2). There was an unexplained difference in the sex incidence in the 5th and 6th decades with a higher rate among
17
573
Section ofProctology
Table 1
30. a
LARGE BOWEL
Crude incidence rates for urban and rural population
m SMALL BOWEL Area Belfast Londonderry City Antrim Armagh Londonderry County Down Tyrone Fermanagh
20
v5
z
10-
I
t. 0-9
-I
_ 10-19
MIi r4l
I 20-29
30-39
40-49
50-59
60-69
7G-79
80+
Fig 3 Anatomical distribution of Crohn's lesion in the large and small bowel in relation to age ofpatients (by decades)
females. The mean annual incidence of new cases was 1.3/100 000 (1.13 for men and 1.46 for women). It has been suggested that Crohn's disease is increasing in the community (Kyle 1971, Norlen et al. 1970). There was no appreciable difference in the number of new cases presenting each year of this survey period. Anatomical Distribution of the Lesion Of 159 cases studied, the disease process was present in the small intestine alone in 57 (34%), in the large intestine alone in 69 (45%) and as a combined lesion in 33 (21 %) cases. Of 88 cases of small intestinal disease, the terminal ileum was involved in 81, the jejunum in 1 and both ileum and jejunum in 6 instances. In the colon, 102 lesions were recorded, 22 in the ascending colon, 7 in the transverse colon, 61 in the left colon and rectum. Total colonic involvement was present in 7 patients and anal lesions only were seen in 5 cases. The pattern of small bowel disease conforms to the classical pattern with young adults being most affected (Fig 3). There was a peak incidence of large bowel disease in the third decade. In the later decades colonic
Expected incidence 38 5 37 14 13 32 14 5
Observed incidence 63 9 28 12 2 33 8 4
Comment A group of patients suffering from chronic Crohn's disease has been defined in Northern Ireland. The proportion of females affected is higher than in most reported series although less than that reported in North East Scotland (Kyle 1971). The patients tended to present early for diagnosis and treatment. There was a marked difference between the urban and rural incidence. Colonic involvement accounted for most of the cases in the later decades. This pattern is similar to that described by Kyle (1971) and Krause (1971). There was no apparent increase in the disease in the period of this survey though other observers have noted a continued upward trend in recent years (Kyle 1971, Norlen et al. 1970). The calculated annual incidence rate of new cases 1.3/100 000 of the population places the Northern Irish in the lower half of Table 2. The figures are not strictly comparable but indicate the approximate incidence of Crohn's disease in other areas, derived by similar methods. We suggest that Crohn's disease has a comparable distribution to other areas of the United Kingdom and is not rarer in the Northern Irish than in the population of the British Isles as a whole.
Table 2
Scotland N. Ireland
Date of Survey
1960-63 1951-60
Vastmanland
1963-68 1966 1956-61 1962-67
Aberdeen
1955-61 1962-68 1966-73
100 000/year 3.50 2.16 0.90 1.21 1.16 0.29 0.54 1.00
disease predominated among females in the survey. Geographically, the cases were found to be distributed in the areas of high population density. The recorded incidence was much greater than expected in the towns and much less than expected in the rural areas (Table 1). The difference was statistically significant.
Average crude incidence rates of Crohn's disease (per 100 000 per year). Modified from Evans (1972) Town Country United States of America Baltimore Oxford England Leeds Gloucester Sweden Uppsala and
Incidencel
Male Female Author 2.5 Monk et al. (1969) 1.2 Evans &Acheson (1965) 0.8 0.8 de Dombal (1971) Tresadern et al. (1973) Krause (1971) 26 2.4 2 Norlen et al. (1970) | 2 1.9 1.4 Kyle (1971), Kyle & Blair (1965) 1.6 3.0 1.13 1.46
Total 1.8 0.8 3.4 1.5
2.5 2.2 1.3
574 Proc. roy. Soc. Med. Volume 68 September 1975 REFERENCES Barber K W, Waugh J M, Beabrs O H & Sauer W G (1962) Annals of Surgery 156,472-480 Crohn B B, Ginzberg L & Oppenheimer G D (1932) Journal of the American Medical Association 99, 1323-1329 de Dombal F T (1971) Proceedings ofthe Royal Society ofMedicine 64, 161 de Dombal F T, Burton I, Clamp S E & Goligher J C (1974) Gut 15,435-443 Edwards H C (1964) Journal ofthe Royal College ofSurgeons of Edinburgh 9, 115-127 Evans J G (1972) Clinics in Gastroenterology 1, 335-347 Evans J G & Acheson E D (1965) Gut 6, 311-324 Krause U (1971) In: Regional Enteritis (Crohn's Disease). Ed. A Engel & T Larsson. Skandia International Symposia. Nordiska Bokhandelns Forlag, Stockholm; pp 142-151 Kyle J (1971) Gastroenterology 61, 826-833 Kyle J & Blair D W (1965) British Journal of Surgery 52, 215-217 Monk M (1969) Gastroenterology 56, 847-857 Norlen B J, Krause U & Bergman L (1970) Scandinavian Journal of Gastroenterology 5, 385-390 Tresadern J C, Gear M W L & Nicholl A (1973) British Journal ofSurgery 60, 366-368
18
internal sphincter but partly from the conjoined longitudinal muscle. Its arrangement is such that its contraction would cause a flattening and bracing of the cushions in the anal canal. Although hwmorrhoids, in the 80 patients examined, occurred mainly in the right anterior, right posterior and left lateral positions, their arrangement was found to have nothing to do with the branching behaviour of the superior rectal artery (which exhibited a great variety of patterns). Another belief dispelled was that piles progress through a stage when they bleed to one when they prolapse. Of the 66 consecutive patients questioned, the great majority were certain that prolapse had been the first symptom. The histological appearance of hemorrhoidectomy specimens was similar to that of normal anal cushions obtained from cadavers, any differences being readily attributable to the effects of the trauma or prolapse. One such was capillary dilatation and hyperplasia in the lamina propria seen in some specimens. It seems they are the usual source of bleeding in piles, the venous The Nature and Cause of H1emorrhoids dilatations lying deep to the muscularis mucose. Finally a comparison of bowel histories showed a much greater prevalence of constipation and by W H F Thomson' MS FRCS straining in patients with hemorrhoids than in (Southampton General Hospital, those without. Shirley, Southampton, S09 4XY) The normal anal canal, then, is lined by cushions Presented is an anatomical and clinical study of of specialized tissue in the same positions thatthe anorectum with special reference to the piles occur. The cushions are supported by a nature of hvmorrhoids. Material for the in- strategic arrangement of smooth muscle and vestigation was provided by 95 cadaveric elastic tissue and the evidence suggests that piles anorectal specimens, 25 hemorrhoidectomy are merely the outward manifestation of their specimens, 80 patients with prolapsing piles, and downward displacement. They would be more likely to be pushed out by a large hard stool 42 normal controls. The main findings were as follows: the anal which perhaps accounts for their association submucosa is not a uniform layer. It is thickened with constipation. Straining might cause suffusion into 3 main cushions, more or less discrete, of the venous dilatations with resultant swelling which occur regularly in the left lateral, right of the cushions and increased likelihood of their anterior and right posterior positions, each expulsion on defaecation. Such a concept has implications for the managecushion extending above and below the pectinate line. Because of them, the anal lumen appears ment of haemorrhoids. It means, since the on proctoscopic examination as a triradiate slit. cushions assist in continence, that piles should be The cushions are composed of a plexus of veins treated as conservatively as possible. If bleeding supported by a stroma of smooth muscle and is the main complaint, the responsible pile may elastic tissue, and have a highly developed be recognized by its inflamed appearance or arterial supply. Direct arteriovenous communica- friable surface and perhaps most appropriately tions were found in their substance. The veins dealt with by elastic band ligature. If prolapse in both infants and adults are a complex (ending needs treatment, redundant tissue only should be at the anal verge) of discrete dilatations so that a excised. The operation should not only be conmechanism is available whereby great changes in sidered 'over' when 'it looks like a clover'; if only the size of the cushions can take place. The one cushion is prolapsing, then it is the only 'pile' finding supports the concept that the anal sub- which needs excision. If the patient's main mucosa assists in anal closure. The smooth complaint is of discomfort on defaecation or muscle component is derived mainly from the difficulty with replacing prolapsed piles then forcible manual dilatation of the anus may be 'Present address: Gloucester Royal Hospital, best. Southgate Street, Gloucester
An Epidemiological Survey of Crohn's Disease in Northern Ireland by W G Humphreys FRCS (The Queen's University of Belfast, Belfast, BT12 6BJ)
Regional ileitis is now accepted as a distinct clinical and pathological entity following the classical description of 13 cases in 1932 by Crohn et al. With increasing interest in the regional and racial incidence of this disease it is essential that groups of patients are defined by strict criteria. This paper presents the results of a retrospective survey of Crohn's disease in Northern Ireland for the years 1966-73 inclusive. It has been suggested that Crohn's disease is rare in the Irish (Edwards 1964). No previous survey has been carried out to confirm or refute this suggestion. A group of patients suffering from chronic Crohn's disease was defined by the following criteria:
16
in the latter half there was a large migration of population within the Belfast area. General practitioner and specialist facilities are available to the whole population and it was thought unlikely that patients would have moved into the towns to be nearer the treatment facilities. It has been assumed that the majority of patients with chronic Crohn's disease present or are referred to hospital for investigation and treatment (Barber et al. 1962, de Dombal et al. 1974). Most cases were accepted into the survey on the basis of a positive histological diagnosis or on a histological diagnosis suggestive of Crohn's disease with characteristic clinical or radiological findings. A few cases were accepted on their clinical and radiological findings alone. A total of 299 cases were traced with a provisional diagnosis of Crohn's disease; 135 were excluded because the diagnosis could not be substantiated when strict criteria were applied; 159 cases fulfilled the criteria for the study. In 5 other cases the records could not be traced. 3.0 7
(1) Clinical history of colicky abdominal pain, diarrhoea and weight loss. (2) Typical histological features in biopsy or resected specimens. (3) Characteristic radiological findings.
§ 2.C
8 Z
1.0
z
Data were obtained from the hospital case notes and these were traced through Hospital Central Fig 2 Incidence of new cases o Crohn's disease per 100 000 of the population at risk per year. The Records and the histopathology biopsy reference histograms represent the total incidence; the incidence in Belfast and Londonderry. both systems for males andfemales is shown separately by the Northern Ireland is a circumscribed area of dottedand solid lines 5459 square miles with a population of 1 536 000 in 1971 (Census). This figure was taken as the A number of cases were excluded from the mean population for the study as it was the most survey, i.e. those who were non-residents of accurate figure available. The population was Northern Ireland, those diagnosed and treated numerically stable throughout this period though outside Northern Ireland and patients with acute terminal ileitis. Patients with a recrudescence of 30 pre-1966 Crohn's disease or those with previously diagnosed ulcerative colitis subsequently reFEMALE designated Crohn's disease were omitted. 1ZMALE 20
zE
0-9
10-19
20-29.
30-39
40-49
X
AGE (YEARS)
Fig
Age (by decades)
atfirst presentation ofmales
and females with Crohn's disease
Results Of 159 cases examined in detail, 68 (43%) were male and 91 (57%) were female. Fig 1 shows the number of cases of Crohn's disease by sex and age (in decades) at the first presentation. It can be seen that there was a peak in the third decade with a preponderance of males and a second peak in the 5th to 7th decades with a marked preponderance of females. When the agespecific incidence rates per 100 000 of the population at risk was calculated there were true peaks in the 3rd and 7th decades (Fig 2). There was an unexplained difference in the sex incidence in the 5th and 6th decades with a higher rate among
17
573
Section ofProctology
Table 1
30. a
LARGE BOWEL
Crude incidence rates for urban and rural population
m SMALL BOWEL Area Belfast Londonderry City Antrim Armagh Londonderry County Down Tyrone Fermanagh
20
v5
z
10-
I
t. 0-9
-I
_ 10-19
MIi r4l
I 20-29
30-39
40-49
50-59
60-69
7G-79
80+
Fig 3 Anatomical distribution of Crohn's lesion in the large and small bowel in relation to age ofpatients (by decades)
females. The mean annual incidence of new cases was 1.3/100 000 (1.13 for men and 1.46 for women). It has been suggested that Crohn's disease is increasing in the community (Kyle 1971, Norlen et al. 1970). There was no appreciable difference in the number of new cases presenting each year of this survey period. Anatomical Distribution of the Lesion Of 159 cases studied, the disease process was present in the small intestine alone in 57 (34%), in the large intestine alone in 69 (45%) and as a combined lesion in 33 (21 %) cases. Of 88 cases of small intestinal disease, the terminal ileum was involved in 81, the jejunum in 1 and both ileum and jejunum in 6 instances. In the colon, 102 lesions were recorded, 22 in the ascending colon, 7 in the transverse colon, 61 in the left colon and rectum. Total colonic involvement was present in 7 patients and anal lesions only were seen in 5 cases. The pattern of small bowel disease conforms to the classical pattern with young adults being most affected (Fig 3). There was a peak incidence of large bowel disease in the third decade. In the later decades colonic
Expected incidence 38 5 37 14 13 32 14 5
Observed incidence 63 9 28 12 2 33 8 4
Comment A group of patients suffering from chronic Crohn's disease has been defined in Northern Ireland. The proportion of females affected is higher than in most reported series although less than that reported in North East Scotland (Kyle 1971). The patients tended to present early for diagnosis and treatment. There was a marked difference between the urban and rural incidence. Colonic involvement accounted for most of the cases in the later decades. This pattern is similar to that described by Kyle (1971) and Krause (1971). There was no apparent increase in the disease in the period of this survey though other observers have noted a continued upward trend in recent years (Kyle 1971, Norlen et al. 1970). The calculated annual incidence rate of new cases 1.3/100 000 of the population places the Northern Irish in the lower half of Table 2. The figures are not strictly comparable but indicate the approximate incidence of Crohn's disease in other areas, derived by similar methods. We suggest that Crohn's disease has a comparable distribution to other areas of the United Kingdom and is not rarer in the Northern Irish than in the population of the British Isles as a whole.
Table 2
Scotland N. Ireland
Date of Survey
1960-63 1951-60
Vastmanland
1963-68 1966 1956-61 1962-67
Aberdeen
1955-61 1962-68 1966-73
100 000/year 3.50 2.16 0.90 1.21 1.16 0.29 0.54 1.00
disease predominated among females in the survey. Geographically, the cases were found to be distributed in the areas of high population density. The recorded incidence was much greater than expected in the towns and much less than expected in the rural areas (Table 1). The difference was statistically significant.
Average crude incidence rates of Crohn's disease (per 100 000 per year). Modified from Evans (1972) Town Country United States of America Baltimore Oxford England Leeds Gloucester Sweden Uppsala and
Incidencel
Male Female Author 2.5 Monk et al. (1969) 1.2 Evans &Acheson (1965) 0.8 0.8 de Dombal (1971) Tresadern et al. (1973) Krause (1971) 26 2.4 2 Norlen et al. (1970) | 2 1.9 1.4 Kyle (1971), Kyle & Blair (1965) 1.6 3.0 1.13 1.46
Total 1.8 0.8 3.4 1.5
2.5 2.2 1.3
574 Proc. roy. Soc. Med. Volume 68 September 1975 REFERENCES Barber K W, Waugh J M, Beabrs O H & Sauer W G (1962) Annals of Surgery 156,472-480 Crohn B B, Ginzberg L & Oppenheimer G D (1932) Journal of the American Medical Association 99, 1323-1329 de Dombal F T (1971) Proceedings ofthe Royal Society ofMedicine 64, 161 de Dombal F T, Burton I, Clamp S E & Goligher J C (1974) Gut 15,435-443 Edwards H C (1964) Journal ofthe Royal College ofSurgeons of Edinburgh 9, 115-127 Evans J G (1972) Clinics in Gastroenterology 1, 335-347 Evans J G & Acheson E D (1965) Gut 6, 311-324 Krause U (1971) In: Regional Enteritis (Crohn's Disease). Ed. A Engel & T Larsson. Skandia International Symposia. Nordiska Bokhandelns Forlag, Stockholm; pp 142-151 Kyle J (1971) Gastroenterology 61, 826-833 Kyle J & Blair D W (1965) British Journal of Surgery 52, 215-217 Monk M (1969) Gastroenterology 56, 847-857 Norlen B J, Krause U & Bergman L (1970) Scandinavian Journal of Gastroenterology 5, 385-390 Tresadern J C, Gear M W L & Nicholl A (1973) British Journal ofSurgery 60, 366-368
18
internal sphincter but partly from the conjoined longitudinal muscle. Its arrangement is such that its contraction would cause a flattening and bracing of the cushions in the anal canal. Although hwmorrhoids, in the 80 patients examined, occurred mainly in the right anterior, right posterior and left lateral positions, their arrangement was found to have nothing to do with the branching behaviour of the superior rectal artery (which exhibited a great variety of patterns). Another belief dispelled was that piles progress through a stage when they bleed to one when they prolapse. Of the 66 consecutive patients questioned, the great majority were certain that prolapse had been the first symptom. The histological appearance of hemorrhoidectomy specimens was similar to that of normal anal cushions obtained from cadavers, any differences being readily attributable to the effects of the trauma or prolapse. One such was capillary dilatation and hyperplasia in the lamina propria seen in some specimens. It seems they are the usual source of bleeding in piles, the venous The Nature and Cause of H1emorrhoids dilatations lying deep to the muscularis mucose. Finally a comparison of bowel histories showed a much greater prevalence of constipation and by W H F Thomson' MS FRCS straining in patients with hemorrhoids than in (Southampton General Hospital, those without. Shirley, Southampton, S09 4XY) The normal anal canal, then, is lined by cushions Presented is an anatomical and clinical study of of specialized tissue in the same positions thatthe anorectum with special reference to the piles occur. The cushions are supported by a nature of hvmorrhoids. Material for the in- strategic arrangement of smooth muscle and vestigation was provided by 95 cadaveric elastic tissue and the evidence suggests that piles anorectal specimens, 25 hemorrhoidectomy are merely the outward manifestation of their specimens, 80 patients with prolapsing piles, and downward displacement. They would be more likely to be pushed out by a large hard stool 42 normal controls. The main findings were as follows: the anal which perhaps accounts for their association submucosa is not a uniform layer. It is thickened with constipation. Straining might cause suffusion into 3 main cushions, more or less discrete, of the venous dilatations with resultant swelling which occur regularly in the left lateral, right of the cushions and increased likelihood of their anterior and right posterior positions, each expulsion on defaecation. Such a concept has implications for the managecushion extending above and below the pectinate line. Because of them, the anal lumen appears ment of haemorrhoids. It means, since the on proctoscopic examination as a triradiate slit. cushions assist in continence, that piles should be The cushions are composed of a plexus of veins treated as conservatively as possible. If bleeding supported by a stroma of smooth muscle and is the main complaint, the responsible pile may elastic tissue, and have a highly developed be recognized by its inflamed appearance or arterial supply. Direct arteriovenous communica- friable surface and perhaps most appropriately tions were found in their substance. The veins dealt with by elastic band ligature. If prolapse in both infants and adults are a complex (ending needs treatment, redundant tissue only should be at the anal verge) of discrete dilatations so that a excised. The operation should not only be conmechanism is available whereby great changes in sidered 'over' when 'it looks like a clover'; if only the size of the cushions can take place. The one cushion is prolapsing, then it is the only 'pile' finding supports the concept that the anal sub- which needs excision. If the patient's main mucosa assists in anal closure. The smooth complaint is of discomfort on defaecation or muscle component is derived mainly from the difficulty with replacing prolapsed piles then forcible manual dilatation of the anus may be 'Present address: Gloucester Royal Hospital, best. Southgate Street, Gloucester
