The inflammatory diseases of the large and small bowel

The inflammatory diseases of the large and small bowel

jJ is Professor of Surgery at The University of Chicago. He received his M.D. from that inst/tution in 1933, and served his internship at Billings Hc...

9MB Sizes 0 Downloads 21 Views


is Professor of Surgery at The University of Chicago. He received his M.D. from that inst/tution in 1933, and served his internship at Billings Hcspital, followed by residency training at The University of Chicago. In 1942 he joined the staff at the University of Illinois College of Medicine as Clinical Associate Professor of Surgery, returning to The University of Chicago in 1954 to accept a Professorship in Surg~-ry on a full-time basis. Doctor Prohaska is a member of most of the national surgical associations, and has contributed extensively to the literature. His major interests lie in the areas of gastroenterology and mammary carcinoma metastases.

THE INFLAMMATORYD I S E A S E S o f the large and small bowel comprise a ~ o u p of diseases im, olving the colon or the small intestine or both in continuity or in segmental fashion. The inflammatory response of the intestine induces a variety of cli r~al, anatomic and histologic signs, complications and sequelae which do not always clearly differentiate one type of inflammatory disease from another. Forthis reason, terms like ulcerative colitis, Crohn's disease of the colon, granulomatous colitis and ileocolitis are sometimes used interchangeably, often with slumbering disregard for discernible factors which could discriminate one from the others. A meaningful record of our present knowledge and the analysis of the problems that exist in the category of these diseases cannot be made without a common language. It is obvious that some kind of uniform classification must be established. The desire for 6 a workable classification becomes even more urgent if one considers thatjn the past 20 years most diseases have ~chieved the dignity of specific etiologic identification, acceptable classification and some degree of stability in therapeutic management. This kind of advancement has not been recorded in the area of inflammatory diseases of the large and small bowel. Today, there remains the outstanding fact that the etiology of the inflammatory diseases is not known. For this reason, many clinicians and investigators remain timid in offering a classification despite its urgent need, urgent in the face of increasing communication related to all aspects of the disease. A classification should be useful for statistical analysis and for the recording of new clinical and experimental data, and for intelligem dialogue between investigators. It should be made immediatelY workable irrespective of possible future modifications should the cause of the diseases ever be discovered. 3


I. Ulcerative colitis, chr0nic, idiopathic (Fig. l ) II. Granulomatous disease of the large and small bowel (Crohn's disease) 1) Regional enteritis (Fig. 2) 2) Ileocolitis (Fig. 3) 3) Granulomatous colitis (Crohn's disease of the colon (Fig. 4, A and B, and Fig. 5, A and B) a) with granulomas b) without granulomas III. Inflammatory disease of the colon 1 ) Mixed variety (see Fig. 6, A and B) 2) Unclassifiable (see Fig. 7, A and B) This classification implies that the process in regional enteritis, ileocolitis and granulomatous colitis is the sanie inflammatory process but expressed in different areas of the intestine. Using this classification it is possible to segregate ulcerative colitis from granulomatous colitis in most cases. The presence or absence of granulomas should not be used as a sole diagnostic criterion; instead, the specific diagnosis must be sought in the total picture of the disease, including clinical, roentgenographic and laboratory expressions of the inflammatory process. It is highly questionable whether a tiny microscopic segment of the disease involving the entire colon or a large portion of the small bowel can really adequately rule out the existence of granulomas in the entire specimen. Rappaport, in his excellent article on the pathology of regional enteritis writes, "tubercle-like, nodular granul6ma observed in almost one half of the cases was a particularly striking picture of the disease (regional enteritis).. These granulomas develop in preexisting hyperplastic lymphoid tissue or in ectopic lymph follicles, and they frequently are observed in or underneath intact mucosa. The evidence suggests that they precede formation of ulcers and that secondary changes such as exudative inflammation ant/fibrosis may render them unrecognizable or lead to their obliteration." In defense of this classification, one must point out the great differences between true ulcerative colitis on the one hand and granulomatous (Crohn's) diseases on the other hand. These differences not only express themselves histologically but also reflect themselves clearly in the gross specimen, in the course of the disease, in the nature of the complications, in roentgenographic examinations, and in the response to seemingly curative surgical procedures. It is, therefore, possible to identify clinical events, anatomic and histologic criteria which in their totality separate the inflammatory diseases into•



FIG 1 (above).. Typical Chronic ulcerative colitis with pseudopolyposis. The entire colon is almost uniformly involved from rectum to,the ileocecal junction. The terminal ileum is normal. The pr0ctocolectomy was performed on a female patienL aged 31, after 8 years of relemless active colitis associated with episodes o f bleeding. Fro. 2 (below).--Regional enteritis involving the lower jejunum. The disease presents itself with narrowing of the lumen, thickened bowel wall, thick mesentery, mesenteric lymphadenopathy and enteroenteric fistula demonstrated by the probe; Specimen removed fror:t a male aged 17. 5

Fxt3. 3.--Ileocolilis in a male patient aged 28. The terminal ileum is markedly involved in stenosis due to submucous fibrosis and edema. The cecum is likewise involved in granul9matous disease. The typical ileocolic fistula is demonstrated by the probe. The patient also had enterocutaneous fistula. 1 ) Ulcerative colitis; or 2) Granulomatous di~ase of the large and small bowel, with or without histologically recognizable granulomas. Ignoring the single criterion of histologic identification of nodular granuloma, it becomes possible to bring together under one heading the various terms used to describe those inflammatory diseases of the large bowel which differ significantly from ulcerative colitis. Tile great differences that exist between them can be described without entanglement in lengthy terms related to granulomatous or agranulomatous states, to this or that segment of tile intestine. The discussion of classification does not end here, for there are instances where the assemblage of all clinical, anatomic and histologic signs produces a confusing picture short of positive identification. Age of onset of he 'd~sease, seventy of ~ts course, antlb~otlc and corticosteroid "therapy, as well as unknown factors, may induce changes in the bowel so that differentiation between ulcerative and granulomatouscolitis i s n o longer possible. The question must be raised whether this atypical ic01itis represents a disease entity, separate and distinct from ulcerative colitis and, yet, different from Crohn's disease, or is it o n e of them superimposed on the other? Obviously, these questions cannot be answered at the present stage of our knowledge. F o r this reason, the mixed forms and the unclassifiable variety are included in the system of classification. Chronic idiopathic ulcerative coIitis is an inflammatory disease typically limited t o the colon and rectum. It is characterized by bloody 6

Fro. 4.: A, chronic granulomatous colitis of l0 Ye~irs duration in a woman aged 46. Note the rigidity of the colonic wall d u e to s u b m u c o u s fibrosis and edema. The specimen shows inadequate resection. Early recurrence followed, ne, cessitating resection of the:residual colon l a n d permanent ileostomy. B, higher magnification of the rectal part of specimen in A. The submucous fibrosis and edema and the cobblestone mucosa distinguish the disease from ulcerative colitis;

diarrheas and a mild chronic course~ marked by periods of remissions and exacerbations insome gases and by acute fulminating episodes in others. Examination of a typical: clinical course demonstrates the dif' ferences between ulcerative and granulomatous colitis. ~The course of ulcerative colitis isnot infrequently marked by dilatation of the colon, go'cailed toxic megacolon, bloody diarrheas, pseudopolyposis and the 7

FIG. 5.--A, chronic granulomatous colitis of 6 years duration in male patient aged41. The long!lohgi.udinal ulcers and the involvement of the terminal ileum differentiate this:r entity from chronic ulcerative colitis. The mucosa is not uniformly involved; there are areas where, grossly, the mucosa appears almost normal. B, higher lmagnification of a section from specimen in A, showing the deep longitudinal ulcer, the cobblestone mucosa and the fibrosis of the submucosa, classic signs of granulomatous disease.


evolution of carcinoma in the colorectal area. Fistulization, which is predicated on slow perforation of the bowel, occurs in ulcerative colitis only in the perineum. It develops in the colon bnlv in cases of acute dilatation (toxic megacolon) and results in generalized fecal peritonitis. Usually the entire colon and rectum are invoivi:d in ulcerations. The small bowel is never involved except by so-call~d backwash ileitis. A roentgenogram reveals a thin-walled colon of flormal length with typical diffuse small ulcerations and often pseudopolyposis. The examination of the colon at laparotomy or the examination of the resected colon in the laboratory shows the colon to have a thin wall, with normal mesentery and in which the mesenteric lymph nodes are not significantly enlarged or increased in number. Tile mucosal surface of the colon shows minute ulcerations over granular mucosa scattered throughout the colon and rectum in a surprisingly uniform fashion. Numerous pseudopolyps may be present. A!longitudinal section reveals the bowel wall and the mucosa to be thin. The submucosa is not replaced by thick rigid fibrous tissue. The dise~ase process appears to be limited to the mucosa and submucosa. The disease does not recur in the small intestine following the resection of the colon and rectum. The operation of total colectomy with resection of tlle rectum and permanent ileostomy cures the disease. On the other hand, granulomatous colitis, in contrast with chronic ulcerative colitis, is almost never complicated by acute dilatation of the colon, seldom produces sudden perforation of the large bowel with generalized peritonitis, and seldom causes persistent hemorrhages comparable to the severe exsanguinating hemorrhages found in ulcerative colitis. The clinical course of Crohn's disease is marked by the development of persistent fistulas which may be enterocutaneous, enteroenteric and enterocolic in type, as well as perineal fistulas, including anorectal and rectovaginal. The disease may be limited to, the terminal ileum, cecum and ascending colon or the entire colon in a segmental fashion or in continuity. The chief finding on roentgenographic examination is a narrowing of the bowel. On barium examination of the colon anal the small bowel, these structures appear narrow, tubular and shortened, often with areas of strictures. The high rise of the splenic flexure is usually absent. In some patients, enteroenteric fistulas may be demonstrated. These enteroenteric fistulas are often associated with tender, palpable, intraabdominal masses composed of adherent loops of bowel, ~,,i~t laparotomy, Crohn's disease of the small bowel reveals itself as a segment of edematous, thickened small intestine with thick mesentery containing many enlarged lymph nodes. The mesenteric fat encroaches on the serosa of the thickenecl bowel. The invol~,ed intestinal segment is not sharply delineated from the normal bowel. The abnormal fades into the normal imperceptibly. The terminal ileum is most frequently in9

FI6. 6.~,4, severe chronic ulcerative colitis complicated by acute toxic megacolon with perforation and fecal peritonitis. The typical ulcerative colitis is associated with granulomatous terminal ileitis. The patient, a woman aged 36, entered the hospital with obvious generalized peritonitis, known mitral stenosis and periodic heart failure. She died 5 days after total colectomy for toxic megacolon with multiple perforations and fecal peritonitis. Autopsy demonstrated severe congestive heart failure and minimal serosanguinous peritonitis. B, section of cecum and terminal ileum. The cecum shows chronic ulcerative colitis with acute dilatation; the ileum, involved over a length of 40 cm. reveals typical granulomatous ileitis (Crohn's disease). Note the rigidity of the ileal wall due to submucous fibrosis. This picture should never pass for so-called backwash ileitis.


volved. The granulomatous disease has been described as affecting the duodenum, stomach and even the esophagus. Involvement of the small bowel may present itself in several areas interrupted by normal segments of intestine. The serosal surface of the small bowel appears hyperemic. The colon, when involved, is thick, shortened and rubbery, with injected serosa, and its mesentery is thickened, cpr~tmnlng many enlarged lymph nodes. The involvement of the colon may also present itself in skip areas. The surgical specimen shows the mucosal surface of the open bowel to contain longitudinal ulcers coursing along the sides of the destroyed lymphoid ridges like dried-up, riverbeds with denuded bottoms. The thickened mucosa between the ul',cerations is arranged in Cobblestone fashion. In other areas an entire section of the mucosa may be lifted from the submucosa to form fibrous tunnels and mucosal bridges. The most significant surgical finding is the marked thickening and fibrosis with edema,of the submucosa. When the surgeon transects the bowel for resection he can see the cross-section and if thickened submucosa is present, it indicates that his resection is inadequate and will most likely result in recurrence at the line of anastomosis. This submucosal fibrosis is thickest in the areas of the greatest bowel involvement. From this point the fibrosis gradually fades into the region of normal bowel with normal-appearing mucosa. It is this marked submucous fibrosis that thickens the bowel and gives it its garden-hose appearance. The surgeon must carry the resection of the bowel well beyond the point of submucosal fibrosis. Another great difference between granulomatous disease and ulcerative colitis is the fact that in granulomatous ileocolitis, seemingly adequate resection of the involved small or large bowel or segment is often followed by the recurrence of the disease, either at the site of the anastomosis or in another part of the intestine. These, then, are the great generai differences that exist between chronic ulcerative colitis and the granulomatous diseases of the small bowel and the colon. It must be added that cancer of the colon or rectum seems to develop only in ulcerative colitis where its incidence rises with the duration of the disease. In the study of 354 patients with inflammatory bowel disease treated by curative surgical procedures, we found no instance of carcinoma in the group of patients with granulomatous colitis or Crohn's disease of the colon, but we did find that 5 % of the patients with chronic ulcerative colitis developed colorectal carcinomas. The distribution of the various types of the inflammatory diseases may now be reviewed in light of the proposed classification. The 354 patients had curative resection of the involved bowel and have been followed for 1-12 years. Recurrence of the inflammatory disease requiring additional surgical procedures and a summary of the surgical ll

Fro. 7 . - - A , an inflammatory disease of the colon difficult to classify. Total colectomy and ileorecta~ariastomosis was performed on this male aged 17. Four years later the rectum became severely involved, necessitating resection and permanent ileostomy. Opinions based on gross and histologic studies remained equally divided between granulomatous and chronic ulcerative colitis. The pseudopolyposis favors a diagnosis of ulcerative colitis; the Jarg e denuded area suggests granulomatous disease. B, segment, from A showing the large area denuded of mucosa and a mucosal bridge. In this segment the mucosa assumes a cobblestone appearance. Is this unclassifiable, o r a mixed variety? "



TABLE l.--DtsTamtrno~ OF VxruotJs TYPES OF I N F L A M M A T O R Y B O W E L DISEASE*

Ulcerative colitis










44 144 26

40% 7%

18 2

12% 4%

2 0








} { 100

Regional enteritis _ Totals Granulomatous colitis Unclassified inflammatory colitis Ulcerative colitis with regional enteritis To'raks


1 354




6.7 %




100% 3.7%

*Data are based on 354 consecutive i',atients treated by resection of diseased bowel. N o t e the relative rate of recurrence a n d m o r t a l i t y between true ulcerative colitid at:,d the g r a n u l o m a tous diseases. T h e 4 recurrences in ulcerative colitis were observed only in patients who h a d terminal ileum a n a s t o m o s e d to a p p a r e n t l y n o r m a l rectum at the time of total colectomy.

mortality have been added to Table 1 for a convenient initial scanning of the over-all results. The long observation following surgical resection makes the data related to recurrences meaningful. The 4 recurrences in ulcerative colitis were due to intentional preservation of seemingly normal rectum. The operative procedure in these patients consisted of total colectomy and ileorectal anastomosis. F o r the purpose of clarity, the various inflammatory diseases of the large and small bowel will now be discussed in the order in which they are listed in the classification table.

CHRONIC ULCERATIVE COLITIS Chronic idiopathic ulcerative colitis is characterized by bloody diarrhea and a mild chronic course, interrupted by periods of remissions and exacerbations. In some patients, the disease remains as a smoldering, inactive process, allowing the patient to live in symbiosis with his chronic disease of the colon. In others, the disease erupts in acute fulminating complication, requiring urgent medical or surgical management. The disease m a y begin at an early age with a peak age distribution between file ages of 18 and 40. In a study of 180 patients with ulcerative colitis, there were approximately an equal number of patients in each decade. After age 42, the incidence drops precipitously. No patients over 66 were recorded in this ~ o u p of, 180 surNcal cases. The sex distribution 13

is approximately equal. Ulcerative colitis has a world-wide distribution but appears to be prevalent in the United States, England, Scandinavian countries and Northern Germany. It is less common in the Central European states and occurs with decreasing frequency in South America, Asia and Africa. People of Jewish orion appear to be slightly more often involved with ulcerative colitis than other national groups.

ETIOLOGY The etiology of ulcerative colitis is not known. It is logical that the initial steps in the search for an etmlogic agent would relate to microorganisms. So far, no pathogen has been found as the causal agent of ulcerative colitis. The disease and its relapses are often associated with emotional stress, suggesting a psychogenic cause. Be that as it may, the pathways by which the physical changes in the colon are induced by psychogenic upheavals are not known. Lysozymes have been found in excessive amounts in feces of patients with ulcerative colitis. The lysozyme theory had its period of popularity for awhile but it was soon abandoned because excessive lysozymes are, most likely, the byproduct of inflammation and pus cells. Autoimmune reactions have been intensively studied as the possible cause of ulcerative colitis. The initial immune reaction is believed to produce sensitization of the colonic mucosa so that subsequent impact of the same autoimmune agent leads to destruction of the nmcosa. Infection established on this necrotized mucosa then perpetuates the colitis. No definite proof exists that this series of events causes ulcerative colitis. Furthermore, the circulating ,-mtibodies to the colon found in ulcerative colitis have been found in patients with diseases other than colitis. An excellent investigation of the immune basis for ulcerative colitis has been conducted by Joseph Kirsner, who states: "Present data do not demonstrate an immune basis for ulcerative colitis, though clinical and laboratory features continue to implicate hypersensitivity influences, perhaps as secondary factors. The colon is capable of operating a variety of immtmologic reactions, but attempts to produce unequivocally a n immune colitis by injection of antisera to animal colon remain inconclusive." The high titer reactions to milk protein in ulcerative colitis gives some support that sensitivity to milk proteins is one of the etiologic factors in the disease. The pathogenesis of ulcerative colitis is not known. The ideas that mechanical injury produced by hypersensitivity induced by reactions such as the Arthus', Shwartzman's and Auer's phenomena do not fully elucidate the pathologic events taking place in ulcerative colitis. Based on electron microscopy studies, Donnellan and Beal described early changes in ulcerative colitis as degeneration of the reticulin fibers of •



the colonic mucosa and dilatation of the subepithelial capillaries. This early process is followed by progressive lymphocytes and plasma cell infiltration of the lamina propria. It is postulated that the occlusion of small mucosal vessels by agglutinated platelets is responsible for the ulcerations of the mucosa. The infarction of the muco~a is followed by edema, polymorphonuclear neutrophil and eosinophili¢ i:avasion which completes the initial steps of a process which is to become a life-long inflammatory disease. The inflammatory cells form abscesses at the bases of crypts of Lieberkiihn. The inflammatory cells destroy the villi; the accumulation of pus lifts the mucosa, forming mucosal bridges. The developing granulations and a feeble attenipt of healing by fibrosis give rise to a pseudopolyp. The classic descriptions of the pathologic events, as important as they are, nevertheless remain incomplete until the agent or agents responsible for the initiation and perpetuation of the inflammatory process are known. The current problems in ulcerative colitis are not only our ignorance of its etiology and the reluctance to accept any system oil classification but other problems generally related to the management of the disease and specifically to the timing of surgical intervention in certain rapidly debilitating complications such as the acute dilatation of the colon in ulcerative colitis, severe hemorrhages and the end of tolerance of the chronic or recurrent disability by the patient. During the past 12 years, an interval of time between 1956 and 1968, 354 patients were operated on for inf/ammatory bowel disease at tile University of Chicago Hospitals. The surgical group was derived from a reservoir of 1,400 patients under observation by the medical gastroenterolo~, section of Joseph B. Kirsner. This medically-treated group consists of patients with chronic ulcerative colitis, granulomatous colitis (Crohn's disease of the colon), ileocolitis and regional enteritis. Most of these patients live satisfactory active lives under medical management ranging from modest dietary restriction to bland diet, sedation, periodic hospital observation, salicylazosulfapyridine (Azulfidine) and corticosteroid therapy. The group has achieved an impressive degree of numeric stability by admitting approximately 160 new patients per year. Such a yearly replacement more than compen, sates for numeric decrease created by departures from the geographic area and by an occasional death. The incidence of complications and sequelae requifi~lg surgical intervention calculated from the medically treated ulcerative colitis popu, lation is 23 %. Calculating the complications and sequelae from the number of patients (160) seen currently durra= an interval of i year, the incidence is 20%. In the total surgical group of 354 patients, there were 180 patients classified as ulcerative colitis and 174 patients as inflammatory disease of the small and large bowel lo~her than ulcerative colitis. This latter 15




Hemorrhage Pseudopolyposis Carcinoma Toxic megacolon Perforation and peritonitis Progression of disease Deaths

45 38 9 30 25 60 I0


25% 21% 5% 17% 13% 33% 5.5%

group included patients with granulomatous colitis (Crohn's disease), ileocolitis, regional enteritis, unclassified and mixed variety of inflammatory diseases of the colon. The group of 180 patients classified as having ulcerative colitis was analyzed with respect to complications and sequelae requiring surgical intervention. These Complications and sequelae are summarized in Table 2.

HEMORRHAGE Hemorrhage occurred in 45 patients (25 % ) and was observed most frequently in those patients who had acute dilatation of the colon or so-called toxic megacolon. In this group of patients, hemorrhages were severe (all requiring daily blood transfusions) and bleeding continued despite complete medical management, supported by nasogastric intuba,ion, parenteral fluids, including water, electrolytes, colloids, and blood transfusions. All but 2 patients with colonic bleeding requiring 500 ml. of blood daily to maintain stable hematocrit had one-stage total colectomy, resection of the rectum and permanent ileostomy. Most of these patients were operated on within a period of 6 days. Patients who, from the first bleeding episode, bled so severely as to require multiple blood transfusions in the course of 1 day, were oper,~ ated on as soon as hemodynamic stability was secured and adequate blood replacement was on hand. The 2 patients who had total colectomy:with ileoproctostomy suffered rebleeding from the rectum within 1 year. I n both cases the bleeding was severe enough to require resection of the rectum and creation of a permanent ileostomy. In patients with colonic bleeding not associated with toxic dilatation of the colon and perforation (peritonitis), the surgical mortality was zero. I n the granulomatous diseases, the hemorrhages were infrequent, not severe, and often subsided on medical management. 16

PSEUDOPOLYPOSIS Patients referred for operation because of extensive pseudopolyposis comprised 2 1 % - o f the entire surgical group of 180 patients. In this group the pseudopolyposis was not always the priml~r~ or the only reason for surgical intervention. The indications besieges pseudopolyposis were: hemorrhages, suspected c a r c i n o m a and progression of the colitis in spite of medical therapy. The older the disease, the more prevalent was the pseudopolyposis. No 'definite association was found between pseudopolyposis and colorectal carcinoma.


C a r c i n o m a (Fig. 8) was observed in 9 patients out of ~180, an incidence of 5 % . The rectum was the most c o m m o n site of the carcinoma. In 2 of the 9 patients ('Fable 3 ) , the carcinoma was muitiple, involving the rectum and the right colon. T h e r e were 7 colorectal carcinomas and 2 extracolonic neoplasms. T h e extracolonic carcinomas were found in the cxtrahepatic biliary ducts in 2 patients at necropsy. These patients had biliary cirrhosis as a sequel of long-standing ulcerative







23 yrs,




1. LI. colectomy 2. Proctectomy 3. Appendectomy Rt. colectomy 1. Rt, colectomy 2. Coloproctectomy T. colectomy T. colectomy + rectum T. colectomy + rectum T. colectomy + rectum T. colectomy + rectum T. colectomy + rectum

I yr.


I0 yrs. 8 yrs.

Alive Alive

I yr. 1 yr.

Died Died

I yr.


2 yrs.


4 yrs.


6 yrs.



I. Colon 2. Rectum

54 46


15 yrs, 26 yrs,

45 42


15 yrs, 27 yrs.

3. Appendix Cecum I. Colon 2. Rectum Bile duct Rectum



20 yrs.

Bile duct



20 yrs.



15 yrs.

Colon, multiple Colon



! 2 yrs.


*Note that survival is from date of diagnos."6 of the malignant lesion. Three out of the 7 patients (43%) w i t h colorectal carcinoma are surviving 5 or more years from date of coloproctectomy.





,<~" , "-


:':'." ~,

~,'i~ ~e,~'a~e.~,~,~r~-~


. . . .





Fla. 8. Carcinoma in lower sigmoid in a male, 60, who had history of quiescent ulcerative colitis for 23 years. The specimen is that of left hemicolectomy and shows a mucus-producing carcinoma. The resection terminated at the upper rectum. The carcinoma was 12 cm. fl'om the distal resected margin. Pathologic examination revealed metastases to 1 of 18 lymph nodes. Four months later the patient developed an infiltrating carcinoma of the lower rectum treated by combined abdominoperineal resection of the rectum and appendectomy. The pathologic examination showed ulcerative colitis, infiltrating adenocarcinoma of the rectum, ulcerative colitis of the appendix with metastatic adenocarcinoma of the appendix, periappendiceal abscess and metastatic carcinoma in the peritoneum. Patient survived less than 1 year from date of the first operation. Total coloproctectomy as the primary procedure might have kept the patient alive.

colitis. T h e colonic or rectal c a r c i n o m a s are best treated by total col~ctomy with resection of the rectum because of the possibility of simultaneous m u l t i p l e carcinomas or carcinomas arising d e n o v o in the residual colon or r e c t u m when those parts are not resected. The m a l i g n a n t process in the r e c t u m and colon arises in patients in w h o m the ulcerative colitis has been in existence for several years. The peak i n c i d e n c e is at a younger age level than the peak incidence in the g e n e r a l p o p u l a t i o n . T h e colorectal cancer is often multiple; 5-year survival is only 4 3 % .

T o x I c MEGACOLON T h e acute dilatation of the colon occurring in ulcerative colitis (Figs. 9 and 10) was the most d y n a m i c and the most fearful comFitcation encountered. T h e over-all g r o u p of patients with i n f l a m m a t o r y disease of the small and large bowel kept u n d e r medical observation at T h e University o f Chicago Clinics n u m b e r s a p p r o x i m a t e l y 1,400 patients. Only 30 patients were referred for operation because of toxic 18

FI~. 9 (above).--Plain roentgenogram of the abdomen is the simplest and the most reliable criterion in diagnosing acute dilatation of the colon in ulcerative colitis. Progressive abdominal distention, abdominal pain, fever and leukocytosis are the clinical signs which call for roentgenographic confirmation. Corticosteroid therapy may mask toxic symptoms and signs other than the dilatation. Note that the grealest dilatation is in the transverse colon and there is no sign of mechanical obstruction, Fx~. ]0 (below).--Gross specimen of:totalcoloproctectomy showing acute dilatation of the colon in a male, aged 27, who taad had recent ulcerative colitis. 19

megacolon during the interval of 10 years. Another 12 patients developed symptoms and signs of toxic megacolon showing dilatation of the transverse colon on roentgenography. In these, the complications disappeared within a few days without surgical intervention. The calculated probable incidence in a highly selected population of patients with inflammatory disease of the small and large bowel is 3%. The statistical data presented here indicate that toxic megacolon occurs only in ulcerative colitis. Since the ratio of ulcerative colitis patients to the number of patients with inflammatory disease other than ulcerative colitis is approximately 1:1, the actual incidence should be calculated only from the ulcerative colitis group. One may postulate that the 1,400 patients medically managed include approximately one half or 700 patients with true ulcerative colitis. This estimate is supported by the statistical analysis of 354 patients from this group who underwent surgical operations and in whom dcfinite diagnosis could be established by pathologic examination of the resected bowel. Therefore, the corrected incidence of toxic megacolon in ulcerative colitis is approximately 42 cases in 700 patients or 6%. In the 180 patients undergoing total colectomy or proctocolectomy for ulcerative colitis, there were 30 patients with toxic megacolon, an incidence of 15% in patients requiring surgical intervention. The analysis of the 30 patients is presented in Tables 4, 5 and 6. Patients with toxic megacolon and fecal peritonitis were treated by total colectomy and ileostomy. The rectum was not resected in order not to infect the uncontaminated presacral space and to limit the surgical trauma. All other ulcerative colitis patients had total one-stage colectomy with resection of the rectum and the construction of permanent ileostomy. Table 6 indicates that a significant reduction of postoperative mortality was achieved in those patients whose fecal peritonitis was irrigated with physiologic saline solution containing I Gin. of oxytctracycline per L. of the solution. The reduction of postoperative mortality .








- -









PREOV~Va'nVES~AT~ NO p e r f o r a t i o n o f c o l o n C o l o n i c p e r f o r a t i o n and peritonitis Steroid t h e r a p y Emergency operation Scheduled o p e r a t i o n s





11 27 12 18

37% 90% 40% 60%

*The incidence of colonic perforations decreased from I962 to 1968. The key diagnostic procedures are a plain roentgenogram of the abdomen in supine and, if the condition of the patient permits, a film in upright position.



I- 4 5-I0 11-15 16-23



17 4 3 6

57% 13% 10% 2-0%

years years years years

*The greatest incidence of acute dilatation of the colon in ulcerative colitis occurs in the early years of the disease. After 5 years, the incidence of toxic megacolon levels off, probably due to the removal of the fulminating cases either by surgical cures or by death of the patient. i



_ _

• .











from 83% to 20% appears signilicant. A~:tually, the 1 patient who died in the oxytetracycline lavage group did not die from peritonitis but from sudden exsanguinating gastric hemorrhage due io stress ulcer. The 5 deaths in the group not treated with oxytetracycline were due to generalized peritonitis and overwhelming sepsis. The peritoneal cavity which is heavily contaminated with liquid feces due to multiple perforations of acutely dilated colon must be thoroughly cleaned by removing all particulate matter. This may be gently accomplished by copious use of irrigating saline solution and progressive removal of the dihlted feces by suction. The addition of 1 Gin. of oxytetracycline to 1 L. of saline solution has been done on over 200 patients without a reaction. As much as 3 L. of the antibiotic-containing solution has been used to achieve complete debridement of the fecally-contaminated peritoneal cavity. In the final steps of the lavage, the peritoneal cavity is sucked dry, and no antibiotic is instilled into the peritoneal cavity. The abdominal incision is lavaged after closure of each layer from peritoneum to the subcutaneous tissue and skin. The abdominal wounds have healed per prima.


Toxic megacolon (all patients) With colon perforation and peritonitis Without perforation and without peritonitis Peritonitis not treated with oxytetracycline Peritonitis treated with oxytetracycline* Patients without toxic megacolon All ulcerative colitis patients



30 11 19 6 5 150 180

7 6 1 5 1 3 I0


(%) 23% 55% 5% 83% 20% 2% 5.5%

*Patients were treated with peritoneal oxytetracycline lavage, followed by intravenously administered oxytetracycline.


The histologic changes in acute dilatation of the c ~ o n were studied in 16 consecutive resected colon specimens. In all, the ganglion cells in the submucous plexus of Meissner and the myenteric plexus of Auerbach were found to be viable and in normal concentration. Destruction of the ganglion cells, therefore, is not the mechanism by which the dilatation is produced in toxic megacolon. The inflammatory process extends through the muscularis to the serosa causing dcstruction of the muscularis and'resulting in an inadequate peristalsis (Fig. I1). The data presented here relate to pure ulcerative c,.~litis carefully differentiated from other forms of inflammatory bowel disease. It is interesting'to compare these data to those often reported from European countries. A survey of the data analysis substantiates the impression that experiences with ulcerative colitis differ in different countries. In a recent editorial, Maratka writes: "Most American authors report that an emergency ileostomy is attended by a very high mortality, whereas subtotal colectomy is a safe curative procedure. However, in our experience, primary one-stage colectomy in debilitated cases resulted in such postoperative mortality that multiple-stage oFerations had to be resorted to." Analyzing this quotation, one must add that ~!iverting iicos-

FiG, 11.1.... ]In toxic megacolon the ganglion cells in the submucous plexus of Meissner and in the myenteric plexus of Auerbach appear not to have suffered a0 irreversible damage; some swell, others remain normal, In the low-power photomierograph the large ganglion cells are present and viable in an area of severe inflammation. 22

tomies are done rarely by American surgeons and, if resorted to under conditions of risk other than the colitis itself such as heart disease, etc., they are performed for severe debilitating ulcerative colitis with'out toxic dilatation of the colon. In toxic megacolon, ileos'tomies alone are not adequate to stop the progressive and rapid deterioration of the colon leading to multiple perforations and death from ~?eritonitis. In contrast t o the experiences in some European countries, t.otal colectomy for toxic megacolon in our hands has given a mortality of only 5% for patients in whom the complication was not associated with fecal peritonitis• Furthermore, the data presented here testify that one-stage total proctocolectomy performed on patients de:~perately ill from complications other than toxic megacolon resulted in surgical mortality of only 2%. This is less than the mortality in medically managed patients. The high surgical mortality ( 5 5 % ) occurred only in patients whose toxic megacolon was associated with generalized fecal peritonitis. These patients would have suffered a I00% imonality had they not had surgical intervention. Simple diverting ileostomies were done on 2 of our toxic megacolon patients. Four days later they developed multiple colonic perforations and died after total colectomy. Deaths were due to generalized peritonitis. Needless to say, the total colectomies were done at a time when the patients were more seriously deteriorated than they had been at the time of the initial ileostomies. When all the factors related to the different experiences with this disease in tlle various geographic areas are known, the factor of differential.ion of ulcerative colitis from gra~lulomatous colitis will play an impo~nant role. The author has tried to limit the analysis to a series of patients in whom the differentiation could be clearly established. Perhaps complete differentiation of ulcerative colitis from granulomatous ,colitis will never be possible. However, the stage of our presentday knowledge is such that we should be able to rule out clear-cut manifestations of one from the other. If this is done, then our scientific dialogue relative to these differences will become meaningful and diversities which now exist less obvious. Finally, one must be aware of the possibility that variations in severity and extent of ulcerative colitis really do exist in the several geographic areas. The reporting of complete cures following surgical procedures less than total proctocolectomy supports such a contention. Certairtly, the incidence varies with racial, natioiaal and geographic factors. The midcentral United States is a region with high incidence of ulcerative colitis. The disease seems to be more severe than t h a t reported from some European centers where successful cures are achieved with surgical procedures less extensive than those generally practiced in the United States.



CORTICOSTEROID-TREATED PATIENTS q'he use of corticosteroids as an essential part of the total medical management has become a universal pattern of therapy for ulcerative colitis. In some areas, ulcerative colitis patients receive as much as 400 nag. of hydrocortisone daily during periods of exacerbation of the inflammatory process. Such high doses are scaled down if the patient continues to improve, until a maintenance dose is established. The patient may be receiving oral corticosteroids for long periods, varying from a few months to 6 years. A dangerous situation attends the acutely ill patient who shows little or no response to the increasing doses of corticosteroids and who has to undergo a surgical operation at the height of the hormone therapy. Parenthetically, the intlammatory bowel diseases are of course not the only group of diseases in which corticosteroids are used for long periods. All these therapeutic trends have created a large group of patients sharing the common complication of medically induced hypercortisonism. The most common side effect observed in patients receiving exogenous corticosteroids for ulcerative colitis is the Cushingoid appearance: moon face, water-logged obesity, cutaneous striae and acne vulgaris (Fig. 12). These visible effects are coupled with hypertension, hyperglycemia, osteoporosis and electrolyte abnormalities mainly due to mobilization of proteins from muscle, bone and skin. The amino acids derived from protein catabolism are delivcrcd to the liver where they are deaminated and converted to glucose, hence the hyperglycemia and decreased protein synthesis. Surgical trauma alone induces catabolic events. Operation therefore magnifies the catabolic response in corticosteroid-treated patients. In addition, wound healing is delayed i n corticosteroid-treated patients. Even the most uneventful wound healing in patients having steroid therapy results in the formation of keloid. The keloid may be minimized by bridging the wound with adhesive straps after the skin sutures had been removed; other' wise the epithelial bond across the scar keeps tearing, and more and more fibrous reparative tissue is deposited, resulting in a thick keloid. The best clinical estimate is~that the wound healing on postoperative day 9 of a normal patient corresponds to day 20 of a corticosteroidtreated patient. Therefore, skin sutures, in these patients, should be removed about day 18. Some surgeons believe that, corticosteroid-treated patients are more vulnerable to venous thrombosis than are normal individuals subjected to similar surgical trauma. We have not observed a single case of venous thrombosis in 320 corticosteroid-treated patients undergoing major abdominal procedures. Nevertheless, the surgeon must be able to take proper measures against all the abnormalities induced by cor24

Fro. 12.---This 12-year-old girl received corticosteroids for 2 years. She had the typical moon face, buffalo neck, obesity, cutaneous stria and acne vulgaris. She underwent uneventful total coloproctectomy.

ticosteroid therapy. The surgeon must not only appreciate the side reaction induced by prolonged use of the hormone: he must also realize t h a t a sudden or recent interruption of corticosteroid therapy converts the patient into an acute or chronically smoldering state of adrenal insufficiency. H o w long an interval must elapse between cessation o f steroid therapy and the escape from hypoadrenalism effects under stress is not known. In adrenal deficiency the following abnormalities exist: 1 ) Decrease in serum concentration of sodium, chloride, and bicarbonate; 2) tendency to acidosis; 3) increase in serum potassium; 4) decrease in serum water; 5) decrease in plasma volume; 6) reduction of renal hemodynamics (filtration and renal flow are reduced) ; 25

7) impairment of water excretion (water intoxication) ; and 8) hypotension (impaired reaction to vasoconstrictor stinmli). Tile most important measure is to prepare the patient preoperatively with proper cortisone cover so that tile stress and strain of surgery does not produce circulatory collapse during the operation. Even though the patient may have received his last corticosteroids as far back as 2 years before scheduled operation, he still should be given cortisone before and during the operation. The corticosteroid-treated patients should receive 100 mg. of hydrocortisone intranmscularly the night before operation and 100 mg. of hydrocortisone sodium succinate (Solu-Cortef) intravenously as slow drip during the operation. The Solu-Cortef is continued during tile first 24 hours as a slow drip, so that the patient may receive as much as 300 mg. of Solu-Cortef intravenously during the first 24 hours. Thereafter, patients receive 50 rag. of hydrocortisone intramuscularly ever 3, 6 hours for approximately 10 days followed by oral prednisone, 10 rag. t.i.d.; the dosage then is slowly reduced over a period of approximately 2 months from date of the operation. Additional measures must be taken to overcome the problem of delayed wound healing. These steps consist of meticulous closure ot" the peritoneum with continuous chromic catgut and closure of the fascia with interrupted Mersilene or fine stainless steel wire sutures. Retention wire sutures placed over buttons and directed through skin, subcutaneous fat and fascia may also be used for added security in selected patients. The subcutaneous tissue is closed with interrupted plain catgut and the skin is quickly closed with continuous 000 nylon suture.


This study of the course, complications and tile fate of 180 patients operated on for ulcerative colitis lends itself to critical analysis of some of the existing problems in the hope that errors, if now committed, may be corrected. The question of timing surgical intervention in patients with ulcerative colitis has long been an unsettled problem. It is true that ulcerative colitis is a medical disease and its management should remain in the hands of gastroenterologists. The series of patients described here represents only 20% of the total "ulcerative colitis" population observed medically at The University of Chicago Hospitals and Clinics. Patients are channeled to surgery only because of the development of serious .complications and sequelae. The remaining 80% of patients apparently live satisfactory lives under constant medical management. It has 26

been pointed out that the large pool of patients with inflammatory bowel disease is not a stable or static "group; it is roughly stable only in its numeric aspect. The patients that are referred for operation come mainly from the yearly residue of 80%. The decrease in the total group, in tuna, is made up to its numerical stability by the flow of new patients. The data available from such a fluid grd~up cannot answer the question as to how many ulcerative colitis patiehts survive without surgical intervention and for how long. Yet the problem of timing surgical intervention is pre-eminent in the light of the considerations discussed below. PROLONGED CORTICOSTEROIDTHERAPY. The price of medical management with prolonged use of corticosteroids is not always without irreparable sequelae. Osteoporosis induced by exogenous corticosteroid therapy sooner or later results in compression fractures of the vertebrae. The induction of osteoporosis associated with compression fractures of the vertebrae in the growing individual exaggerates the deformity and contributes to the reduction of body height. There is no absolute proof that corticosteroids alone in ulcerative colitislare totally responsible for growth retardation. This hesitation is based only on the fact that a disease like ulcerative colitis is associated with chronic or recurrent nutritional deficits which, if sustained over the developmental period, can by themselves induce growth retardation. The evidence that corticosteroids are implicated in growth retardation is impressive. It has been shown that cortisone given to children in amounts exceeding 45 rag. per sq. M. of body surface per day.will inhibit growth. The suppression of skeletal growth may be interrupted by reducing the daily dose of cortisone below the critical level. When the dose of cortisone is reduced below the level which suppresses growth, there ensues a period of compensatory growth acceleration which, if long enough, will make up the developmental deficit. It is, therefore, important that children having ulcerative colitis who are treated with corticosteroids have the medication interrupted or reduced at intervals in order to provide periods for skeletal growth spurts to take place. It is obvious that no ulcerative colitis patient should be, managed with corticosteroid therapy if the price of escape from surgical intervention is an irreversible skeletal defect in the adult and growth retardation in the developing child. It is encouraging to note that these side effects of corticosteroid therapy are well supervised by gastroenterologists, and fewer and fewer patients are referred to surgery with irreversible skeletal changes. EVOLUTION OF C A R C I N O M A . - -In this series of 180 surgically treated patients, colorectal or biliary duct carcinoma did not develop until the disease was in at least its 12th year. The incidence in this surgical group was 5 % . The consensus of present-day writers is that the incidence of colorectal carcinoma arising in ulcerative colitis is several 27

times that of the incidence in the general population. The projected number of new cases of colorectal car~:inoma in the United States for 1967 is stated by the American Cancer Society to be in the neighborhood of 73,000. The estimated number of new cases of colorectal carcinoma in the United States is 40 per 100,000 population, or 0.04%. Goldgraber and Kirsner, writing in 1964 on cancer incidence, in ulcerative colitis and using the same over-all series as this monograph (the series then numbering 1,200 patients), reported a total of 33 cases or an incidence of 2 . 8 % . T h e s e estimates s u r e s t an overwhelming probability that carcinoma in ulcerative colitis evolves in the colon and rectum with frequency many times that found in the general population. Yet, the precise incidence of colorectal carcinoma in chror/ic ulcerative colitis remains speculative and uncertain. Welch and Hedberg, employing several statistical maneuvers, estimate that colorectal carcinoma in chronic ulcerative colitis is I0 times more common than in the general population. The data available in this study do not permit an accurate deduction as to the true incidence of cancer in ulcerative colitis. This uncertainty is refleEted in the differences in incidence as reported by several authors (Table 7). The striking differences in incidence (which ranged front 2% to 18.1% ) are not thff only disturbing observations. The 5-year survival rates after total coloproctectomy for cancer with ulcerative colitis are amazingly low as reported by several authors and illustrated in Table 8. The 43 % 5-year survival reported by the author is certainly disproportionate with the rest of the series.. This disproportion is, no doubt, due to the small number of cases (7). When the author's series is enlarged by the addition of all other cases in the same institution (Goldgraber's series of 23 patients), the 5-year survival is lowered to I 5 % representing roughly the 5-year mean survival rate quoted in Table 8. It is obvious that the incidence of carcinoma in patients with ulcerative

- -




i i i ii I


r i








Bacon, Barlo.w, IBerkley Dennis and Kai'lson Downing and Ferguson Edward and Truelove Goldgraber and Kirsner Lindner, King, Bolt Prohaska W6Ich and Hedberg Wheelock and Warren

1964 196I 1964 1964 1964 1960 1969 1965 1955

11.3% 11.3% 7.3% 3.5% 2.8 % 2.0% 5.0% 3.3% 18.1%















Bacon Bargen Dawson Dennis Garlock Goldgraber Welch Wheelock Prohaska

14 101 19 16 9 23 23 31 7


3 32 1 1 1 3 4 2 3


21.4 % 32.0% 5.7% 6.0% 11.0% 15.0 % 17.4% 6.0% 43.0%

*The series discussed in this monograph is too small to be a significant factor. I












colitis is much higher than that in the general population. The duration of ulcerative colitis is an important factor in establishing the'time limit beyond which the patient is increasi0gly more vulnerable to a development of colorectal carcinoma. After ulcerative colitis has existed f o r I0 years, awareness and actual diagnostic vigilance for colorectal cancer should become a routine part of periodic examinations. Unfortunately, the Symptoms of cancer are masked by the symptoms of colitis and diagnosis is therefore not easily made. Total coloproctectomy for uncomplicated chronic ulcerative colitis is possible with surgical mortality of less than 2%. It seems plausible, therefore, to recommend this procedure for all patients whose ulcerative colitis remains an inflammatory disease for 10 years or longer. SURGICAL I N T E R V E N T I O N I n TOXIC M E G A C O L O N . - - T h e timing of surgical intervention in acute dilatation of the colon must be directed toward prevention of colonic perforation with consequent fecal peritonitis. The eradication of the septic component in a debilitatingdisease reduces the surgical mortality from total colectomy to an acceptable level of 2%. This has been indicated in Table 6 (page 21) where the mortality of total colectomy for toxic megacolon with co'Ionic perforation and peritonitis is shown to be 5 5 % , whereas the surgical mortality for total coloproctectomy for toxic megacolon without fecal peritonitis is only 5%. These impressive statistics indicate that total coloproctectomy should be performed as soon as :the primary diagnosis of ulcerative colitis and the associated colonic dilatation are established. It must b e remembered that corticosteroids may mask the initial inflammatory impact o f feces spilling into the peritoneal cavity, delay the shock and mature and fix the linevitably fatal peritonitis. In this series there were 30 patients operated on for toxic megacolon and 29

12 patients recovered front the acute dilatation of the colon without surgical intervention. The number of deaths in the total group was 7 patients. If all the patients had had total coloproctectomy before coIonic perforation took place (assuming a surgical mortality of 5% ), only 2 patients would have died instead of 7. (See Table 6, p. 21 ). It was not always possible to exercise the ideal timing of surgical intervention because a number of desperately sick patients were admitted with colons that had already suffered perforation.


The major points of argument, strangely enough, are not concerned so much with the technic of the resection of the colon and rectum as with the construction of an ileostomy. The making of a proper ileostomy is therefore emphasized in this presentation. THE OPERATIVE PROCEDURE.---The patient is placed in the lithotomy position with the trunk of the body in a slight Trendelenburg incline. A soft roll of a laparotomy sheet is used to elevate the hips. Tile knees are slightly flexed over well-padded lithotomy braces. The feet should be slightly higher than the level of the chest to facilitate venous return from the lower extremities. The thigh and trunk ~.' the body should form a slight obtuse angle in order to provide adequate room at the abdomen for the surgical team. The abdomen, pubis, upper thighs, buttock and anus are then scrubbed with Septisol (2% hexachlorophene). Areas like the abdomen, harboring fewer microorganisms, should be scrubbed first or separately from the areas where bacteria are abundant, such as the buttock, pubis and anus. The abdomen, pubis, upper thighs, buttock and anus, in that order, are then prepared with thimerosal (Merthiolate). It is important that the Merthiolate be applied in the order mentioned to minimize the transfer of microorganisms from buttock areas to the abdomen. This is particularly essential when the ulcerative colitis is complicated by anorectal fistulas. A Foley urethral catheter is inserted and reflected over the left inguinal area where it is attached to a tube for continuous collection of urine. The abdomen, left side of the pubis, buttock and anus are draped in continuity. The lower pubis and buttock are then walled off by a separate towel. This may be removed at the completion of the abdominal dissection or~when the perineal dissection begins. The resection.--A midline in~sion is made from the pubis through the umbilicus to about 4 cm° below the xiphoid (Fig. 13). The midline incision saves time, preserves abdominal muscles and leaves the right and left lower quadrants free of scars for placement of the ileostomy. The abdominal viscera are quickly explored. Expioration of the peritoneal cavity is traumatic and, therefore, should be done expertly and 30

/ !

sf I

,.s~ f

.'2, ,¢

\ ,I

I ,

Fit;. 13.--The midline incision designed for total one-stage coloproctectomy passes from pubis through the umbilicus to 3-4 cm. below the xyphoid process. This incision does not destroy nerves or muscles, and leaves the r;ight and left lower quadrants without scars. The division of the umbilicus, which has been well cleaned with ethyl acetate, does not increase the hazards of wound infection. Wound disruption has not been observed in the most recent 100 consecutive surgical procedures using the midline, transumbilical incision.

in minimum time; it should be completed within 2 minutes. During this time, the surgeon must also direct his thoughts to the operative plan best applicable in the situation encountered. The surgical attitude should be one of removal of the entire colon with minimal trauma and with minimal blood loss. The cecum is mobilized upward, bringing with it the terminal ileum. This brings the lateral and the caudal peritoneal attachments of the cecum to tension. This peritoneal attachment is incised, and I~y blunt dissection a considerable portion of the cecum and terminal ileum, as well as the ascending colon, may quickly be mobilized. The dissection then proceeds along the lateral right peritoneal attachments up to the hepatic flexure. The ascending colon is covered on its anterior and lateral aspect by peritoneum. The posterior aspect of the ascending colon is connected to the abdominal wall only by loose areolar tissue which attaches the ascending colon to the iliacus and quadratu~ lumborum muscles. Decision must be made whether to preserve the greater omentum or remove it with the colon. In true ulcerative colitis the omentum is usually normal and, therefore, should be preserved. The tiny vascular attachments between the transverse colon and the greater omentum may usually be divided without clamps as the bleeding is negligible. An occasional bleeding point is ligated separately. The freed o m e n ~ m is placed into the upper abdominal cavity and observed af end of operation for vascularity. Omerttum which has suffered vascular damage should be excised because, subsequent necrotic changes will 31

generate adhesions with the possibility of intestinal obstruction. If the omentum is removed with the specimen (routine for granulomatous colitis), the surgeon then proceeds along the transverse colon; the lesser peritoneal sac is entered. At this point the entire cecum, ascending colon and at least half of the transverse colon are liberated so that the remaining attachment of this segment may be viewed from either side. This remaining attachment is essentially the transverse meso,colon. The ileocolic, the right colic and the middle colic vessels may now be visible under the serosal surfaces of the mesocolon (see Fig. 14 below). The right colic and the middle colic vessels are ligated near their points of origin and divided. The ileocolic artery, when it exists, is preserved. This gives us greater freedom in deciding how much terminal ileum is to be preserved. As much terminal ileum is preserved as is technically possible. By leaving the ileocolic artery and vein intact, one also preserves a greater portion of the mesoileum, which then is easier to mobilize in closing the space between the ileostomy and the lateral parietal peritoneum. The disengagement of the left half of the transverse colon continues, taking along the greater omemum. If the splenic flexure is situated relaFie. 14 (le/t).--The construction of, a permanent ileostomy requires cutting a well-designed pattern of the mesentery of the terminal ileum. The right colon has been mobilized from its peritoneal attachment and lifted from its bed. Note that the mesentery of the terminal ileum has been well preserved and shaped in a design to facilitate the anchoring of the ileum as it will emerge from the abdominal wall through a round opening. The remainder of the mesentery is utilized for closure of the space in the right peritoneal gutter and for peritonealization of the right colonic bed. Fro. 15 (right).---The ileum has been fixed by its mesentery to the parietal peritoneum, at the same time obliterating the right gutter space, The remainder of the raw colonic bed has been covered.


tively high, it is better to stop the dissection at this point and begin the dissection of the upper segment of the descending colon just beneatla the splenic flexure. Again, the lateral peritoneum prodfices a rather avascular attachment. It can be incised, and by scissor dissection, the splenic flexure is mobilized from below as well as from abcve. The phrenicocolic ligament, a peritoneal fold which attaches th~ splenic flexure to the diaphragm at about the tenth or eleventh rib, is divided between clamps and ligated. This ligament may be identified by noting that it forms a support for tile lower pole of the spleen. Once the splenic flexure is mobilized, the entire left upper colon may be reflected downward and to the right. The mesenteric vessels, including the inferior mesenteric artery and vein, are then progressively clamped, divided and ligated. The terminal ileum is divided between Kocher clamps; the division is approximately 4-5 cm. proximal to the entrance of the ileum into the cecum. The superior hemorrhoidal artery and vein are ligated, the pelvic peritoneum is incised and the posterior sacral hollow entered. From this point on the dissection continues close to the rectosigmoid and tile rectum. This ~s llot a cancer operation and does not require the cleardng out of the lymph nodes situated in tile pelvic hollow. By staying close to the rectosigmoid and the rectum, the surgeon avoids vesical disturbances, and in males avoids the creation of impotence in these often rather young individuals. Dissection of the posterior hollow of the sacrum should continue on to the levator muscles. With gentle traction, the anterior peritoneum, reflecting from the urinary bladder over the anterior surface of tile rectum, can be stretched and easily and safely incised. The anterior attachment of the rectum to the urinary bladder and the prostate or the cervix and posterior vaginal ~wall, as the case may be, is separated by blunt dissection. One often encounters a rather impressive lateral suspensory ligament of the rectum containing the middle l~emorrhoidal vessels. This should be clamped,-divided and securely tied. If it is torn by blunt dissection, the vein ot~ten tears off at its entrance to the internal lilac (hypogastric) vein. This causes troublesome bleeding, in a location equally difficult to reach from the abdominal or the perineal approach. At this point tile surgeon's assistant has closed the anus with strong braided silk purse-string. An elliptic incision is made around the closed anus; the posterolateral portion 6f this incision is then extended until the levator ant muscles are exposed. By blunt and sharp dissection, the rectum is freed from the levator ant muscles; the presacral space is entered. Blunt and sharp dissection then frees the rectum from its anterior attachment to the vagina or Prostate and urethra. The entire colorectal specimen is delivered, either from below or via the abdomen, whichever appears thesafer maneuver. 33

Closure o] the perineal floor.--While a member of the surgical team is closing the perineal floor by approximating the levator ani muscles from below, the surgeon reconstructs the peritoneal reflexion deep in the pelvic cavity from above. The perineal team, before closing the perineal incision with 00 interrupted chromic, places a ~ 2 2 Foley catheter into the presacral space. The balloon is inflated to 15 cc. with sterile saline. The catheter is kept on intermittent mild suction postoperatively for 4-8 days. The space previously occupied by the rectum is thus kept collapsed and primary, healing of the perineum is achieved in the majority of cases. In the past 3 years, not a single case of persistent perineal sinus was noted in approximately 80 such operations. The peritoneal gutters.--Most surgeons leave the peritoneal gutters open. In total colectomies and hemicolectomies, the patients have invariably done better when the lateral peritoneal gutters were closed. Closure is particularly feasible in total colectomy for chronic ulcerative colitis, since the resection of the peritoneum is not as radical as in operations performed for cancer. We close the right and left peritoneal ~ t t e r s with meticulous care, using 00 or 000 catgut on an atraumatic needle and placing the suture in a continuous fashion. These patients seem to have less postoperative pain and much less postoperative paralytic ileus, a sign of continuing peritoneal irritation, than those whose peritoneal gutters are left open. This procedure may also be responsible for our low incidence of postoperative bowel obstruction. R E L U C T A N C E OF PATIENTS TO ACCEPT A P E R M A N E N T ILEOSTOMY.

A survey of preoperative patients undergoing operation for inflammatory disease of the large and small bowel indicates that the most frequent concern expressed by the patient and the relatives is the fear of an abdominal stoma, the ileostomy. This fear is over and beyond the natural and normal resistance to the acceptance of such an abnormality. It appears to be generated by the reputation that the ~bdominal stoma is subject to many complications and troublesome experiences, requiring frequent revisions; and that it affects employability, social and marital life. The surgeon must take a convincing role in dispelling these fears by always striving to construct an ileostomy which this patient can handle and live with. The site of the permanent ileostomy should be designated preoperatively and the patient instructed about its nature and the reason for the abdominal stoma. The right lower quadrant is the most logical site for the stoma because of the location and direction of the terminal ileum (Fig. 13). A small fold of the skin is grasped with Kocher forceps, the skin lifted upward t o produce gentle steady tension and, with a scalpel, a circular area is excised. The circular area in the skin should measure approximately 2 ½ era.: in diameter but not more. The subcutaneous fat in this exposed area should not be excised as this would produce a 34

depression and retraction of the ileostomy. T h e s u b c u t a n e o u s fat is gently divided by cruciate incision to expose the anterior fascia over the rectus muscle. A n o t h e r cruciate incision is m a d e in the fascia, the m u s c l e fibers of the rectus are simply spread and the 13eritoneum then is opened. T h u s a r o u n d e d tunnel is m a d e through the entire a b c o m i nal wall; this tunnel should admit, snugly, two fingers. T h e terramal ileum is b r o u g h t through this tunnel to p r o t r u d e at least 5 cm. above the level of the skin. T h e opening that now exists b e t w e e n the lateral parietal p e r i t o n e u m a n d the mesoileum and the ileum itself is closed with c o n t i n u o u s c h r o m i c 00 suture. In patients who have been on corticosteroid t h e r a p y for tong periods, this should be reinforced by an occasional interrupted 00 silk suture. Corticosteroid-treated patients heal poorly and the delayed healing may cause disruption of the simple c h r o m i c suture. T h e terminal ileum, n o w extending through the o p e n i n g in ithe lateral a b d o m i n a l wall, is m a d e ready for the construction of the ileF16. 16.--The incision for the ileostomy is made in the right lower quadrant in an area free from scars, and preoperatively tested as a satisfactory location. The incision is a round opening 21/2 cm. in diameter, produced by lifting a fold of skin and by short sawlike movements excising a button of the skin. Usually, the cut skin presents itself with a slight beveled edge visible as a white ring of dermis. This is desirable because the sutures attaching the stoma pass through the dermis rather than through ihe entire thickness of the skin. This technic prevents the transplantation of mucosal cells through skin suture holes where they grow and form ugly weeping islands around the stoma. It is important not to dissect or injure the subcutaneous fat because this creates a depression for the stoma, making appliance fixation difficult. The subcutaneous fat, the fascia, the rectus muscle and the peritoneum are simply cut in 4 places in pie-crust fashion and spread to a round opening by finger.


/' ;"

/ / / ¢'

,\ "1\

,, \


E v e r t e d rl'l LI COS(3

d(l 3 hlner surface of g u t

- ;~'

FiG. 17.--Seromuscular cuff 21/2 cm. in length is peeled off from the terminal end of the ileum, as indicated in step I and 2. The mucosaI cuff is everted over the remaining protruding ileum which is about 2 ½ cm. in length. F o u r sutures of chromic catgut (000) ftx the everted mucosa to the seromuscular coat near the skirt edge. The sutures with their atraumatic needles are left uncut. When all 4 sutures have been placed and the eversion of the mucosa fixed, the uncut sutures are picked up,again. The needle now passes through the beveled dermal edge as shown in tl'ie round insert. Each suture is then tied fixing the stoma to the round cutaneous opening. Usually 2--4 simple sutures of 000 chromic catgut m a y be added after this p r i m a r y attachment. T h e completed ileostomy (3) is 2-3 cm. above the level o f the skin, is symmetrical and admits the surgeon's index finger. Note again that the anchoring sutures do not pass through the entire thickness of the skin.


ostomy. Any fatty attachments or parts of mesoileum which may cause curling of the protruding ileum should be removed. One should end up with approximately 5 cm. of freed, viable terminal ileum. A finger is then placed into the lumen of the ileum, and the seromuscular coat opposite the mesenteric attachment isiincised without perforating the mucosa. This incision should extend frbm the end of the stoma to approximately 3 cm. down. The seromuscular coat is easily peeled off in circumcision-like fashion. The final attachments of the seromuscular coat at the mesenteric border are removed with scissors. This then gives a protruding terminal ileum, the proximal 2 cm. of which has all the coats and tile distal 3 cm. only tile mucosa and submucosa. The mucosa is now everted and sutured:with fine 000 chromic catgut to the seromuscular coat below and this in turn is sutured to the edges of the skin. Approximately 8-10 such interrupted sutures are used. Tile sutures anchoring the ileostomy ,;hould not penetrate the skin because this causes transplantation of the mucosa with troublesome islands of mucosa growing from tile stitch holes. The suture should pass only through the dermis on the undersurface of the skin (Figs. 1 4 - I 7 ) . The final result is a stoma with everted mucosa so that no serosa at any point remains exposed. This ileostomy is referred to as the Turnbull type, from his original description in 1956. The Turnbull type of ileostomy is possible in about 40% of the patients. In the others, the seromuscular coat of the terminal ileum does not lend itself to easy removal. Under these circumstances, the entire thickness of the ileal wall is everted.

THE GRANULOMATOUS DISEASES OF THE LARGE AND SMALL BOWEL The granulomatous diseases of the large and small bowel include regional enteritis, ileocolitis and granulomatous colitis--Crohn's disease. When involving the colon, they are usually discussed under the general term of ulcerative colitis. In recent years, more and more data have been accumulated to indicate that the proportion of ulcerative colitis to the granulomatous diseases is 1 to 1. In the series presented here there were 180 patients with ulcerative colitis and 174 with granulomatous diseases. It is not known whether this increasing incidence is a real trend or whether it is due to increased effort and ability to differentiate one from the other. The principal surgical problems related to granulomatous disease are: 1. fistula of the intestine; 2. abscesses; 3. partial intestinalobstruction; and 4. recurrence after resection of involved bowel. 37


Regional enteritis lleocolitis Granulomatous colitis TOTAL Ulcerative colitis

9 . ~ F I s l ' u I _ A AND ABSCESS FORMATION*






40 I00

15 ........ 3 8 % 58 ...... 5 8 %

10 50

26 174

I0 .... 83-

10--~4()% 70. .40rF:




40% 47% 6.6%

25% ~ 50 :,c,

(). . . .




SURGICAL NIORI k l . l I Y "~-,

2~. 5% 1 6 ~ ~ - 16 %


2 ~ 20-~I

4f:k 2%

() 1.2%

4 ~


5.5 %

'~Major pathologic marks of progressive inflammalory, granulomalous process. N ot e the great differences between ulcerative colitis and the granulomatous diseases.

The distribution of the 174 patients in reference to the above complications is summarized in Table 9. It is not difficult to tabulate intestinal fistulas because all of the reseeted specimens are examined in the laboratory of surgical patholo~, and all specimens are photographed. Tile tabulations of abscesses may not be accurate because the definition of an abscess is often based on opinion only. Tile recurrences listed are those that required reoperation for serious symptoms and those with roentgenographic evidence of active disease, usually at the anastomotic site. The intestinal fistulas were enteroenteric, enterocolic, enterocutaneous, enterocavitary-cutaneous, enterovesical, enteroduodenai, perineal, including rectovaginal and multiple. Table 9 compares the relative incidence of the fistulas to fistulas found in true ulcerative colitis. It is not difficult to imagine a high rate of listula formation in the inflammatory diseases of the large and small bowel, if one considers the destruction of the mucosa, the increased fecal traffic and the chronieity of the disease. A fistula whether in true ulcerativecolitis or in the granulornatous variety (Crohn's syndrome), obviously indicates that a bowel perforation exists. It is evident from the clinical analysis herein presented that the two diseases, ulcerative colitis on the one hand and granulomatous enterocolitis on the other, pass through different pathologic events in setting the stage for perforation of the bowel wall. In ulcerative colitis the perforation is acute and leads to generalized fecal peritonitis. No internal fistula has been observed in 180 patients operated on for ulcerative colitis. Acute perforations with peritonitis occurred only in cases of acute dilatation of the colon (toxic megacolon). In regional enteritis and ileocolitis, the majority of fistulas were internal, that is, between loops of the intestine or from an intestinal segment to the abdominal wall. These fistulas were not associated with generalized peritonitis but often with an abscess between the loops of adherent bowel. In granulomatous disease, the process of fistula forma38

tion is a slow one and is preceded by initial adherence of segmenls of bowel and followed by eventual f, stulous comnmnication. The evolution of an .anorectal fistula in-a normal person is well known. It is initiated by a traumatic erosion of the mucosa o]' the anal crypt, giving rise to local inflammation, infection, abscess formation and drainage to ~ s i d e b.y a fistulous tract.iP,.'rineal fistulas found in ulcerative colitis are 6f the saffle kind as those tound in normal persons; the difference lies only in tl'~e high incidence found in patients with ulcerative colitis ( 6 . 6 % ) . This in'creased incidence is due to the existence of ulcerative'disease in the rectum a&gravated byfrequent traumatizing stools. in regit~nal et~teritis,, ileocolitis and granulomatous colitis, the incidence of pe'rineal fistula is signifidantly higher than .in ulcerative colitis. The high incidence of .perineal, anorectal-and rectovaginal, fistulas in ileocolitis has been noted .by several writers. Jn a series of pa'tients treated for ulcerative colitis and ilebeolitis, Cornes and Stecker report an incidence of 66% of perineal .and rect0vaginal fistulas found in Crohn's disease. In a study con~lucted by Davis, fistula-in-ano occurred in 28% of 141 patients with Cr0hn's disease,. A retrospective study by Barber, Waugh', Beahrs and Sauer reveals'that 29.6% of 257 patients with regional enterftis had internal or external fistula. It is logical to assume an~t to expect that anorectal fisiula in granulcmatous ileocolitis develops in an uninvolved re~ztum by the same mechanism as in normal per.sons. Such a fistula should respond to the usual surgical excision with the same degree of success as an ordinary tistula-in-ano, provided the defecation trauma is the same. However, this is not always the case. In many patients, the fistula tends to recur after seemingqy adequate surgical excision; occasionally it is made worse, and only infrequently does it heal after local fistulectomy. The reason for the chronicity and recurrence after operation for perineal fistula, particularly in ileocolitis, is the existence of an area of granulomatous involvement in the rectum. The author has followed from 6 to 14 years 8 patients whose initial diagnosis was ileocolitis. All of these patients had repeated roentgenograms showing normal left colons.They had multiple anorectal fistulas with otherwise normal rectums. The usual fistulectomies failed to heal the fistulous process. Following right hemicolectomy, at which time the left colon was found normal, the granulomatous process continued its downward spread, and in 1-4 years residual, proctocolectomy had to be performed. Skipped areas and patches of granulomatous disease feeding the persistent fistulas were found i n the resected specimen. This experience suggests an interesting speculation that granuloma, tous ileocolitis associated with progressive anorectal fistulization eventually ends up with a skip or total involvement of the residual colon and rectum. 39

The recurrence rate of 12% listed for all the granulomatous diseases of the small and large bowel in Table 9 (see page 38) is probably low because it does not include patients being followed for periodic abdominal distress,, without roentgenographic evidence of recurrent disease and whose symptoms, medically controlled, do not require surgical intervention. In 1967, the author reported a recurrence rate of 20% in 100 patients. It is possible that.the rate of recurrence has been reduced by emphasizing two very important factors. The first factor is an adequate resection of the bowel in both directions from its maximal involvement. The second factor is an immediate histologic examination of the proxima! and distal ends of the resected intestine. When inflammatory changes are found at the end of the resected bowel, an additional segment is resected and again examined histologically for inflammatory changes. Such a rapid histologic examination is not fully reliable but is better than gross examination alone. We must realize that the most common site of recurrence following resection is at the anastomosis. The circumstantial evidence, therefore, suggests that the trauma incident to the resection and the execution of the anastomosis somehow generates a recurrence at one end or both ends of the intestine brought together by anastomosis, or that unsuspected disease is still present at the anastomotic end. This concept would dictate a resection far away from the visible involvement, and the frequency with which the surgeon could resect beyond the involved segment of the bowel is enhanced by conducting a rapid histologic examination, for detection of inflammatory tissue reaction. It is not always possible to resect all the small intestine, the mesentery o f which contains enlarged and numerous lymph nodes. To do so would leave the patient a nutritional cripple for the rest of his life. There is no scientific evidence that the enlarged lymph nodes present in the mesentery belonging to ~ossly normal bowel are the forerunners of recurrent enteritis. Nevertheless, the present stage of our knowledge suggests that mesentery containing large lymph nodes should be included in the resection whenever possible, and whenever enough normal bowel remains for satisfactory function. Regional enteritis, o r enterocolitis associated with enteroenteric or enterocolic fistulas, is attackedsurgically in the same way. Adherent loops of bowel usually can be mobilized en masse and resected en bloc, so that. hea!thy-appearing normalends of bowel can be anastomosed end'to-end. The enterocutaneous fistulas require a special technic in entering the abdomen, The cutaneous opening has to b e carefully walled off. and the abdomen entered through a clean incision, the resection performed, a n d the resected specimen removed, together with the cutaneous fistulous opening. The operation can then be completed using a clean set of gowns, gloves and instruments. There isno good answer for the surgical treatment of regional enter40

itis associated with skipped areas• Often the whole l e n ~ h of the intestine is involved, with segments of normal bowel between areas of involvenlent. If the patient is operated on for intolerable complications such as an obstruction or enteroenteric or enterocutaneous fistula associated with skip areas, then one must select a cpnservative procedure to correct the intolerable situation and plan th~ :~peration in such a way as to make the anastomos~s between two laealtny ends of intestine. It is true that some skip areas will remain, but this cannot be helped. In the presence of skip areas that cannot be removed surgically, because to remove them would require the resection of the entire small bowel, the author prefers to be conservative in the extent of the resection• Conservatism in these patients is plausible because their rate of recurrence is high and the site of the next recurrence requiring operation is not predictable. Be that as it may, the fact is that the recurrence after a well-executed and seemingly curative resection remains very!high: indeed a tragic situation, for which there is neither an adequate explanation nor effective prevention. REFERENCES 1. Aylett, S. O.: Three hundred cases of diffuse ulcerative colitis treated by total colectomy and ileorectal anastomosis, Brit. M. J. 1:1001, I966. 2. Bacon, H. E., Bralow, S. P., and Berkley, J. L.: Rehabilitation and longterm survival after" colectomy for ulcerative colitis, J.A.M.A. 172: 324, 1960. 3. Bacon, H. E., and Pezzutti, J. E.: Granulomatous ileocolitis. Report of 61 cases, J.A.M.A. 198:1330, 1966. 4. Barber, K. W.. Jr.. Waugh, J. M., Beahrs, O. H., and Sauer, W. G.: Indications for and the results of the surgical treatment of regional enteritis, Ann. Surg. 156:472, 1962. 5. Barber, K. W., Jr., Waugh, J. M., Beahrs, O. H., and Sauer, W. G.: In retrospect---257 patients. Indications for and the results of the surgical treatment of regional enteritis, Ann. Surg. 156:472; 1962. 6. Belin. R. P., Walder, A. I., and Griffen, W. O.: Chronic ulcerative colitis complicated by toxic megacolon, Arch. Surg. 93:631, 1966. 7. BeZtarello, A., Martins-Campos, J. V., de Arauj~o, L. P., and Fernandes Pontes, J.: Ulcerative colitis: a clinical study on 95 patients, Arq. Gastroenterol. 4:93, 1967. 8. Billinghurst, J. R., and Welchman, J. M.: Idiopathic ulcerative colitis in the African, Brit. M. J. i :21 I, 1966. 9. Binder, H. J., Gryboski, J. D., Thayer, W. R., Jr., and Spiro, H. M.: Intolerance to milk in ulcerative colitis. A preliminary report, Am. J. Digest. Dis. 11:858, 1966. 10. Brooke, B. N.: Management of ileostomy, Lancet 2:102, 1952. 11. Brooke, B. N., and Chir, M.: Indication for emergency colectomy in ulcerative colitis, Dis. Colon & Rectum, 11 : 85, 1968. 12. Brown, J., and Pearson, C. M.: Clinical Uses o] Adrenal Steroids (New York: McGraw-Hill Book Co., Inc., 1962). 13. Cattell, R. B., and Boehme, E. J.: Importance of malignant degeneration as a complication of chronic ulcerative colitis, Gastroenterology 8:695, 1947.


14. Christeas, N., and Georgiades, N.: Total colectomy with ileo-rectal ulcerative colitis, Acta chir. hellen. 13:695, 1966. 15. Cornes, J. S., and Stecher, M.: Regional enteritis, regional ileitis, and ulcerative colitis comparison of case reports, primary Crohn's disease of the colon and rectum, Gut 2:189, 1961. 16. Crohn, B. B.: Regional Ileitis-Segmental Enteritis. In: Lewis" Practice o/ Surgery (Hagerstown, Md.: W. F. Prior Co., Inc., I963), rot. 6, p. 3. 17. Davis, J. M.: Large bowel complications of Crohn's disease of the small intestine, Proc. Roy. Soc. Med. 55: 704, I962. 18. de Dombal, F. T., Geffen, N., Darnborough, A., Watkinson, G., and Goligher, J. C.: Radiological appearances of ulcerative colitis: Evaluation of their clinical significance, Gut 9: 157, 1968. 19. de Dombal, F. T., Watts, J. McK., Watkinson, G., and Goligher. J. C.: Local complications of ulcerative colitis: stricture, pseudopolyposis, and carcinoma of colon and rectum, Brit. M. J. 1 : 1442, 1966. 20. Donnellan, W. L.: Early histological changes in ulcerative colitis, Gastroenterology 50: 519, 1966. 21. Edwards, F. C., and Truelove, S. C.: Course and prognosis of ulcerative colitis. Part III, Complications and Part IV, Carcinoma of the colon, Gut 5: 1, 1964. 22. Farmer, R. G., Hawk, W. A., and Turnbull, R. B., Jr.: Regional enteritis of the colon: Clinical and pathological comparison with ulcerative colitis, Am. J. Digest Dis. 13:501, 1968. 23. Feldman, F., Cantor, D., Soil, S., and Bachrach, W.: Psychiatric study of a consecutive series of 34 patients with ulcerative colitis, Brit. M. J. 3: 14. 1967. 24. Fennessy, J. J., Sparberg, M., and Kirsner, J. B.: Early roentgen manifestations of mild ulcerative colitis and proctitis, Radiology 87:848, 1966. 25. Fink, S., and Mais, R. M.: Cell-mediated immune reactions to autologous colon, J./_.ab. & Clin. Med. 70:868, I967. 26. Geffen, N., Darnborough, A., de Dombal, F. T., Watkinson, G., and Goligher, J. C.: Radiological signs of ulcerative colitis: Assessment of their reliability by means of observed variation studies, Gut 9:150. 1968. 27. Glotzer, D. J., Stone, P. A., and Patterson, J. F.: Prognosis after surgic:~l treatment of granulomatous colitis, New England J. Med. 277:273, i967. 28. Goldgraber, M. B., and Kirsner, J. B.: Carcinoma of colon complicating ulcerative colitis: Report of ten cases, Dis. Colon & Rectum 7:336, 1964. 29. Goligher, J. C.: Treatment o / C h r o n i c Ulcerative Colitis, CURRENT PROI3LEMS IN SURGERY (Chicago: Year Book Medical Publishers, Inc., August, 1965). 30. Goligher, J. C., de Dombal, F. "F., Graham, N. G., and Watkinson, G.: Early surgery in the management of severe ulcerative colitis, Brit. M. J. 3: 193, 1967. 3I. Gonzales~-Licea, A., and Yardley, J. H . : Nature of the tissue reaction in ulcerative colitis. Light and electron microscopic findings, Gastroenterology 51:825, 1966. 32. Gray, J. G;: Immunology and ulcerative colitis, Brit. J. Clin. Pract. 16:929, I962. 33. Greco, V., Lauro, G., Fabbrini, A., and Torsoli, A.: Histochemistry of the colonic epithelial mucins in normal subjects and in patients with ulcerative colitis: Qualitative and histophotometric investigation, Gut 8:49I, 1967. 34. Gump, F., and Lepore, M. J . : Regional enteritis. Prognosis. Prognosis in acute and chronic regional enteritis, Gastroenterology 39:694, 1960. 35. Hammarstrom, S., Lagercrantz, R., Perlmann, P., and Gustafsson, B. E.: Immunological studies in ulcerative colitis, II. "Colon" antigen and human 42

36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58.

blood group A- and H-like antigens in germfree rats, J. Exper. Med. 122: 1075, 1965. Hammer, B., Ashurst, P.. and Naish, J.: Diseases associated with ulcerative colitis and Crohn's disease, Gut 9:17, 1968. Harrison. M. J. G., and Truelove, S. C.: Cerebral venous thrombosis as a complication of ulcerative colitis, Am. J. Digest. Dis, 12: 1025, 1967. Hessler, C.: Surgery and the prevention of cancer, ~e,lvet. chir. acta 29: 358, 1962. Jacobs, E., and Elbaum, S.: Ulcerative colitis., case reports. Study of cz:ses of ulcero-hemorrhagic rectocolitis observed at the University Hospital SaintPierre in Bruxelles, Acta gastroenterol, belg. 24:470, 1961. Kaiser, C.: The liver-colitis syndrome, Gastroenterologia (Basel) 105:335, 1966. Karush. A.. Daniels, G. E., O'Connor, J. F., and Slern, L. O.: The response to psychotherapy in chronic ulcerative colitis. 1. Pretre~tmenl factors, Psychosom. Med. 30:255, 1968. Kirsner, J. B.: Immune mechanisms. Immunologic considerations in ulcerative colitis, S. Clin. North America 42: I 115, 1962. Kirsner. J. B.: The immunologic response of the colbn, J.A.M.A. 191"809, 1965. Kraft, S. C.. Bregman, E., and Kirsner, J. B.: Evaluation of autoimmunity. Criteria for evaluating autoimmune phenomena in ulcerative colitis, Gastroenterology 43:330, 1962. Kraft, S. C., et a/.: lmmunohistoctaemical studies of the colon in ulcerative colitis, Arch. Path. 82:369, 1966. I_ambling. A., and Bernier, J. J.: Is ulcerative colitis a disease caused by antibodies? Bull. Acad. nat. reed. 146:494. 1962. l_aurence, A. E.: Colectomy for ulcerative colitis with preservation of the rectum (Spanish). Bol. Soc. Cir. Buenos Aires 46:258, 1962. Lindner, A. E.. King. R. C.. and Bolt, R. J.: Chronic ulcerative colitis: Clinical appraisal and follow-up study. Gastroenterology 39: 153, 1960. Lindner, A. E.. and Marshak, R. H.: Clinical and radiologic features of ulcerative colitis, GP 37:111, 1968. Maratka, Z.: Recent s~udies on inflammatory intestinal diseases in Czechoslovakia. Am. J. Proclol. 19: 103, 1968. Maratka, Z.: Recent studies on ulcerative colitis in Czechoslovakia, Am. J. Gastroenterol. 45: 460, 1966. Maratka, Z.. Nedbal, J., Novak, J., Capek, V., and Chytilova, D.: Colectomy for ulcerative colitis with ileostomy or with ileorectal anastomosis? Am. J. Proctol. 19:138. 1968. Marder, L.: Symposium: Psychiatric aspects of ulcerative colitis, South. M. J. 60:1281, 1967. Marshak, R. H., and Lindner, A. E.: Ulcerative and granulomatous colitis, J. Mt. Sinai Hosp. 33:444, 1966. Matts, S. G. F.: Fulminating ulcerative colitis, combined steroid therapy for fulminating ulcerative colitis, Brit. M. J. 1:1045, 1962. McMillan, W. O., Jr., Garbutt, J. T., and Ruflin, J. M . : Problem of cancer in ulcerative colitis, South. M. J. 61:526, 1968. Mendeloff, A. I., Monk, M., Siegel, C. I., and Lilienfeld, A.: Some epidemiological features of ulcerative: colitis and regional enteritis. A preliminary report, Gastroenterology 51 : 748, 1966. Mistilis, S. P., Skyring, A. P.,, and Goulston, S. J. M.: Effect of long-term tetracycline therapy, therapy and colectomy in pericholangitis associated with ulcerative colitis, Australasian Ann. Med. 14:286, 1965. ,43

59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78.

79. 80. 81.

Mitschke, H., and Kracht, J.: Experimental studies on the immunopathology of the colon, Beitr. path. Anat. 136:261, 1968. Morson, B. C.: Inflammatory diseases of the large intestine, Arch. real. appar, digest. 51:714, 1962. Morson, B. C., and Pang, L. S. C.: Rectal biopsy as an aid to cancer control in ulcerative colitis, Gut 8:423, 1967. Mosbech, J.: Mortality in Denmark~ Mortality from ulcerative colitis in Denmark, Gastroenterology 39:690, 1960. Myers, R. T., and Hightower, F.: Critical. follow-up of surgically treated ulcerative colitis, Ann. Surg. 167: 920, 1968. Nedbal, J., Kudrmann, J., and Maratka, Z.: Liver dislurbances in ulcerative colitis: Effect of surgical treatment, Am. J. Proctol. 19: 130, 1968. Nedbal, J., and Maratka, Z.: Ulcerative proctocolitis in Czechoslovakia, Am. J. Proctol. 19: 106, 1968. Oppolzer, R.: Ileoanastomosis with interposition of an an!iperistaltic section of small intestine, following colectomy in a patient with ulcerative colitis, Chirurg. 37: 219, 1966. Perlmann, P., et al.: Antigen from colon of germfree rats and antibodies in human ulcerative colitis, Ann. New York Acad. Sc. 124:377, 1965. Polcak, J., and Skalova, M.: Immunologic manifestations of healthy consanguineous relatives of patients suffering from ulcerative colitis, Ant. J. Proctol. 19:197, 1968. Prohaska, J. V.: Development and fate of experimentally induced enteritis. Gastroenterology 5 ] :913, 1966. Prohaska, J. V., Greet, D., Jr., and Ryan, J. F.: Acute dil~ttation of the colon in ulcerr, tive colitis, Arch. Surg. 89:24, 1964. Prohaska, J. V., and Houttuin, E.: Enterocolic surgery based on presentday concepts of inflammatory diseases of the bowel, S. Clin. North America 47:157, 1967. Rappaport, H., Burgoyne, F. H., and Smetana, H. F.: The pathology of regional enteritis, Military Surgeon 109:463, 195 I. Sparberg, M., and Kirsner, J. B.: Corticosteroid therapy of ulcerative colitis: Status after 15 years, Current Therap. Res. 7:324, 1965. Tachev, T., Nedkova-Bratanova, N., Kras:eva. A., and Pelrounov, V.: Causes and frequency of milk intolerance in patients with ulceralive colitis, Acta gastroenterol, belg. 3l : 149, 1968. Tandon, B. N., Mathur, A. K., Mohapatra, L. N., Tandon, H. D., and Wig. K. L.: A study of the prevalence and clinical pattern of non-specific ulcerative colitis in Northern India, Gut 6:448, 1965. Truelove, S. C.: Medical management of ulcerative colitis, Brit. M. J. 2:605, 1968. Turnbull, R. B., Jr.: Colectomy with simultaneous ileostomy as the surgical treatment of diffuse ulcerative colitis, Surgery 4:843, 1957. Vaides-Dapena, A. M.: Special Features of Cancer of the Colon following Ulcerative Colitis, Blakemore, W. S. and Ferguson, L. K. (eds.): Current Perspectives in Gastroenterology (Current Perspectives in Surgery) (New York: Hoebe+r Med. Div. Harper and Row, 1967), pp. 59-65. Van Der Zypen, E.: Light and electron microscopic findings in the autonomic nervous system of the colon in ulcerative colitis in man, Deutsch. Ztschr. Nervenh. 187:787, 1965. Welch, C. E., and Hedberg, S. E.: Colonic cancer in ulcerative colitis and idiopathic colonic cancer, J+A.M.A. 191:815, 1965. Wheelock, F. C., and Warren, R.: Ulcerative colitis, New England J. Med. 252:421, 1955.