- Email: [email protected]
Diverticular Disease of the Colon: Current Concepts and Management
Symposium on Surgical Practices at the Lahey Clinic II
Diverticular Disease of the Colon Current Concepts and Management
Alan W. H ackford, M.D., * and Malcolm C. Veidenheimer, M.D.t
Diverticulitis appears to be a disease of modern Western civilization with an increasing incidence as dietary habits have become more refined and as life expectancy is extended. It was briefly described by Littre 32 at the beginning of the eighteenth century, but Cruveilhierl7 is credited with the first in-depth description of diverticula of the colon as a pathologic entity in 1849. Another half century elapsed before diverticulitis was described by Graser25 as a clinical entity, and it was not until the beginning of the twentieth century that the progression of diverticulitis leading to perforation and fistula was described by Beer. 5 Sexual predilection for diverticular disease has changed. Early in the century the disease was reported in two or three men for every affected woman. 24, 59 In the 1950s no difference in incidence between men and women was noted. 6,29 Recently a female predominance (three to two) has been described. 47 With the increase in the aged population the risk of the development of diverticula, once a pathologic curiosity, is now estimated to exceed 50 per cent. In large series 50 of radiographic and postmortem studies the overall incidence of diverticulosis is between 5 and 6 per cent-5.7 per cent in more than 24,000 barium enema studies and 5.2 per cent in almost 2000 autopsies. Without doubt the incidence increases with age and usually occurs after the age of 40 years. Parks 47,48 estimated an incidence of 5 per cent in the fifth decade of life, rising to 50 per cent in the ninth decade. Welch et al. 64 found that two thirds of patients had diverticula by age 85. The incidence of diverticulitis also increases with the age of the patient. Allen and Behringerl estimated that the incidence of specific radiologic evidence for diverticulitis in patients with diverticulosis will rise from 20 per cent in the fifth decade of life to 33 per cent in the ninth decade, In Horner's study29 of 503 patients with diverticulosis the incidence of di-
*Currently, Section of Colon and Rectal Surgery, Department of Surgery, New England Medical Center Hospitals, Boston, Massachusetts tChairman, Department of Surgery, Lalley Clinic Medical Center, Burlington, Massachusetts
Surgical Clinics of North America-Vol. 65, No.2, April 1985
347
348
ALAN
W.
HACKFORD AND MALCOLM
C.
VEIDENHEIMER
verticulitis was about 10 per cent after 5 years, 25 per cent after 10 years, and greater than 35 per cent at 20 years. Close to 5 per cent of patients were less than 40 years of age.23 In this group, however, diverticular disease tended to be an aggressive form that is more symptomatic and has a high rate of complications. Hannan et al.28 reported in 1961 that almost one half of their 23 patients less than 45 years of age required operation for complications after the first and only attack of diverticulitis. These published reports are in uniform agreement that the sigmoid colon alone is involved in almost 50 per cent of patients. In 147 patients radiographically studied at the Lahey Clinic61 55 per cent showed involvement of the sigmoid alone (Table 1). In 95 per cent of patients reported in the literature29 , 33, 47, 48 the sigmoid is involved either alone or in conjunction with other areas of the colon, The usual impression is that progression of the disease is within the affected segment rather than by spread to other segments, As a result, younger patients with severe localized disease requiring resection will infrequently need subsequent resection for recurrent disease. Of 147 patients undergoing elective resection between 1957 and 1975 at the Lahey Clinic61 for the removal of only the clinically involved portion of the bowel, no patient has yet required a second operation, PATHOGENESIS Current concepts about the development of diverticula are based on two principles: the first is a pressure gradient from lumen to serosa, and the second is a relative weakness in the bowel wall. Morson 41 found disease in every one of 155 operative specimens with a clinical and radiographic diagnosis of diverticulitis, He described what was thought to be an abnormality of muscle: the teniae assumed a cartilaginous consistency, and the circular muscle was thickened causing the bowel wall to have a concertinalike appearance, In the absence of histologic abnormalities Morson 41 considered the muscle thickening to be secondary to shortening rather than to hyperplasia or hypertrophy, Shortening would also explain the redundant folds of mucosa found in diverticular disease, Morson42 , 43 recently confirmed his inability to find histologically abnormal muscle, and he has further described a marked increase in elastic tissue specific to teniae, This teniae-specific elastosis causes longitudinal foreshortening of the colon, ac-
Table 1.
Radiographic Extent of Disease in 147 Patients
EXTENT
NUMBER OF PATIENTS
Sigmoid colon only Sigmoid and descending colon Sigmoid, descending, and transverse colon Pan diverticular disease
81 48 8 10
DIVERTICULAR DISEASE
349
centuating the semicircular corrugations of the shortened circular muscle (Fig. 1). Up to one third of the operative specimens did not have inflammation but showed the muscular thickening described, an abnormality that may precede the actual development of diverticula. Using a combination of perfusion manometry and cineradiography, Painter44-46 demonstrated that waves of high pressure coincide with bandlike contractions, which occlude short segments of bowel and form closed cells in which extraordinarily high pressure can be generated (Fig. 2). Almy2 suggested in 1965 that this abnormality and narrowing of the lumen afforded a straightforward explanation for the predominance of the sigmoid colon in this disorder. Based on the law of Laplace, in which tension in the wall of a cylinder is directly proportional to pressure within the cylinder multiplied by the radius of the cylinder (and inversely proportional to wall thickness), the intraluminal pressure will be greatest where the lumen is narrowest and the wall is thickest. This is in the sigmoid colon in the normal individual. An additional motility disorder has been suggested by more recent myoelectric studies. 58 Asymptomatic patients demonstrate normal motility patterns, but patients with symptomatic diverticular disease have an abnormal slow-wave pattern with a predominant frequency of between 12 and 18 cycles per minute. This is not the pattern seen with irritable bowel syndrome, which shows increased basal electrical rhythms of 3 cycles per minute. 57 A further distinction between diverticular disease and irritable colon is that the abnormal myoelectric activity reverts toward normal when bran is given to patients with diverticular disease, but it does not do so in patients with irritable bowel syndrome. No evidence was found for a causal relationship between the two conditions in a number of clinical surveys.21. 4748 A noteworthy point made by Almy and Howell,3 however, is that since patients with asymptomatic diverticulosis demonstrate no mo-
Figure 1. Macroscopic appearance of diverticular disease with foreshortening and accordionization of the colon.
ALAN W. HACKFORD AND MALCOLM C. VEIDENHEIMER
350
I
it
+.
'."
.
Figure 2.
High-pressure cell formation.
'"
+•
tility changes with dietary management, the effectiveness of such management in preventing serious complications remains uncertain. A weakness in the wall of the intestine is the second major factor in formation of diverticula. Extensive anatomic dissection by Slack54 suggested that diverticula may occur at four sites between the mesenteric and antimesenteric teniae at points of penetration of intramural blood vessels (Fig. 3). Because this is a mucosal herniation these are, by definition, pseudodiverticula. PATHOLOGY OF DIVERTICULITIS
In simple diverticulosis the mucosal sac freely communicates with the lumen of the bowel. Increasing intraluminal pressure may force a fecal mass to become inspissated at the neck and cause obstruction. Subsequent inflammation begins at the apex and spreads to adjacent tissue, either mes-
DIVERTICULAR DISEASE
Figure 3. formation.
351
Sites of diverticula
enteric or pericolic fat, producing a peridiverticulitis. 42 Abscess may occur and extend deep into the mesentery, to adjacent structures, or along the longitudinal muscle layer, rupturing back into the colon to form intramural fistulas.
DIAGNOSIS
As has already been suggested, the diagnosis of diverticulitis carries an appreciable rate of error. Morson42 reported that one third of operative specimens of patients with diverticulitis showed no evidence of inflammation. In a postmortem study Hughes 30 found a converse rate of error with 12 per cent of patients who gave no history of an inflammatory problem having evidence of diverticulitis. Clearly, spasm of the sigmoid with highpressure segmentation can produce pain in the left lower quadrant and occasionally a mass in the left lower quadrant that can mimic an inflammatory process. Reliance on indications of inflammation, such as leukocyte count and an elevated temperature, may also cause error. A recent study62 from Sweden found that leukocyte count and sedimentation rate were of little value in the evaluation of acute diverticulitis. Of 130 patients at the Lahey Clinic who had acute complications of diverticular disease at operation 64 per cent had a leukocyte count within a normal range of 5000 to 12,000. 27
ALAN W. HACKFORD AND MALCOLM C. VEIDENHEIMER
352
The barium enema study has been considered the prime diagnostic modality for the evaluation of diverticular disease. We usually obtain this study 1 to 10 days after the process of acute inflammation subsides with conservative therapy. Diverticula are easily identified, but inflammation is not as evident. In patients with considerable muscular abnormality and pleating of the mucosa an irregular sawtooth pattern will develop. This has been called the prediverticular state and does not necessarily reflect inflammation, but it may be the only finding in patients with severe clinical disease. On the rare occasion when only one diverticulum is present, the results of barium enema study may be normal. Only radiologic demonstration of an abscess cavity or intramural or extramural fistulization can be taken as a certain sign of diverticulitis (Fig. 4). Of 105 contrast studies performed at the Lahey Clinic for acute disease27 only one third demonstrated these specific findings. In up to 25 per cent of the patients the radiographic evaluation raised the possibility of a carcinoma because of obstruction or stricture. Marshak et al. 36 stated that stricture secondary to diverticulitis tends to be a long segment with normal-appearing mucosa, whereas that secondary to carcinoma tends to be a short segment with overhanging edges and ulceration of the mucosa. At times, however, the distinction cannot be made even with the aid of a colonoscope, and a resection is required. AI-
Figure 4.
Contrast study demonstrating perforation and abscess formation.
353
DIVERTICULAR DISEASE
though the diagnosis of diverticulitis is determined clinically, taking into consideration radiographic and laboratory data, it is based principally on the clinical findings of left lower quadrant tenderness sometimes with localizing peritoneal signs and sometimes with the presence of a mass.
MEDICAL MANAGEMENT Not infrequently, particularly in patients less than 40 years of age, the first attack of diverticulitis may result in serious and potentially fatal complications. A period of observation and conservative management is necessary to determine the outcome of a particular attack. Colcock12 estimated that 45 per cent of these patients have recurrent problems. Parks 47 found that a similar number of patients (40 per cent) continue to have symptoms despite medical management. The risk of complication rises with recurrent attacks. Marshall37 reported that complications developed in 22 per cent of his patients with a single attack of diverticulitis, and this incidence rose to almost 60 per cent in patients having two or more attacks. Parks 47 demonstrated that patients whose diverticular disease involves the whole colon do not require operation any more frequently than those with involvement of the sigmoid colon only. Approximately 25 per cent of the latter group required surgical procedures. In the series of Jordan and Boles 31 16 of 73 patients (24 per cent) required operation. In the conservative management of diverticulitis, bowel rest, intravenous fluids, and broad-spectrum antibiotics form the mainstay of therapy. Intraluminal antibiotics are of little use because the inflammatory process is pericolic rather than intraluminal. Almy and Howell3 caution that "considering the frequency of spontaneous recovery, the routine use of antibiotics in diverticulitis generates unjustified confidence in their efficacy." Certainly, if these agents are used, the physician must constantly make clinical judgments in monitoring the progression of an acute attack. If improvement is not seen within 48 hours, serious consideration should be given to operative intervention. As mentioned previously, electromyographic data in patients with acute diverticulitis support the concept of hypermotility. This has provided the rationale for the common use of anticholinergic drugs, but their efficacy in the acute setting remains unsubstantiated. Glucagon may also be helpful as a smooth muscle relaxant. However, the source of pain in patients who improve following this type of regimen is more likely to be sigmoid muscle spasm than an inflammatory process. The use of morphine in this setting is contraindicated because of the well-document~d increases in intraluminal pressure after its administration. Fiber and osmotic bulking agents have little place in the management of acute diverticulitis. The normalization of abnormal electromyographic patterns is only of theoretical importance at this point. The assumption that dietary fiber will ameliorate the pain and bowel dysfunction of diverticular disease is based on several poorly controlled trials. From a pathophysiologic standpoint, however, after the acute attack has settled, in-
354
ALAN W. HACKFORD AND MALCOLM C. VEIDENHEIMER
crease of fecal bulk leads to an expansion of sigmoid diameter and to a reduction in intraluminal pressure (according to the law of Laplace), thus theoretically conferring some protection against fecal inspissation.
SURGICAL MANAGEMENT In 1907 William Maydl8 presented to the American Surgical Association a review of the literature on 19 patients with diverticulitis and described five of his own patients. At that time he advocated primary resection. In 1942 Smithwick55 stated that treatment with colostomy and drainage was insufficient. In 42 patients he found a 5 per cent operative mortality rate but a subsequent late mortality rate of 12 per cent. Furthermore, the majority of his patients showed little improvement clinically. Accordingly, he developed a three-stage approach that resulted in a 6 per cent operative mortality rate but no late deaths and a good clinical course in 82 per cent of the patients. He concluded that the optimum treatment was a multistage approach that included removal of the diseased segment at the second stage. By 1950 support for an elective surgical approach was inevitable because of the availability of antibiotics. In 1953 Welch et al.,64 reporting on 114 resections of the colon, stated that uncomplicated diverticulitis should be considered an important. indication for operation. In papers from the Lahey Clinic, Colcockl 2-15 presented an aggressive surgical approach for diverticular disease that had a low surgical mortality rate. Review of experience at the Lahey Clinic in a recent series60 of 133 patients undergoing elective operation showed no deaths and complications in only 10 patients. Few would now take issue with the concept of elective primary resection and anastomosis for uncomplicated disease. This is not necessarily true in patients in whom complications occur, however. In 1960 Smithwick56 continued to advocate the three-stage approach. In comparing two surgical series approximately 8 years apart, Botsford et al. 7 noted that as primary resection became the preferred treatment the surgical mortality rate of patients increased. Yet, 1 year later Byrne and Garick, 8 found a 26 per cent mortality rate with the three-stage approach when carried out for an abscess and a 30 per cent mortality rate in the presence of peritonitis. This kind of data prompted many surgeons to argue that such patients were too sick to tolerate anything other than primary resection. 34 Those opposed to this concept pointed to the high incidence of anastomotic problems, which led to the application of the Hartmann operation with closure of the rectum and end sigmoid colostomy as the procedure of choice for complicated disease. In a recent series reported by Eisenstat et al. 22 44 patients with complicated diverticulitis were treated by a Hartmann closure resulting in a mortality rate of only 4.5 per cent. In 1969 Dandekar and McCann 18 reviewed 26 patients operated on with diffuse peritonitis. Operations were staged in 19 patients with a morbidity rate of 53 per cent and a mortality rate of 21 per cent after the first stage alone, and 7 patients underwent primary resection with a morbidity
DIVERTICULAR DISEASE
355
rate of 33 per cent and a mortality rate of 14 per cent. They were also among the first to comment on length of hospitalization (52 days) and disability (26 weeks) for the staged approach as compared with that of the group undergoing primary resection (25 days and 7 weeks, respectively). Rodkey and Welch52 in 1974 found no essential difference in morbidity and mortality rates between the use of primary resection and the staged approach in patients who had localized peritonitis with abscess formation. However, they did find a similar disparity between duration of hospitalization and disability with 50 days and 32 weeks, respectively, for the staged approach and 25 days and 5 weeks, respectively, for primary resection. Greif et al. 26 presented a review of world medical literature in 1980 in which they identified 1363 cases of diverticulitis. Of 510 patients with perforation, abscess formation, and localized peritonitis, 204 underwent resection with a mortality rate of 2 per cent, and 306 were treated with colostomy with a mortality rate of 12 per cent. Of 833 patients with diffuse peritonitis, 316 underwent resection with a mortality rate of 12 per cent, and 505 treated with proximal colostomy incurred a mortality rate of 29 per cent. Greif et al. 26 concluded that the results were considerably better if the operative approach for acute diverticulitis included resection. Previously from the same institution McSherry et al. 39 had compared 150 staged patients with 50 patients treated with resection and found that the staged group had it mortality rate of 6 per cent and a morbidity rate of 31 per cent as opposed to the primary resection group, which had a mortality rate of 18 per cent and a morbidity rate of 46 per cent. They concluded that the problem with the emphasis on primary resection was that it was not the surgeon but the patient who was at risk. Classen et al. ll pointed out that in the last decade the same factors that have improved the survival of patients undergoing primary resection, such as the great advances in intensive care and in antibiotic therapy, have also affected the survival of patients treated according to the three-stage procedure. They cited an overall mortality rate of 11 per cent against a historical rate of 21 to 45 per cent for the three-stage approach, which represents the best data in the literature. Classen et al.l! support the use of primary resection when feasible. However, in a recent study from Scandinavia by Wara et al. 63 83 patients with complications were examined, and of 25 patients with generalized peritonitis who were managed with proximal colostomy alone, 11 died. This was contrasted with a mortality rate of 8 per cent for patients who were treated by an emergency resection. In the three-stage group only 58 per cent went on to the second stage of treatment, and only 46 per cent had eventual restoration of intestinal continuity. We have recently reviewed the Lahey Clinic experience with surgical management of complicated diverticulitis between 1967 and 1982 (Table 2).27 Of 140 patients with peritonitis, phlegmon, or formation of abscess or fistula, 86 were treated with primary resection and anastomosis. An additional 13 patients underwent formation of a proximal stoma. Evaluation of the pathologic findings in these two groups showed that abscess was most frequently (46 per cent) associated with a proximal stoma. Phlegmon alone
356
ALAN
Table 2.
W.
HACKFORD AND MALCOLM
Primary resection (86 patients) Primary resection and proximal colostomy (13 patients) Stage 1 Stage 2 Primary resection with Hartmann closure or mucous fistula formation (19 patients) Stage 1 Stage 2 Three-stage procedure (22 patients) Stage 1 Stage 2 Stage 3
VEIDENHEIMER
Lahey Clinic Experience, 1967-1982 MAJOR COMPLICATIONS
DEATHS
PROCEDURE
C.
NUMBER OF OPERATIONS
Number
Mortality Rate (%)
Number
Morbidity Rate (%)
86
(14)*
1
1
18
18
32t 13 13
0 0 0
0 0 0
7
(6)
4 3
22:j: 21 23
30t 19 10
(1)
3 3 0
16:j: 16 0
7 7 0
23:j: 35 0
3 1 2 0
14:j: 5 10 0
14 8 5 1
24:j: 31 25 8
59t 22 20 13
(4)
*Numbers in parentheses indicate additional operations. tTotal number. :j:Cumulative rate.
was present in 42 per cent of patients without a stoma. Formation of a proximal stoma was also utilized in 8 per cent of patients with localized perforation and in 16 per cent of patients with peritonitis. A Hartmann procedure was performed in 19 patients. Both abscess and phlegmon were present in 80 per cent of these patients, and 32 per cent had peritonitis. A three-stage approach was used in 22 patients with a pathologic profile similar to that of the group undergoing the Hartmann operation. Pertinent to the decision to perform anastomosis was the ability to achieve bowel preparation. This was accomplished in 96 per cent of the group who underwent primary resection, in 70 per cent of patients who had a proximal stoma, and in only 30 to 40 per cent of patients who underwent a Hartmann operation or the three-stage approach. Morbidity and mortality rates were examined for each group (see Table 2). The single-stage group had a 1 per cent mortality rate and an 18 per cent morbidity rate. Patients who underwent first and second stages consisting of resection, anastomosis, and creation of a proximal stoma had a 22 per cent morbidity rate and no deaths. The Hartmann group had a 16 per cent mortality rate and a 35 per cent morbidity rate in the first stage. Since no additional (second-stage) complications occurred, the cumulative morbidity rate was 23 per cent. These results are similar to those seen with the three-stage approach, which yielded 14 per cent mortality and 24 per cent morbidity rates overall.
357
DIVERTICULAR DISEASE
An enormous amount of effort has gone into the appraisal of surgical management of diverticulitis. What May038 said in 1907 remains true today: If considerable disease is present and the symptoms do not disappear, primary resection of the affected part of the bowel is indicated before abscess and fistula supervene. In fact, the selection of patients who are at risk for complications is probably the best way to reduce the morbidity rate of this disease. We have modified very little the indications developed by Colcock15 over 10 years ago (Table 3). A number of operations are available for the management of diverticular disease (Table 4). Use of the EEA (end-to-end anastomosis) stapler (United States Surgical Corporation, Stamford, CT) has substantially simplified the second stage of the Hartmann operation and undoubtedly has shortened the length of hospitalization during the second stage. The Mikulicz exteriorization of mucous fistula formation is difficult to perform because the inflammation and foreshortening of the bowel mesentery make delivery of bowel without tension to the abdominal wall a challenge. Our data and that of a number of other series 18, 26, 34, 51, 52, 63 dispute the theory that operations performed in multiple stages offer the patient the best chance for survival. The classic first stage of colostomy and drainage may leave the patient with a hole in the sigmoid colon and a column of stool on the left side. In this situation persistent sepsis is a predictable sequela. Smithwick56 demonstrated in 1960 that 25 per cent of patients with diverticulitis showed evidence of active postoperative inflammation despite construction of a proximal colostomy. In a series presented by Dawson et aI., 20 even after a 2-week period of time 11 per cent of patients treated by the staged approach with an initial proximal colostomy had persistent fever, rigor, and wound drainage-ample evidence of ongoing sepsis. If we accept the theory of Painter et al. 44--46 of compartmentalization of the sigmoid with high-pressure cells, it is difficult to envision any type of colostomy alone being able to vent the pressure that drives stool through a perforated diverticulum. In fact, recurrent diverticulitis in the defunctionalized bowel has been described by several authors. 1g, 35, 53 Diverting colostomy as a treatment for colovesical fistula was popularized by Cripps16 in 1888, a full decade before diverticulitis was even described as a clinical entity. In the 1940s more complicated procedures involving proximal colostomy were popular. Yet, as pointed out by Carpenter et al. g, 10 colostomy does not allow for closure of the fistula.
Table 3.
Indications for Elective Resection
Two or more attacks ofleft lower quadrant pain associated With fever, leukocytosis, and radiographic evidence of diverticulitis One attack associated with radiographic or clinical evidence of perforation One attack associated With radiographic or clinical evidence of obstruction One attack associated with urinary tract symptoms One attack in patient less than 55 years old Inability to rule out carcinoma
358 Table 4.
ALAN
W.
HACKFORD AND MALCOLM
C.
VEIDENHEIMER
Choices of Operation for Perforation or Abscess of Sigmoid Colon*
Resection with primary anastomosis Resection with primary anastomosis plus proximal colostomy Mikulicz resection Hartmann operation Resection with colostomy and formation of mucous fistula Drainage with proximal colostomy
*Adapted from Veidenheimer,
M. C.: Clinical presentation and surgical treatment of complicated diverticular disease. In Allan, R. N., et al. (eds.): Inflammatory Bowel Diseases. Edinburgh, Churchill Livingstone, 1983, p. 525.
Even Smithwick56 indicated that fistulas could also develop secondary to diverticular disease after proximal colostomy. Review of the pathologic findings of our series 27 indicates that with decompression into the urinary bladder abscess and inflammation around the colon subside dramatically. Frequently, little phlegmon is present at the time of operation, making the bowel and fistula particularly suited for primary resection and anastomosis. In first-stage colostomy and drainage the timing of the second-stage resection is often a problem because of the persistence of inflammation. If the second stage is undertaken before the inflammatory process has resolved, the anastomosis is often just as difficult to perform as it would have been at initial presentation. In some of the early data the results of three-stage operations were skewed optimistically because the cumulative morbidity and mortality of all stages were not considered. In a separate review of colostomies performed at the Lahey Clinic from 1963 to 197440 for a variety of indications the mortality rate in patients undergoing colostomy closure alone was 4.2 per cent. The morbidity rate after an interval of 4 months to a year was 34 per cent and rose to greater than 50 per cent when the interval between formation and closure was less than 3 months. Examination of cumulative rates in our current series 27 shows little difference between patients undergoing the three-stage approach and a similar group of patients undergoing two-stage Hartmann closure or mucous fistula formation, despite a higher first-stage mortality rate in the latter (see Table 2). Our data for duration of hospitalization and disability are consistent with data published 14 years ago by Ponka49 who coined the 20-40-60 rule, by which one can anticipate a 20-day hospitalization for the one-stage approach, a 40-day hospitalization for the two-stage procedure, and 60 days of hospitalization for the classic three-stage approach. In our series 27 a single-stage approach was associated with an average duration of 21 days in hospital and 3 weeks of disability (Fig. 5). For patients who underwent two-stage operations with anastomosis the length of hospitalization rose to 31 days with 12 weeks of disability. Hartmann operations required 39 days in hospital, and the average period of disability was 22 weeks. Patients who underwent three-stage operations incurred 52 days in hospital and 7 months of disability. In light of the current concern for medical economics this cannot be considered a trivial issue.
DIVERTICULAR DISEASE
359
III 11 50 40
•
1 Stage 2 Stage Anastomosis & Colostomy A Hartmann Procedure B 3 Stage
Days of Hospitalization
30 20 10 0 Weeks of Disability
10 20 30 40 50
Figure "5 Duration of hospitalization and disability classified according to the number of stages performed.
It would be a mistake to conclude from these data that primary resection as a single-stage treatment should always be chosen for complicated diverticular disease. As Byrne and GarickB pointed out, the main difficulty in attempting to reconcile the divergent opinions in the literature is the problem of classifying this disease in terms of either physiologic or anatomic severity. An attempt to systematize the decisions made by the surgeon during operation for acute diverticulitis was presented by Auguste and Wise4 in 1981. They classified their patients according to four clinical stages (Table 5). Our experience supports the conclusions that disease is localized in stages I and II and that primary resection and anastomosis can be performed safely. When considerable contiguous inflammation is present or when adequate bowel preparation has not been accomplished, proximal colostomy may be indicated. Stages III and IV represent acute diffuse peritoneal inflammatory processes, and an end-colostomy with Hartmann closure of the rectum is currently considered the procedure of choice. In our series,27 however, 16 per cent of patients who underwent a primary anastomosis with a proximal stoma had generalized peritonitis. All patients survived, and the morbidity rate of 22 per cent is essentially equivalent to 23 per cent for the Hartmann operation. The number of patients is too small to draw valid conclusions, but this may be a safe and even preferable alternative to the more conventional two- and three-stage approaches.
360
ALAN
W.
HACKFORD AND MALCOLM
C.
VEIDENHEIMER
Table 5 Pathologic Classification of Acute Diverticular Disease* Stage I Stage II Stage III Stage IV
Pericolic abscess or phlegmon formation Pelvic, intra-abdominal, or retroperitoneal abscess resulting from perforation of a confined pericolic abscess Generalized purulent peritonitis caused by rupture of stage I or stage II abscess Generalized fecal peritonitis
*Adapted from Auguste, L. J., and Wise, L.: Surgical management of perforated diverticulitis. Am. J. Surg., 141:122, 1981. SUMMARY Diverticular disease of the colon is being seen with increasing frequency. An acute complicated presentation of the disease occurs in a minority of patients. In contrast to a previous studyoo in which we found that 70 per cent of patients had had prior episodes, our most recent study27 revealed that for nearly 50 per cent of the patients with acute diverticular disease a complicated attack was the initial manifestation of the disease. Because these patients are more likely to have concomitant medical problems, aggressive elective surgical management is appropriate. This approach is now associated with a mortality rate of less than 1 per cent in patients with uncomplicated disease. Even in patients with complicated active disease, a mortality rate of less than 4 per cent can be anticipated when bowel preparation can be achieved. In patients below the age of 55 resection is advocated after a single attack because the rate of recurrence in this group may be as high as 50 per cent. In the setting of stage I or stage II disease primary resection with anastomosis is safe and should be performed. Proximal colostomy formation may be carried out at the discretion of the surgeon if warranted by such local circumstances as contiguous inflammation or macroscopic contamination. For patients with stage III and stage IV disease end-colostomy with Hartmann closure of the rectum is the procedure of choice, although anastomosis with proximal stoma may prove to be an acceptable alternative. The morbidity and mortality rates associated with the classic three-stage approach are similar to those with two-stage management, but the latter is associated with a substantially shorter duration of hospitalization and disability. The best form of management of diverticular disease is prevention. It is appealing to embrace high-bulk dietary management as a prophylaxis based on current knowledge of pathophysiologic principles, but good prospective randomized data are not yet forthcoming.
REFERENCES 1. Allen, A. W., and Behringer, G. E.: Diverticulosis and diverticulitis of the colon. Pa. Med., 59:1345-1350, 1956.
DIVERTICULAR DISEASE
361
2. Almy, T. P.: Diverticular disease of the colon-the new look (Editorial). Gastroenterology, 49:109-112, 1965. 3. Almy, T. P., and Howell, D. A.: Diverticular disease of the colon. N. Eng!. J. Med., 302:324-331, 1980. 4. Auguste, L. J., and Wise, L.: Surgical management of perforated diverticulitis. Am. J. Surg., 141:122--127, 1981. 5. Beer, E.: Some pathological and clinical aspects of acquired (false) diverticula of the intestine. Am. J. Med. Sci., 128:135--145, 1904. 6. Boles, R. S., Jr., and Jordan, S. M.: The clinical significance of diverticulosis. Gastroenterology, 35:579--582, 1958. 7. Botsford, T. W., Zollinger, R. M., Jr., and Hicks, R.: Mortality of the surgical treatment of diverticulitis. Am. J. Surg., 121 :702--705, 1971. 8. Byrne, J. J., and Garick, E. 1.: Surgical treatment of diverticulitis. Am. J. Surg., 121 :379384, 1971. 9. Carpenter, W. S., Allaben, R. D., and Kambouris, A. A.: Fistulas complicating diverticulitis of the colon. Surg. Gyneco!. Obstet., 134:625--628, 1972. 10. Carpenter, W. S., Allaben, R. D., and Kambouris, A. A.: One-stage resections for colovesical fistulas. J. Uro!., 108:265--267, 1972. 11. Classen, J. N., Bonardi, R., O'Mara, C. S., et a!.: Surgical treatment of acute diverticulitis by staged procedures. Ann. Surg., 184:582--586, 1976. 12. Colcock, B. P.: Surgical management of complicated diverticulitis. N. Eng!. J. Med., 259:570-573, 1958. 13. Colcock, B. P.: Recent experiences in the surgical treatment of diverticulitis. Surg. Gyneco!. Obstet., 121 :63-69, 1965. 14. Colcock, B. P.: Surgical treatment of diverticulitis. Am. J. Surg., 115:264-270,1968. 15. Colcock, B. P.: Diverticulitis: A surgical challenge. Surg. Clin. North Am., 51:791-797, 1971. 16. Cripps, W. H.: The Passage of Air and Faeces from the Urethra. London, J. A. Churchill, 1888. 17. Cruveilhier, J.: Traite d'Anatomie Pathologique Generale, Volume 1. Paris, Bailliere, 1849, p. 592. Cited by Localio, S. A., and Stahl, W. M., Reference 33. 18. Dandekar, N. V., and McCann, W. J.: Primary resection and anastomosis in the management of perforation of diverticulitis of the sigmoid flexure and diffuse peritonitis. Dis. Colon Rectum, 12:172--175, 1969. 19. Dardik, H., Delany, H. M., and Hurwitt, E. S.: Recurrent diverticulitis in a defunctionalized colonic loop. Am. J. Surg., 108:914-916, 1964. 20. Dawson, J. L., Hanon, I., and Roxburgh, R. A.: Diverticulitis coli complicated by diffuse peritonitis. Br. J. Surg., 52:354-357, 1965. 21. Edwards, D. A.: Techniques and interpretations of measurement of gastro-intestinal pressures. Ann. R. Coli. Surg. Engl., 37:275--284, 1965. 22. Eisenstat, T. E., Rubin, R. J., and Salvati, E. P.: Surgical management of diverticulitis: The role of the Hartmann procedure. Dis. Colon Rectum, 26:429--432, 1983. 23. Eusebio, E. B., and Eisenberg, M. M.: Natural history of diverticular disease of the colon in young patients. Am. J. Surg., 125:30~311, 1973. 24. Fraser, I.: Causation, Pathology and Treatment of Diverticula of the Large and Small Intestine. M. D. Thesis, Belfast, 1933. Cited by Parks, T. G., Reference 48. 25. Graser, E.: Das falsche Darmdivertike!. Arch. Klin. Chir., 59:63~647, 1899. 26. Greif, J. M., Fried, G., and McSherry, C. K.: Surgical treatment of perforated diverticulitis of the sigmoid colon. Dis. Colon Rectum, 23:483-487, 1980. 27. Hackford, A. W., Schoetz, D. J., Jr., Coller, J. A., et a!.: Surgical management of complicated diverticulitis: The Lahey Clinic experience, 1967 to 1982. Dis. Colon Rectum (In Press). 28. Hannan, C. E., Knightly, J. J., and Coffey, R. J.: Diverticular disease of the colon in the younger age group. Dis. Colon Rectum, 4:419--423, 1961. 29. Horner, J. L.: Natural history of diverticulosis of the colon. Dig. Dis. Sci., 3:343-350, 1958. 30. Hughes, L. E.: Postmortem ~urvey of diverticular disease of the colon. Gut, 10:336-351, 1969. 31. Jordan, S. M., and Boles, R. S., Jr.: Diverticulitis. Modern Medical Monographs 21. New York, Grune & Stratton, 1960.
362
ALAN W. HACKFORD AND MALCOLM C. VEIDENHEIMER
32. Littre, A., 1700. Cited by Localio, S. A., and Stahl, W. M., Reference 33. 33. Localio, S. A., and Stahl, W. M.: Diverticular disease of the alimentary tract. Part I: The colon. Curro Probl. Surg., 4(12):1-78, 1967. 34. Madden, J. L.: Primary resection and anastomosis in the treatment of perforated lesions of the colon. Am. Surg., 31:781-786, 1965. 35. Madden, J. L., and Tan, P. Y.: Primary resection and anastomosis in the treatment of perforated lesions of the colon, with abscess or diHUsing peritonitis. Surg. Gynecol. Obstet., 113:64&-650, 1961. 36. Marshak, R. H., Lindner, A. E., and Maklansky, D.: Diverticulosis and diverticulitis of the colon. Mt. Sinai J. Med., 46:261-276, 1979. 37. Marshall, S. F.: Earlier resection in one stage for diverticulitis of the colon. Am. Surg., 29:337-346, 1963. 38. Mayo, W. J.: Acquired diverticulitis of the large intestine. Surg. Gynecol. Obstet., 5:815,1907. 39. McSherry, C. K., Grafe, W. R., Jr., Perry, H. S., et al.: Surgery of the large bowel for emergent conditions: Staged vs. primary resection. Arch. Surg., 98:749--753, 1969. 40. Mirelman, D., Corman, M. L., Veidenheimer, M. C., et al.: Colostomies-indications and contraindications: Lahey Clinic experience, 1963-1974. Dis. Colon Rectum, 21:172-176, 1978. 41. Morson, B. C.: The muscle abnormality in diverticular disease of the sigmoid colon. Br. J. Radiol., 36:385-392, 1963. 42. Morson, B. c.: Pathology of diverticular disease of the colon. Clin. Gastroenterol., 4:3752,1975. 43. Morson, B. C.: Pathogenesis of diverticular disease of the colon. Presented at the 83rd Annual Convention, American Society of Colon and Rectal Surgeons, New Orleans, Louisiana, May 6-11, 1984. 44. Painter, N. S.: Diverticular disease of the colon. Br. Med. J., 3:475-479,1968. 45. Painter, N. S., Truelove, S. C., Ardran, G. M., et al.: Effect of morphine, prostigmine, pethidine, and probanthine on the human colon in diverticulosis studied by intraluminal pressure recording and cineradiography. Gut, 6:57-63, 1965. 46. Painter, N. S., Truelove, S. C., Ardran, G. M., et al.: Segmentation and the localization of intraluminal pressures in the human colon, with special reference to the pathogenesis of colonic diverticula. Gastroenterology, 49: 169--177, 1965. 47. Parks, T. G.: Natural history of diverticular disease of the colon: A review of 521 cases. Br. Med. J., 4:639--642, 1969. 48. Parks, T. G.: Natural history of diverticular disease of the colon. Clin. Gastroenterol., 4:53-69, 1975. 49. Ponka, J. L.: Emergency surgical operations for diverticular diseases. Dis. Colon Rectum, 13:235-242, 1970. 50. Rankin, F. W., and Brown, P. W.: Diverticulitis of the colon. Surg. Gynecol. Obstet., 50:836-847, 1930. 51. Risholm, L.: Primary resection in perforating diverticulitis of the colon. World J. Surg., 6:490-491, 1982. 52. Rodkey, G. V., and Welch, C. E.: Colonic diverticular disease with surgical treatment: A study of 338 cases. Surg. Clin. North Am., 54:655-674, 1974. 53. Roxburgh, R. A., Dawson, J. L., and Yeo, R.: Emergency resection in treatment of diverticular disease of colon complicated by peritonitis. Br. Med. J., 3:465-466, 1968. 54. Slack, W. W.: The anatomy, pathology, and some clinical features of diverticulitis of the colon. Br. J. Surg., 50:185-190, 1962. 55. Smithwick, R. H.: Experiences with surgical management of diverticulitis of the sigmoid. Ann. Surg., 115:969--985, 1942. 56. Smithwick, R. H.: Surgical treatment of diverticulitis of the sigmoid. Am. J. Surg., 99:192-205, 1960. 57. Snape, W. J., Jr., Carlson, G. M., and Cohen, S.: Colonic myoelectric activity in the irritable bowel syndrome. Gastroenterology, 70:326-330, 1976. 58. Taylor, I., and Duthie, H. L.: Bran tablets and diverticular disease. Br. Med. J., 1:988990, 1976. 59. Telling, W. H. M., and Gruner, O. C.: Acquired diverticula, diverticulitis, and peridiverticulitis of the large intestine. Br. J. Surg., 4:468--530, 1917.
DIVERTICULAR DISEASE
363
60. Veidenheimer, M. C.: Clinical presentation and surgical treatment of complicated diverticular disease. In Allan, R. N., Keighley, M. R. B., Alexander-Williams, J., et al. (eds.): Inflammatory Bowel Diseases. Edinburgh, Churchill LiVingstone, 1983, pp. 519-528. 61. Veidenheimer, M. C.: Surgical management of diverticulitis and its complications. In Kodner I. J., Fry, R. D., and Roe, J. P. (eds.): Controversies in Colon and Rectal Surgery. St. Louis, C. V. Mosby, 1985 (In Press). 62. Wiihlby, L., and Knutsen, O. H.: Leukocyte counts, ESR and fever in the diagnosis of diverticulitis. Acta Chir. Scand., 148:623--624, 1982. 63. Wara, P., S0rensen, K., Berg, V., et al.: The outcome of staged management of complicated diverticular disease of the sigmoid colon. Acta Chir. Scand., 147:209-214, 1981. 64. Welch, C. E., Allen, A. W., and Donaldson, G. A.: An appraisal ofresection of the colon for diverticulitis of the sigmoid. Ann. Surg., 138:332-343, 1953. Department of Surgery Lahey Clinic Medical Center 41 Mall Road, Box 541 Burlington, Massachusetts 01805
Diverticular Disease of the Colon Current Concepts and Management
Alan W. H ackford, M.D., * and Malcolm C. Veidenheimer, M.D.t
Diverticulitis appears to be a disease of modern Western civilization with an increasing incidence as dietary habits have become more refined and as life expectancy is extended. It was briefly described by Littre 32 at the beginning of the eighteenth century, but Cruveilhierl7 is credited with the first in-depth description of diverticula of the colon as a pathologic entity in 1849. Another half century elapsed before diverticulitis was described by Graser25 as a clinical entity, and it was not until the beginning of the twentieth century that the progression of diverticulitis leading to perforation and fistula was described by Beer. 5 Sexual predilection for diverticular disease has changed. Early in the century the disease was reported in two or three men for every affected woman. 24, 59 In the 1950s no difference in incidence between men and women was noted. 6,29 Recently a female predominance (three to two) has been described. 47 With the increase in the aged population the risk of the development of diverticula, once a pathologic curiosity, is now estimated to exceed 50 per cent. In large series 50 of radiographic and postmortem studies the overall incidence of diverticulosis is between 5 and 6 per cent-5.7 per cent in more than 24,000 barium enema studies and 5.2 per cent in almost 2000 autopsies. Without doubt the incidence increases with age and usually occurs after the age of 40 years. Parks 47,48 estimated an incidence of 5 per cent in the fifth decade of life, rising to 50 per cent in the ninth decade. Welch et al. 64 found that two thirds of patients had diverticula by age 85. The incidence of diverticulitis also increases with the age of the patient. Allen and Behringerl estimated that the incidence of specific radiologic evidence for diverticulitis in patients with diverticulosis will rise from 20 per cent in the fifth decade of life to 33 per cent in the ninth decade, In Horner's study29 of 503 patients with diverticulosis the incidence of di-
*Currently, Section of Colon and Rectal Surgery, Department of Surgery, New England Medical Center Hospitals, Boston, Massachusetts tChairman, Department of Surgery, Lalley Clinic Medical Center, Burlington, Massachusetts
Surgical Clinics of North America-Vol. 65, No.2, April 1985
347
348
ALAN
W.
HACKFORD AND MALCOLM
C.
VEIDENHEIMER
verticulitis was about 10 per cent after 5 years, 25 per cent after 10 years, and greater than 35 per cent at 20 years. Close to 5 per cent of patients were less than 40 years of age.23 In this group, however, diverticular disease tended to be an aggressive form that is more symptomatic and has a high rate of complications. Hannan et al.28 reported in 1961 that almost one half of their 23 patients less than 45 years of age required operation for complications after the first and only attack of diverticulitis. These published reports are in uniform agreement that the sigmoid colon alone is involved in almost 50 per cent of patients. In 147 patients radiographically studied at the Lahey Clinic61 55 per cent showed involvement of the sigmoid alone (Table 1). In 95 per cent of patients reported in the literature29 , 33, 47, 48 the sigmoid is involved either alone or in conjunction with other areas of the colon, The usual impression is that progression of the disease is within the affected segment rather than by spread to other segments, As a result, younger patients with severe localized disease requiring resection will infrequently need subsequent resection for recurrent disease. Of 147 patients undergoing elective resection between 1957 and 1975 at the Lahey Clinic61 for the removal of only the clinically involved portion of the bowel, no patient has yet required a second operation, PATHOGENESIS Current concepts about the development of diverticula are based on two principles: the first is a pressure gradient from lumen to serosa, and the second is a relative weakness in the bowel wall. Morson 41 found disease in every one of 155 operative specimens with a clinical and radiographic diagnosis of diverticulitis, He described what was thought to be an abnormality of muscle: the teniae assumed a cartilaginous consistency, and the circular muscle was thickened causing the bowel wall to have a concertinalike appearance, In the absence of histologic abnormalities Morson 41 considered the muscle thickening to be secondary to shortening rather than to hyperplasia or hypertrophy, Shortening would also explain the redundant folds of mucosa found in diverticular disease, Morson42 , 43 recently confirmed his inability to find histologically abnormal muscle, and he has further described a marked increase in elastic tissue specific to teniae, This teniae-specific elastosis causes longitudinal foreshortening of the colon, ac-
Table 1.
Radiographic Extent of Disease in 147 Patients
EXTENT
NUMBER OF PATIENTS
Sigmoid colon only Sigmoid and descending colon Sigmoid, descending, and transverse colon Pan diverticular disease
81 48 8 10
DIVERTICULAR DISEASE
349
centuating the semicircular corrugations of the shortened circular muscle (Fig. 1). Up to one third of the operative specimens did not have inflammation but showed the muscular thickening described, an abnormality that may precede the actual development of diverticula. Using a combination of perfusion manometry and cineradiography, Painter44-46 demonstrated that waves of high pressure coincide with bandlike contractions, which occlude short segments of bowel and form closed cells in which extraordinarily high pressure can be generated (Fig. 2). Almy2 suggested in 1965 that this abnormality and narrowing of the lumen afforded a straightforward explanation for the predominance of the sigmoid colon in this disorder. Based on the law of Laplace, in which tension in the wall of a cylinder is directly proportional to pressure within the cylinder multiplied by the radius of the cylinder (and inversely proportional to wall thickness), the intraluminal pressure will be greatest where the lumen is narrowest and the wall is thickest. This is in the sigmoid colon in the normal individual. An additional motility disorder has been suggested by more recent myoelectric studies. 58 Asymptomatic patients demonstrate normal motility patterns, but patients with symptomatic diverticular disease have an abnormal slow-wave pattern with a predominant frequency of between 12 and 18 cycles per minute. This is not the pattern seen with irritable bowel syndrome, which shows increased basal electrical rhythms of 3 cycles per minute. 57 A further distinction between diverticular disease and irritable colon is that the abnormal myoelectric activity reverts toward normal when bran is given to patients with diverticular disease, but it does not do so in patients with irritable bowel syndrome. No evidence was found for a causal relationship between the two conditions in a number of clinical surveys.21. 4748 A noteworthy point made by Almy and Howell,3 however, is that since patients with asymptomatic diverticulosis demonstrate no mo-
Figure 1. Macroscopic appearance of diverticular disease with foreshortening and accordionization of the colon.
ALAN W. HACKFORD AND MALCOLM C. VEIDENHEIMER
350
I
it
+.
'."
.
Figure 2.
High-pressure cell formation.
'"
+•
tility changes with dietary management, the effectiveness of such management in preventing serious complications remains uncertain. A weakness in the wall of the intestine is the second major factor in formation of diverticula. Extensive anatomic dissection by Slack54 suggested that diverticula may occur at four sites between the mesenteric and antimesenteric teniae at points of penetration of intramural blood vessels (Fig. 3). Because this is a mucosal herniation these are, by definition, pseudodiverticula. PATHOLOGY OF DIVERTICULITIS
In simple diverticulosis the mucosal sac freely communicates with the lumen of the bowel. Increasing intraluminal pressure may force a fecal mass to become inspissated at the neck and cause obstruction. Subsequent inflammation begins at the apex and spreads to adjacent tissue, either mes-
DIVERTICULAR DISEASE
Figure 3. formation.
351
Sites of diverticula
enteric or pericolic fat, producing a peridiverticulitis. 42 Abscess may occur and extend deep into the mesentery, to adjacent structures, or along the longitudinal muscle layer, rupturing back into the colon to form intramural fistulas.
DIAGNOSIS
As has already been suggested, the diagnosis of diverticulitis carries an appreciable rate of error. Morson42 reported that one third of operative specimens of patients with diverticulitis showed no evidence of inflammation. In a postmortem study Hughes 30 found a converse rate of error with 12 per cent of patients who gave no history of an inflammatory problem having evidence of diverticulitis. Clearly, spasm of the sigmoid with highpressure segmentation can produce pain in the left lower quadrant and occasionally a mass in the left lower quadrant that can mimic an inflammatory process. Reliance on indications of inflammation, such as leukocyte count and an elevated temperature, may also cause error. A recent study62 from Sweden found that leukocyte count and sedimentation rate were of little value in the evaluation of acute diverticulitis. Of 130 patients at the Lahey Clinic who had acute complications of diverticular disease at operation 64 per cent had a leukocyte count within a normal range of 5000 to 12,000. 27
ALAN W. HACKFORD AND MALCOLM C. VEIDENHEIMER
352
The barium enema study has been considered the prime diagnostic modality for the evaluation of diverticular disease. We usually obtain this study 1 to 10 days after the process of acute inflammation subsides with conservative therapy. Diverticula are easily identified, but inflammation is not as evident. In patients with considerable muscular abnormality and pleating of the mucosa an irregular sawtooth pattern will develop. This has been called the prediverticular state and does not necessarily reflect inflammation, but it may be the only finding in patients with severe clinical disease. On the rare occasion when only one diverticulum is present, the results of barium enema study may be normal. Only radiologic demonstration of an abscess cavity or intramural or extramural fistulization can be taken as a certain sign of diverticulitis (Fig. 4). Of 105 contrast studies performed at the Lahey Clinic for acute disease27 only one third demonstrated these specific findings. In up to 25 per cent of the patients the radiographic evaluation raised the possibility of a carcinoma because of obstruction or stricture. Marshak et al. 36 stated that stricture secondary to diverticulitis tends to be a long segment with normal-appearing mucosa, whereas that secondary to carcinoma tends to be a short segment with overhanging edges and ulceration of the mucosa. At times, however, the distinction cannot be made even with the aid of a colonoscope, and a resection is required. AI-
Figure 4.
Contrast study demonstrating perforation and abscess formation.
353
DIVERTICULAR DISEASE
though the diagnosis of diverticulitis is determined clinically, taking into consideration radiographic and laboratory data, it is based principally on the clinical findings of left lower quadrant tenderness sometimes with localizing peritoneal signs and sometimes with the presence of a mass.
MEDICAL MANAGEMENT Not infrequently, particularly in patients less than 40 years of age, the first attack of diverticulitis may result in serious and potentially fatal complications. A period of observation and conservative management is necessary to determine the outcome of a particular attack. Colcock12 estimated that 45 per cent of these patients have recurrent problems. Parks 47 found that a similar number of patients (40 per cent) continue to have symptoms despite medical management. The risk of complication rises with recurrent attacks. Marshall37 reported that complications developed in 22 per cent of his patients with a single attack of diverticulitis, and this incidence rose to almost 60 per cent in patients having two or more attacks. Parks 47 demonstrated that patients whose diverticular disease involves the whole colon do not require operation any more frequently than those with involvement of the sigmoid colon only. Approximately 25 per cent of the latter group required surgical procedures. In the series of Jordan and Boles 31 16 of 73 patients (24 per cent) required operation. In the conservative management of diverticulitis, bowel rest, intravenous fluids, and broad-spectrum antibiotics form the mainstay of therapy. Intraluminal antibiotics are of little use because the inflammatory process is pericolic rather than intraluminal. Almy and Howell3 caution that "considering the frequency of spontaneous recovery, the routine use of antibiotics in diverticulitis generates unjustified confidence in their efficacy." Certainly, if these agents are used, the physician must constantly make clinical judgments in monitoring the progression of an acute attack. If improvement is not seen within 48 hours, serious consideration should be given to operative intervention. As mentioned previously, electromyographic data in patients with acute diverticulitis support the concept of hypermotility. This has provided the rationale for the common use of anticholinergic drugs, but their efficacy in the acute setting remains unsubstantiated. Glucagon may also be helpful as a smooth muscle relaxant. However, the source of pain in patients who improve following this type of regimen is more likely to be sigmoid muscle spasm than an inflammatory process. The use of morphine in this setting is contraindicated because of the well-document~d increases in intraluminal pressure after its administration. Fiber and osmotic bulking agents have little place in the management of acute diverticulitis. The normalization of abnormal electromyographic patterns is only of theoretical importance at this point. The assumption that dietary fiber will ameliorate the pain and bowel dysfunction of diverticular disease is based on several poorly controlled trials. From a pathophysiologic standpoint, however, after the acute attack has settled, in-
354
ALAN W. HACKFORD AND MALCOLM C. VEIDENHEIMER
crease of fecal bulk leads to an expansion of sigmoid diameter and to a reduction in intraluminal pressure (according to the law of Laplace), thus theoretically conferring some protection against fecal inspissation.
SURGICAL MANAGEMENT In 1907 William Maydl8 presented to the American Surgical Association a review of the literature on 19 patients with diverticulitis and described five of his own patients. At that time he advocated primary resection. In 1942 Smithwick55 stated that treatment with colostomy and drainage was insufficient. In 42 patients he found a 5 per cent operative mortality rate but a subsequent late mortality rate of 12 per cent. Furthermore, the majority of his patients showed little improvement clinically. Accordingly, he developed a three-stage approach that resulted in a 6 per cent operative mortality rate but no late deaths and a good clinical course in 82 per cent of the patients. He concluded that the optimum treatment was a multistage approach that included removal of the diseased segment at the second stage. By 1950 support for an elective surgical approach was inevitable because of the availability of antibiotics. In 1953 Welch et al.,64 reporting on 114 resections of the colon, stated that uncomplicated diverticulitis should be considered an important. indication for operation. In papers from the Lahey Clinic, Colcockl 2-15 presented an aggressive surgical approach for diverticular disease that had a low surgical mortality rate. Review of experience at the Lahey Clinic in a recent series60 of 133 patients undergoing elective operation showed no deaths and complications in only 10 patients. Few would now take issue with the concept of elective primary resection and anastomosis for uncomplicated disease. This is not necessarily true in patients in whom complications occur, however. In 1960 Smithwick56 continued to advocate the three-stage approach. In comparing two surgical series approximately 8 years apart, Botsford et al. 7 noted that as primary resection became the preferred treatment the surgical mortality rate of patients increased. Yet, 1 year later Byrne and Garick, 8 found a 26 per cent mortality rate with the three-stage approach when carried out for an abscess and a 30 per cent mortality rate in the presence of peritonitis. This kind of data prompted many surgeons to argue that such patients were too sick to tolerate anything other than primary resection. 34 Those opposed to this concept pointed to the high incidence of anastomotic problems, which led to the application of the Hartmann operation with closure of the rectum and end sigmoid colostomy as the procedure of choice for complicated disease. In a recent series reported by Eisenstat et al. 22 44 patients with complicated diverticulitis were treated by a Hartmann closure resulting in a mortality rate of only 4.5 per cent. In 1969 Dandekar and McCann 18 reviewed 26 patients operated on with diffuse peritonitis. Operations were staged in 19 patients with a morbidity rate of 53 per cent and a mortality rate of 21 per cent after the first stage alone, and 7 patients underwent primary resection with a morbidity
DIVERTICULAR DISEASE
355
rate of 33 per cent and a mortality rate of 14 per cent. They were also among the first to comment on length of hospitalization (52 days) and disability (26 weeks) for the staged approach as compared with that of the group undergoing primary resection (25 days and 7 weeks, respectively). Rodkey and Welch52 in 1974 found no essential difference in morbidity and mortality rates between the use of primary resection and the staged approach in patients who had localized peritonitis with abscess formation. However, they did find a similar disparity between duration of hospitalization and disability with 50 days and 32 weeks, respectively, for the staged approach and 25 days and 5 weeks, respectively, for primary resection. Greif et al. 26 presented a review of world medical literature in 1980 in which they identified 1363 cases of diverticulitis. Of 510 patients with perforation, abscess formation, and localized peritonitis, 204 underwent resection with a mortality rate of 2 per cent, and 306 were treated with colostomy with a mortality rate of 12 per cent. Of 833 patients with diffuse peritonitis, 316 underwent resection with a mortality rate of 12 per cent, and 505 treated with proximal colostomy incurred a mortality rate of 29 per cent. Greif et al. 26 concluded that the results were considerably better if the operative approach for acute diverticulitis included resection. Previously from the same institution McSherry et al. 39 had compared 150 staged patients with 50 patients treated with resection and found that the staged group had it mortality rate of 6 per cent and a morbidity rate of 31 per cent as opposed to the primary resection group, which had a mortality rate of 18 per cent and a morbidity rate of 46 per cent. They concluded that the problem with the emphasis on primary resection was that it was not the surgeon but the patient who was at risk. Classen et al. ll pointed out that in the last decade the same factors that have improved the survival of patients undergoing primary resection, such as the great advances in intensive care and in antibiotic therapy, have also affected the survival of patients treated according to the three-stage procedure. They cited an overall mortality rate of 11 per cent against a historical rate of 21 to 45 per cent for the three-stage approach, which represents the best data in the literature. Classen et al.l! support the use of primary resection when feasible. However, in a recent study from Scandinavia by Wara et al. 63 83 patients with complications were examined, and of 25 patients with generalized peritonitis who were managed with proximal colostomy alone, 11 died. This was contrasted with a mortality rate of 8 per cent for patients who were treated by an emergency resection. In the three-stage group only 58 per cent went on to the second stage of treatment, and only 46 per cent had eventual restoration of intestinal continuity. We have recently reviewed the Lahey Clinic experience with surgical management of complicated diverticulitis between 1967 and 1982 (Table 2).27 Of 140 patients with peritonitis, phlegmon, or formation of abscess or fistula, 86 were treated with primary resection and anastomosis. An additional 13 patients underwent formation of a proximal stoma. Evaluation of the pathologic findings in these two groups showed that abscess was most frequently (46 per cent) associated with a proximal stoma. Phlegmon alone
356
ALAN
Table 2.
W.
HACKFORD AND MALCOLM
Primary resection (86 patients) Primary resection and proximal colostomy (13 patients) Stage 1 Stage 2 Primary resection with Hartmann closure or mucous fistula formation (19 patients) Stage 1 Stage 2 Three-stage procedure (22 patients) Stage 1 Stage 2 Stage 3
VEIDENHEIMER
Lahey Clinic Experience, 1967-1982 MAJOR COMPLICATIONS
DEATHS
PROCEDURE
C.
NUMBER OF OPERATIONS
Number
Mortality Rate (%)
Number
Morbidity Rate (%)
86
(14)*
1
1
18
18
32t 13 13
0 0 0
0 0 0
7
(6)
4 3
22:j: 21 23
30t 19 10
(1)
3 3 0
16:j: 16 0
7 7 0
23:j: 35 0
3 1 2 0
14:j: 5 10 0
14 8 5 1
24:j: 31 25 8
59t 22 20 13
(4)
*Numbers in parentheses indicate additional operations. tTotal number. :j:Cumulative rate.
was present in 42 per cent of patients without a stoma. Formation of a proximal stoma was also utilized in 8 per cent of patients with localized perforation and in 16 per cent of patients with peritonitis. A Hartmann procedure was performed in 19 patients. Both abscess and phlegmon were present in 80 per cent of these patients, and 32 per cent had peritonitis. A three-stage approach was used in 22 patients with a pathologic profile similar to that of the group undergoing the Hartmann operation. Pertinent to the decision to perform anastomosis was the ability to achieve bowel preparation. This was accomplished in 96 per cent of the group who underwent primary resection, in 70 per cent of patients who had a proximal stoma, and in only 30 to 40 per cent of patients who underwent a Hartmann operation or the three-stage approach. Morbidity and mortality rates were examined for each group (see Table 2). The single-stage group had a 1 per cent mortality rate and an 18 per cent morbidity rate. Patients who underwent first and second stages consisting of resection, anastomosis, and creation of a proximal stoma had a 22 per cent morbidity rate and no deaths. The Hartmann group had a 16 per cent mortality rate and a 35 per cent morbidity rate in the first stage. Since no additional (second-stage) complications occurred, the cumulative morbidity rate was 23 per cent. These results are similar to those seen with the three-stage approach, which yielded 14 per cent mortality and 24 per cent morbidity rates overall.
357
DIVERTICULAR DISEASE
An enormous amount of effort has gone into the appraisal of surgical management of diverticulitis. What May038 said in 1907 remains true today: If considerable disease is present and the symptoms do not disappear, primary resection of the affected part of the bowel is indicated before abscess and fistula supervene. In fact, the selection of patients who are at risk for complications is probably the best way to reduce the morbidity rate of this disease. We have modified very little the indications developed by Colcock15 over 10 years ago (Table 3). A number of operations are available for the management of diverticular disease (Table 4). Use of the EEA (end-to-end anastomosis) stapler (United States Surgical Corporation, Stamford, CT) has substantially simplified the second stage of the Hartmann operation and undoubtedly has shortened the length of hospitalization during the second stage. The Mikulicz exteriorization of mucous fistula formation is difficult to perform because the inflammation and foreshortening of the bowel mesentery make delivery of bowel without tension to the abdominal wall a challenge. Our data and that of a number of other series 18, 26, 34, 51, 52, 63 dispute the theory that operations performed in multiple stages offer the patient the best chance for survival. The classic first stage of colostomy and drainage may leave the patient with a hole in the sigmoid colon and a column of stool on the left side. In this situation persistent sepsis is a predictable sequela. Smithwick56 demonstrated in 1960 that 25 per cent of patients with diverticulitis showed evidence of active postoperative inflammation despite construction of a proximal colostomy. In a series presented by Dawson et aI., 20 even after a 2-week period of time 11 per cent of patients treated by the staged approach with an initial proximal colostomy had persistent fever, rigor, and wound drainage-ample evidence of ongoing sepsis. If we accept the theory of Painter et al. 44--46 of compartmentalization of the sigmoid with high-pressure cells, it is difficult to envision any type of colostomy alone being able to vent the pressure that drives stool through a perforated diverticulum. In fact, recurrent diverticulitis in the defunctionalized bowel has been described by several authors. 1g, 35, 53 Diverting colostomy as a treatment for colovesical fistula was popularized by Cripps16 in 1888, a full decade before diverticulitis was even described as a clinical entity. In the 1940s more complicated procedures involving proximal colostomy were popular. Yet, as pointed out by Carpenter et al. g, 10 colostomy does not allow for closure of the fistula.
Table 3.
Indications for Elective Resection
Two or more attacks ofleft lower quadrant pain associated With fever, leukocytosis, and radiographic evidence of diverticulitis One attack associated with radiographic or clinical evidence of perforation One attack associated With radiographic or clinical evidence of obstruction One attack associated with urinary tract symptoms One attack in patient less than 55 years old Inability to rule out carcinoma
358 Table 4.
ALAN
W.
HACKFORD AND MALCOLM
C.
VEIDENHEIMER
Choices of Operation for Perforation or Abscess of Sigmoid Colon*
Resection with primary anastomosis Resection with primary anastomosis plus proximal colostomy Mikulicz resection Hartmann operation Resection with colostomy and formation of mucous fistula Drainage with proximal colostomy
*Adapted from Veidenheimer,
M. C.: Clinical presentation and surgical treatment of complicated diverticular disease. In Allan, R. N., et al. (eds.): Inflammatory Bowel Diseases. Edinburgh, Churchill Livingstone, 1983, p. 525.
Even Smithwick56 indicated that fistulas could also develop secondary to diverticular disease after proximal colostomy. Review of the pathologic findings of our series 27 indicates that with decompression into the urinary bladder abscess and inflammation around the colon subside dramatically. Frequently, little phlegmon is present at the time of operation, making the bowel and fistula particularly suited for primary resection and anastomosis. In first-stage colostomy and drainage the timing of the second-stage resection is often a problem because of the persistence of inflammation. If the second stage is undertaken before the inflammatory process has resolved, the anastomosis is often just as difficult to perform as it would have been at initial presentation. In some of the early data the results of three-stage operations were skewed optimistically because the cumulative morbidity and mortality of all stages were not considered. In a separate review of colostomies performed at the Lahey Clinic from 1963 to 197440 for a variety of indications the mortality rate in patients undergoing colostomy closure alone was 4.2 per cent. The morbidity rate after an interval of 4 months to a year was 34 per cent and rose to greater than 50 per cent when the interval between formation and closure was less than 3 months. Examination of cumulative rates in our current series 27 shows little difference between patients undergoing the three-stage approach and a similar group of patients undergoing two-stage Hartmann closure or mucous fistula formation, despite a higher first-stage mortality rate in the latter (see Table 2). Our data for duration of hospitalization and disability are consistent with data published 14 years ago by Ponka49 who coined the 20-40-60 rule, by which one can anticipate a 20-day hospitalization for the one-stage approach, a 40-day hospitalization for the two-stage procedure, and 60 days of hospitalization for the classic three-stage approach. In our series 27 a single-stage approach was associated with an average duration of 21 days in hospital and 3 weeks of disability (Fig. 5). For patients who underwent two-stage operations with anastomosis the length of hospitalization rose to 31 days with 12 weeks of disability. Hartmann operations required 39 days in hospital, and the average period of disability was 22 weeks. Patients who underwent three-stage operations incurred 52 days in hospital and 7 months of disability. In light of the current concern for medical economics this cannot be considered a trivial issue.
DIVERTICULAR DISEASE
359
III 11 50 40
•
1 Stage 2 Stage Anastomosis & Colostomy A Hartmann Procedure B 3 Stage
Days of Hospitalization
30 20 10 0 Weeks of Disability
10 20 30 40 50
Figure "5 Duration of hospitalization and disability classified according to the number of stages performed.
It would be a mistake to conclude from these data that primary resection as a single-stage treatment should always be chosen for complicated diverticular disease. As Byrne and GarickB pointed out, the main difficulty in attempting to reconcile the divergent opinions in the literature is the problem of classifying this disease in terms of either physiologic or anatomic severity. An attempt to systematize the decisions made by the surgeon during operation for acute diverticulitis was presented by Auguste and Wise4 in 1981. They classified their patients according to four clinical stages (Table 5). Our experience supports the conclusions that disease is localized in stages I and II and that primary resection and anastomosis can be performed safely. When considerable contiguous inflammation is present or when adequate bowel preparation has not been accomplished, proximal colostomy may be indicated. Stages III and IV represent acute diffuse peritoneal inflammatory processes, and an end-colostomy with Hartmann closure of the rectum is currently considered the procedure of choice. In our series,27 however, 16 per cent of patients who underwent a primary anastomosis with a proximal stoma had generalized peritonitis. All patients survived, and the morbidity rate of 22 per cent is essentially equivalent to 23 per cent for the Hartmann operation. The number of patients is too small to draw valid conclusions, but this may be a safe and even preferable alternative to the more conventional two- and three-stage approaches.
360
ALAN
W.
HACKFORD AND MALCOLM
C.
VEIDENHEIMER
Table 5 Pathologic Classification of Acute Diverticular Disease* Stage I Stage II Stage III Stage IV
Pericolic abscess or phlegmon formation Pelvic, intra-abdominal, or retroperitoneal abscess resulting from perforation of a confined pericolic abscess Generalized purulent peritonitis caused by rupture of stage I or stage II abscess Generalized fecal peritonitis
*Adapted from Auguste, L. J., and Wise, L.: Surgical management of perforated diverticulitis. Am. J. Surg., 141:122, 1981. SUMMARY Diverticular disease of the colon is being seen with increasing frequency. An acute complicated presentation of the disease occurs in a minority of patients. In contrast to a previous studyoo in which we found that 70 per cent of patients had had prior episodes, our most recent study27 revealed that for nearly 50 per cent of the patients with acute diverticular disease a complicated attack was the initial manifestation of the disease. Because these patients are more likely to have concomitant medical problems, aggressive elective surgical management is appropriate. This approach is now associated with a mortality rate of less than 1 per cent in patients with uncomplicated disease. Even in patients with complicated active disease, a mortality rate of less than 4 per cent can be anticipated when bowel preparation can be achieved. In patients below the age of 55 resection is advocated after a single attack because the rate of recurrence in this group may be as high as 50 per cent. In the setting of stage I or stage II disease primary resection with anastomosis is safe and should be performed. Proximal colostomy formation may be carried out at the discretion of the surgeon if warranted by such local circumstances as contiguous inflammation or macroscopic contamination. For patients with stage III and stage IV disease end-colostomy with Hartmann closure of the rectum is the procedure of choice, although anastomosis with proximal stoma may prove to be an acceptable alternative. The morbidity and mortality rates associated with the classic three-stage approach are similar to those with two-stage management, but the latter is associated with a substantially shorter duration of hospitalization and disability. The best form of management of diverticular disease is prevention. It is appealing to embrace high-bulk dietary management as a prophylaxis based on current knowledge of pathophysiologic principles, but good prospective randomized data are not yet forthcoming.
REFERENCES 1. Allen, A. W., and Behringer, G. E.: Diverticulosis and diverticulitis of the colon. Pa. Med., 59:1345-1350, 1956.
DIVERTICULAR DISEASE
361
2. Almy, T. P.: Diverticular disease of the colon-the new look (Editorial). Gastroenterology, 49:109-112, 1965. 3. Almy, T. P., and Howell, D. A.: Diverticular disease of the colon. N. Eng!. J. Med., 302:324-331, 1980. 4. Auguste, L. J., and Wise, L.: Surgical management of perforated diverticulitis. Am. J. Surg., 141:122--127, 1981. 5. Beer, E.: Some pathological and clinical aspects of acquired (false) diverticula of the intestine. Am. J. Med. Sci., 128:135--145, 1904. 6. Boles, R. S., Jr., and Jordan, S. M.: The clinical significance of diverticulosis. Gastroenterology, 35:579--582, 1958. 7. Botsford, T. W., Zollinger, R. M., Jr., and Hicks, R.: Mortality of the surgical treatment of diverticulitis. Am. J. Surg., 121 :702--705, 1971. 8. Byrne, J. J., and Garick, E. 1.: Surgical treatment of diverticulitis. Am. J. Surg., 121 :379384, 1971. 9. Carpenter, W. S., Allaben, R. D., and Kambouris, A. A.: Fistulas complicating diverticulitis of the colon. Surg. Gyneco!. Obstet., 134:625--628, 1972. 10. Carpenter, W. S., Allaben, R. D., and Kambouris, A. A.: One-stage resections for colovesical fistulas. J. Uro!., 108:265--267, 1972. 11. Classen, J. N., Bonardi, R., O'Mara, C. S., et a!.: Surgical treatment of acute diverticulitis by staged procedures. Ann. Surg., 184:582--586, 1976. 12. Colcock, B. P.: Surgical management of complicated diverticulitis. N. Eng!. J. Med., 259:570-573, 1958. 13. Colcock, B. P.: Recent experiences in the surgical treatment of diverticulitis. Surg. Gyneco!. Obstet., 121 :63-69, 1965. 14. Colcock, B. P.: Surgical treatment of diverticulitis. Am. J. Surg., 115:264-270,1968. 15. Colcock, B. P.: Diverticulitis: A surgical challenge. Surg. Clin. North Am., 51:791-797, 1971. 16. Cripps, W. H.: The Passage of Air and Faeces from the Urethra. London, J. A. Churchill, 1888. 17. Cruveilhier, J.: Traite d'Anatomie Pathologique Generale, Volume 1. Paris, Bailliere, 1849, p. 592. Cited by Localio, S. A., and Stahl, W. M., Reference 33. 18. Dandekar, N. V., and McCann, W. J.: Primary resection and anastomosis in the management of perforation of diverticulitis of the sigmoid flexure and diffuse peritonitis. Dis. Colon Rectum, 12:172--175, 1969. 19. Dardik, H., Delany, H. M., and Hurwitt, E. S.: Recurrent diverticulitis in a defunctionalized colonic loop. Am. J. Surg., 108:914-916, 1964. 20. Dawson, J. L., Hanon, I., and Roxburgh, R. A.: Diverticulitis coli complicated by diffuse peritonitis. Br. J. Surg., 52:354-357, 1965. 21. Edwards, D. A.: Techniques and interpretations of measurement of gastro-intestinal pressures. Ann. R. Coli. Surg. Engl., 37:275--284, 1965. 22. Eisenstat, T. E., Rubin, R. J., and Salvati, E. P.: Surgical management of diverticulitis: The role of the Hartmann procedure. Dis. Colon Rectum, 26:429--432, 1983. 23. Eusebio, E. B., and Eisenberg, M. M.: Natural history of diverticular disease of the colon in young patients. Am. J. Surg., 125:30~311, 1973. 24. Fraser, I.: Causation, Pathology and Treatment of Diverticula of the Large and Small Intestine. M. D. Thesis, Belfast, 1933. Cited by Parks, T. G., Reference 48. 25. Graser, E.: Das falsche Darmdivertike!. Arch. Klin. Chir., 59:63~647, 1899. 26. Greif, J. M., Fried, G., and McSherry, C. K.: Surgical treatment of perforated diverticulitis of the sigmoid colon. Dis. Colon Rectum, 23:483-487, 1980. 27. Hackford, A. W., Schoetz, D. J., Jr., Coller, J. A., et a!.: Surgical management of complicated diverticulitis: The Lahey Clinic experience, 1967 to 1982. Dis. Colon Rectum (In Press). 28. Hannan, C. E., Knightly, J. J., and Coffey, R. J.: Diverticular disease of the colon in the younger age group. Dis. Colon Rectum, 4:419--423, 1961. 29. Horner, J. L.: Natural history of diverticulosis of the colon. Dig. Dis. Sci., 3:343-350, 1958. 30. Hughes, L. E.: Postmortem ~urvey of diverticular disease of the colon. Gut, 10:336-351, 1969. 31. Jordan, S. M., and Boles, R. S., Jr.: Diverticulitis. Modern Medical Monographs 21. New York, Grune & Stratton, 1960.
362
ALAN W. HACKFORD AND MALCOLM C. VEIDENHEIMER
32. Littre, A., 1700. Cited by Localio, S. A., and Stahl, W. M., Reference 33. 33. Localio, S. A., and Stahl, W. M.: Diverticular disease of the alimentary tract. Part I: The colon. Curro Probl. Surg., 4(12):1-78, 1967. 34. Madden, J. L.: Primary resection and anastomosis in the treatment of perforated lesions of the colon. Am. Surg., 31:781-786, 1965. 35. Madden, J. L., and Tan, P. Y.: Primary resection and anastomosis in the treatment of perforated lesions of the colon, with abscess or diHUsing peritonitis. Surg. Gynecol. Obstet., 113:64&-650, 1961. 36. Marshak, R. H., Lindner, A. E., and Maklansky, D.: Diverticulosis and diverticulitis of the colon. Mt. Sinai J. Med., 46:261-276, 1979. 37. Marshall, S. F.: Earlier resection in one stage for diverticulitis of the colon. Am. Surg., 29:337-346, 1963. 38. Mayo, W. J.: Acquired diverticulitis of the large intestine. Surg. Gynecol. Obstet., 5:815,1907. 39. McSherry, C. K., Grafe, W. R., Jr., Perry, H. S., et al.: Surgery of the large bowel for emergent conditions: Staged vs. primary resection. Arch. Surg., 98:749--753, 1969. 40. Mirelman, D., Corman, M. L., Veidenheimer, M. C., et al.: Colostomies-indications and contraindications: Lahey Clinic experience, 1963-1974. Dis. Colon Rectum, 21:172-176, 1978. 41. Morson, B. C.: The muscle abnormality in diverticular disease of the sigmoid colon. Br. J. Radiol., 36:385-392, 1963. 42. Morson, B. c.: Pathology of diverticular disease of the colon. Clin. Gastroenterol., 4:3752,1975. 43. Morson, B. C.: Pathogenesis of diverticular disease of the colon. Presented at the 83rd Annual Convention, American Society of Colon and Rectal Surgeons, New Orleans, Louisiana, May 6-11, 1984. 44. Painter, N. S.: Diverticular disease of the colon. Br. Med. J., 3:475-479,1968. 45. Painter, N. S., Truelove, S. C., Ardran, G. M., et al.: Effect of morphine, prostigmine, pethidine, and probanthine on the human colon in diverticulosis studied by intraluminal pressure recording and cineradiography. Gut, 6:57-63, 1965. 46. Painter, N. S., Truelove, S. C., Ardran, G. M., et al.: Segmentation and the localization of intraluminal pressures in the human colon, with special reference to the pathogenesis of colonic diverticula. Gastroenterology, 49: 169--177, 1965. 47. Parks, T. G.: Natural history of diverticular disease of the colon: A review of 521 cases. Br. Med. J., 4:639--642, 1969. 48. Parks, T. G.: Natural history of diverticular disease of the colon. Clin. Gastroenterol., 4:53-69, 1975. 49. Ponka, J. L.: Emergency surgical operations for diverticular diseases. Dis. Colon Rectum, 13:235-242, 1970. 50. Rankin, F. W., and Brown, P. W.: Diverticulitis of the colon. Surg. Gynecol. Obstet., 50:836-847, 1930. 51. Risholm, L.: Primary resection in perforating diverticulitis of the colon. World J. Surg., 6:490-491, 1982. 52. Rodkey, G. V., and Welch, C. E.: Colonic diverticular disease with surgical treatment: A study of 338 cases. Surg. Clin. North Am., 54:655-674, 1974. 53. Roxburgh, R. A., Dawson, J. L., and Yeo, R.: Emergency resection in treatment of diverticular disease of colon complicated by peritonitis. Br. Med. J., 3:465-466, 1968. 54. Slack, W. W.: The anatomy, pathology, and some clinical features of diverticulitis of the colon. Br. J. Surg., 50:185-190, 1962. 55. Smithwick, R. H.: Experiences with surgical management of diverticulitis of the sigmoid. Ann. Surg., 115:969--985, 1942. 56. Smithwick, R. H.: Surgical treatment of diverticulitis of the sigmoid. Am. J. Surg., 99:192-205, 1960. 57. Snape, W. J., Jr., Carlson, G. M., and Cohen, S.: Colonic myoelectric activity in the irritable bowel syndrome. Gastroenterology, 70:326-330, 1976. 58. Taylor, I., and Duthie, H. L.: Bran tablets and diverticular disease. Br. Med. J., 1:988990, 1976. 59. Telling, W. H. M., and Gruner, O. C.: Acquired diverticula, diverticulitis, and peridiverticulitis of the large intestine. Br. J. Surg., 4:468--530, 1917.
DIVERTICULAR DISEASE
363
60. Veidenheimer, M. C.: Clinical presentation and surgical treatment of complicated diverticular disease. In Allan, R. N., Keighley, M. R. B., Alexander-Williams, J., et al. (eds.): Inflammatory Bowel Diseases. Edinburgh, Churchill LiVingstone, 1983, pp. 519-528. 61. Veidenheimer, M. C.: Surgical management of diverticulitis and its complications. In Kodner I. J., Fry, R. D., and Roe, J. P. (eds.): Controversies in Colon and Rectal Surgery. St. Louis, C. V. Mosby, 1985 (In Press). 62. Wiihlby, L., and Knutsen, O. H.: Leukocyte counts, ESR and fever in the diagnosis of diverticulitis. Acta Chir. Scand., 148:623--624, 1982. 63. Wara, P., S0rensen, K., Berg, V., et al.: The outcome of staged management of complicated diverticular disease of the sigmoid colon. Acta Chir. Scand., 147:209-214, 1981. 64. Welch, C. E., Allen, A. W., and Donaldson, G. A.: An appraisal ofresection of the colon for diverticulitis of the sigmoid. Ann. Surg., 138:332-343, 1953. Department of Surgery Lahey Clinic Medical Center 41 Mall Road, Box 541 Burlington, Massachusetts 01805