The mortality associated with pelvic fractures.

The mortality among 604 patients with pelvic fractures was 12%. Pedestrian accidents were the etiologic agent in 27% of the patients, but accounted for 49% of the deaths and for 73% of the deaths primarily due to pelvic fractures. Although 71 of the 72 patients who died sustained concomitant major injuries (mean, 3.1), 60% of the deaths (43 patients) were attributed entirely or in part to pelvic fractures. Of particular interest were the 26 patients in whom the pelvic fracture was the primary cause of death. Ninety-three percent were in shock or had clinical evidence of hypovolemia at the time of admission. Eighteen patients (69%) exsanguinated from their pelvic fractures shortly after hospital admission (mean, 9 hours). They were more elderly than the eight patients who survived their initial resuscitation, but subsequently died of sepsis or of renal failure (mean, 62 vs. 38 years). Sepsis arising in the pelvic hematoma and acute renal failure induced by pelvic hemorrhage and/or pelvic sepsis each accounted for 15% of the deaths. Ninety-one percent of the patients who died primarily of their pelvic fracture had a single or double break in the pelvic ring. Thirty-one precent had open pelvic fractures, and injury associated with a 50% mortality. Twenty-three percent had pelvic fracture related iliac or femoral vessel disruptions, an injury associated with a 75% mortality. Mortality in these patients clearly resulted from ineffective control of pelvic hemorrhage and from the inability to prevent sepsis in the pelvic hematoma.

Villous adenomas of the rectum: the accuracy of clinical assessment.

To avoid an unnecessary radical operation, it is important for surgeons to identify a clinically benign villous tumor of the rectum, especially in the middle third area, where a transanal approach may not be feasible. If the high accuracy of this clinical impression can be achieved, alternative methods such as piecemeal snare excision, or electrocoagulation, or both are justified. To evaluate the diagnostic accuracy of a benign rectal villous tumor, 151 patients with totally excised rectal tumors were reviewed. All of these patients had soft and nonulcerated lesions and were judged to be benign. Induration and ulceration of the lesions signified malignancy and were excluded. One hundred and fourteen patients (76%) had benign villous adenomas, 23 patients (15%) had superficial carcinomas, and 14 patients (9%) had invasive carcinomas. Hence the accuracy of detecting a clinically benign villous tumor of the rectum was 91%. This is high enough to avoid a more radical procedure when the clinical impression is that of a benign villous tumor of the rectum.

Ileal pouch-anal anastomosis. Patterns of failure.

During the period from 1980 through 1990, our institution constructed 253 ileoanal reservoirs in 253 patients, of whom 25 (9.9%) experienced pouch failure. A poor functional result was the most common cause of pouch failure (seven [28%] of 25 patients). Unsuspected Crohn's disease became manifest in 13 (5%) of the 253 patients, resulting in pouch loss due to perianal sepsis or pouch fistulas in six patients (24% of 25 failures), and resulted in a significantly increased risk of pouch failure compared with that of the non-Crohn's population. Pouchitis occurred in 78 patients (31%) and accounted for four (16%) of 25 failures, but patients with pouchitis were not at higher risk for pouch failure than were patients who did not have pouchitis (failure rates of 6.4% vs 10.4%, respectively; not significant). Significant pelvic sepsis in the absence of Crohn's disease developed in 13 patients, five (38%) of whom lost their pouches. Poor functional results, pelvic sepsis, and unsuspected Crohn's disease were the major causes of pouch failure, while pouchitis was not.

Palliation for rectal cancer. Resection? Anastomosis?

There is no agreement regarding the proper management of patients with advanced carcinoma of the rectum. We performed a study to clarify whether palliative resection with or without primary anastomosis is worthwhile and safe. Among 679 patients managed for cancer of the rectum, 125 were considered incurable and underwent palliative procedures. High and low anterior resections were performed in nine and 57 cases, respectively, abdominoperineal resection in 26, Hartmann's procedure in three, simple diverting colostomy in 17, and transanal excision in 13. The overall postoperative mortality rate was 0.8%. Postoperative morbidity was 18% in abdominal operations and none in local excisions. Among patients treated by abdominal resections, only one required subsequent reoperation for colonic obstruction secondary to local recurrence. The median survival was 6.4 months for patients treated by diverting colostomy, 14.8 months for abdominally resected cases, and 14.7 months for transanal excisions. We conclude that palliative resection, often with primary anastomosis or local transanal excision, can be done safely in patients with incurable rectal cancer. We believe this approach improves the quality of the remaining life for these patients.

The Hartmann procedure. First choice or last resort in diverticular disease?

OBJECTIVE: To critique changing trends in the surgical management of diverticular disease. DESIGN: Case series. Two hundred twenty-seven consecutive patients required surgery for diverticular disease from 1988 to 1993. Patient records were reviewed retrospectively. Operative procedures included primary resection in all patients with either anastomosis, anastomosis with proximal ileostomy, or the Hartmann procedure. Morbidity, mortality, and length of stay were then compared with each operative procedure and stage of disease. Patients were categorized according to the following pathologic stages: stage 0, no inflammation; stage I, chronic inflammation; stage II, acute inflammation with or without microabscesses; stage III, pericolonic or mesenteric abscess; stage IV, pelvic abscess; and stage V, purulent or feculent peritonitis. SETTING: A university hospital and private affiliated hospitals in a large metropolitan area. MAIN OUTCOME MEASURES: Study outcome parameters included mortality, morbidity, length of hospital stay, and leak rates. These outcomes were then compared with different disease stages and treatments. RESULTS: Mean patient age was 66 years (range, 25-98 years). Male-female ratio was 84:143. Mean follow-up was 23 months (range, 1-132 months). There were 50 fistulas: 24 colovesical, 21 colovaginal, 3 colocolonic, 1 coloenteric, and 1 colouterine. Surgery was categorized as elective for 196 patients (86%), urgent for 12 (5%), and emergent for 19 (8%). Primary resection was performed in all cases. Primary anastomosis was performed in 200 patients (88%), 183 without and 17 with proximal diversion. Twenty-seven patients (12%) underwent a Hartmann procedure with colostomy; 19 patients (70%) have since undergone colostomy closure. Morbidity occurred in 52 patients (23%), including 4 anastomotic leaks (2%). There were 3 perioperative deaths (1%). Mean length of initial hospital stay was 11 days (range, 4-59 days). Length of stay was 5 days (range, 4-7 days) for ileostomy closure (7% morbidity) and 13 days (range, 7-35 days) for the colostomy closure after the Hartmann procedure (33% morbidity). CONCLUSIONS: Primary resection is virtually always possible in complicated diverticular disease. Primary anastomosis, with or without proximal diversion, is safe for patients with no abscesses or localized abscesses and should be considered on an individual basis for patients with pelvic abscesses and peritonitis. Colostomy closure after the Hartmann procedure is associated with significant length of hospitalization and morbidity and leaves one third of patients with permanent stomas.

Surgical management of left colon obstruction: the University of Minnesota experience.

BACKGROUND: Management of left-sided colonic obstruction is a surgical challenge. This study was performed to review our management of patients with left colon obstruction presenting to the University of Minnesota Hospitals over a 10-year period, 1985 to 1994. STUDY DESIGN: We did a retrospective chart review of 143 patients (48 male and 95 female; mean age 70 years). RESULTS: Sites of obstruction were rectosigmoid, 40%; sigmoid colon, 47%; descending colon, 5%; and splenic flexure, 8%. Fifty-two percent of patients had obstructing colorectal cancer. Two patients presented with generalized peritonitis secondary to colonic perforation. The majority (n = 121, 85%) of patients underwent resection (subtotal in 39 [32%], and segmental in 82 [68%]) and anastomosis in a single stage after appropriate resuscitation. Intraoperative colonic cleansing was undertaken in 40 patients (28%). Morbidity within 30 days of operation was 11%, including 1 anastomotic leak, and mortality was 3%. The 4 deaths occurred in patients over 75 years of age and were not from anastomotic complications. CONCLUSIONS: A single stage resection and an anastomosis facilitated by intraoperative colonic cleansing in one-third of cases was performed in 85% of patients presenting with left colon obstruction. One anastomotic leak occurred. Our current policy of strongly favoring a single stage, definitive operation for patients presenting with left colon obstruction appears reasonable on the basis of this retrospective review of our experience.

Prognostic significance of tumor markers in colorectal cancer patients: DNA index, S-phase fraction, p53 expression, and Ki-67 index.

Risk of colorectal cancer recurrence has traditionally been determined by use of pathologic staging. However, it is apparent that subgroups of patients exist within tumor stages whose clinical behavior differs. This study was undertaken to identify tumor-associated factors that might be predictive of outcome in patients with intermediate stages who will benefit the most from postsurgical adjuvant therapy. Seventy patients with stage II and III colorectal cancer were assessed for DNA index, S-phase fraction, p53 expression, and Ki-67 index. Tumor recurrence was analyzed by means of nonparametric tests and Cox proportional hazard models incorporating standard clinical and pathologic criteria. Of the four prognostic markers evaluated, Ki-67 index was significantly associated with disease recurrence (P = 0.02), whereas DNA index, S-phase fraction, and p53 expression were not. After stratification by tumor stage, significant associations between Ki-67 index and disease recurrence were retained in stage II tumors (P = 0.01) but not in stage III tumors (P = 0.23). Cox proportional hazard regression analysis indicated that among stage II patients, those with a Ki-67 index >45% were associated with 6.5 times greater risk for disease recurrence than those with a Ki-67 index >/=45%. It was concluded that an elevated Ki- 67 index is associated with an increased risk of tumor recurrence in stage II colorectal cancer.

Major vascular injuries secondary to pelvic fractures: an unsolved clinical problem.

Twelve patients sustained sixteen pelvic fracture-related iliac and femoral arterial (5) and venous (11) injuries. Death was due in large part to delays in recognition and direct operative control of the major vascular disruption. Prompt operative exploration of all pedestrians admitted in hemorrhagic shock will open pelvic fractures characterized by a double break in the pelvic ring should reduce the 83 per cent mortality currently associated with this combination of injuries.

The ParkS S ileal pouch and anal anastomosis after colectomy and mucosal proctectomy.

Attention to detail is crucial to the success of the operation described. Surgeons contemplating performing it should first be experts in pelvic surgery and are advised to personally observe and participate in the procedure performed by surgeons currently experienced in this technique.

Tube decompression of the dilated colon.

With the technique described herein, decompression of a massively dilated colon is effectively accomplished, allowing resection to proceed safely. This method of decompression has allowed us to perform colonic resection in all patients with toxic megacolon seen in recent years. It is desirable to remove the infected and inflamed colon in such a circumstance. We have not resorted to blowhole colostomies in cases of toxic megacolon, as this leaves the colon, which is the septic source, within the abdominal cavity. This technique has also been used successfully to expedite subtotal colectomy and ileosigmoid anastomosis in patients with obstructing lesions of the left side. It allows colonic decompression and on-table bowel preparation by irrigation with antiseptics such as povidone-iodine (Betadine).

The ileoanal reservoir.

One hundred nine men and 71 women with a mean age of 31 years had construction of 164 S, 2 J, and 14 other ileoanal reservoirs. Postoperative gastrointestinal complications included small bowel obstruction in 11 percent and ileus, hemorrhage, and sepsis in 6 percent, 5 percent, and 11 percent, respectively. There was a 13 percent incidence of miscellaneous postoperative complications. Pouch perianal fistulas developed in 5 percent of patients, and pouch vaginal and other pouch fistulas developed in an additional 4 percent. During long-term follow-up, small bowel obstruction developed in 27 percent of patients, and enterolysis or enterectomy was required in 15 percent of patients. One hundred fourteen patients who were followed for a mean length of 5 years after ileostomy closure (range 16 to 88 months) were evaluated for functional outcome. Function improved with time in 63 percent of patients and remained stable in another 33 percent; only 4 percent had long-term deterioration. Ninety-five percent of patients would again choose an ileoanal reservoir over a permanent ileostomy. This long-term assessment shows that although the ileoanal reservoir is a viable option in the management of mucosal ulcerative colitis, it should not be recommended to every patient.

Restorative proctocolectomy with ileal reservoir and ileoanal anastomosis.

An initial experience with a technique of restorative proctocolectomy utilizing a rectal mucosectomy, total colectomy, and ileal reservoir (Parks S-pouch) with ileoanal anastomosis for patients with ulcerative colitis and familial polyposis is presented. Although there were no deaths, significant morbidity did occur and was attributed to the use of a temporary loop ileostomy which may not be necessary. Early functional results are promising and to date, patient satisfaction is very high.

Pelvic floor physiology.

Pelvic floor physiology is poorly understood. The funnel shape of the pelvic floor and anal canal is uniquely developed to provide discriminatory continence of gas, liquid, and solid. Proximally, the pelvic floor consists of the pubococcygeus and iliococcygeus muscles. Distally, the anal canal is surrounded by the internal and external sphincter muscles. The anorectal ring is situated between the proximal pelvic floor and the distal anal canal. It is the site of the puborectalis muscle, which is anatomically, neurologically, and functionally merged with the deep portion of the external sphincter muscle. It is at this site that unique forces act to create both a flutter valve and the anorectal angle with the flap valve. Extrinsic pressures at this level reinforce both the flap valve and the flutter valve. Intrinsic pressures are generated by all of the surrounding muscles to produce a high-pressure zone. These factors are critical, but many other factors, such as rectal capacity, compliance, colonic transit, motility, and sensory mechanisms, also interact in a complex way to provide normal continence and defecation. Not surprisingly, no single test allows a complete assessment of the interactions of all these factors. Nevertheless, analysis of components thought to be important in pelvic floor physiology has contributed significantly to the understanding of normal as well as abnormal physiology. Although clinical evaluation continues to be the cornerstone of the diagnosis of pelvic floor disorders, anorectal physiological testing has contributed significantly to our understanding of the dynamics of the pelvic floor. With the refinement of existing techniques and the addition of new investigative tools, it is anticipated that knowledge of pelvic floor physiology will continue to grow.

Surgery for complicated diverticulitis.

The surgical treatment of acute, complicated diverticulitis remains controversial. No randomized studies have been performed to clarify which operative procedure best fits each situation. As a result, the surgeon must use accumulated knowledge and judgment to make the correct decisions for an individual patient. The morbidity and mortality of patients with complicated diverticular disease in 1993 depend, not so much on the operative procedure, but on the severity of the disease and the associated comorbid conditions, namely the presence of fecal or purulent peritonitis, past medical problems, immune status, and nutritional status. However, adherence to the several principles detailed in this report will minimize morbidity and mortality. The surgeon should always attempt to convert the patient from an emergency to an urgent or elective operative status. In the absence of free perforation, this goal usually can be achieved. Rushing into surgery in patients with a normal immune system is generally ill advised. It is far preferable to stabilize the patient, percutaneously drain abscesses if possible, prepare the bowel before exploration, and thus keep the option of primary anastomosis open. A primary anastomosis done first thing in the morning is far preferable to an end-stoma created in the middle of the night in an emergency situation. The algorithm displayed in Figure 1 provides a useful guideline for treating patients with complicated diverticulitis.

Rectovaginal fistulas. Therapeutic options.

The management of RVF depends on size, location, and cause; anal sphincter function and overall health status of the patient; and the skill and judgment of the surgeon. Careful preoperative assessment of the fistula, surrounding tissues, and anal sphincter and exclusion of associated disease are essential. With thorough evaluation, thoughtful consideration of treatment options, and meticulous operative technique, patients can be assured of an optimal outcome. Success in treatment of patients with RVF should be measured not just in terms of successful closure of the fistula but also in terms of patient satisfaction with postoperative anal continence.

Immunohistochemical localization of insulin-degrading enzyme along the rat intestine, in the human colon adenocarcinoma cell line (Caco-2), and in human ileum.

Insulin-degrading enzyme (IDE) has been implicated in the intracellular degradation of insulin in insulin target cells. Knowledge of the existence of this enzyme in the intestine will be beneficial to the achievement of clinical oral efficacy of insulin. A comparative study was conducted with rat intestine, human colon adenocarcinoma (Caco-2) cells, and human ileum. Confocal microscopy analysis using the anti-IDE antibody showed that IDE was localized in the mucosal cells of rat and human intestines, as well as in Caco-2 cells. Immunostaining of this enzyme was homogeneous throughout the cell excluding nucleus, indicating a typical cytosolic distribution in rat and human enterocytes and in Caco-2 cells.

Intra-operative endoscopy. University of Minnesota experience.

This review of intraoperative endoscopies was undertaken to identify the accepted and the controversial indications, to report the results and complications, and to determine the impact the procedure had on the operation performed. This retrospective review identified 78 patients who had undergone 82 intraoperative endoscopic exams between 1981 and 1987. Sixty-nine patients underwent intraoperative colonoscopy (84%), six had upper enteroscopy (7%), and seven had total gut endoscopy (9%). Indications were as follows: the inability to fully colonoscope preoperatively (29%), Crohn's disease (19%), malignant polypectomy site (12%), gastrointestinal bleeding (12%), routine screening to rule out synchronous lesions (12%), non palpable colonic lesions (11%), and miscellaneous (4%). Intraoperative endoscopy was successful in 89 per cent of patients. The exams provided information that altered the planned operation in 27 per cent of cases. There were four complications that may have been related to the endoscopic exams. All resolved without the need for further surgery. It is concluded that intraoperative endoscopy can greatly influence the operation performed in a significant percentage of cases. The clear cut indications for its use at this time are location of malignant polypectomy sites, adhesions or tortuous colon that prevents complete preoperative colonoscopy, chronic gastrointestinal bleeding, and location of non palpable bowel lesions. Controversial indications that we feel will prove useful include Crohn's disease and acute gastrointestinal bleeding. Intraoperative endoscopy is best used as an adjunct to preoperative endoscopy and not as a substitute for it.

The presence of insulin-degrading enzyme in human ileal and colonic mucosal cells.

The aim of this research is to characterize the presence of insulin-degrading enzyme in human colon and ileal mucosal cells. Biochemical studies, including the activity-pH profiles, the effects of enzyme inhibitors, immunoprecipitation and western blots, were conducted. The majority of insulin-degrading activity in colon mucosal cells was localized in the cytosol. In both colon and ileum, cytosolic insulin-degrading activities had a pH optimum at pH 7.5, and were extensively inhibited by each of N-ethylmaleimide, p-chloromercuribenzoate, and 1,10-phenanthroline, but were very weakly affected by each of leupeptin, chymostatin, diisopropyl phosphofluoridate and soybean trypsin inhibitor. In the colon and ileum, more than 93% and 96%, respectively, of cytosolic insulin-degrading activities were removed by the mouse monoclonal antibody to human RBC insulin-degrading enzyme, as compared with less than 20% by the normal mouse IgG for both tissues. Further, a western blot analysis revealed that a cytosolic protein of 110 kD, in both human colon and ileum, reacted with the monoclonal antibody to insulin-degrading enzyme. It is concluded that insulin-degrading enzyme is present in the cytosol of human colon and ileal mucosal cells.

Neonatal death following in utero traumatic splenic rupture.

Perinatal death due to maternal injury is unusual unless associated with extensive maternal trauma or death. An unusual case of neonatal death due to in utero traumatic splenic rupture in the absence of significant maternal injury is presented. The case alerts physicians responsible for neonatal care to the existence of treatable causes of neonatal distress following maternal trauma.