Conservative surgical treatment of diffuse peritonitis

Conservative surgical treatment of diffuse peritonitis

Conservative surgical treatment of diffuse peritonitis C.A. Seiler, MD, L. Brügger, MD, U. Forssmann, MD, H.U. Baer, MD, and M.W. Büchler, MD, Bern, S...

44KB Sizes 50 Downloads 65 Views

Conservative surgical treatment of diffuse peritonitis C.A. Seiler, MD, L. Brügger, MD, U. Forssmann, MD, H.U. Baer, MD, and M.W. Büchler, MD, Bern, Switzerland

Background. Peritonitis is, even today, a significant source of death and complications. The objective of this study was to determine the morbidity and mortality rates, the incidence of reoperations, and the need for additional treatment strategies (on demand) in patients with diffuse peritonitis. Methods. Prospective analysis including all patients (n = 258) with diffuse peritonitis admitted to our surgical service between November 1993 and April 1998 who underwent a uniform surgical treatment concept of peritonitis including early intervention, source control, and extensive intraoperative lavage. Results. The 258 patients with diffuse peritonitis averaged a mean Mannheim Peritonitis Index of 27.1 points (range, 11-43 points). Source control at the initial operation was possible in 230 of the patients (89%), of those, 21 patients (9%) needed reintervention. In 28 patients (11%), source control was not possible at the initial operation. Twenty of these patients (71%) had to undergo additional treatment strategies (on demand) such as continuous lavage and/or laparostomy. Overall 228 of the 258 patients (88%) needed just 1 initial surgical intervention. The overall morbidity rate was 41%; the rate of reoperation was 12%, and the hospital mortality rate was 14%. Conclusions. A conservative surgical treatment concept supplemented with “extensive” intraoperative lavage reduces the reoperation rate compared with other treatment standards of peritonitis and achieves a low mortality rate in patients with diffuse peritonitis. (Surgery 2000;127:178-84.) From the Department of Visceral and Transplantation Surgery, University of Bern, Inselspital, Bern, Switzerland

EARLY IN THE LAST CENTURY, mortality rates for peritonitis were recognized to be very high, reaching rates of up to 50% to 80%. With the steady progress in the supportive treatment of peritonitis (ie, antibiotics), the evolution of intensive care medicine, the improvement of anesthesiology, and optimized surgical techniques and material, mortality rates have fallen to 20%. The pathologic features and evolution of peritonitis, however, have remained the same. We today only have more efficient and potent means to interrupt the vicious cycle of peritonitis: gastrointestinal perforation, local peritonitis, diffuse peritonitis, release of endotoxins and other mediators, all of which may cumulate in multiple organ failure and death.1-3 The surgical approach to peritonitis was made easier with the introduction of anesthesia in 1846. Indeed one of the first treatment concepts of peritonitis had been proposed by the surgeon Mikulicz in 1889. His strikingly easy concept was early operAccepted for publication July 18, 1999. Reprint requests: M.W. Büchler, MD, Department of Visceral and Transplantation Surgery, University of Bern, Inselspital, CH-3010 Bern, Switzerland. Copyright © 2000 by Mosby, Inc. 0039-6060/2000/$12.00 + 0

178 SURGERY

11/56/101583

ation, elimination of focus, and lavage.4 Between 1970 and 1980, several new and more radical treatment concepts were introduced (eg, continuous postoperative lavage,5,6 staged reintervention,7-9 open abdomen, and scheduled reoperation8-10). However, the importance of these concepts is still debated. We present a prospective study that includes 258 patients with diffuse peritonitis who were treated according to a more conservative treatment concept with early operation, elimination of focus, and extensive intraoperative lavage. Interestingly, the results of this conservative surgical treatment concept are comparable with the data from the more aggressive approach. PATIENTS AND METHODS Of the 423 patients whose cases were prospectively analyzed, 165 patients (39%) had local peritonitis, and 258 patients (61%) had diffuse peritonitis. The 258 patients with diffuse peritonitis, who were treated at our institution between November 1993 and April 1998 (54 months), were included in this report. There were 141 male patients and 117 female patients. The median age of the patients was 63.5 years (range, 16-95 years). Diffuse peritonitis was defined as the spread of disease over several quadrants (2-4 quadrants).

Seiler et al 179

Surgery Volume 127, Number 2

Fig 1. Treatment concept in diffuse peritonitis.

Abscesses were placed into the group of localized peritonitis. Patients with acute pancreatitis were excluded from this study. The severity of peritonitis was evaluated and graded by the Mannheim Peritonitis Index (MPI; Table I), a peritonitis score that has been validated internationally.11-15 Organ failure was defined, according to the criteria published by Deitch2 in 1992, included pulmonary, hepatic, renal, intestinal, hematologic, neurologic, and cardiovascular disorders. Special attention was given to the cause and source of peritonitis. All 258 patients with diffuse peritonitis underwent surgical intervention generally through a midline celiotomy as early as possible, with source control whenever possible, followed by extensive intraoperative lavage of all 4 abdominal quadrants with a minimum 20 to 30 L of (antibiotic free) warm saline solution, which, if needed, continued until the aspiration fluid was clear. Only in cases in which the source of infection (focus) could not be controlled (eg, resection, excision) during the operation was the strategy changed, and continuous postoperative lavage through Salem Sump tube catheters (Argyle; Sherwood Medical, Petit Rechain, Belgium) with 20 L of standard peritoneal dialysis fluid (antibiotic free) per day was added. In the 4 patients (1.5%) in whom postoperative abdominal closure was no longer anatomically possible, treatment with “open abdomen” was applied. All patients were treated

with an antibiotic regimen that consisted of imipenem/cilastatin, or piperacillin plus ornidazole or ceftriaxone plus ornidazole, respectively (ornidazole is metronidazole with a longer halflife). As soon as bacteriologic analysis from the initial operation was available, antibiotic treatment was adapted. 16-20 Nutrition was primarily parenteral and as soon as possible (3-6 days after the operation) was given orally or enternally through the nasogastric tube or a feeding jejunostomy. Death was defined as total in hospital death. Statistical analysis was applied with the chisquared test of independence and the MannWhitney U Test, respectively. Significance was set at a P value of less than .05. RESULTS Of the 258 patients, the source of peritonitis was gastrointestinal in 236 patients (91%). The main origin of peritonitis was the lower gastrointestinal tract (53%) followed by the upper gastrointestinal tract (38%). Sources other than gastrointestinal sources were mainly urologic and retroperitoneal sources of infections in 9% of the patients. In most cases, the cause of peritonitis was a spontaneous secondary peritonitis, and one third of the patients had postoperative/postinterventional/posttraumatic peritonitis (34%) (Table II). The overall severity of peri-

180 Seiler et al

Surgery February 2000

Table I. MPI scoring system Risk factors

Maximum points given

Age, > 50 years Female Organ failure (preoperative)* Malignancy Duration of peritonitis, > 24 h (onset→operation) Origin, not colon Spread diffuse Intra-abdominal fluid Clear Cloudy, pus Fecal TOTALS

Table III. Surgical treatment performed in diffuse peritonitis Surgical treatment

5 5 7 4 4

Resection + stomy Resection + anastomosis Excision + suture Resection (ie, gallbladder, appendix) Stomy Drainage Others Source control possible

4 6 0 6 12 0-47

N (of 258 patients) Percentage 96 32 55 51

37 13 21 20

8 13 3* 230

3 5 1 89

*Suture

alone (2 patients) in colon and stomach, respectively, and removal of peritoneal dialysis catheter (1 patient).

*Pulmonary,

hepatic, renal, intestinal, hematologic, neurologic, and cardiovascular disorders.2

(From Nathens AB, Rotstein OD. Therapeutic options in peritonitis. Surg Infect 1994;74:677-92. By permission.)

Table II. Cause of diffuse peritonitis Cause

N (of 258 patients) Percentage

Spontaneous (secondary) peritonitis Gastric perforation Diverticular perforation Appendicitis perforation Gallbladder perforation Tumor perforation Others Postoperative/postinterventional and posttraumatic peritonitis Anastomotic break-down Stomach Small bowel Colon Perforation gastrointestinal tract ` Posttraumatic peritonitis Postinterventional peritonitis Others

171

66

41 43 33 9 20 25 87

16 16 13 3 8 10 34

1 5 19 17

0.5 2 7.5 6.5

15 9 21

6 3.5 8

tonitis according to the MPI was 27.1 points (range, 11-43 points). The subgroup of postoperative diffuse peritonitis scored an MPI of 29.6 points (range, 16-43 points; P = .002). There was a difference in the MPI scores in patients for whom the source control was “possible” (MPI = 26.7 points; range, 11-43 points) versus source control “not possible” (MPI = 30 points; range, 16-43 points; P = .01). The MPI scores of the selected patients who died of diffuse peritonitis was again significantly higher with 32.7 points (range, 22-39 points) compared with the survivors with an MPI of 26.1 points (range, 11-43

points; P = .000045). There was no significant difference in the MPI scores based on the organ source of peritonitis. However, the time of operation after onset of symptoms had a significant influence on the MPI score: an operation with less than 24 hours had an MPI score of 23.2 points (range, 11-43 points) versus operation with more than 24 hours had an MPI score of 29 points (range, 16-43 points; P = .0000007). The type of surgery of peritonitis consisted whenever feasible of source control which was possible in 89% of the cases. Table III reflects the surgical procedure being performed in regard to source control. There was no significant difference in the surgical treatment in patients where the source control was possible versus the 28 patients where the source control was not possible, except for the fact that in the latter patients a continuous postoperative lavage was established in 20 patients (71%). Table IV reflects the causes and the treatment of all 30 patients who needed reoperation. The overall morbidity rate in diffuse peritonitis was 41%. The relaparotomy rate was 12%. The different types of surgical and medical complications are shown in Table V. The overall mortality rate from diffuse peritonitis was 14% (37/258 patients). The mortality rate for local peritonitis (165 patients not mentioned further in this article) during the same study period was 3%. In cases with diffuse peritonitis where the source control was possible, the mortality rate was 13% (29/230 patients) versus 27% (8/28 patients; P < .05) in patients where the source control was not possible. Postoperative peritonitis scored a 20% mortality rate (11/56 patients). The cause of death was multiorgan failure and sepsis in 27 patients (71%) followed by myocardial infarctions in 4 patients, pulmonary embolism in 2 patients, tumor recurrences in 2 patients, respiratory failure in 1 patient, and renal failure in 1 patient. In 1 patient with sepsis, 1

Seiler et al 181

Surgery Volume 127, Number 2

Table IV. Causes for reoperation and treatment performed in the 30 patients who underwent reoperation Cause

n

Treatment

n

Reoperations in the “source control possible” group (21/230 patients) Abcess 6 Drainage Breakdown of suture 4 Resection/stomy Stomy Suture Evisceration 6 Closure Laparostomy Hematoma 2 Drainage Others 3 Others

6 2 1 1 5 1 2 3

Reoperations in the “source control not possible” group (9/28 patients) Abscess 2 Drainage Evisceration 4 Closure Laparostomy Enterocutaneous fistula 1 Suture/enterostomy Hematoma 1 Drainage Others 1 Others

2 1 3 1 1 1

patient with cancer, 1 patient with respiratory insufficiency, and 1 patient with myocardial infarction, the treatment was abandoned as the result of the patient’s or the relatives’ wishes. The mean hospital stay was 18 days (range, 5-88 days), including 5 days (range, 1-70 days) in the intensive care unit (Table VI). The flow chart (Fig 1) summarizes the patient management for diffuse peritonitis in this series. DISCUSSION Endotoxin is produced by gram-negative bacteria and becomes life-threatening when it is spread into the peritoneal cavity during hollow-organ perforation. It then acts as one of the most important mediators for sepsis and multiorgan failure in diffuse peritonitis.21,22 Therefore a possible strategy to reduce the production of endotoxin would be, first, source control and, second, neutralization of already released mediators as early as possible. To reduce morbidity and death from diffuse peritonitis,22,23 source control as a therapeutic aim has been advocated now for more than 100 years.4 In 80% to 90% of the patients with spontaneous perforation of a hollow-organ or with postoperative/postinterventional/posttraumatic peritonitis, source control is feasible either by resection of the perforated segment or by some sort of repair or decompression. In our own experience, source control was not possible in only 28 of 258 patients (11%) with diffuse peritonitis. However, this group, unfortunately, is prone to carry a high mortality rate because persistent peritonitis means persistant sepsis and multiorgan failure, which is reflected in a more than doubled mortality rate in our cases.3,24-26 Based on the historic experience with source control alone (still

carrying a mortality rate of 20%-40%), other additive treatment options, concepts, and strategies were developed within the 1970s (eg, continuous postoperative lavage,5,6 scheduled reoperation, and open packing7-10). However, these treatment concepts were mainly based on the fear of secondary foci or, in other words, the fear of insufficient control over potential secondary foci at the initial operation. But these additional treatment concepts aimed at prevention of recurrent infection or secondary foci by repeated or prolonged interventions may harbor a considerable amount of treatment-related morbidity and even deaths through multiple and often truly unnecessary reoperations, which may in turn cause secondary problems (eg, perforations) because of iatrogenic surgical lesions or tubes, finally often needing reconstructive surgery (eg, in the cases of planned relaparotomy). Indeed looking at our own results, it seems to be unnecessary to reoperate on most of these patients to eliminate secondary foci because the safest and most successful way to prevent secondary foci is a sufficient initial operation. Thanks to the improvement in intensive care medicine, especially in regards to the treatment of multiorgan failure1,2,21,22 and improvement and optimization in antibiotic regimens,16 the surgical treatment of diffuse peritonitis can nowadays be less aggressive, thus avoiding unnecessary “white” reoperations and only allowing for reoperation on evident indications. This “on demand” approach inevitably helps to reduce procedure-related morbidity. The key maneuver, however, is the initial operation with an adequate source control and, as a point of major importance, extensive intraopera-

182 Seiler et al

Surgery February 2000

Table V. Morbidity from diffuse peritonitis Morbidity Surgical (n = 65 patients; 25%) Breakdown of suture Evisceration Wound infection Abscess Sepsis Others Medical (n = 41 patients; 16%) Lungs Cardiocirculatory Others Overall morbidity Rate of reoperation

Table VI. Hospital stay

N (of 258 patients) Percentage

4 9 22 3 13 17

2 3 8 1 5 6

21 13 7 106 30

8 5 3 41 12

tive lavage to reduce the incidence of secondary foci.2,13,21,22,27-29 According to Hunt,30 he was not able to find any improvement in morbidity and mortality rates with intraoperative irrigation. However, investigations from Kleine and Beger27,28 could clearly demonstrate that endotoxin levels that were extremely high in infectious ascites could be significantly reduced and even brought to normal levels by extensive intraoperative lavage. On the basis of these results, our prospective treatment protocol included extensive intraoperative lavage in patients with diffuse peritonitis since 1993. All patients underwent intraoperative lavage with a minimum of 20 to 30 L of warm (37°C) saline solution. Every compartment of the abdomen from the diaphragm to the pelvis was systematically irrigated, including every small bowel loop. Irrigation was continued until the aspiration fluid was clear, sometimes requiring up to 60 L.29 This protocol was applied in every patient with peritonitis, requiring the surgeon-on-call to be knowledgeable about the amount and the quality of intraoperative irrigation. No antibiotics were added to the irrigation fluid. With this treatment concept, all the additional treatment strategies, such as continuous postoperative peritoneal lavage or staged reintervention, were omitted and only performed on demand; thus only 20 of 258 of our patients (8%) needed an additional treatment option, mainly in the cases in which the initial operation source control could not be achieved sufficiently. In these cases, a continuous postoperative peritoneal irrigation system was installed. In the few patients for whom primary closure of the wound was technically not possible (5/258 patients; 2%) or in cases of secondary evisceration when a direct closure was no longer possi-

Mean ICU Mean hospital stay (days) stay (days) Overall (range) Postoperative peritonitis (range) Source control possible (range) Source control not possible (range)

5 (1-70) 11 (2-70)

18 (5-88) 30 (5-88)

4 (1-33)

16 (5-70)

13 (1-70)

29 (6-87)

ble (4/258 patients; 1.5%) open abdomen was established. With this treatment strategy, an overall mortality rate of 14% from diffuse peritonitis could be achieved. The mortality rate was 13% where source control was possible and 27% (P < .05) where source control at the initial operation was not possible. A factor that influenced survival in diffuse peritonitis is the time of the initial operation, as already stated by Mikulicz4 more than 100 years ago. The shorter the time after onset of disease or perforation, respectively, the less time is left to start the evolution of infection and sepsis cumulating in multiple organ failure. In our series, 85 of 258 patients (33%) could be brought to surgery within the first 24 hours after the onset of disease or first symptoms, respectively. This early, timely intervention attributed positively to the results of this treatment concept. To make our series comparable to other peritonitis series, we chose the MPI as our scoring system.11-13 It is an internationally validated and reliable peritonitis scoring system, comparable to other scoring systems. It especially shows a good correspondence with the Acute Physiology and Chronic Health Evaluation II scoring system.12-15,31 Both scoring systems (APACHE II and MPI) are able to correctly grade intra-abdominal infection severity, and both are strongly and independently associated with the outcome; however, the latter score has the evidence of being easier to calculate.13 Validation studies correlate an MPI of 26 points with a potential mortality rate of over 30%.12 A later study of more than 2000 patients correlated an MPI score of less than 21 points with a mortality rate of 2.3% (range, 0%-11%), an MPI score of 21 to 29 points with a mortality rate of 22.5% (range, 10.6%-50%), and an MPI of more than 29 points with a mortality rate of 59% (41%-87%). Although the sensitivity, specificity, and accuracy of predicting death with these scores is improving, there are still considerable differences between different centers and different publications. MPI proved to be a very valuable tool in predicting the severity of complications

Seiler et al 183

Surgery Volume 127, Number 2

in our hands; however, decision making on the individual case, especially in the instance of limitation of therapy, was never based on the MPI score alone. The MPI in our experience was sensitive to screening those patients who were at risk on the basis of their scoring points: median MPI overall score, 27.1 points (mortality rate, 14%); postoperative peritonitis median MPI score, 29.6 points (mortality rate, 20%). All patients who died had a significantly higher median MPI score of 32.7 points compared with the survivors (median MPI score, 26.1 points), which reflected the severity of peritonitis in this series. The most common origin of peritonitis was perforation of the colon because of diverticulitis followed by perforation of a duodenal or gastric ulcer. This corresponds to the experience of others.22 Diffuse peritonitis after perforated appendicitis is the least dangerous cause of peritonitis, with a median MPI score of 23.7 points and a mortality rate of 0% in our hands. Corresponding with the experience of others, however, postoperative/postinterventional/posttraumatic peritonitis (prognostically rated with an MPI score of 29.6 points and reaching a mortality rate of 20%) is much more dangerous.22 The patients most in danger are those in whom source control could not be achieved, resulting in a mortality rate of 27%, which is reflected by the median MPI score of 30 points.25 The surgical morbidity rate of 25% and the mortality rate of 14% demonstrate the effectiveness of this treatment concept, which in fact is a conservative surgical treatment concept including early operation, elimination of focus, and intraoperative lavage. The extensive intraoperative lavage, according to our opinion, seems to effectively reduce the incidence of secondary foci by its physiologic early debridement of fibrin, blood, bacteria, and debris. In addition, this concept harbors none of the relevant secondary morbidity associated with the other more aggressive treatment concepts. In the future, surgical prevention of infectious complications in peritonitis will concentrate more and more on extensive intraoperative lavage where, for example, with water-jet devices, improved mechanical debridement can be achieved.32 We believe that the indications for planned relaparotomy, staged intervention, open abdomen, and continuous postoperative lavage should be reassessed. In our experience, 88% of the patients were adequately treated with our conservative initial treatment concept. Only 8% (20/258 of patients) needed the option of on-demand supportive measures, in the cases where source control was not possible. In these cases, a continuous postoperative irrigation system was inserted at the ini-

tial operation, but the mortality rate in this group was still an unsatisfactory 27%. Open abdomen and staged reintervention were needed in 2% (5/258 patients) where (at the initial operation) the abdomen could no longer be closed; therefore we were forced to alter our treatment plan. The indication for reoperation should be based on the clinical or radiologic positive evidence of secondary foci or suspicion of uncontrolled conditions only. Additive treatment modalities have not been performed routinely or scheduled but only on demand. The rate of reintervention in our series was 12% (30/258 patients). According to our experience the treatment concept of peritonitis following the 1890 rules of Mikulicz,1 only modified by the extensive intraoperative lavage, allows even seriously sick patients with diffuse peritonitis to achieve an acceptable morbidity and a low mortality rate combined with a low reoperation rate and a short intensive care unit and hospital stay. This easy conservative surgical concept also helps to reduce treatment costs in comparison with the staged relaparotomy concept. Extensive intraoperative lavage, seems to be extremely potent and easy to perform and shows no side effects. It is inexpensive and can be performed in every hospital without additional resources or personnel needed. After our experience, the conservative treatment concept should challenge the discussion about the need for additional surgical treatment concepts in patients with diffuse peritonitis. REFERENCES 1. Goodwin CW. Multiple organ failure: clinical overview of the syndrome. J Trauma 1990;30(suppl):163-5. 2. Deitch EA. Multiple organ failure: pathophysiology and potential future therapy. Ann Surg 1992;216:117-34. 3. Anderson D, Fearon KCH, Grant IS. Laparotomy for abdominal sepsis in the critically ill. Br J Surg 1996;83:535-9. 4. Mikulicz J. Weitere Erfahrungen über die operative Behandlung der Perforationsperitonitis. Arch Klin Chir (Berl) 1889;39:756-84. 5. Hallerback B, Andersson C, Englund N, Glise H, Nihlberg A, Solhang J, et al. A prospective randomized study of continuous peritoneal lavage postoperatively in the treatment of purulent peritonitis. Surg Gynecol Obstet 1986;163:433-6. 6. Leiboff AR, Soroff HS. The treatment of generalized peritonitis by closed postoperative peritoneal lavage: a critical review of the literature. Arch Surg 1987;122:1005-10. 7. Hau T, Ohmann C, Wolmershäuser A, Wacha H, Yang Q. Planned relaparotomie vs relaparotomie in the treatment of intra-abdominal infections. Arch Surg 1995;130:1193-6. 8. Schein M. Planned reoperations and open management in critical intra-abdominal infections: prospective experience in 52 cases. World J Surg 1991;15:537-45. 9. Teichmann W, Wittmann DH, Andreone PA. Scheduled reoperation (Etappen-lavage) for diffuse peritonitis. Arch Surg 1986;121:147-52. 10. Mughal MM, Bancewicz J, Irving MH. “Laparostomy”: a

184 Seiler et al

11.

12.

13.

14.

15.

16.

17.

18.

19. 20.

technique for the management of intractable intra-abdominal sepsis. Br J Surg 1986;73:253-59. Linder MM, Wacha H, Feldmann U, Wesch G, Streifensand RA, Gundlach E. Der Mannheimer Peritonitis-Index: Ein Instrument zur intraoperativen Prognose der Peritonitis. Chirurg 1987;58:84-92. Demmel N, Maag K, Osterholzer G. Wertigkeit klinischer Parameter zur Prognosebeurteilung der Peritonitis: Validierung des Mannheimer Peritonitis-Index. Langenbecks Arch Chir 1994;379:152-8. Pacelli F, Doglietto GB, Alfieri S, Piccioni E, Sgadori A, Gui D, et al. Prognosis in intra-abdominal infections. Arch Surg 1996;131:641-5. Bosscha K, Reijnders K, Hulstaert PF, Algra W, Van der Werken C. Prognostic scoring systems to predict outcome in peritonitis and intraabdominal sepsis. Br J Surg 1997;84,1532-4. Billing A, Fröhlich D, Schildberg FW, and the Peritonitis Study Group. Prediction of outcome using the Mannheim peritonitis index in 2003 patients. Br J Surg 1994;81,209-13. Bohnen JMA, Solomkin JS, Dellinger EP, Bjornson HS, Page CP. Guidelines for clinical care: anti-infective agents for intra-abdominal infection. Arch Surg 1992;127:83-9. Christou NV, Turgeon P, Wassef R, Rotstein O, Bohnen J, Potvin M, and the Canadian Intra-abdominal Infection Study Group. Management of intra-abdominal infections: the case for intraoperative cultures and comprehensive broad-spectrum antibiotic coverage. Arch Surg 1996;131:1193-201. Schein M, Assalia A, Bachus H. Minimal antibiotic therapy after emergency abdominal surgery: a prospective study. Br J Surg 1994;81:989-91. Carrico CJ. The elusive pathophysiology of the multiple organ failure syndrome. Ann Surg 1993;218:109-10. Quinn JP. Rational antibiotic therapy for intra-abdominal infections. Lancet 1997;249:517-8.

Surgery February 2000 21. Marshall JC, Christou NV, Meakins JL. The gastrointestinal tract: the “undrained abscess” of multiple organ failure. Ann Surg 1993;218:111-9. 22. Wittmann DH, Schein M, Condon RE. Management of secondary peritonitis. Ann Surg 1996;224:10-8. 23. Nathens AB, Rotstein OD. Therapeutic options in peritonitis. Surg Infect 1994;74:677-92. 24. Farthmann EH, Schöffel U. Principles and limitations of operative management of intraabdominal infections. World J Surg 1990;14:210-7. 25. Schein M, Decker GAG. Gastrointestinal fistulas associated with large abdominal wall defects: experience with 43 patients. Br J Surg 1990;77:97-100. 26. Schöffel U, Jacobs E, Ruf G, Minerswa F, v Specht BU, Farthmann EH. Intraperitoneal micro-organisms and the severity of peritonitis. Eur J Surg 1995;161:501-8. 27. Kleine HO, Beger HG. Edotoxinelimination durch intraund postoperative Lavage bei diffuser bakterieller Peritonitis. [abstract] Arch Klin Chir (Berl) 1988;II(suppl):654. 28. Kleine HO, Beger HG. Endotoxin in the peritoneal fluid of bacterial peritonitis. Chir Forum 1988:301-4. 29. Seiler CA, Balsiger B, Feodorovici M, Baer HU, Büchler MW. Extensive intraoperative lavage: the key-maneuver in the treatment of severe peritonitis. Dig Surg 1996;13:400-4. 30. Hunt JL. Generalized peritonitis: to irrigate or not to irrigate the abdominal cavity. Arch Surg 1982;117:209-11. 31. Nespoli A, Ravizzini C, Trivella M, Segala M. The choice of surgical procedure for peritonitis due to colonic perforation. Arch Surg 1993;128:814-8. 32. Katsuhiko S, Mitsuhiro H, Tsunemasa T, Takashi O, Seiichi S, Akira K. Mechanically assisted intraoperative peritoneal lavage for generalized peritonitis as a result of perforation of the upper part of the gastrointestinal tract. J Am Coll Surg 1994;179:443-8.