J Hepatobiliary Pancreat Surg (2001) 8:211–215

Prophylaxis for septic complications in acute necrotizing pancreatitis Claudio Bassi, Gerardo Mangiante, Massimo Falconi, Roberto Salvia, Isabella Frigerio, and Paolo Pederzoli Surgical and Gastroenterological Department, Hospital “G.B. Rossi”, University of Verona, 37134 Verona, Italy

Abstract Because the mortality of severe pancreatitis is higher when infected necrosis supervenes, prevention of infections has become a relevant endpoint for management. The “ideal” drug should be characterized by specific activity against the bacteria known to be responsible for infection and should be able to penetrate the gland in a sufficient concentration. To date there have been eight prospective trials with antibiotics, one on selective digestive decontamination, and others with enteral nutrition. A meta-analysis regarding experiences with antimicrobial drugs reports a significant reduction in the incidence of infected necrosis and pancreatic abscesses during severe pancreatitis. In conclusion, among the several options aimed at reducing infections during necrotizing pancreatitis, the prophylactic use of antibacterial drugs is the only one to have been tested to date in several randomized studies. Strong consideration should be given to treating patients with severe pancreatitis with broadspectrum antibiotics, selective digestive decontamination, and enteral nutrition. Key words Acute pancreatitis · Antibiotics · Enteral nutrition · Infected pancreatic necrosis · Pancreatic abscess

Introduction Acute pancreatitis ranges in form from mild to severe fatal disease. During the course of the severe form of the disease, infected pancreatic necrosis, pancreatic abscess, and septic complications develop in 40%–70% of patients. The development of secondary infection, particularly between the third and fifth weeks, has now emerged as the principal determinant of survival. The identification of pancreatic necrosis and the prevention and treatment of the associated infections today represent the most important fields of experimental and

Offprint requests to: C. Bassi Received: June 5, 2000 / Accepted: December 28, 2000

clinical research in acute pancreatitis; advances in these fields will, it is speculated, reduce the mortality.1–5 The concept of preventing superinfections applies only to the severe and necrotic form of acute pancreatitis. Previous prospective studies of the use of antibiotics in acute pancreatitis failed to show benefits, because all patients with acute pancreatitis, whether the disease was severe (fewer than 10%) or uncomplicated (90%) were enrolled.6,7 However, the patients in whom prophylactic antibiotics would be of potential use are only those in whom necrotizing pancreatitis is present, and these patients represent fewer than 15%–20% of all patients with acute pancreatitis. It is the necrotic tissue in such patients that constitutes the ideal pabulum for substrate bacterial colonization.1 There are different hypothetical mechanisms by which bacteria may enter pancreatic and peripancreatic necrotic tissue: the hematogenous pathway via the circulation, transmural migration through the colon8 via the colonic translocation of bacteria to the lymphatics,9 via ascites, via the biliary duct system,10 and from the duodenum via the main pancreatic duct.8 Because most pathogens in pancreatic infection are of gastrointestinal origin, the gut seems to be the principal source, and it is reasonable that bacterial translocation may be defined as the passage of bacteria, exotoxins, endotoxins, and cell wall fragments from the intestinal lumen to sterile extraintestinal sites. Three potential mechanisms exist for the pathogenesis of bacterial translocation in acute pancreatitis: altered permeability of the intestinal mucosa, caused by ischemic injury, with a consequent failure of mucosal barrier function; decreased host defense; and the disruption of indigenous gut flora. The impairment of intestinal motility plays a pathophysiological role in the bacterial overgrowth, resulting in bacterial translocation to mesenteric lymph nodes.11 To summarize, Gram-negative pathogens isolated in infected necrosis are believed to reach the pancreas by

212

C. Bassi et al.: Prophylaxis of infected pancreatic necrosis and pancreatic abscess

Table 1. Penetration of different antibiotics into the pancreas Good penetration

Varying penetration

Poor penetration

Clindamycin Fluoroquinolone Imipenem Metronidazole Meziocillin

Cefoxitin Ceftazidime Chloramphenicol Clotrimoxazole Streptomycin

Aminoglycosides Ampicillin Cephalosporins Moxalactan Tetracyclines

bacterial translocation, whereas Gram-positive organisms appear to superinfect the necrotic tissues during bacteremia from remotes sites.1,12–14 The most frequently involved bacteria are therefore of intestinal origin, and this, therefore, should be a rationale for prophylactic antibiotic treatment, or for bowel sterilization. The ideal drug to use should be characterized by: • Specific activity against the bacteria known to be responsible for pancreatic infections • Ability to penetrate the pancreatic tissue, pancreatic exocrine secretions, and peripancreatic fluid/exudate at therapeutic mean inhibitory concentrations (MIC) • Ability to penetrate the pancreas during acute pancreatitis • Clearcut clinical capacity to reduce the development of infected necrosis.1,12,14–16 Following the recent introduction of modern antibiotic prophylaxis, the occurrence of staphylococcal species and even primary fungal infection is slowly, but steadily increasing.1,5,17,18 This changing bacteriologic spectrum suggests the need for continuous reassessment of the potential roles of other specific drugs.12,15,19–22 With regard to the ability of the antibiotics to penetrate the pancreatic tissues, there appears to be some kind of blood/pancreatic juice barrier within the pancreas.1,3,4,12,15,16,21–23 Table 1 illustrates the overall pancreatic penetration of different antibiotics in humans.

Antibiotic prophylaxis We were the first to address, in a prospective randomized clinical trial,24 the role of an effective antibiotic (imipenem, 500 mg every 8 h) given prophylactically in patients with severe acute pancreatitis. We admitted only patients with documented necrotizing pancreatitis, in whom the antibiotic was begun within 72 h of the onset of the disease and continued for 2 weeks. This regimen decreased the incidence of infected necrosis (confirmed by fine-needle aspiration and/or intraoperative tissue culture), when compared with randomized patients not receiving the antibiotic (12% vs 30%; P ,

0.01). The beneficial effects were maximal (no infections in 27 consecutive patients) when the necrosis involved less than 50% of the glandular volume. Sainio and colleagues25 demonstrated a significant reduction in mortality rates (3% vs 23%) in a trial with cefuroxime (4.5 g/day). A letter commenting on their trial26 points out that the inclusion of patients dying of untreatable systemic inflammatory response syndrome in the early phases of pancreatitis undermines any possibility of properly assessing the benefit of antibiotic therapy aimed at preventing a later phenomenon; namely, infected necrosis. In effect, the trial reported a statistically significant difference in mortality between treated and untreated patients, after including two patients with fulminant pancreatitis in the control group. Moreover, three-quarters of the patients had their antibiotic changed during the course of the disease. There was no significant difference in the incidence of specific pancreatic infections between the treated and untreated groups (9 versus 12 cases of infected necrosis or pancreatic abscess), although there was a significant difference in extrapancreatic (particulary urinary tract) infections. In his reply to the letter, Sainio asserts that “to exclude patients from analysis after randomization is not justified”. While this statement may be impeccable from the theoretical standpoint, it fails to make allowances for the clinical complexity of the pathophysiology of acute pancreatitis. Another important study on this topic was carried out in Japan by Takeda et al.,27 and is reviewed by these authors in this issue of the Journal of Hepatobiliary Pancreatic Surgery.28 Delcenserie and coworkers,29 in a smaller series, tested combinations of different antibiotics (ceftazidime, 2 g every 8 h; amikacin, 7.5 mg every 8 h for 10 days). No infection occurred in treated patients, versus a 58.3% infection rate in the untreated control group of 12 patients. Recently, the concept of prevention of pancreatic superinfection has been addressed by a different approach. Luiten et al.30 reported a reduction in mortality rate (35% versus 22%) in patients with severe pancreatitis treated with selective intestinal decontamination. This effect was related to a reduction in late deaths (more than 2 weeks) and was secondary to a decrease in Gramnegative pancreatic infections (P , 0.05). The timing of the start of treatment in relation to the onset of the disease was not given, nor was any standard duration of treatment in relation to the onset of the disease stated, nor yet was there stratification of patients in regard to the extent of necrosis. Finally, in a recent multicenter study, we18 compared imipenem versus perfloxacin (400 mg twice a day). Despite its theoretical potential, pefloxacin was not a valid alternative, because there was a significantly lower in-

C. Bassi et al.: Prophylaxis of infected pancreatic necrosis and pancreatic abscess

fection rate detected in the imipenem group (10% versus 37%).

Enteral nutrition (EN) Although prognosis has been improved with the administration of new antibiotic drugs, which reduce infection, and, consequently, the mortality rate, to date, the incidence of infected necrosis is still high and, as already stressed, in the near future the rate of resistance to antimicrobials will increase more and more as new pathogens emerge. Several experimental studies support the hypothesis that, in severe pancreatitis, disruption of the intestinal flora, damage to the enteric mucosa, and impaired host defenses promote bacterial translocation from the gut, with consequent infection of the necrotic pancreas.31–33 Previous reports have demonstrated that enteral nutrition (EN) for critically ill patients significantly reduce septic complications when compared with total parenteral nutrition (TPN).34 In severe pancreatitis, Kalfarentzos35 showed a reduction of infection in patients treated by EN, while Windsor et al.36 demonstrated a significant increase in antiendotoxin antibodies and a fall in total antibody capacity in TPN versus EN patients. Even if we cannot reach definitive conclusions, as these series are too small, at least we note that EN was cheaper than TPN in most series.37,38 In any case, to prevent bacterial translocation, the intestinal mucosa must be kept viable by a continuous supply of nutrients directly into the lumen. This enteral supply must be delivered as soon as possible, because the therapeutic window is very narrow before the cytokines cascade begins, this being responsible for the inflammatory response syndrome. As soon as the cascade starts, damage to the mucosa and intestinal villi, as well as bacterial translocation, occur extremely early.39 In this view, EN must be considered as a drug.

Probiotics: the future? If antibiotic drugs become more and more obsolete, we can adopt an old, neglected strategy to fight bacteria: microbial interference therapy. The concept is to enhance the enterofermentative flora in contrast to putrefactive flora. In fact, some experimental experiences demonstrate that some strains of Lactobacillaceae seem to prevent bacterial translocation by the suppression of pathogenic intestinal flora. These probiotic bacteria can adhere to the enteric mucosal layer via agonistic mechanisms

213

against Gram-negative pathogens such as Escherichia coli.40 Lactobacillus rhamnosus and L. plantarum can survive gastric acidity and reach the small and large intestines.41 They are able to ferment oat fibers, producing short chain fatty acid (SCFA), which is the real fuel of colic cells.42 Bacteriocins have been found in lactic acid bacteria,43 while the exogenous administration of L. reuterii has been demonstrated to reduce the effects of chemically induced colitis,44 and L. plantarum administration in healthy volunteers showed a significant reduction in Gram-negative bacteria.42 In our model of severe pancreatitis in mice, we demonstrated a significant reduction in bacterial translocation, not only in mesenteric nodes but also in pancreatic necrosis, with the supply for 4 days before and 4 days after the induction of the pancreatitis of large numbers of L. plantarum 299, a strain with a capacity for very high adherence to colic mucosal cells, and showing antagonism to Escherichia coli. Rayes et al. (Personal communication, 1999), reported excellent results in reducing the rate of infectious complications after liver transplants by the administration of fibers and Lactobacillaceae. In any case, before the clinical use of bacterial antagonism during acute disease can be established, we must solve the main problems, such as the pathogenicity of the probiotic agents in critically ill patients, the amounts of probiotics to be administered during illness, and, obviously, the timing of administration. Doses administered to animals are, in our experience, too high for use in human beings, while, for the timing, we can envisage, in future, administration before such invasive diagnostic procedures as endoscopic retrograde cholangio pancreatography (ERCP), to prevent infection in acute pancreatitis. The idea of replacing the prophylactic role of antibiotic drugs with probiotic agents is intriguing.45–47 In experimental studies, only a single strain of probiotics has been administered; however, we believe that a mixture of probiotics, such as Bifidobacteria and Lactobacillaceae, which are resistant to gastric, biliary, and pancreatic juices, could be more useful for the prevention of infections and as alternatives to antibiotics. A mixture that mimics and replaces the normal endogenous flora seems to better reflect a healthy intestinal ecoenviroment than a single strain of probiotic. Obviously, the quantity must be enough to provide an antagonist action against pathogenic flora. The introduction of probiotic agents belongs to a new concept of nutritional therapy: ecoimmunonutrition, i.e., the restoration of the ancestral ecosystem destroyed not only by antibiotic drugs but also by the modern diet.47 The main role of the gut in the immune system

214

C. Bassi et al.: Prophylaxis of infected pancreatic necrosis and pancreatic abscess

of the body could represent the new frontier of this antiinfectious therapy.

Conclusion The results of the studies we have mentioned, taken together, prompt us to conclude that we have no choice today but to regard “antimicrobial” treatment as mandatory in severe pancreatitis. But with which drugs? Alone or in combination? Together with gut decontamination and/or enteral nutrition? When to start prophylactic treatment? How long to keep it up? These issues are not yet resolved. Despite these still unanswered questions, we can currently state that our understanding of the incidence, pathogenesis, and significance of infected necrosis has improved markedly with the more recent advances in research. Similarly, more precise knowledge has been gleaned in both the microbiological and pharmacological fields, so that we now have a series of drugs that are potentially active for prophylaxis and, in selected patients, even for therapy. References 1. Bassi C (1994) Infected pancreatic necrosis. Int J Pancreatol 16:1– 10 2. Isnemann R, Buchler MW (1994) Infection and acute pancreatitis. Br J Surg 81:1707–1708 3. Bassi C, Falconi M, Sartori N, Salvia R, De Sanctis L, Pederzoli P (1996) Role of antibiotics in acute severe pancreatitis. In: Dervenis C (ed) Advances in pancreatic diseases. Georg Thieme Verlag, Stuttgart, chapter 1, pp 64–69 4. Bassi C, Pederzoli P (1993) Antibiotics in acute pancreatitis. In: Bradley EL III (ed) Acute pancreatitis: diagnosis and therapy. Raven, New York, pp 93–98 5. Uomo G, Visconti M, Manes GP, Calise F, Laccetti M, Rabbitti PG (1996) Nonsurgical treatment of acute necrotizing pancreatitis. Pancreas 12:142–148 6. Howes R, Zuidema GD, Cameron JL (1975) Evaluation of prophylactic antibiotics in acute pancreatitis. J Surg Res 18:197–200 7. Finch WT, Sawyers JL, Schenker S (1976) A prospective study to determine the efficacy of antibiotics in acute pancreatitis. Ann Surg 183:667–671 8. Widdison AL, Karanjia ND, Reber HA (1994) Routes of spread of pathogens into the pancreas in a feline model of acute pancreatitis. Gut 35:1306–1310 9. Runkel NS, Moody FG, Smith GS, Rodriguez LF (1991) The role of the gut in the development of sepsis in acute pancreatitis. J Res Res 51:18–23 10. Webster MW, Pasculle AW, Myerovitz RL, Rao KN, Lombardi B (1979) Postinduction bacteremia in experimental acute pancreatitis. Am J Surg 138:418–420 11. Tarpila E, Nystrom PO, Franzen L, Ihse I (1993) Bacterial translocation during acute pancreatitis in rats. Eur J Surg 159:109–113 12. Bradley EL III (1989) Antibiotics in acute pancreatitis. Current status and future direction. Am J Surg 158:472–477 13. Bassi C, Falconi M, Girelli R (1989) Microbiological findings in severe pancreatitis. Surg Res Commun 5:1–4

14. Barie PS (1996) A critical review of antibiotic prophylaxis in severe acute pancreatitis. Am J Surg 132:38–43 15. Buchler M, Malfertheiner P, Friess H (1989) Human pancreatic tissue concentration of bactericidal antibiotics. Gastroenterology 103:1902–1908 16. Bassi C, Pederzoli P, Vesentini S (1994) Behavior of antibiotics during human necrotizing pancreatitis. Antimicrob Agents Chemother 38:830–836 17. Ho HS, Frey CF (1997) The role of antibiotic prophylaxis in severe acute pancreatitis. Arch Surg 132:487–493 18. Bassi C, Falconi M, Talamini G (1998) Controlled clinical trial of pefloxacin vs imipenem in severe acute pancreatitis. Gastroenterology 115:1513–1517 19. Mithòfer K, Fernandez del Castillo C, Ferraro MJ (1996) Antibiotic treatment improves survival in experimental acute necrotizing pancreatitis. Gastroenterology 110:232–240 20. Tyden G, Battstrom C, Malborg AS (1995) Early antibiotic treatment in acute necrotizing pancreatitis. Lancet 346:1375–1376 21. Battstrom C, Malmborg AS, Tyden G (1988) Penetration of ciprofloxacin and ofloxacin into human allograft pancreatic juice. J Antimicrob Chemother 22:213–219 22. Battstrom C, Malmborg AS, Tyden G (1990) Penetration of pefloxacin into human allograft pancreatic juice. J Antimicrob Chemother 25:393–397 23. Bertazzoni Minelli E, Benini A (1996) Pefloxacin penetration into human necrotic pancreatic tissue. J Antimicrob Chemother 38:237–243 24. Pederzoli P, Bassi C, Vesentini S, Campedelli A (1993) A randomized multicenter clinical trial of antibiotic prophylaxis of septic complications in acute necrotizing pancreatitis with imipenem. Surg Gynecol Obstet 176:480–483 25. Sainio V, Kempainnen E, Puolakkainen P (1995) Early antibiotic treatment in acute necrotizing pancreatitis. Lancet 346:663–667 26. Baudin F (1995) Early antibiotic treatment in acute necrotizing pancreatitis (letter). Lancet 346:1374 27. Takeda K, Matsuno S, Sunamura M, Kakugawa Y (1996) Continuous regional arterial infusion of protease inhibitor and antibiotics in acute necrotizing pancreatitis. Am J Surg 171:394–398 28. Takeda K, Matsuno S, Ogawa M, Watanabe S, Atomi Y (2001) Continuous regional arterial infusion (CRAI) therapy reduces the mortality rate of acute necrotizing pancreatitis: results of a cooperative survey in Japan. J Hepatobiliary Pancreat Surg 8:216– 220 29. Delcenserie R, Yzet T, Ducroix JP (1996) Prophylactic antibiotics in the treatment of severe acute alcoholic pancreatitis. Pancreas 13:198–201 30. Luiten EJT, Hop WCJ, Lange JF, Bruning HA (1995) Controlled clinical trial of selective decontamination for the treatment of severe acute pancreatitis. Ann Surg 222:57–65 31. Bengmark S (1996) Econutrition and health maintenance — a new concept to prevent GI inflammation, ulceration and sepsis. Clin Nutr 15:1–10 32. Runkel NSF, Moody FG, Smith GS, Rodriguez LF, LaRocco MT, Miller TA (1991) The role of the gut in the development of sepsis in acute pancreatitis. J Surg Res 51:18–23 33. Gianotti L, Munda R, Alexander JW (1993) Bacterial translocation: a potential source for infection in acute pancreatitis. Pancreas 5:551–558 34. Moore FA, Feliciano DV, Andrassy RJ (1992) Early enteral feeding, compared with parenteral, reduces postoperative septic complications. The result of a meta-analysis. Ann Surg 216:172– 183 35. Kalfarentzos F, Kehagias J, Mead N (1997) Enteral nutrition is superior to parenteral nutrition in severe acute pancreatitis: results of a randomized prospective trial. Br J Surg 84:1665–1669 36. Windsor ACJ, Kanwar S, Li AKG (1998) Compared with parenteral nutrition, enteral feeding attenuates the acute phase response and improves disease severity in acute pancreatitis. Gut 42:431–435

C. Bassi et al.: Prophylaxis of infected pancreatic necrosis and pancreatic abscess 37. Clave SA, Greene LM, Snider HL (1997) Comparison of the safety of early enteral vs parenteral nutrition in mild acute pancreatitis. J Parent Ent Nutr 21:14–20 38. Mangiante G, Marini P, Fratucello GB (1999) The role of early enteral nutrition (EEN) in severe pancreatitis. Riv Ital Nutr Parent Enter 17:78–84 39. Norman J (1998) The role of cytokines in the pathogenesis of acute pancreatitis. Am J Surg 175:76–83 40. Elmer GW, Surawicz CM, Mc Farland LV (1996) Biotherapeutic agents: a negleted modality for the treatment and prevention of selected intestinal and vaginal infections. JAMA 275: 870–876 41. Bengmark S (1998) Immunonutrition: role of biosurfactants, fiber and probiotic bacteria. Nutrition 14:585–594 42. Goldin BR, Gorbach SL, Saxelin M, Barakat S, Gualtieri L, Salminen S (1992) Survival of Lactobacillus species (strain GG) in human gastrointestinal tract. Dig Dis Sci 37:121–128

215

43. Johansson ML, Molin G, Jeppsson B, Nobaek S, Ahrné S, Bengmark S (1993) Administration of different Lactobacillus strains in fermented oat meal soup: in vivo colonization and effect on the indigenous flora. Appl Environment Microbiol 59:15–20 44. Nes IF, Diep DB, Havarstein LS (1996) Biosynthesis of bacteriocins in lactic acid bacteria. Antonie Van Leeuwenhoek 70:113– 128 45. Fabia R, Ar-Rajab A, Johansson ML (1993) Impairment of bacterial flora in human ulcerative colitis and experimental colitis in the rat. Digestion 54:248–255 46. Adawi I, Ahrnè S, Johansson ML, Molin G, Wold A (1996) A mannose-specific adherence mechanism in Lactobacillus plantarum conferring binding to the human colonic cell line HT-29. Appl Environment Microbiol 62:2244–2251 47. Bengmark S (1996) Econutrition and health maintenance. A new concept to prevent GI inflammation, ulceration and sepsis. Clin Nutrition 15:1–10