How To Diagnose Secondary Bacterial Peritonitis In Cirrhosis

Transcription

1 Secondary bacterial peritonitis in cirrhosis: A retrospective study of clinical and analytical characteristics, diagnosis and management Germán Soriano 1,2, *, José Castellote 3, Cristina Álvarez 1, Anna Girbau 3, Jordi Gordillo 1, Carme Baliellas 3, Meritxell Casas 1, Carles Pons 3, Eva María Román 1,2, Sandra Maisterra 3, Xavier Xiol 3, Carlos Guarner 1,2 1 Department of Gastroenterology, Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona, Sant Antoni Maria Claret, 167, Barcelona, Spain; 2 Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Universitat de Barcelona, Spain; 3 Liver and Liver Transplantation Unit, Department of Gastroenterology, IDIBELL, Hospital Universitari de Bellvitge, L Hospitalet de Llobregat, Universitat de Barcelona, Spain See Editorial, pages 7 9 Background & Aims: Secondary bacterial peritonitis in cirrhotic patients is an uncommon entity that has been little reported. Our aim is to analyse the frequency, clinical characteristics, treatment and prognosis of patients with secondary peritonitis in comparison to those of patients with spontaneous bacterial peritonitis (SBP). Methods: Retrospective analysis of 24 cirrhotic patients with secondary peritonitis compared with 106 SBP episodes. Results: Secondary peritonitis represented 4.5% of all peritonitis in cirrhotic patients. Patients with secondary peritonitis showed a significantly more severe local inflammatory response than patients with SBP. Considering diagnosis of secondary peritonitis, the sensitivity of Runyon s criteria was 66.6% and specificity 89.7%, Runyon s criteria and/or polymicrobial ascitic fluid culture were present in 95.6%, and abdominal computed tomography was diagnostic in 85% of patients in whom diagnosis was confirmed by surgery or autopsy. Mortality during hospitalization was higher in patients with secondary peritonitis than in those with SBP (16/24, 66.6% vs. 28/106, 26.4%) (p < 0.001). There was a trend to lower mortality in secondary peritonitis patients who underwent surgery (7/13, 53.8%) than in those who received medical treatment only (9/11, 81.8%) (p = 0.21). Considering surgically treated patients, the time between diagnostic paracentesis and surgery was shorter in survivors than in non-survivors (3.2 ± 2.4 vs. 7.2 ± 6.1 days, p = 0.31). Keywords: Cirrhosis; Surgery; Secondary peritonitis. Received 3 June 2009; received in revised form 11 July 2009; accepted 20 August 2009; available online 23 October 2009 * Corresponding author. Address: Department of Gastroenterology, Hospital de la Santa Creu i Sant Pau, Barcelona, Universitat Autònoma de Barcelona, Sant Antoni Maria Claret, 167, Barcelona, Spain. Tel.: ; fax: address: gsoriano@santpau.cat (G. Soriano). Abbreviations: SBP, spontaneous bacterial peritonitis; CT, computed tomography; LDH, lactate dehydrogenase; ROC, receiver operating characteristic; MELD, model for end stage liver disease; AUC, area under curve; ERCP, endoscopic retrograde cholangiopancreatography; HIV, human immunodeficiency virus; INR, international normalized ratio. Conclusions: Secondary peritonitis is an infrequent complication in cirrhotic patients but mortality is high. A low threshold of suspicion on the basis of Runyon s criteria and microbiological data, together with an aggressive approach that includes prompt abdominal computed tomography and early surgical evaluation, could improve prognosis in these patients. Ó 2009 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Introduction Spontaneous bacterial peritonitis (SBP) is a frequent complication in cirrhotic patients with ascites and morbidity and mortality remain significantly high even though advances in the management of this complication have improved survival in recent years [1 7]. In contrast with SBP, secondary bacterial peritonitis consists of ascitic fluid bacterial infection due to an intraabdominal infectious source such as gastrointestinal tract perforation or abscess [8 13]. In cirrhotic patients, secondary peritonitis is much less frequent than SBP and therefore much less reported [5,8 14]. There are two main concerns in secondary peritonitis. The first is the differential diagnosis with SBP [8 10,12,13] as surgical treatment should be considered in secondary peritonitis but never in SBP [2,9,15 16] and furthermore, an unnecessary laparotomy carries a high risk of mortality in cirrhotic patients [17]. Consequently, highly sensitive and specific diagnostic criteria are needed [9,10,16]. The second major concern is the high mortality rate, reported to range from 50% [9] to 80% [8,13]. It should also be pointed out that most articles on secondary peritonitis were published two decades ago [8 13]. In recent years, besides the improvement in the general management of cirrhotic patients, abdominal computed tomography (CT) [15] with the possibility of percutaneous drainage of abscesses [18] has become common practice, and more effective antibiotics are available [19]. It is therefore of interest to examine the mortality Journal of Hepatology 2010 vol. 52 j 39 44

2 of cirrhotic patients with secondary peritonitis in recent years, and determine which patients could benefit from surgery. The objectives of the present study were to: (1) analyze the frequency of secondary peritonitis among cirrhotic patients with peritonitis, (2) describe the clinical and analytical characteristics, treatment and prognosis of cirrhotic patients with secondary peritonitis in comparison with patients with SBP, (3) evaluate the accuracy of ascitic fluid analysis and abdominal CT in the diagnosis of secondary peritonitis, and (4) analyze which patients could benefit from surgery. Table 1. Characteristics of patients with secondary peritonitis and SBP. Patients and methods We retrospectively analyzed all cirrhotic patients with ascites who presented secondary bacterial peritonitis between January 2000 and December 2007 at two university hospitals in Barcelona, Spain: Hospital Universitari de Bellvitge and Hospital de la Santa Creu i Sant Pau. Secondary bacterial peritonitis was diagnosed when patients fulfilled the three following criteria: ascitic fluid neutrophil count was P250/mm 3, a positive ascitic fluid culture, and evidence of an intraabdominal source of infection, demonstrated by surgery, autopsy [9,10] or abdominal CT. Patients with recent abdominal surgery (<1 month) were excluded. Patients with secondary bacterial peritonitis were treated with broad-spectrum antibiotics [16,19] and surgery was considered by the clinicians in charge of the patient. Patients with secondary peritonitis were compared with a cohort of consecutive cirrhotic patients with SBP diagnosed between January 2001 and December 2004 in Hospital de la Santa Creu i Sant Pau. SBP was diagnosed when ascitic fluid neutrophil count was P250/mm 3 and there was no evidence of an intraabdominal source of infection [2,16]. To avoid including secondary peritonitis misdiagnosed as SBP, we excluded patients with suspected SBP when data to assess Runyon s criteria for secondary peritonitis were not available. Patients with Runyon s criteria or polymicrobial culture in whom imaging techniques or autopsy were not performed were also excluded. Patients with SBP were treated with 2 g/day of intravenous ceftriaxone [2,16,20] and albumin was administered [7] or not based on the decision of the clinician. We considered secondary peritonitis or SBP as community-acquired when infection was present at admission and nosocomial when diagnosis was made after 72 h of admission or when a previous neutrophil count during admission was <250/mm 3 [5,21]. The presence of two of the following three parameters in ascitic fluid was considered indicative of secondary peritonitis suspicion: total protein >10 g/l, glucose <2.7 mmol/l, and lactate dehydrogenase (LDH) > upper limit of normal for serum (Runyon s criteria) [8,9]. Microbiological study of ascitic fluid included: Gram stain, conventional culture, and inoculation of ascitic fluid into two blood culture bottles (aerobic and anaerobic, 10 ml each), when possible, in view of the quantity of ascitic fluid obtained at paracentesis [22,23]. Renal failure at diagnosis of peritonitis was considered when serum creatinine values were above 133 lmol/l and/or serum urea values were above 11 mmol/l [7,24]. Septic shock was defined as a decrease in systolic blood pressure below 90 mm Hg or a reduction of more than 40 mm Hg from baseline, despite adequate fluid resuscitation, accompanied by tachycardia and oliguria (urine output less than 20 ml/h) or anuria in the absence of other causes of shock [25]. SBP was considered resolved when signs and symptoms of infection had disappeared, ascitic fluid neutrophil count was <250/mm 3, and ascitic fluid culture was negative [16]. Statistical analysis was performed with the Student s t test and Mann Whitney test to analyze quantitative variables, while the v 2 test and Fisher test were used for qualitative variables to compare secondary peritonitis patients with SBP episodes and to compare survivors and non-survivors with secondary peritonitis. The results are expressed as mean ± standard deviation or frequencies. Multivariable logistic regression was performed with significant variables from univariate analysis to identify diagnostic parameters that would differentiate between secondary peritonitis and SBP. A forward procedure with Wald test was used to determine the best model. The predictive ability of the model was evaluated using the area under the receiver operating characteristic (ROC) curve. Calibration was assessed using the Hosmer Lemeshow goodness-of-fit test. Calculations were performed with the SPSS Statistical Package (version 15.0, 2006; SPSS Inc., Chicago, IL). A p value < 0.05 was considered statistically significant. Results Between January 2000 and December 2007, 24 cirrhotic patients fulfilled the diagnostic criteria for secondary bacterial peritonitis SBP, spontaneous bacterial peritonitis; HIV, human immunodeficiency virus; MELD, model for end stage liver disease. Data are mean ± SD or frequency (%). a Overall number of patients with previous complications of cirrhosis. in the two hospitals: 14 were diagnosed at Hospital de Bellvitge and 10 at Hospital de la Santa Creu i Sant Pau. In addition, a total of 124 episodes of SBP were diagnosed in 108 cirrhotic patients between January 2001 and December 2004 in the Hospital de la Santa Creu i Sant Pau. From these 124 SBP episodes, 18 were excluded because: (1) parameters to calculate Runyon s criteria were not available (17 episodes) or (2) Runyon s criteria were present and the patient died in the 24 h after diagnosis of peritonitis without imaging techniques being performed (1 episode). Therefore, 106 SBP episodes in 93 patients were finally included in the study. These data represent 26.5 SBP episodes/year as compared to1.25 episodes of secondary peritonitis/year at Hospital de la Santa Creu i Sant Pau. Secondary peritonitis thus accounted for 4.5% of all peritonitis in cirrhotic patients in this center. Patients characteristics The characteristics of cirrhotic patients with secondary peritonitis and those with SBP are shown in Tables 1 and 2. Patients with SBP showed a statistically significant more advanced liver disease indicated by Child-Pugh class distribution. However, the model for end stage liver disease (MELD) score was higher in patients with secondary peritonitis, probably due to a higher incidence of renal failure in this group, although these differences did not reach statistical significance. Serum creatinine and urea levels were higher in patients with secondary peritonitis, but statistical significance was only achieved for urea levels. As shown in Table 2, patients with secondary peritonitis showed a higher incidence of abdominal pain, higher blood leukocyte count, and a higher mean arterial pressure than patients with SBP. Ascitic fluid analysis As observed in Table 2, patients with secondary peritonitis had a higher neutrophil count, higher total protein and LDH, and lower 40 Journal of Hepatology 2010 vol. 52 j 39 44

3 Table 2. Clinical and analytical characteristics at diagnosis of the episode of peritonitis in patients with secondary peritonitis and SBP. JOURNAL OF HEPATOLOGY Considering patients with secondary peritonitis, 22/23 (95.6%) presented Runyon s criteria and/or polymicrobial culture. As 12/ 106 patients with SBP also met these criteria, the specificity of this combination of parameters was 88.6%. Multivariable analysis showed that among parameters with statistical significance in the univariate analysis to differentiate secondary peritonitis and SBP patients (Tables 1 and 2), Runyon s criteria (OR 61.2, 95% CI , p < 0.001) and polymicrobial culture (OR 587.5, 95% CI , p < 0.001) were the only independent predictive factors associated to secondary peritonitis diagnosis. The result of the Hosmer Lemeshow test was p = and AUC of ROC curve Abdominal CT and final diagnosis SBP, spontaneous bacterial peritonitis; INR, international normalized ratio; LDH, lactate dehydrogenase. Data are mean ± SD or frequency (%). glucose levels in ascitic fluid than patients with SBP. Ascitic fluid total protein and glucose levels were available in all patients with secondary peritonitis, but ascitic fluid LDH was not determined in seven patients. However, four of these seven patients fulfilled Runyon s criteria for suspected secondary peritonitis as they showed total protein >10 g/l and glucose <2.7 mmol/l in ascitic fluid. Taken together, at least two out of three Runyon s criteria were present in 14 of 21 (66.6%) patients with secondary peritonitis. Considering only the 13 patients with perforation, two patients were not suitable for evaluation, and 7/11 (63.6%) fulfilled at least two of Runyon s criteria. Considering SBP episodes, 11 of these patients fulfilled two Runyon s criteria (10.3%) (p < with respect to secondary peritonitis). Thus, the sensitivity of Runyon s criteria for diagnosing secondary peritonitis in the present study was 66.6% and specificity was 89.6%. A second paracentesis to check the evolution of ascitic fluid neutrophil count was performed 48 h after the initial tap in 14 patients with secondary peritonitis; seven patients (50%) failed to show a decrease P25% with respect to basal values. SBP resolution was demonstrated in 90/106 (85%) of SBP episodes. Gram stain and culture were more frequently positive in secondary peritonitis patients. Bacteria isolated in secondary peritonitis were 20 Gram-negative bacilli, 16 Gram-positive cocci, four anaerobes, and three fungi. In SBP 28 Gram-negative bacilli, 17 Gram-positive cocci, one anaerobe, and one fungus were isolated. In secondary peritonitis, culture was more frequently polymicrobial (62.5% vs. 0.9%, p<0.001) and anaerobes and/or fungi were more frequently isolated (25% vs. 1.8%, p < 0.001). An abdominal CT was performed in 23/24 patients with secondary peritonitis and was diagnostic in 21/23 (91.3%). Only one patient had percutaneous radiology-guided abscess drainage. Final diagnosis of secondary peritonitis was obtained by abdominal CT in 21 patients, abdominal ultrasound in 1 and surgery in 2. Abdominal CT was performed in 14 of 15 patients in whom secondary peritonitis was later confirmed by surgery (13 patients) or by autopsy (two patients), and it provided the diagnosis of secondary peritonitis in 12/14 (85.7%). The time between the first paracentesis and final diagnosis of secondary peritonitis was 2.5 ± 3.2 days. We classified secondary peritonitis into two groups: patients with free perforation and patients without free perforation. Free perforation was demonstrated in 13 patients: peptic ulcer (5), intestinal diverticula (3), appendicitis (2), intestinal ischemia (1), perforation during gastroscopy (1), and intestinal perforation of unknown cause (1); the remaining cases without free perforation [11] were secondary to: intestinal ischemia [4], cholecystitis [3], appendicitis [3], and post-ercp abscess [1]. Seven of the 11 patients with SBP episodes that fulfilled Runyon s criteria underwent abdominal CT and no features of secondary peritonitis were found. In the remaining four patients, abdominal ultrasound was not indicative of secondary peritonitis, the evolution was satisfactory with ceftriaxone, and the patients were discharged with resolution of SBP confirmed by subsequent paracentesis. Another patient who had SBP and polymicrobial culture died during hospitalization and autopsy failed to demonstrate secondary peritonitis. Clinical course and mortality Surgery was performed in 13/24 (54.1%) patients with secondary peritonitis, nine with free perforation and four without. Seven operated patients were Child-Pugh A or B and six were Child- Pugh C. Total hospitalization time was 22.6 ± 21.4 days in patients with secondary peritonitis and 21.0 ± 14.7 in patients with SBP (p = 0.65). Considering only survivors, the hospitalization time was 38.8 ± 28.1 days for patients with secondary peritonitis and 22.9 ± 14.9 for patients with SBP (p = 0.07). Thirty-eight (35.8%) patients with SBP received albumin expansion. Mortality during hospitalization was higher in patients with secondary peritonitis (16/24, 66.6%) than in patients with SBP (28/106, 26.4%) (p < 0.001). Three-month mortality was 66.6% in the former and 29.2% in SBP patients (p = 0.001). In patients with SBP, mortality was lower for Child-Pugh A or B patients (8/51, 15.6%) than in Child-Pugh C patients (20/55, 36.3%) (p = 0.02). Journal of Hepatology 2010 vol. 52 j

4 Table 3. Characteristics of patients with secondary peritonitis according to hospital survival (I). Table 4. Characteristics of patients with secondary peritonitis according to hospital survival (II). SBP, spontaneous bacterial peritonitis; MELD, model for end stage liver disease. Data are mean ± SD or frequency (%). a Overall number of patients with previous complications of cirrhosis. In contrast, in patients with secondary peritonitis, mortality was identical in Child-Pugh A or B (10/15, 66.6%) and Child-Pugh C patients (6/9, 66.6%). Mortality showed a trend to be lower in surgically treated patients (7/13, 53.8%) than in patients receiving medical treatment only (9/11, 81.8%), although this did not reach statistical significance (p = 0.21). Considering Child-Pugh A or B patients, mortality was 3/7 (42.8%) in operated patients and 7/8 (87.5%) in non-operated patients (p = 0.11). In Child-Pugh C patients, mortality was 4/6 (66.6%) in operated patients and 2/3 (66.6%) in non-operated patients. There was a tendency for mortality to be higher in patients with free perforation than in patients without (10/13, 76.9% vs. 6/11, 54.5%, p = 0.39). All non-operated perforated patients died. Considering patients without free perforation, mortality was 5/7 (71.4%) in those who did not undergo surgery and 1/4 (25%) in operated patients (p = 0.24). The time elapsed between the paracentesis that diagnosed peritonitis and surgery in the 13 operated patients was shorter in the six survivors than in the seven non-survivors (3.2 ± 2.4 vs. 7.2 ± 6.1, (p = 0.31). Tables 3 and 4 show the characteristics of patients with secondary peritonitis according to in-hospital survival. The variables associated with survival included higher mean arterial pressure (p = 0.01) and higher ascitic fluid total protein (p = 0.04). Discussion INR, international normalized ratio; LDH, lactate dehydrogenase. Data are mean ± SD or frequency (%). The true incidence of secondary peritonitis in cirrhotic patients is not well known, although it has been suggested to represent approximately 10% of all peritonitis in this setting [5,8,14]. The present study included 24 patients, the largest series of cirrhotic patients with secondary bacterial peritonitis reported until now, and we confirm the rarity of this complication as it represented 4.5% of all peritonitis in cirrhotic patients with ascites. We could have underestimated the incidence of this complication because we excluded patients with recent abdominal surgery and we considered a positive ascitic fluid culture as necessary diagnostic criteria, a parameter that is not considered by all authors [5,10]. However, as our two centers are tertiary referral hospitals, patients with secondary peritonitis could have been preferentially referred thus increasing the percentage of these patients in the present series. In the present study, patients with secondary peritonitis showed less advanced liver disease than patients with SBP. This finding could have been expected, as SBP occurs spontaneously due to bacterial translocation, mainly in patients with advanced cirrhosis [1 3,26]. In contrast, secondary peritonitis would be more a consequence of chance in a cirrhotic patient with otherwise still preserved liver function [8 10]. The significantly higher incidence of abdominal pain in patients with secondary peritonitis than in patients with SBP was likely the result of a more severe local inflammatory response in the former. This is in agreement with the higher neutrophil count, higher total protein and LDH levels and lower glucose levels in ascitic fluid in these patients, as has been previously reported [8 10,13]. The direct passage of a greater amount of bacteria to the peritoneal cavity from the infectious focus would explain not only this more severe local inflammatory response but also the higher positivity rates of ascitic fluid Gram stain and culture observed in the present study, in contrast with SBP, where the bacterial inoculum is considered to be very low [22]. In agreement with previous studies [2,8 10,12,22,23,27], ascitic fluid culture was more frequently polymicrobial in patients with secondary peritonitis than in SBP episodes. Moreover, anaerobes and/or fungi were more frequently isolated in secondary peritonitis. It is interesting to point out that in spite of the less severe local inflammatory response, patients with SBP showed a more severe hemodynamic impairment reflected by lower mean arterial pressure than patients with secondary peritonitis. 42 Journal of Hepatology 2010 vol. 52 j 39 44

5 Runyon s criteria [8] reflect the more severe local inflammatory response in secondary peritonitis than in SBP, and we observed a sensitivity of 66.6% and specificity of 89.6% for these criteria diagnosing secondary peritonitis, similar to the figures previously reported [9]. Considering patients with secondary peritonitis, 95.6% presented Runyon s criteria and/or polymicrobial ascitic fluid culture, in contrast to only 11.3% of SBP episodes. This combination of parameters could therefore be highly sensitive and specific in detecting suspected secondary peritonitis. In addition, with the automated microbial detection systems currently available, the presence of more than one bacterium in ascitic fluid culture in some cases are known in the first 12 h after paracentesis [27]. However, as the culture frequently takes several days to become positive it would be of interest to find other parameters that might indicate the possibility of secondary peritonitis. Wu et al. found carcinoembryonic antigen and alkaline phosphatase levels in ascitic fluid were useful in the diagnosis of secondary peritonitis due to viscus perforation, although their study included only a small number of cirrhotic patients [10]. The definitive diagnosis of secondary peritonitis relies on imaging techniques, surgery or autopsy findings [8 10,13]. We found no data in the literature regarding the usefulness of abdominal CT specifically in the diagnosis of secondary peritonitis in cirrhosis. In the present study, abdominal CT yielded the diagnosis of secondary peritonitis in most patients in whom it was later confirmed by surgery or autopsy: 12/14 (85.7%). Early diagnosis of secondary peritonitis permits timely initiation of broad-spectrum antibiotic treatment [16,19] and prompt consideration for surgery [2,8 10,16] thereby preventing patient deterioration. Therefore, the early use of appropriate imaging techniques such as abdominal CT seems to be a crucial point in the management of these patients. In the present study, the time elapsed between paracentesis and definitive diagnosis of secondary peritonitis was 2.5 days. We believe this time could be shortened by promoting awareness of this complication, particularly in areas other than gastroenterology departments. Mortality during hospitalization was higher in patients with secondary peritonitis than in patients with SBP (66.6% vs. 26.4%). In the study of Akriviadis and Runyon [9] mortality was 53%; prognosis in cirrhotic patients with secondary peritonitis does not therefore seem to have improved over the last two decades. Mortality was the same in Child-Pugh A or B and Child-Pugh C patients and tended to be lower in patients who were surgically treated than in those who received medical treatment only. This trend towards lower mortality in operated patients was mainly observed in Child-Pugh A or B patients but surgery did not increase mortality in Child-Pugh C patients. These data suggest that when considering survival in cirrhotic patients with secondary peritonitis, solving the intraabdominal source of infection would be more important than the degree of liver insufficiency, contrary to what is observed in patients with SBP [28]. Moreover, the time between diagnostic paracentesis and surgery was shorter in survivors than in non-survivors, although this did not reach statistical significance likely due to the small number of patients. These findings seem to support early consideration for surgical treatment in these patients before the patient s condition deteriorates. When comparing characteristics of survivors and non-survivors with secondary peritonitis, the only significant differences between the two groups were mean arterial pressure and ascitic fluid total protein levels. As the Child-Pugh score was similar in the two groups, the lower mean arterial pressure in non-survivors was probably due to more severe sepsis [1,3,25,26] thus emphasizing that the most important negative variable for the outcome of these patients was the severity of sepsis. Lower ascitic fluid total protein could suggest a more impaired antimicrobial capacity of ascitic fluid [29]. Although our study has several limitations, such as the retrospective design and the possible bias in the selection of patients, we consider it provides valuable clinical information concerning this infrequent complication in cirrhotic patients. Our findings confirm the rarity of secondary peritonitis among cirrhotic patients with peritonitis and its poor prognosis. A low suspicion threshold on the basis of clinical and analytical features, such as Runyon s criteria and microbiological data, and an aggressive approach that includes prompt abdominal CT and early consideration for surgery could improve outcome in these patients. Acknowledgements The authors who have taken part in this study declared that they do not have anything to disclose regarding funding from industries or conflict of interest with respect to this manuscript. We thank Carolyn Newey for English language revision and Ignasi Gich from Clinical Epidemiology Department of Hospital de la Santa Creu i Sant Pau for assistance with statistical analysis. References JOURNAL OF HEPATOLOGY [1] Tandon P, García-Tsao G. Bacterial infections, sepsis, and multiorgan failure in cirrhosis. Semin Liver Dis 2008;28: [2] Guarner C, Soriano G. Spontaneous bacterial peritonitis. Semin Liver Dis 1997;17: [3] Wong F, Bernardi M, Balk R, Christman B, Moreau R, García-Tsao G, et al. Sepsis in cirrhosis: report on the 7th meeting of the International Ascites Club. Gut 2005;54: [4] Lontos S, Gow PJ, Vaughan RB, Angus PW. Norfloxacin and trimethoprimsulfamethoxazole therapy have similar efficacy in prevention of spontaneous bacterial peritonitis. J Gastroenterol Hepatol 2008;23: [5] Fernández J, Navasa M, Gómez J, Colmenero J, Vila J, Arroyo V, et al. Bacterial infections in cirrhosis: epidemiological changes with invasive procedures and norfloxacin prophylaxis. Hepatology 2002;35: [6] Fernández J, Navasa M, Planas R, Montoliu S, Monfort D, Soriano G, et al. Primary prophylaxis of spontaneous bacterial peritonitis delays hepatorenal syndrome and improves survival in cirrhosis. Gastroenterology 2007;133: [7] Sort P, Navasa M, Arroyo V, Aldeguer X, Planas R, Ruíz-del-Árbol L, et al. Effect of intravenous albumin on renal impairment and mortality in patients with cirrhosis and spontaneous bacterial peritonitis. N Engl J Med 1999;341: [8] Runyon BA, Hoefs JC. Ascitic fluid analysis in the differentiation of spontaneous bacterial peritonitis from gastrointestinal tract perforation into ascitic fluid. Hepatology 1984;4: [9] Akriviadis EA, Runyon BA. Utility of an algorithm in differentiating spontaneous from secondary bacterial peritonitis. Gastroenterology 1990;98: [10] Wu SS, Lin OS, Chen YY, Hwang KL, Soon MS, Keeffe EB. Ascitic fluid carcinoembryonic antigen and alkaline phosphatase levels for the differentiation of primary from secondary bacterial peritonitis with intestinal perforation. J Hepatol 2001;34: [11] Pinzello G, Simonetti RG, Craxi A, Di Piazza S, Spanò C, Pagliaro L. Spontaneous bacterial peritonitis: a prospective investigation in predominantly nonalcoholic cirrhotic patients. Hepatology 1983;3: [12] Runyon BA, Hoefs JC. Spontaneous vs. secondary bacterial peritonitis. Differentiation by response of ascitic fluid neutrophil count to antimicrobial therapy. Arch Int Med 1986;146: [13] Caralis PV, Sprung CL, Schiff ER. Secondary bacterial peritonitis in cirrhotic patients with ascites. South Med J 1984;77: Journal of Hepatology 2010 vol. 52 j

6 [14] Castellote J, López C, Gornals J, Tremosa G, Fariña ER, Baliellas C. Rapid diagnosis of spontaneous bacterial peritonitis by use of reagent strips. Hepatology 2003;37: [15] Pieracci FM, Barie PS. Management of severe sepsis of abdominal origin. Scand J Surg 2007;96: [16] Rimola A, García-Tsao G, Navasa M, Piddock LJ, Planas R, Bernard B, et al. Diagnosis, treatment and prophylaxis of spontaneous bacterial peritonitis: a consensus document. J Hepatol 2000;32: [17] Teh SH, Nagorney DM, Stevens SR, Offord KP, Therneau TM, Plevak DJ, et al. Risk factors for mortality after surgery in patients with cirrhosis. Gastroenterology 2007;132: [18] Mehendiratta V, McCarty BC, Gómez L, Graviss EA, Musher DM. Computerized tomography (CT)-guided aspiration of abscesses: outcome of therapy at a tertiary care hospital. J Infect 2007;54: [19] Wong PF, Gilliam AD, Kumar S, Shenfine J, O Dair GN, Leaper DJ. Antibiotic regimens for secondary peritonitis of gastrointestinal origin in adults. Cochrane Database Syst Rev. 2005;(2):CD doi: / CD pub2. [20] Gómez-Jiménez J, Ribera E, Gasser I, Artaza MA, Del Valle O, Pahissa A, et al. Randomized trial comparing ceftriaxone with cefonicid for treatment of spontaneous bacterial peritonitis in cirrhotic patients. Antimicrob Agents Chemother 1993;37: [21] Guarner C, Solà R, Soriano G, Andreu M, Novella MT, Vila MC, et al. Risk of a first community-acquired spontaneous bacterial peritonitis in cirrhotics with low ascitic fluid protein levels. Gastroenterology 1999;117: [22] Runyon BA, Canawati HN, Akriviadis EA. Optimization of ascitic fluid culture technique. Gastroenterology 1988;95: [23] Castellote J, Xiol X, Verdaguer R, Ribes J, Guardiola J, Giménez A, et al. Comparison of two ascitic fluid culture methods in cirrhotic patients with spontaneous bacterial peritonitis. Am J Gastroenterol 1990;85: [24] Follo A, Llovet JM, Navasa M, Planas R, Forns X, Francitorra A, et al. Renal impairment after spontaneous bacterial peritonitis in cirrhosis: incidence, clinical course, predictive factors and prognosis. Hepatology 1994;20: [25] American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992;20: [26] Guarner C, Soriano G. Bacterial translocation and its consequences in patients with cirrhosis. Eur J Gastroenterol Hepatol 2005;17: [27] Ortiz J, Soriano G, Coll P, Novella MT, Pericas R, Sàbat M, et al. Early microbiologic diagnosis of spontaneous bacterial peritonitis with BacT/ ALERT. J Hepatol 1997;26: [28] Toledo C, Salmerón JM, Rimola A, Navasa M, Arroyo V, Llach J, et al. Spontaneous bacterial peritonitis in cirrhosis: predictive factors of infection resolution and survival in patients treated with cefotaxime. Hepatology 1993;17: [29] Runyon BA. Patients with deficient ascitic fluid opsonic activity are predisposed to spontaneous bacterial peritonitis. Hepatology 1988;8: Journal of Hepatology 2010 vol. 52 j 39 44