Factors Affecting Mortality and Morbidity of Patients With Cirrhosis Hospitalized for Spontaneous Bacterial Peritonitis

POSter PreSentAtIOn Factors Affecting Mortality and Morbidity of Patients With Cirrhosis Hospitalized for Spontaneous Bacterial Peritonitis Fatih Ensaroğlu, 1 Murat Korkmaz, 1 Ali Ümit Geçkil, 2 Serkan Öcal, 1 Bengisu Koç, 3 Özgün Yıldız, 3 Fatma Büşra Atalay, 3 Emine Gül Taş, 3 Mehmet Haberal 4 Abstract Objectives: Spontaneous bacterial peritonitis, unless originating from surgery or an intra-abdominal source, is an infection diagnosed by neutrophil counts greater than 250/mm 3 in ascites. Spontaneous bacterial peritonitis is the most common infection among patients hospitalized with cirrhosis, with a prevalence of 9% and a risk of development among all patients with cirrhosis within 1 year of 10%. No valid parameters have been defined to predict the mortality related to spontaneous bacterial peritonitis. Unless it is treated, the mortality rate as a result of spontaneous bacterial peritonitis is 50%, and serious complications may arise. Materials and Methods: Medical records from 29 patients on the deceased-donor transplant waiting list and receiving treatment at the Başkent University Hospital Gastroenterology Clinic for cirrhotic ascites infection between 1996 and 2013 were analyzed. Demographic information, para - centesis findings, clinical follow-up, and treatment results were reviewed and collected from patient medical records, with data recorded to the research form. Results: In our patient group, 72.4% were men and the average age was 46.6 years. Most of our patients were at advanced stage, with 55.2% having From the 1 Department of Gastroenterology, Baskent University Faculty of Medicine; the 2 Department of Family Medicine, Baskent University Faculty of Medicine; the 3 Baskent University Faculty of Medicine; and the 4 Department of General Surgery, Baskent University Faculty of Medicine, Ankara, Turkey Acknowledgements: The authors declare that they have no sources of funding for this study, and they have no conflicts of interest to declare. Corresponding author: Fatih Ensaroğlu, Department of Gastroenterology, Baskent University Faculty of Medicine, Fevzi Çakmak Cad, Ankara 06490, Turkey Phone: +90 312 203 6868 Fax: +90 312 223 7333 E-mail: fatihensaroglu@gmail.com Experimental and Clinical Transplantation (2015) Suppl 3: 131-136 a Child-Pugh score of C and an average Model for End-Stage Liver Disease score of 17 ± 4.1. We found that 34.5% of the patients received prophylactic treatment for spontaneous bacterial peritonitis, 72.4% received a proton pump inhibitor, and 82.8% had treatment with intravenous albumin support at the time of diagnosis. Albumin treatment showed no effect on mortality. Mortality rate among patients with Child-Pugh score of C was 81.3%. Conclusions: Existence of chronic renal failure, liver graft surgery, and hepatocellular cancer did not seem to have a significant effect on patient mortality. The albumin treatment protocol showed no significant difference despite common belief among physicians. Key words: Ascites, Albumin treatment, Liver graft surgery, Child-Pugh score Introduction Spontaneous bacterial peritonitis (SBP) is an infection diagnosed by a neutrophil count of greater than 250 neutrophils/μl3 in ascites unless it originates from surgery or an intra-abdominal source. 1 No valid parameters to predict SBP-related mortality have been identified. 2 However, if not treated, SBP can be fatal in 50% of patients and serious complications may arise. Spontaneous bacterial peritonitis is a common bacterial infection among patients with cirrhosis, with a prevalence of 10% to 30% in hospitalized patients and 1.5% to 3.5% in patients who are not hospitalized. 3 Spontaneous bacterial peritonitis is one of the leading causes of mortality and morbidity. 4 All patients with cirrhosis and ascites are at risk of developing SBP. One-year mortality after first diagnosis of SBP is shown to be between 31% and 93%. 5 However, early diagnosis and early Copyright Başkent University 2015 Printed in Turkey. All Rights Reserved. DOI: 10.6002/ect.tdtd2015.P71

132 Fatih Ensaroğlu et al/experimental and Clinical Transplantation (2015) Suppl 3: 131-136 Exp Clin Transplant antibiotic treatment has decreased this rate to 20% to 30% in recent years. 1,5 Spontaneous bacterial peritonitis is a mono - microbial infection of ascites, which may occur by infection, translocation, or hematogenous dissemination of intestinal flora. 4 Bacterial trans location is the major mechanism of SBP. 3 Moreover, excess bacterial production in the intestine triggers trans - location. 4 The frequency of ascites development in patients with compensated cirrhosis in a 5-year span is about 30%. Prognosis of patients with cirrhosis significantly worsens after ascites development. Although the 1-year survival rate in patients with compensated cirrhosis is above 90%, the rate decreases to 50% in decompensated patients. After ascites development, 10% to 30% of patients develop SBP within 1 year. Patients who had 1 SBP attack and survived are at risk of relapsing within 1 year. 6 About 83% to 93% of patients recover from their first SBP occurrence when treated with cefotaxime, a third-generation cephalosporin. Mortality of patients treated in a hospital setting with cefotaxime was found to be 29%, a consistent rate in many other studies. 4-6 However, for patients treated in the hospital setting with both cefotaxime and albumin, the mortality decreased to 10%. Albumin may decrease the risk of hepatorenal syndrome development, a serious complication, 6 by increasing the effective circulating blood volume. Serum albumin concentration is responsible for 75% of plasma osmotic pressure. As a result, plasma is used for volume replacement. Moreover, albumin is a multifunctional protein that plays a role in anti - oxidation, immunomodulation, and detoxification. Liver damage, which prevents albumin to act as an immunomodulator, is 1 factor leading to SBP development. Hemodynamic disorders and multiorgan failure may occur as a result of hypo - albuminemia. Studies show that albumin infusion given to patients with SBP during antibiotic treatment decreases renal failure incidence and mortality. 7 In a study by Cho and associates, 8 patients with SBP were given intravenous albumin treatment (1.5 g/kg/d). This treatment resulted in patients not needing the withdrawal of large amounts of fluid (> 5 L) during paracenteses. 8 The production of C-reactive protein (CRP) in the body is affected by liver function; however, its clinical importance among patients with cirrhosis has not been clarified. In a study showing the relation between serum CRP levels and efficiency of treatment among 182 patients, 54.2% of the patients who died from SBP had responded to antibiotic therapy. It is also shown that high CRP levels cause low response to antibiotic therapy. 8 The use of prophylactic treatment in patients with ascites and SBP is an important issue. Only patients at high risk for development of SBP should receive prophylactic treatment because the long-term use of antibiotics for prophylactic use can cause resistance to bacterial emergence. 9 Pathogens responsible for SBP are usually gramnegative bacteria that have translocated from the intestines. These pathogens include Escherichia coli, Klebsiella pneumonia, and Streptococcus pneumonia. As a result, norfloxacin and trimethoprim-sulfa - methoxazole are appropriate for prophylaxis. 9 Most centers, including ours, prefer norfloxacin as firstline treatment. Bacterial overgrowth and decreased small intestine motility are more common among patients with SBP than in those without the disease. Bacterial colonization is much easier because the immune systems of patients with SBP are weaker. 10 Proton pump inhibitors and H2 receptor antagonists are drugs used to suppress gastric acid. Treatment of patients with cirrhosis with acid suppressors can cause bacterial overgrowth and translocation. Risk of SBP development in patients with cirrhosis who are not being treated with proton pump inhibitors is 3 times more likely than in those who are receiving this treatment. 11 Although some studies have reported that the risk is not increased, 10 indications for proton pump inhibitors have to be questioned first. When to start and whether long-term use is appropriate require careful consideration. 11 In this study, we aimed to outline our experiences with SBP in patients with cirrhosis. Materials and Methods We reviewed the hospital registry database of patients with cirrhotic ascites infection who had received inpatient treatment at the Başkent University Hospital Gastroenterology Clinic between 1995 and 2013. The following data were collected: demographics; cause of cirrhosis (hepatitis B virus, hepatitis C virus, alcohol, metabolic, vascular, cryptogenic); reason for hospital admission

Fatih Ensaroğlu et al/experimental and Clinical Transplantation (2015) Suppl 3: 131-136 133 (shortness of breath, renal failure, decompensated ascites, abdominal distension, fever, stomach ache, change in cognitive status, asymptomatic, coincidence); examination findings (blood pressure, pulse, body temperature, respiration, abdominal sensitivity, rebound sensitivity, encephalopathy); existence of systemic disease and hepatocellular carcinoma; Child-Pugh and Model for End-Stage Liver Disease scores; serum values (prothrombin time, leukocyte count, total protein, international normalized ratio, and serum albumin, serum urea nitrogen, creatinine, and sodium levels); serum ascites albumin gradient; CRP levels (before and after treatment); paracentesis findings (ascites protein, ascites albumin, ascites leukocyte level, ascites neutrophil level, culture); diagnosis time of cirrhosis and SBP; treatment; existence of control paracentesis after treatment; existence of complications during therapy; intensive care admission history; status of prophylaxis; whether albumin was given; use of proton pump inhibitor; and time of death or whether death did not occur. Exclusion criteria were presence of chronic renal failure, secondary peritonitis, patients who were on peritoneal dialysis, and missing information. Patients were reviewed retrospectively. The study was approved by the ethics committee, and all protocols conformed with the ethical guidelines of the 1975 Helsinki Declaration. Patients provided written informed consent to be included on the registry. Statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 17.0, IBM Corporation, Armonk, NY, USA). Continuous variables are presented as average ± SD (or median and range from minimum to maximum), whereas categorical variables are presented as numbers and percentage. The ratios of patient mortality, cause of cirrhosis, history of liver transplant surgery, existence of hepatocellular carcinoma, Child-Pugh score, Model for End-Stage Liver Disease score, use of albumin replacement, use of proton pump inhibitors and cephalosporin treatment, and whether patients received prophylactic treatment were compared within each other using chi-square test. Wilcoxon rank sum test was used to compare CRP levels before and after treatment and paracentesis findings (the presence of proteins, albumin, leukocytes, neutrophils, culture). P <.05 was considered statistically significant in all analyses. Results In our study, 40 of 351 patients with cirrhosis (11.3%) at the Gastroenterology Clinic at Başkent University Hospital developed SBP. After application of exclusion and inclusion criteria, 29 patients were included in our study. General characteristics of patients and reasons for hospital admission are listed in Tables 1 and 2. The demographic traits and admission reasons of most of the patients in our study. The patient examination results and laboratory findings are listed in Tables 3 and 4. Hepatitis B virus was the leading cause of cirrhosis (34.5% of patients), with hepatitis C virus as the second leading cause (24.1% of patients) (Table 1). Cirrhosis from alcohol use was found in 13.8% of our study patients (Table 1). table 1. Patient Characteristics Characteristic Number or Percent of Patients (N = 29) Age at cirrhosis diagnosis (y) 46.66 ± 19.48 Age at development of spontaneous bacterial peritonitis diagnosis (y) 49.91 ± 18.75 Sex (male) 72.4% Cause of cirrhosis Hepatitis B virus 34.5% Hepatitis C virus 24.1% Hepatitis B virus and hepatitis D virus 0% Alcohol 13.8% Metabolic 17.2% Cryptogenic 34.5% Existence of another disease Hepatocellular carcinoma 27.6% Liver transplant 20.7% Child-Pugh score Class A 6.9% Class B 37.9% Class C 55.2% Model for End-Stage Liver Disease score 17 ± 4.1 Results are shown as means ± SD or percent. table 2. Reason for Hospital Admission Reason for Hospital Admission Percent of Patients (N = 29) Bleeding 13.8 Shortness of breath 34.5 Renal failure 10.3 Ascites 34.5 Abdominal distension 55.2 Fever 37.9 Stomach ache 41.4 Cognitive change 34.5 No symptoms 34.5 Diagnosis of SBP by culture-positive variant occurred in 17% of patients, with the diagnosis by neutrophil count occurring in 83% of patients (mean neutrophil count of 560/μL). The antibiotic regimens included third-generation cephalosporin (24% of patients), other antibiotic regimens (45% of patients),

134 Fatih Ensaroğlu et al/experimental and Clinical Transplantation (2015) Suppl 3: 131-136 Exp Clin Transplant table 3. Patient Examination Results Test Number or Percent of Patients (N = 29) Systolic blood pressure (mm Hg) 115.03 ± 23.53 Diastolic blood pressure (mm Hg) 72.07 ± 13.12 Pulse (beats/min) 87.24 ± 16.6 Body temperature ( C) 36.72 ± 0.9 Respiration (breaths/min) 27.10 ± 18.45 Abdominal sensitivity 51.7% Rebound 20.7% Encephalopathy 51.7% Results are shown as average ± SD or percent. table 4. Patient Laboratory Results Laboratory Test in Serum Average ± SD (N = 29 Patients) PT (seconds) 20.29 ± 7.1 INR 1.72 ± 0.99 Leukocyte (count/µl) 7.58 ± 3.71 Total protein (g/dl) 5.91 ± 1.25 Albumin (g/l) 25.5 ± 6.7 Total bilirubin (µmol/l) 4.83 ± 5.95 Serum urea nitrogen (mmol/l) 29.7 ± 17.7 Creatinine (mg/dl) 1.24 ± 1.3 Sodium (mmol/l) 131.45 ± 6.9 Ascites albumin gradient (mg/dl) 2.15 ± 0.63 C-reactive protein before treatment (nmol/l) 545.7 ± 509.5 C-reactive protein after treatment (nmol/l) 223.05± 203.72 Abbreviations: INR, international normalized ratio; PT, prothrombin time and a combination of third-generation cephalosporin plus other antibiotics (31% of patients). In the 29 patients, 6.9% of patients were scored as Child-Pugh Class A, 37.9% as Child-Pugh Class B, and 55.2% as Child-Pugh Class C, with the high number with C score because patients presented with poor medical conditions. In our study, there was no mortality among patients with Child-Pugh Class A. However, the mortality rate was 81.3% for those with Child-Pugh Class C. In our study group, average Model for End-Stage Liver Disease score was 17 ± 4.1. Review of patient medical records showed a 1-year mortality rate of 44.83%. We found no significant correlation in terms of mortality between patients receiving proton pump inhibitors, thirdgeneration cephalosporin, albumin infusion, or thirdgeneration cephalosporins plus albumin infusion treatment. Discussion According to the literature, after development of SBP, 29% of the patients die within 1 year. 1 Spontaneous bacterial peritonitis is a complication that is life threatening and seen in cirrhotic progress. According to the literature, 8% to 27% of patients admitted with ascites to a hospital have SBP. 12 Hepatitis B virus, hepatitis C virus, hepatitis B and hepatitis D virus, alcohol-related, metabolic, and cryptogenic reasons are among the causes of cirrhosis. Although 1 study has stated that alcohol is the primary reason, another study has claimed hepatitis B virus as the principle cause of cirrhosis. 6,12 These differences may be related to the geographic region where the studies were performed. Although alcohol remains in the foreground versus viral hepatitis in Western Europe and the United States, hepatitis B virus remains the primary reason in Turkey and in other Eastern communities. In our study, the patients presented with advanced stages of cirrhosis and 20.7% had received a liver transplant, with death occurring in 66.7% of patients receiving a transplant. This high rate can be explained by the poor conditions of patients. In a previous study of 40 patients who had developed SBP, 25.7% were score as Child-Pugh Class B and 74.3% with Child-Pugh Class C. In the same study, 27.5% of the patients later developed hepatocellular carcinoma, reflecting the poor condition of patients, similar to our study. 13 Measured levels of CRP at the beginning of treatment, existence of hepatocellular carcinoma, and Child-Pugh scores are independent risk factors for mortality for hospitalized patients. 8 In our study, patients had an average CRP level of 51.7 mg/dl before treatment. After treatment, this average level was 179.05 nmol/l. We had expected a fall in CRP levels at the end of antibiotic treatment. Our findings were statistically significant and supported our expectations (P =.042). C-reactive protein levels increase in response to infection in patients with cirrhosis, although their liver s ability to synthesize decreases. A high CRP level indicates that the infection is more serious than initially predicted and may result in a less successful treatment. The prevalence of hepatocellular carcinoma is high among patients with cirrhosis and SBP. According to a previous study, hepatocellular carcinoma occurs in 41.5% of these patients.1 Moreover, in a similar study, 35 of 168 patients (20.8%) with cirrhosis and SBP were diagnosed with hepatocellular carcinoma. 9 In our study, hepato - cellular carcinoma was found in 27.6% of patients. Albumin increases the effective blood volume and blood flow in the splanchnic area; as a result, this avoids development of hepatorenal syndrome. Albumin administration to patients with serum

Fatih Ensaroğlu et al/experimental and Clinical Transplantation (2015) Suppl 3: 131-136 135 albumin levels below 25 g/l can decrease the probability of developing hepatorenal syndrome. In our study, 82.8% of patients received albumin treatment. The mortality rate in patients who received albumin treatment was 66.7%, whereas this rate was 80% in those who did not receive this treatment. Therefore, albumin replacement treatment in patients with cirrhosis and SBP can decrease the mortality rate. In addition, a related study showed that mortality rate was 16% for patients who had albumin replacement treatment and 35.4% for those who did not have this treatment. 14 Standard treatment of SBP is third-generation cephalosporin (ceftriaxone), administered intra - venously at 1 to 2 grams, 3 times per day, for 5 days. Mortality for hospitalized patients treated with cefotaxime was shown to be 29%, which is similar to the rate shown in most studies on SBP. 9,10 However, the mortality rate for hospitalized patients treated with a combination of cefotaxime and albumin was 10%. 8 According to another study, the mortality rate for hospitalized patients treated with cefotaxime only was 41%; however, this rate was 22% in patients treated with both cefotaxime and albumin. 15 In our study, 24% of patients received treatment with cefotaxime, 45% received antibiotics, and 31% received both cefotaxime and other antibiotics. Patients who were treated with both cephalosporin and albumin had a mortality rate of 75%, whereas this rate was 64.7% in those who were not treated. Because most of our patients had advanced-stage cirrhosis, cephalosporin was given along with albumin. However, the mortality rate was higher than we expected. Prophylactic treatment is given to patients with cirrhosis and ascites who have not developed SBP but are at a high risk of advancing to SBP. Administration of 400 mg norfloxacin 1 time per day is recommended for these patients indefinitely to lower the risk of SBP. 9 In a recent study of norfloxacin in 35 patients with cirrhosis, 14 had infection (40%) and 2 had SBP (6%). This study showed the mortality rate for patients who took prophylaxis to be 29%; however, those who did not had a mortality rate of 39.4%. 16 In our study, 34.5% of patients received prophylactic treatment, with mortality rate in these patients of 70%. In contrast, the rate among patients who did receive prophylactic treatment was 68%. A high mortality rate despite prophylaxis can be generally attributed to the advanced stage of cirrhosis in our patients. In addition, because the numbers of quinolone-resistant bacteria increase in fecal flora of patients, the activity of norfloxacin decreases with time. 16 In a study of 184 patients, 30-day mortality rate was 27% in patients with a Model for End-Stage Liver Disease score of 20. 2 Our study also included patients in poor condition with correspondingly high Model for End-Stage Liver Disease scores. 2 A limitation of our study was the inability to include patients who met the inclusion criteria but had medical records that were inaccessible or had chronic renal failure. In conclusion, most of our patients were at advanced stage of cirrhosis, with treatment efficiency below our expectations. References 1. Tsung PC, Ryu SH, Cha IH, et al. Predictive factors that influence the survival rates in liver cirrhosis patients with spontaneous bacterial peritonitis. Clin Mol Hepatol. 2013;19(2):131-139. 2. Tandon P, Kumar D, Seo YS, et al. The 22/11 risk prediction model: a validated model for predicting 30-day mortality in patients with cirrhosis and spontaneous bacterial peritonitis. Am J Gastroenterol. 2013;108(9):1473-1479. 3. Joseph T, Sobhan P, Bahuleyan S, et al. Non-typhoidal salmonella: an unusual cause of spontaneous bacterial peritonitis in decompensated cirrhosis. Gastroenterol Rep (Oxf ). 2014;2(3):242-244. 4. Sheikhbahaei S, Abdollahi A, Hafezi-Nejad N, Zare E. Patterns of antimicrobial resistance in the causative organisms of spontaneous bacterial peritonitis: a single centre, six-year experience of 1981 samples. Int J Hepatol. 2014;2014:917856. 5. Oladimeji AA, Temi AP, Adekunle AE, Taiwo RH, Ayokunle DS. Prevalence of spontaneous bacterial peritonitis in liver cirrhosis with ascites. Pan Afr Med J. 2013;15:128. 6. Koulaouzidis A, Bhat S, Karagiannidis A, Tan WC, Linaker BD. Spontaneous bacterial peritonitis. Postgrad Med J. 2007;83(980): 379-383. 7. Garcia-Martinez R, Caraceni P, Bernardi M, Gines P, Arroyo V, Jalan R. Albumin: pathophysiologic basis of its role in the treatment of cirrhosis and its complications. Hepatology. 2013;58(5):1836-1846. 8. Cho Y, Park SY, Lee JH, et al. High-sensitivity C-reactive protein level is an independent predictor of poor prognosis in cirrhotic patients with spontaneous bacterial peritonitis. J Clin Gastroenterol. 2014;48(5):444-449. 9. Hsu SJ, Huang HC. Management of ascites in patients with liver cirrhosis: recent evidence and controversies. J Chin Med Assoc. 2013;76(3):123-130. 10. Chang CS, Chen GH, Lien HC, Yeh HZ. Small intestine dysmotility and bacterial overgrowth in cirrhotic patients with spontaneous bacterial peritonitis. Hepatology. 1998;28(5):1187-1190. 11. Deshpande A, Pasupuleti V, Thota P, et al. Acid-suppressive therapy is associated with spontaneous bacterial peritonitis in cirrhotic patients: a meta-analysis. J Gastroenterol Hepatol. 2013;28(2):235-242. 12. Hoefs JC, Canawati HN, Sapico FL, Hopkins RR, Weiner J, Montgomerie JZ. Spontaneous bacterial peritonitis. Hepatology. 1982;2(4):399-407. 13. Musskopf MI, Fonseca FP, Gass J, de Mattos AZ, John JA, de Mello Brandao AB. Prognostic factors associated with in-hospital mortality in patients with spontaneous bacterial peritonitis. Ann Hepatol. 2012;11(6):915-920.

136 Fatih Ensaroğlu et al/experimental and Clinical Transplantation (2015) Suppl 3: 131-136 Exp Clin Transplant 14. Salerno F, Navickis RJ, Wilkes MM. Albumin infusion improves outcomes of patients with spontaneous bacterial peritonitis: a meta-analysis of randomized trials. Clin Gastroenterol Hepatol. 2013; 11(2):123-130. 15. Sort P, Navasa M, Arroyo V, 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(6):403-409. 16. Fernandez J, Navasa M, Planas R, et al. Primary prophylaxis of spontaneous bacterial peritonitis delays hepatorenal syndrome and improves survival in cirrhosis. Gastroenterology. 2007;133(3):818-824.