Abstract
Background
Nosocomial acquisition of spontaneous bacterial peritonitis (SBP) is debated as having a different microbial etiology and prognosis. Identification of clinical, laboratory predictors of mortality and appropriate empirical antimicrobial selection is necessary to prevent early mortality and morbidity. We aimed to find the clinical and bacteriological profile in nosocomial and community acquired SBP and its variants, and the predictors of mortality.
Material and methods
One hundred and fifty patients with 162 discrete episodes of different types of SBP were analyzed. Relevant clinical and laboratory data were analyzed. SBP was diagnosed according to standard criteria and classified as community acquired if the infection detected within 48 h of admission and as nosocomial after 48 h of admission to the hospital.
Results
Eighty seven percent had community acquired SBP (CSBP), 13% had nosocomial SBP (NSBP). Patients of NSBP were older, had more episodes of GI bleed and higher previous episodes of encephalopathy. Patients who died were older, had worse encephalopathy. NSBP had higher one month mortality. Age, serum sodium, encephalopathy and NSBP predicted mortality. Culture positivity was 22.22%. Escherichia coli was the commonest organism isolated. There was no difference in the bacteriological profile between CSBP and NSBP. E. coli showed up to 48% resistance to third generation cephalosporins. Overall sensitivity to aminoglycosides was more than 75%.
Conclusions
Overall mortality was 59%. NSBP had significantly high one month mortality. Age, serum sodium, encephalopathy and NSBP were predictors of mortality. Bacteriological profile was similar between CSBP and NSBP.
Abbreviations: CNNA, culture negative neutrocytic ascites; HCV, Hepatitis C virus; SBP, spontaneous bacterial peritonitis
Keywords: SBP, CSBP, NSBP, ascitic fluid
Spontaneous bacterial peritonitis (SBP) is a frequent and severe complication of cirrhotic ascites.1 SBP is the most common infection in cirrhotics (25%) followed by urinary tract infection (20%) and pneumonia (15%).2
SBP is a landmark event in the natural history of cirrhosis of liver with more than 50% mortality at one year.3 Proper identification and treatment of SBP reduces in hospital and short term mortality significantly.4
Enteric gram negative bacilli are the commonest organisms isolated in ascitic fluid of patients with SBP. However there is increased incidence of gram positive organisms isolated in hospitalized patients with cirrhosis.5 Studies done to assess the importance of acquisition site in prognosis of SBP have shown conflicting results. This study was done to compare the clinical, profile in CSBP and NSBP, predictors of mortality and to find out difference if any in the bacteriological profile between the two groups.
Material and Methods
This is a prospective study done from June 2010 to August 2012. Patients with SBP and its variants admitted to Department of Gastroenterology and Hepatology, St Johns medical college hospital, Bangalore were studied. The study was approved by institutional ethical review board (IERB). An informed written consent was obtained from all the patients. All patients under went detailed clinical examination, laboratory study and abdominal imaging. Ascitic fluid was collected from all the patients during admission at bedside and analyzed for total leukocyte count, differential count along with protein, albumin, glucose, amylase and 10 ml of ascitic fluid was injected into the BACTEC bottle at the bedside for culture and sensitivity. Diagnosis of cirrhosis was made based on clinical, radiological and endoscopic findings.
Past history of SBP, encephalopathy and GI bleed were recorded. Risk factors like EGD, diagnostic paracentesis, therapeutic paracentesis in the preceding one month of admission were noted.
SBP was defined as an ascitic fluid PMN count > 250 with or without positive culture. Patients with ascitic fluid PMN count > 250 and negative culture were labeled as culture negative neutrocytic ascites (CNNA), were grouped with culture positives for analysis. Patients with positive culture and PMN count <250 were labeled as Monomicrobial bacterascites and were excluded from final analysis. SBP was considered community acquired if the infection was detected within 48 h of admission and as nosocomial if it was detected after 48 h after admission to the hospital.6, 7, 8 All the patients were followed up for 6 months from the date of admission, those went against medical advice were called by telephone to determine the status after discharge.
Statistical Analysis
Data were analyzed with statistical software StataIC-12. The results were reported as mean ± SD or Median (Range) for the continuous variables. The Student t-test or Wilcoxon rank sum and Chi-square test were used to compare quantitative data and qualitative variables. Logistic regression was done to assess the clinically and statistically significant factor associated with mortality separately after adjusted odds ratio was calculated. The P-value of <0.05 was considered statistically significant.
Results
One hundred and fifty patients with 162 episodes of spontaneous bacterial peritonitis and its variants were studied among 706 patients of cirrhotic ascites. Incidence of SBP was 22.94%. Of the 162 episodes, 141 episodes (87.04%) were due to CSBP and 21 (12.96%) were secondary to NSBP. Mean age was 48.4 ± 14 and male to female ratio was 6:1.
One hundred and twenty six episodes (77.7%) were due to culture negative neutrocytic ascites (CNNA), rest were culture positive. Among them thirty three were due to spontaneous bacterial peritonitis and three episodes were due to Monomicrobial bacterascites.
Clinical characteristics, laboratory data, risk factors before an episode of SBP, past history and mortality comparing CSBP and NSBP are shown in Table 1.
Table 1.
Clinical, Laboratory Data, Past History, Risk Factors and Mortality in Comparison Between CSBP vs NSBP.
| CSBP n (%) | NSBP n (%) | P value | |
|---|---|---|---|
| Symptom | |||
| Ascites | 130 (92.2) | 13 (61.9) | 0.001 |
| Jaundice | 59 (41.8) | 12 (57.1) | 0.18 |
| Pain abdomen | 60 (42.5) | 11 (52.38) | 0.39 |
| GI bleed | 37 (26.3) | 6 (28.6) | 0.01 |
| Fever | 50 (35.5) | 9 (42.7) | 0.22 |
| Vomiting | 13 (9.2) | 5 (23.8) | 0.04 |
| Diarrhea | 23 (16.3) | 0 (0) | 0.04 |
| Renal impairment | 34 (24.1) | 3 (14.2) | 0.41 |
| Altered sensorium | 37 (26.2) | 8 (38) | 0.29 |
| Mean Age (years) | 48 | 54 | 0.07 |
| Past history | |||
| SBP | 19 (13.4) | 3 (15) | 0.73 |
| GI bleed | 29 (20.7) | 6 (28.5) | 0.41 |
| Encephalopathy | 13 (9.2) | 5 (23.8) | 0.04 |
| Norfloxacin prophylaxis | 18 (12.7) | 3 (15) | 0.7 |
| Clinical signs | |||
| Pallor | 64 (45.7) | 11 (52.4) | 0.57 |
| Edema | 108 (76.6) | 17 (80.95) | 0.79 |
| Septic shock | 7 (4.9) | 1 (4.7) | 1 |
| Hypothermia | 2 (1.4) | 1 (4.7) | 0.34 |
| Encephalopathy | 47 (33.3) | 7 (33.3) | 1 |
| Splenomegaly | 38 (26.9) | 3 (14.2) | 0.28 |
| Abdominal tenderness | 28 (20) | 4 (19) | 1 |
| BMI | 22.9 ± 4.59 (mean ± SD) | 25.5 ± 4.23 (mean ± SD) | 0.01 |
| Laboratory data | |||
| PT seconds | 21.5 ± 8.1 | 20.1 ± 5.9 | 0.47 |
| INR | 1.7 (0.9–10.1) | 1.7 (1.1–3.0) | 0.57 |
| Creatinine (mg/dl) | 1.2 (0.5–6.0) | 1.3 (0.7–4.2) | 0.56 |
| Serum sodium (mEq/l) | 128.7 ± 6.79 | 130.1 ± 6.77 | 0.34 |
| Total bilirubin (mg/dl) | 3.4 (0.2–41.3) | 7.2 (0.5–34.04) | 0.06 |
| Albumin (g/dl) | 1.85 ± 0.58 | 1.92 ± 0.47 | 0.58 |
| CTP | |||
| A | 7 (5.1) | 0 (0) | 0.5 |
| B | 27 (19.7) | 5 (25) | |
| C | 103 (75.2) | 15 (75) | |
| MELD score | 21.68 ± 8.54 | 23.50 ± 8.93 | 0.38 |
| Risk factor | |||
| EGD | 10 (7.1) | 8 (38.1) | <0.001 |
| Diagnostic paracentesis | 17 (12.06) | 12 (57.1) | <0.001 |
| Others | 3 (2.13) | 0 (0) | 1 |
| Mortality | |||
| In hospital | 16 (22.2) | 31 (43) | 0.01 |
| Discharge to 1 month | 25 (34.8) | 10 (62.5) | 0.01 |
| 1 month to 6 months | 31 (43) | 1 (6.3) | 0.01 |
Patients of NSBP were older, had more episodes of GI bleed, more episodes of vomiting and higher previous encephalopathy episodes compared to CSBP. Patients with CSBP presented with diarrhea more often than NSBP. Before an episode of SBP, endoscopic interventions and diagnostic paracentesis were more in NSBP compared to CSBP. Other risk factors were ERCP, ruptured urethra and pleural tap. Ascites was common in CSBP than NSBP. Few patients of NSBP developed ascites after admission and later showed SBP. CSBP had lower BMI in spite of more ascites compared to NSBP. Patients with NSBP had higher bilirubin compared to CSBP. Patients with NSBP had higher amount of alcohol use per day and higher cumulative dose compared to CSBP.
Etiologies for liver cirrhosis are depicted in Table 2. Ethanol was the commonest etiology followed by cryptogenic cirrhosis. Other etiologies were, Hepatitis B virus (HBV) in 8.6% (n = 14), Hepatitis C virus (HCV) in 4.3% (n = 7), Combination of HBV and ethanol in 2.4% (n = 4), Budd chairi syndrome in 3% (n = 5) and others in 4.3% (n = 7). There was no statistically significant difference between CSBP vs NSBP and culture positive vs culture negative with respect to etiology of cirrhosis.
Table 2.
Common Etiologies of Liver Cirrhosis in CSBP vs NSBP and Culture Positive vs Culture Negative.
| Etiology of cirrhosis | CSBP n (%) |
NSBP n (%) |
P value | Culture positive n (%) |
Culture negative n (%) |
P value |
|---|---|---|---|---|---|---|
| Ethanol | 85 (60.3) | 10 (47.6) | 0.27 | 25 (69.4) | 70 (55.6) | 0.18 |
| Cryptogenic | 24 (17) | 4 (19) | 0.76 | 6 (16.7) | 22 (17.5) | 1 |
| HBV | 11 (7.8) | 3 (14.3) | 0.39 | 2 (5.6) | 12 (9.5) | 0.73 |
HBV – Hepatitis B virus.
Overall mortality was 59% at six months. Clinical characteristics, laboratory data, risk factors and past history comparing those who survived and died are shown in Table 3. Patients who died were older, had more renal impairment and had worse sensorium than those who survived. Patients who survived presented with fever more often than those who died. Patients who died had more endoscopic interventions, had higher number of patients with edema, encephalopathy, had higher bilirubin, creatinine, higher number of patients with CTP class C and lower serum sodium compared to those who survived. NSBP had significantly higher early mortality (less than one month) compared to CSBP. There was no statistically significant difference in mortality between culture positives and negatives.
Table 3.
Clinical, Laboratory Data, Past History and Risk Factors in Comparison Between Survivors and Dead.
| Survived n (%) |
Died n (%) |
P value | |
|---|---|---|---|
| Symptom | |||
| Ascites | 63 (85.1) | 80 (90.9) | 0.25 |
| Jaundice | 28 (37.8) | 43 (48.9) | 0.16 |
| Pain abdomen | 32 (43.2) | 39 (44.3) | 0.89 |
| GI bleed | 24 (32.4) | 19 (21.6) | 0.29 |
| Fever | 33 (44.6) | 26 (29.6) | 0.04 |
| Vomiting | 6 (8.1) | 12 (13.6) | 0.26 |
| Diarrhea | 11 (14.9) | 12 (13.6) | 0.82 |
| Renal impairment | 11 (14.9) | 26 (29.5) | 0.02 |
| Altered sensorium | 14 (18.9) | 31 (35.2) | 0.02 |
| Mean Age (years) | 45 | 51 | 0.002 |
| Past history | |||
| SBP | 8 (11) | 14 (15.9) | 0.36 |
| GI bleed | 17 (23) | 18 (20.7) | 0.72 |
| Encephalopathy | 5 (6.8) | 13 (14.9) | 0.1 |
| Norfloxacin prophylaxis | 7 (9.4) | 14 (15.9) | 0.3 |
| Clinical signs | |||
| Pallor | 40 (54.8) | 77 (39.8) | 0.47 |
| Edema | 52 (70.3) | 73 (83) | 0.05 |
| Septic shock | 1 (1.3) | 7 (8) | 0.07 |
| Hypothermia | 0 (0) | 3 (3.4) | 0.25 |
| Encephalopathy | 13 (17.6) | 41 (57.6) | <0.001 |
| Splenomegaly | 19 (25.7) | 22 (25) | 0.92 |
| Abdominal tenderness | 14 (19.2) | 18 (20.5) | 0.84 |
| BMI | 23.10 ± 4.74 (mean ± SD) | 23.52 ± 4.47 (mean ± SD) | 0.54 |
| Laboratory data | |||
| PT seconds | 20.05 ± 6.7 | 22.35 ± 8.7 | 0.06 |
| INR | 1.6 (0.9–10.1) | 1.8 (1.1–5.8) | 0.11 |
| Creatinine (mg/dl) | 1.2 (0.6–3.3) | 1.35 (0.5–6.0) | 0.02 |
| Serum sodium (mEq/l) | 131 ± 5 | 126.9 ± 7.4 | <0.001 |
| Total bilirubin (mg/dl) | 3.25 (0.20–23.40) | 4.30 (0.50–41.3) | 0.009 |
| Albumin (g/dl) | 1.94 ± 0.64 | 1.79 ± 0.49 | 0.22 |
| CTP | |||
| A | 6 (8.3) | 1 (1.2) | 0.02 |
| B | 18 (25) | 14 (16.5) | 0.02 |
| C | 48 (66.7) | 70 (82.4) | 0.02 |
| MELD score | 21.52 ± 8.9 | 22.23 ± 8.36 | 0.6 |
| Risk factor | |||
| EGD | 4 (5.4) | 14 (16.1) | 0.04 |
| Diagnostic paracentesis | 9 (12.2) | 20 (22.7) | 0.08 |
| Others | 2 (2.7) | 1 (1.1) | 0.59 |
Univariate and multivariate logistic regression of predictors of mortality are shown in Table 4. In univariate analysis age, serum sodium, serum creatinine, encephalopathy, bilirubin, NSBP, history of endoscopic procedure and AST predicted mortality. However on multivariate logistic regression age, serum sodium and encephalopathy predicted mortality.
Table 4.
Mortality Predictors in Univariate and Multivariate Regression.
| Variable | Odds ratio | Confidence interval | P value |
|---|---|---|---|
| Univariate analysis | |||
| Age | 1.03 | 1.01–1.06 | 0.003 |
| Serum sodium | 0.89 | 0.85–0.95 | 0.00001 |
| Serum creatinine | 1.98 | 1.27–3.08 | 0.002 |
| Total Bilirubin | 1.08 | 1.02–1.13 | 0.004 |
| AST | 1.003 | 1.00–1.007 | 0.037 |
| History of EGD | 3.35 | 1.05–10.69 | 0.041 |
| Hepatic encephalopathy | 4.09 | 1.97–8.49 | 0.001 |
| Nosocomial SBP | 3.06 | 1.06–8.82 | 0.03 |
| Multivariate regression | |||
| Age | 1.029 | 1.003–1.057 | 0.02 |
| Serum sodium | 0.92 | 0.866–0.977 | 0.007 |
| Serum creatinine | 1.169 | 0.69–1.972 | 0.559 |
| Total bilirubin | 1.055 | 0.966–1.118 | 0.06 |
| History of EGD | 2.9 | 0.81–10.36 | 0.1 |
| Hepatic encephalopathy | 2.73 | 1.19–6.29 | 0.01 |
| Nosocomial SBP | 2.53 | 0.71–9.00 | 0.15 |
Ascitic fluid positive culture yield was 22.22% (36 patients). Thirty three had SBP (20.37%). Three patients had Monomicrobial bacterascites (1.8%). The type of organisms in CSBP and NSBP are shown in Table 5. Escherichia coli was the commonest organism isolated. S. aureus or MRSA was not isolated in any of the patients. Culture positivity in CSBP was 22.69% and in NSBP was 19.04% (P = 0.59). No significant difference noted in bacteriological profile of CSBP and NSBP. All patients (n = 3) with Monomicrobial bacterascites were CSBP, however they were excluded from final analysis. Even though small number of culture positives in NSBP and E. coli was the commonest organism isolated, 50% isolates were due to gram positives.
Table 5.
Number and Percentage of Organisms Isolated in CSBP and NSBP.
| Organism | CSBP (n = 29) (%) |
NSBP (n = 4) (%) |
|---|---|---|
| E. coli | 17 (58.6) | 2 (50) |
| Klebsiella spp. | 4 (13.8) | 0 (0) |
| Non-fermenting GNB | 2 (6.9) | 0 (0) |
| α-Hemolytic Streptococci | 2 (6.9) | 1 (25) |
| CoNS | 3 (10.3) | 1 (25) |
| Enterobacter spp. | 1 (3.4) | 0 (0) |
P = 0.59.
GNB – gram negative bacilli.
CoNS – coagulase negative staphylococci.
Antimicrobial susceptibility of common organisms are depicted in supplement 1. Alfa-hemolytic Streptococci showed 100% sensitivity to penicillin, gentamicin, vancomycin and Teicoplanin. Since the number of culture positives in NSBP group was small, intergroup comparison between CSBP and NSBP with respect to drug susceptibility and resistance was not possible. The resistance pattern of organisms to various drugs as follows. E. coli showed 75% resistance to ciprofloxacin, around 50% resistance to third generation cephalosporins. Klebsiella showed around 25% resistance to third generation cephalosporins. Non fermenting GNB showed around 50% resistance to third generation cephalosporins.
Discussion
The incidence of SBP in our study of 162 episodes in 150 patients was 22.9%. CNNA was more common, and culture positivity was seen in 22.2%. Eighty seven percent had CSBP and 13% had NSBP. Patients with NSBP were older, had more episodes of GI bleed, higher previous episodes of encephalopathy, more diagnostic paracentesis and more endoscopic procedures compared to CSBP. Patients with NSBP had higher amount of alcohol intake compared to CSBP. Both the groups did not show any difference in previous episodes of SBP or patients on norfloxacin prophylaxis. Age, serum sodium and hepatic encephalopathy were the predictors of mortality in multivariate logistic regression. NSBP had significant high one month mortality and was a predictor of mortality in both univariate and multivariate analysis. E. coli was the common organism isolated in 58.3% patients. There was no statistically significant difference in bacteriological profile between CSBP and NSBP. Third generation cephalosporin resistance was noted in up to 48% for E. coli. Overall aminoglycoside sensitivity was seen in up to 75%.
There are limited studies comparing CSBP and NSBP from India. Mixed results are largely due to geographical differences in etiology with hepatitis B virus common in Asia and alcohol in the west, ethanol was the commonest etiology in our series which is similar to other Indian studies of SBP, however Indian studies comparing CSBP and NSBP are lacking.7, 8, 10, 11, 12
No clinical feature is unique to CSBP or NSBP. In study by Cheong et al. fever was more common in NSBP and abdominal pain was more common in CSBP, although there were no significant other differences.7 Our study vomiting was more common in NSBP and diarrhea in CSBP. In our study patients who survived presented with more fever episodes than who died, underlining the fact that fever likely triggers more rapid analysis of ascitic fluid and hence early diagnosis and treatment.
Study by Kim et al., patients of NSBP had lower serum sodium compared to CSBP but there were no difference in CTP scores or mortality.8 In our study we did not notice any difference in laboratory parameters between CSBP and NSBP except NSBP had higher bilirubin.
Patients with NSBP had more episodes of GI bleed and endoscopic interventions and higher prior diagnostic paracentesis compared to CSBP suggesting higher interventions contributing to NSBP. These interventions had valid indications including diagnostic and therapeutic. However there was no difference in the isolated bacteriological profile of organisms isolated between CSBP and NSBP.
Patients who died had severe liver dysfunction having more patients with CTP class C. which is a consistent predictor of mortality in all most all studies. In addition, we found older age, encephalopathy, low sodium was associated with higher mortality and NSBP had higher one month mortality. Another observation noted was that patients with NSBP had higher overall mortality than CSBP if gram positive organism was isolated more and there was no difference in mortality in studies where gram negative organism was isolated more.7, 8, 9, 10
Campillo et al., described 70% NSBP were due to gram positive organisms, 24% from Methicillin resistant staph aureus (MRSA) infection. Mortality was higher in NSBP and especially with MRSA infection.9 Our study showed that NSBP had significantly higher 30 days mortality compared to CSBP. NSBP itself was a predictor of mortality. In a similar study by Cheong et al., a third of whom had third generation cephalosporin resistance.7 Some studies did not show difference in the mortality between CSBP and NSBP.10
E. coli was the commonest organism isolated in our study followed by Klebsiella, with resistance to third generation cephalosporin in up to 48%, which is similar to other studies.7, 10, 11, 12
There is increased isolation of gram positive organisms in SBP in west, particularly in patients on quinolone prophylaxis.13 There was no statistically significant difference noted between CSBP and NSBP in our study with respect to number of patients on norfloxacin prophylaxis, however they were seen in up to 15% patients and most of the infections were first episodes. This may explain gram negative predominance in our study.
A recent study from Italy showed a higher incidence of gram positive organisms (62%), with response to a combination of Meropenem and Daptomycin and a recommendation was made to use it as an empirical treatment.14 In our study Meropenem sensitivity to E. coli was seen in 66.6% and for Piperacillin tazobactum in 76%. Even though small numbers up to 50% of NSBP showed gram positives, larger study with more NSBP is required to suggest empirical gram positive cover in Indian patients. The variable antibiotic sensitivity calls empirical antibiotic choice to be made according to local sensitivity pattern.
Antimicrobial sensitivity of organisms are shown in supplement 1. Klebsiella spp. showed 25% resistance to third generation cephalosporins and quinolones in our study. All patients with CoNS (coagulase negative staphylococci) were sensitive to aminoglycosides in our study even though they showed variable resistance to other antibiotics. Overall Aminoglycoside sensitivity was seen in more than 75%. A study done by Puri et al., it was noticed that all patients with monomicrobial bacterascites were due to gram positive cocci, however we found gram negative predominance in the organisms isolated in monomicrobial bacterascites.15
Our study has several unique features like including a large number of patients of NSBP with alcoholic cirrhosis. There is a concern of NSBP with increasing interventions in hospitalized patients. Our study showed that apart from severe liver dysfunction which predict long term mortality, NSBP predict one month mortality. Awareness and proper early intervention lead to better survival. Enteric gram negatives like E. coli were still the commonly isolated organisms unlike the west where there is a trend of gram positive isolates. Since there is regional variation in bacteriological profile and sensitivity, empirical antibiotics to be chosen according to local sensitivity patterns. Limitations of our study included non-availability of species identification of non-fermenting gram-negative bacilli.
Conclusions
Patients of NSBP were older, had more episodes of GI bleed, endoscopic interventions and previous episodes of encephalopathy. Patients who died had worse liver function. Age, serum sodium, encephalopathy were predictors of mortality. Patients of NSBP had higher one month mortality, and itself was a predictor of mortality. CSBP and NSBP showed similar bacteriological profile. E. coli was the common organism isolated in both. Third generation cephalosporin resistance was seen in 48%. Empirical antibiotics to be chosen according to local sensitivity pattern.
Ethical Approval
This study was approved by Institutional ethical review board of St Johns Medical College Hospital, Bangalore.
Conflicts of Interest
The authors have none to declare.
Footnotes
Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jceh.2017.03.005.
Appendix A. Supplementary data
The following are the supplementary data to this article:
References
- 1.Conn H.O. Spontaneous peritonitis and bacteremia in Laennec's cirrhosis caused by enteric organisms. A relatively common but rarely recognized syndrome. Ann Int med. 1964;60(4):568–580. doi: 10.7326/0003-4819-60-4-568. [DOI] [PubMed] [Google Scholar]
- 2.Fernandez J., Navasa M., Gomez J., Colmenero J., Vila J., Arroyo V. Bacterial infections in cirrhosis: epidemiological changes with invasive procedure and norfloxacin prophylaxis. Hepatology. 2002;35(1):140–148. doi: 10.1053/jhep.2002.30082. [DOI] [PubMed] [Google Scholar]
- 3.Altman C., Grange J.D., Amiot X., Pelletier G., Lacaine F., Bodin F. Survival after a first episode of spontaneous bacterial peritonitis. Prognosis of potential candidates for orthotopic liver transplantation. J Gastroenterol Hepatol. 1995;10(1):47–50. doi: 10.1111/j.1440-1746.1995.tb01046.x. [DOI] [PubMed] [Google Scholar]
- 4.Garcia-Tsao G. Current management of the complications of cirrhosis and portal hypertension: variceal hemorrhage, ascites, and spontaneous bacterial peritonitis. Gastroenterology. 2001;120:726–748. doi: 10.1053/gast.2001.22580. [DOI] [PubMed] [Google Scholar]
- 5.Campillo B., Dupeyron C., Richardet J.P. Epidemiology of hospital acquired infections in cirrhotic patients: effect of carriage of methicillin-resistant Staphylococcus aureus and influence of previous antibiotic therapy and norfloxacin prophylaxis. Epidemiol Infect. 2001;127:443–450. doi: 10.1017/s0950268801006288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bert F., Andreu M., Durand F., Degos F., Galdbart J.O., Moreau R. Nosocomial and community acquired spontaneous bacterial peritonitis: comparative microbiology and therapeutic implications. Eur J Clin Microbiol Infect Dis. 2003;22(1):10–15. doi: 10.1007/s10096-002-0840-z. [DOI] [PubMed] [Google Scholar]
- 7.Cheong H.S., Kang C.I., Lee J.A., Moon S.Y., Joung M.K., Chung D.R. Clinical significance and outcome of nosocomial acquisition of spontaneous bacterial peritonitis in patients with live cirrhosis. Clin Infect Dis. 2009;48(9):1230–1236. doi: 10.1086/597585. [DOI] [PubMed] [Google Scholar]
- 8.Kim S.U., Chon Y.E., Lee C.K., Park J.Y., Kim D.Y., Han K.H. Spontaneous bacterial peritonitis in patients with hepatitis B virus-related liver cirrhosis: community-acquired versus nosocomial. Yonsei Med J. 2012;53(2):328–336. doi: 10.3349/ymj.2012.53.2.328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Campillo B., Richardet J.P., Kheo T., Dupeyron C. Nosocomial spontaneous bacterial peritonitis and bacteremia in cirrhotic patients: impact of isolate type on prognosis and characteristics of infection. Clin Infect Dis. 2002;35:1–10. doi: 10.1086/340617. [DOI] [PubMed] [Google Scholar]
- 10.Song J.Y., Jung S.J., Park C.W., Sohn J.W., Kim W.J., Kim M.J. Prognostic significance of infection acquisition sites in spontaneous bacterial peritonitis: nosocomial versus community acquired. J Korean Med Sci. 2006;21:666–671. doi: 10.3346/jkms.2006.21.4.666. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Purohith P.H., Malek S.S., Desai K.J., Sadadia M. A study of bacteriological profile of ascitic fluid in suspected clinical cases of spontaneous bacterial peritonitis at a tertiary care hospital in India. Int J Med Sci Public Health. 2015;4(4):496–501. [Google Scholar]
- 12.Bhat G., Vandana K.E., Bhatia S., Suvarna D., Pai C.G. Spontaneous ascitic fluid infection in liver cirrhosis: bacteriological profile and response to antibiotic therapy. Indian J Gastroenterol. 2013;32(5):297–301. doi: 10.1007/s12664-013-0329-y. [DOI] [PubMed] [Google Scholar]
- 13.Alexopoulou A., Papadopoulos N., Eliopoulos D.G., Alexaki A., Tsiriga A., Toutouza M. Increasing frequency of gram-positive cocci and gram-negative multidrug-resistant bacteria in spontaneous bacterial peritonitis. Liver Int. 2013;33(7):975–981. doi: 10.1111/liv.12152. [DOI] [PubMed] [Google Scholar]
- 14.Piano S., Fasolato S., Salinas F., Romano A., Tonon M., Morando F. The empirical antibiotic treatment of nosocomial spontaneous bacterial peritonitis: results of a randomized, controlled clinical trial. Hepatology. 2016;63(4):1299–1309. doi: 10.1002/hep.27941. [DOI] [PubMed] [Google Scholar]
- 15.Puri A.S., Puri J., Ghoshal U.C., Sharma B.C., Saraswat V.A., Ayyagari A. Frequency, microbial spectrum and outcome of spontaneous bacterial peritonitis in north India. Indian J Gastroenterol. 1996;15(3):86–89. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.