Prevalence and predictors for spontaneous bacterial peritonitis in cirrhotic patients with ascites admitted at medical block in Korle-Bu Teaching Hospital, Ghana
Amoako Duah1,&, Kofi Nyaako Nkrumah2
1Department of Medicine, St Dominic Hospital, Akwatia, Ghana, 2Department of Medicine and Therapeutictics, School of Medicine and Dentistry, College Of Health Science, University Of Ghana, Korle-Bu, Accra, Ghana
Amoako Duah, Department of Medicine, St Dominic Hospital, Akwatia, Ghana
spontaneous bacterial peritonitis (SBP) is one of the most common and life-threatening complications of patients with cirrhotic ascites. Recognition and prompt treatment of this condition is essential to prevent serious morbidity and mortality. This study aimed to determine the prevalence of SBP among in-patients with cirrhotic ascites attending our facility and to determine the clinical and laboratory parameters associated with SBP.
a cross-sectional study was conducted involving one hundred and three (103) patients admitted at medical block in the Korle-Bu Teaching Hospital (KBTH) with cirrhotic ascites from 25th March, 2016 to 25th November, 2016. Demographic and clinical data were collected using a standardized questionnaire. Ascitic fluid culture and cell count were conducted. Positive ascitic fluid culture and/or ascitic polymorphonuclear leukocyte ≥ 250cells/mm3 were diagnostic for SBP.
of the 103 patients with cirrhotic ascites, the mean age was 43.5 ± 12.2 years. There were fifty eight (58) male patients. The prevalence of SBP was 25.24% (26/103). Majority, 5 (55.6%) of the bacteria isolated from ascitic fluid with SBP was Escherichia coli. Severe ascites and high INR were found to be independent predictors of SBP.
SBP is common among patients with cirrhotic ascites admitted at KBTH. Severe ascites and high INR were highly suggestive of SBP. Diagnostic paracentesis should be done immediately on admission to confirm the diagnosis irrespective of the clinical characteristics as part of baseline investigation.
Spontaneous bacterial peritonitis (SBP) is a common and serious infection occurring in patients with cirrhosis and ascites . Numerous studies suggest that 10-30% of hospitalized patients and 3.5% of outpatients with cirrhosis and ascites have SBP, with in-hospital mortality ranging from 20-40% [2-5]. SBP may be the precipitating factor for the occurrence of kidney failure, hepatic encephalopathy, gastrointestinal bleeding, hypervolemic hyponatremia and development of acute on chronic liver failure, systemic sepsis and poor survival . An increase in the permeability of the intestinal wall leads to translocation of bacteria and subsequent development of SBP [7, 8]. Intestinal bacterial overgrowth and uncontrolled bacterial growth in ascitic fluid then occur, as a result of an impaired host immune response [9, 10]. Factors associated with the risk of developing SBP in cirrhotic patients include upper gastrointestinal bleeding, poor liver function, low ascitic fluid protein levels, prior SBP and hospitalization [11, 12].
Bacteria most commonly isolated from ascitic fluid in patients with SBP are usually those of the normal intestinal flora [13, 14]. More than 92% of all cases are monomicrobial with aerobic gram negative bacilli being responsible for more than two thirds of cases . Escherichia coli accounts for nearly half of these cases followed by Klebsiella species and other gram negative bacteria. Twenty-five percent of cases are caused by gram positive organisms with Streptococcus species being the most common .
The symptoms and signs of SBP are subtle compared with those of patients who have surgical peritonitis in the absence of ascites. SBP may be asymptomatic in about 10-32% of cases, particularly in outpatients [4, 14, 15]. Symptoms and signs patients with SBP normally present with include fever, abdominal pain, altered mental status, abdominal tenderness, diarrhoea, paralytic ileus, hypotension and hypothermia (17%) .
Early diagnosis and prompt management of SBP once it has developed and preventing infections in high risk groups by giving prophylactic antibiotics are measures that can reduce morbidity and mortality in patients with liver cirrhosis. Because of the significant risk of adverse outcomes related to SBP, identifying predisposing factors is of utmost urgency. There are no recent published studies that exclusively address the prevalence and predictors of SBP in a hospital population in Ghana. This study aims to determine the prevalence and predictors of SBP among in-patients with cirrhotic ascites attending a tertiary referral center in Accra, Ghana. This will help with identification of high risk patients for early and appropriate treatment to reduce morbidity and mortality.
A formal approval of this study was obtained from the Ethical and
Protocol Committee of the University of Ghana School of Medicine and Dentistry.
This study was conducted in accordance with the Helsinki Declaration. The
research design was a cross -sectional hospital-based study, carried out
at the Department of Medicine, Korle-Bu Teaching Hospital (KBTH), Accra,
from 25th March, 2016 to 25th November, 2016. One
hundred and three (103) patients with cirrhotic ascites admitted to the
medical block of KBTH were consecutively recruited. All adult patients above
18 years with cirrhotic ascites who provided informed consent were included.
Exclusion criteria were patients who had already been started on antibiotics
at the time of recruitment or who had taken antibiotics up to 2 weeks preceding
recruitment, as well as refusal of consent. Diagnosis of liver cirrhosis
was made based on the clinical features, laboratory investigations and abdominal
ultrasound findings suggestive of liver cirrhosis. Patients' medical records
were reviewed to exclude those with cirrhotic ascites who were on antibiotics
or had been on antibiotics in the preceding two weeks to the date of recruitment.
Relevant history including alcohol use and clinical features of liver cirrhosis
(spider angioma, palmar erythema, ascites, asterixis, hepatomegaly, splenomegaly
and abdominal vein collaterals) were obtained. Ascites was graded as mild
(detectable only on ultrasound), moderate (visible moderate symmetrical
abdominal distension) or severe (marked abdominal distension). After thoroughly
explaining the study to patients, individuals who gave informed consent
were recruited and a questionnaire was administered (socio-demographic data
and clinical history of the patients were obtained). A sample of 15mls of
venous blood was taken for haematological, biochemical and serological investigations.
Abdominal paracentesis was performed using an aseptic technique at the right
or left iliac fossa, 3cm above and 3cm medial to the anterior superior iliac
spine. Exactly 15mls of ascitic fluid was collected using a sterile syringe
for culture, cell count and differentials, albumin and protein. Furthermore,
an abdominal ultrasound scan was performed for all patients. The following
details were recorded: maximum vertical span of the liver; nodularity of
liver surface; spleen size (length of its longest axis); and presence of
Ascitic fluid work up:
10mls of ascitic fluid was inoculated into a blood
culture bottle at the bed side and 5mls of ascitic fluid was inoculated into
a sterile EDTA bottle for culture, cell count and differentials, albumin and
protein. Ascitic fluid culture was performed by an experienced laboratory technologist
by inoculating the ascitic fluid into Blood Agar and MacConckey Agar. Preliminary
results were obtained after 48 hours, followed by conventional biochemical identification
tests. SBP diagnosis was based on neutrophil count in ascitic fluid ≥ 250
and/or positive ascitic fluid culture. If ascitic fluid
cultures were positive and the neutrophils count was > 250cells/mm3
patients were diagnosed as having culture-positive neutrocytic ascites. If ascitic
fluid cultures were negative in the presence of neutrophils > 250 cells/mm3
patients were characterized as having culture-negative neutrocytic ascites (CNNA).
Patients with positive cultures on ascitic fluid but without neutrocytic ascites
were classified as having monobacterial bacterascites (MNB).
hematological, biochemical and serological workup included
measurement of total serum bilirubin, serum albumin, alanine aminotransferase
(ALT), alkaline phosphatase (ALP), gamma glutamyl transferase (GGT), aspartate
aminotransferase (AST), serum creatinine, serum sodium (Na+
heamoglobin (HB), platelet count, white blood cells (WBC), and prothrombin time
with international normalized ratio (INR). Serological work up included hepatitis
B surface antigen (HBSAg) and anti-bodies to hepatitis C virus (anti-HCV Ab).
Anti-nuclear antibodies (ANA), immune globin G (IgG) and anti-smooth muscle antibodies
(ASMA) were done for some of the patients with suspicion of autoimmune hepatitis.
Model for end Stage Liver Disease Sodium (MELD-Na+) score was calculated based
on laboratory parameters (bilirubin, creatinine levels, sodium and INR) collected
at admission and determined by using the UNOS Internet site MELD-Na+ calculator.
Child-Pugh Score (CPS) was calculated based on grade of ascites and hepatic encephalopathy,
serum albumin, bilirubin and INR collected at admission using
QxMD calculator on the net.
data obtained were analyzed using STATA 15 statistical
software. Descriptive statistics was run for all the variables. The prevalence
of SBP and other categorical variables were expressed as proportions. Biochemical
parameters and the clinical scoring systems (MELD-Na+
Score and CPS)
were reported as Mean ± SD (normal data) and median (IQR) (non-normal data).
The Student t-test or Wilcoxon rank sum test were used to test the difference
in means. Chi-squared test and the Fishers exact tests were used to determine
the association of categorical variables and SBP. Univariable and multivariable
logistic regression models were used to determine the strength of association
between CPS, MELD-Na+
score, laboratory and other clinical parameters
to SBP. This was presented as crude and adjusted odds ratios on a 95% confidence
interval. For all analysis, p-values < 0.05 were considered statistically significant.
One hundred and three patients with cirrhotic ascites were recruited
for the study with a mean age of 43.5 ± 12.2 years (age range 18 to 74)
years. Fifty eight (58) (56.3%) patients were males and 44 (43.4%) were
females with male to female ratio of 1.7:1. HBV infection was the cause
of liver cirrhosis in 53.4% when acting alone and alcohol alone accounted
for 21.4% of causes of liver cirrhosis. HCV infection was the cause in 8.9%
of cases and HBV infection in combination with alcohol accounted for 6.9%.
Autoimmune hepatitis, fatty liver disease and congenital atresia were uncommon
The prevalence of SBP was 25.24% (95% CI=17.69-34.67). Of the 26 patients that
had spontaneous bacterial peritonitis, culture positive SBP was present
in 26.9% (9/26) while CNNA was found in 65.4% (17/26). The prevalence of
MNB was 7.7% (2/26) in this study. The major isolates from the positive
cultures were Escherichia coli (5/9, 55.6%) and Klebsiella spp.
(2/9, 22.2%). Staphylococcus epidermidis and Corynebacterium spp.
accounted for 1/9 (11.1%) each (Table
Results from the univariable logistic regression showed that patients who had
SBP had increased odds of having severe ascites (cOR=4.20, 95% CI=1.52-11.66),
fever (cOR=3.14, 95% CI=1.25-7.81), chills (cOR=4.42, 95% CI=1.61-10.55)
and encephalopathy (cOR=3.03, 95% CI=1.16-7.92). There was a 4% reduced
odds of SBP with increasing age (cOR=0.96, 95% CI=092- 0.99) (Table 3).
The multiple logistic regression model was fitted whilst controlling for
age, sex and the other variables in Table 3. The clinical feature significantly associated with SBP from the
multivariable analysis was severe ascites (aOR=5.82, 95% CI=1.51-22.42)
Laboratory parameters associated with SBP from the univariable analysis were
INR (cOR=2.09, 95%CI=1.28-3.45), CPS (cOR=1.34, 95%CI=1.07-1.69) and MELD-Na+ score
(cOR=1.09, 95%CI=1.03-1.16). The adjusted multivariable logistic regression
model fitted using a stepwise approach (backward selection). The variables
that remained in the final model were Fluid Albumin, Meld-Na+ Score
and INR, though only INR (aOR=1.90, 95%CI=1.08-3.35) was a significant predictor
of SBP (Table
This study aimed to determine the prevalence, clinical and laboratory
features predicting the presence of SBP among inpatients with cirrhotic
ascites in a tertiary care center in Accra, Ghana. The prevalence of SBP
in this study was 25.24%. This is similar compared to 10-30% found by most
studies from the developed world [12, 14, 17, 18]
but lower than one reported by Oladimeji et al. .
This may be due to the severity of liver cirrhosis involved in the study.
Oladimeji et al.  stated that almost all
the patients recruited in their study were in Child's grade C compared with
this study with majority of the participants in child's grade B and C The
diversity of the clinical and laboratory parameters that is associated with
the presence of SBP has been reported in various literatures. This justifies
the indication for diagnostic paracentesis in all patients with decompensated
cirrhosis with ascites admitted in the hospital. In a study by Evan's et
al.  no clinical or laboratory parameters were
found to be associated with the presence of SBP whiles Figueiredo et
al.  identified serum albumin, complement C4
of ascitic fluid and upper gastrointestinal bleeding as independent predictors
for the diagnosis of SBP. Hypoalbuminaemia, high MELD score, C-reactive
protein, Child-Pugh stage C, low protein level in ascitic fluid, low prothrombin
concentration, increased serum aspartate aminotransferase levels, high serum
bilirubin, low platelet count, hepatic encephalopathy and abdominal pain
have also been reported by other studies to predict SBP [21-24].
In this study fever, chills, hepatic encephalopathy, severe ascites, lower
age group, high INR, high CPS and high MELD-Na+ predicted the presence of
SBP. However, only severe ascites and high INR showed strong independent
association with SBP from the multivariate analysis. Severe ascites and
high INR are among five markers used to stage the severity of liver disease
according to Child-Pugh rankings . The higher the
CPS, the greater the risk of SBP . This helps to
explain why 70% of cases of SBP are seen in patients with Child-Pugh class
C cirrhosis . In the current study, though the
Child-Pugh score was not an independent predictor of SBP, but 80% of the
patients with SBP were in Child-Pugh grade C.
HBV (52.4%) was the major cause of liver cirrhosis in this study followed by alcohol (21.4%). This compared similarly with other studies done in Africa [27-29
] but differed from reports from the western countries [30
]. The most common causes of liver cirrhosis globally are thought to be HBV, HCV and alcohol but the causes vary from country to country and from region to region. In countries where alcohol consumption is more common, alcohol is the commonest cause of liver cirrhosis and in countries where chronic HBV infection is endemic, HBV is the commonest cause of liver cirrhosis [32
]. HBV infection is endemic in sub-Saharan Africa including Ghana and this makes it the major cause of cirrhosis in this study. Alcohol abuse was the second commonest cause of liver cirrhosis in this study which implies that alcohol is a significant cause of liver cirrhosis in patients attending clinic at KBTH. This is of public health concern; therefore society should be educated on the harmful effects of alcohol abuse on the liver.
In this study E. coli
(55.6%) was the most common organism isolated followed by Klebsiella spp
(22.2%). The isolation of these organisms is consistent with studies conducted by Bhuva et al.
] and Oladimeji et al.
] which shows E. coli
as the dominant bacteria cultured in patients with spontaneous bacterial peritonitis.
The common causes of decompensated liver cirrhosis with ascites admitted at Korle-Bu Teaching Hospital were chronic HBV and chromic alcohol abuse. Spontaneous bacterial peritonitis was common among patients with cirrhotic ascites admitted at Korle-Bu Teaching Hospital. Fever, chills, hepatic encephalopathy, severe ascites, high INR, high MELD-Na+ and CPS were predictors of SBP but only severe ascites and high INR were independent predictors. Therefore diagnostic paracentesis should be done immediately for all patients on admission to confirm the diagnosis of SBP preferably before starting empiric antibiotic therapy especially in patients with severe ascites and high INR.
What is known about this topic
- There is diversity of clinical and laboratory parameters associated with the presence of SBP;
- Upper gastrointestinal bleeding is a risk factor associated with the risk of developing SBP.
What this study adds
- Only severe ascites and high INR were independent predictors associated with SBP;
- Patient presented with upper gastrointestinal bleeding were not at high risk of developing SBP.
The authors declare no competing interests.
Amoako Duah, the principal investigator of the project, was involved
in concept design, patient recruitment, data analysis and drafted
the manuscript. Kofi Nyaako Nkrumah guided the principal investigator
in concept design,
assisted in the analysis and interpretation of the data, critically
revised the article. All authors have read and agreed to the final
I would like to acknowledge Dr. Rafiq Okine, for his assistance during analysis of project data.
Table 1: causes of liver cirrhosis
Table 2: bacteria isolated from ascitic fluid
bivariate analysis (clinical and demographic parameters)
Table 4: univariate and multiple logistic regression (socio-demographic and clinical)
Table 5: univariate and stepwise multiple logistic regression
- Mowa CT, Stanley AJ. Review article: spontaneous bacterial peritonitis-diagnosis treatment and prevention. Aliment Pharmacol Ther. 2001; 15(12): 1851-9. PubMed | Google Scholar
- Ogutu EO. Spontaneous bacterial peritonitis in patients with liver diseases and ascites as seen at Kenyatta National Hospital. East African Medical Journal. 1988; 4: 547-51.
- Alaniz C, Regal RE. Spontaneous Bacterial Peritonitis: a review of treatment options. PT. 2009; 34(4): 204-10. PubMed
- Evans LT, Kim WR, Poterucha JJ, Kamath PS. Spontaneous
bacterial peritonitis in asymptomatic outpatients with
cirrhotic ascites. Hepatology. 2003; 37(4): 897-901. PubMed | Google
- Andreu M, Sola R, Sitges-Serra A, Alia C, Gallen M, Vila MC et al. Risk factors for spontaneous bacterial peritonitis in cirrhotic patients with ascites. Gastroenterology. 1993; 104(4): 1133-1138. PubMed | Google Scholar
- Jalan R, Javier Fernandez J, Wiest R, Schnabl B, Moreau R, Angeli P et al. Bacterial infections in cirrhosis: a position statement based on the EASL Special Conference 2013. Journal of Hepatology. 2014; 60(6): 1310-1324. PubMed | Google Scholar
- Wiest R, Garcia-Tsao G. Bacterial translocation (BT) in cirrhosis. Hepatology. 2005; 41(3): 422-433. PubMed | Google Scholar
- Llovet JM, Bartoli R, March F, Planas R, Vinado B, Cabre E et al. Translocated intestinal bacteria cause spontaneous bacterial peritonitis in cirrhotic rats: molecular epidemiologic evidence. Journal of Hepatology. 1998; 28(2): 307-313. PubMed | Google Scholar
- Fiuza C, Salcedo M, Clemente G, Tellado JM. "In vivo neutrophil dysfunction in cirrhotic patients with advanced liver disease." Journal of Infectious Diseases. 2000; 182(2): 526-533. Epub 2000 Jul 24. PubMed | Google Scholar
- Cereto F, Molina I, Gonz'alez A, Del Valle O, Esteban R, Guardia J et al. Role of immunosuppression in the development of quinolone-resistant Escherichia coli spontaneous bacterial peritonitis and in the mortality of E. coli spontaneous bacterial peritonitis. Alimentary Pharmacology and Therapeutics. 2003; 17(5): 695-701. PubMed | Google Scholar
- Fernandez J, Gustot T. Management of bacterial infections in cirrhosis. J Hepatol. 2012; 56 Suppl 1: S1-12. PubMed | Google Scholar
- Tandon P, Garcia-Tsao G. Bacterial infections, sepsis and multiorgan failure in cirrhosis. Semin Liver Dis. 2008; 28(1): 26-42. PubMed | Google Scholar
- Fernández J, Bauer TM, Navasa M, Rhodes J. Diagnosis, treatment and prevention of spontaneous bacterial peritonitis. Best Pract Res Clin Gastroenterol. 2000; 14(6): 975-90. PubMed | Google Scholar
- Rimola A, Soto R, Bory F, Arroyo V, Piera C, Rodes J. Reticuloendothelial system phagocytic activity in cirrhosis and its relation to bacterial infections and prognosis. Hepatology. 1984; 4(1): 53-58. PubMed | Google Scholar
- Nousbaum JB, Cadranel JF, Nahon P, Khac EN, Moreau R, Thevenot T et al. Diagnostic accuracy of the Multistix 8 SG-reagent strip in diagnosis of spontaneous bacterial peritonitis. Hepatology. 2007; 45(5): 1275-1281. PubMed | Google Scholar
- McHutchison JG, Runyon BA. Spontaneous bacterial peritonitis. In: gastrointestinal and Hepatic Infections, Surawicz CM, Owen RL (Eds), WB Saunders, Philadelphia. 1994; 455.
- Almdal TP, Skinhoj P. Spontaneous bacterial peritonitis in cirrhosis. Incidence, diagnosis and prognosis. Scand J Gastroenterol. 1987; 22(3): 295-300. PubMed | Google Scholar
- Pinzello G, Simonetti RG, Craxi A, Di Piazza S, Spano C, Pagliaro L. Spontaneous bacterial peritonitis: a prospective investigation in predominantly nonalcoholic cirrhotic patients. Hepatology. 1983; 3(4): 545-549. PubMed | Google Scholar
- 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. PubMed | Google Scholar
- Figueiredo FAF, Coelho HSM, Soares JAS. Spontaneous bacterial peritonitis in hepatic cirrhosis: prevalence, predictive, predictive factors and prognosis. Rev Assoc Med Bras. 1999; 45(2): 128-36. PubMed | Google Scholar
- Guarners C, Sola R, Soriano G, Andreu M, Novella MT, Villa MC et al. Risk of a first community-acquired spontaneous bacterial peritonitis in cirrhotics with low ascetic fluid protein levels. Gastroenterology. 1999; 117(2): 414-19. PubMed | Google Scholar
- Kasztelan-Szczerbinska B, Slomka M, Celinski K, Serwacki M, Szczerbinski M, Cichoz-Lach H. Prevalence of spontaneous bacterial peritonitis in asymptomatic inpatients with decompensated liver cirrhosis: a pilot study. Adv Med Sci. 2011; 56(1): 13-7. PubMed | Google Scholar
- Gayatri AA, Suryadharma IG, Purwadi N, Wibawa D. The relationship between a model of end stage liver disease score (MELD score) and the occurrence of spontaneous bacterial peritonitis in liver cirrhotic patients. Acta Medica Indonesiana. 2007; 39(2): 75-78. PubMed | Google Scholar
- Kraja B, Babameto AE, Burazeri G. Predictive Value of the Model of End-Stage Liver Disease in Cirrhotic Patients with and without Spontaneous Bacterial. Gastroenterology Research and Practice. 2012; 2012: 539059. PubMed | Google Scholar
- Pugh RN, Murray-Lyon IM, Dawson JL. Transection of the esophagus for bleeding esophageal varices. Br J Surgery. 1973; 60(8): 646-9. PubMed
- Such J, Runyon BA. Spontaneous bacterial peritonitis. Clin Infect Dis. 1998; 27(4): 669-74; quiz 675-6. PubMed | Google Scholar
- Blankson A, Wiredu EK, Gyasi RK, Adjei A, Tettey Y. Sero -prevalence of hepatitis B and C viruses in cirrhosis of the liver in Accra, Ghana. Ghana Med J. 2005; 39(4): 132-137. PubMed
- Iloh GU, Ikwudinma AO. Sero-epidemiology of Hepatitis B surface antigenaemia among adult Nigerians with clinical features of liver diseases attending a primary-care clinic in a resource-constrained setting of eastern Nigeria. North Am J Med Sci. 2013; 5(4): 293-300. PubMed
- Ayele AG, Solomon GS. Prevalence and risk factors
of hepatitis B and hepatitis C virus infections among
patients with chronic liver diseases in public hospitals
in Addis Ababa, Ethiopia. ISRN Tropical
Med. 2013; e56: 3821. Google
- Patricia LG, Maria da PZG, Carla CM, Ana TM, Carlos SG, Fausto ELP. Etiology of liver cirrhosis in Brazil: chronic alcoholism and hepatitis viruses in liver cirrhosis diagnosed in the state of Espfrito Santo. Clinics. 2013; 68(3): 291-295. PubMed | Google Scholar
- Wolf DC. cirrhosis. Medscape. August 5th, 2013.
- Lim YS, Kim WR. The global impact of hepatic fibrosis
and end-stage liver disease. Clin Liver Dis. 2008;
12(4): 733-746. PubMed | Google
- Bhuva M, Ganger D, Jensen D. Spontaneous bacterial peritonitis: an update on evaluation, management, and prevention. Am J Med. 1994; 97(2): 169-75. PubMed | Google Scholar