ORIGINAL ARTICLE
The Relationship Between a Model of End Stage Liver
Disease Score (MELD Score) and The Occurrence of Spon-
taneous Bacterial Peritonitis in Liver Cirrhotic Patients
AA. Ayu Yuli Gayatri*, IGA. Suryadharma**, N. Purwadi**, IDN. Wibawa**
ABSTRACT
Aim: to determine relationship between MELD score
and the occurrence of SBP, prevalence of SBP, pattern of
bacterial culture and antibiotic susceptibility of causative
bacteria of liver cirrhotic patients at Sanglah Hospital.
Methods: study design was a cross-sectional analytic
study. The population in this study consists of liver cirrhotic
patients admitted at Sanglah Hospital Denpasar from June
2005 to February 2006.
Results: this result conrmed that the MELD score is a
reliable index of disease severity & that higher MELD scores
had a signicantly more frequent SBP prevalence in patients
with a MELD score of 18 or more compared with that in
patients with a MELD score of 17 or less (p=0.01; 95% CI
= 1.379-15.537). Prevalence of SBP was 30.6%. Thirteen
patients (68.4%) had monomicrobial positive culture of
Aerob bacteria, consisting of Gram negative bacterias in 10
(77%), with Escherichia coli and Acinettobacter baumanii
being the most frequent, and 3 (23%) had Gram positive
bacteria. High sensitivity to Cefoperazone, Cefotaxim, Cef-
tazidime, Cefpirome, Ciprooxacin, Fosfomicin, Meropenem,
Streptomycin, Gentamycin, Trimethoprim/Sulphamethoxazole,
were shown.
Conclusion: based on this study we concluded that
severe liver cirrhosis with MELD score > 18 was associated
with an increase risk of SBP, with a prevalence of 30.6%.
Common causes of SBP mostly were Escherichia coli and
Acinettobacter baumanii, which were sensitive to antibiotic
treatment of Cefoperazone, Cefotaxime and Ciprooxacin.
Key words: SBP, MELD score, pattern of bacterial culture,
antibiotic.
* Department of Internal Medicine Udayana University-Sanglah
Hospital, Denpasar Bali, ** Division of Gastroenterohepatology,
Department of Internal Medicine Udayana University-Sanglah
Hospital, Denpasar Bali
INTRODUCTION
Patients with end stage liver cirrhosis may frequently
develop a variety of complications, including spontaneous
bacterial peritonitis (SBP). Importantly, these complica-
tions may occur repeatedly and may eventually lead
to mortality. From various epidemiological studies in
Indonesia, the prevalence of SBP is approximately
10-30% with an associated mortality rate of 20-40%.
The diagnosis is based on the presence of PMN count
> 250 cells/mm 3 or indicated by monomicrobial
culture.1,2,3
In the current study, it is shown that the SBP patients
who died had a signicantly higher baseline MELD
score compared to those who survived. MELD is the
latest attempt to rene the Child Turcotte Pugh score by
removing more subjective clinical variables and using
an index consists of bilirubin, INR and serum creatinin
only.4,5,6 The key of successful treatment of SBP is the
appropriate antibiotic regimens and an anticipation of
high risk of infection. This study was conducted to
determine the relationship between MELD score and the
occurrence of SBP. We also investigated the prevalence
of SBP, causative microorganism and its sensitivity
to antibiotics. Hopefully, it would provide additional
information to improve better management of liver
cirrhotic patients.
METHODS
The design of this study was a cross-sectional
analytic study, which was conducted at Sanglah Hospital
Denpasar. Samples were collected consecutively from
June 2005 to February 2006. Adult patients with liver
cirrhosis and ascites were eligible for enrollment,
and they were asked to sign informed consent forms.
Excluded from this study were patients with DIC,
infection of abdominal wall, poor cooperative patients,
intestinal obstruction and history of abdominal surgery.7
Study subjects consisted of 62 liver cirrhotic patients
admitted to Sanglah Hospital Denpasar during that
75
A A Ayu Yuli Gayatri, et al
period of time. Diagnostic abdominal paracentesis was
performed in all of subjects. Ascitic uid sample was
collected in Brain Heart Infusion transport media and
continued with Blood Agar and McConkey media.
Antibiotic sensitivity test was performed by inoculation
of the samples in Muller Hilton Agar.8 The MELD
equation used to calculate the severity score for liver
cirrhosis was as follows:
0.957x log (creatinine, mg/dl) + 0.378x log (bilirubin,
mg/dl)
+ 1.120x log (INR) + 0.643
The maximal serum creatinin level considered
within MELD score equation was 4.0 mg/dl.9 Data
management and analysis were performed using the
SPSS Software version 10 for windows. The association
between MELD score and occurrence of SBP was
analyzed by Chi Square test. The signicance level used
in this study was 5%.
RESULTS
In this study, the prevalence of SBP was 30.6%.
Of the 62 patients (mean age 55.23, 80.6% male), the
etiology of liver cirrhotic was chronic hepatitis B in
45.2% and hepatitis C in 23% of patients. The details of
patient characteristics are shown in table 1.
Of the 19 SBP patients, there were 13 (68.4%)
positive monomicrobial culture, consisting of Gram
negative 10 (77%) and Gram positive 3 (23%). The most
common infecting microorganisms were Escherichia
coli (15.4%), Acinettobacter baumanii (15.4%) and
Streptococcus β haemolitycus (15.4%). (Table 2)
Those microorganisms showed sensitivity to
Cephalosporins (Cefoperazone, Cefotaxime, Cef-
tazidime, Cefpirome), Quinolone (Ciprofloxacine),
Fosfomicin, Meropenem, Streptomicin, Gentamicin,
Trimethoprim/Sulphamethoxazole, Tetracycline, and
Chloramphenicole. The Gram positive microorganisms
also showed sensitivity to antibiotics such as Amoxicillin,
Ampicillin, Linezolid and Meticillin. (Table 3 and 4)
76
Acta Med Indones-Indones J Intern Med
In this study, based on MELD score, most patients
had MELD score < 17 (74.2%). Among the 46 patients
who had MELD score > 18, there were 9 (14.5%)
patients with SBP. The correlation between MELD score
and occurrence of SBP are given in table 5.
The study found a signicant relationship between
occurrence of SBP and MELD score of 18 or more
compared with a MELD score of 17 or less (p= 0.01;
95% CI= 1.379-15.537).
DISCUSION
Establishment of prognostic factors is the key towards
evaluating clinical interventions in any disease. The
most commonly used prognostic model in patients with
cirrhosis is Child-Turcotte-Pugh (CPS). In order to
overcome the limits of the CPS score, previous studies
have evaluated a “combined score” with quantitative
liver function test, have created new scores, or have
applied scores that were originally formulated to evaluate
multi-organ insufficiency in critically ill patients to
cirrhotic patients. A new approach is the MELD formula
which is used in patients with end stage cirrhosis
awaiting liver transplantation. Previously, a MELD score
18 was identied as a threshold for high risk patients with
a median survival time of three months or less.4,5
In this study we conrmed that the MELD score
is a reliable index of disease severity and that higher
MELD scores had a signicantly more frequent SBP
prevalence in patients with a MELD score of 18 or more
compared with that in patients with a MELD score of 17
or less (p=0.01; 95%CI= 1.379-15.537).
Prevalence of SBP was 30.6%, which was not much
different from the other local or foreign studies. The
number and species distribution of the isolated
microorganisms from SBP patients did not change over
Vol 39 • Number 2 • April - June 2007 The Relationship Between a Model of End Stage Liver Disease Score
the last 20 years period. Most infections were caused by
Enterobacteriaceae and their percentage did not change
signicantly. Aerob Gram negative bacteria, primarily
Escherichia coli are the most common isolates, although
the frequency of episodes caused by Gram positive
bacteria has recently increased.10
Among 19 SBP patients on this study, thirteen
patients (68.4%) had positive monomicrobial culture of
Aerob bacteria, consisting of Gram negative bacteria
10 (77%), with Escherichia coli and Acinettobacter
baumanii being the most frequent gram negative
bacteria. The rest were gram positive bacteria found
in 3 patients (23%). Many studies on microorganisms’
pattern have been conducted and reported, but this
would be the rst from Sanglah Hospital Denpasar, and
hopefully would provide additional information for
clinicians. Good susceptibility to cefoperazone, cefo-
taxim, ceftazidime, cefpirome, ciprooxacin, fosfomicin,
meropenem, streptomycin, gentamycin, trimethoprim/
sulphamethoxazole, were shown. The gram positive
bacteria also showed sensitivity to amoxicillin, ampicil-
lin, linezolid, and meticillin. Among all of antibiotics
evaluated, the third generation cephalosporine and
quinolone as the recommended therapy for SBP by
PPHI, and also the most commonly used antibiotics
77
A A Ayu Yuli Gayatri, et al Acta Med Indones-Indones J Intern Med

regimen at Sanglah Hospital Denpasar, still showed
good clinical response.3
CONCLUSION
From this study, it can be concluded that increase
risk of SBP occurs along with increase of MELD score,
while according to severity of liver cirrhosis, the MELD
score >18 is associated with the increase risk of SBP
with prevalence of 30.6%. It is also shown that the most
common microorganisms among SBP patients were
Escherichia coli and Acinettobacter baumanii. Cefo-
perazone, Cefotaxime and Ciprooxacin were among
the most frequently used antibiotics at Sanglah Hospital
Denpasar and treatment result was favorable.
REFERENCES
1. Evans LT, Kim WR, Poterucha JJ, Kamath PS. Spontaneous
bacterial peritonitis in asymptomatic outpatients with cirrhotic
ascites. Hepatology. 2003;37:897-901.
2. Gines P, Cardenas A, Arroyo V, Rodes J. Management of
cirrhosis and ascites. N Engl J Med. 2004;350:1646-54.
3. PPHI. Diagnosis dan terapi peritonitis bakteri spontan pada
sirosis hati. Konsensus Perhimpunan Peneliti Hati Indonesia,
2001.
4. Botta F, Gannini E, Romagnoli P, Fasoli A, Malfatti F,
Chiarbonello B, et al. MELD scoring system is usefull for
predicting prognosis in patients with liver cirrhosis and is
correlated with residual liver function: a European study.
Gut. 2003;52:134-9.
5. Angermayr B, Cejna M, Karnel F, Gschwantler M, Koenig F,
Pidlich J, et al. Child-pugh versus MELD score in predicting
survival in patients undegoing transjugular intrahepatic
portosystemic shunt. Gut. 2003;52:879-85.
6. Ferral H, Gambora P, Postrack DN, Albernaz VS, Young CR,
Speeg KV, et al. Survival after elective transjugular intrahepatic
portosystemic shunt creation: Prediction with model for end-
stage liver disease score. Radiology. 2004;231:231-6.
7. Saadeh S, Davis G. Management of ascites in patients
with end stage liver disease. Rev Gastroenterol Disord.
2004;4(4):175-85.
8. Laboratorium/Instalasi Patologi Klinik RS Sanglah Denpasar.
Pemeriksaan bakteriologik cairan asites. Protap mikrobiologi.
Denpasar: RS Sanglah; 2004.
9. Coombes JM, Trotter JF. Development of the allocation
system for decreased donor liver transplantation. Clinical
Medicine and Research. 2005;3(2):87-92.
10. Garcia-Tsao G. Current management of the complications
of cirrhosis and portal hypertension: Variceal haemorrhage
ascites and spontaneous bacterial peritonitis. Gastroenterol.
2001; 120:726-48.

78