M J M L Hakeem, D N Bhattacharyya
Department of Infectious Diseases, Victoria Hospital Hayfield Road Kirkcaldy Fife KY2 5AH
Correspondence to: Dr M J M L Hakeem, Staff Grade in Infectious Diseases, Fife Acute Hospitals Victoria Hospital, Hayfield Road Kirkcaldy, Fife KY2 5AH
E-mail: nancy.steele@faht.scot.nhs.uk
SMJ 2006 51(1): 57
Abstract
Cholestatic
jaundice is a well recognised complication of late sepsis.
However, cholestatic liver injury preceding sepsis is less well known and
appreciated. This leads to delays
in diagnosis in clinical practice and may result in significant morbidity and
mortality. The development of a
cholestatic blood picture should be considered as an early warning sign of an
underlying infection, even in the absence of signs or symptoms of sepsis.
This report describes the case of a previously healthy patient who
presented with cholestatic jaundice. Staphylococcal
osteomyelitis involving the left sacroiliac joint was diagnosed.
Literature relating to cholestatic jaundice in sepsis is reviewed.
Case
Report
A 47 year old Caucasian male with no significant past medical history presented to hospital with a five day history of jaundice, fever, poor appetite and rigors. He worked in Angola as an engineer and had returned to the United Kingdom about six weeks prior to being admitted. He had then gone on a hill walking holiday to the Spanish-French border and was admitted to a hospital in Spain with back pain after falling. He was treated with analgesia and bed rest and was subsequently discharged after a few days. X-ray of his lumbar sacral region was normal at that stage. He had not had any vaccinations prior to travelling. He was on Diclofenac and Diazepam on admission.
Physical
examination revealed a high temperature of 38.5oC.
He was clearly jaundiced. His
blood pressure was 130/70, pulse rate was 80 beats per minute, heart sounds were
pure. Chest sounded clear. There
was no evidence of organomegaly or tenderness on abdominal examination.
He did complain of some mild sacral tenderness, however he did not appear
to be in any distress on examination. There
was no focal neurological deficit.
Admission bloods showed an albumin of 31 g/L, alkaline phosphatase of 1866 U/L, gamma glutamyl transferase of 336 U/L, alanine transaminase of 138 U/L and bilirubin of 80 umol/L. He also had a leucocytosis of 17,500 cells/ml with a neutrophilia. C-reactive protein was raised at 220. He had an abdominal ultrasound scan which was normal. Lumbar spine x-ray was repeated and this again did not reveal any obvious destructive lesions or abnormalities. Although the patient was jaundiced and pyrexial, he otherwise appeared remarkably well. He was seen by the admitting physicians in the Acute Medical Receiving ward and a provisional diagnosis of viral hepatitis was made and bloods were sent for hepatitis serology, EBV and CMV IgM levels. The patient was discharged two days after admission with a follow-up appointment because he was not keen to stay in hospital. However, he was readmitted the following day to the Infectious Diseases Unit with worsening symptoms. The patient remained deeply jaundiced and was running high temperatures. He was tachycardic with a pulse rate of 100 beats per minute and had a relatively low blood pressure of 110/60 mmHg. A septic screen was organised and he had multiple blood cultures. He continued to complain of sacral back pain. He was started on broad spectrum antibiotics (ceftriaxone 2 g) and was resuscitated with IV fluids.
His
serology results for hepatitis A, B and C were negative and his EBV and CMV IgM
antibodies were negative. However,
his blood cultures were positive for Staphylococcus aureus.
There was no evidence of any skin lesions but he continued to complain of
back pain. A bone scan was
therefore organised and this showed an increased uptake around the left sacro-iliac
joint. An MRI scan later confirmed
an increased signal at the left sacro-iliac joint consistent with osteomyelitis,
as shown in figure 1.
He
was therefore diagnosed to have Staphylococcal osteomyelitis involving the left
sacro-iliac joint. His antibiotics
were changed to intravenous flucloxacillin and his condition improved with this.
He was discharged home after making a remarkable recovery.
His liver function tests showed near complete normalisation and his CRP
was <4 mg/L prior to discharge.
Discussion
This
is a case of Staphylococcal osteomyelitis presenting with cholestatic jaundice.
Intrahepatic cholestasis is a well recognised complication of
extrahepatic bacterial infections. The
association of hyperbilirubinaemia and extrahepatic bacterial infection was
recognised as early as 1837,1-2 and occurrence of jaundice in
pneumonia was also described by Osler in his first edition of Principles and
Practice of Medicine.3
Cholestatic
jaundice has been reported to occur in 1–6% of adults with bacterial sepsis
with a slightly higher prevalence in neonates.4
It is said to occur in up to 55% of patients with pre-existing
hepatobiliary disease or malignant neoplasm where hyperbilirubinaemia is more
pronounced.5 Persistent or increasing hyperbilirubinaemia indicative of
ongoing active infection was correlated with the worst outcome approaching 100%
mortality.1-6 However,
mortality rate was considerably lower in patients without pre-existing biliary
disease or malignant neoplasms. Furthermore,
in this group of patients, elevations in serum aspertate aminotransferase,
alanine aminotransferase and alkaline phosphatase levels (present in 65% of
patients) were more frequent than hyperbilirubinaemia.1-7
Although
it is generally accepted that jaundice may complicate many systemic infections,
there are few detailed reports of its nature in adults.8-10
The degree of jaundice is usually mild, occasionally the jaundice may be
severe. This severe jaundice may be
misdiagnosed as a sign of primary hepatic or biliary tract disease and thus
divert attention from the underlying infection and lead to inappropriate
treatment.8,11 Clinically it is important to differentiate infection in an
obstructed biliary tree (cholangitis) from hepatic dysfunction due to sepsis.
The former usually presents with fever, rigors, right upper quadrant
tenderness and jaundice. The
diagnosis could be supported by a biliary ultrasound and if doubt remains, an
endoscopic retrograde cholangiography may be helpful.12
Sepsis
is a frequent cause of death in hospitals.
One factor leading to the high morbidity and mortality of sepsis may be
the failure to recognise alterations in cellular function that herald the early
stages of sepsis, leading to delays in diagnosis and appropriate management of
these patients.13 Subtle
hepatocellular dysfunction and cholestatic jaundice associated with bacteraemia
can occur from one to nine days before the initial positive blood cultures in
more than a third of patients.13
Therefore development of a cholestatic blood picture should be considered
as an early warning sign of an underlying infection even in the absence of
fever, leucocytosis or any other signs or symptoms.1
Some
of the systemic infections causing jaundice include pneumonia, pyelonephritis
and soft tissue infections.4 Although
cholestatic jaundice is more commonly associated with gram negative bacteraemia,
it is also known to occur with gram positive bacteraemia.14-16
E.coli and Klebsiella pneumoniae are common gram negative organisms
associated with jaundice, also organisms like Streptococcus and Staphylococcus
are common gram positive organisms known to cause jaundice.12
The presence of hyperbilirubinaemia in patients with Staphylococcus
aureus sepsis may be associated with significantly poorer prognosis.1
The
pathogenesis of cholestatic jaundice in sepsis is not well understood.
Hepatic hypoxia, haemolysis, direct hepatic toxicity from endotoxins and
lipopolysaccharides (LPSs) contained within bacterial cell wall components are
all thought to contribute. Endotoxins
and bacterial cell wall component LPSs are
potent inducers of pro-inflammatory cytokine production, including tumour
necrosis factor a
(TNFa),
interleukin-1 (IL1) and interleukin 6 (IL6).
Recent studies point to endotoxinaemia and subsequent release of
cytokines during infections as the major factor in sepsis-associated
cholestasis.1,17 Endotoxin
and the pro-inflammatory cytokine, interleukin-1b (IL-1b) stimulate Kupffer cells to release TNFa,
IL-6 and IL-8. endotoxins and TNFa
up regulate the expression of both intracellular adhesion molecule-1 (ICAM-1) on
sinusoidal endothelial cells, Kupffer cells, hepatocytes and the complimentary
ligand, MAC-1 on neutrophils. Neutrophil
adhesion within the sinusoids is thereby favoured and subsequent release of
superoxide radicals and enzymes such as elastase and proteases promote hepatic
injury. Similar events also occur
in the portal tract as a result of cytokine induced enhancement of ICAM-1
expression on bile duct epithelial cells leading to neutrophil attack.18
Histologically
the most prominent finding in bacteraemic patients with jaundice is intrahepatic
cholestasis. In addition, Kupffer
cell hyperplasia, portal mononuclear cell infiltrates, focal hepatocyte dropout
and steatosis have been described.1
This
case illustrates that jaundice is not always indicative of primary hepatic or
biliary tract disease and that cholestatic jaundice is a recognised complication
of sepsis. An extrahepatic source
of sepsis should be considered in any febrile jaundiced patient with normal
ultrasound examination. The
development of cholestatic blood picture should be considered as an early
warning sign of an underlying infection even in the absence of
any signs or symptoms of sepsis.
Acknowledgement:
References
1.
Moseley PH. Sepsis
and cholestasis. Clinics in liver
disease. 1999 August; 3(3):465-75.
2.
Garvin IP. Remarks
on pneumonia biliosa. S Med Surg J,
1837;1:536.
3.
Osler W. The
principles and practice of medicine. New
York: Appleton, 1892.
4.
Wong EW, Cohen AJ. A
case of pyelophlebitis presenting with cholestatic jaundice.
American Society of Emergency Radiology 2000;7:56-58.
5.
Franson TR, Hierholzer WJ Jr, La Brecque DR.
Frequency and Characteristics of hyperbilirubinaemia associated with
bacteraemia. Reviews of Infectious Diseases 1985;7:1.
6.
Franson TR, La Brecque DR, Buggy BP et al.
Serial bilirubin determinations as a prognostic marker in clinical
infections. American Journal of
Medical Science 1989;297:149.
7.
Sikuler E, Gvetta V, Keynan A, et al.
Abnormalities in bilirubin and liver enzyme levels in adult patients with
bacteraemia. Arch Inter Med
1989;149:2246
8.
Narita R, Murata M, Kihara Y, et al.
Sepsis presenting with severe jaundice.
American Journal of Gastroenterology (US) Nov 2001;96(11):3214-5.
9.
Vermillion SE, Gregg JA, Baggenstoss AH et al.
Jaundice associated with bacteraemia.
Arch Intern Med 1969;124:611-8.
10.
Miller DJ, Keeton GR, Webber BL et al.
Jaundice in severe bacterial infections. Gastroenterology 1976;71:94-7.
11.
Eley A, Hargreaves T, Lambert HP. Jaundice
in severe infection. British
Medical Journal 1965;10:75-7.
12.
Stuart C, Gordon MD. Jaundice
and cholestasis. Postgraduate
Medicine 1991;Vol 90(4):Sep 15.
13.
Moseley RH. Sepsis
associated cholestasis. Gastroenterology
1997;112(1):302-305.
14.
Shamir R, Maayan-Metzger A, Bujanover Y et al.
Liver enzyme abnormalities in gram negative bacteraemia of premature
infants. Pediatric Infectious
Disease Journal (U.S.) June 2000;19(6):495-8.
15.
Minuk GY, Ruscanin N, Sarjeant ES et al.
Sepsis and cholestasis. The
in vitro effect of bacterial products on 14-C–taurocholate uptake by isolated
rat hepatocytes. Liver
1986;6:199-204.
16.
Rubin E, Faber JL. eds pathology. Philadelphia:
Lippincott-Raven, 1998;766.
17.
Jansen LM, Muller M. Early
events in sepsis associated cholestasis. Gastroenterology
1999;116(2):486-488.
18. Crawford JM, Boyer JL. Clinicopathology Conferences: inflammation induced cholestasis. Hepatology 1998;28:257.