Liver Abscess

Dr. Johnny Cai presented a case of a woman with a history of a pancreatic tail cyst who had two weeks of progressive RUQ pain, fever, and night sweats. She was hemodynamically stable, but her labs revealed a hepatocellular liver injury and a severe leukocytosis. RUQ ultrasound revealed pneumobilia without common bile duct dilatation, cholelithiasis or pericholecystic fluid.

Pneumobilia, you say?

The differential for pneumobilia is short -- post-procedural vs BADNESS (like gas-producing bacterial infection).


The team then proceeded with a CT (shown here):

“gas and fluid filled structure 5x4 cm in segment 4 of the liver consistent with abscess without clear source.”


When thinking about a liver abscess, it can be helpful to try to classify it further into pyogenic (most common) vs less common causes such as amebic or unusual infections (TB, Echinococcus, candidal). Your history should alert you if you should be thinking about atypical liver abscesses. Echinococcus has a very distinct look on imaging and should not be aspirated for culture as this can be lethal.

You should also ensure you get blood cultures as liver abscesses can form from hematogenous spread.

A liver abscess?

The patient was started on pipercillin/tazobactam and IR aspirated the liver abscess and placed a drain. The aspirate culture never grew anything (suspected sterilized with antibiotics).

Her leukocytosis was not improving and thus she was -reimaged without change in abscess and had another drain placed. THIS IS A GOOD TEACHING POINT: if the clinical picture isn’t improving, make sure you truly have source control (drain not in right place, another process going on, wrong antibiotics, etc.) After her clinical picture started to improve, she was transitioned to metronidazole and moxifloxacin to complete a 6 week course.


  • UptoDate “Pyogenic Liver Abscess”

  • Lübbert C, Wiegand J, Karlas T. Therapy of Liver Abscesses. Viszeralmedizin. 2014;30(5):334–341

  • Rathimian, J et al. Clinical Infectious Diseases. 2004 Dec 1;39(11):1654-9

  • Tan YM, Chung AY, Chow PK, et al. Ann Surg. 2005;241(3):485–490

  • Wong, Wai Man et al. Journal of Gastroenterology and Hepatology. (2002) 17, 1001–1007

Fatigue and Hemochromatosis

Dr. Chandrashekar presented a case of a man with heavy alcohol use who presented to primary care clinic for fatigue. Work-up was revealing for hepatocellular liver injury and an elevated transferrin saturation leading to a workup for hereditary hemochromatosis.

Transferrin saturation= Iron/TIBC x 100

As a reminder, there are many causes of elevated ferritin (including liver disease) which was highlighted on another blog post from June. Ferritin is an acute phase reactant and it is not enough to diagnosis someone with hemochromatosis.

To diagnosis someone with iron overload you need a transferrin saturation of >45% (AASLD). Many labs do not calculate a transferrin sat so you may miss this if you aren't thinking about hemochromatosis when your iron studies return. Transferrin sat can be falsely elevated in settings where there is acute increase in iron (such as hemolysis or oral iron intake).

The work up for fatigue starts with a good history. In patient's with fatigue, hemochromatosis should be considered particularly if a patient has liver disease or diabetes without traditional risk factors. As highlighted in this case, many patient's liver disease is attributed to their alcohol use when in reality hemochromatosis is also at play.

The illness script for hemochromatosis:

Screen Shot 2019-09-05 at 10.16.28 PM.png

40-50 yo Caucasian man (or post-menopausal female) with fatigue, chronic liver disease, diabetes, arthritis, and hyperpigmentation. Other organs include endocrine glands (pituitary, pancreas, thyroid, hypothyroid, etc) and infiltrative cardiomyopathy.

More on Hereditary Hemochromatosis:

  • the HFE gene is the most common cause of hereditary hemochromatosis. The HFE gene is autosomal recessive. However, the gene has low penetrance, so even patients with homozygosity may not have phenotypic disease.
  • alcohol use is common in patient hemochromatosis with evidence of liver disease in part because of the "two hit" hypothesis. This concept is common in hepatology in which a liver is vulnerable due to an underlying disease (such as hemachromatosis) and another injury (such as alcohol) causing inflammation and injury can lead to clinicial disease.

One more small but important hint in this case:


on the peripheral smear.

What are pappenheimer bodies you may be asking?They are iron inclusion bodies in RBCs (look like Heinz bodies but stain positive for iron). If you see this on a peripheral smear you should question if the patient has splenic dysfunction or if there is an issue with hemolysis or excess iron.

Pappenheimer bodies can be seen in:

  • asplenia
  • spenic dysfunction such as MDS
  • sickle cell disease
  • sideroblastic anemia
  • lead poisoning
  • hemochromatosis