Discussion
To our knowledge, this is the first reported case of multiple PLA caused by Y. pseudotuberculosis in Denmark. Our patient presented with six weeks of intermittently fever, fatigue, and unattended weight loss. A CT-scan performed in the General Practitioner setting had revealed the diagnosis of multiple hepatic abscesses. She was successfully treated with percutaneous drainage combined with antibiotic therapy with piperacillin/tazobactam and ciprofloxacin for a total of eight weeks.
In general, approximately half of PLA patients have secondary bacteremia.8 While there are only a few case-reports of Y. pseudotuberculosis PLA,9-14 it seems this also applies to Y. pseudotuberculosis PLA (in four cases the bacteria was cultured from blood).9-11,13 In one case, the diagnosis was based on antibody titers for Y. pseudotuberculosis only.12 In our case, blood culture and stool were culture negative, but only the first blood culture set was taken prior to antibiotic therapy. Overall, cryptogenic liver abscesses are most often monomicrobial.8 Prior case-reports of Y. pseudotuberculosis PLA9-14and ours, suggest that Y. pseudotuberculosis PLA are also monomicrobial.
The majority of PLA occurs in patients with underlying hepatobiliary or pancreatic disease through ascending cholangitis or hematogenous spread though the portal vein,8 others may be induced by trauma (e.g., post-surgical), and a substantial part remains cryptogenic. While our patient had choledocholithiasis, she had no signs of cholecystitis or cholestasis. Diabetes mellitus is associated with an increased risk of PLA,18 and also a risk factor for systemic Y. pseudotuberculosis infection, including bacteremia.2 Our patient had a newly diagnosis of type 2 diabetes mellitus, which also was reported in three of six prior cases of Y. pseudotuberculosis PLA,9,11,13 and in a patient with a splenic abscess.13 Iron overload has been suggested to predispose for Y. enterocolitica liver abscesses and systemic Y. pseudotuberculosis infection (owing to lack of sophisticated iron metabolism pathways).19 One of six prior cases of Y. pseudotuberculosis PLA had a diagnosis of genetic hemochromatosis.10 While our patient had a slightly elevated level of plasma ferritin at admission, this was interpreted as an acute-phase reactant, as she had no other signs of hemochromatosis.
Y. pseudotuberculosis is hosted by various animals with main reservoir in rodents, deer and wild birds.2Multiple Y. pseudotuberculosis outbreaks has been reported from contaminated food such as carrots and lettuce,20,21 but no Danish outbreaks have yet been identified. We were not able to confirm the source of infection in our patient, and she had no close contacts to animals.
Treatment of PLA includes drainage (either percutaneous or surgical), and antibiotic therapy. The usual regime includes two weeks of parental antibiotic treatment followed by four to six weeks of oral antibiotics.5 The optimal antibiotic treatment and duration of antimicrobial treatment for Y. pseudotuberculosis PLA is unknown, and was heterogeneously reported in the published cases.9-14 Y. pseudotuberculosis is usually in vitro susceptible to ampicillin, cephalosporins, tetracycline, ciprofloxacin and aminoglycosides,1,2 and the recommended antibiotic therapy in case of bacteraemia or systemic infection includes a third-generation cephalosporin such as ceftriaxone 2g per day (or alternatively ciprofloxacin) combined with a daily dose of gentamycin 5mg/kg.2,22,23 The treatment regimens of the prior cases of Y. pseudotuberculosis PLA all differed, but included ampicillin, amoxicillin-clavulanic acid, ceftriaxone, quinolones (ciprofloxacin and ofloxacin), gentamicin, and in one case troleandomycin.9-14
Our isolate of Y. pseudotuberculosis was susceptible for all tested antibiotics including ampicillin, but owing to worsening on the clinical condition after drainage, the empiric piperacillin-tazobactam treatment was supplemented with ciprofloxacin. After ribosomal 16S PCR analysis confirmed monomicrobial finding of Y. pseudotuberculosis , oral treatment was finalized with ciprofloxacin alone.
In conclusion, this is the first Danish case of Y. pseudotuberculosis PLA. Y. pseudotuberculosis was identified in pus after liver abscess drainage but not in blood cultures, highlighting the need for drainage for both treatment and for identification of the aetiology. Of note, our patient had diabetes mellitus, which was also reported in prior cases of Y. pseudotuberculosis PLA.