Citrobacter freundii-induced cold agglutinin hemolysis

Abstract

Dear Editor, A 64-year-old Hispanic woman with a past medical history of hypertension, diabetes mellitus type 2, and dyslipidemia was admitted with symptoms of fever, left-sided flank pain, and blood in her urine of 2 days duration. Seven days before admission, the patient had been treated as outpatient for urinary tract infection with trimethoprim/sulfametoxazole (Bactrim). Physical exam revealed left costovertebral angle tenderness and some scleral icterus. The patient was awake, alert and oriented, and showed no changes in mental status. Her temperature was febrile, 38.5°C. Initial laboratory results revealed a hemoglobin (Hgb) of 8.2 g/dL, hematocrit of 23.6 g/dL, platelet count of 524,000×10/μL, creatinine of 1.1 mg/dL, a total bilirubin of 1.6 mg/dL, haptoglobin of 5.8 mg/dL, and a lactic dehydrogenase of 699 U/L. Blood cultures returned positive for Citrobacter freundii within 24 h of collection. A direct Coombs’ test (DAT) was positive for C3d but negative for IgG, and high-titer cold agglutinins were detected. Tests for influenza A and B, Epstein–Barr virus, and Mycoplasma were negative. The patient was treated for Citrobacter infection and transfused with warm blood. She was kept in a heated room with body warming blankets. Two days after admission, the patient’s Hgb dropped to 5.4 g/dL, with increasing lactic dehydrogenase levels and indirect bilirubin suggesting an ongoing active hemolysis. The patient was started on intravenous methylprednisolone 80 mg IV daily for 10 days, and the hemolysis resolved. Her Hgb improved gradually to 7.8 g/dL, and repeat blood cultures as well as a repeat DAT were negative. Subsequently, she was discharged home on a tapering dose of oral corticosteroids. Four weeks after discharge, corticosteroids were completely stopped without any relapse of cold agglutinins. Several viral and bacterial pathogens have been reported in the literature as causative factors for secondary cold agglutinin disease. Cytomegalovirus, parvovirus B19, and varicella-zoster virus are commonly associated viral etiologies. Mycoplasma pneumoniae, Chlamydia, Legionella, and Leptospira have been reported as bacterial predisposing agents to cold agglutinin disease as well [1, 2]. To our knowledge, ours is the first case of Citrobacterassociated cold agglutinin hemolysis to be reported in the literature. Citrobacter bacteria are gram-negative rods belonging to the Enterobacteriaceae family and are often found in water supplies as well as food sources, in addition to other parts of the environment [3]. In some humans, Citrobacter is part of the normal bacterial flora of the colon, but this occurs in only a small percentage of the population. In our patient, the concomitant sepsis with C. freundii is the likely precipitating factor for the cold agglutinininduced hemolytic anemia. The spontaneous resolution of cold agglutinins as well as hemolysis after the treatment of Citrobacter sepsis supports Citrobacter as the most likely trigger [4]. C. freundii has been reported in the past as one of the causes of Shiga toxin-mediated hemolytic uremic syndrome [5]; however, cold agglutinin disease has never been associated with it. Like in cases associated with Mycoplasma pneumoniae, it is possible that some polyclonal IgM cold agglutinins arise in association with Citrobacter A. Kumar (*) Department of Medicine, Saint Michael’s Medical Center, 111 Central Avenue, Newark, NJ 07101, USA e-mail: Ice9996@aol.com