A patient on hydroxyurea for sickle cell disease who developed an opportunistic infection

Recent reports1-7 indicate that hydroxyurea (HU) is safe and effective in the treatment of patients with severe sickle cell disease (SCD). In particular, there have been no reports of any patient developing an opportunistic infection while taking this drug. Below, we report a case of Cryptosporidium infection in a patient with SCD on HU treatment.

A 36-year-old African American woman with known sickle β⁺thalassemia with past medical history significant for multiple hospitalizations for severe pain crises was admitted to the hospital for a severe painful episode. In addition, this time she complained of epigastric pain with nausea, vomiting, and watery diarrhea for 2 days prior to admission. Review of systems was negative for other symptoms. Her last blood transfusion was 6 months prior to admission, and she had only rarely required blood transfusions in the past. Her medications were 500 mg HU (for 20 months), acetaminophen with codeine, methadone, folic acid, conjugated estrogens, and ranitidine. Upon physical examination, she was afebrile and the significant findings were tachycardia (pulse rate, 109/min) and epigastric tenderness to palpation without guarding or rebound. Laboratory data were significant for a low white blood cell count of 3 × 10⁹/L with a normal differential and hemoglobin level of 9 g/dL. Initial stool tests for fecal leukocytes, ova and parasites, andClostridium difficile toxin was negative.

She was treated symptomatically for a presumptive diagnosis of viral gastroenteritis. One week after admission, after she failed to respond to treatment, she underwent colonoscopy. Colonic mucosal biopsies were positive for Cryptosporidium. Subsequently, her stool culture results came back as positive for Cryptosporidium as well. Further workup revealed a negative test for the human immunodeficiency virus (HIV). But her absolute CD4 count was low at 0.234 × 10⁹/L (normal, 0.4 to 1.77 × 10⁹/L). She was treated with paromomycin and octreotide. The patient required multiple subsequent admissions for similar symptoms and was treated in similar fashion. She later developed cholecystitis and required cholecystectomy. Pathology of the gallbladder revealed eosinophilic cholecystitis, consistent with a parasitic infection.8 Following removal of the gallbladder, which can act as a reservoir ofCryptosporidium,8 the patient's symptoms resolved. The patient's subsequent absolute CD4 count normalized (0.455 × 10⁹/L) 3 months after discontinuation of HU. A repeat value was 0.553 × 10⁹/L 13 months after stopping HU.

HU is a commonly used drug in SCD and has been reported to be safe. To our knowledge there has been no reports of opportunistic infections associated with HU, and it is rare to see such infections in patients with SCD. The finding in our patient of Cryptosporidiumgastroenteritis, which is an opportunistic infection, raised the question of whether there is any association between HU and such infections.

Opportunistic infections are usually associated with immunosuppression and/or low CD4 counts. We have reported that CD4 counts are normal in sickle cell patients both during steady state and painful crisis.9 There has been a report that HU can lower CD4 counts in HIV patients.10 Our hypothesis is that HU lowered the CD4 count in our patient, perhaps predisposing her toCryptosporidium gastroenteritis. Currently we are testing this hypothesis by checking CD4 counts in patients that are being treated with HU and those who are not.