A 79-year-old man, wheelchair-bound because of residual right hemiparesis from remote stroke, presented with increased fatigue, weakness, and recent history of febrile illness that had resolved spontaneously. Results of physical examination were normal except for neurologic findings related to the stroke. Medications included fluoxetine hydrochloride (Prozac, Dista Products, Indianapolis, Indiana), warfarin, glipizide, diphenhydramine, and tobramycin-dexamethasone ophthalmic drops.

Hemogram showed a leukocyte count of 2800 cells/mm³ with granulocytopenia and significant monocytosis (0% segmented cells, 11% band cells, 2% metamyelocytes, 45% lymphocytes, 36% monocytes, and 6% eosinophils). Hemoglobin level (16.2 g/dL), platelet count (277 000 cells/mm³), and mean corpuscular volume (92 fL) were normal.

Therapy with all medications was discontinued, and the absolute neutrophil count rapidly returned to normal Figure 1 Therapy with fluoxetine hydrochloride was restarted because we found no convincing published evidence that this drug could cause neutropenia. After three daily 20-mg doses of the drug, however, severe neutropenia recurred; this condition resolved rapidly after therapy was discontinued (Figure 1). Glipizide and warfarin were then administered, and neutropenia did not recur.

View larger version (14K): [in this window] [in a new window]   Figure 1. Clinical course of patient with neutropenia associated with fluoxetine hydrochloride (Prozac). Total leukocyte (solid line) and absolute neutrophil counts (dotted line) are shown.

Serum drug-dependent neutrophil antibodies, obtained just before therapy with fluoxetine hydrochloride was restarted, were not detected by in vitro testing in the presence or absence of fluoxetine hydrochloride, glipizide, or diphenhydramine (assays were done at the Blood Center of Southeastern Wisconsin, Milwaukee, Wisconsin). Nonetheless, the rapidity with which neutropenia recurred after rechallenge with fluoxetine hydrochloride is consistent with a drug-related antineutrophil antibody mechanism for the neutropenia.

To our knowledge, this constitutes the first well-documented case of fluoxetine hydrochloride-associated neutropenia. It is important to recognize that severe neutropenia may complicate the administration of this commonly prescribed agent.