Many chemotherapy regimens are associated with variable periods of mye
losuppression. In cancer patients, neutropenia (less than 500 neutroph
ils/mu L) is the most important risk factor for infections. The incide
nce and severity of infectious complications are related to depth and
duration of neutropenia, with the highest risk if neutrophils are less
than 100/mu L for more than a week. The period required for neutrophi
l recovery is usually short with standard regimens, but prolonged afte
r high dose chemotherapy followed by autologous bone marrow transplant
(ABMT) or peripheral blood stem cell (PBSC) infusion. Under these con
ditions, the administration of granulocyte colony-stimulating factor (
G-CSF) or granulocyte-macrophage colony-stimulating factor (GM-CSF) ac
celerates neutrophil recovery and shortens the duration of hospitaliza
tion. In standard chemotherapy settings, prophylactic use of CSF's is
a matter of debate. Several studies have reached contrasting conclusio
n, but, combining effectiveness and costs, it results that this use of
CSF'S is not to be recommended unless the risk of infections (elderly
patients, reduced marrow reserve) is high. The administration of G-CS
F or GM-CSF to a febrile neutropenic patient (cfr CSF's therapy) short
ens the duration of neutropenia, although no great clinical benefits a
re evident. Nevertheless the identification of subsets of patients wit
h additional risk factors (i.e. absolute neutrophil count <100/mu L at
the onset of fever or delayed neutrophil recovery) should be helpful
in establishing the role of CSF's therapy. When prolonged periods of s
evere neutropenia (less than 500 neutrophils/mu L) are expected, antib
iotics should be prophylactically administered. Fluoroquinolones seem
to be the optimal choice in heavily myelosuppressed patients (ie. bone
marrow transplant recipients). Fluoroquinolones are effective in redu
cing the frequency of gram-negative bacteremia, but, because of incomp
lete coverage, gram-positive infections are becoming increasingly prob
lematic. The association with an agent that can be absorbed orally, ac
tive against gram-positive cocci, seems to be an effective strategy. F
ungal infections are an important cause of morbility and mortality in
severely neutropenic patients. Safety and efficacy of antifungal triaz
oles and the lipid formulations of amphotericin B used prophylacticall
y still require investigation. In patients at high risk for fungal inf
ections, monitoring cultures are predictive for systemic mycoses and s
hould guide prophylactic and therapeutic choices. The standard treatme
nt of oncologic patients with potential infectious neutropenia complic
ations is admission to the hospital and treatment with broad-spectrum
intravenous antibiotics. Until third generation cephalosporin and carb
apenems became available, most neutropenic febrile patients were treat
ed with associations of an aminoglycoside plus a betalactam. Monothera
py with the new antibiotics has proven to be effective as an associati
on therapy and offers advantages in therms of cost and tolerability. W
hether or not vancomycin is included in the initial antibiotic regimen
should be decided on the basis of epidemiological consideration (i.e.
prevalence of meticillin-resistent Staphylococcus aureus or Staphyloc
occus mitis in certain centers). Antifungal therapy is indicated in ne
utropenic patients who remain febrile after one week of broad-spectrum
antibiotics or have recurrent fever. Amphotericin B should be promptl
y administered in patients suspected of invasive mycoses. Selected pat
ients with fever and neutropenia, that can be identified on the basis
of reduced risk of severe complications, do not need hospitalization.
In the first reports, outpatient treatment has proven to be effective,
cost saving and well received by patients but further studies are nee
ded to accurately define low risk status and the optimal home antibiot
ic regimens.