Fever and neutropenia in cancer patients: the diagnostic role of cytokines in risk assessment strategies.

Cancer patients treated with chemotherapy are susceptible to bacterial infections. Therefore, all neutropenic cancer patients with fever receive standard therapy consisting of broad-spectrum antibiotics and hospitalization. However, febrile neutropenia in cancer patients is often due to other causes than bacterial infections. Therefore, standard therapy should be re-evaluated and new treatment strategies for patients with variable risk for bacterial infection should be considered. This paper reviews the changing spectrum of microorganisms and resistance of microorganisms to antibiotics in infection during neutropenia and discusses new strategies for the selection of patients with low-risk for bacterial infection using clinical and biochemical parameters such as acute phase proteins and cytokines. These low-risk patients may be treated with alternative therapies such as oral antibiotics, early discharge from the hospital or outpatient treatment.

Comparison of imipenem versus cefuroxime plus tobramycin as empirical therapy for febrile granulocytopenic patients and efficacy of vancomycin and aztreonam in case of failure.

143 aplastic episodes with fever in 91 haematological patients with granulocytopenia were treated empirically in a randomized prospective study using either imipenem (Imi) or a combination of tobramycin and cefuroxime (T/C). Response after 72 h was significantly better in patients receiving Imi (44/75 vs 27/68, p < 0.05). This was seen especially in patients with bacteriologically proven infections where the isolated staphylococci and streptococci were more susceptible to Imi. In both groups, patients who failed to respond to the initial antibiotic therapy were given vancomycin and aztreonam (V/A). The response rate after another 72 h, measured using the same criteria as after the first 72 h, did not differ statistically between the groups. One patient in each study group died from the bacterial infection, both from Gram-positive bacteraemia. Duration of fever was significantly shorter in the Imi group (4 days vs 7 days, p < 0.04). Serum peak and trough concentrations of the antibiotics were comparable. Both regimens were well tolerated. Our results show that monotherapy with imipenem is superior to the combination of tobramycin and cefuroxime during the first 72 h of therapy and can be safely administered to neutropenic patients with predominantly Gram-positive infections. A combination of vancomycin and aztreonam, given when initial imipenem treatment has failed, was effective in only a few patients. Adjuvant glycopeptide therapy from the outset in the treatment of febrile granulocytopenic patients did not seem worthwhile.

Selective oral antimicrobial prophylaxis for the prevention of infection in acute leukaemia-ciprofloxacin versus co-trimoxazole plus colistin. The EORTC-Gnotobiotic Project Group.

230 leukaemic patients were entered into a randomised, prospective, multicentre trial of either ciprofloxacin (1 g/day) or co-trimoxazole (1920 mg/day) plus colistin (800 mg/day) for the prevention of infection during granulocytopenia. Bacteraemia due to resistant gram-negative rods occurred only in the co-trimoxazole-colistin group though both regimens were effective for selective gastrointestinal tract decontamination. However, there were fewer patients without any infective complications (31% vs. 18%: P = 0.02), fewer febrile days [mean (S.D.) 5.9 (1.1) vs. 8.2 (1.4): P = 0.0242], a lower proportion of infective events (0.9 (0.16) vs. 1.2 (0.18): P = 0.005) and fever occurred later (median 19 vs. 14 days: 0.025 less than P less than 0.05) in the co-trimoxazole-colistin group. The choice of prophylactic regimen therefore appears to depend upon whether or not protection against gram-negative infection is required or better systemic prophylaxis overall.

Prevention of infection in acute leukemia.

In a randomized study comparing cotrimoxazole plus colistin with ciprofloxacin, each in combination with nonabsorbable antimycotics, the incidence of major infections in terms of septicemias and pneumonias as well as of minor infections and episodes of unexplained fever (FUO) was higher in patients treated with ciprofloxacin. In cases of microbiologically documented infections, gram-positive cocci dominated by far. In surveillance cultures of oral washings and of feces, gram-negative enterobacteria were only rarely detected; however, large numbers of cultures were positive for Acinetobacter species. There were four cases of documented Pneumocystis carinii pneumonia in patients not receiving cotrimoxazole. The incidence of documented mycotic infections as well as the detection of fungi in surveillance cultures was similar in both treatment groups. A decrease in the number of adverse events, especially of allergic reactions, could not be achieved by the administration of ciprofloxacin. In conclusion, cotrimoxazole plus colistin in combination with nonabsorbable antimycotics remains the standard regimen for prevention of infection in patients with acute leukemia undergoing aggressive remission induction therapy. A detailed analysis of study II will be prepared for publication.

Ciprofloxacin and norfloxacin for selective decontamination in patients with severe granulocytopenia.

In a randomized multicenter study, ciprofloxacin and norfloxacin, each in two different dose regimens and in combination with non-absorbable antimycotics, were administered to 51 patients with acute leukaemia undergoing aggressive remission induction chemotherapy for infection prevention. Both drugs showed an effective elimination of gram-negative potential pathogens and Staphylococcus aureus not affecting the anaerobic flora of the gastrointestinal tract. A low incidence of side effects and a satisfactory patient compliance could be observed. A daily dosage of 1,000 mg ciprofloxacin or 800 mg norfloxacin is recommended for infection prevention in severely granulocytopenic patients.