Aztreonam therapy in neutropenic patients with cancer

Combinations of aztreonam/vancomycin, aztreonam/vancomycin/amikacin, and moxalactam/ticarcillin were compared in a prospective randomized trial as empiric therapy for febrile neutropenic cancer patients. Vancomycin was added to aztreonam to provide coverage against gram-positive organisms. Of 535 febrile episodes included in the study, 455 were evaluable. The aztreonam/vancomycin and aztreonam/vancomycin/amikacin combinations were both more effective than the moxalactam/ticarcillin combination in a total of 244 episodes of documented infection. The difference was due to the fact that both aztreonam-containing combinations were more effective than the moxalactam/ticarcillin combination in documented gram-positive infections. The three regimens were equally effective in 67 documented infections due to a single gram-negative bacterial species. (The response rates were 87, 86 and 94 percent for the aztreonam/vancomycin, aztreonam/vancomycin/amikacin, and moxalactam/ticarcillin combinations, respectively.) Aztreonam was effective as the single active antibiotic in the treatment of gram-negative infections in neutropenic patients; however, it must be used in combination with another antibiotic to provide gram-positive coverage.     

肿瘤患者粒细胞缺乏症并感染的经验性抗生素治疗回顾性分析

目的　总结肿瘤患者粒细胞缺乏症并感染的经验性抗生素治疗的临床经验。方法　回顾性分析 119例肿瘤患者 2 4 8例次的临床资料 ,观察头孢哌酮 (CFP)、氧哌嗪青霉素 (PPC)、头孢他定 (CTZ)、亚胺培南 (IMP)与丁胺卡那霉素 (AMK)分别组成BA、PA、CA、TA方案的疗效及副作用。结果　BA、PA、CA、TA方案分别治疗 10 3、87、35和 2 3例次 ,治疗的中位时间为 7～ 8d ,有效率分别为 6 3%、4 8%、6 8%和 74 % ,PA方案明显为低 (P <0 0 5 )。治疗无效者调整治疗 :改用CA或TA方案加减去甲万古霉素 (NVC)及抗真菌治疗 ;总有效率分别为 88%、83%、86 %和 91%。最常见副作用为胃肠功能紊乱 ,腹泻为 12 % ,皮疹和肝转氨酶轻度升高分别为 4 % ;2例NVC和AMK治疗者出现明显的听力下降和肾脏毒性。结论　BA、CA、TA方案具有相似的疗效 ,可作为一线经验性治疗方案。经验性治疗无效者 ,应改用具有更高抗菌活性的方案或尽早采用NCV和 (或 )抗霉菌治疗     

Piperacillin plus amikacin vs. piperacillin plus amikacin plus teicoplanin for empirical treatment of febrile episodes in neutropenic patients receiving quinolone prophylaxis.

A prospective, randomized trial was initiated to evaluate the efficacy of two antibiotic regimens, differing in the agent included with activity against gram-positive bacteria, for the empirical treatment of febrile episodes in neutropenic patients with hematologic malignancies (group 1, piperacillin plus amikacin; group 2, piperacillin plus amikacin plus teicoplanin). After 72 hours of therapy, patients in group 1 who were still febrile were administered teicoplanin and those in group 2 were administered amphotericin B. A total of 158 evaluable episodes were observed within 8 months. The success rate was 50.6% in group 1 and 60% in group 2. The response rate among patients who did not respond to the original regimen increased to 86.7% with the addition of teicoplanin (group 1) and to 90% with the addition of amphotericin B (group 2). There were 86 unexplained febrile episodes and 56 documented episodes of bacteremia (34 caused by gram-positive organisms). Our results indicate that teicoplanin is safe, well tolerated, and effective for the treatment of documented episodes of gram-positive bacteremia and as an empirical agent. The inclusion of teicoplanin in the initial empirical regimen appears unnecessary if a combination of antibiotics active against gram-positive organisms is used, unless infections are due to oxacillin-resistant staphylococci.     

Using teicoplanin for empiric therapy of febrile neutropenic patients with haematological malignancies

S ummary The cumulative experience with teicoplanin in treating febrile neutropenic patients included in three different comparative clinical trials conducted at a single institution during a 3-year period, is presented. 1 52 febrile episodes in 129 neutropenic patients were treated with iv. teicoplanin (6 mg/kg/d) combined with amikacin (15 mg/kg/d) plus ceftazidime (90 mg/kg/d). The study population comprised 75 patients with acute leukaemia and 77 marrow recipients: 53(% (81/152) had a central venous catheter in place and 68%) (103/152) had severe neutropenia (< 100/mm³) at the beginning of the febrile episode.
The overall response rate of the evaluiable febrile episodes was excellent: 88% (107/122) improved. Bacteraemias due to Gram-positive cocci accounted for 75% of the total (42/56) and pathogens in the blood isolates were mostly staphylo-cocci (coagulase-negative 14, coagulase-positive 13) and streptococci (131. The response rate of Gram-positive bacteraemias was good: 88% (37/42) improved and 75% (9/12) of Gram-positive bacteraemias having teicoplanin as the only antibiotic with in vitro activity against the infective strains were cured. Death due to infection accounted for 7% of total febrile episodes (11/152). Side effects were documented in 14(%, of the episodes. In a setting of high prevalence of Grampositive infections caused by strains with a high rate of resistance to aminoglycoside and betalactam antibiotics, there may be an advantage in including teicoplanin in the initial empiric antibiotic regimen for febrile neutropenic cancer patients.     

Ceftazidime plus amikacin versus ceftazidime plus vancomycin as empiric therapy in febrile neutropenic children with cancer

Two antibiotic regimens, ceftazidime plus amikacin and ceftazidime plus vancomycin, were compared in a prospective, randomized clinical trial as empiric therapy in febrile granulocytopenic children with cancer. The rate of response was similar in the two groups (66% vs. 77%). The prevalence of secondary gram-negative bacteremia was higher--but not significantly higher--in the group receiving vancomycin. Adverse reactions also occurred more often in the latter group (35% vs. 4%). Mortality did not differ significantly in the two groups. Adjustment for independent predictors of response to treatment by means of multivariate analysis confirmed the lack of any remarkable difference between the responses to the two regimens. We conclude that the use of vancomycin instead of amikacin in combination with ceftazidime does not significantly improve the outcome of treatment of fever and infection in granulocytopenic children with cancer and that the use of vancomycin is associated with an increased frequency of both secondary infections due to gram-negative bacteria and adverse reactions.     

Vancomycin as part of initial empirical antibiotic therapy for febrile neutropenia in patients with cancer: pros and cons.

Gram-positive organisms predominate as the bacterial pathogens identified in episodes of febrile neutropenia. This has led to increased use of antibiotics with efficacy against gram-positive organisms (often vancomycin) as part of empirical antibiotic regimens for treating febrile neutropenia. Among 101 children randomized to receive amikacin, ticarcillin, and vancomycin or ticarcillin/clavulanate and amikacin along with vancomycin placebo, treatment success in those treated with vancomycin was higher (85% vs. 62%). In 1990, the European Organization for Research and Treatment of Cancer-National Cancer Institute of Canada Clinical Trials Group compared amikacin and ceftazidime with and without vancomycin and concluded that there was no need to include vancomycin in initial empirical antibiotic therapy. Results from another study and a retrospective review of a large clinical trial also support the previous conclusion. In 1999, most experts in the field recommend vancomycin not be part of the initial empirical therapy regimen for treating febrile neutropenia in patients with cancer.     

Combination of Amikacin and either Ampicillin or Cephalotin as Initial Treatment of Febrile Neutropenic Patients

ABSTRACT. A prospective, randomized trial of two antibiotic combinations (amikacin plus either ampicillin or cephalotin) was performed on 39 consecutive episodes of fever in 30 patients with neutropenia and hematological malignancy. Infections were documented as the cause of fever in 37 episodes (95%): in 21 episodes (54%) bacteria or a virus (n=1) were isolated, and in 16 (41% of all episodes) the infection was documented clinically but no pathogen was isolated. The most frequently isolated bacteria were Staph, aureus (38% of all strains), E. coli (13%), and Pseudomonas aeruginosa (13%). Bacteremia occurred in 18% of the febrile episodes. Improvement followed treatment with the combination amikacin plus ampicllin in 73% of 19 cases, and with amikacin plus cephalotin in 55% of 20 cases (p>0.05), giving a total Improvement rate of 64%. Failure of therapy was seen in episodes caused by multiple bacteria or Pseudomonas infections. Mild signs of nephrotoxicity were noted in 13% during both regimens. Audiograms were normal in all but two patients who showed slight high-frequency hearing loss. A second infection occurred in 7 episodes (18%). Thus, the combination of amikacin plus ampidllin was as efficient (but less expensive) as amikacin plus cephalotin in the initial treatment of febrile episodes in neutropenic patients with hematological malignancies.     

Ceftriaxone and amikacin as single daily dose in the empiric therapy for febrile episodes in neutropenic patients

One hundred thirty-three febrile episodes in 115 neutropenic patients with hematologic malignancies were empirically treated with ceftriaxone and amikacin in a single daily dose. An indwelling central venous catheter (CVC) was present in 44 cases. Septicemia was documented in 18 (41%) patients with CVC (13 gram-positive, 5 gram-negative and 1 fungus) and in 30 (34%) patients without CVC (19 gram-positive, 10 gram-negative and 2 fungi). Coagulase-negative staphylococcus was observed in 10 out of 19 blood isolates in the presence of a CVC and in 6 out of 31 blood isolates in patients without CVC. Empiric therapy was successful in 56.4% of cases. Improvement after the addition of vancomycin or teicoplanin was observed in 38.6% of cases with a CVC and in 13.5% of those without (p less than 0.02). Only two patients died from gram negative septicemia, and the substitution of ceftriaxone with another beta-lactam was necessary in only 6% of the cases. Empiric therapy with single daily-dose ceftriaxone and amikacin appears to be effective in febrile neutropenic patients; our data, however, show the high incidence of Staphylococcus epidermidis septicemia and the frequent need to add an anti-gram-positive drug in patients with an indwelling CVC.     

Meropenem monotherapy versus combination therapy with ceftazidime and amikacin for empirical treatment of febrile neutropenic patients

 Infections remain the major cause of morbidity and mortality among neutropenic cancer patients. The current study addresses the question whether monotherapy with the new broad-spectrum carbapenem meropenem exhibits efficacy comparable to that of the standard combination therapy with ceftazidime and amikacin for empirical treatment of febrile neutropenic patients. Seventy-one patients with hematological malignancies (55%) or solid tumors (45%), neutropenia <500/μl, and fever <38.5  °C were randomly assigned to either meropenem (1 g every 8 h) or ceftazidime (2 g every 8 h) and amikacin (15 mg/kg/day) intravenously. Meropenem (n=34) and ceftazidime/amikacin (n=37) were equivalent with respect to the clinical response at 72 h (62% versus 68%) (p<0.05) and at the end of unmodified therapy (59% versus 62%). Gram-positive bacteremia responded poorly in the meropenem and ceftazidime/amikacin group (29% versus 25%), whereas all gram-negative bacteremias responded except for one in the meropenem group caused by Pseudomonas aeruginosa. All patients survived to 72 h. One patient in each group died of gram-positive sepsis resistant to study medication. No significant side effects occurred in any regimen. This study suggests that meropenem monotherapy might be as effective as combination therapy with ceftazidime and amikacin for the empirical treatment of febrile neutropenic patients. Received: 13 June 1997 / Accepted in revised form: 5 December 1997     

Predictive factors of septic shock and mortality in neutropenic patients

Neutropenia is a major risk factor for developing a serious infection. Bacteremia still causes significant mortality among neutropenic patients with cancer. The purpose of this study was to identify risk factors for septic shock and for mortality in neutropenic patients with leukemia and bacteremia. Consecutive samples from 20 patients with acute myeloid leukemia and bacteremia were studied during a 1 year period (January-December 2003). All patients received empirical antibiotic therapies for febrile episodes using ceftazidime plus amikacin. About 110 neutropenic febrile episodes were noted: clinically documented 14.54%, microbiologically documented 16.36% and fever of unknown origin 69.09%. Gram-negative organism caused eight febrile episodes: Pseudomonas (5), Klebsiella (3). Gram-positive organism caused 10 episodes: Staphylococcus (6), Streptococci (2), Enterococci (2). Pulmonary infection accounted for 25% of clinically documented infections. About 14 of the 110 febrile episodes were associated with septic shock causing mortality in 7 patients. In a univariate analysis variables associated with septic shock were: pulmonary infection (OR = 17, p = 0.001), serum bicarbonate < 17 mmol/l (OR = 68, p < 0.001) and serum lactate >3 mmol/l (OR = 62, p < 0.001). Variables associated with mortality were: pulmonary infection (OR = 83, p < 0.001) and serum bicarbonate < 17 mmol/l (OR = 61, p < 0.001). In a multivariate analysis two variables were associated with septic shock: pulmonary infection (OR = 5, p = 0.043) and serum lactate >3 mmol/l (OR = 10, p = 0.003). An elevated serum lactate (>3 mmol/l) and low serum bicarbonate ( < 17 mmol/l) at the onset of bacteremia are useful biomarkers in predicting septic shock and mortality in neutropenic patients.     

Empirical antibiotic therapy in febrile neutropenic patients with acute leukemia

One hundred and ninety-five episodes of fever during the neutropenic phase of chemotherapy in 49 patients with acute leukemia from 1984 to 1987 were analyzed with the following results: 1) Febrile episodes occurred in 80 percent of the neutropenic (less than 500/microliters) phase lasting more than 7 days after chemotherapy. 2) Febrile episodes consisted of 44 (22%) of established septicemia and 111 (57%) of suspected septicemia. 3) The pathogens causing septicemia were 8 GPC, 38 GNB (22 Pseudomonas species) and 6 fungi. Fungemia was confirmed on an average of 4.8 days after the onset of fever. The mortality of septic events was 10 out of 17 episodes (59%) when treated with antibiotics alone, while 8 out 27 (30%) with the combination of antibiotics plus antifungal drugs. 4) The mortality of suspected sepsis was only 2 out of 111 episodes. Eighty-three (75%) of these 111 episodes responded to antibiotics alone, while 26 (23%) cases needed antibiotics plus antifungal drugs. Our results suggest that in febrile neutropenic patient empiric broad-spectrum antibiotic therapy should be initiated which is especially effective for Pseudomonas species, but if fever persists despite more than 4 or 5 days of antibiotic therapy, additional antifungal therapy should be considered.     

Ceftazidime as initial therapy in febrile patients with acute leukemia during induction chemotherapy. Leukemia Group of Middle Sweden.

We studied the efficacy of ceftazidime as initial monotherapy in 82 adult patients with acute leukemia who developed 123 febrile episodes during induction chemotherapy. 88% of the patients survived their febrile episode(s), whereas 10% died of infection. When assessed at 72 h after initiation of treatment (early evaluation), 43/123 episodes (35%) had been successfully treated with ceftazidime. These 43 favourable responses were seen in 15/47 (32%) microbiologically documented infections, 20/46 (43%) clinically defined infections, and 8/30 (27%) fever of unknown origin (FUO). At the resolution of fever (late evaluation) 115 episodes were evaluable, and 48% had responded successfully to ceftazidime. Successful treatment was most frequently observed in FUO, 18/29 (62%). In contrast, only 19/44 (43%) microbiologically documented infections and 18/42 (43%) clinically defined infections were cured during ceftazidime treatment. In bacteremia the response rate was only 8/26 (31%). Thus, this study shows that although ceftazidime can be safely used for initial empirical monotherapy in neutropenic leukemia patients, the need for therapy modification is high and few patients with serious infections are cured with ceftazidime alone.     

Gram-positive infections and the use of vancomycin in 550 episodes of fever and neutropenia

STUDY OBJECTIVE: To determine the appropriate role for vancomycin in neutropenic patients with cancer. To review the incidence, types, and outcome of gram-positive infections in a series of neutropenic patients with cancer. DESIGN: Retrospective review. SETTING: Inpatient units of the Medical and Pediatric Oncology Branches of the National Cancer Institute. PATIENTS: Five hundred and fifty consecutive episodes of fever and neutropenia in patients with cancer randomized prospectively on another study to receive either ceftazidime alone or combination antibiotics for initial empirical therapy. INTERVENTION: Intravenous vancomycin (dosage adjusted by serum levels). MEASUREMENTS AND MAIN RESULTS: Gram-positive organisms were the commonest of the bacterial pathogens isolated (63%). Of the 53 gram-positive organisms accounting for primary infections (isolated at initial presentation), there were 36 staphylococcal isolates (19 coagulase-negative and 17 coagulase-positive), 13 streptococcal isolates (8 non-group D and 5 group D), and 4 polymicrobial isolates. Of the 22 secondary gram-positive infections (occurring after institution of initial antibiotics), there were 10 streptococcal isolates (9 group D and 1 non-group D), 7 staphylococcal isolates (6 coagulase-negative and 1 coagulase-positive), and 5 polymicrobial isolates. Vancomycin was used to treat 26 of the 53 primary infections, but was begun only after knowledge of the isolate in 25. Vancomycin was used to treat 17 of the 22 secondary infections, and begun only after knowledge of the isolate in 14. This approach resulted in no treatment failures for the primary infections, and a single microbiological failure for the secondary infections. There was a tendency towards a greater proportion of secondary gram-positive infections in the monotherapy group compared to the combination therapy group (16 of 282 compared with 6 of 268 respectively, P2 = 0.04 by the chi-squared test); but all were treated successfully. CONCLUSION: Vancomycin need not be included in routine empirical therapy for febrile neutropenic patients, but should be added when clinical or microbiological data suggest the need.     

A randomized comparative trial of three aminoglycosides--comparison of continuous infusions of gentamicin, amikacin and sisomicin combined with carbenicillin in the treatment of infections in neutropenic patients with malignancies.

Continuous infusions of gentamicin, amikacin or sisomicin combined with carbenicillin were compared in a randomized study in the treatment of 572 febrile episodes in 281 patients with cancer. The three treatments (C+A, C+A and C+S) were equally effective with no significant differences in response rate overall (67%, 68%, 67%) or in any infection, except septicemia where C+G had a significantly lower response rate than the other two groups. Pneumonia, the most common infection, had the lowest response rate for all three groups (45-50%). Klebsiella spp. were the most common pathogens and showed a lower response rate than other gram-negative bacilli (P = 0.003). Patients with persistent severe neutropenia had a response rate of 56%. Azotemia was significantly less common in patients with documented infection treated with C+A than in the C+S group. Combinations of carbenicillin plus an aminoglycoside antibiotic are effective for the treatment of infections in neutropenic patients.     

Penicillin or vancomycin plus ceftazidime in febrile neutropenic patients after stem cell transplantation

The response of gram-positive cocci to third generation cephalosporin therapy in febrile neutropenic patients is not optimal. We evaluated the safety and efficacy of ceftazidime plus penicillin (C + P) and compared it to ceftazidime plus vancomycin (C + V) in febrile neutropenic patients. The study includes 64 patients admitted to the Department of Haematology. Thirty-six patients were treated with C + V and 28 patients with C + P. Control of infection was observed in 78% of the patients in the C + V group and in 57% of the patients in C + P group (p = 0.5). Infection was the cause of the death of 1 patient in each group. We conclude that the combination of C + P has the same activity as the combination of C + V in febrile neutropenic patients. The morbidity and mortality were identical in both groups but cost effectiveness was in favour of the C + P group.     

Piperacillin–Tazobactum Plus Amikacin Versus Ceftazidime Plus Amikacin as Empirical Therapy for Fever in Neutropenic Patients with Hematological Malignancies

Infections are one of the main cause of death in cancer patients particularly when granulocytopenia is present. A number of drugs have been used for the treatment of neutropenic patients with fever. Most published literature has shown piperacillin–tazobactum in combination with amikacin to be significantly more effective than ceftazidime plus amikacin in empirical treatment of febrile episodes in patients with neutropenia. In view of the reported literature we have tried this combination in our febrile neutropenic patients with haematological malignancies at PGIMS Rohtak. It was an open randomized trial. Patients were divided into two groups of 20 each. In the first group (group A) piperacillin–tazobactum (4 + 0.5 g 6 hourly) with single daily dose of amikacin 20 mg/kg was given. In the second group (group B) ceftazidime 40 mg/kg every 8 hourly with single daily dose of amikacin 20 mg/kg was given. The most common site of infection was blood followed by urinary tract, respiratory tract and oral cavity. 13 (65%) patients in group A and 12 (60%) patient in group B showed clinical success. In our study however in our patients a better response was seen in patients with piperacillin–tazobactum + amikacin (65% vs. 60%). So it is recommended that piperacillin–tazobactum + amikacin should be given in febrile neutropenic patients with haematological malignancies.     

Ceftazidime as first-line therapy for fever in acute leukaemia

Fifty patients with acute non-lymphocytic leukaemia were treated by random allocation with either ceftazidime alone or a combination of piperacillin, netilmicin and cefotaxime for 65 febrile neutropenic episodes. Nineteen of 33 patient episodes (58%) responded to ceftazidime alone compared with 21 of 32 episodes (66%) treated with the combination. There was one infective death in a patient given the combination; rates of documented superinfection were low. The treatment groups appeared identical in terms of patient demography, underlying disease and other risk factors, though patients with a clinical site of infection responded more slowly than those without. Bacteraemia per se did not appear to influence outcome. Bactericidal serum concentrations greater than or equal to 8 X the minimum bactericidal concentration were predictive of a rapid response (within 4 days) to antibiotics. Furthermore, serum from patients treated with ceftazidime maintained adequate cidal activity against Pseudomonas aeruginosa for longer than that obtained from patients treated with the three-drug combination. Ceftazidime was shown to be a safe and effective alternative to the three-drug combination for the initial management of febrile neutropenic episodes in leukaemic patients.     

Imipenem-cilastatin for empirical therapy in neutropenic patients with fever: an open study in patients with hematologic malignancies

In adult neutropenic patients with hematological malignancies, we explored imipenem-cilastatin as empirical antibiotic therapy in a dose of 500 mg four times daily. Changing to second-line treatment was only resorted to if clinical deterioration, new infections or recurrence of fever occurred. A clinically or microbiologically documented infection was apparent in 115 of 150 episodes studied (76.7%). Imipenem-cilastatin was successful in 70.7% of all episodes and in 67.8% of all proven infections. Modification was necessary and successful in 22.0% of all episodes. 11 patients died while still febrile, 6 of them by infection (4%) and 5 because of progressive disease (3.3%). Imipenem-cilastatin is safe initial therapy in neutropenic febrile patients.     

Teicoplanin in the treatment of Gram-positive bacteraemia in neutropenic patients

S ummary The increasing incidence of bacteraemia caused by Gram-positive bacteria in neutropenic patients prompted the authors to evaluate, in a prospective trial, the role of teicoplanin in the treatment of this infection. Over a 15-month period. 76 cases of bacteraemia (out of 265 evaluable episodes of fever) were observed at the Division of Haematology. University La Sapienza. Kome. Of the 76 cases studied, 46 (60%) were caused by Gram-positive bacteria and 28 (37%) were caused by Gram-negative bacilli. All febrile episodes were treated randomly and empirically with piperacillin plus amikacin with or without teicoplanin. Overall, 41 (54%) of the 76 cases of bacteraemia responded to the initial antibiotic regimen: with subsequent modifications the response rate rose to 96%.
In the treatment of Gram-positive bacteraemia. first-line administration of teicoplanin was found to be associated with early defervescence and with a significantly higher rate of success without modification of treatment (P < 0.01). Addition of teicoplanin as second-line therapy produced a favourable outcome in 12 (70%) out of 17 cases of bacteraemia unresponsive to the initial piperacillin + amikacin regimen. No cases of Gram-positive bacteraemia associated with septic shock or adult respiratory distress syndrome were observed in either treatment group. Only two late deaths were observed, and these occurred in patients with streptococcal septicaemia who were not receiving early teicoplanin.
The above data do not endorse the use of glycopeptide antibiotics in the early treatment of fever in neutropenic patients: rather, these compounds should be reserved for proven or presumed Gram-positive infections which do not respond to initial beta-lactam/aminoglycoside treatment.     

Clinical Infections and Bloodstream Isolates Associated with Fever in Patients Undergoing Chemotherapy for Acute Myeloid Leukemia

Summary Background: Patients with acute myeloid leukemia (AML) are at high risk for infections. The aim of this study was to identify the sources of fever and the type of pathogens that cause bloodstream infection in patients with AML undergoing cytotoxic chemotherapy and antibiotic prophylaxis. Patients and Methods: The source of fever and the type of pathogens causing bloodstream infection were identified for 129 febrile episodes experienced by 42 patients with AML receiving cytotoxic chemotherapy and antibiotic prophylaxis. Results: A source of fever was identified in 81% of all febrile episodes. Mucositis (21.7%), pneumonia (13.2%), central venous catheter infection (12.4%), neutrophenic enterocolitis (9.3%) and invasive fungal disease (9.3%) were the most common sources of fever. Of 16 central venous catheter infections, seven (43.8%) were not associated with local signs. 49 febrile episodes (37.9%) were associated with bloodstream infections, of which 14 (28.6%) were polymicrobic and seven (14.3%) had an undefined source of infection. Bloodstream infection was commonly associated with cellulitis (60%), mucositis (57.1%), central venous catheter infection (55.6%), neutropenic enterocolitis (41.7%) and invasive fungal disease (41.7%). Gram-positive microorganisms were the most common blood isolates (75.8%). Gram-negative bacteremic infections occurred in eight episodes (12.1%) experienced by patients who were nor receiving ciprofloxacin prophylaxes at the time of bacteremia. Noninfectious sources of fever accounted for 23 (17.8%) of the 129 febrile episodes. Conclusion: Although the spectrum of pathogens that cause infection in this group of patients has shifted from gram-negative to gram-positive bacteria, the most common sources of infection remain the same as previously described and they mainly involve integumental surfaces. Received: December 5, 1999 · Revision accepted: August 8, 2000