Rev Esp Quimioterap, Septiembre 2002; Vol.15 (Nº 3): 257-263
ã 2002 Prous Science, S.A.- Sociedad Española de Quimioterapia
Original
Immediate vs. delayed imipenem treatment
in cancer patients with profound neutropenia
induced by high-dose chemotherapy:
Results of a randomized study
J.A. García-Sáenz, M. Martín, A. Casado, P. Pérez-Segura, I. Manrique, L. Flores, J.A. Macias, J.C. Cámara,
C. Perezagua and E. Díaz-Rubio
Department of Medical Oncology, Hospital Universitario San Carlos, Madrid, Spain
SUMMARY A study was carried out to compare the use of prophylactic imipenem administered at the onset of profound neutropenia (immediate) with therapeutic imipenem administered at the onset of neutropenic fever (delayed) in cancer patients treated with high-dose chemotherapy. A total of 65 patients who were scheduled to receive two cycles of high-dose cyclophosphamide, etoposide, cisplatin (CEP) chemotherapy were randomized to receive imipenem either at presentation of neutropenia (immediate imipenem arm, prophylactic arm) or at commencement of neutropenic fever (delayed imipenem arm, therapeutic arm). Treatment was crossed over when the second CEP chemotherapy cycle was received. Of the 65 patients, 41 received the two planned cycles and 24 received only the first. Compared with the delayed imipenem arm, the immediate imipenem arm was associated with lower fever incidence (86.3% vs. 100%, p=0.0142) and Gram-negative bacteria infection [4/51 (7.8%) vs. 14/55 (25.5%), OR=0.24, p =0.031]. There were fewer episodes of pneumonia (2% vs. 12.7%), septic shock (0% vs. 3.6%) and deaths from infection (0% vs. 3.6%), but these differences did not reach statistical significance. With regard to delayed imipenem, for every seven patients with immediate imipenem, one episode of febrile neutropenia was avoided; for every six patients administered immediate imipenem, one case of Gram-negative infection was avoided; and for every nine patients administered immediate imipenem, one episode of pneumonia was avoided. There were no differences in the incidence of Gram-positive infections nor in the length of hospitalization between the two treatment arms. In conclusion, compared to its conventional delayed use, immediate imipenem significantly reduces the frequency of febrile neutropenia and Gram-negative infections in patients with high-dose chemotherapy.
Tratamiento inmediato o tardío con imipenem
en pacientes con cáncer y neutropenia
por quimioterapia a dosis altas:
Resultados de un estudio comparativo aleatorizado
RESUMEN Se realizó un estudio para comparar el uso profiláctico de imipenem cuando aparece neutropenia intensa (tratamiento inmediato) y el uso terapéutico de imipenem cuando se inicia una fiebre neutropénica (tratamiento tardío) en pacientes con cáncer tratados con quimioterapia a dosis altas. Participaron 65 pacientes que iban a recibir dos ciclos de quimioterapia con ciclofosfamida, etopósido y cisplatino (CEP) a dosis altas. Se distribuyó aleatoriamente a los pacientes en dos grupos para recibir imipenem al aparecer neutropenia (brazo tratado con imipenem inmediato, brazo con tratamiento profiláctico) o bien al inicio de una fiebre relacionada con la neutropenia (brazo tratado con imipenem tardío, brazo con tratamiento terapéutico). Cuando se llevó a cabo el segundo ciclo de quimioterapia CEP, se cruzaron los tratamientos con imipenem. De los 65 pacientes, 41 recibieron los dos ciclos previstos y 24 sólo el primer ciclo. En comparación con el brazo de imipenem tardío, el brazo tratado con imipenem inmediato se asoció a una menor incidencia de fiebre (86,3% frente al 100%, p=0,0142) y a menos infecciones por gramnegativos (4/51 [7,8%] frente a 14/55 [25,5%], OR=0.24, p=0.031). Hubo menos episodios de neumonía (2% frente a 12,7%), "shock" séptico (0% frente a 3,6%) y muertes debidas a infección (0% frente a 3,6%), aunque estas diferencias no llegaron a ser estadísticamente significativas. Con respecto al imipenem de administración tardía, por cada siete pacientes tratados con imipenem inmediato se evitó un episodio de fiebre relacionada con la neutropenia; por cada seis pacientes tratados con imipenem inmediato se evitó un caso de infección por gramnegativos; y por cada nueve pacientes tratados con imipenem inmediato se evitó un episodio de neumonía. No hubo diferencias en la incidencia de infecciones por grampositivos ni en la duración de la hospitalización entre los dos brazos de tratamiento. En conclusión, y comparado con el uso tardío convencional, el tratamiento con imipenem inmediato reduce la frecuencia de neutropenia febril y las infecciones por gramnegativos en los pacientes que reciben quimioterapia a dosis elevadas.
Cancer patients who present with granulocytopenia as a result of intensive chemotherapy have a high risk of developing infections which can be lethal if the empiric antibiotic treatment is not implemented at the first signs of infection; this is usually fever without an apparent focus of infection. Patients with febrile neutropenia have a probability of 48-60% of presenting infection. With a neutrophil count of <100 cells/µl, it is likely that 16-20% of such febrile episodes are associated with bacteremia (1). If the bacteremia is due to a Gram-negative organism, the risk of death is high. Strategies for preventing such infections in neutropenic patients include environmental protection measures and the use of prophylactic antimicrobial agents, usually administered orally with the aim of selective decontamination of the gastrointestinal tract. However, the administration of oral prophylactic antimicrobial agents in the treatment of neutropenic fever continues to be controversial. Conventionally, fluoroquinolones and trimethoprinsulfamethoxazole have been used (2). The use of these agents does not eliminate the risk of infection, it can favor the appearance of resistant Gram-negative microorganisms and, perhaps, an excess of infections by Gram-positive bacteria. As such, other agents and routes of administration would appear to be needed for this patient population
Once the first signs of infection, usually fever, appear, the standard clinical practice is to implement empiric intravenous broad-spectrum antibiotic administration (3). It is clear that empiric intravenous administration of an antibiotic active against Gram-negative bacteria great diminishes the mortality rate in these patients (4, 5).
Traditionally, combinations of b-lactam and aminoglucosides have been used as the empiric treatment of established neutropenic fever. Current availability of low-toxicity, broad-spectrum antibiotics active against Gram-negative and anaerobic bacteria has been instrumental in their widespread use in monotherapy (6-9). Imipenem is a carbapenem antibiotic active against the majority of Gram-negative and anaerobic bacteria (10). A meta-analysis of imipenem as empiric treatment of febrile neutropenia indicates it as an effective alternative to other more classical forms of treatment (11). Studies that have compared imipenem against cephalosporins in monotherapy (12-14), or against a combination of antimicrobial agents (15-20), conclude that imipenem has the same, or better efficacy, when used to treat these patients.
Imipenem is used in our hospital in monotherapy as an empiric treatment of febrile neutropenia induced by high-dose cyclophosphamide, etoposide, cisplatin (CEP) chemotherapy; this is because of the unacceptable renal toxicity induced by aminoglucosides in patients administered cisplatin (21).
In 1992, we initiated a prospective, randomized crossover study to compare imipenem administered in the standard manner: i) when the neutropenic patient presents fever (therapeutic imipenem), with its immediate administration from the onset of profound neutropenia; ii) without awaiting the onset of fever (prophylactic imipenem), in patients with solid tumors who had received high-dose CEP chemotherapy.
PATIENTS AND METHODS
Patients eligible for the present study presented solid tumors in partial or complete response to prior conventional chemotherapy, and were scheduled to receive intensive consolidation chemotherapy with two courses of CEP regimen, with or without autologous hemopoietic progenitor cell support. Other inclusion criteria included: age between 16 and 65 years; no other serious concomitant illnesses; ECOG performance status of 0-1; adequate cardiac, renal and hepatic functions; neutrophil count of >2000/µl; hemoglobin >10 g/dl; and platelets >100´109/l.
The CEP regimen consisted of cisplatin 75 mg/m2 on days 1 and 5, etoposide 150 mg/m2/12 h on days 1-3, and cyclophosphamide 2.25 g/m2 on days 4-5 with filgrastim granulocyte-colony stimulating factor (G-CSF) support. Furthermore, patients with diminished bone marrow reserve received an infusion of autologous peripheral blood progenitor cells (PBPC) previously obtained by cytapheresis using mobilization with filgrastim. Patients were hospitalized in single rooms during treatment. During aplasia, patients were kept in isolation and the usual methods of infection prevention, i.e., diet of low bacterial content, hand washing and use of masks by health workers and patients' visitors, were employed. Patients were discharged when they no longer required platelet transfusions and had neutrophil counts of >500/µl on three consecutive days and no fever.
Neutropenia was defined as an absolute neutrophil count of <500 cell/µl [neutropenia grade IV on the World Health Organization (WHO) scale]. Febrile neutropenia was defined as neutropenia grade IV with fever of >38.5 °C on one occasion, or fever of >38 °C on three occasions in 24 h, not related to the administration of any blood derivatives or other pyrogen-contaminated products.
In accordance with the standard procedures at our institution, from the start of chemotherapy treatment all the patients received oral prophylactic antibiotics: ciprofloxacin 500 mg/12 h; aciclovir 750 mg/8 h; and fluconazole 200 mg/12 h.
The study protocol was approved by the hospital ethics committee, and all patients provided their written informed consent to participate in the trial prior to enrollment.
Laboratory measurements
Each patient had a daily hematogram and general biochemistry analyses (including urea and electrolytes, creatinine, calcium, magnesium, total proteins, albumin, total bilirubin, alkaline phosphatase, lactic dehydrogenase, and serum levels of aspartate aminotransferase and alanine aminotransferase). In each episode of febrile neutropenia, hematocultures were performed on samples taken from the central Hickman catheter inserted in all patients and from a peripheral vein. Other culture samples were obtained from urine, sputum (if applicable), as well as any other site that was considered appropriate. Chest x-rays and other imaging diagnostic tests were performed as considered clinically necessary.
In the case of thrombopenia (<15 platelets/109/l or <20 platelets/109/l with signs of hemorrhage or fever), the appropriate units of platelets were infused. Similarly, when the hematocrit was <25 %, the appropriate concentrated red cells were administered.
Study objectives
The objective of the study was to compare the effectiveness of both schemes of imipenem in the management of neutropenia induced by myelosuppressive chemotherapy. The principal end-points of the study were the incidence of Gram-negative infections, septic shock, and deaths attributable to infection. The secondary end-points of the study were the number of days with fever, and the number of days of hospitalization.
Study design
The study was prospective and randomized with a crossover design. In the first cycle of CEP, patients were chronologically assigned to a random number list and were then randomized to receive one of the two following options: i) control arm, with the imipenem 1 g/8 h from the start of the febrile neutropenia (delayed imipenem); or ii), the experimental arm, with imipenem 1 g/8 h from the start of grade IV neutropenia without awaiting the development of fever (immediate imipenem).
If the fever persisted for 48 h following the first febrile episode, in addition to imipenem patients received vancomycin 1 g/12 h with serum level monitoring. If the fever persisted 120 h from the first febrile episode, amphotericin B was added to the above agents. Specific antibiotic treatment was also added to imipenem when the specific bacterial cause was identified in cultures. In the second cycle of CEP, the patients received the schedule of imipenem conversely, to that received in the first cycle.
The results were analyzed with respect to the effect of the intervention, the time-course effect and the sequence effect. For the effect of the intervention, data were analyzed from 106 cycles of CEP in which the results of the empiric arm were compared with those from the prophylaxis arm. For the time-course effect, as with the sequence effect, the data were obtained from 82 cycles corresponding to the 41 patients who completed the two cycles of CEP. In the time-course effect, the results obtained from the first cycle of CEP were compared with those obtained from the second. In the sequence effect, the data obtained from the administration of immediate-delayed antibiotic administration were compared with the inverse sequence.
All analyses of data were carried out using the EPIDAT2 program for Windows. The means of the results of the qualitative variables were evaluated using odds-ratio (OR) and a 95% confidence interval (CI). Chi-squared contingency tables were used with the Yates correction when contents squares had less than 5 values.
The OR and 95% CI limits were calculated for each of the following variables: episodes of febrile neutropenia; incidence of pneumonia; number of positive cultures; species of bacterial organisms isolated; incidence of septic shock; and death from infection. For the quantitative variables, i.e., days of hospitalization, duration of fever, transfusions of platelets and blood products, tests of comparisons of means were performed with these independent variables.
A current parameter of considerable clinical importance is that of the number-needed-to-treat (NNT) or number of patients who must be treated for a given period to achieve or to prevent an event. The NNT is the reciprocal of the absolute risk reduction. The absolute risk reduction is the difference in event rates for the two groups, treatment and control (22).
RESULTS
Sixty-five patients were included in the study from December 1992 to July 1998. The two therapeutic groups were well balanced with respect to demographic parameters. Forty-one patients received the two planned cycles of CEP, while 24 received only one cycle. In 20 of the 106 cycles administered, PBPC infusions were administered in a total of 14 patients. The median age of the patients was 47 years (range: 17-63 years). The characteristics of the patients in the study are presented in Table 1 and the results of the effect of the intervention are summarized in Table 2.
With regard to febrile neutropenia, seven of the 51 patients in the immediate imipenem arm (13.7%) did not have neutropenic fever, compared with 100% of the delayed imipenem arm (OR=0.05, 85% CI=0.39-0.76; p=0.0142). In the immediate imipenem treatment group, 12 bacterial infections (eight Gram-positive and four Gram-negative) were isolated, compared with 21 (seven Gram-positive and 14 Gram-negative) in the delayed imipenem treatment group. The difference between the two groups with respect to Gram-negative infection was statistically significant (OR=0.24, 95% CI=0.06-0.90; p=0.031). In the comparison of means of duration of fever, duration of neutropenia and the number of platelet transfusion requirements, there were no statistically significant differences.
The incidence of pneumonia was lower in the prophylactic treatment group compared with the empiric treatment group (2% vs. 12.7%), but the difference did not reach statistical significance. Two episodes of septic shock (3.6%) were reported in the empiric treatment group and none in the prophylactic treatment group. However, the difference did not reach statistical significance. Two deaths resulted from Gram-negative infection in the empiric treatment group compared with none in the prophylactic treatment group, but the difference did not reach statistical significance. There were no statistically significant differences in the duration of hospitalization.
According to the NNT calculations, the administration of immediate imipenem treatment would preempt one episode of febrile neutropenia, one of pneumonia or one of Gram-negative infection for every seven, nine and six patients, respectively (Table 3).
The infusion of PBPC did not affect the results of the other variables. No statistically significant differences were encountered with regard to the time-course effect or the sequence effect variables in the study.
DISCUSSION
This study was designed to evaluate the effectiveness of imipenem given immediately at the start of grade IV neutropenia caused by high-dose CEP chemotherapy administered to patients with solid tumors. The control arm was the conventional administration of the antibiotic at the onset of neutropenic fever. The study demonstrated that the immediate administration of imipenem resulted in a statistically significant reduction in the number of infections caused by Gram-negative bacteria, as well as in the number of episodes of febrile neutropenia experienced by the study population. Furthermore, statistically nonsignificant evidence was found in the reduction of the duration of the fever; in the number of episodes of pneumonia; in the number of episodes of septic shock; and in the number of deaths due to Gram-negative bacterial infections in the patients who received prophylactic imipenem treatment. Prophylactic treatment with imipenem did not reduce the number of days of hospitalization, nor did it influence the transfusion of platelets. The positive results observed in the most relevant clinical parameters suggest an interesting field of investigation in antimicrobial prophylaxis in these types of patients.
In current clinical practice, the standard approach to the prophylaxis and treatment of infections occurring as a result of intensive chemotherapy is the administration of oral antibiotics to reduce or eliminate potential gastrointestinal tract pathogens because the damage of the mucosa by the cytotoxic pharmacological agents favor the invasion by opportunistic agents at this site (6, 7). If fever still appears, then broad-spectrum i.v. antibiotics are administered.
Several different oral prophylactic regimens have been studied and they demonstrated a capacity to reduce, but not eliminate, the risk of infection. However, the benefits derived from these agents are frequently counterbalanced by their toxic side effects or because of the possibility of creating resistant bacteria and the possibility of additional fungal infection. The prophylactic regimens most commonly used to reduce the risk of infection have included cotrimoxazole and oral quinolones. With the former, the experts differ in their recommendations, however, in institutions with a high prevalence of fungal infections and in patients with high risk of Pneumocystis carinii, cotrimoxazole appears to be the drug of choice (8).
Prophylaxis with quinolones reduces infections by Gram-negative bacteria including Pseudomonas aeruginosa and, currently, the accumulated experience with its use has been sufficiently encouraging to make it almost the standard treatment for bone marrow receptor (23). Prophylaxis with quinolones does not reduce the number of infections by Gram-positive flora and this has contributed to the appearance of resistant strains of microorganisms (6, 24, 25). To resolve the problem of Gram-positive infections, some studies investigated combination prophylaxis, e.g., ciprofloxacin plus rifampicin or penicillin, and reductions in the incidence of fever from infection in oncology patients treated with intensive chemotherapy were reported (26-33).
Chemoprophylaxis therapy of quinolones added to a Gram-positive agent in patients subjected to high-dose chemotherapy is controversial. Some authors have reported that chemoprophylaxis with quinolones plus a Gram-positive anti-microbial agent reduced the incidence of Gram-positive infection. However, because an overall decrease in the incidence of febrile neutropenia, infections or mortality in the patient population was not observed, this treatment is not translated into a clear clinical benefit (30-33).
In our study, use of the prophylactic antibiotic treatment (immediate imipenem) led to a statistically significant reduction in the number of episodes of febrile neutropenia and of infections caused by Gram-negative bacteria; these are both parameters of considerable clinical importance. The reduction in the number of episodes of pneumonia, septic shock and deaths from infections due to Gram-negative bacteria was not statistically significant. This was probably due to the low incidence of these events and to the sample size of the study population. However, these could represent additional benefits of prophylactic treatment with imipenem. While no reduction in the incidence of infections by Gram-positive agents with immediate imipenem was observed, nor was there an increase.
It is important to note that despite the use of oral prophylactic ciprofloxacin in patients in the delayed imipenem control group treated according to standard practice, they presented neutropenic fever, thus indicating that this treatment is insufficient in preempting neutropenic fever in protracted aplasias.
With immediate imipenem, an absolute reduction of 13.7% in the incidence of febrile neutropenia was observed. With regard to the NNT calculations, this meant that for every seven patients receiving immediate imipenem, one case of febrile neutropenia was avoided. Similarly, one case of infection by Gram-negative bacteria was avoided for every six patients treated with prophylactic imipenem.
In patients receiving high-dose chemotherapy, the pathogens most frequently causing infection are Gram-positive aerobic cocci and Gram-negative aerobic bacilli (8), followed by fungi and viruses. Gram-negative bacteria infections continue to be the main cause of mortality in these patients, while septic shock and pneumonia are the infections which most frequently cause death in Gram-negative infections. In the present study, we observed a reduction in the number of episodes of pneumonia and septic shock in the patients who received imipenem from the onset of neutropenia (<500 cells/µl). Although this difference was not statistically significant it is noteworthy because, in the NNT calculation, it meant that one episode of pneumonia was avoided for every nine patients receiving prophylactic imipenem.
We can conclude that immediate/prophylactic use of imipenem in cancer patients subjected to high-dose chemotherapy results in an effective prevention of neutropenic fever and a reduction in infection by Gram-negative bacteria. These data introduce a novel approach in the management of febrile neutropenia in patients receiving intensive chemotherapy. The early use of a broad-spectrum antibiotic against Gram-negative infection from the first signs of neutropenia results in greater clinical benefit than its conventional use when the febrile neutropenia is established. Despite the administration of imipenem from the moment neutropenia appeared, 86% of the patients in our study developed fever during the neutropenia and, in four cases (7.8% of the cycles), this was documented as being due to Gram-negative bacterial infection. This could suggest that to be more effective, prophylactic imipenem might need to be administered even earlier than the appearance of the first signs of neutropenia. The ideal time for the empiric antibiotic administration of broad-spectrum antibiotics to patients receiving intensive chemotherapy needs to be investigated in further randomized studies.
Correspondence: Miguel Martín, Servicio de Oncología Médica, Hospital Universitario San Carlos, 28040 Madrid, Spain. Tel: +34-33303546. e-mail: [email protected]
REFERENCES
1. Hughes, W.T., Armstrong, D., Bodey, G.P. et al (on behalf of the Infectious Diseases Society of America). Guidelines for the use of antimicrobial agents in neutropenic patients with unexplained fever. J Infect Dis 1990; 161: 381-396.
2. Holland, H.K., Saral, R. Infectious diseases. In: Armitage, J.O., Antman, K.H. (Eds.). High-Dose Cancer Therapy. Williams & Wilkins, city? 1995; 508-526.
3. Talcott, J.A., Siegel, R.D., Finberg, R. et al. Risk assessment in cancer patients with fever and neutropenia: A prospective, two-center validation of a prediction rule. J Clin Oncol 1992; l0: 316-322.
4. Schimpff, S.C. Overview of empiric antibiotic therapy for the febrile neutropenic patient. Rev Infect Dis 1985; 7(Suppl. 4): S734-740.
5. Pizzo, P.A. Granulocytopenia and cancer therapy. Past problems, current solutions, future challenges. Cancer 1984; 54: 2649-2661.
6. Verhoef, J. Prevention of infections in the neutropenic patient. Clin Infect Dis 1993; 17: S359-367.
7. Hunter, A.E., Haynes, A.P., Russell, N.H. Bone marrow transplantation: Current situation, complications and prevention. J Antimicrob Chemother 1995; 36(Suppl. B): 119-133.
8. Hughes, W.T., Armstrong, D., Bodey, G.P. et al. Guidelines for the use of antimicrobial agents in neutropenic patients with unexplained fever (Infectious Diseases Society of America). Clin Infect Dis 1997; 25: 551-573.
9. Sanders, J.W., Powel, N.R., Moore, R.D. Ceftazidime monotherapy for empiric treatment of febrile neutropenic patients: A meta-analysis. J Infect Dis 1991; 164: 907-916.
10. Wade, J.C., Standiford, H.C., Drusano, G.L. et al. Potential of imipenem as single-agent empiric antibiotic therapy of febrile neutropenic patients with cancer. Am J Med 1985; 78: 62-72.
11. Deaney, N.B., Tate, H. A meta-analysis of clinical studies of imipenem-cilastatin for empirically treating febrile neutropenic patients. J Antimicrob Chemother 1996; 37: 975-986.
12. Biron, P., Fuhrmann, C., Cure, H. et al. Cefepime versus imipenem-cilastatin as empirical monotherapy in 400 febrile patients with short duration neutropenia. CEMIC (Study Group of Infectious Diseases in Cancer). J Antimicrob Chemother 1998; 42: 511-518.
13. Liang, R., Yung, R., Chiu, E. et al. Ceftazidime versus imipenem-cilastatin as initial monotherapy for febrile neutropenic patients. Antimicrob Agents Chemother 1990; 34: 1336-1341.
14. Freifeld, A.G., Walsh, T., Marshall, D. et al. Monotherapy for fever and neutropenia in cancer patients: A randomized comparison of ceftazidime versus imipenem. J Clin Oncol 1995; 13: 165-176.
15. Winston, D.J., Ho, W.G., Bruckner, D.A. et al. Beta-lactam antibiotic therapy in febrile granulocytopenic patients. A randomized trial comparing cefoperazone plus piperacillin, ceftazidime plus piperacillin, and imipenem alone. Ann Intern Med 1991; 115: 849-859.
16. Mortimer, J., Miller, S., Black, D. et al. Comparison of cefoperazone and mezlocillin with imipenem as empiric therapy in febrile neutropenic cancer patients. Am J Med 1988; 85: 17-20.
17. Norrby, S.R., Vandercam, B., Louie, T. et al. Imipenem/cilastatin versus amikacin plus piperacillin in the treatment of infections in neutropenic patients: A prospective, randomized multi-clinic study. Scand J Infect Dis Suppl 1987; 52: 65-78.
18. Leyland, M.J., Bayston, K.F., Cohen, J. et al. A comparative study of imipenem versus piperacillin plus gentamicin in the initial management of febrile neutropenic patients with haematological malignancies. J Antimicrob Chemother 1992; 30: 843-854.
19. Rolston, K.V., Berkey, P., Bodey, G.P. et al. A comparison of imipenem to ceftazidime with or without amikacin as empiric therapy in febrile neutropenic patients. Arch Intern Med 1992; 152: 283-291.
20. Cornelissen, J.J., de Graeff, A., Verdonck, L.F. et al. Imipenem versus gentamicin combined with either cefuroxime or cephalothin as initial therapy for febrile neutropenic patients. Antimicrob Agents Chemother 1992; 36: 801-807.
21. Haas, A., Anderson, L., Lad, T. The influence of aminoglycosides on the nephrotoxicity of cis-diamminedichloroplatinum in cancer patients. J Infect Dis 1983; 147: 363.
22. McQuay, H.J., Moore, R.A. Using numerical results from systematic reviews in clinical practice. Ann Intern Med 1997; 126: 712-720.
23. Cruciani, M., Rampazzo, R., Malena, M. et al. Prophylaxis with fluoroquinolones for bacterial infections in neutropenic patients: A meta-analysis. Clin Infect Dis 1996; 23: 795-805.
24. Cometta, A., Calandra, T., Bille, J. et al. Escherichia coli resistant to fluoroquinolones in patients with cancer and neutropenia. N Engl J Med 1994; 330: 1240-1241.
25. Trucksis, M., Hooper, D.C., Wolfson, J.S. Emerging resistance to fluoroquinolones in staphylococci: An alert. Ann Intern Med 1991; 114: 424-426.
26. Meisenberg, B., Gollard, R., Brehm, T. et al. Prophylactic antibiotics eliminate bacteremia and allow safe outpatient management following high-dose chemotherapy and autologous stem cell rescue. Support Care Cancer 1996; 4: 364-369.
27. Gilbert, C., Meisenberg, B., Vredenburgh, J. et al. Sequential prophylactic oral and empiric once-daily parenteral antibiotics for neutropenia and fever after high-dose chemotherapy and autologous bone marrow support. J Clin Oncol 1994; 12: 1005-1011.
28. International Antimicrobial Therapy Cooperative Group of the European Organization for Research and Treatment of Cancer. Reduction of fever and streptococcal bacteremia in granulocytopenic patients with cancer. A trial of oral penicillin V or placebo combined with pefloxacin. JAMA 1994; 272: 1183-1189.
29. Broun, E.R., Wheat, J.L., Kneebone, P.H. et al. Randomized trial of the addition of gram-positive prophylaxis to standard antimicrobial prophylaxis for patients undergoing autologous bone marrow transplantation. Antimicrob Agents Chemother 1994; 38: 576-579.
30. Bow, E.J., Mandell, L.A., Louie, T.J. et al. Quinolone-based antibacterial chemoprophylaxis in neutropenic patients: Effect of augmented gram-positive activity on infectious morbidity. National Cancer Institute of Canada Clinical Trials Group. Ann Intern Med 1996; 125: 183-190.
31. Murphy, M., Brown, A.E., Sepkowitz, K.A. et al. Fluoroquinolone prophylaxis for the prevention of bacterial infections in patients with cancer - is it justified? Clin Infect Dis 1997; 25: 346-348.
32. Momin, F., Chandrasekar, P.H. Antimicrobial prophylaxis in bone marrow transplantation. Ann Intern Med 1995; 123: 205-215.
33. Gómez-Martín, C., Sola, C., Hornedo, J. et al. Rifampicin does not improve the efficacy of quinolone antibacterial prophylaxis in neutropenic cancer patients: Results of a randomized clinical trial. J Clin Oncol 2000; 18: 2126-2134.
Table 1. Characteristics of patients in each intervention arm.