Originalia F. Menichetti, A. Del Favero, G. Bucaneve, V Minotti, L. Patoia, S. Pauluzzi
Ceftriaxone versus Aztreonam plus Cefazolin for Infections in Cancer Patients with Adequate Neutrophil Counts Summary: In a prospective randomized trial, 154 febrile episodes in cancer patients with adequate neutrophil counts (>1,000 cells/ram3) were treated with either ceftriaxone (72 episodes) or aztreonam plus cefazolin (82 episodes). Documented infections represented almost half of the febrile episodes. The overall response rates among the 144 evaluable episodes were similar for the two regimens: 76% (51/67) with ceftriaxone versus 82% (63/77) with aztreonam plus cefazolin (p = 0.41, not significant). Although not statistically significant, the response rate of the microbiologically documented infections was slightly better in patients treated with the double 13-1actam combination (85% vs. 65%, p = 0.16) and clinically documented infections showed a better response in the group of patients receiving monotherapy (87% vs. 59%, p = 0.12). No serious adverse effects were observed during this study and both regimens were well tolerated. Ceftriaxone or the combination of aztreonam plus cefazotin showed a similar efficacy as empirical therapy for infections in cancer patients with adequate neutrophil counts.
Zusammenfassung: Ceftriaxon im Vergleich zu Aztreonam plus Cefazolin zur Behandlung yon Infektionen bei Krebspatienten mit adiiquaten Neutrophilenzahlen. 154 Fieberschtibe bei Krebspatienten mit adfiquaten Neutrophilenzahlen (>1.000 Zellen/mm3) wurden in einer prospektiven randomisierten Studie entweder mit Ceftriaxon (72 Episoden) oder Aztreonam plus Cefazolin (82 Episoden) behandelt. Nahezu die Hfilfte der fieberhaften Episoden waren dokumentierte Infektionen. Die Gesamtansprechrate bei den 144 auswertbaren Episoden war fiir beide Therapien Nmlich: 76% ftir Ceftriaxon (51/67) und 82% f'tir Aztreonam plus Cefazolin (63/77) (p = 0,41; nicht signifikan0. Bei mikrobiologisch dokumentierten Infektionen war die Ansprechrate unter der Doppel-Betalaktam-Kombination etwas besser (85 % im Vergleich zu 65%; p = 0,16), doch war dieser Unterschied nicht signifikant. Bei den klinisch dokumentierten Infektionen zeigten die mit der Monotherapie behandelten Patienten bessere Ansprechraten (87% gegeniiber 59%; p = 0,12). W~ihrend der Studie wurden keine ernsthaften Nebenwirkungen beobachtet, beide Therapien wurden gut vertragen. In der empirischen Therapie yon Infektionen bei Krebspatienten mit adaquaten Neutrophilenzahlen erwiesen sich Ceftriaxon und die Kombination yon Aztreonam plus Cefazolin als vergleichbar wirksam.
44 / 166
Introduction Bacterial infections occur most frequently in cancer patients during periods of neutropenia; however, serious infections also occur in those with adequate neutrophil counts (>1,000 cells/mm3) [1, 2]. Gram-negative bacilli have been responsible for the majority of these infections, but in recent years gram-positive cocci have caused an increasing number of infections [3]. Owing to the fact that the infection-related morbidity and mortality in non-neutropenic cancer patients seems to be lower compared to those with neutropenia, a less aggressive, toxic and expensive therapeutic approach than that usually recommended in neutropenic patients [4] could be evaluated. A possible choice is monotherapy with a third generation cephalosporin [1, 5, 6], imipenem [7] or a double 13-1actamcombination [2, 8]. We therefore conducted a prospective randomized clinical trial to compare the efficacy and toterability of two empiric therapeutic regimens in febrile cancer patients without neutropenia: monotherapy with ceftnaxone versus the double 13-1actam combination aztreonam plus cefazolin. Ceftriaxone is a third generation cephalosporin that combines a broad antibacterial spectrum with a unique pharmacokinetic characteristic that allows once-daily administration and has been used as effective initial monotherapy in the treatment of severe infections [9]. Aztreonam, a monobactam antibiotic, showed good therapeutic activity in the treatment of gram-negative infections when used alone or in combination in neutropenic cancer patients [10]. Cefazolin was chosen for combination with the monobactam due to its good antibacterial activity directed against gram-positive cocci, including susceptible staphylococci, and its low capability of inducing [3-1actamase production.
Patients and Methods Patient selection: Consecutive cancer patients with adequate neutrophil counts (> 1,000 cells]ram~) admitted to the Departments of Internal Medicine, Oncology and Infectious Diseases of the Perugia General Hospital were eligible for this study if they had a proved or presumed infection, or an axillary temperature of 38°C or more persisting for more than two hours in the absence of obvious noninfective causes. Excluded from randomization were
Received: 27 October/Revision accepted: 4 April 1990 F. Menichetti, M. D., Prof. S. Pauluzzi, Institute of Infectious Diseases; Prof. A. Del Favero, G. Bucaneve, M. D., L. Patoia, M. D., 1st Institute of Internal Medicine, University of Perngia Medical School, Regional Hospital, via Brunarnonti, 1-06100 Perugia; V. Minotti, M. D., Oncology Division, Regional Hospital, via Brunamonti, 1-06100 Perugia, Italy.
Infection 18 (1990) No. 3 O MMV Medizin Verlag GmbH Mtinchen, Mtinchen 1990
F. Menichetti et al.: Empiric Antibiotics for Infections in Cancer Patients Table 1 : Patient characteristics.*
Table 2: Overall response to therapy: improved/total (%).
Febrile episodes(patients)
67 (64)
77 (75)
Mean age (years) Range
56 I2 to 79
60 19 to 83
45/27
58/24
25 23 19
35 15 27
7 (1-17)
7 (1-16)
Males/Females Diagnosis Solid tttmors Lymphomas Haematologicmalignancies Duration of therapy in days mean (range)
Microbiologically documented -with bacteraemia
5/9 (56)
11/14 (79)
withoutbacteraemia
6/8 (75)
11/12 (92)
Clinicallydocumented
13/15 (87)
10/17 (59)
Possible
8/11 (73)
7/7 (t00)
Doubtful
19/24 (79)
24/27 (89)
Totalno.
51/67 (76)
63/77 (82)
-
* Data refers to I44/154 episodesincludedin the efficacyanalysis. patients expected to develop neutropenia due to chemotherapy, those who, in the previous five days, had received systemic antibiotics and those who had a history of allergy to a ~-lactam antibiotic. For entry into the trial the following requirements were necessary for each patient: informed consent, a complete history and physical examination, blood cultures (at least three separate venipunctures), cultures of urine, and of any clinically suspicious source, chest X-ray, complete blood cell count including differential leukocyte count and determination of serum electrolytes, creatinine and 12-channel blood chemistry. Regimens: Randomization was achieved by using consecutive sealed envelopes assigning patients either to ceftriaxone or to aztreonam plus cefazolin regimen as determined by random permuted blocks. Ceftriaxone was given at a dosage of 30 mg/kg/once daily (maximum daily dose 2 g); aztreonam and cefazolin were both given at dosage of 50 mg/kg/day divided into three equal doses (maximum daily dose: 3 g). Each antibiotic was separately dissolved in 50 ml 0.9% saline and administered i. v. over 10-15 min. Microbiology: All bacterial isolates were identified by standard techniques and antimicrobial susceptibility tested by the KirbyBauer method. Blood cell count and differential leukocyte count, determination of serum electrolytes and creatinine, blood cultures and culture specimens from infection sites were obtained every three days during the study and within 72 h after discontinuation of study drugs. Blood cultures positive for coagulase-negative staphylococci were considered only if the organism was isolated from at least two blood culture specimens drawn with separate venipunctures within 24 h. Pneumonia was considered microbiologically documented only when a pathogen was cultured from blood and/or from bronchoscopically sampled specimens. Evaluation: Febrile episodes were classified as previously described [11]. Overall response to therapy was classified as: 1) improvement (disappearance of fever and other clinical and microbiological evidence of infection); 2) failure (the patient died during therapy or had no response to assigned regimen i. e., the microorganism or fever or both persisted and the patient's clinical condition did not improve, requiring a change in antibiotic therapy). One or more of the investigators examined al! study patients daily: patients who responded to therapy remained on the study drugs for at least four days after all signs of infection had dis-
appeared; a change in therapy was taken into consideration 72-96 h after randomization in patients evaluated as failures. The final outcome of the patient at the end of each febrile episode was registered as survival or death from infective or non-infective causes. Non-bacterial infection or protocol violation were excluded from the efficacy analysis. The Two-Tailed Fisher's Exact Test was used for comparing differences in proportion. Results
A total of 154 febrile episodes were randomized and 144 were evaluable for efficacy analysis. Reasons for exclusion from analysis were (ceftriaxone/aztreonam plus cefazolin): nonbacterial infections (3/3) and protocol violation (2/2). Characteristics of the evaluable patients were similar in the two treatment groups (Table 1 ). The majority of patients (42%) had solid tumors (mainly lung cancer, 26/60) or lymphomas (26%). The other haematologic malignancies were represented by acute leukemia [15], chronic leukemia [14], multiple myeloma [11], and other conditions [6]. None of the patients had recently received chemotherapy and none became neutropenic following the initiation of antibiotics. None of the patients received prophylactic antibiotics and only three patients (two in aztreonam plus cefazolin group and one in ceftriaxone group) had a central venous catheter in place at the onset of the febrile episode. The overall response to therapy was similar in both regimens. Improvement was achieved in 76% (51/67) of the episodes treated with monotherapy versus 82% (63]77) of those receiving combined treatment (Table 2). The response rate of microbiologically documented infections was slightly better with the aztreonam plus cefazolin regimen (85%, 22/26) than with monotherapy (65%, 11/17), but this difference was not statistically significant (p = 0.16). The double [3-1actam combination showed better results than monotherapy both in infections caused by gram-negative bacilli (92% versus 78%) and in those
Infection 18 (1990) No. 3 © MMV Medizin Verlag GmbH Mtinchen, Mthnchen1990
167 ]45
F. Menichetti et al.: Empiric Antibiotics Ibr Infections in Cancer Patients Table 3: Response to therapy in microbiologically documented infections (bacteraemic): improved/total,
Table 4: Site of infect{on (bacteraernic) and response to therapy: improved/total.
Pneumonia Pseudomonasaerugillosa
1/1
Othergram-negatives
3/3 (3/3)
2/2 (2/2)
Totalgram-negatives
7/9 (4/6)
11/12 (6/6)
Staphylococci
2/4 (1/2)
Other gram-positives
2/3
Totalgram-positives
4/7 (1/2)
Polymicrobialor doubleinfections Cn-andtotal
-
-
2/2 (1/I)
11/15 (1/I) -
t2/t9 (2/3)
Urinarytract
3/3
Other infection
7/8 (1/2)
6/8 (3/4)
Bacteraemiaof unknown origin
3/.5 (3/5)
5/6 (5/6)
4/6 (2/4)
Doubleinfections
0/1 (0/1)*
1/1 # -
5/6 (2/3)
Total
24/32 (5/9)
8/9 (1/I)
32/43 (tl/14)
9/12 (4/7)
* Bacteraemia by gram-negativeanaerobic bacilli plus UTI by Pseudo-
0/1 ~ (0/1)
2/2 # (1/1)
monas aeruginosa; # Cellufitisby Corynebacterium sp. plus UTI by Citrobacter sp.
11/17 (5/9)
22]26 (11/14)
*Bacteraemiaby gram-negativeanaerobes+ UTI by Pseudomonas aeruginosa; #Bacteraemia by Klebsiella sp. + Staphylococcus aureus; UTI by Citrobacter sp. + cellulifisby Corynebacterium sp.
triaxone, and one by enterococci resistant to both drugs. The other eight failures occurred in documented infections represented by six pneumonias (one of these caused by susceptible E. coli), by an upper urinary tract infection and an upper respiratory tract infection. Three failures occurred in doubtful infections. Among ceftriaxone treated patients, infective death occurred in two patients with bacteraemias (caused by an E. coli and a gram-negative anaerobe respectively), and in one patient with a pleural empyema caused by enterococci. Two pneumonias (one caused by E. coli) were responsible for the infective deaths in patients receiving aztreonam plus cefazolin. Among the 154 enrolled episodes no side effects requiring treatment interruption occurred. One case only of superinfection was documented, an oral candidiasis occurring in a ceftriaxone treated patient. (Tablos 1-4)
caused by gram-positive organisms (75% versus 57%). However, none of these differences was statistically significant (Table 3). Monotherapy with ceftriaxone showed better results in clinically documented infections (87% versus 59%) but again, this difference was not statistically significant (p = 0.12). No difference in response rate was found between the two regimens as regards the infection site (Tabte 4). Final evaluation of patients showed that the survival rate was 87% in both groups, being 58/67 in ceftriaxone versus Discussion 67/77 in the aztreonam plus cefazolin group. Thirty episodes failed to respond to empirical therapy (16 The empirical antibiotic combinations used in febrile in the ceflriaxone and 14 in the aztreonam plus cefazolin cancer patients usually include an aminoglycoside plus a group); bacteraemias [7] and pneumonias [10] accounted beta-lactam [4]. Since the aminoglycosides have been assofor the majority of failures among the documented infec- ciated with oto- and nephrotoxicity and, in addition, they are difficult to administer because the dosing schedules tions. The failures encountered among patients treated with cef- must be carefully monitored and adjusted according to their triaxone were four bacteraemias caused by two susceptible serum level, the evaluation of antibiotic regimens that do Escherichia coli, one susceptible coagulase-negative sta- not contain these drugs is important. Our study shows that phylococcus and one anaerobic gram-negative bacillus re- monotherapy with ceftriaxone or the use of the double betasistant to ceftriaxone; three pneumonias (one of these lactam combination aztreonam plus cefazolin have similar caused by susceptible Staphylococcus aureus) and a pleural overall efficacy in the empiric treatment of febrile episodes empyema by enterococci resistant to ceftriaxone. The other in cancer patients with adequate neutrophil counts. Our reeight failures occurred in non-documented infections (three sults are similar to those obtained employing other betalactams, such as single agent [1, 5-7] or in double combinapossible and five doubtful infections). In the group of patients treated with the double 13-1actam tion [2, 8], providing evidence that aztreonam plus cefacombination failures were represented by three bacteraemic zolin and ceftriaxone can be added to the list of drugs that infections, two caused by staphylococci susceptible to cef- can be safely used on these patients. 46 / 168
Infection 18 (1990) No. 3 © MMV MedizinVerlag GmbHMtinchen,Mfinchen1990
F. Menichetti et al.: Empiric Antibiotics for Infections in Cancer Patients
The response to therapy obtained with the two regimens (ceftriaxone versus aztreonam plus cefazotin) was good both at overall evaluation of improvement rate (76% versus 82%) and at evaluation of survival rate at the end of febrile episode (87% in both groups). The response rate in microbiologically documented infections was slightly better in patients receiving the double I~-lactam combination (85% versus 65 %, p = 0.16)both in gram-negative (92% versus 78%, p = 0.56) and in grampositive infections (75% versus 57%, p = 0.61). The lack of a statistically significant difference between the two treatment groups may be related to the weak power of the study. In fact with an expected improvement rate of 75%, in order to detect a clinically significant difference of 20% in the response rate of microbiologically documented infections between the two groups, the sample size of our study should have included 65 patients with microbiologically documented infection in each treatment arm (alpha error =
References 1. Roiston, K., Bolivar, R., Fainstein, V., Jones, P., Elting, L., Bodey, G.P.: Cefotaxime: single agent therapy for infections in cancer patients with adequate granulocyte counts. J. Antirnicrob. Chemother. 15 (1985) 91-96. 2. Rolston, K., Fainstein, V., Elting, L., Bodey, G. P.: Ceftizoxime plus ticarcillin:double beta-lactam therapy for infections in cancer patients. J. Antimicrob. Chemother. 19 (1987) 367-71.
3. Del Favero, A., Meniehetti, F., Bucaneve, G., Minotti, V., Pauluzzi, S.: Septicaemia due to gram-positive cocci in cancer patients. J. Antimicrob. Chemother. 21 (Suppl. C)(1988) 157-162. 4. Young, L. S.: Empirical antimicrobial therapy in the neutropenic host. N. Engl. J. Med. 315 (1986) 580.581.
5. Fainstein, V., Bolivar, R., Eiting, L., Valdivieso, M., Bodey, G. P.: Ceflizoxime in the treatment of infections in patients with cancer. J. Anfimicrob. Chemother. 10 (Suppl. C) (1982) 167-173.
0.05, beta error = 0.10). Drug resistance during therapy was not a problem in this study, but it remains a potentially dangerous complication of single or double beta-lactam therapy. No serious adverse effects occurred during this study and both regimens were very well tolerated. In conclusion, our study shows that documented infections represent almost half of the febrile episodes occurring in a population of non-neutropenic cancer patients and provides evidence of the good efficacy of ceftriaxone or the combination aztreonarn plus cefazolin as empirical therapy for infections in those patients.
Acknowledgements We thank Enio Rossi and Amedeo Moretti for their skilled technical assistance, and Prof. G. Patrizi, Squibb Italia, for the aztreonam supply.
7. Bodey, G. P., Alvarez, M. A., Jones, P. J., Rolston, K. V. I., Steelhammer, L., Fainstein, V.: Imipenem-cilastatinas initial therapy for febrile cancer patients. Anfimicrob. Agents Chemother. 30 (1986) 211-14.
8. D'Olimpio, J. T., Miller, M., Sheridan, C. A., Carlisle, P., Wollner, D. L, McKitrick, J. C., Wiernik, P.: A comparative trial of moxalactam plus ticarcillin and clavutanic acid or piperacillin as empiric antibiotic therapy for febrile cancer patients. J. Clin. Pharmacol. 27 (1987) 673-677. 9. Baumgartner, J. D., Glanser, M. P., Single daily dose treatment of severe refractory infections with ceftriaxone. Cost savings and possible parenteral outpatient treatment. Arch. Intern. Med. 143 (1983) 1868-1873.
10. Jones, P. G., Rolston, K. V. I., Fainstein, V., Elting, L., Waiters, R. S., Bodey, G.P.: Aztreonam therapy in neutropenic patients with cancer. Am. J. IVied.81 (1986) 243-248.
6. Piccart, M., Klastersky, J., Meunier, F., Lagast, H., Van Laethem,
11. Menichetti, F., Del Favero, A., Guerciolini, R., Tonato, M., Aversa, F., Roila, R., FrongiUo, R. F., Martelli, M. F., Davis, S., Pauluzzi,
Y., Weerts, D.: Single-drug versus combination empirical therapy for gram- negative bacillary infections in febrile cancer patients with and without granutocytopenia. Antimicrob. Agents Chemother. 26 (1984) 870-875.
S.: Empiric antimicrobial therapy in febrile granulocytopenic patients. Randomized prospectivecomparison ofamikacin plus piperacillinwith or without parenteral ~ethoprim/sulphamethoxazole. Infection 14 (1986) 261-267.
Infection 18 (1990) No. 3 © MMV Medizin Verlag GmbH Mttnchen, Mttnchen 1990
169 ]47
Ceftriaxone versus Aztreonam plus Cefazolin for Infections in Cancer Patients with Adequate Neutrophil Counts Summary: In a prospective randomized trial, 154 febrile episodes in cancer patients with adequate neutrophil counts (>1,000 cells/ram3) were treated with either ceftriaxone (72 episodes) or aztreonam plus cefazolin (82 episodes). Documented infections represented almost half of the febrile episodes. The overall response rates among the 144 evaluable episodes were similar for the two regimens: 76% (51/67) with ceftriaxone versus 82% (63/77) with aztreonam plus cefazolin (p = 0.41, not significant). Although not statistically significant, the response rate of the microbiologically documented infections was slightly better in patients treated with the double 13-1actam combination (85% vs. 65%, p = 0.16) and clinically documented infections showed a better response in the group of patients receiving monotherapy (87% vs. 59%, p = 0.12). No serious adverse effects were observed during this study and both regimens were well tolerated. Ceftriaxone or the combination of aztreonam plus cefazotin showed a similar efficacy as empirical therapy for infections in cancer patients with adequate neutrophil counts.
Zusammenfassung: Ceftriaxon im Vergleich zu Aztreonam plus Cefazolin zur Behandlung yon Infektionen bei Krebspatienten mit adiiquaten Neutrophilenzahlen. 154 Fieberschtibe bei Krebspatienten mit adfiquaten Neutrophilenzahlen (>1.000 Zellen/mm3) wurden in einer prospektiven randomisierten Studie entweder mit Ceftriaxon (72 Episoden) oder Aztreonam plus Cefazolin (82 Episoden) behandelt. Nahezu die Hfilfte der fieberhaften Episoden waren dokumentierte Infektionen. Die Gesamtansprechrate bei den 144 auswertbaren Episoden war fiir beide Therapien Nmlich: 76% ftir Ceftriaxon (51/67) und 82% f'tir Aztreonam plus Cefazolin (63/77) (p = 0,41; nicht signifikan0. Bei mikrobiologisch dokumentierten Infektionen war die Ansprechrate unter der Doppel-Betalaktam-Kombination etwas besser (85 % im Vergleich zu 65%; p = 0,16), doch war dieser Unterschied nicht signifikant. Bei den klinisch dokumentierten Infektionen zeigten die mit der Monotherapie behandelten Patienten bessere Ansprechraten (87% gegeniiber 59%; p = 0,12). W~ihrend der Studie wurden keine ernsthaften Nebenwirkungen beobachtet, beide Therapien wurden gut vertragen. In der empirischen Therapie yon Infektionen bei Krebspatienten mit adaquaten Neutrophilenzahlen erwiesen sich Ceftriaxon und die Kombination yon Aztreonam plus Cefazolin als vergleichbar wirksam.
44 / 166
Introduction Bacterial infections occur most frequently in cancer patients during periods of neutropenia; however, serious infections also occur in those with adequate neutrophil counts (>1,000 cells/mm3) [1, 2]. Gram-negative bacilli have been responsible for the majority of these infections, but in recent years gram-positive cocci have caused an increasing number of infections [3]. Owing to the fact that the infection-related morbidity and mortality in non-neutropenic cancer patients seems to be lower compared to those with neutropenia, a less aggressive, toxic and expensive therapeutic approach than that usually recommended in neutropenic patients [4] could be evaluated. A possible choice is monotherapy with a third generation cephalosporin [1, 5, 6], imipenem [7] or a double 13-1actamcombination [2, 8]. We therefore conducted a prospective randomized clinical trial to compare the efficacy and toterability of two empiric therapeutic regimens in febrile cancer patients without neutropenia: monotherapy with ceftnaxone versus the double 13-1actam combination aztreonam plus cefazolin. Ceftriaxone is a third generation cephalosporin that combines a broad antibacterial spectrum with a unique pharmacokinetic characteristic that allows once-daily administration and has been used as effective initial monotherapy in the treatment of severe infections [9]. Aztreonam, a monobactam antibiotic, showed good therapeutic activity in the treatment of gram-negative infections when used alone or in combination in neutropenic cancer patients [10]. Cefazolin was chosen for combination with the monobactam due to its good antibacterial activity directed against gram-positive cocci, including susceptible staphylococci, and its low capability of inducing [3-1actamase production.
Patients and Methods Patient selection: Consecutive cancer patients with adequate neutrophil counts (> 1,000 cells]ram~) admitted to the Departments of Internal Medicine, Oncology and Infectious Diseases of the Perugia General Hospital were eligible for this study if they had a proved or presumed infection, or an axillary temperature of 38°C or more persisting for more than two hours in the absence of obvious noninfective causes. Excluded from randomization were
Received: 27 October/Revision accepted: 4 April 1990 F. Menichetti, M. D., Prof. S. Pauluzzi, Institute of Infectious Diseases; Prof. A. Del Favero, G. Bucaneve, M. D., L. Patoia, M. D., 1st Institute of Internal Medicine, University of Perngia Medical School, Regional Hospital, via Brunarnonti, 1-06100 Perugia; V. Minotti, M. D., Oncology Division, Regional Hospital, via Brunamonti, 1-06100 Perugia, Italy.
Infection 18 (1990) No. 3 O MMV Medizin Verlag GmbH Mtinchen, Mtinchen 1990
F. Menichetti et al.: Empiric Antibiotics for Infections in Cancer Patients Table 1 : Patient characteristics.*
Table 2: Overall response to therapy: improved/total (%).
Febrile episodes(patients)
67 (64)
77 (75)
Mean age (years) Range
56 I2 to 79
60 19 to 83
45/27
58/24
25 23 19
35 15 27
7 (1-17)
7 (1-16)
Males/Females Diagnosis Solid tttmors Lymphomas Haematologicmalignancies Duration of therapy in days mean (range)
Microbiologically documented -with bacteraemia
5/9 (56)
11/14 (79)
withoutbacteraemia
6/8 (75)
11/12 (92)
Clinicallydocumented
13/15 (87)
10/17 (59)
Possible
8/11 (73)
7/7 (t00)
Doubtful
19/24 (79)
24/27 (89)
Totalno.
51/67 (76)
63/77 (82)
-
* Data refers to I44/154 episodesincludedin the efficacyanalysis. patients expected to develop neutropenia due to chemotherapy, those who, in the previous five days, had received systemic antibiotics and those who had a history of allergy to a ~-lactam antibiotic. For entry into the trial the following requirements were necessary for each patient: informed consent, a complete history and physical examination, blood cultures (at least three separate venipunctures), cultures of urine, and of any clinically suspicious source, chest X-ray, complete blood cell count including differential leukocyte count and determination of serum electrolytes, creatinine and 12-channel blood chemistry. Regimens: Randomization was achieved by using consecutive sealed envelopes assigning patients either to ceftriaxone or to aztreonam plus cefazolin regimen as determined by random permuted blocks. Ceftriaxone was given at a dosage of 30 mg/kg/once daily (maximum daily dose 2 g); aztreonam and cefazolin were both given at dosage of 50 mg/kg/day divided into three equal doses (maximum daily dose: 3 g). Each antibiotic was separately dissolved in 50 ml 0.9% saline and administered i. v. over 10-15 min. Microbiology: All bacterial isolates were identified by standard techniques and antimicrobial susceptibility tested by the KirbyBauer method. Blood cell count and differential leukocyte count, determination of serum electrolytes and creatinine, blood cultures and culture specimens from infection sites were obtained every three days during the study and within 72 h after discontinuation of study drugs. Blood cultures positive for coagulase-negative staphylococci were considered only if the organism was isolated from at least two blood culture specimens drawn with separate venipunctures within 24 h. Pneumonia was considered microbiologically documented only when a pathogen was cultured from blood and/or from bronchoscopically sampled specimens. Evaluation: Febrile episodes were classified as previously described [11]. Overall response to therapy was classified as: 1) improvement (disappearance of fever and other clinical and microbiological evidence of infection); 2) failure (the patient died during therapy or had no response to assigned regimen i. e., the microorganism or fever or both persisted and the patient's clinical condition did not improve, requiring a change in antibiotic therapy). One or more of the investigators examined al! study patients daily: patients who responded to therapy remained on the study drugs for at least four days after all signs of infection had dis-
appeared; a change in therapy was taken into consideration 72-96 h after randomization in patients evaluated as failures. The final outcome of the patient at the end of each febrile episode was registered as survival or death from infective or non-infective causes. Non-bacterial infection or protocol violation were excluded from the efficacy analysis. The Two-Tailed Fisher's Exact Test was used for comparing differences in proportion. Results
A total of 154 febrile episodes were randomized and 144 were evaluable for efficacy analysis. Reasons for exclusion from analysis were (ceftriaxone/aztreonam plus cefazolin): nonbacterial infections (3/3) and protocol violation (2/2). Characteristics of the evaluable patients were similar in the two treatment groups (Table 1 ). The majority of patients (42%) had solid tumors (mainly lung cancer, 26/60) or lymphomas (26%). The other haematologic malignancies were represented by acute leukemia [15], chronic leukemia [14], multiple myeloma [11], and other conditions [6]. None of the patients had recently received chemotherapy and none became neutropenic following the initiation of antibiotics. None of the patients received prophylactic antibiotics and only three patients (two in aztreonam plus cefazolin group and one in ceftriaxone group) had a central venous catheter in place at the onset of the febrile episode. The overall response to therapy was similar in both regimens. Improvement was achieved in 76% (51/67) of the episodes treated with monotherapy versus 82% (63]77) of those receiving combined treatment (Table 2). The response rate of microbiologically documented infections was slightly better with the aztreonam plus cefazolin regimen (85%, 22/26) than with monotherapy (65%, 11/17), but this difference was not statistically significant (p = 0.16). The double [3-1actam combination showed better results than monotherapy both in infections caused by gram-negative bacilli (92% versus 78%) and in those
Infection 18 (1990) No. 3 © MMV Medizin Verlag GmbH Mtinchen, Mthnchen1990
167 ]45
F. Menichetti et al.: Empiric Antibiotics Ibr Infections in Cancer Patients Table 3: Response to therapy in microbiologically documented infections (bacteraemic): improved/total,
Table 4: Site of infect{on (bacteraernic) and response to therapy: improved/total.
Pneumonia Pseudomonasaerugillosa
1/1
Othergram-negatives
3/3 (3/3)
2/2 (2/2)
Totalgram-negatives
7/9 (4/6)
11/12 (6/6)
Staphylococci
2/4 (1/2)
Other gram-positives
2/3
Totalgram-positives
4/7 (1/2)
Polymicrobialor doubleinfections Cn-andtotal
-
-
2/2 (1/I)
11/15 (1/I) -
t2/t9 (2/3)
Urinarytract
3/3
Other infection
7/8 (1/2)
6/8 (3/4)
Bacteraemiaof unknown origin
3/.5 (3/5)
5/6 (5/6)
4/6 (2/4)
Doubleinfections
0/1 (0/1)*
1/1 # -
5/6 (2/3)
Total
24/32 (5/9)
8/9 (1/I)
32/43 (tl/14)
9/12 (4/7)
* Bacteraemia by gram-negativeanaerobic bacilli plus UTI by Pseudo-
0/1 ~ (0/1)
2/2 # (1/1)
monas aeruginosa; # Cellufitisby Corynebacterium sp. plus UTI by Citrobacter sp.
11/17 (5/9)
22]26 (11/14)
*Bacteraemiaby gram-negativeanaerobes+ UTI by Pseudomonas aeruginosa; #Bacteraemia by Klebsiella sp. + Staphylococcus aureus; UTI by Citrobacter sp. + cellulifisby Corynebacterium sp.
triaxone, and one by enterococci resistant to both drugs. The other eight failures occurred in documented infections represented by six pneumonias (one of these caused by susceptible E. coli), by an upper urinary tract infection and an upper respiratory tract infection. Three failures occurred in doubtful infections. Among ceftriaxone treated patients, infective death occurred in two patients with bacteraemias (caused by an E. coli and a gram-negative anaerobe respectively), and in one patient with a pleural empyema caused by enterococci. Two pneumonias (one caused by E. coli) were responsible for the infective deaths in patients receiving aztreonam plus cefazolin. Among the 154 enrolled episodes no side effects requiring treatment interruption occurred. One case only of superinfection was documented, an oral candidiasis occurring in a ceftriaxone treated patient. (Tablos 1-4)
caused by gram-positive organisms (75% versus 57%). However, none of these differences was statistically significant (Table 3). Monotherapy with ceftriaxone showed better results in clinically documented infections (87% versus 59%) but again, this difference was not statistically significant (p = 0.12). No difference in response rate was found between the two regimens as regards the infection site (Tabte 4). Final evaluation of patients showed that the survival rate was 87% in both groups, being 58/67 in ceftriaxone versus Discussion 67/77 in the aztreonam plus cefazolin group. Thirty episodes failed to respond to empirical therapy (16 The empirical antibiotic combinations used in febrile in the ceflriaxone and 14 in the aztreonam plus cefazolin cancer patients usually include an aminoglycoside plus a group); bacteraemias [7] and pneumonias [10] accounted beta-lactam [4]. Since the aminoglycosides have been assofor the majority of failures among the documented infec- ciated with oto- and nephrotoxicity and, in addition, they are difficult to administer because the dosing schedules tions. The failures encountered among patients treated with cef- must be carefully monitored and adjusted according to their triaxone were four bacteraemias caused by two susceptible serum level, the evaluation of antibiotic regimens that do Escherichia coli, one susceptible coagulase-negative sta- not contain these drugs is important. Our study shows that phylococcus and one anaerobic gram-negative bacillus re- monotherapy with ceftriaxone or the use of the double betasistant to ceftriaxone; three pneumonias (one of these lactam combination aztreonam plus cefazolin have similar caused by susceptible Staphylococcus aureus) and a pleural overall efficacy in the empiric treatment of febrile episodes empyema by enterococci resistant to ceftriaxone. The other in cancer patients with adequate neutrophil counts. Our reeight failures occurred in non-documented infections (three sults are similar to those obtained employing other betalactams, such as single agent [1, 5-7] or in double combinapossible and five doubtful infections). In the group of patients treated with the double 13-1actam tion [2, 8], providing evidence that aztreonam plus cefacombination failures were represented by three bacteraemic zolin and ceftriaxone can be added to the list of drugs that infections, two caused by staphylococci susceptible to cef- can be safely used on these patients. 46 / 168
Infection 18 (1990) No. 3 © MMV MedizinVerlag GmbHMtinchen,Mfinchen1990
F. Menichetti et al.: Empiric Antibiotics for Infections in Cancer Patients
The response to therapy obtained with the two regimens (ceftriaxone versus aztreonam plus cefazotin) was good both at overall evaluation of improvement rate (76% versus 82%) and at evaluation of survival rate at the end of febrile episode (87% in both groups). The response rate in microbiologically documented infections was slightly better in patients receiving the double I~-lactam combination (85% versus 65 %, p = 0.16)both in gram-negative (92% versus 78%, p = 0.56) and in grampositive infections (75% versus 57%, p = 0.61). The lack of a statistically significant difference between the two treatment groups may be related to the weak power of the study. In fact with an expected improvement rate of 75%, in order to detect a clinically significant difference of 20% in the response rate of microbiologically documented infections between the two groups, the sample size of our study should have included 65 patients with microbiologically documented infection in each treatment arm (alpha error =
References 1. Roiston, K., Bolivar, R., Fainstein, V., Jones, P., Elting, L., Bodey, G.P.: Cefotaxime: single agent therapy for infections in cancer patients with adequate granulocyte counts. J. Antirnicrob. Chemother. 15 (1985) 91-96. 2. Rolston, K., Fainstein, V., Elting, L., Bodey, G. P.: Ceftizoxime plus ticarcillin:double beta-lactam therapy for infections in cancer patients. J. Antimicrob. Chemother. 19 (1987) 367-71.
3. Del Favero, A., Meniehetti, F., Bucaneve, G., Minotti, V., Pauluzzi, S.: Septicaemia due to gram-positive cocci in cancer patients. J. Antimicrob. Chemother. 21 (Suppl. C)(1988) 157-162. 4. Young, L. S.: Empirical antimicrobial therapy in the neutropenic host. N. Engl. J. Med. 315 (1986) 580.581.
5. Fainstein, V., Bolivar, R., Eiting, L., Valdivieso, M., Bodey, G. P.: Ceflizoxime in the treatment of infections in patients with cancer. J. Anfimicrob. Chemother. 10 (Suppl. C) (1982) 167-173.
0.05, beta error = 0.10). Drug resistance during therapy was not a problem in this study, but it remains a potentially dangerous complication of single or double beta-lactam therapy. No serious adverse effects occurred during this study and both regimens were very well tolerated. In conclusion, our study shows that documented infections represent almost half of the febrile episodes occurring in a population of non-neutropenic cancer patients and provides evidence of the good efficacy of ceftriaxone or the combination aztreonarn plus cefazolin as empirical therapy for infections in those patients.
Acknowledgements We thank Enio Rossi and Amedeo Moretti for their skilled technical assistance, and Prof. G. Patrizi, Squibb Italia, for the aztreonam supply.
7. Bodey, G. P., Alvarez, M. A., Jones, P. J., Rolston, K. V. I., Steelhammer, L., Fainstein, V.: Imipenem-cilastatinas initial therapy for febrile cancer patients. Anfimicrob. Agents Chemother. 30 (1986) 211-14.
8. D'Olimpio, J. T., Miller, M., Sheridan, C. A., Carlisle, P., Wollner, D. L, McKitrick, J. C., Wiernik, P.: A comparative trial of moxalactam plus ticarcillin and clavutanic acid or piperacillin as empiric antibiotic therapy for febrile cancer patients. J. Clin. Pharmacol. 27 (1987) 673-677. 9. Baumgartner, J. D., Glanser, M. P., Single daily dose treatment of severe refractory infections with ceftriaxone. Cost savings and possible parenteral outpatient treatment. Arch. Intern. Med. 143 (1983) 1868-1873.
10. Jones, P. G., Rolston, K. V. I., Fainstein, V., Elting, L., Waiters, R. S., Bodey, G.P.: Aztreonam therapy in neutropenic patients with cancer. Am. J. IVied.81 (1986) 243-248.
6. Piccart, M., Klastersky, J., Meunier, F., Lagast, H., Van Laethem,
11. Menichetti, F., Del Favero, A., Guerciolini, R., Tonato, M., Aversa, F., Roila, R., FrongiUo, R. F., Martelli, M. F., Davis, S., Pauluzzi,
Y., Weerts, D.: Single-drug versus combination empirical therapy for gram- negative bacillary infections in febrile cancer patients with and without granutocytopenia. Antimicrob. Agents Chemother. 26 (1984) 870-875.
S.: Empiric antimicrobial therapy in febrile granulocytopenic patients. Randomized prospectivecomparison ofamikacin plus piperacillinwith or without parenteral ~ethoprim/sulphamethoxazole. Infection 14 (1986) 261-267.
Infection 18 (1990) No. 3 © MMV Medizin Verlag GmbH Mttnchen, Mttnchen 1990
169 ]47
