British Journal of Haematology. 1990, 76, Suppl. 2, 49-53
Ceftazidime plus teicoplanin versus ceftazidime plus amikacin as empiric therapy for fever in cancer patients with granulocytopenia F . MEUNIER.* P. V A N DER AUWERA,M. AOUNA N D D. BRONService de Mtdecine Interne et Laboratoire d’lnvestigation Clinique Henri Tagnon, Institut Iules Bordet, Centre des Tumeurs de I’Universitt Libre de Bruxelles, Belgique
Summary. In a prospective randomized study, 100 episodes of fever ( >38°C) and granulocytopenia ( < 100O/p1) in cancer patients were empirically treated with ceftazidime(2 g every 8 h) plus teicoplanin (400 mg every 8 h on day 1: 400 mg every day thereafter) or ceftazidime (2 g every 8 h) plus amikacin (500 mg every 8 h). Bacteraemia. clinically documented infection and possible infection were documented in seven. 11 and 19 patients treated with ceftazidimeplus teicoplanin and in 11, four and 17 patients treated with ceftazidime plus amikacin. Overall, the response rate was similar in the two groups of patients as
was the need for treatment modifications and the rate of death. For documented Gram-positive bacteraemia, the response rate was 2 / 5 patients treated with ceftazidimeplus teicoplanin and 2/7 with ceftazidime plus amikacin: for documented Gram-negative bacteraemia. the response rate was 1/2 and 3/4 patients respectively. N o breakthrough bacteraemia was observed. Tolerance was excellent. although renal toxicity (elevation of serum creatinine) was observed in three patients treated with ceftazidime plus teicoplanin and in none allocated to ceftazidime plus amikacin.
Infections are a major cause of morbidity and mortality in cancer patients. especially in granulocytopenic hosts (EORTC, 1987; Klastersky et al, 1988: Meunier, 1990). Empiric treatment of fever in those patients is mandatory but the optimal regimen is still controversial and usually consists of the administration of a broad-spectrum beta-lactam antibiotic sometimes combined with an aminoglycoside (EORTC. 1987: Klastersk: et al. 1988: Meunier, 1990: Young, 1988). The use of combination therapy is particularly advocated for bacter: kmic infections, especially when due to Gram-negative baci Ci and when severe granulocytopenia persists (EORTC, 1’987: Young, 1988). In some circumstances, monotherapy has been found effective (Pizzo et al. 1986) with no increase in mortality compared with a combination therapy. However, using this therapeutic approach, the number of treatment modifications is uniformly high, vancomycin, amikacin and amphotericin B being the most frequent antimicrobial agents added during treatment. The increasing number of Gram-positive micro-organisms causing bacteraemia in granulocytopenic cancer patients
(Klastersky et al. 1988: Pizzo et al. 1978: Viscoli et al. 1988: Wade et al, 1982: Winston et al. 1983) has prompted a renewed interest in the inclusion of specific anti-Grampositive antimicrobial coverage as empiric therapy together with agents effective against Gram-negative bacilli (Jadeja et al. 1984: Karp et al, 1986; Kramer et al, 1986: Shenep et al, 1988). Whether vancomycin should be added to empiric therapy is still a matter of discussion (Hughes et ai, 1990: Meunier. 1990: Viscoli et al, 1988): in particular, the potential risk of cumulative toxicity in patients treated with vancomycin. aminoglycosides. amphotericin B and cyclosporin A is a matter of concern (Cimino et al, 1987: Eisenberg et al. 1987). Teicoplanin is a glycopeptide antibiotic effective against most Gram-positive cocci. It has a similar spectrum of activity to vancomycin, possibly with less side-effects (Bibler et al, 1987: Cimino et al. 1987: Glupczynski et al. 1986; Verbist et al, 1984).It has been used successfully in a limited number of febrile granulocytopenic patients when administered in combination with ceftazidime and amikacin (Del Favero et al, 1987). Therefore, we were interested in evaluating the potential benefit of empiric therapy of fever in granulocytopenic patients with ceftazidime plus teicoplanin versus ceftazidimeplus amikacin, a regimen generally considered as standard empiric therapy in such circumstances (EORTC. 1987).
* Senior Research Associate, Fund for Medical Scientific Research. Brussels. Belgium.
49
50
F. Meunier et a1
PATIENTS AND METHODS Granulocytopenic cancer patients (PMN less than 1 000/pl) with fever (temperature > 38°C) hospitalized at the Institut J. Bordet, Brussels, Belgium, were eligible for this randomized and prospective study. Patients were allocated to either ceftazidime plus amikacin or ceftazidime plus teicoplanin using sealed envelopes. Ceftazidime was administered in three intravenous doses of 2 g daily and amikacin was administered immediately afterwards (3 times 500 mg daily). Ceftazidime was reconstituted in 50 ml of 5% dextrose in distilled water and infused over 5 min. Amikacin was similarly reconstituted but the infusion time was 30 min. Teicoplanin was also reconstituted in 50 ml of 5% dextrose in distilled water and given as 1 5 min infusion following the ceftazidime infusion. The loading dose of teicoplanin was 4 0 0 mg t.i.d. on day 1. and 400 mg teicoplanin was administered once daily thereafter. A complete history and physical examination were obtained for all eligible patients. Sets of blood cultures were drawn prior to the initiation of antibiotics as well as urine cultures and other relevant samples according to clinical presentation. Chest X-rays were also performed prior to empiric therapy and laboratory investigations including haematology. coagulation and blood chemistry for renal and hepatic functions. Initial serum creatinine > 2 mg/dl was an exclusion criterion.
Nephrotoxicity was assessed by the measure of serum creatinine: a rise of 0.5 mg/dl over baseline was considered as nephrotoxicity. Hepatotoxicity was defined as a rise of transaminases (1.5 times) above baseline values if noted in the absence of other apparent causes. Patients undergoing allogeneic bone marrow transplantation were allocated to ceftazidime plus teicoplanin (two patients). Evaluation of response to empiric therapy was performed according to the criteria used by the EOKTC International Antimicrobial Therapy Co-operative Group (EORTC, 1987: Klastersky et a], 1988). Any addition of an antibiotic was considered as failure of empiric therapy. Patients were considered as unevaluable if an episode of fever was classified as doubtful infection (such as fever due to a drug or blood product transfusion), if the patient had a viral or a fungal infection or a major violation of their protocol occurred. RESULTS Overall 100 patients were eligible. The characteristics of the patients in both groups were very similar for age, sex ratio, underlying disease and previous chemoprophylaxis with antibacterial agents (Table I). The number of unevaluable episodes (doubtful infection.
Table I. Characteristics of the patients
No. of eligible patients Sex: male/female Age in years: median (range) Underlying diseases Leukaemia Hodgkin's or lymphoma Solid tumour Multiple myeloma Allogeneic BMT* Autologous BMT* Types of infection Microbiologically documented infection With bacteraemia Without bacteraemia ClinicaIly documented infection Possible infection Unevaluable patients Doubtful infection Fungal infection Mixed infection (viral and fungal) Viral infection Protocol violation Previous chemoprophylaxis No. of episodes Median duration in days (range)
* BMT= bone marrow transplantation.
Ceftazidime plus teicoplanin
Ceftazidime plus amikacin
50 30120 47 (20-80)
50 27/23 54.5 ( 1 5-75)
25 12 12
26 7 17 0 0 2
1
2 7 9
7 2 11
19
15 I1 4 4 17
3 3 1 2 2
18 8 (1-23)
17 9 (1-14)
Ceftazidime with Teicoplanin or Amikacin for Granulocytopenia
51
Table 11. Results and outcome
Ceftazidime plus teicoplanin
Ceftazidime plus arnikacin
Duration of therapy as allocated at randomization (range)
39 25 11 14 (1-90) 8 (1-21)
36 19 12 14 (2-86) 7 (2-1 3 )
Success/total (%) Overall Bacteraemia Clinically documented infection Possible infection
26/39 (66.6%) 317 (42.8%) 611 1 (54.5%) 16/19 (84.2%)
24/36 167%) 5/11 (45%) 414 (100%) 12/17 (70.5%)
Modilication of antibacterial therapy Death within a month
10 8
No. of evaluable patients No. of patients with PMN< 1OO/jd at randomization No. of patients who recovered PMN > lOOO/pl during therapy Overall duration of granulocytopenia in days (range)
fungal or viral infection and protocol violation) was coniparable. 1 1 in the ceftazidime plus teicoplanin group and 1 4 in the ceftaxidime plus amikacin group. Two episodes in the ceftazidime plus teicoplanin group were unevaluable because of a protocol violation: for one patient, amikacin was added on day 2 after initiation of empiric therapy although the patient was already afebrile. This patient did not develop nt side-effects and survived. Another patient was treated only for a few days according to the protocol and requested to be discharged tc pursue oral treatment at home. Scven episodes in the ceftaz lime plus amikacin group were unevaluable owing to proto ,ol violation: therapy was modified and teicoplanin added iii two patients on day 3 without good reasons. Another patient. afebrile after 2 d. received teicoplanin on day h without consent of the investigator and without good reasons. That patient died on day l h from disseminated pneumonia. Autopsy was not performed. One patient was febrile for 1 d and treatment stopped after 3 d: this patient survived. One patient was afebrile after 3 d but received teicoplanin from day 7 (without appropriate reasons).One patient with perineal abscesses received metronidazole from day I . He survived and was also considered as unevaluable. Finally. one patient was treated with ceftazidime plus amikacin for 3 d, was afebrile but had an uncontrolled leukwnia and was sent home. He died on day I 1. Table I1 shows the duration oftherapy as well as the degree of granulocytopenia and the outcome of patients in both groups. There were a similar number of patients recovering adequate granulocyte counts (I'MN > 1000/pl) during therapy. The overall duration of granulocytopenia was similar (14 d ) in both groups as was the duration of therapy as allocated at randomization: 8 and 7 d respectively in both groups. The improvement rate was similar for the two groups even when the episodes of persisting profound granulocytopenia were considered. Failure was never due to persisting bacteraemia. but to persisting fever which was treated empirically by modifying the antimicrobial regimen. The overall success rate was 66% for ceftazidime plus 1
(25.6%)
9
(25%)
7
teicoplanin as well as for ceftazidime plus amikacin. Success was observed in 3/7 patients with bacteraemia treated with ceftazidimeplus teicoplanin and in 5/11 patients treated with ceftazidime plus amikacin. For patients without microbiologically documented infection, success was observed in 24/ 3 0 patients treated with ceftazidinie plus teicoplanin and in 1 6 / 2 1 patients treated with ceftazidinie plus amikacin. The need for modifications of antibacterial therapy was similar in both groups and was observed in 25% of the patients. The number ofpatients that died within a month after empiric therapy for fever was also similar in both groups. Three patients treated with ceftazidime plus teicoplanin died as a result of their primary infection. One patient died in septic shock (on day 1 ) caused by bacteraemia due to Eschericliia coli (susceptible to ceftazidime). Another patient died (on day 6 ) from bacteraeinia caused by Stuphylococcits aimus and Streptococcusjiwnlis and the third patient died (on day 14) with uncontrolled bacteraemia caused by Staphylococcxs c~pidarriiidis(related to a 'Port a cath'). Among the patients who died while receiving ceftazidime plus amikacin, three also died from their primary infection: one patient died on day 2 from bacteraemia caused by S. epidcvwidis associated with a severe bronchopulmonary infection. Another patient died on day 5 from a polymicrobial bacteraemia caused by S . uuwus. Clostridiuni sporopric's and Micwcxwus sp. and a third patient died on day 8 from bacteraemia due to E. r d i (susceptible to ceftazidime and amikacin). Renal toxicity (elevation of serum creatinine) occurred in three episodes from the ceftazidime and teicoplanin group and none in ceftazidime plus amikacin group. Increases in serum transaniinases were observed in five patients treated with ceftazidime plus teicoplanin and in three patients receiving ceftazidime plus amikacin. DISC l J S S I O N The optimal empiric therapy for fever in cancer patients with granulocytopenia is still a controversial issue (Hughes ct d, 1990; Meunier. 1990: Viscoli at ul. 1988).
52
F. Meunier et al
Various factors have to be taken into consideration before recommending such a regimen. The degree of granulocytopenia ( < 1000 PMN/pI) as well as the duration of profound granulocytopenia have been recognized as poor prognostic factors and particularly for severely granulocytopenic patients, adequate coverage particularly against Gram-negative bacilli is mandatory. For patients with less severe granulocytopenia and short duration of marrow aplasia, the need for a combined regimen is probably less critical. In fact, mainly those patients who have documented Gram-negative bacteraemia seem to benefit from a combined regimen while patients with FUO (fever of unknown origin) seem to have a favourable response to most broad-spectrum antibacterial agents. During the last few years there has been a high incidence of documented infection caused by Gram-positive cocci. These are not necessarily related to a n increased rate of mortality (Viscoliet al. 1988).However, the morbidity related to bacteraemia caused by Gram-positive cocci is important. Prolonged hospitalization and delay in administration of further antineoplastic chemotherapy is often observed as well as pulmonary complications requiring a prolonged course of antibiotic, leading to a potential increased risk of fungal infection. In one study the addition of vancomycin to a classical regimen was associated with a decreased need for amphotericin B (Karpef a/. 1986). In the studies by Shenep et ul(1988) and Kramer e t a / (1986)the vancomycin-containing regimen was superior. However. whether empiric coverage should include vancomycin or teicoplanin is still a matter of discussion (Hughes et a/. 1990; Viscoli et al. 1988). This study confirms the worse prognosis of patients with documented bacteraemia. The limited number of patients with documented Gram-positive coccal bacteraemia does not allow us to make definite recommendations, but it is important to stress that two patients died from Gram-positive bacteraemia in both groups. In addition, the need for modification of therapy occurred in approximately 2 5% of patients treated with any combined regimen. Overall, this study suggests that ceftazidime plus teicoplanin may be a safe alternative to ceftazidime plus amikacin as empiric therapy for fever in granulocytopenic patients. However, each physician should take into consideration local epidemiological factors such as the most commonly offending pathogens and their susceptibility pattern, as well as prognostic factors that include the degree and duration of low white blood cell count.
ACKNOWLEDGMENTS We acknowledge the secretarial help of Mrs A. Meurrens and the excellent assistance of C. Dekoster and M. Janssens, data managers. This research was supported by Merrell Dow, Glaxo and Bristol.
REFERENCES Bibler, M.R., Frame, P.T.. Hagler. D.N..Bode. R.B., Staneck. J.L., Thadiitkul, V., Harris, J.E.. Haregewoin, A. & Bullock, W.E.. Jr
( 1 987) Ciinical evaluation of efficacy, pharmacokinetics. and safety of teicoplanin for serious Gram-positive infections. Antimicrobial Agents and Chemotherapy, 31, 207-212. Cimino. M.A.. Rotsein. C.. Slaughter, R.L. & Emrich. L.J. (1987) Relationship of serum antibiotic concentrations to nephrotoxicity in cancer patients receiving concurrent aminoglycoside and vancomycin therapy. American Journal of Medicine. 83. 10911097. Del Favero, A., Menichetti, F., Guerciolini, R., Bucaneve, G.. Baldelli. F.. Aversa. F.. Terenzi. A,. Davis, S. & Pauluzzi. S. (1987) Prospective randomized clinical trial of teicoplanin for empiric combined antibiotic therapy in febrile, granulocytopenic acute leukemia patients. Antimicrobial Agents and Chemotherapy. 31, 1126-1 129. Eisenberg, J.M.. Koffer. H.A.. Glick. H.A.. Connell. M.L.. Loss. L.E.. Talbot. G.H., Shusterman. N.H. & Strom. B.L. ( I 9 8 7 ) What is the cost of nephrotoxicity associated with aminoglycosides?Annals of Internal Medicine. 107, 900-909. EORTC International Antimicrobial Therapy Cooperative Group (1987) Ceftazidime combined with a short or long course of amikacin for empiric therapy of Gram-negative bacteremia in cancer patients with granulocytopenia. New England Journal of Medicine. 317, 1692-1698. Glupczynski, Y.. Lagast. H.. Van der Auwera. P., Thys, J.P., Crokaert. F., Yourassowsky. E.. Meunier-Carpentier. F.. Klastersky, 1.. Klains. J.P., Serruys-Schoutens.E. & Legrand. J.C. ( 1 9 8 6 )Clinical evaluation of teicoplanin for therapy of severe infections caused by Gram-positivebacteria. Antimicrobial Agents and Chemotherapg. 29, 52-57. Hughes, W.T.. Body, G.P.. Meyers. J.D.. Pizzo. P.A., SchimpfT, S.C.. Shenep. J.L.. Wade, J.C.. Young. L.S. & Yow. M.D. (1990) Guidelines for the use of antimicrobial agents in neutropenic patients with unexplained fever. journal oflnfectious Diseases. 161, 381-396. Jadeja. L.. Bolivar. R.. Fainstein. V.. Keating. M.. McCredie. K.. Hay, M. & Bodey. G. ( 1 9 8 4 ) Piperacillin plus vancomycin in the therapy of febrile episodes in cancer patients. Antimicrobial Agents and Chemotherapy. 26, 295-299. Karp. J.E.. Dick. J.D.. Angelopoulos. C.. Charache. P.. Green, L.. Burke. P.J. & Saral. R. (1986) Empiric use of vancomycin during prolonged treatment-induced granulocytopenia. American Journal o/ Medicine, 81, 237-242. Klastersky. J.. Zinner. S.H.. Calandra. T.. Gaya. H.. Glauser. M.P.. Meunier, F.. Rossi. M.. Schimpff.S.C.. Tattersall. M.. Viscoli. C. and the EORTC InternationalAntimicrobial CooperativeGroup ( 1988) Empiric antimicrobialtherapy for febrile granulocytopeniccancer patients: lesson from four EORTC trials. European Journal of Cancer and Clinical Oncology, 24, S35-S45. Kramer, B.. Ramphal. R. & Rand. K.H. ( 1 986) Randomized comparison between two ceftazidime-containingregimens and cephalothin-carbenicillin in febrile granulocytopenic cancer patients. Antimicrobial Agents and Chemotherapy, 30, 64-68. Meunier. F. (1990) Infections in Patients with Acute Leukemia and Lymphoma, 3rd edn (ed. by G. L. Mandell. R. G. Douglas and J. E. Bennett), pp. 2265-2275. Churchill Livingstone, New York. Pizzo, P.A., Hathorn, J.W.. Hiemenz, J., Browne. M., Commers, J.. Cotton, D., Gress, J., Longd, D., Marshall,D.. McKnight. J.. Rubin. M., Skelton, J., Thaler. M. &Wesley,R. (1986)A randomized trial comparing ceftazidime alone with combination antibiotic therapy in cancer patientswith feverand neutropenia. New England journal of Medicine, 315, 552-558. Pizzo. P.A., Ladisch. S.. Simon, R.M.. Gill, F. & Levine, A.S. (1978) Increasing incidence of Gram-positive sepsis in cancer patients. Medical and Pediatric Oncology, 5 , 241-244. Shenep. J.L., Hughes, W.T., Roberson. P.K., Blankenship, K.R.,
Ceftazidime with Teicoplanin or Amikacin for Granulocytopenia Baker, D.K.. Meyer. W.H., Gigliotti.F.. Sixbey,J.W.,Santana, V.M.. Feldman. S. & Lott. L. (1988) Vancomycin, ticarcillin. and amikacin compared with ticarcillin-clavulanate and amikacin in the empirical treatment of febrile, neutropenic children with cancer. New England Journal of Medicine, 319, 1053-1058. Verbist. L.. Tjandramaga, B.. Hendrickx, B., Van Hecken. A., Van Melle. P.. Verbesselt. R.. Verhaegen. J. & De Schepper. P.J. (1984) In vitro activity and human pharmacokinetics of teicoplanin. Antimicrobial Agents and Chemotherapy. 26, 881-886. Viscoli. C.. Van der Auwera. P. & Meunier. F. (1988)Gram-positive infections in granulocytopenic patients: an important issue? Journal of Antimicrobial Chemotherapy, 21, (Suppl. C). 149-1 56.
53
Wade, J.C., Schimpff, S.C.. Newman, K.A. & Wiernik. P.H. (1982) Staphylococcus epidermidis: an increasing cause of infection in patients with granulocytopenia. Annals of Internal Medicine. 9 6 , l 10. Winston, D.J., Dudnick, D.V.. Chapin. M.. Ho, W.G.. Sale, R.P. & Martin, M.J. (1983) Coagulase negative staphylococcal bacteremia in patients receiving immunosuppressive therapy. Archives of Internal Medicine. 143, 32-36. Young, L.S. (1988) Fever and septicemia. Clinical Approach to Infection in the Compromised Host, 2nd edn (4by .R. H. Rubin and L. S. Young), pp. 75-114. Plenum Medical, New York.
Ceftazidime plus teicoplanin versus ceftazidime plus amikacin as empiric therapy for fever in cancer patients with granulocytopenia F . MEUNIER.* P. V A N DER AUWERA,M. AOUNA N D D. BRONService de Mtdecine Interne et Laboratoire d’lnvestigation Clinique Henri Tagnon, Institut Iules Bordet, Centre des Tumeurs de I’Universitt Libre de Bruxelles, Belgique
Summary. In a prospective randomized study, 100 episodes of fever ( >38°C) and granulocytopenia ( < 100O/p1) in cancer patients were empirically treated with ceftazidime(2 g every 8 h) plus teicoplanin (400 mg every 8 h on day 1: 400 mg every day thereafter) or ceftazidime (2 g every 8 h) plus amikacin (500 mg every 8 h). Bacteraemia. clinically documented infection and possible infection were documented in seven. 11 and 19 patients treated with ceftazidimeplus teicoplanin and in 11, four and 17 patients treated with ceftazidime plus amikacin. Overall, the response rate was similar in the two groups of patients as
was the need for treatment modifications and the rate of death. For documented Gram-positive bacteraemia, the response rate was 2 / 5 patients treated with ceftazidimeplus teicoplanin and 2/7 with ceftazidime plus amikacin: for documented Gram-negative bacteraemia. the response rate was 1/2 and 3/4 patients respectively. N o breakthrough bacteraemia was observed. Tolerance was excellent. although renal toxicity (elevation of serum creatinine) was observed in three patients treated with ceftazidime plus teicoplanin and in none allocated to ceftazidime plus amikacin.
Infections are a major cause of morbidity and mortality in cancer patients. especially in granulocytopenic hosts (EORTC, 1987; Klastersky et al, 1988: Meunier, 1990). Empiric treatment of fever in those patients is mandatory but the optimal regimen is still controversial and usually consists of the administration of a broad-spectrum beta-lactam antibiotic sometimes combined with an aminoglycoside (EORTC. 1987: Klastersk: et al. 1988: Meunier, 1990: Young, 1988). The use of combination therapy is particularly advocated for bacter: kmic infections, especially when due to Gram-negative baci Ci and when severe granulocytopenia persists (EORTC, 1’987: Young, 1988). In some circumstances, monotherapy has been found effective (Pizzo et al. 1986) with no increase in mortality compared with a combination therapy. However, using this therapeutic approach, the number of treatment modifications is uniformly high, vancomycin, amikacin and amphotericin B being the most frequent antimicrobial agents added during treatment. The increasing number of Gram-positive micro-organisms causing bacteraemia in granulocytopenic cancer patients
(Klastersky et al. 1988: Pizzo et al. 1978: Viscoli et al. 1988: Wade et al, 1982: Winston et al. 1983) has prompted a renewed interest in the inclusion of specific anti-Grampositive antimicrobial coverage as empiric therapy together with agents effective against Gram-negative bacilli (Jadeja et al. 1984: Karp et al, 1986; Kramer et al, 1986: Shenep et al, 1988). Whether vancomycin should be added to empiric therapy is still a matter of discussion (Hughes et ai, 1990: Meunier. 1990: Viscoli et al, 1988): in particular, the potential risk of cumulative toxicity in patients treated with vancomycin. aminoglycosides. amphotericin B and cyclosporin A is a matter of concern (Cimino et al, 1987: Eisenberg et al. 1987). Teicoplanin is a glycopeptide antibiotic effective against most Gram-positive cocci. It has a similar spectrum of activity to vancomycin, possibly with less side-effects (Bibler et al, 1987: Cimino et al. 1987: Glupczynski et al. 1986; Verbist et al, 1984).It has been used successfully in a limited number of febrile granulocytopenic patients when administered in combination with ceftazidime and amikacin (Del Favero et al, 1987). Therefore, we were interested in evaluating the potential benefit of empiric therapy of fever in granulocytopenic patients with ceftazidime plus teicoplanin versus ceftazidimeplus amikacin, a regimen generally considered as standard empiric therapy in such circumstances (EORTC. 1987).
* Senior Research Associate, Fund for Medical Scientific Research. Brussels. Belgium.
49
50
F. Meunier et a1
PATIENTS AND METHODS Granulocytopenic cancer patients (PMN less than 1 000/pl) with fever (temperature > 38°C) hospitalized at the Institut J. Bordet, Brussels, Belgium, were eligible for this randomized and prospective study. Patients were allocated to either ceftazidime plus amikacin or ceftazidime plus teicoplanin using sealed envelopes. Ceftazidime was administered in three intravenous doses of 2 g daily and amikacin was administered immediately afterwards (3 times 500 mg daily). Ceftazidime was reconstituted in 50 ml of 5% dextrose in distilled water and infused over 5 min. Amikacin was similarly reconstituted but the infusion time was 30 min. Teicoplanin was also reconstituted in 50 ml of 5% dextrose in distilled water and given as 1 5 min infusion following the ceftazidime infusion. The loading dose of teicoplanin was 4 0 0 mg t.i.d. on day 1. and 400 mg teicoplanin was administered once daily thereafter. A complete history and physical examination were obtained for all eligible patients. Sets of blood cultures were drawn prior to the initiation of antibiotics as well as urine cultures and other relevant samples according to clinical presentation. Chest X-rays were also performed prior to empiric therapy and laboratory investigations including haematology. coagulation and blood chemistry for renal and hepatic functions. Initial serum creatinine > 2 mg/dl was an exclusion criterion.
Nephrotoxicity was assessed by the measure of serum creatinine: a rise of 0.5 mg/dl over baseline was considered as nephrotoxicity. Hepatotoxicity was defined as a rise of transaminases (1.5 times) above baseline values if noted in the absence of other apparent causes. Patients undergoing allogeneic bone marrow transplantation were allocated to ceftazidime plus teicoplanin (two patients). Evaluation of response to empiric therapy was performed according to the criteria used by the EOKTC International Antimicrobial Therapy Co-operative Group (EORTC, 1987: Klastersky et a], 1988). Any addition of an antibiotic was considered as failure of empiric therapy. Patients were considered as unevaluable if an episode of fever was classified as doubtful infection (such as fever due to a drug or blood product transfusion), if the patient had a viral or a fungal infection or a major violation of their protocol occurred. RESULTS Overall 100 patients were eligible. The characteristics of the patients in both groups were very similar for age, sex ratio, underlying disease and previous chemoprophylaxis with antibacterial agents (Table I). The number of unevaluable episodes (doubtful infection.
Table I. Characteristics of the patients
No. of eligible patients Sex: male/female Age in years: median (range) Underlying diseases Leukaemia Hodgkin's or lymphoma Solid tumour Multiple myeloma Allogeneic BMT* Autologous BMT* Types of infection Microbiologically documented infection With bacteraemia Without bacteraemia ClinicaIly documented infection Possible infection Unevaluable patients Doubtful infection Fungal infection Mixed infection (viral and fungal) Viral infection Protocol violation Previous chemoprophylaxis No. of episodes Median duration in days (range)
* BMT= bone marrow transplantation.
Ceftazidime plus teicoplanin
Ceftazidime plus amikacin
50 30120 47 (20-80)
50 27/23 54.5 ( 1 5-75)
25 12 12
26 7 17 0 0 2
1
2 7 9
7 2 11
19
15 I1 4 4 17
3 3 1 2 2
18 8 (1-23)
17 9 (1-14)
Ceftazidime with Teicoplanin or Amikacin for Granulocytopenia
51
Table 11. Results and outcome
Ceftazidime plus teicoplanin
Ceftazidime plus arnikacin
Duration of therapy as allocated at randomization (range)
39 25 11 14 (1-90) 8 (1-21)
36 19 12 14 (2-86) 7 (2-1 3 )
Success/total (%) Overall Bacteraemia Clinically documented infection Possible infection
26/39 (66.6%) 317 (42.8%) 611 1 (54.5%) 16/19 (84.2%)
24/36 167%) 5/11 (45%) 414 (100%) 12/17 (70.5%)
Modilication of antibacterial therapy Death within a month
10 8
No. of evaluable patients No. of patients with PMN< 1OO/jd at randomization No. of patients who recovered PMN > lOOO/pl during therapy Overall duration of granulocytopenia in days (range)
fungal or viral infection and protocol violation) was coniparable. 1 1 in the ceftazidime plus teicoplanin group and 1 4 in the ceftaxidime plus amikacin group. Two episodes in the ceftazidime plus teicoplanin group were unevaluable because of a protocol violation: for one patient, amikacin was added on day 2 after initiation of empiric therapy although the patient was already afebrile. This patient did not develop nt side-effects and survived. Another patient was treated only for a few days according to the protocol and requested to be discharged tc pursue oral treatment at home. Scven episodes in the ceftaz lime plus amikacin group were unevaluable owing to proto ,ol violation: therapy was modified and teicoplanin added iii two patients on day 3 without good reasons. Another patient. afebrile after 2 d. received teicoplanin on day h without consent of the investigator and without good reasons. That patient died on day l h from disseminated pneumonia. Autopsy was not performed. One patient was febrile for 1 d and treatment stopped after 3 d: this patient survived. One patient was afebrile after 3 d but received teicoplanin from day 7 (without appropriate reasons).One patient with perineal abscesses received metronidazole from day I . He survived and was also considered as unevaluable. Finally. one patient was treated with ceftazidime plus amikacin for 3 d, was afebrile but had an uncontrolled leukwnia and was sent home. He died on day I 1. Table I1 shows the duration oftherapy as well as the degree of granulocytopenia and the outcome of patients in both groups. There were a similar number of patients recovering adequate granulocyte counts (I'MN > 1000/pl) during therapy. The overall duration of granulocytopenia was similar (14 d ) in both groups as was the duration of therapy as allocated at randomization: 8 and 7 d respectively in both groups. The improvement rate was similar for the two groups even when the episodes of persisting profound granulocytopenia were considered. Failure was never due to persisting bacteraemia. but to persisting fever which was treated empirically by modifying the antimicrobial regimen. The overall success rate was 66% for ceftazidime plus 1
(25.6%)
9
(25%)
7
teicoplanin as well as for ceftazidime plus amikacin. Success was observed in 3/7 patients with bacteraemia treated with ceftazidimeplus teicoplanin and in 5/11 patients treated with ceftazidime plus amikacin. For patients without microbiologically documented infection, success was observed in 24/ 3 0 patients treated with ceftazidinie plus teicoplanin and in 1 6 / 2 1 patients treated with ceftazidinie plus amikacin. The need for modifications of antibacterial therapy was similar in both groups and was observed in 25% of the patients. The number ofpatients that died within a month after empiric therapy for fever was also similar in both groups. Three patients treated with ceftazidime plus teicoplanin died as a result of their primary infection. One patient died in septic shock (on day 1 ) caused by bacteraemia due to Eschericliia coli (susceptible to ceftazidime). Another patient died (on day 6 ) from bacteraeinia caused by Stuphylococcits aimus and Streptococcusjiwnlis and the third patient died (on day 14) with uncontrolled bacteraemia caused by Staphylococcxs c~pidarriiidis(related to a 'Port a cath'). Among the patients who died while receiving ceftazidime plus amikacin, three also died from their primary infection: one patient died on day 2 from bacteraemia caused by S. epidcvwidis associated with a severe bronchopulmonary infection. Another patient died on day 5 from a polymicrobial bacteraemia caused by S . uuwus. Clostridiuni sporopric's and Micwcxwus sp. and a third patient died on day 8 from bacteraemia due to E. r d i (susceptible to ceftazidime and amikacin). Renal toxicity (elevation of serum creatinine) occurred in three episodes from the ceftazidime and teicoplanin group and none in ceftazidime plus amikacin group. Increases in serum transaniinases were observed in five patients treated with ceftazidime plus teicoplanin and in three patients receiving ceftazidime plus amikacin. DISC l J S S I O N The optimal empiric therapy for fever in cancer patients with granulocytopenia is still a controversial issue (Hughes ct d, 1990; Meunier. 1990: Viscoli at ul. 1988).
52
F. Meunier et al
Various factors have to be taken into consideration before recommending such a regimen. The degree of granulocytopenia ( < 1000 PMN/pI) as well as the duration of profound granulocytopenia have been recognized as poor prognostic factors and particularly for severely granulocytopenic patients, adequate coverage particularly against Gram-negative bacilli is mandatory. For patients with less severe granulocytopenia and short duration of marrow aplasia, the need for a combined regimen is probably less critical. In fact, mainly those patients who have documented Gram-negative bacteraemia seem to benefit from a combined regimen while patients with FUO (fever of unknown origin) seem to have a favourable response to most broad-spectrum antibacterial agents. During the last few years there has been a high incidence of documented infection caused by Gram-positive cocci. These are not necessarily related to a n increased rate of mortality (Viscoliet al. 1988).However, the morbidity related to bacteraemia caused by Gram-positive cocci is important. Prolonged hospitalization and delay in administration of further antineoplastic chemotherapy is often observed as well as pulmonary complications requiring a prolonged course of antibiotic, leading to a potential increased risk of fungal infection. In one study the addition of vancomycin to a classical regimen was associated with a decreased need for amphotericin B (Karpef a/. 1986). In the studies by Shenep et ul(1988) and Kramer e t a / (1986)the vancomycin-containing regimen was superior. However. whether empiric coverage should include vancomycin or teicoplanin is still a matter of discussion (Hughes et a/. 1990; Viscoli et al. 1988). This study confirms the worse prognosis of patients with documented bacteraemia. The limited number of patients with documented Gram-positive coccal bacteraemia does not allow us to make definite recommendations, but it is important to stress that two patients died from Gram-positive bacteraemia in both groups. In addition, the need for modification of therapy occurred in approximately 2 5% of patients treated with any combined regimen. Overall, this study suggests that ceftazidime plus teicoplanin may be a safe alternative to ceftazidime plus amikacin as empiric therapy for fever in granulocytopenic patients. However, each physician should take into consideration local epidemiological factors such as the most commonly offending pathogens and their susceptibility pattern, as well as prognostic factors that include the degree and duration of low white blood cell count.
ACKNOWLEDGMENTS We acknowledge the secretarial help of Mrs A. Meurrens and the excellent assistance of C. Dekoster and M. Janssens, data managers. This research was supported by Merrell Dow, Glaxo and Bristol.
REFERENCES Bibler, M.R., Frame, P.T.. Hagler. D.N..Bode. R.B., Staneck. J.L., Thadiitkul, V., Harris, J.E.. Haregewoin, A. & Bullock, W.E.. Jr
( 1 987) Ciinical evaluation of efficacy, pharmacokinetics. and safety of teicoplanin for serious Gram-positive infections. Antimicrobial Agents and Chemotherapy, 31, 207-212. Cimino. M.A.. Rotsein. C.. Slaughter, R.L. & Emrich. L.J. (1987) Relationship of serum antibiotic concentrations to nephrotoxicity in cancer patients receiving concurrent aminoglycoside and vancomycin therapy. American Journal of Medicine. 83. 10911097. Del Favero, A., Menichetti, F., Guerciolini, R., Bucaneve, G.. Baldelli. F.. Aversa. F.. Terenzi. A,. Davis, S. & Pauluzzi. S. (1987) Prospective randomized clinical trial of teicoplanin for empiric combined antibiotic therapy in febrile, granulocytopenic acute leukemia patients. Antimicrobial Agents and Chemotherapy. 31, 1126-1 129. Eisenberg, J.M.. Koffer. H.A.. Glick. H.A.. Connell. M.L.. Loss. L.E.. Talbot. G.H., Shusterman. N.H. & Strom. B.L. ( I 9 8 7 ) What is the cost of nephrotoxicity associated with aminoglycosides?Annals of Internal Medicine. 107, 900-909. EORTC International Antimicrobial Therapy Cooperative Group (1987) Ceftazidime combined with a short or long course of amikacin for empiric therapy of Gram-negative bacteremia in cancer patients with granulocytopenia. New England Journal of Medicine. 317, 1692-1698. Glupczynski, Y.. Lagast. H.. Van der Auwera. P., Thys, J.P., Crokaert. F., Yourassowsky. E.. Meunier-Carpentier. F.. Klastersky, 1.. Klains. J.P., Serruys-Schoutens.E. & Legrand. J.C. ( 1 9 8 6 )Clinical evaluation of teicoplanin for therapy of severe infections caused by Gram-positivebacteria. Antimicrobial Agents and Chemotherapg. 29, 52-57. Hughes, W.T.. Body, G.P.. Meyers. J.D.. Pizzo. P.A., SchimpfT, S.C.. Shenep. J.L.. Wade, J.C.. Young. L.S. & Yow. M.D. (1990) Guidelines for the use of antimicrobial agents in neutropenic patients with unexplained fever. journal oflnfectious Diseases. 161, 381-396. Jadeja. L.. Bolivar. R.. Fainstein. V.. Keating. M.. McCredie. K.. Hay, M. & Bodey. G. ( 1 9 8 4 ) Piperacillin plus vancomycin in the therapy of febrile episodes in cancer patients. Antimicrobial Agents and Chemotherapy. 26, 295-299. Karp. J.E.. Dick. J.D.. Angelopoulos. C.. Charache. P.. Green, L.. Burke. P.J. & Saral. R. (1986) Empiric use of vancomycin during prolonged treatment-induced granulocytopenia. American Journal o/ Medicine, 81, 237-242. Klastersky. J.. Zinner. S.H.. Calandra. T.. Gaya. H.. Glauser. M.P.. Meunier, F.. Rossi. M.. Schimpff.S.C.. Tattersall. M.. Viscoli. C. and the EORTC InternationalAntimicrobial CooperativeGroup ( 1988) Empiric antimicrobialtherapy for febrile granulocytopeniccancer patients: lesson from four EORTC trials. European Journal of Cancer and Clinical Oncology, 24, S35-S45. Kramer, B.. Ramphal. R. & Rand. K.H. ( 1 986) Randomized comparison between two ceftazidime-containingregimens and cephalothin-carbenicillin in febrile granulocytopenic cancer patients. Antimicrobial Agents and Chemotherapy, 30, 64-68. Meunier. F. (1990) Infections in Patients with Acute Leukemia and Lymphoma, 3rd edn (ed. by G. L. Mandell. R. G. Douglas and J. E. Bennett), pp. 2265-2275. Churchill Livingstone, New York. Pizzo, P.A., Hathorn, J.W.. Hiemenz, J., Browne. M., Commers, J.. Cotton, D., Gress, J., Longd, D., Marshall,D.. McKnight. J.. Rubin. M., Skelton, J., Thaler. M. &Wesley,R. (1986)A randomized trial comparing ceftazidime alone with combination antibiotic therapy in cancer patientswith feverand neutropenia. New England journal of Medicine, 315, 552-558. Pizzo. P.A., Ladisch. S.. Simon, R.M.. Gill, F. & Levine, A.S. (1978) Increasing incidence of Gram-positive sepsis in cancer patients. Medical and Pediatric Oncology, 5 , 241-244. Shenep. J.L., Hughes, W.T., Roberson. P.K., Blankenship, K.R.,
Ceftazidime with Teicoplanin or Amikacin for Granulocytopenia Baker, D.K.. Meyer. W.H., Gigliotti.F.. Sixbey,J.W.,Santana, V.M.. Feldman. S. & Lott. L. (1988) Vancomycin, ticarcillin. and amikacin compared with ticarcillin-clavulanate and amikacin in the empirical treatment of febrile, neutropenic children with cancer. New England Journal of Medicine, 319, 1053-1058. Verbist. L.. Tjandramaga, B.. Hendrickx, B., Van Hecken. A., Van Melle. P.. Verbesselt. R.. Verhaegen. J. & De Schepper. P.J. (1984) In vitro activity and human pharmacokinetics of teicoplanin. Antimicrobial Agents and Chemotherapy. 26, 881-886. Viscoli. C.. Van der Auwera. P. & Meunier. F. (1988)Gram-positive infections in granulocytopenic patients: an important issue? Journal of Antimicrobial Chemotherapy, 21, (Suppl. C). 149-1 56.
53
Wade, J.C., Schimpff, S.C.. Newman, K.A. & Wiernik. P.H. (1982) Staphylococcus epidermidis: an increasing cause of infection in patients with granulocytopenia. Annals of Internal Medicine. 9 6 , l 10. Winston, D.J., Dudnick, D.V.. Chapin. M.. Ho, W.G.. Sale, R.P. & Martin, M.J. (1983) Coagulase negative staphylococcal bacteremia in patients receiving immunosuppressive therapy. Archives of Internal Medicine. 143, 32-36. Young, L.S. (1988) Fever and septicemia. Clinical Approach to Infection in the Compromised Host, 2nd edn (4by .R. H. Rubin and L. S. Young), pp. 75-114. Plenum Medical, New York.
