Volume 15, Issue 4 , Pages 512-516, April 2009
Imipenem/Cilastatin with or without Glycopeptide as Initial Antibiotic Therapy for Recipients of Autologous Stem Cell Transplantation: Results of a Spanish Multicenter Study
Article Outline
Abstract
We analyzed the efficacy of imipenem/cilastatin alone (group I, 197 patients) or in combination with a glycopeptide (group I + G, 231 patients) as first-line antibiotic therapy for 2 consecutive cohorts of autologous stem cell transplantation (ASCT) recipients with febrile neutropenia. From June 2001 to June 2002, patients received imipenem/cilastatin (500 mg/6 hours), and from July 2002 to December 2003, they received imipenem/cilastatin as for group I plus a glycopeptide (vancomycin, 1 g/12 hours or teicoplanin, 400 mg/day). Fever of unknown origin accounted for 33.5% of episodes (66 patients) in group I and 50% of episodes (116 patients) in group I + G (P = .005). Bacteremia occurred in 55 patients (28%) in group I and in 51 patients (22%) in group I + G (P = .16). Resolution of fever without modification of the therapy regimen was observed in 108 patients (55%) and 159 patients (69%) in groups I and I + G, respectively (P = .003). The median interval to defervescence (4 days) and overall mortality were similar between groups. Inclusion of a glycopeptide in the initial antibiotic regimen for febrile neutropenia results in a higher success rate without modifying the regimen. However, glycopeptide inclusion does not improve the interval to defervescence or mortality rate.
Key Words: Imipenem, Glycopeptide, Neutropenia, Autologous transplantation
Introduction
Infection is a frequent complication among patients undergoing an autologous stem cell transplantation (ASCT) 1, 2, 3, 4, 5, 6. Early antibiotic therapy reduces morbidity and mortality, and is standard practice for ASCT recipients who exhibit fever and neutropenia. The standard therapy for such patients is monotherapy with a broad-spectrum antipseudomonal β-lactam antibiotic [7]. Several studies have suggested that the addition of a glycopeptide to β-lactam antibiotics as initial therapy for neutropenic fever did not improve therapy failure rate or mortality rate 8, 9, 10. However, these studies were carried out in heterogeneous populations of cancer patients who may not have been exposed to risk factors for Gram-positive infection frequently present in ASCT patients, such as severe mucositis, the presence of central venous catheters, and parenteral nutrition. To study the efficacy of antimicrobial monotherapy and to assess the role of the empirical use of anti-Gram-positive antibiotics for febrile neutropenic patients undergoing ASCT, we conducted an observational, prospective, multicenter study using 2 consecutive cohorts of ASCT recipients with febrile neutropenia to determine the efficacy of imipenem/cilastatin alone (group I) compared with imipenem/cilastatin plus a glycopeptide antibiotic (group I + G).
Patients and Methods
Between June 2001 and December 2003, an opened, comparative, unblinded, sequential, multicenter study was conducted at 23 Spanish institutions ascribed to the PETHEMA-GETH group (see Acknowledgments). The trial was designed in accordance with guidelines issued by the Immunocompromised Host Society (IHS) Consensus Conference and the European Society of Clinical Microbiology and Infectious Disease (ESCMID) 11, 12, 13.
The study protocol was conducted in accordance with the Declaration of Helsinki, and was approved by the ethics committee of each participating institution; informed consent was provided by all patients who fulfilled the inclusion criteria.
Patients undergoing ASCT were enrolled if they met the following inclusion criteria: (1) hospitalized patients aged 18 years and older; (2) an absolute granulocyte count <1 × 109/L after ASCT, which was anticipated to decrease to <0.5 × 109/L within 24 to 48 hours; and (3) fever, defined as an axillary temperature ≥38°C on 2 occasions at least 1 hour apart, or 38.5°C on 1 occasion. Exclusion criteria were allergy to any of the antibiotics used in the study (imipenem/cilastatin, vancomycin, or teicoplanin) or a history of allergy to β-lactam antibiotics, pregnant or breast-feeding women, a serum creatinine level >200 μmol/L or a creatinine clearance rate of <40 mL/min, concomitant i.v. treatment with an antibiotic or i.v. administration of an antibiotic within 96 hours of admission to the study, and HIV infection.
Specimens for microbiologic analysis were obtained before and during therapy, as clinically indicated. Patients with suspected pneumonia were subjected to bronchoalveolar lavage or fibrobronchoscopy brushing if possible. Samples for blood culture were obtained daily from patients with persistent fever; in cases of established bacteremia, samples were obtained until the blood culture results were negative. All bacterial isolates were tested for in vitro susceptibility to imipenem/cilastatin using the Kirby-Bauer disc diffusion method or by determining minimum internal controls (MICs), as recommended by the National Committee on Clinical Laboratory Standards (NCCLS) [14]. Primary febrile episodes were classified according to guidelines issued by the IHS Consensus Conference and the ESCMID [11].
From June 2001 to June 2002, patients received i.v. imipenem/cilastatin as initial antibiotic therapy at a dosage of 500 mg every 6 hours (group I). From July 2002 to December 2003, patients received imipenem/cilastatin in combination with a glycopeptide consisting of vancomycin (1 g every 12 hours, 99 cases) or teicoplanin (400 mg/day, 132 cases) (group I + G). Patients received antibiotic therapy for a minimum of 7 days in total and for at least 4 days after their last fever. Group I patients were given glycopeptide (vancomycin or teicoplanin) if a fever persisted for 72 hours. Antifungal therapy was commenced if patients remained febrile or if their clinical status worsened after 4 to 6 days of antibiotic therapy.
The primary endpoint of the study was the response rate in the absence of treatment modification 72 hours after the initial antimicrobial treatment. Secondary endpoints were response rates after modification of the initial antibiotic regimen (overall response), response rates in microbiologically documented infection (MDI), fever of unknown origin (FUO), time to response, and overall mortality. Response was assessed 72 hours after treatment (early response) and at completion of therapy (final response). A treatment was categorized having a successful response if all of the following criteria were met: afebrile (<38°C) for 4 consecutive days, abatement of signs and symptoms of infection, eradication of the infecting microorganism (when present), and no recurrence of the primary infection at 1 week after completion of treatment. Treatment failure was defined as: (1) death from primary infection, (2) modification of the initial antibacterial treatment to eradicate the primary infection, and (3) in vitro resistance to the β-lactam agent. A patient's response was considered nonassessable in the following circumstances: (1) coexistent fungal or viral infection, (2) febrile episode not related to infection, and (3) protocol violation (nonadherence to the protocol or early discontinuation of treatment because of severe adverse effects).
Adverse events were graded according to the World Health Organization (WHO) grading system [15]. Secondary infection was defined as an infection caused by a microorganism other than the initial pathogen that developed during antibiotic therapy or within 1 week of discontinuation of antibiotic treatment.
All data were analyzed using the 4F and 3D routines of the Biomedical Data Package (BMDP, University of California, Berkeley, CA) [16]. Fisher's exact test, the chi-square test, Student's t-test, and Mann-Whitney's U test were used to test differences between end point means. Values of P < .05 were considered statistically significant.
Results
Four hundred twenty-eight patients who met all the inclusion/exclusion criteria were included in the study; group I consisted of 197 patients and group I + G consisted of 231 patients. Demographic data are shown in Table 1. All but 10 patients received ASCT involving peripheral blood stem cells (PBSCs). The duration of neutropenia (<1 × 109/L) in group I + G did not differ from that in group I (11.3 days versus 10.9 days, P = .12).
Table 1. Patient Characteristics
| No. of Patients | |||
|---|---|---|---|
| Treatment Group | Total (%) | I (n = 197) | I + G (n = 231) |
| Age (years) | |||
 Mean† | 48.4 | 51.1 | |
| Median | 51 | 53 | |
| Range | 20-70 | 18-74 | |
| Sex | |||
| Male | 114 (58) | 127 (55) | |
| Female | 83 (42) | 104 (45) | |
| Underlying disease‡ | |||
| Multiple myeloma | 179 (42) | 69 (35) | 110 (48) |
| Lymphoma | 167 (39) | 90 (46) | 77 (33) |
| Acute leukemia | 59 (13) | 28 (14) | 31 (13) |
| Other‡ | 23 (6) | 10 (5) | 13 (6) |
| Oral prophylaxis§ | |||
| None | 157 (37) | 76 (39) | 81 (35) |
| Fluoroquinolone | 158 (37) | 57 (29) | 101 (44) |
| Cotrimoxazole | 71 (16) | 36 (18) | 35 (15) |
| Fluoroquinolone + Cotrimoxazole | 35 (8) | 23 (11) | 12 (5) |
| Others | 7 (2) | 5 (3) | 2 (1) |
| Use of growth factors (G-CSF)¶ | |||
| Yes | 357 (85) | 154 (79) | 203 (89) |
| Type of central venous line before onset of fever | |||
| Port-a-Cath or Hickman | 110 (33) | 57 (38.5) | 53 (29) |
| Central | 205 (62) | 80 (54) | 125 (69) |
| Peripheral | 14 (5) | 11 (7.5) | 3 (2) |
| Use of total parenteral nutrition | |||
| Yes | 157 (39) | 70 (37) | 87 (40) |
| Mucositis at start of intravenous antibiotics | |||
| Absent | 67 (16) | 27 (14) | 40 (17) |
| Present | 361 (84) | 170 (86) | 191 (83) |
†P = .04. |
‡P = .008. |
§P = .09. |
¶P = .006. |
FUO accounted for 33.5% of episodes in group I and 50% of episodes in group I + G (P = .005). Clinically documented infections (CDI) were observed in 66 patients and 53 patients in groups I and I + G, respectively (P = .01). The prevalence of bacteremia was 30% in group I and 22% in group I + G (P = .16) (Table 2). There were 6 and 8 cases of polymicrobial bacteremia in the I and I + G groups, respectively. Table 3 lists microorganisms isolated from bacteremia patients according to the treatment groups.
Table 2. Type of Infection According to Treatment Group
| Type of Infection | I (n = 197) No. of Patients (%) | I+G (n = 231) No. of Patients (%) | P |
|---|---|---|---|
| Fever of unknown origin | 66 (33.5) | 116 (50) | .005 |
| Clinically documented infection | 66 (33.5) | 53 (23) | .01 |
| Bacteremia∗ | 55 (28) | 51 (22) | |
| Gram-positive | 34 | 28 | |
| Gram-negative | 15 | 15 | |
| Microbiologically documented infection† | 10 (5) | 11 (5) |
∗There were 6 and 8 cases of polymicrobial bacteremia in Groups I and I + G, respectively. |
†Without bacteremia. |
Table 3. Isolated Pathogens in Bacteremia According to Treatment Group
| Pathogen | I | I + G |
|---|---|---|
| CNS | 29 | 22 |
| Escherichia coli | 7 | 14 |
| KES | 4 | 2 |
| Streptococcus spp | 3 | 7 |
| Pseudomonas spp | 4 | 2 |
| Staphylococcus aureus | 2 | 3 |
| Other Gram-positive | 9 | 6 |
| Other Gram-negative | 6 | 5 |
Breakthrough infections were observed in 56 patients; 29 in group I, and 27 in group I + G (P = ns).
Resolution of fever without modification to therapy was observed in 108 patients (55%) and 159 patients (69%) in groups I and I + G, respectively (P = .003). The most frequent modifications to the initial antimicrobial regimens were the addition of glycopeptide (56 patients in group I) and the addition of amikacin (19 patients in group I and 18 patients in group I + G). The median time to defervescence was 4 days (range: 1-40 days) for group I and 4 day (range: 2-36 days) for group I + G (P = ns).
The response in patients presenting with FUO was 70% in group I and 74% in group I + G (P = ns). With respect to MDI, a higher response rate among Gram-positive bacteremias was observed in group I + G than in group I (68% versus 29%, respectively; P = .0025; Table 4). Eight and 18 patients of groups I and I + G, respectively, received antifungal agents (P = .1). Five (1.2%) of the 428 patients died, and mortality rate did not differ significantly between groups. Adverse events were associated with antibiotics in 7 (3.5%) patients in group I, and in 12 (5.2%) in group I + G.
Table 4. Clinical Response and Time to Defervescence According to Treatment Group
| I (n = 197) No. of Patients (%) | I + G (n = 231) No. of Patients (%) | |
|---|---|---|
| Complete response to first-line antimicrobial therapy | 108 (55) | 159 (69) |
| Complete response in patients with FUO | 46 (70) | 86 (74) |
| Complete response in patients with bacteremia∗ | 20 (36) | 32 (63) |
| Gram-positive† | 10 (29) | 19 (68) |
| Gram-negative | 7 (47) | 8 (53) |
| Polimicrobial | 3 (50) | 5 (62.5) |
| Overall response after modification | 86 (97) | 70 (97) |
| Median time to defervescence (days) | 4 (range 1-40) | 4 (range 2-36) |
∗P = .006. |
†P = .0025. |
The median duration of hospitalization was 26 days (range: 18-96 days) in group I and 26 days (range: 10-84 days) in group I + G (P = ns). The median and range of the duration of hospitalization after commencement of antibiotic administration was 12.6 days (2-42 days) in group I and 12.2 days (2-51 days) in group I + G (P = .28). Fifty-six patients in group I received glycopeptides for 9 days; the median duration of glycopeptide therapy for the 231 patients in group I + G was 8 days. The cost of glycopeptide therapy for group I was 26,495.56 € (135.49 €/patient) and the median cost of glycopeptide therapy for group I + G was 64,283.44 € (278.28 €/patient).
Discussion
This study shows that empirical monotherapy with imipenem/cilastatin is an effective approach for neutropenic patients with fever after ASCT. The addition of a glycopeptide to the initial antimicrobial therapy does not reduce mortality, duration of hospitalization, or time to defervescence, and it increases the cost of therapy.
Our results show that in ASCT recipients, inclusion of a glycopeptide in the initial antibiotic regimen for febrile neutropenia results in a high success rate and abrogates the need to modify the initial regimen. Because coagulase-negative staphylococci are present in most patients with MDI, the inclusion of a glycopeptide in first-line therapy might be considered for ASCT patients. In our series, 56 patients who developed persistent fever and who were not treated initially with glycopeptides subsequently required glycopeptide treatment. However, because of the nonblinded nature of our study, the success rate without modification should be interpreted cautiously. On the other hand, glycopeptide addition had no effect on clinically relevant endpoints such as time to defervescence, incidence of breakthrough infections, and mortality rate, as patients with Gram-positive bacteremia were appropriately treated without an increase in morbidity or mortality.
The higher incidence of FUO in group I + G was probably because of the early addition of a glycopeptide, which probably prevented the growth of Gram-positive bacteria in blood cultures. Conversely, more patients in group I presented with CDI.
Some authors have recommended the administration of glycopeptide for patients with febrile neutropenia who are at high risk of shock or respiratory distress syndrome associated with viridans streptococcal infections 17, 18, 19, 20. However, these complications occur infrequently when a carbapenem is used 21, 22, 23, 24, 25. In fact, our study shows that none of the cases of febrile neutropenia was accompanied by shock or severe sepsis, probably because of the use of imipenem/cilastatin as initial antibiotic regimen. These findings suggest that the systematic use of glycopeptides is not justified in patients undergoing ASCT, at least when treated with imipenem/cilastatin. However, the addition of a glycopeptide and amikacin should be taken into consideration as initial antimicrobial therapy for patients who present with septic shock or acute respiratory distress syndrome.
Our data suggest that it is cost-effective to withhold glycopeptide treatment during the first 3 days of febrile neutropenia or until Gram-positive blood cultures are grown [9]. In addition, a reduction in the use of glycopeptides would help to retard the emergence of glycopeptide-resistant organisms.
In conclusion, our study shows that in ASCT recipients, inclusion of a glycopeptide in the initial antibiotic regimen for febrile neutropenia results in an increase of the antibiotic therapy costs, without benefits in terms of clinically relevant endpoints such as the interval to defervescence, incidence of breakthrough infection, and mortality rate.
Acknowledgments
Financial disclosure: The authors have nothing to disclose.
Study Group Members: The following institutions and individuals participated in this study: Barcelona, H. Sant Pau (Dr. R. Martino); Valencia, H. La Fe (Dr. I. Jarque); Salamanca, H. Clínico (Dr. L Vázquez); Barcelona, H. Clinic (Dr. M. Rovira); Madrid, H. Ramón y Cajal (Dr. J. López); Barcelona, H. Germans Trias i Pujol (Dr. M. Batlle); Madrid, H. La Princesa (Dr. R. de la Cámara); Barcelona, H. Vall d'Hebrón (Dr. A. Juliá); Madrid, H. 12 de Octubre (Dr. J.J. Lahuerta); La Coruña, H. J. Canalejo (Dr. G. Debén); Madrid, H. Clínico (Dr. J. Díaz); Castellón, H. General (Dr. R. García), Valencia, Instituto Valenciano de Oncología (Dr. I. Picón); Murcia, H. Virgen de la Arrixaca (Dr. A. Morales); Lugo, Complejo Hospitalario Xeral (Dr. E. Lavilla); Santiago de Compostela, H. Clínico (Dr. N. Alonso); Málaga, H. Carlos Haya (Dr. M.J. Martínez); Canarias, H. Universitario (Dr. M.T. Hernández); Palma de Mallorca, H. Son Dureta (Dr. A. Novo); Cádiz, H. Puerta del Mar (Dr. J. Capote); Cáceres, H. San Pedro de Alcántara (Dr. M.L. Amigo); Valencia, H. Dr. Peset (Dr. M.J. Sayas); Granada, H. Virgen de la Nieves (Dr. M. Jurado).
References
- . Early infectious complications in autologous bone marrow transplantation: a review of 219 patients. Bone Marrow Transplant. 1996;18:265–271
- . Infectious complications during neutropenia subsequent to peripheral blood stem cell transplantation. Bone Marrow Transplant. 1997;19:143–147
- Treatment-related mortality in 1000 consecutive patients receiving high-dose chemotherapy and peripheral blood progenitor cell transplantation in community cancer centers. Bone Marrow Transplant. 1997;19:671–678
- Infectious complications in 126 patients treated with high-dose chemotherapy and autologous peripheral blood stem cell transplantation. Bone Marrow Transplant. 1999;23:27–33
- . Infectious complications after high-dose chemotherapy and autologous stem cell transplantation: comparison between patients with lymphoma or multiple myeloma and patients with solid tumors. Bone Marrow Transplant. 2001;27:525–529
- . Infectious complication in 314 patients after high-dose therapy and autologous hematopoietic stem cell transplantation: risk factors analysis and outcome. Infection. 2007;35:421–427
- Empirical antimicrobial monotherapy in patients after high-dose chemotherapy and autologous stem cell transplantation: a randomised, multicenter trial. Br J Haematol. 2005;130:265–270
- Analysis of the value of empiric vancomycin administration in febrile neutropenia occurring after autologous peripheral blood stem cell transplantation. Bone Marrow Transplant. 1998;21:923–926
- . Vancomycin as part of initial empirical antibiotic therapy for febrile neutropenia in patients with cancer: pros and cons. Clin Infect Dis. 1999;29:503–507
- . A prospective, randomized, double-blinded, placebo-controlled trial of empirical teicoplanin in febrile neutropenia with persistent fever after imipenem monotherapy. J Antimicrob Chemother. 2000;45:843–849
- . From the Immunocompromised Host Society. The design, analysis, and reporting of clinical trials on the empirical antibiotic management of the neutropenic patient. Report of a consensus panel. J Infect Dis. 1990;161:397–401
- . Evaluation of new anti-infective drugs for the treatment of febrile episodes in neutropenic patients. Infectious Diseases Society of America and the Food and Drug Administration. Clin Infect Dis. 1992;15(Suppl. 1):206–215
- 1997 guidelines for the use of antimicrobial agents in neutropaenic patients with unexplained fever. Clin Infect Dis. 1997;25:573
- National Committee for Clinical Laboratory Standard. Performance Standard for Antimicrobial Susceptibility Testing. Villanova, PA: NCCLS; 1994;
- . Handbook for Reporting Results of Cancer Treatment. Geneva, Switzerland: World Health Organization; 1979;
- . BMDP Statistical Analysis Software. Berkeley, CA: University of California; 1994;
- . Emergence of high rates of antimicrobial resistance among viridans group streptococci in the United States. Amtimicrob Agents Chemother. 1996;18:25–31
- . Septicemia and shock syndrome due to viridans streptococci: a case-control study of predisposing factors. Clin Infect Dis. 1992;14:1202–1207
- . Infections caused by viridans streptococci in patients with neutropenia. Clin Infect Dis. 2002;34:1524–1549
- Epidemiology and risk factors for Gram-positive coccal infections in neutropenia. Toward a more targeted antibiotic strategy. Clin Infect Dis. 2003;36:149–158
- . How should imipenem-cilastatin be used in the treatment of fever and infection in neutropenic cancer patients?. Cancer. 1998;82:2449–2458
- Comparative review of the carbapenems. Drugs. 2007;67:1027–1052
- A multicenter, double-blind, placebo-controlled trial comparing piperacillin-tazobactam with and without amikacin as empiric therapy for febrile neutropenia. Clin Infect Dis. 2001;33:1295–1301
- . Meropenem versus ceftazidime in the treatment of cancer patients with febrile neutropenia: a randomized double-blind trial. J Clin Oncol. 2000;18:3690–3698
- Piperacillin-tazobactam monotherapy in high-risk febrile and neutropaenic cancer patients. Clin Microbiol Infect. 2006;12:212–216
Financial disclosure: See Acknowledgments on page 515.
PII: S1083-8791(08)01137-3
doi:10.1016/j.bbmt.2008.12.505
© 2009 American Society for Blood and Marrow Transplantation. Published by Elsevier Inc. All rights reserved.
Volume 15, Issue 4 , Pages 512-516, April 2009
