Sixty-one febrile episodes among 27 neutropenic patients with malignant hematological diseases were studied during a study period between 1998 and 2001. The results were compared with the previous 3-year period. The main differences in treatment were that we did not use ciprofloxacin prophylaxis routinely and we added granulocyte-macrophage colony-stimulating factor to the antimicrobial treatment regimen. The clinical and microbiological spectra of diseases shifted toward Grampositive bacteremia and lower respiratory infections. The number and rate of infections and the mortality rate did not change significantly.
Febrile neutropenia remains a frequent complication after chemotherapy among patients with malignant hematological diseases. In daily practice, febrile neutropenia represents a significant challenge for hematologists and infectologists.
As mentioned in our previous article,1 the reason for the success achieved in the field of supportive therapy for patients with malignant hematological diseases is that the working groups of different institutions regularly publish the results of microbiological examinations and the effectiveness of the antimicrobial therapy used. Our aim was to survey in our patients the frequency and distribution of infections among febrile neutropenic patients after chemotherapy to estimate the effectiveness of the antimicrobial treatment and prophylactic regimen and the efficacy of the hematopoietic colony-stimulating factor we applied. These results were compared with data we published earlier.1
At the Department of Medicine of the Central Military Hospital of Hungarian Defense Forces, we performed a retrospective study from January 1, 1998, to December 31, 2001. For the study, we selected patients with malignant hematological diseases who developed a fever (twice or more within 24 hours, with an axillary temperature of ≥38°C) within 21 days after chemotherapy and who had an absolute neutrophil granulocyte count of ≤500 cells/mm^sup 3^ during the febrile period. In the inclusion period, a total of 27 patients met the selection criteria (14 women and 13 men; average age, 62.2 years). The distribution of patients according to diagnosis is shown in Table I. By the criteria of the Multinational Association for Supportive Care in Cancer, all of our 27 patients were at high risk.2 We used ciprofloxacin prophylaxis in 14 cases and antifungal prophylaxis in only 2 cases. Table II shows the antimicrobial protocol that was used during the study period for the treatment of febrile neutropenia. During the indicated period, we did not routinely add aminoglycosides to antipseudomonas antibiotics and we did not give ceftriaxone empirically.1
During the study period, we observed a total of 61 febrile neutropenic episodes among 27 patients after chemotherapy (Table III). On average, neutropenia manifested on day 9, whereas the febrile condition manifested on day 12. On average, the febrile episodes lasted 6.6 days. The average number of neutropenic days was 8.3. Neutropenia lasted for 11.8 days in the group of patients with leukemia, whereas neutropenia lasted for only 6.3 days in the non-Hodgkin's lymphoma and multiple myeloma patient groups. Of 61 cases, 30 (49.1%) involved severe neutropenic episodes (absolute neutrophil count 10 days in 10 cases. We lost nine patients during the febrile neutropenia study period, and the cause of death was infection for four patients. Causes of death for the febrile neutropenic patients are listed in Table IV.
We were able to determine the origin of the infection for a large proportion of the cases (Table V). Among the 61 febrile neutropenic episodes, we detected 22 cases of bacteremia (36.1%) and one case of fungemia (1.6%) (Table VI). Of the coagulase-negative Staphylococcus strains (n = 13), four isolates (30.7%) were susceptible to methicillin/oxacillin. The Enterococcus faecalis isolate cultured from the blood was susceptible to ampicillin, the Staphylococcus aureus strain was methicillin/oxacillin resistant, and the Pseudomonas aeruginosa isolates were susceptible to carbapenems and ceftazidime.
Distribution of so-called bacterial/nonbacteremic infections were as follows: one case of a urinary tract infection (E. faecalis), two cases of pneumonia (Acinetobacter Iwojfii and Klebsiella pneumoniae), one case of maxillary sinusitis (S. aureus), one case of antibiotic-associated colitis (Clostridium difficile), and one case of cellulitis (S. aureus). A disseminated infection caused by herpes simplex virus was the reason for one of the febrile conditions. We verified the pulmonary aspergillosis with radiological examinations (chest X-rays and chest computed tomography scans), serological tests, and later pathological examination. In one case, we found P. aeruginosa bacteremia and Candida glabrata fungemia in association with bilateral pleuropneumonia. We present the distribution of clinically defined infections in Table VII. The suspected origins of positive blood cultures are shown in Table VIII.
We used antibiotic prophylaxis with ciprofloxacin during neutropenia in 14 cases. During antibiotic prophylaxis, six bacteremic cases developed (Table IX). We applied antifungal prophylaxis with fluconazole in only two cases. During neutropenia, colony-stimulating factor was administered in 56 cases (in the form of granulocyte-macrophage colony-stimulating factor in 55 cases).
At the time of febrile neutropenic episodes, we found 23 bacteremic cases. The rate of response to empiric antibiotic therapy during febrile neutropenia for the bacteremic patients was 52.2% (12 patients); for 11 patients, modification of therapy was necessary. We lost six patients from this patient group.
The effectiveness of the initial empiric antibiotic therapy was 59.0% for all of the febrile neutropenic episodes. After the first modification, the rate was 80.3%. Initially we administered monotherapy with carbapenem antibiotic in 28 cases (19 cases with imipenem/cilastatin and 9 cases with meropenem), with 58.1% effectiveness. We added vancomycin to the regimen in 19 cases. During empiric antibiotic therapy, we used vancomycin six times initially with broad-spectrum antibiotics. We administered amphotericin B in seven cases, and we lost five patients from this patient group.
Among patients who were receiving chemotherapy for malignant hematological diseases, the degree and duration of neutropenia were correlated with the severity of infection.3,4 Patients with absolute neutrophil counts of
Often fever is the only sign of infection. Approximately onehalf of neutropenic patients with fever have clinically manifest or occult infections. At least one-fifth of patients with neutrophil counts of
During the study period, we observed a significant increase in bacteremia caused by Gram-positive micoorganisms. During the 61 febrile neutropenic episodes, we registered 23 positive blood cultures (37.7%). Eighteen of these bacteremia cases were attributable to Gram-positive pathogens (78.2%). This represents a 22% increase, compared with the earlier study period, whereas the proportions of central line infections (two vs. four cases) and mucositis (three vs. four cases) decreased. This trend is very similar to results of the 11th European Organisation for Research and Treatment of Cancer trial, in which the percentage of Gram-positive bacteremia was 69%.7 Gram-positive pathogens, predominantly coagulase-negative staphylococci and viridans group streptococci, may now account for as many as twothirds of bacteremic episodes among febrile neutropenic cancer patients.8 Concerning the evaluation of Gram-positive bacteremia, we must draw attention to the correct interpretation of the microbiological results; the significant proportions of staphylococci and Corynebacterium spp. reflect contamination and not real pathogens.
We observed one lethal outcome (E. faecalis sepsis) among 18 cases of Gram-positive bacteremia (1 polymicrobial case), whereas 13 bacteremic cases attributable to coagulase-negative staphylococci proved to be benign. According to international practice, if there is no clinical sign of central venous catheterassociated infection, cellulitis, or severe mucositis, we abstain from empiric usage of glycopeptides, although the length of hospital stay still increases.4,6 We must take into consideration the fact that the extensive use of glycopeptides can lead to the risk of emerging vancomycin-resistant bacteremia. '
The 14.7% mortality rate attributable to febrile neutropenia was higher than that in the previous study period (8.4%). The mortality rate in connection with infection was 6.5%, as opposed to the previous period, when the mortality rate in this category was 4.2%.
In the autopsies of four patients who died because of infection, we verified one case of sepsis caused by E. faecalis, one of bilateral pneumonia and pulmonary abscess attributable to P. aeruginosa, one of pulmonary aspergillosis, and one of bilateral pneumonia and Candida sepsis, in accordance with clinical and microbiological data. The reasons for the higher mortality rates in our study were the older age of the patients (62 vs. 47 years) and the more advanced stages of underlying diseases. According to literature on the adult population, the mortality rate for bacteremia with malignant diseases is approximately 18 to 42%.9
Among our patients, P. aeruginosa currently dominate among the Gram-negative agents. In our study, among antipseudomonas jS-lactams, carbapenem (imipenem/cilastatin and meropenem), which we used most commonly, demonstrated resistance for P. aeruginosa on one occasion, whereas all isolates remained sensitive to ceftazidime.
We can observe the worldwide increase in carbapenem resistance of P. aeruginosa. The imipenem resistance among P. aeruginosa isolates from patients in intensive care units in 1999-2000 was 22.5% in the United States and 15.6% in the United Kingdom.10 In our hospital, the proportion of carbapenem-resistant isolates of P. aeruginosa from blood cultures was 18.5%, whereas the rate was only 12% for ceftazidime. For this reason, since 2002 we have used initial empiric therapy with ceftazidime with or without amikacin, according to our antimicrobial guidelines, for neutropenic patients with fever. During the study period, we used a β-lactam (ceftazidime) plus aminoglycoside (amikacin) antibiotic combination in only five cases in our group of patients, in cases of fever related to severe neutropenia. In this group, one patient died. During the treatment of two patients with bloodstream infections caused by P. aeruginosa, we used carbapenem monotherapy. In severe clinical situations, such as cases of bloodstream infections caused by P. aeruginosa, it could be reasonable to add aminoglycosides to β-actam antibiotics with antipseudomonas effect, because of synergistic effects.
In the first study period (1995-1997), we used ciprofloxacin for chemoprophylaxis in 49% of the cases; in the second period, this value was only 23%. Earlier during antibiotic prophylaxis, 12 bacteremic cases developed (34.2%); in this period, 6 cases (42.8%) were observed. Among these bacteremic cases, coagulase-negative staphylococci predominated. Most guidelines recommend that routine antibiotic prophylaxis for afebrile neutropenic patients be avoided, because of the problem of emerging drug-resistant bacteria and fungi attributable to extensive antibiotic use. Therefore, since 2002, with this group of patients we do not use antibiotic prophylaxis. We place great importance on isolation and increased infection control activity.
We start administering fluconazole prophylaxis if we notice multiple colonization. We lost two patients with invasive fungal infections despite early administration of amphotericin B therapy. Later, we do not routinely perform antiviral prophylaxis.
We used mostly colony-stimulating factor (usually granulocyte-macrophage colony-stimulating factor) for our febrile neutropenic patients. In every case, the indication was secondary prophylaxis. We started the administration of colony-stimulating factor if neutrophil counts were
On the basis of recent international guidelines, the administration of colony-stimulating factors could be considered in the following clinical circumstances: in cases of severe neutropenia when a long delay in recovery of the bone marrow is expected and the patient's general condition could worsen, in cases of pneumonia, hypotensive episodes, severe cellulitis or sinusitis, systemic fungal infections, or multiorgan dysfunction secondary to sepsis, and for patients who remain severely neutropenic and have documented infections that do not respond to appropriate antimicrobial therapy.4 Additional studies and observations are necessary to establish the most effective antimicrobial treatment.
1. Rokusz L, Liptay L: Infections of febrile neutropenic patients in malignant hematological diseases. Milit Med 2003; 168: 355-9.
2. Klaslersky J, Paesmans M, Rubenstein EB, et al: The Multinational Association for Supportive Care in Cancer risk index: a multinational scoring system for identifying low-risk febrile neutropenic cancer patients. J Clin Oncol 2000; 18: 3038-51.
3. Bodey GP, Buckley M, Shate YS, Freireich FJ: Quantitative relationship between circulating leukocytes and infection in patients with leukaemia. Ann Intern Med 1966; 64: 328-40.
4. Hughes W, Armstrong D, Bodey GP, et al: 2002 guidelines for the use of antimicrobial agents in neutropenic patients with cancer. Clin Infect Dis 2002; 34: 730-51.
5. Lucas KG, Brown AE, Armstrong D, et al: The identification of febrile, neutropenic children with neoplastic disease at low risk for bacteremia and complication of sepsis. Cancer 1996: 77: 791-8.
6. Hungarian College of Infectology and Hungarian College of Hematology and Transfusiology: Guideline for Prevention and Treatment of Infections in Neutropenic Patients, 2003. Közös szakmai protokoll (in press).
7. Glauser MP, Calandra T: Infections in patients with hematologic malignancies. In: Management of Infections in Immunocompromised Patients, pp 141-88. Edited by Glauser MP, Pizzo PA. London, United Kingdom, WB Saunders, 2000.
8. Garcia-Carbonero R, Paz-Ares L: Antibiotics and growth factors in the management of fever and neutropenia in cancer patients. Curr Opin Hematol 2002; 9: 215-21.
9. Wisplinghoff H, Seifert H, Wenzel R, Edmond MB: Current trends in the epidemiology of nosocomial bloodstream infections in patients with hematological malignancies and solid neoplasms in hospitals in the United States. Clin Infect Dis 2003; 36: 1103-10.
10. Livermore DM: Multiple mechanisms of antimicrobial resistance in Pseudomonas aeruginosa. our worst nightmare? Clin Infect Dis 2002; 34: 634-40.
11. Liptay L, Fürész J, Kádár K, et al: The role of cytokines in the treatment of patients with neutropenic infections after chemotherapy. Infektol Klin Mikrobiol 1999; 6: 41-5.
Guarantor: COL László Rókusz
Contributors: COL László Rókusz; COL László Liptay (Ret.)
First Department of Medicine, Central Military Hospital of the Hungarian Defense Forces, 1553 P.f. 1, Budapest, Hungary.
This manuscript was received for review in January 2004. The revised manuscript was accepted for publication in July 2004.
Reprint & Copyright © by Association of Military Surgeons of U.S., 2005.
Copyright Association of Military Surgeons of the United States Aug 2005
Provided by ProQuest Information and Learning Company. All rights Reserved