39°C and those
patients who develop any of the conditions that would have placed them in the high-risk category
mentioned above.
a Hacettepe University School of Medicine, Department of Medicine, Section of Infectious Diseases, Ankara 06100, Turkey b University of Arkansas for Medical Sciences, Arkansas Cancer Research Center, 4301 West Markham MS 776, Little Rock, AR 72205, USA
Why outpatient therapy for fever during neutropenia?
The dramatic increase in the cost of healthcare has raised the pressure on providers to deliver more cost-effective care. The USA is currently spending more than US $1 trillion annually for healthcare, and hospital stay accounts for a sizable amount of the total cost. In an effort to reduce health expenditures, the USA government has established reimbursement policies that favour outpatient management of patients who were traditionally cared for in the hospital setting. Outpatient therapy has proved to be feasible and successful in a variety of circumstances, including the practice of infectious disease. This approach has led to the development of such concepts as `sequential therapy's and `home intravenous therapy's in some diseases such as osteomyelitis, infective endocarditis, urosepsis and others. 1 ,2 ,3 ,4
Intravenous antibiotic therapy given in the hospital setting has been the standard approach for the management of fever and neutropenia. This strategy resulted from the well-known mortality and morbidity of infection in these patients. 5,6,7 This approach, however, is based on the erroneous assumption that patients with fever during neutropenia constitute a homogenous group with a similar risk for death and serious complications. In addition, inpatient therapy is likely to be associated with superinfection with nosocomial resistant microorganisms, discomfort to patient and family and a high economic burden. These factors, together with the high cost of treating cancer patients, should prompt a full-scale evaluation of outpatient therapy of fever during neutropenia.
Who is a candidate for outpatient therapy? The risk-adjusted approach
A number of factors have now been defined that can help identify those neutropenic patients at low risk of complication from feverish episodes and thus potential candidates for outpatient therapy.
In a study by Talcott et al.,8 261 episodes of fever during neutropenia were retrospectively evaluated. The patients were divided into four groups at presentation according to the presence of major risk factors, including uncontrolled malignancy, the presence of comorbid conditions (such as hypotension, altered mental status, respiratory failure and gross bleeding), the need for parenteral fluid therapy and pain control, and others. Group 1 (101 of 261 patients; 39%) consisted of patients cared for in the hospital setting when fever developed; group 2 were outpatients (22 of 261; 8%) with concurrent morbidity; group 3 were outpatients (26 of 261; 10%) without concurrent morbidity but with uncontrolled cancer; and group 4 were outpatients (112 of 261; 43%) with none of the risk factors of groups 2 or 3. Serious complications developed in 34%, 55%, 31% and 2% of patients, respectively. None of the patients in group 4 died as compared with a mortality rate of 14% to 23% in the other three groups. This risk-assessment model was subsequently validated in a prospective study in two centres; 9 again, serious medical complications were observed more frequently in the high-risk patients (groups 1-3; 34%) than in the low-risk group (group 4; 5%) (P < 0.000001).
Delayed bone marrow recovery and evidence of deep organ infection were additional risk factors for complications in paediatric patients with fever and neutropenia. 10,11,12,13 Mullen & Buchanan 10 demonstrated that monocytosis preceded recovery from neutropenia by a mean of 4.3 days. Griffin & Buchanan11 investigated this phenomenon further and reported that the most sensitive indicator of bone marrow recovery was the absolute phagocyte count (monocytes, neutrophils and band forms). In a similar study by Rackoff et al.,14 three groups of patients at risk for bacteraemia (and thus at risk for complications) were identified. The criteria used for identifying these patients were the absolute monocyte count and body temperature. The risk of bacteraemia was low when the monocyte count was >0.1 x 109/L (regardless of the temperature); intermediate when the monocyte count was <0.1 x 10 9/L but the temperature was <39°C; and high when the monocyte count was <0.1 x 109/L and the temperature >39°C.
Is outpatient therapy for febrile neutropenia feasible and safe?
A large number of trials have suggested that early patient discharge or outpatient therapy for the whole febrile episode are feasible in low-risk patients. 10,11,12,15,16,17,18 ,19,20,21,22,23,24,25 None of these trials, however, directly compared outpatient therapy with the standard inpatient approach.
More solid evidence came from two controlled trials addressing this issue. The first prospective randomized trial to compare outpatient therapy of febrile neutropenia with the standard of care in the inpatient setting26 was conducted in Pakistan, where patients are often too poor to be admitted to hospital for treatment. In this study, the investigators randomized 182 episodes of fever and neutropenia (in 135 low-risk patients) to receive oral ofloxacin either at home or as an inpatient. One-third of the patients had underlying leukaemia or lymphoma, and half were profoundly neutropenic (0.1 x 109/L) at randomization. Overall success was comparable in the two arms (98% and 96% for the inpatient and the outpatient groups, respectively). Twenty-one percent of the patients who had initially been managed as outpatients were later hospitalized because of the persistence of fever. The mortality rate, however, was low in both treatment groups (4% and 2% respectively), and toxicity was negligible. While these results were encouraging, the study suffered two limitations: the outpatient regimen used was not compared with the standard inpatient regimen and a prospective evaluation of the cost-effectiveness of this approach was not conducted.
In a recent prospective randomized study conducted in Greece,27 183 adult low-risk patients with fever and neutropenia were assigned to receive amikacin plus ceftazidime intravenously in the hospital setting or oral ampicillin/ sulbactam and ciprofloxacin on an outpatient basis. The response rates were similar (92% and 91%, respectively), and no deaths occurred. Significant cost savings favouring outpatient therapy (US $1300 versus US $1750 per episode) were noted even when considering the costs incurred when patients who had been assigned to the outpatient arm needed to be admitted for the management of persistent fever.
When and how to deliver outpatient therapy
The selection of patients for outpatient therapy of febrile neutropenia should be based on two considerations, namely, medical and social. Medically, patients should be in the low-risk category mentioned above and should not have any of the following: defined infection, uncontrolled cancer, fevers >39°C in the presence of a monocyte count <0.1 x 109/L, or any medical condition that would itself necessitate admission. Personal and social conditions should also be taken into consideration because they may preclude outpatient therapy even in the low-risk patient. The patient must be mentally competent and have a level of education that would allow full compliance with the requirement of outpatient therapy. In addition, the patient should live within 1 or 2 h travelling time from the hospital and have a 24 h companion, easy access to a telephone and transportation, and home conditions that allow safe administration of iv antibiotics (clean areas, electricity, refrigeration, etc.). These selection criteria for outpatient therapy may, however, be valid only for western countries: in others, medicosocial conditions may dictate alternative criteria.
After a baseline evaluation that includes history, physical examination and an assessment of the risk status and socio-economic condition of the patient, the patient should be observed in the ambulatory setting for a few hours. During this observation period, the clinical condition of the patient should be closely monitored (heart rate and blood pressure with orthostatic measurements, respiratory rate, pulse oximetry, etc.). Clinical and laboratory findings that may preclude outpatient therapy are also sought during this period. If the patient remains clinically stable and fulfils all the medical and social criteria for outpatient therapy, the patient may then be discharged with specific instructions about administering antibiotics and returning for evaluation. Daily evaluation can then be performed in an outpatient setting until the patient is completely afebrile and asymptomatic or a decision is made to hospitalize. This evaluation can be performed by the treating physician (at a house or clinic visit) or by experienced (often privately financed) outpatient therapy teams. These teams may also deliver iv antibiotics, and perform and monitor laboratory investigations, in collaboration with the treating physician.
Considerations for future studies
Substantial data have now accumulated on the feasibility of an effective and safe approach to
outpatient therapy in the low-risk febrile neutropenic patient. However, several issues still need
to be addressed. The duration of observation before discharge, the frequency of visits to the
healthcare facility and the frequency and type of laboratory studies all need to be examined.
Furthermore, the conditions that should trigger hospital admission must be defined carefully and
validated prospectively. At present, it is prudent to hospitalize those patients whose fever persists
and remains unexplained despite careful evaluation or increases to 39°C and those
patients who develop any of the conditions that would have placed them in the high-risk category
mentioned above.
The decision on the type of the antibiotic(s) used for treatment should be based on several factors, namely the local flora and susceptibility patterns of the organisms, the bactericidal activity of the agent and favourable pharmacokinetics that allow once- or twice-daily administration. Fluoroquinolones, glycopeptides, aminoglycosides, ceftriaxone and cefepime could thus be potential candidates for use in outpatient therapy. For an intravenous antibiotic, stability and compatibility particularly with other antimicrobial agents is an important matter. On the other hand, oral intake and gastrointestinal integrity of the patient must be adequate if the treatment regimen is to be administered orally. In general, when possible an oral regimen is preferred and usually consists of clindamycin, (flu)cloxacillin, erythromycin or co-amoxiclav plus a broad-spectrum quinolone, especially the newer agents with improved Gram-positive activity (e.g. levofloxacin or trovafloxacin). When intravenous antibiotics are needed, a glycopeptide (vancomycin or teicoplanin) is usually combined with a quinolone or a broad-spectrum ß-lactam antibiotic (a third-generation cephalosporin, antipseudomonal penicillin, carbapenem or aztreonam). In the absence of renal dysfunction, a once-daily regimen of an aminoglycoside and ceftriaxone often represents the most attractive option for intravenous therapy.
The evaluation of the role (if any) and cost of colony-stimulating factors should be addressed given the lack of evidence that such factors are beneficial when given after the onset of febrile neutropenia.28
Outpatient therapy for viral and fungal infections is another interesting approach that is currently being undertaken, although no controlled trials have been performed. Azoles or amphotericin B may be continued on an outpatient basis after initial inpatient course and the antiviral agents acyclovir, valacyclovir or famciclovir may provide effective outpatient therapy for low-risk varicella-zoster virus infections.
The above-mentioned issues should be assessed with additional prospective, randomized trials of adequate sample size. Haematological malignancies and solid tumours should be evaluated separately, because patients with solid tumours tend to have a shorter duration of neutropenia and, hence, fewer complications. The endpoints of such trials should not be simply the response to a given antimicrobial regimen. Rather, success should be judged by the ability of treating the patient in an outpatient setting regardless of the need for modification in the therapeutic regimen. In other words, the primary endpoints should be survival, hospital admission and total cost of therapy, whereas secondary endpoints might include the need for sequential therapy, colonization by resistant organisms and the acquisition of nosocomial infections. Finally, cost analysis should be an integral part of such a trial. In this context, the method of administration (home nursing, self-administration, or treatment at infusion centres), the availability of reimbursement, the cost of admission if outpatient therapy fails, the cost-effectiveness of colony-stimulating factors, and the out-of-pocket expenses incurred by the patient are all aspects to be included in this analysis. Given the major differences in reimbursement policies and costs between various countries (and even among different regions of the same country), cost analysis studies cannot be automatically extrapolated from studies conducted in other settings.
In conclusion, existing data suggest that outpatient therapy works well for neutropenic patients with fever provided that patients are selected carefully and followed closely. Additional work is needed to delineate further the most appropriate treatment strategy in suitable patients.
Notes
* Corresponding author. Tel: +1-501-686-8250, ext. 1454; Fax:
+1-501-686-6422. 
References
1 . Daly, J. S., Worthington, M. G., Rzvi, S. A. & Robillard, R. (1989). Brief report: intravenous and sequential intravenous and oral ciprofloxacin in the treatment of severe infections. American Journal of Medicine 87, Suppl. 5A, 232S–4S.[Medline]
2 . Jewesson, P. (1994). Cost-effectiveness and value of an IV switch. Pharmacoeconomics 5, Suppl. 2, 20– 6.[Medline]
3 . Vogel, F. (1995). Sequential therapy in the hospital management of lower respiratory infections. American Journal of Medicine 99, Suppl. 6B, 14S– 9S.[Medline]
4
.
Van den Brande, P., Vondra, V., Vogel, F., Sclaeffer, F.,
Staley, H. & Holmes, C. (1997). Sequential therapy with cefuroxime followed by
cefuroxime axetil in community-acquired pneumonia. Chest 112, 406–15.
5 . Anaissie, E., Fainstein, V., Bodey, G. P., Rolston, K., Elting, L., Kantarjian, H. et al. (1988). Randomized trial of ß -lactam regimens in febrile neutropenic cancer patients. American Journal of Medicine 84, 581–9.[ISI][Medline]
6 . De Pauw, B. E., Raemaekers, J. M., Schattenberg, T. & Donnelly, J. P. (1997). Empirical and subsequent use of antibacterial agents in the febrile neutropenic patient. Journal of Internal Medicine 740, Suppl., 69–77.
7 . Pizzo, P. A. (1993). Management of fever in patients with cancer and treatment-induced neutropenia. New England Journal of Medicine 398, 1323–31.
8 . Talcott, J. A., Finberg, R., Mayer, R. J. & Goldman, L. (1988). The medical course of cancer patients with fever and neutropenia. Clinical identification of a low-risk subgroup at presentation. Archives of Internal Medicine 14, 2561–8.
9 . Talcott, J. A., Siegel, R. D., Finberg, R. & Goldman, L. (1992). Risk assessment in cancer patients with fever and neutropenia: a prospective, two-center validation of a prediction rule. Journal of Clinical Oncology 10, 316–22.[Abstract]
10 . Mullen, C. A. & Buchanan, G. R. (1990). Early hospital discharge of children with cancer treated for fever and neutropenia: identification and management of the low-risk patient. Journal of Clinical Oncology 8, 1998–2004.[Abstract]
11 . Griffin, T. C. & Buchanan, G. R. (1992). Hematologic predictors of bone marrow recovery in neutropenic patients hospitalized for fever: implications for discontinuation of antibiotics and early discharge from the hospital. Journal of Pediatrics 121, 28–33.[ISI][Medline]
12 . Bash, R. O., Katz, J. A., Cash, J. V. & Buchanan, G. R. (1994). Safety and cost effectiveness of early hospital discharge of lower risk children with cancer admitted for fever and neutropenia. Cancer 74, 189–96.[ISI][Medline]
13 . Lucas, K. G., Brown, A. E., Armstrong, D., Chapman, D. & Heller, G. (1996). The identification of febrile, neutropenic children with neoplastic disease at low risk for bacteremia and complications of sepsis. Cancer 77, 791–8.[ISI][Medline]
14 . Rackoff, W. R., Gonin, R., Robinson, C., Kreissman, S. G. & Breitfeld, P. P. (1996). Predicting the risk of bacteremia in children with fever and neutropenia. Journal of Clinical Oncology 14, 919–24.[Abstract]
15 . Aquino, V. M., Buchanan, G. R., Tkaczewski, I. & Mustafa, M. M. (1997). Safety of early hospital discharge of selected febrile children and adolescents with prolonged neutropenia. Medical and Pediatric Oncology 28, 191–5.[ISI][Medline]
16 . Gillis, S., Dann, E. J. & Rund, D. (1996). Selective discharge of patients with acute myeloid leukemia during chemotherapy-induced neutropenia. American Journal of Hematology 51, 26–31.[ISI][Medline]
17 . Kaplan, A. H., Weber, D. J., Davis, L., Israel, F. & Wells, R. J. (1991). Short courses of antibiotics in selected febrile neutropenic patients. American Journal of Medical Sciences 302, 353–4.[ISI]
18 . Kaplinsky, C., Druckner, M., Goshen, J., Tamary, H., Cohen, I. J. & Zaizov, R. (1994). Ambulatory treatment with ceftriaxone in febrile neutropenic children. Israel Journal of Medical Sciences 30, 649–51.[ISI][Medline]
19 . Lau, R. C., Doyle, J. J., Freedman, M. H., King, S. M. & Richardson, S. E. (1994). Early discharge of pediatric febrile neutropenic cancer patients by substitution of oral for intravenous antibiotics. Pediatric Hematology and Oncology 11, 417–21.[ISI][Medline]
20 . Malik, I. A., Khan, W. A., Aziz, Z. & Karim, M. (1994). Self-administered antibiotic therapy for chemotherapy-induced, low-risk febrile neutropenia in patients with non-hematologic malignancies. Clinical Infectious Diseases 19, 522–7.[ISI][Medline]
21 . Mustafa, M. M., Aquino, V. M., Pappo, A., Tkaczewski, I. & Buchanan, G. R. (1996). A pilot study of outpatient management of febrile neutropenic children with cancer at low risk of bacteremia. Journal of Pediatrics 128, 847–9.[ISI][Medline]
22 . Petrilli, A. S., Melaragno, R., Barros, V. T., Silva, A. A. M., Kusano, E., Ribeiro, R. C. et al. (1993). Fever and neutropenia in children with cancer: a therapeutic approach related to the underlying disease. Pediatric Infectious Diseases Journal 12, 916–21.[ISI][Medline]
23 . Preis, S., Jürgens, H., Friedland, C., Oudekotte-David, A. A. & Thomas, L. (1993). Ceftriaxone alone or in combination with teicoplanin in the management of febrile episodes in neutropenic children and adolescents with cancer on an outpatient base. Klinische Padiatriae 205, 295–9.
24 . Rubenstein, E. B., Rolston, K., Benjamin, R. S., Loewy, J., Escalante, C., Manzullo, R. et al. (1993). Outpatient treatment of febrile episodes in low-risk neutropenic patients with cancer. Cancer 71, 3640–6.[ISI][Medline]
25 . Santolaya, M. E., Villarroel, M., Avendano, L. F. & Cofre, J. (1997). Discontinuation of antimicrobial therapy for febrile, neutropenic children with cancer: a prospective study. Clinical Infectious Diseases 25, 92–7.[ISI][Medline]
26 . Malik, I. A., Khan, W. A., Karim, M., Aziz, Z. & Khan, M. A. (1995). Feasibility of outpatient management of fever in cancer patients with low-risk neutropenia: results of a prospective randomized trial. American Journal of Medicine 98, 224–31.[ISI][Medline]
27 . Anaissie, E. J., Samonis, G., Kalbakis, K. & Georgoulias, V. (1997). Therapy for low-risk cancer patients with fever and neutropenia: results of a prospective, randomized trial with cost analysis. In Proceedings of the Thirty-Seventh Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, 1997. Abstract LM-51. American Society for Microbiology, Washington, DC.
28 . Anaissie, E. J., Vartivarian, S., Bodey, G. P., Legrand, C., Kantarjian, H., Abi-Said, D. et al. (1996). Randomized comparison between antibiotics alone and antibiotics plus granulocyte macrophage colony-stimulating factor (Escherichia coli-derived) in cancer patients with fever and neutropenia. American Journal of Medicine 100, 17–23.[ISI][Medline]