(2005)日本版FN的临床治疗

International Journal of Antimicrobial Agents26S(2005)S123–S127Clinical guidelines for the management of neutropenic patients with unexplained fever in Japan:validation by theJapan Febrile Neutropenia Study GroupKazuo Tamura∗The First Department of Internal Medicine,School of Medicine,Fukuoka University,Fukuoka,JapanAbstractThe Japan Febrile Neutropenia Study Group(JFNSG)Trial was a multicenter,open,randomized study designed to validate thefirst Japanese guidelines for the management of neutropenic cancer patients with unexplained fever issued in1998.The trial compared cefepime monotherapy with cefepime plus amikacin combination therapy in febrile neutropenic patients with hematological disorders.The JFNSG found that monotherapy with cefepime was,in general,as effective as combination therapy.In terms of subset analyses,defervescence appeared to occur more frequently in leukemic patients and in those with profound neutropenia treated with the dual combination.The conclusion of the trial was that the1998guidelines were applicable to the Japanese febrile neutropenic patient population.The JFNSG met again in2003to revise these guidelines.An important addition to the guidelines was a distinction between low-and high-risk patients.Low-risk febrile neutropenic patients can receive oral ciprofloxacin or levofloxacin,with or without amoxicillin/clavulanic acid,on an outpatient basis,or intravenous (i.v.)monotherapy with cefepime,ceftazidime or a carbapenem.High-risk patients can receive i.v.cefepime,ceftazidime or a carbapenem, or an i.v.dual combination with cefepime,ceftazidime or a carbapenem plus an aminoglycoside.Those patients with a documented infection with methicillin-resistant Staphylococcus aureus should also receive a glycopeptide.It remains to be determined whether existing assessment scoring systems apply to Japanese patients;whether a broad-spectrum cephalosporin plus an aminoglycoside combination is required as the initial management of patients with acute leukemia and/or profound neutropenia;which antibacterial drugs should be used whenfirst-and second-line agents fail;what are the appropriate oral agents and dosing regimens for low-risk patients;whether serology or the polymerase chain reaction should be the preferred marker for initiating pre-emptive antifungal therapy;and whether the azoles or the candins should be the preferred antifungal agents.©2005Elsevier B.V.and the International Society of Chemotherapy.All rights reserved.Keywords:Monotherapy;Combination therapy;Febrile neutropenia;Risk stratification1.IntroductionGuidelines for the management of neutropenic cancer patients with fever of unknown origin werefirst issued in Japan in1998[1].Two separate study groups were sub-sequently formed to validate these guidelines,the Kyushu Hematology Organization for Treatment(K-HOT)Study Group,involving17institutions[2],and the Japan Febrile Neutropenia Study Group(JFNSG)[3],involving30institu-tions.∗Fax:+81928628200.E-mail address:ktamura@fukuoka-u.ac.jp.2.The JFNSG Trial2.1.Background dataThe JFNSG Trial was a multicenter,open,randomizedstudy designed to compare cefepime monotherapy withcombination therapy of cefepime plus amikacin in febrileneutropenic patients with hematological disorders[3].Febrile neutropenia was defined by an axillary temperature ≥37.5◦C,an absolute neutrophil count(ANC)<1000/␮L and increasing levels of C-reactive protein(CRP)in patientswith no other identifiable cause of fever.Patients with blooddisorders meeting these criteria were randomized to receiveempirical treatment with either intravenous(i.v.)cefepime0924-8579/$–see front matter©2005Elsevier B.V.and the International Society of Chemotherapy.All rights reserved. doi:10.1016/j.ijantimicag.2005.08.001S124K.Tamura /International Journal of Antimicrobial Agents 26S (2005)S123–S127Fig.1.Study by the Japan Febrile Neutropenia Study Group (JFNSG)[1,3].CRP,C-reactive protein.monotherapy 1–2g every 12h,or i.v.cefepime plus amikacin combination 100–200mg every 12h,and were re-assessed after 3days.If defervescence occurred at day 4,patients with infections of unknown etiology continued receiving the same treatment for an additional 4days,whereas therapy was adjusted related to the causative microorganism in those with infections of established etiology.The total duration of treatment was 7days.Patients who remained febrile at day 4were re-assessed based on a complete blood count,blood chemistry profile,serology for fungal pathogens,CRP,radiography,etc.(Fig.1)[1,3].If the etiology of infection was established,therapy was adjusted accordingly.Vancomycin therapy was considered for Gram-positive infections.If the etiology of infection remained unknown,patients on monotherapy received an aminoglycoside in addition to cefepime.If afebrile after 48h,the patient continued with the same treatment.If febrile,they received an antifungal agent.Change to another ␤-lactam antimicrobial was considered for patients previously on dual therapy.Fungal serology and/or culture were performed.Amphotericin B was added to patients with positive fungal cultures,amphotericin B or an azole was added to those with positive serology,while an azole was added to those with negative serology.Patients were evaluated at days 3,7,10,14and 30after initiation of empirical therapy.Response to treatment was stratified as:excellent if afebrile within 3days with clini-cal/laboratory improvement;good if afebrile within 7days with clinical/laboratory improvement;fair if fever tended to decrease within 7days with clinical/laboratory improvement;and poor if the initial drug had to be changed within 3days,or if fever had not declined within 7days,with no clini-cal/laboratory improvement.2.2.ResultsOf the 201patients initially recruited,it was possible to evaluate 189(95in the monotherapy arm and 94in the combi-nation therapy arm).Complete defervescence with improved general condition at day 3was obtained in 32.6%of patients in the monotherapy arm compared with 45.7%in the combina-tion therapy arm (not significant).By day 7,64.2%and 76.6%of patients in the cefepime monotherapy 1–2g i.v.every 12h group and the cefepime plus amikacin 100–200mg i.v.every 12h combination group,respectively,had responded (not sig-nificant).The potential impact of three variables,namely body tem-perature,ANC and underlying disease,on treatment response was examined.In general,no significant differences were observed in the response to monotherapy or to combination therapy at day 3and day 7in terms of body temperature below,equal to,or above 38.0◦C.With regard to the ANC at the start of empirical antibiotic treatment,at day 3cefepimeK.Tamura/International Journal of Antimicrobial Agents26S(2005)S123–S127S125alone was equally effective as the combination therapy for patients with ANCs between500/␮L and1000/␮L,whereas the cefepime plus amikacin combination was more effective for those with fewer than500neutrophils/␮L.This suggests that high-risk patients should be further stratified in terms of their ANCs before selecting an empirical treatment approach. Finally,when patients were categorized as leukemic or non-leukemic,the dual combination seemed more effective for leukemic patients.Adverse events were few and generally mild.In the monotherapy group,two patients developed a skin rash and two had liver dysfunction.In the dual therapy arm,one patient each developed a skin rash,liver dysfunction and renal dysfunction.There werefive cases of bacteremia both in the cefepime monotherapy group(Enterobacter cloacae,Staphylococ-cus epidermidis,Pseudomonas aeruginosa,Aeromonas hydrophila and Escherichia coli)and in the combination ther-apy group(P.aeruginosa,E.coli,Stomatococcus mucilagi-nosus and coagulase-negative staphylococci(two patients)). Gram-negative bacteria were more prevalent.One patient in the combination therapy arm developed fungemia by Can-dida inconspicua at day14,and another developed fungemia by an unspecified Candida at day17.2.3.ConclusionsThe JFNSG concluded that:•an axillary temperature≥37.5◦C and an ANC<1000/␮L were appropriate markers for the initiation of antibiotic therapy in febrile neutropenic patients;•patients with ANCs between500/␮L and1000/␮L tend to respond to antibiotics better than other patients;•monotherapy with cefepime is,in general,as effective as combination therapy;and•according to subset analyses,defervescence appears to be attained more frequently with the dual combination in leukemic patients and in those with profound neutropenia, although these two groups did not differ in terms of overall survival or risk for severe complications.Based on these considerations,the JFNSG concluded that the1998febrile neutropenia guidelines were applicable to the Japanese febrile neutropenic patient population.3.The2003revision of the Japanese guidelinesIn2002,the Infectious Diseases Society of America revised the US guidelines for the use of antimicrobial agents in neutropenic patients with cancer[4].Novel characteristics of the revised guidelines included:1.stratification of febrile neutropenic patients into high-versus low-risk subgroups;e of oral antibiotic therapy in carefully selected patientsat low risk of complications;3.an extension from3days to3–5days in the time for eval-uating initial therapy;4.the feasibility of continuing to administer the same initialantimicrobials to stable patients with persistent fever;and 5.the emphasis on an ANC of500/␮L for importantdecision-making in the management of febrile neu-tropenic patients.In2003,the JFNSG met again in Honolulu,Hawaii,to revise the Japanese guidelines[5].The revised Japanese guidelines incorporated a distinction between low-and high-risk patients.Various instruments have been developed to make this distinction,including a scoring system designed to identify low-risk febrile neutropenic cancer patients[4,6]. An important pending task in this respect is the validation of these instruments specifically for the Japanese patient popu-lation.In the revised Japanese guidelines,febrile neutropenic patients at low risk of complications can be given oral ciprofloxacin or levofloxacin,with or without amoxi-cillin/clavulanic acid,on an outpatient basis.Low-risk patients can also be given i.v.monotherapy with cefepime, ceftazidime or a carbapenem.Oral agents used in Japan are different from those used in the Western world.They are usually administered at one-third to one-half the dose used in the USA,e.g.ciprofloxacin200mg every8h.Similarly, i.v.agents,e.g.cefepime,ceftazidime and carbapenems,are administered in Japan at1–2g every12h.Patients at high risk of complications can be given i.v. cefepime,ceftazidime or a carbapenem in a schedule similar to that outlined for low-risk patients,or a dual combination therapy with i.v.cefepime,ceftazidime or a carbapenem plus an aminoglycoside.A glycopeptide such as vancomycin or teicoplanin should be added to the treatment of high-risk patients with documented infection with methicillin-resistant Staphylococcus aureus.Patients should be re-assessed after 3–5days.Those with infections of unknown etiology who become afebrile by days4or5can continue on the initial empirical treatment for at least4additional days.When the etiology of the infection has been established,therapy can be adjusted accordingly.If defervescence fails to occur in4–5days,patients need to be re-assessed,based on a complete blood count,blood chemistry profile,cultures,serology for fungal pathogens, CRP,radiography and other imaging techniques(Fig.2)[5]. Granulocyte colony-stimulating factor or gamma globulin therapy,if not instituted previously,should be considered for patients with persistent fever at days4–5.Patients with infec-tions of unknown etiology who remain stable can continue receiving the initial antibiotic regimen.Those previously on monotherapy who show signs of progressive disease should have an aminoglycoside added to their treatment.A change to alternative monotherapy should be considered, e.g.the initial cephalosporin can be changed to another cephalosporin or a carbapenem,or the initial carbapenem can be changed to a broad-spectrum cephalosporin.ForS126K.Tamura /International Journal of Antimicrobial Agents 26S (2005)S123–S127Fig.2.The 2003Japan Febrile Neutropenia Study Group (JFNSG)new guidelines [5].CBC,complete blood count;CRP,C-reactive protein;G-CSF,granulocyte colony-stimulating factor.patients initially receiving dual therapy,the cephalosporin or carbapenem can be changed as recommended above,whereas the initial aminoglycoside should be changed to another aminoglycoside or to i.v.ciprofloxacin.If the patients are afebrile when re-assessed 48h later,no further changes in therapy are needed.In contrast,a shift to another ␤-lactam antibiotic should be considered for those with persistent fever,as well as for those with signs of progressive disease originally treated with combination therapy.Fungal infection should be evaluated in these patients based on serology and/or culture,and determination of ␤-d -glucan and galac-tomannan antigenemia.If the culture is positive,specific therapy for the fungal pathogen should be initiated.If fungal infection is considered possible or probable,an azole (if not used as prophylaxis),amphotericin B or the echinocandin micafungin (where this product is available)should be added.Patients with infections of established etiology should have their therapy adjusted based on the causative microor-ganism.If this is a Gram-positive bacterium,treatment with a glycopeptide should be considered.4.Issues pending elucidationResearch to be planned for the next few years should address a number of unanswered questions related to the man-agement of infections in febrile neutropenic patients.Some of these are:•Are the scoring systems developed to assess the risk for complications applicable to Japanese patients?•Is a dual combination strategy with a broad-spectrum cephalosporin and an aminoglycoside required as the ini-tial management of patients with acute leukemia and/or profound neutropenia?•Which antibacterial agents should be given to patients who fail to respond to first-and second-line therapies?•Which are the appropriate oral agents and dosage regimens for low-risk patients?•Should serology or the polymerase chain reaction be the preferred marker for initiating pre-emptive antifungal ther-apy?Which should be the preferred antifungals for empir-ical therapy:the azoles or the candins?K.Tamura/International Journal of Antimicrobial Agents26S(2005)S123–S127S127References[1]Masaoka T.Management of fever of unknown origin in theneutropenic patient:the Japanese experience.Int J Hematol 1998;68(Suppl.1):S9–11.[2]Tamura K,Matsuoka H,Tsukada J,et al.Cefepime or carbapenemtreatment for febrile neutropenia as a single agent is as effective asa combination of4th-generation cephalosporin+aminoglycosides:comparative study.Am J Hematol2002;71:248–55.[3]Tamura K,Imajo K,Akiyama N,et al.Randomized trial of cefepimemonotherapy or cefepime in combination with amikacin as empir-ical therapy of febrile neutropenia.Clin Infect Dis2004;39:S15–24.[4]Hughes WT,Armstrong D,Bodey GP,et al.2002guidelines for theuse of antimicrobial agents in neutropenic patients with cancer.Clin Infect Dis2002;34:730–51.[5]Tamura K.Initial empirical antimicrobial therapy:duration and sub-sequent modifications.Clin Infect Dis2004;39:S59–64.[6]Klastersky J,Paesmans M,Rubenstein EB,et al.The MultinationalAssociation for Supportive Care in Cancer risk index:a multinational scoring system for identifying low-risk febrile neutropenic cancer patients.J Clin Oncol2000;18:3038–51.。