Feb 7, 1994 - Universit~itsklinikum Rudolf Virchow,. D-13122 Berlin, Germany. P. Meyer. Medizinische Poliklinik, Universit~itsklinik,. D-97070 Wtirzburg ...
Ann Hematol (1994) 69:231-243
9 Springer-Verlag 1994
H. Link 9 G. Maschmeyer 9 P. Meyer W. Hiddemann 9 W. Stille 9 M. Helmerking D. Adam for the study group of the Paul Ehrlich Society for Chemotherapy
Interventional antimicrobial therapy in febrile neutropenic patients
Received 7 February 1994 / Accepted 28 July 1994
Summary In this prospective multicenter trial, treatment strategies for 1573 patients with neutropenia < 1000hxl and fever _>38.5~ after cytotoxic chemotherapy were compared. Patients with unexplained fever were randomized to a three-phase sequential study for different established drug regimens. If an infection could be defined microbiologically or clinically, treatment modifications were determined. In phase I, treatment for all patients consisted of acylaminopenicillin (PEN) plus aminoglycoside (AMG); or third-generation cephalosporin (CEPH) plus AMG; or PEN plus CEPH. In 800 patients with unexplained fever the response rates were: PEN/AMG (n = 258): 74.4%, CEPH/ AMG (n=252): 73.4%; PEN/CEPH (n=290): 70.0%. Total response rate was 72.5%. In phase II, patients not H. Link (P~) Department of Hematology and Oncology, Medizinische Hochschule Hannover, D-30625 Hanuover, Germany G. Maschmeyer Department of Medical Oncologyand Applied Molecular Biology, Robert-ROssle-Hospital, Universit~itsklinikumRudolf Virchow, D-13122 Berlin, Germany P. Meyer Medizinische Poliklinik, Universit~itsklinik, D-97070 Wtirzburg, Germany W. Hiddemann Department of Hematology and Oncology, Medizinische Universit~itsklinik, D-37075 G6ttingen, Germany W. Stille Center for Internal Medicine, Universit~itsklinik, D-60590 Frankfurt/Main,Germany M. Helmerking Paul Ehrlich Gesellschaftffir Chemotherapy (PEG), Studienzentrale, D-81701 Mtinchen, Germany D. Adam Childrens Hospital, Universit~it Mtinchen, D-80337 Mtinchen
responding after 3 days received PEN/CEPH/vancomycin (n = 70) or PEN/CEPH/AMG (n = 74). The respective response rates were 52.9% and 55.4%, total 54.2%. If fever did not resolve, the patients received either PEN/CEPH (n=40) or imipenem/cilastatin (n=59) both in combination with amphotericin-B/5-flucytosin/ rifampin. The response rates were 62.5% and 79.7%, respectively (p =0.07), total 72.7%. No significant differences between the treatment modalities compared were found. Analyzing all three phases together, 91.3% of patients with unexplained fever were cured. The response rate was also analyzed according to patients with gram-positive bacteremia (n = 183), response rate = 82.5%; gram-negative organisms (n=145) 78.6%; fungemia (n=51) 43.1% (p38.5~C which was not accompanied by either clinical or microbiological evidence of infection or associated with drug administration or transfusion of blood products was defined as unexplained fever.
233
Microbiologically defined infection
was 21 days. Responders to the phase III regimen were maintained on treatment as described above.
Bacteremia and~orfungemia (B/F). This was defined as fever with blood cultures positive for bacteria and/or fungi, but no other clinical signs of infection. For coagulase-negative staphylococci or corynebacteria, this was accepted only where there were at least two separate positive blood cultures.
Clinically and microbiologically defined infection (CMDI). This was defined as fever related to a clinically evident source of infection and a microbiologically defined site of infection (e.g., pneumonia or other source) with or without concomitant bacteremia. Clinically defined infection (CDI) This was defined as fever accompanied by clinical symptoms such as lung infiltrates, skin or soft tissue inflammation, signs of enteritis, perianal infection or urinary tract infection and, in rare cases, other local symptoms such as maxillary sinusitis.
Response criteria In the case of unexplained fever, response was defined as defervescense with no requirement for further antimicrobial treatment for at least 7 days after the termination of treatment. In CDI, resolution of the respective clinical signs was also required, without a recurrence of the infection for at least 1 week after the initial antibacterial regimen had been stopped. The causative pathogen also had to be eliminated. Patients fulfilling response criteria on days 4-6 of the respective treatment phase, but subsequently relapsing with fever and/or clinical symptoms under continuing antimicrobial therapy or within 2 days after cessation of treatment, were regarded as secondary nonresponders. The addition of any antibacterial, a change of the initial therapy, or progress to a subsequent treatment phase, was recorded as nonresponse [15]. Patients with initial response and relapse of fever within 7 days were assessed as nonresponders. Patients who did not respond during any of the study phases or who died as a result of infection were recorded as failures.
Study design This was a prospective randomized study, consisting of a threephase sequential antimicrobial treatment regimen which was adapted to clinical features and previous treatment response. Patients were classified accordingly and are reported in the respective sections of this paper.
Unexplained fever In phase I, all patients were randomized to one of three treatment groups acylaminopenicillin (PEN) plus aminoglycoside (AMG); third-generation cephalosporin (CEPH) plus AMG; or PEN plus CEPH. Initial response was evaluated on days 4-6 of the study (Fig. 1). Among responders, the applied regimen was maintained for at least 2 days post defervescence in patients with increasing neutrophil counts > 1000/p~l or for 7 days in patients with persistent neutropenia (_38.5~ granulocytes < lO00/pl Randomized comparisons Phase I Penicillin Aminoglycoside
Cephalosporin Aminoglycoside
Penicillin Cephalosporin
No response: persistence of fever >_38.5 ~ : day 4 - 6 Phase II Penicillin Cephalosporin Aminoglycoside
Penicillin Cephalosporin Vancomycin
No response: persistence of fever > 38.5 ~ : day 8 -11 Phase III Penicillin, Cephalosporin Rifampin Amphotericin-B, 5-Flucytosine
Imipenem/Cilastatin Rifampin Amphotericin-B, 5-Flucytosine
Defined infections Fever > 38.5~
granulocytes < 1000/pl
Randomized comparisons and treatment strategy Phase I Penicillin Aminoglycoside
Cephalosporin Aminoglycoside
Penicillin Cephalosporin
Phase Ih modification
Pneumonia +Rifampicin
Skin~venous access: + Vancomycin Abdomen, perianal: + Metronidazole
Microorganism Modification ?
No response: persistence of fever _>38.5 ~ : after 4 - 6 days Phase IIh Modification and addition of antifungal therapy Imipenem, Cilastatin Rifampin Amphotericin-B, 5-Flucytosine
patterns of documented micro-organisms and detailed base line data, was completed for each patient. The trend in white blood cell counts (WBC) was measured by comparing the mean WBC on study day 1 or 2 with the mean WBC at the time of treatment termination for each individual patient. Falling WBC were defined as a decrease to below 80% of initial WBC, stable WBC as deviations of less than 20%, and increasing WBC as an increase of more than 20% above the initial WBC, at least to a level of 500/pJ. All patients treated for at least 48 h after study entry were evaluable for response. Patients who died due to infection within 48 h after study entry were included in the analysis. Evaluation For evaluation of response, the patients' definitive allocation to the unexplained fever, microbiologically defined infection, or clinically defined infection group was made according to their state at the end of the study, irrespective of their classification at study entry. Additionally, clinical features and outcome in patients with delayed clinical or microbiological definition of infection (i.e., days 6-21 of the study) were analyzed in comparison to these of patients with earlier documentation (i.e., days 1-5). Patients whose response was not evaluable due to protocol violation, transfer to another hospital unit, etc., were excluded from
the evaluation of complete response rates. Differences between the three treatment groups in phase I (PEN/AMG vs CEPH/ AMG vs PEN/CEPH) and between different antimicrobial substances within these groups were analyzed by the two-tailed Fisher's exact test (complete response vs. nonresponse/death, alpha=0.05). Comparison between the phase-II combinations (PEN/CEPH/VAN vs PEN/CEPH/AMG) stratified according to the three phase-I treatment groups were analyzed using a threedimensional contingency table with a log-linear model (alpha=0.05). Analysis of phase-III results was done in the same way (PEN/CEPH/RIFA/AM-B/5-FC vs IMI/RIFA/AM-B/5-FC). The impact of potential risk factors was analyzed by multivariate analysis. Two dimensional contingency tables were analyzed by Fisher's exact test. In the case of multiple testing, the significance level (alpha=0.05) was secured using the Bonferroni-Holm correction.
Results
F r o m D e c e m b e r 1985 to D e c e m b e r 1991, 1770 p a t i e n t s f r o m 23 p a r t i c i p a t i n g i n s t i t u t i o n s were e n t e r e d into the study. T h e d e m o g r a p h i c data are s u m m a r i z e d in T a b l e
235 1. There was no statistically significant difference between the various randomization subgroups concerning age, diagnosis, stage of disease, leukocyte counts, Karnofsky score, intensive chemotherapy, or preceding radiotherapy (Wilcoxon test). Of 197 patients (11.1%) who were not evaluable for response, 99 cases were due to protocol violation, e.g., noncompliance with entry criteria or randomization error, 53 due to adverse reactions, and 45 due to death from noninfectious causes within 24 h after study entry. Nonevaluable patients were assessable for infections and were distributed equally among the treatment groups, as well as between different participating institutions. Of the remaining 1573 patients evaluablc for response, 800 (50.9%) had unexplained fever. Of the 773 (49.1%) with defined infections, 269 (17.1%) had lung infiltrates (LI), 222 (14.1%) had bacteremia and/or fungemia, 198 (12.6%) had other clinically defined infections (CDI), and 84 (5.3%) had other clinically and microbiologically defined infections (CMDI).
Response rates Of the total of 1573 patients, 1313 (83.5%) responded to therapy, 105 (6.7%) were nonresponders, and 155 (9.9%) died during the study. No statistically significant difference between response, nonresponse, and early death rates was observed among participating institutions.
Unexplained .fever The treatment results of patients with unexplained fever are listed in Table 2. The overall response rate after three phases of sequential antimicrobial therapy was 91.3%. Analysis of the three treatment phases separately revealed response rates of 72.5% in phase I, 54.2% in phase II, and 72.7% in phase III. No significant differences between the randomly compared antimicrobial regimens were observed at any of the three treatment phases. Among the various antibiotics within the respective groups, i.e., different acylaminopenicillins (PEN), third-generation cephalosporins (CEPH), and Table
Defined infections A total of 773 patients (49.1%) had defined infections (DI). Their response rates as compared with those of patients with unexplained fever are shown in Table 3a. The microbiological findings in patients with positive blood cultures are shown in Table 4. Treatment results of all microbiologically defined infections related to the causative pathogens are given in Table 5. The response rates for the different initial treatment groups are shown in Table 3b; the data of patients with bacteremia and clinically defined infections are listed. The response rates in patients with gram-negative bacteremia were higher with PEN/AMG or CEPH/AMG than with PEN/CEPH. However, this difference was not significant (p = 0.22, Fisher's exact test). The grouping of the defined infections (excepting bacteremia, fungemia) into different initial treatment arms resulted in small numbers of patients. Thus, a meaningful analysis was not possible. The response rate for unexplained fever was significantly higher than for patients with any of the DI subtypes. The response rate in lung infiltrates (LI) was significantly lower compared with other DI (p 0.33)
Lung infiltrates
fined LI, fungi were most common (42.7%), followed by gram-negative (31.3%) and gram-positive pathogens (22.3%), while Pneumocystis carinii was detected in four cases (3.9%). The response rate in Li with fungal involvement (46.3%) was significantly inferior to that of LI from bacterial origin (76.3%, p =0.01). A m o n g 44 documented fungal pneumonias, there were 19 cases of aspergillosis; seven of these patients were nonresponders and six of them died during the study. Further details regarding patients with lung infiltrates have been published recently elsewhere [32].
In patients with LI, response rates among the three initial treatment groups in phase I: P E N / A M G ( n = 6 8 ) : 55.9%; vs C E P H / A M G (n =69): 62.2% vs P E N / C E P H (n=72): 63.9%) were not significantly different (p = 0.78). For 60 patients the initial treatment regimen was modified, resulting in a response rate of 60%. For 67 patients, rifampin could not be administered as scheduled, mostly due to elevated liver enzymes. However, they were assessed and evaluated as study patients. Their response rate (58.2%) did not differ from that of patients who received rifampin as scheduled (response rate, 62.4%, p =0.57). In 123 patients, LI were only clinically defined and in 79 patients they were also microbiologically defined. In 67 patients, LI was associated with another defined infection. The response rates among these three subgroups were not significantly different (63.4% vs 58.2% vs 60.8%, p = 0.77). In patients with microbiologically de-
Clinically defined infections In 198 patients with CDI other than LI, i.e., skin or venous access infection (n =113), abdominal or perianal infection (n =57), or other CDI (n =28), no significant difference was detected between response rates (see Table 3a and 3b).
237 Table 3a Response to therapy among patients with unexplained fever or defined infections (NR nonresponse, UF unexplained fever, LI lung infiltrate, B/F bacteremia and/or fungemia, DI deType of infection
n
Unexplained fever Defined infection (total) Lung infiltrate (total) - clinically defined - microbiologically defined - plus other infection Other CDI (total) - venous access infection - abdominal or perianal infection - other Other CMDI (total) - venous access infection - abdominal or perianal infection - urinary tract infection - other Bacteremia/fungemia (total) - single organism - polymicrobial Total
800 773 269 123 79 67 198 113 57 28 84 26 15 18 25 222 197 25 1573
fined infection, CDI clinically defined infection, CMDI clinically and microbiologically defined infection) Response
NR
Death
n (%)
n (%)
n
21 84 46 21 12 13 19 9 6 4 8 2 2 2 2 11 9 2 105
49 106 58 24 19 15 11 5 4 2 7 2 2 1 2 30 24 6 155
730 (91.3) 583 (75.4) 165 (61.3) 78 (63.4) 48 (60.8) 39 (58.2) 168 (84.8) 99 (87.6) 47 (82.5) 22 (78.6) 69 (82.1) 22 (84.6) 11 (73.3) 15 (83.3) 21 (84.0) 181 (81.5) 164 (83.2) 17 (68.0) 1313 (83.5)
(2.6) (10.9) (17.1) (17.1) (15.2) (19.4) (9.6) (8.0) (10.5) (14.3) (9.5) (7.7) (13.3) (11.1) (8.0) (5.0) (4.6) (8.0) (6.7)
(%) (6.1) (13.7) (21.6) (19.5) (24.1) (22.4) (5.6) (4.4) (7.0) (7.1) (8.3) (7.7) (13.3) (5.6) (8.0) (13.5) (12.2) (24.0) (9.9)
Comparison of CR rates (Fisher's exact test): UF UF UF BF
vs DI p
