- Email: [email protected]
Pseudomonas aeruginosa bacteremia: An analysis of 123 episodes, with particular emphasis on the effect of antibiotic therapy
Original
Report
Pseudomonas aeruginosa Bacteremia: An Analysis of 123 Episodes, with Particular Emphasis on the Effect of Antibiotic Therapy Yardena Siegman-Igra, MD;* Ramit Ravona, MD;* Hedva Primerman, MSc;* and Michael Giladi, MD* ABSTRACT Objectives: To review current experience with Pseudomonas aeruginosa bacteremia and compare outcome of patients treated with single-drug, versus combination therapy. Methods: The charts of all patients with P aeruginosa bacteremia between 1990 and 1992 were reviewed, and pertinent demographic, clinical, and bacteriologic data were retrieved. In addition, similar data were collected from a series of patients with P aeruginosa bacteremia from the literature of the past 20 years. Results: One hundred and twenty-three episodes of F! aeruginosa bacteremia in 121 patients were identified. Most patients were older than 70 years, had at least one underlying condition, and had acquired the infection in the hospital. Attributable mortality was 34%. After exclusion for early mortality and inappropriate therapy, 57 patients remained eligible for comparison of outcome according to therapy protocol. Mortality from infection was equal between the group of 42 patients who received monotherapy and the 15 patients who received combination therapy (14% and 13%, respectively). The literature review revealed eight articles describing 21 to 410 spisodes of Pseudomonas bacteremia. The clinical characteristics of these series did not differ significantly from those of the present series. Conclusions: Incidence, epidemiology, clinical characteristics, and outcome of pseudomonas sepsis did not change significantly over the past 2 decades. Appropriate monotherapy was as effective as combination drug therapy for individuals with pseudomonas bacteremia surviving the first 2 days of infection.
Key Words:
Int J Infect
bacteremia, drug therap)/; nosocomial, Pseudomonas aeruginosa, sepsis Dis
1998;
2:211-215.
*Infectious Disease Unit and Microbiology Laboratory, Medical Center, Sackler Faculty of Medicine, Tel-Aviv Received:
October
17,1997;
Address correspondence Unit, Tel-Aviv Sourasky Israel.
Accepted:
to Dr.Yardena Medical Center,
April
Tel-Aviv Sourasky University, Israel.
9, 1998.
Siegman-Igra,
6 Weizman
Infectious Disease Street, Tel-Aviv 64239
Pseudomonas aeruginosa infections continue to be responsible for considerable morbidity and mortality in compromised and hospitalized patients. In spite of longterm recognition of the syndrome of Pseudomonas bacteremia and availability of effective drugs, the treatment of choice is still controversial, especially as to whether combination therapy is superior to monotherapy. A literature review encompassing the past 20 years revealed much information concerning the epidemiology, demography, underlying conditions, place of acquisition, foci of infection, risk factors for mortality, and outcome associated with I! aeruginosa bacteremia, but did not supply a satisfactory answer to the controversy.1-9 The authors conducted a retrospective analysis of 123 episodes of I? aeruginosa bacteremia over a period of 3 years, reviewed epidemiologic and clinical characteristics in comparison with previous similar series, and compared outcome for monotherapy versus combination drug therapy. MATERIAL
and
METHODS
The Tel-Aviv Sourasky Medical Center is a 950-bed tertiary hospital located in central Tel-Aviv, serving a mostly elderly population. All patients who had I? aeruginosa in blood cultures during the period 1990 to 1992 were included in this analysis. Their medical records were carefully reviewed and pertinent demographic and medical data were retrieved. The following definitions were employed: 1. Neutropenia: Polymorphonuclear count of less than 1000&L. 2. Hospital-acquired bacteremia: Bacteremia that occurred more than 48 hours after admission, or less than 48 hours following an invasive procedure. 3. Death attributed to infection: Death occurring before resolution of the septic manifestations, and in the absence of another obvious direct cause of death. 4. Death not associated with infection: All other deaths occurring during hospitalization. 5. Empiric therapy: Antibiotics given from the onset of clinical signs attributed to bacteremia to the day that
211
212
6.
7.
8.
9.
International
Journal
of Infectious
Diseases
/ Volume 2, Number 4, April-June 1998
antimicrobial susceptibility of the isolate was reported. Definitive therapy: Drugs given after antimicrobial susceptibility was reported. The interval between empiric and definitive therapy ranged from 2 to 4 days. Appropriate antipseudomonal therapy. Monotherapy: Either a fluoroquinolone or a third generation cephalosporin or imipenem-cilastatin, to which the isolate was susceptible in vitro, given either alone or in combination with another drug to which the organism was resistant. The use of either a ureidopenicillin or an aminoglycoside alone was not considered appropriate even if the isolate was susceptible to the drug. Combination drug therapy: The use of one of the following drugs: a fluoroquinolone, a third generation cephalosporin, imipenem-cilastatin, or a ureidopenicillin in combination with an aminoglycoside when the isolate was susceptible to both. No therapy: The use of either no antibiotics at all or the administration of antibiotics for less than 24 hours. This occurred mostly in patients who died in the first 48 hours of bacteremia.
Identification of p aeruginosa and antibiotic susceptibility testing was performed using an automated system (Sensititer, Radiometer, Copenhagen, Denmark). Statistical analysis was performed using the chi-squared or Fisher’s exact test as appropriate. A P value of I 0.05 was considered significant. RESULTS One hundred and twenty-three episodes of p aeruginosa bacteremia in 121 patients were identified during the study period (1990- 1992), representing 1.66 episodes per 1000 admissions. The patients’ ages ranged between 12 and 94 years (mean, 70; median, 72); 79 patients (65%) were 70 years or older; 81 patients (67%) were males.
Sources of Infection in Communityand HospitalTable 1. Acquired Pseudomonas Bacteremia in 120 Episodes” Source
of Infection
Urinary tract Respiratory tract Intra-abdominal Vascular lines Soft tissue Perianal Meningitis Unknown *An additional institutions.
three episodes
Community Acquired n = 24 (700%)
Hospital Acquired n = 96 (100%)
3 (12%) 5 (21%) 1 1
37 17 11 4 4 1 1 21
(4%) (4%)
1 (4%) 13 (54%) were excluded
due to acquisition
(39%) (18%) (11%) (4%) (4%) (1%) (1%) (22%)
at other
Table
2.
Comparison Group 1 (emp t/def+)
Number of patients (n = 112) Mean age (70 -t 14 y) Risk factors Malignancy (n = 39) (35%) Neutropenia (n = 12) (11%) Mortality Overall (n = 46) (40%) Infection-related (n = 39) (35%)
of the Four Group 2 (emp-/def+)
23
41
72 +- 16
67i
4 (17%)
14 (34%)
Groups
Group 3 (emp-/def-)
Group 4 (none)
29
19
69211
77 2 9
16 (55%)
5 (26%)
(5%)
5 (17%)
2 (11%)
6 (26%)
11 (27%)
14 (48%)
15 (79%)
4 (17%)
7 (17%)
13 (45%)
15 (79%)
3 (13%)
2
17
Treatment
*Nine patients could not be categorized. emp+ = appropriate empiric therapy; def+ = appropriate definitive therapy; emp- = inappropriate empiric therapy; def- = inappropriate definitive therapy.
The majority of the patients had at least one underlying condition, the most common of which were: malignancy 35% (42/121), steroid therapy 19% (23/121), diabetes mellitus 16% (19/121), and neutropenia 11% (13/121). Of the 123 infectious episodes, 99 (80%) were hospital acquired. The sources of bacteremia were primarily urinary and respiratory tract infections (Table 1). Resistance to antipseudomonal drugs among 96 hospitalacquired isolates was as follows: gentamicin, 43%; piperacillin, 29%; ceftazidime, 18%; ciprofloxacin, 17%; amikacin, 9%; and imipenem, 5%. Resistance among community-acquired isolates was significantly different only for gentamicin (26%) and piperacillin (9%) (P I 0.05). Forty-nine patients (40.5%) died during hospitalization, but only 41 deaths (34% of the total) were attributed to pseudomonas infection. One hundred and twelve patients were divided into the following four groups according to the type of therapy (nine patients were excluded owing to absent or conflicting data): Group 1: appropriate empiric and appropriate definitive therapy (emp+/def+) = 23 patients. Group 2: inappropriate empiric and appropriate definitive therapy (emp-/def+) = 41 patients. Group 3: inappropriate empiric and inappropriate definitive therapy (emp-/def-) = 29 patients. Group 4: no therapy = 19 patients. Mortality from infection was identical (17%) in groups 1 and 2 irrespective of appropriateness of empiric therapy (Table 2). The difference in mortality from infection between patients receiving appropriate definitive therapy (groups 1 and 2 combined) and patients receiving inappropriate definitive therapy (group 3) was statistically significant (1 I/64 [17%] vs. 13/29 [45%], P < 0.05).
Pseudomonas
Table 3. versus
Comparison Combination
Number of patients Age 2 70 years Risk factors Malignant disease Neutropenia Hospital-acquired infection Mortality Overall Infection-related
Two Drugs ___(%) n
n
(%)
42 27
(100%) (64%)
15 9
(100%) (60%)
8 2 37
(19%) (5%) (88%)
7
(47%) (7%) (93%)
0.05 NS NS
7 6
(17%) (14%)
7 2
(47%) (13%)
0.05 NS
1:
P Value NS
NS = not significant.
Obviously, comparing the outcome of patients receiving either monotherapy or combination drug therapy was applicable only to the 64 patients who received appropriate definitive treatment (groups 1 and 2). Seven of these 64 patients were excluded from this analysis because of changes in the definitive therapy either from combination therapy to monotherapy or vice versa. Of the remaining 57 patients, 42 patients received monotherapy and 15 combination drug therapy. These two treatment groups did not differ significantly in terms of age, underlying conditions, or place of acquisition of infection (Table 3), although malignant disease was somewhat more prevalent in the combination drug group (P = 0.05). Mortality from infection was equal in the two groups (14% and 13%, see Table 3).
Table
4.
Characteristics
Baltch and Griffin 2 Study period Number of episodes Male:Female Age Mean Range 270 years (%) Incidence (cases/i 000 admissions) Hospital-acquired (%) Stay to bacteremia (d)+ Focus Urinary (%) Pulmonary (%) Underlying malignancy (%) Chemotherapy (%) Neutropenia (%) Steroid therapy (“Y) Mortality Overall (%) Infection-related (%) *Median;
+mean number
1972-1975 75 1.8:1 59 O-82 24
97
28 21 33 17 13 39 63 44
of days of hospitalization
/ Siegman-Igra
et al
213
DISCUSSION
of Monotherapy Drug Therapy
Monotherapy
aeruginosa
of Pseudomonas
A literature review of the clinical and epidemiologic features of l? aeruginosa bacteremia revealed no significant changes over the past 20 years. Comparison of several large series of P aeruginosa bacteremia published from I976 to 1996 showed striking similarities in clinical features and outcome among the different series (Table 4).2-9 The only differences detected among these series concerned the incidence of Pseudomonas bacteremia per 1000 admissions and the percentage of underlying malignant disease, reflecting most likely, differences in the populations from which the study patients were derived. Interestingly, and in spite of the varying underlying conditions, mortality due to infection remained surprisingly constant (38-48%) throughout these 2 decades, except for that reported in the most recent study, which is, so far, exceptional.” Historically, up to the late 1960s most Pseudomonas infections were fatal. The introduction of gentamicin and later carbenicillin, when used alone, improved prognosis only slightlylO However, the combination of carbenicillin and gentamicin was shown to be synergistic in vitro,” and resulted in a higher survival rate than either drug alone in experimental Paeruginosa infections in animal models.12-14 Human studies involving, primarily, cancer patients with or without neutropenia also favored combination therapy, but included only a small number of cases of Pseudomonas bacteremia among others with gram-negative sepsis.15-” Following these findings, combination therapy with an
Bacteremia
Bodey et al3
McKendrick and Gedde9
Lechi et a/5
1972-1981 410 1.3:1
1974-1980 21 1:l.l
1975-1981 60 1.6:l
44 2-85
65* 18-78
15 32 100 78 69
64 14-87 43 1.41 65 22
14 19 57 43 52 64
17 30 66
27 32
48
75
to bacteremia;
*hematologic
996)
Ma/lo/as et al8
Wdal et al9
Present study
1982-1986 200
1983-1989 274 2.2:1
1991-1994 189 2:i
1990-1992 123 2:l
59 O-96
77 20
27 61 48
41
malignancy
(1976-l
Hi/f et al 7
29 34 24* 34 26” 40
27
38 from admission
1980-1984 96 1.9:1
0.4 85
71
the Literature
Gallaghar and Wa tanakunakorn6
8 4.7 81 12
from
54
70 12-94 65
1.4 80
1.5 83
1.66 80 17
48
16 16 28
33 18 35 17 11 19
23 22 18
only: Q3000
42 neutrophils/pL.
11
40 34
214
International
Journal
of Infectious
Diseases
/ Volume 2, Number 4, April-June 1998
antipseudomonal beta-lactam and an aminoglycoside became the standard therapy for Pseudomonas bacteremia.” Later, when newer drugs, such as ceftazidime, imipenem-cilastatin, and quinolones were shown to be effective as single agents in the therapy of severe Pseudomonas infections, even in neutropenic patients,19-22 the need for a second drug was questioned. To the authors’ knowledge, no prospective randomized comparison between monotherapy and combination drug therapy has been done in a series of patients with Pseudomonas bacteremia. Bodey et al showed equal results when an antipseudomonal beta-lactam antibiotic was used with or without an aminoglycoside in patients with Pseudomonas bacteremia.3,23 However, the authors were cautious about recommending the use of an antipseudomonal penicillin as a single agent to treat serious Pseudomonas infections3 Hilf et al, in a prospective, but not randomized study on Pseudomonas bacteremia, concluded that the use of combination therapy was associated with a significantly lower mortality than monotherapy (27% vs. 47%).’ In their study, 143 patients received combination therapy and 43 monotherapy However, 37 of these 43 patients received aminoglycoside alone, and six patients received beta-lactam alone (mostly a ureidopeni~illin).~~~*~~~ Thus, most of these 43 patients would have been categorized as receiving inappropriate therapy according to the definitions presented in this report, and indeed, the mortality in these 43 patients was 47%, the same as in the inappropriately treated patients in the present study (group 3; 45%). Therefore, no conclusion can be drawn from this study concerning the effectiveness of single-drug therapy. Another retrospective study reviewing 189 episodes of Paeruginosa bacteremia showed no difference in outcome between patients receiving one drug or two or more.9 There are no details as to whether ureidopenicillins were included as monotherapy The present study, in spite of being retrospective and including a relatively small number of patients for comparison, showed that monotherapy (with a cephalosporin, imipenem-cilastatin, or quinolone) given to 42 patients who survived the first 48 hours of I? aeruginosa bacteremia, was as effective as combination drug therapy. Interestingly, in accordance with a recent study,9 and in contrast to an earlier one,3 in the present study no difference was noted in outcome when appropriate therapy was begun early or only after drug susceptibility was reported (group 2 vs. group 1). However, the possible effect of the initial empiric therapy on early survival was not evaluated, and patients who died within the first 48 hours were excluded from the final analysis. In conclusion, this study shows equal results with monotherapy compared to combination drug therapy in patients who survived the first 48 hours of infection of Pseudomonas bacteremia. In spite of the weaknesses of a retrospective analysis, no better data are available.
Prospective, randomized controlled studies are needed to support this conclusion. It should also be stressed that effectiveness is not the only consideration dictating whether or not to use combination therapy, and preventing antibiotic resistance may be another important reason for choosing double therapy Thus, the choice of treatment should probably be tailored individually to each patient, according to various factors, such as the presence of neutropenia or septic shock, being hospitalized in the intensive care unit, length of hospitalization, previous antibiotic therapy, presence of renal failure, etc. It should be kept in mind, however, that certain antipseudomonal drugs are fully effective as monotherapy.
REFERENCES bacteremia. Review of 108 cases.Am J Med 1976; 60:501-508. 2. Baltch AL., Griffh PE. Pseudomonas aeruginosa bacteremia: a clinical study of 75 patients. Am J Med Sci 1977; 274:
1. Flick MR, Cluff LE. Pseudomonas
119-120.
3. Bodey GPJadeja L, Elting L. Pseudomonas bacteremia. Retrospective analysisof 410 episodes.Arch Intern Med 1985; 145;1621-1629. 4. McKendrick Mw; Geddes AM. Pseudomonas septicaemia in a general hospital: seven years’ experience. Q J Med 1981; 199:331-344. 5. Lechi A, Arosio E,PanceraP,et al. Pseudomonas septicaemia. A review of 60 cases observed in a university hospital. J Hosp Infect 1984; 5:29-37. 6. Gallaghar PG,Watanakunakorn C. Pseudomonas bacteremia in a community teaching hospital, 1980-1984. Rev Infect Dis 1989; 11:846-852. 7. Hilf M,Yu VL, Sharp J, Zuravleff JJ,Korvick JA, Muder RR. Antibiotic therapy for Pseudomonas aeruginosa bacteremia: outcome correlation in a prospective study of 200 patients. Am J Med 1989; 87:540-546. 8. Mallolas J, Gate11JM, Miro JM, Almela M, Soriano E. Prognostic factors for Pseudomonas aeruginosa bacteremia [Letter]. Am J Med 1991; 90:134. 9. Vidal F,Mensa J,Almela M, et al. Epidemiology and outcome of Pseudomonas aeruginosa bacteremia, with special emphasis on the influence on antibiotic treatment. Analysis of 189 episodes. Arch Intern Med 1996; 156:2121-2126. 10. Andriole VT. Pseudomonas bacteremia: Can antibiotic therapy improve survival? J Lab Clin Med 1979; 94:196-200. 11. Weinstein RJ,Young LS, Hewitt WL. Comparison of methods for assessing in vitro antibiotic synergism against Pseudomonas and Sewatia. J Lab Clin Med 1975; 86:853-862. 12. Andriole VT. Synergy of carbenicillin and gentamicin in experimental infection with Pseudomonas. J Infect Dis
1971; 124 (suppl):S46-S55. 13. Khakoo RA, Kluge RM. Effectiveness of carbenicillin and gentamicin in a rat model against infection with Pseudomonas aeruginosa resistant to gentamicin or gentamicin and carbenicillin. J Antimicrob Chemother 1979; 5:53-59. 14. Rusnak MG, Drake TA, Hackbarth CJ, Sande MA. Single versus combination antibiotic therapy for pneumonia due to Pseudomonas aeruginosa in neutropenic guinea pigs. J Infect Dis 1984; 149:980-985.
Pseudomonas
15. KIastersky J, Cappel R, Daneau D. Therapy with carbeniciUin and gentamicin for patients with cancer and severe infections caused by gram-negative rods. Cancer 1973; 31:331-336. 16. Klastersky J. Significance of antimicrobial synergism for the outcome of gram-negative sepsis. Am J Medl Sci 1977; 273:157-167. 17. EORTC Antimicrobial Therapy Study Group. Three antibiotic regimens in the treatment of infection in febrile granulocytopenic patients with cancer. J Infect Dis 1978; 137: 14-29. 18. Baltch AL, Smith RP. Combination of antibiotics against Pseudomonas aeruginosa. Am J Med 1985; 79(Suppl lA):8-16. 19. Verhagen C, de Paw BE, Donnelly JP, Williams KJ, de Witte T, Janssen TH. Ceftazidime alone for treating Pseudomonas aeruginosa septicemia in neutropenic patients. J Infect 1986; 13:125-131.
aeruginosa
/ Siegman-Igra
et al
215
20. Liang R,Yung R, Chiu E, et al. Ceftazidime versus imipenemcilastatin as initial monotherapy for febrile neutropenic patients. Antimicrob Agents Chemother 1990; 34:1336-1341. 21. Johnson PR,Yin JA, Tooth JA. High dose intravenous ciprofloxacin in febrile neutropenic patients. J Antimicrob Chemother 1990; 26(Suppl F):lOl-107. 22. GiamareIlou H, Galanakis N. Use of intravenous ciprofloxacin in difficult to treat infections. Am J Med 1987; 82:346-351. 23. Bodey GE Feld R, Burgess MA. Beta-lactam antibiotics alone or in combination with gentamicin for therapy of gram-negative bacillary infections in neutropenic patients. Am J Med Sci 1976; 271:179-186. 24. Bross JE. Combination therapy versus monotherapy for Pseudomonas aeruginosa septicemia [Letter]. Am J Med 1991; 90:135. 25. Yu VL, Hilf M. The reply [Letter]. Am J Med 1991; 90:135.
Report
Pseudomonas aeruginosa Bacteremia: An Analysis of 123 Episodes, with Particular Emphasis on the Effect of Antibiotic Therapy Yardena Siegman-Igra, MD;* Ramit Ravona, MD;* Hedva Primerman, MSc;* and Michael Giladi, MD* ABSTRACT Objectives: To review current experience with Pseudomonas aeruginosa bacteremia and compare outcome of patients treated with single-drug, versus combination therapy. Methods: The charts of all patients with P aeruginosa bacteremia between 1990 and 1992 were reviewed, and pertinent demographic, clinical, and bacteriologic data were retrieved. In addition, similar data were collected from a series of patients with P aeruginosa bacteremia from the literature of the past 20 years. Results: One hundred and twenty-three episodes of F! aeruginosa bacteremia in 121 patients were identified. Most patients were older than 70 years, had at least one underlying condition, and had acquired the infection in the hospital. Attributable mortality was 34%. After exclusion for early mortality and inappropriate therapy, 57 patients remained eligible for comparison of outcome according to therapy protocol. Mortality from infection was equal between the group of 42 patients who received monotherapy and the 15 patients who received combination therapy (14% and 13%, respectively). The literature review revealed eight articles describing 21 to 410 spisodes of Pseudomonas bacteremia. The clinical characteristics of these series did not differ significantly from those of the present series. Conclusions: Incidence, epidemiology, clinical characteristics, and outcome of pseudomonas sepsis did not change significantly over the past 2 decades. Appropriate monotherapy was as effective as combination drug therapy for individuals with pseudomonas bacteremia surviving the first 2 days of infection.
Key Words:
Int J Infect
bacteremia, drug therap)/; nosocomial, Pseudomonas aeruginosa, sepsis Dis
1998;
2:211-215.
*Infectious Disease Unit and Microbiology Laboratory, Medical Center, Sackler Faculty of Medicine, Tel-Aviv Received:
October
17,1997;
Address correspondence Unit, Tel-Aviv Sourasky Israel.
Accepted:
to Dr.Yardena Medical Center,
April
Tel-Aviv Sourasky University, Israel.
9, 1998.
Siegman-Igra,
6 Weizman
Infectious Disease Street, Tel-Aviv 64239
Pseudomonas aeruginosa infections continue to be responsible for considerable morbidity and mortality in compromised and hospitalized patients. In spite of longterm recognition of the syndrome of Pseudomonas bacteremia and availability of effective drugs, the treatment of choice is still controversial, especially as to whether combination therapy is superior to monotherapy. A literature review encompassing the past 20 years revealed much information concerning the epidemiology, demography, underlying conditions, place of acquisition, foci of infection, risk factors for mortality, and outcome associated with I! aeruginosa bacteremia, but did not supply a satisfactory answer to the controversy.1-9 The authors conducted a retrospective analysis of 123 episodes of I? aeruginosa bacteremia over a period of 3 years, reviewed epidemiologic and clinical characteristics in comparison with previous similar series, and compared outcome for monotherapy versus combination drug therapy. MATERIAL
and
METHODS
The Tel-Aviv Sourasky Medical Center is a 950-bed tertiary hospital located in central Tel-Aviv, serving a mostly elderly population. All patients who had I? aeruginosa in blood cultures during the period 1990 to 1992 were included in this analysis. Their medical records were carefully reviewed and pertinent demographic and medical data were retrieved. The following definitions were employed: 1. Neutropenia: Polymorphonuclear count of less than 1000&L. 2. Hospital-acquired bacteremia: Bacteremia that occurred more than 48 hours after admission, or less than 48 hours following an invasive procedure. 3. Death attributed to infection: Death occurring before resolution of the septic manifestations, and in the absence of another obvious direct cause of death. 4. Death not associated with infection: All other deaths occurring during hospitalization. 5. Empiric therapy: Antibiotics given from the onset of clinical signs attributed to bacteremia to the day that
211
212
6.
7.
8.
9.
International
Journal
of Infectious
Diseases
/ Volume 2, Number 4, April-June 1998
antimicrobial susceptibility of the isolate was reported. Definitive therapy: Drugs given after antimicrobial susceptibility was reported. The interval between empiric and definitive therapy ranged from 2 to 4 days. Appropriate antipseudomonal therapy. Monotherapy: Either a fluoroquinolone or a third generation cephalosporin or imipenem-cilastatin, to which the isolate was susceptible in vitro, given either alone or in combination with another drug to which the organism was resistant. The use of either a ureidopenicillin or an aminoglycoside alone was not considered appropriate even if the isolate was susceptible to the drug. Combination drug therapy: The use of one of the following drugs: a fluoroquinolone, a third generation cephalosporin, imipenem-cilastatin, or a ureidopenicillin in combination with an aminoglycoside when the isolate was susceptible to both. No therapy: The use of either no antibiotics at all or the administration of antibiotics for less than 24 hours. This occurred mostly in patients who died in the first 48 hours of bacteremia.
Identification of p aeruginosa and antibiotic susceptibility testing was performed using an automated system (Sensititer, Radiometer, Copenhagen, Denmark). Statistical analysis was performed using the chi-squared or Fisher’s exact test as appropriate. A P value of I 0.05 was considered significant. RESULTS One hundred and twenty-three episodes of p aeruginosa bacteremia in 121 patients were identified during the study period (1990- 1992), representing 1.66 episodes per 1000 admissions. The patients’ ages ranged between 12 and 94 years (mean, 70; median, 72); 79 patients (65%) were 70 years or older; 81 patients (67%) were males.
Sources of Infection in Communityand HospitalTable 1. Acquired Pseudomonas Bacteremia in 120 Episodes” Source
of Infection
Urinary tract Respiratory tract Intra-abdominal Vascular lines Soft tissue Perianal Meningitis Unknown *An additional institutions.
three episodes
Community Acquired n = 24 (700%)
Hospital Acquired n = 96 (100%)
3 (12%) 5 (21%) 1 1
37 17 11 4 4 1 1 21
(4%) (4%)
1 (4%) 13 (54%) were excluded
due to acquisition
(39%) (18%) (11%) (4%) (4%) (1%) (1%) (22%)
at other
Table
2.
Comparison Group 1 (emp t/def+)
Number of patients (n = 112) Mean age (70 -t 14 y) Risk factors Malignancy (n = 39) (35%) Neutropenia (n = 12) (11%) Mortality Overall (n = 46) (40%) Infection-related (n = 39) (35%)
of the Four Group 2 (emp-/def+)
23
41
72 +- 16
67i
4 (17%)
14 (34%)
Groups
Group 3 (emp-/def-)
Group 4 (none)
29
19
69211
77 2 9
16 (55%)
5 (26%)
(5%)
5 (17%)
2 (11%)
6 (26%)
11 (27%)
14 (48%)
15 (79%)
4 (17%)
7 (17%)
13 (45%)
15 (79%)
3 (13%)
2
17
Treatment
*Nine patients could not be categorized. emp+ = appropriate empiric therapy; def+ = appropriate definitive therapy; emp- = inappropriate empiric therapy; def- = inappropriate definitive therapy.
The majority of the patients had at least one underlying condition, the most common of which were: malignancy 35% (42/121), steroid therapy 19% (23/121), diabetes mellitus 16% (19/121), and neutropenia 11% (13/121). Of the 123 infectious episodes, 99 (80%) were hospital acquired. The sources of bacteremia were primarily urinary and respiratory tract infections (Table 1). Resistance to antipseudomonal drugs among 96 hospitalacquired isolates was as follows: gentamicin, 43%; piperacillin, 29%; ceftazidime, 18%; ciprofloxacin, 17%; amikacin, 9%; and imipenem, 5%. Resistance among community-acquired isolates was significantly different only for gentamicin (26%) and piperacillin (9%) (P I 0.05). Forty-nine patients (40.5%) died during hospitalization, but only 41 deaths (34% of the total) were attributed to pseudomonas infection. One hundred and twelve patients were divided into the following four groups according to the type of therapy (nine patients were excluded owing to absent or conflicting data): Group 1: appropriate empiric and appropriate definitive therapy (emp+/def+) = 23 patients. Group 2: inappropriate empiric and appropriate definitive therapy (emp-/def+) = 41 patients. Group 3: inappropriate empiric and inappropriate definitive therapy (emp-/def-) = 29 patients. Group 4: no therapy = 19 patients. Mortality from infection was identical (17%) in groups 1 and 2 irrespective of appropriateness of empiric therapy (Table 2). The difference in mortality from infection between patients receiving appropriate definitive therapy (groups 1 and 2 combined) and patients receiving inappropriate definitive therapy (group 3) was statistically significant (1 I/64 [17%] vs. 13/29 [45%], P < 0.05).
Pseudomonas
Table 3. versus
Comparison Combination
Number of patients Age 2 70 years Risk factors Malignant disease Neutropenia Hospital-acquired infection Mortality Overall Infection-related
Two Drugs ___(%) n
n
(%)
42 27
(100%) (64%)
15 9
(100%) (60%)
8 2 37
(19%) (5%) (88%)
7
(47%) (7%) (93%)
0.05 NS NS
7 6
(17%) (14%)
7 2
(47%) (13%)
0.05 NS
1:
P Value NS
NS = not significant.
Obviously, comparing the outcome of patients receiving either monotherapy or combination drug therapy was applicable only to the 64 patients who received appropriate definitive treatment (groups 1 and 2). Seven of these 64 patients were excluded from this analysis because of changes in the definitive therapy either from combination therapy to monotherapy or vice versa. Of the remaining 57 patients, 42 patients received monotherapy and 15 combination drug therapy. These two treatment groups did not differ significantly in terms of age, underlying conditions, or place of acquisition of infection (Table 3), although malignant disease was somewhat more prevalent in the combination drug group (P = 0.05). Mortality from infection was equal in the two groups (14% and 13%, see Table 3).
Table
4.
Characteristics
Baltch and Griffin 2 Study period Number of episodes Male:Female Age Mean Range 270 years (%) Incidence (cases/i 000 admissions) Hospital-acquired (%) Stay to bacteremia (d)+ Focus Urinary (%) Pulmonary (%) Underlying malignancy (%) Chemotherapy (%) Neutropenia (%) Steroid therapy (“Y) Mortality Overall (%) Infection-related (%) *Median;
+mean number
1972-1975 75 1.8:1 59 O-82 24
97
28 21 33 17 13 39 63 44
of days of hospitalization
/ Siegman-Igra
et al
213
DISCUSSION
of Monotherapy Drug Therapy
Monotherapy
aeruginosa
of Pseudomonas
A literature review of the clinical and epidemiologic features of l? aeruginosa bacteremia revealed no significant changes over the past 20 years. Comparison of several large series of P aeruginosa bacteremia published from I976 to 1996 showed striking similarities in clinical features and outcome among the different series (Table 4).2-9 The only differences detected among these series concerned the incidence of Pseudomonas bacteremia per 1000 admissions and the percentage of underlying malignant disease, reflecting most likely, differences in the populations from which the study patients were derived. Interestingly, and in spite of the varying underlying conditions, mortality due to infection remained surprisingly constant (38-48%) throughout these 2 decades, except for that reported in the most recent study, which is, so far, exceptional.” Historically, up to the late 1960s most Pseudomonas infections were fatal. The introduction of gentamicin and later carbenicillin, when used alone, improved prognosis only slightlylO However, the combination of carbenicillin and gentamicin was shown to be synergistic in vitro,” and resulted in a higher survival rate than either drug alone in experimental Paeruginosa infections in animal models.12-14 Human studies involving, primarily, cancer patients with or without neutropenia also favored combination therapy, but included only a small number of cases of Pseudomonas bacteremia among others with gram-negative sepsis.15-” Following these findings, combination therapy with an
Bacteremia
Bodey et al3
McKendrick and Gedde9
Lechi et a/5
1972-1981 410 1.3:1
1974-1980 21 1:l.l
1975-1981 60 1.6:l
44 2-85
65* 18-78
15 32 100 78 69
64 14-87 43 1.41 65 22
14 19 57 43 52 64
17 30 66
27 32
48
75
to bacteremia;
*hematologic
996)
Ma/lo/as et al8
Wdal et al9
Present study
1982-1986 200
1983-1989 274 2.2:1
1991-1994 189 2:i
1990-1992 123 2:l
59 O-96
77 20
27 61 48
41
malignancy
(1976-l
Hi/f et al 7
29 34 24* 34 26” 40
27
38 from admission
1980-1984 96 1.9:1
0.4 85
71
the Literature
Gallaghar and Wa tanakunakorn6
8 4.7 81 12
from
54
70 12-94 65
1.4 80
1.5 83
1.66 80 17
48
16 16 28
33 18 35 17 11 19
23 22 18
only: Q3000
42 neutrophils/pL.
11
40 34
214
International
Journal
of Infectious
Diseases
/ Volume 2, Number 4, April-June 1998
antipseudomonal beta-lactam and an aminoglycoside became the standard therapy for Pseudomonas bacteremia.” Later, when newer drugs, such as ceftazidime, imipenem-cilastatin, and quinolones were shown to be effective as single agents in the therapy of severe Pseudomonas infections, even in neutropenic patients,19-22 the need for a second drug was questioned. To the authors’ knowledge, no prospective randomized comparison between monotherapy and combination drug therapy has been done in a series of patients with Pseudomonas bacteremia. Bodey et al showed equal results when an antipseudomonal beta-lactam antibiotic was used with or without an aminoglycoside in patients with Pseudomonas bacteremia.3,23 However, the authors were cautious about recommending the use of an antipseudomonal penicillin as a single agent to treat serious Pseudomonas infections3 Hilf et al, in a prospective, but not randomized study on Pseudomonas bacteremia, concluded that the use of combination therapy was associated with a significantly lower mortality than monotherapy (27% vs. 47%).’ In their study, 143 patients received combination therapy and 43 monotherapy However, 37 of these 43 patients received aminoglycoside alone, and six patients received beta-lactam alone (mostly a ureidopeni~illin).~~~*~~~ Thus, most of these 43 patients would have been categorized as receiving inappropriate therapy according to the definitions presented in this report, and indeed, the mortality in these 43 patients was 47%, the same as in the inappropriately treated patients in the present study (group 3; 45%). Therefore, no conclusion can be drawn from this study concerning the effectiveness of single-drug therapy. Another retrospective study reviewing 189 episodes of Paeruginosa bacteremia showed no difference in outcome between patients receiving one drug or two or more.9 There are no details as to whether ureidopenicillins were included as monotherapy The present study, in spite of being retrospective and including a relatively small number of patients for comparison, showed that monotherapy (with a cephalosporin, imipenem-cilastatin, or quinolone) given to 42 patients who survived the first 48 hours of I? aeruginosa bacteremia, was as effective as combination drug therapy. Interestingly, in accordance with a recent study,9 and in contrast to an earlier one,3 in the present study no difference was noted in outcome when appropriate therapy was begun early or only after drug susceptibility was reported (group 2 vs. group 1). However, the possible effect of the initial empiric therapy on early survival was not evaluated, and patients who died within the first 48 hours were excluded from the final analysis. In conclusion, this study shows equal results with monotherapy compared to combination drug therapy in patients who survived the first 48 hours of infection of Pseudomonas bacteremia. In spite of the weaknesses of a retrospective analysis, no better data are available.
Prospective, randomized controlled studies are needed to support this conclusion. It should also be stressed that effectiveness is not the only consideration dictating whether or not to use combination therapy, and preventing antibiotic resistance may be another important reason for choosing double therapy Thus, the choice of treatment should probably be tailored individually to each patient, according to various factors, such as the presence of neutropenia or septic shock, being hospitalized in the intensive care unit, length of hospitalization, previous antibiotic therapy, presence of renal failure, etc. It should be kept in mind, however, that certain antipseudomonal drugs are fully effective as monotherapy.
REFERENCES bacteremia. Review of 108 cases.Am J Med 1976; 60:501-508. 2. Baltch AL., Griffh PE. Pseudomonas aeruginosa bacteremia: a clinical study of 75 patients. Am J Med Sci 1977; 274:
1. Flick MR, Cluff LE. Pseudomonas
119-120.
3. Bodey GPJadeja L, Elting L. Pseudomonas bacteremia. Retrospective analysisof 410 episodes.Arch Intern Med 1985; 145;1621-1629. 4. McKendrick Mw; Geddes AM. Pseudomonas septicaemia in a general hospital: seven years’ experience. Q J Med 1981; 199:331-344. 5. Lechi A, Arosio E,PanceraP,et al. Pseudomonas septicaemia. A review of 60 cases observed in a university hospital. J Hosp Infect 1984; 5:29-37. 6. Gallaghar PG,Watanakunakorn C. Pseudomonas bacteremia in a community teaching hospital, 1980-1984. Rev Infect Dis 1989; 11:846-852. 7. Hilf M,Yu VL, Sharp J, Zuravleff JJ,Korvick JA, Muder RR. Antibiotic therapy for Pseudomonas aeruginosa bacteremia: outcome correlation in a prospective study of 200 patients. Am J Med 1989; 87:540-546. 8. Mallolas J, Gate11JM, Miro JM, Almela M, Soriano E. Prognostic factors for Pseudomonas aeruginosa bacteremia [Letter]. Am J Med 1991; 90:134. 9. Vidal F,Mensa J,Almela M, et al. Epidemiology and outcome of Pseudomonas aeruginosa bacteremia, with special emphasis on the influence on antibiotic treatment. Analysis of 189 episodes. Arch Intern Med 1996; 156:2121-2126. 10. Andriole VT. Pseudomonas bacteremia: Can antibiotic therapy improve survival? J Lab Clin Med 1979; 94:196-200. 11. Weinstein RJ,Young LS, Hewitt WL. Comparison of methods for assessing in vitro antibiotic synergism against Pseudomonas and Sewatia. J Lab Clin Med 1975; 86:853-862. 12. Andriole VT. Synergy of carbenicillin and gentamicin in experimental infection with Pseudomonas. J Infect Dis
1971; 124 (suppl):S46-S55. 13. Khakoo RA, Kluge RM. Effectiveness of carbenicillin and gentamicin in a rat model against infection with Pseudomonas aeruginosa resistant to gentamicin or gentamicin and carbenicillin. J Antimicrob Chemother 1979; 5:53-59. 14. Rusnak MG, Drake TA, Hackbarth CJ, Sande MA. Single versus combination antibiotic therapy for pneumonia due to Pseudomonas aeruginosa in neutropenic guinea pigs. J Infect Dis 1984; 149:980-985.
Pseudomonas
15. KIastersky J, Cappel R, Daneau D. Therapy with carbeniciUin and gentamicin for patients with cancer and severe infections caused by gram-negative rods. Cancer 1973; 31:331-336. 16. Klastersky J. Significance of antimicrobial synergism for the outcome of gram-negative sepsis. Am J Medl Sci 1977; 273:157-167. 17. EORTC Antimicrobial Therapy Study Group. Three antibiotic regimens in the treatment of infection in febrile granulocytopenic patients with cancer. J Infect Dis 1978; 137: 14-29. 18. Baltch AL, Smith RP. Combination of antibiotics against Pseudomonas aeruginosa. Am J Med 1985; 79(Suppl lA):8-16. 19. Verhagen C, de Paw BE, Donnelly JP, Williams KJ, de Witte T, Janssen TH. Ceftazidime alone for treating Pseudomonas aeruginosa septicemia in neutropenic patients. J Infect 1986; 13:125-131.
aeruginosa
/ Siegman-Igra
et al
215
20. Liang R,Yung R, Chiu E, et al. Ceftazidime versus imipenemcilastatin as initial monotherapy for febrile neutropenic patients. Antimicrob Agents Chemother 1990; 34:1336-1341. 21. Johnson PR,Yin JA, Tooth JA. High dose intravenous ciprofloxacin in febrile neutropenic patients. J Antimicrob Chemother 1990; 26(Suppl F):lOl-107. 22. GiamareIlou H, Galanakis N. Use of intravenous ciprofloxacin in difficult to treat infections. Am J Med 1987; 82:346-351. 23. Bodey GE Feld R, Burgess MA. Beta-lactam antibiotics alone or in combination with gentamicin for therapy of gram-negative bacillary infections in neutropenic patients. Am J Med Sci 1976; 271:179-186. 24. Bross JE. Combination therapy versus monotherapy for Pseudomonas aeruginosa septicemia [Letter]. Am J Med 1991; 90:135. 25. Yu VL, Hilf M. The reply [Letter]. Am J Med 1991; 90:135.