M. F e r n f i n d e z - G u e r r e r o , F. G u d i o l , A. R o d r i g u e z - T o r r e s , C. A r n a u , L. V a l d d s , C. Vallv6

Nosocomial Pneumonia: Comparative Multicentre Trial between Monotherapy with Cefotaxime and Treatment with Antibiotic Combinations Introduction Summary: In a multicentre clinical trial involving 32 hospitals, 588 adult patients diagnosed with nosocomial pneumonia and not receiving mechanical ventilation were treated randomly with monotherapy with cefotaxime or the antibiotic combination routinely used in each particular hospital. Both groups of patients were similar regarding demographic data, concurrent diseases, additional therapies and causative organism. Protocol violations were recorded in 40 patients, and these patients were excluded from the evaluation of treatment efficacy. The cure rate was 79% in the cefotaxime group and 71% in the group receiving antibiotic combinations; this difference is statistically significant (p -- 0.03, Fisher's two-tailed test). In the patients receiving combinations of cephalosporins having activity predominantly against gram-positive organisms plus aminoglycosides, the cure rate obtained was very low. The frequency of serious adverse reactions was significantly higher in the group treated with antibiotic combinations. It is concluded that monotherapy with cefotaxime is the regimen that offers better results for the empirical treatment of nosoeomial pneumonia. Zusammenfassung: Behancllung der nosokomtalen Pneumonie mit Cefotaxim allein oder einer Antibiotikakombination. Multizentrische Vergleichsstudie. In einer multizentrischen Studie wurden 588 Erwachsene mit nosokomialer Pneumonie, die nicht beatmungspflichtig waren, in 32 Krankenh/iusern randomisiert entweder mit Cefotaxim allein oder der jeweils fiblichen Kombinationstherapie behandelt. Nach demographischen Daten, Begleitkrankheit, Zusatztherapien und kausalem Erreger waren die beiden Gruppen vergleichbar. 40 der Patienten wurden wegen Protokollverletzung von der Auswertung bezfiglich therapeutischer Wirksamkeit ausgeschlossen. Die Heilungsrate betrug in der Cefotaxim-Gruppe 79%, in der Vergleichsgruppe 71%. Dieser Unterschied ist statistisch signifikant (p = 0,03, Fisher's two-tailed test). Bei Patienten, die mit einer Kombination aus einem vorwiegend gegen grampositive Erreger wirksamen Cephalosporin mit einem Aminoglykosid behandelt wurden, war die Heilungsrate sehr gering. Bei Behandlung mit den Antibiotikakombinationen traten signifikant mehr ernste Nebenwirkungen auf. Ffir die empirische Therapie der nosokomialen Pneumonie bietet die Behandlung mit Cefotaxim folglich bessere M6glichkeiten.

S 320

Nosocomial pneumonia continues to be a diagnostic and therapeutic challenge. The incidence of hospital-acquired pneumonia is estimated to be 0.6% [1] and, in certain high-risk areas, 7 to 20% of hospitalized patients develop lower respiratory tract infections [2]. Mortality rates reported for nosocomial pneumonia have ranged from 20% in community hospitals to 50% in university referral hospitals [3]. Although specialized microbiologic techniques and several methods for obtaining respiratory specimens have been developed to improve diagnostic accuracy [4, 5], these methods may be impractical for widespread use or may not be feasible in some cases. Thus, many patients are treated empirically based on a knowledge of the microbiologic spectrum of the infection. Antimicrobial therapy with combinations of aminoglycosides plus cephalosporins, semisynthetic penicillins or clindamycin have been successfully used for nosocomial pneumonia [6]. Nevertheless, the need for combined therapy is not always clear, and treatment with a thirdgeneration cephalosporin, active against a wide array of microorganisms including nosocomial strains o f Streptococcus pneumoniae, Haemophilus, Moraxella and most species of Enterobacteriaceae, could be equally effective. In this study the efficacy and safety of empirical monotherapy with cefotaxime was compared with that of commonly used antibiotic combinations in the treatment of hospital-acquired pneumonia. A suitably designed study carried out with this purpose would also enable the assessment of the frequency of adverse reactions associated with the use of both treatment schemes.

Patients and Methods A prospective, open:labelled trial was performed, with random allocation of patients to one of the following two treatment regimens: a) monotherapy with intravenous cefotaxime, starting with a dose of 2 g every 8 hours, reduced to 2 g every 12 hours after observing improvement in the clinical picture; or b) an antibiotic combination routinely used in each center taking part in the study. M. Ferndndez-Guerrero, M. D., C. Arnau, Ph. D:, L. Vald&, M. D., C. Vallv~, M. D., Divisionof Infectious Disease, Fundaci6n Jim6nez Dfaz, Madrid; F. Gudiol, InfectiousDisease Service, Hospital Principes de Espafia, Barcelona;A. Rodffguez-Torres., M. D., Department of Clinical

Microbiology,Hospital Cllnico, Valladolid; Spain. Correspondence and reprint requests: Dr. 3,1. Fern(mdez-Guerrero, Fundaci6n Jim6nezDfaz,Avda. ReyesCat61icos2, 28040Madrid, Spain.

Infection 19 (1991) Suppl. 6 © MMV Medizin Verlag GmbH Mtinchen, Miinchen 1991

M. Fern~indez-Guerrero et al.: Cefotaxime in Nosocomial Pneumonia Table 1: Reasons for admission and concurrent diseases in the two patient groups studied.

iiiii!!ii!i!iiiliiiil I~i~i~:~i~:~;~i~i~i~i~:~!~i~i~:~i~'~I~,~ii~i~i~i~i~:~!]~i~!~!~!~ii~ii~!~!~;!~i~ii~I~i~i!~:: 11 Infectious and parasitic diseases 39 Neoplasms 15 Endocrine, nutritional, metabolic and immune diseases 5 Diseases of the blood and hematopoietic organs 1 Mental disorders 3 Diseases of the nervous system and sensory organs 61 Diseases of the cardiovascular system 35 Diseases of the respiratory system 57 Diseases of the digestive system 9 Diseases of the genitourinary system 5 Complications of pregnancy, labour and confinement 1 Diseases of the skin and subcutaneous tissue 6 Diseases of the locomotorsystem and connective tissue 9 Ill-defined symptoms,signs and conditions 20 Wounds, traumas and poisoning 3 No information Total

In both cases treatment was continued until at least three days after clinical remission, X-ray normalization and negativity of the microbiological tests. Diagnostic criteria: a) fever above 38°C, b) lung infiltrate documented by X-ray, and c) onset of symptoms more than 72 hours after admission to hospital. Patients with known hypersensitivity to cephalosporins or penicillins, patients receiving antibiotic therapy in the seven days before the onset of the disease, and patients hospitalized in an intensive care unit were excluded from the study. Upon enrolling each patient in the study, the patient's demographic data, reason for admission, concurrent diseases and additional therapies were recorded. Before starting treatment and 24--48 hours after completing treatment, the body temperature, X-ray finding and, if carried out, the results of the blood culture and antibiogram were recorded. Any possible adverse reactions observed during therapy were also recorded. The number of subjects to be included in the study, 641 in each group, was calculated to detect a difference of 10% in a two-tailed test, with c~ = 0.05 and power = 0.95. As no precise information was available about the cure rate, it was assumed that this would be 50%; this assumption always gives a sample size above the number of patients actually needed. The random allocation of patients to the therapies and the statistical evaluation of the results were done using the procedures ,of the SAS statistical package. The randomization was done in blocks of two, so that the numbers of patients in each treatment group would be similar. Fisher's two-tailed test was used to compare the efficacy and frequency of adverse reactions.

Results The study was conducted between September 1988 and November 1989. A total of 588 patients not receiving mechanical ventilation were included; 280 of these were assigned to the cefotaxime group and 308 to the antibiotic

280

~ii~l~i] 16 60 6 4 5 9 60 33 37 4 0 1 14 8 45 6

7 22 58 11 0 11 113 54 29 15 0 2 10 14 15 0

18 19 56 10 0 14 98 71 32 14 0 5 9 6 23 0

308

361

375

combination group. Thirty-two hospitals took part in the trial, each including between five and 40 patients. Both patient groups were similar regarding age, sex, reason for hospital admission and concurrent diseases. The median age was 67 years (range: 18-94) for the cefotaxime group and 65 years (range: 18-96) for the antibiotic combination group. Regarding sex, of the 280 patients receiving cefotaxime, 190 were males and 85 were females; this information was lacking in five patients. Of the 308 patients receiving the combination of antibiotics, 210 were males and 91 were females; this information was lacking in seven patients. The reasons for admission for each treatment group and the concurrent diseases, as classified by the World Health Organization [7], are listed in Table 1. Five patients in the cefotaxime group and 35 patients in the control group did not fulfill the protocol specifications: in three patients in the cefotaxime group the onset of the disease occurred before 72 hours after admission; one patient had received a previous treatment with antibiotic agents; and in one case data were contradictory. In the control group, 26 patients received only one antibiotic, in five cases data were contradictory, two patients had received previous treatment with antibiotic agents, and in two patients the onset of the disease occurred before 72 hours after admission. These 40 patients were not included in the efficacy evaluation but were included in the adverse reactions evaluation. Of the 548 patients eligible for evaluation, 275 were treated with cefotaxime and 273 with antibiotic combinations; of these, 91 were treated with cefotaxime combined with other antibiotics. The remaining discussion is devoted to the results with regard to withdrawals, deaths, clinical efficacy, bacteriological outcome and adverse reactions. To evaluate clinical efficacy,

Infection 19 (1991) Suppl. 6 © MMV Medizin Verlag GmbH Minchen, Mtinchen 1991

S 321

M. Fermindez-Guerrero et al.: Cefotaxime in Nosocomial Pneumonia Table 2: Causes of death recorded in the patients treated with cefotaxime and with antibiotic combinations.

Septicaemia Mycotic superinfection Malnutrition Diabetic ketoacidosis Gastrointestinal bleeding Hepatic failure Acute abdomen Renal failure Pulmonary embolism Acute lung oedema Respiratory failure Cardiorespiratory arrest Heart failure Sudden death Ictus Coma No information Total

4 0 1 2 1 2 1 2 4 0 6 4 1 1 2 1 4 36 (13 %)

4 1 0 1 2 2 0 2 1 1 18 6 1 0 4 2 7 52 (19 %)

withdrawals from the study, deaths, therapeutic failures and cases without information about clinical outcome areconsidered "failures."

Table 3: Clinical outcome obtained with the various treatment regimens used in this study.

Cefotaxime Antibiotic combinations Cefotaxime + aminoglycosides Cefotaxime + other antibiotics Broad-spectrum penicitlins + aminoglycosides Cephalosporins with action predominantly against gram-positive organisms + aminoglycosides Cephalosporins with action predominantly against gram-negative organisms + aminoglycosides Cephalosporins active against Pseudomonas + aminoglycosides Cephalosporins active against anaerobes + aminoglycosides Clindamycin + aminoglycosides Narrow-spectrum penicillins active against gram-positiveorganisms + aminoglycosides Other antibiotics + aminoglycosides Other antibiotic combinations

275 273 78 13

79% 71% 77% 77%

(74-84) (65-76) (66-86) (46-95)

31

74%

(55-88)

21

48%

(26-70)

24

67%

(45-84)

21

76%

(53-92)

18 61% 12 58%

(36-83) (2~55)

18 15 22

72 % 73 % 73 %

(46-90) (45-92) (50-89)

Total

548

75 %

(71-78)

Withdrawals from the Study Six patients were withdrawn from the group receiving cefotaxime: one due to intolerance, three due to voluntary withdrawal and two due to other reasons that were not specified. Nine patients were withdrawn from the group receiving antibiotic combinations: two due to intolerance, three to voluntary withdrawal and four due to unspecified causes. T h e r e were no significant differences between these two groups regarding withdrawals from the study.

Deaths O f the 548 patients eligible for evaluation, 88 (16%) died because of the causes listed in Table 2. The number of deaths was 36 (13%) in the group treated with cefotaxime alone, and 52 (19%) in the group receiving antibiotic combinations.

In the group of 275 patients treated with cefotaxime alone, 217 (79%) were cured. In the group of 273 patients treated with a combination of antibiotics, 193 (71%) were cured. This 8% difference (95% confidence interval from 1 to 16%) is statistically significant (p = 0.03, Fisher's two-tailed test). Table 3 gives the cure rates, with their 95% confidence intervals, observed with the various treatment regimens used in this trial, classified according to the Anatomical Classification of Pharmaceutical Specialities [8].

Bacteriological Efficacy Figure 1 shows the percentage of cures classified according to the result of blood culture (not done, negative and positive).

Therapeutic Failures In the group treated with cefotaxime, 16 cases were considered therapeutic failures: eight due to lack of in vitro sensitivity of isolates, seven due to clinical worsening and one due to lack of information. In the control group, 19 cases were considered therapeutic failures: seven due to lack of in vitro sensitivity of isolates, eleven due to clinical worsening and one due to lack of information. S 322

Clinical Efficacy

Adverse Reactions Table 4 lists the adverse reactions recorded with the various treatment regimens used in this study. Ten patients reported one adverse reaction in the group treated with cefotaxime alone (n = 280 + 91 = 3 7 1 ) o r

Infection 19 (1991) Suppl. 6 © MMV Medizin Verlag GmbH Miinchen, Mfinchen 1991

M. Fermindez-Guerrero et al.: Cefotaxime in Nosocomial Pneumonia

P for total CTX/AC comparison (3 groups of culture) = 0.04. %

Blood culture not done P = 0.06

Negative blood culture P = 0.18

Positive blood culture P = 0.55

90 t-

t

BO I-

30 I-

0 ~-

t

N=63

Cefotaxime

N=61

67%

77%

59%

N=154

N=152

N=58

N=60

Antibiotic combinations

Figure 1: Percentages of patients cured (95% confidence interval) with cefotaxime (CTX) and antibiotic combinations (AC).

combined (n = 91), and 16 adverse reactions were recorded in the group treated with antibiotic combinations (n = 308 - 91 = 217): 14 patients with one adverse reaction and one patient with two adverse reactions. The table states the antibiotics to which these adverse reactions were attributed and the total number of patients treated with these antibiotics. Serious adverse reactions were more frequent (p = 0.007, Fisher's two-tailed test) in the antibiotic combination group: 5/217 versus 0/371 for the cefotaxime group. Discussion

For years, combined antimicrobial therapy has been considered first-choice treatment for nosocomial pneumonia. The rationale of this common practice has been the need to give the patient broad coverage against the multiple bacterial species causing hospital-acquired lung infection. In addition, the potential for synergistic bactericidal activity has led to the widespread use of cephalosporin-aminoglycoside combinations in this setting [6]. Yet, there is little evidence that two antibiotics with activity against gram-negative organisms offer advantages over a single agent for treating gram-negative pneumonia [91.

In this study monotherapy with cefotaxime was, in terms of clinical efficacy, superior to different combinations of antimicrobial agents used for treatment of nosocomial pneumonia. When one looks at the bacteriologic efficacy, both monotherapy and combined therapy seem equivalent. But, as suggested by the results of in vitro studies [10], the ability of cefotaxime to eradicate specific pathogens such as Staphylococcus aureus or Pseudomonas aeruginosa was less than that of combined therapy that included aminoglycosides plus first-generation cephalosporins, semisynthetic peniciUins or antipseudomonal drugs. This observation suggests that monotherapy with cefotaxime should be avoided in patients with special risk factors for staphylococcal or pseudomonal infections, such as intubated patients, immunosuppressed hosts, and those hospitalized in intensive care units. The introduction of new antimicrobial compounds with an extended spectrum of activity may render the use of combined therapy for nosocomial pneumonia obsolete. When examples are available, the clinical efficacy of mono-drug empiric treatment of hospital-acquired pneumonia has been satisfactory [11-13]. Cefotaxime has been also successfully used in the treatment of nosocomial pneumonia caused by resistant gram-negative organisms [14, 15]. Its activity against emerging nosocomial

Infection 19 (1991) Suppl. 6 © MMV Medizin Verlag GmbH Mtinchen, Mfinchen 1991

S 323

M. F e r m i n d e z - G u e r r e r o et al.: Cefotaxime in Nosocomial P n e u m o n i a

Table 4: Adverse reactions attributed to the antibiotics given.

[iiiiiiii:ii®i!lili i i iiiii iliii!!i!liiliii!i iI iii!iliii!iiiiiiiiiiil!ii!!!;iiii!ii!;i!iii!iiii!ii!i Ni Cefotaxime

i iiiiiii!iiii!iii iiiiiiiiiiii!iiiiiiiiiiiiiii Niiiii ii!! i!i ii!i! !iiiiii ii ii !i i i! iii iii i! iiii ! i i i ! Ii i

3 3 1 1 t 1

Phlebitis Diarrhoea Vomiting Exanthema Ataxia Increased transaminases

68

1 1

Renal failure Increased creatinine

Extended hospital stayb Extended hospital stayb

5

1 1

Epigastralgia Toxicodermia

Extended hospital stayb

Gentamicin

99

1 1 1 2

Renal failure Renal failure Increased creatinine Vertigo and tinnitus

Exitusb Extended hospital stayb Treatment discontinued

Erythromycin

15

1 1

Phlebitis Phlebitis

Treatment discontinued

Amikacin

56

1

Increased BUN

Cloxacillin

4

1

Exantbema

Gentamicin + cefazolin

9

1

Increased BUN

Cefoxitin + tobramycin

10

1

Diarrhoea

Ceftazidime + amikacin

6

1

Phlebitis

Tobramycin Co-trimoxazole

371

!!ii

Lactose intolerance Treatment discontinued Attributed to halloperidol

"Number of patients treated with the corresponding antibiotics, bSerious adverse reaction. BUN = blood urea nitrogen.

pathogens such as S. pneumoniae and Moraxella may be of particular interest. Nosocomial strains of pneumococci are frequently resistant to penicillins, erythromycin, clindamycin and ceftazidime [16]. It should be expected that patients infected with these organisms will follow a downhill course when treated with such antibiotics alone or in combination with aminoglycosides. In terms of safety, monotherapy with cefotaxime may be more a convenient regimen than conventional combined therapy, particularly in the aged population of hospitalized patients. This seems particularly relevant when considering combined therapy that includes aminoglycosides. Some clinical studies have shown the therapeutic importance of obtaining maximal post-infusion levels of aminoglycoside antibiotics. In patients with gram-negative bacillary pneumonia, such

studies have demonstrated that peak plasma levels of gentamicin >7 mg/l, or amikacin > 28 mg/1, were more often associated with successful outcomes than plasma levels below these [17]. It should be expected that patients given high-dose aminoglycoside therapy will exhibit higher rates of renal or auditory toxicity. In this study treatment with cefotaxime caused a few minor side effects. In comparison with patients who received combined therapy, patients treated with cefotaxime showed fewer serious adverse reactions, This was particularly remarkable for renal and vestibular toxicity associated with aminoglycoside treatment. In summary, monotherapy with cefotaxime may be a useful empirical treatment for hospital-acquired pneumonia in non-ventilated immunocompetent patients.

References 1. Centers for Disease Control. National Nosocomial Infections Study Report. Annual Summary 1984. MMWR 35 (1986) 17ss-29ss.

2. Garibaldi, R. A., Britt , M. IL, Coleman, M. L.,Reading, J. C., Pace, N. L.: Risk factors for postoperative pneumonia. Am. J. Med. 70 (1981) 677-680.

3. Gross, P. A., Neu, H. C., Aswaspokee, P., Van Antwerpen, C., Aswapokee, N.: Deaths from nosocomial infections: experience in a

S 324

university hospital and a community hospital. Am. J. Med. 68 (1980) 219-223. 4. Wimherley, N., Faling, L. J., Bartlett, J. G.: A fiberoptic bronchoscopy technique to obtain uncontaminated lower airway secretions for bacterial culture. Am. Rev. Respir. Dis. 119 (1979) 337-343.

5. Chastre, J., Viau, F., Bran, P., Pierre, J., Dauge, M. C., Bouehana, A.,

Infection 19 (1991) Suppl. 6

© MMV Medizin Veflag GmbH Mtinchen. Mtinchen 1991

M. Fernfindez-Guerrero et al.: Cefotaxime in Nosocomial P n e u m o n i a Akesbi, A., Gibert, C.: Prospective evaluation of the protected specimen brush for the diagnosis of pulmonary infections in ventilated patients. Am. Rev. Respir. Dis. 130 (1.984) 924-929. 6. Pennington, J. E.: Nosocomial respiratory infection. In: Mandell, G. L., Douglas, 17. G., Bennett, J. E. (eds.): Principles and practice of infectious disease. Third edition. Churchill Livingstone, New York (1990) pp. 2199-2205. 7. Organizaci6n Mundial de la Salud: Clasificaci6n Internacional de Enfermedades. Washington 1978. 8. Cat~iiogo de especialidades farmac~uticas. Departamento T6cnico. Consejo General de Colegios Oficiales de Farmac6uticos. Madrid 1989. 9. Eliopoulos, G. M., Eliopoulos, C. T.: Antibiotic combinations:should they be tested? Clin. Microbiot. Rev. 1 (1988) 139--156. 10. Todd, P. A., Brogden, R. N.: Cefotaxime. A n update of its pharmacology and therapeutic use. Drugs 40 (1990) 608-651. 11. Cone, L; A., Woodard, D. R., Stolzman, D. S., Byrd, R. G.: Ceftazidime versus tobramycin-ticarcillin in the treatment of pneumonia and bacteremia. Antimicrob. Agents Chemother. 28 (1985) 33-36.

12. Greenberg, R. N., Reiily, P. M., Luppen, K. Lo, Bollinger, M., Mc Millan, R.: Aztreonam therapy for gram-negative pneumonia. Am. J. Med. 78s (1985) 31-33. 13. Salata, R. A., Gebhart, R. L., Palmer, D. L., Wade, B. H., Scheld, W. M., Groschel, D. H. M, Wenzel, R. P., Mandeli, G. L., Duma, R. J.: Pneumonia treated with imipenem/cilastin. Am. J. Med. 78s (1985) 104-109. 14. Carmine, A. A., Brogden, R. N., Heel, R. C., Speight, T. M., Avery, G. S°: Cefotaxime: a review of its antibacterial activity, pharmacological properties and therapeutic use. Drugs 25 (1983) 223-289. 15. Reeves, J. H., Russell, G. M, Cade, J. F., McDonald, M.: Comparison of ceftriaxone with cefotaxime in serious chest infections. Chest 96 (1989) 1292-1297. 16. Pallar~s, R., Gudiol, F., Lifiares, J., Viladrich, P.: Bacteremic pneumonia caused by penicillin-resistant pneumococci. N. Engl. J. Med. 318 (1988) 124-128. 17. Moore, R. D, Smith, C. R , Lietman, P. S.: Association of aminoglycoside plasma levels with therapeutic outcome in gram-negative pneumonia. Am. J. Med. 77 (1984) 657-662.

Infection 19 (1991) Suppl. 6 © MMV Medizin Verlag GmbH Mtinchen, M~inchen 1991

S 325