Scand J Infect Dis 11: 21 1-217, 1979
Chemotherapy of Chronic Infections with Mucoid Pseudomonas aeruginosa in Lower Airways of Patients with Cystic Fibrosis BIRGITTE FRIIS
Scand J Infect Dis Downloaded from informahealthcare.com by University of Otago on 07/20/15 For personal use only.
From Childrens Department TG, Rigshospitalet, Copenhagen, and Department of Clinical Microbiology, Statens Serurninstitut, Hvidovre Hospital, Hvidovre, Denmark
ABSTRACT. The bacteriological effect of chemotherapy against Pseudomonas aeruginosa (Ps.ae.) in lungs of patients with cystic fibrosis is reviewed. During a 5-year period 49 children and adults were treated with 190 courses of different antibiotics. The mucoid strains of Ps.ae. disappeared in 72.0% of the courses in which a combination of tobramycin and carbenicillin was employed. Tobramycin given alone had only bacteriological effect in 26.6 % of the courses. Colimycin alone or in combination with carbenicillin had no effect. In 18 patients who received subsequent courses of tobramycin and combination of tobramycin and carbenicillin a significant difference in favour of the combination therapy was found, also in cases with many precipitins against Ps.ae. in serum. In 74.5% of the initially successful courses the patients were recolonized with Ps.ae. within 1 month. No nephrotoxic or ototoxic side effects were demonstrated in spite of the high doses of tobramycin (10 mg/kg/24 h) employed and the repeated courses.
INTRODUCTION
MATERIAL AND METHODS
During the last 10 years the more intensive general and antimicrobial therapy of children and adolescents with cystic fibrosis (CF) has prolonged their survival time remarkably. One of the greatest problems today in the care of C F is the chronic lung infections with Pseudomonas aeruginosa (Ps.ae.). A recent publication has shown that these infections are highly contributory to the bad prognosis as to life in C F (10). At the C F centre, Rigshospitalet, Copenhagen, C F patients with Ps.ae. infections have been treated very actively with antibacterial chemotherapy because earlier investigations in the centre demonstrated a strong correlation between a bad prognosis and a long duration and high intensity of the Ps.ae. infection. On the other hand we were not fully convinced of the antimicrobial effect of the different chemotherapeutic regimens. Earlier studies have shown different results in the clinical and bacteriological improvement after treatment of Ps.ae. infections in C F (4, 5, 6, 14, 15). With this background 190 courses of antipseudomonas chemotherapy given in a 5-year period from 1970 to 1975 were evaluated, especially with respect to the bacteriological effect of these courses.
49 patients, 22 girls and 27 boys, were treated during the 5-year period. All patients had a history typical for CF. The diagnosis was verified by finding of pathological sweat electrolytes on repeated pilocarpine iontophoresis tests according to Gibson and Cooke (3). They have all been followed monthly as outpatients and the tracheal secretions have been examined by microscopy and bacteriological cultivation on each visit. They were all chronically infected with mucoid Ps.ae. (9, 10). Age range at the first course was 1-26 years, mean 9 years. 190 courses were given in total. The patients received up to 14 subsequent courses of chemotherapy. 16 died during the examination period. All the patients were treated daily with intensive lung physiotherapy and aerosol inhalations with mucolytic agents at home. They were all admitted to the same ward during the antipseudomonas chemotherapy. Indication for antimicrobial therapy included that Ps.ae. had been cultivated as well as found by microscopy according to the principles described below. The antibiotic therapy was in each case decided by the clinical microbiologist in collaboration with the medical staff of the CF centre. Most of the patients were in bad general condition with severe productive cough and grave X-ray lung changes. The indication for initiating chemotherapy against Ps.ae. was a remarkable weight loss, growth stagnation or continuously decreasing lung function measured monthly by peak-flow and forced vital capacity (Vitalor). Later on, the indications for chemotherapy were somewhat changed because it was shown that a correlation
1 4- 791957
Scotid J Infecr Dis I I
212
B . Friis
Scand J Infect Dis Downloaded from informahealthcare.com by University of Otago on 07/20/15 For personal use only.
existed between the number of serum precipitins against Ps.ae. and a bad prognosis (10). Precipitins in blood above 10 or rapidly increasing became a criterion for treatment. Average age at the first colonization with Ps.ae. was 7.25f3.96 years (+S.D.) and at onset of chronic infection with mucoid strains of Ps.ae. 8.74f5.13 years. The average interval between chronic infection and first course of chemotherapy was 15.8f 15.0 months. Bacteriological examination Sputum or tracheal secretion obtained by aspiration through a catheter and collected in sterile tubes were brought to the laboratory in cooled boxes and examined immediately. The most purulent part of the material was examined in the following way: ( I ) Gram-stained films were examined by microscopy by a bacteriologist and the flora associated with areas consisting of respiratory epithelial cells and pus cells without squamous epithelial cells was described, whereas the flora associated with oral epithelial cells was considered to be predominantly of oral or pharyngeal origin, and was not described (8). (2) 4 pieces, the size of half a pea, of the material were inoculated on 2 blood agar plates and one differential medium plate for gram-negative rods. One of the blood agar plates was used for primary testing of antibiotic sensitivity (Neosensitabs@,Rosco A/S, TAstrup, Denmark). The other blood agar plate was furthermore inoculated with a streak of Staphylococcus aureus to enable the diagnosis of Haemophilus influenzae by means of the satellite phenomenon. The plates were inoculated for 20 h at 35°C. The cultivated bacteria were identified using standard methods and secondary sensitivity testing was performed on isolated strains of Ps.ae. Sputum or tracheal secretion was examined as described just before initiation of antimicrobial treatment, on the last day of treatment or 1 day after the treatment was stopped, and 1 month later. Serum concentration of antibiotics Serum tobramycin concentration was normally measured before and 2 h after intravenous or intramuscular injections on the 3rd and 7th day of treatment, sometimes more often. Absorption curves were made in a few patients (5). The mean values of all measurings at a given time were used. When a combination of tobramycin and carbenicillin was used, penicillinase was added before the analysis of serum tobramycin. Serum concentrations of the other antibiotics used were only studied when indicated for clinical reasons. Pseudoinonas aeruginosa precipitins The number of precipitins against Ps.ae. was measured by means of crossed immunoelectrophoresis as described elsewhere (9). Other investigations The urine was examined with microscopy and sticks for glucose and protein in all courses. All patients had normal serum creatinine or serum carbamide on several controls, but the values were not controlled systematically after antimicrobial treatment, except in a few patients (5). The Sctrritl J Iifi,ct Dis I I
liver enzymes were checked after many, although not all courses of therapy. All patients who received tobramycin and who were not too young to cooperate at the examination, were examined with audiometric and vestibular tests in connection with the course and at a follow-up several months later (19). Antibiotics Table I shows the different drugs and drug combinations used. Only antibiotics to which the bacteria were fully sensitive in vitro were given, except in a very few cases in which carbenicillin was included in a combination therapy in spite of a slightly reduced sensitivity. Doses +S.D./kg/ 24 h and duration of the courses were:
Tobramycin (Nebcin@): 9.95 mgf I .27 mg i.m. or i.v. for 10-14 days (10 mg/kg/24 h) Carbenicillin (Fugacillin@): 345 mgf 120.8 mg i.v. for 14 days (300-500 mg/kg/24 h) Colimycin (Colistin@):79000 Uf31000 U i.m. or i.v. for 10-14 days (50 000-90 000 U/kg/24 h) Gentamicin? 3.46 mg f 1.30 mg i.m. for 10-14 days (3-4 mg/kg/24 h) Aerosol inhalation of tobramycin 100 mg twice a day was given to 55 % of the patients treated with tobramycin. This treatment was used during the first part of the period but was stopped as it proved superfluous. Aerosol inhalation of colimycin: 1000000 U twice a day during courses with colimycin Probenecid: 250-500 mglday orally in 24% of courses involving carbenicillin Intercurrent infections caused by other bacteria (e.g. Staph. aureus) were treated independently when necessary.
RESULTS Bacteriology A positive bacteriological effect was registered when the strain disappeared from sputum or tracheal secretion by microscopy as well as by culture at the end of the treatment. Table I shows the results. It is remarkable that the combination therapy of tobramycin and carbenicillin had effect in 72 % of the cases. In contrast, tobramycin alone had only effect in 26.6 %. There was no effect of colimycin or combination of colimycin and carbenicillin. Gentamicin was only given in 15 courses and in a much lower dosage. It can therefore not be compared with tobramycin in this material. There was no difference in the mean dosage and serum concentration of tobramycin when tobramycin was given alone or in combination with carbenicillin (Table 11). There was a slight preponderance of boys in the combination therapy group. Aerosol inhalation was given to the same proportion of pa-
Therapy of pseudomonas infections in cysticfibrosis
213
Scand J Infect Dis Downloaded from informahealthcare.com by University of Otago on 07/20/15 For personal use only.
Table I. The effect of therapy with different antibiotics on disappearance of mucoid Ps. aeruginosa from tracheal secretion in I90 courses to 49 patients with cysticfibrosis Antibiotics
No. of courses
Effect
No effect
Tobramycin Tobramycin + carbenicillin Gentamicin Colimycin Colimycin + carbenicillin Carbenicillin Gentamicin + carbenicillin Tobramycin + colimycin Tobramycin + carbenicillin+ colimycin
64 43 12 30 30 6 3 1 1
17 (26.6%) 31 (72.0%) 0 0 1 (3.3%) 1 1 0 0
47 (73.4%) 12 (28.0%) 12 (100%) 30 (100%) 29 (%.7%) 5 2 1 1
tients in the two groups. The frequency of other antibiotics and anabolic steroid given simultaneously was approximately the same. 18 patients received both tobramycin and combination of tobramycin and carbenicillin in subsequent courses. In total they had 35 courses of tobramycin and 29 courses of tobramycin and carbenicillin. The results were proportionally equivalent to those found in the total material. If the first courses of each of the treatments are compared there was no difference between the results of the two forms of therapy in 7 of 18 patients. In the other 11 patients a difference in the bacteriological effect was found as tobramycin alone had effect in only one patient whereas the combination therapy had effect in 10 patients (Table 111). This difference is statistically significant (P=0.05, sign test). Tobramycin was given as first course in 12 patients while 6 had Table 11. Distribution of sex, doses of antibiotics, serum concentrations and aerosol inhalations during courses of tobramycin and combination of tobramycin and carbenicillin
Femaleslmales Dosaee lmelkrrI24 h) No,of patients ' Mean
S.D. Serum conc. of tobramycin (pglml) No. of patients Mean S.D. Tobramycin inhalation (%)
Tobramvcin
Tobramycin and carbenicillin
42.2%/57.8%
34.9%/65.1%
64 9.72 1.405
10.27 1.141
59 3.64 1.878 54.7
the combination therapy primarily. The interval between the two courses was from 3 months to 3 years, mean 16 months.
Reinfection In 74.5% of the successful courses the patients were recolonized within 1 month. The rest were colonized within 5 months, except one who died 9 months after the treatment without harbouring Ps.ae. He had recurrent infections with H. influenzae and very high number of precipitins against this bacterium. In some cases rough strains of Ps.ae. were cultured at the recurrence, later on changing into mucoid strains. Serum precipitins Fig. 1 shows the number of precipitins against Ps.ae. in relation to the effect in 89 courses of
Table 111. I8 CF patients with chronic infection with mucoid Ps.ae. in the lower airways received subsequent courses of tobramycin given alone and in combination with carbenicillin The divergence (1 + 10= 1 1 cases) in the bacteriolorrical effect of tKe two treatments is significant in favour oT the combination therapy (P=0.05,sign test)
43
40 3.17 1.687
55.8
Tobramycin
345 120.8
No Tobramycin + carbenicillin Effect No effect Total
Effect
effect
Total
4 1 5
10 3 13
14 4 18
Sccind J Iigececr Dis I I
214
B. Friis
5c
w
A 0
Lo
.
0
Tobramycin
A
55
0
50
AAA
L5
0
Tobramycin and Carbeniclllin
A.
30 AAAA
: AA
20
A
fiA
AAA
3 A '
..
A
A
10
A
40
.-ccIn. ..-n u
35
a8
h
Scand J Infect Dis Downloaded from informahealthcare.com by University of Otago on 07/20/15 For personal use only.
AA
0 lobrmwm
o
Cwbmirdlm
* lobramyan
A
.
A
A
Fig. I . Number of serum precipitins against Ps.ae. in relation to the effect of chemotherapeutic courses with tobramycin and combination of tobramycin and carbenicillin.
0 L
225
f
= 2 0 15
I0
tobramycin and combination of tobramycin and carbenicillin. Combination therapy seemed to be the most effective, also in cases with many precipitins. The number of precipitins rose significantly from the first to the second course (P
Chemotherapy of Chronic Infections with Mucoid Pseudomonas aeruginosa in Lower Airways of Patients with Cystic Fibrosis BIRGITTE FRIIS
Scand J Infect Dis Downloaded from informahealthcare.com by University of Otago on 07/20/15 For personal use only.
From Childrens Department TG, Rigshospitalet, Copenhagen, and Department of Clinical Microbiology, Statens Serurninstitut, Hvidovre Hospital, Hvidovre, Denmark
ABSTRACT. The bacteriological effect of chemotherapy against Pseudomonas aeruginosa (Ps.ae.) in lungs of patients with cystic fibrosis is reviewed. During a 5-year period 49 children and adults were treated with 190 courses of different antibiotics. The mucoid strains of Ps.ae. disappeared in 72.0% of the courses in which a combination of tobramycin and carbenicillin was employed. Tobramycin given alone had only bacteriological effect in 26.6 % of the courses. Colimycin alone or in combination with carbenicillin had no effect. In 18 patients who received subsequent courses of tobramycin and combination of tobramycin and carbenicillin a significant difference in favour of the combination therapy was found, also in cases with many precipitins against Ps.ae. in serum. In 74.5% of the initially successful courses the patients were recolonized with Ps.ae. within 1 month. No nephrotoxic or ototoxic side effects were demonstrated in spite of the high doses of tobramycin (10 mg/kg/24 h) employed and the repeated courses.
INTRODUCTION
MATERIAL AND METHODS
During the last 10 years the more intensive general and antimicrobial therapy of children and adolescents with cystic fibrosis (CF) has prolonged their survival time remarkably. One of the greatest problems today in the care of C F is the chronic lung infections with Pseudomonas aeruginosa (Ps.ae.). A recent publication has shown that these infections are highly contributory to the bad prognosis as to life in C F (10). At the C F centre, Rigshospitalet, Copenhagen, C F patients with Ps.ae. infections have been treated very actively with antibacterial chemotherapy because earlier investigations in the centre demonstrated a strong correlation between a bad prognosis and a long duration and high intensity of the Ps.ae. infection. On the other hand we were not fully convinced of the antimicrobial effect of the different chemotherapeutic regimens. Earlier studies have shown different results in the clinical and bacteriological improvement after treatment of Ps.ae. infections in C F (4, 5, 6, 14, 15). With this background 190 courses of antipseudomonas chemotherapy given in a 5-year period from 1970 to 1975 were evaluated, especially with respect to the bacteriological effect of these courses.
49 patients, 22 girls and 27 boys, were treated during the 5-year period. All patients had a history typical for CF. The diagnosis was verified by finding of pathological sweat electrolytes on repeated pilocarpine iontophoresis tests according to Gibson and Cooke (3). They have all been followed monthly as outpatients and the tracheal secretions have been examined by microscopy and bacteriological cultivation on each visit. They were all chronically infected with mucoid Ps.ae. (9, 10). Age range at the first course was 1-26 years, mean 9 years. 190 courses were given in total. The patients received up to 14 subsequent courses of chemotherapy. 16 died during the examination period. All the patients were treated daily with intensive lung physiotherapy and aerosol inhalations with mucolytic agents at home. They were all admitted to the same ward during the antipseudomonas chemotherapy. Indication for antimicrobial therapy included that Ps.ae. had been cultivated as well as found by microscopy according to the principles described below. The antibiotic therapy was in each case decided by the clinical microbiologist in collaboration with the medical staff of the CF centre. Most of the patients were in bad general condition with severe productive cough and grave X-ray lung changes. The indication for initiating chemotherapy against Ps.ae. was a remarkable weight loss, growth stagnation or continuously decreasing lung function measured monthly by peak-flow and forced vital capacity (Vitalor). Later on, the indications for chemotherapy were somewhat changed because it was shown that a correlation
1 4- 791957
Scotid J Infecr Dis I I
212
B . Friis
Scand J Infect Dis Downloaded from informahealthcare.com by University of Otago on 07/20/15 For personal use only.
existed between the number of serum precipitins against Ps.ae. and a bad prognosis (10). Precipitins in blood above 10 or rapidly increasing became a criterion for treatment. Average age at the first colonization with Ps.ae. was 7.25f3.96 years (+S.D.) and at onset of chronic infection with mucoid strains of Ps.ae. 8.74f5.13 years. The average interval between chronic infection and first course of chemotherapy was 15.8f 15.0 months. Bacteriological examination Sputum or tracheal secretion obtained by aspiration through a catheter and collected in sterile tubes were brought to the laboratory in cooled boxes and examined immediately. The most purulent part of the material was examined in the following way: ( I ) Gram-stained films were examined by microscopy by a bacteriologist and the flora associated with areas consisting of respiratory epithelial cells and pus cells without squamous epithelial cells was described, whereas the flora associated with oral epithelial cells was considered to be predominantly of oral or pharyngeal origin, and was not described (8). (2) 4 pieces, the size of half a pea, of the material were inoculated on 2 blood agar plates and one differential medium plate for gram-negative rods. One of the blood agar plates was used for primary testing of antibiotic sensitivity (Neosensitabs@,Rosco A/S, TAstrup, Denmark). The other blood agar plate was furthermore inoculated with a streak of Staphylococcus aureus to enable the diagnosis of Haemophilus influenzae by means of the satellite phenomenon. The plates were inoculated for 20 h at 35°C. The cultivated bacteria were identified using standard methods and secondary sensitivity testing was performed on isolated strains of Ps.ae. Sputum or tracheal secretion was examined as described just before initiation of antimicrobial treatment, on the last day of treatment or 1 day after the treatment was stopped, and 1 month later. Serum concentration of antibiotics Serum tobramycin concentration was normally measured before and 2 h after intravenous or intramuscular injections on the 3rd and 7th day of treatment, sometimes more often. Absorption curves were made in a few patients (5). The mean values of all measurings at a given time were used. When a combination of tobramycin and carbenicillin was used, penicillinase was added before the analysis of serum tobramycin. Serum concentrations of the other antibiotics used were only studied when indicated for clinical reasons. Pseudoinonas aeruginosa precipitins The number of precipitins against Ps.ae. was measured by means of crossed immunoelectrophoresis as described elsewhere (9). Other investigations The urine was examined with microscopy and sticks for glucose and protein in all courses. All patients had normal serum creatinine or serum carbamide on several controls, but the values were not controlled systematically after antimicrobial treatment, except in a few patients (5). The Sctrritl J Iifi,ct Dis I I
liver enzymes were checked after many, although not all courses of therapy. All patients who received tobramycin and who were not too young to cooperate at the examination, were examined with audiometric and vestibular tests in connection with the course and at a follow-up several months later (19). Antibiotics Table I shows the different drugs and drug combinations used. Only antibiotics to which the bacteria were fully sensitive in vitro were given, except in a very few cases in which carbenicillin was included in a combination therapy in spite of a slightly reduced sensitivity. Doses +S.D./kg/ 24 h and duration of the courses were:
Tobramycin (Nebcin@): 9.95 mgf I .27 mg i.m. or i.v. for 10-14 days (10 mg/kg/24 h) Carbenicillin (Fugacillin@): 345 mgf 120.8 mg i.v. for 14 days (300-500 mg/kg/24 h) Colimycin (Colistin@):79000 Uf31000 U i.m. or i.v. for 10-14 days (50 000-90 000 U/kg/24 h) Gentamicin? 3.46 mg f 1.30 mg i.m. for 10-14 days (3-4 mg/kg/24 h) Aerosol inhalation of tobramycin 100 mg twice a day was given to 55 % of the patients treated with tobramycin. This treatment was used during the first part of the period but was stopped as it proved superfluous. Aerosol inhalation of colimycin: 1000000 U twice a day during courses with colimycin Probenecid: 250-500 mglday orally in 24% of courses involving carbenicillin Intercurrent infections caused by other bacteria (e.g. Staph. aureus) were treated independently when necessary.
RESULTS Bacteriology A positive bacteriological effect was registered when the strain disappeared from sputum or tracheal secretion by microscopy as well as by culture at the end of the treatment. Table I shows the results. It is remarkable that the combination therapy of tobramycin and carbenicillin had effect in 72 % of the cases. In contrast, tobramycin alone had only effect in 26.6 %. There was no effect of colimycin or combination of colimycin and carbenicillin. Gentamicin was only given in 15 courses and in a much lower dosage. It can therefore not be compared with tobramycin in this material. There was no difference in the mean dosage and serum concentration of tobramycin when tobramycin was given alone or in combination with carbenicillin (Table 11). There was a slight preponderance of boys in the combination therapy group. Aerosol inhalation was given to the same proportion of pa-
Therapy of pseudomonas infections in cysticfibrosis
213
Scand J Infect Dis Downloaded from informahealthcare.com by University of Otago on 07/20/15 For personal use only.
Table I. The effect of therapy with different antibiotics on disappearance of mucoid Ps. aeruginosa from tracheal secretion in I90 courses to 49 patients with cysticfibrosis Antibiotics
No. of courses
Effect
No effect
Tobramycin Tobramycin + carbenicillin Gentamicin Colimycin Colimycin + carbenicillin Carbenicillin Gentamicin + carbenicillin Tobramycin + colimycin Tobramycin + carbenicillin+ colimycin
64 43 12 30 30 6 3 1 1
17 (26.6%) 31 (72.0%) 0 0 1 (3.3%) 1 1 0 0
47 (73.4%) 12 (28.0%) 12 (100%) 30 (100%) 29 (%.7%) 5 2 1 1
tients in the two groups. The frequency of other antibiotics and anabolic steroid given simultaneously was approximately the same. 18 patients received both tobramycin and combination of tobramycin and carbenicillin in subsequent courses. In total they had 35 courses of tobramycin and 29 courses of tobramycin and carbenicillin. The results were proportionally equivalent to those found in the total material. If the first courses of each of the treatments are compared there was no difference between the results of the two forms of therapy in 7 of 18 patients. In the other 11 patients a difference in the bacteriological effect was found as tobramycin alone had effect in only one patient whereas the combination therapy had effect in 10 patients (Table 111). This difference is statistically significant (P=0.05, sign test). Tobramycin was given as first course in 12 patients while 6 had Table 11. Distribution of sex, doses of antibiotics, serum concentrations and aerosol inhalations during courses of tobramycin and combination of tobramycin and carbenicillin
Femaleslmales Dosaee lmelkrrI24 h) No,of patients ' Mean
S.D. Serum conc. of tobramycin (pglml) No. of patients Mean S.D. Tobramycin inhalation (%)
Tobramvcin
Tobramycin and carbenicillin
42.2%/57.8%
34.9%/65.1%
64 9.72 1.405
10.27 1.141
59 3.64 1.878 54.7
the combination therapy primarily. The interval between the two courses was from 3 months to 3 years, mean 16 months.
Reinfection In 74.5% of the successful courses the patients were recolonized within 1 month. The rest were colonized within 5 months, except one who died 9 months after the treatment without harbouring Ps.ae. He had recurrent infections with H. influenzae and very high number of precipitins against this bacterium. In some cases rough strains of Ps.ae. were cultured at the recurrence, later on changing into mucoid strains. Serum precipitins Fig. 1 shows the number of precipitins against Ps.ae. in relation to the effect in 89 courses of
Table 111. I8 CF patients with chronic infection with mucoid Ps.ae. in the lower airways received subsequent courses of tobramycin given alone and in combination with carbenicillin The divergence (1 + 10= 1 1 cases) in the bacteriolorrical effect of tKe two treatments is significant in favour oT the combination therapy (P=0.05,sign test)
43
40 3.17 1.687
55.8
Tobramycin
345 120.8
No Tobramycin + carbenicillin Effect No effect Total
Effect
effect
Total
4 1 5
10 3 13
14 4 18
Sccind J Iigececr Dis I I
214
B. Friis
5c
w
A 0
Lo
.
0
Tobramycin
A
55
0
50
AAA
L5
0
Tobramycin and Carbeniclllin
A.
30 AAAA
: AA
20
A
fiA
AAA
3 A '
..
A
A
10
A
40
.-ccIn. ..-n u
35
a8
h
Scand J Infect Dis Downloaded from informahealthcare.com by University of Otago on 07/20/15 For personal use only.
AA
0 lobrmwm
o
Cwbmirdlm
* lobramyan
A
.
A
A
Fig. I . Number of serum precipitins against Ps.ae. in relation to the effect of chemotherapeutic courses with tobramycin and combination of tobramycin and carbenicillin.
0 L
225
f
= 2 0 15
I0
tobramycin and combination of tobramycin and carbenicillin. Combination therapy seemed to be the most effective, also in cases with many precipitins. The number of precipitins rose significantly from the first to the second course (P
