THE JOURNAL OF INFECTIOUS DISEASES. VOL. 135. NO.2. FEBRUARY 1977
© 1977 by the University of Chicago. All rights reserved.
Cefamandole and Ampicillin Therapy in Experimental Haemophllus influenzae Meningitis Larry J. Strausbaugh, Christopher D. Mandaleris, and Merle A. Sande
From the Department of Internal Medicine, University of Virginia School of Medicine, Charlottesville, Virginia
=
fluenzae [11]. and its potency against ampicillinresistant strains is equal to that of chloramphenicol. MICs of both drugs are between 0.5 p.gjml and 2.0 p.gjml [10]. The purpose of this study was to evaluate cefamandole and ampicillin in a rabbit model of H. influenzae meningitis. The specific goals of this investigation were to compare (1) the penetration of cefamandole and ampicillin into the cerebrospinal fluid (CSF) in experimental H. infiuenzae meningitis; (2) the bacterial killing by cefamandole and ampicillin in vivo; and (3) the therapeutic efficacy of cefamandole and ampicillin in experimental H. infiuenzae meningitis.
Several case reports of meningitis caused by ampicillin-resistant strains of Haemophilus infiuenzae have been published within the last two years [1-6], and 41 cases from 23 states were confirmed by the Center for Disease Control (CDC; Atlanta, Ga.) between February 1974 and June 1975 [7]. Recommendations for treatment of meningitis in children have changed accordingly [8-9] and now include chloramphenicol, which is uniformly active against ampicillin-resistant strains of H. infiuenzae [10]. Although the ultimate significance of these ampicillin-resistant strains is unknown, an alternative to chloramphenicol is desired because of its toxicity. Cefamandole, a new cephalosporin derivative, is exceptionally active against H. in-
Materials and Methods
Received for publication December 29, 1975, and in revised form April 5, 1976. The data in this paper were presented in part at the 15th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, D.C., September 24-26.1975. We thank Ms. Kip B. Courtney for technical assistance and Ms. Cathy Crider for assistance in the preparation of the manuscript. Please address requests for reprints to Dr. Merle A. Sande, University of Virginia School of Medicine, P.O. Box 251, Charlottesville, Virginia 22901.
Bacteria. The strain of H. infiuenzae used was type B Eagan (obtained through the courtesy of Dr. Arnold L. Smith, Children's Hospital, Boston, Mass.). The clinical source, date, and place of its isolation have been described previously [12]. The MBC, as determined by microdilution studies in broth, was 0.5 p.gjml for cefamandole and 0.12 p.gjml for ampicillin. Bacteria were grown in brain-heart infusion
210
Downloaded from http://jid.oxfordjournals.org/ at East Carolina University on June 30, 2015
Cefamandole and ampicillin were compared in the therapy of experimental Haemophilus influenzae meningitis in rabbits. Three dosages of each drug were administered as a continuous intravenous infusion for 8 hr to 24 infected animals. Samples of serum and cerebrospinal fluid (CSF) were obtained at 0, 4, and 8 hr for determination of antibiotic concentrations and bacterial titers in CSF. Serum levels of cefamandole were higher, but CSF concentrations of both antibiotics were similar. With the 60mg/kg per hr dose, the mean serum level was 106 ± 61 p.g/ml for ce£amandole and 58 ± 32 p.g/ml for ampicillin (P < 0.05). With this dosage the mean level in CSF was 7.3 ± 8,4 p.gjml for cefamandole and 9.5 ± 5,4 p.gjml for ampicillin (P 0.26). The percentage penetration ([concentration in CSF jconcentration in serum] X 1.00%) was higher for ampicillin (mean, 18.8% ± 8.9%) than for cefamandole (mean, 5.6% ± 3.8%) with all dosages tested (P < 0.001). The rate of bacterial killing in vivo during therapy was similar with both drugs. The efficacy of ce£amandole and ampicillin given intramuscularly for five days (250 mg every 8 hr) was examined in 42 rabbits. Twelve of 14 untreated control rabbits died within 24-72 hr of inoculation. In contrast, II of 14 rabbits treated with cefamandole and 10 of 14 rabbits treated with ampicillin were cured of their infections. Cefamandole compared favorably with ampicillin in the therapy of experimental H. infiuenzae meningitis.
211
Cejamandole and Ampicillin in Meningitis
suspension of H. influenzae. The suspension was obtained from a 16-hr culture and contained 9.0 loglo dujml. After inoculation, the cisternal needle was withdrawn, and the animal was removed from the frame and returned to its cage. Within 6 hr the rabbits had meningitis characterized by lethargy, temperature of >40 C, CSF leukocytosis (20-1,500 white blood cells/mm": 90~~ polymorphonudear neutrophils), and bacterial counts in CSF of 4.0-8.0 loglo dujml. If left untreated, 85% of these animals died within 24-72 hr. Penetration into CSF and bacterial killing in vivo. Six to eight hours after inoculation, rabbits were anesthetized iv with 30 mg of sodium pentobarbitol (Barber Veterinary Supply Co., Richmond, Va.). Indwelling femoral arterial and venous catheters (Intramedice polyethylene tubing 7420, Clay-Adams, Parsipanny, N.J.) were placed, and the animals were resuspended in the stereotaxic frame with the spinal needle again positioned in the cisterna magna. The animals were then treated with antibiotics for 8 hr. Cefamandole (supplied by Eli Lilly and Company. Indianapolis, Ind.) and ampicillin (Omnipen®N, Wyeth Laboratories, Philadelphia, Pa.) were administered with a syringe infusion pump (model 351, Sage Instruments, White Plains, N.Y.) via the femoral venous catheter at a constant rate during the treatment period according to the
Table 1. Penetration of cefamandole and ampicillin into the cerebrospinal fluid (CSF) of rabbits with Haemophilus influenzae meningitis. Cefamandole
Ampicillin Dose (mg/ kg per hr], time of No. of sample* animals 15 4 hr 8 hr 30 4 hr 8 hr 60 4 hr 8 hr 90 4 hr 8 hr
Mean concentration ± SD (J,Lg/ml) CSF
Serum
Percentage penetration t
2.4 ± 0.6 8.4 ± 4.3
20 ±6 37 ± 13
12 23
4.2 ± 3.2 19 ± 21
44 ±49 49 ±45
10 38
6.8 ± 3.4 13 ± 5.9
42 ± 18 81 ± 36
16 15
No. of animals
Mean concentration ± SD (J,Lg/ml) Serum
Percentage penetrationt
0.6 ± 0.5 2.7 ± 0.9
46 ±20 83 ±39
1.3 3.2
3.4 ± 1.0 11 ±ll
101 ± 45 111 ± 80
3.4 9.9
8.5 ± 8.3 9.3 ± 4.9
119 ± 61 130 ± 50
7.1 7.2
CSF
4
3
3
6
4
3
*Antibiotics were given by constant iv infusion for 8 hr. Sampling times are given as hr after initiation of infusion. t(Mean drug concentration in CSF /mean concentration in serum) X100%.
Downloaded from http://jid.oxfordjournals.org/ at East Carolina University on June 30, 2015
broth (Difco, Detroit, Mich.) supplemented after autoclaving with nicotinamide adenine dinucleotide and heme (to a volume of 1/300th of the broth volume). Nicotinamide adenine dinucleotide (grade III; Sigma Chemical Co., St. Louis, Mo.j was dissolved in distilled water (1 mg/rnl) and sterilized by filtration through a Millipore membrane filter (pore size, 0.45 ps»; Millipore Corp., Bedford, Mass.) on the day of use. Heme was prepared by addition of one volume of rabbit blood to two volumes of distilled water, centrifugation at 20,000 g for 30 min at 4 C, and filter sterilization. Broth was made monthly and was stored without loss of activity at 4 C. The rabbit model. New Zealand white rabbits weighing 2 kg were prepared according to the method of Dacey and Sande [13]. In brief, a helmet made of dental acrylic was first attached to the animal's skull. This helmet prevented movement of the rabbit's head in a modified stereotaxic frame. Then a spinal needle (25gauge, 3.5 inches) mounted to the frame's geared electrode introducer could be guided through the posterior alantooccipital membrane and positioned in the cisterna magna. The cisternal needle was used initially for inoculation and subsequently for sampling of CSF. Inoculation was accomplished by withdrawal of 0.5 ml of CSF and then injection of 0.3 ml of a
212
one of three treatment groups: control (no treatment), cefamandole, or ampicillin. Therapy was started 6 hr after inoculation. Cefamandole and ampicillin were administered im in a dose of 250 mg every 8 hr for five days. On the day after therapy was stopped, all animals had a cisternal tap to obtain CSF for culture. The animals were then observed for five days. CSF for culture was obtained from three control animals
© 1977 by the University of Chicago. All rights reserved.
Cefamandole and Ampicillin Therapy in Experimental Haemophllus influenzae Meningitis Larry J. Strausbaugh, Christopher D. Mandaleris, and Merle A. Sande
From the Department of Internal Medicine, University of Virginia School of Medicine, Charlottesville, Virginia
=
fluenzae [11]. and its potency against ampicillinresistant strains is equal to that of chloramphenicol. MICs of both drugs are between 0.5 p.gjml and 2.0 p.gjml [10]. The purpose of this study was to evaluate cefamandole and ampicillin in a rabbit model of H. influenzae meningitis. The specific goals of this investigation were to compare (1) the penetration of cefamandole and ampicillin into the cerebrospinal fluid (CSF) in experimental H. infiuenzae meningitis; (2) the bacterial killing by cefamandole and ampicillin in vivo; and (3) the therapeutic efficacy of cefamandole and ampicillin in experimental H. infiuenzae meningitis.
Several case reports of meningitis caused by ampicillin-resistant strains of Haemophilus infiuenzae have been published within the last two years [1-6], and 41 cases from 23 states were confirmed by the Center for Disease Control (CDC; Atlanta, Ga.) between February 1974 and June 1975 [7]. Recommendations for treatment of meningitis in children have changed accordingly [8-9] and now include chloramphenicol, which is uniformly active against ampicillin-resistant strains of H. infiuenzae [10]. Although the ultimate significance of these ampicillin-resistant strains is unknown, an alternative to chloramphenicol is desired because of its toxicity. Cefamandole, a new cephalosporin derivative, is exceptionally active against H. in-
Materials and Methods
Received for publication December 29, 1975, and in revised form April 5, 1976. The data in this paper were presented in part at the 15th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, D.C., September 24-26.1975. We thank Ms. Kip B. Courtney for technical assistance and Ms. Cathy Crider for assistance in the preparation of the manuscript. Please address requests for reprints to Dr. Merle A. Sande, University of Virginia School of Medicine, P.O. Box 251, Charlottesville, Virginia 22901.
Bacteria. The strain of H. infiuenzae used was type B Eagan (obtained through the courtesy of Dr. Arnold L. Smith, Children's Hospital, Boston, Mass.). The clinical source, date, and place of its isolation have been described previously [12]. The MBC, as determined by microdilution studies in broth, was 0.5 p.gjml for cefamandole and 0.12 p.gjml for ampicillin. Bacteria were grown in brain-heart infusion
210
Downloaded from http://jid.oxfordjournals.org/ at East Carolina University on June 30, 2015
Cefamandole and ampicillin were compared in the therapy of experimental Haemophilus influenzae meningitis in rabbits. Three dosages of each drug were administered as a continuous intravenous infusion for 8 hr to 24 infected animals. Samples of serum and cerebrospinal fluid (CSF) were obtained at 0, 4, and 8 hr for determination of antibiotic concentrations and bacterial titers in CSF. Serum levels of cefamandole were higher, but CSF concentrations of both antibiotics were similar. With the 60mg/kg per hr dose, the mean serum level was 106 ± 61 p.g/ml for ce£amandole and 58 ± 32 p.g/ml for ampicillin (P < 0.05). With this dosage the mean level in CSF was 7.3 ± 8,4 p.gjml for cefamandole and 9.5 ± 5,4 p.gjml for ampicillin (P 0.26). The percentage penetration ([concentration in CSF jconcentration in serum] X 1.00%) was higher for ampicillin (mean, 18.8% ± 8.9%) than for cefamandole (mean, 5.6% ± 3.8%) with all dosages tested (P < 0.001). The rate of bacterial killing in vivo during therapy was similar with both drugs. The efficacy of ce£amandole and ampicillin given intramuscularly for five days (250 mg every 8 hr) was examined in 42 rabbits. Twelve of 14 untreated control rabbits died within 24-72 hr of inoculation. In contrast, II of 14 rabbits treated with cefamandole and 10 of 14 rabbits treated with ampicillin were cured of their infections. Cefamandole compared favorably with ampicillin in the therapy of experimental H. infiuenzae meningitis.
211
Cejamandole and Ampicillin in Meningitis
suspension of H. influenzae. The suspension was obtained from a 16-hr culture and contained 9.0 loglo dujml. After inoculation, the cisternal needle was withdrawn, and the animal was removed from the frame and returned to its cage. Within 6 hr the rabbits had meningitis characterized by lethargy, temperature of >40 C, CSF leukocytosis (20-1,500 white blood cells/mm": 90~~ polymorphonudear neutrophils), and bacterial counts in CSF of 4.0-8.0 loglo dujml. If left untreated, 85% of these animals died within 24-72 hr. Penetration into CSF and bacterial killing in vivo. Six to eight hours after inoculation, rabbits were anesthetized iv with 30 mg of sodium pentobarbitol (Barber Veterinary Supply Co., Richmond, Va.). Indwelling femoral arterial and venous catheters (Intramedice polyethylene tubing 7420, Clay-Adams, Parsipanny, N.J.) were placed, and the animals were resuspended in the stereotaxic frame with the spinal needle again positioned in the cisterna magna. The animals were then treated with antibiotics for 8 hr. Cefamandole (supplied by Eli Lilly and Company. Indianapolis, Ind.) and ampicillin (Omnipen®N, Wyeth Laboratories, Philadelphia, Pa.) were administered with a syringe infusion pump (model 351, Sage Instruments, White Plains, N.Y.) via the femoral venous catheter at a constant rate during the treatment period according to the
Table 1. Penetration of cefamandole and ampicillin into the cerebrospinal fluid (CSF) of rabbits with Haemophilus influenzae meningitis. Cefamandole
Ampicillin Dose (mg/ kg per hr], time of No. of sample* animals 15 4 hr 8 hr 30 4 hr 8 hr 60 4 hr 8 hr 90 4 hr 8 hr
Mean concentration ± SD (J,Lg/ml) CSF
Serum
Percentage penetration t
2.4 ± 0.6 8.4 ± 4.3
20 ±6 37 ± 13
12 23
4.2 ± 3.2 19 ± 21
44 ±49 49 ±45
10 38
6.8 ± 3.4 13 ± 5.9
42 ± 18 81 ± 36
16 15
No. of animals
Mean concentration ± SD (J,Lg/ml) Serum
Percentage penetrationt
0.6 ± 0.5 2.7 ± 0.9
46 ±20 83 ±39
1.3 3.2
3.4 ± 1.0 11 ±ll
101 ± 45 111 ± 80
3.4 9.9
8.5 ± 8.3 9.3 ± 4.9
119 ± 61 130 ± 50
7.1 7.2
CSF
4
3
3
6
4
3
*Antibiotics were given by constant iv infusion for 8 hr. Sampling times are given as hr after initiation of infusion. t(Mean drug concentration in CSF /mean concentration in serum) X100%.
Downloaded from http://jid.oxfordjournals.org/ at East Carolina University on June 30, 2015
broth (Difco, Detroit, Mich.) supplemented after autoclaving with nicotinamide adenine dinucleotide and heme (to a volume of 1/300th of the broth volume). Nicotinamide adenine dinucleotide (grade III; Sigma Chemical Co., St. Louis, Mo.j was dissolved in distilled water (1 mg/rnl) and sterilized by filtration through a Millipore membrane filter (pore size, 0.45 ps»; Millipore Corp., Bedford, Mass.) on the day of use. Heme was prepared by addition of one volume of rabbit blood to two volumes of distilled water, centrifugation at 20,000 g for 30 min at 4 C, and filter sterilization. Broth was made monthly and was stored without loss of activity at 4 C. The rabbit model. New Zealand white rabbits weighing 2 kg were prepared according to the method of Dacey and Sande [13]. In brief, a helmet made of dental acrylic was first attached to the animal's skull. This helmet prevented movement of the rabbit's head in a modified stereotaxic frame. Then a spinal needle (25gauge, 3.5 inches) mounted to the frame's geared electrode introducer could be guided through the posterior alantooccipital membrane and positioned in the cisterna magna. The cisternal needle was used initially for inoculation and subsequently for sampling of CSF. Inoculation was accomplished by withdrawal of 0.5 ml of CSF and then injection of 0.3 ml of a
212
one of three treatment groups: control (no treatment), cefamandole, or ampicillin. Therapy was started 6 hr after inoculation. Cefamandole and ampicillin were administered im in a dose of 250 mg every 8 hr for five days. On the day after therapy was stopped, all animals had a cisternal tap to obtain CSF for culture. The animals were then observed for five days. CSF for culture was obtained from three control animals
