Antibiotic concentrations in liquor compared to the minimal inhibitory concentrations of isolates in paediatric bacterial meningitis. The Finnish Study Group

The susceptibilities of 171 bacteria which caused meningitis in 200 children were tested for their susceptibility as minimal inhibitory concentrations (MICs) for the antibiotics used in therapy. These antibiotics were chloramphenicol, ampicillin, cefotaxime and ceftriaxone. The MICs were compared to the minimal concentrations of the drugs seen in the cerebrospinal fluid (CSF) samples. The minimal bacteriostatic capacity (lowest concentration in CSF/MIC) of both cephalosporins was superior to that of chloramphenicol and ampicillin. The correlation of the finding with the speed of liquor sterilization in the treatment groups is discussed.

Antibiotic susceptibility of type b Haemophilus influenzae and Streptococcus pneumoniae, and antibiotic concentration in cerebrospinal fluid

Antibiotic susceptibilities of 38 type b Haemophilus influenzae and 28 Streptococcus pneumoniae strains isolated from cerebrospinal fluid, blood and other specimens between 1973 and 1988 were studied. Minimal inhibitory concentrations (MICs) of ampicillin against 10 beta-lactamase positive and 28 negative H. influenzae isolates were 32-64 and 0.25 micrograms/ml, respectively. The MIC of chloramphenicol against one of the beta-lactamase positive H. influenzae strains was 8 but MICs against the rest of the organisms were 0.5-1 micrograms/ml. MICs of cefotaxime, ceftriaxone and cefuroxime against all H. influenzae strains were 0.016, 0.008 and 0.5 micrograms/ml, respectively. No S. pneumoniae isolates were resistant to penicillin G and MICs of this drug were 0.016-0.032 micrograms/ml. MICs of cefotaxime, cefriaxone and cefuroxime against all S. pneumoniae strains were 0.016-0.032, 0.016-0.032 and 0.032-0.063 micrograms/ml, respectively. MICs of chloramphenicol against 15, 4 and 9 of S. pneumoniae isolates were 2, 8 and 16 micrograms/ml, respectively. Antibiotic concentrations in the cerebrospinal fluid of patients with bacterial meningitis after intravenous administration of ampicillin (50-70 mg/kg x 4/day), penicillin G (31-63 mg/kg x 4/day), cefotaxime (50 mg/kg x 4/day) and chloramphenicol (25 mg/kg x 4/day) were 4.70 +/- 1.83 (n = 11), 0.57 +/- 0.32 (n = 7), 4.97 +/- 2.60 (n = 9) and 8.52 +/- 3.54 micrograms/ml (n = 3), respectively. The initial choice of antibiotics in older children with bacterial meningitis is a combination of ampicillin (75 mg/kg x 4/day) and cefotaxime (50 mg/kg x 4/day) to cover ampicillin-resistant H. influenzae, S. pneumoniae, and Listeria monocytogenes in Japan. These antibiotics should be changed according to the causative organisms and their antibiotic susceptibilities.

Routine methods in toxicology and therapeutic drug monitoring by high performance liquid chromatography. II. A rapid microscale method for determination of chloramphenicol in blood and cerebrospinal fluid

A highly sensitive, specific method for determining chloramphenicol levels in human blood plasma and cerebrospinal fluid is described. The method uses high performance liquid chromatography for the analysis, requires minimal quantities of patient specimen, and thus is suitable for use in newborn and pediatric patients. The method described in this article is specifically developed for routine use in laboratories engaged in therapeutic drug monitoring. It has advantages over other methods because it is less time consuming and can be used with commercially available controls.

Body fluid concentrations and pharmacokinetics of chloramphenicol given to mares intravenously or by repeated gavage

Serum concentrations and the pharmacokinetics of chloramphenicol were determined in 6 healthy mares after a single IV administration (50 mg/kg of body weight) or after the 1st and 5th sequential intragastric (IG) administration (50 mg/kg/6 hours) of chloramphenicol. Synovial fluid, peritoneal fluid, CSF, and urinary concentrations of chloramphenicol after the IG administrations also were determined. Mean (+/- SEM) overall elimination rate constant (K) values for the IV, 1st IG, and 5th IG dosages were 0.42 +/- 0.064/hr, 0.42 +/- 0.049/hr, and 0.29 +/- 0.074/hr, respectively, and were not significantly different from one another (P greater than 0.05). Bioavailability was 40 +/- 8.6% after the 1st IG administration and was 21 +/- 5.2% after the 5th IG administration. Values for the area under the curve (AUC) for the 1st and 5th IG dosages were significantly different from the AUC value for the IV dosage, and the AUC value for the 5th IG dosage was significantly different from that for the 1st IG dosage. Chloramphenicol was administered to 2 mares in 6 consecutive doses; the first and last doses were given IV and the others were given IG. Mean K values after the 2 IV doses were 0.38 +/- 0.112/hr and 0.56 +/- 0.078/hr, which were not significantly different from each other or from the mean value for the IV dosage given to all 6 mares. Absorption of chloramphenicol decreased with repeated IG administrations, resulting in lower concentrations of chloramphenicol with subsequent administrations. Five consecutive IG doses of chloramphenicol were administered to 4 of the mares in a separate experiment and did not alter intestinal xylose absorption.

Continuous sampling for determination of pharmacokinetics in rat cerebrospinal fluid

A method for determining drug concentration relationships between plasma and cerebrospinal fluid (CSF) in rats is described. Continuous CSF samples were collected directly from the third anterior ventricle with an indwelling cannula inserted through the bregma point, and drug concentrations were determined by high-pressure liquid chromatography and radioimmunoassay micromethods. Three antibiotics with different abilities to cross the blood-CSF barrier (chloramphenicol, piperacillin, and gentamicin) were tested. This method was found to be reproducible for each drug even if the antibiotic levels were low and the sample volumes very small. Peak CSF concentrations occurred between 0.75 and 1.25 h after injection for all three antibiotics. Percent penetration values at 1 h were 50, 1.2, and 5.4% for chloramphenicol, piperacillin, and gentamicin, respectively.

Sulbactam/ampicillin vs. chloramphenicol/ampicillin for the treatment of meningitis in infants and children

Eighty-one patients ages one month to 14 years with meningitis were randomized to receive either sulbactam (50 mg/kg per day) and ampicillin (400 mg/kg per day; 41 patients) or chloramphenicol and ampicillin (40 patients). The groups were comparable in terms of sex and degree of illness; however, more patients treated with chloramphenicol/ampicillin than patients treated with sulbactam/ampicillin were younger than 12 months of age (78% vs. 56%). Pathogens were isolated from the cerebrospinal fluid (CSF) of 65 (80%) of the 81 patients. In the sulbactam/ampicillin group, there were 18 Haemophilus influenzae isolates (one resistant to ampicillin), five Streptococcus pneumoniae, five Neisseria meningitidis, one Klebsiella pneumoniae, one Pseudomonas aeruginosa, and one Listeria. In the chloramphenicol/ampicillin group, there were 19 H. influenzae isolates, 10 S. pneumoniae, three N. meningitidis, one Haemophilus parainfluenzae, and one Citrobacter. Of 63 patients with assessable CSF pathogens, one (3%) of 29 treated with sulbactam/ampicillin died (S. pneumoniae) and six (18%) of 34 treated with chloramphenicol/ampicillin died (two, H. influenzae; three, S. pneumoniae; and one, Citrobacter). Twelve percent in the sulbactam/ampicillin group and 18% in the chloramphenicol/ampicillin group had neurologic sequelae. No clinically significant reactions or toxicities were noted. Sulbactam/ampicillin was as effective as chloramphenicol/ampicillin in the treatment of meningitis.

The monitoring of serum chloramphenicol levels in children with severe infections

Serum chloramphenicol levels were evaluated in 52 children with severe infection treated intravenously with chloramphenicol succinate and orally with chloramphenicol palmitate, chloramphenicol monostearoylglycolate or chloramphenicol in capsules. Effective serum levels were recorded with all chloramphenicol preparations. The variability was largest with chloramphenicol monostearoylglycolate. In a case of neonatal Escherichia coli meningitis good serum levels of chloramphenicol were achieved with chloramphenicol palmitate orally, supporting the view that oral chloramphenicol palmitate can be used to treat serious infections in this age group. Our data and those in the literature show that monitoring of serum chloramphenicol levels in neonates is necessary. After the neonatal period monitoring of serum chloramphenicol levels is useful in avoiding too high concentrations. On the other hand, toxic effects of high concentrations can be recognized from reticulocyte and haemoglobin, neutrophil and platelet counts, which should be performed every three to four days.

[20 years (1960-1980) of bacterial meningitis in childhood. I. Epidemiological and clinical data]

One hundred forty six children having suffered bacterial meningitis at any age have been collected and retrospectively studied over a period of a twenty year survey (1960 to 1980). In our series we can confirm the prominent epidemiologic and clinical features emerging from the literature. It is notewborty to outline the great percentage of cases lacking any bacterial identification due to previous antimicrobial treatment for parameningeal foci. This raises some major problems about early diagnosis, validity and signification of lumbar punctures, and usefulness of starting treatment with large spectrum antibiotics.

Gram-negative bacillary meningitis

The incidence of gram-negative bacillary meningitis has increased significantly in the past two decades. Approximately two thirds of all reported cases have occurred after neurosurgical procedures. With the development of the newer cephalosporins, the overall mortality rate has decreased from 40 to 80 per cent to 10 to 20 per cent.

Bactericidal versus bacteriostatic antibiotic therapy of experimental pneumococcal meningitis in rabbits

A rabbit model of pneumococcal meningitis was used to examine the importance of bactericidal vs. bacteriostatic antimicrobial agents in the therapy of meningitis 112 animals were infected with one of two strains of type III Streptococcus pneumoniae. Both strains were exquisitely sensitive to ampicillin, minimum inhibitory concentration (MIC)/minimum bactericidal concentration (MBC)<0.125 mug/ml. The activity of chloramphenicol against the two strains varied: strain(1)-MIC 2 mug/ml, MBC 16 mug/ml; strain(2)-MIC 1 mug/ml, MBC 2 mug/ml. Animals were treated with either ampicillin or chloramphenicol in dosages that achieved a peak bactericidal effect in cerebrospinal fluid (CSF) for ampicillin against both strains. Two different dosages were used for chloramphenicol. The first dosage achieved a peak CSF concentration of 4.4+/-1.1 mug/ml that produced a bacteriostatic effect against strain(1) and bactericidal effect against strain(2). The second dosage achieved a bactericidal effect against both strains (mean peak CSF concentration 30.0 mug/ml). All animals were treated intramuscularly three times a day for 5 d. CSF was sampled daily and 3 d after discontinuation of therapy for quantitative bacterial cultures. Results demonstrate that only antimicrobial therapy that achieved a bactericidal effect in CSF was associated with cure. Over 90% of animals treated with one of the bactericidal regimens (i.e., animals in which the bacterial counts in CSF dropped >5 log(10) colony-forming units [cfu]/ ml after 48 h) had sterile CSF after 5 d of treatment. On the other hand, the regimen that achieved bacteriostatic concentrations (CSF drug concentrations between the MIC and MBC) produced a drop of 2.4 log(10) cfu/ml by 48 h; however, none of the animals that survived had sterile CSF after 5 d. These studies clearly demonstrate in a strictly controlled manner that maximally effective antimicrobial therapy of experimental pneumococcal meningitis depends on achieving a bactericidal effect in CSF.

Efficacy of chloramphenicol in the treatment of neonatal and infantile meningitis: a study of 70 cases

The efficacy of chloramphenicol in the treatment of 21 neonates and 9 infants with proven meningitis and 37 neonates and 3 infants with suspected meningitis was evaluated from mortality and morbidity data, and by assay of the drug in serum and cerebrospinal fluid. Minimum inhibitory concentrations (MICs) were established for ten isolates. 25% of neonates and 50% of infants had subtherapeutic concentrations of chloramphenicol in serum or cerebrospinal fluid. Dosage was less than that currently recommended in over half of these subjects. Mild toxicity (reversible thrombocytopenia) was observed in only 1 of 20 babies being treated at the recommended dose. Toxic reactions, including the grey-baby syndrome, occurred in 10 babies receiving higher doses. In 4 cases, doses up to ten times that prescribed had been given, and death of 1 baby was attributable in part to chloramphenicol toxicity. 5 of 21 neonates and 1 of 9 infants with bacteriologically proven meningitis died, an overall mortality of 20%. Those infected with gram-negative bacteria had a higher mortality than those infected with gram-positive bacteria (p less than 0 . 05). 21% of the survivors had neurological sequelae. Therapeutic concentrations of chloramphenicol will be achieved in serum and cerebrospinal fluid with daily doses of 25 mg/kg in preterm and term infants during the first week of life and 37 . 5-50 mg/kg for older term babies. The drug should be assayed at 48-hour intervals, to maintain concentrations in the therapeutic, non-toxic range. Dosage should be increased when the peak serum concentration falls below 20 mg/l and decreased when the trough serum concentration exceeds 15 mg/l or the peak concentration exceeds 30 mg/l.

Intraventricular cerebrospinal fluid antibiotic concentrations in patients with intraventricular infections

The antibiotic concentration of the fluid from either lateral ventricle was determined 104 times in 37 patients through direct ventricular puncture, external ventricular drainage (EVD), or cerebrospinal fluid shunt sampling. The patients were 1 month to 12 years old. When the patients were receiving maximal intravenous antibiotic therapy alone, the concentrations for the most part were below 5 microgram/ml, whereas patients receiving an antibiotic through direct ventricular puncture, EVD, or a shunt reservoir usually had concentrations over 5 microgram/ml. However, wide variations from patient to patient were found with all forms of treatment despite similar dosages. Clustering of the concentration tended to occur in each individual patient. The authors conclude that, to obtain a high concentration of an antibiotic in the ventricular fluid, one should administer it directly into the ventricle.

Rapid bioassay for chloramphenicol in the presence of other antibiotics

A simple, rapid bioassay for the measurement of chloramphenicol in serum or cerebrospinal fluid was developed using a multiply antibiotic-resistant strain of Escherichia coli. The agar diffusion system involved the addition of patients' specimens and three standard concentrations of chloramphenicol to 7.5 mm diameter wells cut in agar seeded with the test organism. Assays of chloramphenicol using this system could be read routinely in three to four hours and allowed determinations of levels of 5-60 micrograms/ml. Chloramphenicol could be measured accurately in the presence of a variety of beta-lactam, aminoglycoside, sulfonamide, and tetracycline antimicrobial agents, but not cefoxitin or trimethoprim-sulfamethoxazole. Repetitive assays of sera containing known concentrations of chloramphenicol indicated a coefficient of variation of 8%. Seeded assay plates could be stored at 2-8 degrees C for up to five days prior to use.

[Chloramphenicol concentrations in serum and CSF in newborn infants and babies with bacterial meningitis (author's transl)]

Serum and CSF levels of chloramphenicol were determined repeatedly during the course of treatment in 24 premature and full term babies and infants with bacterial meningitis. Variations in chloramphenicol concentrations were caused in the premature babies by a small dose increase or by interaction with other drugs. In the fullterm newborn babies a higher dose could be given but even in these children the increase in concentration after small changes in dosage was marked. Chloramphenicol doses of 100 mg/kg daily could only be given after the sixth or eighth weeks of life. In the mature newborns and in the older babies inaccuracies in the administration of the drug may be a cause of marked variations of serum and CSF concentrations.

Pharmacokinetic comparison of intravenous and oral chloramphenicol in patients with Haemophilus influenzae meningitis

The pharmacokinetics of chloramphenicol following intravenous and oral administration were studied in 14 infants with Haemophilus influenzae meningitis. Following five days of treatment with intravenous chloramphenicol (100 mg/kg/day every six hours), oral chloramphenicol was substituted at the same dose. Multiple serum levels of chloramphenicol were determined after an intravenous dose on day 4 and after an oral dose on day 10. CSF levels were measured six hours after intravenous or oral chloramphenicol dose on those days (CSF trough). Following intravenous administration, the mean peak serum level of 15.0 micrograms/ml was reached at 45 minutes. In comparison, after oral chloramphenicol in the same dosage, the mean peak serum level of 18.5 micrograms/ml was achieved at two to three hours. The mean serum half-life of the drug (6.5 hours) was significantly longer after oral administration than after intravenous chloramphenicol (4.0 hours) (P less than .001). The increased serum half-life following orally administered chloramphenicol was occasionally associated with drug accumulation. In addition, mean trough CSF levels were somewhat higher when the patient received oral medication (6.6 micrograms/ml) compared to intravenous administration (4.2 micrograms/ml) (P less than .001). For any treatment regimen for H influenzae meningitis that includes a period of oral chloramphenicol therapy the patient should be hospitalized to ensure compliance. Because of the wide range of individual variation in serum half-life that may result in accumulation, periodic monitoring of serum chloramphenicol levels is also recommended.

Evaluation of certain current chloramphenicol assay methods

The rapid assay of chloramphenicol in serum and cerebrospinal fluid is an important service to the clinician, especially in the paediatric setting. More than 3 years' experience in the use of chloramphenicol transacetylase in conjunction with 14C-acetyl co-enzyme A has proved that this is an accurate method.

[Penetration of antibiotics into the cerebrospinal fluid. Theoretical and practical aspects (author's transl)]

The penetration of antibiotics into the cerebrospinal fluid (CSF) is governed by several factors, including lipid solubility, degree of ionization, pH gradient, serum protein binding, molecular weight and structure and, above all, degree of meningeal inflammation. The distribution within, and elimination from the subarachnoid spaces depend upon the physiology of the CSF and the route of administration of the drug. CSF penetration studies carried out in patients or volunteers with various groups of antibiotics and antibacterial agents have shown that, except for chloramphenicol and co-trimixazole, passage through the blood-brain barrier is generally poor. However, when the meninges are inflamed therapeutically active concentrations of penicillins and of some cephalosporins can be measured in the CSF.

[Cerebrospinal fluid (CSF) levels of chloramphenicol: a critical review (author's transl)]

Chloramphenicol was given to four neurosurgical patients intravenously. Serum and cerebrospinal fluid (CSF) concentrations were determined with High-pressure-Liquid-Chromatography (HPLC). A comparison with results reported in the literature shows that in patients without meningitis there is markedly lower concentration of chloramphenicol in the CSF than expected, i.e. it penetrates the "blood-brain" barrier much less easily.

Chloramphenicol levels in cerebrospinal fluid in meningitis

Chloramphenicol was found to cross the blood-brain barrier into the cerebrospinal fluid of children with pyogenic meningitis effectively both at days 2 and 10 of therapy. It is recommended as the drug of choice in the treatment of children with Haemophilus influenzae meningitis.

Comparison of cotrimoxazole, ampicillin, and chloramphenicol in treatment of experimental Haemophilus influenzae type B meningitis

To evaluate cotrimoxazole in the treatment of bacterial meningitis, we compared its action with that of ampicillin and chloramphenicol in experimental Haemophilus influenzae type b meningitis. Both trimethoprim and sulfamethoxazole penetrated well into the cerebrospinal fluid of infected rabbits, reaching 40 and 26%, respectively, of their simultaneous serum levels. Levels measured 30 and 60 min after intravenous injection exceeded the minimum inhibitory concentration of this combination for H. influenzae by 10- to 100-fold. The mean ratio of trimethoprim to sulfamethoxazole in cerebrospinal fluid was 1:22. Cotrimoxazole was as effective as ampicillin in therapy of beta-lactamase-negative H. influenzae meningitis and as effective as chloramphenicol for a beta-lactamase positive strain. These findings corroborate favorable preliminary clinical experience reported by others and indicate that cotrimoxazole deserves further study in the therapy of bacterial meningitis.

Comparison of cefamandole, cephalothin, ampicillin, and chloramphenicol in experimental Escherichia coli meningitis

The activities of cefamandole, cephalothin, ampicillin, and chloramphenicol were compared in fulminant and temperate Escherichia coli meningitis in rabbits. Intensive dosing schedules were employed to achieve maximal therapeutic benefits with short-term treatment. In an 8-h schedule chloramphenicol was significantly more effective in sterilizing the cerebrospinal fluid and curing both fulminant and temperate infections than cefamandole or ampicillin. Cephalothin was without effect in fulminant meningitis. Cefamandole and ampicillin were equivalent in activity in this and longer (12- and 24-hr) treatment schedules. The therapeutic benefits of chloramphenicol were purchased via use of doses above those generally regarded as safe for human use. The mean serum, cerebrospinal fluid, and brain concentrations of chloramphenicol, cefamandole, and ampicillin were significantly greater in rabbits with fulminant meningitis than in those with temperate meningitis. The difference was of such magnitude as to support the need to monitor drug concentrations.

Chloramphenicol disposition in infants and children

We studied the pharmacokinetics of chloramphenicol following its intravenous administration as the sodium succinate ester to 54 infants and children. The mean "apparent t 1/2" of chloramphenicol clearance from the plasma was 5.94 hours (range 0.87 to 17.8 hours). The t 1/2 of patients who weighed less than 10 kg was significantly longer than that of those who weighed more than 10 kg (9.02 vs 4.55 hours; P less than 0.0001). There was a discrepancy between the plasma levels of chloramphenicol and total aromatic nitro compounds in four patients; these compounds were cleared at different rates. Repeated dosing (eight to 30 doses) did not produce a consistent effect on the t 1/2. The CSF concentrations of chloramphenicol in 13 patients were 67% (range 45 to 99%) of the simultaneous serum concentrations. We conclude that the marked individual variation in chloramphenicol pharmacokinetics in infants and children requires monitoring of blood concentrations during therapy.

Ventricular fluid levels of chloramphenicol in infants

Investigation of two infants suggested that peak ventricular fluid levels of chloramphenicol after intravenous administration were achieved after 3 h. The penetration of chloramphenicol into ventricular fluid may be unpredictable. In one patient, the peak ventricular fluid level was 57.5% of the peak serum level, and it was only 22.5% in the other patient. This observation may explain some of the past treatment failures when chloramphenicol was utilized in patients with gram-negative ventriculitis and meningitis.

Central nervous system chloramphenicol concentration in premature infants

Four premature infants under 1,500 g were treated with parenteral chloramphenicol for central nervous system infections due to organisms resistant to the penicillins. Serum, cerebrospinal fluid (CSF), and ventricular fluid concentrations of chloramphenicol were measured frequently during therapy and were used to maintain drug dosages in the safe and therapeutic range. Concentrations of chloramphenicol in the lumbar CSF and ventricular fluid had a mean of 23.3 +/- 7.7 micrograms/ml, consistently greater than 45% of peak serum levels. These data show that chloramphenicol enters the CSF in both ventricular and lumbar regions in therapeutic concentrations when administered intravenously. The clinical usefulness of this drug remains to be investigated. The importance of monitoring serum drug levels during therapy is emphasized.

[Local use of levomycetin in neurosurgery]

A method of local application of the unsoluble medicinal form of chloramphenicol during operations has been developed. Administration of the preparation in a dose up to 2g under an osteal or cutaneous graft induces no side complications and forms a depot of the agent therapeutic concentrations for more than 8 days. There exists a direct relationship between the increasing of staphylococcus resistance to chloramphenicol and the quantity of the agent used in patients.

Studies on the concentrations of chloramphenicol in the serum and cerebrospinal fluid of neonates, infants, and small children. Reciprocal reactions between chloramphenicol, penicillin and phenobarbitone

The interactions between chloramphenicol, penicillin and phenobarbitone were investigated in 383 children (premature and neonate children, infants and small children). As expected, the chloramphenicol concentrations in the serum of the newborns was considerably higher than that of infants and small children with the same dosage of chloramphenicol. In the age group of the premature and newborn children and infants there was significantly higher total chloramphenicol concentrations with the chloramphenicol-penicillin combination than with chloramphenicol monotherapy. Addition of phenobarbitone to the combination significantly reduced the chloramphenicol concentrations in the neonates. Lowering of the serum chloramphenicol concentrations by phenobarbitone could not be statistically confirmed in the infant age group. Combinations of chloramphenicol with ampicillin, gentamycin or cephalosporin derivatives showed no influence on serum chloramphenicol concentrations. Transference of chloramphenicol from the serum to the cerebrospinal fluid was about twice as high in the acute inflammatory stage as when the meninges were no longer acutely diseased (60 and 30% respectively of the serum concentrations). The passage of chloramphenicol to the cerebrospinal fluid showed no dependence on age.

[Therapy of meningitis (author's transl)]

For the specific and unspecific treatment of purulent meningitis, penicillin, ampicillin and chloramphenicol are usually sufficient. Only resistant pathogens (Klebsiella, Pseudomonas among others) and meningitis in infants require other treatment. Cephalosporins and aminoglycosides only come into consideration for the treatment of very rare exceptional cases. The still relatively poor prognosis of purulent meningitis is largely independent of the efficacy of the treatment. Delay in beginning therapy, lack of intensive care, primary diseases and complications are principally responsible for this. Medicamentous prophylaxis is only possible for meningococcal meningitis. The distribution of antibiotics in the CSF is irregular. Also the antibacterial activity in the CSF is different from culture media.