[An example of simulation for a better understanding of PK/PD relationship of antibiotics]. / Un exemple de modélisation permettant une meilleure compréhension de la relation PK/PD des antibiotiques.

The interpretation of PK/PD indices is specific to each class of antibiotics. In order to illustrate this, we developed a multidisciplinary tutorial program based on simulation of clinical cases. Three drugs were included in this software: tobramycin, vancomycin and azithromycin. From the dosage regimen proposed by the user, the model simulates a plotting of antibiotic plasma concentrations vs. time (tobramycin, vancomycin and azithromycin) and tissue concentrations (azithromycin). Peak and trough concentrations are calculated at steady-state. A commentary is provided to evaluate the efficacy of treatment and to assist the user in improving his prescription of tobramycin or vancomycin. T(> MIC) (time the concentration remains above the MIC) and AUC(24) (area under the concentration-time curve) are calculated in plasma and tissues for azithromycin. In order to create a link between theoretical pharmacokinetics and clinical practice, we propose this model as a simulation of antibiotic monitoring. We put the emphasis on interactivity and simulation, leading to applied reasoning and decision making. It illustrates (i) the influence of pharmacokinetic parameters, location of infection and bactericidal kinetics on the use of three different classes of antibiotics, (ii) the role of route of administration, dosing and intervals between administrations on therapeutic response and (iii) the influence of erratic administrations on clinical efficacy.

Pharmacokinetic parameters of vancomycin for therapeutic regimens in neutropenic adult patients.

Vancomycin (V) is widely used in neutropenic patients, though its kinetics are known in this type of patient. In the present study, ten patients were included: all of them received an intensive therapy for non-Hodgkin malignant lymphoma, Hodgkin disease, myeloma, acute leukemia, followed by an autologous bone marrow transplantation in 6 cases. All patients were neutropenic (100/mm3). The pharmacokinetic study was done at the first V administration: 1000 mg V were injected as a 1-h infusion. Plasma V concentrations were measured by an enzyme immunoassay (EMIT, Syva, France). V maximal and minimal concentrations were 61.3 +/- 38.6 micrograms/ml and 1.69 +/- 0.77 microgram/ml, respectively. Total V clearance was 158 +/- 51 ml/min, with a creatinine clearance of 141.2 +/- 36.2 ml/min on test day. V plasma kinetics can be described by a biexponential model, with the following parameters: [table: see text] These data show a 3-fold increase of initial volume of distribution and a shortened (3-fold) T1/2 beta, if compared to values obtained in normal subjects. Because the bactericidal effect is time dependent, there can be a risk of insufficient antibiotic effect throughout the day. Our data suggest that new therapeutic regimens are needed for these patients.

Teaching individualized antibiotic dosage regimens by means of two computer-assisted learning programs.

We developed two multidisciplinary tutorial programs (TOBRA-DIDACT and VANCO-DIDACT) for teaching the basic principles of antibiotic drug monitoring by simulation of repeated administrations to fictitious patients whose physio-pathologic characteristics were pre-defined in the programs. To illustrate the two types of bactericidal kinetics, we have chosen one time-dependent (vancomycin) and one concentration-dependent (tobramycin) antibiotic. These computer-assisted programs operate on an interactive mode. In each of them, three main steps are connected: (1) Various types of clinical cases are submitted to the student: for each of them, case report includes clinical characteristics, location of infection, bacterial strain and minimal bactericidal concentration. These data must be taken into account during the following steps. (2) The student has to establish the treatment schedule: route of administration, dose for each injection, intervals between injections and duration of infusion. (3) The result of the dosage scheme proposed by the student is represented by a simulation of plotting antibiotic plasma concentrations vs. time during the first 4 days of treatment. These curves are obtained by a monoexponential (TOBRA-DIDACT) or biexponential (VANCO-DIDACT) pharmacokinetic model. Peak and trough concentrations are calculated at steady-state. An expert system provides a commentary with each result to evaluate the efficacy of the treatment and to assist the student in improving his prescription. TOBRA-DIDACT and VANCO-DIDACT illustrate the influence of age, obesity, renal impairment, location of infection and bacterial strain on antibiotic therapy. They also show the role of route of administration, dosing and intervals between injections on therapeutic response.(ABSTRACT TRUNCATED AT 250 WORDS)

Application of USC*PACK clinical programs to vancomycin in neutropenic patients.

The pharmacokinetics of vancomycin were studied in 10 neutropenic patients (4 male, 6 female) using the USC*PACK Clinical Programs. The experimental data was determined after the first administration of 1000 mg injected as a 1-h infusion. Eight blood samples were collected between 15 min and 11 h after the end of the infusion. Plasma vancomycin concentrations were measured by immunoassay procedure. Creatinine clearance and urine flow were also measured. Pharmacokinetic parameters were computed using a two-compartment model: Vc = 0.270665 +/- 0.161033 (l.kg-1); Kcp = 0.732927 +/- 0.464449 (h-1); Ks = 0.004952 +/- 0.00272 (min.ml-1.h-1); Kpc = 0.470243 +/- 0.194677 (h-1); Ki = 0.011675 +/- 0.004086 (h-1); Ke = 0.644415 +/- 0.239376 (h-1). When we compared this population to the general population of the program, Ke was increased. Elimination constant Ke was not correlated to either creatinine clearance or urine flow. Evaluation of the predictive performance of the Bayesian PC Program for adaptive control of vancomycin therapy in neutropenic patients is the next step of this study.

Plasma concentrations of midazolam after i.v., nasal or rectal administration in children.

Midazolam is used frequently for premedication in children, preferably by non-parenteral administration. We have compared plasma concentrations of midazolam after nasal, rectal and i.v. administration in 45 children (aged 2-9 yr; weight 10-30 kg) undergoing minor urological surgery. General anaesthesia consisted of spontaneous respiration of halothane and nitrous oxide in oxygen via a face mask. After administration of atropine and fentanyl i.v., children were allocated randomly to receive midazolam 0.2 mg kg-1 by the nasal, rectal or i.v. route. In the nasal group, children received 50% of the dose of midazolam in each nostril. In the rectal group, midazolam was given rectally via a cannula. Venous blood samples were obtained before and up to 360 min after administration of the drug. Plasma concentrations of midazolam were measured by gas chromatography and electron capture detection. After nasal and rectal administration, midazolam Cmax was 182 (SD 57) ng ml-1 within 12.6 (5.9) min, and 48 (16) ng ml-1 within 12.1 (6.4) min, respectively. Rectal administration resulted in smaller plasma concentrations. In the nasal group, a plasma concentration of midazolam 100 ng ml-1 occurred at about 6 min. After 45 min, the concentration curves after i.v. and nasal midazolam were similar.

Left ventricular performance during propofol or methohexital anesthesia: isotopic and invasive cardiac monitoring.

Using gated radionuclide ventriculography and invasive cardiac monitoring, the effects of propofol and methohexital on left ventricular volumes and function were investigated in 22 unpremedicated patients (ASA physical status III, 50-78 yr) with chronic coronary artery disease (NYHA class II-III). Anesthesia was induced with either propofol or methohexital (2 mg/kg), followed by a maintenance infusion of 100 micrograms.kg-1.min-1. Vecuronium (0.05 mg/kg) was administered and ventilation (FIO2, 1.0) was manually controlled (FECO2, 0.04-0.05). Data acquisitions were serially obtained over 15 min. Propofol and methohexital anesthesia caused an average 15% decrease in mean arterial pressure, associated with a 20% decrease in cardiac index without a decrease in systemic vascular resistance index. It is interesting that the determinants of these hemodynamic effects were different. Heart rate did not change during propofol infusion despite the decrease in mean arterial pressure, whereas heart rate increased during methohexital infusion. In the propofol group, the decrease in cardiac index was associated with decreases in indicators of preload (end-diastolic volume and pulmonary capillary wedge pressure), whereas end-systolic volume and global ejection fraction did not change statistically. In the methohexital group, the decrease in cardiac index was associated with a decrease in global ejection fraction and an increase in end-systolic volume, whereas indicators of preload remained unchanged. It is concluded that methohexital reduces left ventricular performance. In contrast, propofol preserves left ventricular performance despite a likely negative inotropic effect.

Pharmacokinetics of norfloxacin in the elderly.

9 elderly and 9 younger adult patients, with proven post-operative lower urinary tract infection were treated with 400 mg of norfloxacin twice daily for 5 days. Pharmacokinetics of norfloxacin were measured on days 1 and 5. Compared to the younger adult patients, the elderly showed a decreased creatinine clearance and, following the last dose on day 5, an increased maximum plasma concentration of norfloxacin, an increased area under the concentration-time curve and a decreased total body clearance of norfloxacin. These results confirm that in elderly, as in younger adult patients, the pharmacokinetics of norfloxacin can be described by a linear model and accumulation of the drug during repetitive multiple doses is predictable. The differences between the two groups cannot be considered as clinically significant so that no dose change would be required in elderly patients within the range of creatinine clearance studied.

Pharmacokinetics of teicoplanin in burn patients.

The pharmacokinetics of teicoplanin were studied in 21 adult burn patients and in 5 healthy controls. All subjects were given a single dose of teicoplanin, 10 mg/kg i.v. The pharmacokinetic parameters of teicoplanin were calculated using a tri-exponential model. Three parameters were investigated--burn surface area, creatinine clearance and time of administration after injury (3 days, 8-10 days and 15-20 days respectively). The elimination half-life was not significantly different between the control group (47 +/- 6 hours), the group with a burn surface area greater than 40% (64 +/- 38 hours), and the group with a burn surface area of 25-30% (44 +/- 15 hours). Trough serum concentrations at 12 hours were consistently less than 8 mg/l in all groups. There was some degree of correlation between creatinine clearance and renal clearance of teicoplanin (p = 0.03; r = 0.44). However, there was no correlation between burn surface area and creatinine clearance, or between the time of administration and creatinine clearance. Non-renal clearance of teicoplanin was unaffected by a large burn surface area. The conclusions from this study are as follows. Burn surface area and time of administration after injury are not predictive factors for the determination of an appropriate dose regimen. Trough serum concentrations at 12 hours emphasize the need for a further intravenous bolus. Burn patients with a high creatinine clearance (greater than 140 ml/min) should be carefully monitored.

Femoral nerve blockade in children using bupivacaine.

The authors evaluated the efficacy and incidence of side effects from blockade of the femoral nerve with 0.5% bupivacaine in 14 children with fracture of the middle third of the femoral shaft. In nine of these children, a pharmacokinetic analysis was also performed. The onset of analgesia occurred in 8.0 +/- 3.5 minutes after blockade of the femoral nerve. One block failed, resulting in iv narcotics being administered to alleviate the pain. In the remaining 13 children, pain decreased to nonexistent in 11 of the children and only mild pain with movement in the remaining two children. The level of analgesia did not change when the children underwent radiographic examination (60 +/- 18 min after the femoral nerve block) and application of traction (124 +/- 19 min after femoral nerve block). The maximum bupivacaine plasma concentration was 0.89 +/- 0.37 microgram/ml, obtained 24.4 +/- 12.6 min after the end of the injection. The femoral nerve blockade with bupivacaine provides prompt, effective, and prolonged analgesia in children suffering from fractures of the femoral shaft, allowing transport, radiographic examination, and application of traction in optimal conditions. Although the sample size was small, the side effects appeared to be rare.

Pharmacokinetics of lidocaine and bupivacaine in retrobulbar and facial block.

Analgesia can be obtained during ophthalmic surgery by regional anesthesia using local anesthetic agents. As in other indications, neurological complications may occur, especially because the site of injection is close to the central nervous system. In order to evaluate the risk of retrobulbar and facial block obtained after 40 mg lidocaine and 20 mg bupivacaine injection, pharmacokinetics of both drugs was evaluated in plasma obtained from 11 patients. In addition, 3 cerebrospinal fluid samples were analyzed. Maximal plasma concentration was 0.73 +/- 0.33 micrograms.ml-1 for lidocaine and 0.19 +/- 0.06 micrograms.ml-1 for bupivacaine, obtained 24.7 +/- 23.0 min and 12.0 +/- 3.7 min after the end of injection, respectively. CSF/plasma ratio was in the range 0.05-0.26 for lidocaine and 0.56-1.33 for bupivacaine. In all patients, regional anesthesia was sufficient to perform surgery without any other analgesic drug. No sign of cardiovascular or respiratory toxicity was observed during the study.

Pharmacokinetics and haemodynamic effects of prolonged methohexitone infusion.

1. The use of continuous infusion anaesthesia has only been of interest since the development of short-acting, less cumulative and less toxic drugs. 2. This study aimed to compare pharmacokinetics and haemodynamic effects during and after long time methohexitone constant rate infusion. Sixteen patients were given either 60 or 90 micrograms kg-1 min-1 methohexitone during 14 h. Blood samples were taken hourly during this time and 12 h following the end of infusion. 3. Infusion period was analysed by a single exponential model; post-infusion time showed a three compartment model, the intermediate phase parameters corresponding to those of the infusion period. 4. Methohexitone was haemodynamically well tolerated; prolonged infusion decreases oxygen consumption, mainly by a decrease in oxygen demand. 5. Many patients remained unconscious for unacceptably long periods of time after post-operative sedation by methohexitone.

Quantitative analysis of serum methohexital by GLC using capillary column and nitrogen-selective detection.

A gas chromatographic method for routine quantitation of methohexital in plasma samples is reported. One-step extraction in organic phase, the use of a fused silica capillary column, and nitrogen-selective detection permit simple, precise, and sensitive determination of methohexital in plasma. A linear relationship is described between peak height ratio and methohexital concentrations ranging from 0.125 to 50.0 micrograms/ml (r = 0.998). The sensitivity limit of the assay was 6 ng/ml in plasma. No interfering peak was observed with numerous other drugs. The procedure was successfully applied to the determination of pharmacokinetic parameters of methohexital after IV administration or continuous infusion in a child and an adult.

Rapid gas-chromatographic assay of bupivacaine in plasma.

A method for estimating bupivacaine concentration in human plasma by capillary gas-chromatography using solid injection and nitrogen-specific detection is described. Etidocaine, another anilidetype local anesthetic was used as internal standard and added to the sample before single-step extraction with diethylether. This method demonstrates high sensitivity (6 ng/ml plasma) and combines selectivity, rapidity, and simplicity. Results of this procedure correlate well with those obtained by an HPLC method.

A monitoring study of cardiotonic treatment by immunoenzymologic measurement of digoxinemia (emit).

We performed a study on 96 patients to compare monitoring by immunoenzymologic measurement (EMIT) of digoxinemia. In doing so, we uniquely relied on clinical and electrocardiographic results. Compliance was good because only 9 patients had a digoxinemia equal or below 0.8 ng/ml, but we emphasize that our patients were hospitalized. Correlation between intoxications and plasmatic levels shows that 5 patients presented clinical or electrocardiographic signs indicative of digitalis intoxication with a digoxinemia less than 2.5 ng/ml and 14 patients without intoxication with a digoxinemia higher than 2.5 ng/ml. We point out in this study that for EMIT immunoassay it is better to take 2.5 ng/ml as the concentration limit to be sure to avoid intoxication.

Pharmacokinetics of intravenous and intramuscular methohexitone in dogs.

Twenty-four dogs received methohexitone, either intravenously injected (2 mg kg-1 of a 1% solution) or intramuscularly (10 mg kg-1 of a 2% solution). Plasma methohexitone concentrations were measured by gas/liquid chromatography and pharmacokinetics were obtained from the general equations of a multicompartment model. Peak blood concentrations were equivalent following i.v. (18.2 +/- 9.9 mg 1(-1); at 30 s) and i.m. (19.1 +/- 5.6 mg 1(-1); at 3 min) injections. After i.v. injection a rapid distribution phase (half-life t 1/2 lambda 1; 1.3 +/- 0.5 min) was followed by an elimination phase (elimination half-life t 1/2 lambda z; 26.4 +/- 7.8 min). After i.m. injection the distribution phase was followed by two further phases (half-lives: 10.1 +/- 3.6 min and 75.6 +/- 22.6 min). The authors conclude that an i.m. dose five times as great as the i.v. dose produces equivalent peak blood concentration 30 s after i.v. injection and 3 min after i.m. injection. In addition, after i.m. injection of 10 mg kg-1 an additional compartment was apparent, indication substantial drug uptake by poorly perfused tissues.