Abstract. The goal of the present study was to examine the utility of the conscious dog model by assessing the QT-interval-prolonging potential of ten positive compounds that have been reported to induce QT interval prolongation in clinical use and seven negative compounds considered not to have such an effect. Three doses of test compounds or vehicle were administered orally to male beagle dogs (n = 4), and telemetry signals were recorded for 24 h after administration. All positive compounds (astemizole, bepridil, cisapride, E-4031, haloperidol, MK-499, pimozide, quinidine, terfenadine, and thioridazine) caused a significant increase in the corrected QT (QTc) interval, with a greater than 10% increase achieved at high doses. In contrast, administration of negative compounds (amoxicillin, captopril, ciprofloxacin, diphenhydramine, nifedipine, propranolol, and verapamil) did not produce any significant change in the QTc interval, with the exception of nifedipine that may have produced an overcorrection of the QTc interval due to increased heart rate. The estimated plasma concentrations of the positive compounds that caused a 10% increase in the QTc interval were in good agreement with the plasma / serum concentrations achieved in humans who developed prolonged QT interval or torsade de pointes (TdP). Although careful consideration should be given to the interpretation of QT data with marked heart rate change, these data suggest that an in vivo QT assay using the conscious dog is a useful model for the assessment of QT interval prolongation by human pharmaceuticals. Supplementary material (Appendix): available only at http://dx

The present study confirms that CBA/J mice are susceptible to several clinical isolates of Streptococcus pneumoniae, including four of five penicillin-susceptible and all five penicillin-resistant strains tested, thus providing the first noncompromised animal model for penicillin-resistant S. pneumoniae pneumonia. In this model, doses of penicillin G of 0.6 mg/kg of body weight given six times at 1-h intervals produced effective pulmonary clearance of a penicillin-susceptible strain (penicillin G MIC, 0.015 microgram/ml), while doses of 40 mg/kg given six times at 1-h intervals were required to clear a penicillin-resistant strain (penicillin G MIC, 1 microgram/ml). Imipenem (MIC, 0.25 microgram/ml) was the most active antibiotic tested against the penicillin-resistant strain, with a calculated dose of 0.42 mg/kg given six times at 1-h intervals, resulting in a 2-log decrease in the number of pulmonary bacteria. Comparable effects were seen with vancomycin (MIC, 0.5 microgram/ml), cefotaxime (MIC, 0.5 microgram/ml), and penicillin G at doses of 3.3, 5.5, and 31.0 mg/kg given six times at 1-h intervals, respectively. The pharmacokinetic profile of vancomycin in infected lungs was superior to those of the other antibiotics, especially in regard to the elimination half-life (215.4 min for vancomycin versus 15.0, 14.5, and 14.5 min for penicillin G, cefotaxime, and imipenem, respectively). Both imipenem and vancomycin allowed 90% survival when 40-mg/kg doses were administered twice a day beginning 5 days after infection. Survival rates with penicillin G (160-mg/kg doses) and cefotaxime (40-mg/kg doses) were 40 and 30%, respectively, while no saline-treated mice survived. The present study shows that the CBA/J mouse pneumonia model may be useful for evaluating antibiotic efficacies against penicillin-resistant pneumococcal pneumonia in immunocompetent individuals. Our data suggest that imipenem and vancomycin may be the most active agents against penicillin-resistant S. pneumoniae pneumonia.

Abstract. The purpose of this study was to assess the utility of the isoflurane-anesthetized dog model for detecting the potential for QT interval prolongation by human pharmaceuticals. The effects of 10 positive compounds with torsadogenic potential, 8 negative compounds with little torsadogenic potential, and dl-sotalol as a common positive compound were evaluated in 5 facilities in accordance with the common protocol approved by QT PRODACT. Each test compound was cumulatively infused into male beagle dogs anesthetized with isoflurane. Surface lead II ECG, blood pressure, and plasma concentrations for the positive compounds were measured. Repeated administration of the vehicle examined in each facility before the start of the experiments resulted in a slight, but not significant, change in corrected QT (QTc) interval, indicating that this model only shows slight experimental variation. Although an inter-facility variability in the extent of dl-sotalol-induced QT interval prolongation was observed, dl-sotalol significantly prolonged QTc interval in all facilities. All positive compounds significantly prolonged QTc interval at plasma levels up to 10 times those in patients who developed prolonged QTc interval or TdP, whereas no negative compounds did so. These data suggest that the in vivo QT assay using the anesthetized dog is a useful model for detecting the potential for QT interval prolongation by human pharmaceuticals. Supplementary material (Appendix): available only at http://dx

No abstract

The establishment of a new rate-correction method for the QT interval is presented for long-term telemetry ECG recording in free-moving beagle dogs. First, in order to define the QT-RR relation to derive the correction formula, the diurnal variations of the QT and RR intervals and the influencing factors were analyzed, and the QT-RR regression coefficient beta was estimated under various conditions: steady and non-steady states of animal, light and dark periods, and over 24 h. In the results, the diurnal rhythm of the QT interval was synchronized with the RR interval reflecting the physical and emotional states of the animal. The coefficient beta had considerable variation during the day: beta; 0.276 +/- 0.052 (maximum to minimum: 0.495 to 0.153). Thus, it was considered that the ideal rate-correction technique for telemetry ECG requires the selection of a flexible coefficient beta adjusted to the condition of the measurement. Therefore, rate-correction utilizing analysis of covariance estimating the coefficient beta for each dog, was compared with previously proposed formulas which fix the rate-correction coefficient, based on the capacity to dissociate the effects of heart rate on the QT interval. This was then tested by the levels of discrimination apparent in the QT prolongation caused by a class III antiarrhythmic drug, which ranked the formulas on the levels of correction achieved as follows: covariance adjustment > Matsunaga > Van de Water > Bazett. Thus, the rate-correction method utilizing analysis of covariance is proven adequate for data from telemetry ECG recordings.