Warfarin maintenance dose prediction for Chinese after heart valve replacement by a feedforward neural network with equal stratified sampling.

Patients requiring low-dose warfarin are more likely to suffer bleeding due to overdose. The goal of this work is to improve the feedforward neural network model's precision in predicting the low maintenance dose for Chinese in the aspect of training data construction. We built the model from a resampled dataset created by equal stratified sampling (maintaining the same sample number in three dose-groups with a total of 3639) and performed internal and external validations. Comparing to the model trained from the raw dataset of 19,060 eligible cases, we improved the low-dose group's ideal prediction percentage from 0.7 to 9.6% and maintained the overall performance (76.4% vs. 75.6%) in external validation. We further built neural network models on single-dose subsets to invest whether the subsets samples were sufficient and whether the selected factors were appropriate. The training set sizes were 1340 and 1478 for the low and high dose subsets; the corresponding ideal prediction percentages were 70.2% and 75.1%. The training set size for the intermediate dose varied and was 1553, 6214, and 12,429; the corresponding ideal prediction percentages were 95.6, 95.1%, and 95.3%. Our conclusion is that equal stratified sampling can be a considerable alternative approach in training data construction to build drug dosing models in the clinic.

Induction of heart valve lesions by small-molecule ALK5 inhibitors.

Aberrant signaling by transforming growth factor-ß (TGF-ß) and its type I (ALK5) receptor has been implicated in a number of human diseases and this pathway is considered a potential target for therapeutic intervention. Transforming growth factor-ß signaling via ALK5 plays a critical role during heart development, but the role of ALK5 in the adult heart is poorly understood. In the current study, the preclinical toxicology of ALK5 inhibitors from two different chemistry scaffolds was explored. Ten-week-old female Han Wistar rats received test compounds by the oral route for three to seven days. Both compounds induced histopathologic heart valve lesions characterized by hemorrhage, inflammation, degeneration, and proliferation of valvular interstitial cells. The pathology was observed in all animals, at all doses tested, and occurred in all four heart valves. Immunohistochemical analysis of ALK5 in rat hearts revealed expression in the valves, but not in the myocardium. Compared to control animals, protein levels of ALK5 were unchanged in the heart valves of treated animals. We also observed a physeal dysplasia in the femoro-tibial joint of rats treated with ALK5 inhibitors, a finding consistent with a pharmacological effect described previously with ALK5 inhibitors. Overall, these findings suggest that TGF-ß signaling via ALK5 plays a critical role in maintaining heart valve integrity.

Drug-induced valvulopathy: an update.

Drug-induced valvulopathy is a serious liability for certain compound classes in development and for some marketed drugs intended for human use. Reports of valvulopathy led to the withdrawal of fenfluramines (anorexigens) and pergolide (antiparkinson drug) from the United States market in 1997 and 2007, respectively. The mechanism responsible for the pathogenesis of valvulopathy by these drugs is likely a result of an "off-target" effect via activation of 5-hydroxytryptamine (5-HT) 2B receptor (5-HT2BR) expressed on heart valve leaflets. Microscopically, the affected valve leaflets showed plaques of proliferative myofibroblasts in an abundant extracellular matrix, composed primarily of glycosaminoglycans. However, the valvular effects caused by fenfluramines and pergolide were not initially predicted from routine preclinical toxicity studies, and to date there are no specific validated animal models or preclinical/toxicologic screens to accurately predict drug-induced valvulopathy. This review covers the structure and function of heart valves and highlights major advances toward understanding the 5-HT2BR-mediated pathogenesis of the lesion and subsequently, development of appropriate animal models using novel techniques/experiments, use of functional screens against 5-HT2BR, and more consistent sampling and pathologic evaluation of valves in preclinical studies that will aid in avoidance of future drug-induced valvulopathy in humans.

Nifedipine therapy for preterm labor: effects on placental, fetal cerebral and atrioventricular Doppler parameters in the first 48 hours.

OBJECTIVE: To assess the effect of nifedipine tocolysis on Doppler parameters of the uterine, umbilical and fetal middle cerebral arteries and atrioventricular valves in the first 48 h of therapy. METHODS: Doppler waveforms of uterine, umbilical and middle cerebral arteries and both atrioventricular valves were measured from 28 pregnant women and fetuses prior to and during nifedipine therapy for preterm labor. Maternal and fetal heart rates (FHR), maternal systolic and diastolic blood pressure, and the Doppler pulsatility index (PI) of the uterine, umbilical and middle cerebral arteries were measured. The cerebroplacental ratio (middle cerebral artery PI/umbilical artery PI) was calculated. The total time velocity integrals (TVIs) of tricuspid and mitral valves and their E- and A-wave peak velocity ratio (E/A) were measured. Friedman repeated-measures analysis of variance was used to compare the variables before and after nifedipine therapy. If significant differences were found, Wilcoxon's signed ranks test was used to analyze the difference between the two variables. A P-value of < 0.05 was considered significant. RESULTS: Nifedipine maintenance was associated with a significant decline in maternal systolic and diastolic blood pressure after 24 h, while maternal heart rate and FHR were unaffected. The uterine artery PI had decreased significantly at 24 and 48 h, while the umbilical artery PI did not change significantly. The middle cerebral artery PI had decreased significantly at 24 and again at 48 h. A significant fall in the cerebroplacental Doppler ratio was maintained beyond 24 h. The mean E/A values, TVIs and TVI x FHR values at 24 and 48 h were unchanged from the baseline values. CONCLUSIONS: Nifedipine maintenance tocolysis is associated with a significant decline in uterine artery and middle cerebral artery Doppler indices 24 h after the first dose. Fetal cardiac diastolic function is unaffected and the significant redistribution observed after 24 h is likely to be attributable to altered cerebral blood flow.

Culture results of heart valves resected because of streptococcal endocarditis: insights into duration of treatment to achieve valve sterilization.

OBJECTIVES: To analyse the culture results of heart valves removed following streptococcal endocarditis in order to gain insight into the duration of treatment required for valve sterilization. PATIENTS AND METHODS: Retrospective review of 131 episodes of streptococcal endocarditis: 94 due to alpha-haemolytic streptococci; 15 due to beta-haemolytic streptococci; 10 due to nutritionally deficient streptococci; eight due to the Streptococcus anginosus group and four due to Streptococcus pneumoniae. Patients had their valves removed during antimicrobial treatment. Culture results were analysed with respect to duration of treatment before surgery. RESULTS: For alpha-haemolytic streptococci, 17 (18%) valves were culture-positive and 77 (82%) culture-negative after a median (range) of 4 (1-20) and 16 (4-58) days of treatment, respectively, P < 0.001. For beta-haemolytic streptococci, two valves (13%) were culture-positive; both patients had received < or = 4 days of treatment. Four patients (40%) with nutritionally deficient streptococci were culture-positive, and had received < or = 8 days of treatment. For the S. anginosus group, two valves (25%) were culture-positive; both patients had received < or = 4 days of treatment before operation. Overall, only one of 131 (0.8%) valves was culture-positive after 14 days of treatment. All valves infected with beta-haemolytic streptococci, nutritionally deficient streptococci and the S. anginosus group, who were treated for more than 8 days before surgery, were culture-negative. CONCLUSIONS: Our findings support current treatment guidelines for endocarditis caused by alpha-haemolytic streptococci. We suggest that the recommended duration of treatment for endocarditis resulting from other streptococci may be excessive and treatment trials evaluating 2 and 4 week regimens are justified.

Sulbactam efficacy in experimental models caused by susceptible and intermediate Acinetobacter baumannii strains.

Sulbactam and imipenem were compared in an experimental pneumonia model in immunocompetent mice, using a susceptible strain of Acinetobacter baumannii, and in an experimental endocarditis model in rabbits, using an intermediately susceptible strain. In the former, sulbactam was as efficacious as imipenem in terms of survival, sterility of lungs and in the bacterial clearance from lungs and blood, provided that the t > MIC for sulbactam (1.84 h) was similar to that for imipenem (2.01 h). In the endocarditis model, imipenem (t > MIC, 2.12 h) was more efficacious than sulbactam (t > MIC, 1.17 h) in bacterial clearance from vegetations. These results show the efficacy of sulbactam in infections caused by susceptible strains of A. baumannii, with an MIC up to 4 mg/L, provided that doses reach a t > MIC similar to that of imipenem. The activity of sulbactam was time dependent.