Superiority of disease-specific over conventional formula in predicting creatinine clearance from serum creatinine in patients with liver cirrhosis

To determine if liver cirrhosis may influence the accuracy of formulas estimating creatinine clearance (Clcr) from serum creatinine, we compared the measured and estimated Clcr in 95 male (group A) and 47 female cirrhotic patients (group B). The Clcr values of group A and B patients were estimated with the equations obtained from 93 male (group C) and 86 female patients (group D) who were free of liver disease: the mean estimation errors (+/- SD) for the group C and D patients with the equations obtained from their own population data were 5 +/- 24% and 4 +/- 22%, respectively. However, these equations significantly (p less than 0.01) overestimated the Clcr of cirrhotic patients: the mean estimation errors for the group A and B patients were 35 +/- 43% and 14 +/- 27%, respectively. In contrast, the mean estimation errors for the group A and B patients using cirrhotic patient-specific equations obtained from their own population data were 5 +/- 33% and 3 +/- 25%, respectively. The accuracy of these disease-specific formulas was also confirmed in a prospective manner in 43 male (group E) and 21 female cirrhotic patients (group F): the mean estimation errors for the group E and F patients were 2 +/- 32% and -7 +/- 29%, respectively. We conclude that the Clcr values of cirrhotic patients, particularly male patients, may not be accurately estimated with equations of formulas deriving from liver disease-free patients, but should be estimated using their own population data.     

Comparison of double log-log,mixture response-surface and combined NIBS/Redlich-Kister solubility models

The accuracy and prediction capability of the linear double log-log (LDL-L), mixture response-surface (MR-S) and the combined nearly ideal binary solvent/Redlich-Kister (CNIBS/R-K) solubility equations have been compared using the model parameters calculated from either the whole data or a minimum number of data in an experimental set. The CNIBS/R-K model produced better prediction for some experimental sets than the other two models when the parameters obtained from the whole data in a set were employed, whereas the LDL-L model was superior to the other models when the parameters calculated from a minimum number of data were used, indicating its greatest prediction capability.     

Comparison of Measured and Estimated Creatinine Clearance in Patients With Advanced HIV Disease

Study Objective . To assess the accuracy of five creatinine clearance equations in predicting measured creatinine clearance in hospitalized patients with human immunodeficiency viral (HIV) infection. Design . Prospective evaluation over a 6-month period. Setting . Erie County Medical Center, a 550-bed teaching institution. Patients . Forty-seven HIV-positive patients (39 men, 8 women) who were admitted for a variety of HIV-related illnesses and judged clinically to have stable renal function. Of the 47 original patients, 44 were evaluable based on exclusion criteria. Interventions . Serum creatinine and 24-hour measured creatinine clearance were performed in each patient. Measurements and Main Results . The estimated creatinine clearance from each of the equations (Cockcroft-Gault, two Jeliffe equations, Mawer et al, and Hull et al) was compared with the measured creatinine clearance. Statistical analysis of these comparisons was performed and all of the equations were found to overestimate the measured creatinine clearance (mean error 34–45%). Conclusions . Many HIV-infected patients have a decreased creatinine clearance despite a serum creatinine concentration within the normal range. Each of the equations overestimated the measured creatinine clearance.     

Fitting concentration-time data to biexponential equations

A digital computer study using simulated data containing random error indicates no difference in the precision and accuracy of parameter estimation when the data are fitted to several different equations.This work was supported in part by Grant GM-20852 from the National Institute of General Medical Sciences, National Institutes of Health.     

Analysis of Various Creatinine Clearance Formulas in Predicting Gentamicin Elimination in Patients With Low Serum Creatinine

Predicted gentamicin elimination rate constants (k_els) using creatinine clearance (C1_cr) estimates from seven equations were compared with k_els calculated from steady-state serum gentamicin concentrations in 186 hospitalized patients. In predicting k_el, the equations varied significantly in precision (mean absolute percentage error), and were particularly imprecise among patients with serum creatinine values of 71 μmol/L or less. Significant differences in bias (mean prediction error) were also observed. All equations using serum creatinine as an element showed improved precision, and most showed reduced bias when a minimum value of 71 μmol/L was used. The Cockcroft-Gault normalized to 72 kg and the Hull equations are among the simplest to calculate and, when using a minimum serum creatinine of 71 μmol/L, had significantly greater precision and less bias than several of the equations. We recommend one of these two methods for predicting gentamicin k_el in patients with low serum creatinine values.     

Development of a decision tree to classify the most accurate tissue-specific tissue to plasma partition coefficient algorithm for a given compound

Physiologically based pharmacokinetic (PBPK) modeling is a tool used in drug discovery and human health risk assessment. PBPK models are mathematical representations of the anatomy, physiology and biochemistry of an organism and are used to predict a drug’s pharmacokinetics in various situations. Tissue to plasma partition coefficients (Kp), key PBPK model parameters, define the steady-state concentration differential between tissue and plasma and are used to predict the volume of distribution. The experimental determination of these parameters once limited the development of PBPK models; however, in silico prediction methods were introduced to overcome this issue. The developed algorithms vary in input parameters and prediction accuracy, and none are considered standard, warranting further research. In this study, a novel decision-tree-based Kp prediction method was developed using six previously published algorithms. The aim of the developed classifier was to identify the most accurate tissue-specific Kp prediction algorithm for a new drug. A dataset consisting of 122 drugs was used to train the classifier and identify the most accurate Kp prediction algorithm for a certain physicochemical space. Three versions of tissue-specific classifiers were developed and were dependent on the necessary inputs. The use of the classifier resulted in a better prediction accuracy than that of any single Kp prediction algorithm for all tissues, the current mode of use in PBPK model building. Because built-in estimation equations for those input parameters are not necessarily available, this Kp prediction tool will provide Kp prediction when only limited input parameters are available. The presented innovative method will improve tissue distribution prediction accuracy, thus enhancing the confidence in PBPK modeling outputs.     

Mechanistic Approaches to Volume of Distribution Predictions: Understanding the Processes

PURPOSE: To use recently developed mechanistic equations to predict tissue-to-plasma water partition coefficients (Kpus), apply these predictions to whole body unbound volume of distribution at steady state (Vu(ss)) determinations, and explain the differences in the extent of drug distribution both within and across the various compound classes. MATERIALS AND METHODS: Vu(ss) values were predicted for 92 structurally diverse compounds in rats and 140 in humans by two approaches. The first approach incorporated Kpu values predicted for 13 tissues whereas the second was restricted to muscle. RESULTS: The prediction accuracy was good for both approaches in rats and humans, with 64-78% and 82-92% of the predicted Vu(ss) values agreeing with in vivo data to within factors of +/-2 and 3, respectively. CONCLUSIONS: Generic distribution processes were identified as lipid partitioning and dissolution where the former is higher for lipophilic unionised drugs. In addition, electrostatic interactions with acidic phospholipids can predominate for ionised bases when affinities (reflected by binding to constituents within blood) are high. For acidic drugs albumin binding dominates when plasma protein binding is high. This ability to explain drug distribution and link it to physicochemical properties can help guide the compound selection process.     

Aminoglycoside dosage adjustment in renal failure: A hand-held calculator program

Summary Several problems occur when devising dosage guidelines for aminoglycoside antibiotics in patients with renal failure. To rationally address these problems, dosage guidelines require several steps involving complex calculations, use of graphic charts and/or use of sophisticated computer systems. We describe a practical program for modifying doses of aminoglycosides using a programmable hand-held calculator; this program is based both on pharmacokinetic theory and on applicability to varied clinical settings. The program compiles a series of equations to provide recommended doses, dosing intervals and predictions of serum concentrations of aminoglycosides at various times after a dose. It is hoped that patient care can be improved by using this simple, convenient and low-cost approach which retains efficiency and accuracy as a bed side method of dosage adjustment for aminoglycosides.     

Automated system for analytical microbiology V: calibration lines for antibiotics.

The accuracy of an automated system for the microbiological assay of antibiotics was increased by improvement attendant to connection to an on-line computer. The system was used to investigate the suitability of four forms of interpolation formulas by assaying for chlortetracycline and erythromycin. The calibration lines were prepared as point-to-point straight-line approximations and as cubic equations. Cubic equations through four calibration points were preferred. Since the automated system was a four-channel instrument, a separate response line was prepared for each channel. Combining the four response lines into one could substantially degrade the accuracy and precision of assays. A new general equation relating the response of the test organism to concentrations of active materials was used to account for factors in addition to the antibiotic upon the dose-response line. Some of these factors were: diluents, growth substances, relative proportions of mixed antibiotics, pH and buffer capacities of the sample solution and assay broth, salts, and organic compounds in samples and not in standard solutions. The equation was used to show under what conditions the dose-response lines of mixtures and single-component antibiotics could be the same. It could also account for the nonspecific nature of turbidimetric assays. The equation showed assay biases to be caused not by differences in composition of antibiotics in standards and samples but by differences in other substances affecting growth of the test organism. A new dose-response line applicable to assays using Klebsiella pneumoniae was described.     

General pharmacokinetic equations for linear mammillary models with drug absorption into peripheral compartments

Summary General equations are derived for the disposition functions of any compartment in a linear mammillary model, when the system input occurs into a peripheral compartment. Laplace transforms and matrix algebra are used to derive these equations. Equations describing the time-course of a drug in any compartment are readily obtainable using disposition and input functions.     

Estimation of creatinine clearance in patients with metastatic ovarian cancer

The authors evaluated the predictive performance of four methods used to estimate creatinine clearance (Cl(cr)) in patients with metastatic ovarian cancer. Methods described by Cockcroft and Gault, Jelliffe, and two equations derived from cancer patients by Robinson and Tsubaki, were evaluated. Estimated Cl(cr) values obtained by each method using actual weight (ABW), ideal weight (IBW) and lower of ABW and IBW were compared with measured values determined by a 12- or 24-hour urine collection for 14 patients enrolled in a controlled clinical trial. The mean prediction errors (ME) and mean absolute errors (MAE) were calculated to evaluate the bias and precision, respectively, of each method. The relationship between predicted and measured Cl(cr) is poor (r = 0.38 to 0.54). Cockcroft and Gault using ABW (p = 0.21), Robinson using ABW (p = 0.44), and Jelliffe (p = 0.17) were equally unbiased predictors of measured Cl(cr). All other methods significantly underestimated measured Cl(cr). All methods appeared to be equally imprecise (p<0.05). The use of standard equations for estimating Cl(cr) in patients with ovarian cancer is predictive of the measured 24-hour value. The use of oncology specific equations does not improve the accuracy or precision of these estimates.     

Comparison of Particle Tracking Algorithms in Commercial CFD Packages: Sedimentation and Diffusion

Computational fluid dynamic modeling software has enabled microdosimetry patterns of inhaled toxins and toxicants to be predicted and visualized, and is being used in inhalation toxicology and risk assessment. These predicted microdosimetry patterns in airway structures are derived from predicted airflow patterns within these airways and particle tracking algorithms used in computational fluid dynamics (CFD) software packages. Although these commercial CFD codes have been tested for accuracy under various conditions, they have not been well tested for respiratory flows in general. Nor has their particle tracking algorithm accuracy been well studied. In this study, three software packages, Fluent Discrete Phase Model (DPM), Fluent Fine Particle Model (FPM), and ANSYS CFX, were evaluated. Sedimentation and diffusion were each isolated in a straight tube geometry and tested for accuracy. A range of flow rates corresponding to adult low activity (minute ventilation = 10 L/min) and to heavy exertion (minute ventilation = 60 L/min) were tested by varying the range of dimensionless diffusion and sedimentation parameters found using the Weibel symmetric 23 generation lung morphology. Numerical results for fully developed parabolic and uniform (slip) profiles were compared respectively, to and analytical sedimentation solutions. equations for sedimentation were also compared. Numerical results for diffusional deposition were compared to analytical solutions of for parabolic and uniform profiles. Significant differences were found among the various CFD software packages and between numerical and analytical solutions. Therefore, it is prudent to validate CFD predictions against analytical solutions in idealized geometry before tackling the complex geometries of the respiratory tract.     

Comparison of particle tracking algorithms in commercial CFD packages: sedimentation and diffusion

Computational fluid dynamic modeling software has enabled microdosimetry patterns of inhaled toxins and toxicants to be predicted and visualized, and is being used in inhalation toxicology and risk assessment. These predicted microdosimetry patterns in airway structures are derived from predicted airflow patterns within these airways and particle tracking algorithms used in computational fluid dynamics (CFD) software packages. Although these commercial CFD codes have been tested for accuracy under various conditions, they have not been well tested for respiratory flows in general. Nor has their particle tracking algorithm accuracy been well studied. In this study, three software packages, Fluent Discrete Phase Model (DPM), Fluent Fine Particle Model (FPM), and ANSYS CFX, were evaluated. Sedimentation and diffusion were each isolated in a straight tube geometry and tested for accuracy. A range of flow rates corresponding to adult low activity (minute ventilation = 10 L/min) and to heavy exertion (minute ventilation = 60 L/min) were tested by varying the range of dimensionless diffusion and sedimentation parameters found using the Weibel symmetric 23 generation lung morphology. Numerical results for fully developed parabolic and uniform (slip) profiles were compared respectively, to Pich (1972) and Yu (1977) analytical sedimentation solutions. Schum and Yeh (1980) equations for sedimentation were also compared. Numerical results for diffusional deposition were compared to analytical solutions of Ingham (1975) for parabolic and uniform profiles. Significant differences were found among the various CFD software packages and between numerical and analytical solutions. Therefore, it is prudent to validate CFD predictions against analytical solutions in idealized geometry before tackling the complex geometries of the respiratory tract.     

Accuracy and Reliability of Dosing Equations to Individualize Theophylline Treatment of Apnea of Prematurity

Apnea of prematurity is associated with high morbidity and mortality. Treatment generally includes supplemental oxygen and theophylline or caffeine. The half-life of theophylline is prolonged in newborns because of their immature cytochrome P-450 system, and there is considerable variation in the drug's metabolism in infants. We compared the accuracy, precision, and reliability of two equations that use postnatal age (PNA) to determine a maintenance dosage of theophylline with a standard maintenance dosage (SMD) that produced a steady-state serum theophylline concentration (STC) of 8 μg/ml for apnea of prematurity in 46 infants less than 34 weeks' gestational age (GA) and less than 36 weeks' postconceptional age (PCA). The two equations were mg/kg/day = [(0.2 × PNA in wks) + 5], and mg/kg/day = [(0.3 × PNA in wks) + 8]. Their reliability to predict the SMD was determined by correlation analysis. The precision and accuracy with which they predicted SMD were determined and analyzed by χ². The SMD did not correlate with the maintenance dosages calculated by equations 1 and 2 (r=0.296 and 0.296, p>0.05 in both cases). Multiple linear regression of SMD versus GA, PNA, and PCA was not significant (r=0.33, p=0.32). After stratifying data based on GA and performing correlation analysis of SMD versus PNA, a weak but significant correlation (r=0.42, p=0.0517) was found for infants with GA between 31 and 34 weeks. Poor correlation was found between SMD versus PNA for infants 27–30 weeks' GA. Two new equations of the best fit line were generated using the same data. Currently available equations for determining maintenance dosages of theophylline in infants and children are not reliable for premature newborns, in whom two new equations based on GA and PNA have been developed. The equations will be evaluated for their utility, and employed to accrue data to determine if better ones can be developed.     

二维带动力吸气式高超声速飞行器绕流的PNS-NS混合求解

对于数值模拟高超声速飞行器带动力整机流动,采用单纯的时间迭代方法非常耗时。本文结合空间推进求解PNS方程和时间迭代求解NS方程的CFD方法来模拟高超声速飞行器整机流动。高超声速飞行器的流场是超声速占主导的,但在流场中也存在局部分离以及亚声速占主导的区域。本文采用的CFD方法,在超声速占主导的流动区域采用空间推进求解PNS方程的方法,在亚声速和分离区采用时间迭代求解NS方程的方法,整机流动在划分的各个区域之间自动推进求解。对于求解带动力高超声速二维整机流动,目前的CFD方法在大大节约计算时间和计算资源的同时,可以得到和完全时间迭代方法同等准确的数值模拟结果。     

慢性肾脏病流行病合作工作组方程在慢性肾脏病2~3期老年患者肾功能评估中的意义

目的 评价3种慢性肾脏病流行病合作工作组(the chronic kidney disease epidemiology collaboration,CKD-EPI)方程在慢性肾脏病(chronic kidney disease,CKD)2~3期老年患者肾小球滤过率(glomerular filtration rate,GFR)评估中的意义.方法回顾分析2010年1月至2015年1月在江苏省中医院住院的老年CKD2~3期患者81例,收集临床资料,以3种CKD-EPI方程估算GFR,记录为EPI-血肌酐(EPI-serum creatinine,EPI-SCr)、EPI-胱抑素C(EPI-cystatin C,EPI-CysC)和EPI-肌酐联合胱抑素C(EPI-serum creatinine-cystatin C,EPI-SCr-CysC),以锝[99m Tc]标记的喷替酸盐(Technetium[99m Tc]Pentetate,99m Tc-DTPA)肾动态显像为金标准,比较3种方程评价CKD2~3期老年患者GFR水平的适应性、灵敏度及准确度.结果CKD2~3期老年人群中,与EPI-SCr方程相比,EPI-CysC及EPI-SCr-CysC方程提高了适用性,且其优势主要体现在CKD3期;依据EPI-CysC及EPI-SCr-CysC方程结果对老年患者进行CKD分组,其灵敏度(39.22%比90.20%比86.27%,P<0.001)及准确度(55.56%比82.72%比82.72%,P<0.001)更高,但三个方程特异度之间差异无统计学意义(83.88%比70.00%比76.67%,P=0.475).结论联合CysC的CKD-EPI方程对CKD2~3期老年患者GFR评估和CKD分期更具指导意义.     

Characteristics of an aerosol photometer while automatically controlling chamber dilution-air flow rate

Trials were conducted to determine those factors that affect the accuracy of a direct-reading aerosol photometer when automatically controlling airflow rate within an exposure chamber to regulate airborne dust concentrations. Photometer response was affected by a shift in the aerosol size distribution caused by changes in chamber flow rate. In addition to a dilution effect, flow rate also determined the relative amount of aerosol lost to sedimentation within the chamber. Additional calculations were added to a computer control algorithm to compensate for these effects when attempting to automatically regulate flow based on a proportional-integral-derivative (PID) feedback control algorithm. A comparison between PID-controlled trials and those performed with a constant generator output rate and dilution-air flow rate demonstrated that there was no significant decrease in photometer accuracy despite the many changes in flow rate produced when using PID control. Likewise, the PID-controlled trials produced chamber aerosol concentrations within 1% of a desired level.     

Comparison of dosing recommendations for antimicrobial drugs based on two methods for assessing kidney function: cockcroft-gault and modification of diet in renal disease

STUDY OBJECTIVES: To quantify the difference between glomerular filtration rates (GFRs) estimated by using the Cockcroft-Gault and Modification of Diet in Renal Disease (MDRD) equations, and to determine whether dosing recommendations for four commonly prescribed antimicrobial agents are discordant when determined by using these equations. DESIGN: Prospective, observational study. SETTING: Tertiary-care medical center. PATIENTS: Two hundred seven consecutive adults without normal renal function but not receiving dialysis who were admitted to a non-intensive-care ward and had two consecutive serum creatinine concentration (S(cr)) values measured 20-24 hours apart. MEASUREMENTS AND MAIN RESULTS: The patients' mean +/- SD S(cr) was 1.41 +/- 0.95 mg/dl. Kidney function was estimated by using two versions of the four-variable MDRD equation and four versions of the Cockcroft-Gault equation. Mean estimated GFRs ranged from 52.3-73.1 ml/minute. Dosing for cefepime, levofloxacin, meropenem, and piperacillin-tazobactam was determined using the two equations that had the highest level of correlation; these were the MDRD equation unadjusted for body surface area and the Cockcroft-Gault equation adjusted for ideal body weight and S(cr). When the total daily doses based on these two equations for the four antimicrobials were compared, the discordance rate was 22.8-36.3%, and statistically significant differences were observed for most of the discordant doses. When discordance was present, the MDRD equation resulted in a higher dose of the drug. CONCLUSION: Discordance rates for drug dosing ranged from 22.8-36.3% between the MDRD and Cockcroft-Gault methods for estimating GFR. Although use of the MDRD equation is a well-accepted and accurate method of estimating GFR to stage chronic kidney disease, our results demonstrated a significant difference in drug dosing regimens between the MDRD method and the Cockcroft-Gault method.     

Methodology for using oral dose pharmacokinetic data to select drugs for prolonged release formulations and validation of the method using simulated data

A computer aided method for selecting drugs as potential candidates for oral prolonged release formulations has been previously published. In order to decide whether trial formulations were warranted, prolonged release dosing was simulated to find all release rates and doses producing successful regimens in every subject for whom clinical pharmacokinetic data were available. Intravenous dose data were required because the models and their parameter values could not be assessed with oral doses. The new method uses model independent equations derived from rapidly absorbed oral doses to simulate the administration of prolonged release formulations. Using these equations, repetitive oral dosing regimens were reiteratively simulated to search for all successful release rates and doses in every available subject. Results were validated by comparison to those obtained with the published method using theophylline and hypothetical drugs, which included examples of flip-flop and vanishing exponentials. When the oral reference dose data were reliable, results were similar even when the model independent parameter values did not agree with those used to generate the oral reference dose data. Differences were observed only when the random errors in the oral reference dose data were large. If more than one model independent equation provided equivalent fits to the oral reference dose data, the final results were similar independent of which equation was employed.     

Effects of drugs of abuse on acquisition of behavioral chains in squirrel monkeys

The acute effects of various drugs of abuse on the acquisition of chains of behavior were assessed in squirrel monkeys trained to respond on three keys for food. Each new session the monkeys acquired a different four-response chain by responding sequentially on three keys in the presence of four different stimuli. Incorrect responses inactivated the keys and darkened the chamber for 10 s (time-out). Dose-effect curves were obtained by administering the drugs intramuscularly before the session and recording their effects on the rate and accuracy of responding. Cocaine,d-amphetamine, and ⁹-tetrahydrocannabinol all decreased the accuracy and rate of responding within the dose range of 0.56–3 mg/kg. The highest dose of morphine tested (3 mg/kg) produced parallel decreases in the accuracy and rate of responding in some monkeys but had no effect at lower doses. These drugs decreased within-session accuracy though clearly acquisition did occur, but high doses of caffeine (30 and 56 mg/kg) prevented acquisition and recovery of performance and, furthermore, at 30 mg/kg these effects were observed in the absence of decreases in the rate of responding. The drugs of abuse tested all produced dose-related decreases in both the accuracy and rate of responding, and the decreases in accuracy were primarily observed only at doses that also decreased response rates. Therefore, based on these results from nonhuman primates each of these drugs has the potential to alter learning particularly when doses that disrupt other behaviors are administered.