Search

Application of a General Linear Model (GLM, “Analysis of Covariance”) to the statistical interpretation of stability data combines the methods of regression and analysis of variance in one common model. Expanding the well accepted method of linear regression upon time, the GLM model permits one to include supportive factors which may be either continuous regressors (temperature, humidity, etc.) or class effects (batch number, formulation type, manufacturer, etc.). Using the GLM procedure of SAS as a convenient software tool, the technique is illustrated by several examples. It is concluded that GLM provides a most suitable approach for the interpretation of stability data.

A detailed description of a column (flow-through) type dissolution apparatus is given, including features of automated sampling, assay, and computerized data processing. Standardized testing parameters are presented which permit reproducible performance in different laboratories, under conditions close to those of physiological dissolution and absorption. Without major modifications the same apparatus can be used for various dosage forms such as powders, granules, tablets, coated tablets, and capsules. Examples are given which demonstrate reproducibility as well as correlation with in-vivo data. It is suggested that this methodological approach be regarded as a promising alternative to the present official methods.

Liquid-phase data, gathered in a purely routine manner using common infra-red intensity measurement methods, have shown a scatter of about ±10 per cent from the mean, a value that has been found to coincide quite closely with more elaborately determined integrated intensities. It is believed that uncertainties in baseline recordings contributed greatly to this lack of precision.

The absorption intensities of the CH and CD stretching vibrations of some deuterated benzene monoderivatives have been measured in CCl4 solution. The compounds are of the type 4-D-phenyl-X and 2,4,6-D3-phenyl-X. The observed intensities can be interpreted as sum of contributions Io, Im, and Ip, of the participating CH or CD oscillators. By using also the data for the undeuterated compounds phenyl-X these bond contributions can be evaluated separately. They show a decrease with the T aft substituent parameter σI, that is greatest in ortho and smallest in para position. The isotope intensity sum rule is verified for the CH and CD stretching vibrations in all the cases considered.

The experimental determination of the resolving power of infrared spectrometers by means of sharp absorption lines is discussed. Under certain conditions the convolution of the true line profile with the monochromator function approximates completely to the latter, from where the spectral slit width can be obtained immediately. Some suitable lines in the absorption spectra of small gas molecules are suggested for calibration purposes.

The absorption intensities of the CH- and CD-stretching vibrations of the deuterated benzene isotopes C 6 H 6 , C 6 H 5 D, p -C 6 H 4 D 2 , symC 6 H 3 D 3 , and C 6 D 6 have been measured in CCl 4 solution. Independent of the molecular symmetry and the actual form of the normal vibrations each CH(CD) oscillator gives a constant contribution of 0,099 cm 3 /mol to the integrated intensity in the region of the CH- or CD-stretching vibrations respectively. This is a special case of the “Isotope Intensity Sum Rule”. 1. 1. Die Frequenzen der CH- bzw. CD-Valenzschwingungen der gemessenen Benzol-Isotope fallen auf einen engen Bereich zusammen. Die durch Kopplung bewirkte Frequenzaufspaltung betragt maximal 2% des Absolutwerts. 2. 2. Die Absorptionsschwerpunkte fallen fur alle Isotope zusammen. Das Frequenzverhaltnis ν CH /ν CD ist um nur 2% kleiner, als man fur vollig ungekoppelte Oszillatoren erwartet. 3. 3. Die einzelnen Bindungen geben unabhangige Beitrage zur gesamten integralen Intensitat in ihrem Frequenzbereich. 4. 4. Zur Bestimmung der integralen Intensitat mussen samtliche Banden in dem Frequenzbereich erfasst werden, auch wenn ihre Zahl die Zahl der IR-aktiven Schwingungen ubersteigt. Diese Resultate werden in folgenden Untersuchungen benutzt werden, um den Einfluss von Substituenten auf die einzelnen CH-Bindungen in Benzolderivaten durch Intensitatsmessungen zu erfassen [9]. Wahrend die Bindungsintensitaten bei unsubstituiertem Benzol in allen Bindungen gleich sind, werden sie bei den Derivaten in unterschiedlichem Mas mit der Stellung zum Substituenten variieren.

A new method is described for the assessment of the dissolution behavior of solid dosage forms. The method, which is based on the mass transfer between solid and liquid phase in an exchange column, is shown to avoid some disadvantages of the commonly used beaker methods employing fixed liquid volumes. Its usefulness is demonstrated by results obtained with nondisintegrating and uniform granules of benzoic acid in water. The influence of various external parameters, such as liquid flow rate, cell cross-sectional area, amount of material, and particle diameter, is found to agree with theory and literature data. Because of its reproducibility and the absence of arbitrary external parameters, the method seems to be useful for a meaningful study of dissolution kinetics.

It is shown that to a good approximation the CH-stretching vibrations in benzene derivatives are localized in the CH bonds and that the total intensity of a CH-stretching vibration of such a compound is given by the sum of the squares of the derivatives ∂μ|∂Rj: ICH = j ( ∂μ ∂ R j ) ∂μ|∂Rj is the change of the molecular dipole moment with respect to a change of the stretching coordinate Rj of the jth CH bond. Experimental data demonstrate the validity of this approximation. In continuation of our former investigations on the dependence of the CH-stretching vibrations on different substituents we have examined a considerable number of deuterated benzene monoderivatives, para-disubstituted benzene derivatives and 2-, 3- and 4-substituted pyridine derivatives. We have found a relation of general validity between ICH and the Taft substitution constant σI: ICH = F(σI). The interpretation of this result leads to conclusions concerning the effect of substituents on the CH dipole of benzene derivatives.

The relationship between in vitro and in vivo drug release from Oros systems has been examined by analysing plasma concentration data from two pharmacokinetic studies, using a numerical deconvolution technique. This method generates an input profile by comparing the response with that achieved following an instantaneous reference unit dose. The approach is conceptually simple and does not require compartmental pharmacokinetic modelling or curve fitting. In the analysis of the first study, the plasma profile following intravenous dosing was used as the reference function, allowing the combined release/absorption process to be calculated; for the second, an oral bolus was used, the result of the deconvolution therefore indicating the in vivo dissolution rate of the Oros systems. The in vivo release from Oros in most volunteers followed the same pattern as that measured in vitro; only after 6-8 h was the decline in the in vivo release rate somewhat greater than expected. In a few individuals the cumulative absorption profile reached an early plateau level which coincided, on some but not all occasions, with the premature excretion of the Oros system from the body. The amount of drug in recovered systems agreed reasonably with the prediction of the deconvolution analysis.

In a recent series of papers, Kildsig and coworkers (1–4) investigated dissolution from compressed pellets in a static column of solvent, under conditions where the flat surface of the solid was either the lower or the upper boundary of the column (ascending and descending models, respectively). It was concluded that in the descending case an interfacial concentration, c1, significantly lower than the saturation concentration, c1, is operative and that c1, is related to energy parameters of the solid–liquid interface. The argument of the authors is not conclusive, however, and their experimental data can be interpreted in a more simple and convenient way which does not require the assumption of ct < cs.