A stability study of amphotericin B, colistin and tobramycin in a hydrophilic suspension commonly used for selective decontamination of the digestive tract by HPLC and in vitro potency measurements.

OBJECTIVES: A suspension for oral use which consists of three non-absorbable antibiotics (amphotericin B, colistin and tobramycin) is often used in clinical practice for the selective decontamination of the digestive tract (SDD) of patients in intensive care. Such a therapy is a preventive tool to minimise the risk of pneumonia and bacteraemia in intubated patients. The administration and the treatment results are controversially discussed. One limiting factor for a unique SDD treatment in the hospitals is a lack of adequate data regarding batch formula and stability for such a formulation. Since no detailed procedures, specifications or stability data are available for manufacturing this formulation there may be discrepancies regarding formulation and stability of suspensions prepared in different pharmacies. The aim of this research was to collect the physicochemical and microbiological stability data of a developed, stable standard formulation under defined storage conditions. The effectiveness of the SDD suspension should be preferably proven over a long period. This would help guarantee that all patients receive the same preparation, therefore, ensuring similar efficacy and improved safety. METHODS: An adequate formulation composed of the registered, marketed medicinal product Ampho-Moronal suspension (Dermapharm AG, Germany) and a buffered, preserved aqueous solution of colistin and tobramycin both as sulfates has been developed. A stability study has been performed on two batches of the formulation. During the storage, samples were taken and compatibility was verified by physicochemical and microbiological testing in stability-indicating terms of colour, odour, flavour, pH, chemical and microbiological purity as well as in vitro potency. The test methods were built and tailored to be suitable, reliable and precise for the test needs. RESULTS: The results show the physicochemical and microbiological stability of the described formulation for defined storage conditions. CONCLUSIONS: A standardised formulation with a proven stability for at least 6 months under fridge (5°C±3°C) conditions for the SDD of patients in intensive care was established.

Purity determination of amphotericin B, colistin sulfate and tobramycin sulfate in a hydrophilic suspension by HPLC.

A suspension comprising of the three antibiotic substances amphotericin B, colistin sulfate and tobramycin sulfate is often used in clinical practice for the selective decontamination of the digestive tract of patients in intensive care. Since no detailed procedures, specifications or stability data are available for manufacturing this suspension, there may be discrepancies regarding formulation and stability of suspensions prepared in different pharmacies. The aim of this work is to develop a standardized formulation and to determine its stability under defined storage conditions. This would help guarantee that all patients receive the same preparation, therefore ensuring similar efficacy and improved safety. The first step in this process is to develop the required analytical tools to measure the content and purity of the drug substances in this complex mixture. In this paper, the development and validation of these tools as well as the development of the drug suspension formulation is described. The formulation comprises of Ampho-Moronal(®)-Suspension (Dermapharm) and a buffered, preservated aqueous solution of colistin sulfate and tobramycin sulfate. Two simple, well established high-performance liquid chromatography (HPLC) methods in the European Pharmacopoeia (EP) for impurity profiling of the two active ingredients amphotericin B and colistin sulfate were combined with a newly developed sample extraction procedure for the suspension. Sufficient selectivity and stability-indicating power have been demonstrated. Additionally, a new robust routine method was developed to determine possible degradation products of tobramycin sulfate in the investigated suspension. The specificity, precision, accuracy and linearity of the analytical procedures were demonstrated. The recovery rate was in the range of 90-110%. The precision results for the calculated impurities showed variation coefficients of <10%. The calibration curves were found to be linear with correlation of greater than 0.9994 for all components. The results show the suitability of the methods for the quality control analysis of the suspension.

Development and validation of a rapid stability indicating HPLC-method using monolithic stationary phase and two spectrophotometric methods for determination of antihistaminic acrivastine in capsules.

Simple, rapid and accurate high performance liquid chromatographic (HPLC) and spectrophotometric methods are described for determination of antihistaminic acrivastine in capsules. The first method (method A) is based on accurate, sensitive and stability indicating chromatographic separation method. Chromolith® Performance RP-18e column, a relatively new packing material consisting of monolithic rods of highly porous silica, was used as stationary phase applying isocratic binary mobile phase of ACN and 25 mM NaH2PO4 pH 4.0 in the ratio of 22.5:77.5 at flow rate of 5.0 mL/min and 40°C. A diode array detector was used at 254 nm for detection. The elution time of acrivastine was found to be 2.080±0.032. The second and third methods (methods B and C) are based on the oxidation of acrivastine with excess N-bromosuccinimide (NBS) and determination of the unconsumed NBS with, metol-sulphanilic acid (λmax: 520 nm) or amaranth dye (λmax: 530 nm). The reacted oxidant corresponds to the drug content. Beer's law is obeyed over the concentration range 1.563-50, 2.0-20 and 1.0-10 µg mL(-1) for methods A, B and C, respectively. The limits of detection and quantitation were 0.40, 0.292 and 0.113 µg mL(-1) and 0.782, 0.973 and 0.376 µg mL(-1) for methods A, B and C, respectively. The HPLC method was validated for system suitability, linearity, precision, limits of detection and quantitation, specificity, stability and robustness. Stability tests were done through exposure of the analyte solution for four different stress conditions and the results indicate no interference of degradants with HPLC-method. The proposed methods was favorably applied for determination of acrivastine in capsules formulation. Statistical comparison of the obtained results from the analysis of the studied drug to those of the reported method using t- and F-tests showed no significant difference between them.

Evaluation of (S)- and (R)-misonidazole as GPX inhibitors: synthesis, characterization including circular dichroism and in vitro testing on bovine GPx-1.

Racemic misonidazole, a radiosensitizer formally used in radiation therapy of cancer and to date still applied, was once reported to exhibit strong inhibitory effects on mouse glutathione peroxidases (GPX). This appeared to qualify misonidazole as a lead structure for the development of novel GPX inhibitors to cause oxidative stress in chemotherapy-resistant tumors. A unique feature of misonidazole as an inhibitor of GPX is the absence of a thiol functionality. Therefore, it was expected to selectively target inhibition devoid of promiscuous interactions with cations and sulfhydryl groups. We synthesized the isomers of misonidazole and analyzed the ability of chiroptical high-performance liquid chromatography (HPLC) to identify the particular enantiomers. Due to the chiral pool synthesis, the assignment of the correct configuration could be verified. Finally, we evaluated both isomers for their inhibitory activities on bovine erythrocyte GPx-1, which is 87% homologous to the human enzyme. Despite the previously reported inhibition of racemic misonidazole on the less homologous mouse GPx-1, we did not find any significant inhibitory activity on the bovine enzyme for either isomer. Though misonidazole appears unlikely to be an inhibitor of human GPx-1 activity, we still spotlight misonidazole as a promising fragment-like lead structure in general.

Intergroup cross-comparison for the evaluation of data-interchangeability from various chromatographic tests.

The recently dramatic increase in the available choices of reversed-phase columns could be an advantage of this mode of separation. However, due to the insufficiency of available information in terms of the exact functionality of these phases and the similarities and differences between these newly introduced and conventional reversed-phase columns, it is now somehow problematic to determine which could be the best column for a given analytical problem. There is no single column that will give us a good separation for all applications. As a result, there have been several attempts to develop testing strategies to characterize column chemistries. In this study three of the most widely used and acceptable approaches for the characterization of reversed-phase columns, which are Tanaka, United States Pharmacopeia (USP), and Snyder-Dolan, are systemically applied to investigate the chromatographic properties of calixarene- and resorcinarene-bonded stationary phases, polar-embedded and polar-endcapped stationary phases, phenyl and ether-linked phenyl with the presence of conventional alkyl-bonded phases (octyl- and octadecylsilane). Although all column classification systems aim to evaluate "more or less" the same characteristics, each system uses different test mixtures in different chromatographic conditions. It is therefore very important to evaluate the similarities and differences in the resulted "column parameters" and the possible interchangeability of them. The results of this comparative study show that the used parameters of Tanaka and of Snyder-Dolan have in many cases a good to very good correlation. The USP approach, which is based on single run, is related to Tanaka and Snyder-Dolan only in terms of hydrophobic characters, and no relation could establish in the other parameters. The hydrophobic-subtraction model could be extended to describe the ligand-solute interactions of calixarene- and resorcinarene-bonded stationary phases, which are belonging to reversed phase material. However they show, depending on the analytes, some additional interactions, since their steric, polar and ionic properties are different compared to those of conventional alkyl-bonded phases.

Analytical power of LLE-HPLC-PDA-MS/MS in drug metabolism studies: identification of new nabumetone metabolites.

Nabumetone is a non-acidic, nonsteroidal anti-inflammatory prodrug. Following oral administration, the prodrug is converted in the liver to 6-methoxy-2-naphthylacetic acid (6-MNA), which was found to be the principal metabolite responsible for the NSAID effect. The pathway of nabumetone transformation to 6-MNA has not been clarified, with no intermediates between nabumetone and 6-MNA having been identified to date. In this study, a new, as yet unreported phase I metabolite was discovered within the evaluation of nabumetone metabolism by human and rat liver microsomal fractions. Extracts from the biomatrices were subjected to chiral LLE-HPLC-PDA and achiral LLE-UHPLC-MS/MS analyses to elucidate the chemical structure of this metabolite. UHPLC-MS/MS experiments detected the presence of a structure corresponding to elemental composition C15H16O3, which was tentatively assigned as a hydroxylated nabumetone. Identical nabumetone and HO-nabumetone UV spectra obtained from the PDA detector ruled out the presence of the hydroxy group in the aromatic moiety of nabumetone. Hence, the most likely structure of the new metabolite was 4-(6-methoxy-2-naphthyl)-3-hydroxybutan-2-one (3-hydroxy nabumetone). To confirm this structure, the standard of this nabumetone metabolite was synthesized, its spectral (UV, CD, NMR, MS/MS) and retention properties on chiral and achiral chromatographic columns were evaluated and compared with those of the authentic nabumetone metabolite. To elucidate the subsequent biotransformation of 3-hydroxy nabumetone, the compound was used as a substrate in incubation with human and rat liver microsomal fraction. A number of 3-hydroxy nabumetone metabolites (products of conjugation with glucuronic acid, O-desmethylation, carbonyl reduction and their combination) were discovered in the extracts from the incubated microsomes using LLE-HPLC-PDA-MS/MS experiments. On the other hand, when 3-hydroxy nabumetone was incubated with isolated rat hepatocytes, 6-MNA was detected as the principal metabolite of 3-hydroxy nabumetone. Hence, 3-hydroxy nabumetone could be the missing link in nabumetone biotransformation to 6-MNA (i.e. nabumetone→3-hydroxy nabumetone→6-MNA).

Spectrophotometric and stability-indicating high-performance liquid chromatographic determinations of terbutaline sulfate.

Spectrophotometric and stability-indicating HPLC procedures are described for determination of terbutaline sulfate in bulk powder and dosage form. The first procedure is based on diazo coupling of the phenolic groups of terbutaline sulfate with fast red B salt in the presence of sodium hydroxide. The colored compound developed in alkaline medium was measured at 475 nm. Different variables affecting the reaction were studied. Beer's Law is obeyed in the concentration range of 1-6 microg/mL. In the HPLC procedure, the separation was carried out on a Caltrex AIII column, a relatively new packing material consisting of silica-bonded calix[8]arene, using an isocratic binary mobile phase, acetonitrile-ammonium acetate (50 + 50, v/v), at pH 6.2. A diode array detector was used at 280 nm. The method was validated for system suitability, linearity, precision, LOD, LOQ, specificity, stability, and robustness. The LOD and LOQ were 0.196 and 0.781 microg/mL, respectively. The recovery values of this method were between 98 and 102%, and the reproducibility was within 0.92%. Statistical comparison of the results obtained from the analysis of the studied drug to those of the official British Pharmacopoeia (2007) method using t- and F-tests showed no significant difference between them.

Effect of chromatographic conditions on liquid chromatographic chiral separation of terbutaline and salbutamol on Chirobiotic V column.

In this study, a method for enantioseparation of terbutaline and salbutamol was established using Chirobiotic V column as a stationary phase. Polar ionic mode applying mobile phase containing ammonium nitrate in 100% ethanol, pH 5.1 was found to give the best separation. The salt concentration in the mobile phase and pH value were found to be the most important chromatographic factors affecting separation. Separation of enantiomers of these two basic analytes was complete in less than 10 min without applying ammonium trifluoroacetate (ATFA) or triethylamine (TEA) salts.

Selectivity of calixarene-bonded silica phases in HPLC: Description of special characteristics with a multiple term linear equation at different methanol concentrations.

Retention and selectivity characteristics of different calixarene-, resorcinarene- and alkyl-bonded stationary phases are examined by analyzing a set of test solutes covering the main interactions (hydrophobic, steric, ionic, polar) that apply in HPLC. Therefore Dolan and Snyder's multiple term linear equation has been adapted to fit the properties of calixarene-bonded columns. The obtained parameters are used to describe retention and selectivity of the novel Caltrex(®) phases and to elucidate underlying mechanisms of retention. Here, differences of stationary phase characteristics at different methanol concentrations in the mobile phases are examined. Both selectivity and retention were found to depend on the methanol content. Differences of these dependencies were found for different stationary phases and interactions. The differences between common alkyl-bonded and novel calixarene-bonded phases increase with increasing methanol content.

Characterization of calixarene-bonded stationary phases.

Calixarene-bonded stationary phases received growing interest in HPLC as stationary phases with special retention characteristics and selectivity. The commercially available unsubstituted and p-tert-butyl-substituted Caltrex(®) columns have been intensively studied and characterized in our workgroup. They can be used as reversed phases, yet they support additional interactions. Especially, their steric, polar and ionic properties differ from conventional alkyl-bonded phases. However, also the hydrophobic interaction shows differences since adsorption and partition interactions on or in a bonded layer of calixarenes are not similar to those of alkyl-bonded layers. The relative strength of the hydrophobic properties of the stationary phases has been found depending on the methanol concentration of the mobile phase. Generally, the dependencies of their interaction strengths on mobile-phase conditions, e.g. the change of the intensity of the hydrogen-bonding abilities with decreasing methanol content, are not similar from phase to phase either. This probably gives calixarene-bonded stationary phases enhanced suitability for analyses at extreme compositions of the mobile phase. An overview about the synthesis, retention and selectivity properties of Caltrex(®) columns is given here.

Description of retention characteristics of calixarene-bonded stationary phases in dependence of the methanol content in the mobile phase.

Calixarene-bonded stationary phases in HPLC are known to support additional interactions compared to conventional alkyl-bonded phases (pi-pi interactions, complex-building interactions). Thus it cannot be presumed that the same mechanisms of retention apply and that retention can be predicted in similar ways. Here 31 solutes of highly various molecular structures have been analysed at different mobile phase compositions (0-98% (v/v) methanol) in order to characterise the chromatographic behaviour of the novel stationary phases and to test the applicability of established models predicting retention factors. The influence of a change of the methanol content is discussed for non-polar, polar and ionic solutes and differences of their behaviour on the differing column types are shown. Additionally estimates about underlying retention mechanisms are given.

Selectivity of calixarene-bonded silica-phases in HPLC: description of special characteristics with a multiple term linear equation at two different pH-values.

Six different calixarene-bonded phases were characterized by analyzing 36 and 26 solutes at pH 3 and 7, respectively. Dolan and Snyder's multiple term linear equation was used to correlate retention factors k' to parameters of the solutes and columns. The column parameters have been related to molecular properties of the stationary phases and new suggestions were made for the interpretation of steric selectivity. Ionic and polar interactions have been found dependent on pH value, while steric interactions are less dependent and hydrophobic interactions remain unchanged. Distinct differences of the supported interactions were confirmed between the calixarene-bonded and the common alkyl-bonded silicas. By use of the parameters, values of k' can be estimated with an average deviation of 2.50 and 7.92% at low and neutral pH-value, respectively.

Comparison of the chromatographic behavior of tricyclic neuroleptics on calixarene-bonded, monolithic and conventional RP-HPLC columns.

The effect of different chromatographic conditions, such as buffer concentration and type of organic modifier, on the retention behavior of nine tricyclic neuroleptics on three different RP-HPLC columns was investigated. Two recently developed columns, calixarene-bonded (CALTREX) AIII) and monolithic (Chromolith) Performance RP-18e) columns, were compared with a conventional RP-C18 HPLC column (LiChrospher). The results showed how the mobile phase conditions had different effects on the analyte retention on these three columns. For example, the elution order of some analytes and the initiation of separation of the geometric isomers of the three analytes--which have E/Z-isomers (cis/trans-isomers)--could be altered by changing the conditions and the column type. Under identical conditions, a calixarene-bonded phase was the best for this separation, a monolithic phase gave comparable results and the conventional RP-column was the least effective. Concerning the geometric isomers separation, the Chromolith Performance RP-18e was superior.

Achiral and chiral high-performance liquid chromatographic determination of flubendazole and its metabolites in biomatrices using UV photodiode-array and mass spectrometric detection.

Flubendazole, methyl ester of [5-(4-fluorobenzoyl)-1H-benzimidazol-2-yl]carbamic acid, belongs to the group of benzimidazole anthelmintics, which are widely used in veterinary and human medicine. The phase I flubendazole biotransformation includes a hydrolysis of the carbamoyl methyl moiety accompanied by a decarboxylation (hydrolysed flubendazole) and a carbonyl reduction of flubendazole (reduced flubendazole). Flubendazole is a prochiral drug, hence a racemic mixture is formed during non-stereoselective reductions at the carbonyl group. Two bioanalytical HPLC methods were developed and validated for the determination of flubendazole and its metabolites in pig and pheasant hepatic microsomal and cytosolic fractions. Analytes were extracted from biomatrices into tert-butylmethyl ether. The first, achiral method employed a 250 mm x 4 mm column with octylsilyl silica gel (5 microm) and an isocratic mobile phase acetonitrile-0.025 M KH(2)PO(4) buffer pH 3 (28:72, v/v). Albendazole was used as an internal standard. The whole analysis lasted 27 min at a flow rate of 1 ml/min. The second, chiral HPLC method, was performed on a Chiralcel OD-R 250 mm x 4.6 mm column with a mobile phase acetonitrile-1 M NaClO(4) (4:6, v/v). This method enabled the separation of both reduced flubendazole enantiomers. The enantiomer excess was evaluated. The column effluent was monitored using a photodiode-array detector (scan or single wavelength at lambda=246 nm). Each of the analytes under study had characteristic UV spectrum, in addition, their chemical structures were confirmed by high-performance liquid chromatography-mass spectrometry (HPLC-MS) experiments. Stereospecificity in the enzymatic carbonyl reduction of flubendazole was observed. While synthetic racemic mixture of reduced flubendazole was separated to equimolar amounts of both enantiomers, practically only one enantiomer was detected in the extracts from all incubates.

Effect of chaotropic mobile phase additives on retention behaviour of beta-blockers on various reversed-phase high-performance liquid chromatography columns.

In this study the effect of type and concentration of chaotropic counter anion in buffered mobile phase on retention behaviour of some beta-blockers was studied. Two types of anions (perchlorate, dihydrogen phosphate) were examined under different concentrations. Further different pH-values for each anion were used to investigate the effect of pH changes on chaotropic behaviour of these anions. The role of organic modifier (methanol, acetonitrile) type was also considered. All these parameters were studied for three different RP-columns (calixarene modified silica gel, monolithic RP-column and a conventional RP-column). The results indicate that all studied factors affected the chaotropic behaviour of the tested anions. Also the type of the used stationary phase has found to play an important role in retention behaviour of beta-blockers.

Retention behaviour of beta-blockers in HPLC using a monolithic column.

The chromatographic behaviour of a new HPLC-stationary phase (Chromolith RP-18e) is described for separation of beta-blockers. The effects of the different chromatographic conditions (buffer system, pH value, content of organic modifier, injection volume and flow rate) on the separation behaviour were studied. At higher flow rates the peaks seemed to be more symmetrical than at lower flow rates. The use of buffer or salt in the mobile phase for this separation is found to be very essential and the counter-anion type of this buffer or salt significantly affected the retention behaviour of beta-blockers while the cation type did not play the same important role.

Peak shape improvement of basic analytes in capillary liquid chromatography.

The analysis of bases is of special interest in pharmaceutical research because numerous active substances contain basic functional groups. Capillary and conventional size LC separations of drug substances spiked with potential impurities were compared. In the case of the nonpolar drug levonorgestrel equal separation efficiency was readily attained by both techniques. The peaks of basic substances, however, showed extensive tailing when separated by capillary LC. The peak deformation was attributable to interactions of the basic substances with the polar inner surface of the fused silica capillaries employed in capillary LC and does not appear with the steel tubing generally used in conventional size LC. This drawback of capillary LC was overcome by use of deactivated fused silica capillaries for column hardware and transfer lines.

Evaluation of packed capillary liquid chromatography columns and comparison with conventional-size columns.

Apart from extracolumn effects peak dispersion in liquid chromatographic columns is caused by the column inlet, the packed bed, and the column outlet. A strategy applicable for independent evaluation of the individual sources of column band broadening was developed on the basis of the linear extrapolation method (LEM). This method was applied to compare the performance of packed capillary LC columns from various commercial suppliers with conventional-size columns. The columns differed widely in their performance with respect to peak shapes and widths for standard substances. The capillary columns were found well packed, but in some cases overall performance would benefit from improving the design of the area between the packed bed and the connecting capillaries, containing frits as well as dead volumes.

Extracolumn band broadening in capillary liquid chromatography.

A commercially available capillary LC instrument was modified to investigate and control the contribution of different instrument components on extracolumn band broadening. Quantitative estimations of dispersion induced by several equipmental parts were carried out. Injection parameters could be optimized to achieve the theoretical value of 12 for a profile factor describing a rectangular sample profile. Additionally, an additive injector flow channel dependent dispersion effect was found. A practical approach for minimizing instrumental effects in capillary LC is suggested. The results were compared with those obtained with an HPLC instrument designed for conventional size columns.

Separation of (Z)- and (E)-isomers of thioxanthene and dibenz[b,e]oxepin derivatives with calixarenes and resorcinarenes as additives in nonaqueous capillary electrophoresis.

Five acidic calix[4]arenes with carboxylic or sulfonic groups at either the upper or lower rim of the cavity and one resorc[4]arene were investigated to separate three thioxanthenes (flupentixol, clopenthixol, chlorprothixene) and a dibenz[b,e]oxepin derivative (doxepin) with cis-/trans-isomerism by nonaqueous capillary electrophoresis (NACE). Partial filling of the capillary with the UV-absorbing selectors led to a low detection limit and an advantageous signal-to-noise ratio (S/N). A sufficient electrophoretic mobility of the calixarenes towards the anode was necessary to outweigh the oppositely directed electroosmotic flow (EOF). This depended from the functional groups, the dissociation and the hydrodynamic radius of the cyclophanes. In contrast, the resorcinarene was useable only by addition of sodium dodecyl sulfate (SDS) because only the complex of the two selectors had an anodic apparent electrophoretic mobility. p-Sulfonyl-calix[4]arene (ss-a1) was the most capable selector for all E/Z-isomers with maximal alpha-values ranging from 1.056 for doxepin to 1.224 for chlorprothixene. This was due to the sufficient migration in reversed direction to the EOF even at low pH* values of 3.0. Otherwise, electrostatic as well as hydrophobic interactions with the positively charged isomers seem to contribute to a superior recognition. Increasing the concentration up to 15 mM ss-a1 and using acidic media (pH* 5.0) led to high separation efficiency. Changing the organic solvent provides a powerful tool to improve selectivity with N,N-dimethylformamide-methanol (DMF-MeOH)-mixtures for thioxanthenes. Further electrophoretic parameters were optimized, such as the concentration of the electrolytes, the addition of SDS, the kind of electrolytes and the voltage. Distinct differences in selectivities were found between the derivatives with thioxanthene and dibenzo[b,e]oxepin ring system, respectively. Further, the different basic side chain was responsible for the different selectivity at higher pH* values. In contrast, the substitution at position 2 of the thioxanthenes played a secondary role. Based on the studies of single parameters a method for a simultaneous separation of the four pairs of isomers within 13 min was developed.