The availability of nitroglycerin from parenteral solutions

Food-inspired innovations to improve the stability of active pharmaceutical ingredients

Food-processing and pharmaceutical industries share a lot of stability issues against the same physical, chemical, and microbiological phenomena. They also share some solutions to improve the stability as the use of preservatives and packaging. Ecological concerns lead to the development of tremendous innovations in food. Some of these innovations could also be beneficial in the pharmaceutical domain. The objective of this review is to evaluate the potential application of these findings in the pharmaceutical field and the main limits in terms of toxicity, environmental, economic and regulatory issues. The principal factors influencing the shelf-life were highlighted through the description of the stability studies usually performed in the pharmaceutical industry (according to European guidelines). To counter those factors, different solutions are currently available as preservatives and specific packaging. They were described and debated with an overview of recent food innovations in each field. The limits of the current solutions in the pharmaceutical field and the innovation in the food field have inspired a critical pharmaceutical outlook. The active and intelligent packaging for active pharmaceutical ingredients of the future is imagined.

Investigating the effect of humidity on the α-relaxations of low-density polyethylene using dielectric spectroscopy

A previous work from our lab reported the higher sorption of lipophilic preservatives in LDPE form-fill-seal packs that were stored at 75% relative humidity (RH) as against 25% RH. The aim of the present work was to investigate structural changes taking place in LDPE on exposure to higher humidity. The crystallinity of LDPE, determined by differential scanning calorimetry, was found to be similar for the packs charged at both humidity conditions. Dielectric spectroscopy (1.0E-02 Hz to 1.0E02 Hz in the temperature range of 75-87°C), however, showed faster α relaxation of LDPE films exposed to higher humidity. The activation energy of α relaxations was lower at 75% RH (99.735 kJ/mol) than at 25% RH (113.112 kJ/mol) after two weeks of storage. This work presents previously unreported evidence of the plasticization effect of water on LDPE, using dielectric spectroscopy. Furthermore, changes in α relaxation on exposure to humidity support the latest theory of its origin to be from the constrained amorphous regions. The authors suggest the employment of extreme humidity conditions (low and high), during accelerated stability studies of aqueous products in plastic packs to track the sorption loss of formulation components.

Drug structure-transport relationships

Malcolm Rowland has greatly facilitated an understanding of drug structure–pharmacokinetic relationships using a physiological perspective. His view points, covering a wide range of activities, have impacted on my own work and on my appreciation and understanding of our science. This overview summarises some of our parallel activities, beginning with Malcolm’s work on the pH control of amphetamine excretion, his work on the disposition of aspirin and on the application of clearance concepts in describing the disposition of lidocaine. Malcolm also spent a considerable amount of time developing principles that define solute structure and transport/pharmacokinetic relationships using in situ organ studies, which he then extended to involve the whole body. Together, we developed a physiological approach to studying hepatic clearance, introducing the convection–dispersion model in which there was a spread in blood transit times through the liver accompanied by permeation into hepatocytes and removal by metabolism or excretion into the bile. With a range of colleagues, we then further developed the model and applied it to various organs in the body. One of Malcolm’s special interests was in being able to apply this knowledge, together with an understanding of physiological differences in scaling up pharmacokinetics from animals to man. The description of his many other activities, such as the development of clearance concepts, application of pharmacokinetics to the clinical situation and using pharmacokinetics to develop new compounds and delivery systems, has been left to others.

Surface characterisation of bags for total parenteral nutrition by tensiometry and atomic force microscopy

Bags made of poly-ethylene and poly-vinylchloride and of the copolymer ethylene-vinylacetate were used as containers of perfusion solutions for total parenteral nutrition. The bags were characterised by tensiometry (free energy and its polar and dispersed components) and atomic force microscopy (AFM) before and after various periods of storage of solutions for total parenteral nutrition containing L-aminoacids, electrolytes or glucose. In most of the cases, after storage of these solutions, tensiometric characterisation and atomic force microscopy analysis of the internal surface of bags showed deep modifications which highlight the adsorption of the solutes. The changes of surface characteristics were found to depend on the time of contact, the wettability of the polymer and the compounds present into the solutions, while their concentration has a negligible effect. Generally, the aminoacid solutions produced a higher increase in the polar component even after short storage times. Poly-ethylene and the copolymer ethylene-vinylacetate showed a greater inertia if compared with the poly-vinylchloride bags.

The release of nitroglycerin absorbed into the central venous catheter

This study was conducted to evaluate the release of nitroglycerin (NG) that has been absorbed into the central venous catheter. A 0.05% NG solution was infused through a central venous catheter and the flow rates were set at 1, 5, or 10 ml/h, given over 12, 24, or 48 h. The catheter was flushed with lactate Ringer solution after completion of the NG infusion. The elution of the lactate Ringer solution from the tip of the catheter was then collected and assayed for its NG concentration by high performance liquid chromatography (HPLC). A higher concentration of NG was released with a faster flow rate and a longer infusion. The high level of NG release continued during the first 20 min, and ranged from a minimum of 0.07 mg/ml to a maximum that exceeded 0.15 mg/ml. Subsequently, the NG concentration gradually declined, but low concentrations of 0.006-0.02 mg/ml were still maintained 360 min later. Thus, it is suggested that if a catheter such as the Swan-Ganz continues to be used after the completion of a NG infusion, certain pharmacological effects due to the absorption of NG into the catheter body should be expected for at least 60 min.

Modeling solute sorption into plastic tubing during organ perfusion and intravenous infusions

The uptake of solutes into plastic infusion and perfusion tubing has been well documented, but the kinetics of the uptake process is not well-defined. Three mathematical models have been developed to describe the outflow fraction concentration--time profiles for solutes sorbed into the plastic tubing during infusion and perfusions. The models are referred to as model 1, convection--diffusion; model 2, convention-- interfacial resistance--diffusion; ad model 3, convection--interfacial resistance--infinite sink models. In each model, plug flow is assumed and, in order to minimize the number of variables required, solutions are limited to early times when the plastic behaves as an infinite sink. Initial conditions of (i) no solution in the tubing and (ii) a preloading of tubing with drug solution are considered for each of the three models. Two parameters, one being the transit time of solution through tubing (tmin) and the other a measure of the affinity and diffusivity of the solute in the plastic (SN), are sufficient to describe the outflow concentration--time profiles for solutes with sorption into tubing being limited by diffusion in the plastic (model 1). A single parameter, which is the effective interfacial permeability coefficient (H), is sufficient to describe the outflow concentration--time profiles for solutes with sorption into tubing being limited by an aqueous--plastic interfacial barrier (model 3). The three parameters (tmin, SN, and H) are required when uptake into tubing is limited by a combination of diffusion into plastic and an interfacial resistance (model 3). Each model has a characteristic outflow concentration--time profile determined by the relative magnitude of diffusivity of the solute in the plastic to that across the interfacial barrier. The sorption of nitroglycerin and isosorbide dinitrate are adequately described by the convection--diffusion model (model 1 (ii)) whereas the convection--interfacial resistance--diffusion model (model 2 (ii)) is required to describe the sorption of diazepam and chlorpromazine.

Influence of intravenous administration set composition on the sorption of isosorbide dinitrate

The influence of the composition of administration sets on the sorption of isosorbide dinitrate was investigated in-vitro. Isosorbide dinitrate solutions (250 micrograms mL-1) in 0.9% NaCl or 10% glucose were stored in glass containers and administered at a flow rate of 20 mL h-1. The influence of the concentration of different plasticizers (di-ethylhexylphthalate, tri-ethylhexyltrimelitate) in polyvinylchloride tubings was determined. Polybutadiene tubings of different mol. wt coextruded laminates of these polybutadienes with PVC of different composition and a polyethylene tubing were evaluated. The higher the Shore hardness of the PVC tubing, the lower the sorption. The infusion fluid played an additional role only for the tubings with high Shore hardness (greater than 70). The sorption of isosorbide dinitrate to polybutadiene tubings of different mol. wt was less than 2.5% after 5 h and was comparable with the sorption to the polyethylene tubing. When polybutadiene/PVC laminates were used, the sorption increased significantly and was in most cases dependent on the Shore hardness of the PVC (higher Shore hardness gave lower sorptions) and on the mol. wt of the polybutadiene (lower mol. wt resulted in higher sorption). Sorption was not dependent on the type of PVC plasticizer.

Prediction of solute sorption by polyvinyl chloride plastic infusion bags

The time course of the sorption of drugs by polyvinylchloride infusion bags has been approximated using a diffusion model in which the plastic is assumed to act as an infinite sink. This model appears to be suitable for estimation of storage times relevant to clinical usage and enables the magnitude of the uptake in a specific time to be described by a single parameter, referred to as the sorption number. This parameter is defined by the plastic-infusion solution partition coefficient, the diffusion coefficient in the plastic, the fraction un-ionized in solution, the volume of the infusion solution, and the surface area of the plastic. An approximation of the model allows a ready estimation of the sorption number from the fraction remaining in solution at a given time. The sorption number can be extrapolated to allow prediction of the effects of time, plastic surface area, solution volume, and solution pH on fractional solute loss. A reasonable correlation was established between the logarithm of this parameter and the logarithm of the octanol-water partition coefficients of various solutes. The model allows the fraction of a solute remaining in a plastic infusion bag at a given storage time to be estimated from the octanol-water partition coefficient of the solute and other readily available data.

Prediction of drug loss from PVC infusion bags

PVC:water partition coefficients for a series of 13 drugs have been calculated from literature data and a high degree of correlation with octanol:water partition coefficients demonstrated. The resulting model, log PPVC = -0.35 + 0.69 log P. (r2 = 0.88) has been prospectively tested with 10 drugs. All the test drugs were within the 95% confidence intervals associated with predicted log PPVC values consistent with a valid model. In practice, predicted log PPVC values may be used to estimate drug loss from the aqueous phase of PVC bags at equilibrium. Equations are described which enable calculation of likely drug loss from 100, 500 and 1000 mL PVC bags. It is recommended that this approach is used to identify drugs which are unlikely to be significantly absorbed into PVC.

Miconazole sorption to intravenous infusion sets

The sorption of the antifungal drug miconazole onto the plastics of infusion bags and intravenous administration sets has been investigated. HPLC methodology was applied for the determination of miconazole concentrations. The results obtained indicate that the sorption of miconazole is low and is unlikely to result in significant loss of clinical efficacy.

Clinical pharmacokinetics of glyceryl trinitrate following the use of systemic and topical preparations

Glyceryl trinitrate has been used for more than a century for the treatment of angina pectoris and, more recently, for the treatment of congestive heart failure. The introduction of transdermal delivery systems has renewed the controversy regarding the efficacy of the drug, mainly in the light of the development of tolerance. With concentrations of the order of 1 microgram/L or less, the measurement of glyceryl trinitrate in plasma is not easy: gas chromatography with electron capture detection has been used widely but recently gas chromatography-mass spectrometry has provided satisfactory results. Assay problems are most likely to be responsible for some of the unexpected results reported. Further factors which may confound the results of the study of plasma concentrations are the rapid metabolism of glyceryl trinitrate in blood in vitro, adsorption to containers and infusion sets, and the uptake and/or metabolism in vessel walls. From the intravenous infusion data, the large interindividual variability in plasma concentrations of glyceryl trinitrate is apparent. The plasma half-life is about 2 to 3 minutes; plasma clearance values reported vary from 216 to 3270 L/h, indicating extensive non-hepatic metabolism. With transdermal administration, mainly with the transdermal controlled delivery systems, plasma concentrations of glyceryl trinitrate appear to be maintained for up to 24 hours, with large interindividual variations. Despite the ability to maintain, for example with the transdermal delivery systems, relatively constant concentrations of glyceryl trinitrate, it has not been possible to find a relationship between plasma concentrations and pharmacological or clinical effects. This is in part due to the attenuation of the effects with time; from the available data it is clear that this attenuation occurs at a pharmacodynamic level (reflex adaptation and tolerance) and not at the pharmacokinetic level.

Kinetics of sorption of ionizable solutes by plastic infusion bags

Aqueous solutions of several ionizable substances were stored in plastic infusion bags and the sorption of the substances monitored with time. The substances used were p-nitrophenol, p-toluidine, warfarin sodium [3-(alpha-acetonylbenzyl)-4-hydroxycoumarin sodium salt] and trifluoperazine hydrochloride (10-[3-(4-methyl-1-piperazinyl)propyl]-2-(trifluoromethyl)phenothiazine dihydrochloride). The rate and extent of sorption for each substance varied with pH and was consistent with a preferential uptake of the un-ionized species. The uptake of p-nitrophenol and p-toluidine was adequately described by a diffusion model derived assuming that sorption is rate-controlled by the diffusivity of the solute in the plastic matrix, and that only the un-ionized species was sorbed by the plastic matrix. However, the uptake of warfarin sodium and trifluoperazine hydrochloride was described more accurately by a diffusion model in which the diffusional resistance of the plastic matrix and of an interfacial resistance barrier both contributed to the diffusional resistance encountered in the sorption process. It appeared that the rate of uptake of the un-ionized form of these solutes was diminished due to the influence of interfacial or aqueous diffusional barriers. Solute lipophilicity and degree of ionization appeared to be important factors determining the relative contribution of the respective barriers to the overall diffusional resistance.

Effect of infusion administration set on the delivery rate and plasma concentration of nitroglycerin in dogs

A randomized crossover study with eight dogs was carried out to determine whether the in vitro difference observed in the delivery of nitroglycerin using polyvinyl chloride (PVC) and polyethylene (PE) infusion sets would result in a measurable difference in steady-state plasma concentration. At the same apparent infusion rate of 40 micrograms/min, PVC infusion sets produced steady-state plasma nitroglycerin concentrations that were only 40% of those generated with PE sets (p less than 0.0001). Attainment of steady-state drug levels appeared more rapid with the PE administration sets, but in three out of eight animals, these infusion systems appeared to produce a "bolus" dosing effect. PE sets did not reduce the variability in plasma nitroglycerin concentration observed with PVC sets.

Comparison of models describing the sorption of nitroglycerin and diazepam by plastic infusion systems: diffusion and compartment models

The time course of sorption of diazepam and nitroglycerin from aqueous solutions into plastic materials has been represented by the diffusion and compartmental models for a variety of storage conditions. The diffusion model seemed to be the more satisfactory model in respect to both description and prediction of the drug uptake for all conditions. The compartment model appeared to be useful for describing the drug uptake at earlier times, giving a satisfactory fit to the data and reliable final parameter estimates. However, that model was not able to describe the loss as equilibrium was approached or accurately predict the disappearance profiles for these solutes with alterations in solution volume or infusion bag size. Approximations of the diffusion model gave parameter estimates consistent with those obtained by nonlinear regression using the full equations.

Dose dependent pharmacokinetics of nitroglycerin after multiple intravenous infusions in healthy volunteers

Evaluation of the pharmacokinetics of nitroglycerin has been hindered in the past by the lack of specific and sensitive analytical procedures, and the unavailability of parenteral nitroglycerin and infusion sets which did not adsorb nitroglycerin. The purpose for this present study was to determine the pharmacokinetic parameters of nitroglycerin and the dinitrate metabolites after multiple intravenous infusions of nitroglycerin in healthy volunteers. Six volunteers received variable infusion rates of nitroglycerin. Generally, at 0, 40, 80, and 120 min, the infusion rates were adjusted to 10, 20, 40, and 10 micrograms/min, respectively. Plasma samples were drawn and analyzed for nitroglycerin and its 1,2- and 1,3-dinitrate metabolites using capillary GC. Steady-state nitroglycerin plasma concentrations attained at 10, 20, 40, and 10 micrograms/min were 0.44 +/- 0.31, 1.32 +/- 0.71, 4.23 +/- 1.50 and 1.04 +/- 0.43 ng/ml, respectively. As the infusion rate was increased, the steady-state concentrations increased disproportionately. When the dose was decreased from 40 to 10 micrograms/min, the steady-state nitroglycerin concentrations were always higher than those at the initial low infusion rate. Thus, in the majority of subjects, a hysteretic type of response was present. The hysteresis observed in the dose versus steady-state concentration curve may be explained by either end-product inhibition or saturable binding of nitroglycerin to blood vessels. The clearance values (5.5 to 711/min) were very high and far exceed the maximum possible hepatic clearance suggesting that nitroglycerin is metabolized by organs other than liver. Clearance was not directly related to plasma concentrations but was found to decrease to a constant value (approximately 11 +/- 6 l/min) as nitroglycerin concentrations initially increased.

Specialized delivery systems for intravenous nitroglycerin. Are they necessary?

Nitroglycerin is absorbed in vitro into polyvinyl chloride tubing, and it has been recommended that nitroglycerin be administered intravenously through specialized polyethylene infusion sets. To determine if tubing type is essential to achieve physiologic effectiveness, we studied dose responses to intravenous nitroglycerin in 15 patients with heart failure using standard polyvinyl chloride tubing in seven (group 2) and special polyethylene infusion sets in seven (group 1) (one patient was excluded from analysis because of technical difficulties). We monitored heart rate, blood pressure, right atrial pressure, pulmonary artery pressure, pulmonary capillary wedge pressure, and cardiac output. Cardiac index, systemic and pulmonary vascular resistance, triple index, rate pressure product, stroke volume, stroke volume index, and stroke work index were calculated. Baseline and treatment measurements were obtained from five to 15 minutes after the infusion of 10, 20, 40, and 80 micrograms of nitroglycerin per minute. Over-all, systolic blood pressure decreased (p less than 0.05) about 8 percent and mean blood pressure approximately 12 percent, mean pulmonary artery pressure and mean pulmonary capillary wedge pressure decreased 30 to 40 percent, and the decline in mean right atrial pressure was 35 percent of baseline (all p less than 0.05). Heart rate and cardiac index did not change (p greater than 0.05). Pulmonary vascular resistance decreased slightly (p = 0.07) and systemic vascular resistance significantly (p less than 0.05). When the two groups were compared physiologic changes were virtually identical (p less than 0.05). Two-way analysis of variance for baseline corrected data proved no differences between tubing sets (p less than 0.05), but the infusion concentration rate was highly related to response (p = 0.0001). A significant (p less than 0.05) decrease in mean blood pressure and mean right atrial pressure was noted at lower dose rates (20 micrograms per minute and 40 micrograms per minute, respectively) in group 1. Beneficial hemodynamic effects in heart failure patients can, then, be predicted to occur at 80 micrograms per minute infusion rates; these responses seem independent of the type of infusion tubing system employed. Additionally, when patients given intravenous nitroglycerin for various reasons were followed for 48 hours, the majority receiving infusions via polyvinyl chloride tubing (group 2) did not require dosage adjustments. Also, at lower flow rates, more solution than calculated may be delivered when polyethylene tubing infusion sets are employed with volumetric infusion pumps.

Dynamics of clomethiazole edisylate interaction with plastic infusion systems

The dynamics of the interaction of clomethiazole edisylate (1) with polyvinyl chloride and cellulose propionate, the main plastics used in the manufacture of infusion bags and sets, was examined. An experimental system in which the plastic was either open or closed to the environment was used to determine the relative contribution of the sorption and permeation processes to loss from solutions of clomethiazole edisylate (I) in contact with the plastic infusion systems. Sorption by the plastic infusion materials accounted for most of the drug loss, while permeation into the external environment accounted for the remainder. The sorption and permeation into and through polyvinyl chloride was temperature dependent. The diffusion coefficient and permeation rate constant both increased with temperature, while the polyvinyl chloride-water partition coefficients were independent of temperature. The activation energy for the diffusion in polyvinyl chloride was 13.5 kcal/mol. The permeability of the infusion bag plastic and the evaporation across an unstirred air boundary layer adjacent to the external surface of the plastic both appeared to contribute to the overall diffusional resistance encountered in the permeation process. The plastic-water partition coefficients are independent of initial concentration, suggesting that the concentration-dependent loss of the drug from solutions stored in plastic infusion bags and burets is a result of the greater diffusivity of the drug in the plastic at the higher initial concentrations. Plasticization of the polymers by the drug is indicated by the increase in the diffusivity of the drug in polyvinyl chloride and cellulose propionate, the increase in the rate and extent of sorption of a radiolabeled marker (diazepam) by the plastic, and the decreased stiffness of polyvinyl chloride exposed to higher concentrations of the drug.

Drug interactions with medical plastics

Knowledge about drug interactions with plastic materials used in medical and surgical practice is at an elementary stage. Information that has appeared so far on the sorption of drugs to intravenous fluid containers, delivery sets, syringes, or other plastic apparatus has highlighted that polyvinyl chloride (PVC) is the major offender in this respect. Fortunately, in only a few cases is this sorption phenomenon and loss of drug from fluid likely to present a clinical hazard; in most instances, methods are available to prevent or overcome the problem, providing it is recognized. Current information suggests that the following drugs may exhibit clinically significant sorption to plastic materials: insulin, glyceryl trinitrate (nitroglycerin), diazepam, chlormethiazole, vitamin A acetate, isosorbide dinitrate, and a miscellaneous group of drugs including some phenothiazines, warfarin sodium, hydralazine hydrochloride, and thiopentone sodium. In addition, chloroquine binds strongly to glass and to cellulose acetate, but seemingly not to plastics. Brief details of these interactions and their management are given, together with some preliminary information and warnings on drug interactions (e.g., epinephrine, rifampicin) with hydrophilic contact lenses. The latter interactions may cause irreversible coloration of the lenses.

Toxicology studies with a stable intravenous formulation of nitroglycerin

The preclinical toxicologic profile of Nitrostat, a stable parenteral formulation of nitroglycerin, was determined in mice, rats, rabbits and dogs. Single-dose i.v. studies in rodents yielded LD50 values of 17.3 and 18.2 mg kg-1 in male and female mice, and 24.4 and 23.2 mg kg-1 in male and female rats, respectively. Subacute i.v. studies in rats at doses of 2.5, 5.0 and 10.0 mg per kg per day, and in dogs at doses 1.0 and 3.0 mg per kg per day for two weeks, elicited minimal reactions. In rats, suppression of body-weight gain and food consumption occurred among treated and vehicle-control animals. Mild tissue irritation at injection sites was noted in treated and vehicle-control groups. There were no clearly drug-related clinical or pathological findings in dogs. In rabbits, repeated intravenous administration of Nitrostat did not induce significant local venous irritation. The results of these studies indicated that the stabilized parenteral formulation of nitroglycerin did not elicit unusual toxic properties in intravenous infusion studies.

Pharmaceutical considerations of nitroglycerin

During the past few years, there have been rapid changes in the pharmaceutical uses of nitroglycerin. New dosage forms and new delivery systems have become available, which have resulted in potential confusion to all concerned with the proper use of these systems. The goal of this review is to prevent confusion and to bring all the relevant information together. The various analytical techniques available for quality control of the dosage forms and for the study of the pharmacokinetics are reviewed, with the intent of enabling the reader to identify pertinent references rapidly. The interaction of nitroglycerin with packaging and plastic delivery devices is also reviewed so that the reader can make informed choices. Finally, the clinical pharmacy and pharmacokinetics are reviewed so as to bring the reader up to date in that area. After reading this article, the areas of nitroglycerin research that still need to be explored should be apparent.

The compatibility of diazepam with infusion fluids and their containers

A significant interaction exists between diazepam and infusion fluid containers constructed of PVC plastics. This is attributed to absorption of the drug into the PVC matrix and should preclude the use of this type of container prior to or during infusion of diazepam admixtures. Infusions containing diazepam injection either as a solution (in dextrose 5% or sodium chloride 0.9%) or oil-in-water emulsion (in dextrose 5%) exhibit no significant interaction over 48 h when prepared in polyethylene (Polyfusor) containers. No data were obtained regarding the physical stability of diluted emulsion. The inert nature of polyethylene and an absence of plasticizers or other additives account for a good compatibility with the drug additive. Therefore, diazepam dilutions may be prepared in polyethylene containers as an alternative to glass. However, the interaction with administration sets should not be overlooked when considering these solutions for infusion.

Nitroglycerin intravenous infusion

The treatment of various cardiovascular problems with intravenous nitroglycerin is widely practiced though unapproved. The uses of iv nitroglycerin include improvement of the hemodynamics of left ventricular failure and cardiogenic shock associated with infarction, control of hypertension during coronary artery surgery, and possibly, reduction of acute myocardial infarct size. The popularity of this treatment in the absence of an FDA-approved commercial product has forced some hospitals to prepare the drug extemporaneously, and the advantages and disadvantages of preparation and storage techniques are presented.

Nitroglycerin stability: effects on bioavailability, assay and biological distribution

The solution, absorptive, and adsorptive properties of nitroglycerin are related to problems encountered in the intravenous use of solutions of the drug and in its assay in biological tissues. Solutions of the drug in usual intravenous fluids are quite stable to hydrolysis under neutral conditions. Loss to plastic intravenous delivery set components by rapid adsorption and slower absorption processes present a significant clinical problem. Assay of the drug and its major metabolites is complicated by problems of extraction related to the solubility of nitroglycerin, its metabolites and substances in the plasma. Loss of nitroglycerin incubated with red blood cells is a very rapid process (half-life 4.0 min. at 10 ng/ml). The mechanism appears to be a physical process rather than an enzymatic one. The loss to blood cells and perhaps other biological materials should be considered in an analysis of the distribution of the drug.

Availability of isosorbide dinitrate, diazepam and chlormethiazole, from i.v. delivery systems

The loss of isosorbide dinitrate from aqueous solutions stored in plastic infusion bags and/or infused through plastic giving sets was investigated. During simulated infusions, the loss of isosorbide dinitrate was found to be flow-rate dependent. The clinical and pharmacokinetic significance of this loss is discussed. Infusion o isosorbide dinitrate from a glass syringe through high density polyethylene tubing overcame the loss associated with its administration via plastic infusion bags and intravenous giving sets. This method was also applied successfully to minimise the previously reported loss of diazepam and chlormethiazole during infusions.