MANUFACTURING OF DRUGS UNDER PRESSURIZE.
Aerosol drug delivery to the lungs has long been the route of choice for the treatment of respiratory diseases, including asthma and chronic obstructive airway disease. Metered dose inhalers (MDIs), dry powder inhalers (DPIs), and nebulizers have been employed to successfully deliver a wide range of pharmaceuticals principally to the lungs for local action. However, with their unique characteristics, the lungs have now begun to be targeted as a means of noninvasive delivery of systemically acting compounds, including genes, proteins, peptides, antibiotics, and other small molecules. The primary function of the respiratory tract is gaseous exchange, transferring oxygen from the inspired air to the blood and removing carbon dioxide from the circulation. This pulmonary circulation offers rapid absorption and systemic distribution of suitable drugs deposited in the airways. Due to its anatomical structure, however, an important secondary role is the protection of the body from inhalation of foreign particles (including aerosol drug particles). The challenge of aerosol drug delivery is to overcome this highly effective barrier and accurately and reproducibly deliver aerosol drug particles in suffi cient doses to their targeted sites within the lungs for either local action or systemic absorption. Effective aerosol drug delivery is tied to the aerosol inhaler that generates and delivers the respirable aerosol. This chapter will primarily focus on aerosol drug delivery devices, their development, and future prospects for pulmonary administration.
HUMAN RESPIRATORY TRACT AND AEROSOL PARTICLE DEPOSITION
Human Respiratory Tract
The human respiratory tract can be divided into three main regions: fi rst, the upper airways, including the nose, mouth and throat (oropharnyx), and the larynx. The conducting airways consist of the regions from the trachea to the respiratory bronchioles and have airway diameters between 0.6 and 20 mm. The alveolar region consists of respiratory bronchioles and alveolar sacs and has airway diameters between 0.2 and 0.6 mm. The lungs are a branching system which commences asymmetrically, dividing fi rst at the base of the trachea. The left and right bronchi branch dichotomously into the conducting airways. There are approximately 23 generations before the respiratory bronchioles give way to the alveoli, the site of gaseous exchange. This branching produces a progressive reduction in airway diameter and also signifi cantly increases the total surface area of the lower airways.
Another important characteristic with respect to drug delivery is the extensive vascular circulation. The blood vessels supplying the conducting airways are part of the systemic circulation. In contrast, the alveolar region is connected to the pulmonary circulatory pathway; drugs absorbed into this circulation will avoid fi rst – pass hepatic metabolism effects.
Mechanisms of Particle Deposition
Aerosol particles are deposited in the lungs by three main mechanisms, and the site of deposition is dependent upon the predominating mechanism. Inertial impaction occurs because a particle traveling in an air stream has its own momentum (the product of its mass and velocity). As the direction of the airfl ow changes due to a bend or obstacle, the particle will continue in its original direction for a certain distance because of its inertia. Particles with a high momentum, due to high velocity or large size, are often unable to change direction before they impact on the surface in front of them. Impaction of particles entering the mouth with a high velocity occurs either at the back of the mouth or at the bend where the pharynx leads to the trachea. Only a small fraction of particles greater than 15 μ m will reach the trachea following mouth breathing. The majority, due to their size, will impact in the oropharyngeal region. Deposition by impaction will also occur as the trachea splits into the left and right bronchus. As the velocity of the particles decreases, inertial impaction becomes a less important mechanism of deposition in the smaller airways.
Following the removal of larger particles in the upper airways by inertial impaction, gravitational sedimentation is the mechanism by which smaller particles (2 – 5 μ m) are deposited in the respiratory bronchioles and alveoli. These particles settle under gravity and accelerate to a steady terminal velocity when the gravitational force is balanced by the resistance of the air through which it is traveling. It is a time – dependent process which is aided by breath holding. Brownian motion or diffusion is a mechanism which signifi cantly affects only particles less than 0.5 μ m in diameter. These particles are subjected to bombardment by surrounding gas molecules causing random movement of the particles. In this situation, the diffusivity of a particle is inversely proportional to its diameter. For an extensive mechanistic review of the area of particle deposition readers should consult Finlay (2001).
Aerosol particle size and polydispersity are major determinants of the site and mechanism of pulmonary deposition. Fundamental deposition studies using monodisperse aerosols together with mathematical models have established the optimum aerosol particle size for lung depositio Aerosols larger than 10 μ m will deposit in the oropharyngeal region and will not be inhaled. Particles less than 3 μ m will be capable of penetrating into the alveolar region. Aerosols in the size range 3 – 10 μ m will be distributed in the central and conducting airways. A polydisperse aerosol containing a range of these particle sizes will allow deposition throughout the lungs. In theory, lung site deposition targeting should be possible by controlling the particle size of the inhalation aerosol. However, a number of other signifi cant variables can affect deposition within the respiratory tract and these often confound any efforts at targeting. The patient ’ s respiratory cycle, both the rate and depth of breathing, will affect aerosol deposition, and this is also the source of large intersubject variability in deposition. Slow and deep inhalations are required for deposition in the peripheral airways, and this is the technique often recommended for inhalation with the MDI. A different technique may be required for DPIs, where the patient ’ s inspiratory effort is often the powder dispersion and delivery force. Flow rates greater than 60 L/min are commonly employed for powder inhalers. A fi nal respiratory maneuver can be employed to promote deposition; breath holding up to 10 s is generally recommended to enhance deposition by sedimentation. Other parameters that will affect lung deposition are the disease state within the lungs and its effect on airway caliber together with the patient ’ s age and airway morphology.
Pharmacokinetics
Once deposited on the surface of the airways, the particle is subject to absorption and clearance processes depending upon its physical properties and the site of deposition. For example, a lipophilic small molecule deposited in the central airways would have a different pharmacokinetic profi le than a 50 – kDa macromolecule deposited in the alveolar region. The former may undergo mucociliary clearance following deposition on a ciliated epithelial cell. Following dissolution, lipophilic drugs may be transported across the epithelium by passive transcytosis, while hydrophilic compounds are taken up by other pathways such as via tight junctions and endocytosis. Having overcome the barrier of the epithelial layer, the drug is available for distribution into the systemic circulation or to its site of action. Finally, the drug may also be subject to metabolism within the airways. For the macromolecule deposited in the peripheral airways, the absorption rate has been shown to be dependent upon molecular size. Larger molecules are subject to active processes such as caveolae or vesicular transport across the cell. Diffusion remains the predominant mechanism for smaller lipophilic macromolecules. Insoluble molecules can be phagocytosed by alveolar macrophages and removed via the lymphatic system or the mucociliary escalator. The pharmacokinetics of inhaled drugs is complicated by the fact that a signifi cant fraction of the delivered dose is deposited in the oropharnyx or removed from the lungs via mucociliary clearance and in both cases subsequently swallowed. An often desired goal for a pulmonary formulation is prolonged action within the lung. Rapid clearance or metabolism results in short duration of action for most inhaled drugs. A number of approaches using formulation excipient additives have been investigated to increase the residency or prolong release of drug at its site of action within the lungs. Microspheres containing nanoparticles have been formulated as dry powders for inhalation offering sustained – release properties. In addition, prodrugs which are activated locally within the lungs have been used in an alternative approach.
The pharmacokinetic process of absorption, distribution, metabolism, and excretion within the lungs is an enormous subject area and readers are referred to specifi c reviews for further details. Of particular interest may be the subject of absorption enhancer methodologies for lung delivery, which is beyond the scope of this chapter.
THERAPEUTIC INDICATIONS FOR AEROSOL DELIVERY
Current Applications
Aerosolized drug delivery is currently used to deliver a limited range of therapeutic classes of compounds. These are mainly for asthma and chronic obstructive airway disease. These classes of compounds include short – and long – acting β – adrenoceptor agonist, corticosteroids, mast cell stabilizers, and muscarinic antagonists. Of recent note is the popularity of combination products. These have obvious advantages from a patient compliance perspective. In addition, certain combinations of drugs have shown synergistic therapeutics benefi ts when compared to the drugs given by separate inhalers. Long – acting β – adrenoceptor agonists and corticosteroids formulated as combination products are available as both MDIs and DPIs. Also recently introduced was a MDI formulation, the R enantiomer of albuterol, which is believed to be mainly responsible for bronchodilation in the racemic mixture.
Zanamivir is licensed in the United States as an inhaled antiviral agent for the treatment of infl uenza. Recombinant human deoxyribonuclease (rhDNAase) is available as a nebulizer product for the treatment of cystic fi brosis, in which it acts to liquefy viscous lung secretions. And recently, insulin was approved as an inhaled powder for glycemic control in type I and II diabetes.
Future Applications
Research and development are presently underway covering a vast array of novel applications. Clark (2004) provides an extensive list of products and their current state of development. A signifi cant future advance will be the development of inexpensive, noninvasive, stable, single – dose vaccine delivery via the lungs.
Efforts in this area are being led by the World Health Organization in the Measles Aerosol Project, and in a separate project, the Grand Challenges in Global Health initiative has funded a program to further develop an inhalation aerosol measles vaccine. Delivery of the measles vaccine via the lungs has been demonstrated to be both safe and effective. Now the challenge of each of these projects is to produce stable inhalation vaccine formulations to be delivered via inexpensive inhalers while maintaining both safety and effi cacy. The use of inhaled vaccinations in the event of a bioterrorism attack is also a potential application.
The use of the inhalation route for the delivery of gene therapy is also an area of signifi cant interest. Cationic liposomes and polymers together with adenoviral vectors containing the reporting genes have been aerosolized using nebulizers for the majority of clinical studies. However, there are a signifi cant number of challenges that must be overcome before pulmonary gene delivery is deemed completely successful, the most important being low gene transfer effi ciency at the cellular level. This problem is not unique to inhalation therapy. Inhalation of a recombinant adenovirus containing the cystic fi brosis transmembrane regulator (Ad2/CFTR) demonstrated the feasibility of this approach for the treatment of cystic fi brosis. However, the limited duration of transfection and low cellular uptake effi ciency still remain a barrier to full utilization of this route. There are a number of reviews that provide updates as to recent developments in this area.
Given the success of delivering insulin, other peptides and proteins are being considered for pulmonary applications. Leuprolide is a nonapeptide which has been investigated as both an MDI and DPI formulation for the treatment of prostrate cancer. Other hormones being investigated include calcitonin for the treatment of Paget disease and osteoporosis, parathyroid hormone to treat osteoporosis, growth hormone releasing factor for the treatment of pituitary dwarfism, and vasoactive intestinal peptide (VIP) for the treatment of pulmonary diseases.
Other potential inhalation applications include drugs for both local and systemic delivery. Inhaled tobramycin is being investigated for the treatment of Pseudomonas aeruginosa exacerbations in cystic fi brosis. Liposomal ciprofl oxacin is being developed as a fi rst – line defense against biowarfare agents (e.g., anthrax). Inhaled cyclosporine has been shown to improve survival rates and extend periods of chronic rejection – free survival in lung transplant patients. Apomorphine has been proposed as an inhalation formulation for the treatment of erectile dysfunction. Aerosol delivery of chemotherapeutic drugs has been advocated for the treatment of lung cancer. Morphine and fentanyl have been investigated for alternative routes of administering analgesics. Heparin and low – molecular – weight heparins have been aerosolized and advocated for the treatment of emphysema and thrombosis. Iloprost, a stable prostacyclin analog, has been aerosolized by nebulization for use in the treatment of pulmonary hypertension. This list of potential new treatments approached via the inhalation route is not exhaustive; among the other compounds under investigation are α1 – antitrypsin, sumatriptan, ergotamine, nicotine as replacement therapy, pentamidine, and ribavirin. Readers should be aware that a large number of these examples are proof – of – concept studies that may not get beyond in vitro experiments and animal studies.
AEROSOL DRUG DELIVERY DEVICES
As can be seen from the previous section, aerosol drug delivery continues to be an area of intensive research and development for the pharmaceutical industry. Not only are new applications for the pulmonary route being investigated, but also new delivery technologies are under development. The reformulation of MDIs with hydrofl uoroalkane (HFA) propellants together with the potential of using the inhalation route as a means of systemic administration has led to signifi cant technological advances in delivery devices. In parallel to MDI research DPIs have been developed from breath – actuated single – dose devices to both multiple – dose inhalers and active – dispersion DPIs. There is an extensive literature detailing the fundamental mechanisms of powder dispersion aimed at improving pulmonary deposition from powder inhalers. In addition, novel particle production technologies have been developed that provide alternatives to the traditional micronized powder for formulation in both MDIs and DPIs. Nebulizer technology has evolved from previously nonportable devices into high – effi ciency, hand – held nebulizers that offer alternatives to the MDI and DPI for certain treatment regimes. Finally, novel soft mist inhalers that generate aerosols by solution atomization have emerged on the inhaler landscape.
All this research has focused on improving aerosol deposition effi ciency and reproducibility within the lungs, together with targeting the peripheral lungs for systemic absorption. The efforts of the last decade culminated in two signifi cant events. First, the regulatory approval of Proventil HFA and QVAR, the fi rst suspension and solution HFA MDIs, respectively. Second, in 2006, the U.S. and European regulatory authorities approved Exubera, an insulin DPI for the systemic treatment of type I and II diabetes. Exubera offered a noninvasive alternative to subcutaneous injections of insulin.
Characteristics of Ideal Delivery Device
With these developments, innovation continues toward development of the ideal inhaler. A number of authors have compiled lists of desired characteristics for an aerosol inhaler. These can be grouped into patient – desired or industry – driven properties. From the patient ’ s perspective the overriding requirement is a device that is simple to operate. This is becoming increasing diffi cult to achieve as evidenced by the intensive patient education initiative that is being planned for the launch of the Exubera insulin inhaler. Poor compliance and adherence to prescribed therapy may be related to patients ’ failure to use the inhaler correctly. Inhalers should be portable and contain a large number of doses. The device should also give some indication to the patient when it is empty. The inhaler should be suitable for use by all of the population, especially children and the elderly. Ganderton (1999) cited that from a device perspective aerosol generation should be independent of the patient ’ s inhalation and should continue for a substantial portion of the inspiratory cycle. This would minimize the reliance on coordinating inhalation and actuation of the device. Breath – actuated devices have been developed to address this issue. In order to achieve lung deposition targeting, the particle size distribution of the aerosol should be capable of being altered depending upon the specifi c target region. For example, the central airways may be targeted with a 3 – 5 μ m aerosol for the treatment of acute bronchoconstriction, while a smaller aerosol (1 – 3 μ m) might be used for deep lung deposition and subsequent systemic absorption. In addition, the dose should be delivered reproducibly with minimal oropharyngeal deposition, perhaps as a low – velocity aerosol. There should be a minimal number of small parts in the inhaler, and it should be robust and reliable when placed “ in use. ” The manufacturer has the option of producing a disposable or refi llable unit; however, the inhaler should protect the formulation from environment and not affect its stability.
Dolovich et al. (2005) have provided an extensive evidence – based evaluation of aerosol drug devices. They concluded that when selecting an inhalation delivery system the following should be considered: device and drug combination availability, clinical setting, patient age, the ability of the patient to use the device correctly, device use with multiple medications, cost and reimbursement, drug administration time, convenience in outpatient and inpatient settings, and patient and physician
preference. Other reviews have compared the benefi ts and disadvantages of inhalers from clinical and patients ’ perspectives.
METERED DOSE INHALERS
Since their development, MDIs have been widely used for pulmonary aerosol drug delivery. Despite their recognized limitations, they remain the device of choice for many physicians around the globe. From a patient ’ s perspective, they are light, portable, and robust and contain multiple doses of medication. They are also relatively simple to operate (press and fi re); however, signifi cant numbers of patients experience diffi culties correctly using the MDI due to coordination problems.
To maximize lung drug deposition, actuation (pressing the MDI canister) by the patient must be coordinated with a slow, deep inhalation. Studies have reported that 51% of patients fail to operate the MDI correctly. This leads to low lung deposition, high oropharyngeal deposition, and ultimately perhaps therapeutic failure. From the pharmaceutical industry perspective, the components are relatively inexpensive; however, the formulation and manufacturing have become increasingly complex. There are numerous studies describing the multifaceted and interactive effects of propellant, excipient, metering valve, and actuator on the aerosol particle size characteristics of the MDI.
To date, the success of the MDI has relied in part on the potency and relative safety of the bronchodilators and corticosteroids commonly used for the treatment of respiratory disorders rather than its delivery effi ciency. The relatively low and often variable aerosol deposition effi ciency, only around 10 – 20% of the nominal dose being delivered to the lungs, is the challenge that is beginning to be addressed as the MDI looks to enter the next 50 years of aerosol drug delivery.
Metered Dose Inhaler and HFA Reformulation
The basic design and operation of the MDI has changed little over its lifetime. Aerosols are generated from a formulation of drug (0.1 – 1% w/w) either suspended or in solution in the liquefi ed propellant. The formulation is held under pressure in a canister.
Aerosol generation takes place when the canister is pressed against the actuation sump by the patient. Actuation causes the outlet valve to open and the liquefi ed propellant formulation is released through the actuator nozzle and subsequently through the mouthpiece to the patient. Metered volumes between 20 and 100 μ L are dispensed, and as the pressurized propellant is released, it forms small liquid droplets traveling at high velocity. These droplets evaporate to leave drug particles for inhalation. Purewal and Grant (1998) have assembled a defi nitive reference source for issues relating to the design, manufacturing, and performance of MDIs.
The currently marketed MDIs may look similar to the devices that were fi rst developed by Riker in 1950. However, due to the replacement of the ozone – depleting chlorofl uorocarbon (CFC) propellants with HFA propellants, virtually all of the components of the MDI have been altered. In 1987, the Montreal Protocol was drawn up, leading to the eventual phase – out of CFC propellants. MDIs contain
FIGURE 1 Schematic of MDI.
ing CFC propellants were granted essential – use exemptions until viable alternatives became available. Therefore, with this impending withdrawal, a consortium of pharmaceutical companies (IPACT – I and IPACT – II) worked to identify and toxicologically test alternative propellants for MDIs. HFA 134a and HFA 227 were identifi ed as viable alternatives and the task of reformulation began. At fi rst look, it appeared that the most expeditious route to replacing a CFC product would be to produce a suspension HFA MDI with exactly the same in vitro characteristics as the CFC MDI. This would prove to be a time – consuming route. While some manufacturers focused on producing HFA products with identical characteristics to the current CFC versions to accelerate the pathway through clinical testing to market.
Others undertook extensive research and development in the area of HFA formulation options, and this has led to the possibility of utilizing the MDI for both local and systemic administration. During this reformulation effort, the industry has taken the opportunity to address some of the other shortcomings of the MDI.
Among these issues were poor peripheral lung delivery, variable dose delivery, and limitations as to the dose capable of being delivered to the lung (typically about 1 mg).
The replacement of CFC MDIs with inhalers formulated with the HFA propellants is now well underway in Europe. Although progress in the United States has been slower, with the introduction of suitable alternatives for albuterol inhalers, the FDA has ordered that CFC albuterol MDIs be withdrawn from the market by the end of 2008. Examples of reformulated products available in the United States include Ventolin HFA, which is a suspension albuterol sulfate formulation using HFA 134a alone. ProAir is an alternative albuterol sulfate product manufactured by Ivax which contains ethanol and HFA 134a. Xopenex HFA has recently been approved for marketing in the United States. This product contains levalbuterol tartrate (R – albuterol enantiomer) together with HFA 134a, dehydrated alcohol, and oleic acid as a suspension formulation. Table 1 summarizes the HFA products currently available in the United States and their excipients. The following section will focus on the current options for formulation of drugs in HFA propellant systems and the challenges that are encountered as products are reformulated as HFA formulations.
Propellants
The CFC propellants primarily used in MDI formulations were CFC 11, 12, and 114.
Blends of these propellants were held liquefi ed under pressures of 50 – 80 psig within the canister. Flocculated drug suspensions in CFC propellants were formulated using a surfactant (e.g., oleic acid and lecithin). In a suspension formulation, the aerosol particle size is dependent upon the initial micronized drug particle size (typically between 2 and 5 μ m) and the evaporation of the propellant droplets. It has long been recognized that changes in CFC propellant vapor pressure result in changes in droplet size and velocity of the aerosol. Newman et al. (1982) showed that increasing the vapor pressure of the propellant blend in the MDI signifi cantly increased whole – lung deposition and reduced oropharyngeal deposition.
The fi rst challenge encountered during the reformulation with HFA propellants was the altered physicochemical properties of HFA 134a and HFA 227 compared to the CFC propellants.
The potential to formulate as a solution offered a number of advantages together with signifi cant problems. Perhaps most importantly, changing from a suspension to a solution formulation altered the mechanism of aerosol particle formation.
In the case of solution formulations, drug is dissolved in the liquefi ed propellant and a suitable cosolvent (if necessary) and particle formation takes place during evaporation of the propellant. This leads to much smaller particles being formed when propellant evaporation is complete. Stein and Myrdal (2006) recently described the MDI aerosol generation for solution formulations as a two – step process.
Droplet formation takes place as millions of atomized droplets are produced after the formulation exits the metering valve through the actuator sump. Initial droplet size is dependent upon the vapor pressure and surface tension of the formulation, valve size, and actuator orifi ce diameter. The second step is an evaporative or “ aerosol maturation ” phase, as the propellant and cosolvents (e.g., ethanol) rapidly evaporate leaving inhalable drug particles. The fi nal size of these particles is dependent upon the initial droplet size, the vapor pressure of the formulation mixture, and the proportion of nonvolatiles in the formulation.
Leach et al. (1998) compared the pulmonary deposition of a suspension CFC formulation of beclomethasone diproprionate with a solution HFA formulation. The marketed CFC product had a mass median aerodynamic diameter (MMAD) of 3.5 μ m compared to 1.1 μ m for the solution formulation, refl ecting the altered aerosol formation mechanism. The gamma scintigraphy profi le for the solution formulation showed the drug and label to be diffusely deposited throughoutthe airways with approximately 55 – 60% deposited in the lungs. In contrast, the CFC product was deposited mainly in the mouth and throat (90 – 94%), with only 4 – 6% being deposited in the airways. In many ways, this study summarized the defi ciencies of the suspension formulation CFC MDI and offered the alternative of improved delivery effi ciencies with the solution HFA formulation. A signifi cant conclusion from this and other studies supported the hypothesis that improved pulmonary deposition and reduced oropharyngeal losses of aerosols would allow reduction in the dose required by the patient to achieve the same therapeutic effect.
The altered solubility profi le of the HFA propellant, while providing attractive characteristics for solution formulations, also provide signifi cant challenges with respect to their interactions with the basic MDI components. Leachables are compounds that can be transferred from MDI component parts to the formulation during the shelf life of the product. Berry et al. (2003) postulated that MDI orientation could affect the amount of leachables that entered a formulation and affect the particle size distribution of aerosol. Extensive efforts are now required for extractable and leachable testing of the component materials prior to formulation of an MDI. Another by – product of replacing the CFC propellants was to tighten the impurity specifi cations required for the new propellants. A proposed U.S. Pharmacopeia (USP) monograph for HFA 134a has now been published detailing the impurity profile.
Manufacturing processes for MDIs have also required adapting for the use with the new propellant system. There are two main manufacturing processes used for MDIs: cold fi lling and pressure fi lling. Cold fi lling requires cooling the propellants to below − 50 ° F and fi lling at that temperature prior to crimping the valve onto the canister. Pressure – fi lling techniques for MDIs are most commonly employed. These can be accomplished in either a one – or two – step process. In the single – step process, the formulation is placed in a pressurized mixing vessel. The empty canister is purged with propellant to remove the air. The valve is then crimped onto the canister and the formulation is metered through the valve. The absence of a HFA propellant that was liquid at room temperature was a major difference compared to the process employed for CFC manufacturing. In the two – step process, the formulation (excluding the propellant) are mixed together to form a concentrate.
Previously, liquefi ed CFC 11 was used in this step of the process. However, there is no suitable HFA propellant that is liquid at room temperature. Therefore, cosolvents such as ethanol and glyercol are employed during this step to form the product concentrate. The concentrate is metered into the empty canister. The valve is then crimped onto the canister and the propellant is fi lled through the valve.
Wilkinson (1998) provides an extensive history and review of the manufacturing procedures for MDIs .
Excipients
A number of excipients have been included in MDI formulations; however, the nature of the excipients has changed with the introduction of the HFA propellant aerosols. Oleic acid and sorbitan trioleate (SPAN 85) and lecithins were used in CFC suspension MDIs as suspending agents and valve lubricants. Typical concentrations ranged from 0.1 to 2.0% w/w. Ethanol is now being used in HFA formulations as a cosolvent for suspension and solution formulations. The addition of ethanol to the formulation has a number of effects. Increasing the ethanol concentration has been shown to increase the initial droplet size. In addition, ethanol can increase the hydrophilicity of the formulation and increase moisture uptake. Glycerol and polyethyleneglycol have also been added as cosolvents but also have the effect of increasing the residual droplet particle size due to their lower volatility. In general, a relationship can be observed between the fraction of nonvolatile components (drug and nonvolatile excipients) in a solution HFA formulation and the fi nal particle size of the aerosol. The MMAD was observed to be linearly proportional to the cube root of the nonvolatile concentration. Oligolactic acids (OLAs) have been investigated for their use in a variety of functions in HFA formulations. OLAs with repeating units of 6 – 15 units have been proposed as suspending agents. They are readily soluble in both HFA 134a and 227. These molecules have also been shown to act as ion pair solubilizers for certain drugs (e.g., albuterol). The addition of ethanol to these OLA formulations synergizes the solubilizing effect.
A word of caution is required when considering introduction of novel excipients into any inhalation drug product formulation. Due to the unique toxicological challenges associated with administration and clearance from the lung, the qualifi cation of novel excipients for inhalation has proven to be an expensive and time – consuming challenge. This has led to a limited number of compounds with an extensive “ in – use ” profi le being commonly employed.
Valves
Metering valves are required to accurately meter and dispense the formulation upon MDI actuation. In addition, they perform an important contact closure role preventing moisture ingress and minimizing propellant evaporation. Figure 2 and Table 3 show the basic components of the metering valve. Currently, the most common valve type is the retention valve, consisting of a plastic metering chamber and two rubber gaskets. The remaining valve components are manufactured from plastic, metal, and elastomeric materials.
Material component evaluation and selection are critical steps in the development of a MDI formulation. The materials must be chemically resistant and compatible with all components of the formulation. Gaskets must have appropriate mechanical properties and work effectively as a seal, preventing leakage of the formulation and moisture ingress. While the basic components themselves have
FIGURE 2 Schematic of components of metering valve. (Courtesy of Valois Pharm.)
remained unchanged during the introduction of the HFA propellants, the materials used to manufacture the components have required signifi cant adaptation. Nitrile was the most commonly employed elastomer in CFC MDIs; it has good mechanical and elastic properties. However, it has been shown to swell when in contact with HFA propellants and ethanol. Newer elastomers such as ethylene propylene diene monomer (EPDM), chloroprene, and bromobutyl are now used in HFA MDIs.
The ideal universal elastomer has yet to be developed and the newer materials must be assessed on a case – by – case basis for formulation compatibility and the desired moisture ingress characteristics. Among the many issues to be considered when screening materials are formulation – material compatibility, extractable profi les, and mechanical resistance. Manufacturers such as Valois, Solvay, and Bespak have extensive knowledge of drug/excipient/material component compatibility and should be used as the fi rst point of reference when considering a MDI formulation project.
Another concern to formulators is the ingress of moisture into HFA – formulated MDIs [143] . HFA propellants have a higher moisture affi nity compared to the CFC propellants, especially HFA 134a. In addition, the inclusion of ethanol in some formulations increases its hydrophilicity. Moisture entering the canister can have several effects; it may alter the physical or chemical stability of the formulation and aerosolization performance of HFA MDIs. Due to its lower volatility compared to the other components, water may affect aerosol generation and alter aerosol particle size. The increased water content may increase the solubility of suspended polar drug particles or decrease the solubility of hydrophobic compounds.
Corrosion in aluminum canisters may also increase over the shelf life of the product. Williams and Hu (2000) reported that HFA 134a had a greater tendency to take up moisture during storage than did HFA 227. The issue of moisture ingress during storage has led to certain HFA MDIs being stored in moisture – protecting pouches prior to initial use (e.g., Ventolin HFA). An alternative approach to minimize the effects of moisture ingress has been taken by SkyePharma, which has incorporated subtherapeutic doses of cromolyn sodium into its HFA MDI formulations.
Cromolyn sodium is used as a hygroscopic excipient to scavenge any moisture that penetrates into the formulation. Cromolyn sodium has been used widely by inhalation over the past 30 years and has an excellent safety profi le via the inhalation route. Burel et al. (2004) reported that for a HFA 134a MDI formulation the inclusion of a polyamide (nylon 66) molded ring around the valve body reduced both the initial water content and the fi nal water content (6 months) when stored under stress conditions [40 ° C and 75% relative humidity (RH)]. A combination of a thermoplastic elastomer sealing gasket in the MDI valve and a polyamide ring produced the lowest water ingress under these stress conditions. For formulations that might be susceptible to water – induced stability issues, HFA 227 may be considered a more suitable propellant than HFA 134a. Given the possibility of moisture ingress, there is also the issue of propellant leakage. Leak testing is among the array of in – process quality assurance tests that are required. These include assay of the suspension or solution, moisture level, consistency of fi lling of both the concentrate and the propellant, valve crimp measurements, quality of sealing, in – line leak testing under stress conditions, and performance of the valve.
Another signifi cant issue encountered during use of MDIs was related to loss of prime and dose reproducibility. Loss of prime relates to the fact that in conventional capillary retention metering valves the dose is fi lled into the valve immediately following the last actuation. Capillary retention valves require priming with one or two sprays prior to their fi rst use. In addition, if there is a signifi cant interval between the actuations and the inhaler is stored upside down or on its side or shaken, then the metering valve may actually partially empty, resulting in a low and variable dose being delivered to the patient. A review of patient information leafl ets indicated varying instructions on priming MDIs. This ranged from Atrovent CFC and Combivent CFC requiring priming with 3 sprays “ after 24 hours of nonuse. ”
Ventolin HFA and Proventil HFA both required priming with 4 sprays after “ 2 weeks of nonuse. ” Flovent CFC required priming with 4 sprays after “ 4 weeks of nonuse. ” Clearly, such instructions add to the complexity for patients using MDIs and also contribute to drug waste issues. Loss of prime is also a signifi cant issue for breath – actuated MDIs, where the opportunity to prime the inhaler is not readily possible. A number of new valve designs have been developed to address this issue.
The fast – fi ll, fast – empty valves offer a solution to the priming and loss of prime issues. In these valves [(e.g., 3M Shuttle valve (3M), 3M Face Seal valve (3M), ACT (Valois), and Easifi ll valve (Bespak)], the metering valve is only isolated from the formulation canister reservoir immediately prior to dose actuation. Therefore, the metering chamber can be emptied and refi lled with a fresh dose from the reservoir simply by shaking the canister prior to use.
Actuators
Nonvolatile component concentration has previously been described as one of the primary determinants of the initial droplet size for HFA solution formulations.
Perhaps, equally important is the MDI actuator. The actuator consists of the sump block into which the metered dose is immediately delivered during MDI actuation. As expansion and vaporization of the propellant take place, the aerosol exits the sump via the actuator nozzle and then is inhaled through the actuator mouthpiece. From a practical perspective, in general, reducing the size of the orifi ce diameter for HFA solution formulations produced a relatively slower spray emitted with less force compared to marketed CFC products. The nozzle orifi ce diameter has been considered to be the most important, although not the only, actuator variable determining the particle size distribution of HFA solution formulations. Recently, Smyth et al. (2006) described three critical components of the actuator that could affect the aerosol performance of a solution HFA formulation.
n addition to the orifi ce diameter, sump depth (and hence the expansion chamber volume) together with orifi ce length was observed to have signifi cant effects on the aerosol particle size distribution and should be considered for optimization with an HFA formulation. It has also been recognized that the electrostatic charge of all components of the MDI and its formulation may affect the aerosolization properties of the aerosol spray.
Canisters
Aluminum canisters are widely used in commercial MDI products mainly due to their inert characteristics. Other materials, including stainless steel and glass, can be employed depending upon the particular formulation characteristics. These canisters were usually uncoated. Changes to the canister may be required when the formulation interacts with the interior surface of the canister altering the chemical stability of the formulation. The presence of ethanol in HFA formulations has also increased the risk of metal corrosion. Drug migration or absorption to the metal components of the canister and also the metal valve components has also been reported. The loss of drug to the walls of the canister will result in variability in the delivered dose from the MDI during the shelf life of the inhaler. The use of canister coating and anodized canisters has been advocated to mitigate this problem.
Breath Actuation
In order to overcome the problems associated with many patients ’ inability to coordinate actuating the MDI and inhaling, breath – actuated MDIs were developed.
These devices allow the MDI to be automatically actuated only when the patient commences inhaling through the mouthpiece. Of critical importance here is ensuring that the patient has suffi cient inspiratory fl ow rate to trigger actuation. While these devices offer little improvement for patients with a good inhaler technique, it has been shown that patients with poor coordination did have signifi cantly greater lung drug deposition when inhaling using a breath – actuated MDI. The 3M Autohaler was the fi rst device marketed using this technology. In Europe, the Easibreathe and Autohaler breath – actuated MDIs are used to deliver β agonists and corticosteroids for the treatment of obstructive airway. Recently, in the United States, the MD Turbo has been launched, a device that allows patients to take their regular MDI canister and actuator and insert it into the MD Turbo. The MD Turbo acts as a generic breath actuator for a number of marketed MDIs and also incorporates a dose counter.
Spacers
Spacer devices have been developed as another alternative to overcome the problems associated with patients coordinating the beginning of their inspiratory effort with actuation of the MDI. This problem is extenuated by the fact that the MDI emits a high – velocity, short – duration aerosol cloud. On actuation, the propellant spray is delivered into the spacer that often incorporates a one – way inhalation valve. The patient is now able to inhale the aerosol cloud. The large – volume spacers have an additional effect in that they allow evaporation of large propellant droplets prior to inhalation. These high – velocity droplets would previously have had a high probability of impacting in the patient ’ s throat. Figure 3 shows the large number of spacer chambers that are available. Spacers are advocated for use by children and elderly patients and people who experience diffi culty coordinating actuation of the MDI. The use of spacers for the delivery of corticosteroids also minimizes oral deposition of the inhaled dose and therefore reduces the incidence of steroid – related side effects. Both in vitro and clinical studies have shown the effectiveness of spacers with CFC MDIs. It has been shown that electrostatic charge can have a signifi cant effect on the performance of a spacer chamber and where possible the charge should be minimized to maximize drug delivery. Finally, it should be noted that the use of any particular spacer – MDI combination should be evaluated at least in vitro to confi rm the benefi cial effect, especially when employed with solution – based HFA MDI formulations.
FIGURE 3 Example spacer chambers available for use with MDIs. (Reproduced from ref. 154 with permission of Pharmacotherapy .)
Dose Counters
A guidance document from the Food and Drug Administration (FDA) recommends the addition of a dose counter to the MDI. This would overcome a long – standing problem with the MDI, the inability of a patient to accurately know the number of doses remaining in the canister. Dose counters have been incorporated successfully into multiple – dose DPIs. In general, the counter should give a clear indication of when approaching end of life and the actual end of life. It should be either numeric or color coded. If numeric, it should count downward and should be 100% reliable and avoid undercounting.
DRY POWDER INHALERS
Dry power inhalers have been in use for over 40 years. They were developed as an environmentally friendly alternative to the MDI. The early DPIs were simple in design, portable, but again, a relatively ineffi cient means of delivering drugs to the lungs for local action. The Spinhaler, the fi rst DPI, has been prescribed in Europe since the late 1960s. In general, the acceptance and use of DPIs is much greater in Europe than in the United States. However, with the reformulation efforts for MDIs, there are an increasing number of DPIs becoming available in the United States (Figure 4).
Research and development for dry powder inhalers have two main focuses: the optimization of the powder formulation for use in these inhalers and investigations of novel DPI device designs and technology. An enormous literature now exists in each of these areas; for more extensive reviews readers should consult refs. 33, 167, or 168 .
FIGURE 4 Example DPIs available in United States.
Size Reduction and Particle Formation Technologies
Dry powder inhaler formulations consist usually of either a drug – only formulation or an ordered mixture of drug and excipient, most commonly lactose monohydrate.
In both cases, the fi rst challenge is the production of drug particles with suitable size characteristics for inhalation (i.e., 1 – 5 μ m). Traditionally, micronization or jet – milling methods have been employed as the method of choice for conventional small molecules. This method is identical to that employed for the production of fi ne particles for suspension MDIs. Using this method it is possible to produce primary particles between 1 – 5 μ m. However, as a consequence of the particle size reduction there are a number of undesirable effects with respect to the powder properties. Micronized powders possess high intramolecular forces and are cohesive.
They readily form aggregates that are diffi cult to disperse to the primary particles.
Dispersion to its primary particle is essential for successful pulmonary deposition. In addition, they often possess high inherent electrostatic charges which cause particle adhesion to the components of the dry powder inhaler. The high – energy micronization process also causes disruption of the crystal lattice and results in the formation of amorphous regions which may affect the long – term stability of the formulation. Finally, it is not possible to control the drug particle morphology.
Despite all of these problems, micronization remains the most common technique employed for respirable particle formation. Modifi cations to conventional micronization techniques have been investigated as alternative methods of particle size reduction.
A number of novel particle formation technologies now exist that are able to produce respirable drug particles for formulation in both DPIs and MDIs. Depending on the method of preparation, these particles offer unique and potentially advantageous physical and aerodynamic properties compared to conventional crystallization and micronization techniques. Some investigators have advocated that major improvements in aerosol particle performance may be achieved by lowering particle density and increasing particle size, as large, porous particles display less tendency to agglomerate than (conventional) small and nonporous particles. Also, large, porous particles inhaled into the lungs can potentially release therapeutic substances for long periods of time by escaping phagocytic clearance from the lung periphery, thus enabling therapeutic action for periods ranging from hours to many days.
Many of these techniques involve particle formation from solution formulations that contaiovel excipients. Spray drying is the most advanced of these technologies and has been used to produce the powder formulation in the Exubera inhaler. Various modifi cations of this basic technique, including co – spray drying with novel excipients, have been employed.
AIR particles are low – density lipid – based particles that are produced by spray drying lipid – albumin – drug solutions. These particles are characterized by their porous surface characteristics and large geometric diameter while having a low aerodynamic diameter. This technology has been used to produce porous particle powder formulations of L – dopa that have been investigated for the treatment of Parkinson ’ s disease.
Pulmospheres are produced using a proprietary spray drying technique, with phosphatidylcholine as an excipient to produce hollow and porous particles with low interparticulate forces. These particles have been formulated as suspended particles in HFA MDIs. In comparison with conventional suspension MDIs, the Pulmosphere MDI exhibited signifi cantly higher fi ne particle fractions. This technology has been used to produce cromolyn sodium, albuterol sulfate and formoterol fumarate microspheres. Pulmospheres powder formulations containing tobramycin and budesonide have also been tested clinically.
Technosphere technology has been developed as an alternative porous particle for pulmonary delivery. These porous microspheres are formed by precipitating a drug – diketopiperazine derivative from an acidic solution. Para – thryroid hormone (PTH) Technospheres have been investigated for the treatment of osteoporosis following aerosol delivery.
The use of supercritical fl uid processing technology has also been widely used for its application in controlled microparticle formation. Conventional small molecules and proteins for inhalation have been generated and formulated as powders for inhalation.
The application of pulmonary delivery of nanoparticles ( < 1 um) for pharmaceuticals remains to be developed.
Drug – Lactose Formulations
The most common means of overcoming cohesion problems is by incorporation of a carrier excipient. Lactose monohydrate is used most often; it is inert, cheap, widely available, and a GRAS (generally regarded as safe) non – toxic excipient. A signifi – cant area of research has been undertaken to optimize the critical parameters involved in the formulation of drug – lactose blends. Micronized drug is typically blended with lactose (50 – 100 μ m) to produce an ordered mix. The blend ratio is fi xed depending upon the dose of drug to be delivered and the mass of powder blend in each dosage unit (typically between 5 and 25 mg). The aerosolization properties of the blend are related to the adhesive forces between the drug and lactose together with the cohesive forces between the drug particles. Reproducible dispersion of the blend either by the dry powder inhaler (active DPI) or by the patients ’ inspiratory effort (passive DPI) is required. This allows the detached micronized drug to be inhaled and deposited in the respiratory tract while the larger lactose particles are deposited by inertial impaction in the oropharnyx.
Formulators have become increasingly aware of the criticality of the drug and lactose powder surface characteristics and their relationship to the aerosolization performance in a DPI. A number of investigators have shown in vitro the importance of controlling the size of the lactose and the amount of “ fi nes ” (lactose particles less than 5 μ m in size) in the drug – lactose blend. Inherent fi nes are present in all lactose powders, and the fi nes are usually adhered to the surface of the larger lactose particle. These fi nes are believed to occupy “ active ” or high – energy sites on the lactose particle surface. Occupation of these sites by the lactose fi nes prevents the micronized drug from adhering to these positions. This allows the drug to adhere to less active sites and become detached easier from the lactose surface during inhalation. Obviously any signifi cant change in the quantity of fi nes present in the lactose may alter the distribution of the micronized drug on the lactose particle and therefore the aerosolization characteristics of the powder blend. Batch – to – batch control of the fi nes content of inhalation lactose has been recognized as critical to ensuring reproducible in vitro emitted and fi ne particle doses. Jones and Price (2006) have recently surveyed the literature in this area and provided a comprehensive review. Modifi cation of the surface characteristics of the lactose particle has been used as an alternative approach to control the adherence of drug particles to the lactose surface. Alternative sugar carriers have also been investigated; these appear to possess many of the same performance – limiting characteristics as lactose. Finally, tertiary additives have also been used to improve the aerosolization properties of DPI formulations. The majority of the studies described above relate to in vitro testing of DPI formulation performance, and little is known about the clinical signifi cance of these studies.
Moisture ingress into a powder formulation is a particular concern as it may signifi cantly decrease the aerosolization performance of the formulation.
Increased adhesion of particles is often seen following exposure to high – RH environments. Moisture ingress has also been shown to affect drug stability. The pharmaceutical industry has used a number of approaches to protect powder formulations from the ingress of moisture during storage and for their “ in – use ” life. The Turbuhaler incorporates a desiccant in the base of the inhaler to keep the power reservoir free from moisture. Unit – dose blisters used in the Diskus are sealed in a foil strip pack to protect each individual dose prior to inhalation. It is also essential that the patient not exhale into the DPI immediately prior to inhaling the dose.
Electrostatic charge can also infl uence the performance of DPI formulations. A number of studies have investigated the interactions of drug and lactose particle charge with respect to aerosolization properties and drug retention by the plastic components within the inhaler.
Dry Powder Inhaler Design
Inhalation Flow Rate The main function of a DPI is to facilitate dispersion and delivery of inhalable drug particles. An extensive patent and scientifi c literature exists describing the ever – increasing number of DPI device designs. Powder dispersion in the early passive DPIs was provided in part by the inspiratory effort of the patient. This removed the necessity to coordinate patient inhalation with actuation and delivery of the dose (in contrast to MDIs). These passive DPIs were “ breath actuated, ” with the patients ’ inspiratory effort dispersing, aerosolizing, and delivering the powder during the inhalation cycle. The airfl ow rate through the inhaler was determined by the inherent device resistance and the inspiratory force exerted by the patient. Devices such as the Spinhaler, Rotahaler, and Diskhaler are low – resistance devices requiring relatively high inspiratory fl ow rates to disperse the powder formulations by turbulent deaggregation. These simple devices have low aerosolization effi ciencies with only 5 – 20% of the dose being delivered to the lung. The inhalation fl ow rate dependence of passive DPIs has been cited as a potential problem in their use, especially given the large intersubject fl ow rate variability within the patient population (especially for the young and older patients).
In vitro testing revealed that for certain DPIs there was large variability in both the delivered dose and the aerodynamic particle size distribution as a function of the inhalation fl ow rate. Similar clinical studies also revealed a fl ow rate dependence for certain DPIs while others were observed to perform with a degree
FIGURE 5 Exubera Inhaler. (Reprinted from ref. 175 . Courtesy of Mary Ann Liebert, Inc.)
of fl ow rate independence. When choosing a DPI, the effect of inhalation fl ow rate should be assessed on a case – by – case basis for each individual DPI, and readers should be aware of contradictory studies, especially when comparing in vitro and clinical performance. The Turbuhaler is one such example, where some in vitro studies show high variability; however, this is not refl ected in clinical studies.
From these and many other studies it can be concluded that a desirable characteristic for any DPI is that its dose delivery performance is independent of inhalation fl ow rate. A second generation of DPIs have been developed that incorporate a combination of improved powder formulations, more effective turbulent dispersion within the inhaler, and in some cases an active dispersion mechanism. The Exubera inhaler releases a bolus of compressed air through the formulation and actively generates an aerosol cloud from the powder (Figure 5 ). The cloud is held within a reservoir chamber from which the patient then inhales the insulin dose. Active dispersion improves device aerosolization effi ciency, with greater than 50% of the dose being deposited in the lungs, while minimizing the reliance on the patients ’ inspiratory effort.
Single – and Multiple – Dose DPI s Inhalation powder dose metering is one of the problems encountered by DPI formulators. The powder dose can range from 250 μ g in the drug – only Pulmicort Turbuhaler formulation to 25 mg in the lactose – blended Spinhaler formulations. In each case, accurate and reproducible metering of the powder is required for regulatory approval and therapeutic effi cacy. This proved to be a technological challenge that was solved in a number of ways. Single – unit – dose inhalers were the fi rst generation of DPIs, the unit dose being metered in the factory and subsequently loaded into the inhaler by the patient immediately prior to each dosing. Because metering takes place prior to batch release by the manufacturer, this allows for quality control and release testing, ensuring that dosage units were within acceptable criteria. Procedures such as capsule fi lling were common for early devices such as the Spinhaler and Rotahaler. This approach is still used by some of the newer devices being developed (e.g., Aerohaler and Cyclohaler).
While popular with the pharmaceutical industry, the single – unit – dose device required signifi cant patient handling to load and empty the inhaler for each inhalation (unlike the MDI, which often contained up to 200 doses available for inhalation on demand).
Two approaches were taken toward the design of multiple – dose DPIs; the multiple – unit – dose DPI (e.g., Diskhaler and Diskus) and the powder reservoir multidose DPI (e.g., Turbuhaler). For the multiple – unit – dose DPI, manufacturers sought to address the requirement for multiple doses while retaining the control of factory premetering. Perhaps the most successful DPI in this respect is the Diskus, in which the dose is premetered into a coiled foil covered strip containing individually sealed blister reservoirs. Each blister is opened immediately prior to inhalation and up to 60 doses can be help in each foil strip. For the powder reservoir multidose DPI, volumetric dose metering of the powder takes place within the DPI immediately prior to inhalation in a manner analogous to MDIs. Among the devices that use this approach are the Turbuhaler, Clickhaler, Pulvinal, and Easyhaler. The Turbuhaler is used with a drug – only formulation (although lactose blends have also been used) that employs a proprietary powder agglomeration process to produce loosely bound aggregates that are easily dispersed by the patient ’ s inhalation and by the turbulent fl ow path encountered in the DPI.
Besides the Diskus and Turbuhaler, there are four other devices currently available in the United States, the Asmanex Twisthaler, the Foradil Aerolizer, the Relenza Diskhaler, and the Spiriva Handihaler. Other devices in development include the Novolizer, a multidose, refi llable, breath – actuated DPI that delivers up to 200 metered doses of drug from a single cartridge. The Ultrahaler offers yet another alternative DPI. The Taifun inhaler, the JAGO inhaler, and the Airmax are other multidose DPIs.
Exubera
Systemic delivery of drugs via the lungs offers a noninvasive route of administration.
Perhaps the most important and widely investigated molecule considered for this route has been insulin. Following over a decade of development, in January 2006, Pfi zer and its partner Nektar received marketing approval for Exubera, their insulin DPI. This offered diabetics a noninvasive route of insulin administration rather than repeated subcutaneous injections. Exubera has been indicated for the treatment of adult patients with diabetes mellitus for the control of hyperglycemia. It has an onset of action similar to rapid – acting insulin analogs and has a duration of glucose – lowering activity comparable to subcutaneously administered regular human insulin. Patton et al. (2004) provided an extensive review of the clinical pharmacokinetics and pharmacodynamics of inhaled insulin. In patients with type I diabetes, Exubera should be used in regimens that include a longer acting insulin. In patients with type II diabetes, Exubera can be used as monotherapy or in combination with oral agents or longer acting insulins. Studies revealed that the same level of blood sugar control was achieved following inhalation compared to subcutaneous injection, although different nominal doses were required due to lung bioavailability issues. The therapeutic effi cacy and safety of inhaled insulin appears to have been proven, although there are a signifi cant number of issues with its administration via this route. It has beeoted that asthmatics absorb less insulin from the lungs thaonasthmatics. In addition, smokers absorb more insulin thaonsmokers. Small and reversible changes in pulmonary lung function have been observed in some studies with
inhaled insulin. Each of these issues has led to the development of specifi c prescribing guidelines and an intensive physician/patient education program for the inhaled insulin product. The Exubera insulin formulation is a spray – dried, amorphous insulin powder containing 60% insulin in a buffered, sugar – based matrix.
Other pharmaceutical companies are also continuing to develop their own inhalation insulin products. Aradigm and NovoNordisk are using a liquid insulin formulation in combination with the AERx IDMS inhaler. Alkermes and Lily are developing an insulin product derived from their research on geometrically large, low – density particles that are formed by a spray drying process incorporating a natural phospholipid. MannKind is using its Technosphere technology to produce low – density porous insulin particles. This formulation is delivered using the MedTone inhaler. Other companies working in this area include Kos Pharmaceuticals, Mircodose Technologies, Coremed, and Biosante.
NEBULIZERS
Nebulization of liquid formulations has long been established as an effective, if not effi cient, means of pulmonary drug delivery. The basic principle of nebulizer aerosol generation has remained unchanged; however, a number of technological advances have been made which have improved effi ciency and reduced variability. Aerosols that were previously delivered in a continuous inhalation mode over 5 – 15 min are now delivered only during the inspiratory cycle, thus reducing drug waste. In general, nebulizers convert a liquid into a fi ne droplet mist, either by means of a compressed gas (jet nebulizer) or by high – frequency sound (ultrasonic nebulizer). Ultrasonic nebulizers use a piezoelectric source within the formulation reservoir to induce waves at the surface of the nebulizer formulation. Interference of these waves induces the formation of droplets which are then carried in a fl owing air stream that is passed over the formulation. These devices are not suitable for the nebulization of suspension formulations. Rau (2002) also observed that ultrasonic nebulizers can increase the solution reservoir temperature and may cause drug degradation. In the case of the jet nebulizer, an aerosol is produced by forcing compressed air through a narrow orifi ce which is positioned at the end of a capillary tube. The negative pressure created by the expanding jet causes formulation to be drawn up to the capillary tube from the reservoir in which it is immersed. As the liquid emerges from the tip of the capillary, it is drawn into the air stream and broken up into droplets by the jet to produce an aerosol. Baffl e structures within the nebulizer fi lter the large droplets from the aerosol by impaction and the deposited drug solution is recycled back into the drug reservoir. Only the small aerosol droplets evade impaction on the baffl es and are delivered to the patient for inhalation. Jet nebulizers can be categorized by function, for example, the DeVilbiss 646 is a conventional jet nebulizer with continuous drug output resulting in signifi – cant waste during exhalation. The Pari LC Plus system incorporated a valve system and operates as an active venturi jet nebulizer; although drug output is continuous, there is an increased output during inhalation. The patient ’ s inspiratory effort increases the nebulizer airfl ow, thus increasing drug output for these breath – enhanced nebulizers. Finally, dosimetric jet nebulizers such as the Ventstream use a one – way valve system to emit aerosol only during inspiration and are also breath – enhanced nebulizers [256, 257] . It is this last type of nebulizer that offers the most signifi cant advances in technology.
Jet nebulizers are commonly used in nonambulatory settings such as hospitals or the patient s home. In vitro studies comparing the performance of commercial nebulizers have concluded that there were large differences in drug delivery betweeebulizers of different classes and even betweeebulizers of apparently the same class. The aerosolization performance of different nebulizers has been found to be dependent upon a number of factors, including the drug being aerosolized, the formulation fi ll volume, the compressed airfl ow rate, and breathing pattern. These parameters ultimately control the aerosol droplet size and rate of drug output. However, probably the size of the conventional nebulizer, the duration of the treatment cycle (5 – 15 min), and the cost of the nebulizers are the main reasons that they are usually reserved for nonambulatory settings and remain less popular than the MDI and DPI.
Solutions or suspensions are available as nebulizer formulations. Due to the relative simplicity in formulating a liquid nebulizer formulation and because of the relatively large range of doses available for delivery, the nebulization method is often chosen as the aerosol method for proof – of – concept investigational studies.
Nebulizer technology continues to be developed to miniaturize and lower the cost of the devices while maintaining the quality of the aerosols generated. The Halolite incorporates adaptive aerosol delivery which monitors patients inspiratory cycle and delivers drug to patients during the fi rst 50% of their inspiratory cycle. The Pari eFlow is a hand – held device that uses a vibrating membrane nebulizer to generate a respirable aerosol. Aerogen (now part of Nektar) has a range of nebulizer – based technologies, including the Aeroneb and Aerodose devices. Aerosols are generated as a liquid formulation passes through vibrating apertures.
EMERGING TECHNOLOGIES
Soft Mist Aerosols
In recent years research has focused on a new method of pharmaceutical aerosol generation that involves passing a solution formulation through a nozzle or series of nozzles to generate a “ soft mist ” aerosol as a bolus dose. Aerosol generationis achieved by mechanical, thermomechanical or electromechanical processes depending upon the particular technology employed. It is worth noting that these devices are bolus dose delivery inhalers, rather than the new continuous – generatioebulizers which generate aerosols by vibrating porous membranes at ultrasonic frequencies. Such devices include the eFlow and Aerodose, which were described earlier.
While the precise mechanism of soft mist aerosol generation may differ between inhalers, a number of common characteristics can be observed. They are propellant free and produce slow – moving aerosols over an extended duration with high in vitro fi ne – particle fractions compared to MDIs and DPIs. The aerosols are often generated from simple solution formulations containing pharmaceutically acceptable excipients. Water and ethanol are the most commonly employed vehicles for soft mist aerosols. Perhaps the most simple and advantageous vehicle is water.
There is often a well – known and established stability profi le of many pharmaceuticals in aqueous solutions, accelerating the route to the clinic in any development program. Drug solubility can be manipulated by choice of water, ethanol, or mixtures of the two to increase formulation options and doses. In multidose reservoir – type devices, a preservative would be required to prevent microbial contamination.
This is in addition to the current federal regulations that all aqueous – based drug products for oral inhalation must be manufactured to be sterile.
Respimat
The Respimat inhaler was recently launched in Germany as a combination product of fenoterol and ipratropium hydrobromide (Berodual) and was licensed for th treatment of chronic obstructive airway disease. A large body of literature now exists documenting the aerosol characteristics and clinical performance of the Respimat inhaler with a number of different drugs. Aerosolized formulations include the steroids budesonide and fl unisolide in addition to the β agonist fenoterol as well as the commercially available combination product of fenoterol and ipratropium bromide.
The Respimat device is a multidose reservoir system that is primed by twisting the device base (Figure 6 ). This compresses a spring and transfers a metered volume of formulation from the drug cartridge to the dosing chamber. The metered volume is between 11 and 15 μ l depending upon the drug formulation. When the device is actuated (in coordination with the patient ’ s inspiration), the spring is released. This forces a micropiston into the dosing chamber and pushes the solution through the uniblock. The uniblock is the heart of the aerosol generation system and consists of a fi lter structure with two fi ne outlet nozzle channels. The uniblock produces two fi ne jets of liquid that converge at a precisely set angle and then collide. This collision aerosolizes the liquid to form an aerosol.
Aerosols generated from the Respimat inhaler have been characterized as having a prolonged aerosol cloud duration compared to MDIs and have a slower cloud velocity as measured using video camera imaging. Hochrainer et al. (2005) measured the cloud duration of the Respimat aerosol to be 0.2 – 1.6 s compared to less than 0.2 s for HFA and CFC MDIs. Aerosol velocities have been reported as less than 1 m/s for the Respimat, compared to 6 – 8 m/s for CFC MDI inhalers. While a degree of patient coordination is required to actuate the Respimat and to inhale,
FIGURE 6 Respimat Inhaler. (Courtesy of Boehringer Ingelheim.)
the longer duration of aerosol cloud generation makes this maneuver less critical than with MDIs.
Aqueous and ethanolic formulations have been employed with the Respimat and the in vitro aerosol performance determined. Zierenberg (1999) reported fi ne – particle fractions of 66% for an aqueous fenoterol formulation and 81% for an ethanolic fl unisolide formulation. The respective MMADs were 2.0 Ѓ} 0.4 μ m for the aqueous formulation and 1.0 Ѓ} 0.3 μ m for the ethanolic formulation.
AER x
The AERx system was developed for the systemic delivery of insulin. Unit – dose aqueous solution formulations were produced in a blister strip design. The fi rst – generation AERx device is a battery – operated device that guides the patient through the inhalation technique required to successfully deliver a dose. It can also monitor dose times and frequency together with the facility to download dosing data in the clinic. A number of macromolecules, including insulin, and traditional small molecules (e.g., morphine) have been investigated using the AERx technology.
Aerosol generation using the AERx system is achieved by mechanically forcing a dose of the liquid formulation though a nozzle array in its disposable unit – dose blisters. The electronic version of the AERx inhaler guides the patient to inhale at the required fl ow rate. A cam – operated piston mechanism is actuated to compress the blister and extrude the dose as an aerosol through the nozzle array into warmed fl owing air. The nozzle array consists of a number of laser – drilled holes. Nozzle design characteristics can be altered depending upon the formulation characteristics and the desired droplet particle size. The single – use nature of the blister avoids potential problems such as microbial contamination from a dosing solution reservoir and nozzle – clogging issues.
A number of prototype versions of the AERx system have been investigated. In general, the in vitro aerosol characteristics revealed that about 50 – 60% of the loaded dose was emitted from the device, of which over 90% was respirable. MMADs ranged from 1 to 3 μ m depending upon the formulation and nozzle array. In a scintigraphic study, lung deposition following inhalation from the AERx was 53.3% (expressed as a percentage of the radioactivity in the AERx blister) compared to 21.7% for an MDI [285] .
A number of clinical studies delivering insulin to diabetic patients using the AERx system are currently ongoing. Hermansen et al. (2004) concluded that in type II diabetics, preprandial inhaled insulin via the AERx was as effective as preprandial subcutaneous insulin in achieving glycemic control. Clinical studies with morphine revealed comparable analgesic effi cacy for a matched dose of inhaled and intravenous morphine in a postsurgical pain model. In addition, the AERx inhaler has been employed for the topical delivery of rhDNase to cystic fi brosis patients. A mean relative increase in forced expiratory volume in 1 s (FEV1) of 7.8% was observed after 15 days treatment compared to control.
Mystic
The Mystic inhaler offers a soft mist aerosol generated from solution or suspension formulations. Unlike the previously described soft mist inhalers which use purely mechanical forces to generate the aerosol, the Mystic inhaler applies an electric fi eld to the formulation within the spray nozzle. An electric charge builds on the fl uid surface and, as the droplets exit the nozzle, the repelling force of the surface charge overcomes the surface tension of the droplets to form a soft mist droplet aerosol. This process is known as electrohydrodynamic aerosolization or electrospray.
The particle size characteristics of the aerosol can be controlled by adjusting the physical and chemical characteristics of the formulation together with the formulation fl ow rate and electrical fi eld properties. The inhaler consists of a number of components, a drug containment system, metering system, aerosol nozzle, power supply, and microprocessor, all enclosed in a housing. To date, Ventaira reports that the inhaler has been successfully employed to generate aerosols from small – molecule formulations (albuterol, triamcinolone, cromolyn, budesonide, and terbutaline) and macromolecules, including insulin.
Capillary Aerosol Generator
In the capillary aerosol generator (CAG) system, the aerosol is formed by pumpingяthe drug formulation through a small, electrically heated capillary. Upon exiting the capillary, the formulation is rapidly cooled by ambient air to produce an aerosol.
The generated aerosol characteristics are dependent upon the formulation employed.яUsing propylene glycol as a condensing vehicle, drug containing condensation aerosolsяare generated. When using water, ethanol, or combinations of both asяnoncondensing excipients, a stream of solid particles is delivered as a soft mistяaerosol. In vitro studies using budesonide, cromolyn sodium, buprenophine, albuterol,яand insulin have been performed to demonstrate various applications of theяCAG technology. These studies are characterized by high emitted doses and highяfi ne – particle fractions. Using noncondensing excipients, it is possible to produceяaerosols with vastly different size characteristics, depending upon the requiredяapplication.
Staccato
This technology utilizes a rapid heating technique to vaporize a thin fi lm of drug.
Following vaporization, the drug particles condense in the inhalation fl ow streamяto form a respirable aerosol and are inhaled. Single – and multiple – dose breath -яactuated inhalers are currently in development. As with any method involvingяheating of a formulation, drug degradation must be minimal. Rabinowitz et al.я(2006) described the absorption of prochlorperazine from human lungs as similarяto the pharmacokinetic profi les observed following intravenous administration.
Pharmaceutical aerosol drug delivery has been established for over 50 years. Pulmonaryяadministration remains the route of choice for local treatment of respiratoryяdiseases. Over the past decade there have been changes in both the diseases treatedяby this route and the devices used for aerosol generation. Future advances will seeяpulmonary delivery of gene therapy and vaccines, together with improved drug targeting within the respiratory tract using novel inhalers.
Content of drugs under pressure is emulsion. An emulsion is a heterogeneous preparation composed of two immiscible liquids (by convention described as oil and water), one of which is dispersed as fine droplets uniformly throughout the other. Emulsions are thermodynamically unstable and revert back to separate oil and water phases by fusion or coalescence of droplets unless kinetically stabilized by a third component, the emulsifying agent. The phase present as small droplets is called the disperse, dispersed, or internal phase and the supporting liquid is known as the continuous or external phase. Droplet diameters vary enormously, but in pharmaceutical emulsions they are typically polydispersed with diameters ranging from approximately 0.1 to
Multiple emulsions of the oil-in-water-in-oil (o/w/o) type are w/o emulsions in which the water globules themselves contain dispersed oil globules; conversely, water-in-oil-in-water (w/o/w) emulsions are those where the internal and external aqueous phases are separated by the oil (Fig.
Emulsions are formulated for virtually all the major routes of administration, and there are a number of dermatological, oral and, parenteral preparations available commercially. The internal phase may contain water-soluble drugs, preservatives, and flavoring agents whilst the oil phase may itself be therapeutically active or may act as a carrier for an oil-soluble drug.
Such preparations provide an effective approach to many of the problems in drug delivery, often showing distinct advantages over other dosage forms by way of improved bioavailability and/or reduced side effects.
However, despite such advantages, emulsions are not used as extensively as other oral or parenteral dosage forms due to the fundamental problems of emulsion instability that result in unpredictable drug release profiles and possible toxicity. The full potential of emulsions will not be realized until stable systems are developed with predictable in vitro and in vivo release patterns. Much of the emulsion research over the past decade is based on attempts to understand the relationships between emulsion stability, physicochemical properties, and biological fate. Multiple emulsions are even more difficult to stabilize, and characterize and although there is an increasing interest in their potential applications for drug delivery, at present there are no commercial preparations available.[2]
PHARMACEUTICAL APPLICATIONS
The current and potential pharmaceutical applications of emulsions have been the subject of a number of general reviews.[3–6] Traditionally the term ‘‘emulsion’’ is restricted to mobile emulsions for internal use; emulsions for external use are described by their pharmaceutical types as liniments, lotions, and creams. This tends to conceal the fact that by far the largest group of emulsions currently used in pharmacy and medicine are dermatological emulsions for external use.[7,8] Both oil-in-water and water-in-oil emulsions are extensively used for their therapeutic properties and/or as vehicles to deliver drugs and cosmetic agents to the skin. The
emulsion facilitates drug permeation into and through the skin by its occlusive effects and/or by the incorporation of penetration-enhancing components. Particular attention is paid to patient acceptance of such formulations, which range in consistency from mobile liniments and lotions to semisolid ointments and creams.
In the past, the development of dermatological emulsions was essentially empirical with only a limited understanding of the underlying principles. Today, although the microstructure of many of these complex formulations is now better understood,[9,10] the mechanisms by which the structure of an emulsion can influence drug bioavailability are far from clear and much of the literature on the role of emulsions in drug release to the skin is contradictory. Confusion arises because the majority of investigations concerning in vitro vehicle effects on drug release are only on bulk formulation. As most emulsions are applied to the skin as a thin film, the drug delivery system is not one of bulk emulsion, but rather a dynamic evaporating system in which phase changes can occur as the preparation is rubbed into the skin and the relative concentrations of volatile ingredients alter. Droplet size appears to influence drug delivery to the skin, with submicron lipid emulsions enhancing the transcutaneous permeation and efficiency of a number of lipophilic drugs.[11]
Oral emulsions are almost exclusively of the oil-inwater type. They provide a degree of taste masking as the aqueous external phase effectively isolates the oil from the tongue. Mineral and castor oils have been emulsified in water and administered orally for the local treatment of constipation for many years (cf. Mineral Oil Emulsion USP) as have various nutritional oils from fish liver (generally halibut or cod) or vegetable origin to produce oral liquid food supplements. It has long been established that the use of o/w emulsions as carriers for lipophilic drugs may improve oral bioavailability and efficacy.[3–6] For example, griseofulvin formulated as an o/w emulsion has enhanced gastrointestinal absorption when compared with suspensions, tablets, or capsule dosage forms.[12] The mechanisms by which emulsions modify and improve absorption processes are complex and not fully understood, although the oil itself influences gastric motility. Fats and oils are solubilized by the bile salts so that the administration of already emulsified oil droplets containing a high concentration of drug may increase the likelihood of further droplet and drug solubilization and transport across the GI tract by the fat absorption pathways.
The type of emulsion used parenterally depends on the route of injection and the intended use.[13–15] Oil-inwater emulsions are administered by all the major parenteral routes whereas water-in-oil emulsions are generally reserved for intramuscular or subcutaneous administration where sustained release is required.
Drug action is prolonged in such oily emulsions because the drug has to diffuse from the aqueous dispersed phase through the oil-continuous environment to reach the tissue fluids. Water-in-oil emulsions are used to disperse water-soluble immunizing antigens in mineral oil for injection via subcutaneous or intramuscular routes as adjuvant preparations where they prolong and enhance the antigenic stimulus and increase the antibody titer. Oily emulsion formulations also show promise in cancer chemotherapy as vehicles for prolonging drug release after intramuscular or intratumoral injection, and as a means of enhancing the transport of anticancer agents via the lymphatic system.[16]
Water-in-oil emulsions for sustained release are often difficult to inject because of the high viscosities of the oily continuous phases. Although these problems can be overcome by reemulsification of the primary w/o emulsion to produce a less viscous multiple w/o/w emulsion, a study using 5-fluorouracil implied that sustained release was actually less marked with multiple emulsions.[17]
Sterile parenteral oil-in-water emulsions have been used extensively for over 40 years for the intravenous administration of fats, carbohydrates, and vitamins to debilitated patients. Several vegetable oil-in-water emulsions are now available commercially with droplet sizes similar to that of chylomicrons (approximately 0.5–2 mm), the natural fat droplets in the blood that transport ingested fats to the lymphatic and circulatory systems (Table 1). More recently, such emulsions have been employed as intravenous carriers for poorly water-soluble lipophilic drugs such as vitamin K (e.g., Sterile Phytonadione Injection U.S.P.) diazepam (e.g., Diemuls_), vitamin A (Vitlipid N_), and profonol (Diprovan_) as alternatives to the traditionally used cosolvent, surfactant solubilized, or pH controlled parenteral solutions. The drug dissolved in the oil phase of the emulsion is unlikely to precipitate and cause pain when diluted by blood on injection, and if susceptible to hydrolysis or oxidation, it will be protected by the non-aqueous environment. Emulsion formulations of diazepam and more recently clarithromycin have been clinically shown to be less painful than solubilized preparations[18,19] while emulsions containing amphoteracin B are less toxic.[20] This emulsion was also shown to be an equally effective, cheaper, and more elegant alternative to a liposomal system.
The enormous literature on the potential of lipid emulsions for drug delivery and targeting is discussed in a recent book.[21]
Radiopaque emulsions, which have long been used as contrast media in conventional X-ray examinations of body organs, are finding further application with more sophisticated techniques including computed tomography, ultrasound, and nuclear magnetic resonance.
Perfluorochemical emulsions are used as artificial blood substitutes. The potential advantages of such systems over donated blood are enormous with the elimination of major donor associated problems such as blood group incompatibilities and blood disease.
The first commercial product, Fluosol-DA_ (Green Cross Corporation, Osaka, Japan) was licensed several years ago in a number of countries to reduce myocardial ischaemia in patients undergoing angioplasty; however, Fluosol-DA was not a commercial success due to its slow excretion rate and to its marked instability, which meant that it had to be stored in the frozen state. In addition, some patients were sensitive to one of the emulsifiers, pluronic F68_. Currently, a second generation of emulsions is being evaluated to resolve the problems encountered with Fluosol[22] and these are discussed in another chapter of this encyclopedia.
There are only a few studies on the ocular and nasal applications of emulsions. Lipid (submicron) emulsions exhibited a long-lasting antidepressant effect on the intraocular pressure of rabbits after a single application when used as carriers for lipophilic antiglaucoma drugs.[23] Medium-chain triglyceride emulsions formulated at pH 8 show potential as controlled release formulations for nasal delivery[24,25] for they give prolonged drug residence in the nasal cavity (Fig. 2).
Enhanced nasal delivery of insulin was observed when insulin was incorporated into the continuous phase of an o/w emulsion, but not when incorporated into the aqueous phase of a w/o emulsion.[26]
FORMULATION CONSIDERATIONS
The choice of oil, emulsifier, and emulsion type (o/w, w/o, or multiple) is limited by its ultimate use and route of administration. Potential toxicity and chemical incompatibilities in the final formulation must be taken into account as must processing details for these also affect the variables that control emulsion stability and therapeutic response such as droplet size distributions and rheology. The design of stable emulsions with the correct pharmacokinetic characteristics and tissue distribution is currently an area of enormous interest, particularly for parenteral IV emulsions.
Immediately after injection, the surface of the parenteral emulsion droplets is altered by adsorption of blood components (optosonation) and they are then distributed rapidly through the circulation. Their subsequent fate depends on whether they are treated by the body in the same manner as chylomicrons, or whether they are recognized as foreign particles and cleared by the RES. Many factors, including droplet size and charge, the type of lipid, and the emulsifier composition influence their fate. A major factor to be considered in the formulation of oral preparations is the low pH and high ionic strength of stomach fluids, which may destabilize the emulsion by its effect on the emulsifier.
Pharmaceutical Oils
Oils used in the preparation of pharmaceutical emulsions are of various chemical types, including simple esters, fixed and volatile oils, hydrocarbons, and turpenoid derivatives. The oil itself may be the medicament, it may function as a carrier for a drug, or even form part of a mixed emulsifier system as in the case of some fixed oils that contain sufficient free fatty acids.
Many oils, particularly those of vegetable origin, are liable to autooxidation with subsequent rancidity, and it is frequently necessary to add an antioxidant and/or preservative to inhibit this degradation process. For externally applied emulsions, mineral oils, either alone or combined with soft or hard paraffins, are widely used both as the vehicle for the drug and for their occlusive and sensory characteristics. The most widely used oils in oral preparations are non-biodegradable mineral and castor oils that provide a local laxative effect, and fish liver oils or various fixed oils of vegetable origin (e.g., arachis, cottonseed, and maize oils) as nutritional supplements.
The choice of oil is severely limited in emulsions for parenteral administration for reasons of toxicity.
Purified soybean, sesame, safflower, and cottonseed oils composed mainly of long-chain triglycerides have been used for many years as they are resistant to rancidity and show few clinical side effects. More recently, it has been recognized that the structure of the oil will influence the fate of emulsion droplets after injection.
Mixtures containing both long- and medium-chain triglycerides are not only better energy sources for nutritional purposes but they are also cleared more rapidly from the circulation;[27] such mixtures are now used in commercial preparations (c.f. Table 1).
Structured triglycerides, formed by modifying the oil enzymatically to produce 1,3-specific triglycerides are an area of increasing interest because of their influence on the in vivo circulation time of an emulsion.[28]
Purified mineral oil is used in some water-in-oil depot preparations where mineral toxicity (e.g., abscess formation at the injection site) must be carefully balanced against efficiency. Emulsified perfluorochemicals are considered acceptable for IV use provided that they are excreted relatively fast. A major problem in the formulation of the early perfluorocarbon emulsions was that the oils that form the most stable emulsions were not cleared rapidly from the body.
Pharmaceutical Emulsifiers
Emulsifying agents are used both to promote emulsification at the time of manufacture and to control stability during a shelf life that can very from days for extemporaneously prepared emulsions to months or years for commercial preparations. In practice, combinations of emulsifiers rather than single agents are used.
The emulsifier also influences the in vivo fate of lipid parenteral emulsions by its influence on the surface properties of the droplets and on the droplet size distributions.
For convenience, most pharmacy texts classify emulsifiers into three groups: i) surface active agents; ii) natural (macromolecular) polymers; and iii) finely divided solids.
Active Agents
The range of surfactant emulsifiers used in pharmaceutical preparations is illustrated in Table 2. Surfactants are manufactured from a variety of natural and synthetic sources and consequently they show considerable batch-to-batch variations in their homologue compositions and in trace impurities from the starting material. For example, batch variations in the number of neutral phospholipids occur in lecithin surfactants and non-ionic polyethylene surfactants show variations in the number of moles of ethylene oxide. The mechanisms by which such batch variations lead to differences in emulsifying properties are now better understood.[29]
Although synthetic and semisynthetic surfactants form by far the largest group of emulsifiers studied in the scientific literature and many of them are available commercially, their use in pharmaceutical emulsions is limited by the fact that the majority are toxic (i.e., haemolytic) and irritant to the skin and mucous membranes of the gastrointestinal tract. In general, cationic surfactants are the most toxic and irritant and non-ionic surfactants the least. Surfactants are therefore used mainly at relatively low concentrations in topical preparations. The quaternary ammonium compounds constitute an important group of cationic emulsifiers in dermatological preparations because they have antimicrobial properties in addition to their o/w emulsifying action. There are many non-ionic surfactants with different oil and water solubilities available commercially because for each fatty starting material the polyoxyethylene chain length can be modified by the systematic addition of ethylene oxide groups. However, a limited number of polysorbate surfactants are used in oral emulsions, and parenteral preparations appear to be based only on the lecithins from plant or animal sources and the non-ionic polyoxyethylene oxide/polyoxypropylene oxide block copolymer poloxomer 188 (Poloxamer F68_), although some patients using the first generation of perfluorochemical emulsions were sensitive to this poloxomer. The emulsifier influences both emulsion stability and in vivo disposition by its influence on droplet surface properties.
Natural macromolecular materials and finely divided solids Materials derived from natural sources (Table 3) may originate from animal or vegetable sources and many of these products are susceptible to degradation. For example, depolymerization (the polysaccharides) or hydrolysis (the steroids) usually lead to loss in emulsifying power. Some of these materials, polysaccharides and proteins in particular, provide good culture medium for microorganisms, and therefore preservation of emulsions containing them is imperative.
To overcome these problems, a number of purified and semisynthetic derivatives are available, including various purified wool fat derivatives and semisynthetic celluloses such as methylcellulose and sodium carboxymethylcellulose.
These are generally more stable than the unmodified materials. These celluloses are used in oral preparations; they are less suitable for topicals because of their unpleasant feel. Finely divided solids such as clays are used in dermatological preparations as structuring agents.
Preservatives
It is essential that emulsions are formulated to resist microbial attack, as this not only can affect the physicochemical properties of the formulation, causing color, odor, or pH changes and even phase separation, but may also constitute a health hazard. The potential sources of contamination can be from raw materials (especially if these are natural products), water, manufacturing and packaging equipment, or patients themselves. W/o emulsions are less susceptible to attack than o/w emulsions because the aqueous continuous xternal phase can produce ideal conditions for the growth of bacteria, moulds, and fungi. Preservatives are not used in parenteral emulsions, which are sterilized, generally by autoclaving, but sometimes by using sterile components and aseptically assembling the final emulsion.
There is no simple way of predicting the ideal preservative for a particular emulsion. In addition to requiring a wide spectrum of activity against bacteria, yeasts, and molds, the preservative should be free from toxic, irritant, or sensitizing activity. Some commonly used preservatives in oral and topical preparations include phenoxyethanol, benzoic acid, parabenzoates, and chlorcresol. Emulsions are heterogeneous products, and the preservative partitions between the oil and aqueous phases. As a sufficient aqueous concentration of the active (usually unionized) form must be present to ensure proper preservation, pH is an additional factor to be considered. Problems often arise because many of the materials used in emulsion formulation, for example hydrocolloids or polyoxyethylene surfactants, can interact with the preservatives, thus depleting their activity. The use of a single preservative is often considered unrealistic, and attention is being increasingly focused on the use of mixtures for a wider spectrum of activity, although this may introduce additional compatibility problems.
Antioxidants and Humectants
Antioxidants are added to many pharmaceutical preparations to prevent oxidative deterioration on storage of the oil, emulsifier, or the drug itself. Such deterioration, as well as destabilizing the formulation, imparts an unpleasant odor or taste. Some oils are supplied containing antioxidants already. Those commonly used in pharmacy include butylated hydroxyanisole (BHA) and butylated hydroxytoluene (BHT) at concentrations up to 0.2%, and the alkyl gallates. Humectants such as propylene glycol, glycerol, and sorbitol (5%) are often added to dermatological preparations to reduce the evaporation of water from the emulsion during storage and use. They are sometimes claimed to prevent evaporation of water from the surface of the skin, although their use at high concentrations would be expected to have the opposite effect (i.e.,remove moisture and dehydrate the skin).
EMULSION FORMATION
There are essentially two major considerations in emulsification: first, the formation of emulsions of the correct type, oil-in-water, water-in-oil, or multiple emulsion with the required droplet size distributions and second, the stabilization of the dispersed droplets so formed. When given amounts of two immiscible liquids are mixed or mechanically agitated in the absence of other additives, both phases tend to form droplets of various sizes. The size distributions are related to the forces involved during the agitation process, and the number of droplets of each liquid depends on its relative volume. The surface free energy of the system, which is dependent on both total surface area and interfacial tension is raised by the increase in surface area produced during dispersion, and the system is thermodynamically unstable. To reduce this, highenergy droplets first assume a spherical shape, as this gives the minimum surface area for a given volume, and then on collision the droplets will coalesce (fuse) to reduce the interfacial area, the interfacial tension remaining constant.
The type of emulsion that forms, either o/w or w/o, depends on the relative rates of coalescence of each type of droplet, with the more rapidly coalescing droplets forming the continuous phase. Generally, this is the liquid present in the larger amount because higher number of droplets increase the probability of collision and coalescence. With phase volumes of oil and water close to 50%, other factors such as the order and rate of addition of each liquid are important. If agitation ceases, coalescence will continue until complete phase separation—the state of minimum free energy—is reached. Thus, emulsification can be considered as the result of two competing processes, namely the disruption of bulk liquids to produce fine droplets and the recombination of the droplets to give back the original bulk liquids. With the inclusion of surfactants or other classes of emulsifiers, the type of emulsion that forms is no longer simply a function of the phase volume and the order of mixing, but also the relative solubilities of the emulsifier in oil and water. In general, the phase in which the emulsifier is most soluble becomes the continuous phase (Bancroft’s Rule); thus, hydrophilic polymers and surfactants promote o/w emulsions whereas lipophilic surfactants promote w/o emulsions.
Preferential coalescence of the phase in which the emulsifier is most soluble occurs because when droplets collide, the emulsifier is easily displaced from the interface into the droplet, thus providing little protection against coalescence. Theoretically, the disperse phase of an emulsion can occupy up to 74% of the phase volume, and such high internal phase o/w emulsions have been produced with suitable surfactants. It is more difficult to formulate w/o emulsions with greater than 50% disperse phase because of the steric mechanisms involved in their stabilization (discussed later), and the addition of extra water sometimes causes inversion to an o/w emulsion.
Emulsion Characteristics
In general, an emulsion exhibits the characteristics of its external phase. Several methods are available for identifying the emulsion type. Dilution tests are based on the fact that the emulsion is only miscible with the liquid that forms its continuous measurements rely on the poor conductivity of oil compared with water, and give low values in w/o emulsions where oil is the continuous phase. Staining tests in which a water-soluble dye is sprinkled onto the surface of the emulsion also indicate the nature of the continuous phase. With an o/w emulsion there is rapid incorporation of the dye into the system whereas with the w/o emulsion the dye forms microscopically visible clumps. The reverse happens on addition of an oil-soluble dye. These tests essentially identify the continuous phase and do not indicate whether a multiple emulsion has been produced. This can be resolved by microscopy.
Rheology
The rheological properties of emulsions are influenced by a number of interacting factors, including the nature of the continuous phase, the phase volume ratio, and to a lesser extent, particle size distributions.
A variety of products ranging from mobile liquids to thick semisolids can be formulated by altering the dispersed phase volume and/or the nature and concentration of the emulsifiers. For low internal phase volume emulsions, the consistency of the emulsion is generally similar to that of the continuous phase; thus, w/o emulsions are generally thicker than o/w emulsions, and the consistency of an o/w system is increased by the addition of gums, clays, and other thickening agents that import plastic or pseudoplastic flow properties. Some mixed emulsifiers interact in water to form a viscoelastic continuous phase to give a semisolid o/w cream.[7]
Droplet Size Distributions
Droplet size distributions in pharmaceutical emulsions are important from both stability and biopharmaceutical considerations. The larger the particle size, the greater the tendency to coalesce and further increase droplet size. Thus, fine particles generally romote better stability. Size distributions are influenced by the characteristics of the emulsifier and the method of manufacture. From a biological point of view, fine emulsification enhances gastrointestinal absorption and whilst this is desirable with oral formulations of nutrient oils alone or with drugs dissolved in them, it may give adverse clinical effects with mineral oils that are used for a local effect and are toxic if absorbed. Droplets in emulsions used as contrast media incomputed tomography are approximately 1–3 mm. Parenteral emulsions should be formulated so that the dispersed droplet sizes range from approximately 100nm to
Drug delivery from dermatological preparations also appears to be improved in lipid emulsions containing submicron droplets.[11]
EMULSION STABILITY
A stable emulsion is considered to be one in which the dispersed droplets retain their initial character and remain uniformly distributed throughout the continuous phase for the desired shelf life. There should be no phase changes or microbial contamination on storage, and the emulsion should maintain elegance with respect to odor, color, and consistency. Instabilities of both chemical and physical origins can occur in emulsion formulations. Chemical instabilities, such as the development of rancidity in natural oils due to oxidation by atmospheric oxygen, the depolymerization of macromolecular emulsifiers by hydrolysis, or microbial degradation can be minimized by the addition of suitable antioxidants and preservatives.
More general chemical instabilities involving interactions between the drug and emulsion excipients or between the excipients themselves may lead to physical instabilities. If these interactions involve the emulsifying agent, they may destroy its emulsifying properties, causing the emulsion to break. For example, interactions between phenolic preservatives and polyoxyethylene non-ionic emulsifiers may lead to loss of emulsifying power as well as poor preservation.
Physical Instability
As emulsions are inherently unstable, they eventually revert to the original state of two separate liquids, that is, will break or crack. In the presence of an emulsifier and other additives, this state is approached via several distinct processes, some of which are reversible such as creaming and flocculation and others irreversible such as coalescence and Ostwald ripening. Phase inversion when an oil-in-water emulsion inverts to form a water-in-oil emulsion or visa versa is a special case of irreversible instability that occurs only under welldefined conditions such as a change in emulsifier solubility due to specific interactions with additives or to a change in temperature (Fig. 3).
Flocculation describes a weak reversible association between emulsion globules separated by thin films of continuous phase. The individual droplets retain their separate identities, but each floccule or cluster of droplets behaves physically as a single unit. The association arises from the interaction of attractive and repulsive forces between droplets and is reversible in the sense that the original dispersion can generally be regained by mild agitation. Flocculation is generally regarded as a precursor to the irreversible process of coalescence, although sometimes the time scale between flocculation and coalescence can be extended almost indefinitely by the adsorbed emulsifier, giving a kinetically stable emulsion.
Coalescence, where dispersed phase droplets merge to form larger droplets, takes place in two distinct stages. It begins with the drainage of liquid films of continuous phase from between the oil droplets as they approach one another and ends with the rupture of the film when a critical thickness is reached. The approaching droplets may deform as the opposing surfaces distort to either flatten (small droplets) or dimple (larger droplets) under the hydrodynamic pressures generated by viscous flow of the continuous phase.
Coalescence is not the only mechanism by which dispersed phase droplets increase in size. If the emulsion is polydispersed and there is significant miscibility between the oil and water phases, then Ostwald ripening, where droplet sizes increase due to large droplets growing at the expense of smaller ones, may also occur. This destabilizing process is a result of the Kelvin effect and occurs when small emulsion droplets (less than
Creaming or sedimentation occurs when the dispersed droplets or floccules separate under the influence of gravity to form a layer of more concentrated emulsion, the cream. Generally a creamed emulsion can be restored to its original state by gentle agitation.
This process, which inevitably occurs in any dilute emulsion if there is a density difference between the phases as a consequence of Stokes law, should not be confused with flocculation which is due to particle interactions resulting from the balance of attractive and repulsive forces. Most oils are less dense than water so that the oil droplets in o/w emulsions rise to the surface to form an upper layer of cream. In w/o emulsions, the cream results from sedimentation of water droplets and forms the lower layer. According to Stokes Law, the rate of creaming can be minimized by reducing droplet sizes and/or thickening the continuous phase. Adjustment of the densities of the two phases has received little attention.
The destabilization processes are not independent and each may influence or be influenced by the others.
For example, the increased droplet sizes after coalescence or Ostwald ripening will enhance the rate of creaming, as will the formation of large floccules which behave as single entities. In practice, creaming, flocculation, and Ostwald ripening may proceed simultaneously or in any order followed by coalescence.
Coalescence and Ostwald ripening are obviously the most serious types of instability as they result in the formation of progressively larger droplets and ultimately lead to phase separation. Creaming and flocculation, on the other hand, are more subtle forms of instability, for although they represent potential steps towards coalescence and breaking due to the close proximity of the droplets, many practical emulsions remain in this state for long periods of time without significant coalescence and can be redispersed simply by shaking the container.
EMULSION STABILIZATION
Emulsifiers stabilize emulsions in a number of different ways, all of which act to prevent or delay the various destabilization processes described previously.
The emulsifier may form an interfacial film at the oil–water interface and/or structure (i.e., thicken) the continuous phase. The interfacial film introduces additional repulsive (e.g., electrostatic, steric, or hydrational) forces between droplets to counteract attractive van der Waals forces and inhibit the close approach of droplets. It may also provide a barrier to the coalescence of droplets in close proximity, particularly if the film is close-packed and elastic.
Surfactant interfacial films also lower the interfacial tension between oil and water. Although this effect is important during the emulsification process where it facilitates the breakup of droplets, it is not a major factor inmaintaining the long-term stability of emulsions.
Inemulsions that are thickened by the emulsifier, the interfacial film does not play the dominant role in maintaining stability; rather, it is the structured continuous phase that forms a rheological barrier to prevent the movement and hence the close approach of droplets and also inhibits Ostwald ripening.
Classical (Interfacial) Theories
Classical theories of emulsion stability focus on the manner in which the adsorbed emulsifier film influences the processes of flocculation and coalescence by modifying the forces between dispersed emulsion droplets.
They do not consider the possibility of Ostwald ripening or creaming nor the influence that the emulsifier may have on continuous phase rheology. As two droplets approach one another, they experience strong van der Waals forces of attraction, which tend to pull them even closer together. The adsorbed emulsifier stabilizes the system by the introduction of additional repulsive forces (e.g., electrostatic or steric) that counteract the attractive van der Waals forces and prevent the close approach of droplets. Electrostatic effects are particularly important with ionic emulsifiers whereas steric effects dominate with non-ionic polymers and surfactants, and in w/o emulsions. The applications of colloid theory to emulsions stabilized by ionic and non-ionic surfactants have been reviewed as have more general aspects of the polymeric stabilization of dispersions.[4,31,32]
The DLVO theory, which was developed independently by Derjaguin and Landau and by Verwey and Overbeek to analyze quantitatively the influence of electrostatic forces on the stability of lyophobic colloidal particles, has been adapted to describe the influence of similar forces on the flocculation and stability f simple model emulsions stabilized by ionic emulsifiers.
The charge on the surface of emulsion droplets arises from ionization of the hydrophilic part of the adsorbed surfactant and gives rise to electrical double layers. Theoretical equations, which were originally developed to deal with monodispersed inorganic solids of diameters less than
The theory states that the forces between droplets can be considered as the sum of an attractive van der Waals part VA and a repulsive electrostatic part VR when identical electrical double layers overlap. As the origin of each force is independent of the other, each is evaluated separately, and the total potential of interaction VT between the two droplets as a function of their surface-to-surface separation is obtained by summation VT ј VA ю VR
A schematic potential energy of interaction with distance plot is shown in Fig. 4A. It can be seen that a weak attraction occurs at large droplet separations represented by the secondary energy minimum, and a very strong attraction at small droplet separations hence the very deep primary minimum. At intermediate distances, double-layer repulsion dominates and there is a maximum in the curve. Flocculation occurs in the secondary minimum, where the attractive forces are relatively weak and floccules are easily separated by low energy agitation. Once flocculated, droplets are prevented from approaching closer by the potential energy barrier. If they have sufficient energy to overcome the barrier, the process of coalescence commences as the droplets move closer together. Once in the primary minimum the aggregates formed are separated by only a small distance so that stability against coalescence is determined by the resistance of the interfacial film to rupture.
The height of the energy barrier, which is crucial to emulsion stabilization, depends on the state of ionization of the emulsifying agent. Most surfactants are used at pH values where they are totally ionized so that the surface potential is high, giving a correspondingly high energy barrier. The surface potential cannot be measured directly, but can be estimated from the experimentally derived zeta potential. In lipid emulsions for parenteral nutrition, the electrostatic barrier is provided by the ionization of the negatively charged phospholipids in the emulsifier film at the oil droplet–water interface. At physiological pH, a typical fatemulsion carries a negative charge with the zeta potentialbetween 30 and 60mV. This is sufficient to ensurestability because of the high potential energy barrier.
The addition of electrolytes or a change in pH can havea devastating effect on emulsion stability by compressing the double layers, thus reducing the zeta potential and energy barrier and allowing droplets to move into the primary minimum. Thus, great care must be exercised when electrolytes are added nutritional emulsions.
With emulsifiers such as proteins and gums, ionization, and hence emulsifying activity, is also pH dependent (c.f. Table 3).
The DLVO theory does not explain either the stability of water-in-oil emulsions or the stability of oilin-water emulsions stabilized by adsorbed non-ionic surfactants and polymers where the electrical contributions are often of secondary importance. In these, steric and hydrational forces, which arise from the loss of entropy when adsorbed polymer layers or hydrated hains of non-ionic polyether surfactant intermingle on close approach of two similar droplets, are more important (Fig. 4B). In emulsions stabilized by polyether surfactants, these interactions assume importance at very close distances of approach and are influenced markedly by temperature and degree of hydration of the polyoxyethylene chains. With block copolymers of the ethylene oxide–propylene oxide type, such as the poloxamers, the hydrated polyoxyethylene chains extend into the continuous phase to provide steric stabilization and the hydrophobic propylene oxide portion is anchored onto the droplet surface to form a strong protecting layer against coalescence. Stability is optimized when the droplet surfaces are completely coated by polymer chains so that desorption and lateral movement of the polymer is inhibited. With w/o emulsions, steric hindrance of the adsorbed chains of emulsifier can also result in entropic repulsion effects at small distances ofseparation.
Some natural polymeric emulsifiers such as the gums, in addition to forming steric and electrostatic barriers form thick multilayered films that are very resistant to film rupture. They may also thicken the continuous phases of o/w emulsions, thereby
reducing the rate of film drainage in the initial stages of coalescence. Small solid particles may stabilize emulsions if they are wetted by both phases and possess sufficient adhesion for one another to form a coherent interfacial film. The film serves as a mechanical barrier to prevent the coalescence of droplets, and if charged, electrostatic mechanisms further assist in the stabilization of the emulsion. Although solids are not generally sufficient to stabilize emulsions on their own, they often reinforce the effectiveness of other emulsifiers.
Stabilization by Mixtures of Emulsifiers
Most pharmaceutical emulsions, whether dilute mobile systems for internal use or thick semisolid creams for application to the skin, contain mixtures of emulsifiers, as these provide more stable preparations. For example, traditional oral preparations are sometimes stabilized by mixtures of gums such as acacia and tragacanth and mixtures of non-ionic surfactants of high and low hydrophile–lipophile balance (HLB) generally form more stable emulsions than a single surfactant. The ecithins used to stabilize parenteral emulsions are usually mixtures of neutral and charged lipids as are the partially neutralized glyceryl esters such as selfemulsifying glyceryl monostearate. Combinations of sparingly soluble long-chain acids, alcohols, or glyceryl esters with more soluble ionic and non-ionic surfactants are widely used in dermatological o/w lotions and creams, where they are sometimes added in the form of a preblended emulsifying wax (Table 4). Surfactant/ fatty acid combinations are also present in traditional liniment and lotion emulsion formulations prepared by the nascent soap method and in preparations where triethanolamine soaps are formed in situ from the interaction of triethanolamine and excess fatty acid.
Equations from the DLVO theory even if modified to allow for the steric repulsive forces cannot cope with mixtures of emulsifiers. Increased stability in model emulsions (c.f. Fig. 1A) is attributed not so much to the control of flocculation (although this does occur), but rather to the prevention or retardation of coalescence by closer packing of the molecules in the adsorbed monolayer to form a more rigid and condensed film. There is now substantial evidence that interactions between emulsifier components to form specific lamellar phases, either liquid crystalline or gel, that are capable of incorporating large volumes of water are important for the stability of many parenteral and dermatological emulsions. Mobile parenteral injections stabilized by phospholipid mixtures usually contain swollen lamellar liquid crystals[34] whereas a swollen gel phase which generally provides better stability as well as a means of controlling rheological properties dominates in semisolid dermatological emulsions prepared with emulsifying waxes. The relevance of bilayer gel and liquid crystalline phases in dermatological and parenteral emulsions have been discussed in reviews.[10,29] Much of the information about their structures was obtained from investigations of the phase behaviour of emulsifiers and their components in water over the ranges of concentration and temperature relevant to the manufacture, storage, and use of the formulations. It is interesting to note that the same electrostatic, hydrational, and steric forces that operate in simple emulsions also dominate the stability and properties of the lamellar phases.[35]
Ostwald Ripening
Ostwald ripening has not been studied as extensively in emulsions as has coalescence, although it is a major mechanism for instability in lipid and perfluorochemical emulsions with submicron droplet sizes where a condensed monolayer is not always necessary for emulsion stability.[36] Although surfactant interfacial films protect against flocculation and coalescence, Ostwald ripening may in fact be enhanced if the surfactant is above the critical micelle concentration (cmc) because of the diffusion of solubilized oil through the continuous phase. The addition of a third component to the emulsion that has a lower vapor pressure and solubility
than the disperse phase will also inhibit Ostwald ripening. The addition of long-chain alkanes to comparatively unstable oil-in-water emulsions prepared with sodium dodecyl sulfate resulted in marked increases in stability even though the alkanes do not effect the composition or mechanical properties of the oil–water interface.[37] The stability of pure perfluorodecalin emulsions used as blood substitutes is enhanced by the addition of a small quantity of perfluorotributylamine, and lipid emulsions containing local anaesthetic/analgesic drugs show enhanced stability in the presence of hydrophobic excipients of lower solubility than the disperse phase.[38] Polymeric emulsifiers possibly stabilize emulsions against Ostwald ripening by increasing the viscosity of the continuous phase. The relative lack of Ostwald ripening in emulsions prepared from oils immiscible with water, such as mineral oil, may partly explain why they are easier to emulsify than are more miscible vegetable oils used in parenteral preparations.
Selection of Emulsifier
Over the years there have been many attempts to find systemic methods for screening potential emulsifiers from the enormous number of surfactants available commercially. Although the mechanisms governing the stability of emulsions, including the complex multiple phase systems of pharmacy are becoming clearer, there are still few scientific guidelines to assist in the proper selection of emulsifiers for a particular emulsion.
Semiempirical methods based on both interfacial considerations and the phase behavior of the emulsifiers are considered briefly next.
The hydrophile–lipophile balance (HLB) concept
Griffin devised the concept of hydrophile–lipophile balance (HLB) and its additivity many years ago for selection of non-ionic emulsifiers and this rather empirical method is still widely used. The enormous literature on the HLB of surfactants has been reviewed by Becher.[39] Each surfactant is allocated an HLB number usually on a scale of 0–20, based on the relative proportions of the hydrophilic and hydrophobic part of a molecule. Water-in-oil emulsions are formed generally from oil-soluble surfactants of low HLB umber and oil-in-water emulsions from more hydrophilic surfactants of high HLB number. The method of selection is based on the observation that each type of oil will require an emulsifying agent of a specific HLB number to produce a stable emulsion. Thus, oils are often designated two ‘‘required’’ HLB numbers, one low and one high, for their emulsification to form water-in-oil and oil-in-water emulsions respectively. A series of emulsifiers and their blends with HLB values close to the required HLB of the oil are then examined to see which one forms the most stable emulsion (c.f. Fig. 1A).
Although the HLB concept narrows the range of emulsifiers to select and provides a schematic approach for the formulator, it is limited by its strict relation to molecular structure of the individual surfactants. The concept does not consider the total emulsion and is therefore insensitive to interactions between emulsifier components, the influence of temperature changes, or the presence of additional ingredients in the emulsion.
Consequently, not all emulsifier blends of the correct HLB form stable systems. For example, when surfactants of widely different HLB numbers are blended to give the optimum theoretical HLB, the high solubility f the surfactant in the oil and aqueous phases change the balance of the molecules at the interface and unstable emulsions may result. Similarly, if the added surfactants form intermolecular associations at the interface, the association complex is unlikely to have properties that are related in any simple way to the individual properties of the constituent molecules.
The phase inversion temperature (PIT) method (HLB-temperature) A complementary means of emulsifier selection, the phase inversion temperature (PIT), which employs a characteristic property of the emulsion rather than the properties of the emulsifiers in isolation, was introduced by Shinoda.[40] The method uses the fact that the stabilities of oil-in-water emulsions containing nonionic surfactants are closely related to the degree of hydration of the interfacial films. Emulsion stability is reduced by increase in temperature or added salts because these decrease the extent of interfacial film hydration. Phase inversion, due to a change from preferential water solubility of the emulsifier film at low temperature to preferential oil solubility at high temperature, will occur at a specific temperature unique to the particular emulsion and this can be determined experimentally. As a general rule, relatively stable oilin-water emulsions are obtained when their temperatures during storage and use are between 20 and 65_C below the PIT, presumably because the films are sufficiently hydrated. Mixtures of emulsifiers with identical HLBs produce emulsions with quite different PITs because additives and interactions between the components affect PIT but not HLB.
Microscopic selection for multiple phase emulsions
The better understanding of the mechanisms of stability incomplex dermatological emulsions stabilized by surfactants and amphiphiles has enabled the development of a rapid microscopic method for evaluation of potential emulsifiers. The method is based on the observation that good emulsifier blends that stabilize emulsions by the formation of multilayers of stable gel phase also swell spontaneously in water at ambient temperature and this process can be observed microscopically.
Mixtures that do not form gel phase or form metastable gels only after a heating and cooling cycle cannot be observed to swell spontaneously at ambient temperature.[4]
Emulsification Techniques
Emulsions are usually prepared by the application of mechanical energy produced by a wide range of agitation techniques. These disrupt droplets by the application of either shear forces in laminar flow or inertial forces in turbulent flow. Emulsifying devices ranging from simple hand mixers and stirrers to the use of propeller or turbine mixers, static mixers, colloid mills, homogenizers, and ultrasonic devices have been used.
Emulsifiers also have an important role in the process of emulsification. Surfactant emulsifiers reduce interfacial tensions during emulsification, making droplets easier to break up as well as reducing the tendency for recombination. Other emulsifiers such as the polymer macromolecules alter the hydrodynamic forces during the agitation process by their influence on rheological properties. Scale-up procedures from the laboratory to manufacture can introduce a number of problems due to the difficulties in matching the exact conditions of mixing, and, because of entrapment-of air, especially in emulsions of high consistency that have a yield value. Along with being inelegant, even traces of atmospheric air can cause decomposition in drugs or excipients susceptible to oxidation.
There are additional constraints when manufacturing parenteral emulsions that must be sterile and of fine particle size. Perfluorochemical and fat emulsions are usually prepared by homogenization at high temperature and pressure, as a large output of energy is required to produce droplet sizes considerably less than
Although heat sterilization is widely used, this places a severe test on the stability, and emulsions are sometimes prepared from sterile components under strict aseptic conditions and further sterilized by filtration.[15]
Processing variables
Differences in manufacturing techniques such as the rate of the heating and cooling cycle, the extent and order of mixing can cause variations in the consistency and rheology of the resulting emulsions. The initial particle size of the emulsion depends on the emulsifiers used, the emulsification equipment, the addition speed, and the phase volume. If the surfactant is placed in one of the phases prior to emulsification, it will migrate to the other to establish equilibrium. Thus, emulsification temperatures and cooling rates are important and the time of the mixing should be sufficient to allow the surfactant to migrate to and equilibrate at the interface throughout the process. Oilin-water emulsions are sometimes prepared by the phase inversion technique, where the aqueous phase is added to the oil phase to form a w/o emulsion that inverts to an o/w emulsion on addition of further amounts of water. This process is claimed to give finer emulsions.
Preparation techniques, in particular cooling rates and mixing procedures, have a marked effect on initial and final consistencies of emulsions prepared with nonionic emulsifying waxes. For example, ‘‘shock’’ cooling and limited mixing initially produces very mobile systems whereas slow cooling with adequate mixing produces semisolid emulsions. Mixing time, when the emulsifiers are in the molten state, influences the distribution of surfactant within the molten masses and bilayers and the relative lamellar order within thesystem. With ionic emulsifying waxes, different preparation techniques cause comparatively minor variations in the consistency of the final product. It was shown that differences are not due to the gel phase component of cationic ternary systems, but rather due to the variations in size of the crystalline alcohol that precipitates after manufacture. Systems formed by a rapid ‘‘shock’’ cooling method exhibited smaller but greater numbers of cetostearyl alcohol crystals and were thicker than similar ternary systems manufactured by a more lengthy procedure.[29]
