Dry powder inhaler

Abstract

The present invention provides a dry powder inhaler comprising one or more doses of powdered medication comprising a pharmaceutically active ingredient; an outlet, such as a mouthpiece or nose piece, through which a user may inhale a dose of medication; and an airway comprising a cyclone chamber. The cyclone chamber has one tangential powder inlet channel, two tangential bypass (clean) air inlet channels and an axial exit opening connected to the outlet of the inhaler. The cross-sectional area of the powder inlet channel is not equal to the cross-sectional areas of the bypass inlet channels. When a user inhales on the outlet to create an air flow, medication is entrained in the air flow and flows through the airway via the cyclone chamber, and out through the outlet.

Description

[0001] Dry powder inhaler

[0002] Technical Field of the invention

[0003] The present invention relates to a dry powder inhaler for oral or nasal delivery of medicament in powdered form.

[0004] Background to the invention

[0005] Dry powder inhalers (DPIs) are commonly used to deliver a pharmaceutically active ingredient to the lungs of a patient in order to treat local respiratory illnesses, such as asthma. There is also interest in using DPIs to deliver drugs systemically to the bloodstream via the lungs, for treating other medical conditions.

[0006] In order to be effective, the powder must be aerosolized (e.g. from a blister or capsule) and deagglomerated so that the fine particles are dispersed into the air flow and transported effectively to the user's lungs. Some inhalers have a cyclone (vortex) chamber for deagglomerating the powder. For example, WO 2010 / 086285 discloses an inhaler with a cyclone chamber that has inlet and outlet ports at opposite ends of a generally cylindrical chamber. Drug-laden air enters axially at the inlet port, passes through the chamber, and exits at the outlet port. The inhaler also has tangential bypass air inlets for the flow of clean (i.e. not drug-laden) air to create a cyclone in the chamber. The cyclone causes the drug-laden air entering through the inlet port to rotate and follow a generally helical path towards the outlet port. The drug-laden air experiences increased shear forces and differential velocities as it flows through the chamber, which further deagglomerates the particles, resulting in an increased fine particle fraction of the emitted dose. Similarly, WO 2005 / 037353 discloses an inhaler which, in one embodiment, has a cyclone chamber with an axial inlet for drug-laden air from a blister and a tangential bypass inlet for clean air. The clean air entering through the bypass inlet meets the drug-laden air flow at a tangent so as to create a cyclonic effect which assists deagglomeration. US 2004 / 0118399 describes an inhaler with a dispersion chamber that contains one or more beads. When the user inhales, air carrying the powder is drawn inlet into the dispersion chamber via a tangential inlet. The air causes the beads to circulate rapidly around in the chamber, which disperses the powder. The air and powder then move out of the chamber through a mouthpiece for inhalation by the user. The inhaler may have openings for bypass air that moves generally parallel to the cylindrical side-walls of the mouthpiece. This bypass air forms a sheath around the powder-laden air and helps to reduce deposition of powder.

[0007] FR 2 352556 discloses an inhaler with a cyclone chamber having a tangential bypass inlet and an opposite tangential powder inlet of equal width. Swirling flow is established in the cyclone by the clean air through the bypass inlet and the air through the powder inlet which has passed through a capsule and is loaded with powder. The cyclone chamber is said to result in a homogeneous mixing of the powder with the air, dissociation of the agglomerates by impacts between powder particles against the walls of the cyclone and selection of particles according to their size. Thus, only the finest particles are inhaled while the largest particles remain in the cyclone chamber.

[0008] Introducing powder through a tangential inlet at the edge of a cyclone chamber (as in the inhaler of FR 2 352 556) is believed to result in better deagglomeration than introducing it axially at the centre (as in the inhaler of WO 2010 / 086285) because the highest velocity / shear is at the periphery of the cyclone chamber. Moreover, introducing the powder at the periphery prevents the powder from bypassing the cyclonic flow and flowing straight along the axis to the outlet.

[0009] Inhalers with multiple cyclone chambers are known. For example, W02004 / 110538 discloses an inhaler with two or more cyclone chambers. While having more than one cyclone chamber in parallel may reduce the air flow resistance of the inhaler while deagglomerating larger amounts of powder, this has the disadvantage that the volume of the inhaler is increased to accommodate the cyclone chambers, which is undesirable for the user. Alternatively, the chambers must be quite small in order not to increase the volume of the inhaler, but this results in less effective deagglomeration. Powders that require large volumes for each dose and / or which are cohesive present particular challenges, both in evacuating the powder from the blister and also in deagglomerating the powder as it flows through the device prior to entry into the patient's airway. Known inhalers, even those with deagglomeration mechanisms such as cyclone chambers, are not capable of aerosolizing these powders, at least across the wide range of inhalation flow rates that different patients are capable of. Thus, there remains a need for an inhaler that addresses these drawbacks, in particular that is capable of delivering a large volume of powder with a high fine particle dose across a range of inhalation flow rates.

[0010] Brief Description of the invention

[0011] The present inventors have identified an optimum configuration for the cyclone chamber. Accordingly, the present invention provides a dry powder inhaler comprising:

[0012] • one or more doses of powdered medication comprising a pharmaceutically active ingredient;

[0013] • an outlet, such as a mouthpiece or nose piece, through which a user may inhale a dose of medication;

[0014] • an airway having a single cyclone chamber, the cyclone chamber having:

[0015] • one tangential powder inlet channel;

[0016] • two tangential bypass (clean) air inlet channels; and

[0017] • an axial exit opening connected to the outlet of the inhaler; wherein, when a user inhales on the outlet to create an air flow, medication is entrained in the air flow and flows through the airway via the cyclone chamber, and out through the outlet, characterized in that the cross-sectional area of the powder inlet channel is not equal to the cross-sectional areas of the bypass inlet channels.

[0018] The present inventors have found that increasing the number of inlets, in particular the number of bypass inlets (i.e. inlets for clean air), results in better deagglomeration. However, increasing the number of bypass inlets also reduces both the airflow resistance of the inhaler (defined as the ratio of the square root of the differential pressure to the volumetric flow rate) and the proportion of flow through the powder inlet. This could result in the total flow rate through the inhaler being too high and the powder not being evacuated properly from its container (e.g. a blister). These effects could be in principle countered by reducing the width of the bypass inlets compared to the powder inlet. However, if the number of bypass inlets is greater than two, the bypass inlets would need to be rather narrow, which would increase cost and complexity of manufacturing the inhaler, due to the inherent difficulty in moulding narrow channels. Moreover, the inventors have found that an inhaler with a single cyclone chamber, one powder inlet and two bypass inlets provides good performance (e.g. good aerosolization and deagglomeration of the powder), whilst remaining compact in size. Furthermore, the performance of the cyclone chamber can be optimized for powders with different properties by adjusting the sizes of the powder and bypass inlet channels.

[0019] The term "one tangential powder inlet channel" means that the cyclone chamber has a single (i.e. exactly one) tangential powder inlet channel. It does not include cyclone chambers with more than one tangential powder inlet channel. The term "two tangential bypass inlet channels" means that the cyclone chamber has exactly two tangential bypass inlet channels. It does not include cyclone chambers with more than two tangential bypass inlet channels. In other words, the tangential inlet channels to the cyclone chamber consist of one powder inlet channel and two bypass inlet channels.

[0020] The term “tangential powder / bypass inlet channel" means that the channel directs the air flow in a substantially tangential direction into the cyclone chamber. Thus, the inlet channels are preferably tangential to the cyclone chamber, but they do not have to be exactly tangential.

[0021] Preferably the cyclone chamber has no axial clean air inlets situated in the base of the cyclone chamber.

[0022] The inhaler may comprise a container for the powdered medication, such as a blister, capsule or reservoir. The inhaler may have a passage that connects the container to the powder inlet channel of the cyclone chamber via an entrance from the passage to the powder inlet channel. The entrance to the powder inlet channel may be located generally towards the centre of the passage, rather than at the edge of the passage. For example, the entrance to the power inlet channel may be located in the central 50% of the passage, i.e. between 25% and 75% of the width (span) of the passage. For example, the entrance may be at a distance of about one third of the width of the passage from the edge.

[0023] The powder inlet channel and bypass inlet channel may have constant cross-sectional areas, at least in the region of each channel that is immediately upstream of (i.e. adjacent to) the cyclone chamber. Thus, the powder inlet channel and bypass inlet channel may have constant widths and constant heights, at least in these regions. The term "cross-sectional area of the powder inlet channel" refers to the smallest cross-sectional area of the powder inlet channel between the dose of medication and the cyclone chamber. Similarly, the term "cross-sectional areas of the bypass inlet channels" refers to the smallest cross-sectional area between the external air and the cyclone chamber of each of the bypass inlets individually (i.e. it does not refer to the total cross-sectional area of the two bypass channels together). The air flow resistance increases when the cross-sectional area decreases. The largest contribution to the air flow resistance is the size (i.e. the cross-sectional area) of smallest constriction.

[0024] The heights (i.e. the dimension parallel to the axis of the cyclone chamber) of the powder inlet channel and the bypass inlet channel are preferably equal, so that the ratio of the cross- sectional area of the powder inlet channel to the cross-sectional area of each bypass inlet channel corresponds to the ratio of their widths.

[0025] The powder inlet channel and the bypass inlet channels may be from 0.5mm to 5mm in width, preferably from 1.0 to 4.0mm, such as about 1.5mm, 2.0mm, 2.5mm, 3.0mm or 3.5mm.

[0026] The two bypass inlet channels preferably are of equal width.

[0027] The ratio of the width of the powder inlet channel to the width of the bypass inlet channels (Ri) may be from about 1:5 to 5:1, such as 1:3 to 3:1 or 1:2 to 2:1, provided that Ri is not 1:1. In other words, 0.2 < Ri < 5.0, such as 0.33 < Ri < 3.0, or 0.5 < Ri < 2.0, provided that Ri 1.0.

[0028] In one embodiment, the width of the bypass inlet channels is less than the width of the powder inlet channel. For example, Ri may be given by 1.0 < Ri < 2.0, preferably 1.1 < Ri < 1.7, more preferably 1.2 < Ri < 1.5. The inventors have found that for powders that have relatively poor flowability (e.g. powders with a Carr's index of greater than about 25, such as greater than 30), the highest fine particle dose at low flow rates is achieved when the bypass inlets are narrower than the powder inlet.

[0029] In another embodiment, the width of the bypass inlet channels is greater than the width of the powder inlet channel. For example, Ri may be given by 0.2 < Ri < 1.0, preferably 0.25 < Ri < 0.75, more preferably 0.25 < Ri < 0.5.

[0030] The flow rates through the powder and bypass inlet channels depend on the air flow resistances of the powder flow path and the bypass flow path. If the resistances are unequal, then the flow rates are also unequal. The air flow resistances of the inlet channels depend on their cross-sectional areas. The powder flow path has additional resistance resulting from the powder container (e.g. a blister or capsule) and the passage between the container and the powder inlet channel (which may include e.g. a piercing element), whereas the bypass air may flow into the cyclone chamber through the bypass inlets without having to pass through any other components inside the housing of the inhaler. This additional resistance may be minimized by maximizing the size of the powder container, the openings through which air enters and leaves the powder container, and the passage. The largest contribution to the additional air flow resistance is the size (i.e. the cross-sectional area) of the smallest constriction in the powder flow path. In particular, if the smallest constriction in the powder flow path is large in comparison to the size of the powder inlet channel, then the additional air flow resistance of the powder flow path is relatively small. Consequently, the flow rate into the cyclone chamber through the powder inlet will be determined mainly by the cross- sectional area of the powder inlet channel.

[0031] The inventors have found that the overall air flow resistance of an inhaler with a cyclone chamber having a tangential powder inlet and two tangential bypass inlets is lowest when the air flow is split evenly between the inlets. Thus, the highest flow rate is achieved for a given pressure drop when the flow rates through the inlets are equal. The highest flow rate results in the highest particle velocities inside the cyclone chamber, and hence good deagglomeration. Moreover, the inventors have found that when the air flow rates through the inlets are unequal, the centre of the vortexed air flow (i.e. the eye of the cyclone) is not located at the centre of the cyclone chamber. This could result in the finest powder being hindered from leaving the cyclone chamber because the centre of the vortex does not correspond to the exit opening. Thus, very unequal airflows could lead to sub-optimal deagglomeration. Consequently, it is preferable that the additional airflow resistance in the powder airway is minimized by maximizing the size of the powder container, the openings through which air enters and leaves the powder container, and the passage. The ratio of the cross-sectional area of the smallest constriction in the container or the passage to the cross- sectional area of the smallest constriction in the powder inlet channel may be greater than 2. The passage may have a smallest cross-sectional area of at least 50mm2, preferably at least 75mm2, or at least 100mm2. The air entry opening and the powder exit opening in the container may each have an area of at least 30mm2, preferably at least 40mm2, or 50mm2. The smallest cross-sectional area of the container through which the air flows may be at least 50mm2, preferably at least 75mm2, such as about 100mm2.

[0032] Preferably the inlets are spaced approximately equiangularly, for example at angular spacings of between 90° and 150°, preferably between 105° and 135°, or between 115° and 125°, such as about 120°. The optimum configuration may not be exactly equiangular. If the air flow resistances of the powder inlet and the bypass inlets (which depend on the width of the inlets, and for the powder inlet, the upstream resistance of the container and passage) are unequal, then the air flow rates through the inlets will also be unequal. This can result in the centre of the vortexed airflow (i.e. the eye of the cyclone) not being located at the centre of the cyclone chamber. A small offset relative to the centre of the cyclone chamber is usually acceptable since the centre of the vortexed air flow is still within the exit opening. However, a larger displacement could hinder the finest powder particles at the centre of the vortex from leaving the cyclone chamber through the exit opening. This can be counteracted by adjusting the angular spacing of the inlets to compensate for the different flow rates. For example, if the bypass inlets are significantly larger than the powder inlet, the angular spacing between the bypass inlets could be increased above 120° (and the angular spacings between the powder inlet and each bypass inlet correspondingly decreased). This can balance the air flows so that the centre of the vortexed air flow is close to the centre of the cyclone chamber. Conversely, if the powder inlet is significantly larger than the bypass inlets, the angular spacing between the bypass inlets could be decreased below 120°. In any case, the two bypass inlets are preferably of equal width so that the flows through them are equal (if the flows through the bypass inlets were different, there would be an additional source of asymmetry that would affect the vortexed air flow).

[0033] The axial exit opening is preferably located centrally, i.e. the cyclone chamber and the axial exit opening are concentric. However, another approach to address a non-central vortexed air flow is for the exit opening to be located off-centre, so that it corresponds to the centre of the vortex instead of the centre of the cyclone chamber.

[0034] The diameter of the cyclone chamber may be from 10mm to 40mm, such as 12mm to 30mm, for example from 15mm to 20mm. The diameter of the axial exit opening may be from 5mm to 15mm, such as 7mm to 12mm, for example 8mm to 10mm.

[0035] Preferably the ratio of the diameter of the cyclone chamber to the diameter of the exit opening is (R2) is given by 1.0 < R2< 3.0, preferably 1.2 < R2< 2.5. In one embodiment, R2is given by 1.5 < R2< 2.0, more preferably 1.6 < R2< 1.9. In another embodiment, R2is given by 2.0 < R2< 2.5, more preferably 2.1 < R2< 2.3.

[0036] Brief Description of the Figures

[0037] Figure 1 shows an inhaler.

[0038] Figures 2 shows the inhaler removed from the cover.

[0039] Figure 3A is an expanded view showing the components of the inhaler.

[0040] Figure 3B shows a cross-section through the inhaler.

[0041] Figure 4 shows the central part of the inhaler with a cyclone chamber.

[0042] Figure 5 shows another central part of the inhaler with a second embodiment of a cyclone chamber.

[0043] Figure 6 shows a central part of the inhaler with a cyclone chamber not according to the invention. Detailed Description of the invention

[0044] Figure 1 shows an inhaler according to the invention. The inhaler 1 has a housing 3 and a mouthpiece 4. A cover 2 holds and protects the inhaler. In particular, the cover has an extension 7 that extends over the mouthpiece 4, thereby preventing foreign material from entering the mouthpiece before use. The inhaler has a pair of grips 5 located on either side of the housing, and the cover has a pair of grips 6, for removing the inhaler from the cover.

[0045] To prepare the inhaler for use, the user holds the grips 5 of the housing 3 between the finger and thumb of one hand, and the grips 6 on the cover 2 between the finger and thumb of their other hand. The inhaler 1 is pulled out of the cover which uncovers the mouthpiece as shown in Figure 2. This action also causes a blister containing the medication to be pierced, thereby avoiding the need for any further user steps (such as pressing a button or lever to cause piercing) before use. The user then inhales on the mouthpiece to receive the medication.

[0046] Figure 3A is an expanded view of the components of the inhaler 1. Figure 3B shows a crosssection through the inhaler. The inhaler has an upper housing part 10 with the mouthpiece 4, a central part 30, a mixing element 35, a blister 40 and a lower housing part 20. The central part 30 has two pairs of piercing elements 31 (for example of the type described in WO 2014 / 114916) on its lower surface.

[0047] The central part 30 is fixed (e.g. clipped or welded) in the upper housing part 10 so that the internal surface of the upper housing part and the upper surface of the central part together define an airway that fluidically connects the blister (once it has been pierced) to the mouthpiece 4 via a passage 32 and a cyclone (deagglomeration) chamber 33. The mixing element 35 is located inside the cyclone chamber 33. An exit opening with a mesh 12 is formed in the upper housing part; this connects the cyclone chamber to the mouthpiece 4.

[0048] The blister 40 has a lid 41 and a base 42 with a rim 43 which fits into slots 21 on either side of the lower housing part 20, thereby holding the blister in place in the lower housing part.

[0049] The upper 10 and lower 20 housing parts are movable relative to each other. When the inhaler is in the cover 2, the upper and lower housing parts are held in an initial position in which the piercing elements 31 are held spaced apart from the lid 41 of the blister 40. The cover prevents the upper and lower housing parts from accidentally being pushed together before use. When the inhaler is removed from the cover, the cover interacts with the upper and lower housing parts, so that they are pushed together into an actuated position, in which the piercing elements 31 pierce the lid and enter the blister. The mechanism for this consists of a pair of cams 8 in the form of pegs on the inside of the cover and two sloping cam surfaces 22, one in each side of the lower housing part 20. As the inhaler 1 is removed from the cover 2, the cams 8 slide along the cam surfaces 22, pushing the lower housing part 20 upwards into the upper housing part 10. One pair of piercing elements 31 creates an air entry opening in the lid 41 of the blister 40, and the other pair creates a powder exit opening.

[0050] Once the inhaler has been removed from the cover, the user inhales on the mouthpiece 4. This creates an air flow into the blister through the air entry opening which aerosolizes the powder, and then carries it out through the powder exit opening into the passage 32 and then into the cyclone chamber 33.

[0051] The central part 30 is shown from above in Figure 4. The powder-containing air flows from the passage 32 through a powder inlet channel 36 into the cyclone chamber 33. External (i.e. powder-free) air also enters the cyclone chamber through two bypass inlet channels 37, 38. The powder inlet channel 36 and the bypass inlet channels 37, 38 are located 120° apart from each other and are tangential to the wall of the cyclone chamber, so that the air flow into the chamber creates a vortex. The entrance to the powder inlet channel 36 is located generally towards, but not exactly at, the centre of the passage 32. In this case, it is located somewhat off-centre, at about one third of the width of the passage from the edge.

[0052] Figure 5 shows a second embodiment of a central part with a cyclone chamber according to the invention. As before, the cyclone chamber 33 has a powder inlet channel 36 and two bypass inlet channels 37, 38. The powder inlet channel 36 and the bypass inlet channels 37, 38 are located 120° apart from each other and are tangential to the wall of the cyclone chamber, so that the air flow into the chamber creates a vortex. However, the entrance to the powder inlet channel 36 is located at the edge of the passage 32 rather than generally centrally (as in Figure 4). The inventors have found that the performance of the cyclone chamber can be optimized for different powders by adjusting the sizes of the powder and bypass inlet channels.

[0053] The vortex, in combination with the mixing element 35 (not shown in Figures 4 and 5) which is freely movable within the cyclone chamber, breaks the powder up into fine particles. The inlets are tangential so that the powder is exposed to the maximum shear, in order to achieve the greatest deagglomeration. Having a tangential powder inlet (instead of an axial powder inlet as in WO 2010 / 086285) has the advantage that the particles, especially the largest ones, impact against the wall of the cyclone as a result of their momentum as they enter the cyclone chamber, which helps to disaggregate them. This was confirmed by using high-speed video photography to visualize the flow of powder in a cyclone chamber with a tangential powder inlet and two tangential bypass inlets. Large particles were observed to impact against the wall of the cyclone chamber multiple times whilst being fragmented into smaller particles which exited the cyclone chamber. This effect was also observed in computational fluid dynamics simulations.

[0054] The diameter of the mixing element is larger than the widths of the powder inlet channel 36 and the bypass inlet channels 37, 38 at the points where they enter the cyclone chamber, so that the mixing element cannot enter the inlet channels. The height of the cyclone chamber is relatively short and is less than its diameter. The cyclone chamber must be tall enough that the mixing element can move freely. However, if the cyclone chamber were relatively tall (compared to the diameter of the mixing element), the mixing element would mostly move around the upper part of the cyclone chamber. Consequently, it would not be very effective at dislodging powder from the lower part of the cyclone chamber.

[0055] The aerosolized fine powder then leaves the cyclone chamber 33 via the mesh at the exit opening 11 and flows out through the mouthpiece 4 to the user's lungs. The mesh prevents any large lumps of powder from leaving the cyclone chamber. The mesh also reduces the vorticity of the air. Without this rectification of the air flow, the powder would spread outwards after it leaves the inhaler, resulting in deposition in the patient's mouth and throat. The inhaler described above has a blister that contains the dry powder medication, which is pierced by the piercing elements. However, the invention also encompasses other containers, such as a capsule, and other opening mechanisms, such as peeling a lid off, pulling two halves of a capsule apart or piercing a capsule by means of needles. Regardless of which type of opening mechanism is used, the inhaler may be automatically actuated by the action of removing it from the cover.

[0056] The inhaler may have a single dose of medication, and may be pre-loaded with a blister or capsule. The inhaler may be re-usable, so that a new blister or capsule in inserted each time it is to be used. Alternatively, the inhaler may be a multi-dose device and contain a number of doses, for example 30 or 60 doses in a blister strip, dose disk or a reservoir, along with a suitable mechanism for preparing each dose.

[0057] The blister, capsule or other container contains a dry powder medication for inhalation. The medication comprises a pharmaceutically active ingredient and may also comprise one or more pharmaceutically acceptable excipients. The excipient may be present in relatively small amounts, so that the powder may comprise at least 50%, at least 70%, at least 90% or more of the active ingredient. The amount of the powdered medication in the blister, capsule or other container may be 5-300 mg, preferably 20-200 mg, more preferably 30-150 mg, even more preferably 40-100 mg. For example, there may be about 50, 60, 70, 80 or 90 mg of powder in the blister, capsule or other container.

[0058] Each dose may be delivered from the inhaler in a single inhalation or in two or more inhalations. The width of the inlets, in particular the powder inlet channel, can be chosen in order to achieve a desired rate of flow of powder from the blister. Thus, the inlet channels can be narrower if the dose is intended to be delivered over two inhalations instead of one inhalation.

[0059] The medication may be capable of treating or preventing a thromboembolic event. The pharmaceutically active ingredient in the medication may be an antiplatelet drug. For example, the pharmaceutically active ingredient may be a non-steroidal anti-inflammatory drug (NSAID). Preferably, the pharmaceutically active ingredient is a salicylate (a salt or ester of salicylic acid), most preferably acetylsalicylic acid or a pharmaceutically acceptable salt thereof. The pharmaceutically active ingredient may be another type of NSAID. For example, the pharmaceutically active ingredient may be Celecoxib (Celebrex), Dexdetoprofen (Keral), Diclofenac (Voltaren, Cataflam, Voltaren-XR), Diflunisal (Dolobid), Etodolac (Lodine, Lodine XL), Etoricoxib (Algix), Fenoprofen (Fenopron, Nalfron), Firocoxib (Equioxx, Previcox), Flurbiprofen (Urbifen, Ansaid, Flurwood, Proben), Ibuprofen (Advil, Brufen, Motrin, Nurofen, Medipren, Nuprin), Indomethacin (Indocin, Indocin SR, Indocin IV), Ketoprofen (Actron, Orudis, Oruvail, Ketoflam), Ketorolac (Toradol, Sprix, Toradol IV / IM, Toradol IM), Licofelone (under development), Lomoxicam (Xefo), Loxoprofen (Loxonin, Loxomac, OxOrjo), Lumiracoxib (Prexige), Meclofenamic acid (Meclomen), Mefenamic acid (Ponstel), Meloxicam (Movalis, Mel ox, Recoxa, Mobic), Nabumetone (Relafen), Naproxen (Aleve, Anaprox, Midol Extended Relief, Naprosyn, Naprelan), Nimesulide (Sulide, Nimalox, Mesulid), Oxaporozin (Daypro, Dayrun, Duraprox), Parecoxib (Dynastat), Piroxicam (Feldene), Rofecoxib (Vioxx, Ceoxx, Ceeoxx), Salsalate (Mono-Gesic, Salflex, Disalcid, Salsitab), Sulindac (Clinoril), Tenoxicam (Mobi flex), Tolfenamic acid (Clotam Rapid, Tufnil), or Valdecoxib (Bextra). The pharmaceutically active ingredient may be an alternative to an NSAID. Such alternatives include P2Y12 inhibitors. Examples of P2Y12 inhibitors include Plavix (clopidogrel), ticlopidine, ticagrelor, prasugrel, and cangrelor. Other pharmaceutically active ingredients may include COX-2 inhibitors, and Nattokinase (an enzyme (EC 3.4.21.62, extracted and purified from a Japanese food called natto). The medication may comprise both acetylsalicylic acid, or a pharmaceutically acceptable salt thereof, and a P2Y12 inhibitor.

[0060] The pharmaceutically active ingredient may alternatively be a bronchodilator, such as a beta- 2 agonist or anticholinergic for the treatment of an asthma exacerbation; adrenaline and / or atropine for the treatment of cardiac failure, cardiac dysfunction, cardiac arrest, anaphylaxis, drug overdose or the like; glucose and / or glucagon for the treatment of hypoglycaemia, diabetes induced coma or the like; benzodiazepine, phenytoin or anti-seizure medications for the treatment of seizure; di hydroergotamine for the treatment of migraine; naloxone for treating an opioid overdose; insulin for managing blood sugar level or the like.

[0061] The medication may include one or more agents for inducing an immune response, e.g. a vaccine, such as a measles vaccine, a Hepatitis B vaccine, or an influenza vaccine. The medication may include a natural or synthetic cannabinoid, such as Cannabidiol (CBD) or Tetrahydrocannabinol (THC).

[0062] Examples

[0063] Example 1: single vs double bypass inlet

[0064] Figure 6 shows a central part with a cyclone chamber similar to that of FR 2 352 556 (i.e., not according to the invention). The cyclone chamber 33 has a powder inlet channel 36 and one bypass inlet channel 37. The powder inlet channel 36 and the bypass inlet channel 37 are located 180° apart and are tangential to the wall of the cyclone chamber, so that the air flow into the chamber creates a vortex.

[0065] The performance of an inhaler with a double bypass cyclone chamber (Figure 5) was compared with that of an inhaler with a single bypass cyclone chamber (Figure 6). The powder inlet channel and the bypass inlet channel of the single bypass inlet cyclone chamber (Figure 6) had almost the same width as the powder inlet channel (3.65mm and 3.6mm respectively), whereas the bypass inlet channels of the double bypass inlet cyclone chamber (Figure 5) were narrower (2.5mm) than the powder inlet channel (3.65mm). This is to ensure that the air flow resistance of the two cyclone chambers, and hence of the two inhalers is the same. This means that, when a chosen pressure drop is applied to each inhaler for a chosen time, the air flow rate and total volume of air that passes through each inhaler is the same.

[0066] Apart from the single / double bypass inlets, the inhalers were identical. The cyclone chambers were 20mm in diameter and had a central (axial) circular exit opening to the mouthpiece 10mm in diameter. The exit opening contained a mesh in the form of a square grid with a spacing of 0.75mm. The cyclone chambers contained a solid spherical ball with diameter of 6mm as the mixing element. Blisters were filled with 85mg of a powder consisting of micronized acetylsalicylic acid powder (99.5% ASA) and magnesium stearate (0.5% MgSt). The bulk density (po) and tapped density (pt) of the powder were measured, and the Carr's Index of the powder (given by (pt- po) / Pt x 100) was calculated to be 30. The Carr's Index is a measure of the compressibility of the powder and is indicative of the flowability of the powder: a low Carr's Index indicates a flowable powder, whereas a high Carr's Index indicates poor flowability. A Fast Screening Impactor (FSI, Copley Scientific) was used to measure the fine particle dose (< 5pm) delivered from each inhaler. The FSI has a filter which captures emitted aerosol particles of less than 5pm in size. Simulated inhalation manouevres with flow rates of 30 and 60 litres per minute and a total volume of 2 litres were performed on each inhaler. The FPD was determined gravimetrically by weighing the filter before and after each inhalation. The results are shown in Table 1.

[0067] Table 1: FSI results for micronized powder

[0068] Table 1 demonstrates that the double bypass cyclone resulted in a higher FPD than the single bypass cyclone at both flow rates. Without wishing to be limited by theory, this is believed to be because the aerosolized powder gets a "kick" each time it passes an inlet. Having more inlets means that the powder gets more "kicks", which results in a more uniform cyclonic flow within the chamber. As a result, the powder stays in the cyclone chamber for longer before being drawn towards the centre of the cyclone chamber and out through the exit opening.

[0069] Example 2: position of powder inlet channel

[0070] The performance of two inhalers with a double bypass cyclone chamber were compared. One inhaler had a cyclone chamber with the entrance to the powder inlet channel located away from the edge of the passage (as in Figure 4) and the other had the entrance to the powder inlet channel at the edge of the passage (as in Figure 5). The cyclone chambers were 17.5mm in diameter and had a central (axial) circular exit opening to the mouthpiece with a diameter of 10mm. The exit opening contained a mesh in the form of a square grid with a spacing of 0.75mm. The cyclone chambers contained a solid spherical ball with diameter of 6mm as the mixing element. Blisters were filled with 150mg of the powder used in example 1. Simulated inhalation manoeuvres with flow rates of 30 and 60 litres per minute and a total volume of 2 litres were performed on each inhaler. The FPD was determined using the FSI as in Example 1. The results are shown in Table 2.

[0071] Table 2: FSI results for micronized powder

[0072] The performance of two further inhalers with a double bypass cyclone chamber were compared. One inhaler had a cyclone chamber with a generally central entrance to the powder inlet channel (as in Figure 4) and the other had the powder inlet channel entrance at the edge of the passage (as in Figure 5). The cyclone chambers were 17.5mm in diameter and had a central (axial) circular exit opening to the mouthpiece with a diameter of 7mm. The exit opening contained a mesh in the form of a grid formed from circular rings with a spacing of 0.85mm and radial ribs, as described in our co-pending application PCT / EP2025 / 065843. The cyclone chambers contained a mixing element in the form of three orthogonal circular discs with diameter of 6mm and a common centre, as shown in Figures 3A and 3B and described in our co-pending application PCT / EP2025 / 065842. Blisters were filled with 85mg of a sprayed- dried powder (A) consisting of 80% acetylsalicylic acid powder and 20% leucine. Simulated inhalation manoeuvres were performed on each inhaler with a total volume of 2 litres and pressure drops of lkPa and 4 kPa. The FPD was determined using the FSI as in Example 1. The results are shown in Table 3.

[0073] Table 3: FSI results for spray-dried powder A Tables 2 and 3 demonstrate that both cyclone chambers resulted in good FPDs, but that the cyclone chambers with the generally central entrance (as in Figure 4) resulted in slightly higher FPDs than the cyclone chamber with the edge entrance (as in Figure 5) at both flow rates and for both types of powder. Without wishing to be limited by theory, this is believed to be because having a generally central entrance helps to evacuate powder from the blister, particularly at low flow rates.

[0074] Example 3: ratio of the cyclone chamber diameter to the exit opening diameter

[0075] The performance of three inhalers with a double bypass cyclone chamber were compared. All of the inhalers had a cyclone chamber with a powder inlet channel that enters the cyclone chamber at the edge of the passage (as in Figure 5). The diameter of the cyclone chamber differed in each inhaler (15mm, 17.5mm and 20mm), while the diameter of the exit opening was fixed (10mm), so that the ratio of the diameters varied from 1.5:1 to 2:1. The exit opening contained a mesh in the form of a square grid with a spacing of 0.75mm. The cyclone chambers contained a solid spherical ball with diameter of 6mm as the mixing element. Blisters were filled with 150mg of the powder used in example 1. Simulated inhalation manoeuvres with flow rates of 30 and 60 litres per minute and a total volume of 2 litres were performed on each inhaler. The FPD was determined using the FSI as in Example 1. The results are shown in Table 4.

[0076] Table 4: FSI results for micronized powder

[0077] The FPD was observed to vary with the diameter of the cyclone. Without wishing to be limited by theory, this is believed to be because the smallest particles tend to be in the centre of the vortex, whereas larger particles are closer to the edge of the cyclone. As a result, a small exit opening relative to the size of the cyclone chamber means that only the very fine particles are able to pass out though the exit. However, a small exit opening also increases the air flow resistance, and hence decreases the flow rate through the inhaler. The optimal size for the exit opening is a compromise between these two effects. The size of the cyclone chamber is limited by the need to keep the overall size of the inhaler reasonably small.

[0078] Example 4: width of the powder and bypass inlet channels with micronized powder

[0079] The performance of nine inhalers with a double bypass cyclone chamber were compared. All of the inhalers had a cyclone chamber with the entrance from the powder inlet channel at the edge of the passage (as in Figure 5). The cyclone chambers were 17.5mm in diameter and had a central (axial) circular exit opening to the mouthpiece with a diameter of 10mm. The exit opening contained a mesh in the form of a grid formed from circular rings with a spacing of 0.85mm and radial ribs, as described in our co-pending application PCT / EP2025 / 065843. The width of the powder inlet channel (3.5, 3.0, 2.5, 2.0 and 1.5mm) and of the bypass channels (2.5, 2.0, 1.5 and 1.0mm) differed in each inhaler. In each inhaler, the two bypass channels had the same width. The cyclone chambers contained a mixing element in the form of three orthogonal circular discs with diameter of 6mm and a common centre, as shown in Figures 3A and 3B and described in our co-pending application PCT / EP2025 / 0658432.

[0080] Initially, the air flow resistance of each inhaler was determined. The two inhalers with the narrowest bypass channels (1.0mm) were rejected because the air flow resistance was too high. If the air flow resistance is too high (e.g., if a pressure difference of greater than about 4kPa is required to generate a flow rate of 90 L / min), patients who are only able to inhale relatively weakly would not be able to evacuate the powder from the blister. Another inhaler with a relatively narrow inlet (2.0mm) and bypass (1.5mm) was also found to have an air flow resistance that was too high. The inhaler with the widest bypass channels (2.5mm) was rejected because the air flow resistance was too low. The large bypass channels have the effect that less air flows through the blister, so that the powder might not be evacuated sufficiently. Blisters were filled with 150mg of the powder used in example 1. Simulated inhalation manoeuvres with flow rates of 30 and 60 litres per minute and a total volume of 2 litres were performed on the remaining five inhalers. The FPD was determined using the FSI as in Example 1. The results are shown in Table 5.

[0081] Table 5: FSI results for micronized powder

[0082] Table 5 demonstrates that the remaining five cyclone chambers resulted in good FPDs at both flow rates. The cyclone chamber with a powder inlet channel width of 2.5mm and bypass channel width of 2.0mm resulted in the highest FPD at the lower flow rate of 30 litres per minute. Increasing the FPD at low flow rates, such as 30 L / min, is particularly important because the FPD at higher flow rates, such as 60 L / min, is already quite high. The FPD at the higher flow rate was observed to increase as Ri decreased, i.e. as the width of the powder inlet channel decreased relative to the width of the bypass channels.

[0083] Table 5 also shows the difference between the FPDs at 60 L / min and 30L / min for each cyclone chamber, and the ratio between the widths of the powder inlet and bypass channels. The difference between the FPDs generally increases as the ratio of the widths decreases, i.e. as the inlet becomes narrower and the / or the bypass becomes wider. In other words, the dependency of the FPD on the flow rate was reduced for higher ratios. A weaker dependency of the FPD on the flow rate may be advantageous because different patients may inhale at very different flow rates.

[0084] The particle size distribution (PSD) of the powder was determined by laser diffraction (HELOS BR, Sympatec GmbH). The powder was introduced by the ASPIROS feeder into the RODOS dry dispersion unit and dispersed at two pressures; 0.2 bar and 3 bar, and measurements (three replicates) were conducted using R5 and R1 lenses respectively. The results were as follows: 0.2 bar: Dio=1.63, Dso=7.93, D90=39.06; 3 bar: Dio=0.50, D50=1.40, D90=3.24. The fact that the particle sizes are quite different at 0.2 bar and 3.0 bar shows that that the powder has not been fully dispersed at the lower pressure, i.e. it is relatively hard to disperse. This is consistent with the observation in the FSI data that the FPD at 60L / min increases as Ri decreases. When the bypass inlets are relatively large, a greater proportion of the air flow is available to deagglomerate the powder in the cyclone chamber.

[0085] Example 5: width of the powder and bypass inlet channels with micronized powder

[0086] The performance of two inhalers with a double bypass cyclone chamber were compared. Both of the inhalers had a cyclone chamber with the entrance to the powder inlet channel away from the edge of the passage (as in Figure 4). The cyclone chambers were 17.5mm in diameter and had a central (axial) circular exit opening to the mouthpiece with a diameter of 10mm. The exit opening contained a mesh in the form of a grid formed from circular rings with a spacing of 0.85mm and radial ribs, as described in our co-pending application PCT / EP2025 / 065843. The width of the powder inlet channel was 2.5mm in one inhaler and 2.0mm in the other. In both inhalers, the two bypass channels had the same width of 2.0mm. The cyclone chambers contained a mixing element in the form of three orthogonal circular discs with diameter of 6mm and a common centre, as shown in Figures 3A and 3B and described in our co-pending application PCT / EP2025 / 065842.

[0087] Blisters were filled with 150mg of the powder used in example 1. Simulated inhalation manoeuvres with flow rates of 30 and 60 litres per minute and a total volume of 2 litres were performed on each inhaler. The FPD was determined using the FSI as in Example 1. The results are shown in Table 6.

[0088] Table 6: FSI results for micronized powder

[0089] Table 6 demonstrates that both cyclone chambers resulted in good FPDs at both flow rates. The inhaler with equal powder and bypass inlet widths resulted in a slightly higher FPD at each flow rate. The difference between the FPDs at the two flow rates was smaller for the inhaler with the larger powder inlet (i.e. the higher ratio), i.e. it was less flow rate dependent.

[0090] Example 6: width of the powder and bypass inlet channels with spray-dried powder

[0091] The performance of two inhalers with a double bypass cyclone chamber were compared. Both inhalers had a cyclone chamber with the entrance to the powder inlet channel away from the edge of the passage (as in Figure 5). The cyclone chambers were 17.5mm in diameter and had a central (axial) circular exit opening to the mouthpiece with a diameter of 10mm. The exit opening contained a mesh in the form of a grid formed from circular rings with a spacing of 0.85mm and radial ribs, as described in our co-pending application PCT / EP2025 / 065843. The widths of the powder inlet channels were 2.0 or 1.5mm, and the widths of the bypass channels were 2.0 or 3.5mm (the two bypass channels in each inhaler had the same width). The cyclone chambers contained a mixing element in the form of three orthogonal circular discs with diameter of 6mm and a common centre, as shown in Figures 3A and 3B and described in our co-pending application PCT / EP2025 / 065842. Blisters were filled with 85mg of a sprayed-dried powder (B) consisting of 80% acetylsalicylic acid powder and 20% leucine prepared under different conditions from that used in Example 2. The spray-dried powder was more free-flowing than the micronized powder used in Examples 4 and 5. Simulated inhalation manoeuvres were performed on each inhaler with a total volume of 2 litres and pressure drops of lkPa and 4 kPa. The FPD was determined using the FSI as in Example 1. The results are shown in Table 7.

[0092] Table 7: FSI results for spray-dried powder B

[0093] Table 7 demonstrates that the cyclone chamber with the narrower powder inlet channel and wider bypass channels performed better with the spray-dried powder than the cyclone chamber with inlet and bypass channels of equal width. In both cases, the difference between the FPDs at 60 and 30 L / min was smaller than with the micronized powder.

[0094] Example 7: width of the powder and bypass inlet channels with spray-dried powder

[0095] The performance of three further inhalers with a double bypass cyclone chamber were compared. All of the inhalers had a cyclone chamber with the entrance to the powder inlet channel away from the edge of the passage (as in Figure 5). The cyclone chambers were 17.5mm in diameter and had a central (axial) circular exit opening to the mouthpiece with a diameter of 8mm. The exit opening contained a mesh in the form of a grid formed from circular rings with a spacing of 0.85mm and radial ribs, as described in our co-pending application PCT / EP2025 / 065843. The widths of the powder inlet channels were 1.0 or 1.5mm, and the widths of the bypass channels were 3.5 or 4.0mm (the two bypass channels in each inhaler had the same width). The cyclone chambers contained a mixing element in the form of three orthogonal circular discs with diameter of 6mm and a common centre, as shown in Figures 3A and 3B and described in our co-pending application PCT / EP2025 / 065842. Blisters were filled with 85mg of the powder used in Example 6. Simulated inhalation manoeuvres were performed on each inhaler with a total volume of 2 litres and pressure drops of lkPa and 4 kPa. The FPD was determined using the FSI as in Example 1. The results are shown in Table 8.

[0096] Table 8: FSI results for spray-dried powder B

[0097] Table 8 demonstrates that the cyclone chamber with a narrower powder inlet channel and wider bypass channels performed better with the spray-dried powder than the other cyclone chambers. Again, the differences between the FPDs at 60 and 30 L / min were smaller than with the micronized powder.

[0098] The cyclone chamber with powder inlet and bypass inlet channel widths of 1.5mm and 3.5mm was tested again using a blister containing 170mg of a further spray-dried powder (C). The powder had the same composition but was produced under different spray-drying conditions.

[0099] Table 9: FSI results for different spray-dried powder C

[0100] Table 9 shows that the FPDs were significantly higher at both flow rates than in Table 8. This is largely because the amount of powder in the blister was doubled (170mg rather than 85mg). Part of the difference may also be a result of the fact that the powder was produced under different spray-drying conditions.

[0101] Example 8: width of the powder and bypass inlet channels with further spray-dried powders

[0102] The performance of three further inhalers with a double bypass cyclone chamber were compared. All of the inhalers had a cyclone chamber with the entrance to the powder inlet channel away from the edge of the passage (as in Figure 4). The cyclone chambers were 17.5mm in diameter and had a central (axial) circular exit opening to the mouthpiece with a diameter of 8mm. The exit opening contained a mesh in the form of a grid formed from circular rings with a spacing of 1.15mm and radial ribs, as described in our co-pending application PCT / EP2025 / 065843. The widths of the powder inlet channel and bypass inlet channels were respectively: (i) 3.5mm, 1.5mm, (ii) 2.0 mm, 2.0mm, and (iii) 1.5mm, 3.5mm. The two bypass channels in each inhaler had the same width. The cyclone chambers contained a mixing element in the form of three orthogonal circular discs with diameter of 6mm and a common centre, as shown in Figures 3A and 3B and described in our co-pending application PCT / EP2025 / 065842. The three inhalers had approximately the same airflow resistance, namely 0.028, 0.029 and 0.024 VkPa.min / L respectively.

[0103] Three different spray-dried powders were used. Powder D consisted of 80% trehalose and 20% leucine; Powder E consisted of 50% trehalose and 50% leucine; and Powder F consisted of 90% trehalose and 10% trileucine. The powders were produced by spray-drying aqueous feedstock solutions using a Niro Mobile Minor spray dryer with two-fluid atomisation to generate powders suitable for inhalation i.e. Dv,9o particle size <6pm. The processing parameters are set out in Table 10.

[0104] Table 10: Spray-drying process parameter

[0105] The particle size distribution of the powders was determined by laser diffraction (HELOS BR, Sympatec GmbH). Each powder was introduced by the ASPIROS feeder into the RODOS dry dispersion unit and dispersed at two pressures; 0.2 bar and 3 bar, for measurement. All measurements were conducted using an R1 lens unless larger particle sizes were detected, in which case a suitable alternative e.g. R5 lens, was adopted. Measurements are based on three separate replicates and are shown in Table 11.

[0106] Table 11: Particle size distribution measurements for spray-dried powders D, E and F

[0107] The difference between the particle sizes measured at 0.2 bar and 3.0 bar provides an indication of how easy or difficult the powders are to disperse. A small difference (as for powder D) indicates that the powder has already been dispersed at the lower pressure, i.e. it is relatively easy to disperse. A larger difference (as for powder F) indicates that the powder has not been fully dispersed at the lower pressure, i.e. it is relatively hard to disperse. Blisters were filled with 82.5mg of each powder. Simulated inhalation manoeuvres were performed on each inhaler with a total volume of 2 litres and pressure drops of lkPa and 4 kPa. The FPD was determined using the FSI as in Example 1. The results are shown in Table 12. The Emitted dose (ED), i.e. the amount of powder evacuated from the blister and the Fine Particle Dose (FPD) are expressed as a percentage of the filled mass. The Fine Particle Fraction (FPF) is the FPD as a percentage of the ED.

[0108] Table 12: FSI results for spray-dried powders D, E and F at constant fill mass

[0109] Table 12 demonstrates that at the low pressure drop (i.e. low flow rate), only the inhaler with the large powder inlet (3.5-1.5) results in EDs > 80%. The other inhalers show a low ED, and consequently a low FPD, even though the FPF is relatively high (>66%). This shows that these inhalers are not able to fully evacuate the powders from the blister at low flow rates, which results in poor performance, i.e. a low FPD. In other words, at low flow rates, it is advantageous for a greater proportion of the air flow to go through the blister to evacuate the powder than into the cyclone to disperse the powder.

[0110] At the high pressure drop (i.e. high flow rate), all of the inhalers result in ED of > 75%, apart from powder D in the 1.5-3.5 inhaler, which still has a reasonably high ED of about 60%. This shows that all of the inhalers are able to evacuate the powder from the blister at high flow rates.

[0111] Powder D, which is relatively easy to disperse, has similar FPFs (60%, 63% and 64%) in each inhaler. This shows that even with the narrowest bypass inlets, there is sufficient energy in the cyclone to disperse the powder. The 3.5-1.5 and 2.0-2.0 inhalers therefore both result in good FPDs (> 50%), while the 1.5-3.5 inhaler has a lower FPD due to the lower ED. Powder F, which is relatively hard to disperse, has quite different FPFs (44%, 53% and 68%) in each inhaler, showing that the 3.5-1.5 inhaler does not provide sufficient energy in the cyclone (due to its relatively narrow bypass inlets) to fully disperse the powder. The 2.0-2.0 and 1.5-3.5 however both result in good FPDs. In fact, the 1.5-3.5 inhaler has the highest FPD, even low it has the lowest ED, because of the very effective dispersion in the cyclone chamber.

[0112] Powder E, which is intermediate between powders D and F in terms of dispersibility, performs similarly in each device. There is a trade-off between increasing ED (larger powder inlet) and increasing FPF (larger bypass inlets). For powder E, these effects largely counterbalance each other.

[0113] The experiments were repeated using a fixed fill volume of 0.45mL of powder instead of a fixed mass. Since the densities of the powders differ, the fill masses differ. The FSI results are shown in Table 13.

[0114] Table 13: FSI results for spray-dried powders D, E and F at constant fill volume

[0115] Table 13 demonstrates the same trends as Table 12. At the low pressure drop, only the 3.5- 1.5 inhaler results in an ED > 80%. The other inhalers show a low ED, and consequently a low FPD, even though the FPF is relatively high (>66%). This confirms that these inhalers are not able to fully evacuate the powder from the blister at low flow rates. At the high pressure drop, 1 all the inhalers result in ED of > 75%, apart from powder D in the 1.5-3.5 inhaler, which still has a reasonably high ED of 56%. This shows that all of the inhalers are able to evacuate the powder from the blister at high flow rates. Powder D, which is relatively easy to disperse, performs well in the 3.5-1.5 and 2.0-2.0 inhalers, but less well in the 1.5-3.5 inhaler due to the lower ED. Powder F, which is relatively hard to disperse, performs best in the 1.5-3.5 inhaler.

[0116] The observation that the constant mass and constant volume experiments show the same trends indicates that the density of the powder does not have a major effect on the performance.

[0117] Example 9: width of the inlet and bypass channels with a lactose-based powder

[0118] The three inhalers used in Example 8 were also tested using a fixed mass (82.5mg) of a powder consisting of a blend of 0.8 wt% micronized fluticasone furoate (FF) and 99.2% lactose carrier. The micronized FF particles were < 5pm in size, and the lactose was in the form of large carrier particles with Dso of 50-70 pm and D90 of 130- 135pm. The emitted dose, fine particle fraction and fine particle dose of the FF are shown in Table 14.

[0119] Table 14 FSI results for the lactose-based powder at constant fill volume

[0120] Table 14 shows that at the low pressure drop, only the 3.5-1.5 inhaler results in an ED > 70%. The other inhalers show a low ED, and consequently a low FPD. This confirms that these inhalers are not able to fully evacuate the powder from the blister at low flow rates. At both flow rates, the FPFs are very similar for all of the inhalers, so the FPD depends largely on the ED. Consequently, the 3.5-1.5 inhaler performs best at both flow rates. The Examples demonstrate that the optimal cyclone configuration depends on the nature of the powder and the inhalation strength (flow rate / pressure drop). The powder needs first to be aerosolized from its container and then disaggregated into fine particles in the cyclone chamber.

[0121] Thus, with powders that are relatively difficult to aerosolize, more of the air flow should flow through the container to ensure that the powder is fully aerosolized. Consequently, the cyclone chamber should have a relatively large power inlet channel and relatively small bypass inlet channels. The emitted dose, i.e. the total mass of powder that is emitted from the inhaler is a measure of the ease of evacuation. The ED can be determined using the FSI as in the Examples from measuring the total mass of powder that enters the impactor (instead of the mass captured by the filter).

[0122] On the other hand, for powders that are relatively difficult to disaggregate, more air flow is needed to disaggregate the aerosolized powder in the cyclone chamber. In this case, the powder inlet channel should be relatively small, and the bypass inlet channels should be relatively large. The difference in the particle size distributions of a powder dispersed at two different pressures is a measure of the ease of disaggregation. The PSDs can be determined using the Sympatec as in the Examples. Furthermore, with powders that are difficult to disaggregate, it is advantageous to have a relatively small exit opening in order to increase the residence time of the powder in the cyclone chamber.

[0123] Thus, the powder and bypass inlet widths can be optimised for any powder by first measuring the ED from an inhaler with a narrow powder inlet channel and a wide bypass inlet channel (i.e. low value of Ri) in which most of the energy from the user's inhalation is directed into the cyclone chamber. If the resulting ED is high, then the powder has been successfully evacuated from the blister. The FPD is likely to be maximised, because the maximum energy has been put into the cyclone to deagglomerate the powder. On the other hand, if the ED is low, then more energy needs to be out into evacuating the powder from the blister, so the width of the powder inlet should be increased relative to the width of the bypass inlet (i.e. increasing Ri), until a high ED is achieved. Increasing Ri further is unlikely to improve the ED (because good evacuation has already been achieved) and could reduce the FPD, since less energy would be available for deagglomeration.

[0124] Thus, the performance of the cyclone chamber can be optimized for powders with different properties by adjusting the sizes of the powder and bypass inlet channels and the size of the exit opening.

Claims

Claims1. A dry powder inhaler comprising:• one or more doses of powdered medication comprising a pharmaceutically active ingredient;• an outlet, such as a mouthpiece or nose piece, through which a user may inhale a dose of medication;• an airway having a single cyclone chamber, the cyclone chamber having:• one tangential powder inlet channel;• two tangential bypass air inlet channels; and• an axial exit opening connected to the outlet of the inhaler; wherein when a user inhales on the outlet to create an air flow, medication is entrained in the air flow and flows through the airway via the cyclone chamber, and out through the outlet, characterized in that the cross-sectional area of the powder inlet channel is not equal to the cross-sectional areas of the bypass inlet channels.

2. A dry powder inhaler according to claim 1, wherein the inhaler further comprises a container for the powdered medication, and the airway has a passage that connects the container to the powder inlet channel at an entrance to the power inlet channel.

3. A dry powder inhaler according to claim 2, wherein the entrance is located in the central 50% of the passage.

4. A dry powder inhaler according to any of claims 1 to 3, wherein the widths of the powder inlet channel and the bypass inlet channels are from 0.5mm to 5mm.

5. A dry powder inhaler according to any of claims 1 to 4, wherein the two bypass inlet channels have equal widths.

6. A dry powder inhaler according to claim 5, wherein the ratio of the width of the powder inlet channel to the width of each bypass inlet channel (Ri) is from 1:5 to 5:1.

7. A dry powder inhaler according to claim 6, wherein the width of the bypass inlet channels is less than the width of the powder inlet channel.

8. A dry powder inhaler according to claim 7, wherein the powdered medication has a Carr's Index of greater than 25, preferably greater than 30.

9. A dry powder inhaler according to claim 7 or claim 8, wherein 1.0 < Ri < 2.0.

10. A dry powder inhaler according to claim 6, wherein the width of the bypass inlet channels is greater than the width of the powder inlet channel.

11. A dry powder inhaler according to claim 10, wherein 0.2 < Ri < 1.0.

12. A dry powder inhaler according to any of claims 1 to 11, wherein the inlets are at angular spacings of between 90° and 150°, preferably at 120°.

13. A dry powder inhaler according to any of claims 1 to 12, wherein the diameter of the cyclone chamber is from 10mm to 40mm and the diameter of the axial exit opening is from 5mm to 15mm.

14. A dry powder inhaler according to any of claims 1 to 13, wherein the ratio of the diameter of the cyclone chamber to the diameter of the axial exit opening is ( R2) is given by 1.0 < R2 < 3.0.

15. A dry powder inhaler according to claim 2 or according to any of claims 3 to 14 when dependent on claim 2, wherein the ratio of the cross-sectional area of the smallest constriction in the container or the passage to the cross-sectional area of the smallest constriction in the powder inlet channel is greater than 2.

Images