Introduction

Hard gelatin capsules have been applied in single dose dry powder inhalers for several decades. Recently, hydroxyl-propyl methylcellulose (HPMC) capsules are increasingly being used in DPI devices due to their favourable properties, i.e. lower moisture content and mechanical characteristics less sensitive to variations in humidity. The capsules are externally lubricated during production to prevent sticking to each other in capsule filling operations and subsequent packaging [1], [2]. In order to emit the powder from the device, capsules are pierced by one or more needles equipped in inhaler device during dose administration manoeuvres [3], [4]. Different lubricants could alter adhesion to the capsule material surface and thus mechanical properties of capsules. This could in turn influence the puncturing properties and subsequently shape and size of the opening and the flap and subsequent powder release. This study attempts to evaluate the puncturing properties of the RS01 Plastiape® device of gelatin and HPMC capsules treated with different external lubricants. Moreover, the goal is to investigate the relationship between the mechanical properties of differently lubricated capsules and their puncture characteristics and to indirectly compare those to the aerodynamic performance.

Material and Methods

Size 3 capsules of gelatin and HPMC were obtained from Qualicaps (Qualicaps Europe, S.A.U., Spain). These capsules were externally lubricated with magnesium stearate (MgSt), sodium lauryl sulphate (SLS) and carnauba wax (CW) and were compared to unlubricated capsules (w/o). After a storage period of one week in dry (22% RH) and humid (51% RH) conditions, the capsules (n=6) were punctured using a compression and tensile measuring device (Instron 5943, Instron® GmbH, Germany). The needle from an inhaler (RS01^®, Plastitape, Italy) was disassembled from the device and attached to the force transducer. The applied forces (N) were recorded, as previously described by Torrisi et al. [3]. The needle moved with 1mm/sec towards the centre of the capsule lid and the maximum force applied was referred to as the breakthrough force of the capsule. Subsequently, macroscopic pictures of the created openings were taken using a DSLR camera with a reversed 18mm objective (Canon 60D, 0.6sec, ISO500) and the contours were traced using ImageJ software. The number of the pixels in the opening area was calculated and compared to an area of the external diameter of the needle (1mm, 100% of the opening area), resulting in a calculated opening value. Mechanical properties of the conditioned capsules were measured at ambient conditions (22°C ± 2°C, 35% RH ± 3%) using a MCR compact Rheometer (MCR 300, Anton Paar, Austria). The resulting normal force (N) and displacement (mm) values were recorded every 0.06 mm, resulting in the stress-strain curves. The Young modulus was determined by linear regression from the slope of the elastic region of the stress-strain curves. Subsequently, these results were indirectly compared to the release of a powder mixture from the capsules after a filling process. For this purpose, the unconditioned capsules were filled with 25mg of a model blend at the previously mentioned room conditions (22% and 51% RH) with a capsule filling machine (FlexaLab, MG2, Pianoro, Italy) [5]. The model blend consisted of 1% (w/w) Budesonide (Laboratorios Liconsa, Spain) and 99% (w/w) Inhalac 230 (MEGGLE, Germany). The machine settings were maintained for all capsule types and aerodynamic performance was measured via the fast screening impactor (FSI).

Results and Discussion

Puncture profiles for gelatin and HPMC capsules were very different and were influenced largely by the storage conditions, which has also been previously described [3], [4]. Gelatin capsules required overall higher piercing forces (average of all capsules: 6.86N ± 1.14N at 22% RH to 5.90N ± 1.20N at 51% RH) than HPMC capsules (3.89N ± 0.65N at 22% RH to 2.99N ± 0.41N at 51% RH). The piercing forces decreased with increased humidity for both types of capsules. Even though there were slight differences in puncture properties between the differently lubricated capsules of the same material, this differences were not significant (data not shown). Our results showed that the small amount of lubricant does not significantly impact puncturing properties (shown in Table 1). This could be explained by the very low amount of lubricant used in the after-treatment of the production (2g lubricants on 10kg of capsules), which gives results in a very small layer in the nanometer scale.

Further, taking the size and shape of the openings into account, visible differences between gelatin and HPMC could be noticed. While gelatin capsules have relatively large openings under all conditions (90.8% opening ± 6.2% (average of all capsules)), the position of the flap in HPMC capsules was closing the puncture to a greater extent, resulting in a lower value for the openings (39.5% ± 18.9% (average of all capsules)). Provided that the flap might return towards the capsule material between the time point of the puncturing measurements (Instron) and the time point when the pictures were taken, the size of the opening was calculated without the area covered by the returning flap. HPMC capsules still showed substantially lower opening values than gelatin capsules (77.3% ± 8.5% (average of all capsules/conditions). It is assumed that, when capsules are opened within an inhaler, the flap has substantially less time to return. Also, these experiments were performed in absence of airflow, which is actually the driver of in vitro or in vivo aerosolisation. Therefore, we assume that the opening of HPMC capsules in the actual use within inhaler could be larger as well. Observing further the shape of the opening, it was evident that the edges of the punctures become smoother with increasing storage humidity (Figure 1).

The percentage values on Figure 1 show larger openings at lower humidity for all gelatin capsules and most of the HPMC capsules (exception were HPMC capsules lubricated with CW). However, looking into differently lubricated capsules, only few differences in opening sizes may be noticed for both materials. While at 22%RH, HPMC w/o (80.72% ± 8.98%), SLS (85.4% ± 10.41%) and MgSt (78.1% ± 15.38%) showed similarly large openings, HPMC capsules lubricated with CW (61.8% ± 11.75%) resulted in much smaller openings. Differences in the size and shape of the openings among the differently lubricated gelatin capsules were less pronounced. All gelatin capsules stored at 22% RH exhibited similarly large openings, (SLS 95.87% ± 5.09%, CW 95.76% ± 8.42%, w/o 92.26% ± 4.85% and MgSt 88.55% ± 2.99%). At 51% RH storage, opening area seemed to slightly decrease (w/o 91.76% ± 5.17%, SLS 90.17 ± 7.31%, CW 83.99% ± 7.16% and MgSt 87.94% ± 8.40%). The force needed for puncturing correlated well with the elasticity properties of the capsules as observed by Young modulus described previously by authors [6]. Elasticity of gelatin capsules increased with higher storage humidity and is probably the consequence of the increase of the amount of water molecules bound to the helical fragments of proteins in gelatin. The elasticity of HPMC capsules was found to increase at 51% RH, indicating the potential plasticization of the material [6]. Lubricants did not show large effect on the elasticity of both capsule types.

The aerodynamic performance of all filled capsules was tested using FSI measurements and was related to the capsule openings Figure 2. It was found that the main difference in API release from capsules was arising from the capsule material, although, there were some differences observed as a consequence of storage and lubrication. In general, gelatin capsules resulted in smaller fine particle dose (FPD) compared to HPMC capsules, suggesting that complete opening of the capsules is not necessarily indicator of the higher FPD. It is possible that due to the modified opening structure on both materials, different air flows are created in the inhaler, resulting in different shear forces directly at the exit point, which influence the detachment of an API. On the other hand structural differences of the different capsule materials on the inner surface could potentially influence the release of finer particles. Interestingly, opening size for HPMC and gelatin capsules did not seem to have that much impact on the FPD (Figure 2). HPMC capsules without lubricants filled at 51% RH and capsules with MgSt and SLS filled at 22% RH achieved a FPD of almost 60µg, whereas the average for gelatin was only about 25µg.

Conclusion

In summary, the use of different external lubricants on the surface of capsules could not be directly related to the force required to pierce capsules. However, it showed slight variations in the puncture sizes and shapes. The piercing forces were higher for gelatin than for HPMC capsules and have decreased with increased RH for both materials. The aerodynamic performance of both capsules showed that the larger size of the opening is not necessarily decisive for the higher dose delivery, however, both the capsule material and the storage humidity had an important role on the effective dose delivery from the device.