Benchtop-magnetic resonance imaging (BT-MRI) characterization of push–pull osmotic controlled release systems Vincent Malaterre a,b , Hendrik Metz c , Joerg Ogorka a , Robert Gurny b , Nicoletta Loggia a , Karsten Mäder c, ⁎ a Novartis Pharma AG, Technical R&D, Fabrikstrasse 2, CH-4056 Basel, Switzerland b School of Pharmaceutical Sciences, Ecole de Pharmacie Genève-Lausanne (EPGL), University of Geneva, Quai Ernest-Ansermet 30, CH-1211 Geneva 4, Switzerland c Institute of Pharmacy, Martin-Luther-University of Halle, Wolfgang-Langenbeck-Str. 4 D-06120 Halle/Saale, Germany abstract article info Article history: Received 29 April 2008 Accepted 4 September 2008 Available online 20 September 2008 Keywords: MRI NMR imaging Osmotic drug delivery Osmotic pumps Push–pull osmotic systems GITS Controlled release Drug delivery The mechanism of drug release from push–pull osmotic systems (PPOS) has been investigated by Magnetic Resonance Imaging (MRI) using a new benchtop apparatus. The signal intensity profiles of both PPOS layers were monitored non-invasively over time to characterize the hydration and swelling kinetics. The drug release performance was well-correlated to the hydration kinetics. The results show that (i) hydration and swelling critically depend on the tablet core composition, (ii) high osmotic pressure developed by the push layer may lead to bypassing the drug layer and incomplete drug release and (iii) the hydration of both the drug and the push layers needs to be properly balanced to efficiently deliver the drug. MRI is therefore a powerful tool to get insights on the drug delivery mechanism of push–pull osmotic systems, which enable a more efficient optimization of such formulations. © 2008 Elsevier B.V. All rights reserved. 1. Introduction Controlled drug delivery systems for oral applications are widely used clinically to decrease the frequency of administration or to reduce side effects that are related to peak plasma concentrations (C max ). Many of them are based on matrix tablets. In many cases, their drug release rates are either dependent on pH and / or on shear forces of the local environment leading to a so-called food effect and in vivo variability [1]. A reproducible zero-order release profile of poorly and pH-dependently soluble drugs is a challenge. Osmotic [2,3], multiparticulate [4,5] and, more recently, special erosion controlled delivery systems (Egalet™) [6] have been developed to overcome these limitations. Osmotic pumps also known as oral osmotic systems (OROS™) were reported as suitable to deliver poorly soluble compounds such as nifedipine, isradipine or doxazosin [7]. This controlled release technology was initially developed by Theeuwes and associates in the 1970s as a result of an ultimate simplification of Higuchi–Leeper pump design [8]. Two delivery systems based on OROS technology were mainly developed and marketed [9]. The elementary osmotic pump is based on a single tablet core and suitable for highly soluble drugs. The second type is the push–pull osmotic system (PPOS) based on a bilayer tablet core for poorly soluble compounds (Fig. 1). The delivery principle of all osmotic systems involves controlled water diffusion through a semipermeable membrane and the drug release through a laser-drilled orifice [2]. Several mathematical hydration models were proposed for single core systems [2,9] as well as for bilayer PPOS [10]. These approaches were based on the Starling equation (Eq. (1)) describing the flow rate (dV/dt) through a semipermeable membrane as: dV dt ¼ AdL p h σ Á Δπ-ΔP ð Þ ð1Þ with the membrane thickness (h) and surface (A), the water permeability (L p ), the difference of hydraulic pressure (ΔP) and the osmotic gradient (σ · Δπ). Eq. (2) was adapted to bilayer PPOS by Wong et al. [11]: dV dt ¼ σ dL p h A P H ð Þdπ p þ A-A P H ð Þ ð ÞÁ π D -ΔPH ð Þ Â Ã ð2Þ with the degree of hydration (H), the layer surfaces (A x ) and the osmotic pressure (π x ) of the push and the drug layers indexed with P and D respectively. These models were used to explain the effect of some parameters from a qualitative point of view. However, knowledge of the detailed mechanisms underlying the release process from PPOS is still limited due to a lack of experimental data despite the clinical value and long history of oral osmotically driven drug delivery systems. Therefore, the aim of the present study was to investigate the hydration kinetics of push–pull osmotic systems in more detail. For this purpose, a marketed formulation was compared to several laboratory formulations. The drug layer composition was modified with respect to drug load and polymer Journal of Controlled Release 133 (2009) 31–36 ⁎ Corresponding author. Tel.: +49 345 5525167; fax: +49 345 5527029. E-mail address: Karsten.Maeder@pharmazie.uni-halle.de (K. Mäder). 0168-3659/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2008.09.007 Contents lists available at ScienceDirect Journal of Controlled Release journal homepage: www.elsevier.com/locate/jconrel