Using deconvolution to understand the mechanism for variable plasma concentration–time proﬁles after intramuscular injection Adnan H. Mahmood a, b , Xin Liu a , Jeffrey E. Grice a , Gregory A. Medley a , Michael S. Roberts a, c, * a Therapeutics Research Centre, UQ School of Medicine, Translational Research Institute, 37 Kent St., Woolloongabba, Qld 4102, Australia b The Technical Institute of Baqubah, Foundation of Technical Education, Iraq c School of Pharmacy & Medical Sciences, University of South Australia, Adelaide, SA, Australia A R T I C L E I N F O Article history: Received 28 July 2014 Received in revised form 6 January 2015 Accepted 24 January 2015 Available online 27 January 2015 Keywords: Deconvolution Intramuscular injection Antibiotics A B S T R A C T To introduce better antibiotics for the treatment of some infectious diseases in sheep and to expand the range of antibiotics available for veterinary medicine, pharmacokinetics of two antibiotics marboﬂoxacin (MBX) and trovaﬂoxacin (TVX) were investigated in sheep after intramuscular injection. Variable and irregular plasma concentration–time proﬁles were observed for TVX but not for MBX. To understand the mechanism of this phenomenon, intravenous studies were performed for both drugs and data were analyzed using a population approach. Deconvolution was then performed using various approaches to obtain absorption proﬁles of both drugs in sheep after intramuscular injection. The Loo–Riegelman and staircase deconvolution function methods were found to provide more reliable estimates of absorption rate than the Spath-spline and B-spline constraining break points deconvolution methods. The absorption proﬁles resulting from deconvolution indicated a zero-order absorption process for TVX and a ﬁrst-order process for MBX. Precipitation of TVX at the injection site was suspected to cause the pseudo zero-order absorption. This hypothesis was supported by the observation of crystalline deposits of TVX in sheep meat after direct injection, using reﬂectance confocal microscopy. ã 2015 Elsevier B.V. All rights reserved. 1. Introduction There is a need to expand the range of antibacterial treatments for veterinary medicine in countries with persistent endemic bacteria diseases. In particular, more effective therapies than currently used classical antibiotics are required to treat the main respiratory infectious diseases in sheep, such as those caused by Mannheimia (Pasteurella) haemolytica and Mycoplasma ovipneu- moniae (Skoufos et al., 2007). Marboﬂoxacin (MBX) and trova- ﬂoxacin (TVX) were chosen as candidates as they have been shown to be effective for a range of infections, including intracellular bacterial diseases such as Salmonellosis, Campylobacteriosis and Brucellosis in sheep (Neu and Chin, 1994). MBX has been used in veterinary medicine to treat respiratory infections in cattle, pigs, dogs and cats, but with limited use in farm animals (Skoufos et al., 2007; Sidhu et al., 2010). TVX was previously used in human, but was withdrawn following reports of idiosyncratic hepatotoxicity (Lucena et al., 2000). These two drugs are also interesting because of their different physiochemical properties (Table 1). In general, it is preferred to administer antibiotics intramuscu- larly (IM) in ruminants as oral antibiotics can kill rumen ﬂora. IM administration may result in variable pharmacokinetics (PK) as a consequence of erratic absorption in some cases, especially for drugs with low solubility. For example, the absorption of phenytoin (Dam and Olesen, 1966) and diazepam (Bell et al., 1991; Gamble et al., 1975) after IM injection has been found to be erratic and slower than after oral administration (Jusko et al., 1973; Kostenbauder et al., 1975; Mandelli et al., 1978). In our study, unusual and variable plasma concentration–time proﬁles were observed for TVX after IM injection to sheep, but not for MBX. This paper details our effort to understand the mechanism for the variability of the TVX pharmacokinetic proﬁle and the different pharmacokinetic behavior of MBX and TVX after IM injection. Pharmacokinetic proﬁles of both drugs after intravenous (IV) injection were obtained to deﬁne the disposition processes of these drugs in sheep. The deconvolution method was used to access the absorption temporal proﬁle since there is not enough information for mechanistic modeling. * Corresponding author at: Therapeutics Research Centre, UQ School of Medicine, Translational Research Institute, 37 Kent St., Woolloongabba, Qld 4102, Australia. Tel.: +61 411264506. E-mail address: m.roberts@uq.edu.au (M.S. Roberts). http://dx.doi.org/10.1016/j.ijpharm.2015.01.046 0378-5173/ ã 2015 Elsevier B.V. All rights reserved. International Journal of Pharmaceutics 481 (2015) 71–78 Contents lists available at ScienceDirect International Journal of Pharmaceutics journal homepage: www.elsev ier.com/locate /ijpharm