Magréault S (1,2), Berrah R (1), Kerroumi Y (3), Salmon D (4), Marmor S (3), Zeller V (3,4), Jullien V (1,2)
(1) Université Paris Cité and Université Sorbonne Paris Nord, Inserm, IAME, F-75018 Paris, France (2) Unité Fonctionnelle de Pharmacologie, GHU Paris Seine Saint-Denis, Assistance Publique–Hôpitaux de Paris (APHP), Université Sorbonne Paris Nord; Bondy, France; (3) Centre de Référence des Infections Ostéo-Articulaires Complexes (CRIOAc), Groupe Hospitalier Diaconesses–Croix Saint-Simon, Paris, France; (4) Service de Médecine Interne, Hôpital Cochin, (APHP), Paris, France; (5) Service de Médecine Interne et Infectiologie, Groupe Hospitalier Diaconesses–Croix Saint-Simon, Paris, France
Background: Clindamycin is an antibiotic frequently used in the treatment of bone and joint infections (BJIs) and is often co-administered with rifampin to ensure efficacy and avoid the emergence of resistance in Staphylococcus spp infections. However, an important clindamycin-rifampicin pharmacokinetic (PK) interaction has been reported, particularly when these molecules were used orally (PO) [1]. Intravenous (IV) administration of clindamycin by continuous infusion (ci) can therefore lead to a satisfactory concentration (in terms of pharmacodynamics (PD)), even when co-administered with rifampicin [1,2]. This is probably because rifampicin has a greater impact on the bioavailability of clindamycin than on its clearance, but no model with sufficient PO and IV data has been able to demonstrate this.
Objectives: The aim of the present study was to develop a PK interaction model of IV and PO clindamycin when combined with rifampicin, and to determine whether bone clindamycin concentrations equal to at least four times the minimal inhibitory concentration (MIC) could be achieved for different doses and administration routes of clindamycin.
Methods: Data were prospectively obtained from 124 patients treated for bone and joint infections. Thirty-nine patients were treated by continuous infusion (ci) of clindamycin and then orally. Two and 7 samples were drawn during the IV and PO treatments, respectively. Twenty patients received clindamycin without rifampicin, 19 received clindamycin concomitantly with rifampicin. The remaining 85 patients received clindamycin by ci successively without and then with rifampicin. Clindamycin concentration was determined before and at least 10 days after rifampicin initiation.
Population PK analysis was performed using the Monolix v2023R1 Stochastic Approximation Expectation Minimization (SAEM) algorithm. Monte-Carlo simulations were run to determine the dose regimens and route of administration allowing the achievement of satisfactory clindamycin concentrations, with or without rifampicin. Dose regimen from 600 mg q8h to 1200mg q6h were tested using PO, intermittent IV or ci administrations. Global probabilities of target attainment (PTAs) were based on the MIC distribution of Staphylococcus aureus.
Results: A linear one-compartment model with first-order absorption and elimination was developed using 518 plasma concentrations. Concomitant administration of rifampicin increased clindamycin clearance by an average factor of 3 and the impact on clindamycin bioavailability was dose-dependent, with a decrease from 11.1 to 3.8 % for rifampicin doses of 600 mg and 900 mg q12h, respectively.
Global PTAs were highly variable according to the mode of administration. When administered orally, global PTAs were equal to over 75% if clindamycin was given alone, but when given together with rifampicin, it fell to less than 1%. Discontinuous IV administration was satisfying in only half of patients and only in cases of low MICs (≤ 0.125 mg/L) combined with high clindamycin doses (1200 mg q 6h). Continuous infusion (4800 mg/d) was adequate even for MICs close to the breakpoints.
Conclusions: This model satisfyingly described the differential effect of rifampicin on the bioavailability and clearance of clindamycin. As previously described, this combination must not be taken orally. However, administration of high clindamycin doses by intermittent or, preferably, continuous infusion, can balance the impact of rifampicin.
References:
[1] Zeller V et al., Clin Microbiol Infect. 2021 Dec;27(12):1857.e1-1857.e7.
[2] Mimram L et al., J Antimicrob Chemother. 2023 Dec 1;78(12):2943–9.
Reference: PAGE 32 (2024) Abstr 10760 [www.page-meeting.org/?abstract=10760]
Poster: Clinical Applications