Is the drug-drug interaction between rifampin and linezolid worth worrying about? In this article two infectious diseases pharmacists discuss existing data on the topic.
Authored By: Ryan Stevens, Pharm.D., BCPS, BCIDP and Christina Rivera, Pharm.D., BCPS, AAHIV-M
The management of deep seated endovascular and osteoarticular infections has long hinged on the administration of intravenous (IV) antibiotics [1-4]. Therapy with highly bioavailable oral agents has been recommended as a reasonable option in some circumstances with these infections; however, this has seemingly failed to gain steam in real world clinical practice.
From a pharmacokinetic stand point, administration of antibiotics with high bioavailability (>90%) will produce plasma concentrations roughly equivalent to those of intravenous administration, and, thereby, oral administration may potentially avoid the complications associated with long-term, indwelling central lines and outpatient or infusion center IV antibiotic administration. This concept makes the oral administration of antibiotics such as fluoroquinolones, tetracyclines, clindamycin, and oxazolidinediones attractive even in deep seated infections, particularly those requiring long durations. These agents also have also proven to have adequate bone penetration [5]. Thus, the high bioavailability lends potential application in endovascular infections and the bioavailability coupled with the bone penetration provides promise in the treatment of osteomyelitis.
Two important studies regarding the efficacy of oral antibiotics in the treatment of deep seated endovascular or osteoarticlular infections have recently been published, and it would seem that many are re-evaluating the need for long term IV antibiotic use in these syndromes.
The first, in left sided endocarditis, evaluated the outcomes of patients who were treated initially with IV antibiotics and then switched to oral antibiotics versus those continued on IV antibiotics for the duration of therapy [6]. This multicenter, randomized noninferiority trial enrolled 400 patients (roughly 200 in each arm), and demonstrated no statistically significant difference between groups in the rate of primary composite endpoint of all-cause mortality, unplanned cardiac surgery, embolic events, or relapse of bacteremia with a primary pathogen (IV group 12.1% vs. PO group 9%, 3.1% difference, 95% CI -3.4 to 9.6). Roughly 27% of patients in each group had prosthetic heart valves in place and the oral regimens used (found in the supplemental material) widely varied.
The second study evaluated outcomes of management of patients with bone and joint infections with oral antibiotics [7]. This study included 1054 patients with 527 being managed with oral antibiotics and 527 managed with IV antibiotics. For enrollment, patients were required to be within seven days of definitive surgery or, for those being medically managed, seven days of antibiotic initiation. They found, in their intention to treat analysis, antibiotic failure rates within one year of randomization to be 14.6% and 13.2% in the intravenous and oral antibiotic groups, respectively (-1.4% difference, 95% CI -5.6 to 2.9).
As the confidence in the utilization of oral antibiotics for endocarditis and osteoarticular infections grows the oxazolindinedione antibiotic linezolid has been a notable player in many discussions. The potential benefits to the use of linezolid for these infections included its high oral bioavailability, twice daily dosing regimen and broad gram-positive spectrum of activity. These benefits must be balanced with the limitations of relatively high rates of hematologic toxicity beyond fourteen days of therapy and notable drug-drug interactions [8]. Linezolid carries the risk of serotonin syndrome when administered with other serotonergic medications given the agent’s inhibition of monoamine oxidase A [9]. However, a lesser known interaction is that which must be considered when linezolid is co-administered with rifampin.
Rifampin is commonly administered as a part of a regimen in staphylococcal endocarditis and osteoarticular infections when prosthetic material (e.g., heart valves or joint hardware, is involved )[2-4]. The addition of rifampin is aimed to decrease biofilm production and increase penetration of co-administered antibiotics into biofilms.
Rifampin has long been known to enhance the metabolism of many drugs through its potent induction of the cytochrome p450 system [8, 10]. Given the lack of metabolism of linezolid through the cytochrome p450 system, the interaction mechanism is proposed to be related to inductions in p-glycoprotein expression with rifampin which may increase the clearance of linezolid and thereby reduce the area under the curve by roughly 30% [10, 11]. This interaction between linezolid and rifampin is problematic and could be theorized to lead to higher rates of treatment failure if this combination oral regimen is selected, especially in the treatment of serious infections with indwelling prosthetic material. This is also important given that in the aforementioned study evaluating the use of oral antibiotics for the treatment of left-sided endocarditis, a total of 14 patients received a regimen of co-administered linezolid and rifampicin.
In the below tables we highlight the available evidence surrounding observed decreases in linezolid area the curve (AUC) when co-administered with rifampin and the associated clinical outcomes of patients treated with this regimen.
Table 1. Case Reports.
Abbreviations: IVC = inferior vena cava, CVVHD = continuous venovenous hemodiafiltration, MRSA = methicillin-resistant Staphylococcus aureus, PJI = prosthetic joint infection, MRSE = methicillin resistant Staphylococcus epidermidis
Table 2. Pharmacokinetic Studies.
Abbreviations: MIC = minimum inhibitory concentration, D1 = day 1, PK = pharmacokinetic, Pg-P = P-glycoprotein, AUC = area under the curve
Pharmacists are likely to be called upon to offer management strategies for patients with a strong indication for an antimicrobial program involving both linezolid and rifampin. Though the pharmacokinetic evidence of interaction is clear, the clinical outcomes data related to combination use is mixed and limited to case reports and case series. Tertiary sources identify the potential for sub-therapeutic linezolid exposure and loss of efficacy, but offer only caution with combination use and monitoring for clinical failure as mitigating strategies. Linezolid assays are commercially available, though a send out test for many institutions, thus limiting utility. Turning to the primary literature, no pre-emptive linezolid dosing strategy has clearly emerged.
What appears consistent from the available studies is a general recommendation for linezolid serum monitoring. A specific approach for the practicing clinician is to assess a linezolid trough on day 3 (Cmin, ss) and titrate dose to goal trough of 2-7 μg/mL. Linezolid assays are commercially available, though a send out test for many institutions, which limits its utility. A reasonable dose adjustment in response to a low trough on standard linezolid dosing would be to increase to linezolid 600 mg every 8 hours. Once the target linezolid trough is obtained, consider repeating linezolid trough if any concern for clinical failure arises.
For critically ill patients requiring combination of linezolid and rifampin, an initial aggressive dosing of linezolid 600 mg every 8 hours that can later be adjusted if hematologic or pharmacokinetic monitoring parameters indicate seems justifiable. In addition to lack of real-time serum analysis availability, another notable limitation to serum monitoring approach is an unclear interplay of synergy effect between linezolid and rifampin. A standard target linezolid trough of 2-7 μg/mL does not take into account this in vitro demonstrated effect and thus calls into question serum target levels applicability in this specific setting.
An alternative approach is to consider proceeding with linezolid monotherapy. In vitro models have shown variable activity of linezolid monotherapy against biofilm forming and embedded isolates of Staphylococcus epidermidis, Staphylococcus aureus, and Enterococcus species, as compared to other agents considered standard of therapy in endovascular and osteoarticular infections [12-15]. A propensity-matched cohort study recently evaluated the efficacy of managing patients with Staphylococcus aureus bacteremia with an early switch to oral linezolid [16]. Of importance, this study specifically excluded those patients with complicated endovascular or osteoarticular infections; however, they did not identify differences in 90-day relapse or 30 day mortality in patients managed with early transition to oral linezolid versus those receiving standard parenteral regimens. A majority of the patients in the study had a primary source of infection of either a central venous catheter or skin and soft tissue infection and a majority of isolates were methicillin sensitive.
A systematic review of the use of linezolid in endocarditis showed a cure rate of 63.6%, and overall and endocarditis relative mortality rates of 33.3% and 12.1%, respectively [17]. The review included 33 patients from case reports/series. Methicillin resistant Staphylococcus aureus and vancomycin intermediate Staphylococcus aureus were the most commonly isolated organisms. Prosthetic valve endocarditis was present in 25% of the 33 patients, and linezolid was administered as monotherapy in 66.7% of the 33 cases. The mean duration of linezolid exposure was 36 days though the range was between 7 and 148 days. Data pertaining to development of hematologic toxicity was reported in 26 patients, of which 30.8% were noted to develop thrombocytopenia.
With regards to orthopedic infections involving implants the data also leaves some questions at hand. In one case series 14 patients received linezolid monotherapy for 6 weeks in the management of orthopedic implant infections and were followed for a duration of 6 months [18]. All 14 patients experienced cure with no relapse of infection; however, they also all received early debridement and removal of the orthopedic implants in addition to antimicrobial therapy. In another case series 20 patients with osteoarticular infections involving prosthetic joints were managed with linezolid monotherapy [19]. In this series the implanted prosthetics were retained, and patients were treated for a mean duration of 7.2 weeks with a range of 6-10 weeks of linezolid therapy. Four of the 20 patients experienced treatment failure, and there were no observed hematologic adverse effects that led to discontinuation of treatment.
Given the available evidence surrounding linezolid’s variable biofilm activity in comparison to other antimicrobials and the relatively weak literature supporting the use of linezolid monotherapy in the case of endovascular or osteoarticular infections where implanted prosthetics are retained, it would seem that linezolid monotherapy in this scenario may be an option, albeit non-ideal and weakly supported. Utilization of this regimen would require careful consideration, and may be best limited to cases where other therapeutic alternatives and/or the inability to co-administer rifampin is present if prosthetic material is retained.
A third, previously unexplored option is the consideration of rifabutin in place of rifampin to use with linezolid as it is a relatively weaker inducer than rifampin. Given a total lack of evidence in this specific context, rifabutin substitution cannot be recommendation at this time but represents an interesting area for future exploration.
In conclusion, the dawning era of oral antimicrobials for serious endovascular and osteoarticular infections presents interesting challenges and opportunities for infectious diseases pharmacists. Operating at the frontier of new treatment paradigms invites pharmacists to help cultivate future practice standards.
REFERENCES
9. FDA drug safety communication: serious CNS reactions possible when linezolid (Zyvox) is given to patients taking certain psychiatric medications. United States Food and Drug Administration website. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-serious-cns-reactions-possible-when-linezolid-zyvox-given-patients. Last updated: December 14th, 2017. Accessed: August 10, 2019.
RECOMMENDED TO YOU
