Next Article in Journal
Staphylococcus aureus Isolated from the Oral Cavity: Phage Susceptibility in Relation to Antibiotic Resistance
Next Article in Special Issue
Cerebrospinal Fluid Concentrations of Meropenem and Vancomycin in Ventriculitis Patients Obtained by TDM-Guided Continuous Infusion
Previous Article in Journal
Carvacrol Essential Oil: A Natural Antibiotic against Zoonotic Multidrug-Resistant Staphylococcus Species Isolated from Diseased Livestock and Humans
Previous Article in Special Issue
Prolonged Infusion of β-Lactams Decreases Mortality in Patients with Septic Shock: A Retrospective before-and-after Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Antibiotics
Volume 10
Issue 11
10.3390/antibiotics10111330
antibiotics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Clinically Relevant Interactions with Anti-Infectives on Intensive Care Units—A Multicenter Delphi Study

by
Joachim Andreas Koeck
1,2,3,4,*,†,
Heike Hilgarth
1,5,6,7,†,
Andreas von Ameln-Mayerhofer
1,8,
Damaris Meyn
1,9,
Ruediger Warlich
1,10,
Andreas Münstedt
11,
Dagmar Horn
1,11,‡ and
Christina König
1,6,7,‡
1
Committee for Intensive Care Medicine and Clinical Nutrition, German Association of Hospital Pharmacists (ADKA), 10559 Berlin, Germany
2
Department of Intensive Care and Intermediate Care, RWTH Aachen University Hospital, 52074 Aachen, Germany
3
Hospital Pharmacy, RWTH Aachen University Hospital, 52074 Aachen, Germany
4
Institute of Clinical Pharmacology, RWTH Aachen University Hospital, 52074 Aachen, Germany
5
Committee for Pharmacists’ Interventions, German Association of Hospital Pharmacists (ADKA), 10559 Berlin, Germany
6
Hospital Pharmacy, University Medical Center Hamburg-Eppendorf, 20246 Hamburg, Germany
7
Department of Intensive Care Medicine, University Medical Center Hamburg-Eppendorf, 20246 Hamburg, Germany
8
Hospital Pharmacy, Sindelfingen-Boeblingen Medical Center, 71065 Sindelfingen, Germany
9
Hospital Pharmacy, Gesundheit Nordhessen Holding AG, 34121 Kassel, Germany
10
Hospital Pharmacy, Sana Medical Center Offenbach, 63069 Offenbach, Germany
11
Hospital Pharmacy, University Hospital Muenster, 48149 Muenster, Germany
 
add Show full affiliation list
 
remove Hide full affiliation list
*
Author to whom correspondence should be addressed.
†
contributed equally as first authors.
‡
contributed equally as last authors.
Antibiotics 2021, 10(11), 1330; https://0-doi-org.brum.beds.ac.uk/10.3390/antibiotics10111330
Submission received: 30 September 2021 / Revised: 22 October 2021 / Accepted: 26 October 2021 / Published: 31 October 2021
(This article belongs to the Special Issue TDM, Individual Dosing and Application of Antibiotics/Antiinfectives in Critically Ill Patients)
Browse Figure
Versions Notes

Abstract

:
Patients in intensive care units (ICUs) are at high risk of drug–drug interactions (DDIs) due to polypharmacy. Little is known about type and frequency of DDIs within German ICUs. Clinical pharmacists’ interventions (PI) recorded in a national database (ADKA-DokuPIK) were filtered for ICU patients. Binary DDIs involving ≥1 anti-infective agent with >1 database entry were selected. A modified two-step Delphi process with a group of senior hospital pharmacists was employed to evaluate selected DDIs for clinical relevance by using a five-point scale and to develop guidance for clinical practice. In total, 16,173 PI were recorded, including 1836 (11%) DDIs in the ICU setting. Of the latter, 41% (756/1836) included ≥1 anti-infective agent, 32% (590/1836) were binary DDIs, and 25% (455/1836) were listed at least twice. This translates into 88 different DDIs, 74% (65/88) of which were rated as being clinically relevant by our expert panel. The majority of DDIs (76% [67/88]) included macrolides, antifungals, or fluoroquinolones. This percentage was even higher in DDIs being rated as clinically relevant by the experts (85% [55/65]). It is noted that an inter-professional discussion and approach is needed in the individual patient management of DDIs. The guidance developed might be a tool for decision support.

1. Introduction

Critically ill patients typically receive polypharmacy and are at high risk for drug–drug interactions (DDIs) [1,2,3]. DDIs occur between two or more drugs, which may result in an altered effect or even toxicity of at least one drug. Two recent reviews reported a prevalence of DDIs between 58–67% during ICU stays [4,5]. Due to the patients’ critical condition and continuous monitoring in the ICU, not all DDIs are considered clinically relevant [6]. However, 38% of ICU patients will experience at least one clinically relevant DDI during their ICU stay [6].
Since infections are an important issue in the ICU, a significant number of patients will receive at least one anti-infective agent at some stage during their treatment [7]. In consequence, adequate knowledge of most common interactions between anti-infectives and other drugs and their clinical consequences is necessary [7]. If undetected, DDIs can lead to patient harm or to potentially fatal adverse events [7].
The two main mechanisms known for DDIs are pharmacodynamic and pharmacokinetic alterations. Many of the major pharmacokinetic interactions affect hepatic cytochrome P-450 (CYP) enzymes [8,9]. Interactions involving CYP generally are of two types. The first one is enzyme induction, which leads to an increase of hepatic metabolism, resulting in decreased serum levels of CYP substrates, increasing the risk of therapeutic failure [8,9]. The second type is enzyme inhibition, leading to a reduced metabolism, an increase of serum levels resulting in an increased risk of overdosing and manifestation of potentially fatal adverse events (AE) [8,9].
A frequently recorded pharmacodynamic interaction involves QTc-prolonging drug–drug combinations. In contrast to its high prevalence, only 0.7% of all hospitalized patients in a six-month period developed a severe acquired prolonged long QT-syndrome (ALQTS, i.e., ≥500 ms) [10]. Furthermore, even in the case of an ALQTS, the prevalence of syncope and life-threatening arrhythmia (torsade de pointes) was found to be only 5.8% [10]. Additional risk factors must be considered when assessing the clinical relevance of these DDIs in the individual patient, such as serum potassium and magnesium levels, heart failure, acute myocardial infarction, or acute sepsis. The Tisdale score takes these risk factors into account and has been validated in cardiac and medical ICU patients [11,12]. Thus, it should be applied to daily practice to prevent over-alerting due to QTc-related DDIs.
Serotonin syndrome is known as a potentially life-threatening condition associated with increased central and peripheral serotonergic activity in the central nervous system [13]. It can manifest with therapeutic medication use and especially as a consequence of DDIs [13,14]. Although commonly described as the triad of mental status changes, autonomic hyperactivity, and neuromuscular abnormalities, serotonin syndrome can occur in the absence of an elevated temperature or monoamine oxidase inhibitor treatment, and fast onset cannot be regarded as a reliable clinical sign [13,14].
In order to limit the high number of possible interactions to those which are of clinical significance, our study group used a two-step Delphi process with the aim to describe clinically relevant DDIs involving anti-infective agents, which commonly occur in critically ill patients. In addition, we developed guidance on how to manage these in clinical practice.

2. Results

2.1. Drug–Drug Interactions from the ADKA-DokuPIK Database

The German ADKA-DokuPIK database comprised 16,173 PI from ICUs that were recorded over 13.5 years until 2021. Of these, 11% (1836/16,173) described a DDI, of which 41% (756/1836) involved at least one anti-infective agent. A total of 32% (590/1836) were binary drug combinations, with 455 DDIs (455/1836 [25%]) being recorded at least twice (see Figure 1). Out of 455 DDIs, 88 different binary drug–drug combinations were identified (see Table S1).

2.2. Ratings of the Expert Panel within the Modified Two-Step Delphi Process

The expert panel comprised of senior clinical pharmacists working on interdisciplinary (6/7 pharmacists), neurology (3/7), and neonatal and pediatric (1/7) ICUs as well as on burns units (1/7), with a professional experience of at least 10 years (IQR 10–16 years).
Within the first Delphi round, consensus was achieved on 59% (52/88) of DDIs, increasing to 93% (82/88) by the end of the second round. Low agreement was attained for 7% (6/88) (Table S3).
In total, 74% of DDIs (65/88) were rated as clinically relevant with sufficient agreement (Table 1 and Table 2). Macrolides (29/88), antifungals (22/88), and fluoroquinolones (16/88) were involved in 76% (67/88) of all DDIs and in 85% (55/65) of those DDIs with clinical relevance according to our expert panel. Acknowledging the initial dataset from the ADKA-DokuPIK database, DDIs rated as clinically relevant frequently included fluoroquinolones (15/16, 94%), antifungals (19/22, 86%), macrolides (21/29, 72%), and rifampicin (4/6, 67%), whereas interactions with, e.g., linezolid were considered less relevant by our expert panel (1/7, 14%). Of all DDIs, 19% (17/88) were rated as not clinically relevant by the authors (see Table S2). Among these, seven were “not relevant at all” and ten “relevant but with low risk for AE due to routine monitoring”. Only for 7% of all DDIs (6/88), the expert panel determined a “low agreement” (see Table S3). The discussion that led to the low agreement is briefly described in Table S3 to provide guidance in daily practice. Of note, all 88 DDIs were initially considered clinically relevant by the pharmacist in the individual patient context that entered the DDI into the database.
For most DDIs (46/65) rated clinically relevant (Category 3), additional monitoring could help to limit toxicities (see Table 2). Nineteen DDIs required therapy modification as they may not be controlled by additional monitoring (Categories 4 and 5).
In total, the expert panel developed 81 recommendations for 65 clinically relevant DDIs. Therapeutic drug monitoring (TDM), electrocardiogram (ECG) monitoring for QTc-prolongation, and monitoring of creatine kinase (CK) or withholding a drug was recommended for 25, 22, and 14 DDIs, respectively. Therapy modification (e.g., switching to an alternative drug) was recommended for seven DDIs.

2.3. Interaction Fact Sheets

2.3.1. Cephalosporins

Ceftriaxone and Calcium-Containing Intravenous Solutions

Co-administration of intravenous ceftriaxone and calcium-containing solutions (i.e., calcium administration for therapeutic purposes or as part of a solvent) may result in precipitation of ceftriaxone–calcium salts. This interaction may also occur when both drugs are administered via Y-site infusion [16]. Therefore, this DDI was rated relevant consistently by all consulted databases.
Ceftriaxone is mainly excreted via renal and biliary pathways. There are pediatric reports showing ceftriaxone–calcium precipitates that resulted in nephrolithiasis [17], biliary sludge or stones [18,19]. In neonates up to an age of 28 days, ceftriaxone is contraindicated according to the summary of product characteristics (SmPC), as precipitation may occur even when ceftriaxone and calcium are administered at different sites [16]. In patients >28 days, ceftriaxone and calcium may be administered at different sites or when infusion lines are sufficiently flushed between administration of these drugs. In case of symptoms for nephrolithiasis or biliary sludge/stones, ultrasound imaging is recommended, and discontinuation of ceftriaxone should be considered.

2.3.2. Carbapenems

Carbapenems and Valproic Acid

The interaction of carbapenems and valproic acid (VPA) is significant, of rapid onset, and should be avoided if possible [20]. It leads to decreased serum levels of VPA and may result in loss of seizure control and an increase of seizure frequency [21,22]. The exact pharmacokinetic mechanism is poorly understood. Animal studies suggest a reduced intestinal absorption and enterohepatic recirculation [23]. An increase of glucuronidation and a decrease of hepatic hydrolysis resulting in an increased renal clearance of VPA-glucuronide have been postulated [24,25,26,27]. Inhibition of efflux of VPA from erythrocytes and its accumulation have also been described [28,29].
The decrease of VPA levels was highest with meropenem (77%), followed by ertapenem (71%) and imipenem (52%) [30]. Serum levels declined within 24–72 h and were found to be subtherapeutic within 4 days. VPA levels remained low despite VPA dose-increase and were not dependent on meropenem dosages [22,31]. After discontinuation of carbapenem therapy VPA levels returned to the therapeutic range after 8–14 days [30,32]. A case series illustrated that even a short course of meropenem may have long-lasting effects (4 weeks) on VPA serum levels [33].
The concomitant use of carbapenems and VPA should be avoided based on the SmPC and database recommendations. Selection of an alternative anti-infective agent and/or (additive) antiepileptic drug should be discussed based on patients’ individual characteristics (e.g., organ function, microbiology results, seizures frequency/type, or drug history). The additive antiepileptic therapy should be continued for up to 7 days after discontinuation of carbapenem therapy, and VPA serum levels should be checked regularly [32].

2.3.3. Fluoroquinolones

Fluoroquinolones and QTc Prolonging Drugs

Fluoroquinolones can prolong the QTc interval, with moxifloxacin posing the greatest risk [34]. Concomitant use of other drugs prolonging the QTc interval such as class III antiarrhythmics (e.g., amiodarone), SSRIs (e.g., citalopram), Noradrenaline and specific serotonergic antidepressants (e.g., mirtazapine), tricyclic antidepressants (e.g., amitriptyline), and antipsychotics (e.g., haloperidol) can increase the risk of arrhythmias [35,36]. In general, the SmPC advises caution and additional monitoring. Particularly, combining a fluoroquinolone with amiodarone is not recommended by the SmPC, and alternative medication should be discussed.
Despite conflicting database ratings, one can conclude that additional measures such as ECG monitoring are useful to detect prolonged QTc intervals in order to avoid AE. Moreover, prior to the administration of QTc prolonging drug combinations, the screening for risk factors by applying the Tisdale score is reasonable to evaluate the potential threat (see Section 1: Introduction) [11].

Fluoroquinolones and Polyvalent Cations or Simvastatin

DDIs with polyvalent cations such as iron and calcium with fluoroquinolones do appear frequently when using oral anti-infective therapy and can be avoided by using separate dosing schedules (see Table 2). Moreover, patients using a combination of ciprofloxacin (CYP 3A4 inhibitor) and simvastatin (CYP 3A4 substrate) should be monitored for tendinopathies and CK-increases (for more details, see Section 2.3.4 Macrolides).

2.3.4. Macrolides

Macrolides and CYP 3A4 Substrates

Unlike azithromycin, both clarithromycin and erythromycin inhibit CYP 3A4 enzymes resulting in reduced metabolism of various statins (simvastatin and atorvastatin) [37]. Drug exposure is markedly increased, and the risk of myopathy and rhabdomyolysis rises [38]. The combination of CYP 3A4 inhibiting macrolides and statins is usually not recommended by SmPCs. When possible, affected statins should be withheld until the end of the macrolide therapy [39]. If concurrent use is unavoidable, the statin therapy should be switched to another statin where dosing recommendations are available (e.g., atorvastatin) and monitoring of elevated CK and muscle tenderness should be performed. Alternatively, therapy can be changed to a statin, which is not substrate to CYP 3A4 (e.g., rosuvastatin, fluvastatin or pravastatin).
The combination of erythromycin or clarithromycin with tacrolimus and/or cyclosporine (both CYP 3A4 substrates) will lead to increased concentrations of the immunosuppressants [40]. This may result in potentially toxic serum levels, nephrotoxicity, and prolonged immunosuppression. A combination of CYP 3A4 inhibiting macrolides with immunosuppressants (e.g., tacrolimus and cyclosporine) should be avoided if possible. If concurrent use is unavoidable, immunosuppressant serum levels should be frequently monitored and dosages adjusted accordingly [39].

Macrolides and Antidepressants or Antipsychotics

All macrolides are associated with prolongation of the QTc interval and have different cardiac safety profiles. In vitro studies show a variety of causative mechanisms such as formation of reactive oxygen species, block of potassium channels, as well as effects in the cardiomyocyte mitochondria being responsible for their cardiotoxic adverse effects [41,42]. Moreover, macrolide antibiotics showed different potential in causing arrhythmias (erythromycin > clarithromycin > azithromycin) [43]. A recently published meta-analysis evaluated patients receiving erythromycin or clarithromycin being at a higher risk of myocardial infarction (OR = 1.58 and OR = 1.41) when compared with azithromycin [44]. In combination with other QTc prolonging agents such as antidepressants and antipsychotics (e.g., quetiapine, melperone, haloperidol, or citalopram) the risk of cardiac adverse events such as torsade de points increases. If macrolides are used in combination with other QTc-prolonging agents, the approach as suggested in Section 1: Introduction should be used [11].

2.3.5. Antifungals

Echinocandins

Caspofungin undergoes slow metabolic transformation but uses hepatic transporters such as the OATP-1B1 (Organic Anion Transporting Polypeptide) [45]. Therefore, coadministration of caspofungin and cyclosporine (substrate of OATP-1B1-transporters) increased the AUC of caspofungin by 35% [46]. In contrast, cyclosporine serum levels were not increased [46,47]. By an unknown mechanism, caspofungin also has the potential to alter tacrolimus pharmacokinetics as AUC and Cmin of tacrolimus were found to be reduced by about 25% [46]. These are preliminary results and still under debate [47]; thus, their clinical relevance is uncertain. A higher exposure of caspofungin appears uncritical due to its low potential for adverse effects. However, regular monitoring of the serum levels of cyclosporine and tacrolimus seems to be reasonable when using caspofungin.

Azole Antifungals

The azole antifungals itraconazole, fluconazole, posaconazole, voriconazole, and isavuconazole are frequent partners in clinically relevant DDIs due to their inhibition of fungal and human CYP enzyme systems (see Table 3) [48].
A combination of azoles causing moderate to strong inhibition of CYP 3A4 with statins such as atorvastatin and simvastatin (CYP 3A4 substrates) is usually not recommended (SmPC) due to an increased risk of statin toxicity. Clinical management is described in the Section 2.3.4. Macrolides.
A combination of immunosuppressants such as cyclosporine or tacrolimus with CYP inhibiting azoles results in a marked increase of drug exposure of the immunosuppressants with potentially toxic effects [49,50,51]. Hence, a concurrent use of azoles and cyclosporine or tacrolimus should be avoided. If unavoidable, an initial dose reduction of the immunosuppressants by about 30–50% is recommended with further therapy being guided by TDM (SmPC). A similar approach applies to sildenafil (CYP 3A4 substrate) as azoles increase its serum level 10-fold, leading to a risk of AE (e.g., excessive blood pressure-lowering) [52]. Therefore, concurrent administration of an azole is not recommended by the SmPC, and alternative antifungal agents such as echinocandins or liposomal amphotericin B should be discussed.
Concurrent use of azoles (e.g., posaconazole or fluconazole) with carbamazepine (CYP 3A4 substrate and strong inducer) results in altered pharmacokinetics of both substances. Inhibition of carbamazepine metabolism increases its serum level up to 140%, resulting in a high risk for AE. On the other hand, carbamazepine co-administration can result in decreased azole serum levels (e.g., voriconazole), increasing the risk for therapeutic failure. Close monitoring for increased carbamazepine levels and toxicity (e.g., ataxia, drowsiness, and vertigo) as well as azole levels is highly recommended.
The use of azoles (CYP substrate, e.g., voriconazole and isavuconazole) in rifampicin (CYP 3A4 inducer)-treated patients leads to an inadequate exposure and increased risk of failure of the antifungal therapy. The SmPC recommends avoiding this combination. When antifungal therapy cannot be switched to alternatives (e.g., echinocandins or liposomal amphotericin B), a suitable alternative for rifampicin should be discussed such as fosfomycin in the case of biofilm-producing microorganisms [53]. Alternatively, azole TDM can be performed especially in patients with extra-corporeal organ support and with regard of reduced azole exposition due to DDI [54]. If TDM is not available, SmPC for fluconazole recommends a dose increase of 25% in combination with rifampicin.
The oral liquid of posaconazole shows a highly variable bioavailability especially when administered on an empty stomach, indicating pH dependent absorption. By increasing gastric pH, e.g., by proton-pump inhibitors (PPI), absorption of posaconazole suspension is markedly reduced [55]. To avoid therapeutic failure, concurrent use of a PPI should be avoided or posaconazole administered as a modified-release tablet (when swallowing is possible) [56] or as an intravenous infusion.

2.3.6. Miscellaneous Anti-Infectives

Daptomycin

Both statins and daptomycin can cause an increase in CK and are independently associated with myopathy and rhabdomyolysis [57]. When used concurrently, the SmPC recommends twice-weekly CK-monitoring and consideration of temporarily withholding statins; in the case of a CK increase to five times its normal value, both the statin and daptomycin should be stopped. Additional risk factors (e.g., age > 65 years, female sex, and untreated hypothyroidism) for myopathy and rhabdomyolysis should also be considered. Because of high inter-individual serum level ranges of daptomycin, TDM can help to achieve effective serum levels whilst minimizing the risk of toxicity [58,59].

Rifampicin

Rifampicin is a strong inducer for CYP 3A4 and P-gp. Thereby, the concomitant administration can result in altered metabolism or transportation. During concurrent use of cyclosporine (CYP 3A4, P-gp substrate) and rifampicin, serum levels of cyclosporine can decrease to subtherapeutic levels, increasing the risk of therapeutic failure. If an alternative immunosuppressant therapy is ineligible, cyclosporine concentrations should be closely monitored. A two- to threefold increase in doses of cyclosporine can be required in these situations [60].
After five days of concurrent use with rifampicin, Cmax and AUC of simvastatin dropped by about 90% and 87% in healthy adults [61]. The SmPC recommends avoiding concurrent use of simvastatin and rifampicin. When statin therapy is unavoidable, a statin that is not substrate to CYP 3A4 (e.g., fluvastatin and pravastatin) may be used.
The oral P2Y12 receptor antagonist ticagrelor is substrate to CYP 3A4, and both the parent compound and its active metabolite are substrates of P-gp. Concurrent administration significantly decreased Cmax by 73%, AUC by 86%, and the serum half-life of ticagrelor by 67%. Moreover, the active metabolite decreased by 46% [62]. Therefore, concurrent use can result in a marked decline of platelet inhibition and should be avoided.
The prodrug clopidogrel is primarily metabolized by CYP 2C19 into its active metabolite. In combination therapy with rifampicin, Cmax and AUC of the active metabolite were increased up to fourfold in healthy adults, resulting in an increased platelet inhibition [63]. Because of the high risk for bleeding complications, this combination should be monitored closely (SmPC) or if possible avoided. For both ticagrelor and clopidogrel, prasugrel (minor substrate of CYP 3A4) may be an alternative as rifampicin does not significantly alter its metabolism [64]. For information on DDIs between rifampicin and azole antifungals, see Section 2.3.5 Antifungals.

Linezolid and Serotonergic Agents

Linezolid was originally investigated as a psychotropic agent with antidepressant effects through mild reversible nonselective inhibition of monoamine oxidase (MAO). Moreover, it was also found to have anti-infective properties against drug-resistant gram-positive cocci [65]. While some case reports showed the occurrence of a serotonin syndrome solely with mirtazapine [66,67], there is one case where serotonin syndrome developed under the concomitant therapy of linezolid, mirtazapine, and citalopram [68]. Therefore, a combination of serotonergic agents and linezolid should be avoided. On the other hand, the mediQ® database and Fong et al. estimated the risk of serotonin syndrome with mirtazapine as low, because of its antagonistic impact on 5-HT2-receptors, which might provide a protective effect [69]. If strictly indicated, linezolid may be used in mirtazapine-treated patients, provided there is close monitoring for symptoms of serotonin syndrome (e.g., perspiration, fever, and tachycardia) (see Section 1: Introduction).
As there is a high inter- and intra-individual range of linezolid serum levels and an evident risk of dose-dependent AE, dosing of linezolid should be guided by TDM where available [70].

3. Discussion

To the best of our knowledge, this is the first publication to review and interpret clinically relevant DDIs occurring regularly in ICU patients in Germany. PIs, documented in a national database, described regular clinical pharmacists’ activity and input on patient and drug safety. Monitoring and managing DDIs is only one aspect of clinical pharmacy practice on German ICUs. However, clinical pharmacists are an essential team member when detecting, evaluating, and managing DDIs according to clinical relevance. Considering an individual patient’s condition and requirements, they support other health care practitioners by recommendations on monitoring or suggesting therapeutic alternatives [71].
This study has several strengths. Firstly, by using the previously described ADKA DokuPik® database [72], we were able to describe the natural heterogeneity of critically ill patients. We were able to include PI from a wide range of pharmacists throughout the country. Therefore, we do present a high diversity of DDIs within the medical therapy of critically ill patients with various comorbidities within several ICU types. Secondly, we identified the main anti-infective agents accountable for a considerable number of DDIs in ICUs.
Among all substances that have been recorded in the database three substance classes, namely macrolides, antifungals, and fluoroquinolones were involved in about 76% of all documented DDIs and in 85% of all by our experts clinically relevant rated DDIs. This is in line with several other studies that investigated relevant DDIs in critically ill patients where the same substance classes have been reported as being highly relevant or frequently causing adverse effects [4,6,73]. This parallelism in findings may be due to the nature of the interactions, i.e., the P-gp and/or CYP interaction potential [74,75]. With the essential knowledge of these three groups, the physicians and clinical pharmacists can deal with the most frequent DDIs.
Modern patient data management systems (PDMS) and electronic health records do provide automated DDI checks, often resulting in a flood of drug-associated warnings. One of the fundamental roles of a clinical pharmacist is to become a gatekeeper in the over-alerting of DDI warnings. Clinical pharmacists can support the physicians in benefit-risk assessments, when strictly required polypharmacy raises both the risks for therapeutic failure and potentially fatal AE. Therefore, we have developed guidance on how to manage these DDIs in clinical practice based on SmPC, currently available literature, and DDI databases. It should be particularly emphasized that clinical decision-making is a process requiring interdisciplinary collaboration. With regard to the urgent needs and problems of ICU patients, DDIs must be discussed with the treating physicians. In this context, as reported previously by Tisdale et al., risk factors such as QTc prolongation should be assessed in order to evaluate and minimize the occurrence of severe adverse events [11]. Managing CYP interactions is usually well-described in SmPC as well as in clinical studies. However, when clinical or patient data are not available (e.g., in the case of new drugs), in vitro or in silico estimations based on the metabolism and extent of CYP inhibition/induction are a common approach to estimate the effect [76,77].
Last, our clinical relevance definition was based on a Delphi process assessed by a senior expert panel. This method has been described and carried out before [15].
A significant limitation of this work is that the number of pharmacists working on German ICUs is not known. Data suggests that 16.8% of all PI entered in DokuPIK® are generated on ICU/intermediate care units or operating theatres [78]. As the data entry in DokuPIK® is voluntary, data might have been entered by more engaged pharmacists and therefore the results might not be fully applicable to ‘clinical pharmacy practice in Germany’ as a whole. However, the extracted binary DDIs are suitable to provide reasonable and sufficiently reliable results. Within a duration of more than 13 years, only 16,173 PI were entered, which could be due to time restraints and lack of legal requirements for clinical pharmacists to document PI as well as small numbers of registered users. However, 11% of PI were logged as DDI. Hence, this represents only a small proportion of a pharmacist’s daily work.
It must be noted that we only included binary DDIs. DDIs between three or more drugs were excluded as evidence for relevance, and guidance is sparse. Moreover, we did not include all possible DDIs but only those that were included in the ADKA DokuPIK® database. There might be clinically relevant DDIs that we did not include using this method.
At last, the expert panel was formed by senior hospital pharmacists regularly working in ICU but with no further specialist qualification for critical care pharmacy, because there is no structured training in this regard in Germany. We did not involve other groups of health care professionals. Additionally, the senior pharmacists rated the DDIs according to their individual opinion, based on their expertise in the field. This may have influenced the assessment of a DDI being clinically relevant. However, the Delphi process aimed at minimizing this effect, in which an anonymized rating of DDIs by multiple experts took place.
Patients in critical care are especially vulnerable due to life-threatening illness, complex pharmacotherapy, and medical care requirements, in addition to extensive monitoring including diagnostic and laboratory data. In the case of clinically relevant DDIs, the multidisciplinary team will responsibly decide whether monitoring is feasible, and drug-combinations may be used in these special circumstances and comply with local legal requirements.

4. Materials and Methods

The national anonymous self-reported online documentation system DokuPIK®, hosted by the German Association of Hospital Pharmacists (ADKA), was used for data acquisition. In DokuPIK®, pharmacists voluntarily document medication errors (ME) and PIs describing type of intervention, reasons, actions taken, and outcomes. A more detailed description of the DokuPIK® database has been published previously [79].
From DokuPIK®, we extracted data for PI entered as binary DDI (interactions between only two drugs) in the ICU setting from 2007 until 2021. Using Microsoft Excel® (Redmond, Washington), binary interactions with at least one anti-infective agent (ATC code “J”—anti-infectives for systemic use [80]) were selected. Combinations of more than two substances were excluded. DDIs had to be listed at least twice to be included in the two-step Delphi process.

4.1. Expert Panel and Modified Two-Step Delphi Process

The resulting list of DDIs was assessed by a seven-member expert panel of the ADKA committee for intensive care medicine and clinical nutrition using a modified two-step Delphi process [81]. Rating of clinical relevance was based on five mode categories as reported by Bakker et al. [15]:
  • Mode category 1: “Not clinically relevant, since adverse effects of this DDI are negligible”.
  • Mode category 2: “Clinically relevant; the adverse effects of this DDI will, however, be limited since routine monitoring to timely identify adverse effects is present”.
  • Mode category 3: “Clinically relevant; the adverse effects of this DDI can, however, be limited by additional monitoring and/or changes in dosage/frequency/timing”.
  • Mode category 4: “Clinically relevant, the adverse effects of this DDI on the patient can be substantial; however, these effects are acceptable and treatable”.
  • Mode category 5: “Clinically relevant; the adverse effect of this DDI on the patient should preferably be avoided”.
A clinically relevant DDI required the majority of experts to decide for Mode Category 3–5 and excluded low agreement. Low agreement was defined in line with Bakker et al. by at least one rating in Mode Categories 1–2 and 5 [15]. If a DDI was scored in Mode Category 3, it was required to provide additional information on preventing or monitoring AE.
For the recommendation and guidance, experts reviewed current literature and SmPCs and consulted six international DDI databases (Medscape® drug interaction checker (United States); Lexicomp® Drug Interactions (Netherlands); Stockely’s Drug Interactions® (Great Britain); mediQ® (Switzerland); AiDKlinik® (Germany); ID-Pharma® (Germany)).
Finishing the first round, all DDIs with low agreement (criteria see above) were extracted and included in the second round of the Delphi process via video conferencing. Those selected were discussed, and the rating was confirmed or revised. In both steps of the Delphi process, confidentiality of expert’s votes was provided to avoid influencing each other.
Some DDIs are referred to as contraindication by manufactures. For these, a multidisciplinary discussion and individual risk–benefit evaluation and documentation were recommended. It was suggested by our expert panel that the following questions should be used to support the multidisciplinary dialog:
  • Are both drugs indispensable for the individual patient?
  • Is there a therapeutic alternative available for one of the drugs?
  • Is there any feasible monitoring available to early identify AE?

4.2. Interaction Fact Sheets

To guide clinical management of frequently identified relevant DDIs, a detailed review of mechanism, toxicities, management, and monitoring was prepared.

4.3. Data Analysis

The ratings for clinical relevance were ordinal scaled. Therefore, the group ratings were determined via mode. The guidance on preventing or monitoring AE, listed for DDIs in Category 3, was iteratively assigned to five categories. Metric data were displayed with percentages or median and interquartile range, whatever was applicable.

5. Conclusions

Clinically relevant DDIs do appear as a result of polypharmacy in ICU. Especially macrolides, antifungals, and fluoroquinolones are often involved in clinically relevant DDIs and require special attention. However, many clinically relevant DDIs can be avoided by choosing an alternative agent. Others can be attenuated by additional or precautionary monitoring such as TDM or ECG. In addition, regular surveillance of DDIs by a clinical pharmacist as part of the multidisciplinary team in the ICU facilitates other healthcare professionals to manage DDIs and can therefore substantially support patient and drug safety.

Supplementary Materials

The following are available online at https://0-www-mdpi-com.brum.beds.ac.uk/article/10.3390/antibiotics10111330/s1, Table S1: Number of 88 different binary drug–drug interactions listed in ADKA-DokuPIK database at least twice and contained at least one anti-infective agent (total n = 455 interactions). Table S2: DDI (n = 17) including anti-infectives rated not clinically relevant by expert panel. Table S3: DDI (n = 6) containing anti-infectives resulting in low agreement of the expert panel.

Author Contributions

Conceptualization and methodology J.A.K., H.H., A.v.A.-M., D.M., R.W., D.H., and C.K.; formal analysis, investigation, and data curation, J.A.K., and H.H.; resources, H.H.; writing—original draft preparation, J.A.K., H.H., A.v.A.-M., D.M., R.W., A.M., D.H., and C.K.; writing—review and editing, J.A.K., H.H., D.M., A.M., D.H., and C.K.; visualization, J.A.K.; supervision, D.H. and C.K.; project administration, J.A.K., H.H., A.v.A.-M., D.M., R.W., D.H., and C.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding. Submission fees were paid by the German Association of Hospital Pharmacists (ADKA), Berlin, Germany.

Institutional Review Board Statement

Ethical review and approval were waived for this study, due to anonymous patient data entry into the ADKA-DokuPIK database.

Informed Consent Statement

Each participant in the Delphi process had significant contributions to the publication at hand and thus was named as author. Therefore, an informed consent statement was waived.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Acknowledgments

The authors thank C. Langebrake and the Committee of Pharmacist’s intervention of the German Association of Hospital Pharmacists for providing us with the data of the DokuPIK® database. We also thank Nicola Jane Young for her helpful discussion and valuable contributions at the initial stage of the project. The authors wish to thank the German Association of Hospital Pharmacists (ADKA) for financial and intellectual support.

Conflicts of Interest

The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results. J.A.K., H.H., A.v.A.-M., D.M., R.W., A.M., and C.K. have declared no conflict of interest concerning representation or interpretation of reported research results. D.H. has received lecture fees from Pfizer, Astellas, Novartis, and Specialty Diagnostics.

References

  1. Plaza, J.; Alamo, M.; Torres, P.; Fuentes, A.; López, F. Drug interactions and adverse events induced by drugs used in an intensive care unit. Revista Medica de Chile 2010, 138, 452–460. [Google Scholar]
  2. Cvikl, M.; Sinkovič, A. Interventions of a clinical pharmacist in a medical intensive care unit—A retrospective analysis. Bosn. J. Basic Med. Sci. 2020, 20, 495. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  3. Masjedi, M.; Mirjalili, M.; Mirzaei, E.; Mirzaee, H.; Vazin, A. The effect of different intensivist staffing patterns on the rate of potential drug-drug interactions in adult trauma intensive care units. Ther. Adv. Drug Saf. 2020, 11, 2042098620980640. [Google Scholar] [CrossRef]
  4. Fitzmaurice, M.G.; Wong, A.; Akerberg, H.; Avramovska, S.; Smithburger, P.L.; Buckley, M.S.; Kane-Gill, S.L. Evaluation of potential drug-drug interactions in adults in the intensive care unit: A systematic review and meta-analysis. Drug Saf. 2019, 42, 1035–1044. [Google Scholar] [CrossRef] [PubMed]
  5. Zheng, W.Y.; Richardson, L.C.; Li, L.; Day, R.O.; Westbrook, J.I.; Baysari, M.T. Drug-drug interactions and their harmful effects in hospitalised patients: A systematic review and meta-analysis. Eur. J. Clin. Pharmacol. 2018, 74, 15–27. [Google Scholar] [CrossRef] [PubMed]
  6. Bakker, T.; Abu-Hanna, A.; Dongelmans, D.A.; Vermeijden, W.J.; Bosman, R.J.; de Lange, D.W.; Klopotowska, J.E.; de Keizer, N.F.; Hendriks, S.; Ten Cate, J.; et al. Clinically relevant potential drug-drug interactions in intensive care patients: A large retrospective observational multicenter study. J. Crit. Care 2021, 62, 124–130. [Google Scholar] [CrossRef] [PubMed]
  7. Pereira, J.M.; Paiva, J.A. Antimicrobial drug interactions in the critically ill patients. Curr. Clin. Pharmacol. 2013, 8, 25–38. [Google Scholar] [PubMed]
  8. Tanaka, E. Clinically important pharmacokinetic drug-drug interactions: Role of cytochrome P450 enzymes. J. Clin. Pharm. Ther. 1998, 23, 403–416. [Google Scholar] [CrossRef]
  9. Polasek, T.M.; Lin, F.P.; Miners, J.O.; Doogue, M.P. Perpetrators of pharmacokinetic drug-drug interactions arising from altered cytochrome P450 activity: A criteria-based assessment. Br. J. Clin. Pharmacol. 2011, 71, 727–736. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  10. Yu, H.; Zhang, L.; Liu, J.; Liu, Y.; Kowey, P.R.; Zhang, Y.; Chen, Y.; Wei, Y.; Gao, L.; Li, H.; et al. Acquired long QT syndrome in hospitalized patients. Heart Rhythm 2017, 14, 974–978. [Google Scholar] [CrossRef]
  11. Tisdale, J.E.; Jaynes, H.A.; Kingery, J.R.; Mourad, N.A.; Trujillo, T.N.; Overholser, B.R.; Kovacs, R.J. Development and validation of a risk score to predict QT interval prolongation in hospitalized patients. Circulation 2013, 6, 479–487. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  12. Su, K.; McGloin, R.; Gellatly, R.M. Predictive validity of a QTc interval prolongation risk score in the intensive care unit. Pharmacotherapy 2020, 40, 492–499. [Google Scholar] [CrossRef] [PubMed]
  13. Scotton, W.J.; Hill, L.J.; Williams, A.C.; Barnes, N.M. Serotonin syndrome: Pathophysiology, clinical features, management, and potential future directions. Int. J. Tryptophan. Res. 2019, 12, 1178646919873925. [Google Scholar] [CrossRef] [Green Version]
  14. Werneke, U.; Truedson-Martiniussen, P.; Wikström, H.; Ott, M. Serotonin syndrome: A clinical review of current controversies. J. Integr. Neurosci. 2020, 19, 719–727. [Google Scholar] [CrossRef] [PubMed]
  15. Bakker, T.; Klopotowska, J.E.; de Keizer, N.F.; van Marum, R.; van der Sijs, H.; de Lange, D.W.; de Jonge, E.; Abu-Hanna, A.; Dongelmans, D.A. Improving medication safety in the Intensive Care by identifying relevant drug-drug interactions—Results of a multicenter Delphi study. J. Crit. Care 2020, 57, 134–140. [Google Scholar] [CrossRef] [PubMed]
  16. Ceftriaxone (rocephin): Incompatible with Solutions Containing Calcium. Available online: https://www.gov.uk/drug-safety-update/ceftriaxone-rocephin-incompatible-with-solutions-containing-calcium (accessed on 30 September 2021).
  17. Avci, Z.; Koktener, A.; Uras, N.; Catal, F.; Karadag, A.; Tekin, O.; Degirmencioglu, H.; Baskin, E. Nephrolithiasis associated with ceftriaxone therapy: A prospective study in 51 children. Arch. Dis. Child. 2004, 89, 1069–1072. [Google Scholar] [CrossRef] [Green Version]
  18. Voeten, M.; Landstra, A.M.; Maseland, M.H.; van Setten, P.A. Serious side effects of frequently used antibiotics in childhood: Biliary sludge or stones induced by ceftriaxone and thrombocytopenia induced by co-trimoxazole. Nederlands Tijdschrift voor Geneeskunde 2007, 151, 1299–1303. [Google Scholar] [PubMed]
  19. Bartkowska-Śniatkowska, A.; Jończyk-Potoczna, K.; Zielińska, M.; Rosada-Kurasińska, J. Adverse reaction to ceftriaxone in a 28-day-old infant undergoing urgent craniotomy due to epidural hematoma: Review of neonatal biliary pseudolithiasis. Ther. Clin. Risk Manag. 2015, 11, 1035–1041. [Google Scholar] [CrossRef] [Green Version]
  20. Al-Quteimat, O.; Laila, A. Valproate interaction with carbapenems: Review and recommendations. Hosp. Pharm. 2020, 55, 181–187. [Google Scholar] [CrossRef] [PubMed]
  21. Carnovale, C.; Pozzi, M.; Mazhar, F.; Mosini, G.; Gentili, M.; Peeters, G.G.; Clementi, E.; Radice, S. Interactions between antiepileptic and antibiotic drugs: A systematic review and meta-analysis with dosing implications. Clin. Pharmacokinet. 2019, 58, 875–886. [Google Scholar] [CrossRef]
  22. Chai, P.Y.-C.; Chang, C.-T.; Chen, Y.-H.; Chen, H.-Y.; Tam, K.-W. Effect of drug interactions between carbapenems and valproate on serum valproate concentration: A systematic review and meta-analysis. Expert Opin. Drug Saf. 2021, 20, 215–223. [Google Scholar] [CrossRef] [PubMed]
  23. Kojima, S.; Nadai, M.; Kitaichi, K.; Wang, L.; Nabeshima, T.; Hasegawa, T. Possible mechanism by which the carbapenem antibiotic panipenem decreases the concentration of valproic acid in plasma in rats. Antimicrob. Agents Chemother. 1998, 42, 3136–3140. [Google Scholar] [CrossRef] [Green Version]
  24. Yokogawa, K.; Iwashita, S.; Kubota, A.; Sasaki, Y.; Ishizaki, J.; Kawahara, M.; Matsushita, R.; Kimura, K.; Ichimura, F.; Miyamoto, K.I. Effect of meropenem on disposition kinetics of valproate and its metabolites in rabbits. Pharm. Res. 2001, 18, 1320–1326. [Google Scholar] [CrossRef] [PubMed]
  25. Nakajima, Y.; Mizobuchi, M.; Nakamura, M.; Takagi, H.; Inagaki, H.; Kominami, G.; Koike, M.; Yamaguchi, T. Mechanism of the drug interaction between valproic acid and carbapenem antibiotics in monkeys and rats. Drug Metab. Dispos. 2004, 32, 1383–1391. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  26. Yamamura, N.; Imura, K.; Naganuma, H.; Nishimura, K. Panipenem, a carbapenem antibiotic, enhances the glucuronidation of intravenously administered valproic acid in rats. Drug Metab. Dispos. 1999, 27, 724–730. [Google Scholar]
  27. Suzuki, E.; Nakai, D.; Ikenaga, H.; Fusegawa, K.; Goda, R.; Kobayashi, N.; Kuga, H.; Izumi, T. In vivo inhibition of acylpeptide hydrolase by carbapenem antibiotics causes the decrease of plasma concentration of valproic acid in dogs. Xenobiotica 2016, 46, 126–131. [Google Scholar] [CrossRef]
  28. Omoda, K.; Murakami, T.; Yumoto, R.; Nagai, J.; Maeda, Y.; Kiribayashi, Y.; Takano, M. Increased erythrocyte distribution of valproic acid in pharmacokinetic interaction with carbapenem antibiotics in rat and human. J. Pharm. Sci. 2005, 94, 1685–1693. [Google Scholar] [CrossRef] [PubMed]
  29. Ogawa, K.; Yumoto, R.; Hamada, N.; Nagai, J.; Takano, M. Interaction of valproic acid and carbapenem antibiotics with multidrug resistance-associated proteins in rat erythrocyte membranes. Epilepsy Res. 2006, 71, 76–87. [Google Scholar] [CrossRef] [PubMed]
  30. Huang, C.-R.; Lin, C.-H.; Hsiao, S.-C.; Chen, N.-C.; Tsai, W.-C.; Chen, S.-D.; Lu, Y.-T.; Chuang, Y.-C. Drug interaction between valproic acid and carbapenems in patients with epileptic seizures. Kaohsiung J. Med Sci. 2017, 33, 130–136. [Google Scholar] [CrossRef] [PubMed]
  31. Wu, C.-C.; Pai, T.-Y.; Hsiao, F.-Y.; Shen, L.-J.; Wu, F.-L.L. The effect of different carbapenem antibiotics (ertapenem, imipenem/cilastatin, and meropenem) on serum valproic acid concentrations. Ther. Drug Monit. 2016, 38, 587–592. [Google Scholar] [CrossRef] [PubMed]
  32. Bates, D.; Parkins, M.; Duggan, K. Ertapenem-induced reduction in valproate levels: Case report and review of the literature. Can. J. Hosp. Pharm. 2010, 63, 315. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  33. Bede, P.; Lawlor, D.; Solanki, D.; Delanty, N. Carbapenems and valproate: A consumptive relationship. Epilepsia Open 2017, 2, 107–111. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  34. Khatib, R.; Sabir, F.R.N.; Omari, C.; Pepper, C.; Tayebjee, M.H. Managing drug-induced QT prolongation in clinical practice. Postgrad. Med. J. 2021, 97, 452–458. [Google Scholar] [CrossRef]
  35. Armahizer, M.J.; Seybert, A.L.; Smithburger, P.L.; Kane-Gill, S.L. Drug-drug interactions contributing to QT prolongation in cardiac intensive care units. J. Crit. Care 2013, 28, 243–249. [Google Scholar] [CrossRef]
  36. Wiśniowska, B.; Tylutki, Z.; Wyszogrodzka, G.; Polak, S. Drug-drug interactions and QT prolongation as a commonly assessed cardiac effect—Comprehensive overview of clinical trials. BMC Pharmacol. Toxicol. 2016, 17, 12. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  37. Kantola, T.; Kivistö, K.T.; Neuvonen, P.J. Erythromycin and verapamil considerably increase serum simvastatin and simvastatin acid concentrations. Clin. Pharmacol. Ther. 1998, 64, 177–182. [Google Scholar] [CrossRef]
  38. Grunden, J.W.; Fisher, K.A. Lovastatin-induced rhabdomyolysis possibly associated with clarithromycin and azithromycin. Ann. Pharmacother. 1997, 31, 859–863. [Google Scholar] [CrossRef] [PubMed]
  39. Westphal, J.F. Macrolide–induced clinically relevant drug interactions with cytochrome P-450A (CYP) 3A4: An update focused on clarithromycin, azithromycin and dirithromycin. Br. J. Clin. Pharmacol. 2000, 50, 285–295. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  40. Yee, G.C.; McGuire, T.R. Pharmacokinetic drug interactions with cyclosporin (Part I). Clin. Pharmacokinet. 1990, 19, 319–332. [Google Scholar] [CrossRef]
  41. Salimi, A.; Eybagi, S.; Seydi, E.; Naserzadeh, P.; Kazerouni, N.P.; Pourahmad, J. Toxicity of macrolide antibiotics on isolated heart mitochondria: A justification for their cardiotoxic adverse effect. Xenobiotica 2016, 46, 82–93. [Google Scholar] [CrossRef] [PubMed]
  42. Volberg, W.A.; Koci, B.J.; Su, W.; Lin, J.; Zhou, J. Blockade of human cardiac potassium channel human ether-a-go-go-related gene (HERG) by macrolide antibiotics. J. Pharmacol. Exp. Ther. 2002, 302, 320–327. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  43. Milberg, P.; Eckardt, L.; Bruns, H.J.; Biertz, J.; Ramtin, S.; Reinsch, N.; Fleischer, D.; Kirchhof, P.; Fabritz, L.; Breithardt, G.; et al. Divergent proarrhythmic potential of macrolide antibiotics despite similar QT prolongation: Fast phase 3 repolarization prevents early afterdepolarizations and torsade de pointes. J. Pharmacol. Exp. Ther. 2002, 303, 218–225. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  44. Gorelik, E.; Masarwa, R.; Perlman, A.; Rotshild, V.; Muszkat, M.; Matok, I. Systematic review, meta-analysis, and network meta-analysis of the cardiovascular safety of macrolides. Antimicrob. Agents Chemother. 2018, 62, e00438-18. [Google Scholar] [CrossRef] [Green Version]
  45. Lewis, R.E.; Beyda, N.D.; Kontoyiannis, D.P. Caspofungin. In Kucers’ The Use of Antibiotics: A Clinical Review of Antibacterial, Antifungal, Antiparasitic, and Antiviral Drugs, 7th ed.; Grayson, M.L., Cosgrove, S.E., Crowe, S., Eds.; CRC Press: Boca Raton, FL, USA, 2017; Volume 7, pp. 2659–2680. [Google Scholar]
  46. Preston, C.L. (Ed.) Stockley’s Drug Interactions; Phamaceutical Press: London, UK, 2019; Volume 12. [Google Scholar]
  47. Groetzner, J.; Kaczmarek, I.; Wittwer, T.; Strauch, J.; Meiser, B.; Wahlers, T.; Daebritz, S.; Reichart, B. Caspofungin as first-line therapy for the treatment of invasive aspergillosis after thoracic organ transplantation. J. Heart Lung Transplant. 2008, 27, 1–6. [Google Scholar] [CrossRef] [PubMed]
  48. Lomaestro, B.M.; Piatek, M.A. Update on drug interactions with azole antifungal agents. Ann. Pharmacother. 1998, 32, 915–928. [Google Scholar] [CrossRef] [PubMed]
  49. Toda, F.; Tanabe, K.; Ito, S.; Shinmura, H.; Tokumoto, T.; Ishida, H.; Toma, H. Tacrolimus trough level adjustment after administration of fluconazole to kidney recipients. Transplant. Proc. 2002, 34, 1733–1735. [Google Scholar] [CrossRef]
  50. Sugar, A.M.; Saunders, C.; Idelson, B.A.; Bernard, D.B. Interaction of fluconazole and cyclosporine. Ann. Intern. Med. 1989, 110, 844. [Google Scholar] [CrossRef] [PubMed]
  51. Mihara, A.; Mori, T.; Aisa, Y.; Yamazaki, R.; Iketani, O.; Tanigawara, Y.; Ikeda, Y.; Okamoto, S. Greater impact of oral fluconazole on drug interaction with intravenous calcineurin inhibitors as compared with intravenous fluconazole. Eur. J. Clin. Pharmacol. 2008, 64, 89–91. [Google Scholar] [CrossRef]
  52. Muirhead, G.J.; Wulff, M.B.; Fielding, A.; Kleinermans, D.; Buss, N. Pharmacokinetic interactions between sildenafil and saquinavir/ritonavir. Br. J. Clin. Pharmacol. 2000, 50, 99–107. [Google Scholar] [CrossRef]
  53. Mihailescu, R.; Furustrand Tafin, U.; Corvec, S.; Oliva, A.; Betrisey, B.; Borens, O.; Trampuz, A. High activity of Fosfomycin and Rifampin against methicillin-resistant staphylococcus aureus biofilm in vitro and in an experimental foreign-body infection model. Antimicrob. Agents Chemother. 2014, 58, 2547–2553. [Google Scholar] [CrossRef] [Green Version]
  54. Ullmann, A.J.; Aguado, J.M.; Arikan-Akdagli, S.; Denning, D.W.; Groll, A.H.; Lagrou, K.; Lass-Flörl, C.; Lewis, R.E.; Munoz, P.; Verweij, P.E.; et al. Diagnosis and management of Aspergillus diseases: Executive summary of the 2017 ESCMID-ECMM-ERS guideline. Clin. Microbiol. Infect. 2018, 24, e1–e38. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  55. Alffenaar, J.W.; van Assen, S.; van der Werf, T.S.; Kosterink, J.G.; Uges, D.R. Omeprazole significantly reduces posaconazole serum trough level. Clin. Infect. Dis. 2009, 48, 839. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  56. Kraft, W.K.; Chang, P.S.; van Iersel, M.L.; Waskin, H.; Krishna, G.; Kersemaekers, W.M. Posaconazole tablet pharmacokinetics: Lack of effect of concomitant medications altering gastric pH and gastric motility in healthy subjects. Antimicrob. Agents Chemother. 2014, 58, 4020–4025. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  57. Dare, R.K.; Tewell, C.; Harris, B.; Wright, P.W.; Van Driest, S.L.; Farber-Eger, E.; Nelson, G.E.; Talbot, T.R. Effect of statin coadministration on the risk of daptomycin-associated myopathy. Clin. Infect. Dis. 2018, 67, 1356–1363. [Google Scholar] [CrossRef]
  58. Bhavnani, S.M.; Rubino, C.M.; Ambrose, P.G.; Drusano, G.L. Daptomycin exposure and the probability of elevations in the creatine phosphokinase level: Data from a randomized trial of patients with bacteremia and endocarditis. Clin. Infect. Dis. 2010, 50, 1568–1574. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  59. Galar, A.; Muñoz, P.; Valerio, M.; Cercenado, E.; García-González, X.; Burillo, A.; Sánchez-Somolinos, M.; Juárez, M.; Verde, E.; Bouza, E. Current use of daptomycin and systematic therapeutic drug monitoring: Clinical experience in a tertiary care institution. Int. J. Antimicrob. Agents 2019, 53, 40–48. [Google Scholar] [CrossRef]
  60. Finch, C.K.; Chrisman, C.R.; Baciewicz, A.M.; Self, T.H. Rifampin and rifabutin drug interactions: An update. Arch. Intern. Med. 2002, 162, 985–992. [Google Scholar] [CrossRef] [PubMed]
  61. Kyrklund, C.; Backman, J.T.; Kivistö, K.T.; Neuvonen, M.; Laitila, J.; Neuvonen, P.J. Rifampin greatly reduces plasma simvastatin and simvastatin acid concentrations. Clin. Pharmacol. Ther. 2000, 68, 592–597. [Google Scholar] [CrossRef] [PubMed]
  62. Teng, R.; Mitchell, P.; Butler, K. Effect of rifampicin on the pharmacokinetics and pharmacodynamics of ticagrelor in healthy subjects. Eur. J. Clin. Pharmacol. 2013, 69, 877–883. [Google Scholar] [CrossRef]
  63. Wang, Z.Y.; Chen, M.; Zhu, L.L.; Yu, L.S.; Zeng, S.; Xiang, M.X.; Zhou, Q. Pharmacokinetic drug interactions with clopidogrel: Updated review and risk management in combination therapy. Ther. Clin. Risk Manag. 2015, 11, 449–467. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  64. Farid, N.A.; Jakubowski, J.A.; Payne, C.D.; Li, Y.G.; Jin, Y.; Ernest, I.C.; Winters, K.J.; Brandt, J.T.; Salazar, D.E.; Small, D.S. Effect of rifampin on the pharmacokinetics and pharmacodynamics of prasugrel in healthy male subjects. Curr. Med. Res. Opin. 2009, 25, 1821–1829. [Google Scholar] [CrossRef] [PubMed]
  65. Moellering, R.C. Linezolid: The first oxazolidinone antimicrobial. Ann. Intern. Med. 2003, 138, 135–142. [Google Scholar] [CrossRef] [PubMed]
  66. Hernández, J.L.; Ramos, F.J.; Infante, J.; Rebollo, M.; González-Macías, J. Severe serotonin syndrome induced by mirtazapine monotherapy. Ann. Pharmacother. 2002, 36, 641–643. [Google Scholar] [CrossRef] [PubMed]
  67. Ubogu, E.E.; Katirji, B. Mirtazapine-induced serotonin syndrome. Clin. Neuropharmacol. 2003, 26, 54–57. [Google Scholar] [CrossRef] [PubMed]
  68. DeBellis, R.J.; Schaefer, O.P.; Liquori, M.; Volturo, G.A. Linezolid-associated serotonin syndrome after concomitant treatment with citalopram and mirtazepine in a critically ill bone marrow transplant recipient. J. Intensive Care Med. 2005, 20, 351–353. [Google Scholar] [CrossRef] [PubMed]
  69. Foong, A.L.; Grindrod, K.A.; Patel, T.; Kellar, J. Demystifying serotonin syndrome (or serotonin toxicity). Can. Fam. Physician 2018, 64, 720–727. [Google Scholar] [PubMed]
  70. Boyer, E.W.; Shannon, M. The serotonin syndrome. N. Engl. J. Med. 2005, 352, 1112–1120. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  71. Lodes, U.; Troeger, U.; Neuling, F.; Jacob, D.; Meyer, F. Drug Interaction Stewardship (DIS) and therapeutic drug monitoring for anti-infective treatment in intensive care, a single centre observational study. Zentralblatt fur Chirurgie 2020, 145, 176–187. [Google Scholar] [CrossRef] [PubMed]
  72. Ihbe-Heffinger, A.; Langebrake, C.; Hohmann, C.; Leichenberg, K.; Hilgarth, H.; Kunkel, M.; Lueb, M.; Schuster, T. Prospective survey-based study on the categorization quality of hospital pharmacists’ interventions using DokuPIK. Int. J. Clin. Pharm. 2019, 41, 414–423. [Google Scholar] [CrossRef] [PubMed]
  73. Loj, P.; Olender, A.; Slezak, W.; Krzych, L.J. Pharmacokinetic drug-drug interactions in the intensive care unit—single-centre experience and literature review. Anaesthesiol. Intensive Ther. 2017, 49, 259–267. [Google Scholar] [CrossRef] [Green Version]
  74. Lin, J.H.; Yamazaki, M. Role of P-glycoprotein in pharmacokinetics: Clinical implications. Clin. Pharmacokinet. 2003, 42, 59–98. [Google Scholar] [CrossRef] [PubMed]
  75. Zhou, S.; Yung Chan, S.; Cher Goh, B.; Chan, E.; Duan, W.; Huang, M.; McLeod, H.L. Mechanism-based inhibition of cytochrome P450 3A4 by therapeutic drugs. Clin. Pharmacokinet. 2005, 44, 279–304. [Google Scholar] [CrossRef] [PubMed]
  76. Bertz, R.J.; Granneman, G.R. Use of in vitro and in vivo data to estimate the likelihood of metabolic pharmacokinetic interactions. Clin. Pharmacokinet. 1997, 32, 210–258. [Google Scholar] [CrossRef] [PubMed]
  77. Koinuma, K.; Tsuchitani, T.; Imaoka, A.; Akiyoshi, T.; Ohtani, H. Relative contributions of metabolic enzymes to systemic elimination can be estimated from clinical DDI studies: Validation using an in silico approach. Int. J. Clin. Pharmacol. Ther. 2021, 59, 231–238. [Google Scholar] [CrossRef]
  78. Langebrake, C.; Ihbe-Heffinger, A.; Leichenberg, K.; Kaden, S.; Kunkel, M.; Lueb, M.; Hilgarth, H.; Hohmann, C. Nationwide evaluation of day-to-day clinical pharmacists’ interventions in german hospitals. Pharmacotherapy 2015, 35, 370–379. [Google Scholar] [CrossRef] [PubMed]
  79. Langebrake, C.; Hohmann, C.; Lezius, S.; Lueb, M.; Picksak, G.; Walter, W.; Kaden, S.; Hilgarth, H.; Ihbe-Heffinger, A.; Leichenberg, K. Clinical pharmacists’ interventions across German hospitals: Results from a repetitive cross-sectional study. Int. J. Clin. Pharm. 2021, 1–8, online ahead of print. [Google Scholar] [CrossRef]
  80. ATC/DDD Index 2021. Available online: https://www.whocc.no/atc_ddd_index/ (accessed on 30 September 2021).
  81. Boulkedid, R.; Abdoul, H.; Loustau, M.; Sibony, O.; Alberti, C. Using and reporting the delphi method for selecting healthcare quality indicators: A systematic review. PLoS ONE 2011, 6, e20476. [Google Scholar] [CrossRef]
Antibiotics 10 01330 g001 550
Figure 1. Selection of binary DDIs out of pharmacist’s interventions on German intensive care units recorded in the ADKA-DokuPIK database. DDI = drug–drug interaction.
Figure 1. Selection of binary DDIs out of pharmacist’s interventions on German intensive care units recorded in the ADKA-DokuPIK database. DDI = drug–drug interaction.
Antibiotics 10 01330 g001
Table
Table 1. Ratings of the expert panel.
Table 1. Ratings of the expert panel.
GroupClinically Relevant (n = 65)Not Clinically Relevant or Low Agreement (n = 23)Total (n = 88)
Penicillins011
Cephalosporins112
Carbapenems202
Fluoroquinolones15116
Macrolides21829
Glycopeptides011
Antifungals19322
Miscellaneous7815
Table
Table 2. DDIs (n = 65) including anti-infectives and rated clinically relevant by expert panel.
Table 2. DDIs (n = 65) including anti-infectives and rated clinically relevant by expert panel.
GroupDrug 1Drug 2Mode CategoryAdditional Strategies to Reduce Patient Risk from Interaction
CephalosporinsCeftriaxoneCalcium (intravenous)3separate administration
CarbapenemsImipenemValproic acid5TDM for valproic acid
MeropenemValproic acid5
FluoroquinolonesCiprofloxacinCalcium3separate administration by two hours, or use different administration routes (oral/intravenous)
Amiodarone3QTc-monitoring, high-normal serum levels of potassium and magnesium
Mirtazapine3
Amitriptyline3
Haloperidol3
Theophylline3QTc-monitoring, high-normal serum levels of potassium and magnesium, TDM for theophylline, use 60% of regular theophylline dose
Iron3separate administration by two hours, or use different administration routes (oral/intravenous)
Melatonin3monitor for sleepiness
Simvastatin3CK-monitoring, monitor for signs of myalgia
LevofloxacinAmiodarone3QTc-monitoring, high-normal serum levels of potassium and magnesium
LevofloxacinHaloperidol3
MoxifloxacinAmiodarone3
Mirtazapine3
Prednisolone4monitor for signs of tendinopathy
MacrolidesAzithromycinCitalopram3QTc-monitoring, high-normal serum levels of potassium and magnesium
Haloperidol3
ClarithromycinSimvastatin5withhold/switch statin
Atorvastatin5CK-monitoring, monitor for signs of myalgia, withhold/switch statin
Tacrolimus3TDM for tacrolimus
Cyclosporine3TDM for cyclosporine
Carbamazepine3where appropriate: switch clarithromycin to azithromycin,
TDM for carbamazepine
Amiodarone3QTc-monitoring, high-normal serum levels of potassium and magnesium
Haloperidol3
Ivabradine5QTc-monitoring, high-normal serum levels of potassium and magnesium, monitor for bradycardia
Moxifloxacin3QTc-monitoring, high-normal serum levels of potassium and magnesium
Theophylline3QTc-monitoring, high-normal serum levels of potassium and magnesium, TDM for theophylline
ErythromycinSimvastatin5withhold/switch statin
Atorvastatin5CK-monitoring, monitor for signs of myalgia, withhold/switch statin
Quetiapine3QTc-monitoring, high-normal serum levels of potassium and magnesium
Tacrolimus3TDM for Tacrolimus
Amiodaron3QTc-monitoring, high-normal serum levels of potassium and magnesium
Carbamazepine5where appropriate: switch erythromycin to azithromycin,
TDM for carbamazepine
Cyclosporine3where appropriate: switch erythromycin to azithromycin,
TDM for cyclosporine
ErythromycinHaloperidol3QTc-monitoring, high-normal serum levels of potassium and magnesium, where appropriate: switch erythromycin to azithromycin
Melperon3QTc-monitoring, high-normal serum levels of potassium and magnesium
AntifungalsCaspofunginCyclosporine3TDM for cyclosporine
FluconazoleCyclosporine3
Tacrolimus3TDM for tacrolimus
Simvastatin5CK-monitoring, monitor for signs of myalgia, withhold/switch statin
Atorvastatin3
Carbamazepine5TDM for carbamazepine
Rifampicin3discuss alternatives
ItraconazoleSimvastatin5withhold/switch statin
PosaconazolePantoprazol3intravenous administration of posaconazole, TDM for posaconazole
Cyclosporine3TDM posaconazole and cyclosporine
Atorvastatin5withhold/switch statin
Carbamazepine3TDM carbamazepine
Tacrolimus3TDM posaconazole and tacrolimus
VoriconazoleTacrolimus3TDM voriconazole and tacrolimus
Simvastatin5withhold/switch statin
Atorvastatin5
Cyclosporine3TDM for voriconazole and cyclosporine
Rifampicin5TDM for voriconazole, discuss alternatives
Sildenafil3dose reduction of sildenafil
MiscellaneousDaptomycinAtorvastatin3CK-monitoring, monitor for signs of myalgia, withhold/switch statin
Simvastatin3
LinezolidMirtazapine3monitor for signs of serotonin syndrome
RifampicinCyclosporine3TDM for cyclosporine
Clopidogrel5monitor for signs of bleeding, discuss alternatives
Simvastatin5withhold/switch statin
Ticagrelor5discuss alternatives
Mode Category 3: “Clinically relevant, the adverse effects of this DDI can however be limited by additional monitoring and/or changes in dosage/frequency/timing.” Mode Category 4: “Clinically relevant, the adverse effects of this DDI on the patient can be substantial, however these effects are acceptable and treatable.” Mode Category 5: “Clinically relevant, the adverse effect of this DDI on the patient should preferably be avoided.” [15]. CK = creatine kinase; TDM = therapeutic drug monitoring; QTc = Rate-corrected QT interval.
Table
Table 3. Metabolism and CYP inhibition of currently marketed azoles.
Table 3. Metabolism and CYP inhibition of currently marketed azoles.
AzoleWeak InhibitionModerate InhibitionStrong InhibitionSubstrate
Itraconazole CYP 3A4CYP3A4
Fluconazole CYP 2C9
CYP 3A4		CYP 2C19
VoriconazoleCYP 2C9CYP 2C19CYP 3A4CYP 3A4
CYP 2C19
CYP 2C9
Posaconazole CYP 3A4P-gp
Isavuconazole CYP 3A4 CYP 3A4
CYP = Cytochrome P-450; P-gp = P-glycoprotein.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Koeck, J.A.; Hilgarth, H.; von Ameln-Mayerhofer, A.; Meyn, D.; Warlich, R.; Münstedt, A.; Horn, D.; König, C. Clinically Relevant Interactions with Anti-Infectives on Intensive Care Units—A Multicenter Delphi Study. Antibiotics 2021, 10, 1330. https://0-doi-org.brum.beds.ac.uk/10.3390/antibiotics10111330

AMA Style

Koeck JA, Hilgarth H, von Ameln-Mayerhofer A, Meyn D, Warlich R, Münstedt A, Horn D, König C. Clinically Relevant Interactions with Anti-Infectives on Intensive Care Units—A Multicenter Delphi Study. Antibiotics. 2021; 10(11):1330. https://0-doi-org.brum.beds.ac.uk/10.3390/antibiotics10111330

Chicago/Turabian Style

Koeck, Joachim Andreas, Heike Hilgarth, Andreas von Ameln-Mayerhofer, Damaris Meyn, Ruediger Warlich, Andreas Münstedt, Dagmar Horn, and Christina König. 2021. "Clinically Relevant Interactions with Anti-Infectives on Intensive Care Units—A Multicenter Delphi Study" Antibiotics 10, no. 11: 1330. https://0-doi-org.brum.beds.ac.uk/10.3390/antibiotics10111330

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop