Drug management in acute kidney disease – Report of the Acute Disease Quality Initiative XVI meeting

Abstract

Aims

To summarize and extend the main conclusions and recommendations relevant to drug management during acute kidney disease (AKD) as agreed at the 16^th Acute Disease Quality Initiative (ADQI) consensus conference.

Methods

Using a modified Delphi method to achieve consensus, experts attending the 16^th ADQI consensus conference reviewed and appraised the existing literature on drug management during AKD and identified recommendations for clinical practice and future research. The group focussed on drugs with one of the following characteristics: (i) predominant renal excretion; (ii) nephrotoxicity; (iii) potential to alter glomerular function; and (iv) presence of metabolites that are modified in AKD and may affect other organs.

Results

We recommend that medication reconciliation should occur at admission and discharge, at AKD diagnosis and change in AKD phase, and when the patient's condition changes. Strategies to avoid adverse drug reactions in AKD should seek to minimize adverse events from overdosing and nephrotoxicity and therapeutic failure from under‐dosing or incorrect drug selection. Medication regimen assessment or introduction of medications during the AKD period should consider the nephrotoxic potential, altered renal and nonrenal elimination, the effects of toxic metabolites and drug interactions and altered pharmacodynamics in AKD. A dynamic monitoring plan including repeated serial assessment of clinical features, utilization of renal diagnostic tests and therapeutic drug monitoring should be used to guide medication regimen assessment.

Conclusions

Drug management during different phases of AKD requires an individualized approach and frequent re‐assessment. More research is needed to avoid drug associated harm and therapeutic failure.

Introduction

Drug dosing in critically ill patients is challenging due to altered pharmacokinetics and pharmacodynamics, multiple drug interactions and limited evidence to guide accurate prescribing 1. Patients with impaired kidney function are particularly at risk for under‐ and over‐dosing 2, 3, 4. Guidelines are available for the drug management in patients with acute kidney injury (AKI) 5 but little is known for patients with acute kidney disease (AKD). The term AKD refers to a condition in which the renal pathophysiological processes are still ongoing and AKI stage 1 or greater [as defined by the Kidney Disease Improving Global Outcome (KDIGO) criteria] is present ≥7 days after an AKI initiating event 6. AKD that persists beyond 90 days is considered to be chronic kidney disease (CKD). Often, there is a continuum from AKI to AKD and CKD 6. Typical scenarios are cases where AKI is observed and the patient remains in AKI stage 1 or greater for >7 days. Another option is a patient who develops AKI during hospitalization and their serum creatinine improves before discharge but the AKD pathophysiological processes continue leading to CKD at 90 days 7. There may also be instances where the onset of AKI was not directly observed (i.e., community acquired AKI) and patients present with impaired renal function that has been present for ≥7 days but <90 days. Conceptually, AKD consists of three phases: deteriorating phase when kidney function is actively worsening; maintenance phase when the injury has ended but kidney function has yet to recover; and improvement phase where renal function begins to recover, ideally towards the preinjury baseline 8. (Figure 1) The duration of these individual phases varies and not every patient progresses through all three phases. Renal function often fluctuates with a variable rate of renal recovery and it can be difficult to determine when sustained recovery has occurred. As a result, patients are at high risk for over‐dosing, under‐dosing and nephrotoxicity 3, 4, 9. Finally, AKI and AKD are syndromes comprising multiple different aetiologies. They rarely occur in isolation but usually in the context of other dynamic acute illnesses and on the background of profound chronic comorbidities, all of which impact the selection of drugs and their pharmacokinetics and pharmacodynamics.

Figure 1.

Drug management during different phases of acute kidney disease 6. AKD = acute kidney disease. (A) AKD begins to improve early in the clinical course; (B) AKD persists for longer and renal function improves only after a considerable decline; (C) AKD is severe; after an extended period of AKD, renal function may recover fully to baseline or only partially or progress to chronic dialysis dependence

The 16th Acute Disease Quality Initiative (ADQI) consensus conference focused on clinically important aspects during the period of AKD 6. The key recommendations related to the definition of renal recovery, criteria for AKD, assessment of renal function and clinical management were published in early 2017. In this report, we summarize and extend the main conclusions and recommendations relevant to drug management.

Methods

The 16^th ADQI conference was held over 2.5 days in San Diego, USA, and included a diverse panel of clinicians and researchers representing relevant disciplines – internal medicine, paediatrics, primary care, nephrology, critical care, pharmacy, epidemiology, health‐services research, biostatistics, bioinformatics and data analytics – from Europe, North America and Australia 8. The meeting followed the established ADQI process, and used a modified Delphi method to achieve consensus 10. Members of the work groups performed reviews of the literature in a systematic manner and developed a consensus of opinion, backed by evidence where possible, to distil the available literature and identify recommendations for clinical practice and future research.

While the considerations for drug dosing during AKD are relevant to any medication, it was decided to focus on main principles and specific issues related to drugs with one of the following characteristics: (i) predominant renal excretion; (ii) nephrotoxicity; (iii) potential to alter glomerular function; and (iv) presence of metabolites that are modified in AKI/AKD and may affect other organs.

Results

General principles of drug selection, dosing and monitoring during AKD

The disposition and effects of drugs given to the patient with AKD are modulated by several factors, including changes in drug clearance (glomerular and tubular kidney function; nonrenal drug metabolism), and altered pharmacokinetic parameters as a result of decreased kidney function (volume overload, metabolic acidosis etc.). Understanding how a particular drug or metabolite is handled is essential to formulate a therapeutic plan for a specific agent across every phase of AKD (deteriorating, maintenance or improvement phase; Figure 1). Depending on the drug, renal elimination may occur through glomerular filtration, tubular secretion or tubular reabsorption (following glomerular filtration). In addition, the kidney may oxidize, reduce, hydrolyse and/or conjugate a particular drug or its metabolite leading to drug activation/inactivation 1. There may also be additional contribution to drug metabolism through the action of renal tubular cytochrome P450 (CYP450) 11. It is crucial to know what phase of AKD (deteriorating, maintenance or improvement) a patient is in and whether changes in function (e.g. glomerular filtration) and/or structural injury to the renal tubules have occurred and the specific site of injury (as measured by damage biomarkers) 12, 13. These details are vital to understand how the different phases of AKD may impact drug dosing.

In conclusion, drug selection and dosing in AKD requires multiple and regular re‐assessment considering patient risk profile, baseline and actual renal function and trajectories, drug mechanisms, biomarkers predicating injury and recovery, nonrenal dysfunction affecting drug clearance, indications and alternative therapies, genetic susceptibility and potential availability of therapeutic drug monitoring 6.

Assessment of renal function during AKD

Functional markers of glomerular filtration (e.g. serum creatinine or cystatin C) are good biomarkers during steady states, but are not ideally suited for acute changes 14, 15, 16. Other methods to assess filtration include timed urinary collection(s) in order to estimate creatinine and urea clearances 17, 18. Multiple kinetic algorithms have been proposed to predict glomerular filtration rate (GFR) in the nonsteady‐state 19, 20, 21, 22, 23, but none is universally accepted. Several models of real‐time GFR have also been investigated including real‐time plasma elimination kinetics of fluorescein isothiocyanate (FITC)‐sinistrin and fluorescence based measures of filterable (inulin) and nonfilterable (500 kDa dextrans) molecules 24, 25. While real‐time GFR measurement may be the ideal biological measure in the setting of AKD, these techniques have yet to be approved for use following validation in humans.

Potential non‐GFR based imaging modalities as measures of renal function in AKD include contrast enhanced ultrasound, positron emission tomography (PET), and blood oxygenation level‐dependent magnetic resonance imaging 26. Similarly, renal blood flow and renal tissue oxygenation assessments have been suggested as potential indicators of underlying intrinsic kidney function 27 but the role of these new techniques outside the research setting remains uncertain.

Novel biomarkers of renal tubular injury have been approved for clinical use around the globe 28, 29, 30, 31, 32, 33, 34, 35. They predict both short and long‐term outcomes after AKI 12, 13, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45 and have also been shown to direct clinical management 46, 47, 48. Although their role in guiding drug dosing remains unclear, it is easy to imagine clinical scenarios where changes in nephron segment‐specific damage biomarkers alert the prescriber that dose adjustments, including drug discontinuation, are needed for drugs handled by specific nephron segments 49. Examples of biomarkers and drugs specific to nephron segments are provided in Table 1.

Table 1.

Examples of biomarkers and drugs specific to nephron segments

Nephron segment	Segment specific biomarker	Drugs handled by relevant nephron segment
Glomerulus	Albumin Cystatin C	Doxrubicin Gold Penicillamine
Proximal tubules	Albumin Clusterin IGFBP7 KIM‐1 L‐FABP NAG NGAL Osteropontin TIMP‐2 Cystatin C	Amikacin Cisplatin Colistin Cyclosporine Gentamicin Hydroxyetheyl starch Tacrolimus Tobramycin Vancomycin
Loop of Henle	Osteopontin	Analgesics (chronic)
Distal tubules	Clusterin NAG NGAL Osteopontin	Amphotericin B Cyclosporine Sulfadiazine Tacrolimus
Collecting duct	Calbindin D28	Ampotericin B Acyclovir Lithium (acute)

IGFBP7, insulin‐like growth factor‐binding protein 7; KIM‐1, Kidney injury molecule‐1; L‐FABP, Liver‐type fatty acid binding protein; NAG, N‐acetyl‐β‐D‐glucosaminidase; NGAL, neutrophil gelatinase‐associated lipocalin; TIMP‐2, Tissue inhibitor of metalloproteinase 2

Current care algorithms rely on imperfect functional markers of glomerular filtration. Incorporation of newer damage biomarkers may decrease exposure to nephrotoxins, facilitate more accurate dosing of drugs cleared by the kidney, and lead to a decreased incidence and severity of AKI. For example, risk stratification with a novel biomarker to employ an AKI risk reduction bundle (including nephrotoxic medication avoidance) led to less AKI after cardiac surgery 46. In another study, a damage biomarker was able to predict supra‐therapeutic concentrations of tobramycin in patients with cystic fibrosis 50. Ideally, dosing of drugs with renal excretion in any AKD phase (deteriorating versus maintenance versus improvement) should prioritize the most dynamic and feasible functional assessments possible. The ADQI expert group recommended that further research related to assessment of renal function during AKD, including the utilization of renal biomarkers was urgently needed 6.

Nephrotoxin management during AKD

Drug‐associated AKI occurs in approximately 25% of critically ill patients, making drugs a common cause of AKI in the intensive care unit 51, 52. The prevalence of drug associated AKI is 2–3 times higher in elderly patients 53, 54, 55. The consequences are severe, with rates of nonrecovery, dialysis dependence and/or mortality similar to AKI from other aetiologies (40–50%) 52.

Evaluation of nephrotoxins as a plausible cause for AKI is the first consideration in medication management and includes assessment of the temporal sequence between administration and event and alternative explanations 56. In general, in all phases of AKD, selection of the least nephrotoxic drug and/or avoidance of a nephrotoxin should be the goal. This approach is supported by the fact that each nephrotoxin administration presents a 53% greater odds of developing AKI 57, and is compounded when patients receive more than one nephrotoxin 58.

Combining drugs can also result in important pharmacodynamic drug interactions 55, 56; for instance, the administration of some macrolide antibiotics (clarithromycin/erythromycin) with a 3‐hydroxy‐3‐methylglutaryl‐coenzyme‐A (HMG‐CoA) reductase inhibitor (statin) may lead to a greater number of hospitalizations for AKI (from rhabdomyolysis), compared to the administration of azithromycin (a macrolide that does not powerfully inhibit CYP450 enzyme CYP 3A4 and affect clearance of a HMG‐CoA reductase inhibitor) 59.

Drug selection, dosing and monitoring in AKI and AKD should be guided by personalized clinical decision making, including considerations for nephrotoxin initiation and discontinuation and the impact of starting alternative medications (Table 2). Factors to consider when starting a nephrotoxin include patient age, history of diabetes mellitus with proteinuria, hypertension, volume depletion, baseline kidney function, trauma, circulatory shock, sepsis, degree of fluid overload and concomitant administration of nephrotoxins 57, 60, 61. In the future, risk assessment is likely to include the detection of kidney damage using novel techniques before a functional change is observed 60.

Table 2.

Considerations for nephrotoxin management during AKD

When to avoid starting a nephrotoxic drug

Patient has known risk factors for kidney injury (i.e. advanced age, previous AKI episode, CKD, diabetes mellitus, proteinuria, hypertension)

A suitable and less nephrotoxic drug is available

Nephrotoxin is considered nonessential

Patient is already receiving a nephrotoxic drug and there is concern for a pharmacokinetic or pharmacodynamic drug interaction

Intended duration of the drug therapy is chronic and the initiation of the drug can be delayed until after the AKD episode

Biomarker indicates renal tubular injury is impending

Biomarker predicts risk of drug accumulation

There is a concern for a lack of appropriate follow up of serum creatinine and/or therapeutic drug concentration monitoring

When to discontinue a nephrotoxin

An evaluation of causal relationship indicates that the nephrotoxin is the potential cause of AKI/AKD

A suitable and less nephrotoxic drug is available

Nephrotoxin is considered nonessential

Biomarker indicates renal tubular injury has occurred

Other considerations for nephrotoxin management

Regular monitoring of functional status while on a nephrotoxin‐ either dose increase or decrease may be necessary

Minimizing the duration and dose of nephrotoxin exposure, if possible

Using evidence dosing guidelines

When nephrotoxin administration is essential – monitor for and manage subsequent adverse drug events

AKI, acute kidney injury; AKD, acute kidney disease; CKD, chronic kidney disease

The maintenance phase of AKD necessitates continued consideration of nephrotoxin avoidance (Figure 1). During the improvement phase of AKD, caution is still applied to nephrotoxin initiation, to prevent re‐injury. An evaluation of the appropriate timing to start or re‐start a drug assumes that a nephrotoxin is essential for the patient. The treatment of an infection with an antibiotic that is necessary for survival should begin immediately, and may even prevent deterioration of renal function. Further thought is given to the best time to restart a chronic therapy, or to increase the dose of renally excreted drugs to avoid a therapeutic failure in the renal recovery phase 62. Also, the interplay with augmented renal clearance during the recovery phase of AKD is unclear 63, 64.

A general statement cannot be made about a functional threshold to avoid or discontinue nephrotoxins. Ideally, nephrotoxic medications or combinations should be avoided in patients with AKD. When nephrotoxic medications are needed for clinically compelling reasons, efforts should be made to mitigate their nephrotoxic effects. The recommendations provided in package inserts or guidelines specific to a drug or drug class may offer guidance. At present there is a significant gap in knowledge to support drug management in AKD as outlined in the research agenda provided in Table 3.

Table 3.

Consensus recommendations of ADQI 16

Drug management in clinical practice

1) We recommend medication assessments at the following instances during the AKD period: medication reconciliation should occur at ICU/hospital admission and discharge medication regimen assessment at AKD diagnosis and change in AKD stage reassessment when patient condition changes

2) Strategies to avoid adverse drug reactions in AKD should seek to minimize both adverse events from overdosing or nephrotoxicity and therapeutic failure from under‐dosing or incorrect drug selection.

3) Medication regimen assessment or consideration of medication introduction during the AKD period should consider: nephrotoxic potential altered drug disposition including renal or hepatic elimination, toxic metabolites and drug interactions altered pharmacodynamics in AKD

4) Medication regimen assessment should be guided by a dynamic monitoring plan to include repeated serial assessment of: clinical features (adverse drug reaction or therapeutic failure) currently available renal diagnostic tests (eg biochemistry, imaging) currently available therapeutic drug monitoring (parent drugs +/− metabolites)

Advocacy and future research

1) We advocate for a clinical pharmacist to undertake medication regimen assessment throughout the AKD period.

2) We advocate for a clinical pharmacist to undertake a formal medication reconciliation and AKD education at hospital discharge

3) We recommend a research agenda that prioritizes federal and industry funding to: develop better drug specific damage markers and therapeutic monitoring assays validate better biomarkers of kidney injury, including imaging undertake trials for drug management in AKD

ADQI, Acute Disease Quality Initiative; AKD, acute kidney disease; ICU, intensive care unit

Management of angiotensin‐converting enzyme inhibitor and angiotensin II receptor blockers

The therapeutic options for supporting transition from AKI to renal recovery are limited and the decision to restrict certain therapies may reflect their nephrotoxic potential. A clinically relevant example is the use of angiotensin converting enzyme inhibitor (ACEi) and angiotensin II receptor blockers (ARB), which are associated with renal functional impairment, particularly in the setting of acute hypovolaemia 65, 66, 67. The indications for this group of medications are broad and include heart failure with documented left ventricular dysfunction, CKD with proteinuria, and hypertension. Consequently, they are frequently prescribed, particularly in the elderly 68. A recent study using routinely‐collected national hospital administrative data showed while ACEi and ARB prescribing increased by 16%, hospital admissions complicated by AKI rose by 50% in the same time period 69. Despite the clear benefits of these drugs in the chronic setting, the risk–benefit ratio during AKD is uncertain.

It has been suggested to routinely stop ACEi and ARBs during any intercurrent illness, despite the lack of any evidence supporting these recommendations 70. This could potentially lead to worsening in hypertension or deterioration in symptoms of congestive cardiac failure. It is not known if withholding or re‐starting these drugs during AKD results in better outcomes, and at what stage they are ideally re‐started post AKI/AKD. Two studies evaluating nonresumption of ACEi after surgery demonstrated an increase in 30‐day mortality 71, 72. Although re‐introduction of such therapies is often part of a multidisciplinary approach to AKI care, it is usually considered when the GFR has stabilized and volume status is optimized. ACEi‐ or ARB‐ associated hypotension and decreased filtration fraction are recognized as adverse effects that can cause or exacerbate AKI. The risk–benefit ratio for their use in AKD must be carefully considered to personalize therapy. While chronic tolerance to reversible decrements in filtration fraction and GFR caused by ACEi and ARB may be desirable in congestive cardiac failure and CKD, such effects may not be tolerable and beneficial in AKD.

Similarly, there is a significant risk of potential therapeutic failure caused by under‐dosing or avoidance of the most effective drugs in patients with AKD (particularly in the recovery phase) but this is rarely recorded 62. A recent single‐centre observational study including 396 patients highlighted that 7.5% of patients had a therapeutic failure defined as subtherapeutic drug levels and underdosing of medications, with 30% of events being life threatening and primarily related to antibiotics 73. Underdosing of antibiotics in particular, is a serious problem during periods of fluctuating renal function. More than most other drug classes, antibiotics have a clear concentration‐effect relationship where effect can be defined as time‐dependent or concentration‐dependent killing. Maintaining drug concentrations according to pharmacodynamics and antibiotic characteristics is essential. Numerous factors contribute to subtherapeutic concentrations, including inappropriate drug dosing, unrecognized recovery of native renal function, and changes in pharmacodynamics and nonrenal clearance 74.

In patients with AKD, the decision to discontinue, introduce and/or re‐introduce medications with important effects on other organs needs to be individualized and discussed with the appropriate interdisciplinary care providers. This also includes the effects of AKD on drug metabolites, including nonrenal metabolism.

Effects of AKD on drug pharmacokinetics

The effects of CKD on drug metabolism are relatively well established but little is known with regard to the impact of AKD. The extrapolation of data from CKD is flawed given that the time course of disease progression is different and renal function may be fluctuant in AKD. During critical illness, organ cross‐talk may also influence drug metabolism, particularly if involving the liver and the kidney 10, 75. This may reflect the impact of AKD on hepatic blood flow, changes in protein binding and drug distribution, and the increasingly recognized effects on CYP450 activity 75, 76. Impairment of CYP450 activity, as well as effects on drug transporters, may also account for some of the pharmacodynamic effects during AKD. Many critically ill patients have AKI with varying volumes of distribution and protein concentrations that affect a drug's pharmacokinetics. The volume of distribution for hydrophilic drugs such as β‐lactams is typically increased in critically ill patients who have an increase in total body water, especially patients with AKI. As the patient recovers from their critical illness and AKI, the fluid shifts and volume decreases, but the extent and the speed of improvement vary, thus making dosing during AKD a challenge. Further, in patients with renal disease, volume of distribution is increased as a result of reduced protein binding and increased tissue binding. Exact prediction of the effects of AKD on the pharmacokinetics of drugs remains challenging and further in‐depth research is urgently required.

Conclusions

Drug management during the phases of AKD requires an individualized approach using general principles as outlined in Table 3. More research is urgently needed to avoid therapeutic failure and mitigate drug associated harm affecting chances of renal recovery during important phases of acute renal disease.

Competing Interests

There are no competing interests to declare.

We would like to thank R. Bellomo, A. Bihorac, E.D. Siew, S.M. Bagshaw, D. Bittleman, D. Cruz, Z. Endre, R.L. Fitzgerald, E. Hoste, K.D. Liu, E. Macedo, R. Mehta, M. Nadim, P.M. Palevsky, N. Pannu, M. Rosner, R. Wald and A. Zarbock for their invaluable suggestions and contributions towards the document.

Ostermann, M. , Chawla, L. S. , Forni, L. G. , Kane‐Gill, S. L. , Kellum, J. A. , Koyner, J. , Murray, P. T. , Ronco, C. , Goldstein, S. L. , and ADQI 16 workgroup (2018) Drug management in acute kidney disease – Report of the Acute Disease Quality Initiative XVI meeting. Br J Clin Pharmacol, 84: 396–403. doi: 10.1111/bcp.13449.