Drug stewardship for people with chronic kidney disease; towards effective, safe, and sustainable use of medications

Abstract

People living with chronic kidney disease (CKD) often experience multimorbidity and polypharmacy, the pharmacokinetics and pharmacodynamics of medications differ for them, and the risks and benefits of medications may differ or be less well understood. This review aims to serve as a clinician’s resource for optimal drug stewardship for patients with CKD. Drug stewardship aims to maximise medication safety and effectiveness, through medication reconciliation, medication selection, dose adjustment for glomerular filtration rate, monitoring for effectiveness and safety, and discontinuation (deprescribing) when no longer necessary. This review includes special considerations for medication use pregnancy and lactation, acute illness, and people with cancer, as well as, responsible use of over-the-counter drugs, herbal remedies, and supplements and sick day rules. This review also highlights the inequities in medication access worldwide, suggesting policies to ensure access to quality and essential medications for all persons with CKD. Further strategies to promote drug stewardship include patient education and engagement, using digital health tools, shared decision-making and collaborating within interdisciplinary teams. Throughout, we position the person with CKD at the center of all drug stewardship efforts.

1. Introduction

People with chronic kidney disease (CKD) are prescribed, on average, 12 medications per day in addition to nonprescription therapies, creating a complex polypharmacy¹. Evidence of harms and effectiveness is often indirect, requiring judgement and interpretation. The exclusion of people with advanced CKD from clinical trials² creates uncertainty about the generalizability of results. Because of the central role of the kidneys in the elimination of many medications and their metabolites, many medications require dose adjustment or discontinuation in people with low glomerular filtration rate (GFR)³. CKD increases the nephrotoxicity of some medications, and nephrotoxicity is critical to avoid in people with low GFR. People with CKD have increased susceptibility to problems with medications, which may further contribute to suboptimal pharmacological management^4,5.

Drug stewardship typically describes responsible use of antimicrobials;⁶ however, we apply this term to describe effective, safe, and sustainable use of medications for people with CKD.⁷ This review discusses key concepts and related definitions for effective drug stewardship in adults with CKD, with a focus on all medications, not only those used in CKD management. We do not cover peculiarities in in drug stewardship for children or patients undergoing kidney replacement therapy, but recognise that many of the principles discussed are universal. Because this review was developed alongside the KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of CKD,⁸ readers are encouraged to refer to those guidelines for a practical summary of medication management recommendations.

2. Burden of medication-related problems in people with CKD

Medication-related problems are common in people with CKD.¹ Inappropriate medication dosing occurs in nearly 70% of studies in hospital settings and 34% of studies in long-term care settings.⁹ The risk of an adverse drug event is 3 to 10 times higher for a person with CKD than for an age-matched person without CKD.¹⁰ Medication-related problems are traditionally classified into nine categories (Table 1).^11,12 They are consistently associated with worse outcomes, including decreased quality of life, increased hospitalization rates, increased mortality and increased healthcare costs.¹

Table 1.

Medication-Related Problems, Descriptions, and Key Examples in the setting of CKD

Medication-related problem	Description	Example
1. Indication without drug therapy	Patient is not receiving available drug therapy for a diagnosed medical condition	Uncontrolled hypertension that is not addressed with medication therapy when warranted223
2. Drug use without indication	Use of a medication without a medically valid indication	Off-label use of sodium polystyrene sulphonate for hyperkalemia prevention 224
3. Improper drug selection	Medication of choice is not being used	Poor uptake of guideline-recommended pharmacotherapies across stages of CKD severity 4,5
4. Subtherapeutic dosage	Patient has a medical problem that is being treated with too little of the correct drug	Sodium bicarbonate dose <0.5–1.0 mEq/kg and serum bicarbonate remains <22 mmol/l on repeat check 225
5. Overdose	Patient has a medical problem that is being treated with too much of the correct drug (toxicity)	Increased risk of hypoglycemia associated with overdose of oral hypoglycemic agents in people with pre-dialysis CKD
6. Adverse drug reaction	Drug effects that are unwanted, unpleasant, or harmful	Increased hyperkalemia risks attributed to RAS inhibitors across lower eGFR 226
7. Drug interaction	Negative effects of drug-drug, drug-disease, or drug-food interaction	AKI risks associated to the triple-therapy combination of diuretics with ACEi/ARBs and NSAIDs227
8. Failure to receive drug	Patient is not receiving prescribed medication(s), e.g. due to nonadherence or economic reasons	Increased risks of primary non-compliance and early treatment interruption of guideline recommended therapies in people with heart failure and CKD 4
9. Inappropriate laboratory monitoring	Patient is not receiving appropriate laboratory tests to adequately monitor medication therapy or to determine if comorbid conditions are being treated properly	Inadequate monitoring of eGFR during lithium treatment 103 or during the initial weeks of RASi treatment 228.

Risk factors for medication-related problems in CKD include comorbid conditions, complex treatment regimens and polypharmacy, frequent medication changes, use of medications which have a narrow therapeutic window, drug interactions, complex care teams with multiple clinicians involved, and challenges with treatment burden, cost, and nonadherence.¹ These risks likely increase as GFR declines with the change in pharmacokinetics: absorption, distribution, metabolism and excretion. Re-evaluation of drug and dose appropriateness in this setting is essential to patient safety.

3. Informing clinical decisions through both trials and the analysis of routine care data

Poor representation of people with CKD in clinical trials has led to less direct evidence for this group². Even when trials have been conducted, participants may not be representative of the general population in whom they will be used¹³, and the size or duration of the trials may not be enough to explore effects on risk of kidney failure. Where studies exist, some are old, originating from eras when co-interventions were different and polypharmacy less common, or they focus on inpatients, where different medications and monitoring protocols are used¹⁴. Fortunately, the recognition by medical agencies ¹⁵ of changes in GFR decline and albuminuria as surrogate endpoints for kidney failure, has allowed in the recent years for the performance of paramount trials that open long awaited opportunities to delay the progression of this disease.

Acute nephrotoxicity events are rarely well characterised in clinical trials,^16,17 but between 14%–26% of acute kidney injury (AKI) events in adults are attributable to medications.^18–20 Evidence of these and other adverse events often originates from pharmacovigilance systems (e.g., FDA MedWatch), which rely heavily on voluntarily-reported cases identified by patients and healthcare professionals. Fewer than 1% of adverse drug reactions²¹ are detected through spontaneous reporting, which may be influenced by external factors such as media interest or safety alerts (notoriety bias). These systems cannot reflect the true incidence of adverse drug reactions and susceptibility factors. Individual patient chart review by pharmacists is a more comprehensive method to identify adverse drug reactions, but prohibitively expensive and time-consuming for research or administrative use.

Additional data to guide prescribing strategies derives from post marketing surveillance.²² Studies of routine care provide, in many cases, the best source of information to guide practice where trials are not available or sufficient, and are a critical source of information for rare harms. For example, diabetic ketoacidosis or Fournier gangrene are serious but rare adverse events of sodium-glucose cotransporter-2 inhibitors (SGLT2i) that were identified through observational reports from routine care.^23,24 Large datasets, innovative systems for, and innovative approaches to, evaluating causal inference from observational data have improved the validity of conclusions drawn²⁵ and represent a critical complement to classic randomised controlled trials (RCTs; Table 2). As a case example, recent trials like STOP-ACEi and efforts within the target-trial emulation framework^26,27, have allowed debunking long-standing myths of medication-induced harm originated by biased observational research^27,28. Trials and routine care data complement each other and assist healthcare professional in making evidence-based decisions.

Table 2.

Use cases when observational studies can be used to inform decision making for people with CKD.

Use case	Example
Providing evidence where trials are unlikely to happen	Randomizing malnourished people on dialysis to nutritional support would not be ethical, yet we lack proof that treating malnutrition in CKD is effective. An observational study benefiting from a policy change, observed improved survival in hypoalbuminaemic people who received (vs not) oral nutritional supplements.229
Anticipating to ongoing trials	Findings from an observational study investigating stopping vs. continuing RASi was published almost two years earlier than the STOP ACEi trial, while reaching similar conclusions26,27.
Transporting evidence to heterogeneous clinical practice	Observational studies have found that hyperkalemia risk after initiation of mineralocorticoid receptor antagonists is much higher in routine clinical practice167 than the highly controlled settings of randomised trials230.
Expanding evidence to underrepresented or excluded populations	Randomised trials of RASi excluded people with advanced CKD. Observational studies have confirmed the benefits of renin-angiotensin system inhibitors in these trial under-represented populations231.
Evaluating endpoints for which trials were underpowered	Pivotal trials of glucagon-like peptide-1 receptor agonists (GLP-1RA) were not powered to evaluate kidney outcomes. An observational study reported beneficial effects on eGFR decline by GLP1-RA use 232 of identical magnitude to the pooling of existing clinical trials 2 years after 233.
Identifying rare but severe medication harms	Diabetic ketoacidosis or lower limb amputations are rare adverse side effects of SGLT-2i, and trials do not have sufficient power to assess these.234 Increased risk of hematuria, proteinuria, and KFRT associated with the use of rosuvastatin vs atorvastatin, particularly in advanced CKD. These outcomes occurred in <3% of the population during median 3 years of follow-up 235

4. Drug dosing for people with CKD

GFR is considered the best surrogate for renal clearance and therefore is central to determining eligibility and dose for drugs that are cleared by the kidneys. Failure to account for GFR is a risk for treatment failure and adverse drug events^10,29.

4.1. Biomarkers used to guide drug dosing in CKD and their limitations

The standard approach for dosing and monitoring medications in CKD is based on concentration of serum creatinine. Creatinine is freely filtered at the glomerulus and undergoes 10–40% proximal tubular secretion. Creatinine clearance therefore modestly overestimates true GFR, which becomes more pronounced as GFR declines³⁰. Select medications (e.g., trimethoprim) interfere with tubular secretion of creatinine and alter estimated GFR (eGFR) but not clearance. Serum creatinine is the terminal byproduct of skeletal muscle catabolism. Non-renal factors including diet, sex, body composition (i.e., skeletal muscle mass), diseases like cirrhosis, malnutrition or cachexia^31,32and factors that modify extrarenal elimination can affect serum creatinine concentrations³³. In these situations, creatinine-based eGFR (eGFRcr) may overestimate GFR.

Because of the recognised limitations of creatinine, alternative filtration markers have been sought as adjunct or alternatives to estimate GFR for drug dosing and monitoring. While, cystatin C is now the most widely available among these alternatives, serum cystatin C concentrations may be affected by adiposity, smoking, hypo- and hyperthyroidism, glucocorticoid excess, chronic inflammation, and malignancy, independent of GFR.^34–45 Estimated GFR from cystatin C in combination with creatinine (eGFRcr-cys) is recommended as a supportive test for eGFRcr when eGFRcr is thought to be inaccurate and clinical decisions are impacted by the level of GFR.⁸

4.2. Mismatch between eGFR equations in regulatory approvals and drug dosing in clinical practice

Recommendations for medication use and dosing in CKD vary across pharmacopeias and medication agencies.⁴⁶ Only relatively recently did regulatory agencies require pharmacokinetic studies for approval of a kidney-cleared drug. Available studies are generally small, short, and conducted among patients with minimal comorbidities.⁴⁷ In many pharmacokinetic studies, and historically, GFR was estimated with the Cockcroft-Gault (CG) creatinine clearance equation.⁴⁸ The CG equation was developed before creatinine standardization in a small population of white men and validated in the Modification of Diet in Renal Disease dataset.⁴⁹ More accurate equations have since been developed to assess GFR in people of different sex, age groups, ancestry, ethnicity, and weight. A recent large validation study confirmed that the CG equation, compared with measured GFR (mGFR) is less precise, more biased, and associated with a lower percentage with eGFR within ±30% of mGFR (referred to as P30) compared with other more contemporary eGFR equations (i.e., MDRD, CKD-EPI, Lund-Malmo-Revised [LMR], and European-Kidney-Function-Consortium [EKFC] creatinine-based equations).⁵⁰ Because of these advancements, clinicians should use equations that optimize accuracy and minimize bias for optimal drug dosing.

Major regulatory agencies now recognise that “any contemporary, widely accepted, and clinically applicable eGFR equation is considered reasonable to assess kidney function in pharmacokinetic studies”.^51–53 For drugs approved using the CG equation, it is not likely that inconsistencies in the development of dosing recommendations can be improved. Post-marketing observational studies can clarify medication safety for people with CKD. For example, metformin was originally contraindicated in people with serum creatinine >1.5 mg/dL (133 μmol/L) because of the risk for accumulation, lactic acidosis, and diabetes-related death.^{54 55} Dosing by serum creatinine was identified as inconsistent with best practices. Observational studies defined dosing thresholds based on GFR, identifying an eGFR threshold of <30 mL/min per 1.73m² as higher risk.^56,57

Practical challenges also exist with translation of approved labelling to real-world practice. Drug dosage labels rarely include information on which eGFR equation was used in the pharmacokinetic studies used to derive dosing instructions. Although many medications were historically approved based on the CG equation, laboratory reporting of eGFR typically uses equations such as the MDRD⁵⁸ or CKD-EPI^59,60 equations. Electronic health record and clinical decision support tool calculations allow clinicians to apply the local eGFR preferences to dose recommendations without consideration of the original measurement method. For equations based on serum creatinine (i.e., CG, CKD-Epi, LMR, EKFC) the impact of this difference is likely modest, though it is worth noting that the units of CG are usually mL/min and those of the other equations mL/min per 1.73m² (see section 4.4).⁶¹ Calculation of eGFR using alternative biomarkers such as cystatin C (eGFRcr-cys or eGFRcys) and applying it to thresholds based on creatinine only calculations may result in different drug dose recommendations. This might be appropriate if eGFRcr is not an accurate reflection of GFR. Nevertheless, a better understanding of these differences and implications is needed as cystatin-C based measures become widely used in clinical practice.

4.3. eGFR selection for drug dose adjustments in people with CKD

Guidelines recommend that GFR assessment for drug dosing should be individualised and consider the risks and benefits of the therapy in question and the health status of the patient.⁸ Our suggested approach for drug dosing is to use initial and supportive testing to develop a final assessment of the most likely true GFR (Figure 1). The initial test for evaluating drug dosing in most circumstances will be creatinine-based eGFR (eGFRcr). The eGFRcr is widely available, routinely evaluated as part of the basic metabolic panel, is low cost, and has sufficient precision for most medication dose-adjustment. If eGFRcr is expected to be inaccurate due to factors other than GFR, measurement of cystatin C could be considered to calculate cystatin-C–based eGFR alone (eGFRcys) or in combination with creatinine (eGFRcr-cys), which is the most accurate estimate in most populations studied thus far. An eGFRcys often provides discordant GFR estimates to eGFRcr. In an evaluation of Stockholm’s healthcare⁶², one in four patients with same-day testing of creatinine and cystatin C differed in their eGFR estimates by 30% or more, leading to uncertainty about which eGFR estimate to apply. In a qualitative study of acute care providers, some considered a variety of factors in their drug dose decision making (e.g., creatinine, cystatin C, urine output, severity of illness), but many ‘choose the lowest eGFR number’ to be conservative⁶³. When discrepant eGFR estimates exist, use of a lower value would increase the likelihood of dose reductions. This prioritization of safety over effectiveness may not be clinically appropriate. For example, it may be appropriate in a patient with severe infection, if using an antibiotic with a wide therapeutic window, to use a higher dose.

Figure 1.

Suggested approach to GFR evaluation for drug dosing.

While Cystatin-C may not be available in all geographic areas, this algorithm describes the approach to the evaluation of GFR and how it can be applied to drug dosing. *Consider eGFRcys rather than eGFRcr-cys in otherwise healthy populations with decreased creatinine generation due to reduced muscle mass or decreased creatinine secretion or extra-renal elimination due to use of specific medications. ** Use eGFRcr or eGFRcr-cys depending on discordance between eGFRcr and eGFRcys.

In general, if concern about the appropriateness of eGFRcr has led to measurement of cystatin C, we suggest remeasuring creatinine at the same time and using a combination equation (eGFRcr-cys). Use of both biomarkers in the eGFR equation to some extent manages or mitigates the non-renal determinants of either alone. A recent evaluation of accuracy of eGFR equations versus iohexol clearance identified that concordant eGFRcr and eGFRcys values (within 15 mL/min per 1.73 m² or 20–30% of each other) resulted in similar accuracy of eGFRcr, eGFRcys and eGFRcr-cys,⁶⁴ and consistent dose recommendations across the equations. When eGFRcr and eGFRcys were discordant, eGFRcr-cys was generally more accurate than either eGFRcr or eGFRcys.⁶⁴ A study where eGFRcr-cys was used to individualise drug doses resulted in improved pharmacokinetic and pharmacodynamic target attainment of vancomycin therapy compared with standard approaches using eGFRcr.⁶⁵ If eGFRcr-cys is expected to be inaccurate, or if more accurate assessment of GFR is needed for medication dosing in people with CKD, then a measured (mGFR) using plasma or urinary clearance of exogenous filtration markers should be considered. Ultimately, local resources and turnaround time, and the need for additional visits for blood draws and the time commitment of mGFR impact the ability to apply these non-creatinine alternatives to drug dosing in people with CKD.

4.4. Normalised and non-normalised eGFR equations

Drug dosage adjustments should be based on GFR in units of mL/min since clearance of medications is related to the individual person’s absolute level clearance, not compared with another person. The native units for eGFRcr, eGFRcys, and eGFRcr-cys for many contemporary equations are normalised to a nominal standard body surface area (BSA) of 1.73m² and expressed as mL/min per 1.73m². To re-express the result in non-normalised, absolute terms (mL/min) clinicians should use the equation = eGFR in mL/min/1.73 m² × [BSA (m²)/1.73 m²]. The greatest impact of this nuance will be in patients with a body surface area that is markedly different than 1.73m².⁶⁶

4.5. Sex and gender

Women are less likely to be recognised as having CKD ⁶⁷ and thus be at risk of potentially inappropriate prescribing ⁶⁸. Sex differences in drug safety and efficacy in people with CKD are understudied, and very little is known about the optimal assessment of GFR in people who are transgender.⁶⁹ Compared to men, women are also more likely to report adverse drug reactions to commonly used CKD medications.⁷⁰ It has been postulated excess risk of adverse drug reactions in women to be explained by overdosing, given that most drugs are prescribed to women and men at the same dose (i.e. one pill per day) and pharmacokinetics are often higher in women ⁷¹. Sex differences in body weight and composition as well as physiology indeed impacts on drug metabolism and response, and sex hormones appear to play an important role in modulating sex-based differences in pharmacokinetics across the life cycle.^72,73 Interestingly, studies suggest a survival benefit for women with heart failure due to reduced heart function who have lower (vs. higher) doses of RASi.^74,75

4.6. Special circumstances for drug dose adjustments in people with CKD

Acute illness

Acute illness leads to rapid changes in physiology that alter drug pharmacokinetics and pharmacodynamics. For example, large volume fluid resuscitation or use of extracorporeal devices such as extracorporeal membrane oxygenation alters the volume of distribution of certain drugs.³ Acid-base abnormalities and accumulation of uremic toxins alter drug protein binding.³ AKI decreases kidney elimination of drugs and their renally-eliminated metabolites. For example, although midazolam is primarily eliminated hepatically, the major metabolite 1-hydroxy-midazolam is pharmacologically active and 45–75% eliminated renally, thus risking accumulation and over-sedation in critically-ill people with kidney dysfunction.⁷⁶

Further complicating pharmacokinetic evaluation in people with CKD who are acutely ill, serum concentrations of kidney filtration markers fluctuate because of changes in true GFR, in non-GFR determinants of the marker, or in the volume of distribution of the marker. For example, decreased serum creatinine production in acutely ill patients with skeletal muscle catabolism and deconditioning could lead to over-estimation of GFR. Acute-care drug use may compete for tubular secretion of serum creatinine at tubular transporters (e.g., high-dose sulfamethoxazole/trimethoprim [SMZ-TMP] for Pneumocystis jirovecii pneumonia) making it difficult to distinguish between pseudo- and true nephrotoxicity.^77,78 Non-renal determinants of cystatin C are also widespread in acute care. More than 70% of hospitalised patients in whom a cystatin C concentration was assessed were receiving concurrent corticosteroids, had an elevated C-reactive protein concentration, had an abnormal thyroxine concentration, were obese, or had malignancy.⁷⁹ Profound inflammation due to widespread infection could lead to non-renal elevations in cystatin C, underestimation of GFR, and insufficient doses of lifesaving antimicrobials.⁸⁰ Factors influencing GFR estimating equations will be highly variable between and, over time, within individuals, and more pronounced than in outpatients with CKD.⁸¹ Dosing decisions during critical illness must balance all the factors which may alter the accuracy of estimating equations, consider the risks and benefits of the medication for the patient, and iteratively re-evaluate progress toward therapeutic goals. Drug level monitoring should be used, where available.^{6, 57} When difficult decisions have to be made (e.g., continuing or discontinuing foscarnet is a patient with disseminated cytomegalovirus and foscarnet-associated AKI), it is essential to identify and include patient’s preferences and values.

Cancer

Decreased GFR is associated with adverse clinical outcomes in people with cancer, and eGFR is relevant to drug selection, dosing, and eligibility for clinical trials. No universal guidelines exist to select the preferred approach to GFR estimation in people with cancer. Despite its relative inaccuracy compared with other validated eGFR equations, the CG equation continues to be one of the most-commonly used eGFR methods for these patients.^82,83 Non-GFR determinants of both creatinine and cystatin C may be more profound in people with cancer: Cachexia and malnutrition are common in people with cancer and lead to decreased creatinine production and overestimation of GFR; Cystatin C production may be increased in high-cell turnover diseases, in the presence of certain types of cancer cells, and in people receiving high-dose corticosteroids as part of the chemotherapy. These increases in cystatin C concentrations may lead to an under-estimation of eGFR. Alternatively, cathepsin-D–mediated proteolysis with targeted cancer treatment could lead to a reduction in cystatin C concentrations and an over-estimation of eGFR.^84,85 In a pharmacokinetic evaluation of patients treated with high-dose methotrexate for central nervous system lymphoma, the eGFRcys best predicted methotrexate clearance.⁸⁶ An evaluation of eGFR equation performance in patients with solid tumors observed the eGFRcr (CKD EPI) and the eGFRcr-cys (CKD-EPI) predicted mGFR with greater accuracy than CG.⁸⁷ Thus, it would appear that a general approach to GFR and drug dosing (Figure 1) may be adopted in oncology practice and clinical trials. Recent international guidelines on cancer-dosing support this view and recommend the use of validated eGFR equations for most situations.⁸⁸ For specific situations including, but not limited to, dosing of carboplatin, cisplatin, fludarabine, ifosfamide, and melphalan, non-normalised eGFR (section 4.4) may be the preferred method.⁸⁹

Pregnancy

Usually, creatinine decreases physiologically during pregnancy because of glomerular hyperfiltration, and BSA varies, impacting eGFR.⁹⁰ For people who require medications that are impacted by or could impact GFR, clinicians should regularly assess risks and benefits. When prescribing medications to people with CKD for whom pregnancy is a possibility, it is necessary to review teratogenicity and provide regular reproductive counselling in accordance with the values and preferences of the person with CKD; people should not be denied medications because of the possibility of pregnancy.

5. Drug stewardship for safe choice of medications

When new medications are started, healthcare providers should consider comorbidities, functional status and psychosocial factors (e.g., socioeconomic status, self-efficacy) for each person with CKD. Medicines must be evaluated and coordinated for indications, drug-specific risks (e.g., semaglutide is contraindicated in patients with medullary thyroid cancer or pancreatitis), interactions, dosing, frequency, pill burden, and side effects, including potential nephrotoxicity.^1,10

Central to drug stewardship for people with CKD is to minimize use of medications in patients who have contraindications. Contraindications may be absolute, or when risk of using the medication clearly outweigh any potential benefit or relative, when caution should be used when the medication is prescribed. Medications with relative contraindications may be prescribed provided no safer alternative exists and there is reasonable expectation of efficacy; when used, clinicians should consider whether intensive monitoring for adverse effects or dose adjustments are warranted, and iteratively evaluate the clinical need. A proposed general approach to prescribing medications that may carry contraindications in people with CKD is given in Figure 2. Reassessment over time as GFR and health status changes is essential.

Figure 2.

Algorithms proposing decision making process for medications that may carry contraindications in people with CKD. Panel A (upper panel) discusses steps for medications directly related to the management of CKD (e.g., ACE-inhibitors), while Panel B (lower panel) discusses steps for medications prescribed for other indications than CKD (e.g., metformin, antibiotics)

5.1. Nephrotoxic medications

Drug-associated nephrotoxicity is an important contributor to new or worsening CKD.⁹¹ Multiple mechanisms of nephrotoxicity have been reported including hemodynamic changes, vascular endothelial injury, glomerular disease, direct tubular toxicity including acute tubular injury or necrosis, acute interstitial nephritis, osmotic nephrosis, and crystal deposition.⁹² Patient characteristics that increase the risk for nephrotoxicity include pre-existing CKD, concomitant use of other nephrotoxins (e.g., vancomycin with piperacillin-tazobactam⁹³ or vancomycin with aminoglycosides⁹⁴), high-doses or prolonged duration of nephrotoxins, older age, comorbidities (including diabetes, heart and liver failure, and solid organ transplantation) and acute illnesses, especially when associated with intravascular volume depletion and hypotension (e.g., sepsis, shock)^95,96.

Common nephrotoxic medications to be aware of and potential alternatives are listed in Supplemental Table 1. Between 18%–20% of people with CKD G3-G5 receive at least one potentially-inappropriate nephrotoxic medication annually, primarily non-steroidal anti-inflammatory drugs (NSAIDS), antivirals, fenofibrates, bisphosphonates, and proton pump inhibitors (PPI).^68,97,98 Nephrotoxicity can be minimized by a combination of judicious prescription, appropriate dosing and monitoring in people at risk for or with CKD.⁹⁶ For example, PPI are associated with AKI and CKD due to tubulointerstitial nephritis, but the absolute risk is relatively small (e.g., one in thousands) and associated with longer duration of use. In some people with or at risk for CKD, use with careful monitoring may be reasonable, with discontinuation when no longer indicated.^99,100 While some potentially nephrotoxic medications have good alternatives, in other cases the alternatives may be less potent, more costly, or their safety and effectiveness may be less studied; knowing the patient’s social circumstances and their access to drug benefits may be essential. Sometimes the use of a potentially nephrotoxic medication is unavoidable (e.g., calcineurin inhibitor use in a patient with nephrotic syndrome). In this case, close monitoring for nephrotoxicity and risk mitigation using therapeutic drug monitoring is reasonable.

5.2. Medications with contraindications or altered risk benefit profile in declining GFR

Some medications have altered risk-benefit ratio as GFR declines. This is sometimes a known issue and sometimes a lack of data resulting from the exclusion of people with advanced CKD from pivotal trials. Supplemental Table 2 provides selected examples of common medications prescribed to people with CKD that require dose adjustment or discontinuation at lower levels of GFR. Recommendations may change as new trial evidence becomes available.

Monitoring intensity (GFR, electrolytes, and drug concentrations, where available) should incorporate patient preferences and be tailored to patient susceptibility factors, therapeutic window of the medication, risk for adverse effects. Key examples include the need to monitor potassium and GFR during the initial weeks of treatment with RASi¹⁰¹ or during prolonged use of SMZ-TMP or PPI (Supplemental Table 3).¹⁰² Lithium requires annual monitoring of GFR and more frequent drug concentrations to avoid toxicity episodes.^103,104 Warfarin requires frequent monitoring: maximizing time in therapeutic range is more challenging in people with advanced CKD.¹⁰⁵

5.3. Over-the-counter (OTC) medications, herbal remedies, and dietary supplements

OTC medications, herbal remedies, and dietary supplements can cause CKD, AKI and progression of CKD. The use of these therapies is observed in 25–70% of people with CKD.^106–108 NSAIDS, the most commonly used OTC medication, are associated with altered intraglomerular hemodynamics, interstitial nephritis, analgesic nephropathy, hypertension, and heart failure.^109,110 Chronic OTC NSAID use has been associated with a higher risk of kidney failure compared with non-use.^111–114 However, pain is prevalent in people with CKD and its patient-centered management is complicated: for some patients judicious NSAID use with monitoring of eGFR may be preferred to other pain medications, such as opioids, that have stronger associations with adverse events.^115,116

Herbal remedies are also common, and in some countries are used by most of its population.¹¹⁷ Often used in an unmonitored setting, many of these remedies are composed of natural compounds with complex active ingredients that have not been evaluated in people with or without CKD. The frequency of CKD associated with herbal remedy use is not known and likely varies geographically and culturally. Aristolochic acid nephropathy or nephrotoxicity due to alkaloid compounds found in Chinese herbal remedies is the best known example,¹¹⁸ but nephrotoxicity has been reported for many other herbal remedies globally.^117,119,120 Toxicity may be enhanced by volume depletion and by other illnesses or medications.

Dietary supplements are likewise readily available and widely used as alternative or complementary therapy, without an evidence base. Patients and providers commonly incorrectly assume that the widespread availability of these unregulated products indicates safety and effectiveness. However, their activity and pharmacokinetics may be unknown and potential toxicity unstudied (e.g., creatine supplements for body building, associated with acute interstitial nephritis^121,122; high-dose vitamin C (ascorbic acid) leading to tubular calcium oxalate crystal deposition¹²³).

Patients should routinely be asked about use of non-prescription products. Unprescribed products that may threaten kidney health or interact with essential therapies for CKD should be discussed, and the healthcare worker should clearly advocate for their discontinuation, within a framework that recognizes and respects the patient’s perspective, which may be culturally or socially informed, and which balances patient autonomy, beneficence, and non-maleficence. The Natural Medicines Comprehensive Database curates useful information, but is not open access.¹²⁴ Figure 3 summarises common herbal remedies and dietary supplements arranged by the countries where the adverse effects have been identified.

Figure 3.

Selected herbal remedies and dietary supplements with evidence of nephrotoxicy, grouped by continent where case reports come from.

From 2024 KDIGO guidelines on CKD screening, detection and management.

6. Medication Review and Medication Reconciliation

Medication review is essential for minimizing the occurrence of medication-related problems.¹ Medication review is also a critical opportunity for clinicians to optimize medication regimens by prescribing medications that have shown positive impact on kidney function (e.g., ACEi, SGLT2i). Medications that are no longer needed can be considered for discontinuation (e.g., PPI) and others will require dose reduction or discontinuation when GFR falls (e.g., metformin). Medication review by clinical pharmacists in people with CKD is associated with reductions in the use of inappropriate medications and medication-related problems, in both outpatient and inpatient settings.^125–127

Medication reconciliation, the identification of the most complete and accurate list of medications, is the first step of the medication review process (Figure 4)^128,129. After achieving agreement as to which medications the patient is taking, clinicians can take steps to minimise medication-related problems. Best practices for medication reconciliation and review includes documentation and communication (Box 1)^{130, 131}.

Figure 4.

Suggested steps in the process of medication review. From the 2024 KDIGO guidelines on CKD screening, detection, and management

Box 1. Best Practices for Medication Reconciliation and Review.

• Obtain an accurate medication list, including over-the-counter medications, from the patient and/or available sources (e.g., caregiver, electronic medical record, smartphone applications).

• Evaluate whether all medications remain medically necessary or whether any other medication is required.

• Assess whether current medication(s) is the optimal medication(s) for each indication, individualised for each patient.

• Discuss reproductive options and plans with patient and adjust medications accordingly

• Evaluate the medication dosage and regimen, taking into consideration related factors such as liver dysfunction, patient size or weight (e.g., amputation, muscle wasting, over- or underweight).

• Review the medication list for drug interactions, including drug-drug, drug-disease, drug-laboratory, and drug-food interactions.

• Ensure that proper monitoring takes place.

• Determine whether there are any barriers to patient adherence (including cost or availability of medications) and evaluate relevant laboratory values.

• Identify and resolve any discrepancies between the medications list and the one in the medical record.

• Communicate medication changes to other clinicians and to pharmacy.

Because studies of the effectiveness of this practice are lacking, CKD guidelines do not make recommendations about medication reconciliation and review. It seems likely that medication reconciliation and review routinely, at clinic visits, and at care transitions such as hospital discharge facilitates the achievement of benefits of evidence-based therapy^132,133. Clinical pharmacists are ideal for conducting medication reconciliation and review, but other approaches, such as a multidisciplinary team consisting of nurse, pharmacist, and physician are also acceptable^132,133. Section 7.2 highlights resources, including digital tools, that facilitate effective medication review.

6.1. Guidance on prescribing cascades

A prescribing cascade is the circumstance of initiation of additional medication in response to the effects of an initial medication. Prescribing cascades can be appropriate or inappropriate¹³⁴. An appropriate prescribing cascade involves intentional prescribing of a new medication to manage a known adverse drug event caused by a medication, such that the benefits of both medications outweigh the risks. For example, it would be appropriate to provide potassium supplements if diuretics cause hypokalemia¹³⁴; In patients with hyperkalemia associated with RASi, the addition of another medication, such as a potassium binder or chlorthalidone, could reduce potassium and allow continued use of RASi.¹³⁵ However, appropriate prescribing cascades may become problematic when the secondary medications continue after the initial medications are discontinued.¹³⁶

An inappropriate prescribing cascade is a sequence of events that begins when an adverse drug event is misinterpreted as a new medical condition or exacerbation of an existing condition and a subsequent drug is prescribed to treat this new or worsening condition;¹³⁷ they increase the risk of new adverse drug events, increase pill burden, reduce quality of life and increase costs to both the individual and health care system .^138–141 One example of a prescribing cascade is calcium-channel blocker-related peripheral edema is not recognized as a side effect of the medication, but a new medication, a diuretic, is prescribed to treat the peripheral edema (Supplemental Figure 1). More than 20 prescribing cascades have been identified in the literature; many of these are because of medications prescribed for common conditions such as hypertension, diabetes, dementia and chronic pain.^142–144 Clinicians should scan medication lists for potential cascades, and, for new symptoms, consider whether they represent a side effect of an existing medication instead of a new medical condition. The potentially-offending medication may be reduced, stopped, or non-pharmacologic strategies may be sufficient to manage the symptom. Monitoring for improvement after employing these strategies informs next steps and a prescribing cascade may be averted.¹³⁹

6.2. Safe deprescribing

Deprescribing, “the planned and supervised process of discontinuing or reducing medications that may be causing harm or are no longer providing benefit”¹⁴⁵ is integral to drug stewardship, having the potential to minimise medication-related problems and medication burden. Deprescribing often is reactive (i.e., occurring after an adverse event or symptom); however, proactive deprescribing is a hallmark of effective drug stewardship.

Deprescribing interventions have demonstrated effectiveness in improving appropriateness of medication regimens and reducing medication count without increased risk for long-term adverse outcomes.^145,146 These interventions demonstrate potential improvements in medication costs and quality of life. While there is no clear evidence on the impact of deprescribing on CKD progression, deprescribing remains relevant to drug stewardship. There are five steps to the deprescribing process (Box 2). Communication between all clinicians and pharmacies is critical throughout. Clinicians in primary care often do not know when or how to do deprescribing in people with CKD;¹⁴⁷ therefore, optimal deprescribing should involve clinician education, decision support tools, and/or pharmacist engagement.

Box 2. Key steps in the process of deprescribing medications in people with CKD.

1. Obtain an accurate medication list, do medication reconciliation and include over-the-counter medications, herbal remedies, and supplements.

2. Identify medications that could be deprescribed: a) medications that may no longer be indicated, b) medications that are potentially inappropriate, c) medications with questionable efficacy. Deprescribing resources, including algorithms, help clinicians confirm that an individual patient is eligible for deprescribing.

3. Conduct shared decision-making discussion about whether to deprescribe a medication: engage the patient and/or caregiver for their input on deprescribing the medication. Elicit the patient’s goals of care and describe the risks and benefits of the medication. Describe the deprescribing process so patients can make informed decisions.

4. Initiate deprescribing of medication with tapering regimen: establish plan to achieve lower dose or complete discontinuation. Identify alternative therapy, if relevant. Adopt existing algorithms and/or engage a pharmacist to create tapering regimen, if indicated.

5. Monitor for symptoms: watch for symptoms that may represent drug withdrawal or ineffectiveness of the lower dose, with prompt patient-centred resumption if this occurs.

Broadly, clinicians can identify medications to deprescribe by looking for “legacy prescribing” (i.e., medications that are no longer needed)¹⁴⁸ or target specific medications . Additional medications to target for deprescribing are potentially inappropriate medications (PIMs), defined as those that carry more risk of harm than benefit, particularly in older adults. Because many PIMS affect kidney function or might cause adverse outcomes more commonly in people with CKD,^149,150 deprescribing PIMs is clinically relevant in nephrology practice. Example sources of guidance in this regard include the American Geriatrics Society Beers Criteria and the Screening Tool of Older Person’s Prescriptions (STOPP).^151,152

6.2.1. Planned deprescribing before elective surgeries or procedures

The rationale for temporary discontinuation of certain medications before elective surgery or procedures (e.g., contrast imaging) is to prevent perioperative AKI and other complications such as hypotension, metabolic acidosis, or hyperkalemia during the perioperative period. Medications that are recommended to be temporarily discontinued in the 48 to 72 hours prior to elective surgery include RASi and diuretics, to minimise risk of hypotension or volume depletion; metformin and SGLT2 inhibitors to minimise risk of lactic acidosis if GFR falls and euglycemic ketoacidosis with fasting, respectively; NSAIDs and aminoglycosides to minimise the risk of nephrotoxicity in the event of intra-operative hypotension.^153,154 Although clinically sound, available evidence to support these practices is of low-to-moderate quality.

Withholding medications before contrast imaging:

Recent evidence suggests there may be little benefit to withholding RASi before contrast exposure. In a 2017 systematic review of 3 small RCTs and 2 cohort studies of holding RASi before angiography, there were non-significantly fewer events in patients in whom RASi was held.¹⁵⁵ A summary adjusted observational studies of contrast nephrotoxicity post CT identified little additional risk in people with GFR>30 mL/min/1.73m², and that increased risk from contrast was inconsistently identified in people with GFR<30 mL/min/1.73m².¹⁵⁶ Further work on the effectiveness of withholding drugs and following up kidney function for contrast CT is warranted, stratified by GFR: it seems likely that the threshold for concern is <30 mL/min/1.73m². The Royal College of Radiologists alongside the Royal College of Emergency Medicine in the United Kingdom have recently published an advisory statement: pre-existing CKD, diabetes mellitus, or medications such as metformin should not delay emergency iodinated intravenous contrast CT imaging scanning. ¹⁵⁷

Withholding medications before major surgery:

A meta-analysis of RCTs and cohort studies showed that compared with continuing, holding RASi at the time of non-cardiac surgery was associated with no difference in MACE or mortality, but a 32–37% reduction in intra-operative hypotension.^158,159 A subsequent small RCT found no difference,¹⁶⁰ with another cohort study suggesting a trend to increase risk for AKI with continued use of RASi prior to surgery.¹⁶¹ However, no such trend was observed for biomarkers of structural kidney injury, suggesting that might be hemodynamic and more likely reversible.¹⁶¹ Other medications like sulphonylureas, metformin, and SGLT2i would be held because of fasting, and NSAIDs because they affect the risk of bleeding, postoperative GI ulceration and AKI. Case reports of a few patients developing ketoacidosis after withholding SGLT2i only for 24h before surgery lead some experts to recommend discontinuation for 48–72 hours.¹⁶²

These medications can safely be restarted after the procedure when there is normal blood pressure and oral intake; however, it is usual practice to continue to hold these medications should AKI occur. Medication reconciliation at discharge, if hospitalised, or specific information for primary care physicians to consider restarting, can help ensure medications are restarted and minimise the unintentional harm of discontinuation of evidence-based therapy. For example, discontinuation of RASi was associated with higher risk of heart failure hospitalization, cardiac events, CKD progression, AKI, and all-cause mortality.¹⁶³

6.2.2. Unplanned deprescribing during the treatment and resolution of adverse drug reactions

Temporary discontinuation of medications to manage adverse events is indicated in most cases. However, fear for adverse event recurrence often results in failure to resume treatments. In people with CKD, hyperkalemia or AKI are not uncommon adverse effects of RASi treatment, for which clinical guidelines recommend discontinuation of RASi and reinitiation at lower dose when the event is resolved.^164–166 Despite this advice, permanent discontinuation of RASi seems to be the most common clinical reaction to occurrence of adverse events.^167,168 While there are no clinical trials addressing whether to continue or to stop RASi after hyperkalemia, carefully-conducted observational studies consistently suggest that withholding RASi medication compared with continuing treatment after these adverse events is associated with a lower recurrence of adverse events, but a higher risk of MACE and death, for which RASi is mainly indicated.^169–172 In all these situations, enhanced communication with patients, and between inpatient and outpatient teams is necessary to ensure resumption of medications in a timely manner.

6.3. Deprescribing in special circumstances

During acute illness

People with CKD who experience acute illness are at heightened susceptibility for medication therapy problems due to the underlying disease state, volume status changes, hemodynamic alterations, altered end organ function, polypharmacy, drug-drug interactions, and, at times, uncertainty about their home medication regimen.¹⁷³ Many chronic medications are withheld in the context of acute illness because of altered risk-benefit, absence of indication (e.g., antihypertensives in shock) and drugs are substituted for more predictable bioavailability (e.g., parenteral therapies), more rapid onset, greater ease of titration (e.g., continuous infusion), or available companion diagnostics to monitor their safe use (e.g., therapeutic drug monitoring, Figure 5). Sedating medications eliminated by the kidney such as gabapentinoids¹⁷⁴ may be acutely contraindicated in patients admitted with altered mental status. People with CKD admitted with AKI and hyperkalemia may need RASi, aldosterone antagonists, diuretics and potassium supplementation held; whereas those admitted with volume overload may need augmentation of these medications. Finally, fasting or starvation, acute surgery, intrabdominal pathology such as gastroenteritis or pancreatitis, acute intoxication, and infection or sepsis potentiate the risk for ketoacidosis with SGLT2i.¹⁷⁵

Figure 5.

Impact of acute illness on medication management in CKD.

Patients with CKD use chronic therapies to prevent long term morbidity mortality (e.g., hypertension management, diabetes management). When an acute illness occurs, chronic medications may be continued, dose adjusted, or discontinued. New medications which are renally active may be introduced alongside devices to support management of the acute illness. As the patient’s condition stabilises and they transition from acute care to the post-acute care setting, the modifications to their medication program may unintentionally or intentionally persist; The vignette illustrates a case example of a patient with CKD and diabetes, with a history of neuropathy and coronary artery disease that present to the hospital with a myocardial infarction, cardiogenic shock, and type 1 cardiorenal syndrome. Chronic therapies such as RASi, SGLT2i, and loop diuretics would typically be held in favor of acute interventions to stabilise the shock including vasopressor therapy. Reduced GFR would prompt dose adjustment for medications eliminated by kidney like gabapentin to limit oversedation. As the patient’s condition stabilises, therapies like diuretics may be reintroduced to facilitate decongestion. At the point of discharge, drugs may or may not be resumed depending on resolution of ongoing kidney dysfunction, and existence of new comorbidities. There may also be inadvertent medication errors at care transition due to the plethora of changes which occurred as a function of the acute event. Abbreviations: CKD: Chronic kidney disease; RASi: Renin Angiotensin System inhibitor; SGLT2i: Sodium Glucose Co-transporter 2 Inhibitors.

The acute illness and altered level of consciousness may limit the patientś capacity to provide an accurate account of home prescription medications, OTC therapies, herbal remedies or dietary supplements. This may lead to error, including failure to recognise withdrawal syndromes. During acute illness, a patient’s clinical status and GFR may vary day-to-day requiring daily drug stewardship to minimise errors, especially through care transitions and rapid changes in health status.¹⁷⁶

Parenteral medications (e.g., parenteral vancomycin or aminoglycosides) introduced in acute illness may be nephrotoxic and influence the patient’s cumulative nephrotoxin burden.¹⁷⁷ While the risk for medication therapy problems is heightened, there is a greater intensity of monitoring in the acute care setting. Vital sign evaluation and laboratory monitoring occur regularly, which aids in improving medication safety.

During pre-conception, pregnancy and lactation

For people with established CKD, pregnancy is associated with risk of CKD progression.¹⁷⁸ Guidelines advise that some evidence-based medications that slow or prevent CKD progression, should not be prescribed during pregnancy because they are teratogenic (e.g., RASi,¹⁷⁹ mycophenolate mofetil¹⁸⁰) or have not been studied in this population.^90,181 Some CKD-specific medications can be safely continued during pregnancy, including hydroxychloroquine, tacrolimus, cyclosporin, eculizumab, prednisone, azathioprine, colchicine, and intravenous immunoglobulin. Biologics such as rituximab, belimumab and abatacept do not cross the placenta until the 15^th week of gestation,¹⁸² and discussion of the benefits and risks of these medications in the context of conception and pregnancy is essential for shared decision-making. Some patients will wish to replace teratogenic drugs when they start trying to conceive. Others, for whom the beneficial effects of the drugs in question are needed during the uncertain and often prolonged interval of trying to conceive, will prefer a careful monitoring plan that takes into account cycle irregularity, and discontinue teratogenic medications at conception.¹⁸² Given the importance of disease control during gestation in terms of pregnancy outcomes, it may be most prudent to continue some medications throughout pregnancy, highlighting the critical nature of shared decision-making between patient and health care practitioner. During lactation, thoughtful consideration of the risks and benefits of the drug and of lactation (to the patient and their child) and the patient’s preferences about lactation is essential, along with the recognition that some medications suitable for use during pregnancy may not be appropriate for lactation, and vice versa.¹⁸³ Multidisciplinary care with obstetrics, midwifery, lactation consultants and other subspecialty care will be essential at different points in the journey through pre-conception, pregnancy and lactation.⁹⁰

7. Strategies to promote Drug Stewardship in CKD

7.1. Patient education and patient empowerment

People with CKD share responsibility for drug stewardship. They can communicate the CKD diagnosis with non-nephrology healthcare providers for awareness and an assessment of drug selection and dose appropriateness. To facilitate medication safety, people with CKD should be informed and regularly reminded of the expected benefits and possible risks of their medications, carry an up-to-date list of medications, know their GFR, and recall prior adverse drug reactions.

Medication adherence rates for chronic medical conditions are often estimated at around 50% with little evidence of change with time.^184–186 Because medication adherence is critical for CKD self-management, it is essential to provide education on medication benefits, show patients that the treatment is working, and alleviate concerns about adverse effects.¹⁸⁷ Patient education should be inclusive for each population (i.e., literacy level, languages). While brochures and conversations may be useful, interactive digital health applications have been shown to be acceptable to patients and may lead to more effective knowledge gain.^188–192 Practical implementation tips involve printing out the results of the most recent eGFR for the patient to share in future healthcare consultations, and providing a written list of medications and medication changes.

There are growing concerns regarding the use of falsified and substandard medications in LIC and LMIC which pose potential harm, particularly to those people at risk of, and with, CKD. Patients and their families should be aware that medication falsification is often associated with illicit internet supply. Many vulnerable communities, people with low health literacy, and those in countries with less rigorous regulatory systems are more at risk of medication falsification. Increased global awareness is important and healthcare workers should be aware of and alert patients on these issues.

7.1.1. Implementation of sick day rules

Sick day rules refer to patient instructions to temporarily discontinue specific medication in the setting of a dehydrating illness to minimise the risk of AKI or medication accumulation in the setting of AKI. The acronym SADMAN(S) (sulphonylureas, ACEi, diuretics & direct renin inhibitors, metformin, angiotensin receptor blockers [ARB], NSAIDs, to which SGLT2i were subsequently added) was developed to help practitioners recall the classes of drugs to be temporarily discontinued – the actual drug names are discussed with the patient and written on patient education materials.^193,194 Sick day rules have been consensus-based and included in practice guidelines^195,196; however, evidence in support of sick day rules to prevent AKI or other clinically relevant outcomes is lacking.¹⁹⁷ Data instead suggest potential harm if people with CKD make mistakes in recognizing dehydrating illness or about which drugs to stop.^198–200 The complexity of sick day rules is challenging for any patient; challenges that are magnified by cognitive impairment, low literacy, or visual impairment in the patient or caregiver.²⁰⁰ The most reported problem is failure to restart medication.¹⁹⁹ Figure 6 shows the steps that must occur correctly for sick day rules to be implemented appropriately. If sick day rules are to be provided, along with conversations to optimise understanding, patient education materials on sick day rules are critical and must include written material with actual medication names.^195,199,200 The plan to restart medications should be detailed in the medical records and clearly communicated to patients. A structured medication review within one month may be warranted to ensure appropriate medications are restarted. Supplemental Table 4 critically examines the evidence to date on benefits and harms of sick day rules, which in our opinion is insufficient to justify widespread adoption.

Figure 6.

Essential steps for appropriate sick day rule implementation

This figure shows the steps which have to be undertaken correctly by patients for sick day rules to be implemented as intended. Patients need to be able to recognise a dehydrating illness, identify the medication(s) to hold, stop the medication(s), and understand the medication(s) should be restarted after they recover. From the 2024 KDIGO guidelines on CKD screening, detection and management

7.2. Clinical resources and collaboration with pharmacists

Healthcare providers should establish collaborative relationships with pharmacists and use tools to ensure and improve drug stewardship in people with CKD to enhance management of their complex medication regimens. Strategies to improve drug stewardship by multidisciplinary interactions between nephrologists and clinical pharmacists provide safe and cost-effective care in people with CKD.^201–203

Clinical decision-support systems can facilitate drug stewardship through medication dose recommendations in CKD, nephrotoxin surveillance, deprescribing guidance, and medication review and reconciliation prompts at care transitions. Interventions involving electronic clinical decision support systems and/or electronic prescribing have demonstrated improvements in medication error rates, drug-drug interactions, communication among clinicians and patients, and appropriate dosing of medications primarily excreted by the kidneys.^204–209 A paramount example of electronic tools for improving drug stewardship in CKD was pioneered by Chertow et al.²¹⁰ who reported that a trigger software to inform on the need of dose adjustments for people with CKD in an urban tertiary care teaching hospital reduced the fraction of prescriptions with overdosing for the patient’s kidney function. More recently, the Nephrotoxic Injury Negated by Just-in time Action (NINJA) program²¹¹ reported potential for prevention of nephroxic medication induced AKI. NINJA was a clinical-decision electronic triggering system that screened non-critically ill hospitalized children in 9 centers in the US for the presence of over three nephrotoxic medications on the same day or an intravenous aminoglycoside over three consecutive days. If encountered, the software then recommended obtaining a daily serum creatinine level for the duration of, and two days after, exposure ending. Additionally, substitution of equally efficacious but less nephrotoxic medications for exposed patients was also recommended when possible. Implementation of NINJA resulted in a 20% decrease in the rate of nephrotoxic medication associated AKI ²¹¹.

Other forms of digital health may optimize drug stewardship in all clinical settings²¹². An understudied resource that may enhance monitoring of medications and their effects is point-of-care testing.²¹³ Across a given healthcare system, care continuity and transdisciplinary collaboration is facilitated through a shared electronic health record. Pharmacists and other healthcare providers in the community environment may have less access to comprehensive patient information, making assessment of drug appropriateness in CKD more difficult. Efforts to minimize fragmentation in medical records are needed to optimally engage the entire health care team in drug stewardship.

7.3. Health system and health policy barriers to drug stewardship in CKD

One barrier to drug stewardship is limited medication access. Access to medications varies globally, with many options still unavailable or unaffordable to people living in lower- (LIC) and lower middle-income (LMIC) countries²¹⁴. The International Society of Nephrology (ISN) reported recently that only 35% of people in LIC countries have, for example, access to RASi, statins, and insulin.²¹⁵ There are also numerous barriers to additional important medications for management of CKD complications, such as erythropoietin analogues and intravenous iron infusion. Despite the demonstrated cardiorenal benefits of SGLT2i and glucagon-like peptide-1 receptor agonists,^216–219 these medications are still not available in many jurisdictions, nor considered in the national essential medication lists.²²⁰ Additional barriers for medication access involve healthcare reimbursement policies, insurance coverage or lack of affordability for out-of-pocket costs, which perpetuate health inequalities. Clinicians in these settings must focus on CKD prevention through lifestyle modification and risk factor reduction. More broadly, the nephrology community has a responsibility to address this inequity. This may involve direct and transparent dialogue with pharmaceutical companies to ensure global access to lifesaving and life-changing medications for those with CKD. Indefinite coverage of immunosuppression for recipients with kidney transplants should be mandated globally. Within countries, healthcare workers and patients should campaign for universal drug coverage that is free at the point of use. The WHO and individual countries should expand their lists of essential kidney-related medications. A global program that utilises generic formulations should also be considered as a tool to reduce the costs of drug delivery.

Slow adoption of evidence-based interventions (i.e., new therapies for CKD management) is also a barrier to drug stewardship, perpetuating health inequities and disparities in access to quality care. There are many barriers to the implementation and translation of evidence-based interventions into clinical practices and settings. For example, poor uptake and adoption of clinical practice guidelines owing to the lack of appropriate education among patients, health professionals (such as family physicians) and other stakeholders regarding new research findings and their effectiveness, the clinical inertia, feasibility and scalability of applying the interventions in ‘real-life’ settings. Implementation research, which considers the local environment in which these interventions will be delivered and embraced by the community, patients and policy-makers, needs to be funded and supported alongside clinical trials to optimise the likelihood of successful and sustainable implementation of evidence-based interventions.^221,222

8. Conclusion

People with CKD are particularly vulnerable to medication-related problems. Drug stewardship is essential to both minimizing medication-related problems and optimizing health for people with CKD. Current evidence summarized here provides a foundation for effective drug stewardship; however, knowledge gaps and health system and policy barriers persist. Where knowledge gaps exist, drug stewardship principles around drug dosing and close monitoring are paramount. There is a critical need to address key barriers to drug stewardship: limitations in medication access, and slow implementation of evidence-based care. Medication access inequities remain a priority for the promotion of health. Drug stewardship seeks to maximise the benefits and minimise risks in the context of the resource constraints affecting the patient and practitioner and seeks a more equitable future.

Key Points.

• Medication reconciliation and review is an essential first step in patient-centered drug stewardship.

• Doses should be adjusted according to GFR.

• For many medications, different creatinine-based equations to estimate GFR (CG, CKD-Epi, LMR, EKFC) are interchangeable, and most pharmacokinetic information available derives from creatinine-based equations.

• For patients with extreme body habitus (i.e., BSA is not close to 1.73m²), calculation of a non-normalised eGFR based on actual body surface area, using the formula eGFR in mL/min = eGFR in mL/min/1.73 m² × [BSA (m²)/1.73 m²] increases accuracy.

• When precision is needed due to narrow therapeutic range, equations that utilize both creatinine and cystatin C, or directly measuring GFR should be advocated.

• Drug choice should consider relative and absolute contraindications by GFR, and the increased nephrotoxicity of nephrotoxic medications to people with CKD.

• Acute illness and fluctuating GFR should prompt frequent reassessment of GFR and medications.

• For people with CKD who might become pregnant, education around potential teratogens is important, but people should not be denied medication on the basis that they might become pregnant.

• Pregnancy and lactation require re-evaluation of medications and consideration of patients’ perspectives.

• Culturally-appropriate multi-faceted educational and empowerment activities may include written instructional materials and digital technologies.

• Implementing routine sick day rules has a substantial cost or opportunity cost, is not supported by evidence, and may cause more harm than good.

• Healthcare workers and patients should collaborate to promote more equitable access to evidence-based medications within their countries and globally.

Box 3. Proposed policy changes to improve access to essential medications that prevent CKD progression and treat CKD complications.

1. Affirmation that access to essential medications as the fundamental human rights.

2. Call on individual countries and states to develop, promote access to and prescribe generic essential medications, while acknowledging intellectual property rights.

3. Development of a global essential medicines pricing monitoring system.

4. Development of transparent pricing policies with pharmaceutical companies

5. Advocate for universal and indefinite coverage for essential and life-sustaining medications such as immunosuppression for kidney transplant recipients.

6. Engagement with international nephrology societies such as the ISN to create a global accountability mechanism for monitoring access to essential medicines in people with CKD. This may include working with pharmaceutical companies and individual jurisdictions to waive taxes and tariffs, and reduce the ‘mark-up’ pricing along the supply chain.

7. Global research and development partnership between the private and public sector to develop and deliver essentials medicines that are both affordable and sustainable in all resource settings.

Glossary

Medication-related problem

types of problems involving medications that can interfere with patient outcomes

Prescribing Cascade

a circumstance when an additional medication is initiated in response to side effects of an initial medication

Potentially inappropriate medications

a term used to describe medications that carry greater risk of harm than benefit in older adults

Absolute Contraindication

a patient circumstance in which risk of medication use clearly outweigh any potential benefit

Relative Contraindication

a patient circumstance in which clinicians should take caution when a medication is prescribed

Deprescribing

a systematic process for reducing or discontinuing a medication

Adverse event

An undesired effect of a drug. Adverse events can range from mild to severe and can be life-threatening. Also called adverse effect and adverse reaction