Whichestimate of renal function should be used when dosing patients with renal impairment?
Prepared by UK Medicines Information (UKMi) pharmacists for NHS healthcare professionals
Before using this Q&A, read the disclaimer at
Date prepared: February 2018
Background
Accurate measurement of renal function is essential in patients with renal impairment (RI) so that drug dosages can be adjusted accordingly.Direct measure of Glomerular Filtration Rate (GFR) using plasma or urinary clearance is considered the best overall index of renal function. However, this is difficult to do in practice1.
Answer
Clinical laboratories routinely report renal function in adults based on estimated glomerular filtration rate (eGFR) normalised to a body surface area of 1.73 m2 – this is derived from either the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula or the Modification of Diet in Renal Disease (MDRD) formula[1].Current NICE guidance recommends the use of the CKD-EPI creatinine equation to estimate GFR or the use of equations incorporating an alternative marker – cystatin C[2].
All creatinine-based equations for estimating GFR have limitations, although the CKD-EPI equation is more accurate than MDRD for values of GFR >60mL/min1,[3],[4]or in people at low risk of CKD[5].Drug dosing calculations for patients with RI have traditionally been based on estimations of CrCl using Cockcroft & Gault (C&G)[6] and the vast majority of published drug dosing information is based on C&G estimation of CrCl1,[7].
Despite this, advice on adjustment of drug doses in RI in the BNF is usually expressed in terms of eGFR1. Although the two measures of renal function are not interchangeable, the BNF advises that, for most drugs and for most adult patients of average build and height, eGFR(rather than CrCl) can be used to determine dosage adjustments1. The National Kidney Disease Education Program in the USA has also endorsed the use of eGFR for drug dosing in renal disease4,[8].Exceptions to the use of eGFR include toxic drugs, in elderly patients and in patients at extremes of muscle mass where calculation of CrCl is recommended1.There is no compelling evidence to support the superiority of any given method for drug dosing in all patient populations or clinical situations1.
Estimating renal function in patients at extremes of body weight
Because serum creatinine levels are dependent on muscle mass and diet,in some individualse.g. frail, elderly, critically ill or patients with cancer or muscle wasting diseases, body builders, amputees, vegans, the use of creatinine-based equations will lead to an over- or under- estimation of the CrCl1,4,[9].
The BNF advises that in patients at both extremes of muscle mass (BMI of less than 18kg/m2 or greater than 40kg/m2) the absolute GFR or CrCl (calculated from the C&G formula) should be used to adjust drug dosages1.The C&G equationuses body weight as a marker of muscle mass (creatinine being a breakdown product of muscle)6. Therefore in obese or extremely underweight patients there is also potential for over- or under- estimation of CrCl. Ideal body weight (IBW) should be used when calculating CrCl using C&G particularly in oedematous patients and patients with ascites9. Where the patient's actual body weight is less than their IBW, actual body weight (ABW) should be used instead1.For obese patients IBW can be used, but some experts have suggested that an adjustment factor of 40% be applied to the patient’s excess weight over their ideal weight i.e. adjusted ideal body weight = [IBW + (0.4x ABW-IBW)] 9,[10],[11].Clinical judgement is needed, e.g. if a patient's excess weight is due to high muscle mass not excess body fat, ABW should be used6. Others have proposed the use of a CrCl range for drug dosing purposes, with the lower boundary defined by using IBW in the CG equation and the upper boundary by using ABW[12].
The absolute GFR is determined by removing the normalisation for BSA from the eGFR using the following formula1:
GFR (Absolute) = eGFR x (individual's body surface area / 1.73)
The ideal body weight is calculated as follows:
Ideal body weight (kilograms) = Constant + 0.91 (Height - 152.4)
Where:
•Constant = 50 for men; 45.5 for women
•Height in centimetres
Failure to correct to absolute, non-normalised GFR in patients with a BSA smaller than 1.73m2 will overestimate GFR and potentially result in drug overdosing. Conversely, in patients with a BSA greater than 1.73m2 this will underestimate GFR and will potentially result in drug under-dosing6,7.
Estimating renal function in elderly patients
The production and excretion of creatinine decline with age, therefore normal serum creatinine values may not represent normal renal function in older patients[13].The normal process of ageing involves the loss of nephrons and therefore it is reasonable to assume that all elderly patients have some degree of renal impairment9.
There is evidence that in elderly patients with CKD the use of the MDRD14-17or CKD-EPI18 (adjusted or non-adjusted18) equations to calculate eGFR instead of an estimated CrCl overestimates GFR and leads to the calculation of higher than recommended doses of renally excreted drugs such as enoxaparin, gentamicin, digoxin, amantadine, gabapentin, non-Vitamin K antagonist oral anticoagulants (NOACs), tramadol and ramipril in a large proportion of patients[14],[15],[16],[17],[18],[19]. For example, a general-practice based study in over 4100 patients with atrial fibrillation has shown that using MDRD instead of C&G to calculate eligibility and dose of dabigatran or rivaroxaban would have led to 14.9% of patients aged ≥ 80 being incorrectly judged eligible for dabigatran treatment. Furthermore, in patients aged <80, 13.5% of patients would have received too high a dose of rivaroxaban based on the MDRD equation.Alternative equations designed to improve the accuracy of estimating GFR in older patients are being developed, but further study is needed to verify if this could be translated into recommendations for drug dosing [20],[21]. The BNF advises the use of C&Gto estimate renal function in elderly patients aged 75 years and over1.
Estimating renal function in critically ill patients
Estimating CrCl from a serum creatinine level assumes that renal function is stable9over several days or moreand that the serum creatinine level is fairly constant. With rapidly changing renal function the serum creatinine levels will no longer reflect the true creatinine clearance rate, e.g. in acute kidney injury (AKI) where estimation of GFR will overestimate or underestimate kidney function during the evolution of injury and recovery, respectively[22].Serum creatinine levels lag behind the acute injury process, with levels only rising a day or so after the initial renal insult[23].Also patients with hepatorenal syndrome may have renal dysfunction that is substantially more severe than issuggested by the serum creatinine. Both urea and creatinine production may be substantially reduced in this setting, due to the liver disease and to decreased muscle mass and decreased protein intake1,[24].
Interpretation of resultsand choice of drug dose
In a study comparing three equations for estimating GFR in 128,805 patients undergoing percutaneous coronary intervention, there was considerable discrepancy in the classification of patients into stages of CKD. The equations used were C&G, CKD-EPI and MDRD. Equation choice affected drug-dosing recommendations for antiplatelet and antithrombotic agents, with the formulas agreeing for only 34% of patients with a GFR of <30mL/min/1.73m2 (as estimated by at least one equation). The authors comment that the eGFR equation selected to determine drug dosing in the clinic ideally would be identical to the equation used in the drug’s original pharmacokinetic study; although many pharmacokinetic studies have used the C&G equation there is no standard approach[25]. For new drugs most studies are designed and performed by the manufacturer, therefore the manufacturer's dosage in RI recommendations should be followed18. A study comparing secondary sources of prescribing information for patients with RI found a considerable degree of variability amongst the definitions and recommendations in four different standard sources[26],[27].
Published data on drug dosage adjustment in RI include mainly case reports and pharmacokinetic studies in small numbers of subjects28. It is also worth noting that historically, there has been substantial variability in serum creatinine values reported by different clinical laboratory creatinine measurement methods. Consequently, the results of pharmacokinetic studies on which this dosing information was based, were dependent upon the particular method for measuring serum creatinine used in a given study4,[28]. However it is not possible or practical to repeat all the studies using a standardised creatinine measurement method4,28. The estimated GFR based on current standardised creatinineassays is likely to lead to different dosage recommendations from those intended by the original study, even if the same estimating equation is used, because of the change in analytical methodology28.
In addition the level of RI is often defined differently among the pharmacokinetic studies and each category (‘mild’, ‘moderate’, ‘severe’) often encompasses a broad range of kidney function. The drug dosage adjustment recommendations that use broad ranges of kidney function may not be optimal for all patients whose kidney function lies within the specified range, especially for drugs that have a narrow therapeutic index28.
One author argues that the most appropriate formula for GFR estimation for drug dosing decisions in all patientsis the CKD-EPI formula with removal of BSA normalisation if needed (i.e. calculate BSA and then multiply the CKD-EPI equation by BSA/1.73)[29].
Where an accurate GFR is deemed necessary e.g. in chemotherapy dosing, an isotope GFR determination should be performed7.
In all patients clinical judgement should be used alongside any estimates derived from equations as, in some patients, the clinical circumstance may suggest that a lower or higher dose be used than is indicated by the drug dosing guidelines28.
Summary
The vast majority of published drug dosing in RI information is based on Cockcroft & Gault (C&G) estimation of creatinine clearance (CrCl). However it has been replaced in clinical practice by the Modification of Diet in Renal Disease trial (MDRD) and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations for estimating GFR.
Although these equations are not interchangeable, the British National Formulary(BNF) advises that for most drugs and for most adult patients of average build and height, dosage adjustment based on the eGFR is acceptable.
Exceptions to the use of eGFR include toxic drugs, in elderly patients and in patients at extremes of muscle mass where calculation of CrCl is recommended.
The BNF advises the use of C&Gto estimate renal function in elderly patients aged 75 years and over.
The absolute GFR or CrCl calculated by the C&G formula should be used to adjust drug dosages in patients at extremes of body weight (BMI <18kg/m2 or >40kg/m2). If using C&G for these patients it may be necessary to base the calculation on ideal body weight (IBW) or adjusted ideal body weight. Where the patient’s actual body weight (ABW) is less than their IBW, ABW should be used.
One author argues for the use of the CKD-EPI formula to estimate GFR for drug dosing decisions in all patients, with removal of BSA normalisation if needed depending on the patient’s size (i.e. calculate BSA and then multiply the CKD-EPI equation by BSA/1.73).
In critically ill patients with rapidly changing renal function the serum creatinine levels will no longer reflect the true creatinine clearance rate.
Where an accurate GFR is considered necessary e.g. in chemotherapy dosing, an isotope GFR determination should be performed.
The eGFR equation selected to determine drug dosing in practice ideally would be identical to the equation used in the drug’s original pharmacokinetic study.
In the past, pharmacokinetic studies have been subject to variability in serum creatinine assays performed by different clinical laboratory creatinine measurement methods and published data on drug dosage adjustment in RI include mainly case reports and pharmacokinetic studies in small numbers of subjects.
The manufacturer's SPC will often provide guidance on dosing in RI, especially for new drugs.
Clinical judgement should be used alongside any estimates derived from equations.
Limitations
This Q&A is only applicable to adult patients with renal impairment(RI). For information on estimation of renal function or drug dosage adjustment in children or in pregnant patients with RI, please consult the latest BNF, BNF for children, SPC and/or specialist sources of information[30].
References
Available through Specialist Pharmacy Service at
([1]) Joint Formulary Committee. British National Formulary (online). Prescribing in renal impairment. London: BMJ Group and Pharmaceutical Press. Accessed via: 15.02.18.
([2]) National Institute for Health and Care Excellence. Clinical Guideline CG182. Chronic kidney disease in adults: assessment and management. 23 July 2014. Accessed via on 15.02.18.
([3]) 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease KDIGO. Kidney International 2013; 3: supplement. Accessed via on 21.02.18.
([4]) National Institute of Diabetes and Digestive andKidney Diseases. National Kidney Disease Education Program. CKD and drug dosing: information for providers. Revised April 2015. accessed on 16.2.16 and 06.02.18.
([5]) Rule AD. The CKD-EPI equation for estimating GFR from serum creatinine: real improvement or more of the same? Clin J Am Soc Nephrol 2010; 5: 951-53.
([6])Brown C. Prescribing principles for patients with chronic kidney disease. Pharmacy in Practice January/February 2008p.23-27 09.03.18.
([7]) Ashley C, Dunleavy A. editors. Renal Drug Handbook. 4th Edition 2014. Radcliffe Publishing Ltd p. xiii – xvi.
([8]) Stevens LA, Nolin TD, Richardson MM et al. Comparison of drug dosing recommendations based on measured GFR and kidney function estimating equations. Am J Kidney Dis 2009; 54: 33-42.
([9]) Wills S, Badiani A, Power J. Southampton Medicines Advice Service. Medicines Learning Portal. Clinical topics. Assessing renal function. Last updated: Sunday, July 12, 2015. Accessed via: on 15.02.18.
([10])Kane SP. ClinCalc LLP. Drug Dosing in Obesity Reference Table. Accessed via on 22.01.16 and 06.02.18.
([11]) Bouquegneau A, Vidal-Petiot E, Moranne O et al. Creatinine-based equations for the adjustment of drug dosage in an obese population. Br J Clin Pharmacol 2015; 81: 349-61.
([12]) Brown DL, Masselink AJ, Lalla CD. Functional range of creatinine clearance for renal drug dosing; a practical solution to the controversy of which weight to use in the Cockcroft-Gault equation. Ann Pharmacother 2013; 47: 1039-44.
([13]) Munar MY, Singh H. Drug dosing adjustments in patients with chronic kidney disease. Am Fam Physician 2007;75:1487-96.
([14]) Denetclaw TH, Oshima N, Dowling TC. Dofetilide dose calculation errors in elderly associated with use of the Modification of Diet in Renal Disease equation. Ann Pharmacother 2011; 45: e44.
([15])Gill J, Malyuk R, Djurdjev O et al. Use of GFR equations to adjust drug doses in an elderly multi-ethnic group-a cautionary tale. Nephrol Dial Transplant 2007; 22: 2894-99.
([16]) Hellden A, Odar-Cederlof I, Nilsson G et al. Renal function estimations and dose recommendations for dabigatran, gabapentin and valaciclovir: a data simulation study focused on the elderly. BMJ Open 2013; 3: e002686.
([17]) Cabello-Muriel A, Urbieta-Sanz E, Iniesta-Navalon C et al. Influence of equation used to estimate the renal function in dosage potentially nephrotoxic drug. Eur J Hosp Pharm 2015; 22:23-27.
([18]) Cartet-Farnier E, Goutelle-Audibert L, Maire P et al. Implications of using the MDRD or CKD-EPI equation instead of the Cockcroft-Gault equation for estimating renal function and drug dosage adjustment in elderly patients. Fundamental & Clinical Pharmacology 2017; 31: 110-119.
([19]) MacCallum PK, Mathur R, Hull SA et al. Patient safety and estimation of renal function in patients prescribed new oral anticoagulants for stroke prevention in atrial fibrillation: a cross-sectional study. BMJ Open 2013;3:e003343. Doi:10.1136/bmjopen-2013-003343.
([20]) Garasto S, Fusco S, Corica F et al. Estimating glomerular filtration rate in older people. Biomed Res Int. 2014; Article ID:916542. doi: 10.1155/2014/916542. Epub 2014 Mar 20.
([21]) Lamb EJ, Stevens PE, Deeks JJ. What is the best glomerular filtration marker to identify people with chronic kidney disease most likely to have poor outcomes? BMJ2015;350:g7667 (Published 12 January 2015) Accessed via on 03.12.15.
([22]) Wagner LA, Tata LA, Fink JC. Patient safety issues in CKD. Am J Kidn Dis 2015; 66: 159-169
([23]) Chadwick L, Macnab R. Laboratory tests of renal function. Anaesth Intens Care Med 2015;16:257-61.
([24])Runyon BA. Hepatorenal syndrome in UpToDate. Wolters Kluwer. accessed via on 08.02.18.
([25]) Parsh J, Seth M, Aronow H et al. Choice of estimated glomerular filtration rate equation impacts drug-dosing recommendations and risk stratification in patients with chronic kidney disease undergoingpercutaneous coronary interventions. J Am Coll Cardiol 2015; 65: 2714–23.
([26]) Vidal L,Shavit M, Fraser A et al. Systematic comparison of four sources of drug information regarding adjustment of dose for renal function. Brit Med J 2005; 331: 263-6.
([27]) Mehta DK, Sweetman SC, McEvoy GK et al. Dose adjustment in renal impairment. Brit Med J 2005; 331: 292-4.
([28]) Matzke GR, Aronoff GR, Atkinson Jr AJ et al. Drug dosing consideration in patients with acute and chronic kidney disease – a clinical update from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney International 2011; 80: 1122-37.
([29]) Jones GRD. Estimating renal function for drug dosing decisions. Clin Biochem Rev 2011; 32: 81-88.
([30]) Rodieux F, Wilbaux M, van den Anker JN et al. Effect of kidney function on drug kinetics and dosing in neonates, infants, and children. Clin Pharmacokinet 2015; 54:1183–1204.
Quality Assurance
Prepared by
Julia Kuczynska, South West Medicines Information and Training, Bristol
Date Prepared
16th February 2018
Checked by
Michèle Skipp, South West Medicines Information and Training, Bristol
Date of check
22nd March 2018
Search strategy
- Embase: [exp *KIDNEY FAILURE or exp *ACUTE KIDNEY FAILURE or exp *CHRONIC KIDNEY FAILURE ] and [exp *DRUG ADMINISTRATION or exp *PHARMACOKINETICS] [Limit to: Publication Year 2015-2018 and HUMAN ] 16.01.18
- Medline:exp *RENAL INSUFFICIENCY + [exp *DRUG ADMINISTRATION SCHEDULE or exp *PHARMACOKINETICS] [Limit to: Publication Year 2015-2018 and Humans] 11.01.18 (HDAS/Proquest – first 125 references; PUBMED – 52 references)
- In-house drug dosing in renal failure database and resources 17.01.18
- Internet Search: Google; DRUG DOSING KIDNEY FAILURE or DRUG DOSING KIDNEY FAILURE site: nhs.uk 17.01.18 (first 50 references)
- Renal Drug Database 24.01.18
- NHS Evidence 25.01.18 “PRESCRIBING KIDNEY” (first 50 references)
- NHS Evidence(KIDNEY FAILURE and [PHARMACOKINETICS or DRUG ADMINISTRATION]) 07.02.18(first 50 references)