Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Estimating glomerular filtration rate in changing drug dosage regimens
- Jeffrey K Aronson (10 June 2007)
-
Benefits of eGFR reporting
- Charlotte E Bebb, Richard P Burden (12 June 2007)
-
How wrong can we go?
- Jo Richardson (13 June 2007)
-
Perhaps just with the formula
- Michael A Innes (18 June 2007)
-
Estimating GFR – time to move forwards
- Edmund J Lamb, Charlie Tomson, Edmund Lamb, Kathryn Griffith, Donal O'Donoghue, John Feehally (19 June 2007)
-
Protagonists must demonstrate the value of eGFR
- James A Dickinson (20 June 2007)
-
Estimated GFR in primary care: the value of awareness!
- Liam G Glynn, Andrew W Murphy, Professor of General Practice, National University of Ireland, Galway, Ireland (25 June 2007)
-
How eGFR has improved patient management
- Shelagh E O'Riordan, Claire Beeson (5 July 2007)
-
Creatinine-based equations and the true cost of convenience
- Charles J. Diskin (5 July 2007)
|
|
|||
|
Jeffrey K Aronson, Reader in Clinical Pharmacology Department of Clinical Pharmacology, Radcliffe Infirmary, Woodstock Road, Oxford OX2 6HA
Send response to journal:
|
In their excellent article on estimated glomerular filtration rate (eGFR), Giles and Fitzmaurice did not discuss another use of estimated renal function—to guide changes in the dosage regimens of drugs that are eliminated unchanged by the kidneys, or that have active metabolites that are eliminated by the kidneys, or whose pharmacodynamic effects are affected by renal insufficiency.1 This is particularly important for drugs that have a low therapeutic index. Recommendations about drug dosage regimens given in manufacturers' Summaries of Product Characteristics and in secondary sources, such as the British National Formulary, are based on creatinine clearance, not eGFR. It is therefore customary when altering dosage regimens in renal insufficiency to estimate GFR using calculated creatinine clearance, usually derived from the Cockcroft–Gault equation for adults2 or the Schwartz–Haycock equation for children.3 The eGFR estimated by the modified four-variable MDRD (Modification of Diet in Renal Disease) equation underestimates true GFR more than the Cockcroft–Gault equation does in younger patients and less in older patients; overall, MDRD underestimates true GFR more than Cockcroft–Gault does.4 There are further differences in critically ill patients with burns, in whom the Cockcroft–Gault equation gives a better prediction of GFR than the modified MDRD equation.5 There is currently no information on how to use the eGFR to calculate changes in drug dosage regimens. We should therefore continue to use calculated creatinine clearance to guide changes in drug therapy. Clinical biochemistry laboratories would help doctors if they reported not only the MDRD-derived eGFR in ml/min/1.73 m˛, but also the Cockcroft–Gault estimated creatinine clearance in ml/min/70 kg, for which only the age and sex of the patient are needed (and not also ethnicity, as for eGFR). General practitioners could programme the appropriate equations into their computerized records. Given both estimates of GFR, doctors could decide for themselves which to use in specific cases, adjusting the respective values, if need be, for the patient’s actual surface area or weight, the latter being more reliably measurable. Overestimating creatinine clearance for changing drug dosage regimens can lead to overtreatment and toxicity, while underestimating it can lead to undertreatment or even unnecessary avoidance of drugs for which the advice “avoid in severe renal impairment” is given in standard texts (for example, sulphonylureas, warfarin). It should therefore be stressed that dosage changes that are made on the basis of any surrogate measure of GFR should be regarded as initial estimates. Further dosage changes should be made, if necessary, through careful monitoring of beneficial and adverse effects by measuring clinical progress, or pharmacodynamic biomarkers of the actions of the drug, or serum drug concentrations, as relevant. References 1. Aronson JK. Drug therapy in kidney disease. Br J Clin Pharmacol 2007; 63: 509-11. Competing interests: JKA is a member of the Joint Formulary Committee of the British National Formulary and the Paediatric Formulary Committee of the British National Formulary for Children. However, the opinions expressed here do not necessarily reflect those of other members of those committees. |
|||
|
|
|||
|
Charlotte E Bebb, Consultant Nephrologist Nottingham University Hospitals (City Hospital Campus), Hucknall Road, Nottingham, NG5 1PB., Richard P Burden
Send response to journal:
|
We disagree with some of the points made by Giles and Fitzmaurice (1); they overlooked one of the main aims of the recent guidelines on Chronic Kidney Disease (CKD) (2) and did not taken into account the accepted definition of CKD stages 1 and 2, nor the recommendations on screening. Reducing late referral of people who are heading towards dialysis (and avoiding the associated poor outcome) was one of the intentions of the guidelines(2). The main reason for late referral is that glomerular filtration rate (GFR) can be very low when the serum creatinine is only modestly increased and the severity of the kidney disorder is underestimated. In spite of its shortcomings eGFR reporting is the best method available to aid interpretation of serum creatinine. A preliminary analysis of referrals to our Unit suggests that improvements are being made. Since the introduction of eGFR reporting (together with a programme of education in primary care), the proportion of new dialysis patients who had been referred late (defined as within 90 days) has fallen from 38% to 25% (p<0.01). We accept the limitations of GFR that are described but the diagnosis of CKD stages 1 and 2 does not depend on GFR alone. Stages 1 and 2 refer to people known to have another kidney problem – either functional (eg proteinuria) or structural (eg polycystic disease) - and in whom the GFR is at least 60 mL/min (normal or nearly so). The guidelines did not advocate a screening programme but recommended testing for kidney disease in those at risk, including patients with diabetes and hypertension in whom such testing has been normal practice for several years. The suggested improvements in sample collection and analysis deserve attention in their own right, irrespective of policies on eGFR reporting. While we acknowledge that there are improvements to be made, patient anxiety over borderline abnormal results should be avoidable by appropriate explanation, and timely referral at later stages should improve outcomes and reduce patient anxiety and pressures on services that late referral cause. References 1.Giles PD, Fitzmaurice DA. Formula estimation of glomerular filtration rate: have we gone wrong? BMJ 2007;334:1198-1200 (9 June). 2.Chronic kidney disease in adults: UK guidelines for identification, management and referral. London: Royal College of Physicians 2006. Competing interests: RB was a member of the Chronic Kidney Disease Guideline Development Committee. |
|||
|
|
|||
|
Jo Richardson, GP London E14
Send response to journal:
|
There is only one answer to Giles and Fitzmaurice`s question "have we gone wrong", with regard to the place of estimated glomerular filtration rate ( eGFR) in routine practice (1) - and that is a resounding "yes" . As the authors state, eGFR has been introduced as a population screening test ( the population being pretty much anyone deemed by a health professional to need "blood tests", irrespective of any risk factors for kidney disease), without the test meeting appropriate and accepted criteria (2). Until a better performing population screening test is available, one approach might be for laboratories to decouple requests for " u and e" from an automatic report of eGFR, so that clinicians may choose to make a positive request for eGFR in those patients whom they consider at risk, such as those described in the Renal Association guidelines. The ethics of the current situation demand that we halt the unecessary anxiety currently caused to large numbers of patients who receive false positive results. 1. Giles PD, Fitzmaurice DA. Formula estimation of glomerular filtration rate:have we gone wrong? BMJ 2007; 334:1198-1200 2. Richardson J. How to measure renal function in clinical practice. Estimated glomerular filtration rate in general practice. BMJ 2006;333:918 Competing interests: None declared |
|||
|
|
|||
|
Michael A Innes, GP Stirchley Medical Practice, Sandino Road, Stirchley. Telford. TF3 1FB
Send response to journal:
|
Giles and Fitzmaurice have questioned the widespread use of eGFR calculated using the modified MDRD equation in primary care, because of its unsatisfactory performance with lower ranges of creatinine. They have given evidence, of the fact, but not clearly stated that the equation was derived from a population solely with kidney disease, which partly explains the reason for its limited performance.(1,2) It is my understanding from personal communication with my local lab that, in calculating and reporting eGFR on all requests for U&E as directed by the national service framework, labs add a local correction factor that further modifies the MDRD equation to align it better with the local population. This may mean that results are more accurate than we are given to believe by the authors. There has been work to develop an equation that is based on a population that includes a range of kidney function from normal to abnormal, which is more representative for primary care, though still not ideal.(2,3) I agree that we need a measure that performs well if we are expected to use eGFR extensively for practical purposes, and there is great potential in that, not only for the later stages, but also for earlier intervention appropriate for primary care. Perhaps labs should change to using the quadratic GFR instead of the modified MDRD formula with fiddle factor. 1. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D: A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 130:461-470, 1999. 2. http://www.kidney.org/professionals/KDOQI/guidelines_ckd/p5_lab_g4.htm accessed 15/6/07 3. Rule AD, Larson TS, Bergstrath EJ, SlezaK JM, Jacobsen SJ, Cosio FG. Using serum Creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease. Ann Intern Med. 2004;141:929-937. 4. Jones J. Comparison of glomerular filtration rate estimation equations for the measurement of longitudinal renal disease outcomes. Proceedings of NAPCRG 2005 conference. Quebec City. Competing interests: None declared |
|||
|
|
|||
|
Edmund J Lamb, Consultant clinical scientist Kent and Canterbury Hospital, CT1 3NG, Charlie Tomson, Edmund Lamb, Kathryn Griffith, Donal O'Donoghue, John Feehally
Send response to journal:
|
Giles and Fitzmaurice1 rehearse a series of familiar2, 3 but flawed arguments in their article, designed to persuade BMJ readers that the implementation of estimated GFR (eGFR) reporting has introduced a screening programme by the back door, will pressurise specialist services and cause unnecessary anxiety and harm amongst patients (e.g. difficulty in getting life insurance, inappropriate ACEI prescription). Whilst implementation has caused a marked increase in referrals of patients newly recognised to have chronic kidney disease (CKD), this is likely to be a temporary increase caused by referral of prevalent patients. The UK Guidelines for Identification, Management, and Referral of Adults with Chronic Kidney Disease4 are constructed to ensure that only patients who will receive added value from a specialist opinion are referred; most patients with CKD can safely and more efficiently be managed in primary care.5 We are unaware of evidence that detection of CKD causes difficulties in getting life insurance; most patients diagnosed as having CKD as a result of eGFR reporting are older, few of whom will take out new life insurance. ACEI prescription is only indicated in the presence of hypertension (on QOF markers) and this is entirely in keeping with current NICE guidance. Implementation of eGFR reporting has improved the clinical interpretation of a test (serum creatinine) that was already being done,6 with a significant aim being to reduce the morbidity and mortality associated with late referral to nephrology services of patients with significant kidney disease.7 The only reason serum creatinine concentration is measured is to assess kidney function. The indications for testing serum creatinine concentration have not been changed by the implementation of eGFR reporting; most of the existing recommendations for measurement of serum creatinine in the UK Guidelines are well accepted and endorsed by other national and international bodies. It is known that the MDRD equation does not provide a perfect estimate of GFR. Improved assay precision, specificity and standardisation will help. Currently, harmonisation through the United Kingdom National Quality Assessment Scheme has helped laboratories achieve between- laboratory agreement (coefficient of variation) of approximately 6% at GFRs around 60 mL/min/1.73 m2. It is flawed logic to use the limitations of the MDRD equation as an argument against the implementation of eGFR reporting. As the authors acknowledge, it is useful for staging disease in patients with stage 3-5 CKD, as required by the QOF. We believe that eGFR reporting, for all its limitations, is a significant step forward in the identification of patients with CKD. CRV Tomson,1 EJ Lamb,2 K Griffith,3 D O'Donoghue,4 J Feehally5 1Renal Medicine, Southmead Hospital, Bristol, BS10 5NB, UK, E-mail:
Charlie.tomson@nbt.nhs.uk
References 1. Giles PD, Fitzmaurice DA. Formula estimation of glomerular filtration rate: have we gone wrong? BMJ 2007;334:1198-200. 2. Clase CM, Garg AX, Kiberd BA. Classifying kidney problems: can we avoid framing risks as diseases? BMJ 2004;329:912-5. 3. Clase CM. Glomerular filtration rate: screening cannot be recommended on the basis of current knowledge. BMJ 2006;333:1030-1. 4. Joint Specialty Committee on Renal Medicine of the Royal College of Physicians and the Renal Association, and the Royal College of General Practitioners. Chronic Kidney Disease in Adults: UK Guidelines for Identification, Management and Referral. London: Royal College of Physicians, 2006. 5. Mitra PK, Tasker PR, Ell MS. Chronic kidney disease. BMJ 2007;334:1273. 6. Wyatt C, Konduri V, Eng J, Rohatgi R. Reporting of estimated GFR in the primary care clinic. Am J Kidney Dis 2007;49:634-41. 7. Roderick P, Jones C, Tomson C, Mason J. Late referral for dialysis: improving the management of chronic renal disease. QJM 2002;95:363-70. Competing interests: None declared |
|||
|
|
|||
|
James A Dickinson, Professor of Family Medicine University of Calgary Medical Clinic, #1707, 1632 14th Ave NW Calgary AB T2N 1M7 Canada
Send response to journal:
|
Giles and Fitzmaurice have done a great service by asking hard questions about the formula estimation of glomerular filtration rate (eGFR). This concept has been introduced in Canada also, and in a practice with many older patients, it is problematic to receive frequent recommendations to refer patients to nephrologists. I am happy to do so for those who are likely to benefit from preparation for dialysis, and the advice of nephrologists, but many seem to gain little benefit, except for repetition of the advice that we are giving anyway. This applies particularly to those where renal disease is only a small component of their overall problem list. How risky is it to not take the advice gratuitously sent by the laboratory with every test? As Giles and Fitzmaurice note, creatinine is not a good predictor of glomerular filtration rate above 60 ml per minute. However, though they say that the points cluster more closely below that level, their Fig 1 appears to show that this only appears because a linear scale is used. This crowds all the points together in the small space between the axes, close to the origin. If a ratio or logarithmic scale were used, I suggest that it could demonstrate greater dispersion (and worse correlation) at the lower end of the range. If so, we need a much better test that truly correlates with function at the low end where we must make the decisions. Since eGFR is effectively a screening test, its protagonists should demonstrate that for each cutoff level, the recommended action will achieve their end of benefiting the patients substantially enough to outweigh the harms caused by unnecessary referral and subsequent actions. We await this data with great interest. Competing interests: None declared |
|||
|
|
|||
|
Liam G Glynn, GP Principal, Ballyvaughan, Co Clare and Lecturer in Primary Care Department of General Practice, National University of Ireland, Galway, Ireland, Andrew W Murphy, Professor of General Practice, National University of Ireland, Galway, Ireland
Send response to journal:
|
The controversy surrounding the applications of estimated glomerular filtration rate (GFR) that Giles and Fitzmaurice[1] refer to, may exist more for clinical biochemists and nephrologists than for many others managing patients with chronic kidney disease (CKD). As a clinician and researcher working in primary care in a non-incentivised environment close to the UK, one is acutely aware of the position of CKD as the poor relation in the family of common chronic diseases which includes hypertension, cardiovascular disease and diabetes. This is despite the growing body of evidence demonstrating significantly increased cardiovascular risk associated with chronic kidney disease (defined according to MDRD[2] estimated GFR!) in the general population[3, 4], in patients with established cardiovascular disease [5-7] as well as in patients with diabetes[8]. There are many reasons for this but chief among them has been the inaccuracy of the screening tool available to us historically, namely, serum creatinine measurement, which does not accurately reflect GFR[9, 10], the best indicator of renal function in health and disease[11]. Clearly, there are limitations to the estimation equations we use which make the application of estimated GFR as a universal screening tool inadvisable. However, poor physician awareness of chronic kidney disease and its association with excess cardiovascular morbidity and mortality remains a significant problem[12]. The very low rate of recording of chronic kidney disease in patients found to have chronic kidney disease indicates scope for improving detection and early intervention[13]. Estimated GFR, with its acknowledged limitations, should improve the ability of the primary care physician to identify, prioritise and manage these patients. We must not underestimate the positive effect on physician awareness, and ultimately patient identification and treatment, associated with the arrival of estimated GFR to the community setting. There is a lot of catching up to do in the management of CKD and estimated GFR is one of the tools which, when applied in a selective manner, will help that process within the primary care setting, where the vast majority of patients with CKD will be managed. 1. Giles, P.D. and D.A. Fitzmaurice, Formula estimation of glomerular filtration rate: have we gone wrong? 10.1136/bmj.39226.400694.80. BMJ, 2007. 334(7605): p. 1198-1200. 2. Levey, A.S., et al., A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med, 1999. 130(6): p. 461-70. 3. Muntner, P., et al., Renal Insufficiency and Subsequent Death Resulting from Cardiovascular Disease in the United States. J Am Soc Nephrol, 2002. 13(3): p. 745-753. 4. Go, A.S., et al., Chronic Kidney Disease and the Risks of Death, Cardiovascular Events, and Hospitalization. N Engl J Med, 2004. 351(13): p. 1296-1305. 5. Tonelli, M., et al., Proteinuria, impaired kidney function, and adverse outcomes in people with coronary disease: analysis of a previously conducted randomised trial. BMJ, 2006. 332(7555): p. 1426. 6. Anavekar, N.S., et al., Relation between Renal Dysfunction and Cardiovascular Outcomes after Myocardial Infarction. N Engl J Med, 2004. 351(13): p. 1285-1295. 7. Glynn, L.G., et al., Chronic kidney disease and mortality and morbidity among patients with established cardiovascular disease: a West of Ireland community-based cohort study 10.1093/ndt/gfm222. Nephrol. Dial. Transplant., 2007: p. gfm222. 8. Henry, R.M.A., et al., Mild renal insufficiency is associated with increased left ventricular mass in men, but not in women: An arterial stiffness-related phenomenon: The Hoorn Study. Kidney Int, 2005. 68(2): p. 673-679. 9. Levey, A.S., Measurement of renal function in chronic renal disease. Kidney Int, 1990. 38(1): p. 167-84. 10. Swedko, P.J., et al., Serum Creatinine Is an Inadequate Screening Test for Renal Failure in Elderly Patients. Arch Intern Med, 2003. 163(3): p. 356-360. 11. Smith, H., Diseases of the kidney and urinary tract, in The Kidney: Structure and Function in health and disease. 1951, Oxford University Press: New York. p. 836-87. 12. Perazella, M.A. and S. Khan, Increased Mortality in Chronic Kidney Disease: A Call to Action.[Review]. American Journal of the Medical Sciences., 2006. 331(3): p. 150-153. 13. de Lusignan, S., et al., Identifying patients with chronic kidney disease from general practice computer records. Fam. Pract., 2005. 22(3): p. 234-241. Competing interests: None declared |
|||
|
|
|||
|
Shelagh E O'Riordan, Consultant Geriatrician East Kent Hospitals NHS Trust Canterbury, CT1 3NG, Claire Beeson
Send response to journal:
|
It surprises me that Giles and Fitzmaurice title their talk- Formula estimation of glomerular filtration rate: have we gone wrong?1 In the body of the paper they highlight many good reasons for using eGFR , particularly agreeing it can detect changes in renal function over time and say in stages 3-5 CKD, where it is most important, it agrees closely with reference methods. Where they do however disagree with the use of eGFR is in screening patients. No one has, however, advocated using eGFR as a screening test. The UK CKD Guidelines2 specifically state that it is not a screening test and advocate using it only in at risk populations. What this truly means is that when we suspect kidney disease may be present; we have access to a better, if not perfect test, to tell us the answer. If you don’t wish to know the renal function of a patient, don’t request a serum creatinine and you won’t get an eGFR. Where eGFR has really improved clinical management is in older patients. This group were the most likely to have serum creatinine in the normal range or just above it when they had significant impairment of kidney function. Below are some recent cases in whom knowing the eGFR has helped our management. 1. A woman in her 90’s was admitted with suspected pulmonary embolism. Her serum creatinine was 155 µmol/l but her eGFR was 29 ml/min/1.73 m2. It was immediately clear that we needed to reduce the dose of clexane to avoid toxicity. We also decided to do a ventilation perfusion scan instead of a CT pulmonary angiogram because of the risk of contrast nephropathy in chronic kidney disease. 2. An woman in her 80’s was admitted with pneumonia and found to have a haemoglobin of 9.8 g/dl. The MCV was towards the lower end of normal and one of the juniors suggested we investigate using a colonoscopy and a gastroscopy. However, knowing her eGFR was on 35 ml/min/m2 we decided to see if she had renal anaemia before exposing her to invasive investigations. 3. A woman in her 80’s with a new vertebral crush fracture was admitted in pain. Knowing on day 1 that she had significant CKD prompted us to investigate for renal bone disease before starting her on bisphosphonate treatment and standard calcium and vitamin D supplementation. Before the introduction of eGFR we only had access to a less sensitive test informing us of kidney function. We now have automatic access to a better if not perfect test- surely this is an improvement? References 1. Paul D Giles and David A Fitzmaurice Formula estimation of glomerular filtration rate: have we gone wrong? BMJ 2007; 334: 1198-1200 2. Chronic kidney disease in adults: UK guidelines for identification, management and referral. London: Royal College of Physicians 2006. Competing interests: SO'R was the British Geriatric Society representative on the UK CKD Guidelines and on the CKD section of the NSF for Renal Services |
|||
|
|
|||
|
Charles J. Diskin, CLINICAL ASSOCIATE PROFESSOR 36801
Send response to journal:
|
I would like to applaud Drs. Giles and Fitzmaurice for their analysis on the recent trend towards the use of creatinine-based formulae as a substitute for glomerular filtration rate (GFR).1 As they have correctly noted, while there are more accurate markers of GFR, creatinine was chosen for convenience.2 Since the relationship of creatinine clearance to glomerular filtration depends upon a tenuous balance of counterbalancing factors, even excellent clinicians can be led astray by failure to consider how metabolic states, disease, medications, diet, and time of day affect that relationship.3 While the averaging of variations over time helped avoid some of the problems with diurnal variations in the past, the present day reliance upon equations based upon a solitary serum value is likely to amplify those problems. Creatinine clearance is not now and never was GFR, 4 and no amount of regression analysis will make it so due to its fundamentally flawed relationship to GFR.5 As Drs. Giles and Fitzmaurice astutely note, the cost of the continued use of creatinine- based formulae as a substitute for a true measured GFR will include the inconvenience, anxiety and mismanagement of a certain percentage of patients. I personally know of no one who has made any estimate of that cost, but like Drs. Giles and Fitzmaurice, I expect the numbers to be substantial. Perhaps a large-scale multinational analysis should be conducted in order to show us the true cost of our convenience. 1. Giles PD, Fitzmaurice DA. Formula estimation of glomerular filtration rate: have we gone wrong? BMJ. 2007 Jun 9;334(7605):1198-200. 2. Diskin CJ. Creatinine and glomerular filtration rate: evolution of an accommodation. Ann Clin Biochem 2007;44:16-9. 3. Diskin CJ. More on case 7-2006. N Engl J Med. 2006 Oct 5;355(14):1502-3 4. Carrie BJ, Golbetz HV, Michaels AS and Myers BD. Creatinine: an inadequate filtration marker in glomerular disease. Am J Med 1980;69:177- 182. 5. Diskin CJ. Creatinine and GFR: an imperfect marriage of convenience. Nephrol Dial Transplant. 2006 Nov;21(11):3338-9. Competing interests: None declared |
|||