Re: Chronic kidney disease controversy: how expanding definitions are unnecessarily labelling many people as diseased
Thank you to the authors for opening up this confusing issue for wider discussion.I share the concerns of the authors to a huge extent and having read the rapid responses, felt that I should be voicing my views.
To be very clear about the rest of this post, I am sure nobody wants to stop monitoring eGFR. Everybody should continue to adjust drug doses and watch out for AKI in those with low eGFR. That is not the issue here. The focus here is on the word "disease".
Over-diagnosis and medicalisation of every measurable abnormality is increasingly becoming fashionable, if not the norm. Arguing that the label of a "disease" is required to trigger concern and interest in physicians to address the associated cardiovascular issue is not reasonable. That would merely be a reflection of our current sweeping responses to anything and everything that we cannot police or influence on a day to day basis. Specialists will have to balance the vested interests that they have in trying to help identify all eligible patients with high inclusivity against the need to be pragmatic for the sake of the generalists who will be dealing with much of the population to whom the controversial diagnostic criteria will apply to, while taking care not to dumb down criteria to cater to those less engaged. A balance is also required to ensure that specificity is not sacrificed for the sake of sensitivity, lest the false positive diagnosis of a potentially progressive chronic disease causes great emotional distress with knock on effects on various aspects ranging from loss of a productive life to treatment-seeking behaviour.
The major criticism of the paper has been its lack of recognition of the CVD predictive value of CKD. If we are interested in a low eGFR due to its impact on future CVD rather than ESRD (accepting that many do not progress to ESRD, but all have higher risk for CVD), are we then labelling a risk factor as a disease? Is there an IFG and IGT analogy given that there is higher risk for CVD being predicted by all levels of eGFR in a progressive fashion? Are we getting too wound up about a "surrogate marker' of a generalised vascular problem? Should there be a push to label microalbuminuria as a disease since it predicts CVD?
eGFR, like any other value, will have to be treated on a case by case basis. It will have to be interpreted in the different contexts of risk prediction, prognostication, trend monitoring, drug dose adjustments, as well as institution of CVD preventative measures. This is particularly relevant when discussing the issue of whether to label a stable marginal decline of eGFR below 60 ml/minute in a very elderly person without a demonstrable abnormality in the urine or in renal anatomy, as a disease. Many of us would like to believe that a drop in eGFR with age is a manifestation of the natural ageing process. People who offer counter-arguments that low eGFR predicts increased CVD risk and hence is unlikely to be a natural ageing process but instead is an indicator of vascular damage seem to conveniently forget the fact that vascular damage is also part of ageing. The current extreme desire in the medical community to prevent and treat vascular damage in all instances has made it politically incorrect to even suggest that it would be reasonable to avoid cardiovascular preventative measures in the very elderly. If our real concern is the possibility of denying cardiovascular preventative treatment to huge numbers of frail elderly, then we should be looking harder at the NNT(numbers needed to treat) concept to guide us through these difficult discussions. While we know that low HDL predicts higher cardiovascular mortality, we have not shown that treating it improves CVD mortality. Have we shown that treating CKD improves mortality? Given that CKD implies lack of improvement of renal function further-allowing for recognised fluctuations within a stage- any impact of treatment on CVD improvement would be due to appropriate statin, anti platelet, anti-hypertenisve and anti-hyperglycaemic usage. Do we really need a "disease" label to make us use all those? Aren't there enough other markers that can trigger their use appropriately?
We probably will never know the NNT for 80 and 90 year olds to get CVD benefits with current treatment, although all of us suspect triple figure numbers. But one thing that many of us who deal with elderly patients on a daily basis have a gut feeling for is the NNH (Numbers Needed to Harm). NNH figures are likely to be much lower (greater harm) in the real world than in research settings, as elderly patients with multiple co-morbidities are exposed to polypharmacy with monitoring tempered by a multitude of social reasons.
While the desire to highlight the importance of low eGFR from a cardiovascular view point is very justifiable, we should resist the temptation to forget primum non nocere and avoid accepting heavy collateral damage at an individual level in the search for improved public health and life prolongation. Can we talk about CKC or CKA? Chronic Kidney Change or Chronic Kidney Abnormality? Could we introduce the requirement of a second factor of microalbuminuria for diagnosis in the very elderly in the CKD IIIa stage? Should we be revisiting the age related eGFR trends set up years ago when lower life expectancies were the norm?
Rapid Response:
Re: Chronic kidney disease controversy: how expanding definitions are unnecessarily labelling many people as diseased
Thank you to the authors for opening up this confusing issue for wider discussion.I share the concerns of the authors to a huge extent and having read the rapid responses, felt that I should be voicing my views.
To be very clear about the rest of this post, I am sure nobody wants to stop monitoring eGFR. Everybody should continue to adjust drug doses and watch out for AKI in those with low eGFR. That is not the issue here. The focus here is on the word "disease".
Over-diagnosis and medicalisation of every measurable abnormality is increasingly becoming fashionable, if not the norm. Arguing that the label of a "disease" is required to trigger concern and interest in physicians to address the associated cardiovascular issue is not reasonable. That would merely be a reflection of our current sweeping responses to anything and everything that we cannot police or influence on a day to day basis. Specialists will have to balance the vested interests that they have in trying to help identify all eligible patients with high inclusivity against the need to be pragmatic for the sake of the generalists who will be dealing with much of the population to whom the controversial diagnostic criteria will apply to, while taking care not to dumb down criteria to cater to those less engaged. A balance is also required to ensure that specificity is not sacrificed for the sake of sensitivity, lest the false positive diagnosis of a potentially progressive chronic disease causes great emotional distress with knock on effects on various aspects ranging from loss of a productive life to treatment-seeking behaviour.
The major criticism of the paper has been its lack of recognition of the CVD predictive value of CKD. If we are interested in a low eGFR due to its impact on future CVD rather than ESRD (accepting that many do not progress to ESRD, but all have higher risk for CVD), are we then labelling a risk factor as a disease? Is there an IFG and IGT analogy given that there is higher risk for CVD being predicted by all levels of eGFR in a progressive fashion? Are we getting too wound up about a "surrogate marker' of a generalised vascular problem? Should there be a push to label microalbuminuria as a disease since it predicts CVD?
eGFR, like any other value, will have to be treated on a case by case basis. It will have to be interpreted in the different contexts of risk prediction, prognostication, trend monitoring, drug dose adjustments, as well as institution of CVD preventative measures. This is particularly relevant when discussing the issue of whether to label a stable marginal decline of eGFR below 60 ml/minute in a very elderly person without a demonstrable abnormality in the urine or in renal anatomy, as a disease. Many of us would like to believe that a drop in eGFR with age is a manifestation of the natural ageing process. People who offer counter-arguments that low eGFR predicts increased CVD risk and hence is unlikely to be a natural ageing process but instead is an indicator of vascular damage seem to conveniently forget the fact that vascular damage is also part of ageing. The current extreme desire in the medical community to prevent and treat vascular damage in all instances has made it politically incorrect to even suggest that it would be reasonable to avoid cardiovascular preventative measures in the very elderly. If our real concern is the possibility of denying cardiovascular preventative treatment to huge numbers of frail elderly, then we should be looking harder at the NNT(numbers needed to treat) concept to guide us through these difficult discussions. While we know that low HDL predicts higher cardiovascular mortality, we have not shown that treating it improves CVD mortality. Have we shown that treating CKD improves mortality? Given that CKD implies lack of improvement of renal function further-allowing for recognised fluctuations within a stage- any impact of treatment on CVD improvement would be due to appropriate statin, anti platelet, anti-hypertenisve and anti-hyperglycaemic usage. Do we really need a "disease" label to make us use all those? Aren't there enough other markers that can trigger their use appropriately?
We probably will never know the NNT for 80 and 90 year olds to get CVD benefits with current treatment, although all of us suspect triple figure numbers. But one thing that many of us who deal with elderly patients on a daily basis have a gut feeling for is the NNH (Numbers Needed to Harm). NNH figures are likely to be much lower (greater harm) in the real world than in research settings, as elderly patients with multiple co-morbidities are exposed to polypharmacy with monitoring tempered by a multitude of social reasons.
While the desire to highlight the importance of low eGFR from a cardiovascular view point is very justifiable, we should resist the temptation to forget primum non nocere and avoid accepting heavy collateral damage at an individual level in the search for improved public health and life prolongation. Can we talk about CKC or CKA? Chronic Kidney Change or Chronic Kidney Abnormality? Could we introduce the requirement of a second factor of microalbuminuria for diagnosis in the very elderly in the CKD IIIa stage? Should we be revisiting the age related eGFR trends set up years ago when lower life expectancies were the norm?
Competing interests: No competing interests