Blog entry by Meguid El Nahas
A recent publications by Tonelli and his colleagues in Edmonton, Canada in KI December 2011, highlights the risks associated with a high estimated GFR (eGFR). Using data from a population data base of 1,526,437 patients, they conclude that a high eGFR is associated with an increased risk of all cause mortality. They estimated that those with an eGFR of 60-74.9ml/min had the lowest risk compared to higher hazard ratio of 3.7 and 1.8 in those with eGFR>105 and 90-104.9ml/min respectively. Such risk is enhanced in those with dipstick positive proteinuria.
In their discussion they offer a number of possible explanations.
I have reservations about a number of points related to that publication:
1. It relied on a ESTIMATED GFR based on the MDRD equation known to be inaccurate at a measured GFR >60ml/min; in fact many biochemistry laboratories estimate the inaccuracies and imprecision of MDRD formula at eGFR>90 so significant that they do not give a quantitative value above that level of calculation >90ml/min. So to differentiate with a single Cr estimation and an inaccurate formula measured GFRs of 74.9 to 104.9 to >105 is effectively a misleading waste of time...!? The authors somewhat validated their findings with the CKD EPI equation although I an uncertain about that equation adequacy in those with mGFRs between 75-105ml/min?
2. ESTIMATED GFR is not MEASURED GFR; whils this seeems obvious they have become increasingly interchangeable in recent nephrology literature..? Even clinical trials interchange eGFR with mGFR....like the recently published BEAM stduy where the lowering of sCR by the trial agent was automatically assumed to reflect an improvement in kidney function rather than tubular damage and increased urinary leakage of creatinine...?
As we all know, eGFR is a reflection of serum creatinine levels that can be affected by a number of factors including wasting, sarcopenia and overal body muscle mass. The eGFR formula corrects and adjusts for body surface area but not body muscle mass. Patients with low eGFR may have lower serum creatinine and lower muscle mass due to a number of factors independently of their true GFR.
In fact, it is well known that low serum and urinary creatinine measurements as well as low muscle mass are associated with increased mortality risk. Whilst the authors acknowledge that wasting associated with underlying illness/comorbidity may explain their findings, this is not the thrust of their discussion.
A better title for this publication would have been: "Low serum creatinine may be associated with increased risk of adverse outcomes..."
It is high time that we go back to basics and remember the ESTIMATED GFR is what it is...an estimate based on serum creatinine levels and NOT a automatic reflection of TRUE GFR; above a GFR of 60-75ml/min it is not even an ESTIMATE but more likely to be a GUESTIMATE....!!!!
Let's not put too much emphasis on guestimates and pay more attention to underlying co-morbidities that explain wasting, low serum creatinine and high risk of mortality in some members of our communities.