Blog entry by Meguid El Nahas
Susantitaphong and colleagues in Boston have published this month in the June issue of the AJKD a meta-analysis of 16 cohort studies and 1 RCT (n= 1,081,116) relating to impact of the level of GFR at initiation of RRT (HD or PD) on all-cause mortality.
The authors concluded that a higher ESTIMATED GFR (eGFR) was associated with HIGHER mortality.
On the other hand, in a subgroup analysis, higher CALCULATED GFR was associated with LOWER mortality.
Confused you may be...!!!!!
Well, this observation confirms impressions that GFR estimated (eGFR) by the MDRD equation is inaccurate in ESRD as the confounder of serum creatinine leads to this equation, as well as that of Cockcroft-Gault, being more a reflection of the clearance of endogenous creatinine rather than true GFR; clearance of endogenous creatinine is affected by GFR as well as muscle mass and tubular secretion of creatinine. Whilst the authors stress repeatedly that this observation was independent of nutritional parameters; this may reflect visceral nutritional markers such as serum albumin/protein and not necessarily markers of poor somatic nutritional status such as creatinine itself.
On the other hand, calculated GFR derived from a 24h urine collection and the derived mean of urea and creatinine clearances seems to be a better marker of true GFR that is less influenced by muscle mass. Higher calculated GFR, unlike estimated GFR, was in a subgroup analysis (486 patients) associated with LOWER mortality.
This is an important observation putting in context a number of publications on the topic of GFR and timing of RRT. It also warns nephrologists NOT to put too much emphasis on GFR estimation/calculation when deciding when to start RRT but instead use a more holistic approach to the patient and his clinical status as well as co-morbidities.
Lets move away as a profession from these artificial and meaningless equations and go back to the old fashion serum creatinine, serum urea, clinical examination as well as good clinical judgement.
After all we, as Nephrologists, are Physicians and not Mathematicians....