Fascinating study which may have more stories to tell
John Pickering, University of Otago Christchurch
27 June 2013
Congratulations to the authors on this fascinating study. We have been waiting a long time for someone to attempt true GFR measurements on AKI patients, not an easy task.
The authors note that creatinine clearance may be overestimated because of secretion of creatinine. Studies involving the use of cimetidine seem to confirm this, at least for patients in steady state. However, I wonder if the data presented here may contradict that, or at least fail to confirm it? I took the liberty to run a statistical test comparing the GFR (51Cr-EDTA) with the CrCl data from table 2 (Wilcoxon matched pairs sign rank test) and found that there was no difference. Also the bias (Figure 1) was small and favoured a greater GFR51.
The authors considered if loss of muscle mass resulting in loss of creatinine production may in part explain the poor performance of estimating equations. We¿ve noted recently that post cardiac attack there is a potential very rapid loss of creatinine production, probably much more rapid than that of muscle loss (Pickering et al Crit Care 2013:R7).
Finally, our own work on 4 hour creatinine clearance related to AKI suggested some potential clinical utility (Pickering et al Crit Care 2012:R107). I wonder if the authors would like to make some comment concerning to what extent the clinical utility of creatinine clearance may be limited by the duration of urine collection?
Critical Care
Fascinating study which may have more stories to tell
27 June 2013
Congratulations to the authors on this fascinating study. We have been waiting a long time for someone to attempt true GFR measurements on AKI patients, not an easy task.
The authors note that creatinine clearance may be overestimated because of secretion of creatinine. Studies involving the use of cimetidine seem to confirm this, at least for patients in steady state. However, I wonder if the data presented here may contradict that, or at least fail to confirm it? I took the liberty to run a statistical test comparing the GFR (51Cr-EDTA) with the CrCl data from table 2 (Wilcoxon matched pairs sign rank test) and found that there was no difference. Also the bias (Figure 1) was small and favoured a greater GFR51.
The authors considered if loss of muscle mass resulting in loss of creatinine production may in part explain the poor performance of estimating equations. We¿ve noted recently that post cardiac attack there is a potential very rapid loss of creatinine production, probably much more rapid than that of muscle loss (Pickering et al Crit Care 2013:R7).
Finally, our own work on 4 hour creatinine clearance related to AKI suggested some potential clinical utility (Pickering et al Crit Care 2012:R107). I wonder if the authors would like to make some comment concerning to what extent the clinical utility of creatinine clearance may be limited by the duration of urine collection?
Regards
John Pickering
Competing interests
No competing interests