IN THE NEWS: New NKF recommendations for screening

Based on some new findings in a recent AJKD study, the National Kidney Foundation now recommends annual screening with a simple urine albumin test that checks for protein in the urine, in specific high-risk groups. 

These include adults with:

• Diabetes
• High blood pressure
• Age 60 years or older
• Family history of kidney failure requiring dialysis or transplantation

The entire manuscript is in on AJKD website.  

Diabetes, HTN and family history of kidney disease makes sense and has always been part of the screening for kidney disease in past. Now, we are adding age >60 years or older based on the significantly high finding of >50% of americans likely to develop lifetime risk of CKD. The risk was higher in women. How do the authors come up with this risk? They used the current CKD prevalence rate of Americans with CKD and used the Markov chain model ( random process with memorylessness) to come up with statistics to detect the future risk. It is a model use to predict events in the future as a process moves in time. So in this case, current CKD prevalence using this model can predict what it could be based on risk factors and prediction of events what the CKD status would be years from now? The model is like the "drunkard walk" and how with each step, the position may change by +1 or −1 with equal probability. A use in medicine of this model can be found here and here. 

Urine dipstick screening

Early awareness of chronic kidney disease and prevention strategies have led to the mass screening at many places of urine dipstick for albuminuria in the general population. While in diabetics and hypertensives, this might be a very prudent approach, does mass screening really change outcomes? A recent commentary in CJASN discusses the screening concerns from a primary care's perspective. Some key points the authors make are:
1. Based on what a good screening test should be, urine dipstick doesn't cut it. The seven criteria for a good screening test are: target disease is prevalent, morbid; the screening test has to be low risk, cost effective and accurate and acceptable to patients; and we should have the ability to change the outcomes.
2. Many false positives are generate requiring not required follow ups ( especially in a time when we might not have that many nephrologists in near future)
3. Interestingly, are we really changing outcomes. Are we preventing them from getting to ESRD?

This might be a tide changer for many but some points they make are critical to look at. The threshold for screening needs to be higher ( DM, HTN, FMHx, edema - anything that raises the possibility of proteinuria) but as a routine screen for unselected people with no identified risk, the test (which is semi-quantitative and variable at best) probably cause more problems than it solves.  Generating inappropriate work up and referrals from a positive dipstick might be interesting to look at closely. Indication is most important in most cases.