Bad medicine or an opportunity to integrate

60 mils/min/1.73m2

Maximum estimated Glomerular filtration rate that is associated with all cause and cardiovascular mortality in the general population

The Week in Numbers
British Medical Journal, 9 October 2010

The accompanying lead editorial by Vlado Perkovic and Alan Cass at the George Institute for Global Health starts with the statement – “prevention of stroke, through the identification and management of risk factors is a public health priority.” They are commenting on the linked systematic review by Lee and colleagues that assesses the association between kidney function defined by estimated Glomerular filtration rate (eGFR) and the subsequent risk of stroke. The overall risk of stroke was 40% higher in people with an eGFR below 60 mils/min/1.73m2. This risk was graded, with a 28% higher risk for people with an eGFR of 40-60 mils/min/1.73m2 but 77% higher for those with an eGFR below 40 mils/min/1.73m2. The editorial comments that the persistent association after adjustment for known cardiovascular risk factors increases the likelihood that dissociation is real.

The findings of the Chronic Renal Insufficiency Cohort recently published in the American Heart Journal (2010: 145 pg 690-694) showing that 18% of individuals with mild to moderate CKD (mean eGFR 43 mils/min/1.73m2) also had atrial fibrillation may explain some of the increased risks. This rate of Atrial fibrillation, if you will forgive the pun, is 2 to 3 times greater than would expect in a non-CKD population .

In the same issue of the British Medical Journal as Lee’s paper on CKD and the risk of stroke the Reykjavik prospective study from Iceland reports that even the earliest stages of CKD are associated with excessive risk of subsequent coronary heart disease. Advanced stages of CKD are also associated with non-vascular mortality, particularly deaths not attributed to cancer. This is a large, long term study. Men and women who were resident in Reykjavik were recruited between 1967 and 1991 and were followed up for a median of 24 years. Importantly, the participants in the study provided urine for proteinuria estimates as well as blood for eGFR. The analysis shows that proteinuria, even in the absence of a lowered eGFR, is associated with an increased vascular risk of between 55 – 72%.

These studies were quite different in design and scope: the first was a retrospective meta- analysis, the other two were prospective studies. Lee’s study involved 284,672 participants from a variety of countries. The Chronic Renal Insufficiency Cohort was from North American and the Reykjavik study was a prospective cohort study in 16,958 individuals were followed for nearly a quarter of a century. Together they add important information on the risk stratification of patients with CKD. Perkovic and Cass ask the question “what are the implications for clinical practice?”. The answer: at the very least, evidence suggests the presence of CKD (either reduced eGFR or albuminuria but especially both) should act as a “red flag” that triggers cardiovascular risk assessment and implementation of an appropriate preventative strategy.

It is “bad medicine” just to take a single low eGFR measurement and label someone as having kidney disease. The methodological challenges of measuring serum creatinine and the bias within the various eGFR formulae are well know. A low eGFR result should trigger thinking – why was the kidney function checked?; could this be acute kidney injury?; what drugs might be implicated?; what’s the level of albuminuria?; and are there earlier serum creatinine measures to provide the confirmatory evidence of chronic kidney disease?

Mainstreaming kidney disease management with cardiovascular risk prevention strategies not only holds out the prospect of a preventative dividend in terms of end stage renal failure, it also provides the opportunity to integrate risk assessment and detection of early disease, clinical management, patient and public engagement across the whole panoply of vascular disease.