COMMENTARY

Proteinuria: A Cheaper and Better Cholesterol?

Lynda A. Szczech, MD; Donal J. O'Donoghue, MB, ChB

Disclosures

November 02, 2012

This feature requires the newest version of Flash. You can download it here.

Lynda A. Szczech, MD: Hello. My name is Lynda Szczech. I'm a nephrologist in Durham, North Carolina, and President of the National Kidney Foundation. This is another Skype chat with Donal O'Donoghue.

In this particular chat we will talk about preventive screening for kidney disease, and this is motivated largely by a recent report in the Annals of Internal Medicine[1] with the recommendations from the US Preventive Service Task Force (USPSTF) on whether or not to screen for kidney disease.

Donal J. O'Donoghue, MB, ChB: Hi. I am Donal O'Donoghue, a nephrologist in Manchester, United Kingdom. I am the National Clinical Director for Kidney Care [in the United Kingdom]. It is a pleasure to be on this Skype chat. This is a very interesting area with a range of opinions from many different perspectives.

Dr. Szczech: [In this new report,] the USPSTF states that we have no evidence to justify the routine screening for kidney disease in asymptomatic adults. There are many subtleties in that report, but if you read it superficially that [statement about the lack of evidence in favor of screening] is the brick that falls out. Tell me, Dr. O'Donoghue What is your experience in the National Health Service of the UK of how and when to screen the general population in primary care offices? I think you have done something that is really interesting that has helped.

Dr. O'Donoghue: We have a national screening committee, akin to what you have in the States, that looks at screening with a specific public health point of view, with the traditional World Health Organization screening criteria about treatments and benefits. This committee focuses on a totally asymptomatic population that is not in contact with clinical care at all. The concept of screening or the precise definition of screening is quite restrictive when compared with what the man on the street would understand by screening and how the working nephrologist or primary care physician might interpret screening.

In the UK, we screen for chronic kidney disease (CKD) through the National Screening programs that screen for breast cancer and some other cancers, congenital hypothyroidism, etc. However, as part of the chronic disease management of people with diabetes or hypertension, we incentivize primary care to monitor kidney function as part of management of the chronic disease. There is more than a subtle difference, but a patient with a known [chronic] disease would be assessed for CKD. In reality, what does that look like? Few people with diabetes or hypertension have not had a serum creatinine level checked, and therefore their estimated glomerular filtration rate (eGFR) is captured by the system. If it drops below normal, that patient is identified and put on a CKD regimen for kidney disease management.

What about the rest of the population? In the nondiabetic, nonhypertensive population, the guidelines would not suggest that kidney disease needs to be looked for. Of course, many people have their blood drawn [for other reasons], millions per year. And if a creatinine is measured in an adult, and an eGFR is reported to primary care and it is low, then those people are identified and should be assessed for kidney disease and placed on a chronic disease register and managed in that way. Without screening, we have been able to identify 4.3% of the adult population as having CKD [severity stage] 3 to 5. We know from the epidemiologic studies and other research that roughly 6.5% to 7% of the population have CKD stage 3 to 5, and we have picked up two thirds of those by opportunistic screening.

What we miss, however, is a focus on proteinuria. We know from an epidemiologic study that proteinuria is occurring earlier in our population. The question is, should we be screening for proteinuria in people without other known vascular disease at a particular age? We do not have the evidence to answer this question. I think we need to invest in the studies to find answers --what is the population, what is the impact of interventions in that truly asymptomatic population, and what are the benefits of those interventions over time?

Dr. Szczech: You have raised a couple points that are worth going over to make sure that everyone listening really understands the subtleties of the message. Proteinuria is a marker of kidney disease and a marker of vascular disease because it is a predictor of mortality. What is the barrier to "market" proteinuria as a cheaper but better risk-stratifier than serum cholesterol, as some studies suggest?

Dr. O'Donoghue: I have a slide that says that proteinuria is the new cholesterol, proteinuria is "the new black." It is remarkable that even in diabetes, where the evidence base is so strong, even in diabetes, looking for and intervening in proteinuria is the last measure in the basket of measures that need to be put in place. I struggle with why we have not been able to get that message across, or maybe the message is there, but the logistics of collecting urine in the office [are too prohibitive]. We know from the Groningen [The Netherlands] studies that people can collect the urine at home and post it in the mail with no problems. The assessment and management of proteinuria in people with diabetes is the least measured [parameter] in that population, but it is the most improving measure. From 2003-04 to most recently, we have gone from roughly 15% to 20% of people with known diabetes having an albumin and creatinine checked once per year to now close to 65%.[2] That is still a 35% missed opportunity, but a big increase that I think reflects the broader CKD story, to give us the impetus to say that we can prevent some of the advanced kidney disease. The evidence base for people with diabetes is incontrovertible, and we have healthy economic information from Tonelli's group in Canada[3] and the Hunt study[4] in Norway and elsewhere to show that, without a doubt, people with diabetes should be screened. I think that is very clear. Then the question is, if they are screened, what do I need to put in place to accomplish that economically, and how do we achieve the goal that we set ourselves?

Dr. Szczech: When I read the USPSTF recommendations, and when you get right down to exactly what they say, I do not necessarily disagree with them, because the report includes a lot of caveats. Those guidelines do not include people with diabetes or people with hypertension, because we have great guidelines that include those patients, so this is for everyone else. However, they clearly state that proteinuria is not the best predictor of kidney disease, period. Why is that? Because of the competing risk that you and I are talking about, that people die before they have the opportunity to progress to dialysis, and that is obviously well known in people with type 2 diabetes. My biggest concern is that the casual reader may miss that and it may stifle the recognition that urine is really a cheaper and better cholesterol.

Dr. O'Donoghue: I think that's right. The message you take from the headline or the superficial read would be, we don't need to bother about this; [screening for CKD is not supported by] an evidence base. But in actual fact, and this is not a subtlety, the report clearly states that there are populations that need to have their proteinuria checked and managed. It comes down to the meaning of the word "screening," and screening means different things to different medical populations. So the message is [potentially] quite dangerous.

Even if the evidence base in [patients with] nondiabetic proteinuria and stage I/II kidney disease is [minimal], we do know that there are a range of particular populations with very high renal morbidity and mortality and very high healthcare costs. Perhaps we could look at those populations as the place to start. To my mind, that is a very, very important public health target. Moreover, if you add proteinuria to the vast array of risk factors we have now and you cannot show significance -- I turn that on its head and say, let's start with proteinuria and see how many people with proteinuria we pick up. That is, perhaps, an easier way of starting to look at vascular risk than starting with a Framingham equation or one of the newer equations that has to acquire different information. To me, that would be a fairer comparison. I am certainly not suggesting we throw out cholesterol or blood pressure, but I think if you want to actually compare them head-to-head, then the way this has been done in the theoretical discussions to date has not been on a fair playing field. We are back to the issue of not having the evidence base to answer the question. In this case, the recommendation message should be, let's get this evidence in specific populations and in the general population.

Dr. Szczech: As our last parting shot, I want to throw a trick question your way. I know what my answer to this is because I have been thinking about these guidelines for a while and I read them closely, because superficially they are troubling to a nephrologist. What is the definition of "asymptomatic"? Or, what are the symptoms of kidney disease? If we talk about not having evidence to screen asymptomatic individuals, then we have to be able to define the symptoms of kidney disease. What are the symptoms?

Dr. O'Donoghue: Kidney disease is a silent killer, and people do not have discomfort from their kidneys -- we know that -- but they do have a latent period during which we would intervene and prevent the devastating consequences. In the 21st century, we need to think about health and health protection, vascular risk, and have an understanding of vascular risk, which is about exercise; it's about what we eat; it's about smoking -- that public health part of the agenda. In this discussion, it is about proteinuria and blood pressure. Is hypertension a disease? Is hypercholesterolemia a disease? The word "disease" in our population generates a whole series of negative feelings. I am not wedded to the word "disease," but I am wedded to early detection and avoiding the missed opportunities that lead us to spend our lives working in the dialysis unit. All of those patients have experienced some missed opportunities that could have delayed the time to end-stage renal failure. We have not had a chance to explore particular populations; clearly, people with diabetes and first-degree relatives of people with advanced kidney disease are a key group. Those populations are in a high-risk situation and are very grateful when we can offer something and identify things early.

Dr. Szczech: I think kidney disease is like depression. It is a great pretender. Some people will have periorbital swelling and lower-extremity edema, hypertension, hematuria. But most will have fatigue, malaise, and decreased appetite. When I sit down and think about all the carve-outs in those guidelines, I am left to conclude that there is no evidence to screen the young 20-something person that gave me my latte this morning and who is otherwise without chronic condition and is asymptomatic. But the rest of us who always have a few aches and pains, a little fatigue, and sometimes do not feel like ourselves -- for these people, I do not think we should be so quick as to move it off the diagnostic palette.

Dr. O'Donoghue: I agree completely. One of the advantages of knowing early that people have kidney disease is clearly the prevention of vascular events and renal events. We normally think of that [in the context of] progression of chronic kidney disease. But what are the common renal events? We must Skype shout about it soon-- is this acute kidney injury, or acute kidney injury in people with mild asymptomatic kidney disease but who are on a range of medicines because they have the aches and pains you have just talked about, and they are taking an over-the-counter nonsteroidal in addition to an ACE [angiotensin-converting enzyme inhibitor] or an ARB [angiotensin II receptor blocker]? That is a very, very important population that needs to know their kidney function, their kidney risks, and needs to be able to share that information so that healthcare providers can make judgments about continuing the medicines or not. We do not have the evidence base, but what do we do in that situation to reduce acute kidney injury? These are interesting questions.

Dr. Szczech: Thank you, Donal. It has been an absolute pleasure talking with you. And thank you to everyone who listened to this Skype chat. I hope you enjoy them, that they are informative, and that you keep tuning in. This is Lynda Szczech.

Dr. O'Donoghue: And this is Donal O'Donoghue signing off. If you have ideas for challenges for future Skype chats, let us know.

processing....