New kidney disease strategy aimed at earlier intervention
By Deborah Gesensway
By the time the 80-year-old patient saw Edgar V. Lerma, FACP, a Chicago nephrologist, she had a blood creatinine of 2.0 mg/dL. The level hadn’t seemed worrisome enough to the woman’s primary care physician to warrant a referral earlier. If that patient had been young or male, that may have been true. In this case, however, the woman’s kidneys were only working at 25% of their capacity, and she was close to needing dialysis.
This is exactly the kind of patient that Dr. Lerma—and nephrologists across the country along with the federal government’s National Kidney Disease Education Program (NKDEP) and the National Kidney Foundation (NKF)—think a fairly new kidney disease care strategy based on the NKF's 2002 guideline is starting to help (see sidebar).
At the heart of the strategy is a transformation in the language to discuss kidney disease, using words and numbers that make more sense both to patients and physicians—from “renal” to “kidney,” from “insufficiency” to “failure.”
The objective is to help everyone understand they are dealing with a progressive chronic disease that advances through predictable, definable stages. So instead of waiting until they have to start dialysis, the goal is to slow, halt or even prevent or partially reverse the disease through medical treatments and lifestyle changes such as adjusting blood pressure medications or modifying use of non-steroidal painkillers.
The NFK's strategy also focuses on changing how patients are screened for the chronic disease and categorized according to the stage of its progression—a challenge for internists who play a crucial role in making that part of the strategy pay off.
Creatinine levels not enough
The NKF's practice guidelines call for calculating a patient’s estimated glomerular filtration rate (GFR) using the MDRD equation (preferred to the Cockcroft-Gault equation) rather than relying on blood creatinine levels to diagnose kidney damage.
The estimated GFR comes from a formula that combines the creatinine level with the patient’s age, race and gender. Chronic kidney disease (CKD) is defined as the presence of kidney damage or level of GFR less than 60 ml/min per 1.73 m2—with or without kidney damage—that is present for three or more months.
“What we’ve learned is that if you look at the serum creatinine level alone, you may not get the best idea of kidney function,” explained Mahboob Rahman, ACP Member, associate professor of medicine at Case Western Reserve University, and author of ACP’s PIER module on chronic kidney disease.
“You can have what looks like a normal creatinine, but there can be serious kidney problems. We need to get away from using serum creatinine only and more toward using estimated GFR.”
In the four years since the guideline was released, kidney disease experts say, anecdotally, that they are getting earlier referrals than before. However, they emphasize, the medical profession still has a long way to go.
Getting up to speed
Experts say two factors are slowing progress: First, although more and more clinical laboratories are reporting GFR automatically, many still are not, primarily because of equipment snafus. And second, even when they do receive the estimates, many primary care practitioners just don’t know how to interpret and act on the figures.
A study published in the August 2006 issue of the American Journal of Kidney Diseases found that many experienced general internists and family physicians could not recognize chronic kidney disease when given a questionnaire describing a primary care physician caring for a patient with progressing CKD. While 97% of the nephrologists in the study recognized CKD, only 78% of the general internists and 59% of the family physicians did. Interestingly, the primary care physicians who had been in practice for more than 10 years were the least likely to recognize CKD.
Kidney disease experts hope encouraging widespread adoption of GFR will help physicians. “The problem with kidney disease is that it is asymptomatic,” explained Lesley Stevens, MD, assistant professor of medicine in the division of nephrology at Tufts-New England Medical Center, Boston, and the lead author of a June 2006, review article in The New England Journal of Medicine about estimated GFR. “We as nephrologists never see a patient until he’s referred to us, so we need [general internists and other primary care physicians], the doctors who are seeing asymptomatic patients, to be able to detect it.”
The difficulty has been that using serum creatinine tests alone are misleading for everyone except nephrologists who use them all the time and “can detect the nuances of the creatinine level really well,” Dr. Stevens said. “We can look at a person and a creatinine level and say, 'I don’t quite believe it, the kidney function is probably lower than suggested by the creatinine level.' But that’s not a good test to use if we want to have non-nephrologists be able to identify kidney disease early.”
Overcoming GFR flaws
Some doctors have been held back by the fact that the GFR estimating equations are complicated and flawed.
The NKF and NKDEP began to deal with the first issue a few years ago by working to convince clinical laboratories to automate the calculation and regularly report the information back to physicians whenever they order a serum creatinine test for their patients, Dr. Stevens said.
Now that primary care physicians are beginning to receive that information from their labs, the next critical step—educating primary care physicians about how to interpret that information and adjust to the flaws in the estimates—is underway in earnest.
The bottom line, Dr. Stevens said, is that physicians need to interpret the results of the test in the context of the patient. “As clinicians, we do this all the time with other things—for example, ST depression has different meanings depending on who the patient is—and it’s the same with GFR,” she said.
The estimated GFR equation works well in patients with kidney disease, but is less accurate in people without kidney disease, she explained. Specifically, a reading of 70 may mean that someone without kidney disease really has a GFR of 80, “but it doesn’t matter if they are 70 or 80 if there is no other evidence of kidney damage. I will treat them the same way,” she said. “In particular, although these people will not be classified as having kidney disease, it may be worthwhile to be careful with the use of medications that can harm the kidney, such as anti-inflammatories."
For this reason, NKDEP recommends that estimated GFR levels above 60 only be reported as “greater than 60.” For those lower than that, the specific number can help, particularly when deciding on appropriate management of the patient’s kidney disease or tracking progression, Dr. Stevens said.
Primary care physicians need to understand that the number is an estimate; the exact figure rarely matters, said Derrick L. Latos, MACP, a nephrologist in Wheeling, W. Va., ACP’s representative to the NKDEP Advisory Panel and a member of the NKF's Kidney Disease Outcomes Quality Initiative advisory group. Whether an estimated GFR is 34 or 38 is not important clinically, he noted. "The values may not be exact, but we know that the patient has substantially diminished GFR.”
In addition, he said, he has had to explain to primary care physicians that use of estimated GFR in patients already undergoing dialysis is inappropriate.
“We are using the estimating equation to identify individuals who may have a decreased level of kidney function and in whom that fact was not recognized previously because their serum creatinine might well have been in the normal range because it didn’t consider the age of the patient and muscle mass,” Dr. Latos said.
Having reliable estimating equations for GFR is most important in screening for evidence of kidney disease, Dr. Latos continued. "As the equations become further refined, by using standards for creatinine assays and by obtaining additional information for other types of patients—elderly, obese, renal transplants, and those with no other evidence of kidney disease, for example, we can expect to be able to use them for more precise determination of the level of kidney function."
Dr. Lerma’s 80-year-old patient is the perfect example of the problem. A doctor looking at a creatinine clearance rate of 2.0 mg/dL on paper might not be alarmed. If the patient had been a 45-year-old man, for instance, it may have only meant the man’s kidney disease is at a moderate level (or Stage 3), with his kidneys working at about a 40% level. For the older woman, the same creatinine level meant she was in Stage 4 of her kidney disease, and it was time to discuss renal replacement therapies and prepare her for dialysis.
Early intervention works
The consensus is that early physician intervention can lead to positive outcomes. For example, when Dr. Lerma—clinical assistant professor of medicine at the University of Illinois at Chicago College of Medicine—first saw one patient in her 50's, she had about 30% kidney function. Her blood pressure and diabetes were not controlled. She was taking NSAIDs for her arthritis, and she had some protein in her urine.
"We changed her medications, and four to six months later, her kidney function went from 30% to 58%," he said. "It didn’t get any better than that, but it hasn’t gotten worse. We are gaining her an indefinite number of years.”
Early referrals also allow nephrologists to spend extra time educating patients about dialysis options, preparing them emotionally and psychologically for the grueling treatment regimen and getting them physically ready for hemodialysis by creating arteriovenous fistulas to provide the best possible access.
Around 40% of all new hemodialysis patients have fistulas; the Centers for Medicare and Medicaid Services' (CMS) goal is 66% of all patients. "We can’t do anything about that as nephrologists if we get the patient and he needs dialysis tomorrow,” Dr. Latos explained. “The idea of early referral is important. We can sit down with patients and explain how they can save a vein so that at an appropriate time we can create a fistula."
In March 2005, CMS announced its “Fistula First Breakthrough Initiative” to improve fistula use rates to levels seen in Europe and Asia, which average 70% and 80%, respectively. Starting in January 2006, the Centers for Disease Control and Prevention also revised its ICD-9-CM diagnostic codes to allow physicians to report the stage of kidney disease they are treating.
Efforts to get screening and treatment up to speed are spurred by new thinking about CKD:
First, it is now considered a cardiovascular disease risk factor equal to diabetes. (The American Heart Association’s Scientific Statement on “Kidney Disease as a Risk Factor for Development of Cardiovascular Disease” is online.)
And experts know that the prevalence of kidney disease is on the upswing. NIH’s National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) estimated that 8 million adults (or 4.5% of the U.S. adult population) have a moderately or severely reduced GFR. About 11% of adults are in the early stages of the disease.
Of the nearly half million U.S. residents undergoing treatment in 2003 for end-stage renal disease, about 60% had resulted primarily from two diseases—diabetes and hypertension, according to NIDDK—diseases that could be controlled, averting associated kidney failure.
The fact that the disease tends to be asymptomatic until the very end puts even more pressure on the medical profession to develop a means for detecting silent signs of the disease as early as possible.
Dr. Lerma recalled just such a patient, a 38-year-old homemaker who had come to him with an extremely high blood creatinine level and needing immediate dialysis. Although her primary care physician had been tracking her creatinine levels for some time and was getting more and more concerned, the woman hadn’t come to him earlier because she misunderstood her lack of symptoms.
“She, like a lot of patients, had the misconception that if she urinates 24 hours a day without problems her kidneys must be good. She was waiting for symptoms before she came to me,” he said.
“After we started her on dialysis, she told me she had never felt so good in her life. She didn’t even know she hadn’t been feeling good.”
The National Kidney Foundation's 2002 practice guidelines laid out a new five-stage system for defining the disease and called for an enhanced effort to identify people at increased risk and then to screen them for undetected—and usually asymptomatic—kidney damage at earlier stages. (The guideline is online and was published in the July 15, 2003, issue of Annals of Internal Medicine.)
According to the National Kidney Disease Education Program, physicians should:
Test patients—both by serum creatinine (used to estimate GFR) and by a urinalysis to check for protein or albumin—if patients have a family history of kidney failure or if they have hypertension or diabetes.
Think about screening people older than 60, or if they are African-American, Hispanic or Native American, or if they have any other reason for concern.
Test diabetics once a year; all others every one to three years based on the physician’s discretion.
The two most important numbers to look for when analyzing an estimated glomerular filtration rate (GFR) is 60 mL/min/1.73 m², which puts a patient in Stage 3 or higher chronic kidney disease (CKD) when complications appear and should be treated, and 30 mL/min/1.73 m², when Stage 4 begins and physicians should begin preparing patients for the possibility of eventual kidney replacement therapy. Patients with Stage 4 disease should be referred to a nephrologist.
Once primary care physicians identify an asymptomatic patient with diminished kidney function, there is much they can do. According to the guidelines and nephrologists, treatment of early stage chronic kidney disease should include the following:
Control blood pressure, using whatever agent works, to below 130/80 mmHg. Hypertension can cause kidney disease, and kidney disease can worsen hypertension, “so you can be in a downward spiral,” explained William L. Henrich, MACP, dean of the school of medicine and vice president of medical affairs at the University of Texas Health Science Center at San Antonio.
Stress tight blood sugar control for diabetics, who account for nearly half of all new cases of kidney failure, with the goal of achieving a hemoglobin A1C level of less than 7%.
Prescribe ACE inhibitors or angiotensin receptor blockers even if blood pressure is normal if a patient has proteinuria. The goal should be to lower the protein-creatinine ratio to <500 to 1000 mg/g.
Control lipids. “You have to do it aggressively, using statins, weight control, exercise, and tobacco smoking cessation,” said Derrick Latos, MACP, a nephrologist in Wheeling, W.Va. Tobacco smoking, he said, correlates with more rapid decline in kidney function.
Warn patients that common medications can be a problem if they have decreased kidney function. Many—notably NSAIDs—are toxic to the kidneys. Patients also need to limit the number of contrast studies using dyes, such as a CT scans or angiography, they get; if they need them, they should have renal prophylaxis measures with them to reduce the chance of damaging their kidneys further.
Treat anemia with erythropoietin and iron if necessary and watch out for bone disease.
Internist Archives Quick Links
Have questions about the new ABIM MOC Program?
One Click to Confidence - Free to members
ACP Smart Medicine is a new, online clinical decision support tool specifically for internal medicine. Get rapid point-of-care access to evidence-based clinical recommendations and guidelines. Plus, users can easily earn CME credit. Learn more