Twenty years ago, acute kidney injury (AKI) wasn't a significant problem in hospitals, and not just because the term hadn't been invented yet. There were also far fewer patients showing sudden decreases in their renal function.
A study of hospitalized Medicare beneficiaries, published in the Journal of the American Society of Nephrology in 2006, found that incidence of AKI increased at about 11% per year between 1992 and 2001. Another analysis, published in the same journal the same year, found that discharges with AKI rose from 61 per 100,000 people in 1988 to 288 per 100,000 in 2002.
As with many conditions, the aging of the U.S. population is a contributor to this increase, but advances in medical care may also be responsible. “Twenty years ago, we wouldn't be doing cardiopulmonary bypass on a 95-year-old person. We wouldn't be putting patients through three stents and wouldn't be doing the fourth angiogram,” said Charuhas Thakar, MD, an associate professor of nephrology at the University of Cincinnati. “Sometimes we are victims of our own success.”
Successful efforts to raise clinicians' awareness of the perils of AKI may also explain part of the increase. “A rise in creatinine as little as 27 µmol/L or 0.3 mg/dL constitutes acute kidney injury, and … only now is there a true appreciation of the clinical significance of these changes,” said Ron Wald, MD, an assistant professor of nephrology at the University of Toronto. “Acute kidney injury never really had a formal definition or criteria for diagnosis until probably the last ten years.”
The medical community's understanding of AKI recently got even more formal, with last year's publication of a clinical practice guideline for acute kidney injury by a workgroup of Kidney Disease: Improving Global Outcomes (KDIGO). The guideline is targeted especially at nonspecialists who manage patients with AKI—in other words, hospitalists.
“It's not feasible for every patient with mild acute kidney injury to be seen by a nephrologist,” said Dr. Wald. “This is a condition that needs to be appreciated by everyone who manages hospitalized patients, not just nephrologists.”
Even better than appreciating AKI would be preventing it. Hospitalists play a key role in assessing inpatients' potential for kidney injury and taking action on risks and early signs, said nephrologists, who offered some advice on best practices.
The first step is knowing which patients are most likely to develop AKI, although that's not always as simple as it sounds, experts said.
“The KDIGO guideline waxes lyrically—and a large part of that text I wrote, so I can be critical—for many pages on the importance of risk assessment, but at the end is not able to offer very specific details on how one does that,” said John A. Kellum, MD, a professor of critical care medicine at the University of Pittsburgh. “There's no metric. There's no app that I can plug into my iPhone.”
AKI risk typically results from patients' preexisting susceptibilities and conditions or treatments to which they are exposed. The guideline lists factors in each category. Susceptibilities include dehydration or volume depletion, advanced age, female gender, black race, chronic kidney disease, diabetes, cancer, anemia and other chronic diseases.
The exposures highlighted by the guideline are sepsis, critical illness, circulatory shock, burns, trauma, cardiac surgery, major noncardiac surgery, nephrotoxic drugs, radiocontrast agents, and poisonous plants and animals.
Obviously, a high proportion of hospitalized patients are going to fulfill at least one of those criteria, which makes risk assessment tricky. It's the combination of susceptibilities and exposures that matters. “For someone with a lot of susceptibilities, it doesn't take much in the way of exposure to cause AKI, so they're going to be high risk at even very limited amounts of added exposure, versus someone who has very few underlying susceptibilities,” Dr. Kellum said.
Most of the susceptibilities and exposures are not modifiable, so physicians should focus their attention on those that are, experts recommended. “Maintaining euvolemic status is obviously helpful,” said Dr. Wald. Toward that goal, the guidelines recommend using isotonic crystalloids rather than colloids, but Joel Topf, MD, an assistant professor of nephrology at Wayne State University School of Medicine in Detroit, goes a step farther.
Based on some recent research, including a study published in the Oct. 17, 2012, Journal of the American Medical Association finding that a chloride-restrictive fluid strategy reduced AKI and dialysis, his fluid of choice is lactated Ringer's. “This is a very simple thing to do, just choosing a different isotonic crystalloid,” Dr. Topf said. The KDIGO guideline notes that buffered salt solutions are less likely to cause acid-base disturbances but concludes that the effect on patient outcomes is uncertain.
The association between contrast and kidney injury, however, is entirely certain. “The best way to avoid contrast nephropathy is to do your imaging with a modality that doesn't require contrast, so MRI or ultrasound when possible,” said Dr. Topf.
“Does the patient really need this CAT scan? The answer might be yes,” said Dr. Thakar. “But if I as a nephrologist can make a hospitalist think twice before they order a potentially nephrotoxic exposure, that is a huge improvement in our delivery of care.”
Drugs pose the other main nephrotoxin dilemma. “ACE [angiotensin-converting enzyme] inhibitors and ARBs [angiotensin-receptor blockers] have tremendous benefits to the heart and for chronic kidney disease, but they definitely make the kidney much more sensitive to hemodynamic instability and volume changes,” said Dr. Topf.
Dr. Thakar offered an example. “If somebody comes in with pneumonia, has low blood pressure, has diabetes, and they take lisinopril and a water pill for blood pressure, maybe [those drugs] should be held,” he said. “As a hospitalist, [consider] altering the doses of medications that are excreted by the kidney and sometimes not giving certain medications that the patient is prescribed at home.”
In addition to ACE inhibitors and ARBs, Dr. Wald also cited nonsteroidal anti-inflammatory drugs, aminoglycosides and phosphate-containing enemas as potential exposure risks.
“Trying to reduce the number of drugs that people are on that can affect their kidney function would be very helpful and is often not being done with the kind of rigor that we would like to see,” said Dr. Kellum.
Patients at risk of AKI also need to be monitored with greater rigor, experts said, specifically their creatinine measurements and urine output.
The KDIGO definition of AKI starts at a increase in creatinine of 0.3 mg/dL or 50% from baseline. If you don't have a baseline measurement, that's not very helpful, however. “If you've got an otherwise healthy individual who shows up in the ER and their creatinine is 2 mg/dL and we don't have any outpatient records, many people may look at that and think, ‘Oh, they've got some [chronic kidney disease],’” said Dr. Kellum. However, this might also be a case of AKI. A careful history and physical exam and possibly a renal ultrasound to look at kidney size may be helpful to determine whether the problem is acute or chronic, he advised.
On the other end of the spectrum, the development of AKI in patients with an initially low creatinine might slip under the radar because their levels haven't left the normal range. “If creatinine goes from 0.6 mg/dL to 1.2 mg/dL, it often won't be flagged in the medical record, and yet it corresponds to a 50% reduction in renal function” Dr. Kellum said.
The alternative diagnostic method—urine output (defined as AKI at less than 0.5 mL/kg per hour for six to 12 hours)—is another example of medical progress causing new problems. “We've got a push to try to keep Foley catheters out of patients, and I think that's very appropriate to reduce rates of urinary tract infections,” said Dr. Kellum. “But a high-risk patient needs to have their urine monitored for the time that they're at risk.”
Alerts in the electronic health record could potentially assist with this monitoring. “We have one here [at the University of Pittsburgh] that has been running for several months, and various other centers have similar kinds of warnings that are based on changes in creatinine or urine output,” Dr. Kellum said.
Once a patient's AKI has been discovered and efforts have been made to identify any potential causes, hospitalists face the dilemma of how to treat. The avoidance of nephrotoxic exposures becomes even more important, naturally. “If there's low blood pressure, correct it. If they need fluids, if they need antibiotics, so on and so forth,” said Dr. Thakar.
Unfortunately, there's no actual treatment for these early-stage AKI patients. “No single clinical trial has conclusively shown any therapy that is effective in established acute kidney injury,” said Dr. Thakar. The KDIGO treatment recommendations are mostly negative—don't use diuretics, low-dose dopamine, fenoldopam, atrial natriuretic peptide or recombinant human (rh)IGF-1 to treat or prevent AKI.
The evidence base does offer good and bad news about outcomes, however. The good news is that although no successful treatments have been discovered, mortality rates from AKI have been improving. “We as researchers scratch our heads,” said Dr. Thakar. “Maybe overall intensive care has improved. Maybe our hospitalists have recognized not to give certain agents after early recognition of injury. If you take steps to reduce the likelihood that somebody gets to established AKI, then you have just made that patient less likely to die in the hospital.”
The bad news is that the research also shows that patients who do survive to discharge face increased risks, both of kidney problems and death, in the years afterward. “We used to think about this like an ACL [anterior cruciate ligament] injury in an athlete: When they come back, they're going to be fine,” said Dr. Topf. “That doesn't seem to be true. These patients are at future risk for end-stage renal disease and more advanced chronic kidney disease. You don't completely heal from it, though your creatinine may go back down to normal.”
Hospitalists can help these patients by making sure that they receive continued kidney monitoring. “We don't do a good job of following up the patients who are discharged from hospitals with an acute kidney injury,” said Dr. Thakar. “Less than 30% will get their creatinine checked within an optimal time after discharge.”
Acute kidney injury patients will often not be seen by a nephrologist in the hospital unless they require dialysis, so hospitalists should find some way to ensure a future creatinine check, whether it's an outpatient nephrology referral, a request to the primary care physician, or a visit to a hospitalist postdischarge clinic.
“They should be checking their kidney function, and if there is a problem, send them to a nephrologist,” said Dr. Thakar. “This is a very important group of patients that we need to target during transitions of care between in- and outpatient.”