BOSTON, MA — Among patients on long-term warfarin therapy seen in an anticoagulation clinic, those with CAD or moderate renal impairment when therapy was started were most likely to develop severe renal impairment—defined as creatinine clearance (CrCl) <30 mL/min—often within months, in a new study.
This pattern would be similar in patients receiving novel oral anticoagulants (NOACs), although further study is needed to confirm this with NOACs, Dr Christina L Fanola (Boston University School of Medicine, MA) and colleagues write in their study published online November 4, 2016 in the American Heart Journal.
"Moving forward it would be really important to . . . have our study replicated in a real-world population cohort study in patients just on NOACs," Fanola told heartwire from Medscape. In the meantime, "more monitoring is better, at least until we have more real-world cohort data on NOACs."
The message for cardiologists is "here's a subset of [high-risk] patients; if you're already bringing them in for a cardiology appointment, perhaps you should make sure to check their creatinine" to see whether the NOAC dose needs to be adjusted or if the therapy should be changed, she said.
The results emphasize the importance of monitoring renal function in patients on anticoagulants and adjusting the dosing accordingly, Dr Ziad Hijazi (Uppsala Clinical Research Center, Sweden), lead author of a subanalysis of ARISTOTLE that looked at renal function with apixaban, agreed, in a comment to heartwire .
"The important issue," he added, "is to be aware of the different dose-reduction criteria for each NOAC and prescribe accordingly, especially during periods of declining renal function."
The study also suggests that in high-risk patients, yearly CrCl determinations, as recommended in a US guideline and on NOAC labels, may not spot renal-function deteriorations, and instead, more frequent monitoring depending on initial CrCl, as recommended in a European guideline, is a better approach, the researchers say.
Kidney Function With Long-term Anticoagulation
NOACs, unlike warfarin, are at least partly eliminated by the kidneys, so guidelines suggest that the dose may need to be reduced when patients have impaired kidney function.
However, patients with severe CKD were excluded from trials such as RE-LY, ROCKET-AF, ARISTOTLE, and ENGAGE, Fanola noted, and little is known about how stable renal function is among patients in clinical practice who are on long-term anticoagulant therapy.
To investigate this, the researchers identified 787 patients who were seen in their anticoagulation clinic from January 2007 through December 2011 and were on long-term warfarin for atrial fibrillation or venous thromboembolism (VTE).
They excluded 34 patients with severe renal impairment, defined as stage 4 to stage 5 chronic kidney disease (CKD)—that is, CrCl <30 mL/min.
This left a cohort of 324 patients with normal kidney function (CrCl >90 mL/min), 237 patients with mild renal impairment (stage 2 CKD, CrCl 60–89 mL/min), and 174 patients with moderate renal impairment (stage 3 CKD, CrCl 30–59 mL/min).
Slightly more than half of the patients were male. Half were 75 years old or older; a quarter were 65 to 74, and a quarter were younger than 65. The patients were racially diverse: black (45%), white (40%), Hispanic (10%), and other (5%).
They were followed for a median of 2.4 years (maximum 4.1years) and had a median of 10 CrCl determinations.
About one in 10 patients (12%) developed severe renal dysfunction; of these patients, 25% developed it within 5.3 months.
A total of 37% of patients with moderate renal impairment but only 6% and 4% of patients with mild renal impairment or normal renal function, respectively, developed severe renal impairment.
Only initial CAD and moderate renal impairment were independent predictors of this outcome.
Odds of Developing Severe Renal Impairment
|Baseline variable||OR (95% CI)||P|
|CrCl 30–59 mL/min (vs >90 mL/min)||14.5 (6.7–31.3)||<0.0001|
|CAD (vs no CAD)||2.2 (1.3–3.8)||0.004|
"As we expect our results in warfarin patients to be similar [to patients] on NOACs, our results suggest that patients [on NOACs] with moderate CKD and those with CAD may be at highest risk and require more frequent monitoring," Fanola and colleagues write.
It is important to detect any CrCl below 30 mL/min, since this "marks a period of potential hazard, given that the elimination half-lives and plasma drug concentrations of all of the NOACs would increase significantly," they warn.
"Reasonable Rule of Thumb"
The 2015 European Heart Rhythm Association (EHRA) guidelines recommend dividing a patient's baseline CrCl by 10 to determine the monthly interval for monitoring CrCl. For example, a patient with a CrCl of 40 mL/min would need to be monitored every 4 months.
In contrast, the 2014 ACC/AHA/HRS guideline states: "Evaluate renal function before initiation of direct thrombin or factor Xa inhibitors and reevaluate when clinically indicated and at least annually," which is rather "vague," Fanola pointed out.
"We would argue that the ACC/AHA/HRS adopt similar guidelines [to those of the EHRA] to base the frequency of renal-function measurement on baseline creatinine clearance," she said.
This is "is a reasonable rule of thumb," Hijazi agreed, since "these patients also have higher risk of both bleeding and thromboembolic complications."
In ARISTOTLE and ROCKET-AF, about 20% of patients seemed to be at risk of having a clinically significant decline in renal function during follow-up, he noted.
All the data so far point in the same direction, showing that "elderly individuals, those with previously identified renal dysfunction, and those with multiple cardiovascular comorbidities are at higher risk of developing renal impairment over time," according to Hijazi. "It possible that this risk may be higher when on a vitamin-K antagonist, as vitamin-K antagonists have been associated with vascular and glomerular calcification," he added.
Monitoring kidney function is important in patients on anticoagulants, since patients "need to be on the right drug [and the right dose] given the kidney function they have," Fanola summarized.
Fanola has no relevant financial relationships. Disclosures for the coauthors are listed in the article.
Heartwire from Medscape © 2016 Medscape, LLC
Cite this: More Frequent Creatinine Checks for Some Patients on Oral Anticoagulation? - Medscape - Dec 09, 2016.