Can Rivaroxaban Lead to Anticoagulation-Related Nephropathy?

Anticoagulation-related nephropathy (ARN) is an understudied renal complication of anticoagulant therapy that is characterized by acute kidney injury (AKI) defined as an increase in baseline serum creatinine ≥ 0.3 mg/dL, without any alternate etiology, in the setting of a supra-therapeutic International Normalized Ratio (INR) greater than 3.0 [1-7]. ARN has mainly been studied in relation to warfarin, it usually occurs in the first two months of starting anticoagulant therapy and retrospective studies have estimated the prevalence to range from 16% to 37% [4-6]. The strongest risk factor for ARN is pre-existing chronic kidney disease (CKD); other risk factors include older age, diabetes mellitus and hypertension [1,4]. Patients with ARN have an accelerated progression of CKD and a retrospective study of more than 15,000 patients on warfarin showed a 65% increase in mortality in patients with ARN [4,6].

The non-vitamin K oral anticoagulants (NOACs) are replacing warfarin in the clinical setting for long term anticoagulation because of the advantages of faster, reliable onset of action, and unlike warfarin, they do not require dose-response monitoring [8-10]. Rivaroxaban is the most commonly prescribed NOAC at Einstein Medical Center Philadelphia. A few animal models and case reports have shown that these NOACs (particularly dabigatran and apixaban) can lead to deterioration of kidney function, but there is limited epidemiologic data on ARN in NOACs and the incidence of ARN in NOACs is unknown [11-14]. A proposed guideline to diagnose ARN recommends that serum creatinine and urinalysis be checked monthly in the first few months of anticoagulation [1]. Clinical trials that compared warfarin to NOACs specifically studied stroke risk and bleeding outcomes and did not document the incidence of ARN because they did not make these regular creatinine and urinalysis measurements in the first few months of anticoagulation [15-20]. Our study intends to make these necessary measurements in a selected group of high risk patients started on rivaroxaban and hence determine the incidence of ARN in this group.

This study will be the first epidemiological study using a prospective design to measure the incidence of ARN in a population receiving NOACs. Given the high morbidity and mortality associated with ARN, the findings will enhance patient safety, enabling patients and health providers to make informed decisions with their patients regarding the choice of anticoagulants. It will also establish baseline data that will serve as foundation for future studies.

The specific aims of this study will be to -

1. Determine the incidence of ARN in high risk patients on rivaroxaban. Hypothesis - the incidence of ARN in rivaroxaban is less than the previously reported incidence of ARN in warfarin.
2. Describe demographic and urinalysis-associated predictors of ARN in patients on rivaroxaban.

SIGNIFICANCE

Anticoagulant-related nephropathy (ARN) is a relatively new and underdiagnosed complication of anticoagulant therapy with the potential to accelerate chronic kidney disease and lead to increased mortality [4,6]. Brodsky, et al, performed a retrospective study of more than 15,000 subjects and showed that the 5-year Kaplan-Meier survival rate was significantly lower in the ARN cohort compared to the patients without ARN with a 65% increased risk of mortality [4]. This dramatic increase in mortality that is attributable to an episode of ARN in a patient taking an anticoagulant is the reason why this study is very significant.

The clinical trials that compared the NOACs to warfarin did not specifically study ARN and did not repeat creatinine and urinalysis testing in the first two to three months of therapy to establish possible ARN diagnosis [15-20]. In most of these studies, only baseline creatinine and creatinine levels 6 to 12 months later were analyzed, hence it is possible that ARN was under-diagnosed as most cases of ARN are detected in the first eight weeks of therapy. Our study aims to accurately capture ARN cases by making these measurements monthly in the first three months of starting the anticoagulant.

This study is pivotal because it will for the first time, define the incidence of ARN in patients on rivaroxaban therapy. If the investigators find that ARN incidence in patients started on rivaroxaban is less than that seen with warfarin, this can help support evidence that this NOAC is safer for the kidneys compared to warfarin and this data can enable us make evidence based decisions regarding the continued use of the medication to help our patients. If on the other hand the investigators find that the medication leads to ARN, then the investigators will be contributing important post-marketing surveillance data that can also help us make safety decisions about the use of the medication in the future.

INNOVATION

Previous studies on ARN have utilized retrospective study designs which have many epidemiological pitfalls including the cause versus effect bias and the inability to ascertain relative risk [1]. Our study is unique because the investigators intend to utilize a prospective epidemiological design to study this renal complication, a first in the study of ARN. This is important because ARN is a diagnosis of exclusion and hence a prospective study is able to more accurately define cases of the condition compared to the assumptions that are made in retrospective studies where further work up cannot be done in retrospect.

This will be the first epidemiological study focused on ARN in a NOAC. Even though NOACs are new, they are becoming ubiquitous and a mainstay in clinical practice and not just an alternative to warfarin. For example, the recent CHEST guidelines recommend them as first line therapy in the treatment of venous thromboembolism [8]. Hence it will be important to clearly delineate their effects on the kidneys.

OBJECTIVES

In summary, this is a proposal for a prospective study to investigate the incidence of anti-coagulant nephropathy, a dreaded renal complication of anticoagulation. Our literature search has shown that there are no prospective studies of this condition in the new oral anticoagulants and clinical trials did not perform the necessary testing in the appropriately timed interval to diagnose ARN. The investigators chose to study rivaroxaban, the most commonly prescribed NOAC in our academic medical center and they anticipate that the result of this study will contribute safety data that will lead to better care of our patients and great cost savings by avoiding the morbidity and mortality associated with renal disease.