Article Text
Abstract
Objective Non-vitamin K oral anticoagulants (NOACs) require dose adjustment for renal function. We sought to investigate change in renal function over time in patients with atrial fibrillation (AF) and whether those on NOACs have appropriate dose adjustments according to its decline.
Methods We included patients with AF enrolled in the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation II registry treated with oral anticoagulation. Worsening renal function (WRF) was defined as a decrease of >20% in creatinine clearance (CrCl) from baseline. The US Food and Drug Administration (FDA)-approved package inserts were used to define the reduction criteria of NOACs dosing.
Results Among 6682 patients with AF from 220 sites (median age (25th, 75th): 72.0 years (65.0, 79.0); 57.1% male; median CrCl at baseline: 80.1 mL/min (57.4, 108.5)), 1543 patients (23.1%) experienced WRF with mean decline in CrCl during 2 year follow-up of −6.63 mL/min for NOACs and −6.16 mL/min for warfarin. Among 4120 patients on NOACs, 154 (3.7%) patients had a CrCl decline sufficient to warrant FDA-recommended dose reductions. Of these, NOACs dosing was appropriately reduced in only 31 (20.1%) patients. Compared with patients with appropriately reduced NOACs, those without were more likely to experience bleeding complications (major bleeding: 1.7% vs 0%; bleeding hospitalisation: 2.6% vs 0%) at 1 year.
Conclusions In the US practice, about one-fourth of patients with AF had >20% decline in CrCl over time during 2 year follow-up. As a result, about 3.7% of those treated with NOACs met guideline criteria for dose reduction, but of these, only 20.1% actually had a reduction.
- atrial fibrillation
Statistics from Altmetric.com
Introduction
The American Heart Association (AHA)/American College of Cardiology (ACC)/Heart Rhythm Society (HRS) guidelines for the management of atrial fibrillation (AF) have been recently updated to recommend non-vitamin K oral anticoagulants (NOACs) over warfarin in patients who are eligible for NOAC therapy.1 Because all NOACs are partially eliminated through the kidneys, a different dose is labelled and recommended for each NOAC according to renal function or a combination of renal function and other factors including age and body weight.2 Incorrect NOAC dosing may be associated with adverse outcomes and needs to be avoided.3
Decline in renal function is common in patients with AF. A meta-analysis of four trials comparing the efficacy and safety of NOACs relative to warfarin showed that estimated glomerular filtration rate (eGFR) declined in patients treated with either warfarin or NOACs (−3.66 mL/min/1.73 m² for warfarin; −2.54 mL/min/1.73 m² for NOACs at 30 months after randomisation).4 Such declines are expected to be greater in routine practice as patients in clinical trials tend to be younger and healthier. Consequently, dose adjustment in patients treated with NOACs is recommended in the current US and European guidelines to avoid overanticoagulation.1 5 The extent to which this occurs in routine practice, however, is unknown.
The aims of this study were (1) to evaluate changes in renal function in patients with AF over time, (2) to understand what frequency of patients required NOACs dose adjustment due to the change in renal function over time and (3) to understand how often patients actually receive dose adjustment in NOACs.
Methods
Data source and study population
The patients in this analysis were enrolled in the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation II (ORBIT-AF II) registry. The rationale, design and methods of the ORBIT-AF II registry have been published previously.6 Briefly, ORBIT-AF II is a prospective, multicentre nationwide registry of patients with AF. Patients were enrolled at 244 sites by a diverse group of healthcare professionals including internists, cardiologists and electrophysiologists up to 2 years from February 2013 to July 2016. ORBIT-AF II enrolled only patients who either had a new diagnosis of AF within the previous 6 months or were started on NOACs for AF within the previous 3 months. Eligible patients had to be 21 years of age and older with electrographically documented AF. Moreover, eligible patients had to be able to adhere to local follow-up every 6 months. Patients with AF due to a reversible cause (eg, pulmonary embolism, acute thyrotoxicosis or postoperative status), solitary atrial flutter without AF or a life expectancy of <6 months were excluded. A web-based case report form was used to collect information on patient demographics, medical and surgical history, medications, vital signs, laboratory data and imaging and electrocardiographic parameters. Changes in pharmacotherapy including the dose of NOACs, cardiac rhythm and subsequent cardiovascular events and procedures were identified during follow-up and collected via a web-based case report form.
For the purpose of evaluating changes in renal function in patients with AF over time, we excluded (1) patients who were not on OACs at baseline, (2) patients who were on more than one OACs at baseline, (3) patients with missing baseline or follow-up creatinine clearance (CrCl), (4) patients who were on haemodialysis or CrCl <15 mL/min at baseline, (5) patients who discontinued OACs, changed from NOACs to warfarin, or changed from warfarin to NOACs before any follow-up serum creatinine (Cr) measurements and (6) patients without follow-up data after the first follow-up serum Cr measurement (figure 1).
For the analysis of evaluating how changes in renal function may have impacted the dose of NOACs, we identified separate cohort of patients on NOACs who were eligible for dose adjustment due to a decline in renal function by applying the following exclusion criteria: (1) patients who were not on NOACs at baseline, (2) patients who were not on the standard dose according to the US Food and Drug Administration (FDA)-approved package inserts (online supplementary eAppendix 1) at baseline, (3) patients who discontinued OACs, changed from NOACs to warfarin or switched the type of NOACs before any follow-up serum Cr measurements (online supplementary eFigure 1).
Supplemental material
Definitions
Serum Cr was collected at baseline and at follow-up visits every 6 months up to 2 years. Because the measurement and the timing of follow-up visits were at health providers’ discretion, serum Cr was opportunistically collected as available. The Cockcroft-Gault equation was used to calculate CrCl.7 Worsening renal function (WRF) was defined as a decrease of >20% in CrCl from baseline serum Cr measurement at any time during the study period.8 Patients were considered to be in the stable renal function (SRF) group if they did not meet the definition of WRF throughout the study period. Two additional renal outcomes were separately evaluated: a decrease of >30% in CrCl from baseline serum Cr measurement at any time during the study period and an absolute increase of 0.3 mg/dL in serum Cr from baseline at any time during follow-up.9 US FDA-approved package inserts were used to define the reduction criteria of NOACs dosing (online supplementary eAppendix 1).
Statistical analysis
The study population was stratified into WRF and SRF groups and baseline characteristics were presented as counts (percentages) and median (IQR) for categorical and continuous variables, respectively. Time in the therapeutic range was calculated in patients treated with warfarin from the first international normalised ratio (INR) after baseline until the patient is censored using the Rosendaal method. Censoring included discontinuation of OAC, switch to NOAC, lost to follow-up or death. Gaps between INR measurements >8 weeks (56 days) were also excluded.
Mean changes in CrCl from baseline over time were analysed by using a restricted maximum likelihood-based repeated measures approach to account for the correlation of multiple serum Cr measurements over time within an individual and presented by type of OAC. The model included the fixed effects of baseline CrCl, type of OAC (NOACs vs warfarin), time point (6, 12, 18, 24 months) and the type of OAC by time point interaction. An unstructured covariance was used to model the within-patient measurements, and the Kenward-Roger approximation was used to estimate the denominator df.
The frequency and percentage of patients for whose CrCl declined >20% from baseline, CrCl declined >30% from baseline and whose absolute Cr increased >0.3 mg/dL from baseline are presented by type of OAC and compared using the χ2 test.
To understand how changes in renal function may have impacted the dose of NOACs, the frequency and percentage of patients whose CrCl fell below the threshold for dose adjustment is presented overall and by NOAC type. Additionally, the frequency and percentage of patients whose dose was reduced after meeting the threshold for dose adjustment is presented. Among patients whose CrCl fell below the threshold for dose adjustment, baseline characteristics, provider specialty and adverse event rates were compared by patients whose dose was appropriately reduced and patients whose dose was not appropriately reduced. Adverse event of interest included major adverse cardiovascular and neurological events (MACNE), its components, major bleeding and bleeding-related hospitalisation. MACNE was defined as the occurrence of cardiovascular death, myocardial infarction, stroke/non-central nervous system systemic embolism or transient ischaemic attack or new-onset HF. Major bleeding was defined by the International Society of Thrombosis and Haemostasis (ISTH) criteria: fatal bleeding, and/or symptomatic bleeding in a critical area or organ, and/or bleeding causing a fall in haemoglobin level of 2 g/dL or more, or leading to transfusion of two or more units of whole blood or red cells.10
The de-identified, aggregate data were analysed by the Duke Clinical Research Institute using SAS software (V.9.4; SAS Institute) and two-tailed p value <0.05 was considered significant for all statistical tests.
This research was done without patient involvement. Patients were not invited to comment on the study design and were not consulted to develop patient-relevant outcomes or interpret the results. Patients were not invited to contribute to the writing or editing of this document for readability or accuracy.
Results
The ORBIT-AF II registry enrolled 13 394 patients with AF from February 2013 to July 2016 at 244 sites within the USA. After excluding patients not on OAC or on haemodialysis at baseline, a total of 911 (7.8%) patients and 2562 (21.8%) patients did not have serum Cr measurement at baseline and at any follow-up visits, respectively. After the exclusion criteria were applied, the remaining 6682 patients (median age (25th, 75th): 72.0 years (65.0, 79.0); 57.1% male; median CrCl at baseline: 80.1 mL/min (57.4, 108.5)) from 220 sites were analysed in this study (figure 1). Of these, 5566 (83.3%) patients were treated with NOACs, whereas 1116 (16.7%) patients were treated with warfarin with median (IQR) time in the therapeutic range of 65.3% (50.3%, 78.8%). Overall, 1543 (23.1%) met the definition of WRF during the follow-up. Median follow-up period was 361 (IQR 360–549) days. Numbers of patients who were censored at each time-point (6, 12, 18 and 24 months) and numbers of patients who were defined as SRF and WRF at each time-point are summarised in online supplementary eTables 1 and 2, respectively.
The baseline characteristics between those who experienced WRF and those not (SRF) are shown in table 1. Patients with WRF were older, more likely to be female and more likely to have a history of anaemia and cardiovascular comorbidities, such as congestive heart failure, coronary artery disease. In addition, they had higher CrCl at baseline (84.3 (25th, 75th: 59.9, 115.1) mL/min vs 79.3 (56.8, 106.5) mL/min). Patients who developed WRF had higher CHA2DS2-VASc scores, the Anticoagulation and Risk Factors in Atrial Fibrillation (ATRIA) bleeding scores and ORBIT bleeding scores.
Worsening renal function over time
Table 2 summarises the adjusted mean changes in CrCl from baseline by the type of OAC. A continuous decline in CrCl over time was observed regardless of the type of OAC, with an average of >6 mL/min at 2 years from baseline. The decline was not different between patients treated with NOACs and those with warfarin (p values at all time points >0.5).
The rates of all renal outcomes including decline in CrCl >20% from baseline, decline in CrCl >30% from baseline and absolute increase in Cr >0.3 mg/dL from baseline were higher in patients who were on warfarin than in those on NOACs (table 3).
Impact of renal function over time on the dose of NOACs
To understand how changes in renal function may have impacted the dose of NOACs, we examined patients whose CrCl fell below the threshold for dose adjustment. Among 4120 patients who were on standard dose of NOACs at baseline (online supplementary eFigure 1), a total of 154 (3.7%) patients met the dose reduction criteria based on FDA dosing guidelines. Of these, only 31 (20.1%) NOACs dosing was appropriately reduced (table 4). Fifteen patients on edoxaban were excluded from this analysis due to small sample size. Patients with appropriately reduced NOACs were more likely to have a history of cerebrovascular disease and chronic kidney disease. In terms of provider specialty, electrophysiology subspecialty was less likely to reduce NOAC dose based on reduction criteria (online supplementary eTable 3). Compared with patients with appropriately reduced NOACs, those without were more likely to experience bleeding complications (ISTH major bleeding: 1.7% vs 0%; bleeding hospitalisation: 2.6% vs 0%) at 1 year (table 5). Among 154 patients who met the dose reduction criteria, 145 patients had follow-up data after meeting the dose reduction criteria and, of these, 34 (23.4%) recovered renal function afterwards well enough not to need dose reduction. The rate of renal function recovery was not different between patients with appropriately reduced NOACs and those without (24.1% vs 23.3%) (table 6).
Discussion
Using the US multicentre AF registry, we examined declines in renal function over time in patients with AF in US clinical practice. There are several major findings from this analysis. First, 23% of patients had >20% decline in renal function during 2-year follow-up period. Second, changes in renal function did not differ in those treated with NOACs versus warfarin but adverse renal outcomes were more common in those treated with warfarin than in those with NOACs. Third, among those treated with NOACs, about 3.7% of them met guideline criteria for dose reduction of NOACs during 2 years follow-up, but of these, only 20.1% actually had a reduction.
The decline in renal function over time in our cohort was greater than previously observed in landmark randomised controlled trials comparing NOACs and warfarin in patients with non-valvular AF.4 8 11 12 In our analysis, the mean decline in CrCl during 24-month follow-up was −7.33 mL/min for NOACs and −6.36 mL/min for warfarin. A similar degree of decline in CrCl was also reported from another ‘real-world’ dataset.13 In contrast, in the Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared with Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation trial, renal function declined −3.5 mL/min in patients treated with rivaroxaban and −4.3 mL/min in those treated with warfarin despite enrolment of high-risk patients. Similarly, in general population, the mean decrease in CrCl was 0.75 mL/min/year, which was much smaller than that in our analysis.14 Patients with AF often have multiple comorbidities and risks for chronic renal function; therefore, it is not surprising to see decline in renal function over time.
Subanalysis from the Randomised Evaluation of Long Term Anticoagulation Therapy trial showed the decline in renal function was greater in patients treated with warfarin than those with NOACs.11 Although the adjusted mean changes in CrCl did not differ in those treated with NOACs versus warfarin in our study, the rates of adverse renal outcomes including decline in CrCl >20% from baseline, decline in CrCl >30% from baseline and absolute increase in Cr >0.3 mg/dL from baseline were consistently higher in patients treated with warfarin than those with NOACs, which are consistent with the previous report. The reason why the adjusted mean changes in CrCl did not differ between NOACs versus warfarin may be due to censoring during follow-up. In our cohort, more than half of patients were censored before 12 months follow-up (online supplementary eTable 1). This implies that some high-risk patients for decline in renal function could have been dead or lost to follow-up or discontinued medication before the follow-up measurements of serum Cr, resulting in underestimating the risk of decline in renal function. Warfarin-related nephropathy due to overanticoagulation can be a cause of decline in renal function in patients treated with warfarin; however, median time in the therapeutic range in our cohort (65.3%) was similar to that in landmark clinical trials (ranging from 58% to 68.4%).15
The 2019 update to the AHA/ACC/HRS guidelines for the management of AF recommends monitoring renal function at least once a year for patients treated with NOACs to prevent complications from thromboembolism and bleeding.1 Similarly, the European Heart Rhythm Association practical guide on the use of NOACs recommends annual monitoring for renal function.5 Despite the awareness of importance of monitoring renal function in patients who are on NOACs, data from New Zealand indicated that routine monitoring of renal function is not adequately performed.16 In our study, serum Cr had never been checked during 2-year follow-up in about 20% of patients. Among those with regular serum Cr monitoring, dose adjustment was rarely performed: only one in five opportunities. These findings suggest that there remains a lot of opportunities to improve quality of care in patients with AF, although further studies are required to evaluate whether appropriate dose adjustment based on routine monitoring for renal function improves clinical outcomes.
Limitations
There are several potential limitations in our study. First, our sample size was more modest than the large randomised NOAC trials. Second, we only used the Cockcroft-Gault methodology for determination of renal function, since this is the method used in the pivotal clinical trials and recommended in the label for the NOAC agents. We did not use the Modification of Diet in Renal Disease (MDRD) Study equation17 and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation,18 which may have yielded different results. Third, we did not take into account recovery of renal function or duration of renal impairment, which may partly explain the lack of appropriate dose reduction of NOACs. However, the rate of renal function recovery was similar between patients with appropriately reduced NOACs and those without (table 6), although about one-fourth of patients who met the dose reduction criteria recovered renal function afterwards. This suggests renal function recovery afterwards was not a main reason behind lack of dose reduction.
Conclusions
In the US clinical practice, about one-fourth of patients with AF had >20% decline in renal function over time during 2-year follow-up. As a result, 3.7% of those treated with NOACs met guideline criteria for dose reduction, but of these, only 20.1% actually had a reduction. Further evidence is warranted to see whether appropriate dose reduction based on routine serum Cr monitoring improves patient clinical outcomes.
Key messages
What is already known on this subject?
Non-vitamin K oral anticoagulants (NOACs) require dose adjustment for renal function.
What might this study add?
One-fourth of patients with atrial fibrillation who were on oral anticoagulants had >20% decline in renal function during 2-year follow-up period.
Among those treated with NOACs, about 3.7% of them met guideline criteria for dose reduction of NOACs during 2 years follow-up, but of these, only 20.1% actually had a reduction.
How might this impact on clinical practice?
Further efforts to improve quality of care in patients with atrial fibrillation in terms of dose adjustment of NOACs based on regular serum creatinine monitoring are warranted.
References
Footnotes
Twitter @taku_inohara, @gcfmd, @ba_steinberg, @JonPicciniSr
Contributors TI and JPP had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: TI, KP and JPP. Acquisition, analysis or interpretation of data: all authors. Drafting of the manuscript: TI, DNH, KP and JPP. Critical revision of the manuscript for important intellectual content: all authors. Statistical analysis: DNH and KP. Obtained funding: JPP. Administrative, technical or material support: RGB and JPP. Study supervision: KP, EDP and JPP.
Funding The Outcomes Registry for Better Informed Treatment of Atrial Fibrillation is sponsored by Janssen Scientific Affairs LLC.
Disclaimer The funders had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review or approval of the manuscript and decision to submit the manuscript for publication.
Competing interests TI: Research Grant from Boston Scientific. LA: Contract with Janssen and Novartis. GF: Consultant/Advisory Board support from Janssen Pharmaceuticals. BJG: Member of a Data Safety Monitoring Board for Mount Sinai St. Lukes, Boston Scientific Corporation, Teva Pharmaceutical Industries, St. Jude Medical, Janssen Research & Development, Baxter Healthcare Corporation and Cardiovascular Research Foundation. Consultant/Advisory Board for Janssen Scientific Affairs, Cipla Limited, Armetheon Inc and Medtronic. EMH: Consultant: Bayer, Boehringer Ingelheim, Bristol-Myers Squibb/Pfizer, Daiichi-Sankyo, Johnson & Johnson; Research Grant from Bristol-Myers Squibb/Pfizer, Johnson & Johnson. MDE: Consultant/Advisory Board for Boehringer Ingelheim, Daiichi-Sankyo, Pfizer, Bristol-Myers Squibb and Janssen Scientific Affairs. PRK: Consultant for Johnson & Johnson. JAR: Research support from Janssen Pharmaceuticals and Medtronic Inc. Consultancies with Medtronic Inc, Acesion Pharma Aps, and Correvio Pharma Corp. GN: Research Grant from Janssen. Consultant/Advisory Board for Janssen and Daiichi-Sankyo. PC: Employee of Janssen. Consultant for Optum Rx and Johnson & Johnson. KWM: Financial disclosures can be viewed at http://med.stanford.edu/profiles/kenneth-mahaffey. DES: Consultant/Advisory Board for Boehringer Ingelheim, Bristol-Myers Squibb, Merck, Johnson & Johnson, Pfizer and Medtronic. Research Grants from Boehringer Ingelheim and Bristol-Myers Squibb. JF: Consultant/Advisory Board for Janssen Scientific. BAS: Research support from Boston Scientific and Janssen. Consult for Janssen. Speakers’ bureau income Biosense Webster. EDP: Research Grant from Janssen Pharmaceuticals and Eli Lilly. Consultant for Janssen Pharmaceuticals and Boehringer Ingelheim. JPP: Research grant from Agency for Healthcare Research and Quality, ARCA biopharma, Boston Scientific, Gilead Sciences, Janssen Pharmaceuticals, Johnson & Johnson, ResMed, Spectranetics and St Jude Medical. Consultant/Advisory Board for BMS/Pfizer, GlaxoSmithKline, Janssen Pharmaceuticals, Johnson & Johnson, Medtronic and Spectranetics.
Patient consent for publication Not required.
Ethics approval The study was approved by the institutional review board at Duke University (the coordinating centre) and each participating centre obtained local institutional review board approval. Informed consent was obtained from each participant.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement No data are available.