Oral anticoagulation treatment in atrial fibrillation - To bleed or not to bleed, that is the question

Detta är en avhandling från Department of Clinical Sciences, Lund University

Sammanfattning: Background: Oral antigoagulation treatment (OAT) with warfarin have a narrow therapeutic window and patients exhibit a highly variable doseresponse that is attributable to genetic, disease-related,and environmental factors as well as prescription and nonprescription drugs, dietary vitamin K and alcohol. The effect of warfarin treatement on blood coagulation is measured using international normalized ratio (INR). Monitoring and tight control of anticoagulation treatment reduces the risk of both thrombosis and bleeding. The time in therapeutic range (TTR) has been validated as a quality indicator of the anticoagulation treatment given, correlating inversely to complication frequencies. The Swedish national quality registry for atrial fibrillation and anticoagulation (AuriculA) was founded in 2006 and has by August 2011 enrolled 68,000 patients, close to 40 % of all patients on anticoagulation in Sweden. Patients with end stage renal disease are at high risk for both bleeding and thrombosis and even a moderate decrease in glomerular filtration rate (GFR) is associated with thromboembolic and bleeding complications. Impaired kidney function has also been shown to be associated with a greater need of warfarin dose adjustment, poor TTR, and an increased risk of bleeding in patients taking warfarin. New anticoagulants, most of them eliminated by renal route, can be administered in a fix dose, have few drug and no dietary interactions, has been shown to be at least inferior and in some cases superior to OAT with warfarin. Methods:
The Anticoagulation Clinic in Malmö participates in AuriculA, and uses this registry for prospective follow-up and as a tool for OAT dosage. This thesis has utilized data from AuriculA for epidemilogical analysis of impaired kidney function, using two equations for estimated glomerular filtration rate (eGFR), TTR, and the correlation of these two markers to major bleeding and thrombo-embolic events. Results: In Paper I, TTR in 18391 patients in the whole Swedish AuriculA population was 76.2%. Compared to recent prospective randomized trials of warfarin treatment, TTR in the AuriculA population was higher. Adherence to treatment, as measured by TTR, was higher in elderly patients with a significant correlation with increasing age. In 4273 patients from two centers in AuriculA the frequency of major bleedings and venous/arterial thromboembolism were 2.6 % and 1.7 % and for atrial fibrillation (AF), 2.6 % and 1.4 %, per treatment year respectively. In Paper II, the fraction of 2603 AF patients on warfarin with eGFR <30 and <45 ml/min/1.73 m2 were 8.1% and 22.9% with the LM and 4.3% and 16.3% with the MDRD equation, respectively, and significantly higher than corresponding values in a reference population. In paper III, major bleeding and thromboembolic events in 3536 patients on OAT with warfarin were registered. Bleeding events were significantly related to age and percentage of time with INR >3.0 (p<0.001 for both). There was no correlation between age, eGFR and thromboembolic events. The prevalence of eGFR <45ml/min/1.73m2 was 52% in patients aged ≥75 years with major bleeding. eGFR levels <30 ml/min/1.73m2 were particularly associated with high risk of bleeding in elderly patients. No correlation between eGFR and thromboembolic events was seen. In Paper IV, 397 patients on OAT with warfarin, there was a positive correlation between results from the point-of-care device (POC) CoaguChek XS and the Owren-type PT assay for INR measurement (r=0.94;p<0.001) and concordance of 88.2%. In patients with 152 double samples analyzed with the CoaguChek XS, a positive correlation of 0.99 was seen; p<0.001. Conclusions: The quality of OAT with warfarin in Sweden is high and comparable to prospective randomized trials of warfarin treatment. Complications were low, probably due to the organization of anticoagulation treatment in Sweden, although the AuriculA dosing algorithm could have contibuted by keeping dosing regimens consistent over all centers. The POC-device CoaguChek XS presents reproducible results, highly comparable with Owren PT at therapeutic levels of INR, offering a more convenient method of monitoring, compared to regular venous INR measurement in patients on OAT with warfarin. Severe renal impairment is common among AF patients on OAT with warfarin, especially at higher ages, indicating one important difference between a ‘real world’ clinical population and those of randomized controlled trials of new oral anticoagulant drugs, where patients with severe renal failure (eGFR <30 ml/min/1.73 m2) were excluded. Given the strong correlation between eGFR and major bleeding events in patients on anticoagulation treatment demonstrated in Paper III, caution is advised in the upcoming era of new oral anticoagulants with elimination by renal route. Monitoring of renal function should be implemented in clinical practice for AF patients treated with new anticoagulants eliminated by the kidneys and registries like AuriculA can be used for prospective follow-up of these patients.