A prospective observational study of haematuria and acute kidney injury in elderly patients presenting to hospital with supratherapeutic anticoagulation with warfarin.

Abstract

Aims: To determine the relationship between the International Normalised Ratio (INR) and the development of haematuria and acute kidney injury (AKI) in warfarin-treated patients admitted to hospital. Background(s): An acute rise in serum creatinine following warfarin overanticoagulation (INR > 3) has been termed warfarin nephropathy. It may be due to tubular obstruction by red cells or direct toxicity. Studies of warfarin nephropathy have been retrospective and the diagnosis presumptive, without distinguishing glomerular from non-glomerular bleeding. There may be unaccounted confounders for the development of AKI and a high INR. Elderly patients represent a susceptible group due to a higher prevalence of warfarin use and chronic kidney disease. Method(s): We conducted a single-centre, prospective, observational study of warfarin-treated patients admitted under General Medicine. Urine was analysed by phase contrast microscopy and albumin-creatinine ratio. AKI was defined by the RIFLE criteria. Patients who developed AKI or haematuria were reassessed in 4-6 weeks. Result(s): After screening 246 patients, 150 were included. The cohort was elderly (80 +/- 7 years) and mostly male (62%). 54% of patients had an INR > 3 on admission. The incidence of isolated AKI, isolated haematuria or both was 18.7%, 12% and 10% respectively. Antibiotic use increased the risk of overanticoagulation while non-urinary infection increased the risk of haematuria. Many patients with haematuria demonstrated non-glomerular morphology by microscopy. The presence of haematuria, infection and pre-renal factors were associated with a higher risk of AKI. An INR > 3 did not increase the risk of AKI or haematuria. The majority of AKI and haematuria resolved at follow-up. Conclusion(s): Elderly warfarin-treated patients admitted with infection were more likely to develop haematuria and AKI. An INR > 3 did not contribute to this risk in our cohort.