Statistics from Altmetric.com
Editor,—McNulty et alreport improved management of atrial fibrillation in a busy district general hospital by increasing the percentage of patients at high risk of stroke who receive oral anticoagulation.1They suggest that, based on the results of large randomised prospective studies reported in the late 1980s and early 1990s, a similar improvement in management of all hospital patients with atrial fibrillation in the UK would avoid approximately 1400 strokes annually.
The present recommended international normalised ratio (INR) therapeutic range of 2–3 for atrial fibrillation patients on oral anticoagulation is aimed at reducing the risk of thromboembolic disease while minimising the risk of major bleeds.2 However, maintaining orally anticoagulated patients in the desired therapeutic INR range in clinical practice appears to be much more difficult than INR control in carefully selected, well motivated, and closely monitored study subjects.3 For example, in a study of consecutive hospital admissions of over 300 patients to six academic hospitals in the USA, only approximately a third of patients had an INR between 2 and 3 at the time of admission.4 The reasons for poor anticoagulant control are numerous but the major underlying cause remains poor patient compliance.
Unfortunately, McNulty et al fail to provide information on the percentage of time orally anticoagulated patients in their audit were in therapeutic INR range or on the incidence of thrombotic stroke and major haemorrhage in this patient group. As implementation of antithrombotic management in atrial fibrillation patients in their hospital remains suboptimal, even on follow up audit, INR control in the orally anticoagulated patient group could also be inadequate, thus jeopardising the hoped for reduction in strokes in this high risk atrial fibrillation group. In addition, the increase in patients on oral anticoagulation in the period between the two audits may have adversely affected the delivery of acceptable INR control by increasing demands on time and resources in a presumably already overstretched district general hospital haematology laboratory.
We suggest that, in future, similar audits of management of atrial fibrillation should include information on the quality of INR control in orally anticoagulated patients. If such patients are found to be inadequately maintained in therapeutic INR range, increased resources should be allocated to improve the efficacy of INR control. Possible areas of improvement in anticoagulant control might include the upgrading of laboratory coagulation equipment, better communication between clinicians and the anticoagulation clinic, adequate laboratory, clerical, nursing and medical staffing, ongoing and more effective patient and hospital staff education, computerised dosage systems5 and, in selected cases, patient self monitoring and self management.