A Simple Scheme to Initiate Oral Anticoagulant Treatment in Outpatients With Nonrheumatic Atrial Fibrillation Vittorio Pengo, MD, Alessandra Biasiolo, DSci, and Cinzia Pegoraro, MD A trial fibrillation (AF) is a common arrhythmia associated with an increased incidence of stroke, 1 which is significantly reduced with oral anticoagulant treatment. 2 Diagnosis of AF is often made by means of ambulatory electrocardiography, with oral antico- agulant treatment initiated on an outpatient basis. The most popular protocol to initiate warfarin treatment considers prothrombin time-international normalized ratio (PT-INR) determinations every day for the first 4 days. 3 This protocol was shown to be effective in a rela- tively young group of inpatients but not in older groups. 4,5 Moreover, the inconvenience of daily monitoring of treatment renders this scheme impractical for outpa- tients with AF. We have recently developed a scheme for early indication of warfarin maintenance dose on the basis of PT-INR at day 3 after 2 consecutive daily 10-mg doses of warfarin. 6 However, it was recently shown that a 5-mg loading dose produces less excess of anticoagulation, 7 reinforcing the advice to initiate anticoagulation on an outpatient basis with an antici- pated maintenance warfarin dose of 4 to 5 mg/day. 8 Because excess anticoagulation and older age are as- sociated with increased bleeding complications, 9 we decided to shift from the total 20-mg warfarin dose in 2 days (induction scheme) into a total of 20 mg of warfarin administered over 4 consecutive days. Simi- lar to the induction scheme, we then established a correlation between the INR on the fifth day and the warfarin maintenance dose. ••• From January 1998 to June 1999, all consecutive patients with nonrheumatic AF referred to the Throm- bosis Center of the Padova Civic Hospital for initia- tion of oral anticoagulant treatment were considered for the study; no patient was receiving heparin when evaluated for the study. Patients were excluded if they (1) were treated with drugs known to interfere with warfarin (in particular amiodarone, which is often prescribed to these patients), (2) had an associated disease known to affect coagulation, (3) had a relative contraindication to treatment, (4) had undergone a previous course of anticoagulant treatment, (5) had a basal INR that was above 1.2, or (5) refused their informed consent to participate. All patients followed the basic educational program at the center. Patients received a prescription of 5 mg/day of warfarin for 4 consecutive days. PT-INR determination was sched- uled on the fifth day. Dose prescription on the fifth day was freely decided by the doctor in charge. Thereafter, the date of the next PT-INR determination and pre- scription was recommended to be 1 week apart for the first 2 weeks. INR determinations were obtained by using recom- binant thromboplastin (Instrumentation Laboratory, Milan, Italy) and an Electra 1400 coagulometer (Hemoliance, Milan, Italy) with an International Sen- sitivity Index value close to 1.0, certified for each batch by the manufacturer and a national authority. External quality control and clinical and laboratory follow-up of patients attending the center were in line with the recommendations of the ltalian Federation of Anticoagulation Clinics. 10 Follow-up was continued for up to 3 months to establish the warfarin mainte- nance dosage; the warfarin dose was considered to be stabilized when the INR values were between 2.0 and 3.0 (therapeutic range) on 3 consecutive occasions, 1 week apart. Major and minor bleedings as well as thromboembolic complications were recorded and de- fined as previously reported. 9,11 The relation between the INR at day 5 (after 4 consecutive daily doses of 5 mg of warfarin) and the weekly maintenance dose of warfarin in patients was evaluated. To test the validity of an early maintenance dose, an additional group of patients with nonrheu- matic atrial fibrillation were prescribed the predicted maintenance dose after 4 days of treatment according to the obtained scheme. To expedite estimation of the warfarin mainte- nance dose, each patient’s INR value on day 5 was plotted against the weekly maintenance dose. The best fit was obtained by means of minimal squares. The mean warfarin dosage in groups of patients was com- pared by the unpaired t test with Welch’s correction. The agreement between the estimated and actual maintenance dose was assessed by plotting the data according to Bland and Altman. 12 ••• From a total of 105 candidate outpatients with AF, 14 were excluded (8 for associated interfering drugs, 3 for basal INR 1.2, and 3 for an associated dilated cardiomyopathy). After informed consent, the remain- ing 91 patients (mean age 71 years, range 42 to 88) received 5 mg/day of warfarin for 4 consecutive days. Fifty-four (59%) were men, and major risk factors for systemic embolism (hypertension, reduced left ven- tricular ejection fraction, previous stroke) were present in 50 (55%), 21 (23%), and 9 (10%) patients, respectively. Thirty patients were not included in data analysis; 16 of these patients completed the 3 months From the Department of Clinical and Experimental Medicine, Cardiol- ogy Section, University of Padova, Padova, Italy. Dr. Pengo’s address is: Department of Clinical and Experimental Medicine, Thrombosis Center, University of Padova School of Medicine, “Ex Busonera” Hospital, via Gattamelata, 64 I-35128 Padova, Italy. E-mail: vittorio.pengo@unipd.it. Manuscript received May 29, 2001; revised manuscript received and accepted July 6, 2001. 1214 ©2001 by Excerpta Medica, Inc. All rights reserved. 0002-9149/01/$–see front matter The American Journal of Cardiology Vol. 88 November 15, 2001 PII S0002-9149(01)02069-0