Evaluation of Anticoagulation Control among Patients Taking Warfarin in University of Gondar Hospital, Northwest Ethiopia

Introduction Warfarin is a widely used oral anticoagulant in clinical practice. It has variable intraindividual and interindividual dose response and a narrow therapeutic index. Therefore, it requires frequent and regular international normalized ratio (INR) determination to maintain the INR within the therapeutic range. The study evaluated parameters of anticoagulation control among patients on warfarin. Methods A cross-sectional study was conducted at University of Gondar hospital. A consecutive sampling method was used to recruit study subjects. The anticoagulation control was evaluated by determining the proportion of desired INRs and the proportion of time spent in the therapeutic range (TTR). Logistic regression analysis was used to identify associated factors with adequate TTR. A P value <0.05 was used to declare significant association. Result A total of 338 study subjects were included in the study. The mean age of patients was 48.8 (SD = 16.4) years. Atrial fibrillation was the commonest indication for warfarin therapy. One-third (33%) of study subjects achieved the desired INRs of 2.0–3.0, while about one-tenth (13%) of patients attained good INR control (TTR ≥ 65%). Multivariate logistic regression analysis revealed no significant association of sociodemographic and clinical characteristics with good TTR outcome. Conclusion The level of anticoagulation control with warfarin among study subjects was very low. The authors recommend to implement a validated warfarin-dose titration protocol and to establish anticoagulation clinics to mitigate the low anticoagulation level.

is study aimed to determine the magnitude of proportion of INRs within the desired range, proportion of time spent in the therapeutic INR range, and factors associated with quality of anticoagulation outcomes in patients taking warfarin in a hospital setting, in Northwest Ethiopia.

Study Setting.
A hospital-based cross-sectional study was conducted between November 1, 2019, and October 31, 2020, at Cardiology and Hematology Clinics, University of Gondar hospital, Northwest Ethiopia. e Cardiology and Hematology units provide healthcare services for outpatients who were followed up at the clinics. Both clinics were run by internists, medical residents, and respective unit nurses. Patients were admitted every one to three months based on their severity of illness.

Study Population and Study
Subjects. Patients older than 18 years and who were on warfarin with follow-up at Cardiology and Hematology Clinics, University of Gondar hospital, during the study period were the study population.

Inclusion Criteria.
Patients aged 18 years old and above, who were on warfarin for at least 6 months, had at least 6 INR determinations, and had each consecutive INR determination less than 2 months apart were included in the study.

Exclusion Criteria.
Patients who did not give consent to the study, refused to be included in the study, and had incomplete data were excluded from the study. Independent variables included the following: (1) sociodemographic characteristics including age, gender, occupation, marital status, educational level, income level, residence, and religion; (2) clinical characteristics including indication for warfarin, warfarin dosage, concomitant drugs intake (antiplatelets, statins, antihypertensive drugs, antithyroid drugs, antiretroviral drugs), adherence to warfarin, INR monitoring interval, co-existing comorbidities, body mass index, and alcohol intake.

Sample Size and Sampling Procedure.
e sample size was calculated using a single population proportion formula with the assumption of 95% confidence level and 5% margin of error and taking 30% for TTR in the sub-Saharan African region [12]. A consecutive sampling method was used to recruit 338 study subjects.

Data Collection Instrument and
Procedures. Data were collected through an investigator-administered predesigned questionnaire. e questionnaire was prepared in English and translated into local language (Amharic) for data collection and then retranslated back to English while maintaining its consistency. e questionnaire had been pretested on 34 patients in a similar setup before the actual data collection was commenced to check for consistency and reliability of the questionnaire. Patients were interviewed to obtain sociodemographic data. Relevant medical history and laboratory parameters were obtained from patients' records. One supervisor (MD+) and two data collectors (MD) participated in the data collection process. Quality of data was ensured through training and supervision of data collectors. Patient characteristics were reported as counts (percentages) for categorical variables and mean with standard deviation for continuous variables. Bivariate and multivariate logistic regression models were constructed to identify associated factors with anticoagulation outcome. Crude odds ratio (COR) and adjusted odds ratio (AOR) were reported. A P value <0.05 was used to declare a significant association.

Ethical Considerations.
e research protocol complied with the Declaration of Helsinki, and ethical clearance was obtained from the Institutional Review Board (IRB) of College of Medicine and Health Sciences, University of Gondar (04/09/2012; IRB No. 09/20/627/12). Study subjects were recruited only after informed written consent was obtained. All data obtained were treated confidentially. During the data collection process, those patients who were found to have atrial fibrillation and venous thromboembolism were taken care of as per the recommendations of AHA/ACC guidelines and American Society of Hematology guideline, respectively [2, 3].

Definition of Terms.
e proportion of INRs within the desired range of 2.0-3.0 (2.5-3.5 for mechanical prosthetic valves) is defined as the number of INRs within the target range divided by the number of INR measurements per patient [22]. e proportion of time spent within the therapeutic INR range (TTR) is defined as the duration of time for which the patient's international normalized ratio (INR) values were within a therapeutic range of 2.0-3.0 (2.5-3.5 for mechanical prosthetic valves). It was calculated using Rosendaal's method, which has used linear interpolation to assign an INR value to each day between successive observed INR values. TTR was calculated as the number of person-days with therapeutic INR range divided by the total number of person-days on warfarin. TTR ≥ 65% declares "good" INR control, while TTR < 65% affirms "poor" INR control [23].
Warfarin adherence: "continuous, single interval measure of medication gaps" was used to assess medication refill for warfarin. It was calculated as the sum of the days a patient was late for warfarin pick-up appointments in each month of the year, divided by the total number of days between pickup periods in the year of study. Nonadherence was defined as more than one-third of days late for warfarin pick-up appointments [24].
Major bleeding in nonsurgical patients is defined as fatal bleeding; and/or symptomatic bleeding in a critical area or organ, such as intracranial, intraspinal, intraocular, retroperitoneal, intra-articular or pericardial, or intramuscular with compartment syndrome; and/or bleeding causing a fall in hemoglobin of 2 gm/dl or more, or leading to transfusion of 2 or more units of whole blood or red cells [25].
Clinically relevant nonmajor bleeding in nonsurgical patients is defined as any sign or symptom of hemorrhage (e.g., more bleeding than would be expected for a clinical circumstance, including bleeding found by imaging alone) that does not fit the criteria for the ISTH definition of major bleeding but does meet at least one of the following criteria: requiring medical intervention by a healthcare professional; leading to hospitalization or increased level of care; prompting a face-to-face (i.e., not just a telephone or electronic communication) evaluation [26].

Proportion of Time Spent in the erapeutic INR Ranges.
e proportion of time spent within the therapeutic range (TTR) was determined for all study participants (Figure 2). More than one-tenth (44/338, 13%) of patients achieved good INR control (TTR ≥ 65%). More than one-third (125/ 338, 37%) of study subjects had poor TTR control (TTR � 36-65%). Half (169/338) of the study subjects attained TTR < 35%, which was interpreted as no survival benefit from warfarin therapy.

Discussion
Atrial fibrillation (75%) was the commonest indication for warfarin use, followed by venous thromboembolism and prosthetic heart valves. is finding was shared by reports from other sub-Saharan African countries [12][13][14][15][16]. e proportion of adequate anticoagulation outcome (TTR ≥ 65%) in the study subjects was 13%. e finding was on the lower margin of previous reports in sub-Saharan Africa, where the proportion of adequate INR level was 15-25% [12][13][14]. However, the United States and European studies have shown that the proportion of adequate anticoagulation level was 50-70% [7][8][9][10][11]. e reason for varied magnitude of anticoagulation control among countries was partly explained by their difference in anticoagulation management services. Reports from the Western world had shown that establishing anticoagulation clinics, implementing computer-assisted warfarin dosing, self-testing and self-management among motivated patients, and instituting coordinated follow-up clinics had improved the anticoagulation control [27,28]. In this study, only one-third (33%) of study subjects attained the desired INR value (INR � 2.0-3.0) and majority (59%) were on low-dose warfarin (<5 mg, daily). Physicians might tend to undertreat patients because of fear of bleeding [29]. A quarter (24%) of patients were nonadherent to their medication. Patients' comprehension on health benefits of the oral anticoagulant might be low [30]. ere was no validated warfarin-dose titration protocol in the hospital. Warfarin-dose titration was based on conventional clinical practice. "Anticoagulation clinic" for patients' close follow-up was not available.
ese challenging conditions might have contributed to poor anticoagulation control. On bivariate analysis, VTE as indication for warfarin; warfarin dose, 5 mg PO daily; and absence of heart failure were predictors of good INR outcome (TTR ≥ 65%). Patients with VTE might not have pill fatigue, since time to treatment was limited. Even though patients on higher warfarin dosage might achieve an acceptable INR target, variant alleles of the CYP2C9 and VKORC1 genes among patients might affect the warfarin-dose requirements, sensitivity to warfarin, and quality of anticoagulation. Few studies documented the presence of genetic polymorphisms to warfarin among African descents [31]. Patients with no heart failure might not be on multiple drugs, which then limited drug-drug interactions. On multivariate analysis, explanatory sociodemographic and clinical characteristics showed no

Limitation of the Study.
Since it was a cross-sectional study, the determined anticoagulation level might not be a true reflection of what happens all the time. Generalizability to the study population was limited, since the nonprobability sampling method was used to recruit study subjects.

Conclusion
e quality of anticoagulation control with warfarin in the study subjects was very low. e authors recommend to implement a validated warfarin-dose titration protocol and to establish anticoagulation clinics to mitigate the low anticoagulation level.

ACC:
American College of Cardiology AHA: American Heart Association AOR: Adjusted odds ratio ART: Antiretroviral therapy CI: Confidence interval COR: Crude odds ratio DOACs: Direct oral anticoagulants INR: International normalized ratio IRB: Institutional Review Board PTU: Propylthiouracil TTR: Time spent within the therapeutic range VTE: Venous thromboembolism.

Data Availability
All data generated and analyzed are included within this research article.

Ethical Approval
Ethical approval was obtained from the Institutional Review Board of College of Medicine and Health Sciences, University of Gondar. Formal letter of permission was obtained from the University of Gondar hospital administrative body.

Consent
Study subjects were recruited only after informed written consent was obtained. Written informed consent for publication was also obtained from study subjects.

Disclosure
A preprint has previously been published [32]. e funding body had no role in the design of the study, data collection, analysis and interpretation of the data.

Conflicts of Interest
e authors declare that they have no conflicts of interest.  6 Advances in Hematology