Effective and Safe Management of Oral Anticoagulation Therapy in Patients Who Use the Internet-Accessed Telecontrol Tool SintromacWeb

Background: Despite the existing evidence that highlights the benefits of oral anticoagulation therapy (OAT) self-testing and self-management by patients in comparison with conventional control, significant progress is still needed in the implementation of computer-based, Internet-assisted systems for OAT within health care centers. The telecontrol tool “SintromacWeb” is a previously validated system for OAT management at home, which is currently operative and accessed by patients through a hospital Web portal. Objective: The intent of the study was to assess the effectiveness and safety of OAT management in patients using the SintromacWeb telecontrol system in reference to control in patients using the conventional system (management at the hematology department), in terms of time in therapeutic range (TTR) of International Normalized Ratio (INR). Methods: In this observational prospective study, patients were identified by their physician and divided in two groups according to the OAT management system that they were already using (conventional control or telecontrol with SintromacWeb). For 6 months, patients were required to visit the hematology department every time their physician considered it necessary according to usual clinical practice. Sociodemographic and clinical variables for the study were collected at first visit (baseline) and at those visits closest to 2, 4, and 6 months after first visit. Results: A total of 173 patients were evaluated, 87 with conventional control and 86 with telecontrol. Follow-up time was a median of 6.3 (range 5.2-8.1) months. The average time of OAT treatment prior to enrollment was 9.2 (SD 6.4) years. Patients in the telecontrol group tested their INR a median of 21 (range 4-22) days versus a median of 35 (range 14-45) days in patients in the conventional control group (P<.001). TTR in the telecontrol group was 107 (SD 37) days versus 94 (SD 37) days in the conventional control group (an increase of 12.6%; P=.02). In all visits, the percentage of TTR was higher in the telecontrol group (at the third visit: 59% vs 48%; P=.01). Higher TTR (positive coefficient) was associated with patients under OAT telecontrol (P=.03). Under-anticoagulation (INR<1.5) and over-anticoagulation (INR>5) were observed in 34 (19.7%, 34/173) and 38 (22.0%, 38/173) patients, respectively (no differences between treatment groups). Seven thrombotic and/or bleeding events were serious, 12 were non-serious, and most of them (5 and 10, respectively) occurred in the conventional control group. Conclusions: In clinical practice, OAT management with the Internet-based tool SintromacWeb is effective and safe for those patients who are eligible for OAT telecontrol. (Interact J Med Res 2015;4(2):e10) doi: 10.2196/ijmr.3610


Introduction
Oral anticoagulation therapy (OAT) with vitamin K antagonists effectively reduces the risk of thromboembolism in patients with hereditary or acquired thrombophilia, heart valve replacement, atrial fibrillation, and other conditions [1]. Studies carried out in Spain and Italy showed a prevalence for OAT of 1.32% and 0.81%, respectively, with atrial fibrillation as the main cause of an indication for anticoagulation therapy (47.1% and 45.6%, respectively) [2,3].
The INR (International Normalized Ratio) is a standardized number obtained by means of a laboratory test that determines the degree of anticoagulation level achieved by the vitamin K antagonist [4]. The goals of OAT are both preventing thromboembolism and minimizing the risk of bleeding complications by reaching and maintaining the INR within the appropriate range for each patient, depending on their disease [5]. To summarize the INR control over time, percent time in therapeutic range (TTR) of INR is used.
Monitoring of OAT patients is conventionally carried out in hospitals and primary care centers, which are in contact with the referring hematologist in specialized hospitals. However, there is extensive literature that supports effectiveness and safety of OAT self-management by patients at home and shows that self-management at home is similar or even more effective and safer than in conventional control [6][7][8][9]. Decentralized management not only lightens the burden in health centers, but also prompts fewer monitoring visits to specialized centers, resulting in more freedom for the patient and improved quality of life [10][11][12]. Furthermore, that the patient is provided with greater responsibility in measuring and dosing their own INR can increase awareness, commitment, and interest in the management of their disease.
According to biomedical literature, any new management model should demonstrate anticoagulation control levels over 60% for TTR of INR to be considered safe and be within at least 5 to 10% compared with routine monitoring to declare it as a better model [6][7][8]. Since TTR is a parameter strongly associated with the occurrence of clinical events, its use as a primary endpoint in clinical trials of anticoagulation is recommended [13]. Few randomized trials assessing TTR-based effectiveness and/or safety of OAT self-testing programs have been performed so far [14][15][16][17][18][19][20]. Importantly, no trial has tested a validated, already implemented system in a hospital for OAT telecontrol at home. "SintromacWeb" is a new-generation, Internet-based system developed by Grifols (Barcelona, Spain) as an alternative tool to conventional OAT management. SintromacWeb has already been successfully tested and validated in terms of reliability, consistency, and patient satisfaction [21]. The system is currently operative in the La Fe University Hospital (Valencia, Spain), and it is accessed by patients through the hospital Web portal.
Designed as the natural extension of its predecessor [21], this study prospectively assessed both effectiveness, in terms of TTR, and safety of OAT management in patients using the SintromacWeb system in clinical practice, compared to patients using management at the hematology department, a system considered conventional.

Study Design and Objectives
This prospective, observational study was conducted at the La Fe University Hospital in Valencia, Spain. The study was performed in accordance with the Declaration of Helsinki and approved by the Ethics Committee for Clinical Investigation by the La Fe Hospital.
The main objective of this study was to assess the effectiveness and safety of OAT management in patients using the SintromacWeb compared to OAT management in patients using the conventional control (management at the hematology department). Therefore, the TTR of the INR in both groups were compared. The safety objective was to assess the presence of abnormal coagulation events such as the number and severity of thromboembolism and bleeding events during the follow-up period.

SintromacWeb System Description
The SintromacWeb system for OAT telecontrol at home consists of two key elements: a point-of-care (POC) device for patients' INR self-testing and software that allows online interaction with physicians.
The POC device used was the HemoSense INRatio (Philips Remote Cardiac Services, Windsor CT), which is a monitoring system for INR home testing. The INRatio is capable of receiving the prothrombin time and INR results in less than a minute by using a small blood drop. A test strip is inserted into the INRatio monitor and a sample of fresh whole blood (15 µL) received from a finger prick is applied to the test strip. Blood is drawn into the test area by capillary action where it mixes with coagulation inducing reagents. The monitor performs the test and determines whether the controls are within pre-set limits.
The SintromacWeb software allows OAT patients to communicate with their doctors online, at home, or wherever an Internet access point is available. The SintromacWeb site is hosted by the same server as the La Fe Hospital website and is accessed through the hospital portal. Patients are provided with a username and password, which allows them to enter their personal area (Figure 1 A). The current medication schedule can be viewed, and according to the self-testing program, INR results are introduced and sent to the health care center. As a result, the hematologist can connect to the system, analyze the data, and introduce a new medication schedule for the patient (Figure 1 B). After the doctor has updated the schedule, an email is sent to the patient informing them that the treatment recommendations are available in their personal area of SintromacWeb.

Patients and Recruitment
All patients were enrolled at the Hemostasis and Thrombosis Unit of the Hematology Service of the La Fe Hospital. The enrollment period lasted 3 months and patients were followed for 6 months.
The main inclusion criteria were: patients aged 18 years or older, under OAT treatment (either telecontrol or conventional control) for at least 3 months before inclusion in the study, and expected to be under OAT for at least 6 months after the inclusion in the study. The main exclusion criteria included patients who did not follow criteria for treatment adherence according to the investigator, patients who were not able to follow the visit schedule, and patients participating in another clinical trial during the study period.
Patients who met the criteria were identified by their physician when they came to the hematology department for a routine visit. Candidates were invited to participate in the study, and after acceptance, written informed consent was obtained. Patients were divided in two groups according to the OAT management that they were already following (conventional control or telecontrol with SintromacWeb). Patients using the telecontrol management system came from the cohort of the previous study in which the main inclusion criteria were patients aged 18 years or older with adequate technical facilities at home (computer and connection to the Internet) to run the Web-based telecontrol tool [21]. Recruitment ceased when the calculated minimal population size for each treatment group was reached.

Variables of the Study
During the follow-up period, patients were required to visit the hematology department for OAT every time their physician considered it necessary, according to usual clinical practice. Variables of interest for the study were collected at four time points according to the following schedule: first visit (baseline) at the beginning of the study and at three follow-up visits (those closest to 2, 4, and 6 months after first visit). Variables corresponding to the 3 months prior to the first visit were obtained from the patient's clinical records.
Information was gathered in a Case Report Form (CRF) specifically designed for the study. Confidentiality of the patient's identity was preserved according to Spanish law for personal data protection.
Clinical data related to the study included starting date of OAT, indication for OAT, frequency scheduled by the physician for INR measurement, and target INR range and its changes during the follow-up period.
At the three follow-up visits, INR measurements and adverse events were collected. Adverse events included the presence of thromboembolism and its seriousness (major: deep thrombosis or non-transient ischemia; minor: surface thrombosis or transient ischemia ) as well as the presence of bleeding/hemorrhage and its seriousness (major: fatal and/or symptomatic bleeding into a critical area or organ, and/or bleeding that causes a decrease in hemoglobin ≥20 g/L, or which requires the transfusion of two or more units of whole blood or red blood cells, and/or gastrointestinal, urogenital, hemoptysis; minor: other not included in the described concepts).

Time in Therapeutic Range (TTR) Assessments
For the purpose of this study, two therapeutic INR target ranges were defined: (1) INR 2-3 for OAT indications of deep vein thrombosis/pulmonary embolism, atrial fibrillation, heart valve disease, and heart valve bioprosthesis, and (2) INR 2.5-3.5 for indications of mechanical prosthetic heart valve, and venous thrombosis or stroke repetition in the context of antiphospholipid syndrome [22,23].
The primary endpoint of the study, TTR, was calculated using the approach proposed by Rosendaal [4], in which it is assumed that the INR value between two consecutive determinations varies linearly and the value of INR estimated for each of the days between consecutive measurements at intervals of increase/decrease of INR is ≥0.1. The Rosendaal method INR-specific for person-time takes into account the frequency of INR determinations and their actual values. By considering the individual values of INR collected at each visit with respect to the previous visit and the time elapsed between the two determinations, the TTR was calculated as the percentage of days within the therapeutic range out of the total days of treatment.

Statistical Analysis
Population sample size was calculated by taking into consideration that the estimated average TTR is approximately 65% with a standard deviation (SD) of 20. Calculations showed that 174 patients (87 patients in each group) were needed to detect a difference of 8.5% in TTR between the two groups of OAT control (to meet the 5-10% difference between methods as recommended in the literature [6][7][8]) with a power of 80% and a significance value of P=.05. The analysis was stratified according to the type of OAT management to allow for the comparison between groups.
Continuous variables are presented as the mean (SD) or as median and range (minimum-maximum

Results
A total of 175 patients who met the inclusion criteria were enrolled in the study (88 in the conventional OAT control group and 87 in the telecontrol group). Two patients died before 3 months of follow-up: one in the conventional control group (an 84-year old female with obesity, diabetes, dyslipidemia, hypertension, and heart failure; death after aortic valve prosthesis implant) and one in the telecontrol group (a 57-year old male with obesity, dyslipidemia, hypothyroidism, heart failure, smoking habit, and chronic obstructive pulmonary disease; death from sudden cardiac arrest with anoxic encephalopathy).
At the end of the study, 173 patients were evaluated: 87 in the conventional control group and 86 in the telecontrol group. Follow-up time was a median of 6.3 (range 5.2-8.1) months with 6.4 (range 5.7-8.1) months in the conventional control group and 6.2 (range 5.2-7.1) months in the telecontrol group. Figure 2 shows the flow of patients through the study.  Table 1. Variables in both treatment groups were balanced with the exception of sex (more males in the telecontrol group) and family environment (more patients living alone in the conventional control group).
The average time of OAT treatment prior to enrollment in the study was 9.2 (SD 6.4) years (8.1 years in the conventional control group and 10.2 years in the telecontrol group). Most patients were in the range of 5 to <10 years of treatment (31%, 27/87 in the conventional control group and 42%, 36/86 in the telecontrol group), followed by the range of 10 to <20 years (20%, 17/87 in the conventional control group and 33%, 28/86 in the telecontrol group), and the range of 1 to 5 years (31%, 27/87 in the conventional control group and 15%, 13/86 in the telecontrol group). A total of 14.4%, 25/173 of patients were <1 year or ≥20 years in treatment.

OAT Control Effectiveness
The mean TTR of INR-the primary endpoint of the study-was 8% higher (95% CI: 1.82-13.86) in the telecontrol group than in the conventional control group (62%, SD 21% vs 54%, SD 19%, respectively; P=.01). This difference represents a relative increase of 12.6%. The distribution of patients according to ranges of TTR is shown in Table 3. In absolute terms, TTR in the telecontrol group was 107 (SD 37) days while in the conventional control group TTR was 94 (SD 37) days (P=.02).
According to visits, TTR in the telecontrol group was always higher than in the conventional control group-the maximal difference being observed at the third visit (59% vs. 48%, respectively; P=.01).  Figure 3 illustrates the evolution of the percentage of time in under-and over-anticoagulation during the 6-month follow-up. The mean percentage of time in which INR was <1.5 or >5 was 1.1% (SD 3.4%) and 1.4% (SD 3.6%), respectively. Table 3. Distribution of patients of the two oral anticoagulant therapy (OAT) treatment groups (conventional control and telecontrol with the SintromacWeb system) according to ranges of percentage of time in therapeutic range (TTR) during the follow-up period.

OAT Control Safety
A total of 19 thrombotic and/or bleeding events during the follow-up period occurred in 13% (11/87) patients of the conventional control group, but only in 4% (3/86) patients of the telecontrol group (P=.03). The distribution of the events according to seriousness is shown in Table 4. Globally, 7 events were serious, 12 were non-serious, and most of them (5 and 10, respectively) occurred in the conventional control group.
The most common thrombotic event was transient ischemic attack (2 episodes), while there were single episodes of pulmonary thromboembolism, deep vein thrombosis, superficial venous thrombosis, and ischemic heart disease. Conversely, unspecified minor bleeding (9 episodes) and unspecified major bleeding (2 episodes) were the most common bleeding/hemorrhagic events. There was also one episode of gastrointestinal bleeding and one episode of urogenital bleeding.

Principal Findings
Vitamin K antagonists are still the anticoagulant of choice for many patients, even in the face of the pharmacologic advantages shown by new oral anticoagulants targeting either thrombin or factor Xa. The higher drug cost of new oral anticoagulants often restricts their use to patients with risk of failure to maintain the INR in the therapeutic range (eg, with atrial fibrillation at risk for stroke). In addition, vitamin K antagonists are preferred for patients with uncertainties for dosing, such as those with renal dysfunction and those in marked extremes of body weight [24]. Therefore, the development of OAT self-testing and self-management systems to improve patient's quality of life is warranted [6][7][8][9][14][15][16][17][18][19][20][25][26][27]. However, widespread implementation of computer-based, Internet-assisted systems within health care centers has not happened. This is the first study in which a previously described, validated, fully operative and currently in use telecontrol system in a hospital for OAT management at home (SintromacWeb) has been assessed in a prospective trial. Our results indicate that OAT management with the SintromacWeb is effective and safe in terms of clinical practice.
The observational nature of this study can be considered a limitation when interpreting the results. Investigators did not modify their usual clinical practice. In addition, since patients in the telecontrol group were recruited under defined clinical criteria as described in the previous study [21], some differences between the treatment groups were not unexpected. Two patient demographic variables were dissimilar between groups: the proportion of male/female and the proportion of patients living alone/accompanied. This is possibly linked to the fact that patients using the telecontrol tool were required to have a computer and a connection to the Internet at home [21]. The aim of the study was not to determine which OAT management system was superior, but to determine whether the telecontrol system is effective and safe in those patients who are able to use it based on clinical criteria in reference to patients using the conventional system, regardless of the patient profile in each group. In addition, the analysis was stratified to allow the comparison between groups.
Overall, the primary characteristics of the patients in our study were typical as a target of OAT treatment and within the clinic population values and ranges shown by other trials, such as mean age (64 years; range 57-75 in other trials) and percentage of males (50%; range 43-72% in other trials) [8,[14][15][16][17][18][19][20]. The follow-up period of 6 months is also similar to most of those trials.
Patients in the telecontrol group achieved a relative 12.6% greater TTR value than patients in the conventional control group. Moreover, longer TTR in the telecontrol patients was consistently observed in all the scheduled visits during the study follow-up. Therefore, patients using the telecontrol system can be considered to have a good control of their OAT management. In some randomized trials, INR self-testing has been reported to be as good as [14,16,26] or better [8,15,[17][18][19][20]25,27] than conventional testing. In particular, in a study performed with an Internet-based system for the supervised remote management of patients on OAT, Ryan et al indicated a TTR improvement of 15.4% in patients on INR self-testing [8]. However, since our study was observational, any comparison must be viewed with caution.
In our study, the frequency of INR measurements was higher in patients of the telecontrol group than in patients of the conventional control group (around 3 weeks vs 5 weeks, respectively). Although there is no consensus between the major guidelines on the optimal frequency of INR testing to achieve good INR control, it has been described that self-testing patients have increased frequency of INR measurements [28,29], which can be related to an improvement of the clinical outcomes of OAT.
In contrast with another trial [8], our study showed that both under-anticoagulation (INR <1.5) and over-anticoagulation (INR >5.0) occurred equally frequently in both the telecontrol and the conventional control groups. Our study, however, was not powered to detect differences under this approach.
Safety results indicated that patients in the telecontrol group showed significantly fewer adverse events than patients in the conventional control group. Similar results have been described in other trials [15,17,18], and other studies have demonstrated that the number of complications increases in parallel with the time that patients are outside therapeutic INR target range and with the occurrence of serious under-and over-anticoagulation [13,30,31]. Nevertheless, some other trials have shown that OAT self-control is as safe as conventional control [8,14,16,19,20].

Conclusion
In conclusion, results of this study indicate that OAT management with the Internet-based telecontrol tool SintromacWeb is an effective and safe management system for those patients who are able to use it, based on clinical criteria. SintromacWeb is a system that has been previously validated, and it is fully operative and currently in use in a hospital, which adds value to the applicability of the study results to OAT patients eligible for telecontrol management.