Comparison of prothrombin time and international normalized ratio values using point-of-care system with a standardized laboratory method in patients on long-term oral anticoagulation – A prospective study

Raghuram Palaparti; Gopala Krishna Koduru; Sudarshan Palaparthi; PS S Chowdary; Purnachandra Rao Kondru; Somasekhar Ghanta; Mannuva Boochi Babu; Prasad Maganti; Sasidhar Yendapalli; KV N Srikanth; Rohini Medasani

doi:10.4103/JCPC.JCPC_55_19

ORIGINAL ARTICLE

Year : 2020 | Volume : 9 | Issue : 1 | Page : 25-30

Comparison of prothrombin time and international normalized ratio values using point-of-care system with a standardized laboratory method in patients on long-term oral anticoagulation – A prospective study

Raghuram Palaparti MD, DM ¹, Gopala Krishna Koduru MD, DM ¹, Sudarshan Palaparthi MD, DM ¹, PS S Chowdary MD, DM ¹, Purnachandra Rao Kondru MD, DM ¹, Somasekhar Ghanta MD, DM ¹, Mannuva Boochi Babu MD, DM ¹, Prasad Maganti MBBS PGDCC ¹, Sasidhar Yendapalli MBBS PGDCC ¹, KV N Srikanth MD ², Rohini Medasani MSc ¹
¹ Department of Cardiology, Aayush Hospital, Vijayawada, Andhra Pradesh, India
² Department of Pathology, Aayush Hospital, Vijayawada, Andhra Pradesh, India

Date of Submission	02-Dec-2019
Date of Acceptance	21-Jan-2020
Date of Web Publication	16-Mar-2020

Correspondence Address:
Dr. Raghuram Palaparti
Department of Cardiology, Aayush Hospitals, Ramachandra Nagar, Vijayawada - 520 008, Andhra Pradesh
India

Source of Support: None, Conflict of Interest: None

Check

DOI: 10.4103/JCPC.JCPC_55_19

Abstract

Background: In our country, methods of prothrombin time (PT) and international normalized ratio (INR) testing have not been standardized across various centers. Often, we find disparity in INR values from different laboratories, posing a challenge to make an appropriate clinical decision. Objective: The objective of the study was to compare and correlate the PT/INR values using a point-of-care (POC) system with standardized laboratory (SL) testing and to evaluate the efficiency of the POC system in monitoring patients. Methods: We have prospectively compared PT/INR values between a commercially available POC, CoaguChek XS System, Roche, and SL testing in 205 patients and 353 samples from July 2017 to April 2018. Results: The overall strong correlation between POC PT/INR values and simultaneous standard laboratory values was noted. The overall coefficient of correlation among the two groups was 0.919 (P = 0.001) for INR values. In INR range of 2–3.5, the values correlated well with a coefficient of correlation, 0.756 (P = 0.001). In INR range <2, the values correlated with a coefficient of 0.98. The correlation was poor when the INR values were >4. Conclusions: In this comparison study, statistical analysis yielded a good correlation between POC PT/INR values and SL values in therapeutic and subtherapeutic range. POC testing is a good alternative to SL testing. If widely available, POC testing may enable patient self-testing and self-monitoring. In our experience, POC testing had added benefits in emergency settings. However, clinicians and laboratory professionals should be aware of the occasional disagreement between POC INR and standard laboratory INR values.

Keywords: Long-term oral anticoagulation, point-of-care testing, prothrombin time/international normalized ratio values, standard laboratory testing

How to cite this article:
Palaparti R, Koduru GK, Palaparthi S, S Chowdary P S, Kondru PR, Ghanta S, Babu MB, Maganti P, Yendapalli S, N Srikanth K V, Medasani R. Comparison of prothrombin time and international normalized ratio values using point-of-care system with a standardized laboratory method in patients on long-term oral anticoagulation – A prospective study. J Clin Prev Cardiol 2020;9:25-30

How to cite this URL:
Palaparti R, Koduru GK, Palaparthi S, S Chowdary P S, Kondru PR, Ghanta S, Babu MB, Maganti P, Yendapalli S, N Srikanth K V, Medasani R. Comparison of prothrombin time and international normalized ratio values using point-of-care system with a standardized laboratory method in patients on long-term oral anticoagulation – A prospective study. J Clin Prev Cardiol [serial online] 2020 [cited 2023 Jun 9];9:25-30. Available from: https://www.jcpconline.org/text.asp?2020/9/1/25/280820

Introduction

Physicians worldwide prescribe Vitamin K antagonists (VKA) for chronic anticoagulant therapy for a number of clinical indications. However, numerous patient-specific factors influence warfarin sensitivity. These include age, body mass, nutritional status, hepatic function, and genetic variations in the cytochrome P450 complex and Vitamin K epoxide reductase complex 1.^[1] In addition to these patient-specific factors, warfarin has a narrow therapeutic window that requires frequent international normalized ratio (INR) monitoring to minimize hemorrhagic and thromboembolic complications.

Point-of-care (POC) testing of the INR method was developed to standardize prothrombin time (PT) results between different PT analyzers and thromboplastin reagents.^[1] Over the past two decades, different manufacturers have developed several POC INR devices that use finger-stick samples of the whole blood to measure the clotting time. These instruments then estimate INR using conversion formulas derived by the device manufacturers from the comparison of clotting times to a reference PT assay. Although commercially available, these devices are being less commonly utilized for various reasons. We sought out to compare and correlate the PT/INR values using POC system with standardized laboratory (SL) testing and evaluate the efficiency of POC system for PT/INR monitoring in patients on oral anticoagulant therapy.

Methods

The study has been conducted at a tertiary care hospital after institutional ethics committee approval. Informed and written consent has been obtained from all of the patients. We tested PT/INR in 205 consecutive patients on oral anticoagulation (OAC) visiting the cardiology outpatient department and 353 samples from July 2017 to April 2018 (over 10-month period) in a prospective manner. A capillary blood sample was tested using the POC CoaguChek XS system, Roche, and a simultaneous venous sample was tested using our SL method which served as a control. In disagreement of POC and SL values, we used laboratory values for clinical decision making. We included consecutive patients on chronic VKA therapy visiting the cardiology OPD at our institute for routine PT/INR monitoring. Patients who were unable to or refused to give consent and critically ill patients were excluded from the study. Patients on novel oral anticoagulants (NOACs) for nonvalvular atrial fibrillation (AF) were also excluded.

Laboratory testing was done using a commercially available HemosILTM system which uses a very high-sensitivity calcium thromboplastin for simultaneous determinations of PT and fibrinogen for evaluation of the extrinsic coagulation pathway and monitoring OAC therapy in human citrated plasma on the IL coagulation systems. This system uses a reagent PT-FiB HS PLUS, which is a lyophilized rabbit brain extract with an optimal concentration of calcium ions.^[2] This product is manufactured with high sensitivity to Factors II, V, VII, and X, comparable to international reference preparations making it particularly suitable for monitoring VKA therapy. Our SL system had both internal and external quality controls. The internal quality control was run every day (10 samples before usage) and external quality control periodically (quarterly) from an international reference laboratory (the Christian Medical College, Vellore).

The CoaguChek XS System [Figure 1]a and [Figure 1]b is a commercially available third-generation handheld portable monitor used for PT testing.^[3],[4] The results were provided in PT (in seconds) and INR in units. This POC system is intended for use by health-care professionals to evaluate the PT/INR of individuals using oral VKA therapy. The system uses a lot-specific code chip and takes a single drop of capillary or venous whole blood. It displays the result in approximately 1 min and can be configured to display results in INR, seconds, or %Quick. The CoaguChek XS PT test strips are manufactured with a human recombinant tissue factor and have an international sensitivity index (ISI) of 1.0. The handheld system weighs 175 g and fits the size of the palm. The measuring range of this POC system for INR is 0.8–8.0. The strips have a long shelf life of approximately 18 months. The method of testing is heparin insensitive. The test strip contains reagent in dried form. Reactive components of the thromboplastin reagent are human recombinant tissue factor, phospholipids, and a peptide substrate, Electrocyme TH, which can be used for the determination of serine proteases such as thrombin. When a sample is applied, the tissue factor activates the coagulation cascade leading to the formation of thrombin. The enzyme thrombin cleaves Electrocyme TH into a residual peptide and electrochemically active phenylenediamine, thereby generating an electrical signal, which is registered. The system algorithms analyze the signal, use the conversion formulae, and display the result on monitor usually within 30 s.

Figure 1: (a) Image showing Coaguchek XS system with strips. (b) Image showing point-of-care testing with the system

Click here to view

POC testing and SL testing were done at different sites within the hospital with POC testing done usually in the physician chamber itself. Sampling was done by different groups of health-care professionals who were completely blind to each other's results. Other important details such as demographic characteristics; indication for OAC therapy; details of underlying heart disease; details of comorbidities such as diabetes, hypertension, smoking habit, dyslipidemia, and history of MI; type and duration of AF; the presence of prosthetic valves; CHA₂ DS₂-VASc score, HAS-BLED score; and medication being used were recorded.

Statistical methods

Statistical analysis was done using IBM SPSS Statistics, Version 25.0. Armonk, NY: IBM Corp., and the significance level was set as P <.05 in all statistical analyses. The variables were investigated with the use of analytical methods (Kolmogorov–Smirnov/Shapiro–Wilk tests) to determine whether they were normally distributed. Descriptive statistics were reported as mean with standard deviation for normally distributed continuous variables and as the frequency with percentages for the categorical variables. Various tests used were independent t-test, paired t-test, and Pearson's correlation test. The correlation between methods was assessed using the Pearson correlation test. Scatter correlation plots were used to depict the agreement of the PT/INR results between CoaguChek XS and HemisIL systems and between various subgroups of the study population. Subset analysis was performed in the following groups (i) between males and females, (ii) age ≤65 and >65 years, (iii) CHA₂ DS₂-VASc score >2 and <2, (iv) between patients with nonvalvular AF and valvular AF with or without prosthetic valves, the overall correlation and difference were also compared in three INR ranges (INR <2.0, INR 2–3.5, and INR >3.5), and (v) INR range groups <2, 2–3.5, and >3.5. Pearson correlation coefficient (r) was used for assessing the degree of agreement between methods. We considered values of <0.65 as poor correlation, 0.65–0.8 as good correlation, and values >0.8 as excellent correlation for this study purpose.

Results

The study population constituted 96 males (47%) and 109 female patients. The mean age of the population was 57.8 ± 26 years. One hundred forty-three patients (70%) were of age ≤65 years. The remaining 62 patients (30%) were of age ≥66 years. The most common indications for VKA therapy in our study population were pure AF and prosthetic valves with or without AF. The rest of the indications comprised pulmonary thromboembolism, deep-vein thrombosis, severe idiopathic pulmonary arterial hypertension, and severe left ventricular dysfunction [Figure 2]. Many of the patients underwent sampling more than once during their subsequent outpatient visits. Hence, the total number of samples was 353.

Figure 2: Pie chart showing various clinical indications for Vitamin K antagonists therapy in the study population

Click here to view

Overall, there was a strong correlation between POC PT and INR values when compared to simultaneous values of the standard laboratory. The coefficient of correlation for INR values was 0.919 with P < 0.001 and the coefficient of correlation for PT values was 0.941 with P < 0.001 [Figure 3]a and [Figure 3]b.

Figure 3: (a) Correlation plot of international normalized ratio values-point of care versus standardized laboratory for the study population. (b) Correlation of prothrombin time values-point of care versus standardized laboratory values for the study population

Click here to view

Correlation between males and females

In both male and female patients, the POC PT/INR values correlated strongly with the standard laboratory values with correlation coefficients 0.910 and 0.922, respectively (P < 0.001). There was no statistically significant difference among both the groups [Figure 4]a and [Figure 4]b.

Figure 4: (a) Correlation of point of care versus standardized laboratory values in males. (b) Correlation of point of care versus standardized laboratory values in females

Click here to view

Correlation between patients with age ≤65 and ≥66

The correlation was strong in both the age groups ≤65 and ≥65 with P < 0.001. The correlation coefficients for males and females were 0.902 and 0.916, respectively. There was no significant statistical difference among the two groups [Figure 5]a and [Figure 5]b.

Figure 5: (a) Correlation in patients with age <65 years (b) Correlation in patients with age more than 65 years

Click here to view

Correlation between patients with nonvalvular atrial fibrillation and valvular atrial fibrillation with or without prosthetic heart valves

In patients with nonvalvular AF, POC INR correlated significantly with laboratory INR with a coefficient of correlation 0.912 (P < 0.001). In patients with valvular AF with or without prosthetic heart valves, the POC INR correlated significantly with laboratory INR values with a coefficient of correlation 0.932 (P < 0.001). There was no significant statistical difference among the two groups [Figure 6]a and [Figure 6]b.

Figure 6: (a) Correlation in patients with nonvalvular atrial fibrillation. (b) Correlation in patients with valvular atrial fibrillation with or without prosthetic heart valves

Click here to view

Correlation between patients with CHA₂ DS₂-VASc score ≤2 and ≥3

In both the groups of patients with CHA₂ DS₂-VASc score ≤2 and ≥3, POC PT/INR values correlated significantly with laboratory values (P < 0.001), coefficient of correlation being 0.909 for patients with score ≤2 and 0.915 for patients with score ≥3. There was no significant statistical difference among the two groups [Figure 7]a and [Figure 7]b.

Figure 7: (a) Correlation in patients with CHA2DS2Vasc score ≤2. (b) Correlation in patients with CHA2DS2Vasc score ≥3

Click here to view

Correlation in different ranges of international normalized ratio

In patients with INR range of 2–3.5, the values correlated well and the coefficient of correlation was 0.756, with P < 0.05. In patients with INR <2, the values correlated extremely well with the coefficient of 0.98 with P < 0.001. The correlation was poor when the INR values were more than 3.5 with a correlation coefficient of 0.65 with a P = 0.07 [Figure 8]a and [Figure 8]b].

Figure 8: (a) Correlation in patients with international normalized ratio values between 2 and 3.5. (b) Correlation in patients with international normalized ratio <2 and >3.5

Click here to view

We did not have any patient samples with INR more than 10 in standard laboratory testing results. Four samples were detected to have INR value >8 in POC testing and showed as ”high.” All these four patients had INR values between 8 and 10 in the standard laboratory testing.

Discussion

Even with the advent of NOACs and their proven superiority or noninferiority over VKA therapy with respect to clinical efficacy or bleeding complications in most of the clinical scenarios, VKA therapy still may have to be used in many patients, particularly in patients with prosthetic heart valves and valvular AF. Valvular AF is still more common than nonvalvular AF in India, unlike the West, primarily due to the burden of rheumatic heart disease. The higher cost of NOAC agents also limits their usage in India. One of the major disadvantages of VKA therapy is the requirement for periodic anticoagulant monitoring.

The standard test for the monitoring of oral VKA therapy is the plasma-based PT test. POC testing of PT/INR is a relatively new method of anticoagulation monitoring available in India for patients on VKA therapy. It has not been validated in real-world settings in our population, particularly cardiac patients, though several POC machines are commercially available. The disparity in the INR values from different laboratories without proper standardization poses a major challenge in making timely clinical decisions. An easier, reliable, and cost-effective method of INR monitoring is desirable in low-resource settings like in our country. Moreover, rheumatic heart disease in India is still predominantly a disease of younger and lower socioeconomic population, making meticulous INR monitoring in these patients with prosthetic heart valves much more difficult. These patients are also more likely to have poor access to optimal health-care monitoring. This problem is magnified in patients who are either transportation challenged or homebound.

Potential advantages of POC INR testing include rapid turnaround times (usually within seconds) and ease of out-of-hospital testing. These qualities greatly help in the INR monitoring of homebound patients and in those for whom drawing venous blood samples is difficult. Furthermore, POC INR devices enable some patients to self-test and self-manage their warfarin therapy with favorable outcomes.^[5] Several anticoagulation management services, mainly in the West, also have reported that POC INR devices improve clinical efficiency and patient comfort and satisfaction compared to traditional venipuncture methods.^[6] Despite these benefits, POC INR testing has several disadvantages. Severe anemia or polycythemia (hematocrit below approximately 25% or above about 55%), coadministration of other OAC with warfarin (e.g., low molecular-weight heparin), fibrinogen level, and antiphospholipid inhibitors interfere with the results.

The CoaguChek XS system evaluated in the present study is suited for both health-care professional and patient use. Our study demonstrated a good correlation of the CoaguChek XS system using finger-stick whole blood specimens with an SL plasma test as in several independent studies.^{[7],[8],[9],[10],[11],[12]} Lakshmy and Kumar from New Delhi, India, studied PT/INR values in 42 patients and concluded that POC systems are reliable and accurate and can be used by patients for monitoring of OAC therapy.^[13] To the best of our knowledge, our study is the largest in India which validated the POC INR values with SL INR values in patients on long-term VKA therapy.

Proper INR monitoring is mandatory in patients on long-term oral VKA therapy to maintain appropriate anticoagulation levels specific for each patient's condition, thereby reducing the incidence of complications. If an abnormal value is identified on routine monitoring, appropriate therapeutic intervention may prevent untoward events. Our study showed that POC testing accurately assesses the therapeutic status and is an effective method of anticoagulation monitoring. However, careful examination of the correlation plots shows that there are many patients in out-of-range INR where there is a disparity between the POC values and SL values. The correlation was poor when the INR values are >3.5, and similar findings have been noted in various previous studies. Even in the absence of the interfering conditions described above, sometimes significant disparities occur between SL INR and POC INR values, especially when INR levels are >3.0.^{[14],[15],[16]} Different ISI values to thromboplastin reagents, the positive bias up to 0.5 INR units in the POC systems and different reagents and instrument combinations have been attributed as the causes for this disagreement in a previous study.^[17] Awareness of this disagreement is important for physicians to further confirm the findings and ensure appropriate clinical intervention. Most of the current generation POC INR devices use either an electrochemical sensor or a mechanical method to determine the test endpoint, thereby estimating the INR.^[18] The system we used in our study estimates INR by the electrochemical sensor and supposedly more accurate than mechanical methods.

Using controlled and standardized POC measurements could lead to better anticoagulation management and patient outcome. POC testing eliminates variables associated with delayed sample transportation and specimen handling also. In the physician's office or clinic, a POC system for immediate dosage adjustment and patient counseling is possible, without the delay associated with routine laboratory testing. Furthermore, in some clinical scenarios where rapid decision-making is of utmost importance, like for example, in a patient with acute ischemic stroke on OAC therapy and in window period, eligible for thrombolytic therapy, POC testing gives immediate results and rules out out-of-range INR. However, we recommend validating POC PT/INR with hospitals SL before initiating routine clinical use.

Although direct costs may be higher for POC PT testing, it is cost-effective, if we consider indirect expenses of hospital staff, resource utilization and management of adverse events. Indirect and total costs are better controlled under a comprehensive anticoagulation management program.^[19] More frequent patient self-monitoring has been shown to help patients with labile INR and reduce unnecessary dosage adjustments and complications.

Limitations of the study

We did not assess time in therapeutic range for our patients, as serial INR data of all the patients were not available, however, that is not the objective of the present study. We also did not look for various conditions interfering with POC PT/INR testing in our study population comprehensively. The upper limit of detection of INR values in the POC system used in our study is eight, and patients with values >8 will be shown as ”High.” Fortunately, the number of patients with high INR values in our study population was quite low. Moreover, in these patients, INR values were confirmed with the SL method for clinical decision making. Our study was not powered enough to detect the impact of disagreement between the POC and SL INR values on the clinical outcomes.

Conclusions

In this largest comparison study from India, we found a good correlation between POC PT/INR values and SL values. This real-world patient data and various other previous studies findings support the fact that POC PT/INR testing with the currently existing machines is a viable alternative to drawing a venous sample and sending it to the laboratory, in terms of both accuracy and cost-effectiveness. The other greatest advantage of POC testing, we feel, is its ability to give rapid and reliable results in the setting of emergencies and allows us to decide whether the patient is in range INR or to rule out ”out-of-range” INR. POC INR testing for chronic anticoagulation monitoring when widely available in the country may enable patient self-testing and self-monitoring. It is also useful in patients with labile INR for frequent out-of-hospital monitoring. However, physicians should be aware that the potential disparity may occur between POC and SL at higher INR values and work toward developing institutional protocols for confirmatory testing. With the wide usage of various POC systems, standardization across different manufacturers and reagents can be anticipated for more accurate results.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Pendleton RC, Rodgers GM. Disorders of hemostasis and coagulation. In: Greer JP, Arber DA, Glader B, List AF, Means RT, Paraskevas F, et al., editors. Wintrobe's Clinical Hematology. 13^th ed. Philadelphia: Lippincott Williams & Wilkens; 2014. p. 1241-3.
2.	Miale JB, Kent JW. Standardization of the therapeutic range for oral anticoagulants based on standard reference plasmas. Am J Clin Pathol 1972;57:80-8.
3.	Bereznicki LR, Jackson SL, Peterson GM, Jeffrey EC, Marsden KA, Jupe DM. Accuracy and clinical utility of the CoaguChek XS portable international normalised ratio monitor in a pilot study of warfarin home-monitoring. J Clin Pathol 2007;60:311-4.
4.	Kalçık M, Yesin M, Gürsoy MO, Gündüz S, Karakoyun S, Astarcıoǧlu MA, et al. Comparison of the INR values measured by CoaguChek XS coagulometer and conventional laboratory methods in patients on VKA therapy. Clin Appl Thromb Hemost 2017;23:187-94.
5.	Heneghan C, Ward A, Perera R, Self-Monitoring Trialist Collaboration, Bankhead C, Fuller A, et al. Self-monitoring of oral anticoagulation: Systematic review and meta-analysis of individual patient data. Lancet 2012;379:322-34.
6.	Chaudhry R, Scheitel SM, Stroebel RJ, Santrach PJ, Dupras DM, Tangalos EG. Patient satisfaction with point-of-care international normalized ratio testing and counseling in a community internal medicine practice. Manag Care Interface 2004;17:44-6.
7.	Giles TD, Roffidal L. Results of the prothrombin office-testing benefit evaluation (PROBE). Cardiovas Rev Rep 2002;23:27-33.
8.	Bussey HI, Chiquette E, Bianco TM, Lowder-Bender K, Kraynak MA, Linn WD, et al. A statistical and clinical evaluation of fingerstick and routine laboratory prothrombin time measurements. Pharmacotherapy 1997;17:861-6.
9.	Kaatz SS, White RH, Hill J, Mascha E, Humphries JE, Becker DM. Accuracy of laboratory and portable monitor international normalized ratio determinations. Comparison with a criterion standard. Arch Intern Med 1995;155:1861-7.
10.	Campbell PM, Radensky PW, Denham CR. Economic analysis of systematic anticoagulation management vs. routine medical care for patients on oral warfarin therapy. Dis Manage Clin Outcomes 2000;2:101-8.
11.	Jacobson AK, Guilloteau FR, Campbell PM, Denham CR. Comparison of point-of-care testing and standard reference laboratory testing for PT/INR measurements in patients receiving routine warfarin therapy: An engineering work process flow study. Dis Manag Clin Outcomes. 2000;2:109-16.
12.	Franke CA, Dickerson LM, Carek PJ. Improving anticoagulation therapy using point-of-care testing and a standardized protocol. Ann Fam Med 2008;6 Suppl 1:S28-32.
13.	Lakshmy R, Kumar AS. Comparative evaluation of point Of Care coagulation monitoring by coaguchek XS-comparison with standard laboratory method. Indian J Thorac Cardiovasc Surg 2010;26:125-8.
14.	Hur M, Kim H, Park CM, La Gioia A, Choi SG, Choi JH, Moon HW, Yun YM. Comparison of international normalized ratio measurement between CoaguChek XS Plus and STA-R coagulation analyzers. BioMed research international 2012;2013.
15.	Moore GW, Henley A, Cotton SS, Tugnait S, Rangarajan S. Clinically significant differences between point-of-care analysers and a standard analyser for monitoring the International normalized ratio in oral anticoagulant therapy: A multi-instrument evaluation in a hospital outpatient setting. Blood Coagul Fibrinolysis 2007;18:287-92.
16.	Moiz B, Rashid A, Hasan M, Jafri L, Raheem A. Prospective comparison of point-of-care device and standard analyzer for monitoring of international normalized ratio in outpatient oral anticoagulant clinic. Clin Appl Thromb Hemost 2018;24:1153-8.
17.	Gosselin R, Owings JT, White RH, Hutchinson R, Branch J, Mahackian K, et al. A comparison of point-of-care instruments designed for monitoring oral anticoagulation with standard laboratory methods. Thromb Haemost 2000;83:698-703.
18.	Donaldson M, Sullivan J, Norbeck A. Comparison of international normalized ratios provided by two point-of-care devices and laboratory-based venipuncture in a pharmacist-managed anticoagulation clinic. Am J Health Syst Pharm 2010;67:1616-22.
19.	Sharma P, Scotland G, Cruickshank M, Tassie E, Fraser C, Burton C, et al. The clinical effectiveness and cost-effectiveness of point-of-care tests (CoaguChek system, INRatio2 PT/INR monitor and ProTime Microcoagulation system) for the self-monitoring of the coagulation status of people receiving long-term vitamin K antagonist therapy, compared with standard UK practice: Systematic review and economic evaluation. Health Technol Assess 2015;19:1-72.