Comparison and Evaluation of Two Point-of-Care Coagulation Analyzers for Monitoring Warfarin Oral Anticoagulant Therapy

Introduction

Primary care, urgent care and other point-of-care (POC) locations, call for fast and dependable prothrombin time/international normalized ratio (PT/INR) test results in order to support the monitoring of oral anticoagulant therapy (OAT) with vitamin K antagonists such as warfarin.

Siemens Healthcare Diagnostics has developed Xprecia Stride™ Coagulation Analyzer* – an advanced, handheld POC device that is capable of producing fast and quantitative PT/INR test results from fresh fingerstick blood samples.

Performed under ICH-GCP guidelines, this external validation study compared the significant clinical equivalence of the Xprecia Stride analyzer PT/INR test with an extensively employed FDA-approved POC methodology (COAGUCHEK XS System, Roche Diagnostics).

Summary

Method comparison with Roche Diagnostics’ COAGUCHEK XS system helped to validate the performance of the Xprecia Stride analyzer in accordance with its proposed use with fingerstick blood samples across the 0.8–4.5 INR range.

PT/INR test results from Xprecia Stride analyzer showed good correlation with Roche Diagnostics’ COAGUCHEK XS system (r2 = 0.93). Bias was within the acceptable limits. Using an INR therapeutic range of 2.0–4.0, clinical concordance of results produced by both POC methods was 96.7%.

Repeatability of Xprecia Stride analyzer PT/INR results was found to be well within a coefficient of variation (%CV) of ≤10% – an industry-standard acceptance criterion – across the three reporting ranges (results yielding ≤5.8% CVs). Likewise, evaluation of two levels of liquid quality control (LQC) showed intermediate/reproducibility precision that was well within this same industry standard (results yielding within-laboratory CVs of ≤8.3%).

The growing demand for oral anticoagulant therapy (OAT)

Patients suffering from recurrent abnormal blood clotting or those at high risk of developing clots are prescribed OAT on a long-term basis. Atrial fibrillation (AF) is a common cardiac arrhythmia that affects over 2.6 million in the USA and 6 million people in Europe.1 Individuals suffering from this condition are at high risk for blood clots, which also include clots that cause ischemic stroke.

OAT is often prescribed to reduce the risk of stroke, but patients using these drugs have to be frequently monitored through blood tests.

Globally, over 800 million PT/INR tests are performed every year.2 An increasing population of patients on warfarin (COUMADIN) therapy and trend toward testing outside of the central laboratory have escalated the demand for PT/INR results at the point of care.

Xprecia Stride Analyzer: Clinical utility

The Xprecia Stride analyzer has been specifically developed for near-patient monitoring of PT/INR on fresh fingerstick blood samples. The handheld instrument is accurate, convenient, and user-friendly and comes with improved safety features that protect operators during the course of the testing process:

  • Quick, quantitative results across the 0.8–4.5 INR reportable range
  • User-friendly color touchscreen interface and vivid display of results as INR units
  • Onboard animated tutorial shows key functions and guides users during operation through step-by-step pictorial prompts
  • Small, 6 μL sample size
  • Built-in bar-code scanner to enable error-free data capture
  • Biohazard exposure is minimized with push-button ejection of used test strips
  • Secure, seamless bidirectional data transfer through USB connection

For more than three decades, Siemens Healthcare Diagnostics has been a leading company in laboratory hemostasis testing and now offers a POC system for warfarin monitoring. With the Xprecia Stride Coagulation Analyzer extends, the company has extended its hemostasis expertise into the POC market and provides customers with a wide range of hemostasis analyzers under one roof.

                   

Xprecia Stride Analyzer: Test technology

Using single-use reagent test strips and electrochemical technology, the Xprecia Stride analyzer measures the prothrombin time. Within the test strip, a sample chamber is filled with a fresh blood sample via capillary action.

The test strip contains dried reagents that mainly include an electroactive thrombin substrate called thromboplastin as well other reagents. An electroactive group is discharged from the thrombin substrate, which is electrochemically detected at the electrodes provided in the strip; coagulation time is determined by analyzing the generated current by an algorithm.

Two on-strip quality control checks are performed to examine each test strip when applying a sample to check for the presence of sufficient reagent volume and sample on the test strip and also for test strip degradation caused by exposure to environmental factors.

If both controls fail, the Xprecia Stride analyzer will report an error and cancel the test. Operators can also use the LQC material according to local, state, national or federal guidelines.

The diversity of coagulation testing platforms and reagents

Analyzers that are commercially available in the market are used to carry out PT/INR testing. These instruments use a wide range of different reagents such as Thromborel® S, Dade® Innovin®, STAGO STA NEOPLASTINE CI PLUS, Owren's PT (combined thromboplastin) and IL HemosiIL RecombiPlasTin. Variations in instruments, reagents and other variables will not only influence the PT/INR results but also increase the difficulty of method performance comparisons.

The study assessed results reported by platforms employing the same test reagent (Dade Innovin recombinant human tissue thromboplastin). This eliminates a possible area for variability when comparing both techniques.

The importance of clinical concordance

Reliable patient treatment decisions can be made based on the assurance of INR clinical concordance, especially when decisions depend on newly developed analytical methodology. Not all methods applied for measuring PT can create matching results. While the INR helps to standardize the interpretation of PT test results, differences still exist between the methods, and reference ranges can differ between sites and clinical contexts.

Through clinical concordance analysis, method agreement can be assessed with regards to patient management decisions and this proves useful when direct comparison of results is not possible.

The current study assessed the clinical concordance between PT/INR results to establish the utility of the Xprecia Stride analyzer for patient management as opposed to an FDA-cleared POC device.

Study Purpose

The PT/INR test performed on the Xprecia Stride Coagulation Analyzer is required to show considerable clinical equivalence to a proven POC methodology. Accurate and reliable results enable optimal usage of the analyzer for making assured patient treatment decision. The study assessed the Xprecia Stride analyzer PT/INR result:

  • Clinical concordance verus the COAGUCHEK XS system
  • Bias ad agreement verus the COAGUCHEK XS system
  • Repeatability (whole blood)
  • Predicted range for study subjects who are not on OAT
  • Intermediate precision (LQC level 1 and level 2)

Methods

General

Study subjects were enrolled at four clinical sites over a period of 11 months. These included both set of patients – those receiving warfarin therapy and those who were not on warfarin therapy. Two separate fingerstick blood samples were collected from the subjects.

The first drop of fingerstick blood was applied to the COAGUCHEK XS system within 15 seconds, and the second drop was applied to the Xprecia Stride analyzer within 15 seconds. At each site, three varying lots of Xprecia Stride analyzer test strips were employed in a randomized way.

All samples of blood were collected and processed in accordance with CLSI guideline H21-A5. Universal Biosensors Pty Ltd (Rowville, Victoria, Australia) performed the subsequent data analysis. The study was performed using analyzers, reagent test strips and liquid quality control materials developed on validated production lines.

Method comparison

Pooled results from the first drop of fingerstick blood from study subjects were used to determine agreement and bias in PT/INR measurement verus the COAGUCHEK XS system (n = 365). The results obtained were subsequently distributed across INRs of 0.9–4.5.

A Passing-Bablok regression analysis was performed using the results. Slope (95% CI), y-intercept (95% CI), correlation coefficient (r), as well as coefficient of determination (r²) were calculated. Passing-Bablok regression acceptance criteria were defined as:

  • Intercept: +0.5 to -0.5
  • Slope: 95% confidence interval within 0.80–1.20
  • Coefficient of determination (r2) ≥0.80

Bias of the Xprecia Stride technique was calculated at two medical decision points (INR = 4.5 and INR = 2.0, as shown in Table 1).

Expected range

The expected range was evaluated using the Xprecia Stride analyzer PT/INR results (n = 120) from samples collected from subjects not on OAT. The reference range was then reported as the upper and lower INR values covering 95% of results.

Repeatability

Repeatability was assessed by using the difference between the samples results. These sample pairs were collected from two individual fingersticks and run on the same analyzer. For repeatability data analysis, valid pairs of samples from at least 100 study subjects at each clinical site were applied. For each three INR ranges (<2.0, 2.0–3.0 and 3.1–4.5, shown in Table 3), mean INR, %CV and standard deviation were measured.

Intermediate precision

Qualified operators at each clinical site generated intermediate precision data by conducting tests on the Xprecia Stride analyzer using Xprecia Stride Coagulation Analyzer PT LQC levels 1 and 2 in duplicate for a 20 day period. Using three lots of PT liquid quality control kits and three lots of reagent test strips, testing was performed across the four clinical sites. As shown in Table 4, LQC levels 1 and 2 were measured from complete datasets obtained for each site’s analyzer.

Results

Method comparison

Shown in Figure 1 is the all-site Passing-Bablok regression fit for Xprecia Stride analyzer PT/INR results as opposed to the COAGUCHEK XS system. Regression statistics and calculated bias are shown in Table 1.

         

Figure 1. Xprecia Stride analyzer versus COAGUCHEK XS system PT/INR results: regression fit (red) and line of identity (gray).

Table 1. Xprecia Stride analyzer versus COAGUCHEK XS system method comparison regression statistics and calculated bias.

             

This study showed excellent correlation between the COAGUCHEK XS system method (r2 = 0.93) and the Xprecia Stride analyzer PT/INR results. The low bias of the test was also demonstrated in the study. Calculated bias was -0.1 INR and -0.3 INR at the medical decision points of 2.0 INR and 4.5 INR.

Clinical concordance

Table 2. Xprecia Stride analyzer versus COAGUCHEK XS system PT/INR result clinical concordance analysis.

               

  • Clinical concordance of results generated by the two POC methods was 96.7% using a 2.0–4.0 INR therapeutic range which is commonly accepted.
  • Clinical concordance analysis showed high correlation for Xprecia Stride analyzer result classification for most of the patients (INR within, above, or below reference range, as indicated in gray shading).
  • Clinical concordance of Xprecia Stride analyzer results over an extended INR therapeutic range of 2.0–4.5 (data not shown) was 95.9%.

Expected range

For fingerstick capillary blood samples on the Xprecia Stride analyzer, a 0.9–1.1 INR range covered 95% of results for individuals not on OAT.

Repeatability

Data analysis showed that repeatability CVs were ≤5.8% for all three INR ranges; these CVs were well below the industry-standard acceptance criterion of CV ≤10%.

Reproducibility/intermediate precision

Table 3. PT/INR results for the Xprecia Stride analyzer on paired samples of fingerstick blood across all four sites.

Table 4. Reproducibility/Intermediate precision of LQC testing for each study location.

             

Data analysis showed that intermediate precision CVs were ≤8.3% for all four sites; these CVs were well below the industry-standard acceptance criterion of CV ≤10%.

Overview and conclusions

The Xprecia Stride Coagulation Analyzer was therefore verified in accordance with its proposed use with fingerstick blood samples for clinical PT/INR results across the reportable INR range of 0.8–4.5. The analyzer met all acceptance criteria in an external study conforming to ICH-GCP guidelines:

  • Passing-Bablok regression analysis produced a slope of 0.93 and an intercept of 0.0, with a coefficient of determination (r2) of 0.93 across the 0.9–4.5 INR range in a demonstration of performance equivalency with an FDA-approved alternate POC methodology (COAGUCHEK XS system) using the Dade Innovin reagent.
  • At major medical decision points, bias versus the COAGUCHEK XS system was low (-0.3 at 4.5 INR and -0.1 at 2.0 INR).
  • Clinical concordance of Xprecia Stride analyzer PT/INR results verus the COAGUCHEK XS system was 97.0% using the 2.0–4.0 INR therapeutic range, which is commonly accepted.
  • Repeatability showed CVs ≤5.8% across the reportable INR range, which were well below the ≤10% CV acceptance criterion.3
  • Intermediate precision was ≤8.3% CV, also meeting the acceptance criterion of ≤10%.

The Xprecia Stride analyzer delivers similar performance to an FDA-approved PT/INR POC test. The Xprecia Stride analyzer test can be confidently used by trained healthcare professionals at the point of care to monitor patients on warfarin OAT. The efficiency, speed, simplicity and overall practicality of the Xprecia Stride analyzer in POC settings complement the consistent, lab-like performance of the Xprecia Stride analyzer.

*Product availability varies by country.

Evaluation of Precision Performance of Quantitative Measurement Methods (EP05-A2), Measurement Procedure Comparison and Bias Estimation Using Patient Samples (EP09-A3), and Defining,

Establishing, and Verifying Reference Intervals in the Clinical Laboratory (EP28-A3c).

Loma Linda VA Healthcare System, Loma Linda, CA 92357 USA (Investigators: Ronald Fernando, MD, and Alan K. Jacobson, MD) DCOL Center for Research, Longview, TX 75605 USA (Principal Investigator: Anita Scribner, MD)

Geisinger Clinic, Cardiovascular Center for Clinical Research, Danville, PA 17822 USA. Kentucky Clinical Trials Laboratory, Louisville, KY 40202 USA

Acknowledgements

Prodiced from materials originally authored by Fernando R,1 Jacobson AK,1 Kennedy S,2 Lessard C,2 Olson K,2 Scribner A.3

1Loma Linda VA Healthcare System, Loma Linda, CA.

2Siemens Healthcare Diagnostics Inc., Norwood, MA.

3DCOL Center for Research, Longview, TX.

References

  1. Kannel WB, Benjamin EJ. Status of the epidemiology of atrial fibrillation. Med Clin North Am. 2008;92:17-40.
  2. Jackson CM, Esnouf MP. Clin Chem. 2005;51(3):483-85.
  3. Clinical laboratory testing and in vitro medical devices—Requirements for in vitro monitoring systems for self-testing of oral anticoagulant therapy. ISO 17593:2007.

About Siemens Healthineers Point of Care Diagnostics

Point-of-care solutions are designed to provide immediate, convenient, and easy-to-use diagnostic testing. From the ED to the physician’s office, clinical management decisions can be made immediately and result in improved patient safety, clinical outcomes, and overall patient satisfaction.


Sponsored Content Policy: News-Medical.net publishes articles and related content that may be derived from sources where we have existing commercial relationships, provided such content adds value to the core editorial ethos of News-Medical.Net which is to educate and inform site visitors interested in medical research, science, medical devices and treatments.

Last updated: Jan 23, 2018 at 11:22 AM

Other White Papers by this Supplier