Comparison of a column agglutination technology-based automated immunohematology analyzer and a semiautomated system in pretransfusion testing.

INTRODUCTION: Semi-automated equipment using Column agglutination technology (CAT) is widely used where centrifugation and incubation are automated but substantial amount of the work is still executed manually. Larger laboratories are moving towards automation to eliminate errors, reducing exposure to bio-hazardous samples, assuring traceability, reliability, turnaround time (TAT) and throughput. Moving towards automation and greater reliability, we therefore, decided to install an automated immunohematological (IH) analyzer "Vision". In this study we evaluated reliability and performance before clearing "Vision" for routine use.
MATERIALS AND METHODS: Study was conducted in the Department of Transfusion-Medicine. The primary objective was to assess the reliability of results and compare with routine use semi-automated BIOVUE system (Reference system). Secondary objective was to evaluate the performance (TAT and throughput) of the Vision to handle routine and emergency workload.
RESULTS: Total of 1276 known samples were used to assess 2640 pre-transfusion tests (1229 ABO/Rh D typing; 1229 antibody screening; 54 antibody identification; 86 crossmatch and 42 DAT). All 1229 ABO Rh typing results were concordant between the two systems. Overall agreement between the Vision IH analyzer and reference system for ABO Rh typing was 99.95%. All antibody screening, crossmatch and DAT results were concordant between the two systems. TAT of Vision was substantially shorter than the reference system for all test profiles.
CONCLUSION: Based on study results, Vision was approved for routine use in laboratory. It was found to be reliable with considerably shorter TAT and capable of handling high throughput of immunohematological tests. Copyright:

Introduction

Column agglutination technology (CAT) has gradually replaced conventional test tube (CTT) in several blood transfusion services in India, especially in metropolitan cities. The introduction of CAT has changed and simplified the practices for pretransfusion laboratories around the world including India. The cassette used in CAT represents one of the most sensitive technologies for antibody detection and identification.[1] These cassettes act as miniature test tube that can be used for all pretransfusion test procedures using different profiles (blood grouping, direct antiglobulin test [DAT], etc.).

Semiautomated equipment using CAT is widely used where centrifugation and incubation are automated, but substantial amount of the work (sample preparation, dispensing, and interpretation) during pretransfusion testing is still executed manually. Results of the interoperator variability are reduced to great extent as compared to CTT, but results of the tests run in CAT may still be interpreted differently by different operators. An error due to manual intervention can potentially result even in a fatal outcome in the transfusion process. In today's world, therefore, larger laboratories are moving toward automation in pretransfusion testing with an aim to minimize or eliminate the risk of errors from manual manipulations, reduce the staff exposure to bio-hazardous samples, assure the traceability of the process, and improve the reliability, turnaround time (TAT), and throughput. On moving toward automation and greater reliability, we therefore decided to install a fully automated immunohematological (IH) analyzer “Vision” (Ortho Clinical Diagnostics, USA). It is a new fully automated IH analyzer developed to meet the needs of IH laboratories performing various steps of pretransfusion testing.

In this study, we evaluated the reliability and performance of new IH analyzer in a preplanned systemic manner[2] before clearing “Vision” for routine laboratory use.

Materials and Methods

Setting

This study was conducted at the Department of Transfusion Medicine in a large tertiary care hospital in National Capital Region of India. Ours is a hospital-based transfusion facility center collecting around 30,000 whole blood donations annually. We have an annual workload of around 90,000 pretransfusion testing samples, mostly blood grouping and antibody screening and crossmatches; few anti-globulin investigations (DAT/indirect antiglobulin test [IAT]), ABO and Rh antibody titrations; and unexpected antibody identification. To handle this large amount of IH work, evaluating the efficiency of newly introduced “Vision” was considered.

Aim

The primary objective of the study was to assess the reliability of tests results and compare with routinely used semiautomated BioVue system in our laboratory. The secondary objective was to evaluate the performance (TAT and throughput) of the IH analyzer (Vision) to handle routine and emergency workload.

Samples used for testing

Patient and donor samples were used for testing. Samples were either ethylenediaminetetraacetic acid (patient/donor samples) or CPD-SAGM (Citrate Phosphate Dextrose- Saline Adenine Glucose Mannitol)-based anticoagulant obtained from donor blood bag (donor samples). All samples were run in parallel with BioVue system.

Analyzer

Vision is a fully automated IH analyzer designed to use six-column CAT cassettes. Analyzer has high throughput and is capable of managing all IH tests ranging from blood typing and investigation of unexpected antibodies to ABO antibody titration studies. The analyzer identifies the samples and reagents with the barcode and manages the batch and expiry. It has automated dilution and dispensing of samples and reagents, incubation, centrifuging, reading, interpretation of results, and generation of reports. Samples can be processed in routine mode and/or STAT mode. “STAT” mode allows the operator to run urgent samples on a priority basis, while the other routine samples are rescheduled accordingly. Generated results are displayed on computer monitor in high color resolution.

Reagents used

ABO blood grouping, reverse diluent and anti-IgG, and C3d polyspecific cassette (Ortho Clinical Diagnostics, India) were used in the study.

Turnaround time

Routine mode

TAT was analyzed and compared with the BioVue system. A total of 42 samples were analyzed in five commonly used test profiles:

Blood grouping (forward and reverse) – (n = 42)

Blood grouping + antibody screening – (n = 42)

Blood grouping + IAT + DAT – (n = 42)

Major crossmatch – (n = 42)

Blood grouping + IAT – (n = 42).

In Vision, all samples were loaded simultaneously, while in BioVue system, the first samples were loaded manually in cassettes and then centrifuged. Since the BioVue centrifuge can accommodate 12 cassettes in one go, it took four cycles of centrifugation to complete testing of all samples. Stopwatch was used to record time and it included the time of preparation of suspensions, loading, incubation, centrifugation, and interpretation. TAT was recorded in terms of completion of results as “time taken to get first result,” “time taken to get 50% of results,” and “Time taken to get complete results.”

STAT mode

TAT was also analyzed using STAT mode, and completion times were recorded.

Stress test

This test was executed with an aim to evaluate the work performance of the Vision IH analyzer. This was an attempt to simulate “real” average workload received in the laboratory on a working day. A total of 126 samples were run in different combinations of test profiles in a set of 42 samples (maximum capacity of analyzer) three times over.

Blood grouping – 63 samples

Major crossmatch – 36 samples (18 donors + 18 patients)

Blood grouping and antibody screen – 15 samples

Blood grouping and IAT – 6 samples

DAT – 6 samples.

Data collection and analysis

All the data were stored in Microsoft Excel sheets (Microsoft Corporation, USA) and finally were analyzed using the Statistical Package for the Social Sciences, version 20.0 (IBM SPSS Statistics for Windows, IBM Corp, Armonk, NY, USA). Agreement analysis was performed to compare the test results of two methods (Vision and BioVue system).

Results

A total of 1276 known patient or donor samples were used in the study for the evaluation of Vision. These samples were used to assess 2640 pretransfusion tests (1229 were for ABO/Rh D typing; 1229 for unexpected antibody screening; 54 for antibody identification; 86 for crossmatch; and 42 for DAT). Overall agreement between Vision IH analyzer and BioVue method was 99.7%. Positive agreement was 99.72%, while negative agreement was 100%. Pretransfusion testing results were concordant in 99.55%.

ABO-Rh typing

A total of 1229 ABO-Rh typing was performed using both the systems (Vision IH analyzer and BioVue system). In 1229 tests, the blood groups were distributed as 221 A, 173 AB, 454 B, and 381 O, while 49 Rh D negative, and 1180 Rh D positive. All 1229 ABO-Rh typing results were concordant between the two systems.

Seven known group discrepancy samples were analyzed on Vision IH analyzer compared with BioVue system. Results are summarized in Table 1.

Table 1

Analysis of blood group discrepancy samples using Vision and BioVue system

Discrepancy type	Known blood group discrepancy	Results (vision IH analyzer)	Results (BioVue IH system)
Type 1	Forward O/reverseAB	ForwardO/reverse AB	ForwardO/reverse AB
	Forward A/reverseAB	ForwardA/reverse AB	ForwardA/reverse AB
Type 2	Weak A positive (subgroup of A)	A Rh D positive	O Rh D positive
	Weak A positive (subgroup of A)	O Rh D positive	O Rh D positive
Type 4	Auto-control positive	Auto-control positive	Auto-control positive
	Alloantibody positive	Alloantibody positive	Alloantibody positive
	A2 with anti-A1	A2 with anti-A1	A2 with anti-A1

On agreement analysis, overall agreement between the Vision IH analyzer and BioVue system for ABO-Rh typing was 99·95%.

Antibody screening and identification

A total of 1229 unexpected antibody screening was performed using both the systems (Vision IH analyzer and BioVue system). All 1229 antibody screening results (8 positive and 1221 negative) were concordant between the two systems. For antibody screening, the overall agreement between the Vision IH analyzer and BioVue system was 100%.

Forty-five known antibody samples were reanalyzed using both the systems (Vision IH analyzer and BioVue system). All samples were correctly identified by Vision IH analyzer, while BioVue BioVue system missed one weak anti-M antibody. For antibody identification, agreement between the Vision and BioVue system was 98·9% [Table 2].

Table 2

Antibody identification results using Vision and BioVue system

Known antibody	n	Vision IH analyzer		BioVue system
		Correctly identified	Not/incorrectly identified	Correctly identified	Not/incorrectly identified
Anti D	13	13	0	13	0
Weak anti D	11	11	0	11	0
Anti Fya	1	1	0	1	0
Anti Jka	1	1	0	1	0
Anti M	5	5	0	5	0
Weak anti M	1	1	0	0	1
Anti E + S	1	1	0	1	0
Anti C + K	1	1	0	1	0
Anti S	1	1	0	1	0
Anti K	1	1	0	1	0
Anti C	1	1	0	1	0
Anti c	1	1	0	1	0
Anti E	1	1	0	1	0
Anti e	1	1	0	1	0
Anti Jkb	1	1	0	1	0
Anti s	1	1	0	1	0
Anti P	1	1	0	1	0
Anti D + C	1	1	0	1	0
Anti JKb + C	1	1	0	1	0

Crossmatch

Eighty-six known crossmatch tests (75 negative, 11 positive) were performed using both the systems (Vision IH analyzer and BioVue system). All test results were concordant between the two systems.

Direct antiglobulin test

Forty-two (37 negative and 5 DAT positive) known samples were analyzed using both the systems (Vision IH analyzer and BioVue system). All five positive and 37 negative samples were correctly identified by both the systems.

For crossmatch and DAT results, the overall agreement between the Vision IH analyzer and BioVue system was 100%.

Turnaround times

A total of 42 samples (in routine mode) were used for each test profile and compared using both the systems. Results are depicted in Figure 1.

Figure 1

Comparison of Turn Around Time (TAT) between Vision and Reference system for different test profiles

In STAT mode, seven samples were analyzed for each test profile. Time taken to get complete the results of all seven samples was 19 min for blood grouping, 43 min for blood grouping and antibody screening, 42 min for blood grouping + IAT + DAT, 41 min for major crossmatch, and 52 min for blood grouping + IAT.

A total of 126 samples were analyzed in the stress test which lasted for 5 h. Sixty-three blood grouping results were completed in 69 min. Major crossmatch, grouping and antibody screen, and grouping and IAT and DAT completion times are shown in Figure 2.

Figure 2

Stress test

Discussion

Handling the large amount of IH workload and maintaining/improving the quality of testing are the first objectives that were expected from automation. In our study, Vision was found to have comparable reliability of test results with routinely used semiautomated BioVue system used in our laboratory. In the various tests executed throughout the study, Vision performance met the expectations. Vision was able to handle the given workload of IH tests yielding reliable results in TAT shorter than the BioVue system.

Reliability

In ABO-Rh typing, the overall agreement between Vision IH analyzer and BioVue method was very good. In the routine blood grouping samples, the results between the two systems were concordant. On analyzing group discrepancy samples, sample 1 and 2 were known Type 1 discrepancy samples where, in both, the sample antibodies were missing. On analysis, sample 1 was typed as forward “O” and reverse “AB” and sample 2 was typed as forward “A” and reverse “AB” on both the platforms (Vision IH analyzer and BioVue system). In these two samples, antibodies were missing which was confirmed by cold enhancement technique.[3] Samples 3 and 4 were known Type 2 discrepancy samples where, in both, the sample A antigen was weak or undetectable (subgroup of A) which was confirmed by cold adsorption and heat elution technique.[4] On analysis, sample 3 was typed as rightly A Rh D positive on Vision with a query of weak expression of “A antigen” as 1+ or weak reaction, while the same sample was typed as O Rh D positive on BioVue system. Similarly, sample 4 was typed as O Rh D positive by both the systems. Samples 5-7 were known Type 4 discrepancy samples where sample 5 was auto-control positive, 6 was alloantibody (anti-M), and 7 was A2 with anti-A1 antibody. Results of all these three samples were concordant between the two systems. As all these seven samples were known group discrepancy, Vision was still able to detect weaker expression of “A-antigen” which was missed by BioVue system. This may due to automated analysis of results in Vision. A total of 1229 antibody screening results (8 positive and 1221 negative) were concordant between the two systems. Forty-five known antibody samples were reanalyzed using both the systems (Vision IH analyzer and BioVue system). All samples were correctly identified by Vision IH analyzer, while BioVue system missed one weak anti-M antibody. This may be due to automated analysis of results in Vision. All crossmatch and DAT results were concordant between the two systems.

Performance

TAT of Vision was substantially shorter than the BioVue system for all test profiles. Time to get first result in Vision was 71 min earlier than BioVue system while analyzing 42 samples for blood grouping. Similar reduction was noticed in other test profiles as well. Likewise, measured TAT with respect to 50% of test results and completion of test results were also favorable with Vision. Shorter TAT is obvious because of robotic arms, multitasking, twin centrifuges, and automated analysis. A total of 126 samples were successfully analyzed by Vision in a period of 5 h (stress test) which is otherwise completed by two technicians in an 8-h shift.

Conclusion

Based on study results in terms of reliability, Vision was found to be equivalent or superior to the BioVue system and was therefore approved for routine use in the laboratory. Moreover, Vision had considerably shorter TAT and capable of handling higher throughput of IH tests.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.