Comparison of Two Different Serological Viral Marker Testing Assays for Screening of Apheresis Donors: Which Assay Provides Optimum Safety for Transfusion?

Abstract

While whole blood testing has evolved over the years, viral marker testing for plateletpheresis donors is still performed by Rapid Diagnostic Tests (RDT). Aim of this study was to compare diagnostic accuracy of RDT and Chemiluminescence Immunoassay (CLIA) in serological testing for HBsAg, anti-HCV and anti-HIV antibodies. A prospective, analytical study was conducted in the department of Transfusion Medicine at a tertiary healthcare center in India between September 2016 and August 2018. Samples were simultaneously tested by CLIA, RDT and a confirmatory test. Sensitivity, specificity, negative and positive predictive values and mean time taken to report results were calculated. A total of 102 (1.48%) of the 6883 samples were found to be reactive by either or both the assays. A total of 74 (1.08%) samples were HBsAg reactive, 23 (0.33%) were reactive for anti-HCV antibodies and 5 (0.07%) were reactive for anti-HIV I and II antibodies. A combined sero-prevalence of 1.05% (72) was observed; 0.78% (54) for HBsAg, 0.26% (18) for anti-HCV antibodies and none for anti-HIV I and II antibodies. Four (3.85%) reactive samples were missed by RDT and therefore sensitivity of RDT was quite less as compared to CLIA. RDT and CLIA both were found to have a statistically significant shorter turnaround time than confirmatory tests. There is increasing need to develop a safe donor screening strategy for plateletpheresis. CLIA offers an excellent alterative to RDT for viral marker testing in terms of sensitivity.

Background

Testing of blood and blood products reduces the inherent risk of transmission of infective diseases considerably; however, even with the current technologies in use, this risk still remains due to variable sensitivity of screening assays and also due to prevalence of asymptomatic carriers. For blood transfusion services, sensitivity of screening assays is of utmost importance to reduce the risk of transfusion transmitted infections (TTI) [1]. According to the National Aids Control Organization (NACO) strengthening quality management systems in blood banks as well as World Health Organization (WHO) recommendations for screening donated blood for transfusion-transmissible infections, both sensitivity and specificity should be the highest possible [2, 3]. As reported by NACO in their Rapid Situation Assessment of Blood Transfusion Services in India, 2014, the use of Rapid Diagnostic Tests (RDTs) still remains the method of TTI testing in many transfusion services across our country [4].

Many a times situations such as ‘directed donations’ arise, where RDTs are the usual means of testing a donation for viral markers. This is especially used for screening of apheresis donors due to lack of voluntary Single Donor Platelet Concentrate (SDPC) inventories, in an attempt to reduce the waiting time of donors and also to save on the cost of plateletpheresis kit. As suggested in the Directorate General of Health Sciences technical manual, apheresis donors should be screened prior to apheresis for markers of infectious diseases transmitted by transfusion of blood in the same manner as for whole blood and each donor must be tested prior to each apheresis unless the donor in undergoing repeated procedures [5]. Since, product testing for plateletpheresis is not feasible, pre-donation testing is the most logical practice. This is followed at most transfusion services where plateletpheresis is performed when required by a patient and use of a point-of-care test like RDT reduces the waiting time for donors.

While whole blood testing has evolved dramatically over the years, with addition of layers of safety in the form of anti-HBc (total) antibody testing and Nucleic Acid Testing (NAT) testing, viral marker testing for SDPC donors is still performed by Rapid Diagnostic Tests (RDT) commonly, for the above-mentioned reasons. According to WHO global status report on blood safety and availability, of the 112.5 million blood donations in 180 countries, 100.6 million were whole blood donations and 11.9 million were apheresis donations [6]. However, the question that needs to be answered is that what are the odds of transfusion of a reactive SDPC with use of RDT only? With this question in mind; to address the safety of transfusion of such SDPC units, a study was conducted to compare the results of RDT and chemiluminescence Immunoassay (CLIA) in screening of donors for plateletpheresis. The primary aim of this study was to compare the diagnostic accuracy of RDT and CLIA in serological testing for HBsAg (Hepatitis B Surface Antigen), anti-HCV antibodies (anti-Hepatitis C antibodies) and anti-HIV antibodies (anti-HIV I and II antibodies).

Materials and Methods

Settings and Design

A prospective, analytical study was conducted in the department of Transfusion Medicine at a large tertiary healthcare setup with an annual SDPC collection of more than 3000 units. The study population included all consecutive plateletpheresis donor samples received for TTI testing before plateletpheresis donation from September 2016 to August 2018. A total of 6883 donor samples were included between September 2016 and August 2018. As per the standard operating procedure (SOP) of the department, all plateletpheresis donors were first screened for medical history and a brief physical examination was performed. Once these donors were found eligible for donation, blood samples were drawn to determine platelet count, TTI screening tests for HIV, HBV, HCV, malaria and syphilis and for blood grouping and unexpected antibody screening. Donors who had donated whole blood or plateletpheresis previously were classified as repeat donors while those who had not donated previously were classified as first time donors.

CLIA

CLIA for HBsAg testing, anti-HCV antibody testing and anti-HIV antibody testing was performed on automated Vitros 3600 Immunodiagnostic system (Ortho Clinical Diagnostics, USA) as per manufacturer’s instructions. Target analyte bound horse radish peroxidase conjugate is measured by a luminescent reaction by adding a signal reagent which contains luminol and an electron transfer reagent. A luminometer measures the light signals and cut‑offs are automatically calculated by the machine.

RDT

RDT was performed by HIV tridot (J. Mitra & Co Pvt Ltd, India) for anti-HIV antibodies, HCV tridot (J. Mitra & Co Pvt Ltd, India) for anti-HCV antibodies and Hepacard (J. Mitra & Co Pvt Ltd, India) for HBsAg as per manufacturer’s instructions. These are immunochromatography based rapid visual tests for the qualitative detection viral markers in human serum. The test sample flows laterally to mix with signal reagent forming an antigen–antibody complex which is trapped leading to a visual indication.

Confirmatory Tests

Confirmatory testing performed included HBsAg neutralization assay (Ortho clinical diagnostics, USA), immunoblot assay (Deciscan, BioRad, Germany) and Western Blot (J. Mitra & Co Pvt Ltd, India) for hepatitis B, hepatitis C and HIV serological testing respectively. The tests selected as confirmatory tests were in accordance with published literature [7, 8]. Confirmatory tests were performed in a batch, generally on the subsequent day for all 6883 samples.

Decision Making Algorithm

For the purpose of this study, the sample drawn for TTI testing of HBsAg, anti–HCV antibody testing and anti–HIV I and II antibody testing was simultaneously performed by CLIA and RDT, where RDT was done in parallel for research purpose only. Donors who tested non-reactive by CLIA and RDT were accepted for donation. Donors who tested reactive with any of the two screening tests were deferred permanently after appropriate counselling. A third confirmatory test was also performed for these three infectious agents at a later point-in-time, usually, on the subsequent day. The testing algorithm has been illustrated in Fig. 1.

Fig. 1.

Algorithm for plateletpheresis donor screening for the period of this study

Indeterminate Results on Confirmatory Tests

An indeterminate result by a confirmatory test was subjected to Nucleic Acid Amplification tests (NAT) (Procleix Panther system, Grifols, USA) for resolution of the discrepancy.

Calculation of Turnaround Time (TAT)

The donor samples received in TTI laboratory for plateletpheresis donor screening were processed as per Fig. 1. The time of receiving these samples was recorded. Once the sample was centrifuged and ready for testing, the time was recorded. Time taken from receiving the sample to the time it was ready for testing was ‘A’.

• TAT for RDT—For every sample, a fresh RDT device was unwrapped, labelled and the test was performed. Time taken till interpretation of result was ‘B’. Total TAT for RDT was A+B.

• TAT for CLIA—Time taken from allocating the sample in the equipment to the time a valid result was displayed on the screen was ‘C’. An invalid result was repeated and the total time of invalid and valid measurement was considered. Total TAT for CLIA was A+C.

• TAT for confirmatory test—Time taken from initiation of confirmatory test to interpretation of final result was ‘D’. Total TAT for confirmatory test was A+D.

Ethical Clearance

Institutional Review Board (IRB) approved the study and informed consent was obtained from all the blood donors for screening.

Statistical Analysis

Data was entered in an MS excel sheet; numerical values, percentages, mean and standard deviation was calculated. Statistical analysis was performed using SPSS software (Version 25.0.0.0, Chicago, USA). Mean time taken to report results by both the assays was calculated and independent student t test was applied to calculate p-value. p-value was considered significant, if found < 0.05. Sensitivity was calculated as TP/(TP + FN) × 100 and specificity was calculated as TN/(TN + FP) × 100. Positive predictive value was calculated as TP/TP + FP and negative predictive value was calculated as TN/TN + FN. (TP: Number of true positive results, FP: Number of false positive results, TN: Number of true negative results, FN: Number of false negative results).

Results

Between September 2016 and August 2018, a total of 6883 apheresis donor samples were simultaneously tested by RDT and CLIA.

Demographic Data

Out of 6883, 356 (5.17%) samples were from female donors and 31.95% donors were repeat donors. All 6883 were replacement donors. A medical and surgical discipline-wise split has been illustrated in Fig. 2.

Fig. 2.

Department-wise distribution of patients for whom donors were screened for plateletpheresis

Results Obtained by CLIA and RDT

Out of the 102 reactive samples, 30 were found reactive by either one and 68 were reactive by both assays. While there were 30 samples which were reactive by ‘CLIA only’ (non-reactive by RDT), none of the samples were reactive by ‘RDT only’. None of the samples was simultaneously reactive for more than one viral marker i.e. co-infection was not detected in any of the donors. These results have been summarized in Table 1.

Table 1.

Number of reactive results for HBsAg, anti-HCV antibodies and anti-HIV I and II antibodies obtained by RDT and CLIA and their comparison with confirmatory tests

	HBsAg				Anti-HCV				Anti-HIV
	CLIA	RDT	Both	N	CLIA	RDT	Both	I	CLIA	RDT	Both	WB
Reactive result	74	51	51	54	23	17	17	18	5	0	0	0
True reactive	54	51			18	17			0	0

N HBsAg neutralization, I Immunoblot, WB Western Blot

Sero-Prevalence

A total of 102 (1.48%) samples were found to be reactive by either RDT or CLIA or both of the assays. A total of 74 (1.08%) samples were HBsAg reactive, 23 (0.33%) were reactive for anti-HCV antibodies and 5 (0.07%) were reactive for anti-HIV I & II antibodies. With respect to results obtained by confirmatory assays, a combined sero-prevalence of 1.05% (72) was observed; 0.78% (54) for HBsAg, 0.26% (18) for anti-HCV antibodies and none of the samples were found to be reactive for anti-HIV I and II antibodies.

Comparison of Screening Tests

Calculation of Sensitivity, Specificity, Positive and Negative Predictive Value

None of the confirmatory tests gave an indeterminate result. Table 2 gives a summary of the sensitivity, specificity, positive and negative predictive value for both RDT and CLIA. Four (3.85%) reactive samples were missed by RDT and therefore sensitivity of RDT was quite less as compared to CLIA which showed a 100% sensitivity for HBsAg and anti-HCV antibody testing.

Table 2.

Sensitivity, specificity, positive predictive and negative predictive value of RDT and CLIA for screening of all the three viral markers

	RDT			CLIA
	HBsAg (%)	Anti-HCV (%)	Anti-HIV (%)	HBsAg (%)	Anti-HCV (%)	Anti-HIV (%)
Sensitivity	94.44	94.44	–	100	100	–
Specificity	100	100	100	99.71	99.93	99.93
PPV	100	100	–	72.97	78.26	–
NPV	99.96	99.99	100	100	100	100

PPV Positive Predictive Value, NPV Negative Predictive Value

TAT for Performing RDT and CLIA Versus Confirmatory Test

Mean TAT for CLIA, RDT and confirmatory tests were calculated for performing all the three tests simultaneously. Mean time for performing all three tests by RDT was 39.2 ± 2 min while it was 77.8 ± 5.8 min for CLIA. Mean time taken for performing confirmatory tests was found to be 366.5 ± 27.8 min. Independent student t-test was applied for comparison and the p-value obtained was found to be significant (< 0.00001); RDT and CLIA both were found to have a statistically significant shorter TAT than confirmatory tests.

Discussion

All modern-day healthcare establishments depend on their Blood Transfusion Services (BTS) for one of the most precious life-saving interventions i.e. blood transfusion and safe transfusions form the most important cornerstone of any transfusion service. With every blood component being transfused, there is 1 in 843,000 to 1 in 1,208,000 residual risk of transmission of Hepatitis B, 1 in 1,149,000 residual risk of transmission of Hepatitis C and 1 in 1,467,000 residual risk of transmission of HIV in USA [9, 10].

The National Blood Policy of India states that every transfusion service should strive to provide safe and quality blood and blood components which are free from transfusion transmitted infections and are stored and transported under optimum conditions to all patients who require transfusion [11]. Donor medical history questionnaire and screening for transfusion transmitted infections (TTI) are therefore important procedures in any BTS that are performed to minimize the risk of TTI. Laboratory testing of blood prior to transfusion is intended to ensure safety. The National Blood Transfusion Council (NBTC) of India in its guidelines for recovery of processing charges for blood and blood components mentions that additional testing can be done by transfusion services to enhance blood safety and it also accepts that there is wide variation in services provided by blood banks across the country due to the cost involved [12]. Detection of infectious diseases like HIV, HBV and HCV pose a major challenge to transfusion services, especially in countries with high sero-prevalence of these diseases in blood donors. It is the duty of the Transfusion Medicine physician in any BTS to choose an appropriate assay from a wide variety available which ensures maximum safety of the blood component being transfused [13].

Sero-Prevalence/true Seroprevalence

Chandekar et al. collected five-year data for calculating sero-prevalence of transfusion transmitted infections in healthy blood donors. A total of 76,653 healthy donors were included and the overall sero-prevalence of HIV, HBsAg and HCV was found to be 0.26%, 1.30% and 0.25% respectively [14]. In the present study, a combined true sero-prevalence of 1.05% (72) was observed; 0.78% (54) for HBsAg, 0.26% (18) for anti-HCV antibodies and none of the samples were found to be reactive for anti-HIV I and II antibodies. They concluded that methods to ensure a safe blood supply should be encouraged such as, screening with a better selection of donors and use of sensitive screening tests, including NAT. The risk of TTI with blood transfusion is directly related to the method used for screening of these units. Transmission of HIV, HBV and HCV via blood transfusion has seen a decline due to stringent donor deferral criteria and availability of better screening tests over the years. Rawat et al. [15] concluded from their study that increase in public awareness regarding voluntary blood donation, meticulous donor screening, counselling and use of highly sensitive tests can help in reducing the risk of TTIs.

TAT

A majority of transfusion services in India are dependent on replacement donations for single donor platelets; only a few centers have voluntary plateletpheresis donations primarily due to the cost involved and also due to the shorter shelf-life as compared to other components. When talking about plateletpheresis donations, screening of the donor is performed before collection in shortest possible span of time so that the component is available for transfusion as early as possible and also to reduce the inconvenience caused to the donor due to the long duration of the process. For this reason, a large number of blood banks use RDTs for screening of viral infections. The primary advantage of it being reduced TAT. However, CLIA also does not have a long TAT, as seen in the present study and the advantage of random access unlike ELISA. The mean time taken by CLIA was close to an hour for all the three assays, which is quite acceptable in return for increasing safety of the component.

CLIA Versus RDT: Which is a Better Screening Tool?

CLIA has some known advantages such as automation, random access and high throughput. RDT on the other hand are easier and require less skill. A preliminary study was therefore conducted by Kanr et al. [16] to evaluate the efficacy of RDT for screening of blood donors and they concluded that while all rapid kits were 100% specific for HIV, HBsAg and HCV, sensitivity for HIV, HCV and HBsAg was 80%, 87.5% and 93.4% respectively. In the present study, CLIA showed a 100% sensitivity for HBsAg and anti-HCV antibody testing as compared to RDT which showed a sensitivity of 94.44% and 94.44% for HBsAg and anti-HCV respectively.

The present study was conducted to evaluate performance and usefulness of CLIA and RDT for apheresis donor screening. Though RDTs have better specificity than CLIA; the fact that CLIA is an extremely sensitive assay and it prevented TTI in four recipients is more important to a BTS. The present study demonstrates that no matter how lucrative RDTs may appear, there were four samples; three HBsAg reactive and one reactive for anti-HCV antibodies which were missed by RDT and could have led to TTI had the components been transfused to patients.

The results obtained by CLIA in the present study demonstrate that it is an extremely sensitive assay and hence false positivity is more than that of RDT. This is primarily due to the high concentrations of IgG in blood. Some of these IgG molecules bind to the surface of microwells giving a false positive reaction. Being a sensitive assay, CLIA does lead to an increased rate of donor deferral or discard of blood component unit. However, as per the NACO HIV testing manual which mentions the various testing strategies, the first test to be employed for screening of donors has to be a sensitive test [1]. Authors of previously published studies have put up suggestions to increase sensitivity and specificity of viral marker testing. Two relevant suggestions found in this context were found to be pertinent. First by Salawu et al., where the authors concluded that some potential blood units containing HBV are being transfused to patients unknowingly by screening for HBsAg only. Screening for other markers of hepatitis B virus may increase the rejection rate, but will reduce HBV transmission [17]. Similarly second by Tiwari et al., where the authors concluded in their study that a sequential serological testing strategy can be adopted for identifying HBV infection [8].

Maity et al. conducted a study for comparing ELISA and RDT and found that ELISA is a good screening assay for markers of HIV, HBV and HCV infections and rapid tests are useful for further detection of false positive samples. ELISA seems the appropriate assay in blood bank [18]. Similarly, Torane et al. [19] concluded in a correspondence that rapid tests are inferior to ELISA and hence are not recommended for use at transfusion centers for routine screening of donors. Majumder et al. [20] compared ELISA and CLIA and concluded that detection of anti-HCV by CLIA is comparable with ELISA but CLIA can help in detecting infection earlier compared to ELISA and is suitable in large sample volume laboratories. The present study also found high sensitivity of CLIA as compared to RDT. There were three HBsAg seropositive samples and one anti-HCV antibody seropositive sample which was missed by RDT and this leads to the same conclusion that, RDTs should not be used as screening tools for detection of transfusion transmitted viral infections.

The transfusion service has screened a total of 36,457 donors for plateletpheresis from setup till August 2018. If RDT were to be used for all these donors, more than five components with risk of post transfusion hepatitis C and 16 components with risk of post transfusion hepatitis B would have been transfused to patients. There are more than 3000 transfusion services across the country. If all centers were to use RDT, the cumulative number of SDPC capable of transmitting viral infections would definitely raise concerns about safety of the component being transfused. Therefore, to increase safety of the SDPC being transfused, more sensitive methods such as CLIA should be used instead of RDT for screening of donors for plateletpheresis.

Moving Towards Safety

While whole blood donations have been added with additional layers of safety in the form of NAT and anti-HBcore testing; plateletpheresis donations are still being tested by RDT alone. SDPC are a standard of care for patients suffering from thrombocytopenia and hence, it is a matter of concern for Transfusion Medicine specialists that this particular component is still being issued to patients on the basis of results obtained by RDTs only.

The major challenges with more sophisticated tests such as CLIA and NAT is that they are more expensive, require use of more instrumentation and require trained personnel [21]. NAT is an advanced screening technology that reduces the window period of transfusion transmissible infections (TTIs) and helps improve blood safety. It is becoming the gold standard for blood donor screening in the developed world particularly because of its ability to reduce the risk of sero-negative, NAT-reactive donations from entering the blood donor pool [22]. The combined NAT yield for all three viruses reported in studies from various parts of the country range from 1 in 476 to 1 in 4403 and according to a recent review, for individual infections, NAT yield from the pooled data showed HIV in 1:66,000, HCV 1:5484 and HBV in 1:1761 sero-negative Indian donors [23]. Previous reports have indicated that implementation of NAT will provide a significant increment in safety in relation to serological screening alone [24, 25]. Hence, steps towards employment of more sensitive tests in detection of infectious agents like HIV, HBV and HCV in addition to serological tests for plateletpheresis donors should be taken to move towards safety. Strengths of this study include a robust sample size and simultaneous testing of all samples with screening and confirmatory assays. A limitation of the study was inability to test all samples by NAT.

Conclusion

There is increasing need to develop a safe blood donor screening strategy for plateletpheresis donors with use of more sensitive methods to ensure the safety of SDPC as well.

Funding

Not applicable.

Declarations

Conflict of interest

The authors declare no conflict of interest.