Letter Annals of Clinical Biochemistry 2014, Vol. 51(4) 515–516 acb.sagepub.com
Authors’ reply We thank Kavsak et al. for their interest in, and response to, our manuscript.1,2 In our study, troponin I (TnI) was analysed using only the 500-test reagent packs for both Abbott Diagnostics’ contemporary sensitive and high-sensitive assays. We did not examine the effect of reagent pack size on the rate of outliers in our study; therefore, we cannot comment on the effect of reagent pack size on the outlier rate in our hands. Kavsak et al. investigated the occurrence of a first elevated troponin result after periods of analyser inactivity. Both our analysers operate in ‘running’ mode 24 h a day, with the exception of scheduled maintenance (usually once per day) or unforeseen downtime (rare). In ‘running’ mode, the micro-particle reagent bottles are continuously mixed, although the conjugate reagent bottles are not. Post-maintenance start-up of the analyser includes an additional mixing period of reagent bottles. We have examined the datasets from our study, and have found no association between the occurrence of the critical outliers and periods of analyser downtime or low activity. The focus of our study was on comparing the critical outlier rate of Abbott Diagnostics’ contemporary sensitive TnI assay with their high-sensitive assay using data collected from a total of 15,622 routine patient samples. The finding of positive outliers in a low TnI patient pool by Kavsak et al. using the contemporary sensitive TnI assay by Abbott reinforces the importance of investigating the precision and robustness of troponin assays as part of the evaluation process. However, pooling plasma samples from different patients can cause interferences due to matrix modification and introduction of unknown factors into the sample, and may not reflect what happens with individual patient samples. It is of interest to note that the first elevated TnI results did not occur with any of the 100-test packs in Kavsak et al.’s study, but only with the 500-test packs. In experiment 1, no first elevated result was observed with the 500-test pack on the ARCHITECT 16200 analyser at Site 2, but the same 500-test pack when transferred to the ARCHITECT 8200 at Site 1 resulted in
first elevated results in all three cycles in experiment 2. No experiments were carried out looking at the performance of the 100-test packs on the ARCHITECT 16200. Therefore, analyser-related error cannot be excluded as the cause of the outliers until all experiments are repeated at Site 2. Interestingly, in experiment 2, the mean TnI of the patient pool was 0.017 mg/L for the 500-test packs, but was higher using the 100-test packs, with a mean of 0.025 mg/L. The 500-test pack loaded at 15:20 gave an initial result of 0.024 mg/L, and was classed as an outlier, but when an initial result of 0.024 mg/L was found with the 100-test packs, this was not deemed to be an outlier. Clearly, Kavsak et al. have highlighted that the potential for variation in low-level TnI results with reagent pack size and periods of analyser inactivity is worth further investigation. Declaration of conflicting interests None.
Funding None.
Ethical approval Not applicable.
Guarantor NS.
Contributorship NS and SV are authors of the letter.
References 1. Sawyer N, Blennerhassett J, Lambert R, et al. Outliers affecting cardiac troponin I measurement: comparison of a new high sensitivity assay with a contemporary assay on the Abbott ARCHITECT analyser. Ann Clin Biochem. Epub ahead of print 23 September 2013. 2. Kavsak et al. (letter).
516
Annals of Clinical Biochemistry 51(4) N Sawyer and S Vasikaran Department of Clinical Biochemistry, PathWest Laboratory Medicine, Royal Perth Hospital, Perth
Corresponding author: N Sawyer, Department of Clinical Biochemistry, PathWest Laboratory Medicine, Royal Perth Hospital, Wellington Street, Perth 6000, Australia. Email: [email protected] ! The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav acb.sagepub.com doi: 10.1177/0004563214534401 available online at http://acb.sagepub.com
Copyright of Annals of Clinical Biochemistry is the property of Sage Publications, Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Authors’ reply We thank Kavsak et al. for their interest in, and response to, our manuscript.1,2 In our study, troponin I (TnI) was analysed using only the 500-test reagent packs for both Abbott Diagnostics’ contemporary sensitive and high-sensitive assays. We did not examine the effect of reagent pack size on the rate of outliers in our study; therefore, we cannot comment on the effect of reagent pack size on the outlier rate in our hands. Kavsak et al. investigated the occurrence of a first elevated troponin result after periods of analyser inactivity. Both our analysers operate in ‘running’ mode 24 h a day, with the exception of scheduled maintenance (usually once per day) or unforeseen downtime (rare). In ‘running’ mode, the micro-particle reagent bottles are continuously mixed, although the conjugate reagent bottles are not. Post-maintenance start-up of the analyser includes an additional mixing period of reagent bottles. We have examined the datasets from our study, and have found no association between the occurrence of the critical outliers and periods of analyser downtime or low activity. The focus of our study was on comparing the critical outlier rate of Abbott Diagnostics’ contemporary sensitive TnI assay with their high-sensitive assay using data collected from a total of 15,622 routine patient samples. The finding of positive outliers in a low TnI patient pool by Kavsak et al. using the contemporary sensitive TnI assay by Abbott reinforces the importance of investigating the precision and robustness of troponin assays as part of the evaluation process. However, pooling plasma samples from different patients can cause interferences due to matrix modification and introduction of unknown factors into the sample, and may not reflect what happens with individual patient samples. It is of interest to note that the first elevated TnI results did not occur with any of the 100-test packs in Kavsak et al.’s study, but only with the 500-test packs. In experiment 1, no first elevated result was observed with the 500-test pack on the ARCHITECT 16200 analyser at Site 2, but the same 500-test pack when transferred to the ARCHITECT 8200 at Site 1 resulted in
first elevated results in all three cycles in experiment 2. No experiments were carried out looking at the performance of the 100-test packs on the ARCHITECT 16200. Therefore, analyser-related error cannot be excluded as the cause of the outliers until all experiments are repeated at Site 2. Interestingly, in experiment 2, the mean TnI of the patient pool was 0.017 mg/L for the 500-test packs, but was higher using the 100-test packs, with a mean of 0.025 mg/L. The 500-test pack loaded at 15:20 gave an initial result of 0.024 mg/L, and was classed as an outlier, but when an initial result of 0.024 mg/L was found with the 100-test packs, this was not deemed to be an outlier. Clearly, Kavsak et al. have highlighted that the potential for variation in low-level TnI results with reagent pack size and periods of analyser inactivity is worth further investigation. Declaration of conflicting interests None.
Funding None.
Ethical approval Not applicable.
Guarantor NS.
Contributorship NS and SV are authors of the letter.
References 1. Sawyer N, Blennerhassett J, Lambert R, et al. Outliers affecting cardiac troponin I measurement: comparison of a new high sensitivity assay with a contemporary assay on the Abbott ARCHITECT analyser. Ann Clin Biochem. Epub ahead of print 23 September 2013. 2. Kavsak et al. (letter).
516
Annals of Clinical Biochemistry 51(4) N Sawyer and S Vasikaran Department of Clinical Biochemistry, PathWest Laboratory Medicine, Royal Perth Hospital, Perth
Corresponding author: N Sawyer, Department of Clinical Biochemistry, PathWest Laboratory Medicine, Royal Perth Hospital, Wellington Street, Perth 6000, Australia. Email: [email protected] ! The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav acb.sagepub.com doi: 10.1177/0004563214534401 available online at http://acb.sagepub.com
Copyright of Annals of Clinical Biochemistry is the property of Sage Publications, Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
