Intraindividual Stability of Human Erythrocyte ... - CiteSeerX

Papers in Press. Published May 10, 2007 as doi:10.1373/clinchem.2006.085258 The latest version is at http://www.clinchem.org/cgi/doi/10.1373/clinchem.2006.085258

Clinical Chemistry 53:7 000 – 000 (2007)

General Clinical Chemistry

Intraindividual Stability of Human Erythrocyte Cholinesterase Activities Lee J. Lefkowitz,1† Joseph M. Kupina,1 Nigel L. Hirth,1 Rachel M. Henry,1 Georgia Y. Noland,2 John Y. Barbee, Jr.,2 Joey Y. Zhou,1 and Coleen B. Weese1* regarding the appropriate frequency of RBC-ChE testing.

Background: Erythrocyte cholinesterase (RBC-ChE) activities serve as useful and sensitive biomarkers to monitor exposure to cholinesterase-inhibiting substances, such as chemical warfare nerve agents and pesticides. Although the interindividual variation of RBC-ChE is well characterized, the magnitude of intraindividual variation for RBC-ChE remains controversial. An accurate measure of intraindividual variation is critical for establishing the appropriate frequency of RBC-ChE testing. Methods: We retrospectively tracked the intraindividual variation of RBC-ChE activities among 46 male nerve agent workers from a single US Army depot that participated in a medical surveillance program requiring periodic RBC-ChE monitoring. All RBC-ChE analysis was performed by the same medical laboratory technician by the delta pH method. Results: A mean of 38 and a median of 37 RBC-ChE measurements were available for each worker. The mean duration of employment for these workers was 20 years (median, 21 years). The mean CV for RBC-ChE in this set of 46 workers was 3.9%. Linear regression analysis of the data for each worker resulted in a mean slope of 0.0010 delta pH units/h per year. Conclusions: RBC-ChE activities increased in each person by a mean of 0.01 delta pH units/h every 10 years, which is a negligible rate. These findings highlight the stability of RBC-ChE activities over time in a given individual and may have important policy implications

© 2007 American Association for Clinical Chemistry

The clinical measurement of cholinesterase (ChE)3 activities represents the most sensitive biomarker of exposure to anticholinergic organophosphorous compounds, such as pesticides and nerve agents. Both the US Department of Defense (DoD) and the California Environmental Protection Agency use ChE activities as a biomarker of exposure to ChE-inhibiting substances. DoD personnel potentially at risk for occupational exposure to nerve agents as a result of their duties at storage depots, chemical demilitarization facilities, research laboratories, and chemical training schools are required to participate in a mandatory medical surveillance program (1 ). Baseline erythrocyte-ChE (RBC-ChE) activities for workers in this program are established as the mean of 2 RBC-ChE values obtained ⬎24 h but ⬍14 days apart. If the 2 values differ by ⬎0.05 delta pH units/h, a 3rd value is obtained and the baseline is calculated as the mean of all 3 values. Baselines are reestablished every 3 years, and any decreases or increases ⬎10% of a person’s baseline value trigger a medical evaluation and investigation. Additional RBCChE testing is performed in the event of a real or potential nerve agent exposure. A similar medical surveillance program is used by the California Department of Pesticide Regulation for agricultural employees who regularly apply ChE-inhibiting pesticides (2 ). These guidelines also require periodic RBC-ChE and plasma ChE monitoring (3 ). Several earlier studies have characterized the interindividual variation for RBC-ChE activities by use of the same method that is still used by the DoD. Limperos and Ranta (4 ) reported mean (SD) RBC-ChE activities of 0.76 (0.03) delta pH units/h for 52 men and obtained identical results for 49 women. Rider et al. (5 ) analyzed specimens

1 US Army Center for Health Promotion and Preventive Medicine, Aberdeen Proving Ground, MD. 2 US Army Occupational Health Clinic, Bluegrass Army Depot, Richmond, KY. † Current affiliation: Walter Reed Army Institute of Research, Silver Spring, MD. * Address correspondence to this author at: US Army Center for Health Promotion and Preventive Medicine, 5158 Blackhawk Rd., Aberdeen Proving Ground, MD 21010-5403. Fax 410-436-4117; e-mail coleen.weese@ us.army.mil. Received January 3, 2007; accepted April 17, 2007. Previously published online at DOI: 10.1373/clinchem.2006.085258

3 Nonstandard abbreviations: ChE, cholinesterase; DoD, US Department of Defense; RBC-ChE, erythrocyte ChE; RCV, reference change value.

1 Copyright (C) 2007 by The American Association for Clinical Chemistry

2

Lefkowitz et al.: Intraindividual Stability of ChE Activities

from blood bank donors and reported mean RBC-ChE activities of 0.766 (0.081) delta pH units/h for men and 0.750 (0.082) delta pH units/h for women. Both of these results are in excellent agreement with our recent findings (McCurdy SA, Arrieta DE, Henderson JD, Lefkowitz LJ, Reitstetter RE, Wilson BW, unpublished results). The reference interval for RBC-ChE values is 0.63– 0.89 delta pH units/h (6 ). Although there is generally good agreement in the literature as to the magnitude of interindividual RBC-ChE variation, the extent of intraindividual RBC-ChE variability remains controversial. Reports in the literature can be found showing that within a given individual RBC-ChE activities increase (7 ), decrease (8, 9 ), and remain the same over time (4, 5, 10 –15 ). Detailed knowledge of this physiological variable is important clinically because the frequency of RBC-ChE monitoring should be linked to the enzyme’s physiological variation or lack thereof. We retrospectively analyzed RBC-ChE values from workers at a single US Army depot that participated in a medical surveillance program requiring periodic RBCChE monitoring during a period of ⬎30 years. To our knowledge, this is the most comprehensive study of intraindividual RBC-ChE activities to date. The study further benefited from the fact that all RBC-ChE measurements were performed by the same medical laboratory technician.

Materials and Methods materials All chemicals were obtained from Sigma.

assay principle The method used by the US DoD to measure RBC-ChE activities, often referred to as the delta pH method, is based on a modification (16 ) of the Michel method (17 ). The assay monitors the decrease in pH that occurs when acetylcholinesterase catalyzes the hydrolysis of acetylcholine into choline and acetic acid. Comparison of the results from the delta pH method with those obtained with the assay of Ellman et al. (18 ) can be performed by use of the conversion factor established by Groff et al. (19 ). Because the delta pH assay has not been described in some time and important factors related to its adoption for use in clinical laboratories were previously omitted, we provide here a detailed description of the delta pH assay and the equipment that is currently used to perform RBC-ChE testing. During the course of this retrospective study, the methodology of RBC-ChE testing remained the same; however, several models of equipment were used. Fortunately, these changes do not appear to have introduced any systematic errors. Random error likely decreased as technological improvements were made in newer models of water baths, dual-syringe diluters, and pH meters.

specimen collection Blood was collected in EDTA Vacutainer tubes (Becton, Dickinson and Company) by trained phlebotomists. Specimens were centrifuged within 4 h of collection to separate erythrocytes from the plasma. The plasma and buffy coat were removed, and if the specimens could not be assayed immediately, they were maintained at 2– 8 °C for up to 14 days.

buffer preparation Stock buffer was composed of 10 mmol/L sodium barbital, 2 mmol/L potassium phosphate monobasic, and 383 mmol/L sodium chloride. The pH was adjusted with 0.1 mol/L hydrochloric acid to yield a final pH of 8.19 to 8.25 (for 2 L stock buffer, 10 mL 0.1 mol/L hydrochloric acid was used). To prepare working buffer, 0.12 g/L saponin was added to an appropriate amount of stock buffer, and the pH was adjusted by dropwise addition of 1.0 mol/L hydrochloric acid to yield a final pH of 8.05– 8.10.

dilution of erythrocytes A dual-syringe diluter (Hamilton 500 Series Microlab) was used to dilute 200 ␮L erythrocytes with 4 mL working buffer. The diluted specimens were placed in 16by 100-mm test tubes and equilibrated at 25 °C in a circulating water bath (Precision Scientific).

assay procedure

The reaction was initiated by the addition of 400 ␮L of 0.11 mol/L acetylcholine bromide (for a final substrate concentration of 10 mmol/L), and the change in pH was monitored for 17 min. The use of 10 mmol/L substrate in the delta pH method exceeds the optimal substrate concentration; however, this concentration of substrate contributes to a linear time course of the enzymatic reaction (17 ). All pH measurements were made with a semimicro combination pH electrode (Ross) attached to a pH meter (Corning 445). To adapt the assay for higher sample throughput, up to 51 samples can be analyzed in 38 min by staggering the addition of substrate. The results are reported in units of delta pH units/h.

selection of study population To evaluate the stability of intraindividual RBC-ChE, we collected data from the medical records of workers at a US Army depot who participated in a DoD-mandated medical surveillance program. Informed consent was not obtained before initiation of this retrospective study; all individually identifiable information was removed before completion of the study. This study was reviewed and approved for data analysis as an exempt human use protocol by the Walter Reed Army Institute of Research Human Use Review Committee. This particular depot was selected because it had a remarkably stable workforce and medical records were readily accessible as far back as 1975. We evaluated records of data collected over a 30-year time period, from January 1975 through March

Clinical Chemistry 53, No. 7, 2007

3

2005. Because we were interested in changes over time, we included personnel with at least 10 years of data. This criterion was met by 46 men who were employed for a mean of nearly 20 years. None of the women working at this facility had been employed for 10 years or had a total of 10 RBC-ChE measurements, so their data were insufficient for analysis.

statistical analysis We used routine statistical methods to calculate mean, SD, minimum, maximum, and range values. We calculated within-person slopes of RBC-ChE activities over time by use of least-squares regression. We assessed the distribution of data sets by the Kolmogorov–Smirnov test provided in SigmaStat, version 3.1.1 (Systat Software). We calculated the index of individuality (20 ) as CVI/CVG, where CVI ⫽ intraindividual CV and CVG ⫽ interindividual CV (21 ). We determined the reference change value (RCV) with the following equation: RCV ⫽ (2)1/2 ⫻ Zp ⫻ (CVA2 ⫹ CVI2)1/2, where Zp is the SD for a given probability of error and CVA is the analytical CV (22, 23 ).

Results To assess the stability of intraindividual RBC-ChE activities over time, we evaluated medical records from a US Army depot and selected only those workers with at least 10 years of employment history and at least 10 RBC-ChE measurements. These criteria provided reasonable longevity to assess intraindividual variation over time and resulted in a final study population of 46 male workers with a collective total of 1730 RBC-ChE measurements. The data for these workers are summarized in Table 1 in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/ vol53/issue7. A mean (median) of 38 (37) RBC-ChE values were recorded for each worker. The mean (median) employment duration was 20 (21) years. The mean RBC-ChE value was 0.77 (range, 0.59 – 0.99) delta pH units/h. The mean SD for all workers in Table 1 in the online Data Supplement was 0.03 delta pH units/h. These values are in excellent agreement with the results of previous studies that measured the normal population of RBC-ChE values by the delta pH method (4, 5 ). Some of the key data presented in Table 1 in the online Data Supplement are shown in Fig. 1. A histogram of mean RBC-ChE values from 46 male workers is shown in Fig. 1A. Assessment by a Kolmogorov–Smirnov test did not demonstrate significant lack of gaussian distribution. The distribution of within-person CVs for each worker is shown in Fig. 1B. Mean % CV was 3.9. The distribution of slopes for within-person RBC-ChE activities over time is shown in Fig. 1C. Mean slope was 0.0010 delta pH units/h per year. Individual RBC-ChE data points were plotted for selected workers in Fig. 2. The minimum observed slope (for worker 23) was ⫺0.0018 delta pH units/h per year (Fig. 2A). The maximum observed slope (for worker 27)

Fig. 1. Distribution of data from Table 1 in the online Data Supplement. The RBC-ChE results and statistical analysis shown in Table 1 in the online Data Supplement are displayed as histograms to aid in the analysis of the data and the assessment of gaussian distribution. Distributions are plotted for the mean RBC-ChE values from 46 male workers (A), the percentage CV for each worker (B), and the slope of worker RBC-ChE values over time (C).

4


was 0.0060 delta pH units/h per year (Fig. 2B). The slope for worker 41 was 0.00105 delta pH units/h per year (Fig. 2C). This slope was closest to the mean slope of 0.0010 delta pH units/h per year and is representative of the intraindividual stability of RBC-ChE activities over time. Based on our study population and the raw data in Table 1 in the online Data Supplement, CVI was found to be 3.9 and CVG was found to be 8.6, to yield an index of individuality of 0.45 (20 ). A low index of individuality (⬍0.6) indicates that individual-specific reference intervals should be used instead of population reference intervals (24 ). For RBC-ChE measured by the delta pH method, CVA was 2.0 (Lefkowitz LJ, unpublished results), CVI was 3.9 as described above, and Zp was set at 95% or 1.96 to give a RCV of 12.1%. Thus, if a test result differs by more than 12.1% from an individual’s mean value, there is a 95% probability that this result is statistically significant.

Discussion

Fig. 2. Stability of RBC-ChE activities over time. Scatter plots of RBC-ChE values vs time are shown for workers with the minimum (A), maximum (B), and mean (C) slopes. The scaling on the axes is uniform to compare the data from different plots. In this study, the start point was defined as January 1, 1975. The majority of workers had their 1st RBC-ChE value measured during the 1980s, and only 2 workers had data points between 1975 and 1980.

A unique aspect of this study was that we were able to retrospectively track serial RBC-ChE measurements for 46 male workers for a mean (median) of 20 (21) years, and all of the analysis was performed by the same medical laboratory technician. The mean CV for RBC-ChE in this set of 46 workers was 3.9%. Linear regression analysis of the data for each worker resulted in a mean slope of 0.0010 delta pH units/h per year. For a pH-based enzymatic assay carried out over 30 years, assay precision was excellent and a testament to the skill of the individual technician analyzing the specimens and to the robustness of the assay. To our knowledge, this study is the longest intraindividual study of RBC-ChE activities to date. Our findings conclusively demonstrate that RBC-ChE values are extremely stable over time and show very little variation even after 20 years. Deviation from baseline in a worker who has not been exposed to ChE inhibitors is likely due to laboratory error and not intraindividual variation, as Coye et al. (25 ) have previously suggested. Recently, there has been considerable confusion within the ChE field regarding the intraindividual stability of RBC-ChE values. A report by Lotti (26 ) indicated that RBC-ChE exhibits large intra- and interindividual variation. As we have shown here and others have shown previously (4, 5, 10 –15 ), intraindividual variation is small for RBC-ChE. It is possible that some of the earlier studies may have been hampered by experimental errors associated with measuring RBC-ChE activities. Although the US DoD relies on methods developed during the 1940s to measure RBC-ChE, this technique is reliable. During the 1970s, reports of unacceptably high interlaboratory variation in RBC-ChE led the DoD to implement a strict quality assurance program. The key components of this program include standardized technician training and certification, development of control materials, written standard operating procedures, standardized laboratory equipment and maintenance schedules, and creation of a DoD Cholines-

5

Clinical Chemistry 53, No. 7, 2007

terase Reference Laboratory that provides quality assurance analysis testing for primary RBC-ChE testing laboratories. In addition, the reference laboratory manages a quarterly proficiency testing program and conducts annual on-site audits at participating primary RBC-ChE testing laboratories. These measures have greatly improved the quality of RBC-ChE results, and other laboratories may benefit from adopting similar quality assurance measures. Recent reports indicate that interlaboratory reproducibility for ChE testing is still problematic within the civilian community (27, 28 ), although the situation appears to be improving (3 ). During the data collection phase of this study, we found a paucity of data from women. At this particular US Army depot, ⬃90% of the personnel that were required to participate in a medical surveillance program for nerve agent workers were men. The 10% that were female tended to either be recently hired (within the past 3 years) or leave the workforce before their male counterparts. We did not have data for any women who had been monitored for RBC-ChE activities for at least 10 years. Because the number of women who work with pesticides and nerve agents is increasing, it is vitally important to capture data on the within-person variation of RBC-ChE activities in this population group. This study will likely have important implications for policymakers. Currently, nerve agent workers are required to update their baselines every 3 years, and in California, pesticide applicators are required to update their baselines every 2 years for RBC and/or plasma ChE. Because we have shown herein that intraindividual RBCChE activities do not change appreciably with time, testing every 2 or 3 years may be unnecessary. On the basis of our study population and the raw data in Table 1 in the online Data Supplement, we calculated an index of individuality of 0.45 and an RCV of 12.1% for RBC-ChE. Currently, when there is a decrease or increase in RBCChE values for DoD personnel of ⬎10% from that person’s baseline RBC-ChE value, an investigation takes place to evaluate whether that person may have been exposed to any organophosphorous anticholinergic compounds. Typically, occupational health monitoring is more rigorous for those classes of employees at greatest risk of occupational exposure to hazardous conditions. For many classes of workers, such as infrequent visitors, first responders, and researchers working with dilute quantities of anticholinergic organophosphorous compounds, an initial baseline measurement may be sufficient. In light of continuing concerns regarding chemical terrorism, RBC-ChE measurements may become more widespread throughout military and civilian populations. The practicality of testing large numbers of people becomes more feasible when test values do not need to be repeated every 3 years. For workers with an increased risk of exposure to anticholinergic organophosphorous compounds, such as nerve agent workers at chemical depots and chemical

demilitarization facilities, more frequent monitoring may remain the normal procedure. Maintaining current practices in which baselines are reestablished every 3 years may be advisable for nerve agent workers in an industrial setting until larger-scale prospective studies corroborate both the male and female intraindividual stability of RBC-ChE.

Grant/funding support: None declared. Financial disclosures: None declared. Acknowledgements: We are grateful to Dr. Roger G. McIntosh for providing helpful guidance and inspiring us to initiate this retrospective study. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the DoD.

References 1. Department of the Army. Occupational Health Guidelines for the Evaluation and Control of Occupational Exposure to Nerve Agents GA, GB, and VX. Pamphlet 40-8. Washington, DC: Headquarters, Department of the Army, 1990. 2. Ngai W, Ames RG, Wisniewski J, Fan AM. Guidelines for Physicians Who Supervise Workers Exposed to Cholinesterase-Inhibiting Pesticides, 4th ed. Oakland, CA: Office of Environmental Health Hazard Assessment, California Environmental Protection Agency, 2002. 3. Wilson BW, Henderson JD, Arrieta DE, O’Malley MA. Meeting requirements of the California cholinesterase monitoring program. Int J Toxicol 2004;23:97–100. 4. Limperos G, Ranta KE. A rapid screening test for the determination of the approximate cholinesterase activity of human blood. Science 1953;117:453–5. 5. Rider JA, Hodges JL, Swader J, Wiggins AD. Plasma and red blood cell cholinesterase in 800 healthy blood donors. J Lab Clin Med 1957;50:376 – 83. 6. Department of the Army. Red Blood Cell-Cholinesterase Testing and Quality Assurance. Technical Bulletin Medical 590. Washington, DC: Headquarters, Department of the Army, 2001. 7. Sidell FR, Kaminskis A. Influence of age, sex, and oral contraceptives on human blood cholinesterase activity. Clin Chem 1975; 21:1393–5. 8. Allison AC, Burn GP. Enzyme activity as a function of age in the human erythrocyte. Br J Haematol 1955;1:291–303. 9. Gage JC. The significance of blood cholinesterase activity measurements. Residue Rev 1966;18:159 –73. 10. Sawitsky A, Fitch HM, Meyer LM. A study of cholinesterase activity in the blood of normal subjects. J Lab Clin Med 1948;33:203– 6. 11. Callaway S, Davies DR, Rutland JP. Blood cholinesterase levels and range of personal variation in a healthy adult population. Br Med J 1951;2:812– 6. 12. Augustinsson K-B. The normal variation of human blood cholinesterase activity. Acta Phys Scand 1955;35:40 –52. 13. Sabine JC. The clinical significance of erythrocyte cholinesterase titers. I. A method suitable for routine clinical use, and the distribution of normal values. Blood 1955;10:1132– 8. 14. Shanor SP, Van Hees GR, Baart N, Erdos EG, Foldes FF. The influence of age and sex on human plasma and red cell cholinesterase. Am J Med Sci 1961;357– 61.

6


15. Sidell FR, Kaminskis A. Temporal intrapersonal physiological variability of cholinesterase activity in human plasma and erythrocytes. Clin Chem 1975;21:1961–3. 16. Ellin RI, Burkhardt BH, Hart RD. A time-modified method for measuring red blood cell cholinesterase activity. Arch Environ Health 1973;27:48 –9. 17. Michel HO. An electrometric method for the determination of red blood cell and plasma cholinesterase activity. J Lab Clin Med 1949;34:1564 – 8. 18. Ellman GL, Courtney KD, Andres V, Featherstone RM. A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol 1961;7:88 –95. 19. Groff WA, Kaminskis A, Ellin RI. Interconversion of cholinesterase enzyme activity units by the manual delta pH method and a recommended automated method. Clin Toxicol 1976;9:353– 8. 20. Harris EK. Effects of intra- and interindividual variation on the appropriate use of normal ranges. Clin Chem 1974;20:1535– 42. 21. Petersen PH, Sandberg S, Fraser CG, Goldschmidt H. Influence of index of individuality on false positives in repeated sampling from healthy individuals. Clin Chem Lab Med 2001;39:160 –5.

22. Harris EK, Yasaka T. On the calculation of a “reference change” for comparing two consecutive measurements. Clin Chem 1983; 29:25–30. 23. Ricos C, Cava F, Garcia-Lario JV, Hernandez A, Iglesias N, Jimenez CV, et al. The reference change value: a proposal to interpret laboratory reports in serial testing based on biological variation. Scand J Clin Lab Invest 2004;64:175– 84. 24. Petersen PH, Fraser CG, Sandberg S, Goldschmidt H. The index of individuality is often a misinterpreted quantity characteristic. Clin Chem Lab Med 1999;37:655– 61. 25. Coye MJ, Lowe JA, Maddy KT. Biological monitoring of agricultural workers exposed to pesticides. I. Cholinesterase activity determinations. J Occup Med 1986;28:619 –27. 26. Lotti M. Cholinesterase inhibition: complexities in interpretation. Clin Chem 1995;41:1814 – 8. 27. Wilson BW, Padilla S, Henderson JD, Brimijoin S, Dass PD, Elliot G, et al. Factors in standardizing automated cholinesterase assays. J Toxicol Environ Health 1996;48:187–95. 28. Wilson BW, Henderson JD, Ramirez A, O’Malley MA. Standardization of clinical cholinesterase measurements. Int J Toxicol 2002; 21:385– 8.