Analysis of Lead in Whole Blood Using the Thermo Electon X Series ICP-MS
M Matusiak
Citation
M Matusiak. Analysis of Lead in Whole Blood Using the Thermo Electon X Series ICP-MS. The Internet Journal of Laboratory Medicine. 2005 Volume 1 Number 2.
Abstract
Lead analysis has been predominantly performed by the Graphite Furnace Atomic Absorption (GFAA) method in the United States. In recent years, a new method of testing known as Inductively Coupled Plasma Mass Spectrometer (ICP-MS) has started replacing the GFAA method.
The study reported here analyzed the percent recovery rates for the Thermo Electron X Series ICP-MS. The premise was to use proficiency-testing data analyzed on the X Series ICP-MS and compare the percent recoveries to those of the consensus groups, all methods and the GFAA method. This study was performed on proficiency-testing samples that were run in a 30-month time frame.
The data suggests the method of testing on the ICP-MS provided recoveries equal to or exceeding the recoveries found on the GFAA. Further study of the method will need to be performed to definitively state the performance for use throughout the United States for whole blood lead testing.
List of Equipment
Thermo Elemental
X Series Inductively Coupled Plasma Mass Spectrometer
U.S. Headquarters
81 Wyman Street
Waltham, MA 02454
781-622-1000
Introduction
The clinical laboratory must be able to perform elemental analysis at the trace amount levels. Elemental analysis in clinical samples using whole blood has received increasing attention in recent years. As more studies are being performed on a pathogenesis of these elements, the demand on the clinical laboratory to perform elemental testing is and will continue to increase as more studies are performed.
Historically, the Graphite Furnace Atomic Absorption (GFAA) has been the standard for the testing of lead in whole blood since the 1980s. However, the GFAA's relatively long analysis times have been a detractor for many clinical laboratories, due to the reduction in personnel, both the lack of adequately trained personnel and the reduction in the number of staff. The GFAA methods are also very analyte dependant in sample preparation, causing increased technologist time performing both single and multiple elemental analyses. Additionally, the GFAA performs testing one analyte at a time, leading to multiple testing runs for patients requiring multiple elemental analyses1.
Serum has been widely used in the detection of elements in the blood. Nuttal, Gordon and Ash showed that the Inductively Coupled Plasma Mass Spectrometer (ICP-MS) could be used in a clinical laboratory using a variety of specimens, to include whole blood, serum and urine2. Since their study, several studies have been conducted on the use of the ICP-MS for trace elemental analysis. However, most studies have been done in a research setting and not a clinical laboratory setting.
The ICP-MS has several advantages over the GFAA. The detection limit for various elements reaches the 0.1 mcg/L level. The ICP-MS testing methods have simple sample preparation procedures. The ICP-MS also has a consistent high though put, about 40 samples per hour. The ability to test more than one element is a favorable advantage of the ICP-MS2.
One distinct disadvantage is the overall capital cost of the ICP-MS. The ICP-MS has a range of $150,000 to $350,000 in capital expenses, whereas the cost of the GFAA is between $20,000 and $50,000, based on the Governmental Services Administration contracts. Most clinical laboratories cannot afford to operate an ICP-MS for one or two basic elements. Additionally, the reimbursement rate from insurance and Medicare/Medicaid is not based off the more expensive ICP-MS. The overall cost of operating the ICP-MS is a deterrent for most clinical laboratories2.
A review of the literature from 1990 through 2005 showed several articles on the use of an ICP-MS for the detection of metals, which includes lead2,3,4,5,6. Furthermore, only three articles were found that specifically identified lead as the investigated metal7,8,9. No article was found comparing the relative percent recovery of the analyte with a known standard over time in a clinical laboratory setting.
Background on Lead Poisoning
The National Institute of Occupational Safety and Health (NIOSH) at the Centers for Disease Control and Prevention (CDC) has several analytical methods for the detection of lead using graphite furnace atomic absorption10). Increasingly, laboratories are using the ICP-MS and other instruments for whole blood lead analysis.
One reason for the increased use of the ICP-MS is that in order to determine elements at trace level values, sensitive analytical techniques are required. Trace level values are typically defined as less than 10 mcg/dL11. ICP-MS has been shown to achieve the level of sensitivity needed to obtain trace level values.
Lead in children is a very serious health concern. The ability for children to ingest lead has been well documented though out the past decades. The CDC noted that between 1999 and 2002 in children aged 1 through 5 years, 1.6% had blood lead levels greater than or equal to 10 mcg/dL, the CDC's level of concern for lead concentrations12.
Current studies are showing that lead ingestion is being found in places that normally would not be initially considered. In 2004, the Food and Drug Administration issued a warning that “The Food and Drug Administration (FDA) is aware of the problem associated with lead contamination of some Mexican candy products being sold in the United States and is advising parents, care providers and other responsible individuals that it would be prudent to not allow children to each these products at this time”13. Additionally, lead contamination was found in the wrappers of candy, causing the United States Consumer Protection Agency to send letters to produces of Mexican candy and importers of candy from areas outside the United States to halt future imports of candy until they [the importers] could ensure the candy wrappers did not contain lead or use lead contaminated ink14,15,16.
Lead toxicity affecting multiple organs is well established. Lead has been shown to mimic calcium, an important mineral in the body. This mimicking can affect processes that rely on calcium and can even stop those processes from occurring17.
The CDC reported that “acute high lead exposure can cause serious physiological effects, including death or long-term damage to brain function and organ systems”18. Additionally, lead is often noted for neurotoxic side effect, leading to other risk factors, such as aggressive behavior, social and school failure, hearing loss, hypertension, cardiovascular disease, renal disease and dental caries19.
These affects can be immediate or remain undetected for several years. Several longitudinal studies have found that neurodevelopmental delays and reduction in IQ even at low levels of lead exposure in children19,20,21,22,23,24,25,26. There is also research to support that the reduction in IQ is greatest when lead levels are low, defined as less than 10 mcg/dL19,26,27,28,29,30,31,32. Furthermore, research also shows that neurological damage appears to be irreversible19,27.
Lead and lead compounds have been recently listed as “reasonable anticipated to be human carcinogens” leading to other health risks33. Blood lead values are a concern in both children and adults. Current U.S. regulations by the Occupational Safety and Health Administration (OSHA) establish that the upper allowable limit for lead in adult blood is 500 mcg/dL34. The CDC has determined that blood lead values over 10 mcg/dL are a level of concern for children18. It must be noted that the CDC guidelines for blood lead levels in children are being continuously revised18. Current thought is that no threshold may be suitable for the level of lead in children19,22,35. It can therefore be determined that the blood lead level of concern should not be interpreted as a toxicity threshold35.
It seems probable that the more we learn about lead toxicity, lower limits of lead will become a greater concern and certainly the acceptable limits will decrease. This situation emphasized the importance to public health that analytical procedures that measure trace limits of metals in blood should be more readily available. An ICP-MS has the capability of measuring trace concentrations, and this analytical sensitivity makes it a vital tool in a clinical laboratory.
The ICP-MS Instrument
The ICP-MS was first obtained from an Induced Coupled Plasma instrument in 197836. The ICP-MS develops a high temperature (6000 K) plasma created by the energetic collisions between argon ions and atoms oscillating in a high-energy radio frequency field. The sample is delivered to the plasma via a peristaltic pump into a nebulizer where the sample is vaporized, atomized and ionized. The ions are then delivered to the mass spectrometer (operating in a vacuum) via the interface. The interface consists of two cones, the sample cone and the skimmer cone. These cones are separated by several millimeters and operating at a vacuum of approximately one torr. The sample cone must be cooled by external means due to the extreme heat of the plasma. The orifice of the skimmer cone opens directly into the vacuum chamber. A quadrupole mass filter is used to provide mass separation. A dual analog/pulse detector detects the ions that exit the quadrupole mass filter37.
Interfering Substances
The ICP-MS discriminates elements on the basis of mass-to-charge rations. Interfering substances primarily come from: (1) polyatomic ions, (2) elements with the same isotopic masses, and (3) doublely charged ions. Matrix effects may also produce inference in detection by suppressing, reducing, altering or enhancing signal ratios in the quadrupole detection system1. Thermo Elemental offers a third generation collision cell technology option that “provides enhanced performance leading to the highest signal/background” ratio1. However, this facility did not purchase this option.
This procedure was exclusively tested on whole blood for the detection of lead. According to Nuttall, Gordon and Ash, “elements >85 amu [atomic mass units] are well-suited for the ICP-MS analysis being essentially free of polyatomic interferences”2. Lead being of a high atomic weight, approximately 208 amu, is a good candidate for analysis by ICP-MS38. It should be noted that the National Institute of Standards and Technology uses ICP-MS technology to validate and determine lead concentrations in whole blood standards39.
Materials and Methods
Reference Sample Data
Data were gathered from proficiency-testing samples over the course of 30-months from the Marion County Health Department, Public Health Laboratory, Indianapolis, Indiana. The Wisconsin State Laboratory of Hygiene provided the proficiency testing samples. The Marion County Health Department, Public Health Laboratory participates in the monthly federally sponsored blood lead proficiency-testing program at the Wisconsin State Laboratory of Hygiene40. This program is approved as a proficiency-testing program under the Clinical Laboratory Improvement Act (CLIA) by the U.S. Department of Health and Human Services41. The Marion County Health Department has a Certificate of Compliance from the Centers for Medicare and Medicaid Services to perform human sample blood lead testing.
Procedure
The procedure used by the Marion County Health Department, Public Health Laboratory is a modification of the Lead by GFAA10 and the Environmental Protection Agency Method 200.8, Water Analysis by ICP-MS42.
Reagents
The reagents used were of high purity and quality. The gas used was Argon liquid 350 psi and/or 4.8 Grade compressed gas in T cylinders from Mittler Gas Supply, Indianapolis, Indiana. Trace metal grade concentrated nitric acid, traceable to NIST, was obtained from Fisher Scientific. Standards used were:
-
Tuning Solution (10mg/L) – CPI P/N 4400-133096, or equivalent containing Ba, Be, Ce, Co, In, Li, Mg, Pb, Rh, Tl, U, Y) – purchased from Fischer Scientific.
-
Lead 1000 mg/L – Spex CLPB2-2Y, or equivalent – purchased from Fisher Scientific.
-
Internal Standard (10mg/L) – Spex CLISS-1, or equivalent, containing Bi, Ho, In, Li6, Sc, Tb, Y) – purchased from Fisher Scientific.
-
Triton X 100, concentrated, or equivalent – purchased from Fisher Scientific.
-
Double Deionized (DDI) Water – manufactured in house by a Millipore Ultra-Pure Water System.
-
Ammonium Phosphate (ACS), dibasic (NH4)2HPO4, 99.99%, or equivalent – purchased from Fisher Scientific.
Reagent Preparation
The following reagents are used in this method and need preparation: (1) Tuning Solution, (2) Internal Working Standard, (3) Working Standard, (4) Matrix Modifier, (5) Rinse Solution, and (6) Cleaning Solution. Reagents preparation can be found in Table 1.
Figure 1
Proficiency Sample Preparation
For this study, proficiency-testing samples, from the Wisconsin State Laboratory of Hygiene, were used to determine the accuracy of the ICP-MS in whole blood lead determination. The samples are received as whole blood and need no rehydration or other preliminary preparation.
Proficiency samples are diluted in a 1:20 matrix modifier with Internal Standard added. The preparation of the samples is detailed in Table 2.
Figure 3
For routine clinical testing, any whole blood sample type (e.g. capillary whole blood or venous whole blood) can be used. Collection devices should be free from any lead contamination. The preparation of the whole blood sample is the same as the proficiency testing sample preparation.
Control Preparation
Three levels of control material from BioRad Laboratories, Inc. are diluted 1:20 in a matrix modifier with Internal Standard added. The manufacturer provides control ranges for the materials. Ranges were verified prior to use by analysis with known value controls. The preparation of the control samples are detailed noted in Table 3.
Figure 4
Calibration Preparation
A calibration curve in micrograms/deciliter (mcg/dL) was established on the instrument using aqueous standards of 0.0 mcg/dL, 1.2 mcg/dL, 1.5 mcg/dL, 3.0 mcg/dL and 6.0 mcg/dL. The preparation of the calibration standards is detailed in Table 4.
Figure 5
Instrument Configuration
The instrument employed in this laboratory is the Thermo Electron (Franklin, MA) X Series ICP-MS. Auxiliaries include a Cetac ASX-100 auto-sampler (AS), onboard peristaltic pump and PlasmaLab 3.51 software. The facility's instrument did not use the collision cell modes or the Peltier cooling unit. However, the facility has purchased the Peltier cooling unit for future testing. Both are optional purchases from Thermo Electron1.
The instrument used a Conikal Nebulizer (part number AR35-5-FC1E) from Glass Expansion, Pocasset, MA43. This nebulizer was used due to its “low-flow” uptake rate (1.0 mL/min). This was found to be advantageous when testing clinical samples from children, as the sample volume is typically less than 1.5 ml of whole blood. The ability to have a “low-flow” uptake rate allowed the conservation of sample volume in the event that repeat testing was required, without the need to re-prepare the sample.
The Cetac ASX-100 auto-sampler was used to transfer the sample from a sample cup to the nebulizer. A peristaltic pump was not used for sample uptake. The use of the CETAC ASX-100 auto-sampler allowed the sample to be taken to the nebulizer by negative pressure from the argon gas intake. This provided better control of sample entry into the nebulizer for analysis.
Instrument Parameters
The instrument was operated using parameters found in Table 5.
Figure 6
Isotopes measured for the pre-experiment report: 7Li, 24Mg, 25Mg, 26Mg, 56Ar O, 59Co, 138Ba++, 115In, 138Ba, 140Ce, 156 Ce O, 206Pb, 207Pb, 208Pb, 209Bi, 238U.
Isotopes measure for the experiment: 206Pb, 207Pb, 115In*, 159Tb*, Bi* (* = internal standard).
Total acquisition time for sample:
Sample Time: 18 seconds
Washout Time: 26 seconds
Data Acquisition Time: 30 seconds
The instrument was tuned and calibrated daily or when operated during the 30-month period of this study. During that time, some of the instrument parameters were altered, specifically the RF generator power, to ensure accurate operating conditions. The tuning of the analyzer was based on the performance of the calibration curve and control result values.
Performance Characteristics
Researchers studied the performance of the ICP-MS over the past decade. Over 50 elements have detection levels in the range of 0.01 to 0.1 mcg/L2. The detection level in this facility was determined by dilutions of 1:20, 1:40 and 1:83 of a manufactured control from BioRad Laboratories. Irvine, CA44.
Statistical Analysis
Statistical analysis was performed by Analyze-It Software (Version 1.71, 2003) for Microsoft Excel, Leeds, England, UK.
Results
Lower Linear Limit
Using the Bio-Rad manufactured whole blood lead controls, a lower linear reporting limit was calculated. The procedure used to determine the method detection limit is detailed in Appendix B of 40CFR136, Definition and Procedure for the Determination of the Method Detection Limit – Revision 1.1145. The ICP-MS detected the presence of lead in the dilution at the 1:83 dilution or a analyzed value of 0.10 mcg/dL. The calculated lower limit using the 1:83 dilution results would be 0.25 mcg/dL. However, the stated lower linear limit at this facility is 1.0 mcg/dL. Data is shown in Table 6.
Figure 7
Accuracy
Using the Bio-Rad manufactured whole blood lead controls, accuracy studies were performed, in accordance with the Clinical Laboratory Standards Institute, Standard EP-15-A, User Demonstration of Performance for Precision and Accuracy, Approved Guideline, Section 6.2, Recovery of Expected Values from Assayed Reference Materials46. Three levels of control samples were prepared in as described in Table 3 and analyzed in duplicate, daily for 10 days (n=20). Evaluation of the data showed that the method remained stable of this time period. Results are detailed in Table 7.
Figure 8
Precision
Using the Bio-Rad manufactured whole blood lead controls, precision studies were performed, in accordance with the Clinical Laboratory Standards Institute, Standard EP5-A2, Evaluation of Precision Performance of Quantitative Measurement Methods Approved Guideline, Second Edition47.
The samples were prepared in the same fashion as the patient/sample preparation method. All three levels of control material were used to validate the precision of the method.
The EP5-A2 precision statistical analysis (N=20) showed level 1's mean = 8.57, SD = 0.80 and CV = 9.3% (target value = 8.6, SD = 0.85); level 2's mean = 26.57, SD = 2.01 and CV = 7.7% (target value = 26, SD = 3); level 3's mean = 65.74, SD = 4.84 and CV = 7.3% (target value = 66, SD = 7). Data graphs on the precision of control levels 1, 2 and 3 are shown in Figures 1, 2 and 3, respectively.
Figure 9
Figure 10
Figure 11
Proficiency Testing Comparison
A total of 103 proficiency-testing samples were analyzed during a 30-month period. 6 results were excluded due to values below the linear reporting limit and reported as less than 1. 3 results were excluded due to a failure to perform sample error in the laboratory. As this facility participated in the monthly federally sponsored blood lead proficiency-testing program, the actual number of proficiency testing samples was higher than in a normal clinical laboratory. Since the monthly federally sponsored blood lead proficiency-testing program is approved for use under the Clinical Laboratory Improvement Act, consensus and reference values are determined and standardized for samples tested during this study. This federal oversight allowed the study to take place with little or no unknown variation in the proficiency testing sample preparation methods.
The Bland-Altman Plot showed was performed between target values and the ICP-MS method (N=94, 95% CI = -0.828 to 0.232, Bias = -0.298); between the ICP-MS method and the GFAA Method (N=94, 95% CI = -0.668 to 0.604, Bias = -0.032); between the ICP-MS and all reported methods (N=94, 95% CI = -0.356 to 0.760, Bias = 0.202). Results are shown in figure 4, figure 5 and figure 6, respectively.
Figure 12
Figure 13
{image:13}
The number of laboratories in the consensus groups varied over the 30 month period, based on the number of laboratories performing the analysis. However, the range of laboratories in the GFAA Method group ranged from 120 to 150, with a mean of 140. With the varying number of participants in the program, the SDI and CVR was unable to be determined for the consensus group. However, acceptable tolerances for acceptable proficiency performance for lead in blood is 4 ug/dL or 10% of the average of the peer and referee groups.48
Discussion and Conclusion
Recoveries were equate to or exceed those of GFAA. Furthermore, the recoveries are consistent with the recommended recovery targets as set by the CLSI. Agreement was shown between the consensus group for all methods, consensus group using the graphite furnace atomic absorption and the facility method using the Thermo Electron X Series ICP-MS. Overall, it can be concluded that the Thermo Electron X Series ICP-MS is an effective instrument for the detection of lead in whole blood. The instrument can analyze samples and provide acceptable recoveries of the actual known true values, as set by the CLSI. The ICP-MS is able to remain compliant with the regulations for proficiency testing during a 30-month period defined by the U.S. Department of Health and Human Services.
The ICP-MS is becoming an instrument of choice for the detection of lead in whole blood. The acceptance of the instrument and the accuracy and sensitivity of testing will become more defined in the future. The Thermo Electron X Series ICP-MS will be able to overcome the challenges of the future and provide quality testing in the process. It is a viable instrument for all clinical and research facilities performing whole blood lead testing.
Acknowledgements
I would like to acknowledge Virginia A. Caine, MD (Director, Marion County Health Department), Jyl Madlem (Manager, Clinical Laboratories, Marion County Health Department), Dave Czerny (Manager, Environmental Laboratories, Marion County Health Department) and the respective staff in the facility for their assistance in the development of this study.
Additionally, I would like to acknowledge John Olson, MD (Contact Pathologist, Marion County Health Department) for his assistance in medically reviewing this study and Simon Nelms (Thermo Electron Corporation, United Kingdom) for technical assistance with the instrument configurations.