Quality Control Specimens Research Articles

Accuracy of five on-site immunoassay drugs-of-abuse testing devices.

Many current "on-site" urine drug-testing products claim performance equivalent to laboratory testing. Five commercially available products (PharmScreen, Roche TestCup, Accusign DOA 2, Status DS, and American Bio Medica-Rapid Drug Screen) were challenged with quality-control specimens of known drug metabolite concentrations, 25% above and 25% below the SAMHSA cutoffs, and with known positive and negative donor specimens previously analyzed by immunoassay and gas chromatography-mass spectrometry. The results indicate discrepancies between claims and performance for all products, particularly with amphetamines. The implications for employer-based drug testing are discussed.

Quality Assurance of Energy Dispersive Spectrometry Systems

Monitoring the performance capabilities of energy dispersive X-ray spectrometers (EDS) and related x-ray analysis electronics and software is important for determining and improving the reliability, sensitivity, and accuracy of the x-ray analysis system. In addition, there is a growing popularity of quality systems through laboratory accreditation and ISO 9000 related programs that require set quality control procedures for analytical instrumentation. Having similar standard procedures amongst labs would allow direct intercomparison of results. This intercomparison would help labs and manufacturers determine what are normal versus abnormal results and lead to higher quality instruments and analyses. We have been developing a standard operating procedure for the characterization of EDS x-ray analysis systems on electron beam instruments.We are designing the procedure to maximize the efficiency of each quality control (QC) measurement so that we spend as little time monitoring the analysis system as is possible. We first chose useful QC specimens and then designed data collection methods.

Reducing the cost of cyclosporin assays: modification of the EMIT 2000 method.

Cyclosporin is the leading immunosuppressant agent in organ transplantation, and therapeutic drug monitoring forms an integral part of patient management in most institutions. In the authors' laboratory, the cost of cyclosporin assays represents a major fraction of total consumable expenditure. At present, an average of 4,300 patient cyclosporin assays are performed annually using the EMIT 2000 method (Behring-Syva) on the Cobas Mira analyser (Roche), at a cost of AUD$50,000 in kits alone. As a means of reducing laboratory costs, the manufacturer's recommended method was modified by decreasing all of the reagent and sample volumes in the "Analytical" section of the Cobas Mira cyclosporin programme by 33%. Assay performance was monitored over a 10-month period and compared to that of the unmodified method. Calibration curves were stable, requiring a one-point correction on average of once every 12 days, and a full calibration once ever 1.7 months. Interassay variability was not different to that previously reported for the unchanged method, with mean (SD, CV) concentrations for trilevel quality control specimens of 86.5 micrograms/L (10.2, 11.9%), 185.9 micrograms/L (11.4, 6.2%) and 408.5 micrograms/l (28.9, 7.1%). From 24 specimens assayed in an international quality assurance programme, the results of 23 were within 1.2 SD of the group mean for the EMIT method, with an average bias of 0.8%. With the current modifications, we were able to perform an average of 105 patient assays per kit compared to the previous 71, equating to an annual saving the AUD$16,600.

Assessment of a Critical Limit Protocol for Point-of-Care Glucose Testing

A critical limit protocol requiring that all point of care glucose meter readings > 22.2 mmol/L (400 mg/dL) and < 2.2 mmol/L (40 mg/dL) be immediately confirmed by the laboratory was assessed. A total of 193 (2%) of 9,523 glucose meter determinations (63 patients) were > 22.2 or < 2.2 mmol/L. One hundred twenty-two (63%) of critically high and low glucose readings were followed up, and 71 (37%) results were not. Seventy-seven percent (55 of 71) of results without follow up were in patients with multiple glucose meter/central lab comparisons, suggesting that users may have thought it unnecessary to confirm such results. Split sample quality control specimens showed good correlation (r = 0.927) between glucose meter and central lab results, whereas correlation for follow-up glucose results was poorer (r = 0.793), perhaps reflecting time delay in obtaining a lab sample. For follow-up results, only 18% of high/low critical limit glucose meter readings were confirmed by drawing a lab specimen within 10 minutes. Fifty-eight percent were in 17 patients with multiple previous glucose meter readings, suggesting that users may have though it less urgent to confirm a sequence of such results. Eleven follow-up results (9%) showed a > 50% discordance between glucose meter/central lab with three (27%) glucose meter errors, emphasizing the need to confirm critically high/low glucose meter results to avoid potential errors. The critical limit protocol now requires that only the initial critically high/low glucose meter reading be confirmed by the lab and that these patients now be followed with lab values until glucose levels are between 5.6-16.7 mmol/L (100-300 mg/dL) before the glucose meter can again be used.

Disparities in Clinical Laboratory Performance for Blood Lead Analysis

To evaluate the validity of blood lead analysis for clinical specimens. We submitted blood lead samples with a known lead concentration, in a blinded fashion, as clinical specimens to 18 laboratories. These laboratories were surveyed for the following characteristics that were hypothesized to be related to assay validity: laboratory ownership (state vs private), participation in the Centers for Disease Control Blood Lead Proficiency Program, assay method, and price. Each laboratory received 6 specimens with an actual blood lead (ABPb) concentration of 0.43 mumol/L (9 micrograms/dL) and 3 additional specimens--each with an ABPb concentration of 0.33, 0.89, and 1.59 mumol/L (6.9, 18.4, and 32.9 micrograms/dL, respectively). Misclassification error rates for reporting an elevation ( > or = 0.48 mumol/L [ > or = 10 micrograms/dL) in the blood lead concentration, the within-laboratory mean and coefficient of variation (CV) (for multiple specimens with an ABPb concentration of 0.43 mumol/L [9 micrograms/dL]), and the adjusted odds of a reported blood lead concentration differing from those of an ABPb concentration by more than 0.14 mumol/L (3 micrograms/dL). Blood lead results were obtained for 157 of 162 submissions. One laboratory reported all blood lead specimens as "below 0.48 mumol/L (10 micrograms/dL)." Two (11%) of 18 specimens with an ABPb concentration of 0.89 mumol/L (18.4 micrograms/dL) and 1 (6%) of 17 with an ABPb concentration of 1.59 mumol/L (32.9 micrograms/dL) were classified as below 0.48 mumol/L (10 micrograms/dL); 2 (11%) of 18 with an ABPb concentration of 0.33 mumol/L (6.9 micrograms/dL) and 44 (42%) of 104 with an ABPb concentration of 0.43 mumol/L (9 micrograms/dL) were classified as 0.48 mumol/L or greater ( > or = 10 micrograms/dL). For specimens with an ABPb concentration of 0.43 mumol/L (9 micrograms/dL), the within-laboratory mean ranged from 0.23 to 0.52 mumol/L (4.8-10.7 micrograms/dL); the CV ranged from 3% to 37%. Laboratories that used anodic stripping voltammetry were 6.3 (95% confidence interval, 1.4-28.6) times more likely to report a specimen that differed from the ABPb concentration by more than 0.14 mumol/L (3 micrograms/dL) than those that used atomic absorption methods. No other laboratory characteristic predicted discordance between the reported blood lead and ABPb concentrations. This study documents wide variation in the validity of the blood lead measurement among clinical laboratories. While the performance of some laboratories far exceeded the criteria of the Centers for Disease Control Blood Lead Proficiency Program, others made large errors that could have resulted in the false-negative misclassification of children with significant lead exposure. Given these differences, the purchasers of laboratory services may require access to laboratory proficiency data to make rational choices among clinical laboratories. Further study of laboratory performance on clinical specimens is required to determine if order-of-magnitude errors occur with sufficient frequency to warrant routine submission of blinded quality control specimens by proficiency programs and to determine the cause of the poor performance of laboratories that used the anodic stripping voltammetry methodology.

Determination of O 6-benzylguanine in human plasma by reversed-phase high-performance liquid chromatography

A high-performance liquid chromatographic assay for O 6-benzylguanine utilizing liquid-liquid extraction and reversed-phase chromatography has been developed. Plasma samples were alkalinized, extracted into ethyl acetate, evaporated, and the residues were constituted and chromatographed. Separation was accomplished by gradient elution with a mobile phase of methanol, acetonitrile, and phosphate buffer, pH 3.2. Eluted compounds were detected spectrophotometrically at 280 nm. Sample quantitation was obtained from the regression line of six-point standard curves ranging from 25 to 400 ng/ml. O 6-Benzylguanine peak heights were compared to peak heights of O 6-( p-chlorobenzyl)guanine (internal standard). The average regression coefficient was 0.999 ( n = 4). High concentration (305 ng/ml) and low concentration (38 ng/ml) quality control samples were determined with a day-to-day relative standard deviation of 7 and 8%, respectively ( n = 18). The within-day relative standard deviations were 2.7 and 3.0% ( n = 18) for the high and low concentration quality control specimens, respectively. Sample quantitation was reliable to 25 ng/ml with a signal-to-noise ratio of 8:1. This method was applied to plasma samples obtained from patients in a clinical trial of O 6-benzylguanine.

Reference materials and analytical standards to stimulate improved laboratory performance: Experience from the external quality assessment scheme for trace elements in biological samples

The performance of a large number of laboratories measuring trace elements in biological fluids has been monitored over many years by examination of their results in the Guildford external quality assessment scheme. Specific experiences of the UK trace elements reference laboratories have been used to stimulate improvements in performance of other participants in the scheme. The key features of these initiatives were: specially prepared reference materials, used as internal quality control specimens within a common procedure, contributed to accuracy control; proposed standards of satisfactory and unacceptable analytical performance associated with a new system for scoring; regular non-threatening open discussion of performance with interested colleagues. The impact of these features is illustrated with reference to measurements of Al and Zn in serum and Pb in whole blood.

Simultaneous Determination of Tacrine and 1‐Hydroxy‐, 2‐Hydroxy‐, and 4‐Hydroxytacrine in Human Plasma by High‐Performance Liquid Chromatography with Fluorescence Detection

An analytical method was developed for the simultaneous reversed‐phase (cyano) high‐performance liquid chromatographic determination with fluorescence detection of tacrine and 1‐hydroxy‐, 2‐hydroxy‐, and 4‐hydroxytacrine in human plasma. Alkalinized human plasma samples were extracted with a mixture of chloroform/l‐propanol (90:10, v/v). Extraction recoveries generally ranged from 68% to 83% for all four compounds and did not appear to be concentration dependent over the range examined. Calibration curves were constructed over the following ranges: 0.5–30.0 ng/mL for tacrine (THA) and 4‐hydroxytacrine (4‐OH‐THA), 1.0–30.0 ng/mL for 2‐hydroxytacrine (2‐OH‐THA), and 0.925–46.2 ng/mL for 1‐hydroxytacrine (1‐OH‐THA). The intra‐ and interassay precision (RSD) for the quality control specimens analyzed during validation were ⩽ 11.8% and ⩽ 9.28%, respectively. The intra‐and interassay accuracy (RE) for the quality control specimens analyzed during validation ranged from 14.1% to –7.5% and 12.1% to ‐3.33%, respectively, for all four compounds. The limit of quantitation was estimated as the level of the lowest concentration in the calibration curve for each compound based on acceptable interassay precision and accuracy statistics generated at this concentration. The sensitivity of the method was adequate for the determination of THA, 1‐OH‐THA, 2‐OH‐THA, and 4‐OH‐THA following oral administration of Cognex (40 mg single dose) to normal volunteers.

High-performance liquid chromatography determination of clonazepam in plasma using solid-phase extraction.

The use of high-performance liquid chromatography for therapeutic drug monitoring of clonazepam has previously been limited by low sensitivity and labor-intensive liquid-liquid extractions. The present method was developed employing a rapid solid-phase extraction, thus minimising sample workup and providing analytical sensitivity down to 2 micrograms/L using 1 ml of plasma. Plasma samples were loaded onto C18 solid-phase extraction columns, and clonazepam and its internal standard (methyl-clonazepam) were eluted with methanol, dried, and reconstituted in 130 microliters of mobile phase. Chromatographic separation was achieved using a 3-microns RP18 column at 40 degrees C and a mobile phase of 32% acetonitrile and 0.5% glacial acetic acid in distilled water at 0.5 ml/min. Detection was carried out using ultraviolet absorbance at 306 nm. Retention times for clonazepam and methyl-clonazepam were approximately 7 and 12 min respectively. Standard curves were linear over a range of 5-200 micrograms/L with intraassay coefficients of variation of 1.2 and 4.8% at 200 and 5 micrograms/L, respectively. Plasma concentrations measured in patient samples were not statistically different from those obtained using an established gas chromatographic method, and quality control specimens from the Heathcontrol EQA Scheme were consistently within +/- 1.2 SD of the group means. There was no chromatographic interference from other benzodiazepines or other drugs used for the treatment of epilepsy.

An assessment of the 4-6-20 rule for acceptance of analytical runs in bioavailability, bioequivalence, and pharmacokinetic studies.

A recent conference report described a decision rule, hereafter referred to as the 4-6-20 rule, for acceptance/rejection of analytical runs in bioavailability, bioequivalence, and pharmacokinetic studies. This procedure requires that quality control specimens at three concentrations (low, medium, and high) be assayed in duplicate in each run. For run acceptance, at least four of the six assay values must be within +/- 20% of their respective nominal concentrations, and at least one of the two values at each concentration must be within these limits. An inherent flaw in this decision rule is that the risk of rejecting runs, when the assay performance has in fact not deteriorated, varies for each assay and is neither known nor controlled. In this paper simulation methods are used to evaluate the operating characteristics of the 4-6-20 rule in comparison to those of classical statistical quality control procedures.

Screening for neonatal hypothyroidism with the Pharmacia ‘DELFIA’ TSH kit: four incubation protocols compared

To efficiently use the Pharmacia `DELFIA' neonatal thyrotropin (TSH) kit for a large-scale screening program, we compared the performance of the TSH results in dried blood spot specimens with four different incubation protocols: (i) 4 h incubation at room temperature with shaking; (ii) 16 h (overnight) at 4°C; (iii) 40 h (over two-nights) at 4°C; and (iv) 64 h (over three-nights) at 4°C. The overall precisions ranged from 13% to 8% for blood TSH concentrations 11.4 to 100 mIU/1. The accuracy evaluated with the Centers for Disease Control (CDC) quality control specimens 011, 012 and 013, were all within the 10% range of the CDC blood TSH-enriched value of 11.4, 18.2 and 36.4 mIU/1. The standard curve, precision, accuracy and frequency distribution studies of the four different incubation protocols for the ‘DELFIA’ neonatal TSH kit are all comparable. The four different incubation protocols can be interchanged with each other, which will add to the flexibility of this kit increase workload productivity for a large-scale screening program.

Quantitative measurement of clenbuterol at the femtomole level in plasma and urine by combined gas chromatography/negative ion chemical ionization mass spectrometry

A highly sensitive and specific assay was developed for the quantitative measurement of clenbuterol at the femtomole level in human plasma and urine. Clenbuterol and the internal standard (2H9)clenbuterol were measured by gas chromatography/negative ion chemical ionization mass spectrometry with methane as the reagent gas. The two compounds of interest were extracted from the biological samples at pH 13 using ethyl acetate. After two subsequent purification steps, the cleaned-up organic extract was derivatized with pentafluoropropionic anhydride. The mass spectrometer was set to monitor the abundant [M-HCl]- ions of the perfluoroacyl derivatives (m/z 368 and 377), which were generated in the ion source by an electron capture process. This assay required 1 ml of plasma or 0.5 ml of urine and the detection limit of the method was 5 pg ml-1 with a 12.8% relative standard deviation. The accuracy of the clenbuterol assay was also tested day to day with quality control specimens spiked blind to the analyst. The mean difference between the theoretical and actual values was lower than 4.1%.

Determination of calcium acetylhomotaurinate in human plasma and urine by combined gas chromatography-negative-ion chemical ionization mass spectrometry

A highly sensitive and specific assay has been developed for the determination of calcium acetylhomotaurinate and the internal standard (LM 3041) at the picomole level in human plasma and urine by gas chromatography-mass spectrometry with methane as the reagent gas. After a multiple-step extraction process, the cleaned-up organic extract was derivatized with pentafluorobenzoyl chloride at ambient temperature. Subsequently, chlorination followed by amidation of the sulphonic acid group led to the N-pentafluorobenzoyl di- n-butylamide derivatives. The mass spectrometer was set to monitor the abundant [M - HF] − ions ( m/z 424 and 438), which were generated in the ion source switched in the negative-ion chemical ionization mode. This assay required 1 ml of plasma or 50 μl of urine, and the detection limit was 1 ng/ml. The accuracy of the assay was tested day by day with quality control specimens spiked blind to the analyst. The mean difference between the theoretical and actual values was less than 8%.

Prevalence of antibodies to human immunodeficiency virus type 1 among blood donors prior to screening. The Transfusion Safety Study/NHLBI Donor Repository.

The Transfusion Safety Study (TSS) and the National Heart, Lung, and Blood Institute (NHLBI) established a repository of approximately 200,000 sera from blood donors in late 1984 and early 1985. Collections were made in the four metropolitan areas with the highest prevalence of AIDS. Retrospective testing showed an overall anti-HIV-1 prevalence of 16 cases per 10,000 donations. In this study, the predictive value of a negative initial enzyme-linked immunoassay was estimated from both quality control specimens and the rescreening of 13,461 sera to be greater than 99.99 percent with respect to technical error. Among anti-HIV-1-positive persons, there was a 1.3- to 1.5-fold excess of first-time donors. The anti-HIV-1 prevalence among donors showed that infection was more common among young men than suggested by national reporting of AIDS cases. Anti-HIV-1 prevalence varied among the four metropolitan areas less than did reported AIDS cases, but, by 1987, the differences in the latter had decreased. Anti-HIV-1 prevalence in collection areas outside of the four major cities differed much more widely than that among the cities themselves. The TSS/NHLBI Donor Repository will remain available for the indefinite future for further evaluation of screening procedures for HIV-1 and other viruses for which transfusion is found to be an important route of transmission.

MODEL FOR THE WAR ON DRUGS?

In 1982, after learning that 48% of its sailors aged 18 to 24 were using illicit drugs, the Navy declared war. The battle plan I directed from 1982 to 1984 reduced that percentage to less than 10% within two years; today it is less than 5%. The Navy has conducted 10 million drug tests each year. An independent survey taken in late 1983 revealed that 83% of the sailors in that age group cited random testing as the number one deterrent to drug abuse. Perhaps even more significant, about 26% said they probably would resume usage if the testing program were dropped. To cut down on "false positive," the Navy's program has called for 6,000 quality-control urine samples to be sent to the laboratory without its knowledge each year. Not one of the more than 36,000 such "blind negative" quality-control specimens tested to date has been incorrectly labeled positive by the laboratory. Have testing mistakes been made? Yes. However, the Navy instituted a thorough review system for any specimen that tests positive for illicit substances. It is based on a philosophy of "if there's any doubt, throw the case out," weighted in favor of the individual. The lab must never be the sole judge. The Navy has done all it can to make sure that the Fourth (search and seizure) and Fifth (self-incrimation) Amendments are protected. It has shown that a combination of compassion, sensitivity and clear, firm rules works.

The stability of human blood lead in storage.

Whole blood specimens from 12 occupationally exposed workers were stored for 10 weeks either in heparin or EDTA at 22 degrees, 4 degrees, and -20 degrees C for the study of the stability of lead in storage. The lead concentrations of these specimens ranged from 226 to 616 micrograms/L. Polypropylene and vacutainer tubes were used for storage. A Zeeman Graphite Furnace Atomic Absorption Spectrophotometer was used to determine the lead concentration. No loss of lead was found on the 12 specimens stored with heparin or EDTA at three different temperatures in all containers over the 10-week period. The variation of the lead concentration is comparable to the average precision of two quality control specimens, Tox-EL 1 and Tox-EL 2, of 7.4% and 6.1%. The use of the original vacutainer tubes with either heparin or EDTA and refrigerated temperature (4 degrees C) for the storage of human whole blood specimens for routine lead determination is recommended.

The implications of assaying external quality control sera under 'special conditions'.

Two identical series of Wellcome quality control specimens were analysed by different procedures. One series was entered into the laboratory in the same way as patients' specimens while the other was analysed under special favourable conditions. This involved assaying the samples in replicate in different batches, omitting transcription errors and outliers, and placing the sample in a position following a calibration standard or an internal control. It was found that a significantly higher position in overall league ranking was achieved by assaying the control under 'special conditions'.

Differences in reaction characteristics between human and bovine-based control materials for iron determination

Reaction rate studies were carried out on various quality control specimens using a ferrozine iron determination kit. The time required for complete color formation differed among the various materials tested. Quality control materials of bovine origin required a much longer incubation time for full color formation. Using a fixed incubation procedure, the bovine controls would give a falsely low result. The present study suggests that caution is indicated when quality control data using bovine materials are interpreted.

Comparison of the chromotropic acid and modified pararosaniline methods for the determination of formaldehyde in air

Follow-up tests were performed in 25 complaint homes previously investigated by the Wisconsin Division of Health to determine ambient formaldehyde concentrations. Four collection methods were compared in each home: (1) midget impingers containing double-distilled water and immersed in ice baths, (2) midget impingers containing 1% sodium bisulfite solution and immersed in ice baths, (3) midget impingers containing cold 1% sodium bisulfite solution (temperature not maintained with an ice bath), and (4) a refrigerated, complete sampling unit developed at Lawrence Berkley Laboratory which utilized double-distilled water as a collection medium. Air was drawn through the midget impingers using personal sampling pumps. All four types of sampling trains were operated simultaneously in a bedroom, collecting specimens within an area of about 4 ft 2, while in the kitchen or living room, all sampling trains except for the impingers containing 1% sodium bisulfite solution and immersed in ice baths were operated together. Analysis of variance indicated that the three samples collected using midget impingers and personal sampling pumps gave similar results. All specimens collected in water were analyzed using both the chromotropic acid and pararosaniline analytical methods. Quality control specimens prepared in the laboratory showed excellent agreement between the two methods; however, field specimens through which air had been drawn were assigned lower values using the pararosaniline method. Special precautions were taken to ensure that specimens collected in water were not contaminated with sodium bisulfite, a serious interference for the pararosaniline test. To rule out the possibility that specimens deteriorated between analysis, test results from each method were confirmed by repeating each method for selected sets of specimens for a period of several days. It was hypothesized that some source of interference was present in the field specimens which affected the pararosaniline analysis. Statistical tests were applied to determine whether the observed discrepancy between the analytical methods was consistent and a regression model was employed to determine whether any of the other environmental parameters measured during sample collection was associated with the discrepancy.

Suitability of control materials. General principles and methods of investigation.

We propose methods for characterizing the behavior of quality-control specimens. Candidate quality-control specimens and authentic patients' specimens were analyzed by various methods. Patients' specimens were chosen to be fully representative of those encountered, including subsets from persons who were healthy, had defined disease states, were in therapy, or whose specimens were lipemic, icteric, etc. The analytical methods chosen include those most commonly used as well as reference analytical methods. Procedures for characterizing the behavior of patients' specimens and candidate quality-control specimens are proposed and their applicability is demonstrated. The linear ratio method is a univariate graphical approach in which differences in accuracy among methods for any specimen or group of specimens are each displayed on a linear scale. Correspondence analysis is a descriptive multivariate statistical technique that allows both the specimens and the analytical methods to be characterized. The statistical techniques, in our application, allow the behavior of quality-control specimens to be assessed with respect to authentic patients' specimens without influencing the assessment process. Correspondence analysis provides a graphic representation by projecting both the specimens and the analytical methods on factorial planes. The appropriateness of te behavior of a quality-control specimen may be inferred from its position relative to those of authentic patients' specimens. These statistical techniques also provide some information regarding the specificity of analytical methods.