Olympus AU Research Articles

Small dense low density lipoprotein-cholesterol distribution in people with normolipidemia and hyperlipidemia and the correlation with serum lipids

Objective To investigate the distribution characteristics of serum small dense low density lipoprotein-cholesterol (sdLDL-C) levels in healthy normolipidemia and hyperlipidemia and analyze the correlation between sdLDL-C and other serum lipids. Methods Totally 1012 normolipidemic subjects (18-93 years old, 503 male, 509 females) were grouped according to gender and age (18-29, 30-69 and ≥70 years old).And 433 hyperlipidemic subjects (23-90 years) were divided into the following 3 groups based on fasting triglyceride (TG) and low density lipoprotein-cholesterol（LDL-C）levels:hypertriglyceridemia (n=165), high-LDL-C (n=129) and combined hyperlipidemia (n=139).The sdLDL-C and other serum lipids were measured by Olympus AU2700 automatic analyzer.Analysis of Variance and Kruskal-Wallis H test was used for statistical analysis. Results The distribution of the sdLDL-C levels in normolipidemic subjects was near normal distribution.The sdLDL-C levels had differences in gender and age.In the 18-29 and 30-69 years old group, the mean values of sdLDL-C were significantly higher in males than in females ［(0.55±0.21) mmol/L vs (0.47±0.22) mmol/L, t=2.212, P=0.028 and (0.66±0.28) mmol/L vs (0.62±0.25) mmol/L, t=2.121，P=0.034］.In the ≥70 years old group, the difference of sdLDL-C levels in gender was not statistically significant ［male (0.54±0.21) mmol/L vs female (0.54±0.22) mmol/L, t=0.022，P=0.982］; the mean value was (0.54±0.22) mmol/L. The hyperlipidemic subjects had conspicuous higher levels of sdLDL-C compared with normolipidemia ［(1.25±0.44) mmol/L vs (0.60±0.26) mmol/L, t=29.306，P＜0.001］.Among all of groups, the combined hyperlipidemia group had the highest sdLDL-C level ［(1.52±0.49) mmol/L, F=525.66, P＜0.001］; the hypertriglyceridemia group had the highest sdLDL-C/LDL-C level (0.47±0.12, F=287.93, P＜0.001) and the high-LDL-C group had the highest level of non-sdLDL-C ［LDL-C subtract sdLDL-C, (2.71±0.52) mmol/L, F=336.32, P＜0.001］.The sdLDL-C showed a good positive correlation with TC, TG, LDL-C, ApoB and BMI (rs=0.66, 0.68, 0.65, 0.79 and 0.27, P＜0.001), and negative correlation with HDL-C and ApoA1 (rs=-0.42 and -0.37，P＜0.001).Based on partial correlative analysis, sdLDL-C showed a different correlation with TG, LDL-C and TC (r=0.42, 0.28 and 0.15, P＜0.001). Conclusions LDL-C and TG are the important factors influencing the sdLDL-C levels.However, TG has greater effect than LDL-C.The sdLDL-C is an appropriate and good index to evaluate the small dense low density lipoprotein （sdLDL） mass and the overall situation of lipid metabolism.In order to make full use of sdLDL in the clinical treatment and health assessment, it is necessary to establish sdLDL-C reference intervals through the survey of distribution of sdLDL-C levels in different geographic areas. （Chin J Lab Med,2012,35:354-358） Key words: Cholesterol，LDL; Triglycerides; Hyperlipidemias; Autoanalysis

Comparison of Outcome Biases Regarding the Two Modes of Sample Aspiration Offered by the Olympus AU2700 Automatic Biochemical Analyzer

Objectives: To evaluate the comparability of the test outcome or the acceptability of the outcome bias regarding the 2 modes of sample aspiration offered by the Olympus AU2700 automatic biochemical analyzer and to determine whether calibration and internal quality control are required when using the sample aspiration mode for a statistics (STAT) table. Methods: Per the procedures outlined in the National Committee for Clinical Laboratory Standards–EP9-A2 protocol, 16 tests were performed for analytes such as alanine aminotransferase, total protein, and urea to evaluate and compare the outcome bias between the mode used with racks and that used with the STAT table. Forty serologic samples were tested. Results: Among the results of the 16 tests, almost all levels of outcome bias were acceptable. The sole exception was that regarding the TG test, in which system error reached 6.89%; this amount exceeds the 6.25% allowable percentage of error at lower concentrations. Conclusion: Based on these data, we conclude that calibration and internal quality control are unnecessary when using the sample aspiration mode of the STAT table.

Self-monitoring of blood glucose

Self-monitoring of blood glucose

Investigation of unusual high serum indices for lipemia in clear serum samples on Siemens analysers Dimension

We present our work of monitoring 202 different patients with markedly elevated serum index for lipemia whereby serum samples were clear. We tried to clarify the cause of occurrence of these indices which were detected in the years 2006-2010 on Siemens Dimension analyzers. In samples with unusual lipemia index we measured the concentration of lipids (total cholesterol, triglycerides, HDL and LDL cholesterol, Lp(a), ApoA1, ApoB), total proteins and checked for possible interferents (rheumatoid factor, immunoglobulins). We performed serum protein and immuno- electrophoresis. We investigated the repeatability of unusual lipemia indices during the day and after different time periods and we compared them on four different analyzers (RXL Max, Vista, Hitachi 911 and former Olympus AU640). In 87% of 202 samples we found a monoclonal or biclonal peak in serum protein electrophoresis. Different types of paraproteins were confirmed with immunofixation electrophoresis. In the remaining 13%, polyclonal elevated concentrations of immunoglobulins were measured. Other parameters had no influence on appearing of these indices. The repeatability of indices was good during the first day of measurements (P values > 0.05) and markedly lower in the next days or after 3 and 12 months (P values <0.05). The indices were elevated only on Dimension analyzers, but not on Hitachi and former Olympus analysers. A markedly elevated lipemia index in a clear serum sample measured on Siemens analyzers Dimension indicates a high possibility for the presence of a paraprotein in the sample.

Analytical performances of cystatin C turbidimetric assay: which impact on accuracy of glomerular filtration rate estimation in renal transplantation?

The potential use of cystatin C has recently emerged to evaluate kidney function following transplantation. Analytical performance of a particle-enhanced turbidimetric immunoassay (PETIA) using Dako Cytomation Cystatin C reagents on an Olympus AU640(®) analyzer was evaluated. Clinical relevance was determined by comparison with a reference method in a kidney transplant cohort. Repeatability and reproducibility were carried out for four levels of cystatin C. Linearity was done by successive dilution of plasma samples with high level of cystatin C. Comparison study was performed against the particle-enhanced nephelemetry immunoassay (PENIA) using Siemens reagents on BNII(®) system. Values of the glomerular filtration rate (GFR) estimated from several predictive cystatin C- and creatinine-based equations were compared to the GFR measured by isotopic method ((99m)Tc-DTPA). These predictive algorithms were analyzed with respect to bias, precision and accuracy. Total intra-assay and inter-assay variation coefficients were below 4% at the four levels tested. Values obtained with Dako-PETIA on AU640(®) were correlated with the Siemens-PENIA method (Dako-Olympus=0.88 Siemens+0.17). The cystatin C alone and the creatinine-cystatin C combined equation allowed reliable assessment of GFR in our population of renal transplants. The cystatin C-based or the combined equation moderately improved determination of GFR (about 10% compared to 175MDRD for classification into the correct K/DOQI CKD stages). The use of algorithms based on cystatin C and creatinine could provide a reliable estimate of the GFR in kidney transplantation. Standardization of cystatin C assays could further improve the predicting value of cystatin C in transplantation.

Method verification and comparison of urine oxalate assay by Olympus AU400 with 560 express plus analyzer

Method verification and comparison of urine oxalate assay by Olympus AU400 with 560 express plus analyzer

Method verification and comparison of beta-hydroxy-butyrate assay by olympus AU400 with 560 express plus analyzer

Method verification and comparison of beta-hydroxy-butyrate assay by olympus AU400 with 560 express plus analyzer

Do the filarial lymphoedema patients’‘3 months recall’ on acute dermato‐lymphangio‐adenitis (ADLA) correlate with streptococcal serology?

The aim of this study was to know the correlation of patients' 3 months recall on acute dermato-lymphangio-adenitis (ADLA) with anti-streptolysin O (ASO) serology and its application as a tool to know the burden of ADLA in the community. Fifty-nine lymphoedema (LE) patients and 27 age matched controls were clinically assessed for LE and the occurrence of ADLA during the previous 3 months was obtained by recall. After obtaining the informed consent, 2 mL of venous blood sample was collected and ASO was quantified in Olympus AU400 auto-analyzer. When the results were computed as two groups, controls and LE patients with no reported ADLA and LE patients with reported ADLA (by 3 months recall), the ASO positivity and ASO titre was significantly higher in the later group (P < 0.05). When the results were computed as three groups, controls with no reported ADLA, LE patients with no reported ADLA and LE patients with reported ADLA, the ASO titre was significantly higher in LE patients reported ADLA (P < 0.05). As ASO was measured in post-infection phase, we relied on the ASO titre for making conclusion. Patients' 3 months recall on ADLA correlates with the ASO titre and therefore, it could be considered as a tool to measure the burden of ADLA in the community. Multicentre community-based studies are needed to ascertain the findings.

Glucose and insulin dynamics in neonatal foals: influence of maternal diet

s / Journal of Equine Veterinary Science 31 (2011) 230-356 277 Nutrition: Oral Presentations II Glucose and insulin dynamics in neonatal foals: influence of maternal diet T.N. Dobbs , C.E. Foote , A.J. Cawdell-Smith , S.T. Anderson , R.C. Boston , and W.L. Bryden 1 1 The University of Queensland, Gatton, QLD, Australia, 2 Equine Consulting Services Pty Ltd, Glenorie, NSW, Australia, 3 University of Pennsylvania, Philadelphia, PA, USA Introduction: Pregnancy is characterized by a progressive decline in insulin sensitivity, an adaptation that parallels growth of the feto-placental unit, ensuring sufficient glucose supply to the fetus [1]. Therefore factors, including nutrition that influence maternal insulin sensitivity will alter glucose availability to the fetus. Dietary energy composition, specifically feeds high in nonstructural carbohydrates, are known to diminish insulin sensitivity, increasing risk for diseases such as obesity and laminitis in the horse [2]. Moreover, changes in glucose and insulin dynamics during gestation may predispose the foal to metabolic disorders associated with insulin resistance later in life. The current study investigates the relationship between the diet fed to the mare during the last trimester of pregnancy and the glucose and insulin dynamics of her subsequent foal. Materials and Methods: Pregnant Standardbred mares were used in the study and offered either a normal/low carbohydrate (16.0 Mcal/day of digestible energy (DE), n 1⁄4 6) or high carbohydrate (25.8 Mcal DE/day, n 1⁄4 6) concentrate diet during the last trimester of pregnancy (200-320 days of gestation) with the remainder of DE requirements available from forage. Diets were balanced for mineral content and satisfied NRC recommendations [3]. Monthly blood samples were collected in themorning (0700 to 0900h) to determine basal insulin and glucose concentrations in conjunctionwith body condition score (BCS) and body weight. The insulin-modified frequently sampled intravenous glucose tolerance test (FSIGT) [4] was applied to mares at 320 days of gestation. Neonatal foal FSIGTs (n 1⁄4 11) were conducted at 14 days of age. Results obtained from the FSIGT were used for minimal model analysis of glucose and insulin dynamics and determined parameters of insulin sensitivity (SI), glucose effectiveness (Sg), acute insulin response to glucose (AIRg) and disposition index (DI) [5]. Plasma insulin concentrations were assayed using RIA (CoatA-Count Insulin, Diagnostics Products Corp., Los Angeles, CA); validated for use with equine plasma [6]. Plasma glucose concentrations were analysed enzymatically (Hexokinase method, Olympus kit on an Olympus AU 400 Analyser, Diagnostics Systems Division, Melville NY). Means ( S.E) were calculated for plasma insulin and glucose concentrations, body weights, BCSs and minimal model parameters derived from FSIGT analysis. Statistical significance was assessed using parametric tests including one-way ANOVA with repeated measures for treatment and time comparisons and unpaired t-test for analysis of individual time points. Results and Discussion: odyweight and BCS increased in pregnant mares in the last trimester, with significantly (P 1 year) to determine long term effects of feeding different calorie sources. Therefore, the objective of this study was to test the hypothesis that young horses would have decreased insulin sensitivity when growing on feed higher in sugar and starch, and that further adaptations in insulin sensitivity due to time may become evident.

Accuracy and Precision Evaluation of Seven Self-Monitoring Blood Glucose Systems

Self-monitoring blood glucose (SMBG) systems play a critical role in management of diabetes. SMBG systems should at least meet the minimal requirement of the World Health Organization's ISO 15197:2003. For tight glycemic control, a tighter accuracy requirement is needed. Seven SMBG systems were evaluated for accuracy and precision: Bionime Rightest(™) GM550 (Bionime Corp., Dali City, Taiwan), Accu-Chek(®) Performa (Roche Diagnostics, Indianapolis, IN), OneTouch(®) Ultra(®)2 (LifeScan Inc., Milpitas, CA), MediSense(®) Optium(™) Xceed (Abbott Diabetes Care Inc., Alameda, CA), Medisafe (TERUMO Corp., Tokyo, Japan), Fora(®) TD4227 (Taidac Technology Corp., Wugu Township, Taiwan), and Ascensia Contour(®) (Bayer HealthCare LLC, Mishawaka, IN). The 107 participants (44 men and 63 women) were between 23 and 91 years old. The analytical results of seven SMBG systems were compared with those of plasma analyzed with the hexokinase method (Olympus AU640, Olympus America Inc., Center Valley, PA). The imprecision of the seven blood glucose meters ranged from 1.1% to 4.7%. Three of the seven blood glucose meters (42.9%) fulfilled the minimum accuracy criteria of ISO 15197:2003. The mean absolute relative error value for each blood glucose meter was calculated and ranged from 6.5% to 12.0%. More than 40% of evaluated SMBG systems meet the minimal accuracy criteria requirement of ISO 15197:2003. However, considering tighter criteria for accuracy of ±15%, only the Bionime Rightest GM550 meets this requirement. Because SMBG systems play a critical role in management of diabetes, manufacturers have to strive to improve accuracy and precision and to ensure the good quality of blood glucose meters and test strips.

Methodology comparison and bias estimation on two different biochemical detecting systems in myocardial enzymes

Objective To provide accurate and reliable experimental data by making methodology comparison and bias estimation to creatine kinase (CK) and creatine kinase MB isoenzyme (CKMB)in the two biochemical detecting systems of Olympus AU640 and Express Plus 560. Methods Take Olympus AU640 as the reference system (X), while Express Plus 560 as the laboratory system ( Y ),according to EP9-A file of NCCLS, detect controller of different levels and fresh mixed serum of CK and CKMB in the two biochemical detecting systems of Olympus AU640 and Express Plus 560, then mark down the detecting results, check the sample breakdown points,calculate the correlation coefficient and linear regression equations and makerelavant statistical analysis by referring to the data. Results The detecting results of both CK and CKMB show significant differences in the two systems ( P< 0.05 ).Differences of detecting results of CK are clinically acceptable, while the degree of precision of CKMB in this detection can not meet the clinical standard and there is no comparability for its clinical acceptability evaluation. Conolusion If there are more than two systems for the same test, it is necessary to do methodology comparison and bias estimation in order to insure the comparability of inspection ressults. Key words: CK; CK-MB; Detecting system; Methodology comparison; Bias estimation; EP9-A

Cystatin C is a reliable marker for estimation of glomerular filtration rate in renal transplantation: validation of a new turbidimetric assay using monospecific sheep antibodies

The potential use of cystatin C was recently assessed in kidney transplantation. A new particle-enhanced turbidimetric immunoassay (PETIA) that uses sheep antibodies (Binding Site human cystatin C immunoassay) has been developed. Analytical performance of this new assay was evaluated. Clinical relevance was determined by comparison with a reference method in a cohort of kidney transplant patients. First, the analytical performance of the Binding Site cystatin C kit was tested on SPA(PLUS)® and Hitachi® analyzers. Second, a comparison study was performed using SPA(PLUS)® analyzer against two other cystatin C methods (the Siemens-PENIA method on BNII® and the Dako-PETIA application on Olympus AU640®). Third, the glomerular filtration rate (GFR) was estimated using several predictive cystatin C- and creatinine-based equations and compared to GFR measured by an isotopic method (⁹⁹(m)Tc-DTPA). These predictive algorithms were analyzed with respect to bias, precision and accuracy. Total intra-assay and inter-assay coefficients of variation were below 5%. Values obtained with the SPA(PLUS)® correlated with the Siemens-PENIA and the Dako-PETIA methods. The creatinine and cystatin C-based equation allowed reliable assessment of GFR in our population of renal transplantation. The use of algorithms based on cystatin C and creatinine could provide a reliable estimate of GFR in kidney transplantation.

Serum biochemical values of mouflon (Ovis orientalis musimon) according to age and sex

Interest for breeding mouflons in Europe is growing by the day, so it is important to extend our knowledge regarding their physiology. To establish reference intervals of biochemical values for mouflon (Ovis orientalis musimon), serum samples from 35 males and 26 females were analyzed using the automatic analyzer Olympus AU 640 for 22 serum biochemistry parameters. All mouflons aged from 1 to 4 years were apparently healthy and in good body condition and kept at an average altitude of 87 m (45°4′N, 18°6′E). Obtained results were tested according to age and sex categories by the repeated measurement model with Tukey’s post hoc test. Data showed a significantly higher (P < 0.01) concentration of total serum bilirubin and chloride in adult pregnant female mouflons compared to adult male mouflons. A significantly higher (P < 0.01) value of sodium and a lower (P < 0.01) value of serum phosphate in adult pregnant female mouflons compared to adult male mouflons have been determined. These data provide the baseline information about serum biochemical parameters which can be used to manage the health in mouflons.

An appraisal of serum preptin levels in PCOS

Plasma preptin levels were higher in women with PCOS compared with the control group (mean±SD, 823.2±140.7 vs. 324.9±147.3 pg/mL). This article suggests a potential role of preptin in the pathogenesis of PCOS.

Analytical evaluation of the clinical chemistry analyzer Olympus AU2700 plus

Background : The objective of this study was to perform the analytical evaluation of the clinical chemistry analyzer Olympus AU2700 plus. The evaluation was performed according to the guidelines of the European Committee for Clinical Laboratory Standards (ECCLS). Materials and methods : The evaluation consisted of determination of within-run and between-run imprecision, inaccuracy and comparison with Olympus AU2700. The tested analytes were: glucose, creatinine, urate, total bilirubin, cholesterol, tryglicerides, calcium, phosphate, iron, unsaturated iron binding capacity, aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, creatine kinase, alpha-amylase, alkaline phosphatase, potassium, sodium, chloride, C-reactive protein, antistreptolysin O and rheumatoid factor. Results : The results showed low within-run and between-run imprecision and acceptable inaccuracy for all analytes. High correlation with AU2700 analyzer was found. Intercept and slope (with 95% con7 dence interval) met the preferences of Passing-Bablok regression for all of the analytes except alanine aminotransferase, sodium, rheumatoid factor, creatine kinase, total bilirubin and tryglicerides. Conclusions : Olympus AU2700 plus analyzer shows acceptable precision and accuracy for majority of analytes, with the exception of sodium and chloride. Instrument is fully comparable with Olympus AU2700 analyzer for all but several analytes (ALT, RF, sodium, CK, total bilirubin and tryglicerides), where only some minor deviations (constant and proportional di> erence) were observed.

Multicenter Evaluation of a Novel Nanoparticle Immunoassay for 5-Fluorouracil on the Olympus AU400 Analyzer

Background: 5-Fluorouracil (5-FU) is the most widely used chemotherapy drug, primarily against gastrointestinal, head and neck, and breast cancers. 5-FU has large pharmacokinetic variability resulting in unexpected toxicity or ineffective treatment. Therapeutic drug management of 5-FU minimizes toxicity and improves outcome. A nanoparticle-based immunoassay was developed to provide oncologists with a rapid, cost-effective tool for determining 5-FU plasma concentrations. Methods: Monoclonal antibodies, bound to nanoparticles, were used to develop an immunoassay for the Olympus AU400. Assay precision, linearity, calibration stability, and limit of detection were run at multiple centers; interference, cross-reactivity, lower limit of quantitation and recovery at 1 center. Clinical samples collected from 4 cancer centers were analyzed for 5-FU concentrations by liquid chromatography-tandem mass spectrometry and compared with the immunoassay results. Results: With calibrators from 0 to 1800 ng/mL 5-FU and autodilution, concentrations up to 9000 ng/mL could be determined. Time to first result was 10 minutes, and 400 samples per hour could be quantitated from a standard curve stored for >30 days. Imprecision across all laboratories was <5%, and the assay was linear upon dilution over the entire range. Cross-reactivities for dihydro-5-FU, uracil, capecitabine, and tegafur were <1%, 9.9%, 0.05%, and 0.23%, respectively. The limit of detection was 52 ng/mL with a lower limit of quantitation of 86 ng/mL. Assay results of clinical samples (93-1774 ng/mL) correlated with liquid chromatography-tandem mass spectrometry results: (R = 0.9860, slope 1.035, intercept 10.87 ng/mL). Conclusions: This novel immunoassay is suitable for quantitating 5-FU plasma concentrations with advantages of speed, small sample size, minimal sample pretreatment, and application on automated instrumentation. These advantages enable efficient therapeutic drug management of 5-FU in clinical practice.

Automation of the Hepascore and validation as a biochemical index of liver fibrosis in patients with chronic hepatitis C from the ANRS HC EP 23 Fibrostar cohort

Background Hepascore combining serum bilirubin, gamma glutamyl transpeptidase, hyaluronic acid (HA) and α2-macroglobulin with age and sex, was reported as relevant in predicting liver fibrosis in patients with chronic HCV infection and was proposed as an alternative to liver biopsy. Methods Since an automated HA assay (Latex method, Wako, Japan) became available, we investigated to automate Hepascore by simultaneous measurements of components using an OLYMPUS AU640 analyzer (Tokyo, Japan). For its clinical evaluation, we considered a cohort of chronic HCV patients included in a multicenter prospective study (ANRS HC EP 23 Fibrostar). Results Automated Hepascore was not significantly different than assayed as previously described. An improvement in HA variability was evidenced. In 512 chronic HCV patients, automated Hepascore, using ROC curves analysis, showed good predictive performances for significant fibrosis (AUROC = 0.81), severe fibrosis (AUROC = 0.82), and cirrhosis (AUROC = 0.88). For significant fibrosis, Hepascore (cut-off = 0.5) had a sensitivity of 0.77, a specificity of 0.70, a positive predictive value of 0.71 and a negative predictive value (NPV) of 0.77. Hepascore < 0.25 could exclude significant fibrosis with a sensitivity of 0.95 and a NPV of 0.90 and Hepascore < 0.75 could exclude cirrhosis with a sensitivity of 0.86 and a NPV of 0.97. Conclusions This study shows that Hepascore, a non-invasive index of liver fibrosis, necessitating only one serum sample, can be totally automated using a single analyzer and confirms that Hepascore accurately predicts liver fibrosis in chronic HCV. Hepascore might be largely used in assessing liver fibrosis as surrogate to the liver biopsy.

The BH3-Only Protein Bid Does Not Mediate Death-Receptor-Induced Liver Injury in Obstructive Cholestasis

The accumulation of bile acids during obstructive cholestasis causes liver injury and fibrosis, which is at least partly mediated by the death receptors Tumor necrosis factor-related apoptosis-inducing ligand, Tumor necrosis factor-alpha, and Fas. The BH3-interacting domain death agonist Bid is a critical mediator of death receptor-induced apoptosis in hepatocytes. Our aim for this study was, therefore, to elucidate whether Bid also mediates death receptor-induced liver injury in obstructive cholestasis. Overall, survival and various aspects of liver injury were analyzed in wild-type and Bid(-/-) mice after bile duct ligation (BDL), a commonly used model to study obstructive cholestasis in mice. Liver injury was examined at 3, 7, and 14 days after BDL. Loss of Bid did not affect the number of bile infarcts, serum aspartate aminotransferase values, or animal survival. Processing of procaspase-3 and procaspase-9, and caspase-3 enzyme activities, were not detectable in either group, and Bid(-/-) mice displayed the same pattern of terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling positive hepatocytes as wild-type controls following BDL. In contrast to Fas-receptor deficient lpr mice, hepatic fibrosis and the inflammatory response was not affected by loss of Bid. Together, these data suggest that Bid is not a downstream target of death receptors in obstructive cholestasis and does not significantly contribute to bile acid induced liver injury and fibrosis.

Investigators Did Not Follow Guidelines for Evaluating Point-of-Care Glucose Testing

In your January 2009 issue, Cook et al1 compared glucose meter results for central venous catheter (CVC) blood and fingerstick to a laboratory method for CVC blood, and observed meter error that purportedly would lead to faulty treatment decisions. Although we agree with the “split-sample” design (ie, testing portions of the same blood sample with the meter and laboratory method), we believe the data were confounded by the authors’ failure to follow Clinical and Laboratory Standards Institute (CLSI, formerly the National Committee for Clinical Laboratory Standards) recommendations in several crucial aspects.2The authors state that the average time from blood sampling to laboratory analysis was less than 1 hour. However, CLSI recommends either blood analysis or centrifugation and erythrocyte removal within 5 minutes of collection, because longer delays can produce marked changes in glucose concentrations due to glycolysis.3 The bias and imprecision of the laboratory instrument (Olympus AU640) were not established, duplicate meter and laboratory tests were not performed, and the hematocrit was not measured for each sample. (Chart recordings were used.) Some patients clearly had hematocrits below the labeled lower limit for the point-of-care meter system used in this study, the SureStepFlexx.The authors state that glucose meters are normally calibrated to capillary blood glucose and must be adjusted to measure venous blood glucose. However, no such adjustments are either necessary or desirable for SureStepFlexx, which, like most glucose meters, displays plasma-equivalent glucose results.Finally, the data analysis is problematic. The authors cite glucose differences greater than 20 mg/dL as significant. However, CLSI defines clinical accuracy as meter results within ±20% of a reference method for glucose =75 mg/dL (or within ±15 mg/dL for glucose <75 mg/dL). When we apply these criteria to Figure 1, most reported differences appear to meet the CLSI standard.In summary, this study contains methodological flaws that may have skewed the data. Investigators should follow published standards and guidelines for glucose meter evaluations.4 Furthermore, the authors provide no evidence for the premise that a device meeting current performance standards may lead to incorrect treatment.

Automated Determination of Serum α1-Antitrypsin by Antitryptic Activity Measurement

Alpha1-Antitrypsin (A1AT) deficiency is currently detectable by protein immunoassay, phenotyping, and genotyping of the S and Z mutations, but no fully automated method for standard biochemical analyzers is yet available. Here, we present a method that measures the antitryptic activity in serum. This method is rapid, automated, and allows the easy evaluation of a large cohort of patients. Our automated assay involves determining serum antitryptic capacity on the Olympus AU 400 autoanalyzer by using trypsin and succinylated gelatin as substrate in the presence of trinitrobenzene sulfonic acid. The results are expressed as a percentage of inhibition of the reaction of trypsin with succinylated gelatin. After we performed analytical validation studies and reference-interval determination based on serum samples from 120 healthy persons, we tested the assay on deidentified samples from 120 patients with various pathologies (primarily pulmonary) of unexplained origin and normal A1AT concentrations and phenotypes. The analysis rate was up to 120 samples per hour. Intraassay CVs ranged from 3.1%-16.2%, and interassay CV was 7.5%. The reference population showed mean (SD) 58.4 (6.7)% inhibition. The detection limit was 9.5% inhibition. The 120 studied patients displayed significantly lower mean activity than 120 healthy individuals (P < 0.0001). This assay is stable, reliable, and easily automated by use of open-system analyzers, allowing for the rapid evaluation of patients. After further validation on a larger randomized cohort, this new approach should function as a useful method to explore A1AT deficiency, especially in large-scale studies.