Volkher Scharnhorst

Therapeutic drug monitoring of infliximab: performance evaluation of three commercial ELISA kits

Abstract Background: Therapeutic drug monitoring (TDM) of infliximab (IFX, Remicade®) can aid to optimize therapy efficacy. Many assays are available for this purpose. However, a reference standard is lacking. Therefore, we evaluated the analytical performance, agreement and clinically relevant differences of three commercially available IFX ELISA kits on an automated processing system. Methods: The kits of Theradiag (Lisa Tracker Infliximab), Progenika (Promonitor IFX) and apDia (Infliximab ELISA) were implemented on an automated processing system. Imprecision was determined by triplicate measurements of patient samples on five days. Agreement was evaluated by analysis of 30 patient samples and four spiked samples by the selected ELISA kits and the in-house IFX ELISA of Sanquin Diagnostics (Amsterdam, The Netherlands). Therapeutic consequences were evaluated by dividing patients into four treatment groups using cut-off levels of 1, 3 and 7 μg/mL and determining assay concordance. Results: Within-run and between-run imprecision were acceptable (≤12% and ≤17%, respectively) within the quantification range of the selected ELISA kits. The apDia assay had the best precision and agreement to target values. Statistically significant differences were found between all assays except between Sanquin Diagnostics and the Lisa Tracker assay. The Promonitor assay measured the lowest IFX concentrations, the apDia assay the highest. When patients were classified in four treatment categories, 70% concordance was achieved. Conclusions: Although all assays are suitable for TDM, significant differences were observed in both imprecision and agreement. Therapeutic consequences were acceptable when patients were divided in treatment categories, but this could be improved by assay standardization.

Differentiating transudative from exudative pleural effusion: should we measure effusion cholesterol dehydrogenase?

Abstract Introduction: Pleural effusions are often classified into transudates and exudates based on Lights criteria. In this study, the diagnostic properties of Lights criteria were compared to those of several other analytes for the classification of pleural fluids into transudative and exudative. Methods: A total of 471 patients with pleural effusions were evaluated. In pleural effusions and simultaneously drawn blood samples, lactate dehydrogenase (LDH), total protein, albumin, cholesterol, amylase, glucose, pH and the cell number were measured. Retrospectively, the clinical records were used to establish a clinical diagnosis. The diagnostic properties of the biochemical tests were calculated using the clinical diagnoses as gold standard. Results: By clinical diagnosis, 108 patients had transudative and 300 patients had exudative pleural effusions. In addition to pleural LDH activity (accuracy 89%, sensitivity 86%, specificity 97%) and fluid to serum LDH ratio (accuracy 89%, sensitivity 91%, specificity 85%), pleural cholesterol concentration readily identified exudates (accuracy 82%, sensitivity 76%, specificity 98%). Combination of these three parameters achieved a higher overall accuracy (accuracy 95%, sensitivity 93%, specificity 100%) than the Lights criteria (accuracy 93%, sensitivity 100%, specificity 73%). Combination of effusion cholesterol concentration and effusion LDH activity had the highest discriminatory potential (accuracy 98%, sensitivity 98%, specificity 95%). Conclusions: Including effusion cholesterol, concentration in the routine biochemical work-up of pleural fluid allows for correct classification of more pleural effusions than achieved by use of Lights criteria. Combination of cholesterol and LDH had the highest discriminatory potential and the added advantage that no patient plasma is needed for correct classification. Clin Chem Lab Med 2007;45:1332–8.

Electrolyte-balanced heparin in blood gas syringes can introduce a significant bias in the measurement of positively charged electrolytes.

Abstract Background: Heparin binds positively charged electrolytes. In blood gas syringes, electrolyte-balanced heparin is used to prevent a negative bias in electrolyte concentrations. The potential pre-analytical errors introduced by blood gas syringes are largely unknown. Here, we evaluate electrolyte concentrations in non-anticoagulated blood compared with concentrations measured in electrolyte-balanced blood gas syringes. Methods: Venous blood was collected into plain tubes. Ioni-zed calcium, potassium, sodium and hydrogen ions were analyzed directly using a blood gas analyzer and the remaining blood was collected into different blood gas syringes in random order: Preset (Becton Dickinson), Monovette (Sar-stedt) and Pico 50-2 (Radiometer). Results: Ionized calcium and sodium concentrations were significantly lower in blood collected in Becton Dickinson and Sarstedt syringes compared to non-heparinized (NH) blood. The mean bias exceeded biological variation-based total allowable error, which in most cases leads to clinically misleading individual results. In contrast, ionized calcium concentrations in blood collected in Pico 50-2 syringes were identical to values obtained from NH blood. Sodium showed a minor, yet statistically significant, bias. Conclusions: Despite the fact that blood gas syringes now contain electrolyte-balanced heparin, one should be aware of the fact that these syringes can introduce pre-analytical bias in electrolyte concentrations. The extent of the bias differs between syringes.

Analytical evaluation of a new point of care system for measuring cardiac Troponin I.

OBJECTIVESnPoint-of-care cardiac troponin testing with adequate analytical performances has the potential to improve chest pain patients flow in the emergency department. We present the analytical evaluation of the newly developed Philips Minicare cTnI point-of-care immunoassay.nnnDESIGN & METHODSnLi-heparin whole blood and plasma were used to perform analytical studies. The sample type comparison study was performed at 4 different hospitals. The 99th percentile upper reference limit (URL) study was performed using Li-heparin plasma, Li-heparin whole blood and capillary blood samples from 750 healthy adults, aging from 18 to 86years.nnnRESULTSnLimit of the blank, limit of detection and limit of quantitation at 20% coefficient of variation (CV) were determined to be 8.5ng/L, 18ng/L and 38ng/L respectively without significant differences between whole blood and plasma for LoQ. Cross-reactivity and interferences were minimal and no high-dose hook was observed. Total CV was found to be from 7.3% to 12% for cTnI concentrations between 109.6 and 6135.4ng/L. CV at the 99th percentile URL was 18.6%. The sample type comparison study between capillary blood, Li-heparin whole blood and Li-heparin plasma samples demonstrated correlation coefficients between 0.99 and 1.00 with slopes between 1.03 and 1.08. The method comparison between Minicare cTnI and Beckman Coulter Access, AccuTnI+3 demonstrated a correlation coefficient of 0.973 with a slope of 1.09. The 99th percentile URL of a healthy population was calculated to be 43ng/L with no significant difference between genders or sample types.nnnCONCLUSIONSnThe Minicare cTnI assay is a sensitive and precise, clinical usable test for determination of cTnI concentration that can be used in a near-patient setting as an aid in the diagnosis of acute myocardial infarction.

Do we measure bilirubin correctly anno 2005

Abstract We observed 30% discrepancy between liquid chemistry and dry chemistry analysers for the determination of total bilirubin in human adult serum samples, which were consistent with a 20% overestimation and 10% underestimation relative to a Jendrassik-Grof reference method, respectively. In contrast, standard reference material SRM916, which was recently recommended as being the most suitable material for attaining interlaboratory agreement, shows very good agreement on both types of analysers, as well as close to 100% recovery with respect to the reference method. We show that the liquid vs. dry bilirubin discrepancies seem to originate in the presence of either conjugated or δ-bilirubin. Our observations make it clear that good interlaboratory (or inter-analyser) agreement between bilirubin reference materials does not guarantee the same for bilirubin concentrations in human serum samples.

Multicenter evaluation of the commutability of a potential reference material for harmonization of enzyme activities

Abstract Standardization of laboratory results allows for the use of common reference intervals and can be achieved via calibration of field methods with secondary reference materials. These harmonization materials should be commutable, i.e., they produce identical numerical results independent of assay principle or platform. This study assessed the commutability of a cryolyoprotectant-containing harmonization material, obtained from the Dutch Foundation for Quality Assessment in Clinical Laboratories, that is intended to harmonize measurements of enzyme activities within the Dutch project “Calibration 2000”. The catalytic concentrations of alkaline phosphatase, aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, γ-glutamyltransferase and creatine kinase were analyzed in pooled patient sera and in the reference material in 14 laboratories. On liquid chemistry analyzers the harmonization material behaves like patient material. The enzyme activities measured in it fall on the regression lines calculated from activities measured in serum samples. For dry chemistry analyzers the activities of all enzymes measured in the harmonizator differ from the serum-based regression line. We show that this is due to the sucrose-containing cryolyoprotectant in the harmonization material. For each enzyme, correction factors were calculated that compensated for the bias and proved to be constant between reagent lots. Depending on the enzyme activity measured, application of these factors leads to 2- to 10-fold reduction of between-laboratory percentage coefficient of variation. Thus, additives to (potential) reference materials may alter their matrix in a way that interferes with analysis on certain test systems. The bias caused may be quantifiable and correctable. Establishment of correction factors leads to analytical uncertainties and costs. Therefore, matrix-based materials without additives should be selected as reference materials.

Moderate elevations of high-sensitivity cardiac troponin I and B-type natriuretic peptide in chronic hemodialysis patients are associated with mortality

Abstract Background: Several biomarkers are associated with mortality in hemodialysis patients. In particular, elevated cardiac troponin T and B-type natriuretic peptide (BNP) are strong predictors of mortality; however, less is known about cardiac troponin I (cTnI). Elevated troponin I is detected in many hemodialysis patients, but the association of moderate elevations with mortality is unclear. Methods: The relation between mortality and cTnI, using a high-sensitivity cTnI assay, as well as BNP and C-reactive protein (CRP) was evaluated in 206 chronic hemodialysis patients. Results: Median follow-up was 28 months with a total mortality of 35%. Mortality was significantly associated with elevated cTnI, BNP and CRP. Even patients with only moderate elevation of cTnI (0.01–0.10 μg/L) showed 2.5-fold increased mortality. Interestingly, hazard ratios for mortality for single (random) measurements were comparable to those for mean/median measurements. Subsequently, subgroup analysis based on combined markers was performed. Patients with both cTnI <0.01 μg/L and BNP in the first quartile had 100% survival. Patients with either cTnI <0.01 μg/L or BNP in the lowest quartile had significantly lower mortality (12% and 13%, respectively) than patients with BNP levels in the second quartile or higher and cTnI of 0.01–0.05 μg/L and patients with cTnI ≥0.05 μg/L (mortality 46 and 58%, respectively). Conclusions: A combination of moderate elevation of cTnI and BNP provided additional prognostic value. A single measurement of these biomarkers performed comparably to the mean/median of multiple measurements.

Urine flow cytometry and detection of glomerular hematuria.

Abstract Background: The UF-100 is a flow cytometer designed for automated cellular urinalysis. In this study, the usefulness of the UF-100 in laboratory investigation into the origin of hematuria was evaluated. Methods: Results from flow cytometric urinalysis were used to classify urinary red blood cells (RBCs) according to glomerular and non-glomerular origin and the classification was compared to the patients clinical diagnosis as the gold standard. In parallel, microscopic sediment analysis was carried out. Results: A total of 206 urine samples from 129 patients were analyzed (127 from patients with glomerular hematuria, 79 from patients with non-glomerular hematuria). Of these, 136 samples (92 patients) showed overt hematuria (≥20RBC/μL). Urine flow cytometry correctly classified 61% (sediment analysis 69%) of urine samples with overt hematuria. If inconclusive results are excluded, the UF-100 correctly diagnosed 85% (sediment analysis 98%) of urine samples with overt hematuria. The UF-100 and microscopic sediment analysis both showed sensitivity of 99% for the detection of glomerular hematuria. The specificity of the UF-100 for the detection of glomerular bleeding was lower (42%) than the specificity of microscopic sediment analysis (93%). Conclusions: Owing to its low specificity, the UF-100 showed limited capacity to discriminate glomerular from non-glomerular causes of hematuria in a population with a high incidence of renal disease. Therefore, extensive microscopic urinalysis remains necessary to assess the origin of hematuria. Clin Chem Lab Med 2006;44:1330–4.

Equal clinical performance of a novel point-of-care cardiac troponin I (cTnI) assay with a commonly used high-sensitivity cTnI assay

BACKGROUNDnEfficient rule-out of acute myocardial infarction (MI) facilitates early disposition of chest pain patients in emergency departments (ED). Point-of-care (POC) cardiac troponin (cTn) may improve patient throughput. We compared the diagnostic accuracy of a novel cTnI test (Minicare cTnI, Philips), with current POC cTnI (I-Stat, Abbott) and high-sensitivity central laboratory cTnI (hs-cTnI; Architect, Abbott) assays.nnnMETHODSnThe clinical performance of the assays were compared in samples from 450 patients from a previous clinical evaluation of Minicare cTnI.nnnRESULTSnMinicare cTnI correlated with Architect hs-cTnI (r2=0.85, p<0.0001) and I-Stat cTnI (r2=0.93, p<0.0001). Areas under the receiver operating characteristics curves were 0.87-0.91 at admission (p=ns) and 0.96-0.97 3h after admission (p=ns). The negative predictive values (NPV) at admission were 95% ((92-97%, 95% CI) for Minicare cTnI and increased to 99% (97-100%) at 2-4h, and similar to Architect hs-cTnI (98%, 96-100%), but higher than I-Stat cTnI (95%, 92-97%; p<0.01). Negative likelihood ratios (LR-) after 2-4h were 0.06 (0.02-0.17, 95% CI) for Minicare cTnI, 0.11 (0.05-0.24) for Architect hs-cTnI (p=0.02) and 0.28 (0.18-0.43) for I-Stat cTnI (p<0.0001). The clinical concordances between Minicare cTnI and Architect hs-cTnI were 92% (admission) and 95% (2-4h), with lower concordances between Minicare cTnI and I-Stat cTnI (83% and 78%, respectively; p=0.007).nnnCONCLUSIONSnThe Minicare cTnI POC assay may become useful for prompt and safe ruling-out of AMI in ED patients with suspected AMI using a guideline supported 0/3h sampling protocol.

Troponin I concentrations in heparinized plasma and serum differ when measured with the Advia Centaur TnI-Ultra assay

Cardiac troponins I (TnI) and T (TnT) are the preferred diagnostic markers of acute myocardial infarction (AMI) [1]. Most laboratories prefer heparinized plasma over serum for the analysis in order to minimize the turnaround time. Siemens Medical Solutions Diagnostics recently marketed a novel TnI-Ultra assay with increased analytical sensitivity. Siemens recommends the use of serum, heparinized plasma or EDTA plasma for the determination of the TnI concentration. The product sleeve [2] of the kit reports a 1 % lower TnI concentration in heparinized plasma than in serum, based on a comparison of serum and heparinized plasma samples obtained from 53 subjects. The Passing–Bablok equation reported was: [heparinized plasma TnI] 5 0.99 6 [serum TnI] + 0.03 mg/L. No details of the group of subjects are provided, and the heparin concentration and brand of tubes used are not reported. During the routine method validation performed at our laboratory, we compared the TnI analysis in heparinized plasma with serum analysis. Blood from 34 patients who had undergone coronary artery bypass grafting (CABG) surgery was drawn into plain gel serum tubes and heparin tubes. The samples were collected within the first 8 h after surgery. A maximum of two samples were collected per patient at different time-points and a total of 58 samples were collected. Subjects volunteering for the study were recruited according to hospital guidelines. Blood samples were collected in vacuum polypropylene tubes containing lithium heparin (BD Vacutainer, 4 mL, LH 68 IU Plus, BD Vacutainer Systems Preanalytical Solutions). The estimated heparin concentration in the sample is 17 IU/mL of whole blood, corresponding to 28 IU/mL in plasma based on a hematocrit of 0.40. Serum samples were collected in plain gel tubes (BD Vacutainer SST II Advance, 4 mL, BD Vacutainer Systems Preanalytical Solutions). Samples were centrifuged at 3000g for 10 min. Plasma was analysed within 4 h of storage at room temperature. The assay was calibrated in accordance with the manufacturer’s instructions. Data was analysed with Analyze-It version 1.72 (Analyze-It Software Ltd, UK). The 99th percentile limit of a selected reference population was 0.06 mg/L and a concentration >0.05 mg/L can be determined with (10 % total precision [3]. As shown in Figure 1, TnI results of heparinized plasma correlated well with the serum values, but were relevantly lower. The median concentration in heparin was 0.65 mg/L and in serum 0.75 mg/L. The difference between the median values was 0.10 mg/L (13 % of the median serum value), which is statistically significant (pv0.0001) as determined with the Wilcoxon signed-rank test. As is also shown in Figure 1, the negative bias of TnI in heparinized plasma as compared to serum was less in the concentration range v0.3 mg/L. Four samples showed a lower concentration in serum than in plasma. The concentration range of v0.3 mg/L represents the high-sensitivity range of the TnIUltra assay, which could not be determined with adequate precision by the predecessor cTnI assay of Siemens [3]. Several studies have reported significantly reduced TnI concentrations in heparinized plasma compared to serum. The Liaison cTnI assay (DiaSorin, Italy) showed a difference of 27 % (in 124 paired samples; 95 % confidence interval 24– 29 %), for which the heparin concentration was not described [4]. For the Immulite cTnI assay (Siemens Medical Solutions Diagnostics, Tarrytown, USA) a difference of 14 % was found (in 20 paired samples; 95 % confidence interval 9–19 %), using a heparin concentration of 60 IU/mL full blood [5]. Another study, however, showed an insignificant difference for the ACS:180 cTnI assay (Bayer Diagnostics, currently Siemens) [6]. The heparin concentration used in our validation is relatively low, i.e. 17 IU/mL in full blood. Relevant