Application of receiver operating characteristics curve (ROC) analysis when definitive and suspect morphologic flags appear in the new Coulter LH 750 analyzer.

Workflow efficiency is important in every laboratory. Manual assessment of white blood cell counts and differentials that have been rejected by an automated analyzer is one of the most time-consuming tasks in the routine hematology laboratory. In this study, receiver operating characteristics (ROC) curve analysis was used for the first time when anomalous distribution and suspect flag alarms appeared in hemograms carried out with the new Beckman Coulter LH 750 analyzer. This article is our second about the LH 750 analyzer published in this journal; we increased the number of cases and describe the novel application of statistical analysis of ROC curves. In processing of specimens from patients with 3% to 6% immature granulocytes (myelocytes + metamyelocytes + bands ), the suspect flag Imm Ne 1 (immature granulocytes) showed 77% diagnostic efficiency with a maximum area under curve (AUC) of 0.71 and a 95% confidence interval (CI) of 0.597 to 0.831 without significant differences between the 3 available levels of alarms in the analyzer (L1, L2, L2). In processing of specimens from patients with >6% immature granulocytes, the Imm Ne 1 flag showed superior diagnostic efficiency of 98% with a maximum AUC of 0.80 and a CI of 0.713 to 0.879. The suspect flag Imm Ne 2 in processing of specimens from patients with >6% of immature granulocytes showed diagnostic efficiency of 92% with a maximum AUC of 0.77 and a CI of 0.665 to 0.871, finding a significant positive difference in level L3 regarding sensitivity in comparison with the other 2 levels of the analyzer (L1, L2). For specimens from patients with >2% blasts, the suspect Blasts alarm showed a diagnostic efficiency of 94%, an AUC of 0.91, and a CI of 0.775 to 1.043; positive differences were observed between the levels L2/L3 and L1. In processing of specimens with variant lymphocytes (large, granular, prolymphocytes, cleaved, chronic lymphocytic leukemia type, and so forth) >10% (x = 14%), the suspect alarm Var Lym (variant lymphocytes) showed a low diagnostic sensitivity of 20% with a maximum AUC of 0.59 and a CI of 0.300 to 0.870 without significant differences between the 3 available levels (L1, L2, L3). However, in processing of specimens presenting values >10% reactive or activated lymphocytes (x = 23%), typical for patients with infectious mononucleosis, the Var Lym flag showed a superior sensitivity of 75% with a diagnostic efficiency of 92% and an AUC of 0.84 with a CI of 0.587 to 1.089. Finally, the laboratory can easily program definitive abnormal morphological flags of distribution (granulocytosis, eosinophilia, monocytosis, and so forth) on the basis of its patient population. In this study we were able to carry out comparisons of AUC and to choose the values for the automated counts in percentage, absolute value, or both. Therefore we were able to define the reliability and impact on the alarm routine to optimize the performance of the user-adjustable definitive alarms for anomalous distribution.

Use of the Advia 120 hematology analyzer in the differential cytologic analysis of biological fluids (cerebrospinal, peritoneal, pleural, pericardial, synovial, and others).

The centralization of our laboratories and the demand for new parameters to measure have led to an increase in the number of biological fluid samples, which are generally sent for urgent analysis. Due to this they cannot be processed by manual methods. Meeting this increased demand for assistance is a challenge for the laboratory, and the challenge has been met by the automated hematology area. A study of the reliability of the Advia 120 hematology analyzer has been carried out through leukocyte and red blood cell counting of 179 biological fluids: cerebrospinal, peritoneal or ascitic, pleural, pericardial, synovial, and others. The automated leukocyte counts of cerebrospinal fluid samples containing up to 0.150 x 10(9) leukocytes/L are correlated with counts obtained with the manual reference method in a Neubauer counting chamber (r = 0.958; P = .0001). Applying Passing-Bablok regression analysis to these results indicates a slope p of 1.155 (95% confidence interval [CI], 0.915-1.347) and an ordinate intercept b of 0.0076 (95% CI, 0.012-0.034), showing the results to be perfectly interchangeable. In the comparison of the manual analysis of the leukocyte differential using the May-Grünwald-Giemsa staining method with the analysis using the automated method, the percentage of polymorphonuclear granulocytes of the Advia 120 basophil/lobularity method is significantly correlated (r = 0.844; P = .0001) with that obtained with the manual count. The results of Passing-Bablok regression analysis (p = 0.859 [95% CI, 0.58-1.190]; b = 8.8 [95% CI, -12.09-24.2]) indicate that these two counting methods are also perfectly interchangeable. Automated leukocyte and differential counts of peritoneal or ascitic fluids also show good correlations with the manual method, and the results are not statistically different. Pretreating synovial fluid samples with hyaluronidase enzyme allows their processing on the Advia 120; no significant differences were found between manual and automated methods with respect to leukocyte counts and differentials. Finally, results with pleural fluid samples indicated that leukocyte and differential counts obtained with the Advia 120 showed significant differences from results obtained with manual methods because of the high incidence of mesothelial, lymphoid, and other tumoral cells in this kind of fluid sample. This result shows that use of hematology analyzers is questionable for these kinds of samples, especially from oncology patients with tumors. A procedure is proposed for the processing of these pleural fluids.

Validation of the Coulter LH 750 in a hospital reference laboratory.

Validation of the Coulter LH 750 was carried out in our central hospital laboratory, which processes 1500 hemograms per day for patients admitted into the 4 hospitals of our hospital complex and their corresponding outpatient departments. It is the reference laboratory for the provincial health care area. The analytical quality and the practical use of the instrument were studied, and we obtained within- and between-run imprecision estimates with reference controls of between 0.24% and 2.56% and inaccuracies of between -1.32% and 3.07% for basic hemogram parameters. Similar results were obtained with specimens from patients, with imprecision estimates between 0.56% and 2.56%. Linearity estimates were between 0.3 x 10(9)/L and 380 x 10(9)/L for leukocytes and between 3 x 10(9)/L and 1900 x 10(9)/L for platelets. The instrument evaluation was completed with a study of interference by bilirubin, lipemia, hemolysis, platelet clumps, and heparin and an examination of other variables, including carryover, detection limits, and the correlation of results with those of the Coulter Gen-S. A special evaluation was made of the new erythroblast count feature; with reference controls, imprecision estimates for this count were 10% to 12%, and with patient specimens imprecision averaged 10.39% up to 4 erythroblasts per 100 leukocytes. We also studied the correlation of LH 750 results and interferences with those of the manual reference method. Finally, National Committee for Clinical Laboratory Standards protocols were used to test for suspect and confirmation flags in leukocyte differential counts for 258 specimens representative of our routine. The practicability of the analyzer was studied in terms of technical difficulty, speed, and cost; also evaluated were new software elements for validation by source clinic and pathology and for reference values based on age. In conclusion, we analyzed the impact and improvements that may be expected in our laboratory (a user of Technicon and Coulter instruments) from introducing the new LH 750 analyzer into our routine.

Estudio de la influencia de la hemólisis en un perfil básico de coagulación / Study of the effect of hemolisys in a basic coagulation panel

Se han procesado 40 muestras de pacientes provenientes de cirugía ambulatoria con un perfil básico de coagulación compuesto por TP (tiempo de protrombina), TTPA (tiempo de tromboplastina parcial activado), fibrinógeno derivado, fibrinógeno VonClaus, TT (tiempo de trombina) y TR (tiempo de reptilase) y ATIII (antitrombina III). Seguidamente se ha realizado una hemólisis controlada "in vitro" de las muestras mediante la aplicación de ultrasonidos y se ha repetido el mismo perfil a las muestras hemolizadas, midiendo la cuantía de la hemólisis mediante la determinación de la Hemoglobina (mg/dl) del plasma sobrenadante. Aplicando la t de Student para muestras apareadas, entre los resultados de la muestra normal y la misma hemolizada se deduce que: las muestras hemolizadas muestran diferencias significativas en el TP medido por metodos ópticos a partir de 150 mg/dl de Hb. El fibrinógeno derivado, presenta descensos muy significativos de su concentración a partir de >75 mg/dl Hb de hemólisis. El TT se ve afectado también significativamente por un pequeño grado de hemólisis. Finalmente se analiza las alternativas tecnológicas que ofrecen los analizadores actuales de coagulación, para la resolución de los problemas de hemólisis, como puede ser la posible utilización alternativa de los métodos mecánicos de medición de la coagulación, frente a los métodos ópticos más generalizados, estableciendo una normativa para la realización correcta, de las pruebas de coagulación de las muestras hemolizadas que nos llegan al Laboratorio Central, remitidas sobre todo, desde centros de extracción periféricos. (AU)

Quantitative detection of hepatitis B virus DNA in serum by a new rapid real-time fluorescence PCR assay.

A sensitive and accurate HBV DNA quantification assay is essential for monitoring hepatitis B virus (HBV) replication. This study evaluated a real-time PCR method performed in the LightCycler analyser for quantitative HBV DNA assay. HBV DNA results with this method were compared with those obtained using a branched-chain DNA (bDNA) solution hybridization assay. Real-time PCR was performed using two adjacent fluorescently labelled probes and primers corresponding to the HBV core gene. The same standard employed in the bDNA assay was used for calibration. Serum samples came from 193 HBV surface antigen (HBsAg)-positive patients (34 HBV e antigen (HBeAg)-positive and 93 with antibody to HBeAg (anti-HBe)), and 66 asymptomatic HBV carriers. In addition, we analysed serum samples from 8 anti-HBe-positive patients who had been receiving lamivudine treatment for more than three years. A linear standard curve was seen in the range from 10(3) to 10(8) copies/mL. In the reproducibility analysis, intra-assay coefficient of variation (CVs) at two known HBV DNA concentrations were 4% and 2% and interassay CVs were 6% and 4%. The median of serum HBV DNA by real-time PCR was 9.2 x 10(8) copies/mL in HBeAg-positive patients with persistently elevated alanine aminotransferase (ALT) levels, 1.3 x 10(7) copies/mL in anti-HBe-positive cases with persistently elevated ALT levels, 3.7 x 10(4) copies/mL in anti-HBe-positive patients with fluctuating ALT levels and 10(4) copies/mL in asymptomatic HBV carriers. The differences in HBV DNA levels among the various groups studied were statistically significant (P < 0.05). The cut-off between chronic hepatitis patients and asymptomatic carriers was found to be at a serum HBV DNA concentration of 5 x 10(4) copies/mL. Of the 109 serum samples with a viral load < 7.5 x 10(5) (negative by bDNA assay) 44 (40%) were positive by real-time PCR: 24 (56%) chronic hepatitis and 20 (33%) asymptomatic carriers. There was a positive association between HBV DNA levels determined by real-time PCR and ALT levels (P < 0.05), which was not observed with the bDNA assay for HBV DNA quantification. At 12 months of lamivudine treatment, 6 patients (75%) showed HBV DNA levels < 5 x 10(4) copies/mL (range < 10(3)-2 x 10(3)), significantly lower than at baseline. At 36 months, 2 of 8 (25%) showed HBV DNA levels persistently lower than 5 x 10(4) copies/mL (1.7 x 10(3), 6 x 10(3)). The LightCycler quantitative real-time PCR is a practical, sensitive, reproducible single-tube assay with a wide dynamic range of detection. The assay is automatic except for DNA extraction and the running time is only 70 min. The LightCycler real-time PCR is useful for identifying different states of HBV infection and for evaluating the efficacy of viral therapy.

Role of hepatitis B, C, and D viruses in dual and triple infection: influence of viral genotypes and hepatitis B precore and basal core promoter mutations on viral replicative interference.

The interactions among hepatitis B virus (HBV), hepatitis C virus (HCV), and hepatitis delta virus (HDV) were studied by measuring HBV-DNA and HCV-RNA levels and by determining the influence of viral genotypes and mutations in HBV basal core promoter (BCP) and precore regions. We included 65 consecutive patients, 25 HBV/HCV, 18 HBV/HDV, and 22 HBV/HCV/HDV. Controls consisted of 55 patients with chronic HBV and 55 with chronic HCV infection. HBV-DNA and HCV-RNA levels were lower in coinfections than in single infections (P <.05). HBV/HCV coinfection was associated with lower HBV viremia (8.2 x 10(4) copies/mL) and lower HCV-RNA levels (7 x 10(5) IU/mL), than the corresponding control group (P <.05), with more marked decrease in HBV replication (P <.05). Moreover, in HBV/HCV coinfection and in triple coinfection we observed an inverse relationship between HBV-DNA and HCV-RNA levels (P <.05). HBV/HDV coinfection was associated with lower HBV viremia (2.5 x 10(4) copies/mL) than that found in HBV infection (P <.05). Patients with triple coinfection showed lower HBV-DNA and HCV-RNA levels than control groups (P <.05). Prevalence of precore mutations was lower in HCV coinfections (P <.05). No significant association was observed between HCV-RNA levels and HBV precore mutations, BCP mutations or HBV genotypes, or between HBV-DNA levels and HCV genotypes (P <.05). In conclusion, HCV exhibited stronger inhibitory action in the reciprocal inhibition seen in HBV/HCV coinfection. HDV was the dominant virus in HBV/HDV coinfection and in triple coinfection, and had a greater unfavorable influence on HCV than on HBV replication. The reciprocal inhibition of viral replication seemed to be little influenced by the inherent genomic factors studied.

Are equally spaced specimen collections necessary to assess biological variation? Evidence from renal transplant recipients.

The established method for determining the components of biological variation (BV) requires equispaced time intervals between samplings. In a previous study, we determined BV in renal post transplantation patients, taking advantage of the samples obtained within their clinical treatment protocol (not necessarily equispaced). To confirm the validity of this practice, we sought to determine if the use of varying sampling intervals has an effect on the results obtained in such biological variation studies. The study included two phases: comparison of the results found with identical and non-identical sampling intervals and correlation between the within-subject BV and the length of the sampling interval. There were no differences in within-subject BV between the groups or correlations with sampling intervals for any of the constituents studied. We conclude that samples acquired within established clinical protocols for kidney transplant recipients can be used for estimating BV.

External quality assessment of urine analysis in Europe. Results of a round table discussion during the symposium 'From uroscopy to molecular analysis', Seeon, Germany, September 18-20, 1999.

Interlaboratory surveys on urine quantities have only recently been introduced in several European countries. Representatives of 10 European countries exchanged their experiences during an international urinalysis meeting held in September 1999. Although still not mandatory in most areas, more than 5000 laboratories participated in external quality assessment programs in the countries represented. Qualitative (test strips and microscopic morphology) as well as quantitative chemical and immunochemical quantities were included. The maximal allowable deviations are reported as well as methods used to determine target values. Consensus scales up to reference methods were applied. The participants agreed that quality criteria need to be defined separate from those already existing for plasma/serum analytes. Besides higher biological variables and different medical needs, less standardisation of methods to quantify urine constituents was observed as a major cause of higher interlaboratory differences.

Fosinopril ameliorates exogenous cholesterol-induced incipient glomerular lesions in obese Zucker rats. Effects on eicosanoid secretion.

BACKGROUND: To date, the role of dietary cholesterol as a risk factor for some diabetic nephrophathy, such as mesangial expansion and glomerular lesions, is unknown. Controversy also exists regarding the effects of prostaglandin-induced changes in glomerular haemodynamics on the appearance of glomerulosclerosis. METHODS: We have used obese Zucker rats (OZRs) as a model of early nephrophathy to evaluate the effect of hypercholesterolaemic diet on glomerular prostaglandin secretion and on the development of glomerular lesions. Due to the role of angiotensin II (Ang II) in glomerular haemodynamics, we have also evaluated its effects on glomerular eicosanoid secretion. Furthermore, as it has been suggested recently by clinical studies that angiotensin-converting enzyme inhibitors (ACEIs) reduce serum lipids associated with proteinuria, we have also evaluated the effect of the ACEI, fosinopril, both in vivo and in vitro, using 24 h glomeruli cultures. RESULTS: Results showed that a cholesterol-rich diet significantly increased serum cholesterol, proteinuria and glomerular eicosanoid secretion, and caused macrophage-ED1 cell deposits in the glomerular mesangium. Segmentary lesions only appeared in those rats with the highest percentage of glomerular xanthomatous (macrophage-ED1) cells. Ang II, per se, caused a marked rise in glomerular prosaglandin E2 and thromboxane B2. The inhibition of Ang II synthesis with fosinopril reduced all the parameters listed above, whereas Ang II (10(-6)M) increased the secretion of TxB2 and tended to increase PGE2 secretion in glomerular culture. CONCLUSIONS: In conclusion, exogenous cholesterol per se may contribute to nephropathy by increasing eicosanoid secretion, serum lipid profile, urinary protein excretion and the development of glomerular lesions. Fosinopril reduced all these parameters probably by its effects on Ang II.

[Interference in blood coagulation tests on lipemic plasma. Correction using n-hexane clearing]. / Interferencias en las pruebas de coagulación sanguínea en plasmas lipémicos. Corrección mediante el aclaramiento con N-hexano.

PURPOSE: To eliminate the bias in the plasmatic coagulation test caused by lipemic samples by means of with addition of n-hexane in the ACL-3000 analyzer (Instrumentation Laboratory). MATERIAL AND METHODS: The turbidity caused by lipemia was simulated by addition of increasing quantities of Intralipid 20% (interferent) to samples from patients with normal pathological coagulation tests (with oral anticoagulant therapy). We compared the results of PT (prothrombin time), fibrinogen (derived) and APTT (activated partial thromboplastin time) before and after the addition of n-hexane. RESULTS: In samples with an elevated concentration of interferent (> 9.79 mmol/L of triglycerides) the analyzer failed to give any results. Once the lipids were cleared by n-hexane, the analyzer offered results for all the samples, which coincided with the original sample after applying a correction factor 1.06 to the TP and 0.88 to the TTPA. CONCLUSION: The addition of n-hexane is a valid method to eliminate the bias caused by the lipids in the coagulation tests.

European multicentre evaluation of the Super Aution SA-4220 urinalysis analyser.

A multicentre evaluation of the urine test strip analyser Super Aution-4220 was carried out in six laboratories. The analytical performance of the instrument with regard to imprecision, linearity, detection limit, drift, carry-over and method comparison was studied. Using the Aution stick 8 test strip the pH, glucose, protein, ketones, bilirubin, blood, urobilinogen and leukocyte esterase were analysed. Specific gravity measurements were performed by refractive index method. Within-run and between-run imprecision determined at three levels of analyte were good. No carry-over was observed. Obtained results were linear through all the described analytical range. No significant drift was detected. Method comparison with some quantitative methods was performed and showed a good correlation with most of the analytes. The study of interferences showed minor interferences by common therapeutic drugs with the measurement of some analytes. During the assessment period of about 6 months no breakdown occurred in any laboratory. The Super Aution urine analyser appeared to be a highly automated analyser of urinary test strips. The operation was simple and the maintenance required only a few minutes a day.

Potential interfering substances on Falcor-600 and Dax-48 analytical systems.

The increasing availability and use of automatic analysers in clinical chemistry have revealed a number of endogenous interferences. We evaluated the effect of bilirubin, haemolysis and lipaemia on the Falcor-600 analytical system (Menarini Diagnostics) and the Dax-48 (Bayer Diagnostic). We studied the potential endogenous interferences in the measurement of serum glucose, urea, creatinine, cholesterol, triacylglycerols, total bilirubin, total protein, aspartate aminotransferase, alanine aminotransferase and gamma-glutamyltransferase on both analysers; and albumin, direct bilirubin, uric acid, inorganic phosphorus, iron, calcium, magnesium, chloride, sodium, potassium, alkaline phosphatase, amylase, lactate dehydrogenase and creatine kinase on the Dax-48. We followed the guidelines of the Spanish Society of Clinical Biochemistry and Molecular Pathology. Bilirubin samples were prepared using bovine bilirubin, and studied in the concentration range of 20 to 400 mumol/l. For haemolysis, the pool was spiked with a diluted haemolysate of human red cells to achieve a concentration range of 10 to 120 mumol/l of haemoglobin. Lipaemia was studied using samples spiked with Intralipid, a fat emulsion, at concentrations from 1 g/l to 6 g/l (3 to 18 mumol/l of triacylglycerols).

Effects of long-term therapy with either heparin or low-molecular-weight heparin on serum lipid levels. A prospective study.

Beside the anticoagulant and antithrombotic activity, heparin also exerts a lipolytic activity. In a prospective study on patients with venous thromboembolism and some contraindications to coumarin therapy, a low-molecular-weight heparin (Fragmin) was compared to unfractioned (UF) heparin in terms of both efficacy and safety. A secondary aim was to study the influence of both types of heparin on serum lipid levels. Sixty-six consecutive patients who were not taking concomitant treatment with lipid-lowering drugs entered the study. Patients received treatment with either UF heparin, 10,000 IU s.c., b.d., or Fragmin 5,000 IU anti-factor Xa s.c., b.d. for a period of 3 or 6 months, according to whether the initial diagnosis was deep venous thrombosis or pulmonary embolism. Each patient was followed up at 6-weekly intervals, and blood samples were obtained at discharge, and then 6 and 12 weeks after discharge. Finally, a further sample was obtained 3 months after therapy was discontinued. Total cholesterol levels increased significantly in both groups of patients: levels increased from 193 +/- 56 to 246 +/- 63 mg/dl in the UF heparin group (p < 0.001), and from 189 +/- 53 to 222 +/- 47 mg/dl in the Fragmin group (p < 0.05). The increase was mostly due to a very strong increase in HDL cholesterol levels in patients receiving UF heparin (from 46 +/- 12 to 71 +/- 23 mg/dl; p < 0.000005). Three months after discharge, HDL cholesterol levels were significantly higher in patients taking UF heparin than in patients in Fragmin (p = 0.006). By contrast, patients on Fragmin exhibited a significant increase in LDL cholesterol levels: from 112 +/- 39 to 139 +/- 37 mg/dl; p < 0.01.

Peritoneal transport evaluation in peritonitis: comparison between methods.

Peritoneal dialysis patients may need solute permeability transport evaluation during acute peritonitis. The aim of this study was to assess if the simplified mass transfer coefficient (MTCS) or the peritoneal equilibration test (PET) was equivalent to the complex MTC (MTCX) in solute transport evaluation during acute peritonitis in continuous ambulatory peritoneal dialysis (CAPD) patients. We studied 15 episodes of peritonitis (PTIS). Results were compared to a baseline patient study (PRE) and a control study done 30 days after diagnosis of peritonitis (POST). All peritoneal evaluation methods showed a significant increase in solute transport during acute peritonitis compared to baseline and control studies. There was an acceptable correlation between MTCX and simplified methods including the PET in the baseline and control studies. However, correlation between MTCX and simplified methods decreased during acute peritonitis. Likewise, the PET showed a better correlation with MTCX than MTCS. We conclude that the PET has an acceptable agreement with MTCX even during acute peritonitis, so the PET can be a useful tool in evaluating peritonitis-induced peritoneal permeability changes.

Mass transfer coefficient: comparison between methods.

The mass transfer area coefficient (MTC) is the best parameter for solute transport evaluation in continuous ambulatory peritoneal dialysis (CAPD) patients. We compared three simplified MTC methods (calculated according Garred, Krediet, or Lindholm) and the peritoneal equilibration test (PET) (Twardowski) to complex MTC (MTCX) (Randerson and Farrell) for urea and creatinine, by means of 29 tests performed in 24 stable CAPD patients. There were no significant differences (paired t-test) between MTCX and each of the simplified MTC, except for creatinine MTC calculated by Krediet's method, which was significantly different (MTCX: 9.36 +/- 4.32, K-MTC: 10.48 +/- 4.55, p < 0.05). Likewise, there was an acceptable correlation between complex MTC and each of the simplified methods including the PET. However, a more detailed study of the MTC's categorizations shows poor agreement with complex MTC categorization. Better results are obtained by PET categorization, which reaches good likelihood ratios either for positive or negative events. We conclude that simplified MTC or the dialysate/plasma ratio at 240 minutes for urea and creatinine has an acceptable correlation with complex MTC and can be useful in clinical practice. There is poor agreement between solute transport categorizations of simplified MTC and complex MTC. There is a better coincidence between the PET (D/P at 240 minutes) and complex MTC categorizations.

Evaluation of the IL-Phoenix chemistry electrolyte analyser.

This paper reports an evaluation of the IL-Phoenix Chemistry/Electrolyte Analyser; the evaluation was carried out in accordance with internationally recognized guidelines. The evaluation was performed in three steps: evaluation in routine conditions; assessment of interferences; and study of practicability. Seven constituents were studied under routine working conditions. Within-run imprecision rangedfrom 0.6% (CV) for chloride to 3.1% (CV) for glucose. Between-run imprecision ranged from 0.9% for sodium to 6.0% (CV) for urea. Sample-related carryover was not significant. The relative inaccuracy was acceptable; drift was negligible; linearity was agreed with the range showed by the supplier. Haemoglobin produced negative interferences with sodium and chloride. Turbidity interfered negatively with sodium, chloride, potassium and total calcium, andpositively with glucose. Bilirubin showed a negative interference with sodium, chloride and creatinine.

Criteria for the assessment of analyser practicability.

This article lists the theoretical criteria that need to be considered to assess the practicability of an automatic analyser. Two essential sets of criteria should be taken into account when selecting an automatic analyser: 'reliability' and 'practicability'. Practibility covers the features that provide information about the suitability of an analyser for specific working conditions.These practibility criteria are classsified in this article and include the environment; work organization; versatility and flexibility; safely controls; staff training; maintenance and operational costs.