Despite some limitations, current molecular diagnostic methods have a great potential to include targets useful in the rapid identification of microorganisms and antimicrobial resistance, to analyse directly unprocessed samples and to obtain quantitative results in pneumonia, an entity of complex microbiological diagnosis due to the features of the pathogens commonly implicated.

by Dr A. Camporese

A change in culture without culture?
Developing accurate methods for diagnosing respiratory tract infections has long been a challenge for the clinical microbiology laboratory [1].

The current semi-quantitative agar-plate based culture method used in most clinical microbiology laboratories for analysing specimens from patients with suspected community-acquired pneumonia (CAP), hospital-acquired pneumonia (HAP), or ventilator-associated pneumonia (VAP), although adequate for recovering and identifying a wide variety of bacterial species from respiratory specimens, is slow, and cannot differentiate between colonization and infection. Moreover, results may be misleading, particularly if a Gram stain is not performed in parallel to ascertain the adequacy of expectorated sputum samples or endotracheal aspirates [2].

As the Infectious Diseases Society of America (IDSA) and American Thoracic Society (ATS) CAP guideline notes, one of the problems with diagnostic tests for respiratory tract infections “is driven by the poor quality of most sputum microbiological samples and the low yield of positive culture results” [3]. Moreover, the highest predictive value of a culture occurs only when Gram stain shows a predominant morphotype, and the culture yields predominant growth of a single recognized respiratory pathogen of that morphotype (e.g., Streptococcus pneumoniae) [2].

Unfortunately, such concordance decreases rapidly when specimens are collected after the initiation of antimicrobial therapy or when their arrival at the microbiology laboratory is significantly delayed.

One approach that may improve the diagnosis of respiratory tract infections and shorten the time necessary to place patients on appropriate therapy is the use of nucleic acid amplification methods.

Straight ahead toward molecular assays
Today clinical microbiologists appear to be on the threshold of a potentially important transition, with a substantial increase in the use of molecular diagnostic tests to replace or augment the century-old methods of culture, as many experts now view traditional microbiology as slow and antiquated, especially when compared with newer technologies used in other areas of laboratory medicine [4].

Traditional methods demonstrated poor sensitivity and specificity for detecting specific pathogens, particularly when the specimen being cultured is from a non-sterile anatomical compartment, such as the respiratory tract.

For this reason, molecular methods are becoming more widely used also for the detection of respiratory pathogens, in part because of their superior sensitivity, relatively rapid turnaround time, and ability to identify pathogens that are slow growing or difficult to culture.

However, to have a positive impact on patient management, molecular tests will need to be easy to use, and provide clear, definitive results that will give physicians the data necessary to start, or in some cases withhold, antimicrobial agents [5].

Further, to be really successful, industry must determine which combination of molecular targets [Table 1] and clinical specimens will produce results that will effectively guide anti-infective therapy regimens for patients with pneumonia or other respiratory tract diseases.

Another key challenge for industry will be to develop assays that are not only rapid, but also readily accessible, because development of an assay that is rapid, but unavailable on evening or night shifts, or at weekends, because of its technical complexity, limits the clinical value of the test.

Moreover, to be successful, molecular assays will need to be perceived by health care systems as cost-effective, but cost-effectiveness should be determined not only by comparison to the costs of performing slower, conventional methods in the laboratory, but also by consideration of the cost savings achieved from optimized antimicrobial therapy, decreased use of additional diagnostic tests, and shorter hospital stays [2].

To address issues on these topics, the IDSA and the Food and Drug Administration (FDA) co-sponsored a workshop on molecular diagnostic testing for respiratory tract infections in November 2009, with the participation of the FDA, industry, authorities in microbiology, statisticians and others. Respiratory tract infections were selected because this is the site of most infections treated with antibiotics in paediatric and adult practice, and they also represent a group of infections in which an etiologic agent is seldom identified in non-research settings [4].

The IDSA believes that patient care could be improved by accurate and rapid identification of pathogens, which would promote more judicious use of antibiotics, permit pathogen directed therapy, and provide potentially important
epidemiologic information.

Thus, the IDSA strongly desires development and implementation of molecular diagnostic tests that are easy, rapid, technically uncomplicated, applicable to specimens that are readily obtained, reasonably priced, sensitive and specific, because such tests will greatly improve antimicrobial stewardship, thereby helping to reduce the spread and impact of antibiotic resistance. Such tests will also facilitate conduct of clinical trials supporting the approval of new antibacterial agents [4].

Respiratory samples suitable for molecular assays
A variety of respiratory samples are amenable to molecular testing, including expectorated sputum, bronchoalveolar lavages (BALs), protected bronchial brushes, and endotracheal aspirates [2, 5, 6]. Of these, expectorated sputum samples are by far the most common respiratory samples submitted to the clinical microbiology laboratory, but are also the poorest in overall quality.

Endotracheal aspirates from ventilated patients are often of better quality than that of expectorated sputum obtained from patients with CAP/HAP, but may still be contaminated with upper respiratory tract flora.

Therefore, obtaining specimens from the site of infection that are not contaminated with upper respiratory tract flora remains to date a real and constant problem. BALs and protected brush samples seem more likely to yield samples from the site of infection, but require significantly more effort to obtain, and thus offer a much smaller market for a new molecular test.

Moreover, there is a significant gap in our knowledge as to how well molecular tests for bacterial pathogens would perform on expectorated sputum samples, compared with performance on BALs or protected brush samples from the same patient collected within a similar period [2].

This knowledge gap is also a barrier to test development, because a molecular test that cannot be performed on expectorated sputum (given all the problems with specimen quality) may not have broad enough appeal among physicians to make it a financially viable product (from the industry perspective).

Technology perspectives
There are a wide array of emerging technologies for the detection and quantification of respiratory pathogens directly from clinical specimens. Some of these technologies, such as real-time PCR, have potential for high-throughput testing, and others will allow rapid near patient testing, but more studies are needed to fully determine their performance characteristics and determine their ideal clinical application [6,7].

Molecular assays may target either a single pathogen or multiple respiratory pathogens in a single assay. There are merits to both single-pathogen and multiplex approaches. Certain bacterial respiratory pathogens cause such distinct clinical syndromes that assays that target them individually still have clinical utility. These include already many organisms, such as Chlamydophila pneumoniae, Mycoplasma pneumoniae, Legionella pneumophila, or Bordetella pertussis [Table 1].

Moreover, some multiplex assays for respiratory tract disease already include many targets for a rapid diagnosis of CAP, HAP, and VAP, but, in designing new assays, it will be critical to understand whether an assay for a determined number of bacterial pathogens will meet physicians’ needs and provide adequate data for initiating or altering anti-infective therapy [7, 8].

Potential and currently available targets for multiplex or individual molecular assays for respiratory tract samples in immunocompetent and/or immunocompromised patients with CAP, HAP, or VAP are presented in Table 1 [7].

Further, in this age of multidrug resistance, expanding the target selection to include key antimicrobial resistance genes that would alter existing therapy or guide empirical therapy, should also be considered [Table 1].

Lastly, if molecular-based diagnostic methods currently available are helpful in detecting single and multiple bacterial pathogens simultaneously, including the most frequent cause of CAP/HAP/VAP, the real-time PCR is also well known for its ability to quantify targets.

Where available, the application of quantitative molecular tests for the detection of key pathogens, such as S. pneumoniae, both in sputum and in blood, defining a threshold for classification, such as a colonizer or as an invasive pathogen, might be relevant in CAP patients, mainly in whom antibiotic therapy has been initiated, and might be a useful tool for severity assessment [9, 10].

Conclusion
Significant progress exists on the development and improvement of molecular-based methods feasible to be applied to the diagnosis of lower respiratory tract infection.

Multiplex assays, user-friendly formats, results in a few hours, high sensitivity and specificity in pathogen identification, detection of antibiotic resistance genes and target quantification, among others, are some of the contributions of novel molecular-based diagnosis approaches.

Developing new molecular tests for other bacterial respiratory pathogens, particularly microorganisms that can be both asymptomatic colonizers and overt pathogens of the respiratory tract, detection of pathogens and new key antimicrobial resistance genes in unprocessed samples, and determination of the microbial load by quantitative multi-pathogen tests will be some of the future challenges of molecular diagnosis in CAP/HAP/VAP.

References
1. Bartlett JG. Decline in microbial studies for patients with pulmonary infections. Clin Infect Dis 2004; 39: 170–172.
2. Tenover FC. Developing molecular amplification methods for rapid diagnosis of respiratory tract infections caused by bacterial pathogens. Clin Infect Dis 2011; 52(S4): S338–S345.
3. Mandell LA, Wunderink RG, Anzueto A, et al. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis 2007; 44(S2): S27–S72.
4. Infectious Disease Society of America (IDSA). An unmet medical need: rapid molecular diagnostics tests for respiratory tract infections. Clin Infect Dis 2011; 52(S4): S384–S395.
5. Caliendo AM. Multiplex PCR and emerging technologies for the detection of respiratory pathogens. Clin Infect Dis 2011; 52(S4): S326–S330.
6. Lung M and Codina G. Molecular diagnosis in HAP/VAP. Curr Opin Crit Care 2012; 18: 487–494.
7. Camporese A. Impact of recent advances in molecular techniques on diagnosing lower respiratory tract infections (LRTIs). Infez Med 2012; 4: 237–244.
8. Johansson N, Kalin M, Tiveljung-Lindell A, et al. Etiology of community-acquired pneumonia: increased microbiological yield with new diagnostic methods. Clin Infect Dis 2010; 50: 202–209.
9. Werno AM, Anderson TP, Murdoch DR. Association between pneumococcal load and disease severity in adults with pneumonia. J Med Microbiol 2012; 61: 1129–1135.
10. Woodhead M, Blasi F, Ewig S, et al. Guidelines for the management of adult lower respiratory tract infections-Full version. Clin Microbiol Infect 2011; 17(S6): E1–E59.

The author
Alessandro Camporese MD
Clinical Microbiology and Virology Department
S. Maria degli Angeli Regional Hospital, Pordenone, Italy

E-mail: alessandro.camporese@aopn.fvg.it

Interference in immunoassay is a well described phenomenon and all clinical immunoassays, including thyroid function tests, are potentially at risk. Spurious results can lead to over investigation or mismanagement if not detected, but a proactive approach by the laboratory will help to identify and resolve these problems.

by Dr Olivia Bacon and Dr David J. Halsall

Background
Thyroid disorders are relatively common, and are associated with long-term morbidity and mortality. Clinical signs and symptoms are often non-specific, so reliable laboratory tests are critical for diagnosis. Therefore, thyroid function tests (TFTs) are frequently requested immunoassays with around 10 million results being reported each year by UK laboratories. In the UK, TFTs typically include a high sensitivity immunoassay for thyroid stimulating hormone (TSH) with an immunoassay estimation of non-protein bound thyroxine (fT4), either run simultaneously or added if the TSH value is outside the reference interval [1].

For the majority of tests, both results will be within the reference interval and thyroid disease can be excluded. In some patients TFTs support the diagnosis of hypothyroidism (raised TSH with fT4 low, or lownormal) or hyperthyroidism (TSH undetectable, and fT4 elevated), and these results will confirm clinical findings.  However, due to the high volume of TFTs performed, it is not unusual for the laboratorian to be faced with a set of TFTs that are either internally inconsistent, or incompatible with the clinical details provided. Many medications can affect the thyroid axis, as can other non-thyroidal pathologies; these are often transient, but can cause unusual patterns of TFT. Much rarer genetic or pituitary conditions can also cause discordant TFTs [2]. However, if drug effects are excluded, it is necessary at this stage for the laboratorian to consider that one of the TFT results is incorrect, as analytic error is at least as common as these rare thyroid conditions. As spurious TFT results can lead to over investigation, or even inappropriate treatment, it is critical, but not trivial, for the laboratory to confirm the analytical validity of the TFT results.

In one study of more than 5000 samples received for TSH analysis, assay interference with the potential to adversely affect clinical care was detected in approximately 0.5% of patients [3].  This equates to a rather alarming 50,000 tests per annum in the UK.

Although assay design is continually improving, no routine immunoassay is currently robust to interference.  Technical errors with many routine chemistry methods caused by inappropriate sample collection or handling, chemical or spectral interference can be detected during result validation. However, detection of spurious TFT immunoassay results is more challenging as there is no automatic ‘flag’ from the analyser, and there is usually a wide range of plausible values for these analytes, making it difficult to question those which are ‘suspicious’.  Consequently clinical validation, where results are checked for discordance with the clinical correlates and other laboratory tests, is used to detect potentially incorrect results before reporting.  For TFTs this is aided by the characteristic reciprocal relationship between TSH and fT4 in patients with an intact pituitary–thyroid axis.

Mechanisms of interference in TSH assays
Endogenous interfering antibodies are a well described cause of immunoassay interference [4].  In TSH assays these antibodies can have affinity for TSH itself or towards assay components.  Anti-reagent antibodies can be ‘anti-animal’ antibodies, specific to the species in which the reagent antibody was raised, or weak, polyspecific ‘heterophilic’ antibodies, which may be part of the natural process of the generation of antibody diversity [5]. Anti-animal antibodies are more prevalent in animal handlers or patients treated with therapeutics based on animal immunoglobulins.
Anti-reagent antibodies can interact with either the capture or detection antibodies in two-site assays, blocking the generation of signal in the presence of analyte (false negative result) or by causing antibody cross-linking in the absence of analyte (false positive result) [Fig. 1].

Anti-TSH antibodies can generate high molecular weight TSH : antibody complexes (‘macro-TSH’). Depending on the exact site of the antibody–analyte interaction, false positive TSH results may occur as the macro-TSH is unlikely to be biologically active [6].

Detection of interference in TSH assays
Once suspected, a robust laboratory strategy is required for confirming or excluding assay interference. Method comparison using an alternative method is often used as the first step. Most laboratories use two-site immunoassays for TSH, but assay formulations, antibody species and incubation times vary between manufacturers. Varying amounts of blocking agents, designed to prevent non-specific binding of heterophile antibodies, may be included. Significant disagreement between two TSH methods is a strong indicator of assay interference.

Dilution studies are a simple but effective tool to investigate the analytical validity of an immunoassay. Non-linearity to dilution suggests a result is unreliable.  However, although a good ‘rule in’ test, linearity to dilution alone cannot be used to exclude interference [3,7].

Immunosubtraction is a useful method to confirm the presence of antibody interference. This can be done crudely using polyethylene glycol (PEG) precipitation or more specifically using anti-immunoglobulin agaroses. Proprietary heterophile blocking tubes can also be used to confirm the presence of this class of interferent [3,4]. 

Once assay interference is established it can still be difficult to determine the correct TSH value, as there is no ‘gold standard’ method for TSH.  However, an alternative immunoassay result which gives the expected responses to dilution and immunosubtraction, and correlates with fT4 results plus clinical findings, can be used with a reasonable degree of confidence.

Mechanisms of interference in fT4 assays
fT4 assays present a particular analytical challenge as >99.9% of T4 in the serum is protein bound, and the unbound T4 fraction must be measured without upsetting the equilibrium between the two fractions [8]. Therefore, an abnormal T4 binding protein, or agent which affects binding protein affinity in vitro, has the potential to generate incorrect results.  Most commercial fT4 assays are one-site immunoassays based on competitive principles, using either labelled T4 analogue or anti-T4 antibodies for detection. Both heterophile and anti-T4 antibodies therefore also have the potential to interfere with these methods [4].

Non-esterified fatty acids (NEFAs) are a common T4 displacing agent as they can release T4 from the low affinity, high capacity T4 binding site on albumin.  NEFAs can be  generated in vitro, usually as a consequence of heparin therapy, which stimulates the action of lipoprotein lipase on triglyceride. Although the measured fT4 result is genuinely high, it does not reflect the in vivo situation [9]. 

Familial dysalbuminaemic hyperthyroxinaemia (FDH) is a benign genetic condition where the affinity of albumin for T4 is increased, such that circulating albumin-bound T4 is elevated. Despite the high total T4 (tT4), concentrations of free hormone in vivo are unaffected due to the homeostatic regulation of the thyroid axis. However, FDH is often associated with falsely high fT4 measurements using commercial immunoassays [10] [Fig. 2]. Both the increased affinity of the variant albumin for some labelled T4 analogues, as well as potential disruption of the T4 : albumin equilibrium during the assay, are likely mechanisms.  The presence of the FDH mutation can be confirmed using molecular genetic approaches.

Detecting interference in fT4 assays
Despite the greater analytical challenge, confirming interference in fT4 assays can be easier than for TSH due to the availability of physical separation methods, such as equilibrium dialysis, as ‘gold standard’ assays [8]. However, these methods are technically difficult and not available in most clinical biochemistry laboratories. Also, these methods do not solve the in vitro problems of hormone displacement. 
Again a first approach is often method comparison, using a different immunoassay architecture. Dilution and immunosubtraction studies can also be informative, although some fT4 methods are not robust to matrix effects so careful control experiments are required.
Measurement of total rather than free T4 can be useful in situations where there is a suspicion of abnormal T4 binding proteins. For example, total T4 will be elevated in the presence of anti-T4 antibodies and in FDH.

Clinical causes of aberrant TFTs
As mentioned above there are well described pharmacological and pathological causes of unusual TFTs; an increased awareness of analytical artefacts should not detract from the detection of these conditions.  For example thyroxine treatment, a TSH secreting pituitary tumour (TSHoma), the genetic condition thyroid hormone resistance, FDH or TFT antibody interference can give elevated fT4 results with a TSH within the reference interval. Attempts by the laboratory to exclude assay interference should complement both the diagnosis of transient and genetic thyroid conditions as well as the more common drug related effects.

Conclusions and future directions
Immunoassay manufacturers have invested considerable resources into reducing the potential for antibody-mediated assay interference, for example by including blocking agents, or using antibody fragments rather than intact antibodies as assay reagents. Although these measures are effective, it is worth bearing in mind that changes to assay formulations may introduce novel types of interference. We have observed negative interference in one fT4 assay which appears related to the presence of a blocking agent introduced to reduce the risk of positive interference in this method [11]. Mass spectrometric methods have been introduced to eliminate antibody interference in both fT4 and tT4 methods, but unfortunately the fT4 methods still require careful optimization to avoid interference caused by binding proteins and displacing agents.

As current TFT methods remain prone to analytical interference the clinical laboratory must remain vigilant to the potential for assay interference, promote effective communication with requesting clinicians, and have procedures in place for investigation of discordant results.
 
References
1. Association for Clinical Biochemistry (ACB), British Thyroid Association (BTA), British Thyroid Foundation (BTF). UK guidelines for the use of thyroid function tests.2006; www.acb.org.uk/docs/TFTguidelinefinal.pdf.
2. Gurnell M, Halsall DJ, Chatterjee VK. What should be done when thyroid function tests do not make sense? Clin Endocrinol. (Oxf) 2011; 74(6): 673–678.
3. Ismail AA, Walker PL, Barth JH, Lewandowski KC, Jones R, Burr WA. Wrong biochemistry results: two case reports and observational study in 5310 patients on potentially misleading thyroid-stimulating hormone and gonadotropin immunoassay results. Clin Chem. 2002; 48(11): 2023–2029.
4. Despres N, Grant AM. Antibody interference in thyroid assays: a potential for clinical misinformation. Clin Chem. 1998; 44: 440–454.
5. Kaplan IV, Levinson SS. When is a heterophile antibody not a heterophile antibody? When it is an antibody against a specific immunogen. Clin Chem. 1999; 45: 616–618.
6. Halsall DJ, Fahie-Wilson MN, Hall SK, Barker P, Anderson J, Gama R, Chatterjee VK. Macro thyrotropin-IgG complex causes factitious increases in thyroid-stimulating hormone screening tests in a neonate and mother. Clin Chem. 2006; 52: 1968–1969.
7. Ross HA, Menheere PP, Thomas CM, Mudde AH, Kouwenberg M, Wolffenbuttel BH. Interference from heterophilic antibodies in seven current TSH assays. Ann Clin Biochem. 2008; 45: 616.
8. Thienpont LM, Van Uytfanghe K, Poppe K, Velkeniers B. Determination of free thyroid hormones. Best Pract Res Clin Endocrinol Metab. 2013; in press.
9. Stockigt JR, Lim CF. Medications that distort in vitro tests of thyroid function, with particular reference to estimates of serum free thyroxine. Best Pract Res Clin Endocrinol Metab. 2009; 23(6): 753–767.
10. Cartwright D, O’Shea P, Rajanayagam O, Agostini M, Barker P, Moran C, Macchia E, Pinchera A, John R, Agha A, Ross HA, Chatterjee VK, Halsall DJ. Familial dysalbuminemic hyperthyroxinemia: a persistent diagnostic challenge. Clin Chem. 2009; 55(5): 1044–1046.
11. Bacon O, Gillespie S, Koulouri O, Bradbury S, O’Toole A, Stuart-Thompson D, Taylor K, Pearce S, Gurnell M, Halsall DJ. A patient with multiple Roche serum immunoassay interferences including false negative serum fT4. Ann Clin Biochem. 2013; 50(Suppl 1): T50.
 
The authors
Olivia Bacon PhD and David Halsall* PhD, FRCPath, CSci
Department of Clinical Biochemistry and Immunology, Addenbrooke’s Hospital, Cambridge, UK

*Corresponding author
E-mail: djh44@cam.ac.uk

The aim of this study was to assess the practicability and evaluate the analytical characteristics of the Mindray BS-2000M, a new automatic chemistry analyser, manufactured by Shenzhen Mindray Bio-Medical Electronics Co., Ltd (Mindray Shenzhen, China).

The evaluation involved 21 clinical chemistry parameters measured using indirect potentiometry, spectrophotometric and immunoturbidimetric assays.

Spectrophotometric assays
Alanine aminotransferase (ALT, according to IFCC with-out pyridoxal phosphate), aspartate aminotransferase (AST, according to IFCC method), gamma-glutamyltransferasa (GGT, according to Szasz), total bilirubin (TBIL, Diazotized sulfanilic acid method), calcium (Ca, Arsenazo III method), creatine kinase (CK, IFCC method), creatinine (Crea, modified Jaffe method), glucose (Glu, Hexokinase method), high density lipoprotein-cholesterol (HDL-C, direct method), magnesium (Mg, Xylidyl blue method), phosphorus (P, Phosphomolybdate method), total cholesterol (TC, Cholesterol oxidase – Peroxidase method), triglycerides (TG, Glycerokinase Peroxidase – Peroxidase method), total protein (TP, Biuret method), uric acid (UA, Uricase-Peroxidase method), iron (Fe, Colorimetric assay-Ferrozine), α-amylase [α-AMY, substrate: 4, 6- ethylidene-(G7)-1, 4-nitrophenyl-(G1) –α, D-maltoheptaoside (EPS-G7), enzymatic colorimetric assay according to IFCC method] and urea (Urea, Urease-glutamate dehydrogenase).

Immunoturbidimetric assays
Apolipoprotein A1 (ApoA1).

Indirect potentiometry assays (ISE)
Sodium (Na), potassium (k) and chloride (Cl).

Analytical evaluation
Among all the available parameters to verify the Mindray BS-2000M analytical performance, those which are more frequently requested in routine clinical practice were selected (e.g., glucose, creatinine, total protein).

An imprecision study was carried out according to the CLSI EP5-A2 guideline [1]. The within run imprecision was evaluated using two control materials. Final results were expressed as coefficient of variation (CV%). We checked that these CV satisfied the allowable maximum imprecision based on biological variability [2]. These data were taken from the listing of biological variation by Ricos et al [3], recently updated in 2012.

The inaccuracy study was done according to the CLSI EP9-A2 guideline [4], measuring at least 40 patient samples for the two analysers (Mindray BS-2000M and ADVIA 2400 Siemens Healthcare Diagnostics, USA) for 5 days.

In the inaccuracy study the mean bias and 95% confidence interval (CI) was calculated with the Bland-Altman plot analysis, and the linear regression was assessed using Passing-Bablok regression method [5-6].

The results of the imprecision study [Table 1] showed that for all the parameters imprecision was lower than the maximal allowable applying to biological variability based criteria, with the exception of sodium (0.7%) and chloride (0,8%) in control 1 and total proteins (1,7%) in control 2. Nevertheless, these three parameters fulfill the commonly used “State of the art” criterion. According to this criterion, the maximal allowable imprecision for physiological concentrations must be less than the 0.20 fractile of the between-run imprecision (CV) of the laboratories in a external quality assessment scheme (7). The CV% limit for these three parameters following this approach would be 0.9% for Na, 1.6% for Cl and 1.7% for TP. 
In the comparison study, a close correlation was observed for all parameters studied (r range: 0.92 – 1.00) [Table 2]. It is noted that there were no significant differences for 11 of the 21 parameters studied. For the other parameters statistically significant differences were found but, except for creatinine, those differences were not considered to have a clinical significance. The constant and proportional differences may be due to different standardization of both procedures. Traceable calibration materials should be used related to the reference method and also switchable materials that reveal the degree of deviation of multiple methods with respect to the true value should be used [8].

References
1. Clinical and Laboratory Standards Institute. Evaluation of precision performance of quantitative measurement methods; approved guideline -second edition. CLSI document EP5-A2. Wayne, PA:CLSI, 2004.
2. Fraser CG, et al. Proposed quality specifications for the imprecision and inaccuracy of analytical systems for clinical chemistry. Eur J Clin Chem Clin Biochem 1992; 30: 311.
3. Ricos C, et al. Desirable quality specifications for total error, imprecision, and bias, derived from biological variation. http://www.Westgard.com/biodatabase1.htm. Accessed on February 15, 2013.
4. Clinical and Laboratory Standards Institute. Method comparison and bias estimation using patient samples; approved guideline – second edition. CLSI document EP9-A2. Wayne, PA: CLSI, 2002.
5. Bland JM, et al. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307.
6. Passing H, et al. A new biometrical procedure for testing the equality of measurements from two different analytical methods. Application of linear regression procedures for method comparison studies in clinical chemistry, Part I. J Clin Chem Clin Biochem 1983; 21: 709.
7. Sebastian-Gámbaro, et al. An improvement on the criterion of the state of the art to estimate the maximal allowable imprecision. Eur J Clin Chem Biochem. 1996; 34: 445.
8. Documento de consenso. Sociedad Española de Bioquímica Clínica (SEQC) y Sociedad Española de Nefrología (SEN). Recomendaciones sobre la utilización de ecuaciones para la estimación del filtrado glomerular en adultos. Química Clínica 2006; 25: 423.

The author
Dr. Jose Luis Bedini,
Hospital Clinic I Provincial De Barcelona,
Barcelona,
Spain

MINDRAY