Conclusions: Analysis of ordered tests allows the identification ofinappropriate use of laboratory tests along with the most problematicprescribers, so education and discussionwith clinicians can be focusedon the most profitable areas, aiming for large-scale economicalsavings and improved patient care.

doi:10.1016/j.clinbiochem.2014.06.045

517Evaluation of a method to detect prozone/hook effect/antigenexcess phenomenon for free light chain quantification using asimple pooling protocolVincent De Guirea, Frank Courjalc, Richard LeBlancb,MarielleMalvasoc,Julie Amyotc, Nicolas TétreaultcaMaisonneuve-Rosemont Hospital, CanadabDepartment of Hematology, Maisonneuve-Rosemont Hospital, CanadacDepartment of Biochemistry, Maisonneuve-Rosemont Hospital, Canada

Objectives: Antigen excess is an important issue that can lead tounderestimation of patient's results and misdiagnosis. With patientsamples which can range from less than 1 mg/L to over 100 000 mg/L,serum free light chain (FLC) measurement is especially prone to thisinterference.

Design andmethods:We propose a methodology based on samplepooling for a fast and cost effective method to detect antigen excessphenomenon for serum free light chain quantification. Our strategywasevaluated on the Immage (Beckman) and on the BNII (Siemens)nephelometers using the Freelite® assay (The Binding Site).

Results: First, we evaluated the sensitivity of our strategy using apool of 15 mixed samples spiked with different concentrations ofkappa or lambda free light chain. Secondly, patients with antigenexcess ranging from 1192 to 8500 mg/L of kappa free light chain wereefficiently detected using our strategy. Finally, a preliminary evalua-tion of the size of the pools was conducted using pools ranging from15 to 43 different samples. Based on the precision of the method, ourpreliminary data suggest that the number of samples in a pool canreach up to 43 different patients to detect an antigen excess of1192 mg/L.

Conclusion: We described and validated a strategy based onsample pooling for detection of antigen excess on free light chainmeasurement. This approach could be suitable for any laboratorymeasuring serum free light chain that does not currently have aninstrument platform for detection of antigen excess.

doi:10.1016/j.clinbiochem.2014.06.046

518A Canada-wide practice survey on serum protein electrophoresis:Opportunity for standardization and improvementP.C. ChanSunnybrook Health Sciences Centre, Canada

Background: Serum protein electrophoresis (SPE) is a long-established technique used primarily for detecting and quantifyingmonoclonal immunoglobulins or components (MC). However, evi-dence suggested that substantial variations in reporting practices stillexist to date. The objectives of this study are to (1) identify areas andextent of variations in the use and reporting of SPE through a Canada-wide practice survey, and (2) seek possible solutions.

Methods:Under the auspices of the CSCCMonoclonal GammopathyInterest Group, we devised and conducted a nation-wide survey whichconsisted of 15 questions each with multiple responses and individualfree-text elaborations. The practice survey was open to all CSCC mem-bers through a web link to the questionnaire on instant.ly. Participantswere instructed not to complete more than one survey from eachlaboratory.

Results: 43 laboratories participated with 29 completed the wholesurvey. Some 60% of the participants used gel electrophoresis whilethe same % resolved serum proteins into 6 fractions. 66% used SPE asthe sole first line test for MC detection, and about one third used SPEalso for investigating non-MC conditions. 78% of the participantswould perform reflexive testing if an MC was present but only 47%would do so based on specific protein patterns while the other halfwould not even comment on findings other than MC.

Conclusions: Substantial variations in test utility and resultreporting, especially in the absence of a discrete MC, were present.A lack of good evidence/data surrounding these practices appearedto be the major obstacle for variation reduction.

doi:10.1016/j.clinbiochem.2014.06.047

519Screening and characterization of vitamin D bindingprotein variantsLei Fua,b, Betty Y.L. Wonga, Chad R. Borgesc, Rashida Williamsb,Thomas O. Carpenterd,e, David E.C. Colea,b,f,gaDepartment of Clinical Pathology, Sunnybrook Health Sciences Centre,Toronto, ON, CanadabDepartments of Laboratory Medicine and Pathobiology, University ofToronto, Toronto, ON, CanadacDepartment of Chemistry & Biochemistry and Center for PersonalizedDiagnostics of the Biodesign Institute, Arizona State University, Tempe,AZ, USAdDepartments of Pediatrics (Endocrinology), Yale University School ofMedicine, New Haven, CT, USAeOrthopaedics and Rehabilitation, Yale University School of Medicine,New Haven, CT, USAfDepartments of Pediatrics (Genetics), University of Toronto, Toronto,ON, CanadagDepartment of Medicine, University of Toronto, Toronto, ON, Canada

Objectives: The gene (GC) for the vitamin D binding protein (DBP)shows a great deal of variation. Two missense mutations (D432E andT436K) are the major common polymorphic variants, and both mayinfluence vitamin D metabolism. However, many other less commonvariants, identified biochemically, have been reported in literature.This study aimed to identify the underlying mutations by molecularscreening and characterize themutant proteins bymass spectrometry.

Methods: Denaturing high performance liquid chromatography(DHPLC)was used for screening genetic variants onmeta-PCR productsof the genomic DNA samples. Sanger sequencing identified the specificmutations. A mass spectrometric immunoassay method was used tocharacterize the DBP protein variants in serum samples.

Results: Molecular screening identified 10 samples (out of 761)containing a recurrent alanine deletion at codon 246 in exon seven(p.Ala246del, c.737_739delCTG) and 1 sample (out of 97) containing acysteine to phenylalanine substitution at codon 311 in exon eight(c.932GNT, p.Cys311Phe). The mutant allele proteins and posttrans-lational modified products were distinguishable from the wild-typeproteins bymass spectra profiling. Loss of disulfide bond due to loss ofcysteine at residue 311was accompanied by the appearance of a novel

Abstracts 1149

mixed disulfide species, consistent with S-cysteinylation of the pro-tein product.

Conclusions: Our results confirm earlier biochemical studies sug-gesting that some deleterious mutations in the DBP gene are notuncommon. Mutant proteins are secreted and can be found incirculation. By combining molecular screening and mass spectromet-ric methods, mutant DBP species can be characterized and theirpotential functional role explored.

doi:10.1016/j.clinbiochem.2014.06.048

520Use of procalcitonin in the acute exacerbation of chronicobstructive pulmonary disease in the emergency room: Acosts/benefits and hospital practice short studyMarie-Eve Habela, Daniel BrazeaubaDepartment of Biochemistry, CSSS de Rimouski-Neigette, CanadabDepartment of EmergencyMedicine, CSSS de Rimouski-Neigette, Canada

Objectives: Utility of procalcitonin (PCT), a new marker allowingearly detection of bacterial infection, in acute exacerbation of COPD(AECOPD) was evaluated in our emergency room (ER). Influence ofPCT results on antibiotics prescription and treatment duration, aswell as costs/benefits ratio, were addressed.

Design and methods: PCT analyses were performed stat on aMini-Vidas (Bio-Mérieux). ER physicians were asked to complete asurvey and to rely on an adaptation of a published algorithm. Follow-ups of hospitalized patients were managed by family physicians. PCTdata and surveys were analysed.

Results: 33/45 (73%) PCT measurements were adequate. Of those,29 had a well completed survey. Four hospitalized patients were alsoincluded in the study even though not consulting initially at the ER.Procalcitonin results guided decision to prescribe antibiotics (or not)in 70% of the cases. Consequently, 61% of patients (20/33) did notreceive antibiotics following a negative PCT result. However, onceantibiotics were initiated, subsequent negative PCT measurementshad little effect on the reduction of treatment duration (11% ofhospitalized patients, and 15% upon discharge). The costs/benefitsratio was poor. Savings on antibiotics represented only 18% of costsgenerated by PCT analyses.

Conclusions: Although not cost-effective in the short term, the useof PCT could lead to significant long-termbenefits and savings throughthe reduction of antibiotics prescription, which could ultimately leadto a reduction in cases of C. difficile infection and antibiotic's resistancein our community.

doi:10.1016/j.clinbiochem.2014.06.049

521The diagnostic role of B-type natriuretic peptide in pediatricpopulations: A systematic reviewStephen Hilla, Kai Fana, Andrew Don-WauchopebaMcMaster University, CanadabRegional Laboratory Medicine Program, Canada

Objectives: B-type natriuretic peptides (BNP) and N-terminalpro-B-type natriuretic peptides (NT-proBNP) have been establishedas biomarkers of heart failure (HF) in the adult population. However,research evidence on the diagnostic significance of these markers in

pediatric and neonatal populations remains limited. This systematicreview examines the diagnostic utility of BNP and NT-proBNP inpediatric and neonatal populations with HF, respiratory (RD), andcongenital heart diseases (CHD).

Design and methods: We searched literature published between1989 and 2013. 332 articles examining pediatric populations wereisolated and screened. We included 19 articles using BNP and NT-proBNP as diagnostic tests in urgent-care and non-emergency settings.

Results: Neonatal patients have markedly elevated serum BNPand NT-proBNP levels compared to pediatric populations across allstudy populations. BNP differentiated cardiovascular-related RDfrom primary RD (decision range: 100 pg/mL to 550 pg/mL, sensitivityrange: 78%–100%). NT-proBNP was able to differentiate CHD fromhealthy cohorts (decision range: 2000–2850 pg/mL, sensitivity range:74–90%) while BNP diagnosed underlying HF in CHD patients(decision range: 30–100 pg/mL, sensitivity range: 62%–100%). BothBNP (decision range: 31.2–313.3 pg/mL, sensitivity range: 83–100%)and NT-proBNP (178–3617 pg/mL, sensitivity range 81–100%) diag-nosed HF of various aetiology and severity in pediatric and neonatalpopulations.

Conclusions: BNP and NT-proBNP may be a useful diagnostic toolin assessing cardiovascular dysfunctions in pediatric patients. Furtherinvestigations with BNP in pediatric urgent-care settings are encour-aged to more pragmatically evaluate its diagnostic significance.

doi:10.1016/j.clinbiochem.2014.06.050

522Impact of a shortened centrifugation protocol for commonchemistry and immunology testing on differentautomation platformsA. Lou, M. Fullerton, M. Elnenaei, B. NassarDivision of Clinical Chemistry, Department of Pathology and LaboratoryMedicine, Capital Health, CanadaDalhousie University, Canada

Objectives: There are no standard centrifugation protocols forprocessing blood specimens in laboratories. Vendors recommendcentrifugation at 1000–1300 g Relative Centrifugal Force (RCF) for10 min. This study evaluates an abbreviated centrifugation protocolat 2600 g RCF for 5 min.

Design and methods: Two pairs of venous blood samples fromfour healthy volunteers were collected into serum separator (SST)and plasma separator tubes (PST) (Becton Dickinson). One pair ofSST and PST was randomly assigned to our current centrifugationprotocol at 1800 g for 10 min. Another pair was subjected to theabbreviated protocol at 2600 g for 5 min, 53 analytes includingimmunology and serology, were analyzed on ARCHITECT ci16200,DxC800, Centaur and Immage platforms. The impact of centrifuga-tion protocols on analyte results was determined using paired t-testand average % difference.

Results: There were no significant differences in the resultsfrom the two centrifugation protocols (P N 0.05), except for ferritin(ARCHITECT i2000); P = 0.02. The average % differences for allanalytes were less than 5%, except for PTH on ARCHITECT i2000 andGGT on DxC800 which were −5.98% and −5.33% respectively.

Conclusions: Our results demonstrate that the shorter centrifu-gation protocol does not adversely affect analytical results for allassays. Faster sample processing will improve turnaround time andefficiency of laboratory operation.

doi:10.1016/j.clinbiochem.2014.06.051

Abstracts1150