7th EFLM Conference on Preanalytical Phase

doi:10.1515/cclm-2025-1331

Article Publicly Available

7th EFLM Conference on Preanalytical Phase

Published/Copyright: November 30, 2025

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information

REPURPOSING CBC EDTA TUBES FOR LEAD BIOMONITORING: A FEASIBILITY STUDY

I. Aníbarro Miralles ¹, G. Velasco De Cos ¹, S. Rubio Lanchas ¹, A. García Varela ¹, P. González Maroto ¹, S. Herrero Castellano ¹, R. Andrés Luis ¹, M.D. Calvo Nieves ¹

¹Clinical Laboratory Department, Hospital Clínico Universitario de Valladolid

BACKGROUND-AIM

Lead (Pb) is a known contaminant with serious health implications. The blood concentration at which it becomes harmful remains under debate, as available evidence is limited. In 2012, the CDC lowered the threshold for the pediatric population to 5 μg/dL. In 2018, Germany proposed even lower limits: 3.5 μg/dL for children and 4 μg/dL for adults.

The need to track environmental exposure to this and other contaminants has motivated population biomonitoring initiatives, such as the European project HBM4EU. One major challenge for these strategies is accessing suitable samples.

This study aims to assess the feasibility of using EDTA tubes from complete blood count (CBC) tests, which are not validated for heavy metal analysis, for biomonitoring of lead levels in whole blood.

METHODS

Two EDTAK2 tubes were collected from each patient, one of which was specific for heavy metal determination. Samples were collected from a total of 60 patients, resulting in 120 samples overall. The study was approved by the hospital’s Ethics Committee.

Lead concentration in whole blood was determined in both types of samples by inductively coupled plasma mass spectrometry (ICP-MS), using a Perkin Elmer ICP-MS analyzer. Statistical analysis was performed using the MedCalc software.

RESULTS

The results showed a mean difference of 0.03 (95% CI: -0.007 to 0.064) using the Bland-Altman plot. The Passing-Bablok regression yielded an equation of x = y, with 95% confidence intervals ranging from -0.06 to 0.00 for the intercept and from 1.00 to 1.06 for the slope. The correlation coefficient obtained was 0.994.

CONCLUSIONS

Based on the results, the confidence interval of the mean differences in the Bland-Altman plot included 0; in the Passing-Bablok regression, the confidence intervals for the intercept and slope included 0 and 1, respectively. In view of these results, we can affirm that it is feasible to conduct a biomonitoring study of lead in whole blood using the batch of EDTA tubes from CBC tests employed in our study.

Keywords: Lead, tubes, biomonitoring

Blood sampling

P004

ANALYSIS OF PRE ANALYTICAL PRACTICES IN BLOOD GAS ANALYSIS

M. Ayoub ¹, S. Aboulkacem ¹, A. Ba ¹, Z. Aouni ¹, M. Chakib ¹

¹BIOCHEMESTRY DEPARTEMENT OF MILITARY HOSPITAL OF TUNIS

BACKGROUND-AIM

Blood gas analysis is a laboratory test that measures gases and acid-base status in the blood, usually arterial blood.

According to the French clinical biology society, the pre-analytical phase is the most critical phase because the majority of errors occur during this stage. It represents about 60 to 70% of the total non-conformities. The objective of our study is to evaluate the professional practices related to the pre-analytical phase.

METHODS

This is an observational study conducted over a period of 3 months in the year 2025 within the biochemistry laboratory of the military hospital of Tunisia.A checklist on pre-analytical conditions has been established in order to detect failures related to medical prescription and blood sampling.

RESULTS

The statistical analysis involved 900 blood gas examinations from different hospital clinics.

The totality of the medical prescriptions includes the patient’s identity as well as the date and time of collection. However, none of the prescriptions included neither FiO2 nor the temperature nor the sampling site. It was found that 10% of the samples contained an insufficient amount and or a coagulated sample which makes analysis impractical and can lead to an incident of obstruction at the analyzer level.

On the other hand, 5% of the samples were sent to the laboratory using the pneumatic system.

Regarding the delivery time to the laboratory and in accordance with the recommendations of the French Society for Clinical Biology (SFBC), compliance was observed in 68% of the samples with duration of less than 30 minutes, while 32% were transported late.

CONCLUSIONS

The evaluation of pre-analytical practices for blood gas highlighted several discrepancies, in particular the lack of traceability on the information related to temperature and FiO2 as well as the delivery time to the laboratory. Given the criticality of this analysis, measures for improving professional practices must be undertaken in order to provide reliable results as soon as possible.

Blood sampling

Optimizing laboratory test utilization can simultaneously improve patient safety, enhance diagnostic relevance, and reduce costs. Prioritizing frequently ordered tests and adopting small-volume collection tubes are practical strategies to sustain effectiveness. These findings highlight the importance of structured QI initiatives for promoting cost-effective, patient-centered care.

Blood sampling

P008

RELIABILITY OF CAPILLARY VERSUS VENOUS BLOOD GLUCOSE MEASUREMENT

S. Hanachi ¹, K. Sifi ¹, D. Abderrezag ², S. Boukhedenna ², K. Boulassel ², I. Boulemia ², S. Zekri ¹, K. Benembarek ¹, N. Abadi ¹

¹1. Biochemistry Laboratory, CHU Constantine 2. Laboratory of Biology and Molecular Genetics, Faculty of Medicine Constantine University3.

²Biochemistry Laboratory, CHU Constantine

BACKGROUND-AIM

The aims of this study, is to attempt to rank four blood glucose meters available on the Algerian market, based on an analytical accuracy study and an analysis of their clinical relevance using the Parkes consensus curve.

METHODS

This is an analytical performance study comparing four blood glucose meters (BIONIME, CHECK-3, DIAGNO CHECK sens, and VITAL CHECK) to a reference system, with an analysis of their clinical relevance.

From January 3, 2022, to June 6, 2022, capillary and venous blood glucose measurements were taken instantly from 100 participants. All blood glucose measurements were performed on the INTEGRA 400 plus analyzer.

RESULTS

Our results show that the mean venous blood glucose levels were lower than the mean capillary blood glucose levels obtained with the CHECK-3 (0.9789 g/l), DIAGNO-CHECK sens (1.0098 g/l) and VITAL CHECK (1.15 g/l) meters, whereas they were higher than the mean capillary blood glucose levels obtained with the BIONIME (0.9003 g/l). Analysis of variance of capillary averages versus reference blood glucose showed a statistically significant difference for BIONIME (p=0.034) and VITAL-CHECK (p=0.035), whereas this difference was not statistically significant for CHECK- 3 (p=0.069) and DIAGNO-CHECK sens (p=0.085).

A difference in performance between the four readers after comparison of the different measurement results with ISO 15197. 2013 and objective, compliant results of 75% with the BIONIME, 73.33% with the DIAGNO-CHECK sens, 66.66% for the CHECK -3 and only 33.33% for the VITAL CHECK for venous glycemia below 1 g/l, whereas for venous glycemia greater than or equal to 1 g/l the compliant results were as follows: 72.50% for DIAGNO-CHECK sens, 62.50% for BIONIME, 55% for VITAL CHECK and 52.50% for CHECK-3.

CONCLUSIONS

In our study, the four readers tested are ranked according to their performance, with DIAGNO-CHECK sens in first place, CHECK-3 in second, followed by BIONIME, leaving VITAL CHECK as the last resort.

Blood sampling

P009

THE IMPACT OF VITAMIN D DEFICIENCY ON AUTOIMMUNE THYROID DYSFUNCTION IN HASHIMOTO’S DISEASE: EVIDENCE FROM A COHORT STUDY IN SOUTHERN, ALGERIA

S. Khercha ¹

¹DiData

BACKGROUND-AIM

Hashimoto’s thyroiditis (HT) is a leading cause of hypothyroidism, marked by thyroid autoimmunity (↑TPOAb). Although Southern algeria has abundant sunlight, vitamin D deficiency remains prevalent, suggesting impaired metabolism may influence HT severity.

Aim

To assess the relationship between serum 25(OH)D₃, TPOAb, and thyroid function (TSH, FT4) in Algerian HT patients.

METHODS

Design & Setting: Retrospective observational study (Jan 2021–Dec 2023) in Laghouat, Algeria.

Participants: 320 patients with Hashimoto’s thyroiditis and hypothyroidism, diagnosed by symptoms and elevated TPOAb; exclusions included recent vitamin D or thyroid treatment (<6 months).

Data Source: Records from Laghouat State Public Hospital and private labs.

Measurements: Serum 25(OH)D₃, TPOAb, TSH, FT4 measured by ECLIA (Roche Cobas e 801) and FEIA (Tosoh AIA 900).

Analysis: Pearson correlation for biomarker associations; Firth’s penalized logistic regression for hypothyroidism risk. Significance at p < 0.05.

RESULTS

Vitamin D deficiency: Present in 76.3% of HT patients (p < 0.001).

Correlations:

• Strong inverse: 25(OH)D₃ vs TPOAb (r = –0.82, p < 0.001).

• Moderate inverse: 25(OH)D₃ vs TSH (r = –0.72, p = 0.01).

• No significant link: 25(OH)D₃ vs FT4 (r = 0.07, p = 0.27).

Autoimmunity severity:

• TPOAb > 500 UI/mL → 82.1% vitamin D deficient.

• TPOAb < 100 UI/mL → 58.6% deficient (p < 0.001).

Regression analysis: Vitamin D sufficiency markedly reduced hypothyroidism risk (OR = 0.00000156, 95% CI: 5.07×10^-9 – 3.03×10^-5).

CONCLUSIONS

Low vitamin D status is strongly linked to higher thyroid autoimmunity and hypothyroidism in patients with Hashimoto’s thyroiditis. Even in a sun-rich region, deficiency is common, suggesting metabolic or environmental factors. Routine vitamin D assessment and correction may help reduce autoimmune activity and improve thyroid function.

Blood sampling

P010

PATIENT PSYCHOLOGICAL PREPARATION AS AN UNDERESTIMATED ASPECT INFLUENCING THE QUALITY OF BIOLOGICAL MATERIALS.

A. Ochocińska ³, L. Staniszewska ³, E. Król ², D. Wojciechowska-Sypuła ¹

¹Blood Collection Point, The Children’s Memorial Health Institute

²Core Laboratory, The Children’s Memorial Health Institute

³Department of Clinical Biochemistry, The Children’s Memorial Health Institute

BACKGROUND-AIM

The psychological preparation of patients for laboratory sample collection is often underestimated, even though it plays a crucial role in the diagnostic process. Anxiety, stress, or tension can affect patient cooperation and consequently hinder proper sample collection. Inadequate emotional preparation may reduce the quality of the specimen, increasing the risk of inaccurate test results. The aim of this study was to assess the impact of stress, pain, and previous negative patient experiences on the quality and quantity of test material obtained.

METHODS

To assess patients’ fear of blood sampling, a custom-designed survey was included with each referral. The study was conducted among 300 pediatric patients visiting the Blood Collection Point on three randomly selected workdays. The survey was completed by the parent and/or the child, depending on their age. The quantity and quality of the sample was assessed by laboratory staff.

RESULTS

Of the 300 biological samples collected, 3 (1%) did not provide sufficient sample material to perform all planned tests. Of the 300 samples collected, 7 (2.3%) were rejected by the laboratory due to inadequate sample quality (4 - hemolysis, 2 - lipemia, 1 - clot). Low-grade hemolysis was noted in 25 (8.3%) samples, which did not result in rejection due to the type of test ordered. In addition to the 300 successfully collected tests, 3 blood draws were postponed due to difficulties with collection caused by fear (including one fainting). In 139 (46.3%) surveys, patients/parents reported fear of the procedure, and in 46 (15.3%) surveys, negative previous experiences were reported. Contrary to the data for rejected samples, 85% agreement was achieved.

CONCLUSIONS

The above data clearly indicate the need to pay attention to the patient’s psychological preparation for testing. This may be the missing piece of the puzzle in the preanalytical phase.

Blood sampling

P011

SMALL HEART - GREAT COURAGE. STRESS-FREE BLOOD COLLECTION FOR CHILDREN IN THE CHILDREN’S MEMORIAL HEALTH INSTITUTE (CMHI) IN WARSAW.

A. Ochocińska ³, L. Staniszewska ³, E. Król ², D. Walecka-Makulec ², D. Wojciechowska-Sypuła ¹

P017

ASSESSMENT OF THE IMPACT OF HEMOLYSIS ON SEVENTEEN BIOCHEMICAL PARAMETERS

S. Abdi ¹, S. Bennouar ¹, A. El Hadj Tahar ¹, N. Remidi ¹, M. Irvin ¹

¹Blida 1 university

BACKGROUND-AIM

Medical laboratories are essential to patient care, influencing 60 to 70% of major clinical decisions. Despite advances in automation and quality assurance, errors persist, particularly during the pre-analytical phase, which is the most vulnerable. Hemolysis is the most common pre-analytical non-compliance, leading to inaccurate results, misdiagnosis, and inappropriate patient management. Routine and emergency medical biochemistry tests are particularly affected.

Objectives: This study aimed to investigate the interference of hemolysis on the measurement of seventeen common biochemical parameters, determining the direction of interference (overestimation or underestimation) and the plasma hemoglobin concentration at which the impact becomes significant.

METHODS

An experimental study was conducted at the central medical analysis laboratory of Blida University Hospital. Samples that were visibly hemolyzed, lipemic, or icteric were excluded. A hemolysate was prepared from whole blood-EDTA, and erythrocytes were lysed with distilled water. This hemolysate was added to plasma pools to create a range of increasing hemoglobin concentrations from 0 to 1 g/dl. The parameters investigated included LDH, CPK, ALAT, ASAT, GGT, PAL, creatinine, urea, glucose, total cholesterol, triglycerides, uric acid, calcium, phosphorus, magnesium, sodium, and potassium.

RESULTS

Parameters were classified into three categories: 1) Not significantly affected: urea and sodium. 2) Positively impacted (overestimation): transaminases, LDH, CPK, total cholesterol, glucose, creatinine, magnesium, phosphorus, triglycerides, uric acid, and calcium. 3) Negatively impacted (underestimation): GGT and PAL.

CONCLUSIONS

Hemolysis affects the plasma concentration of many biochemical parameters, with a significant impact on enzyme activities and potassium. Therefore, controlling the pre-analytical phase is crucial.

Key words: hemolysis, LDH, potassium

Hemolysis, icterus, lipemia

P018

HAEMOLYTIC INDEX AND INTERFEROGRAM ON ROCHE COBAS ANALYSER FOR GENERAL ANALYTES

M.I. Ahmad ¹, F.S. Ahmed ¹, D. Rao ¹, M. Begum ¹, S. Sequeria ¹

¹Dubai Academic health corporation

BACKGROUND-AIM

DOES HEMOLYSIS INTERFERE WITH POTASSIUM CONCENTRATION IN DOG SAMPLES?

B. Beer LjubiĆ ¹, J. AladroviĆ ³, N. PuvaČa ¹, A. Žuvela ¹, V. ĐuriĆ ¹, D. IvŠiĆ Škoda ¹, V. Vidranski ²

¹Clinic for Internal Diseases, Faculty of Veterinary Medicine, University of Zagreb, Croatia

²Department of Clinical Chemistry, Sestre milosrdnice University Hospital Center, Zagreb, Croatia

³Department of Physiology and Radiobiology, Faculty of Veterinary Medicine, University of Zagreb, Croatia

BACKGROUND-AIM

Erythrocyte potassium (K) concentration depends on Na⁺/K⁺ pump activity in the erythrocyte membrane. Human erythrocytes contain high K concentrations, most dog breeds are low-K breeds whereas Akita Inu, Shiba Inu and some Asian breeds are high-K breeds. The aim was to determine K concentration and hemolysis index (HI) in heparin plasma of different dog breeds and humans after artificially induced hemolysis.

METHODS

Twelve non-hemolyzed heparinized whole blood samples from dogs of various breeds and three from healthy human volunteers were analyzed. Hemolysis was induced by aspiration and osmotic shock. Aspiration was performed using a fine-gauge needle, while osmotic shock was induced by serial dilution with distilled water and saline. Potassium and HI were measured on the Architect c4000 (Abbott Diagnostics, USA). Clinical Laboratory Improvement Amendments (CLIA) limits from human medicine (total allowable error, TEa ±0.5 mmol/L or 5%) were applied as acceptance criteria.

RESULTS

After aspiration-induced hemolysis, bias exceeded TEa in three dogs (two mixed breeds and one Shiba Inu), while in the other nine hemolysis did not cause clinically significant interference and bias remained within CLIA limits. In human samples, K increased proportionally with HI and exceeded the acceptance criterion. Osmotic hemolysis of canine samples by dilution with distilled water decreased K, while HI ranged from 2.4 to 32.1 g/L. In contrast, human samples showed parallel increases, with K up to 12.0 mmol/L and HI of 42.0 g/L at 60% dilution.

CONCLUSIONS

Potassium concentration increases in hemolyzed human samples and in high-K dog breeds with genetically higher Na⁺/K⁺ pump activity. In most dog breeds and mixed breeds, hemolysis does not significantly affect potassium, suggesting that in veterinary practice it should not be considered a major interference, except in high-K breeds where reporting the dog breed is essential to avoid preanalytical errors.

Hemolysis, icterus, lipemia

P022

PERFORMANCE OF LIPEMIA CORRECTION METHODS IN CLINICAL LABORATORY ANALYSIS

D. Dimova ¹, G. Chausheva ², S. Shefket ², Y. Bocheva ²

¹Central Clinical Laboratory, St. Marina University Hospital, Varna, Bulgaria

²Medical University “Prof. Dr. Paraskev Stoyanov”, Varna, Bulgaria

BACKGROUND-AIM

Introduction: Lipemia is a common preanalytical interference affecting laboratory accuracy.

Objective: To evaluate the impact of lipemia on laboratory parameters and the performance of correction methods.

METHODS

Sixty serum (Cobas 6000), 10 whole blood (Advia 2120), and 10 citrated plasma (Sysmex CS-2500) samples were analyzed. Lipemia was induced with Intralipid 20% to triglycerides 5, 10, and 20 mmol/L (mild, moderate, severe). Serum was processed by 1:1 saline dilution (SD), PEG precipitation, and high-speed centrifugation (HSC). Whole blood was treated by isovolumetric substitution (IS) and Cellular Hb (CHb) measurement, while plasma was processed by HSC. Statistical analysis used paired t-tests and ANOVA.

RESULTS

Hemoglobin values were significantly higher than native samples at all lipemia levels (bias>2%). CHb results remained reliable up to moderate lipemia, while severe lipemia caused significant bias. IS did not correct the interference, with Hgb bias increasing with lipemia severity.

Results for coagulation parameters (INR, APTT, and Fbg) showed no significant differences compared to native samples in mild lipemia. In moderate lipemia, INR showed a significant positive bias (>6%), which was corrected by HSC. In severe lipemia, pre-HSC results were unavailable, post-HSC - APTT and Fbg values were comparable to native samples, while INR remained underestimated.

Serum analytes - Glu, UN, Crea, UA, Alb, GGT, ALP, Ca, P, Mg, and CRP, remained comparable to native samples at all lipemia levels. Mild lipemia affected DB, ALT, AST, moderate also Na, Cl, severe additionally TP, K. After SD and PEG treatment, significant biases were observed for Na, K, Cl (>3%) and DBil, ALT, AST, TP (>10%). After HSC, all parameters were comparable to native samples, regardless of lipemia severity.

CONCLUSIONS

CHb measurement remains accurate in mild to moderate lipemia, while IS does not effectively correct interference. HSC is the most reliable method for minimizing lipemic interference in serum and citrated plasma.

Hemolysis, icterus, lipemia

P023

FREQUENCY AND PREVENTION OF HEMOLYSIS IN A CLINICAL LABORATORY SETTING – A SINGLE CENTER STUDY

M. Cojic ¹, A. Stojcev ¹, V. Stojiljkovic ², M. Cosic Petkovic ³, V. Cosic ¹

¹Center for Medical and Clinical Biochemistry, University Clinical Center Nis, Serbia

EFFECT OF HEMOLYSIS AND LIPEMIA ON INTERLEUKIN 6 MEASUREMENT IN HUMAN SERUM SAMPLES

T. Bašić ⁴, H. Čičak ⁴, A. Saračević ⁴, A. Dorotić ⁴, L. Drinovac ⁴, A. Glavan ³, E. Galić ¹, I. Alfirević ²

¹a) Internal Medicine Clinic, University Hospital Sveti Duh, Zagreb, Croatia; b) School of Medicine, University of Zagreb, Zagreb, Croatia

P032

ADOPTING OVER-CAUTIOUS HIL INDEX CUT-OFFS CAN DELAY PATIENT CARE BY NOT RELEASING CLINICALLY VALID RESULTS

R. Marrington ¹, F. Mackenzie ¹

¹Birmingham Quality, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK

BACKGROUND-AIM

While Laboratories are aware of between-method differences for Thyroid-Function-Test results they may not be aware of the effect of haemolysis on TSH and fT4 levels and the impact of using haemolysis index cut-offs advised by manufacturers on the number of results suppressed and not reported.

METHODS

The UK NEQAS for Serum Indices is a monthly EQA scheme operated by Birmingham Quality, UHB. It challenges laboratories with three liquid, ready-to-use, specimens. Laboratories analyse all three HIL Indices (haemolysis, icterus and lipaemia). Additionally, laboratories measure a selected analyte (Analyte X) and report back both the numerical concentration obtained and whether Analyte X would be reported based on the HIL values. HIL cut-offs for Analyte X are also collected.

In May and August 2024, Distributions 183 and 188 were dispatched which contained serum with varying concentrations of haemolysate added, with the respective Analyte X’s being TSH and fT4.

RESULTS

Over 550 participants returned results. Minimal differences were observed in TSH and fT4 results as the haemolysis index increased to 6 g/L. There was a slight, not clinically significant, decrease in TSH and fT4 for Abbott Alinity. What was notable was approximately 80% of Abbott Alinity users would not report a TSH or a fT4 result in a haemolysed sample of ∼6g/L haemolysis index. This contrasts with only 5% of Roche users that would not report the results. Beckman Olympus and Siemens Atellica also showed similar post-examination results to Abbott Alinity. This is reflective of the cut-offs in use by laboratories and manufacturers where there is both within and between manufacturer variation.

CONCLUSIONS

Getting the correct result is important, but so is reporting a result in a timely manner, without the need for re-bleeding patients, to allow effective decisions to be made which could directly impact patient care.

(Keywords: HIL, Serum Indices, Thyroid)

Hemolysis, icterus, lipemia

P033

THE IMPORTANCE OF VERIFICATION OF MANUFACTURER HIL CUT-OFFS

R. Marrington ¹, F. Mackenzie ¹

¹Birmingham Quality, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK

BACKGROUND-AIM

Most, if not all, clinical biochemistry analytes analysed on automated analysers have a ‘manufacturer defined’ cut-off for interference by haemolysis (H), icterus (I) and lipaemia (L). Laboratories use these cut-offs within their computer systems which allow auto-comments to be appended and analyte results not reported.

METHODS

The UK NEQAS for Serum Indices Scheme distributes specimens to >700 participants for analysis of HIL and Analyte X — a different analyte each month. Participants are asked to provide their Analyte X HIL cut-off and if they would report Analyte X based on their reported HIL result.

At Distribution 185, three lipaemic specimens were distributed with progesterone as Analyte X. Progesterone (30 nmol/L) had been spiked into each specimen.

RESULTS

As expected, there were wide differences between manufacturers for the lipaemia index: specimen 185A 1–2 mmol/L, 185B 2–4 mmol/L and 185C 3–7 mmol/L. Because the three specimens are not inter-related, concentrations cannot simply be compared. However, relative method biases showed that increasing lipaemia did not significantly impact the progesterone result.

Interestingly, ∼50% of participants would not report progesterone based on the HIL for specimen 185C (this included 80% of the Roche users — the most popular method). Progesterone would also not be reported by ∼80% Roche users for specimen 185B.

CONCLUSIONS

Despite the relatively low cut-offs used by Roche for their lipaemic index for progesterone, there is very little to suggest that there has actually been any detrimental impact on the progesterone results themselves when assays are challenged with a genuine lipaemic specimen.

Every test result has a patient behind it, waiting on a decision for referral/treatment. Some samples cannot be collected again, there may be specific time constraints, as may be the case for Progesterone.

This is only one example, the concept holds true across laboratory medicine.

(Keywords: Lipaemia, Verification)

Hemolysis, icterus, lipemia

P034

EVALUATION OF SAMPLE INTEGRITY USING THE SIM MODULE: A COMPUTER VISION AND MACHINE LEARNING APPROACH

G. Montesano ², G. Gioiello ², G. Maccioni ¹, M. Martines ¹, P. Caropreso ³, G. Mengozzi ²

¹Inpeco Spa, Val della Torre (Turin) Italy

²Laboratory of Clinical Biochemistry, Città della Salute e della Scienza University Hospital of Turin, Turin, Italy / Department of Medical Sciences, University of Turin

³Laboratory of Clinical Biochemistry, Città della Salute e della Scienza University Hospital of Turin, Turin, Italy.

BACKGROUND-AIM

The Sample Integrity Module (SIM), developed by Inpeco, enables the analysis of serum integrity, thereby reducing the impact of out-of-range serum indices on the workflow of tubes sent to analyzers.

METHODS

A collaborative study between Inpeco and the Clinical Biochemistry Laboratory, Città della Salute e della Scienza University Hospital of Turin, validated a vision system and machine learning model for class-based serum index (HIL) estimation. Initial analysis compared SIM results with those of Beckman AU5800 analyser, followed by data from SIM modules integrated with FlexLab and HIL-capable analysers in other centres. The system analyses tube reflection at specific wavelengths under white LED light.

RESULTS

A total of 105,909 tubes were processed. SIM classification results were compared with analyser data using the following thresholds: for hemolysis, 50, 100, 200, 300, and 500 mg/dL; for icterus, 5, 10, 20, and 30 mg/dL. The application of the machine learning algorithm to SIM-acquired data demonstrated high concordance with analyser results.

For hemolysis, 99.31% of samples were correctly classified, with 0.35% underestimated and 0.34% overestimated. Among discordant cases, 99.95% showed minor discrepancies (±1 class), and 0.03% showed major discrepancies (>1 class). For icterus, 99.40% were correctly classified, with 0.18% underestimated and 0.42% overestimated; among discordant samples, 99.91% exhibited minor discrepancies and 0.09% major discrepancies. Lipemia assessment was excluded due to limited sample availability despite high concordance.

CONCLUSIONS

The SIM Module analyses samples directly in the original tube, without aliquoting. Results are expressed in predefined classes and are not equivalent to quantitative analyser measurements. However, this classification offers useful insights into sample quality and supports laboratory workflow and decision-making.

Hemolysis, icterus, lipemia

P035

RELIABILITY OF GEM PREMIER 7000 WITH IQM3 HEMOLYSIS DETECTION IN THE PRESENCE OF INTERFERENTS

P. Pamidi ¹, S. Balasubramanian ¹, L. Adib ¹, N. Mcnary ¹, M. Lagene-Bazille ¹

¹Werfen, Acute Care Dx R&D, Bedford MA, USA

BACKGROUND-AIM

Routine assessment of hemolysis on blood gas samples is now possible with the GEM Premier 7000 with iQM3 (GEM 7000) system. The system offers an integrated hemolysis detection for arterial, venous, and capillary blood gas specimens. One challenge for hemolysis detection is that other endogenous interferents, such as bilirubin and/or lipids, tend to co-exist in whole blood specimens. Here, we aimed to assess the effect of bilirubin and lipids on GEM 7000 hemolysis performance.

METHODS

The GEM 7000 system employs a novel acoustofluidic technology to instantaneously expose a plasma region on a whole blood sample and measure hemolysis using photometric methods. From absorbance measurements, the system quantifies hemolysis and converts it into six categories (i.e., 0-50, 51-115, 116-200, 201-300, 301-400, ≥401 mg/dL). For the interference assessment on hemolysis, remnant clinical bilirubin samples and 20% intravenous Intralipid® solution were used. Two levels of bilirubin or intralipid spiked whole blood sample spanning over four hemolysis ranges were tested in triplicate measurements on the GEM 7000, and the results were evaluated for agreement in hemolysis index categories.

RESULTS

Overall, in the presence of bilirubin and intralipid, the results showed no difference between the expected and flagged categories across the tested hemolysis ranges. Specifically, in the presence of bilirubin concentrations 9.1 and 23.4 mg/dL, native whole blood showed a ≤ 6 mg/dL bias on hemolysis, while most of the hemolyzed blood results remained within ±10% of the target values. Similarly, when Intralipid was added at concentrations of 125 and 217 mg/dL, it resulted in a modest, non-clinically significant bias in native blood, whereas hemolyzed blood results stayed within ±12% of the target values.

CONCLUSIONS

These results demonstrate that the GEM 7000 hemolysis detection is not impacted in the presence of cross-interference such as bilirubin and lipids for providing reliable hemolysis detection in whole blood samples.

Hemolysis, icterus, lipemia

P036

INFLUENCE OF PNEUMATIC TRANSPORT SYSTEMS ON HEMOLYSIS RATES IN CLINICAL LABORATORY SAMPLES

J. Salgado Sanchez ¹, C. Santamaria Arellano ¹, D. Ruedas Lopez ¹, M. Torres Fernandez ¹, L. Criado Gomez ¹

¹Hospital Universitario de Mostoles

BACKGROUND-AIM

Hemolysis is defined as the release of intraerythrocytic components into plasma as a consequence of red blood cell rupture.

In hospital settings, pneumatic tube systems are frequently used for transporting samples. However, it is not fully established whether these systems may influence test results.

The primary objective of this study was to evaluate potential differences in hemolysis indices between samples transported manually and those delivered via the pneumatic tube system.

HEMOLYSIS IN STAT SAMPLES AND PATIENT DEMOGRAPHICS IMPACT

V. Tole ², H. Lame ², A. Coraj ², N. Heta ², I. Korita ², A. Bulo ¹, E. Refatllari ²

¹Laboratory Department, Faculty of Medicine, University of Medicine, Tirane

²Laboratory Department, Faculty of Medicine, University of Medicine, Tirane; Laboratory Networks, Mother Theresa University Hospital Centre

BACKGROUND-AIM

Hemolysis is a common pre-analytical issue in clinical laboratories and can compromise the reliability of results. This study aims to evaluate the incidence of hemolysis in STAT samples and the influence of demographic factors.

METHODS

This cross-sectional study was conducted at Laboratory Networks, University Hospital Center ‘Mother Theresa’, Albania. A total of 700 consecutive STAT samples were analyzed for hemolysis on Alinity c Abbot analyzer. Hemolysis was categorized on a 5-point scale (0–4+). Patient data included gender and age (analyzed both as continuous and grouped into ten 10-year intervals). Associations between gender and hemolysis presence were examined using chi-square tests. Correlations between age and hemolysis degree were evaluated with Kendall’s tau-b and Spearman’s rho. The data were analyzed using IBM SPSS Statistics 26. P-values <0.05 were considered statistically significant.

RESULTS

The mean age of patients was 55.6 ± 24.2 years. The largest age groups were 61–70 years (25.4%) and 71–80 years (23.4%). Of the 700 patients, 317 (45.3%) were female and 383 (54.7%) were male. Hemolysis was present in 17% of the samples. The distribution of hemolysis degrees was 1+ in 13.6%, 2+ in 2.1%, 3+ in 0.9%, and 4+ in 0.4% of samples. A weak but statistically significant positive correlation was found between age and hemolysis degree (Kendall’s tau-b = 0.068, p = .026; Spearman’s rho = 0.084, p = .026). Gender was not significantly associated with hemolysis presence (χ² (1, N = 700) = 0.98, p = .323, φ = 0.04). Hemolysis occurred in 15.5% of female and 18.3% of male samples.

CONCLUSIONS

Hemolysis was present in 17% of STAT samples, but mostly in low degrees. While age showed a statistically significant but weak correlation with hemolysis degree, gender had no significant impact. These findings suggest that hemolysis in STAT samples is only minimally influenced by patient demographics.

Hemolysis, icterus, lipemia

The study of tHyc stability showed that storage conditions have a significant impact. We recommend processing blood samples within one hour of collection. Alternatively, whole blood can be stored at +4°C for up to 4 hours.

Sample stability

Therefore, we propose that cancer antigen (CA-19-9), luteinizing hormone (LH), prolactin (PRL) and free thyroxine (FT4) are not stable at -20 °C and should be kept at -80° C if required.

Sample stability

P046

SEPARATION AND SEASONALITY IN POTASSIUM ANALYSIS: EFFECTS ON PRIMARY-CARE SERVICES ACROSS THREE LABORATORY CONFIGURATIONS

S. Costelloe ¹, M. Cornes ², S. Hepburn ³, M. Myers ⁴

¹Department of Clinical Biochemistry, Cork University Hospital, Wilton, Cork, Republic of Ireland.

²Department of Clinical Biochemistry, Worcestershire Acute Hospitals NHS Trust (WAH), Worcester, United Kingdom (UK)

³Department of Laboratory Medicine, Raigmore Hospital (NHS Highland (NH)), Inverness, UK

⁴NHSE Getting it Right First Time (GIRFT), Pathology

BACKGROUND-AIM

Serum or plasma potassium (K) is sensitive to delayed separation and ambient temperature; we assessed their combined and individual effects at three lab sites serving primary care (PC).

METHODS

PC K requests in 2023 were analysed from WAH, CUH, and NH. Time of collection to centrifugation (TCTC) and Month (considered singly, and grouped (warm (May-Oct) and cold (Nov-April))) were considered for interactions and effects on K endpoints (median K in mmol/L (K̃); and frequency of hyperkalaemia (K >5)). For continuous outcomes by category, Kruskal-Wallis (KW) tests were used; for categorical outcomes, the χ² test was employed; and interaction models assessed joint effects.

RESULTS

All estimable KW and χ² tests yielded p<0.001. TCTC varied by Month: warm vs cold Month medians (h) were WAH: 8.48 & 8.52; CUH: 6.8 & 6.7, and NH: 8.8 & 21.7. Within months, K̃ rose with increasing TCTC at every site. Monthly K̃ ranges were 4.1-4.4 (WAH; Jun→Nov); 4.3-4.7 (CUH; Jun→Nov); and 4.3-4.4 (NH; Nov→Jan). Hyperkalaemia by month spanned 4.1–9.4% (WAH; Jun→Jan), 4.5–23.0% (CUH; Jun→Jan), and 3.2–5.2% (NH; Sep→Jan). The slope of K̃ vs TCTC differed by Month at WAH (interaction p=0.007), but not at CUH or NH (p>0.1), indicating seasonal amplification at WAH and month-stable TCTC effects at CUH. Differences between the highest and lowest TCTCxMonth bins were 0.90, 1.10, and 0.30 at WAH, CUH, and NH, respectively.

CONCLUSIONS

Observed impacts on K seem to depend on service configuration. Where samples arrive uncentrifuged (CUH), TCTC effects are substantial year-round (higher hyperkalaemia burden). At WAH, the effect intensifies during cold months, particularly with TCTC ≥ 12 h. In NH, near-universal pre-centrifugation essentially neutralises TCTC and season effects despite long transit times. Targeted interventions—pre-centrifugation and temperature-controlled courier chains—are likely to reduce spurious hyperkalaemia and downstream clinical noise.

Sample stability

P047

ASSESSING THE TIME-DEPENDENT STABILITY OF KEY ANALYTES IN BIOBANK SAMPLES

E. D’Angelo ¹, A. Aita ², C. Benna ¹, L. Sciacovelli ², D. Basso ², S. Pucciarelli ¹, M. Agostini ¹, M. Montagnana ², G. Spolverato ¹

H. Čičak ⁵, Ž. Šarčević ⁴, A. Dorotić ⁵, L. Dukić ³, A. Saračević ⁵, E. Galić ¹, I. Alfirević ²

¹a) Internal Medicine Clinic, University Hospital Sveti Duh, Zagreb, Croatia; b) School of Medicine, University of Zagreb, Zagreb, Croatia

²a) University Department of Surgery, University Hospital Sveti Duh, Zagreb, Croatia; b) Faculty of Dental Medicine and Health, Josip Juraj Strossmayer University of Osijek, Osijek, Croatia; c) University North, Varaždin, Croatia

SALIVA AND FLOW: WHAT INFORMATION DOES IT PROVIDE ON PEDIATRIC INFLAMMATORY BOWEL DISEASE?

A. Falda ³, N. Contran ¹, E. Nordi ¹, P. Gaio ⁴, M. Cananzi ⁴, G. Paiusco ⁴, G. Musso ², P. Galozzi ², A. Aita ², D. Basso ², M. Montagnana ², A. Padoan ²

¹Department of Biomedical Sciences – DSB, University of Padova, Italy

²Department of Medicine – DIMED, University of Padova, Italy

³Laboratory Medicine Unit, University-Hospital of Padova, Italy

⁴Unit of Pediatric Gastroenterology, Digestive Endoscopy, Hepatology and Care of the Child with Liver Transplantation, Department of Women’s and Children’s Health, University-Hospital of Padova, Italy

BACKGROUND-AIM

Pediatric Inflammatory Bowel Disease (PIBD) is a chronic condition requiring invasive monitoring. This study aimed to standardize a flow cytometry protocol for assessing salivary lymphocytes as a reliable non-invasive biomarker of disease activity.

METHODS

Control saliva was collected using E-saliva, also after brushing or antiseptic application. Filtration and fixation steps were introduced to optimize sample quality. Cell viability and oral squamous epithelial cells (OSEC) were assessed using 7-Aminoactinomycin D (7-AAD) and cytokeratin. Multicolor-stained samples were analyzed on BC DxFlex cytometer and blood lymphocytes were used to refine gating. Lymphocyte counts were determined using beads.

The final cohort included 37 PIBD patients (31 Crohn’s disease, 6 ulcerative colitis) and 23 controls.

RESULTS

Saliva contains abundant OSEC and bacteria, contributing to intense autofluorescence. Filtration and fixation reduced lymphocyte recovery compared to unfiltered and fresh saliva. Viability assessment using 7-AAD revealed that ∼80% of bacteria and OSEC were non-viable within one hour of collection. Pre-collection oral cleaning did not reduce matrix autofluorescence. Counting beads enabled standardized salivary lymphocyte quantification.

Among PIBD patients, 13 (35%) were positive for salivary lymphocytes (pL+), while 24 (65%) were negative; all controls were negative. Crohn’s disease was more frequent in pL+ (84% vs 50%, p = 0.004). In pL+, 30% had active diseases; of the 70% in remission, half had activity within three months.

Saliva showed a median CD4/CD8 ratio of 2.25 and high CD3+ HLA-DR expression (median 57.38%). No correlation was found with blood activation markers; however, salivary lymphocyte counts correlated with ESR and clinical scores (rho = 0.7857 and 0.5752; p < 0.04).

CONCLUSIONS

We demonstrated the potential of fresh saliva, analyzed by flow cytometry using a tailored gating strategy, to detect lymphocytes as markers of disease activity in PIBD.

Sample stability

P054

LOWER 25-OH VITAMIN D VALUES IN HEPARIN PLASMA THAN IN SERUM: A POSSIBLE PREANALYTICAL MECHANISM

N. Fogh-Andersen ¹

¹Department of Clinical Biochemistry, Copenhagen University, Herlev Hospital, Herlev

BACKGROUND-AIM

We investigated the apparent difference between 25-OH vitamin D in serum and heparin plasma.

METHODS

Blood was collected from 15 volunteers in additive-free serum tubes and lithium heparin tubes (Greiner, 10 U/mL). Samples were stored at 20 degrees C for 2 - 4 h prior to centrifugation, separation, and measurement of 25-OH vitamin D on a Siemens Centaur immunoassay.

RESULTS

25-OH vitamin D concentrations were 14% lower in heparin plasma than in serum (p < 0.001, paired t-test). Regression analysis gave P = 0.86*S - 4.1 nmol/L, r =0.99.

CONCLUSIONS

Maehira et al. (1990) demonstrated that heparin can release lipoprotein lipase from monocytes in vitro.

We suggest that in vitro generation of free fatty acids from plasma triglycerides may influence analyte-binding equilibria or optical properties, particularly for lipophilic analytes such as 25-OH vitamin D, steroid hormones, or certain drugs. This may represent a hidden variable affecting serum-plasma comparability in immunoassays and highlights the need for method-specific matrix validation beyond manufacturer claims.

Reference: Maehira F, Miyagi I, Eguchi Y. Biochim Biophys Acta (BBA) - Lipids and Lipid Metabolism 1990;1042:344-51.

Sample stability

P055

A PRE-ANALYTICAL SOLUTION: OVERCOMING 5-FU INSTABILITY IN PRECISION ONCOLOGY

P. Gavel ¹, M. Gurjar ², N. Mohapatra ¹

¹Department of Biochemistry; Homi Bhabha Cancer Hospital and Mahamana Pandit Madan Mohan Malaviya Cancer Centre, Varanasi (A unit of Tata Memorial Centre, Homi Bhabha National Institute, Mumbai) India.

²Department of Clinical Pharmacology; Homi Bhabha Cancer Hospital and Mahamana Pandit Madan Mohan Malaviya Cancer Centre, Varanasi (A unit of Tata Memorial Centre, Homi Bhabha National Institute, Mumbai) India.

BACKGROUND-AIM

The study aimed to address the in-vitro instability of the drug in whole blood and plasma, which is caused by the Dihydropyrimidine dehydrogenase (DPD) enzyme. The hypothesis was that an optimal concentration of a DPD inactivator would improve 5-FU stability, thereby reduce in-vitro variability and enable more accurate therapeutic drug monitoring.

METHODS

A prospective pilot study with 150 participants was conducted from 26-06-2023 to 23-10-2023 after IEC approval. Pooled blood samples were spiked with a 5-FU standard stock solution and were then divided into two parts. In the first experiment, samples were treated with increasing concentrations of a DPD inactivator to determine the optimal dose. In the second experiment, samples with and without the inactivator were analysed at multiple time points to assess drug stability over time. A HPLC system was used for measuring the concentrations of 5-FU.

RESULTS

In the first phase of the experiment, samples containing the DPD inactivator showed a 21.5% and 0.68% increase in mean 5-FU concentrations compared to samples without the inactivator. This indicates that the DPD inactivator was highly effective in preserving the drug.

CONCLUSIONS

The in-vitro instability of 5-FU in human blood was identified as the primary challenge. This instability is a major pre-analytical hurdle because the DPD enzyme, which is abundant in blood, rapidly degrades the drug after the sample is collected. This degradation leads to a significant reduction in the measured 5-FU concentration over time, making it difficult to accurately assess the actual drug levels in the patient’s body. The study was specifically designed to overcome this challenge by introducing a DPD inactivator to the blood sample during collection, thereby stabilizing the 5-FU and ensuring that the measured concentration more closely reflects the true therapeutic level.

Keywords: 5-FU, DPD Enzyme, Cancer, HPLC, Pre-Analytical, Precision Oncology

Sample stability

CONCLUSIONS

Sample temperature plays a pivotal role in maintaining the integrity of blood ammonia measurements. Immediate cooling of samples and prompt analysis are strongly recommended to avoid artifactual elevations that may lead to misdiagnosis and inappropriate clinical decisions. Implementing strict pre-analytical protocols significantly enhances the reliability of blood ammonia testing.

Sample stability

P059

COMPARISON OF ANALYTE STABILITY IN URINE SAMPLES PRESERVED IN VACUETTE® URINE STB TUBE WITH BD VACUTAINER® URINALYSIS PRESERVATIVE PLUS URINE TUBE

S. Jovičić ³, M. Neschi ², B. Seifert ², E. Füreder-Kitzmüller ², A. Verstraete ¹, M. Oyaert ⁵, A. Šimundić ⁴

¹Cerba Healthcare, Gent, Belgium

²Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria

³Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria; Department for Medical Biochemistry, Faculty of Pharmacy, University of Belgrade, Serbia

⁴Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria; Faculty of Pharmacy and Biochemistry, University of Zagreb, Croatia

⁵Department of Laboratory Medicine, Ghent University Hospital, Gent, Belgium

BACKGROUND-AIM

A multicenter comparative analysis was performed to evaluate the stability of quantitative urine analytes, test strips, and particle analysis in the new VACUETTE^® Urine STB Tube (Greiner Bio-One GmbH, Kremsmünster, Austria), against the Vacutainer^® UAP Urine Tube (Becton, Dickinson and Company, Franklin Lakes, NJ, USA).

METHODS

Anonymized urine samples, 343 in total, were transferred to STB and UAP Tubes within 2h of collection. Analyte stability was assessed by comparing initial values (T0) with measurements at 4, 8, 24, 48, and 72 hours at room temperature. Quantification of urinary creatinine, urea, uric acid, total protein, albumin, calcium, potassium, and chloride (Roche Cobas c702) and test strip testing (Roche Cobas U601) was performed on 121 samples; particle counts for erythrocytes, leukocytes, epithelial cells, and bacteria were done using automated microscopy (Roche U701) and flow cytometry (Sysmex UF-5000), in 120 and 102 samples, respectively. For quantitative parameters, bias versus T0 was acceptable if below the Total Change Limit (TCL); for ordinal scale test strips, no more than one level change in 95% of samples was allowed, while qualitative parameters required 95% consistency with T0 results.

RESULTS

All quantitatively measured parameters in samples from both STB and UAP tubes remained stable at room temperature up to 72h. Stability was demonstrated for up to 72h across all test strip parameters and analyzed particle counts using both automated microscopy and flow cytometry. The exception was blood on test strips, which was stable for only 24h in both tubes.

CONCLUSIONS

The STB and UAP Tubes are equivalent in maintaining sample stability, effectively preserving analytes for urine tests up to 72h at room temperature. Blood test strip samples are stable for only 24h in both tubes, likely due to the quick loss of hemoglobin pseudoperoxidase activity.

Sample stability

P060

VACUETTE® URINE STB VS BD VACUTAINER® URINALYSIS PRESERVATIVE PLUS URINE TUBE: QUANTITATIVE CHEMISTRY, TEST STRIPS, AND PARTICLE COUNTS COMPARISON

S. Jovičić ³, M. Neschi ², B. Seifert ², E. Füreder-Kitzmüller ², A. Verstraete ¹, M. Oyaert ⁵, A. Šimundić ⁴

¹Cerba Healthcare, Gent, Belgium

²Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria

³Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria; Department for Medical Biochemistry, Faculty of Pharmacy, University of Belgrade, Serbia

⁴Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria; Faculty of Pharmacy and Biochemistry, University of Zagreb, Croatia

⁵Department of Laboratory Medicine, Ghent University Hospital, Gent, Belgium

BACKGROUND-AIM

The VACUETTE^® Urine STB Tube (Greiner Bio-One GmbH, Kremsmünster, Austria) preserves urine cellular and chemical components for up to 72 hours. This study compares its analytical performance against the BD Vacutainer^® Urinalysis Preservative Plus (UAP) Tube (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) focusing on analytical performance of quantitative chemistry, test strips, and particle counts.

METHODS

A total of 343 anonymized urine samples were transferred to STB and UAP tubes 2h after collection. Analyses for creatinine, urea, uric acid, total protein, albumin, calcium, potassium, and chloride (Cobas c702), and test strip analysis (Roche Cobas U601), were performed on 121 samples. Particle counts for erythrocytes, leukocytes, squamous epithelial cells, and bacteria were determined using automated microscopy (Roche U701) and flow cytometry (Sysmex UF-5000) in 120 and 102 samples, respectively. Tube agreement was assessed by Deming regression and Bland-Altman plots. Clinical bias was compared to predefined Acceptance Limits (CAL) (Rili-Bäk, Royal College of Pathologists of Australasia). No significant difference was noted when mean bias ± 95% CI were within CAL. For ordinal tests and particle counts, ≤1 level deviation between tubes in ≥95% was acceptable; qualitative test strips required ≥95% concordance for positive/negative outcomes.

RESULTS

All quantitative biochemistry parameters showed minimal mean bias between STB and UAP tubes, well below CAL, with Pearson’s r >0.99. Test strip ordinal results matched within one level except for specific gravity, which differed in only 0.7% of samples. Qualitative test strip results showed 100% agreement, and particle counts differed by no more than one level between tubes.

CONCLUSIONS

STB and UAP tubes exhibit high concordance in quantitative, test strip analysis, and particle counts. Statistical analysis demonstrates strong agreement with no significant bias, indicating the STB Tube is a reliable option for urinalysis.

Sample stability

P061

EU INSTAND-NGS4P: NOVEL PRE-ANALYTICAL WORKFLOWS FOR MULTIMODAL NGS RESEARCH AND FUTURE PRECISION CANCER CARE

F. Kaiser ¹, K. Guenther ¹, M. Sprenger-Haussels ¹, U. Oelmueller ¹

¹QIAGEN GmbH

BACKGROUND-AIM

The EU funded Pre-Commercial Procurement project Instand-NGS4P (GA no. 874719) aimed at improving cancer patient’s benefit from NGS by developing standardised workflows from specimen collection to final report. The pre-analytical phase accounts generally for around 60–70% of medical laboratory errors. Multimodal testing involves analysing multiple specimen types from one patient and/or multiple analytes (e.g. gDNA, RNA, ccfDNA) from a single specimen. This is expected to improve diagnostic cancer tests by combining complimentary results and to minimize patient‘s burden when less specimen material is required.

Aim of this project within Instand-NGS4P was to develop pre-analytical NGS workflow solutions for research and future diagnostic use.

METHODS

Workflow solutions were based on market trends, stakeholder needs and other requirements from medical laboratories defined by Instand-NGS4P by Open Market Consultation. Workflows with multimodal capabilities were developed using specimens from healthy donors and tested with patient samples under local ethics committees approvals.

RESULTS

We created 23 novel or optimized workflows for research use in cancer studies and future diagnostics, which were successfully tested by Instand-NGS4P medical laboratories (e.g. UNIFI). Workflows included specimen collection, stabilisation, storage, analyte isolation, quality control (QC), library preparation and sequencing. They were compliant with ISO and CEN pre-analytical as well as NGS-related standards. As a new QC concept, metadata was collected in novel documentation sheets.

CONCLUSIONS

Developed workflows support multi- and single-mode testing. They meet pre-analytical requirements of the EU In Vitro-Diagnostic-Regulation and follow US FDA guidance to encompass the entire workflow from sample collection to assay output. Metadata collection enabled assessment of final NGS results.

Keywords: pre-analytical, specimen quality, multimodal, NGS, workflow

Sample stability

P062

ANALYSIS OF METALOTHIONEIN IN SERUM OF PATIENTS WITH MALIGNANT GALLBLADDER TUMORS: SIGNIFICANCE AND METHODOLOGY

L. Melich ¹, W. Julia ², S. Tomas ¹, K. Ivana ¹, F. Carlos ³, K. Karel ¹, K. Eva ¹, K. Eva ¹, P. Richard ¹, K. Rene ¹

¹Department of Medical Chemistry and Clinical Biochemistry, Second Faculty of Medicine, Charles University in Prague and Motol University Hospital, V Uvalu 84, 150 06 Prague 5, Czechia

²Please remember to include the presenting author Please write full first names rather than initials

³School of Pharmacy and Life Sciences, Robert Gordon University, Garthdee Road, Aberdeen AB107QB, Aberdeen, United Kingdom

BACKGROUND-AIM

Malignant diseases represent a major challenge for the healthcare system in developed countries. Late diagnosis and a high rate of metastasis formation rank gallbladder tumors among the most aggressive malignancies. Metallothionein (MT) is a low-molecular-weight intracellular protein whose primary function is heavy metal homeostasis. The aim of our work was to conduct a pilot study focused on determining the level of MT and to test the stability of the biological sample.

METHODS

Electrochemical methods are very suitable for the determination of thiols. An observed group of patients was established (2024–2025). Blood sera were obtained from these subjects and stored at -20 °C until processing. The automated electrochemical analysis was performed with a 747 VA Stand instrument connected to a 746 VA Trace Analyzer.

RESULTS

The serum was primarily denatured in ultrapure water, at temperatures (60, 70, 80, 90, 99 °C, 450 rpm). The dependence on denaturation time was also monitored for 10, 20, 30, 60 min. The findings in our study demonstrated a 10% increase in signal intensity for every 10 °C rise in temperature. The sample denaturation time reached a maximum around 20–30 min. Furthermore, the dependence of signals on repeated analysis time was studied. The cumulative signals obtained changed between 5–10% over time. Statistical differences P > 0.5. The average amount of MT found in the group was 15±5 µg/l. When distributing the MT values, we can observe three significant clusters around the values of 9 µg/l, 15 µg/l, and 22 µg/l.

CONCLUSIONS

The stability of the blood serum sample was tested during its preparation for analysis. It was found that the samples do not undergo significant changes. Key words: Electroanalytical detection; Automatic analysis; Tumour Disease Markers; Proteins; Metallothioneins. The authors gratefully acknowledge the financial support of the League Against Cancer Prague and CA COST Action CA22125, under which this work was conducted.

Sample stability

P063

ASSESSMENT OF COMPARISON RESULTS OF URINE PROTEIN STABILITY IN TWO TYPES OF NEW ACTI-FINE® VACUUM CONTAINERS WITH BORIC ACID AND PRESERVATIVE STORED AT +20°C VERSUS RESULTS FROM ACTI-FINE® VACUUM CONTAINERS WITHOUT ADDITIVES STORED AT +4°C FOR 7 DAYS

I. Shmidt ², E. Atmadzheva ², S. Kovalevskaya ¹

¹FSBEI HE I.P. Pavlov SPb SMU MOH Russia.

²St. Luke Clinical Hospital, St. Petersburg, Russia

BACKGROUND-AIM

Clinical laboratories frequently need to validate new devices IVD, such as vacuum urine containers with different preservatives, to ensure analytical stability and result reliability.

Aim: to compare the stability of urine protein results in new Acti-Fine® vacuum containers (LLC Granat Bio Tech) with boric acid and preservative (stored at +20°C) versus additive-free containers (stored at +4°C) over a 7-day period.

METHODS

The study was conducted from June to August 2025 at St. Luke Clinical Hospital. Urine samples from 40 patients were collected in sterile containers and aliquoted into three types of 8.0 ml Acti-Fine® vacuum tubes: Type A: Without additives (reference, stored at +4°C), Type B: With boric acid (stored at +20°C) , Type C: With preservative (chlorhexidine-based, stored at +20°C)

Protein concentrations were measured at 0, 24, 72, 120, and 168 hours using Beckman Coulter AU480 analyzer. Protein was determined spectrophotometrically with pyrogallol red-molybdate reagent. Statistical analysis was performed using MedCalc® V20.218.

RESULTS

Protein Stability (g/L): Type B (Boric Acid) minimal decrease, up to -3.6% at 168 hours, Type C (Preservative) slight increase, not exceeding +4.5% at all-time points.

Stability was consistent across all protein concentration ranges (0–0.99 g/L, 1.00–1.99 g/L, 2.00–2.99 g/L, ≥3.00 g/L), with no significant deviations (p > 0.05).

CONCLUSIONS

Urine protein results from Acti-Fine® tubes with boric acid or preservative stored at +20°C show strong correlation (r > 0.999) with results from additive-free tubes stored at +4°C.

Both types of test tubes provide reliable stability for urine protein for at least 7 days, with preservative tubes showing slightly better performance (+3.6% maximum change vs -3.6% for boric acid tubes). The observed changes are not clinically significant, making both container types suitable for routine clinical use.

Sample stability

P064

STABILITY OF PTH IN PARATHYROID FNAB WASHINGS UNDER VARIOUS STORAGE CONDITIONS AND FREEZE-THAW CYCLES

E. Kozłowska ¹, P. Kozłowski ², O. Ciepiela ²

¹Central Laboratory, University Clinical Centre of the Medical University of Warsaw, Warsaw, Poland

²Department of Laboratory Medicine, Medical University of Warsaw, Warsaw, Poland. Central Laboratory, University Clinical Centre of the Medical University of Warsaw, Warsaw, Poland

BACKGROUND-AIM

Measurement of parathyroid hormone (PTH) in saline washings from fine-needle aspiration biopsy (FNAB), performed during parathyroid gland operations, is useful for confirming whether a sampled tissue is parathyroid—both preoperatively and intraoperatively. This study aimed to assess the stability of PTH in FNAB saline under various storage conditions and freeze-thaw cycles, following the validation protocol described by Andreasson et al. (Front. Neurol. 2015).

METHODS

Three FNAB saline samples with low, medium, and high PTH concentrations were divided into 19 aliquots each. Aliquots were stored at -80°C (reference) and subjected to 1–7 freeze-thaw cycles, at room temperature (RT) and 4°C for 1 h, 2 h, 4 h, 24 h, 72 h, and 168 h, and at -20°C for one month. All aliquots were subsequently transferred to -80°C and analyzed using the ECLIA method (Roche Cobas Pro e801) in duplicate. Stability was defined as recovery within ±15% and CV ≤ 15%.

RESULTS

PTH remained stable after seven freeze-thaw cycles and after one month of storage at -20°C. Recovery in RT samples remained within the acceptable range up to 24 hours but declined toward the lower ±15% limit after 72 and 168 hours. Samples stored at +4°C remained stable for up to seven days. All other recovery results, including those after -20°C, were also within the predefined limits.

CONCLUSIONS

PTH in FNAB washings remains stable across a wide range of preanalytical conditions. Storage at -80°C or -20°C preserves sample integrity. After thawing, testing should be performed within 24 hours, or samples can be stored at +4°C for up to seven days. The results confirm that PTH can be reliably measured in FNAB washings after multiple freeze-thaw cycles and under various storage conditions.

Sample stability

P065

ENSURING RELIABLE PSA RESULTS: THE IMPORTANCE OF TIMELY SERUM ANALYSIS

A. Cumplido Portillo ², A. Alonso Llorente ², N.E. Larocca González ², L. Fraile García ², A. Blanquero Porras ², C. Jaén Ruiz ², M. Jiménez Domínguez ², A. Ramos Gorostiaga ², M. Font Font ¹, M. Ibarz Escuer ²

¹Hospital Comarcal del Pallars

²Hospital Universitario Arnau de Vilanova

BACKGROUND-AIM

Prostate-specific antigen (PSA) is a protein produced by the prostate involved in the liquefaction of the seminal clot. It circulates in free form (fPSA) or bound to proteins. Elevated total PSA (tPSA) and a decreased tPSA/fPSA ratio may suggest prostate cancer, though both can be altered by benign conditions. In Spain, prostate cancer is the third most common tumor and the third leading cause of cancer death in men. Serial PSA testing is therefore essential for diagnosis, prognosis, and monitoring of patients with this malignancy.

The objective of the study was to assess whether the stability of serum tPSA and fPSA in samples stored at 2–8°C exceeds 24 hours, as specified by the manufacturer.

METHODS

A prospective analysis was performed on 20 serum samples. Concentrations of tPSA and fPSA were measured upon arrival (0 hours), at 24 and 48 hours using the UniCel DxI analyzer (Beckman Coulter). Stability was assessed according to the SEQC-ML guidelines on defining stability limits in biological samples, and EFLM minimum specifications.

• tPSA: Coefficient of variation (CV) was 5.1%; maximum allowable deviation: CV × 1.65 = 8.42%

• fPSA: CV was 5.3%; maximum allowable deviation: 8.76%

• Stability was determined using mean difference (MD) comparisons.

• Reference values:

o tPSA: <4ng/mL

o tPSA/fPSA ratio: 0.25–1

RESULTS

tPSA Mean Values:

• 0 hours: 4.12ng/mL

• 24 hours: 3.96ng/mL (MD = 3.94%)

• 48 hours: 3.91ng/mL (MD = 5.35%)

fPSA Mean Values:

• 0 hours: 1.20ng/mL

• 24 hours: 1.15ng/mL (MD = 3.88%)

• 48 hours: 1.08ng/mL (MD = 10.61%)

CONCLUSIONS

Serum tPSA levels remained stable for up to 48 hours under refrigeration (2–8 °C). However, fPSA stability decreased significantly after 24 hours, exceeding the allowable variation threshold at 48 hours. As the accuracy of the tPSA/fPSA ratio depends on both measurements, to ensure accurate interpretation, it is recommended that serum samples continue to be analyzed within the first 24 hours after collection.

Key words: tPSA, fPSA, stability.

Sample stability

P066

THIRTY MINUTES OR BUST: MEASURING COMPLIANCE WITH CRITICAL PRE-ANALYTICAL TIMEFRAMES FOR AMMONIA ANALYSIS AT GROOTE SCHUUR HOSPITAL

K.T.R. Magolego ¹

¹Division of Chemical Pathology, Department of Pathology, University of Cape Town and National Health Laboratory Service, South Africa

BACKGROUND-AIM

Ammonia is a laboratory analyte that is highly susceptible to pre-analytical instability. To minimise artefactual elevation, clinical guidelines recommend transporting specimens on ice and delivering them to the laboratory within 30 minutes of collection. Despite these recommendations, real-world adherence to these stringent handling requirements has not been well characterised.

METHODS

A retrospective audit was conducted on all ammonia results generated between 30 June 2024 and 30 June 2025 from the laboratory information system. Turnaround time metrics were calculated for each sample, including intervals from collection to receipt, collection to result, and registration to result, expressed in minutes. Data were processed and analysed in Python (version 3.12) using the pandas and openpyxl libraries.

RESULTS

A total of 312 ammonia results were audited. Collection time was documented for 243 (78%) samples. The median (p25, p75) time from collection to receipt was 19 minutes (13, 40), with 160 (66%) of timed samples meeting the recommended ≤30-minute threshold. The median (p25, p75) time from collection to result was 87 minutes (63, 148), and from receipt in the laboratory to result was 59 minutes (43, 101). Elevated ammonia concentrations (>35 µmol/L) were observed in 236 (76%) samples, of which 62 (26%) were received more than 30 minutes after collection, suggesting a potential contribution of pre-analytical delay to abnormal results.

CONCLUSIONS

Although most ammonia samples reached the laboratory within the recommended 30-minute window, a substantial proportion did not. Delays were also observed among samples with elevated results, suggesting that pre-analytical factors may contribute to potentially misleading interpretations. These findings highlight the need for ongoing clinician education and reinforced adherence to best-practice sample handling protocols to ensure accurate ammonia measurement

Sample stability

P067

ASSESSMENT OF SERUM LIPOPROTEIN(A) STABILITY AFTER STORAGE AT −20 °C

L. Mino ², N. Heta-Alliu ², J. Seiti ¹, S. Graceni ³, M. Lika ⁴, E. Kapllani ⁴, A. Bulo ²

¹Cardiology Department, “Mother Teresa” University Hospital Center, Tirana

²Laboratory Department, University of Medicine, Tirana

³Laboratory Networks, “Mother Teresa” University Hospital Center, Tirana

⁴Laboratory Networks, “Mother Teresa” University Hospital Center, University of Medicine, Tirana

BACKGROUND-AIM

Lipoprotein(a) [Lp(a)] is an established cardiovascular risk marker, but limited evidence exists on its stability in stored clinical samples. Assessing the reliability of Lp(a) measurements after storage is essential for both research and clinical practice, especially in critical care settings.

METHODS

We evaluated the storage stability of Lp(a) in plasma samples from 40 patients admitted to the Cardiology Intensive Care Unit. Lp(a) concentrations were measured at baseline (Day 0) and after storage for two months at −20 °C using a turbidimetric method on the Roche Integra analyzer. Results were reported in nmol/L. The statistical analysis was conducted using Medcalc.

RESULTS

Percentage difference (PD%) between baseline and stored measurements in the full cohort (n = 40) was 2.9% (95% CI: −2.95 to 8.7) which is well below the reported biological variation of Lp(a) (10.2% according to the EFLM database). To address the potential distortion of PD% caused by very low baseline concentrations, we analyzed a subgroup of 32 patients with Lp(a) >10 nmol/L. In this subgroup, PD% was −2.1% (95% CI: −5.8 to 1.5), confirming that changes remained small and clinically insignificant.

Wilcoxon signed-rank testing in the full cohort demonstrated no statistically significant difference between baseline and stored values (p = 0.066; positive differences = 19, negative differences = 21). Regression through origin yielded a strong correlation (R² = 0.997). Two samples (5%) shifted from above to slightly below the positivity cutoff after storage.

CONCLUSIONS

Lp(a) concentrations remained stable in plasma samples stored at −20 °C for two months, with only minimal and non-significant variation. The exclusion of very low baseline values confirmed that observed differences were not clinically relevant. These findings support the use of frozen storage at −20 °C for short-term preservation of samples in clinical and research settings, though borderline results near decision cutoffs should be interpreted with caution.

Sample stability

P068

EFFECTS OF DELAYED CENTRIFUGATION ON SERUM METABOLITE LEVELS

A. Mino’ ², C. Tiberio ², M. Caruso ², A. Di Costanzo ¹, A. Angiolillo ¹

¹Department of Medicine and Health Sciences “V. Tiberio”, Centre for Research and Training in Medicine of Aging, University of Molise, 86100 Campobasso, Italy / Molise Regional Health Service, ASREM, 86100 Campobasso, Italy.

EFFECT OF MULTIPLE FREEZE–THAW CYCLES ON SELECTED BIOCHEMICAL SERUM ANALYTES IN UNALIQUOTED SERUM SAMPLES

A. Shehu ¹, E. Nuredini ²

¹3P Life Logistic Laboratory

²Planta Laboratory

BACKGROUND-AIM

Repeated freeze–thaw (F-T) cycles are a common preanalytical factor that may compromise laboratory results. The aim of this study is to evaluate the effect of repeated (F-T) cycles on common biochemical serum analytes in unaliquoted samples.

METHODS

Samples from 20 patients were collected in serum separator tubes with gel (SST-G) and centrifuged (2236 × g, 10 min). Baseline values (T0) were obtained on a Beckman Coulter DxC 700 AU analyzer by enzymatic and colorimetric methods for glucose, urea, creatinine, ALT, AST, total/direct bilirubin, total cholesterol, triglycerides, HDL, and LDL. Samples were stored at –20 °C in SST-G tubes with the caps closed, without aliquoting. After each cycle, samples were left to thaw passively at room temperature. All subsequent measurements (T1–T5) used SST-G tube, replicating routine laboratory handling. Differences between T0–T5 were evaluated using the Friedman test with Wilcoxon post-hoc comparisons, considering p< 0.05 as significant.

RESULTS

Significant decreases across T0–T5 were found for all analytes (p<0.001). The greatest decline was between T0-T1 (−13.5% to −21.8%), with progressive reduction until T5. Glucose, lipids, and bilirubins showed losses of -15%, -14–21%, and -22–24% at T5, respectively. ALT declined -28%, whereas AST only -8%. Creatinine decreased -11%, while urea showed minimal change (-5%). Visual hemolysis appeared after repeated cycles. Interestingly, thawing by hand warming or in a thermostatic device yielded values identical to baseline (no significant change, p>0.05).

CONCLUSIONS

Repeated freeze–thaw cycles compromise analyte stability in SST-G tubes without aliquoting, with the largest decline observed after the first cycle, providing relevant preanalytical evidence for optimizing laboratory workflow.

keywords: freeze–thaw cycles; unaliquoted serum; sample stability; preanalytical phase

Sample stability

P072

ASSESSMENT OF SERUM SAMPLE STABILITY USING TXRF

B. Šimac ¹, J. Jablan ², B. Bilandžija ², M. Žarak ¹

¹Clinical Department of Laboratory Diagnostics, Dubrava University Hospital, Zagreb

²Department of Analytical Chemistry, Faculty of Pharmacy and Biochemistry, Zagreb

BACKGROUND-AIM

Total reflection X-ray fluorescence (TXRF) allows the simultaneous determination of multiple elements from very small sample amounts with minimal preparation steps in combination with low detection limits and high accuracy, reduced matrix effects, and easy quantification based on the internal standard method. The aim of this study was to evaluate the stability of prepared serum samples that were not analyzed immediately.

METHODS

Following CLSI C38, blood was collected from 20 healthy volunteers in trace element-free tubes (BD Vacutainer®, Becton Dickinson, Franklin Lakes, NJ, USA). After centrifugation, the serum was collected, transferred to plastic tubes, and stored at -20 °C until use. For TXRF, 150 µl serum was weighed into a tube, 5 µl IS of Ga (100 mg/kg) was added, weighed, and homogenized. 10 µl were pipetted onto sample carriers, dried at 30 °C on a hot plate, and measured (600 s) by TXRF. The remaining 10 prepared aliquots were stored at 2–8 °C and analyzed the next day. A two-sample t-test compared Fe, Cu, Zn, Se, and Br between the groups (statistical significance was set at P < 0.05).

RESULTS

All elements were significantly lower after overnight storage at 2–8 °C compared to immediate analysis:

Fe: 0.45 ± 0.10 vs. 1.04 ± 0.28 mg/L (P < 0.001)

Cu: 0.36 ± 0.13 vs. 1.12 ± 0.11 µg/L (P < 0.001)

Zn: 0.26 ± 0.05 vs. 0.73 ± 0.14 mg/L (P < 0.001)

Se: 0.02 ± 0.004 vs. 0.06 ± 0.01 µg/L (P < 0.001)

Br: 1.01 ± 0.14 vs. 2.94 ± 0.49 mg/L (P < 0.001)

CONCLUSIONS

For TXRF trace element analysis, serum should be prepared and measured on the same day. Delayed analysis of prepared aliquots (even at 2–8 °C) resulted in significant decreases, probably due to adsorption of metals to plastic surfaces. If storage is unavoidable, the unprepared serum should be stored (frozen) and IS added immediately prior to spotting and measurement.

Sample stability

P073

STABILITY AT A GLANCE: A SHINY APP FOR PREANALYTICAL DECISIONS

A. Valle Rodríguez ¹, M. Bello Rego ¹, J.A. Fernández Nogueira ¹

¹Hospital Meixoeiro (Vigo)

BACKGROUND-AIM

Preanalytical variability is a major source of laboratory error. Access to clear, consolidated information on analyte stability across storage conditions is essential to ensure compliance with pre-analytical requirements, particularly during emergency/on-call work.The aim is to facilitate rapid and reliable retrieval of sample-stability information so that clinicians and laboratory staff can comply with pre-analytical requirements in routine practice and during on-call shifts.

METHODS

We developed a web application in R/Shiny—SampleStability App (https://laboratoryapps.shinyapps.io/SampleStability/)—that centralizes stability times for room temperature, refrigerated (2–8 °C), and frozen (≤−20 °C) conditions from a curated Excel repository. Data are cleaned and normalized automatically (e.g., “24h”, “7d”, “2–3w” parsed to days), and presented via free-text search, analyte filters, threshold shortcuts (e.g., “Refrigerated ≥14 days”, “Frozen ≥30 days”), and intuitive visual badges and proportional bars.

RESULTS

The application standardizes heterogeneous stability expressions into consistent day-based values and exposes fast filters and search to retrieve the target analyte and condition in seconds. It is lightweight, browser-based, and suitable for point-of-care consultation during on-call duties. No hypothesis testing was performed; this is a development and implementation report.

CONCLUSIONS

By consolidating and simplifying access to stability data, the SampleStability App supports adherence to pre-analytical requirements, helps prevent avoidable pre-analytical errors, and enables quick, informed decisions in both routine and emergency settings.

Keywords: sample stability; pre-analytics; laboratory medicine; quality management; Shiny app.

Sample stability

P074

IMPACT OF COTTON BALL URINE SAMPLING ON KEY BIOMARKERS IN NEONATAL AND PEDIATRIC TESTING

G. Viale ¹, A. Aita ³, G. Ceschia ⁴, A. Tessari ², A. Stefani ², D. Faggian ², M. Mion ², E. Vidal ⁴, M. Montagnana ³

¹Department of Biomedical Sciences - DSB, University of Padua, Italy. Laboratory Medicine Unit, University-Hospital of Padua, Italy.

ROOT CAUSE ANALYSIS OF PRE-ANALYTICAL DELAYS AND THEIR IMPACT ON TURNAROUND TIME IN A HOSPITAL LABORATORY

C. Bouchet Seraphin ¹, B. Konté ¹, C. Colin ¹, B. Phin ¹, k. Peoc’H ¹

¹APHP, HUPNVS, DMU BioGem, Biochimie laboartory, Bichat Hospital Paris, France

BACKGROUND-AIM

Turnaround Time (TAT) is a key performance indicator in clinical laboratory medicine, particularly in hospital settings where reporting times directly affect therapeutic decisions. At Bichat Hospital (Paris, France), TAT is heavily impacted by the pre-analytical phase: (i) prescription transmission, (ii) test registration, and (iii) sample handling (transport, centrifugation, decanting). This study aims to identify root causes of pre-analytical delays using an Ishikawa diagram (fishbone)) and propose improvement strategies.

METHODS

Since January 2024, we have monitored TAT monthly using two indicators via Navify Analytics (Roche Diagnostics):

1/Percentage of non-urgent creatinine tests reported after 4 hours (frequent test, in 80% of patient records);

2/Percentage of emergency department troponin tests reported after 1 hour (delay recommended for acute coronary syndrome diagnosis).

RESULTS

In early 2024, 20% of creatinine results exceeded 4 hours; this has since decreased to 12%. The pre-analytical phase has the greatest impact: 30% of samples arrive at the lab over 2 hours post-collection, versus only 3–5% of delays from analytical/post-analytical steps. Delays were highest during staff shortages in reception (affecting registration) or pneumatic tube failures—both major causes identified in the Ishikawa diagram. For urgent tests, over 86% of troponin results still exceed the 1-hour threshold, indicating a need for rapid improvement.

CONCLUSIONS

The pre-analytical phase plays a major role in overall TAT performance. Refining indicators—such as time from collection to lab arrival (<1 h) and from prescription to registration (<1 h)—is needed to locate bottlenecks. Stronger collaboration between clinical teams and the lab is essential. Given ongoing troponin delays, implementing point-of-care testing in the ED is under consideration. This multidisciplinary quality approach helps optimize workflows and improve lab performance.

Demand management

P081

IMPACT OF REQUEST MANAGEMENT ON PATIENT SAFETY AND VOLUME REDUCTION IN BLOOD COLLECTION IN THE PREANALYTICAL PHASE

S. Carrasco Ignés ¹, A. Beà Sebastià ¹, M. Ibarz Escuer ¹, N. Aixalà Culleré ¹, S. Pico Forniés ¹, J. Cortes Torres ¹

¹Hospital Universitari Arnau de Vilanova de Lleida

BACKGROUND-AIM

Efficient management of test requests and collection tubes prior to phlebotomy, allowing the identification of duplicate requests for the same patient on the same day, is key to minimizing the negative impact of repeated venipunctures and excessive blood volume collection.

The impact of a request and sample management strategy aimed at reducing the volume of blood collected was evaluated, with the objectives of improving patient safety, reducing waste generation and reducing economic impact associated with the cost of these supplies and the cancellation of unnecessary tests. This was achieved by optimizing the use of blood collection materials and reagents linked to tests not performed due to the cancellation of repeated requests, during the period from January to June 2025, in a clinical laboratory.

METHODS

The impact of this management strategy was analysed from January to June 2025, using data from the Laboratory Information System (LIS). The following indicators were calculated: percentage of repeated requests for the same patient on the same day, total blood volume saved, number of tests and tubes avoided due to cancellation, and the corresponding cost savings associated with non-collected tubes.

RESULTS

During the study period, 0.92% of requests (2.603 out of 281.452 total blood test requests received) were duplicate requests for the same patient on the same day. Their cancellation prevented the performance of 46.746 determinations, avoided 6.077 tubes (0.8% of the total), and saved 30,4 litres of blood, with an estimated economic saving of €729 derived from the unused tubes.

CONCLUSIONS

The implementation of the request and sample management strategy is an effective tool to reduce unnecessary blood collections, with a positive impact both on patient safety and on laboratory sustainability.

Demand management

P082

IMPACT OF AGE AND SEX ON CREATININE ASSAY METHOD (JAFFé VS. ENZYMATIC) IN EGFR ESTIMATION AND CKD STAGE CLASSIFICATION USING MDRD AND CKD-EPI EQUATIONS

Z. Chellouai ⁴, H. Abdelwahed ³, W. Benkanouni ³, K. Djoudad ², A. Hendel ⁴, R. Moussaoui ⁴, M. Nachi ¹

¹Department of Biochemistry, EHU Oran / Oran, Algeria

²Department of Nephrology, University Hospital 1st November 1954, Oran, Algeria

³Department of Pharmacy, Faculty of Medicine, University Oran 1 Ahmed Ben Bella , Oran , Algeria

⁴Department of Pharmacy, Faculty of Medicine, University Oran 1 Ahmed Ben Bella / Department of Biochemistry, EHU Oran / Oran, Algeria

BACKGROUND-AIM

Glomerular filtration rate (GFR) estimation is essential for evaluating kidney function, but its accuracy depends on serum creatinine measurement and patient-related variables. The Jaffé and enzymatic methods can yield discrepancies that impact CKD staging. This study aimed to assess the effect of age and sex on estimated GFR and CKD classification using MDRD and CKD-EPI equations when using Jaffé and enzymatic creatinine assays.

METHODS

A simulation study based on 6,532 data points generated from 71 plasma creatinine samples from hospitalized patients was conducted at the Department of Biochemistry, University Hospital 1st November 1954, Oran. Serum creatinine was measured using the Jaffé and enzymatic methods. GFR was calculated with the MDRD and CKD-EPI equations, stratified by age (18 to 65 years) and sex. Stage reclassification between methods was analyzed using Chi² tests, linear regression, and logistic regression to identify predictors of change.

RESULTS

With CKD-EPI, reclassification rose from 18.2% (<30 years) to 41.6% (60–65 years; χ²=227.9, p<0.0001). Using MDRD, it increased from 30.2% (<30 years) to 46.0% (50–59 years; χ²=92.7, p<0.0001; R²=0.98). Sex differences were marked: with CKD-EPI, 41.5% of women changed stage versus 15.9% of men (χ²=520.2, p<0.0001); with MDRD, 57.1% of women and 22.1% of men (χ²=836.4, p<0.0001). Logistic regression confirmed both variables as predictors. For CKD-EPI, male sex (OR=3.95, 95% CI 3.50–4.45) and younger age (OR/year=0.97, 95% CI 0.961–0.970) predicted reclassification. For MDRD, male sex showed stronger impact (OR=4.82, 95% CI 4.32–5.37) while age remained significant (OR/year=0.98, 95% CI 0.976–0.984).

CONCLUSIONS

Age and sex significantly influence eGFR values and CKD staging. The MDRD equation appeared more sensitive to age and sex than CKD-EPI. These findings emphasize the need for standardization of creatinine measurement and the integration of demographic variables when interpreting eGFR, in line with the recent KDIGO recommendations.

Demand management

P083

IMPLEMENTATION OF LEAN SIX SIGMA IN INTRAOPERATIVE PTH DETERMINATION

V. Ćosić ¹, M. Đorđević ², M. Ćosić Petković ¹, M. Colić ¹, A. Stojčev ¹, V. Stoiljković ¹

¹Center of Medical and Clinical biochemistry, University Clinical Center Niš

²Clinic for Endocrine Surgery, University Clinical center Niš

BACKGROUND-AIM

Intraoperative parathyroid hormone (IOPTH) determination is a rapid method performed during a surgical treatment of hyperparathyroidism. Parathyroidectomy (PTx) requires confirmation of the removal of abnormal glands. Dynamic measurement of PTH levels, which have short half-life (3-5 min.), can serve as a reliable indicator of PTx success. This IOPTH process, due to its importance in ensuring efficiency, must adhere to rigorous standardization. The Six Sigma project was developed in laboratory with the aim of bringing benefits to the pre-analytical and analytical process, focusing on enhancing efficiency, quality, and finally, client satisfaction.

METHODS

During PTx, three blood samples were collected from patients: immediately before, 5 and 10 minutes after the start of PTx. The values of PTH were determined on the Roche Cobas e411, with PTH STAT reagents (shorter analysis time 9 vs. 18 min.). The project involved recording times in different centrifugation durations and speeds with plasma or with sera specimen; the idle time of the analyzer were recalculated and we optimized the start of sample processing.

RESULTS

In this project, we achieved a 35% improvement in TAT, with the majority of gains occurring in the pre-analytical phase. The average TAT under the Lean Six Sigma strategy amounts 17.5 +/- 0.6 min. The greatest time savings were accomplished by optimizing the procedure: starting the analyzer prior to sample arrival and eliminating the preparatory phase of the analyzer, restructuring the sequence of centrifugation and secondary barcoding, selecting EDTA plasma and good coordination with the Endocrine Surgery Clinic.

CONCLUSIONS

Implementation of the Lean Six Sigma strategy in the medical laboratory leads to a significant improvement in the pre-analytic phase predominantly, which is of crucial importance for patient outcome: reduction of potential re-intervention and minimizing anesthesia time during surgery.

Key Words: Lean Six Sigma, intraoperative, parathyroid hormone

Demand management

P084

USE OF MINIMUM RE-TESTING INTERVALS IN A TERTIARY REFERRAL HOSPITAL’S ELECTRONIC REQUEST SYSTEM

T. Escartín-Díez ¹, M. Sanz-Felisi ¹, D. Carrasco-Gómez ¹, T. Doblado-González ¹, A. Medina-Pacheco ¹, M.P. Fernández-Bullón ¹, G. Padrós-Soler ¹, M. Dastis-Arias ¹, N. Llecha-Cano ¹

¹Hospital Universitari de Bellvitge, L’Hospitalet de Llobregat (Barcelona)

BACKGROUND-AIM

A high percentage of clinical laboratory tests are inappropriately requested. Proper use of diagnostic tests avoids unnecessary orders, which may lead to interpretative errors and increased healthcare costs.

Implementing strategies for appropriate test ordering can reduce unnecessary testing. Minimum re-testing intervals (MRIs) can be applied in the electronic request system (ERS) to adjust the number of laboratory requests to those truly necessary for proper diagnosis, follow-up, and patient management.

This work presents our ten-year experience implementing MRIs in our hospital’s ERS.

METHODS

Since 2015, MRIs have been progressively implemented in our hospital’s ERS. When a test is requested within the established MRI, the system notifies the physicians, and it is up to them either reject the test or confirm its repetition.

Time intervals applied to each test were defined based on scientific literature and clinical studies, in addition to collaboration with other laboratories and the requesting physicians.

Some tests have lifetime restrictions (e.g. genetic tests, blood group testing), while others have shorter intervals, as brief as one day for certain emergency markers (e.g. procalcitonin, NT-proBNP).

Impact was assessed by analysing early repeat attempts, cancellations, confirmed repeats, and associated cost savings calculated from reagent costs.

RESULTS

Initially, only 10 tests were included in the application to preliminarily assess the system’s performance. Since then, over 700 tests have been added.

Of all the tests requested before the established time concludes, approximately 80% are ultimately cancelled, resulting in annual savings exceeding €400,000.

CONCLUSIONS

Using MRIs, repetition is avoided in most cases where a retest is attempted, proving its convenience and usefulness both from the healthcare and economic perspectives.

Demand management

P085

RATIONALE FOR LABORATORY DEMAND MANAGEMENT OF TOTAL PROSTATE-SPECIFIC ANTIGEN TESTING

M. Jiménez Domínguez ¹, A. Alonso Llorente ¹, N.E. Larocca González ¹, A. Cumplido Portillo ¹, A. Blanquero Porras ¹, A. Ramos Gorostiaga ¹, C. Jaen Ruiz ¹, M. Ibarz Escuer ¹

¹Clinical Laboratory ICS Lleida, Hospital Universitario Arnau de Vilanova de Lleida, Lleida, Spain

BACKGROUND-AIM

Total Prostate-Specific Antigen (PSA) is a widely-used biomarker for early detection and monitoring of prostate cancer. However, laboratory information systems often reject PSA requests from patients registered as female, assuming an ordering error. This approach excludes valid clinical scenarios, such as transgender patients who may benefit from PSA testing.

Our laboratory performs approximately 26,000 PSA tests annually. Until October 2022, female-registered patient PSA requests were automatically rejected.

Since then, a systematic review of these rejections has been implemented to identify clinically relevant cases, such as transgender women.

METHODS

A retrospective descriptive review of rejected PSA requests was conducted from October 2022 up to May 2025 using the Modulab system (Werfen version 4.7.00).

The considered variables included: reason for rejection, patient age and gender, and whether the request was later accepted after manual review or not. Data analysis was performed using Excel (version 2016). PSA testing was performed on the DxI 800 analyzer (BeckmanCoulter).

RESULTS

During the time the review was carried out 210 rejections were identified, mainly due to ordering errors or sample issues. Among them, 52 rejections belonged to patients registered as female. In the latter, three transgender women were identified and one of them was diagnosed with advanced prostate cancer.

CONCLUSIONS

Our findings emphasize the need to adapt automated rules to diverse clinical situations. Systematically rejecting PSA tests for female-registered patients can delay critical diagnoses in transgender individuals. Manual review enabled the detection of significant cases, supporting the need for inclusive protocols, staff training, and technical adjustments to laboratory systems, such as alerts and guidelines based on functional anatomy rather than legal gender. Automation should not compromise diagnostic equity. Adapting technological tools is essential for safe, inclusive, and high-quality healthcare.

Demand management

P086

PROCESS OPTIMIZATION USING THE ‘SERVICEWARE RESOURCES’ MOBILE APP AND ON-SITE LABEL PRINTING

A. Laskewitz ¹, H. Burgers ², R. Krist ¹, L. Dries ¹, W. Pot ¹

¹Medical Laboratory, Nij Smellinghe Hospital, Drachten, The Netherlands

²Serviceware Benelux B.V., Leiden, The Netherlands

BACKGROUND-AIM

Nij Smellinghe Hospital is an regional hospital in the Netherlands with a clinical laboratory that uses advanced total-lab automation. However, the pre-analytical process for at home and regional sample collection remained inefficient using pre-numbered labels. The aim of this project was to optimize the pre-analytical workflow improving sample traceability, reducing manual processing, and enabling real-time registration.

METHODS

A collaborative project was launched between the medical laboratories of Nij Smellinghe Drachten and Antonius Sneek, coordinated by Serviceware. By implementing the Serviceware Resources mobile app and mobile label printers into the on-site phlebotomy process, the traditional request forms and pre-numbered labels were replaced with digital order retrieval.

RESULTS

The mobile app and on-site printing solution was fully adopted from day one. An unexpected surprise was that phlebotomy staff reported faster throughput at on-site phlebotomy locations. More than 90% of incoming regional samples arrived fully registered and ready for automated processing, which significantly reduced midday backlogs and incorporated smart order handling features, e.g. adjustment of test codes and remarks on patient status. Results became available earlier, with the majority completed by 15:00 hrs, enhancing clinician access to timely information. Moreover, fewer orders were missing sample tubes. Additionally, staff experienced reduced pressure during late shifts, and increased job satisfaction.

CONCLUSIONS

The integration of digital registration and on-site labelling in the pre-analytical phase leads to major improvements in workflow efficiency, sample traceability, and staff satisfaction.

Demand management

P087

STRATEGIES TO PROMOTE RATIONAL PRESCRIBING AND REDUCE MISUSE OF 1α,25 (OH)2 VITAMIN D – INSIGHTS FROM RENNES UNIVERSITY HOSPITAL

C.R. Lefèvre ¹, M. Favalelli ¹, N. Collet ¹, A. Gicquel ¹, M. Maubert ¹, I. Abeille ¹, C. Bendavid ¹

¹Biochemistry and Hormonology department, Rennes University Hospital

BACKGROUND-AIM

The number of prescriptions of vitamin D2+D3 (calcidiol) has surged over the past decade at our hospital, increasing by 70% between 2015 and 2022. A consequence is a concurrent 400% rise in 1α,25(OH)2 vitamin D (calcitriol) demand, amounting to 2,300 requests in 2023. Calcitriol is not recommended to assess vitamin D status and should be reserved for specific indications. Most of these requests result from pre-preanalytical errors, leading to economic consequences, as calcitriol test is far more expensive than calcidiol, and clinical repercussions, as patients are not tested for the appropriate biomarker. This study aimed to identify the causes of this misuse, quantify its costs, and apply corrective measures to reduce inappropriate testing.

METHODS

All 1α,25(OH)2 vitamin D prescriptions from 2022 to 2023 were reviewed and discussed with clinicians to identify sources of errors. As a result, the following corrective actions were applied from Sep to Dec 2023:

• Education of departments generating inappropriate prescriptions

• Renaming the test in the electronic prescription system: ‘1,25 dihydroxyvitamin D’ → ‘Dihydroxy Vitamin D (1,25)’

• Relocating the 1α,25(OH)2 vitamin D on manual prescription forms (General Biochemistry → specialized Biochemistry – Hormonology as of Jan 1, 2024)

RESULTS

These actions led to a rapid 75% reduction in 1,25(OH)2 vitamin D prescriptions (from ∼200 to ∼50 monthly requests). Concurrently, requests for 25OH vitamin D2+D3 increased, suggesting an improved biomarker usage. These measures saved approximately €3,000 per month for our centre from a patient-cost standpoint. Additionally, this reduction enabled systematic prescription reviews with associated advisory services and optimized testing schedules, reducing calibrant and control reagent usage.

CONCLUSIONS

Reviewing 1α,25(OH)2 vitamin D test prescriptions at Rennes University Hospital enhanced patient care by promoting rational use of vitamin D testing, reducing costs, and improving clinical utility of Laboratory Medicine.

Demand management

P088

IMPLEMENTATION OF A MINIMUM RETESTING INTERVAL FOR HBA1C IN PRIMARY CARE

S. Lillo ¹, S. Antonsen ², E. Rabing Brix Petersen ¹

¹Lillebælt Hospital- Biochemistry and Immunology Department

²Odense University Hospital and Svendborg Hospital - Biochemistry Department

BACKGROUND-AIM

The minimum retesting interval (MRI) is the minimum time interval before a test should be repeated, based on the physiologic properties of the test and general clinical considerations. We evaluated the effect of introducing an MRI warning in the electronic request system (WebReq) used by 342 Primary Care clinics in the Region of Southern Denmark (RSD). The test under evaluation was HbA1c, for which the recommended MRI in diabetes follow-up is 2–6 months.

METHODS

An MRI warning was implemented on 15 February 2025 in RSD. The intervention period comprised the subsequent six months and was compared with the corresponding six-month period in 2024. A 90-day MRI was applied: when a repeat HbA1c was requested within this interval, the requester was alerted and shown the most recent result. We calculated the proportion of retests within the MRI totally and in cases of the previous result <48 mmol/mol (diagnostic limit), relative to the total number of HbA1c requests.

RESULTS

In the background period, 394,140 HbA1c tests were requested, of which 86,039 (22%) were repeats within 90 days, while 59,491 (15%) had a previous result < 48 mmol/mol. In the intervention period, the corresponding figures were 394,342 tests requested, with 77,431 (20% repeats and 52,027 (13%) were repeats with a previous result < 48 mmol/mol. This corresponds to a significant 13% relative reduction (p<0.05) in retesting within the MRI.

CONCLUSIONS

Introducing an MRI warning that displayed the most recent HbA1c result reduced inappropriate retesting within 90 days. However, there seem to be potential for further reduction. Furthermore, the overall test volume did not decrease. Thus, the warning seems to make General Practitioners retest HbA1c less frequently, but the reduction does not reduce the total number of HbA1c tests requested.

Demand management

P089

IMPACT OF EDUCATIONAL INTERVENTIONS ON COST REDUCTION THROUGH DEMAND MANAGEMENT IN CLINICAL LABORATORIES

A. Saez-Benito ¹, N. Rico Rios ¹, M. Calero Ruiz ¹

¹Puerta del Mar Hospital, Cádiz, Spain

BACKGROUND-AIM

This study evaluates the impact of educational interventions targeting the top 15 cost-intensive laboratory tests on reducing unnecessary demand and associated costs. By aligning test utilization with evidence-based guidelines and involving stakeholders in decision-making, we aimed to achieve significant cost savings.

METHODS

The 15 costliest tests in the first semester of 2024 were identified and analyzed. Reports outlining proposed measures, were prepared for primary care and hospital services. Following stakeholder feedback, the agreed-upon measures were implemented in September 2024: Rejection based on temporal criteria: ProBNP, Procalcitonin and Troponin I. Temporal and clinical criteria; previous value: 25OH-Vitamin D, Vitamin B12, TSH, HbA1c, CRP. Certain clinical conditions: Protrombine time, ferritin.

LIMS data were analyzed for each test, comparing determinations and costs from the first and second semesters of 2024. Reductions in demand and costs were calculated and validated.

RESULTS

Educational interventions and demand management measures led to a significant reduction of 85,386 test requests in the second semester, resulting in cost savings of €235,828.3.

Key reductions included 25OH-Vitamin D (12,340 tests; €49,720), HbA1c (8,872 tests; €29,415), Prothrombin Time (6663 test; €17,253.2), Ferritin (8592 test; €15,594.5), TSH (9415 test; €13.670.6), Vitamin B12 (1438 test; €8,171.91), Procalcitonin (5,432 tests; €21,728), ProBNP (1315 test; €15,418) and Troponin I (6,890 tests; €19,550).

Tests with limited potential for demand management, such as Calprotectin and Tacrolimus, showed minimal reductions due to clinical necessity.

CONCLUSIONS

Educational interventions targeting evidence-based demand management yielded significant cost reductions, demonstrating the value of strategic stakeholder engagement and data-driven decision-making in laboratory operations. These measures not only enhanced resource allocation but also supported high-quality patient care by minimizing unnecessary testing.

Demand management

P090

IMPROVING APPROPRIATENESS OF URINE CULTURE INDICATIONS AND REDUCING UNNECESSARY TESTING THROUGH ONGOING EDUCATION AND FEEDBACK IN A HOSPITAL SETTING

O. Schwartz Harari ¹, G. Frankfurter ¹, Y. Cohen ¹, I. Zohar ¹, Y. Maor ¹, D. Ben David ¹

¹E. Wolfson medical care

BACKGROUND-AIM

Urine cultures are essential for diagnosis of urinary tract infections. However, in elderly patients, the high prevalence of asymptomatic bacteriuria often leads to overdiagnosis, over treatment, and unnecessary antibiotic exposure. More than half of urine culture requests are submitted without appropriate clinical indications, highlighting the need for interventions. This study aimed to evaluate the effectiveness of an educational intervention in improving urine culture ordering in an acute care hospital.

METHODS

A quasi-experimental pre–post intervention study was conducted in six medical wards of a 670-bed hospital. The study period included: Pre-intervention phase:2019 –2021(36 months)Post-intervention phase:2022 –2023(24 months).The intervention included education and training sessions, monitoring, feedback reports, and ongoing reinforcement.

RESULTS

A substantial reduction in urine culture utilization was observed. The median monthly number of cultures decreased from 1,260 to 578 (p < 0.001), and the culture rate declined from 72.7 to 52.7 per 1,000 patient-days (p < 0.001).Patient demographics remained consistent across study periods, with a median age of 81 years and no significant differences in sex, residence in long-term care, or comorbidities. Regarding indications for testing, fever accounted for 37.8% of requests and local urinary symptoms for 20.6%. multivariate analysis demonstrated that the appropriateness of urine culture requests improved significantly, from 43.9% in the pre-intervention period to 56.7% after the intervention (p = 0.015).

CONCLUSIONS

Implementation of a multifaceted educational program, reinforced by monitoring and feedback, led to a significant reduction in unnecessary urine culture testing and improved adherence to evidence-based indications. These findings highlight the impact of stewardship interventions on diagnostic practices. Further research is needed to optimize interventions for elderly patients and those presenting with nonspecific symptoms.

Demand management

P091

FEASIBILITY OF CONDITIONAL URINE CULTURE BASED ON PATHOLOGICAL URINALYSIS: A RETROSPECTIVE DIAGNOSTIC STEWARDSHIP STUDY

M. Shapira ², M. Pitashny ², Y. Tal ¹, R. Cohen ³

¹Clinical Laboratory Division, Hillel Yaffe Medical Center, Hadera, Israel

²Clinical Laboratory Division, Hillel Yaffe Medical Center, Hadera, Israel; Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel

³Infection Control and Infectious Diseases Units, Hillel Yaffe Medical Center, Hadera, Israel; Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel

BACKGROUND-AIM

Overuse of urine cultures can lead to overtreatment of asymptomatic bacteriuria (ASB), increased costs, antimicrobial overuse and microbial resistance. Diagnostic stewardship interventions may reduce unnecessary cultures using conditional testing based on pathologic urinalysis (UA) results

METHODS

We retrospectively analyzed all adult urine cultures and associated UA tests using dipstick, either in Core Laboratory (CL) or point of care (POC), obtained in 2024 at a tertiary hospital. Pathological UA was determined as positive leukocyte esterase (LE+) or positive nitrite. After excluding gynecology and pediatric patients we evaluated the diagnostic performance for the two tests, and the feasibility in terms of saving and safety for conditional urine cultures. We also assessed the clinical significance of selected cases for the adequacy of urine culturing.

RESULTS

Among 9,036 urine cultures, 4,273 (47.2%) had UA tests. Of 2,445 positive cultures, 210 (8.5%) were contaminants (14 commensals and 186 yeasts). About 25% of the cultures had one or more uropathogen, regardless of whether UA was performed or not. With strict rules for ‘true positive’ culture (excluding asymptomatic bacteriuria (ASB), polymicrobial cultures and suspected contaminations), pathologic UA showed 98.3% sensitivity, 50.6% specificity, PPV 33.9%, and NPV 99.2%. Out of 188 positive cultures examined, 121 (64.4%) were obtained without urinary symptoms, representing ASB which was strongly associated with negative UA test (p < 0.0001; OR = 6.0 (95% CI 2.93–12.9)). Based on the proportion of negative UA among cultures performed, conditional cultures could save 3,519 (38.9%) urine cultures and >120 ASB identifications, without missing clinically significant urine culture.

CONCLUSIONS

A UA-based conditional urine culture strategy may reduce laboratory testing and cost without compromising sensitivity. These findings support diagnostic stewardship efforts in culture triage and warrant prospective validation.

Demand management

P092

IMPACT OF VORTEX MIXING ON PLATELET COUNT IN PEDIATRIC CAPILLARY BLOOD

D. Šišmintsev ¹, I. Golovljova ¹

¹Tallinn Children’s Hospital, Tallinn

BACKGROUND-AIM

In pediatric care, capillary blood (cB) is a common sample type for complete blood count (CBC) analysis, despite a higher risk of platelet (PLT) clumping referred to as pseudothrombocytopenia. This condition requires sample recollection and often causes delays in CBC reporting, ultimately leading to patient discomfort.

The aim of this study was to evaluate the effect of vortex mixing on key CBC parameters and assess its potential to reduce unnecessary recollection.

METHODS

A total of 577 pediatric cB-CBC samples flagged for platelet clumping by the Sysmex XN-2000 analyzer were vortex-mixed at maximum speed for 2 minutes, followed by repeat testing. Samples were divided into three groups by initial clumping flag value: group 1, 100–199 (n=148); group 2, 200–299 (n=79); and group 3, 300 (n=350).

RESULTS

Vortex mixing had a clinically negligible (<1,5%) effect on WBC, RBC, RDW-SD, MCV, MCHC, MPV, and PDW. Only the average paired differences for WBC and PDW were not statistically significant (p>0,05).

Overall, vortex mixing increased the average PLT count by 48×10^9/L (95%CI, 42–54×10^9/L). The increase was greatest in group 3 (54 × 10^9/L; 95%CI, 46–61×10^9/L), followed by group 2 (45×10^9/L; 95%CI, 37–54×10^9/L) and group 1 (37×10^9/L; 95%CI, 27–47 × 10^9/L). However, the difference was statistically significant only between groups 1 and 3 (p<0,01).

After vortex mixing cases of clinically important thrombocytopenia (PLT<150×10^9/L) decreased by 38% across all groups, from 145 to 90. Resolution rates were 40% in group 3 (39/98), 34% in group 1 (10/29), and 33% in group 2 (6/18).

None of the 17 critical thrombocytopenia cases (PLT<50×10^9/L) improved after vortex mixing.

CONCLUSIONS

Laboratories performing CBC on capillary blood may benefit from incorporating vortex mixing into the procedure. This approach does not compromise analytical quality and reduces the incidence of thrombocytopenia by 38% in samples with PLT<150×10^9/L flagged for platelet clumping.

Capillary blood; Platelet clumping; Vortex

Demand management

P093

DIFFERENT STRATEGIES TO THE SAME DEMAND MANAGEMENT CRITERIA DEPENDING ON THE HEALTHCARE LEVEL

B. Sufrate-Vergara ¹, D. Prieto ¹, I. Moreno-Parro ¹, R. Mora ¹, P. Fernández-Calle ¹, J. Díaz-Garzón ¹

¹La Paz University Hospital, Madrid

BACKGROUND-AIM

Patient characteristics and available resources vary from primary care (PC) to hospital care (HC).Therefore, for a certain test, demand management strategies may differ depending on the healthcare settings and their available tools.

METHODS

Based on potential patient outcomes, we acted on HbA1c’s demand management applying a minimum retesting interval (MRI) of 90 days. When MRI was not met, two pathways were designed:

HC: when requesting the test, an alert prompts in the HIS warning the physician that it will not be performed unless they provide a clinical justification. A laboratory specialist will assess its appropriateness. We collected the number of requested tests and warnings triggered and justified in one month.

PC: training was given to physicians before implementation. The LIS displays a comment in the laboratory report advising that the test will not be performed, allowing physicians 2 days to justify its need. We collected the number of comments displayed in one month and we compared the total of tests ordered for the same period of time: pre and post applying the strategy.

RESULTS

During May 2025, in HC, 4580 HbA1c tests were demanded. Among these, 608 (13%) did not meet the MRI: 426 (9%) requests were withdrew by clinicians and 182 (4%) justified.

PC: 6827 tests were ordered, of which 338 (5%) did not meet the MRI. The number of requested tests decreased by 28% compared to February of the same year.

CONCLUSIONS

Both strategies contributed to the rationalisation of resources reducing unnecessary tests. In PC a direct effect probably based on the awareness raised among physicians was seen. In HC, results were due to the deterrent effect of the strategy. Our experience shows that, even with the same demand management strategy, different approaches should be designed to fit the profile of the professionals and the IT resources involved in different healthcare levels.

Demand management

P094

ELABORATION OF A SAMPLING MANUAL FOR BIOLOGICAL ANALYSES

I. Louati ², H. Ayari ¹, K. Guesmi ¹, S. Hammami ³, M. Dabboussi ¹, W. Grouze ³, I. Younes ¹, R. Mahjoub ¹, E. Talbi ³

¹Clinical Biology Laboratory, Zouhair Kallel Institute of Nutrition and Food Technology, Tunis, Tunisia

²Sampling unit ; Clinical Biology Laboratory, Zouhair Kallel Institute of Nutrition and Food Technology, Tunis, Tunisia

³UR17SP01- Clinical Biology Laboratory, Zouhair Kallel Institute of Nutrition and Food Technology, Tunis, Tunisia

BACKGROUND-AIM

The pre-analytical phase is a critical stage in the medical biology process, responsible for most of analytical errors. These errors can affect the reliability of results and compromise patient care. In this context, the development of a sampling manual, compliant with the requirements of ISO 15189:2022 and the Guide to Good Laboratory Practice (GBPL), is essential to standardize practices and guarantee the quality of analyses. The aim of this work was to develop a sampling manual adapted to the Clinical Biology Laboratory of the “Zouhair Kallel” National Institute of Nutrition and Food Technology (INNTA), in order to reduce errors in the pre-analytical phase while meeting normative and regulatory requirements.

METHODS

This was a descriptive, observational study conducted over a 12-month period at INNTA’s Clinical Biology Laboratory. The methodology was based on the requirements of the GBPL and ISO 15189:2022, as well as on examples of sampling manuals distributed by other laboratories. The manual was structured into several chapters covering the key stages of the pre-analytical phase: organization, sampling prescription, transport, management of non-conformities, hygiene and traceability.

RESULTS

The manual includes detailed protocols for each type of sample, standardized criteria for accepting or rejecting samples, and clear instructions for incident management. It details all the biological parameters performed in the laboratory, as well as the preanalytical conditions required for their proper execution. Procedures for managing preanalytical non-conformities and complaints will be developed in future versions.

CONCLUSIONS

The development of this manual represents a major step forward in improving the quality and safety of sampling within the laboratory. It enables us to meet regulatory requirements, while enhancing the efficiency and traceability of our pre-analytical processes. This document is an essential tool for continuous improvement and preparation for accreditation to ISO 15189:2022.

Demand management

P095

IS IT VALID TO ADJUST VITAMIN K TESTING BASED ON INR VALUES?

G. Velasco De Cos ¹, I. Aníbarro Miralles ¹, S. Herrero Castellano ¹, A. García Varela ¹, S. Rubio Lanchas ¹, P. Gonzalez Maroto ¹, C. Pérez Gavilán ², M.D. Calvo Nieves ¹

¹Department of Clinical Analysis, University Clinical Hospital of Valladolid

JELLY-LIKE SERUM IN DISSEMINATED INTRAVASCULAR COAGULATION (DIC): A PREANALYTICAL LABORATORY CHALLENGE

S.N. Ab Rahim ⁴, S.H.H. Fauzi ¹, N. Nordin ⁴, M.S. Zahari ², S.S. Zainal Abidin ³, A.Z. Kamarudin ³

¹Anesthesiology and Critical Care Unit, Faculty of Medicine and Defence Health, National Defence University of Malaysia, Kem Sungai Besi, 57000 Kuala Lumpur, Malaysia

²BP Healthcare, 40150 Shah Alam, Selangor Darul Ehsan

³Pathology Department, Hospital Angkatan Tentera Tuanku Mizan, Seksyen 2 Wangsa Maju, 53300 Kuala Lumpur, Malaysia

⁴Pathology Unit, Faculty of Medicine and Defence Health, National Defence University of Malaysia, Kem Sungai Besi, 57000 Kuala Lumpur, Malaysia

BACKGROUND-AIM

Preanalytical factor is a crucial laboratory process. Particularly in patients with coagulopathy like disseminated intravascular coagulation (DIC), the altered serum coagulation composition could hinder sample analysis. This case report describes a patient’s serum exhibiting a thick, jelly-like consistency following blood centrifugation, presenting a significant challenge to standard laboratory procedures.

METHODS

A 63-year-old male with DIC and multiorgan failure due to severe gastrointestinal bleeding underwent serial laboratory blood testing. However, his serum samples in gel separator tubes persistently showed a thick, gelatinous consistency after centrifugation, rendering them unsuitable for analysis. To resolve this, plasma samples were used instead. Blood sent in lithium heparin tubes revealed clear and analyzable plasma. Following this, biochemical markers, including renal function tests like serum creatinine, urea, and electrolytes, were able to be measured using automated analyzers.

RESULTS

The appearance of serum samples with jelly-like consistency in DIC delays timely analysis and patient management. This case demonstrates that DIC interrupts the complete coagulation process in blood, causing an ongoing clot formation even after serum separation. This leaves fibrin network to appear jelly-like in the serum. Meanwhile, plasma samples remain unaffected by coagulation due to heparin, which prevents fibrin clot formation.

CONCLUSIONS

This case emphasizes a preanalytical challenge where DIC causes serum to turn jelly-like, preventing analysis. Using plasma obtained from blood in a lithium heparin tube resolves this, thus ensuring availability of test results that aid patient management. Thus, awareness on coagulation-related preanalytical errors helps yield reliable results, and facilitates timely treatment in critically ill patients with coagulopathies.

Preanalytical cases

P099

SPURIOUS COMPLETE BLOOD COUNT ANALYSIS DUE TO BLOOD COLLECTION TUBES STORAGE AT EXTREME HIGH TEMPERATURE

A. Agorasti ¹

¹Haematology Laboratory, General Hospital of Xanthi, Xanthi, GREECE

BACKGROUND-AIM

International standards underline the importance of the conditions of blood sample transportation; temperature and time of sample storage from collection to analysis. The aim of this study is to present the analytical error of two samples due to preanalytical sample exposure at high temperature.

METHODS

The laboratory reception unit receives blood samples of outpatients, collected and transported by qualified phlebotomists. The name of phlebotomist, the time of blood drawn and the transportation conditions (temperature) are recorded. We present the results of complete blood count analysis (Sysmex, XN-1000™) of samples from two outpatients collected by two different phlebotomists in BD Vacutainer® EDTA Tubes (plastic K2EDTA) within one-week interval.

RESULTS

The results of complete blood count flagged with an (*) are presented (x10^9/L):

Patient A / 1st sample: WBC=13.02, NEUT=3.07, LYMPH=9.16, MONO=0.62, EO=0.17, PLT=1726.00

Patient B / 1st sample: WBC=14.02, NEUT=2.61, LYMPH=10.5, MONO=0.83, EO=0.08, PLT=2988.00

The blood smear examination (Cellavision® DC-1) revealed red blood cells with extremely abnormal membrane, with spikes producing free bubbles and filaments, grey shadows (platelet size) as well as Neutrophils with destroyed nucleus. Therefore, the results were no reportable. The two phlebotomists ensured us that the only parameter that was out of the protocol was the storage conditions of blood collection tubes (in the car during the summer heat with external temperature 40 – 42°C). Hence, we asked for a second blood collection. The corresponding results are presented (x10^9/L):

Patient A / 2nd sample: WBC=4.86, NEUT=2.73, LYMPH= 1.59, MONO=0.35, EO=0.16, PLT=292.00

Patient B / 2nd sample: WBC=5.38, NEUT=2.96, LYMPH=1.84, MONO=0.50, EO=0.06, PLT=239.00

CONCLUSIONS

To ensure the integrity of samples all preanalytical procedures should be recorded upon the sample’s reception. The storage of blood collection tubes at extreme high temperature may happen often during summer.

Preanalytical cases

This case highlights the value of delta checks, fractional excretion of amylase, and PEG precipitation in the investigation of persistent hyperamylasaemia. Delta checks, by demonstrating persistently high but stable amylase, help avoid repeated imaging and unnecessary hospitalisation for suspected recurrent pancreatitis. Recognition of this benign condition prevents misdiagnosis as pancreatitis and underscores the pivotal role of the laboratory when biochemical and clinical findings are incongruent.

Preanalytical cases

P103

A CASE OF REPEATED FALSELY LOW BLOOD GLUCOSE LEVELS

A. Biasotto ¹, F. D’Aurizio ¹, F. Curcio ¹

¹Institute of Clinical Pathology, Department of Laboratory Medicine, University Hospital of Udine - ASUFC, Udine, Italy

BACKGROUND-AIM

Blood glucose measurement plays a central role for the early detection of disorders in glucose metabolism. There are several pre-analytical sources of variation in glucose testing; its concentrations in whole blood samples stored at room temperature (RT) are affected by a decrease of ∼5-7% per hour.

METHODS

Fingerstick glucose was measured by the visiting nurse using a FreeStyle Optium neo H glucose meter (Abbott BV, Hoofddorp, Netherlands). In the Laboratory, glucose concentrations were determined on Cobas c702 analytical platform (Roche Diagnostics International AG, Rotkreuz, Switzerland).

RESULTS

An 88-year-old man diagnosed with Type 2 DM and Chronic myelomonocytic leukemia underwent a routine serum glucose testing, with at-home blood collection. A panic value of 39 mg/dl (reference range 70-104 mg/dl) was communicated via Read-Back to the general practitioner (GP). In the following weeks, repeated samplings resulted in concentrations between 29 and 41 mg/dl, in contrast with the referred domicile readings, consistently 70-80 mg/dl. We investigated with the GP possible causes of falsely low glucose levels; total WBC was 20-25x10³/μl and the visiting nurse collected the blood at 7am, delivering it to the laboratory 5-6 hours later.

To avoid severe ex-vivo glycolysis, glucose concentration was then determined using a VACUETTE® FC Mix tube (Greiner Bio-One GMBH, Kremsmünster, Austria), containing buffered Na₂EDTA, NaF, citric acid and sodium citrate, resulting in 77 mg/dl, in accordance with fingerstick levels.

CONCLUSIONS

We reported a case of repeated falsely low blood glucose levels due to an improper management of different steps in the pre-analytical phase, recognized by using an alternative tube as first investigative step.

Errors in pre-analytical phase are a critical topic in laboratory medicine; the collaboration between laboratory professionals and clinicians is of crucial importance to ensure patient correct management and avoid over-testing.

Preanalytical cases

P104

EEVALUATION OF A NEW URINE DEVICE FOR CHEMICAL ANALYSIS WITH STRIPS AND AUTOMATIC READER

M. Beffa ¹, V. Imperadori ¹, F. Bonometti ¹

¹Copan Italia S.p.A.

BACKGROUND-AIM

Urinalysis is the first test to determine the destiny of a urine sample and decide the attention that a patient must receive. Nowadays a processing workflow that can guarantee a shorter time to result is required. A solution can be a device suitable for a multiple set of test; reducing collection time, consumables cost, and increasing laboratory efficiency.

Urisponge is a device that integrates an innovative formulation for stabilizing the microbial load inside urine samples up to 48 hours; this device can collect a sample of urine that can undergo urine analysis and, subsequently a urine culture depending on results. In this case a single device is required, and once is received in the laboratory can be easily managed inside different areas with easier stocking temperature and time manage than normal urine cup, which not integrates preservatives.

METHODS

In this evaluation 30 urine samples were collected from mixed (sex and ages) donors. Some of the sample were modified to simulate pathogenic urine, like bacteriuria, glycosuria etc.

Urine sample were then splitted into three conditions:

1) keep as it was,

2) processed in the device without preservatives,

3) processed in the device with the preservatives.

A urine strip was immersed in all three Urine sample and then analyzed with the automatic strip reader. Sample collected with Urisponge with preservatives (condition n3) where then analyzed again after 48h of stock at room temperature.

RESULTS

All urine samples provided an acceptable result (no invalid test). All significant parameters, included the one modified, were all detected at all conditions in all samples. Small differences were detected in pH value, with no consistent pattern, indicating an intrinsic variation in the test.

CONCLUSIONS

This experiment shown promising results, indicating that Urisponge can be used in a workflow in which urine analysis and urine culture can be conducted on the same urine sample collected with the same divice, opening to an improved urine workflow.

Preanalytical cases

P105

ANALYTICAL ASSURANCE: INVESTIGATING STABILITY OF ROUTINE BIOCHEMISTRY TESTS ON THE SIEMENS ATELLICA® PLATFORM

S. Chodha ¹, V. Lee ¹, N. Mohamed ¹, L. Burke ¹, L. Hikin ¹

¹University Hospitals of Leicester NHS Trust, Leicester Royal Infirmary, UK

BACKGROUND-AIM

The requesting of ‘add on’ tests by clinicians can be challenging for laboratories to manage; particularly if the stability of assays has not been established. Analysis or rejection of samples without stability data can potentially lead to inaccurate results or patients being re-bled unnecessarily.

Our aim was to determine the stability timeframes for routine analytes in serum and urine matrices. Serum and urine samples (post-centrifugation) were stored at room temperature or refrigerated for a variety of time points.

METHODS

Patient serum and urine samples were analysed for their requested tests within an hour of collection; classed as baseline results (0hr). Six of these samples were stored at room temperature (23-25°C) and six were refrigerated (3-5°C). Samples were then re-analysed at time intervals of 4, 8, 12, 24, 48, 72, 96, 120, 144 and 168 hours (7 days), on Siemens Atellica® analysers. Differences in results that breached the Westgard desirable specification for imprecision were classified as clinically significant (CVA = <0.50 x CVI).

RESULTS

A total of 62 analytes were assessed for stability; the majority (n=41, 66.1%) of serum assays were stable at both temperatures up to 7 days, including potassium, sodium, creatinine, phosphate, triglycerides and glucose.

All urine analytes (n=11) were stable for up to 7 days at both temperatures except for urine protein, which was stable for up to 3 days at room temperature and up to 5 days refrigerated (Desirable CVA = 17.8).

CONCLUSIONS

This study has shown that most serum and urine analytes are stable, at both room temperature and refrigerated for up to 7 days. This data has enabled the automation laboratory at the University Hospitals of Leicester NHS Trust to assess the suitability of add-on test requests, ensuring we provide accurate and precise results, prevent patient re-bleeding and improve care pathways.

Keywords: stability, imprecision, analyte, storage, Atellica

Preanalytical cases

P106

IMPLEMENTATION OF THE PRE-ANALYTICAL ISO STANDARD LEADS TO A MORE FAVORABLE OUTCOME IN PATIENTS WITH SEPSIS

M. Ćosić Petković ², V. Stoiljković ¹, M. Colić ¹, A. Stojčev ¹, V. Ćosić ¹

¹Center for Medical and Clinical Biochemistry; University Clinical Center Niš, Serbia

²Clinic for Infectious Diseases

BACKGROUND-AIM

Pre-analytic laboratory process may have a crucial impact on patient outcome. ISO 15189:2022, the standard for medical laboratories, allows certain exceptions for sample acceptance in the best interests of the patient, particularly in emergency settings such as sepsis, a life-threatening condition characterized by organ dysfunction.

METHODS

We describe a diagnostic pathway for patients in emergency room with suspected severe sepsis. In these cases, we perform arterial blood gas analysis with lactate measurement, complete blood count, rapid blood culture identification and laboratory analyses of biochemical parameters including inflammatory markers (CRP, PCT, IL-6).

RESULTS

A 58-year-old woman was admitted to the Emergency Center of the University Clinical Center Niš with clinical symptoms suggestive of sepsis. Blood samples were taken according to the standard procedure for diagnosing sepsis. During laboratory triage, the technicians noted that the serum samples were inadequate due to marked hemolysis. An additional blood sample collected in an EDTA K₂ tube was also received. After consultation with clinical doctors in emergency room, and considering the best interests of the patient, we decided to determine inflammatory markers from EDTA plasma, since the patient was undergoing imaging examinations at the time. Our laboratory procedures allow some analyzes to be performed on EDTA plasma instead of serum, based on literature data and our own research which showed that there was no significant difference between serum and plasma sample types (p > 0.05) for CRP, PCT, IL-6.

CONCLUSIONS

In order to maintain patient care, laboratory staff, together with clinicians, play an important role in processes that prioritize the best interests of the patient in receiving care when a blood sample has been compromised, according to ISO 15189:2022 standard. Timely laboratory diagnostics contribute to a faster diagnostic decision-making and improved patient outcomes.

Preanalytical cases

P107

INVESTIGATION OF ABNORMAL SEPARATOR GEL FLOTATION AND UNEXPLAINED PERTURBATION OF BIOCHEMICAL PROFILE IN TUBES OF POST-DIALYSIS SAMPLES: MYSTERY SOLVED

V. Bois ², A. Croisonnier ², C. Chirica ², G. Vernin ¹, D. Guergour ²

¹AGDUC (Association Grenobloise Dilaysés et Urémiques Chroniques), Grenoble, France

²Unity of Biochemistry Immunoanalysis, Departement of Institut Biology and pathology , University of Grenoble Alpes, Grenoble, France

BACKGROUND-AIM

We observed 3 cases of separator gel flotation above the plasma layer in end-of-dialysis samples from patients undergoing hemodiafiltration.

Plasma analysis revealed profound hyperproteinemia (>140 g/L), causing gel flotation, hypercalcemia and a decrease in all other biochemical parameters inconsistent with the patients’ clinical status and baseline dialysis assessments, suggesting a probable pre-analytical error.

Although this issue has been reported by several authors, no conclusive findings have been proposed so far.

Ultrafiltration induced by hemodiafiltration causes extensive losses of ions and water, which must be compensated by reinfusion of substitution fluid into the post-dialysis circuit.

We hypothesize that sampling from the venous site of the dialysis circuit, located upstream of the substitution fluid reinfusion site, may account for the observed biochemical discrepancies.

METHODS

We performed end-of-dialysis blood sampling in 4 patients undergoing hemodiafiltration at two simultaneous sites : the arterial site, conventionally located upstream of the dialyzer, and the venous site, located downstream of the dialyzer but upstream of the substitution fluid reinfusion point.

RESULTS

Following centrifugation, samples collected from the venous site reproduced the same abnormalities observed in our 3 patients, including abnormal separator gel flotation. In contrast, samples obtained from the arterial site demonstrated results consistent with a normal end-of-dialysis biochemical profile, without gel flotation anomalies.

CONCLUSIONS

The gel flotation and analytical anomalies can be attributed to the sampling site. Venous site should never be used for blood collection. The observed hemoconcentration solely reflects the ultrafiltration induced by hemodiafiltration and does not indicate dialysis efficacy.

It is imperative to raise awareness among healthcare staff regarding the importance of the correct sampling site to ensure reliable laboratory results for biological monitoring of these patients.

Preanalytical cases

EDTA samples showed the lowest LOD but were linear only up to 1.5 μg/mL, whereas serum and heparinized plasma were linear up to 5.5 μg/mL. Histamine remained stable in all samples for up to 24 h at room temperature, with a subsequent decline of 10–15%. Unexpectedly, stability was higher at room temperature than at °C, where EDTA samples showed >15% reduction from the start. Freezing at -20 °C led to gradual depletion in all specimens from day 7 onward.

CONCLUSIONS

The following preanalytical factors have been identified as having a potential impact on the outcome of histamine testing: EDTA is not an optimal matrix for the determination of histamine levels. Furthermore, it is posited that samples can be stored at ambient temperature for a period of up to one day. The ensuing essay is intended to furnish the reader with a comprehensive overview of the extant literature on the subject.

Preanalytical cases

P111

COMPARISON OF SERUM AND PLASMA TUBES TREATED WITH SODIUM FLUORIDE FOR MEASURING BLOOD GLUCOSE IN PREGNANT WOMEN

A. García Varela ¹, M.I. Vidriales Vicente ¹, I. Anibarro Miralles ¹, G. Velasco De Cos ¹, L. Alonso Sendino ², J. Calvo Cruz ², A. Gómez Esteban ², M.D. Calvo Nieves ¹

¹Servicio de análisis clínicos, Hospital Clínico Universitario de Valladolid, España

²Servicio de extracciones, Hospital Clínico Universitario de Valladolid, España

BACKGROUND-AIM

During pregnancy, blood glucose levels can vary, causing gestational diabetes, which is detected using the O’Sullivan test. Glucose is unstable after extraction, and factors such as the type of tube and the time elapsed before centrifugation can reduce it by 7%/hour due to glycolysis. To avoid this, tubes with a separating gel that isolates the cells or tubes with sodium fluoride, which inhibits enolase, are used.

The study evaluates the interchangeability of blood glucose values obtained from serum samples with separating gel and plasma with sodium fluoride, taken 60´after the O’Sullivan test in the same patient.

METHODS

356 duplicate samples were analysed using an enzymatic method with hexokinase in Roche autoanalysers, employing serum tubes with separating gel and plasma tubes with sodium fluoride. The samples were centrifuged within the first hour after extraction.

Statistical analysis was performed using the Medcal programme. The following statistical tests were used: Passing-Bablok regression, Bland-Altman plot, and McNemar’s test.

RESULTS

The Passing-Bablok regression showed an intercept of 2.33 (0.16–4.00, 95% CI) and a slope of 1.01 (1.00–1.02, 95% CI), indicating a systematic error. The mean difference in the Bland-Altman plot was 3.66 (3.03–4.29, 95% CI). The McNemar test showed a difference of 1.97% (-0.43–3.56, 95% CI) with p=0.1185.

CONCLUSIONS

In our laboratory, according to Passing-Bablok analysis, there is a systematic error of 2 mg/dL between sodium fluoride tubes and serum tubes with separating gel. This error is confirmed by the mean difference of 3.66 in the Bland-Altman plot. However, this difference does not statistically significantly affect diagnosis, according to the McNemar test. The fluoride tube better inhibits glycolysis, but the serum results are clinically acceptable.

Key words: Glucose, gestational diabetes, serum, sodium fluoride tube

Preanalytical cases

P112

PREANALYTICAL ERROR DUE TO IMPROPER TEMPERATURE HANDLING IN PLASMA AMMONIA DETERMINATION

M.M. González Prado ¹, E. Pérez Pegado ¹

¹Hospital de León

BACKGROUND-AIM

Highlight the importance of temperature in the transportation and handling of the sample for the determination of plasma ammonium.

METHODS

Patient’s clinical history and laboratory reports.

RESULTS

A 56-year-old male patient with a history of chronic liver disease due to alcoholic cirrhosis was admitted to the emergency department with confusion, progressive somnolence, and disorientation. A hepatic profile, venous blood gases, and urgent plasma ammonia determination were ordered due to suspected hepatic encephalopathy.

The venous blood sample was collected in an EDTA tube, but the preanalytical protocol was not followed: the tube was not immediately placed on ice, was not centrifuged under refrigerated conditions, and was transported along with routine samples, with processing delayed by approximately 45 minutes.

The laboratory reported a normal plasma ammonia level (24 µmol/L), which initially led to hepatic encephalopathy being ruled out as the cause of the neurological symptoms. However, due to clinical deterioration, the test was repeated following strict preanalytical handling: immediate placement on ice, refrigerated centrifugation within 10 minutes of collection, and prompt analysis. The result showed a markedly elevated ammonia level of 134 µmol/L.

This confirmed the diagnosis of hepatic encephalopathy, and treatment with lactulose was initiated, resulting in clinical improvement within 48 hours. The incident prompted a full review of the preanalytical protocols for critical tests, particularly emphasizing proper temperature management and timely processing.

CONCLUSIONS

Plasma ammonium determination is highly sensitive to preanalytical variations, especially regarding temperature control and time to analysis. Improper handling can result in falsely normal values and lead to significant diagnostic delays or errors. This case highlights the critical importance of strict preanalytical protocols and ongoing staff training for the accurate interpretation of unstable analytes.

Preanalytical cases

PEG precipitation on the Roche Elecsys platform effectively distinguished a macro-TSH case from controls. While control groups showed recovery ratios between 0.70 and 0.84, the macro-TSH case yielded a markedly suppressed ratio. These findings highlight the diagnostic utility of PEG recovery testing in identifying macro-TSH and preventing misinterpretation of elevated TSH in clinical practice.

Preanalytical cases

This case emphasises the continued value of peripheral blood smear examination in complementing automated haematology analysers. The pseudo-thrombocytopenia arising from giant platelets is distinct from EDTA-induced platelet clumping. Laboratory vigilance and a close clinician–laboratory communication are essential to ensure accurate diagnosis and appropriate patient management.

Preanalytical cases

Keywords: saline, pre-analytical errors, blood sampling

Preanalytical cases

CONCLUSIONS

Extreme lipemia, particularly drug-induced, represents a significant preanalytical barrier to accurate laboratory diagnostics. In patients treated with asparaginase, proactive monitoring of lipid levels is essential. Laboratories should implement protocols for managing lipemic samples, and clinicians must be aware of how treatment-related metabolic effects can compromise critical diagnostic data. Communication between clinical and laboratory teams is vital to ensure appropriate interpretation and timely decision-making.

Preanalytical cases

P129

TESTING OF AN IVDR-COMPLIANT PREANALYTICAL WORKFLOW FOR MULTIMODAL NGS-BASED CANCER DIAGNOSTICS IN THE INSTAND-NGS4P EU-PROJECT

P. Pinzani ¹, C. Marchi ¹, S. Gelmini ¹, I. Mancini ¹, L. Simi ¹, A. Bettiol ¹, M. Pazzagli ¹

¹SOD Laboratorio Biochimica Clinica e Molecolare, Azienda Ospedaliera Careggi e Dipartimento di Scienze Biomediche Sperimentali e Cliniche, Università degli Studi di Firenze

BACKGROUND-AIM

INSTAND-NGS4P is a European PCP project aiming to improve cancer diagnostics through standardized NGS workflows. UNIFI and other public institutions defined clinical and technical requirements to guide the development of IVDR-compliant solutions. The workflow covers four Lots (pre-analytics, sequencing, bioinformatics, e-reporting) and four Phases. In Phase 3, prototypes were tested, resulting in two integrated NGS workflows for routine diagnostics of common and rare cancers in adults and children.

METHODS

Ten lung cancer patients from Careggi University Hospital were included. cfDNA from plasma and gDNA from whole blood were collected using PAXgene Blood ccfDNA Tubes. All procedures followed QIAGEN SOPs. DNA extraction was performed using QIAsymphony and QIAGEN kits; quality and quantity were assessed by BioAnalyzer, Qubit, and Nanodrop. Libraries were prepared with QIAseq kits: cfDNA (Human Lung Cancer Panel), gDNA (custom PGX panel). Quality control included BioAnalyzer, qPCR, and Qubit. Metadata were uploaded to QIAGEN’s database for integration with later phases. Comparative analysis was performed with UNIFI’s internal workflow.

RESULTS

The protocol enabled ISO-compliant dual extraction of cfDNA and gDNA from a single sample. Quality control confirmed nucleic acid integrity and quantification. Library preparation included UMIs; total turnaround was ∼2 days, with 4–5 hours hands-on. One cfDNA library failed QC (likely due to low input), while all PGX libraries passed. qPCR was essential for accurate quantification. NGS results matched those from UNIFI’s routine workflow, confirming feasibility.

CONCLUSIONS

QIAGEN’s preanalytical workflow demonstrated robustness, IVDR compliance, and clinical applicability, supporting streamlined and standardized diagnostics for personalized cancer care.

Preanalytical cases

P130

UNRAVELING THE PUZZLE OF HYPERKALEMIA NOT RESPONDING TO TREATMENT

K.D. Rammuthupura ¹, M. Balasuriya ², P. Murugathas ³, G. Katulanda ²

¹Lady Ridgeway Hospital for children, Sri Lanka

BREAKING THE CYCLE OF LAB ERRORS: A DYNAMIC STRATEGY TO MINIMIZE PRE-ANALYTICAL ERRORS

S.R. Santos ², M. Ferreira ¹, O. Rodrigues ², C. Gonçalves ², M. Macedo ², C. Rebelo ², B. Araújo ², A. Mesquita ²

¹Escola de Medicina da Universidade do Minho

²Unidade Local de Saúde de Braga

BACKGROUND-AIM

Continuous education for healthcare professionals is essential given the rapid evolution of scientific evidence, with new studies, methodologies, and guidelines emerging daily. Delayed updates risk suboptimal patient care, potentially resulting in adverse outcomes. This project aimed to develop an engaging, iterative training program for healthcare professionals to prevent and minimize pre-analytical errors.

METHODS

The training program was implemented in a Hospital Center Unit, specifically in the Internal Medicine Service, given its multidisciplinary scope. Nurses were the initial target audience due to their predominant clinical role.

RESULTS

The workshop covered key areas of clinical pathology, hematology, microbiology, and chemistry, using discarded samples to illustrate common mistakes. Theoretical topic selection followed a bottom-up approach, beginning with a broad subject and progressively narrowing down to specific themes. Pre-analytical errors were selected for exemplification due to their frequency and clinical impact. The program consisted of three sessions held in small groups, fostering discussion and refined through participant feedback. A pre- and post-training questionnaire, applied one month apart, assessed knowledge. Eighty nurses completed the questionnaires, and 29 joined the sessions. Scheduled at shift end, the 30-minute format optimized concentration. Brainstorming and case-based questions fostered reflection and collaboration. Attendance improved with in-person distribution, reminders, and posters.

CONCLUSIONS

This project highlights the importance of continuous professional education in healthcare. A bottom-up, hands-on approach fostered engagement, while iterative refinements improved participation and learning. Continued implementation should reinforce best practices and reduce pre-analytical errors. Systematic, practice-oriented training led by the Pathology Service should be expanded across specialties, with content tailored to professional needs and routine procedures.

Preanalytical cases

P134

PRE-BARCODED TUBES: USER PERSPECTIVES ON CHALLENGES AND SOLUTIONS

R. Yasenkov ⁶, E. Ferretti ⁵, S. Jovičić ⁷, M. Berth ², K. Claes ⁹, M. Depoorter ³, G. Driesschaert ¹, H.J. Ruven ⁴, A. Verdonck ⁹, A. Šimundić ⁸

¹Anacura, Evergem, Belgium

²AZ Alma, Eeklo, Belgium

³Centre Hospitalier Régional de la Haute Senne, Soignies, Belgium

⁴Department of Clinical Chemistry, St. Antonius Hospital, Nieuwegein, The Netherlands

⁵Department of Global Market & Product Management, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria

⁶Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria

⁷Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria; Department for Medical Biochemistry, Faculty of Pharmacy, University of Belgrade, Serbia

⁸Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria; Faculty of Pharmacy and Biochemistry, University of Zagreb, Croatia

⁹Department of Laboratory Medicine, UZ Leuven, Leuven, Belgium

BACKGROUND-AIM

Mislabeling is a major preanalytical error in laboratory diagnostics, leading to misdiagnosis, treatment delays, and unnecessary procedures. Pre-barcoded tubes (PBT) and digital solutions address these issues by eliminating manual labeling, enabling automated identification, improving traceability, and reducing errors. Despite clear benefits, there is resistance or even concerns on the use of PBTs especially around contingency plans during system downtime.

The aim of this study was to explore how expert laboratories manage practical challenges of PBT use, focusing on contingency planning during system disruptions.

METHODS

This study employed a qualitative, expert panel approach, conducting in-depth interviews with a limited group of experienced PBT users from five institutions in Belgium and the Netherlands. The survey addressed contingency plan for barcode reader failure, LIS/HIS downtime, and power outages.

RESULTS

All institutions (n=5) reported that in the event of barcode reader failure, the primary contingency involves manual entry of tube alphanumeric codes. PBTs were recognized as safer for tube-patient association, as only one patient file opens at a time. Completing one collection before starting another is standard practice across all sites. Immediate wristband and tube scanning is mandatory in 60% (3/5) of cases. During LIS downtime, 40% used a collection module located on a separate server with offline buffers. If both LIS and HIS were down, all sites used paper documentation and had backup generators for main analyzers. In long-term outages, 80% (4/5) sent samples to partner sites. All participants emphasized the importance of staff training and SOPs to ensure continuity.

CONCLUSIONS

This study provided insights into expert practices for PBT use and contingency planning. PBTs offer superior traceability and error reduction over manual labeling, even during LIS/HIS downtimes and power outages, when supported by robust protocols.

Keywords: pre-barcoded tubes, system downtime, contingency plan

POCT and preanalytics

P135

CONTRIBUTION TO THE STUDY OF THE ACCURACY OF BLOOD GLUCOSE MEASUREMENT USING TEST STRIPS.

N. Belkaid ¹

¹FACULTY OF MEDECINE, MOULOUD MAMMERI UNVIVERSITY

BACKGROUND-AIM

Self-monitoring of blood glucose by test strips aims to improve the metabolic control of patients, and their reliability is therefore a public health concern. The objective of our work was to study the concordance between capillary and venous blood glucose according to the ISO 15197: 2013 standard.

METHODS

The study took place at the Nedir Mohammed University Hospital in Tizi-Ouzou, in 42 diabetic subjects and 40 non-diabetic subjects. Capillary blood glucose was measured by a blood glucose meter with 3 different batches of strips (Vital Check®), while venous blood glucose was determined with an automatic meter using a reference method , hexokinase (HK). To search for possible interferences, other analysis were also carried out. a complete blood count, measurements of triglycerides, uric acid, bilirubin, and HbA1c. Four blood collection tubes were collected for each subject ( lithium heparinate tube, fluoride/oxalate tube, two EDTA tube) .

RESULTS

Comparison of capillary blood glucose to venous blood glucose on heparinized blood: The results of the modified Bland and Altman comparative method show that the mean differences are negative, indicating a slight underestimation of capillary blood glucose values compared to those of heparinized venous blood in the same subject. the results show an agreement of 97.5%. The analysis of capillary blood glucose by batch demonstrated a concordance of the 3 batches of 95.12%, 97.5%, and 97.5% respectively. The results show also a slight underestimation of glucose levels in heparinized tubes compared to fluorinated tubes.

CONCLUSIONS

Analysis of the results showed that capillary blood glucose was underestimated compared to venous blood glucose.However, the difference between the two measurements is acceptable. Our study also showed that there is a difference between blood glucose levels in fluoridated and heparinised blood.

POCT and preanalytics

P136

EVALUATION OF POLYMICROBIAL UTIS CULTURAL TESTING IN SIMULATED DIABETIC URINE TRANSPORTED WITH A BORIC ACID-FREE DEVICE

V. Imperadori ¹, C. Sabelli ¹, F. Bonometti ¹

¹Copan Italia S.p.A.

BACKGROUND-AIM

Urinary Tract Infections (UTIs) are the most common infections in humans. UTIs are mainly caused by Escherichia coli, Klebsiella pneumoniae, Enterococcus faecalis, and Proteus mirabilis. Since UTIs can result in recurrent infections, high healthcare costs and significant morbidity, an efficient diagnostic process is required.

A prompt UTI diagnosis is even more important for diabetic patients. These people, compared to non-diabetic ones, show a higher susceptibility and predisposition to complicated and polymicrobial UTIs.

This study aims to evaluate Copan Urisponge™ effectiveness in transporting and preserving diabetic-like urine specimens intended for culture testing.

METHODS

SELF-COLLECTED SWABS FOR STI DIAGNOSIS: ACCURACY AND PRE-ANALYTICAL CHALLENGES

O. Grari ¹, M. Meziane ¹, I. El Yandouzi ², M. Lahmer ¹, A. Saddari ¹, S. Ezrari ², A. Maleb ¹

¹Microbiology department, Mohammed VI University Hospital, Oujda, Morocco

²Mohammed First University, Faculty of Medecine and Pharmacy, Oujda, Morocco

BACKGROUND-AIM

Sexually transmitted infections (STIs) such as Chlamydia trachomatis and Neisseria gonorrhoeae remain highly prevalent worldwide, often asymptomatic, and associated with adverse reproductive outcomes. Expanding testing strategies is critical, particularly in low- and middle-income countries where barriers include stigma, lack of privacy, and limited laboratory access. Self-collected swabs (SCS) have been proposed as an alternative to clinician-collected specimens.

METHODS

We reviewed evidence from recent systematic reviews, meta-analyses, and qualitative studies assessing the diagnostic accuracy, feasibility, and acceptability of self-collected vaginal, rectal, and pharyngeal swabs for STI testing. Special attention was given to pre-analytical concerns such as sample adequacy, user instructions, and transport conditions.

RESULTS

Meta-analyses show that SCS demonstrate high concordance with clinician-collected specimens across STIs, with comparable sensitivity and specificity for C. trachomatis, N. gonorrhoeae, Trichomonas vaginalis, and Mycoplasma genitalium. Uptake of testing increased 2–3 fold when self-collection was offered, particularly in at-home settings. Acceptability was high among diverse groups, including adolescents, college students, pregnant women, and the general population in sub-Saharan Africa. Reported advantages included privacy, convenience, and autonomy, while challenges related to uncertainty about correct technique, discomfort, and perceptions of reduced accuracy persisted. WHO now endorses SCS as part of self-care interventions to expand STI services.

CONCLUSIONS

Self-collected swabs are accurate, feasible, and acceptable alternatives for STI testing and represent a promising strategy to overcome pre-analytical barriers in both high- and low-resource settings. Future work should focus on optimizing user instructions, ensuring specimen stability during transport, and integrating SCS into routine diagnostic pathways with clear linkage to care.

POCT and preanalytics

P140

DEMOGRAPHIC ERROR IN POCT GLUCOSE ANALYSIS - DOES IT MATTER ?

B. Jackson ¹, V. Florea ¹, C. Cruise ¹, J. Fennell ¹

¹Naas General Hospital, Dublin Midlands Group, Ireland

BACKGROUND-AIM

Patient identification processes in POCT analysis is an integral part of the pre- analytical process and evaluation of error rates is required. Correct patient demographic entry at POCT glucose analysis is essential to providing an overview of inpatient glucose control in the acute hospital. Integrated glucometers with scanning capabilities and connectivity to both PAS and LIS systems exist however, system limitations which allow users to incorrectly capture other barcoded material e.g. ID badges, strip containers and other miscellaneous items, ensure incomplete patient electronic glucose records for patient evaluation.

METHODS

POINT-OF-CARE HEMOSTATIC TESTING IN CARDIAC SURGERY: POSTOPERATIVE COMPLICATIONS

I. Rodríguez Martín ¹, A. Delgado Hernandez ¹, D. Nuñez Jurado ²

¹Hospital Infanta Elena

²Hospital Universitario Virgen Macarena

BACKGROUND-AIM

Viscoelastic tests (rotational thromboelastometry, ROTEM®), together with the implementation of a specific algorithm for coagulation management in cardiac surgery, enable perioperative coagulopathy to be better controlled. Blood products are related to worse cardiac outcomes (hypotension or hypertension, Transfusion Associated Circulatory Overload TACO, transfusion related acute lung injury/TRALI).

METHODS

Retrospective cohort study including 675 patients who underwent cardiac surgery with cardiopulmonary bypass. The incidence of allogeneic blood transfusions and clinical postoperative complications were analyzed before and after ROTEM® implementation.

RESULTS

Following viscoelastic testing and the implementation of a specific algorithm for coagulation management, the incidence of any allogeneic blood transfusion decreased (41.4% vs 31.9%, p=0.026) during the perioperative period. In the group monitored with ROTEM®, decreased incidence of transfusion was observed for packed red blood cells (31.3% vs 19.8%, p=0.002), fresh frozen plasma (9.8% vs 3.8%, p=0.008). In addition, significant reductions were detected in the incidence of cardiac complications: cardiogenic shock (3.9% vs 3.5%, p=0.354), tamponade cardiac (3.0% vs 1.2%, p=0.091), severe arterial hypertension (2.1% vs 0.9%, p=0.155), hypotension (10.1% vs 9.1%, p=0.325), heart failure (9.2% vs 7.4%, p=0.237), arrhythmias (44.0% vs 43.7%, p=0.353), atrial fibrillation (34.5% vs 31.9%, p=0.249), ventricular arrhythmias (3.6% vs 2.7%, p=0.283), arrest cardiorespiratory (2.1% vs 0.9%, p=0.155) and length of Intensive Care Unit (ICU) stay (6.0 days vs 5.3 days, p=0.026).

CONCLUSIONS

The monitoring of hemostasis by ROTEM® in cardiac surgery, was associated with decreased incidence of allogeneic blood transfusion, clinical cardiac postoperative complications (including better hypo/hypertension control) and lengths of ICU stay.

POCT and preanalytics

P144

EVALUATION OF PATIENT-BASED REAL-TIME QUALITY CONTROL IN COMPARATIVE ASSAYS FOR COMMON CLINICAL ANALYTES

A.K. Sah ¹, R.H. Elshaikh ¹

¹Department of Medical Laboratory Sciences, College of Applied and Health Sciences, A’ Sharqiyah University, Ibra, Oman.

BACKGROUND-AIM

Routine multianalyzer comparisons are essential for laboratories to enhance quality management in test systems. This study investigates the use of patient-based real-time quality controls (PBRTQCs) in comparative assays to ensure consistency and reliability across clinical laboratories.

METHODS

This study analyzed 11 commonly tested analytes using three different analyzers. PBRTQC procedures were implemented with exponentially weighted moving average (EWMA) algorithms and assessed through the AI-MA artificial intelligence platform. Comparative assays were conducted on serum samples, with patient data categorized into total patient (TP), inpatient (IP), and outpatient (OP) groups.

RESULTS

Optimal PBRTQC protocols were assessed and selected based on appropriate truncation limits and smoothing factors. Both the EWMA and median methods demonstrated comparable performance in comparative assays. Strong consistency was observed between patient data and serum sample results, while unacceptable bias was identified for alkaline phosphatase (ALP) and gamma-glutamyl transferase (GGT) when using analyzer C. Categorizing patient data and applying specific groups for comparative assays significantly enhanced PBRTQC performance. When monitoring inter- and intra-analyzer stability daily, EWMA proved superior in detecting subtle quality-related changes while minimizing false-positive alarms.

CONCLUSIONS

Our findings suggest that PBRTQCs are a valuable tool for efficiently assessing multianalyzer comparability. To enhance accuracy and reliability, laboratories should consider population variations related to both analytes and analyzers when developing optimized PBRTQC protocols.

Key words: preanalytical interference, laboratory medicine, digital application

Artificial intelligence and preanalitics

P151

DEEP LEARNING-BASED CLASSIFICATION OF SPUN AND UNSPUN BLOOD TUBES

G. Gioiello ³, G. Montesano ³, G. Maccioni ¹, M. Martines ¹, P. Caropreso ², G. Mengozzi ³

DIGITAL APPROACHES USING METADATA AND CARDIOVASCULAR RISK: STANDARDIZATION AND HARMONIZATION OF RESULTS TO ENSURE ACCURATE LABORATORY EXAMS (PADUA-LAB)

I. Talli ¹, C. Cosma ¹, E. Pangrazzi ², A. Padoan ¹

¹Department of Medicine – DIMED, University of Padua, Italy

²Qi.Lab.Med., Spin-off of the University of Padua, Italy

BACKGROUND-AIM

While analytical procedures are often standardized, pre- and post-analytical phases data remain poorly coded, despite being potentially available for usage. The project aims to standardize information related to the Total Testing Process (TTP) to improve result harmonization across laboratories, particularly in cardiovascular (CV) risk management. A proposed solution is to develop a standard code for pre/post-analytical quality to enhance result reliability and comparability.

METHODS

Through extensive and systematic literature research, laboratory tests related to CV risk (glucose, aldosterone, renin, lipid profile, apolipoprotein A, glycated hemoglobin) were identified. Pre-analytical and post-analytical parameters were defined in order to be standardized and harmonized according to the data available in the literature.

RESULTS

All conditions that may occur for the selected parameters in the laboratory process were considered and subsequently encoded in a standard coding system, which should be easily transferable through an information system or AI. This application standard consists of a sequence of numbers and letters referring to specific procedural details, enabling accurate description of each step of the total testing process (TTP). For example, two samples transported for 20 minutes, one at 20°C and the other at 4°C, are coded as T20C20 and T20C04, respectively.

CONCLUSIONS

The development of a standard code to encode all variables impacting the pre-analytical phase is of utmost importance to guarantee accurate results in the management of CV risk in the territory. The developed code includes the most impacting pre- and post-analytical variables, such as type of collection tube, patient conditions, interfering therapies, transport, and centrifugation procedures. Combined with the LOINC, this code will allow monitoring of samples throughout all phases of the TTP, harmonizing the most clinically relevant steps across laboratories to ensure better test comparability in time and space.

Artificial intelligence and preanalitics

P155

TAYA – AN AI-BASED KNOWLEDGE COMPANION FOR RISK MITIGATION IN THE PREANALYTICAL PHASE

G.F. Turkes ¹, O. Yildirim ³, M.A. Aydin ², K. Gurel ⁶, A.R. Sisman ⁴, S. Sevinc ⁵

¹Ankara University, Faculty of Medicine, Department of Medical Biochemistry, Ankara

²Dokuz Eylul University, Department of Computer Science, Izmir

³Dokuz Eylul University, Faculty of Engineering, Department of Computer Engineering, Izmir

⁴Dokuz Eylul University, Faculty of Medicine, Department of Medical Biochemistry, Izmir

⁵Dokuz Eylul University, Labenko Informatic Inc., Izmir

⁶Izmir Institute of Technology, Department of Electronics and Communications, Izmir

BACKGROUND-AIM

The preanalytical phase is the most error-prone segment of the total testing process and can account for up to 70% of laboratory errors. These include mislabeling, incorrect tube selection, and suboptimal handling, which degrade diagnostic accuracy, delay reporting, and increase costs. Our study aims to describe the design and content curation of TAYA, an AI-powered knowledge platform for preanalytical decision support. We describe TAYA, an AI-assisted knowledge platform that provides evidence-based, practical guidance for identifying and mitigating preanalytical risks.

METHODS

Developed by Labenko Informatic Inc. at DEPARK (Izmir, Türkiye), TAYA uses a custom GPT-based interface integrating a curated corpus of ≥300 sources from high-impact journals and authoritative guidelines. Expert screening ensures methodological quality and traceability. The platform synthesizes literature evidence, guideline recommendations, and case-based insights into concise, source-linked outputs. An accompanying image set of common preanalytical errors enables scenario-specific customization and decision support.

RESULTS

Preliminary applications demonstrate TAYA can: (1) increase awareness of high-frequency preanalytical errors, (2) provide real-time recommendations to preserve sample integrity, and (3) translate technical risks into managerial insights by outlining diagnostic and economic implications.

CONCLUSIONS

TAYA exemplifies a configurable, custom GPT-based architecture that institutions can replicate to develop center-specific assistants for training and routine operations. As a reference implementation for preanalytical quality assurance, TAYA represents a novel approach to preanalytical quality assurance, functioning as both a guardian of sample integrity and an advisor to laboratory leadership. It links preanalytical vulnerabilities to diagnostic reliability and cost efficiency, supporting improved patient safety and optimized resource allocation.

Artificial intelligence and preanalitics

P156

DEVELOPMENT OF AN HIL INDEX ANALYSIS FOR THE KANKA SAMPLE-ACCEPTANCE DEVICE: A PRELIMINARY STUDY

G.F. Turkes ¹, O. Yildirim ², A.R. Sisman ³, S. Sevinc ⁴

¹Ankara University, Faculty of Medicine, Department of Medical Biochemistry, Ankara

²Dokuz Eylul University, Faculty of Engineering, Department of Computer Engineering, Izmir

³Dokuz Eylul University, Faculty of Medicine, Department of Medical Biochemistry, Izmir

⁴Dokuz Eylul University, Labenko Informatic Inc., Izmir

BACKGROUND-AIM

Preanalytical errors constitute the majority of laboratory errors, with hemolysis being the most common. Conventional detection of hemolysis, icterus, and lipemia (HIL) relies on analyzer-derived indices, which can prolong turnaround time (TAT), necessitate recollection, and increase costs. We aimed to shift HIL detection to the point of sample acceptance by integrating AI-based image analysis into the KANKA sample-acceptance device.

METHODS

We designed 3D fixtures to standardize photography of blood collection tubes and developed a convolutional image-classification pipeline initialized with pretrained models. The model was trained on ∼150,000 serum-tube images, with HIL categories assigned from visual HIL indices. External validation used multi-angle, de-identified serum-tube images collected from laboratories. Reference HIL status was established by visual assessment. The model achieved training accuracy of 0.99 and validation accuracy of ∼1.00, indicating excellent generalization. Agreement analyses compared the AI model with two independent expert visual assessments.

RESULTS

Among 205 patient samples collected from laboratory waste streams, 97 were HIL-negative and 108 had HIL>0 by the reference method. Substantial agreement was observed between expert visual assessment and analyzer-reported HIL indices (κ=0.738; total accuracy=88%).

CONCLUSIONS

These preliminary findings suggest that reliable triage of HIL status is feasible before tubes reach the analyzer, potentially reducing unnecessary analyzer workload and preventing release of poor-quality results to clinicians. This prototype represents a first step toward integrating AI-driven HIL assessment into the KANKA device (Labenko Informatics Inc., Türkiye). Future work will adopt analyzer-derived HIL indices (semi-quantitative) as the primary reference, expand datasets, and rigorously characterize analytical performance to support full integration and routine use.

Artificial intelligence and preanalitics

P157

TAYA – AN AI-BASED KNOWLEDGE COMPANION FOR RISK MITIGATION IN THE PREANALYTICAL PHASE

G.F. Turkes ¹, O. Yildirim ⁴, M.A. Aydin ³, K. Gurel ⁶, A.R. Sisman ⁵, S. Sevinc ²

¹Ankara University, Faculty of Medicine, Department of Medical Biochemistry, Ankara

²DEPARK, Labenko Informatic Inc., Izmir

³Dokuz Eylul University, Department of Computer Science, Izmir

⁴Dokuz Eylul University, Faculty of Engineering, Department of Computer Engineering, Izmir

⁵Dokuz Eylul University, Faculty of Medicine, Department of Medical Biochemistry, Izmir

⁶Izmir Institute of Technology, Department of Electronics and Communications, Izmir

BACKGROUND-AIM

The preanalytical phase is the most error-prone segment of the total testing process and can account for up to 70% of laboratory errors. These include mislabeling, incorrect tube selection, and suboptimal handling, which degrade diagnostic accuracy, delay reporting, and increase costs. Our study aims to describe the design and content curation of TAYA, an AI-powered knowledge platform for practical guidance for identifying and mitigating preanalytical risks, and preanalytical decision support.

METHODS

Developed by Labenko Informatic Inc. at DEPARK (Izmir, Türkiye), TAYA uses a custom GPT-based interface integrating a curated corpus of ≥300 sources from high-impact journals and authoritative guidelines. Expert screening ensures methodological quality and traceability. The platform synthesizes literature evidence, guideline recommendations, and case-based insights into concise, source-linked outputs. An accompanying image set of common preanalytical errors enables scenario-specific customization and decision support.

RESULTS

Preliminary applications demonstrate TAYA can: (i) increase awareness of high-frequency preanalytical errors; (ii) provide real-time recommendations to preserve sample integrity; and (iii) translate technical risks into managerial insights by outlining diagnostic and economic implications.

CONCLUSIONS

TAYA exemplifies a configurable, custom GPT-based architecture that institutions can replicate to develop centre-specific assistants for training and routine operations. As a reference implementation for preanalytical quality assurance, TAYA represents a novel approach to preanalytical quality assurance, functioning as both a guardian of sample integrity and an advisor to laboratory leadership. It links preanalytical vulnerabilities to diagnostic reliability and cost efficiency, supporting improved patient safety and optimized resource allocation.

Keywords: Artificial Intelligence, preanalytical risk management, diagnostic accuracy

Artificial intelligence and preanalitics

P158

DEVELOPMENT OF AN HIL INDEX ANALYSIS FOR THE KANKA SAMPLE-ACCEPTANCE DEVICE: A PRELIMINARY STUDY

G.F. Turkes ¹, O. Yildirim ³, A.R. Sisman ⁴, S. Sevinc ²

¹Ankara University, Faculty of Medicine, Department of Medical Biochemistry, Ankara

IMPORTANCE OF VENIPUNCTURE TRAINING IN HEMATOLOGY AND CLINICAL BIOCHEMISTRY IN PHARMACY UNDERGRADUATE STUDENTS

C. Alba Betancourt ¹, M.A. Deveze Álvarez ¹, C.L. Mendoza Macías ¹, F.B. Ramírez Montiel ¹

¹University of Guanajuato

BACKGROUND-AIM

The awareness of the venipuncture preanalytical stage to pharmacy undergraduate students (PUS) will favor that in the labor field these errors are minimized and thus subject the patient to less stress, improve the development of the following quality control stages, achieve efficiency in the processes, reliability of results, and minimizing costs

METHODS

The performance of 200 PUS was evaluated from August 2024 to May 2025, covering two educational semesters, from the beginning of their training in the areas of hematology and clinical biochemistry. Performance was assessed based on:

Sampling errors, incidence of accidental needle sticks injuries, use of wrong additives, hemolyzed samples, hematoma formation.

Elements that involve a critical waste of resources, discomfort in patients, as well as invalidation of the following stages of QC.

RESULTS

During each semester, PUS performed 13 laboratory practices in which blood samples were taken. At the beginning, the mistakes resulted in 90% more material waste, increased Bio-Hazardous Waste (BHW, 100% more), increased hematoma formation (60%), hemolyzed samples (50%), incorrect test tube additive (70%). PUS were aware of the importance of needlesticks injuries and were instructed to dispose the used needle immediately, resulting in a 100% decrease.

Timely intervention and the awareness of PUS of the importance of the pre-analytical stage allowed a decrease in errors: 80% of PUS were able to obtain a blood sample with a vacuum blood collection system with one puncture, an extra 30% of wasted material was obtained as well as an extra 25% of BHW disposal, and 70% reduction in hematoma formation.

CONCLUSIONS

At the end of the intervention, the incidence of errors detected in the venipuncture pre-analytical stage decreased by 75%, thus achieving the improvement of the PUS’ skills which is reflected in a better performance at school and subsequently in the labor field, causing less harm to patients, as well as an optimization of university resources.

Errors in pre-analytical phase

CONCLUSIONS

RepSeq showed excellent reproducibility in clonotype number and distribution, with higher technical imprecision affecting clonotype identity. Sequencing itself had minimal impact on variability, whereas preanalytical steps introduced significant noise, especially for low-frequency clonotypes. Standardization of preanalytical processes is therefore critical to ensure reliable immune repertoire analysis.

Errors in pre-analytical phase

P165

RISK ANALYSIS RELATED TO THE PRE-ANALYTICAL PHASE WITHIN A BIOCHEMISTRY LABORATORY

M. Ayoub ¹, S. Aboulkacem ¹, Z. Aouni ¹, A. Ba ¹, C. Mazigh ¹

¹BIOCHEMESTRY DEPARTEMENT OF MILITARY HOSPITAL OF TUNIS

BACKGROUND-AIM

According to ISO 15189: 2022, the LBM must implement a risk prevention strategy that may affect its various processes in order to ensure the safety of staff and patients.

MONITORING PRE-ANALYTICAL NONCONFORMITIES: INSIGHTS FROM A HIGH-VOLUME LABORATORY NETWORK

A. Barbullushi ¹, L. Kolaneci ¹

¹3P Life Logistic

BACKGROUND-AIM

The preanalytical phase is the most error-prone component of the total testing process, with significant implications for patient safety and laboratory efficiency. Continuous surveillance of preanalytical nonconformities is essential for quality improvement and compliance with ISO

METHODS

A retrospective analysis was conducted on all routine samples received at 3P Life Logistic laboratory during 2024. A total of 461,376 patients were evaluated for preanalytical errors. Data were extracted from the laboratory information system and categorized into nine error groups: clotted samples, hemolysis, lipemic samples, duplicate barcodes, empty containers, inappropriate samples, insufficient volume, improper storage at the phlebotomy site, and containers without barcodes. Percentages were calculated relative to the total sample volume.

RESULTS

Overall, 10,891 preanalytical errors were documented (2.36% of total samples). The detailed distribution is presented in the table below.

Type of Pre-analytical Error Number Percentage

PRE-ANALYTICAL PHASE AND OGTT: DIFFERENT STRATEGIES IN ITALIAN SETTINGS

M. Carta ⁵, A. Vero ², M. Montagnana ⁴, I. Cataldo ⁶, A. Ceka ³, G. Bonetti ¹

¹Laboratorio Analisi ASST –Valcamonica

²Laboratorio di Analisi chimico cliniche, AOU Catanzaro

³SOD Medicina di Laboratorio, AOU delle Marche, Ancona

⁴UOC Medicina di Laboratorio, AOU Padova

ERRORS OF PRE-ANALYTICAL PHASE GENERATED AT EMERGENCY ROOM OF LARGE CLINICAL HOSPITAL DURING ONE YEAR

O. Ciepiela ¹, A. Rodziewicz-Lurzynska ²

¹Medical University of Warsaw

²University Clinical Center of Medical University of Warsaw

BACKGROUND-AIM

Laboratory is responsible for monitoring and reporting errors of preanalytical phase and undertaking corrective actions, when applicable. At Central Laboratory of University Clinical Center of Medical University of Warsaw we report preanalytical errors every quarter of the year and, when necessary, conduct additional training for nurses. The aim of this study was to analyze frequency of errors in preanalytical phase from Emergency Room which is responsible for more than 10% of all registered laboratory test in the hospital.

METHODS

In 2024 there were total of 3,100,000 laboratory tests, and 0.59% of them were affected by preanalytical error. Noted errors include: incorrect referral, erroneous registration, doubled referral, patients mismatch, incorrect test tube, lack of sample, unnecessary sample, incorrect volume of sample, hemolysis, blood clot, contamination with infusion fluid, exceeded transport time and lack of transporting container.

RESULTS

In 2024 there were total of 320,285 laboratory tests from ER and 2166 of them were affected by preanalytical error (0.68%). The most frequent error (1243) was a lack of sample that was already registered in HIS. Total number of 380 anticoagulated blood samples contained clot, and 244 blood samples were hemolyzed. The highest frequency of errors was reported in the second quarter of the year (0.72%) and the lowest in the third quartile (0.64%). There was an increasing number of test ordered from ER during the year (I-IV quarter: 76096; 80653; 81137; 82399) but the frequency of preanalytical errors was rather stable (0.66%, 0.72%, 0.64% and 0.69% respectively).

CONCLUSIONS

The frequency of errors of preanalytical phase from ER is higher than from all other hospital wards, however the majority of them refer to lack of sample, what results from difficulties in blood collection from patients entering to ER. Despite regular training for phlebotomists, frequency of error stays stable, mostly due to high turnover of ER personnel responsible for blood collection.

Errors in pre-analytical phase

P178

MANAGEMENT OF PRE-ANALYTICAL INCIDENTS IN THE CLINICAL LABORATORY IN PRIMARY CARE

S.A. Cristina ¹, S.S. Jorge ¹, C.G. Laura ¹

¹Hospital Universitario de Mostoles

BACKGROUND-AIM

There are three stages in a clinical analysis. The most critical one is the pre-analitycal, where around 70% of errors occurs. These incidents have impact on the delay of the results, the work overload and the cost overrun.

METHODS

The laboratory has registered five commons errors between serum and urine samples during 2024 associated with the 7 centres in primary care associated with this hospital. To get to know the most commons ones and if needed display different strategies to improve this inconvenient.

The incidents studied are: (1) % identification error, (2) % serum samples not received, (3) % insufficient serum, (4) % orine simple not received, (5) % haemolysed sample.

The specification followed were the ones provided by SEMEDLAB (Sociedad Española de Medicina de Laboratorio). The intended objectives for each incident are: (1) 0.029%; (2) 0.256%; (3) 0.037%; (4) 0.748%; (5) 0.465%.

RESULTS

Among the seven centres, two fulfilled all the indicators, three of them failed one indicator (5); one of them did not accomplished one (1); and the last one did not succeeded in three parameters (1,3,5).

CONCLUSIONS

Each primary care centre were given a personalized report of the most common incidents and compared them with the other centres along with some personalized recommendations.

With this reports, the pre-analytical phase have been standardized and harmonized. The hospital and the primary care centres have improved in their communications, have lowered the risk, and improved the safety of the patient.

One limitation that we encounter is that, some incidents were manualy registered so we may have infraestimated some of them. In the future, we are trying to automathise with Informatic laboratory System this question so we avoid human error.

KEY WORDS: pre-analytical incidents, primary care, biochemistry lab

Errors in pre-analytical phase

P179

PREANALYTICAL QUALITY INDICATORS: TOWARD A DIALOGUE BETWEEN LABORATORIES AND BIOBANKS

E. Czekuc-Kryskiewicz ¹, J. Kozlowska ²

¹Biobank, The Children’s Memorial Health Institute, Warsaw, Poland

²Pediatric Clinical Trials Support Center, The Children’s Memorial Health Institute, Warsaw, Poland

BACKGROUND-AIM

The preanalytical phase remains a hidden but decisive factor in research reliability. Clinical laboratories adopt IFCC preanalytical QIs to safeguard sample identification, collection, and handling, while biobanks rely on BBMRI recommendations to ensure biospecimen integrity over time. Both aim to minimize errors, yet the translation of lab practices to biobank settings—and vice versa—raises important questions.

METHODS

IFCC and BBMRI preanalytical QIs were systematically compared, focusing on workflows from patient/sample collection to storage. Parameters including labeling, transport, temperature monitoring, and documentation were analyzed. A hybrid framework was proposed to explore feasible integration and identify potential conflicts.

RESULTS

Convergent QIs include sample identification, handling protocols, and processing timelines. Divergent priorities emerge: laboratories emphasize immediate assay readiness, whereas biobanks focus on long-term preservation. A preliminary hybrid framework highlights 12 core preanalytical indicators that could serve both contexts. However, implementation prompts discussion: How should competing priorities be reconciled? Which indicators are universally essential versus context-specific? How can compliance be measured across diverse centers?

CONCLUSIONS

Harmonizing preanalytical QIs offers a pathway to enhanced reproducibility and research reliability, yet practical and conceptual challenges persist. By framing preanalytical quality as a shared responsibility, this work invites the community to debate trade-offs, establish consensus, and test context-adapted standards. Are we ready to rethink the preanalytical phase as a collaborative, rather than siloed, domain? Can a shared QI framework truly satisfy both immediate laboratory needs and long-term biobank goals? These questions aim to spark dialogue and guide future multi-center studies.

Errors in pre-analytical phase

In the central hospital, participants improved scores (60.9%→70.7%), while non-participants declined (47.8%→39.9%). Gains were highest in PAEs (55.2%→86.2%), collection order (55.2%→89.7%), and packaging (30.4%→100%).

In the peripheral hospital, only participants completed both evaluations. Despite no control group, scores rose (65.6%→71.4%), with progress in PAEs (14.7%→32.1%) and packaging (67.6%→85.7%), while strong baseline performance in labeling and material handling was maintained. The improvement pattern was consistent across settings, reinforcing reproducibility.

CONCLUSIONS

Structured, peer-led training improved nurses’ knowledge of pre-analytical procedures in both hospitals, especially in error-prone areas. Even without a control group, consistent trends underscored robustness. Comparing central and peripheral hospitals showed both shared benefits and context-specific differences, supporting regular, standardized, and cost-effective training to reduce errors, enhance patient safety, and promote professional development.

Errors in pre-analytical phase

P183

BLOOD SAMPLING AND CLINICAL RELATIONSHIP FROM A BIOETHICAL PERSPECTIVE: A QUALITATIVE APPROACH IN A VULNERABLE POPULATION

J.A. Duarte ¹

¹Universidad Nacional Autónoma de México

IMPACT OF TEMPERATURE ON LACTATE MEASUREMENT: A CASE REPORT

M.M. González Prado ², E. Perez Pegado ², C. González Prado ¹

¹Hospital Universitario de Burgos

²Hospital Universitario de León

BACKGROUND-AIM

The critical role of low temperature in the transport of arterial blood gas samples for reliable lactate measurement.

METHODS

Patient´s clinical history and laboratory reports.

RESULTS

A 34-year-old male with no relevant medical history presents to the Emergency Department with progressive dyspnea, fever, and productive cough. Upon arrival, he exhibits tachypnea, and a chest X-ray reveals a left-sided pleural effusion. Basic biochemistry and arterial blood gases were requested and personally delivered to the Emergency Laboratory.

The blood gas results were within normal limits except for an elevated lactate (3.8 mmol/L). However, after reviewing the patient’s medical history and hemodynamic parameters, it was decided to request a new refrigerated arterial blood gas sample. In this second sample, no significant differences were observed compared to the previous one, except for the lactate, which was within normal limits (1.2 mmol/L).

The patient was ultimately diagnosed with an uncomplicated parapneumonic pleural effusion and did not require hemodynamic support.

The difference between both measurements highlights the importance of preanalytical conditions of a sample and their impact on patient management. This case illustrates the critical importance of preanalytical handling in lactate determination. Cellular glycolysis continues after sample collection, especially if the sample is not refrigerated, which can result in false positives for hyperlactatemia and lead to inappropriate clinical decisions. Guidelines recommend analyzing the sample within the first 15 minutes or keeping it on ice to prevent metabolic degradation.

CONCLUSIONS

in recent years, it has been recommended that arterial blood gas samples not be refrigerated due to the potential for increased gas exchange. However, when lactate measurement is required, maintaining low temperatures remains crucial.

Errors in pre-analytical phase

Results show improvement in the pre-analytical process – the number of errors is decreasing both absolutely and relatively.

KEY WORDS. Preanalytical errors trends.

Errors in pre-analytical phase

Keywords: CA, RBC, MCH, MCHC, HCT.

Errors in pre-analytical phase

P196

AN OVERVIEW OF PRE-ANALYTICAL ERRORS: FREQUENCY, DISTRIBUTION AND QUALITY IMPLICATIONS, IN LABORATORY NETWORKS, ALBANIA

M. Lika ², N. Heta-Alliu ¹, E. Kapllani ³, L. Mino ¹, H. Lame ⁴, A. Coraj ⁴, V. Tole ⁴, E. Cela ⁴, E. Refatllari ¹, I. Korita ¹, O. Lena ⁵, A. Bulo ¹

¹Laboratory Department, University of Medicine, Tirana

UTILIZING 6 SIGMA METRICS IN PREANALYTICAL PHASE OF CLINICAL BIOCHEMISTRY LAB TO ASSESS QUALITY AND ERRORS IMPACTING IT

A. Mathew ¹, K. Dharwadkar ³, N. Santhosh ¹, N. Solanki ², S. Swarn ¹, K. Sreevalsan ¹

¹Assistant Professor, Department of Biochemistry, Sri Aurobindo Institute of Medical Sciences, Indore

²Associate Professor, Department of Biochemistry, Sri Aurobindo Institute of Medical Sciences, Indore

³Professor & Head, Director Clinical Biochemistry lab, Department of Biochemistry, Sri Aurobindo Institute of Medical Sciences, Indore

BACKGROUND-AIM

Preanalytical phase as per ISO 15189 starts from the time the test has been requested by the doctor and ends when analytical procedure begin. 70% of laboratory errors arise from pre-analytical phase and most of it is not under lab control but weaved between wards and departments. We aim to utilize Six sigma & Pareto principle to determine errors leading to preanalytical errors and how it impacts quality of lab.

METHODS

Study conducted retrospectively in Clinical Biochemistry lab, Sri Aurobindo Institute of Medical College, Indore. Data was collected from time period of November 2023 to July 2024 including both in-patient & out-patient data by analysing the TRF forms & rejection registers of mentioned period. DPMO was calculated & sigma values were calculated using Six sigma calculator in Westgard website. After identifying errors, a fishbone diagram & Pareto chart was prepared to identify possible causes for low six sigma parameters

RESULTS

A total of 89033 samples had been received in Clinical Biochemistry lab both from Inpatient Department and Out Patient Department during chosen time period of study out of which 12111 (13%) were identified to be total preanalytical error. Amongst all errors, lack of diagnosis, mention of requesting doctor name & sample type mention in TRF forms had lowest sigma (Sigma between 2.7-3.7) while Hemolysis contributed to having lowest sigma in sampling process error (Sigma varying from 3.7-4.4) suggesting improper sampling practices especially in ICUs & wards. A fishbone diagram was made & major possible cause of main errors in pre-analytical errors noted was newly appointed doctors & staff who were not properly trained in wards or had not been following SOPs. Pareto chart showed that common error were incomplete TRF form filing & hemolysis.

CONCLUSIONS

Pre-analytical area have highest errors & it impacts patient outcome. Utilizing Six Sigma metrics, Pareto chart & fishbone diagrams helps to identify cause of errors & improve each part of phase to ensure proper patient care

Errors in pre-analytical phase

P200

PREANALYTICAL ERRORS MONITORING TO DEFINE AN EDUCATIONAL PROGRAM IN A GRENOBLE UNIVERSITY HOSPITAL

F. Migeon ¹, J. Tonini ¹, C. Lo Presti ¹, R. Germi ¹

¹Univ. Grenoble Alpes, CHU Grenoble Alpes, C38000 Grenoble, France

BACKGROUND-AIM

Most laboratory errors occur during the preanalytical phase. Since the process is mainly carried out outside the laboratory, it is necessary to monitor the different types of errors upon sample reception. We recorded the occurrence of preanalytical errors over a 10-year period in the Grenoble University Hospital laboratory in order to better identify the training needs of nurses. Biological samples can also be collected by interns in Medical Biology; therefore, we also aimed to assess their educational needs.

METHODS

Preanalytical errors were checked and recorded electronically for each biological sample received in the laboratory. Errors were classified into five types : identification, test ordering and clinical information, compliance with the sampling plan, sample collection, and transportation. Data extraction was performed annually as an indicator of sampling quality.

Among interns, a survey was conducted to better understand the issues related to training in sample collection.

RESULTS

Over a 10-year follow-up period (2014–2024), the rate of preanalytical errors ranged from 1.97% to 2.28%. The most frequent errors were related to compliance with the sampling plan (21% in 2024 vs. 19% in 2014) and sample collection (23% in 2024 vs. 15% in 2014).

Regarding the interns’ survey, 40% reported not having received any training in sample collection during their studies, and 53% reported feeling apprehensive about performing this procedure.

CONCLUSIONS

We identified that sample collection and compliance with the sampling plan together accounted for half of the preanalytical errors reported in our University Hospital. Moreover, few training opportunities are offered to interns, who therefore feel apprehensive about performing biological sampling.

This evaluation represents a starting point for developing an educational program to train both interns and nurses using educational videos and mannequin-based training. The expected outcomes are improved sample quality and better stress management.

Errors in pre-analytical phase

P201

INVESTIGATING EXTRACELLULAR VESICLE DIVERSITY: LIQUID BIOPSY INNOVATIONS FOR NEURODEGENERATIVE DISEASE APPLICATIONS

V. Crapella ⁴, S. Mimmi ³, E. Pingitore ¹, K. Fatima ¹, C. Cristiani ³, L. Scaramuzzino ³, M. Talarico ¹, A. Quattrone ³, A. Quattrone ², E. Iaccino ¹

¹Department of Experimental and Clinical Medicine, University Magna Graecia of Catanzaro, Catanzaro

²Neuroscence Research Center, University Magna Graecia of Catanzaro, Catanzaro

³Neuroscence Research Center, Department of Medical and Surgical Science, University Magna Graecia of Catanzaro, Catanzaro

⁴University Magna Graecia of Catanzaro, Catanzaro

BACKGROUND-AIM

The clinical application of liquid biopsies based on EVs has been challenged by preanalytical variability, which compromises the reliability and consistency of results. Focusing on specific subpopulations of EVs is essential, as different neuronal types can influence their composition and functionality, offering insights into the underlying mechanisms of neurodegenerative diseases. This study aims to investigate the diagnostic potential of TGFβ1 and LAP levels within neuronal-derived extracellular vesicles (NDEVs) as biomarkers for progressive supranuclear palsy (PSP), highlighting the importance of EV diversity and integrity in clinical applications.

METHODS

We analyzed serum samples from a cohort of 33 PSP patients, 39 patients with Parkinson’s disease (PD), 8 patients with multiple system atrophy (MSA), and 50 healthy controls (HC), quantifying TGFβ1 and LAP levels using Simple Plex™ and ELISA. NDEVs were isolated through an L1CAM-based immunoaffinity method and assessed for TGFβ1 and LAP expression.

RESULTS

Initial serum analyses revealed no significant differences in TGFβ1/LAP levels among the groups. In contrast, NDEVs demonstrated strong expression of active TGFβ1 and LAP, with statistically significant differences effectively distinguishing PSP from HC, PD, and MSA.

CONCLUSIONS

The detection of TGFβ1/LAP within NDEVs highlights the innovative potential of liquid biopsies in neurodegenerative disease applications, offering a promising non-invasive biomarker for differentiating PSP from PD and MSA and improving early diagnostic accuracy. Further studies are essential to validate these findings and explore their implications for personalized treatment strategies. Integrating NDEV-derived biomarker data with clinical assessments and neuroimaging could significantly enhance patient management and outcomes in individuals with PSP and related disorders.

Errors in pre-analytical phase

P215

SAMPLE TRANSPORT TIME AND TEMPERATURE MONITORING PRACTICES IN EUROPE AND US: MULTINATIONAL SURVEY STUDY

R. Soldo ⁷, S. De Kozlowski ⁷, S. Jovičić ⁸, M. Budelier ¹¹, M. Cornes ¹², L. Dukić ³, J. Giannoli ¹, Á. González ¹⁰, G. Grzych ⁴, M. Mcguire ², M. Nybo ⁵, H.J. Ruven ⁶, A. Šimundić ⁹

¹Cerballiance Laboratory, CerbaHealthCare, AURA, Lyon, France

²Children’s National Hospital, Washington, DC

³Clinical Department of Laboratory Diagnostics, Clinical Hospital Center Rijeka, Rijeka, Croatia

⁴Department of Biochemistry, Lille University Hospital, Lille, France

⁵Department of Clinical Biochemistry, Odense University Hospital, Odense, Denmark

⁶Department of Clinical Chemistry, St. Antonius Hospital, Nieuwegein, The Netherlands

⁷Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria

⁸Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria; Department for Medical Biochemistry, Faculty of Pharmacy, University of Belgrade, Serbia

⁹Department of Global Medical and Clinical Affairs, Business Unit for Preanalytics, Greiner Bio-One GmbH, Kremsmünster, Austria; Faculty of Pharmacy and Biochemistry, University of Zagreb, Croatia

¹⁰Service of Biochemistry, Clínica Universidad de Navarra, Pamplona, Spain

¹¹TriCore Reference Laboratories, Albuquerque, New Mexico, USA

¹²Worcestershire Acute Hospitals NHS Trust, Worcester, UK

IMPROVING GLUCOSE STABILITY AND REDUCING HEMOLYSIS IN PRIMARY CARE: EVALUATION OF INNOMED BLOOD COLLECTION TUBES UNDER TRANSPORT CONDITIONS

K. Moonla ³, P. Poljaroentanakit ¹, R. Meesri ¹, R. Wiriyaprasit ³, W. Theansun ¹, W. Treebuphachatsakul ²

¹Departement of Medical Technology, Faculty of Allied Health Sciences, Naresuan University, Phitsanulok, Thailand

²Reference Material and Medical Laboratory Innovation Research Unit, Department of Medical Technology, Faculty of Allied Health Sciences, Naresuan University, Phitsanulok, Thailand

³Reference Material and Medical Laboratory Innovation Research Unit, Faculty of Allied Health Sciences, Naresuan University, Phitsanulok, Thailand

BACKGROUND-AIM

Pre-analytical errors are the main source of laboratory inaccuracies, particularly in primary care where blood samples must be transported to referral hospitals. Hemolysis and glycolysis critically affect glucose and biochemical accuracy, potentially leading to misclassification of prediabetes and diabetes. In Thailand, venous blood is routinely collected at subdistrict health-promoting hospitals (RPHs) and transported to hub hospitals under varying distances, packaging, and temperature conditions, further influencing sample stability.

METHODS

This study investigated the performance of Innomed blood collection tubes coated with lithium heparin and antiglycolytic agents in maintaining glucose stability and reducing hemolysis during transport. It comprised three parts: (1) a survey of 251 RPHs in Health Region 2 on current blood collection and transport practices; (2) a comparison of hemolysis index, LDH, AST, and potassium levels in samples collected with Innomed versus conventional lithium-heparin tubes under real and simulated conditions; and (3) an assessment of glucose stability in samples from normoglycemic, prediabetic, and diabetic individuals.

RESULTS

Survey results showed wide variation in transport practices, with many exceeding two hours, increasing risk of pre-analytical errors. Hemolysis indices were generally similar between Innomed and lithium-heparin tubes, though potassium levels varied with time and temperature. For glucose stability, sodium fluoride performed best, but Innomed preserved glucose more effectively than lithium-heparin, especially at ∼25°C for up to eight hours. At 42°C, degradation increased in all tubes, though Innomed still outperformed lithium-heparin.

CONCLUSIONS

In conclusion, Innomed tubes, combined with strict temperature control and reduced transport time, offer a practical option to minimize pre-analytical errors and improve diagnostic confidence in primary care.

Innovations in blood collection

P226

BCMA TARGETING STRATEGIES IN B-CELL MALIGNANCIES: ADVANCEMENTS IN BIOMARKER DISCOVERY FOR ENHANCED PATIENT OUTCOMES

E. Pingitore ², K. Fatima ², V. Crapella ⁵, S. Mimmi ⁵, D. Gramigna ⁴, A. Quinto ⁴, A.M. Tolomeo ¹, D. Maisano ³, S. Ciavarella ⁴, E. Iaccino ²

¹Department of Cardiac, Thoracic and Vascular Science and Public Health, University of Padova, Padua, Italy

²Department of Experimental and Clinical Medicine, University Magna Graecia of Catanzaro, Catanzaro, Italy

³Harvard Medical School, Boston, MA, USA

⁴Istituto Tumori Giovanni Paolo II, Bari, Italy

⁵Neuroscience Research Center, University Magna Graecia of Catanzaro, Catanzaro, Italy

BACKGROUND-AIM

Lymphoproliferative disorders, including multiple myeloma (MM), are marked by uncontrolled lymphocyte proliferation and present clinical challenges due to biological heterogeneity, therapy resistance, and limited biomarkers. Despite advances in monoclonal antibodies, bispecific antibodies, and CAR-T therapies, MM remains largely incurable. B-cell maturation antigen (BCMA), highly expressed on malignant plasma cells, is a key therapeutic target and a promising biomarker. This study investigates soluble BCMA (sBCMA) and BCMA-positive extracellular vesicles (EVs) as circulating biomarkers in MM patients receiving anti-BCMA therapies, and explores high-affinity BCMA-binding peptides for potential drug delivery and immunotherapy applications.

METHODS

Serum samples from 20 MM patients were collected longitudinally. EVs were isolated and analyzed for BCMA expression, and sBCMA levels were quantified. A phage display library (∼10⁹ clones) was screened on BCMA-expressing H929 cells over four biopanning rounds. High-affinity clones were identified by ELISA and flow cytometry, sequenced, and structurally modeled with molecular docking to predict BCMA interactions. Synthetic peptides were validated for specificity and affinity.

RESULTS

Results showed dynamic changes in sBCMA and EV-associated BCMA during treatment, suggesting their utility as real-time indicators of therapy response. Phage screening identified 20 clones, two of which shared a peptide motif with high BCMA-binding affinity. Computational modeling confirmed strong and specific interactions with BCMA, validated experimentally.

CONCLUSIONS

This study highlights BCMA’s dual role as a circulating biomarker and therapeutic target in MM. Monitoring sBCMA and BCMA-positive EVs may improve patient response assessment, while the novel BCMA-binding peptide offers potential for targeted drug delivery and next-generation immunotherapy. Prospective studies are needed to validate these findings and support clinical translation.

Self-sampling

P227

COMPARISON OF PREANALYTICAL INDICATORS IN PATIENT- AND PHLEBOTOMIST-COLLECTED CAPILLARY BLOOD

S. Calleja ¹, J. Maroto ¹, S. Deza ¹, N. Zabalza ¹, M.Á. Arias ¹, C. Gómez ¹, N. Varo ¹, Á. Gónzalez ¹

¹Clínica Universidad de Navarra

BACKGROUND-AIM

Liquid capillary blood sampling can raised as an alternative to venous sampling for clinical analysis. However, pre-analytical should be enough robust to be integrated in the laboratory routine. In this study, indicators of preanalytical quality were compared between capillary blood samples self-collected by patients or collected by a phlebotomist.

METHODS

Capillary blood samples were obtained from the fingertip using a lancet, comprising 1,177 self-collected specimens and 344 collected by phlebotomists . The quality of the preanalytical phase was assessed by applying the Model of Quality Indicator (MQI) (i.e., inappropriate sample type, insufficient volume and samples rejected by hemolysis), by assessing the hemolysis (H) and lipemia (L) indices in a Cobas analyzers (Roche Diagnostics). Statistical analyses were performed using SPSS Statistics.

RESULTS

Enough blood (250 µL) could not be obtained in 1.86 % of self-collected samples and in 2.23 % of phlebotomist-collected samples. Their performance was similar according to MQI, with no statistically significant differences observed between the groups (p=1.00). H and L indices were higher in self-collected samples compared to those collected by a phlebotomist (H: median 0.44 vs. 0.28 g/L; p<0.001; L: median 0.12 vs. 0.08 g/L; p<0.001). Higher rejection rates of samples due to H>0.5 g/L and H>1 g/L to self-collected samples compared with those by phlebotomist (46 % vs 28 %, p<0.001; 28% vs 7%, p<0.001; respectively).

CONCLUSIONS

Capillary blood samples collected by patients can be highly valuable for parameters less interfered to the hemolysis index. No differences are observed compared to capillary samples collected by a phlebotomist in terms of preanalytical indicators, such as adequate volume or sample suitability.

Self-sampling

P228

EFFECT OF THE COLLECTION SITE AND THE TYPE OF MATRIX IN THE QUALITY OF CAPILLARY BLOOD SAMPLE

S. Deza Casquero ¹, N. Zabalza Rodrigo ¹, M.A. Arias Falagán ¹, C. Gómez Gómez ¹, A. Fernández Montero ¹, N. Orbegozo Redín ¹, N. Varo Cenarruzabeitia ¹, Á. González Hernández ¹

¹Clínica Universidad de Navarra

BACKGROUND-AIM

Capillary blood collection offers advantages over venous blood, such as self-collection and serving as an alternative. Capillary blood flow differs between the most common collection sites, fingertip and upper arm, potentially affecting collected blood volume and the degree of hemolysis. Hemolysis (H) may also vary by matrix (serum vs plasma). We evaluated capillary sample quality by puncture site and matrix.

METHODS

We recruited 480 patients provided informed consent. For each patient we collected venous serum and heparinized plasma samples and two of capillary samples. Capillary blood was obtained from the upper arm using TAP Micro Select and from the ring fingertip using a lancet. Three of capillary matrices were studied: serum (670 samples), heparin plasma (130), and K2EDTA plasma (160). Capillary sample quantity was weighed with a precision balance. H index was measured by spectrophotometry. Wilcoxon and Kruskal-Wallis tests were used.

RESULTS

Mean amount of capillary blood from the finger was 314 mg (IQR: 255-375), and from the arm was 203 mg (IQR: 125-300; p<0.01). Also, no sample could be obtained from the arm in 38 patients (8%), and from the finger in one patient (0.2%). H was lower in venous blood samples compared to capillary blood samples, both in serum and plasma (p<0.01). H in finger serum was higher than in arm serum (p<0.01). No differences were observed between both types of capillary plasma either from finger or arm. Considering a cut-off of 90 mg/dL for H index, capillary serum would be rejected in 43% of finger samples and 20% of arm samples.

CONCLUSIONS

Capillary blood quality varies between the arm and finger, and is lower than that of venous blood. No differences were found between matrices.

Key words: capillary blood, sample quality, hemolysis, volumen.

Acknowledgement to Roche Diagnostics for their support.

These abstracts have been reproduced directly from the material supplied by the authors, without editorial alteration by the staff of this Journal. Insufficiencies of preparation, grammar, spelling, style, syntax and usage are the authors’ responsibility.

Published Online: 2025-11-30

Published in Print: 2025-11-25

Articles in the same Issue

https://doi.org/10.1515/cclm-2025-1331